OBG Management

Hartwig TJ, Ambye L, Gruhn JR, et al. Cell-free fetal DNA for genetic evaluation in Copenhagen Pregnancy Loss Study (COPL): a prospective cohort study. Lancet. 2023;401:762-771. https://doi.org/10.1016/S0140-6736(22)02610-1.

Expert Commentary

A devastating outcome for women, pregnancy loss is directly proportional to maternal age, estimated to occur in approximately 15% of clinically recognized pregnancies and 30% of preclinical pregnancies.¹ Approximately 80% of pregnancy losses occur in the first trimester.² The frequency of clinically recognized early pregnancy loss for women aged 20–30 years is 9% to 17%, and these rates increase sharply, from 20% at age 35 years to 40% at age 40 years, and 80% at age 45 years. Recurrent pregnancy loss (RPL), defined as the spontaneous loss of 2 or more clinically recognized pregnancies, affects less than 5% of women.³ Genetic testing using chromosomal microarray analysis (CMA) has identified aneuploidy in about 55% of cases of miscarriage.⁴

Following ASRM guidelines for the evaluation of RPL, which consists of analyzing parental chromosomal abnormalities, congenital and acquired uterine anomalies, endocrine imbalances, and autoimmune factors (including antiphospholipid syndrome), no explainable cause is determined in 50% of cases.³ Recently, it has been shown that more than 90% of patients with RPL will have a probable or definitive cause identified when CMA testing on miscarriage tissue with the ASRM evaluation guidelines.⁵

Details of the study

In this prospective cohort study from Denmark, the authors analyzed maternal serum for cell-free fetal DNA (cffDNA) to determine the ploidy status of the pregnancy loss. One thousand women older than age 18 were included (those who demonstrated an ultrasound-confirmed intrauterine pregnancy loss prior to 22 weeks’ gestation). Maternal blood was obtained while pregnancy tissue was in situ or within 24 hours of passage of products of conception (POC), then analyzed by genome-wide sequencing of cffDNA.

For the first 333 recruited women (validation phase), direct sequencing of the POC was performed for sensitivity and specificity. Following the elimination of inconclusive samples, 302 of the 333 cases demonstrated a sensitivity of 85% and specificity of 93%. In the subsequent evaluation of 667 women, researchers analyzed maternal serum from the gestational age of fetuses ranging from 35 days to 149 days.

Results. In total, nearly 90% of cases yielded conclusive results, with 50% euploid, 46% aneuploid, and 4% multiple aneuploidies. Earlier gestational ages (less than 7 weeks) had a no-call rate (ie, inconclusive) of approximately 50% (only based on 16 patients), with results typically obtained in maternal serum following passage of POC; in pregnancies at gestational ages past 7 weeks, the no-call rate was about 10%. In general, the longer the time after the pregnancy tissue passed, the higher likelihood of a no-call result.

Applying the technology of single-nucleotide polymorphism (SNP)-based CMA can improve identification of fetal and/or maternal sources as causes of pregnancy loss with accuracy, but it does require collection of POC. Of note, samples were deficient in this study, the authors cite, in one-third of the cases. Given this limitation of collection, the authors argue for use of the noninvasive method of cffDNA, obtained from maternal serum.

Study strengths and weaknesses

Several weaknesses of this study are highlighted. Of the validation cohort, one-third of pregnancy tissue could not be analyzed due to insufficient collection. Only 73% of cases allowed for DNA isolation from fetal tissue or chorionic villi; in 27% of cases samples were labeled “unknown tissue.” In those cases classified as unknown, 70% were further determined to be maternal. When all female and monosomy cases were excluded in an effort to assuredly reduce the risk of contamination with maternal DNA, sensitivity of the cffDNA testing process declined to 78%. Another limitation was the required short window for maternal blood sampling (within 24 hours) and its impact on the no-call rate.

The authors note an association with later-life morbidity in patients with a history of pregnancy loss and RPL (including cardiovascular disease, type 2 diabetes, and mental health disorders), thereby arguing for cffDNA-based testing versus no causal testing; however, no treatment has been proven to be effective at reducing pregnancy loss. ●

Hartwig TJ, Ambye L, Gruhn JR, et al. Cell-free fetal DNA for genetic evaluation in Copenhagen Pregnancy Loss Study (COPL): a prospective cohort study. Lancet. 2023;401:762-771. https://doi.org/10.1016/S0140-6736(22)02610-1.

Expert Commentary

A devastating outcome for women, pregnancy loss is directly proportional to maternal age, estimated to occur in approximately 15% of clinically recognized pregnancies and 30% of preclinical pregnancies.¹ Approximately 80% of pregnancy losses occur in the first trimester.² The frequency of clinically recognized early pregnancy loss for women aged 20–30 years is 9% to 17%, and these rates increase sharply, from 20% at age 35 years to 40% at age 40 years, and 80% at age 45 years. Recurrent pregnancy loss (RPL), defined as the spontaneous loss of 2 or more clinically recognized pregnancies, affects less than 5% of women.³ Genetic testing using chromosomal microarray analysis (CMA) has identified aneuploidy in about 55% of cases of miscarriage.⁴

Following ASRM guidelines for the evaluation of RPL, which consists of analyzing parental chromosomal abnormalities, congenital and acquired uterine anomalies, endocrine imbalances, and autoimmune factors (including antiphospholipid syndrome), no explainable cause is determined in 50% of cases.³ Recently, it has been shown that more than 90% of patients with RPL will have a probable or definitive cause identified when CMA testing on miscarriage tissue with the ASRM evaluation guidelines.⁵

Details of the study

In this prospective cohort study from Denmark, the authors analyzed maternal serum for cell-free fetal DNA (cffDNA) to determine the ploidy status of the pregnancy loss. One thousand women older than age 18 were included (those who demonstrated an ultrasound-confirmed intrauterine pregnancy loss prior to 22 weeks’ gestation). Maternal blood was obtained while pregnancy tissue was in situ or within 24 hours of passage of products of conception (POC), then analyzed by genome-wide sequencing of cffDNA.

For the first 333 recruited women (validation phase), direct sequencing of the POC was performed for sensitivity and specificity. Following the elimination of inconclusive samples, 302 of the 333 cases demonstrated a sensitivity of 85% and specificity of 93%. In the subsequent evaluation of 667 women, researchers analyzed maternal serum from the gestational age of fetuses ranging from 35 days to 149 days.

Results. In total, nearly 90% of cases yielded conclusive results, with 50% euploid, 46% aneuploid, and 4% multiple aneuploidies. Earlier gestational ages (less than 7 weeks) had a no-call rate (ie, inconclusive) of approximately 50% (only based on 16 patients), with results typically obtained in maternal serum following passage of POC; in pregnancies at gestational ages past 7 weeks, the no-call rate was about 10%. In general, the longer the time after the pregnancy tissue passed, the higher likelihood of a no-call result.

Applying the technology of single-nucleotide polymorphism (SNP)-based CMA can improve identification of fetal and/or maternal sources as causes of pregnancy loss with accuracy, but it does require collection of POC. Of note, samples were deficient in this study, the authors cite, in one-third of the cases. Given this limitation of collection, the authors argue for use of the noninvasive method of cffDNA, obtained from maternal serum.

Study strengths and weaknesses

Several weaknesses of this study are highlighted. Of the validation cohort, one-third of pregnancy tissue could not be analyzed due to insufficient collection. Only 73% of cases allowed for DNA isolation from fetal tissue or chorionic villi; in 27% of cases samples were labeled “unknown tissue.” In those cases classified as unknown, 70% were further determined to be maternal. When all female and monosomy cases were excluded in an effort to assuredly reduce the risk of contamination with maternal DNA, sensitivity of the cffDNA testing process declined to 78%. Another limitation was the required short window for maternal blood sampling (within 24 hours) and its impact on the no-call rate.

The authors note an association with later-life morbidity in patients with a history of pregnancy loss and RPL (including cardiovascular disease, type 2 diabetes, and mental health disorders), thereby arguing for cffDNA-based testing versus no causal testing; however, no treatment has been proven to be effective at reducing pregnancy loss. ●

Hartwig TJ, Ambye L, Gruhn JR, et al. Cell-free fetal DNA for genetic evaluation in Copenhagen Pregnancy Loss Study (COPL): a prospective cohort study. Lancet. 2023;401:762-771. https://doi.org/10.1016/S0140-6736(22)02610-1.

Expert Commentary

A devastating outcome for women, pregnancy loss is directly proportional to maternal age, estimated to occur in approximately 15% of clinically recognized pregnancies and 30% of preclinical pregnancies.¹ Approximately 80% of pregnancy losses occur in the first trimester.² The frequency of clinically recognized early pregnancy loss for women aged 20–30 years is 9% to 17%, and these rates increase sharply, from 20% at age 35 years to 40% at age 40 years, and 80% at age 45 years. Recurrent pregnancy loss (RPL), defined as the spontaneous loss of 2 or more clinically recognized pregnancies, affects less than 5% of women.³ Genetic testing using chromosomal microarray analysis (CMA) has identified aneuploidy in about 55% of cases of miscarriage.⁴

Following ASRM guidelines for the evaluation of RPL, which consists of analyzing parental chromosomal abnormalities, congenital and acquired uterine anomalies, endocrine imbalances, and autoimmune factors (including antiphospholipid syndrome), no explainable cause is determined in 50% of cases.³ Recently, it has been shown that more than 90% of patients with RPL will have a probable or definitive cause identified when CMA testing on miscarriage tissue with the ASRM evaluation guidelines.⁵

Details of the study

In this prospective cohort study from Denmark, the authors analyzed maternal serum for cell-free fetal DNA (cffDNA) to determine the ploidy status of the pregnancy loss. One thousand women older than age 18 were included (those who demonstrated an ultrasound-confirmed intrauterine pregnancy loss prior to 22 weeks’ gestation). Maternal blood was obtained while pregnancy tissue was in situ or within 24 hours of passage of products of conception (POC), then analyzed by genome-wide sequencing of cffDNA.

For the first 333 recruited women (validation phase), direct sequencing of the POC was performed for sensitivity and specificity. Following the elimination of inconclusive samples, 302 of the 333 cases demonstrated a sensitivity of 85% and specificity of 93%. In the subsequent evaluation of 667 women, researchers analyzed maternal serum from the gestational age of fetuses ranging from 35 days to 149 days.

Results. In total, nearly 90% of cases yielded conclusive results, with 50% euploid, 46% aneuploid, and 4% multiple aneuploidies. Earlier gestational ages (less than 7 weeks) had a no-call rate (ie, inconclusive) of approximately 50% (only based on 16 patients), with results typically obtained in maternal serum following passage of POC; in pregnancies at gestational ages past 7 weeks, the no-call rate was about 10%. In general, the longer the time after the pregnancy tissue passed, the higher likelihood of a no-call result.

Applying the technology of single-nucleotide polymorphism (SNP)-based CMA can improve identification of fetal and/or maternal sources as causes of pregnancy loss with accuracy, but it does require collection of POC. Of note, samples were deficient in this study, the authors cite, in one-third of the cases. Given this limitation of collection, the authors argue for use of the noninvasive method of cffDNA, obtained from maternal serum.

Study strengths and weaknesses

Several weaknesses of this study are highlighted. Of the validation cohort, one-third of pregnancy tissue could not be analyzed due to insufficient collection. Only 73% of cases allowed for DNA isolation from fetal tissue or chorionic villi; in 27% of cases samples were labeled “unknown tissue.” In those cases classified as unknown, 70% were further determined to be maternal. When all female and monosomy cases were excluded in an effort to assuredly reduce the risk of contamination with maternal DNA, sensitivity of the cffDNA testing process declined to 78%. Another limitation was the required short window for maternal blood sampling (within 24 hours) and its impact on the no-call rate.

The authors note an association with later-life morbidity in patients with a history of pregnancy loss and RPL (including cardiovascular disease, type 2 diabetes, and mental health disorders), thereby arguing for cffDNA-based testing versus no causal testing; however, no treatment has been proven to be effective at reducing pregnancy loss. ●

Pain and pruritus are the most common complaints in patients who present to vulvar clinics.¹ These symptoms can be related to a variety of conditions, including vulvar lesions. There are both common and uncommon vulvar lesions. Vulvar lesions can be skin colored, yellow, and red. Certain lesions can be diagnosed with history and physical examination alone. Some more common lesions include acrochordons (skin tags), benign growths that are common in patients with diabetes, obesity, and pregnancy.^2,3 Other common vulvar lesions are papillomatosis, lichen simplex chronicus, and epidermoid cysts. Other lesions include low- and high-grade squamous intraepithelial lesions (HSIL).⁴ These lesions require biopsy for diagnosis as high-grade lesions require treatment. HSIL of the vulva is considered a premalignancy that necessitates treatment.⁵ Other lesions that can present with vulvar complaints are molluscum contagiosum, Bartholin gland duct cyst, intradermal melanocytic nevus, and squamous cell carcinoma. 

Rarely, other less common conditions can present as vulvar lesions. Syringomas are benign eccrine sweat gland neoplasms. They are more commonly found on the face, neck, or chest.⁶ On the vulva they are generally small subcutaneous skin-colored papules.⁷ They may be asymptomatic and noted only on routine examination. 

Vulvar syringomas also may present with symptoms. On the vulva, syringomas often present as pruritic papules that can be isolated or multifocal. Often on the labia majora they range in size from 2 to 20 mm.⁸

They can coalesce to form a larger lesion. They also may be described as painful. When  syringomas are pruritic, the overlying skin may appear thickened from rubbing or scratching, and excoriations may be present. 

Since vulvar syringomas are rare, there is no standard treatment. Biopsy is necessary for definitive diagnosis. For asymptomatic cases, expectant management is warranted. In symptomatic cases treatment can be considered. Treatment options include cryotherapy, laser ablation, and intralesional electrodissection.⁸ Intralesional electrodissection and curettage also has been described as treatment.⁹ Other treatment options include surgical excision of individual lesions or larger excisions if multifocal. 

The case study described in "Case letter: Vulvar syringoma" highlights the diagnostic and therapeutic challenges associated with rare lesions of the vulva. Referral to a specialty clinic may be warranted in these challenging cases. ●

Pain and pruritus are the most common complaints in patients who present to vulvar clinics.¹ These symptoms can be related to a variety of conditions, including vulvar lesions. There are both common and uncommon vulvar lesions. Vulvar lesions can be skin colored, yellow, and red. Certain lesions can be diagnosed with history and physical examination alone. Some more common lesions include acrochordons (skin tags), benign growths that are common in patients with diabetes, obesity, and pregnancy.^2,3 Other common vulvar lesions are papillomatosis, lichen simplex chronicus, and epidermoid cysts. Other lesions include low- and high-grade squamous intraepithelial lesions (HSIL).⁴ These lesions require biopsy for diagnosis as high-grade lesions require treatment. HSIL of the vulva is considered a premalignancy that necessitates treatment.⁵ Other lesions that can present with vulvar complaints are molluscum contagiosum, Bartholin gland duct cyst, intradermal melanocytic nevus, and squamous cell carcinoma. 

Rarely, other less common conditions can present as vulvar lesions. Syringomas are benign eccrine sweat gland neoplasms. They are more commonly found on the face, neck, or chest.⁶ On the vulva they are generally small subcutaneous skin-colored papules.⁷ They may be asymptomatic and noted only on routine examination. 

Vulvar syringomas also may present with symptoms. On the vulva, syringomas often present as pruritic papules that can be isolated or multifocal. Often on the labia majora they range in size from 2 to 20 mm.⁸

They can coalesce to form a larger lesion. They also may be described as painful. When  syringomas are pruritic, the overlying skin may appear thickened from rubbing or scratching, and excoriations may be present. 

Since vulvar syringomas are rare, there is no standard treatment. Biopsy is necessary for definitive diagnosis. For asymptomatic cases, expectant management is warranted. In symptomatic cases treatment can be considered. Treatment options include cryotherapy, laser ablation, and intralesional electrodissection.⁸ Intralesional electrodissection and curettage also has been described as treatment.⁹ Other treatment options include surgical excision of individual lesions or larger excisions if multifocal. 

The case study described in "Case letter: Vulvar syringoma" highlights the diagnostic and therapeutic challenges associated with rare lesions of the vulva. Referral to a specialty clinic may be warranted in these challenging cases. ●

Pain and pruritus are the most common complaints in patients who present to vulvar clinics.¹ These symptoms can be related to a variety of conditions, including vulvar lesions. There are both common and uncommon vulvar lesions. Vulvar lesions can be skin colored, yellow, and red. Certain lesions can be diagnosed with history and physical examination alone. Some more common lesions include acrochordons (skin tags), benign growths that are common in patients with diabetes, obesity, and pregnancy.^2,3 Other common vulvar lesions are papillomatosis, lichen simplex chronicus, and epidermoid cysts. Other lesions include low- and high-grade squamous intraepithelial lesions (HSIL).⁴ These lesions require biopsy for diagnosis as high-grade lesions require treatment. HSIL of the vulva is considered a premalignancy that necessitates treatment.⁵ Other lesions that can present with vulvar complaints are molluscum contagiosum, Bartholin gland duct cyst, intradermal melanocytic nevus, and squamous cell carcinoma. 

Rarely, other less common conditions can present as vulvar lesions. Syringomas are benign eccrine sweat gland neoplasms. They are more commonly found on the face, neck, or chest.⁶ On the vulva they are generally small subcutaneous skin-colored papules.⁷ They may be asymptomatic and noted only on routine examination. 

Vulvar syringomas also may present with symptoms. On the vulva, syringomas often present as pruritic papules that can be isolated or multifocal. Often on the labia majora they range in size from 2 to 20 mm.⁸

They can coalesce to form a larger lesion. They also may be described as painful. When  syringomas are pruritic, the overlying skin may appear thickened from rubbing or scratching, and excoriations may be present. 

Since vulvar syringomas are rare, there is no standard treatment. Biopsy is necessary for definitive diagnosis. For asymptomatic cases, expectant management is warranted. In symptomatic cases treatment can be considered. Treatment options include cryotherapy, laser ablation, and intralesional electrodissection.⁸ Intralesional electrodissection and curettage also has been described as treatment.⁹ Other treatment options include surgical excision of individual lesions or larger excisions if multifocal. 

The case study described in "Case letter: Vulvar syringoma" highlights the diagnostic and therapeutic challenges associated with rare lesions of the vulva. Referral to a specialty clinic may be warranted in these challenging cases. ●

In this piece, Dr. Mickey Karram & Dr. Roger R. Dmochowski discuss how although UTIs have demonstrated widespread occurrence and significant healthcare costs, there is not yet a “gold standard” definition for complicated UTI. To avoid the overuse of antimicrobial agents and their associated issues, it is vital that clinicians evaluate test results in the context of a patient’s overall risk and history of UTIs and current clinical presentation and utilize testing that enables more informed decisions.

Click here to read more

In this piece, Dr. Mickey Karram & Dr. Roger R. Dmochowski discuss how although UTIs have demonstrated widespread occurrence and significant healthcare costs, there is not yet a “gold standard” definition for complicated UTI. To avoid the overuse of antimicrobial agents and their associated issues, it is vital that clinicians evaluate test results in the context of a patient’s overall risk and history of UTIs and current clinical presentation and utilize testing that enables more informed decisions.

Click here to read more

In this piece, Dr. Mickey Karram & Dr. Roger R. Dmochowski discuss how although UTIs have demonstrated widespread occurrence and significant healthcare costs, there is not yet a “gold standard” definition for complicated UTI. To avoid the overuse of antimicrobial agents and their associated issues, it is vital that clinicians evaluate test results in the context of a patient’s overall risk and history of UTIs and current clinical presentation and utilize testing that enables more informed decisions.

Click here to read more

An elevated serum level of cholesterol has been recognized as a risk factor for atherosclerotic cardiovascular disease (ASCVD) since the publication of the Framingham Study in 1961.¹ Although clinical outcomes related to ASCVD have improved in recent decades, ASCVD remains the leading cause of morbidity and mortality across the globe and remains, in the United States, the leading cause of death among most racial and ethnic groups. Much of this persistent disease burden can be attributed to inadequate control of ASCVD risk factors and suboptimal implementation of prevention strategies in the general population.²

The most recent (2019) iteration of the American College of Cardiology/American Heart Association (ACC/AHA) Guideline on the Primary Prevention of Cardiovascular Disease emphasizes a comprehensive, patient-centered, team-based approach to the management of ASCVD risk factors.² In this article, I review how, first, medication to reduce ASCVD risk should be considered only when a patient’s risk is sufficiently high and, second, shared decision-making and social determinants of health should, in all cases, guide and inform optimal implementation of treatment.²

Estimating risk for ASCVD by ascertaining LDL-C

• The Friedewald equation. Traditionally, low-density lipoprotein cholesterol (LDL-C) is estimated using the Friedewald equation^a applied to a fasting lipid profile. In patients who have a low level of LDL-C (< 70 mg/dL), however, the Friedewald equation becomes less accurate; in patients with hypertriglyceridemia (TG ≥ 400 mg/dL),estimation of LDL-C is invalid.
• The Martin–Hopkins equation offers a validated estimation of LDL-C when the LDL-C value is < 70 mg/dL.³ This equation—in which the fixed factor of 5 used in the Friedewald equation to estimate very low-density lipoprotein cholesterol is replaced by an adjustable factor that is based on the patient’s non-HDL-C (ie, TC–HDL-C) and TG values—is preferred by the ACC/AHA Task Force on Clinical Practice Guidelines in this clinical circumstance.⁴
• National Institutes of Health equation. This newer equation provides an accurate estimate of the LDL-C level in patients whose TG value is ≤ 800 mg/dL. The equation has not been fully validated for clinical use, however.⁵
• Direct measurement obviates the need for an equation to estimate LDL-C, but the test is not available in all health care settings.

For adults ≥ 20 years of age who are not receiving lipid-lowering therapy, a nonfasting lipid profile can be used to estimate ASCVD risk and document the baseline LDL-C level. If the TG level is ≥ 400 mg/dL, the test should be administered in the fasting state.⁴

• Apolipoprotein B. Alternatively, apolipoprotein B (apoB) can be measured. Because each LDL-C particle contains 1 apoB molecule, the apoB level describes the LDL-C level more accurately than a calculation of LDL-C. Many patients with type 2 diabetes and metabolic syndrome have a relatively low calculated LDL-C (thereby falsely reassuring the testing clinician) but have an elevated apoB level. An apoB level ≥ 130 mg/dL corresponds to an LDL-C level >160 mg/dL.⁴
• Calculation of non-HDL-C. Because the nonfasting state does not have a significant impact on a patient’s TC and HDL-C levels, the non-HDL-C level also can be calculated from the results of a nonfasting lipid profile.

Non-HDL-C and apoB are equivalent predictors of ASCVD risk. These 2 assessments might offer better risk estimation than other available tools in patients who have type 2 diabetes and metabolic syndrome.⁶

Continue to: Applying the estimate of 10-year ASCVD risk...

Your recommendation for preventive intervention, such as lipid-lowering therapy, should be based on the estimated 10-year risk for ASCVD. Although multiple validated risk assessment tools are available, ACC/AHA recommends the pooled cohort risk equations (PCE), introduced in the 2013 ACC/AHA cholesterol treatment guidelines. The Framingham Heart Study now recommends the ACC/AHA PCE for risk assessment as well.⁷

The PCE, developed from 5 large cohorts, is based on hard atherosclerotic events: nonfatal myocardial infarction, death from coronary artery disease, and stroke. The ACC/AHA PCE is the only risk assessment tool developed using a significant percentage of patients who self-identify as Black.⁸ Alternatives to the ACC/AHA PCE include:

• Multi-ethnic Study of Atherosclerosis (MESA) 10-year ASCVD risk calculator, which incorporates the coronary artery calcium (CAC) score.
• Reynolds Risk Score, which incorporates high-sensitivity C-reactive protein measurement and a family history of premature ASCVD.⁹

How much does lifestyle modification actually matter?

The absolute impact of diet and exercise on lipid parameters is relatively modest. No studies have demonstrated a reduction in adverse cardiovascular outcomes with specific interventions regarding diet or activity.

• Diet. Nevertheless, ACC/AHA recommends that at-risk patients follow a dietary pattern that (1) emphasizes vegetables, fruits, and whole grains and (2) limits sweets, sugar-sweetened beverages, and red meat.

Saturated fat should constitute no more than 5% or 6% of total calories. In controlled-feeding trials,¹⁰ for every 1% of calories from saturated fat that are replaced with carbohydrate or monounsaturated or polyunsaturated fat, the LDL-C level was found to decline by as much as 1.8 mg/dL. Evidence is insufficient to assert that lowering dietary cholesterol reduces LDL-C.¹¹

• Activity. Trials of aerobic physical activity, compared with a more sedentary activity pattern, have demonstrated a reduction in the LDL-C level of as much as 6 mg/dL. All adult patients should be counseled to engage in aerobic physical activity of moderate or vigorous intensity—averaging ≥ 40 minutes per session, 3 or 4 sessions per week.¹¹

Primary prevention: Stratification by age

• 40 to 75 years. ACC/AHA recommends that you routinely assess traditional cardiovascular risk factors for these patients and calculate their 10-year risk for ASCVD using the PCE. Statin therapy as primary prevention is indicated for 3 major groups (TABLE 1).⁴ The US Preventive Services Task Force (USPSTF) recommends a 10-year ASCVD risk ≥ 10%, in conjunction with 1 or more additional CVD risk factors (dyslipidemia, diabetes, hypertension, smoking), as the threshold for initiating low- or moderate-intensity statin therapy in this age group.¹²

In adults at borderline risk (5% to < 7.5% 10-year ASCVD risk) or intermediate risk (≥ 7.5% to < 20% 10-year ASCVD risk), consider risk-enhancing factors to better inform your recommendation for preventive interventions. In these 2 groups, the presence of risk-enhancing factors might justify moderate-intensity statin therapy (TABLE 2⁴).

If your decision regarding preventive intervention remains uncertain, measuring CAC might further guide your discussion with the patient.⁴ When the CAC score is:

• 0 Agatston units and higher-risk conditions (eg, diabetes, family history of premature coronary artery disease, smoking) are absent, statin therapy can be withheld; reassess ASCVD risk in 5 to 10 years.
• 1-99 Agatston units, statin therapy can be started, especially for patients ≥ 55 years of age.
• ≥ 100 Agatston units or ≥ 75th percentile, statin therapy is indicated for all patients, regardless of additional risk factors.⁴

Because statins promote progression from unstable, inflammatory atherosclerotic plaque to more stable, calcified plaque, CAC scoring is not valid in patients already on statin therapy.¹³

In primary prevention, patients who have been classified as having low or intermediate risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have an annual all-cause mortality < 1%, regardless of age and gender. Patients classified as being at high risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have a significantly lower annual mortality than low- or intermediate-risk patients with a CAC score > 0 Agatston units.¹⁴

• 20 to 39 years. Focus on evaluation of lifetime ASCVD risk, rather than short-term (10-year) risk. Lifestyle modification is the primary intervention for younger patients; for those with moderate hypercholesterolemia (LDL-C, 160-189 mg/dL) and a family history of premature ASCVD, however, consider statin therapy. For patients with LDL-C ≥ 190 mg/dL, lifetime ASCVD risk is markedly increased, and high-intensity statin therapy is recommended, regardless of age. In this group, reassess ASCVD risk factors every 4 to 6 years.⁴
• > 75 years, without ASCVD. In this group, the benefit of statin therapy is less clear and might be lessened by an increased potential for adverse effects. A meta-analysis of 28 trials demonstrated that people ages > 75 years had a 24% relative reduction in major coronary events for every 38.7mg/dL (1.0 mmol/L) reduction in LDL-C, which is comparable to the risk reduction seen in people ages 40 to 75 years.¹⁵

With increasing age, however, the relative reduction in major coronary events with statin therapy decreased,¹⁵ although other trials have not demonstrated age heterogeneity.¹⁶ Because people > 75 years of age have a significantly higher ASCVD event rate, a comparable relative rate reduction with statin therapy results in a larger absolute rate reduction (ARR) and, therefore, a smaller number needed to treat (NNT) to prevent an event, compared to the NNT in younger people.

Secondary prevention

ACC/AHA guidelines define clinical ASCVD as a history of:

• acute coronary syndrome
• myocardial infarction
• coronary or other arterial revascularization
• cerebrovascular event
• symptomatic peripheral artery disease, including aortic aneurysm.

High-intensity statin therapy is indicated for all patients ≤ 75 years who have clinical ASCVD. In patients > 75 years, consider a taper to moderate-intensity statin therapy. An upper age limit for seeing benefit from statin therapy in secondary prevention has not been identified.⁴

In high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy, ezetimibe (discussed in the next section) can be added. In very-high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy plus ezetimibe, a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor (also discussed next) can be added. Always precede initiation of a PCSK9 inhibitor with a discussion of the net benefit, safety, and cost with the patient.⁴

Continue to: Options for lipid-lowering pharmacotherapy...

• Statins (formally, hydroxymethylglutaryl-coenzyme A reductase inhibitors) offer the most predictable reduction in ASCVD risk of any lipid-lowering therapy. The evidence report that accompanied the 2016 USPSTF guidelines on statins for the prevention of cardiovascular disease (CVD) stated that low- or moderate-dosage statin therapy is associated with approximately a 30% relative risk reduction (RRR) in CVD events and CVD deaths and a 10% to 15% RRR in all-cause mortality.¹⁷

High-intensity statin therapy reduces LDL-C by ≥ 50%. Moderate-intensity statin therapy reduces LDL-C by 30% to 49% (TABLE 3).⁴

Statins are not without risk: A 2016 report¹⁸ estimated that treating 10,000 patients with a statin for 5 years would cause 1 case of rhabdomyolysis, 5 cases of myopathy, 75 new cases of diabetes, and 7 cases of hemorrhagic stroke. The same treatment would, however, avert approximately 1000 CVD events among patients with preexisting disease and approximately 500 CVD events among patients at elevated risk but without preexisting disease.¹⁸

• Ezetimibe, a selective cholesterol-absorption inhibitor, lowers LDL-C by 13% to 20% and typically is well tolerated. The use of ezetimibe in ASCVD risk reduction is supported by a single randomized controlled trial of more than 18,000 patients with recent acute coronary syndrome. Adding ezetimibe to simvastatin 40 mg resulted in a 2% absolute reduction in major adverse cardiovascular events over a median follow-up of 6 years (NNT = 50), compared to simvastatin alone.¹⁹ ACC/AHA guidelines recommend adding ezetimibe to maximally tolerated statin therapy in patients with clinical ASCVD who do not reach their goal LDL reduction with a statin alone. Ezetimibe also can be considered a statin alternative in patients who are statin intolerant.⁴
• PCSK9 inhibitors. When added to statin therapy, evolocumab and alirocumab—monoclonal antibodies that inhibit PCSK9—offer an incremental decrease in LDL-C of approximately 60%.^20-22 In a meta-analysis of 35 trials evaluating the incremental benefit of PCSK9 inhibitor therapy, a significant reduction in cardiovascular events, including myocardial infarction (ARR = 1.3%; NNT = 77), stroke (ARR = 0.4%; NNT = 250), and coronary revascularization (ARR = 1.6%; NNT = 63) was reported. No significant difference was observed in all-cause or cardiovascular mortality.^21,23
• Inclisiran, an injectable small-interfering RNA that inhibits PCSK9 synthesis, provides an incremental decrease in LDL-C of > 50% in patients already receiving statin therapy. Meta-analysis of 3 small cardiovascular outcomes trials revealed no significant difference in the rate of myocardial infarction, stroke, or cardiovascular mortality with inclisiran compared to placebo. Larger outcomes trials are underway and might offer additional insight into this agent’s role in ASCVD risk management.²⁴
• Omega-3 fatty acids. Multiple trials have demonstrated that adding omega-3 fatty acids to usual lipid-lowering therapy does not offer a consistent reduction in adverse cardiovascular outcomes, despite providing a significant reduction in TG levels. In a high-risk population with persistently elevated TG despite statin therapy, icosapent ethyl, a purified eicosapentaenoic acid ethyl ester, reduced major ASCVD outcomes by 25% over a median 4.9 years (ARR = 4.8%; NNT = 21), and cardiovascular death by 20% (ARR = 0.9%; NNT = 111), compared with a mineral oil placebo.²⁵ Subsequent trials, using a corn oil placebo, failed to duplicate these data²⁶—raising concern that the mineral oil comparator might have altered results of the eicosapentaenoic acid ethyl ester study.^27,28
• Bempedoic acid is a small-molecule inhibitor of ATP citrate lyase that increases LDL uptake by the liver. Pooled data from studies of bempedoic acid show, on average, a 15% reduction in TC, a 23% reduction in LDL-C, and a 6% increase in HDL-C, without a significant change in TG.²⁹ In statin-intolerant patients, bempedoicacid reduced major ASCVD outcomes by 13% over a median 40 months (ARR = 1.6%; NNT = 63), with no significant reduction in cardiovascular death.³⁰
• Niacin. Two large trials failed to demonstrate improvement in major cardiovascular events or other clinical benefit when niacin is added to moderate-intensity statin therapy, despite a significant increase in the HDL-C level (on average, 6 mg/dL) and a decrease in the LDL-C level (10-12 mg/dL)and TG (42 mg/dL).^31,32
• Fenofibrate lowers TG and increases HDL-C but does not consistently improve cardiovascular outcomes.³³ In a trial of patients with type 2 diabetes and persistent dyslipidemia (serum TG > 204 mg/dL; HDL-C< 34 mg/dL) despite statin therapy, adding fenofibrate reduced CVD outcomes by 4.9%—although this absolute difference did not reach statistical significance.³⁴

Neither niacin nor fenofibrate is considered useful for reducing ASCVD risk across broad populations.⁴

Follow-up to assess progress toward goals

Recheck the lipid profile 4 to 12 weeks after starting lipid-lowering therapy to verify adherence to medication and assess response. The primary goal is the percentage reduction in LDL-C based on ASCVD risk. An additional goal for very-high-risk patients is an LDL-C value ≤ 70 mg/dL. If the reduction in LDL-C is less than desired and adherence is assured, consider titrating the statin dosage or augmenting statin therapy with a nonstatin drug (eg, ezetimibe), or both.⁴ ●

CORRESPONDENCE

Jonathon M. Firnhaber, MD, MAEd, MBA, East Carolina University, Family Medicine Center, 101 Heart Drive, Greenville, NC 27834; firnhaberj@ecu.edu

An elevated serum level of cholesterol has been recognized as a risk factor for atherosclerotic cardiovascular disease (ASCVD) since the publication of the Framingham Study in 1961.¹ Although clinical outcomes related to ASCVD have improved in recent decades, ASCVD remains the leading cause of morbidity and mortality across the globe and remains, in the United States, the leading cause of death among most racial and ethnic groups. Much of this persistent disease burden can be attributed to inadequate control of ASCVD risk factors and suboptimal implementation of prevention strategies in the general population.²

The most recent (2019) iteration of the American College of Cardiology/American Heart Association (ACC/AHA) Guideline on the Primary Prevention of Cardiovascular Disease emphasizes a comprehensive, patient-centered, team-based approach to the management of ASCVD risk factors.² In this article, I review how, first, medication to reduce ASCVD risk should be considered only when a patient’s risk is sufficiently high and, second, shared decision-making and social determinants of health should, in all cases, guide and inform optimal implementation of treatment.²

Estimating risk for ASCVD by ascertaining LDL-C

• The Friedewald equation. Traditionally, low-density lipoprotein cholesterol (LDL-C) is estimated using the Friedewald equation^a applied to a fasting lipid profile. In patients who have a low level of LDL-C (< 70 mg/dL), however, the Friedewald equation becomes less accurate; in patients with hypertriglyceridemia (TG ≥ 400 mg/dL),estimation of LDL-C is invalid.
• The Martin–Hopkins equation offers a validated estimation of LDL-C when the LDL-C value is < 70 mg/dL.³ This equation—in which the fixed factor of 5 used in the Friedewald equation to estimate very low-density lipoprotein cholesterol is replaced by an adjustable factor that is based on the patient’s non-HDL-C (ie, TC–HDL-C) and TG values—is preferred by the ACC/AHA Task Force on Clinical Practice Guidelines in this clinical circumstance.⁴
• National Institutes of Health equation. This newer equation provides an accurate estimate of the LDL-C level in patients whose TG value is ≤ 800 mg/dL. The equation has not been fully validated for clinical use, however.⁵
• Direct measurement obviates the need for an equation to estimate LDL-C, but the test is not available in all health care settings.

For adults ≥ 20 years of age who are not receiving lipid-lowering therapy, a nonfasting lipid profile can be used to estimate ASCVD risk and document the baseline LDL-C level. If the TG level is ≥ 400 mg/dL, the test should be administered in the fasting state.⁴

• Apolipoprotein B. Alternatively, apolipoprotein B (apoB) can be measured. Because each LDL-C particle contains 1 apoB molecule, the apoB level describes the LDL-C level more accurately than a calculation of LDL-C. Many patients with type 2 diabetes and metabolic syndrome have a relatively low calculated LDL-C (thereby falsely reassuring the testing clinician) but have an elevated apoB level. An apoB level ≥ 130 mg/dL corresponds to an LDL-C level >160 mg/dL.⁴
• Calculation of non-HDL-C. Because the nonfasting state does not have a significant impact on a patient’s TC and HDL-C levels, the non-HDL-C level also can be calculated from the results of a nonfasting lipid profile.

Non-HDL-C and apoB are equivalent predictors of ASCVD risk. These 2 assessments might offer better risk estimation than other available tools in patients who have type 2 diabetes and metabolic syndrome.⁶

Continue to: Applying the estimate of 10-year ASCVD risk...

Your recommendation for preventive intervention, such as lipid-lowering therapy, should be based on the estimated 10-year risk for ASCVD. Although multiple validated risk assessment tools are available, ACC/AHA recommends the pooled cohort risk equations (PCE), introduced in the 2013 ACC/AHA cholesterol treatment guidelines. The Framingham Heart Study now recommends the ACC/AHA PCE for risk assessment as well.⁷

The PCE, developed from 5 large cohorts, is based on hard atherosclerotic events: nonfatal myocardial infarction, death from coronary artery disease, and stroke. The ACC/AHA PCE is the only risk assessment tool developed using a significant percentage of patients who self-identify as Black.⁸ Alternatives to the ACC/AHA PCE include:

• Multi-ethnic Study of Atherosclerosis (MESA) 10-year ASCVD risk calculator, which incorporates the coronary artery calcium (CAC) score.
• Reynolds Risk Score, which incorporates high-sensitivity C-reactive protein measurement and a family history of premature ASCVD.⁹

How much does lifestyle modification actually matter?

The absolute impact of diet and exercise on lipid parameters is relatively modest. No studies have demonstrated a reduction in adverse cardiovascular outcomes with specific interventions regarding diet or activity.

• Diet. Nevertheless, ACC/AHA recommends that at-risk patients follow a dietary pattern that (1) emphasizes vegetables, fruits, and whole grains and (2) limits sweets, sugar-sweetened beverages, and red meat.

Saturated fat should constitute no more than 5% or 6% of total calories. In controlled-feeding trials,¹⁰ for every 1% of calories from saturated fat that are replaced with carbohydrate or monounsaturated or polyunsaturated fat, the LDL-C level was found to decline by as much as 1.8 mg/dL. Evidence is insufficient to assert that lowering dietary cholesterol reduces LDL-C.¹¹

• Activity. Trials of aerobic physical activity, compared with a more sedentary activity pattern, have demonstrated a reduction in the LDL-C level of as much as 6 mg/dL. All adult patients should be counseled to engage in aerobic physical activity of moderate or vigorous intensity—averaging ≥ 40 minutes per session, 3 or 4 sessions per week.¹¹

Primary prevention: Stratification by age

• 40 to 75 years. ACC/AHA recommends that you routinely assess traditional cardiovascular risk factors for these patients and calculate their 10-year risk for ASCVD using the PCE. Statin therapy as primary prevention is indicated for 3 major groups (TABLE 1).⁴ The US Preventive Services Task Force (USPSTF) recommends a 10-year ASCVD risk ≥ 10%, in conjunction with 1 or more additional CVD risk factors (dyslipidemia, diabetes, hypertension, smoking), as the threshold for initiating low- or moderate-intensity statin therapy in this age group.¹²

In adults at borderline risk (5% to < 7.5% 10-year ASCVD risk) or intermediate risk (≥ 7.5% to < 20% 10-year ASCVD risk), consider risk-enhancing factors to better inform your recommendation for preventive interventions. In these 2 groups, the presence of risk-enhancing factors might justify moderate-intensity statin therapy (TABLE 2⁴).

If your decision regarding preventive intervention remains uncertain, measuring CAC might further guide your discussion with the patient.⁴ When the CAC score is:

• 0 Agatston units and higher-risk conditions (eg, diabetes, family history of premature coronary artery disease, smoking) are absent, statin therapy can be withheld; reassess ASCVD risk in 5 to 10 years.
• 1-99 Agatston units, statin therapy can be started, especially for patients ≥ 55 years of age.
• ≥ 100 Agatston units or ≥ 75th percentile, statin therapy is indicated for all patients, regardless of additional risk factors.⁴

Because statins promote progression from unstable, inflammatory atherosclerotic plaque to more stable, calcified plaque, CAC scoring is not valid in patients already on statin therapy.¹³

In primary prevention, patients who have been classified as having low or intermediate risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have an annual all-cause mortality < 1%, regardless of age and gender. Patients classified as being at high risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have a significantly lower annual mortality than low- or intermediate-risk patients with a CAC score > 0 Agatston units.¹⁴

• 20 to 39 years. Focus on evaluation of lifetime ASCVD risk, rather than short-term (10-year) risk. Lifestyle modification is the primary intervention for younger patients; for those with moderate hypercholesterolemia (LDL-C, 160-189 mg/dL) and a family history of premature ASCVD, however, consider statin therapy. For patients with LDL-C ≥ 190 mg/dL, lifetime ASCVD risk is markedly increased, and high-intensity statin therapy is recommended, regardless of age. In this group, reassess ASCVD risk factors every 4 to 6 years.⁴
• > 75 years, without ASCVD. In this group, the benefit of statin therapy is less clear and might be lessened by an increased potential for adverse effects. A meta-analysis of 28 trials demonstrated that people ages > 75 years had a 24% relative reduction in major coronary events for every 38.7mg/dL (1.0 mmol/L) reduction in LDL-C, which is comparable to the risk reduction seen in people ages 40 to 75 years.¹⁵

With increasing age, however, the relative reduction in major coronary events with statin therapy decreased,¹⁵ although other trials have not demonstrated age heterogeneity.¹⁶ Because people > 75 years of age have a significantly higher ASCVD event rate, a comparable relative rate reduction with statin therapy results in a larger absolute rate reduction (ARR) and, therefore, a smaller number needed to treat (NNT) to prevent an event, compared to the NNT in younger people.

Secondary prevention

ACC/AHA guidelines define clinical ASCVD as a history of:

• acute coronary syndrome
• myocardial infarction
• coronary or other arterial revascularization
• cerebrovascular event
• symptomatic peripheral artery disease, including aortic aneurysm.

High-intensity statin therapy is indicated for all patients ≤ 75 years who have clinical ASCVD. In patients > 75 years, consider a taper to moderate-intensity statin therapy. An upper age limit for seeing benefit from statin therapy in secondary prevention has not been identified.⁴

In high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy, ezetimibe (discussed in the next section) can be added. In very-high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy plus ezetimibe, a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor (also discussed next) can be added. Always precede initiation of a PCSK9 inhibitor with a discussion of the net benefit, safety, and cost with the patient.⁴

Continue to: Options for lipid-lowering pharmacotherapy...

• Statins (formally, hydroxymethylglutaryl-coenzyme A reductase inhibitors) offer the most predictable reduction in ASCVD risk of any lipid-lowering therapy. The evidence report that accompanied the 2016 USPSTF guidelines on statins for the prevention of cardiovascular disease (CVD) stated that low- or moderate-dosage statin therapy is associated with approximately a 30% relative risk reduction (RRR) in CVD events and CVD deaths and a 10% to 15% RRR in all-cause mortality.¹⁷

High-intensity statin therapy reduces LDL-C by ≥ 50%. Moderate-intensity statin therapy reduces LDL-C by 30% to 49% (TABLE 3).⁴

Statins are not without risk: A 2016 report¹⁸ estimated that treating 10,000 patients with a statin for 5 years would cause 1 case of rhabdomyolysis, 5 cases of myopathy, 75 new cases of diabetes, and 7 cases of hemorrhagic stroke. The same treatment would, however, avert approximately 1000 CVD events among patients with preexisting disease and approximately 500 CVD events among patients at elevated risk but without preexisting disease.¹⁸

• Ezetimibe, a selective cholesterol-absorption inhibitor, lowers LDL-C by 13% to 20% and typically is well tolerated. The use of ezetimibe in ASCVD risk reduction is supported by a single randomized controlled trial of more than 18,000 patients with recent acute coronary syndrome. Adding ezetimibe to simvastatin 40 mg resulted in a 2% absolute reduction in major adverse cardiovascular events over a median follow-up of 6 years (NNT = 50), compared to simvastatin alone.¹⁹ ACC/AHA guidelines recommend adding ezetimibe to maximally tolerated statin therapy in patients with clinical ASCVD who do not reach their goal LDL reduction with a statin alone. Ezetimibe also can be considered a statin alternative in patients who are statin intolerant.⁴
• PCSK9 inhibitors. When added to statin therapy, evolocumab and alirocumab—monoclonal antibodies that inhibit PCSK9—offer an incremental decrease in LDL-C of approximately 60%.^20-22 In a meta-analysis of 35 trials evaluating the incremental benefit of PCSK9 inhibitor therapy, a significant reduction in cardiovascular events, including myocardial infarction (ARR = 1.3%; NNT = 77), stroke (ARR = 0.4%; NNT = 250), and coronary revascularization (ARR = 1.6%; NNT = 63) was reported. No significant difference was observed in all-cause or cardiovascular mortality.^21,23
• Inclisiran, an injectable small-interfering RNA that inhibits PCSK9 synthesis, provides an incremental decrease in LDL-C of > 50% in patients already receiving statin therapy. Meta-analysis of 3 small cardiovascular outcomes trials revealed no significant difference in the rate of myocardial infarction, stroke, or cardiovascular mortality with inclisiran compared to placebo. Larger outcomes trials are underway and might offer additional insight into this agent’s role in ASCVD risk management.²⁴
• Omega-3 fatty acids. Multiple trials have demonstrated that adding omega-3 fatty acids to usual lipid-lowering therapy does not offer a consistent reduction in adverse cardiovascular outcomes, despite providing a significant reduction in TG levels. In a high-risk population with persistently elevated TG despite statin therapy, icosapent ethyl, a purified eicosapentaenoic acid ethyl ester, reduced major ASCVD outcomes by 25% over a median 4.9 years (ARR = 4.8%; NNT = 21), and cardiovascular death by 20% (ARR = 0.9%; NNT = 111), compared with a mineral oil placebo.²⁵ Subsequent trials, using a corn oil placebo, failed to duplicate these data²⁶—raising concern that the mineral oil comparator might have altered results of the eicosapentaenoic acid ethyl ester study.^27,28
• Bempedoic acid is a small-molecule inhibitor of ATP citrate lyase that increases LDL uptake by the liver. Pooled data from studies of bempedoic acid show, on average, a 15% reduction in TC, a 23% reduction in LDL-C, and a 6% increase in HDL-C, without a significant change in TG.²⁹ In statin-intolerant patients, bempedoicacid reduced major ASCVD outcomes by 13% over a median 40 months (ARR = 1.6%; NNT = 63), with no significant reduction in cardiovascular death.³⁰
• Niacin. Two large trials failed to demonstrate improvement in major cardiovascular events or other clinical benefit when niacin is added to moderate-intensity statin therapy, despite a significant increase in the HDL-C level (on average, 6 mg/dL) and a decrease in the LDL-C level (10-12 mg/dL)and TG (42 mg/dL).^31,32
• Fenofibrate lowers TG and increases HDL-C but does not consistently improve cardiovascular outcomes.³³ In a trial of patients with type 2 diabetes and persistent dyslipidemia (serum TG > 204 mg/dL; HDL-C< 34 mg/dL) despite statin therapy, adding fenofibrate reduced CVD outcomes by 4.9%—although this absolute difference did not reach statistical significance.³⁴

Neither niacin nor fenofibrate is considered useful for reducing ASCVD risk across broad populations.⁴

Follow-up to assess progress toward goals

Recheck the lipid profile 4 to 12 weeks after starting lipid-lowering therapy to verify adherence to medication and assess response. The primary goal is the percentage reduction in LDL-C based on ASCVD risk. An additional goal for very-high-risk patients is an LDL-C value ≤ 70 mg/dL. If the reduction in LDL-C is less than desired and adherence is assured, consider titrating the statin dosage or augmenting statin therapy with a nonstatin drug (eg, ezetimibe), or both.⁴ ●

CORRESPONDENCE

Jonathon M. Firnhaber, MD, MAEd, MBA, East Carolina University, Family Medicine Center, 101 Heart Drive, Greenville, NC 27834; firnhaberj@ecu.edu

An elevated serum level of cholesterol has been recognized as a risk factor for atherosclerotic cardiovascular disease (ASCVD) since the publication of the Framingham Study in 1961.¹ Although clinical outcomes related to ASCVD have improved in recent decades, ASCVD remains the leading cause of morbidity and mortality across the globe and remains, in the United States, the leading cause of death among most racial and ethnic groups. Much of this persistent disease burden can be attributed to inadequate control of ASCVD risk factors and suboptimal implementation of prevention strategies in the general population.²

The most recent (2019) iteration of the American College of Cardiology/American Heart Association (ACC/AHA) Guideline on the Primary Prevention of Cardiovascular Disease emphasizes a comprehensive, patient-centered, team-based approach to the management of ASCVD risk factors.² In this article, I review how, first, medication to reduce ASCVD risk should be considered only when a patient’s risk is sufficiently high and, second, shared decision-making and social determinants of health should, in all cases, guide and inform optimal implementation of treatment.²

Estimating risk for ASCVD by ascertaining LDL-C

• The Friedewald equation. Traditionally, low-density lipoprotein cholesterol (LDL-C) is estimated using the Friedewald equation^a applied to a fasting lipid profile. In patients who have a low level of LDL-C (< 70 mg/dL), however, the Friedewald equation becomes less accurate; in patients with hypertriglyceridemia (TG ≥ 400 mg/dL),estimation of LDL-C is invalid.
• The Martin–Hopkins equation offers a validated estimation of LDL-C when the LDL-C value is < 70 mg/dL.³ This equation—in which the fixed factor of 5 used in the Friedewald equation to estimate very low-density lipoprotein cholesterol is replaced by an adjustable factor that is based on the patient’s non-HDL-C (ie, TC–HDL-C) and TG values—is preferred by the ACC/AHA Task Force on Clinical Practice Guidelines in this clinical circumstance.⁴
• National Institutes of Health equation. This newer equation provides an accurate estimate of the LDL-C level in patients whose TG value is ≤ 800 mg/dL. The equation has not been fully validated for clinical use, however.⁵
• Direct measurement obviates the need for an equation to estimate LDL-C, but the test is not available in all health care settings.

For adults ≥ 20 years of age who are not receiving lipid-lowering therapy, a nonfasting lipid profile can be used to estimate ASCVD risk and document the baseline LDL-C level. If the TG level is ≥ 400 mg/dL, the test should be administered in the fasting state.⁴

• Apolipoprotein B. Alternatively, apolipoprotein B (apoB) can be measured. Because each LDL-C particle contains 1 apoB molecule, the apoB level describes the LDL-C level more accurately than a calculation of LDL-C. Many patients with type 2 diabetes and metabolic syndrome have a relatively low calculated LDL-C (thereby falsely reassuring the testing clinician) but have an elevated apoB level. An apoB level ≥ 130 mg/dL corresponds to an LDL-C level >160 mg/dL.⁴
• Calculation of non-HDL-C. Because the nonfasting state does not have a significant impact on a patient’s TC and HDL-C levels, the non-HDL-C level also can be calculated from the results of a nonfasting lipid profile.

Non-HDL-C and apoB are equivalent predictors of ASCVD risk. These 2 assessments might offer better risk estimation than other available tools in patients who have type 2 diabetes and metabolic syndrome.⁶

Continue to: Applying the estimate of 10-year ASCVD risk...

Your recommendation for preventive intervention, such as lipid-lowering therapy, should be based on the estimated 10-year risk for ASCVD. Although multiple validated risk assessment tools are available, ACC/AHA recommends the pooled cohort risk equations (PCE), introduced in the 2013 ACC/AHA cholesterol treatment guidelines. The Framingham Heart Study now recommends the ACC/AHA PCE for risk assessment as well.⁷

The PCE, developed from 5 large cohorts, is based on hard atherosclerotic events: nonfatal myocardial infarction, death from coronary artery disease, and stroke. The ACC/AHA PCE is the only risk assessment tool developed using a significant percentage of patients who self-identify as Black.⁸ Alternatives to the ACC/AHA PCE include:

• Multi-ethnic Study of Atherosclerosis (MESA) 10-year ASCVD risk calculator, which incorporates the coronary artery calcium (CAC) score.
• Reynolds Risk Score, which incorporates high-sensitivity C-reactive protein measurement and a family history of premature ASCVD.⁹

How much does lifestyle modification actually matter?

The absolute impact of diet and exercise on lipid parameters is relatively modest. No studies have demonstrated a reduction in adverse cardiovascular outcomes with specific interventions regarding diet or activity.

• Diet. Nevertheless, ACC/AHA recommends that at-risk patients follow a dietary pattern that (1) emphasizes vegetables, fruits, and whole grains and (2) limits sweets, sugar-sweetened beverages, and red meat.

Saturated fat should constitute no more than 5% or 6% of total calories. In controlled-feeding trials,¹⁰ for every 1% of calories from saturated fat that are replaced with carbohydrate or monounsaturated or polyunsaturated fat, the LDL-C level was found to decline by as much as 1.8 mg/dL. Evidence is insufficient to assert that lowering dietary cholesterol reduces LDL-C.¹¹

• Activity. Trials of aerobic physical activity, compared with a more sedentary activity pattern, have demonstrated a reduction in the LDL-C level of as much as 6 mg/dL. All adult patients should be counseled to engage in aerobic physical activity of moderate or vigorous intensity—averaging ≥ 40 minutes per session, 3 or 4 sessions per week.¹¹

Primary prevention: Stratification by age

• 40 to 75 years. ACC/AHA recommends that you routinely assess traditional cardiovascular risk factors for these patients and calculate their 10-year risk for ASCVD using the PCE. Statin therapy as primary prevention is indicated for 3 major groups (TABLE 1).⁴ The US Preventive Services Task Force (USPSTF) recommends a 10-year ASCVD risk ≥ 10%, in conjunction with 1 or more additional CVD risk factors (dyslipidemia, diabetes, hypertension, smoking), as the threshold for initiating low- or moderate-intensity statin therapy in this age group.¹²

In adults at borderline risk (5% to < 7.5% 10-year ASCVD risk) or intermediate risk (≥ 7.5% to < 20% 10-year ASCVD risk), consider risk-enhancing factors to better inform your recommendation for preventive interventions. In these 2 groups, the presence of risk-enhancing factors might justify moderate-intensity statin therapy (TABLE 2⁴).

If your decision regarding preventive intervention remains uncertain, measuring CAC might further guide your discussion with the patient.⁴ When the CAC score is:

• 0 Agatston units and higher-risk conditions (eg, diabetes, family history of premature coronary artery disease, smoking) are absent, statin therapy can be withheld; reassess ASCVD risk in 5 to 10 years.
• 1-99 Agatston units, statin therapy can be started, especially for patients ≥ 55 years of age.
• ≥ 100 Agatston units or ≥ 75th percentile, statin therapy is indicated for all patients, regardless of additional risk factors.⁴

Because statins promote progression from unstable, inflammatory atherosclerotic plaque to more stable, calcified plaque, CAC scoring is not valid in patients already on statin therapy.¹³

In primary prevention, patients who have been classified as having low or intermediate risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have an annual all-cause mortality < 1%, regardless of age and gender. Patients classified as being at high risk, based on ASCVD risk scoring, with a CAC score of 0 Agatston units, have a significantly lower annual mortality than low- or intermediate-risk patients with a CAC score > 0 Agatston units.¹⁴

• 20 to 39 years. Focus on evaluation of lifetime ASCVD risk, rather than short-term (10-year) risk. Lifestyle modification is the primary intervention for younger patients; for those with moderate hypercholesterolemia (LDL-C, 160-189 mg/dL) and a family history of premature ASCVD, however, consider statin therapy. For patients with LDL-C ≥ 190 mg/dL, lifetime ASCVD risk is markedly increased, and high-intensity statin therapy is recommended, regardless of age. In this group, reassess ASCVD risk factors every 4 to 6 years.⁴
• > 75 years, without ASCVD. In this group, the benefit of statin therapy is less clear and might be lessened by an increased potential for adverse effects. A meta-analysis of 28 trials demonstrated that people ages > 75 years had a 24% relative reduction in major coronary events for every 38.7mg/dL (1.0 mmol/L) reduction in LDL-C, which is comparable to the risk reduction seen in people ages 40 to 75 years.¹⁵

With increasing age, however, the relative reduction in major coronary events with statin therapy decreased,¹⁵ although other trials have not demonstrated age heterogeneity.¹⁶ Because people > 75 years of age have a significantly higher ASCVD event rate, a comparable relative rate reduction with statin therapy results in a larger absolute rate reduction (ARR) and, therefore, a smaller number needed to treat (NNT) to prevent an event, compared to the NNT in younger people.

Secondary prevention

ACC/AHA guidelines define clinical ASCVD as a history of:

• acute coronary syndrome
• myocardial infarction
• coronary or other arterial revascularization
• cerebrovascular event
• symptomatic peripheral artery disease, including aortic aneurysm.

High-intensity statin therapy is indicated for all patients ≤ 75 years who have clinical ASCVD. In patients > 75 years, consider a taper to moderate-intensity statin therapy. An upper age limit for seeing benefit from statin therapy in secondary prevention has not been identified.⁴

In high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy, ezetimibe (discussed in the next section) can be added. In very-high-risk patients, if LDL-C remains ≥ 70 mg/dL despite maximally tolerated statin therapy plus ezetimibe, a proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor (also discussed next) can be added. Always precede initiation of a PCSK9 inhibitor with a discussion of the net benefit, safety, and cost with the patient.⁴

Continue to: Options for lipid-lowering pharmacotherapy...

• Statins (formally, hydroxymethylglutaryl-coenzyme A reductase inhibitors) offer the most predictable reduction in ASCVD risk of any lipid-lowering therapy. The evidence report that accompanied the 2016 USPSTF guidelines on statins for the prevention of cardiovascular disease (CVD) stated that low- or moderate-dosage statin therapy is associated with approximately a 30% relative risk reduction (RRR) in CVD events and CVD deaths and a 10% to 15% RRR in all-cause mortality.¹⁷

High-intensity statin therapy reduces LDL-C by ≥ 50%. Moderate-intensity statin therapy reduces LDL-C by 30% to 49% (TABLE 3).⁴

Statins are not without risk: A 2016 report¹⁸ estimated that treating 10,000 patients with a statin for 5 years would cause 1 case of rhabdomyolysis, 5 cases of myopathy, 75 new cases of diabetes, and 7 cases of hemorrhagic stroke. The same treatment would, however, avert approximately 1000 CVD events among patients with preexisting disease and approximately 500 CVD events among patients at elevated risk but without preexisting disease.¹⁸

• Ezetimibe, a selective cholesterol-absorption inhibitor, lowers LDL-C by 13% to 20% and typically is well tolerated. The use of ezetimibe in ASCVD risk reduction is supported by a single randomized controlled trial of more than 18,000 patients with recent acute coronary syndrome. Adding ezetimibe to simvastatin 40 mg resulted in a 2% absolute reduction in major adverse cardiovascular events over a median follow-up of 6 years (NNT = 50), compared to simvastatin alone.¹⁹ ACC/AHA guidelines recommend adding ezetimibe to maximally tolerated statin therapy in patients with clinical ASCVD who do not reach their goal LDL reduction with a statin alone. Ezetimibe also can be considered a statin alternative in patients who are statin intolerant.⁴
• PCSK9 inhibitors. When added to statin therapy, evolocumab and alirocumab—monoclonal antibodies that inhibit PCSK9—offer an incremental decrease in LDL-C of approximately 60%.^20-22 In a meta-analysis of 35 trials evaluating the incremental benefit of PCSK9 inhibitor therapy, a significant reduction in cardiovascular events, including myocardial infarction (ARR = 1.3%; NNT = 77), stroke (ARR = 0.4%; NNT = 250), and coronary revascularization (ARR = 1.6%; NNT = 63) was reported. No significant difference was observed in all-cause or cardiovascular mortality.^21,23
• Inclisiran, an injectable small-interfering RNA that inhibits PCSK9 synthesis, provides an incremental decrease in LDL-C of > 50% in patients already receiving statin therapy. Meta-analysis of 3 small cardiovascular outcomes trials revealed no significant difference in the rate of myocardial infarction, stroke, or cardiovascular mortality with inclisiran compared to placebo. Larger outcomes trials are underway and might offer additional insight into this agent’s role in ASCVD risk management.²⁴
• Omega-3 fatty acids. Multiple trials have demonstrated that adding omega-3 fatty acids to usual lipid-lowering therapy does not offer a consistent reduction in adverse cardiovascular outcomes, despite providing a significant reduction in TG levels. In a high-risk population with persistently elevated TG despite statin therapy, icosapent ethyl, a purified eicosapentaenoic acid ethyl ester, reduced major ASCVD outcomes by 25% over a median 4.9 years (ARR = 4.8%; NNT = 21), and cardiovascular death by 20% (ARR = 0.9%; NNT = 111), compared with a mineral oil placebo.²⁵ Subsequent trials, using a corn oil placebo, failed to duplicate these data²⁶—raising concern that the mineral oil comparator might have altered results of the eicosapentaenoic acid ethyl ester study.^27,28
• Bempedoic acid is a small-molecule inhibitor of ATP citrate lyase that increases LDL uptake by the liver. Pooled data from studies of bempedoic acid show, on average, a 15% reduction in TC, a 23% reduction in LDL-C, and a 6% increase in HDL-C, without a significant change in TG.²⁹ In statin-intolerant patients, bempedoicacid reduced major ASCVD outcomes by 13% over a median 40 months (ARR = 1.6%; NNT = 63), with no significant reduction in cardiovascular death.³⁰
• Niacin. Two large trials failed to demonstrate improvement in major cardiovascular events or other clinical benefit when niacin is added to moderate-intensity statin therapy, despite a significant increase in the HDL-C level (on average, 6 mg/dL) and a decrease in the LDL-C level (10-12 mg/dL)and TG (42 mg/dL).^31,32
• Fenofibrate lowers TG and increases HDL-C but does not consistently improve cardiovascular outcomes.³³ In a trial of patients with type 2 diabetes and persistent dyslipidemia (serum TG > 204 mg/dL; HDL-C< 34 mg/dL) despite statin therapy, adding fenofibrate reduced CVD outcomes by 4.9%—although this absolute difference did not reach statistical significance.³⁴

Neither niacin nor fenofibrate is considered useful for reducing ASCVD risk across broad populations.⁴

Follow-up to assess progress toward goals

Recheck the lipid profile 4 to 12 weeks after starting lipid-lowering therapy to verify adherence to medication and assess response. The primary goal is the percentage reduction in LDL-C based on ASCVD risk. An additional goal for very-high-risk patients is an LDL-C value ≤ 70 mg/dL. If the reduction in LDL-C is less than desired and adherence is assured, consider titrating the statin dosage or augmenting statin therapy with a nonstatin drug (eg, ezetimibe), or both.⁴ ●

CORRESPONDENCE

Jonathon M. Firnhaber, MD, MAEd, MBA, East Carolina University, Family Medicine Center, 101 Heart Drive, Greenville, NC 27834; firnhaberj@ecu.edu

Whole exome sequencing’s role in diagnosing genetic causes of FGR with and without associated anomalies

Mone F, Mellis R, Gabriel H, et al. Should we offer prenatal exome sequencing for intrauterine growth restriction or short long bones? A systematic review and meta-analysis. Am J Obstet Gynecol. Published online October 7, 2022. doi:10.1016/j.ajog.2022.09.045

Multiple factors can play a role in FGR, including inherent maternal, placental, or fetal factors; the environment; and/or nutrition. However, prenatal diagnosis is an important consideration when exploring the underlying etiology for a growth-restricted fetus, especially in severe or early-onset cases. Many genetic conditions do not result in structural anomalies but can disrupt overall growth. Additionally, phenotyping in the prenatal period is limited and can miss more subtle physical differences that could point to a genetic cause. 

When compared with karyotype, chromosomal microarray (CMA) has been shown to increase the diagnostic yield in cases of isolated early FGR by 5%,^1,2 and the incidence of chromosomal abnormalities has been reported to be as high as 19% in this population. Let’s explore the data on exome sequencing for prenatal diagnosis in cases of isolated FGR. 

Meta-analysis details

In this meta-analysis, the authors reviewed 19 cohort studies or case series that investigated the yield of prenatal sequencing in fetuses with intrauterine growth restriction (IUGR) or short long bones, both in association with and without additional anomalies. All cases had nondiagnostic cytogenetic results. Fetal DNA in most cases was obtained through amniocentesis. Variants classified as likely pathogenic and pathogenic were considered diagnostic. The authors then calculated the incremental yield of prenatal sequencing over cytogenetic studies as a pooled value, comparing the following groups:

• isolated FGR
• growth restriction with associated anomalies
• isolated short long bones
• short long bones with additional skeletal features.  

Study outcomes 

The total number of cases were as follows: isolated IUGR (n = 71), IUGR associated with additional anomalies (n = 45), isolated short long bones (n = 84), and short long bones associated with additional skeletal findings (n = 252). Causative pathogenic or likely pathogenic variants were identified in 224 (50%) cases. Apparent incremental yields with prenatal sequencing were as follows for the 4 groups (as illustrated in the FIGURE):

• 4% in isolated IUGR (95% confidence interval [CI], -5%–12%)
• 30% in IUGR with additional anomalies (95% CI, 13%–47%)
• 48% in isolated short long bones (95% CI, 26%–70%)
• 68% in short long bones with additional skeletal changes (95% CI, 58%–77%).

Overall, the authors concluded that prenatal sequencing does not improve prenatal diagnosis in cases of isolated IUGR. The majority of these cases were thought to be related to placental insufficiency.

Strengths and limitations 

The main limitation of this study with regard to our discussion is the small study populationof isolated growth restriction. The authors indicate that the number of cases of isolated IUGR were too small to draw firm conclusions. Another limitation was the heterogeneity of the isolated FGR population, which was not limited to severe or early-onset cases. However, the authors did demonstrate that growth restriction in association with fetal anomalies has very high genetic yield rates with prenatal sequencing. 

Continue to: Can whole exome sequencing diagnose genetic causes in cases of severe isolated FGR?...

Zhou H, Fu F, Wang Y, et al. Genetic causes of isolated and severe fetal growth restriction in normal chromosomal microarray analysis. Int J Gynaecol Obstet. Published online December 10, 2022. doi:10.1002/ijgo.14620

Severe FGR is diagnosed based on an estimated fetal weight (EFW) or abdominal circumference (AC) below the third percentile. As we discussed in the above study by Mone and colleagues, it does not appear that prenatal sequencing significantly improves the diagnostic yield in all isolated FGR cases. However, this has not been previously explored in isolated severe FGR or cases of early-onset FGR (<32 weeks’ gestation). We know that several monogenic conditions are associated with severe and early-onset isolated fetal growth impairment, including but not limited to Cornelia de Lange syndrome, Smith-Lemli-Opitz syndrome, and Meier-Gorlin syndrome. Often, these syndromes can present in the prenatal period without other phenotypic findings. Therefore, this study explored the possibility that prenatal sequencing plays an important role for severe cases of FGR with nondiagnostic CMA and/or karyotype. 

Retrospective study details 

Zhou and colleagues retrospectively analyzed 51 cases of severe (EFW or AC below the third percentile) isolated FGR with negative CMA who underwent trio whole exome sequencing, which includes submitting fetal cells as well as both parental samples for testing. Patients with abnormal toxoplasmosis, rubella, cytomegalovirus, and herpes simplex virus (TORCH) tests; structural anomalies; and multiple gestation were excluded from the analysis. As in the study by Mone et al, variants classified as likely pathogenic and pathogenic were categorized as diagnostic. 

Results

Eight of 51 cases (15.7%) with severe isolated FGR had diagnostic findings on trio whole exome sequencing as shown in the TABLE. Another 8 cases (15.7%) were found to have variants of unknown significance, of which 2 were later determined to be novel pathogenic variants. Genetic conditions uncovered in this cohort include Cornelia de Lange syndrome, pyruvate dehydrogenase deficiency, Dent disease, trichohepaticenteric syndrome, achondroplasia, osteogenesis imperfecta, Pendred syndrome, and both autosomal dominant type 3A and autosomal recessive type 1A deafness. All 10 cases with diagnostic whole exome sequencing or identified novel pathogenic variants were affected by early-onset FGR (<32 weeks’ gestation). Of these 10 cases, 7 patients underwent pregnancy termination.

To summarize, a total of 10 cases (19.6%) of severe isolated early-onset FGR with negative cytogenetic studies were subsequently diagnosed with an underlying genetic condition using prenatal trio whole exome sequencing. 

Strengths and limitations 

This study is retrospective and has a small sample size (n = 51) that was mostly limited to early-onset isolated severe FGR. However, the diagnostic yield (19.6%) of whole exome sequencing after negative CMA testing was noteworthy and shows that monogenic conditions are an important consideration in the evaluation of severe early-onset FGR, even in the absence of structural abnormalities. 

Could epigenetic mechanisms of placental dysregulation explain low birthweight and future cardiometabolic disease?  

Tekola-Ayele F, Zeng X, Chatterjee S, et al. Placental multi-omics integration identifies candidate functional genes for birthweight. Nat Commun. 2022;13:2384.

FGR has been linked to greater mortality in childhood and increased risk for cardiometabolic disease in adulthood. While genomewide associations studies (GWAS) have defined areas of interest linking genetic variants to low birthweight, their relationship to epigenetic changes in the placenta as well as biologic and functional mechanisms are not yet well understood.  

Multiomics used to identify candidate functional genes for birthweight 

This study analyzed the methylation and gene expression patterns of 291 placental samples, integrating findings into pathways of previously defined GWAS variants. Patient samples were obtained from participants in the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies–Singleton cohort. The cohort is ethnically diverse, with 97 Hispanic, 74 White, 71 Black, and 49 Asian participants. Of 286 single nucleotide polymorphisms (SNPs) found to be associated with birthweight, 273 were analyzed as part of the authors’ data set. These were found to have 7,901 unique protein-coding mRNAs (expression quantitative trait loci [eQTL]) and more than 100,000 nearby (within 1 Mb) CpG islands thought to be involved in changes in DNA methylation (methylation quantitative trait loci [mQTL]). Each functionally connected GWAS-eQTL-mQTL association is referred to as a triplet.

The next arm of the study investigatedthe connections and pathways within each triplet. Three possible scenarios were explored for birthweight GWAS SNPs using a causal interference test (CIT):

1. the SNP alters placental DNA methylation, which then influences gene expression
2. the SNP first alters placental DNA expression, which then influences methylation
3. the SNP influences placental DNA expression and methylation independently, with no notable crossover between their pathways.

Triplets were investigated using the Mendelian randomization (MR) Steiger directionality test to validate the directionality of the pathways found by CIT. Lastly, the possibility of linkage disequilibrium was also studied using the moloc test. 

Results

Using CIT, a causal relationship was predicted in 88 of 197 triplets, in which 84 (95.5%) indicated DNA methylation influences gene expression, and 4 (4.5%) indicated gene expression influences DNA methylation. The authors also used the MR Steiger test to investigate triplets to identify possible causal pathways. Using the MR Steiger test, only 3 of 45 (7%) triplets were found to have independent gene expression and methylation pathways. Thirty-eight of 45 (84%) triplets indicated that gene expression influences DNA methylation, and 7 (15%)triplets demonstrated that DNA methylation influences gene expression. Consistent predictions between CIT and the MR Steiger test revealed 3 triplets in which DNA methylation influences gene expression for the following genes: WNT3A, CTDNEP1, and RANBP2. Additionally, a strong colocalization signal was found among birthweight, DNA methylation, and gene expression for the following genes: PLEKHA1, FES, PRMT7, and CTDNEP1. Gene set enrichment analysis was performed as well and found that low birthweight is associated in substantial upregulation of genes associated with oxidative stress, immune response, adipogenesis, myogenesis, and the production of pancreatic ß cells.  

Study strengths and limitations

The study is one of the first to identify regulatory targets for placental DNA methylation and gene expression in previously identified GWAS loci associated with low birthweight. For example, DNA methylation was found to influence gene expression of WNT3A, CTDNEP1, and RANBP2, which have previously been shown in animal studies to impact the vascularization and development of the placenta, embryogenesis, and fetal growth. The study also identified 4 genes (PLEKHA1, FES, PRMT7, and CTDNEP1) thought to have direct regulatory influence on placental DNA methylation and gene expression.

A limitation of the study is that it could not distinguish between whether the epigenetic changes we outlined have a maternal or fetal origin. Another limitation is that tissue used by the authors for analysis was a small placental biopsy, which does not accurately reflect the genetic heterogeneity of the placenta. Finally, this study does not establish causality between the identified epigenetic pathways and low birthweight. ●

Whole exome sequencing’s role in diagnosing genetic causes of FGR with and without associated anomalies

Mone F, Mellis R, Gabriel H, et al. Should we offer prenatal exome sequencing for intrauterine growth restriction or short long bones? A systematic review and meta-analysis. Am J Obstet Gynecol. Published online October 7, 2022. doi:10.1016/j.ajog.2022.09.045

Multiple factors can play a role in FGR, including inherent maternal, placental, or fetal factors; the environment; and/or nutrition. However, prenatal diagnosis is an important consideration when exploring the underlying etiology for a growth-restricted fetus, especially in severe or early-onset cases. Many genetic conditions do not result in structural anomalies but can disrupt overall growth. Additionally, phenotyping in the prenatal period is limited and can miss more subtle physical differences that could point to a genetic cause. 

When compared with karyotype, chromosomal microarray (CMA) has been shown to increase the diagnostic yield in cases of isolated early FGR by 5%,^1,2 and the incidence of chromosomal abnormalities has been reported to be as high as 19% in this population. Let’s explore the data on exome sequencing for prenatal diagnosis in cases of isolated FGR. 

Meta-analysis details

In this meta-analysis, the authors reviewed 19 cohort studies or case series that investigated the yield of prenatal sequencing in fetuses with intrauterine growth restriction (IUGR) or short long bones, both in association with and without additional anomalies. All cases had nondiagnostic cytogenetic results. Fetal DNA in most cases was obtained through amniocentesis. Variants classified as likely pathogenic and pathogenic were considered diagnostic. The authors then calculated the incremental yield of prenatal sequencing over cytogenetic studies as a pooled value, comparing the following groups:

• isolated FGR
• growth restriction with associated anomalies
• isolated short long bones
• short long bones with additional skeletal features.  

Study outcomes 

The total number of cases were as follows: isolated IUGR (n = 71), IUGR associated with additional anomalies (n = 45), isolated short long bones (n = 84), and short long bones associated with additional skeletal findings (n = 252). Causative pathogenic or likely pathogenic variants were identified in 224 (50%) cases. Apparent incremental yields with prenatal sequencing were as follows for the 4 groups (as illustrated in the FIGURE):

• 4% in isolated IUGR (95% confidence interval [CI], -5%–12%)
• 30% in IUGR with additional anomalies (95% CI, 13%–47%)
• 48% in isolated short long bones (95% CI, 26%–70%)
• 68% in short long bones with additional skeletal changes (95% CI, 58%–77%).

Overall, the authors concluded that prenatal sequencing does not improve prenatal diagnosis in cases of isolated IUGR. The majority of these cases were thought to be related to placental insufficiency.

Strengths and limitations 

The main limitation of this study with regard to our discussion is the small study populationof isolated growth restriction. The authors indicate that the number of cases of isolated IUGR were too small to draw firm conclusions. Another limitation was the heterogeneity of the isolated FGR population, which was not limited to severe or early-onset cases. However, the authors did demonstrate that growth restriction in association with fetal anomalies has very high genetic yield rates with prenatal sequencing. 

Continue to: Can whole exome sequencing diagnose genetic causes in cases of severe isolated FGR?...

Zhou H, Fu F, Wang Y, et al. Genetic causes of isolated and severe fetal growth restriction in normal chromosomal microarray analysis. Int J Gynaecol Obstet. Published online December 10, 2022. doi:10.1002/ijgo.14620

Severe FGR is diagnosed based on an estimated fetal weight (EFW) or abdominal circumference (AC) below the third percentile. As we discussed in the above study by Mone and colleagues, it does not appear that prenatal sequencing significantly improves the diagnostic yield in all isolated FGR cases. However, this has not been previously explored in isolated severe FGR or cases of early-onset FGR (<32 weeks’ gestation). We know that several monogenic conditions are associated with severe and early-onset isolated fetal growth impairment, including but not limited to Cornelia de Lange syndrome, Smith-Lemli-Opitz syndrome, and Meier-Gorlin syndrome. Often, these syndromes can present in the prenatal period without other phenotypic findings. Therefore, this study explored the possibility that prenatal sequencing plays an important role for severe cases of FGR with nondiagnostic CMA and/or karyotype. 

Retrospective study details 

Zhou and colleagues retrospectively analyzed 51 cases of severe (EFW or AC below the third percentile) isolated FGR with negative CMA who underwent trio whole exome sequencing, which includes submitting fetal cells as well as both parental samples for testing. Patients with abnormal toxoplasmosis, rubella, cytomegalovirus, and herpes simplex virus (TORCH) tests; structural anomalies; and multiple gestation were excluded from the analysis. As in the study by Mone et al, variants classified as likely pathogenic and pathogenic were categorized as diagnostic. 

Results

Eight of 51 cases (15.7%) with severe isolated FGR had diagnostic findings on trio whole exome sequencing as shown in the TABLE. Another 8 cases (15.7%) were found to have variants of unknown significance, of which 2 were later determined to be novel pathogenic variants. Genetic conditions uncovered in this cohort include Cornelia de Lange syndrome, pyruvate dehydrogenase deficiency, Dent disease, trichohepaticenteric syndrome, achondroplasia, osteogenesis imperfecta, Pendred syndrome, and both autosomal dominant type 3A and autosomal recessive type 1A deafness. All 10 cases with diagnostic whole exome sequencing or identified novel pathogenic variants were affected by early-onset FGR (<32 weeks’ gestation). Of these 10 cases, 7 patients underwent pregnancy termination.

To summarize, a total of 10 cases (19.6%) of severe isolated early-onset FGR with negative cytogenetic studies were subsequently diagnosed with an underlying genetic condition using prenatal trio whole exome sequencing. 

Strengths and limitations 

This study is retrospective and has a small sample size (n = 51) that was mostly limited to early-onset isolated severe FGR. However, the diagnostic yield (19.6%) of whole exome sequencing after negative CMA testing was noteworthy and shows that monogenic conditions are an important consideration in the evaluation of severe early-onset FGR, even in the absence of structural abnormalities. 

Could epigenetic mechanisms of placental dysregulation explain low birthweight and future cardiometabolic disease?  

Tekola-Ayele F, Zeng X, Chatterjee S, et al. Placental multi-omics integration identifies candidate functional genes for birthweight. Nat Commun. 2022;13:2384.

FGR has been linked to greater mortality in childhood and increased risk for cardiometabolic disease in adulthood. While genomewide associations studies (GWAS) have defined areas of interest linking genetic variants to low birthweight, their relationship to epigenetic changes in the placenta as well as biologic and functional mechanisms are not yet well understood.  

Multiomics used to identify candidate functional genes for birthweight 

This study analyzed the methylation and gene expression patterns of 291 placental samples, integrating findings into pathways of previously defined GWAS variants. Patient samples were obtained from participants in the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies–Singleton cohort. The cohort is ethnically diverse, with 97 Hispanic, 74 White, 71 Black, and 49 Asian participants. Of 286 single nucleotide polymorphisms (SNPs) found to be associated with birthweight, 273 were analyzed as part of the authors’ data set. These were found to have 7,901 unique protein-coding mRNAs (expression quantitative trait loci [eQTL]) and more than 100,000 nearby (within 1 Mb) CpG islands thought to be involved in changes in DNA methylation (methylation quantitative trait loci [mQTL]). Each functionally connected GWAS-eQTL-mQTL association is referred to as a triplet.

The next arm of the study investigatedthe connections and pathways within each triplet. Three possible scenarios were explored for birthweight GWAS SNPs using a causal interference test (CIT):

1. the SNP alters placental DNA methylation, which then influences gene expression
2. the SNP first alters placental DNA expression, which then influences methylation
3. the SNP influences placental DNA expression and methylation independently, with no notable crossover between their pathways.

Triplets were investigated using the Mendelian randomization (MR) Steiger directionality test to validate the directionality of the pathways found by CIT. Lastly, the possibility of linkage disequilibrium was also studied using the moloc test. 

Results

Using CIT, a causal relationship was predicted in 88 of 197 triplets, in which 84 (95.5%) indicated DNA methylation influences gene expression, and 4 (4.5%) indicated gene expression influences DNA methylation. The authors also used the MR Steiger test to investigate triplets to identify possible causal pathways. Using the MR Steiger test, only 3 of 45 (7%) triplets were found to have independent gene expression and methylation pathways. Thirty-eight of 45 (84%) triplets indicated that gene expression influences DNA methylation, and 7 (15%)triplets demonstrated that DNA methylation influences gene expression. Consistent predictions between CIT and the MR Steiger test revealed 3 triplets in which DNA methylation influences gene expression for the following genes: WNT3A, CTDNEP1, and RANBP2. Additionally, a strong colocalization signal was found among birthweight, DNA methylation, and gene expression for the following genes: PLEKHA1, FES, PRMT7, and CTDNEP1. Gene set enrichment analysis was performed as well and found that low birthweight is associated in substantial upregulation of genes associated with oxidative stress, immune response, adipogenesis, myogenesis, and the production of pancreatic ß cells.  

Study strengths and limitations

The study is one of the first to identify regulatory targets for placental DNA methylation and gene expression in previously identified GWAS loci associated with low birthweight. For example, DNA methylation was found to influence gene expression of WNT3A, CTDNEP1, and RANBP2, which have previously been shown in animal studies to impact the vascularization and development of the placenta, embryogenesis, and fetal growth. The study also identified 4 genes (PLEKHA1, FES, PRMT7, and CTDNEP1) thought to have direct regulatory influence on placental DNA methylation and gene expression.

A limitation of the study is that it could not distinguish between whether the epigenetic changes we outlined have a maternal or fetal origin. Another limitation is that tissue used by the authors for analysis was a small placental biopsy, which does not accurately reflect the genetic heterogeneity of the placenta. Finally, this study does not establish causality between the identified epigenetic pathways and low birthweight. ●

Whole exome sequencing’s role in diagnosing genetic causes of FGR with and without associated anomalies

Mone F, Mellis R, Gabriel H, et al. Should we offer prenatal exome sequencing for intrauterine growth restriction or short long bones? A systematic review and meta-analysis. Am J Obstet Gynecol. Published online October 7, 2022. doi:10.1016/j.ajog.2022.09.045

Multiple factors can play a role in FGR, including inherent maternal, placental, or fetal factors; the environment; and/or nutrition. However, prenatal diagnosis is an important consideration when exploring the underlying etiology for a growth-restricted fetus, especially in severe or early-onset cases. Many genetic conditions do not result in structural anomalies but can disrupt overall growth. Additionally, phenotyping in the prenatal period is limited and can miss more subtle physical differences that could point to a genetic cause. 

When compared with karyotype, chromosomal microarray (CMA) has been shown to increase the diagnostic yield in cases of isolated early FGR by 5%,^1,2 and the incidence of chromosomal abnormalities has been reported to be as high as 19% in this population. Let’s explore the data on exome sequencing for prenatal diagnosis in cases of isolated FGR. 

Meta-analysis details

In this meta-analysis, the authors reviewed 19 cohort studies or case series that investigated the yield of prenatal sequencing in fetuses with intrauterine growth restriction (IUGR) or short long bones, both in association with and without additional anomalies. All cases had nondiagnostic cytogenetic results. Fetal DNA in most cases was obtained through amniocentesis. Variants classified as likely pathogenic and pathogenic were considered diagnostic. The authors then calculated the incremental yield of prenatal sequencing over cytogenetic studies as a pooled value, comparing the following groups:

• isolated FGR
• growth restriction with associated anomalies
• isolated short long bones
• short long bones with additional skeletal features.  

Study outcomes 

The total number of cases were as follows: isolated IUGR (n = 71), IUGR associated with additional anomalies (n = 45), isolated short long bones (n = 84), and short long bones associated with additional skeletal findings (n = 252). Causative pathogenic or likely pathogenic variants were identified in 224 (50%) cases. Apparent incremental yields with prenatal sequencing were as follows for the 4 groups (as illustrated in the FIGURE):

• 4% in isolated IUGR (95% confidence interval [CI], -5%–12%)
• 30% in IUGR with additional anomalies (95% CI, 13%–47%)
• 48% in isolated short long bones (95% CI, 26%–70%)
• 68% in short long bones with additional skeletal changes (95% CI, 58%–77%).

Overall, the authors concluded that prenatal sequencing does not improve prenatal diagnosis in cases of isolated IUGR. The majority of these cases were thought to be related to placental insufficiency.

Strengths and limitations 

The main limitation of this study with regard to our discussion is the small study populationof isolated growth restriction. The authors indicate that the number of cases of isolated IUGR were too small to draw firm conclusions. Another limitation was the heterogeneity of the isolated FGR population, which was not limited to severe or early-onset cases. However, the authors did demonstrate that growth restriction in association with fetal anomalies has very high genetic yield rates with prenatal sequencing. 

Continue to: Can whole exome sequencing diagnose genetic causes in cases of severe isolated FGR?...

Zhou H, Fu F, Wang Y, et al. Genetic causes of isolated and severe fetal growth restriction in normal chromosomal microarray analysis. Int J Gynaecol Obstet. Published online December 10, 2022. doi:10.1002/ijgo.14620

Severe FGR is diagnosed based on an estimated fetal weight (EFW) or abdominal circumference (AC) below the third percentile. As we discussed in the above study by Mone and colleagues, it does not appear that prenatal sequencing significantly improves the diagnostic yield in all isolated FGR cases. However, this has not been previously explored in isolated severe FGR or cases of early-onset FGR (<32 weeks’ gestation). We know that several monogenic conditions are associated with severe and early-onset isolated fetal growth impairment, including but not limited to Cornelia de Lange syndrome, Smith-Lemli-Opitz syndrome, and Meier-Gorlin syndrome. Often, these syndromes can present in the prenatal period without other phenotypic findings. Therefore, this study explored the possibility that prenatal sequencing plays an important role for severe cases of FGR with nondiagnostic CMA and/or karyotype. 

Retrospective study details 

Zhou and colleagues retrospectively analyzed 51 cases of severe (EFW or AC below the third percentile) isolated FGR with negative CMA who underwent trio whole exome sequencing, which includes submitting fetal cells as well as both parental samples for testing. Patients with abnormal toxoplasmosis, rubella, cytomegalovirus, and herpes simplex virus (TORCH) tests; structural anomalies; and multiple gestation were excluded from the analysis. As in the study by Mone et al, variants classified as likely pathogenic and pathogenic were categorized as diagnostic. 

Results

Eight of 51 cases (15.7%) with severe isolated FGR had diagnostic findings on trio whole exome sequencing as shown in the TABLE. Another 8 cases (15.7%) were found to have variants of unknown significance, of which 2 were later determined to be novel pathogenic variants. Genetic conditions uncovered in this cohort include Cornelia de Lange syndrome, pyruvate dehydrogenase deficiency, Dent disease, trichohepaticenteric syndrome, achondroplasia, osteogenesis imperfecta, Pendred syndrome, and both autosomal dominant type 3A and autosomal recessive type 1A deafness. All 10 cases with diagnostic whole exome sequencing or identified novel pathogenic variants were affected by early-onset FGR (<32 weeks’ gestation). Of these 10 cases, 7 patients underwent pregnancy termination.

To summarize, a total of 10 cases (19.6%) of severe isolated early-onset FGR with negative cytogenetic studies were subsequently diagnosed with an underlying genetic condition using prenatal trio whole exome sequencing. 

Strengths and limitations 

This study is retrospective and has a small sample size (n = 51) that was mostly limited to early-onset isolated severe FGR. However, the diagnostic yield (19.6%) of whole exome sequencing after negative CMA testing was noteworthy and shows that monogenic conditions are an important consideration in the evaluation of severe early-onset FGR, even in the absence of structural abnormalities. 

Could epigenetic mechanisms of placental dysregulation explain low birthweight and future cardiometabolic disease?  

Tekola-Ayele F, Zeng X, Chatterjee S, et al. Placental multi-omics integration identifies candidate functional genes for birthweight. Nat Commun. 2022;13:2384.

FGR has been linked to greater mortality in childhood and increased risk for cardiometabolic disease in adulthood. While genomewide associations studies (GWAS) have defined areas of interest linking genetic variants to low birthweight, their relationship to epigenetic changes in the placenta as well as biologic and functional mechanisms are not yet well understood.  

Multiomics used to identify candidate functional genes for birthweight 

This study analyzed the methylation and gene expression patterns of 291 placental samples, integrating findings into pathways of previously defined GWAS variants. Patient samples were obtained from participants in the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies–Singleton cohort. The cohort is ethnically diverse, with 97 Hispanic, 74 White, 71 Black, and 49 Asian participants. Of 286 single nucleotide polymorphisms (SNPs) found to be associated with birthweight, 273 were analyzed as part of the authors’ data set. These were found to have 7,901 unique protein-coding mRNAs (expression quantitative trait loci [eQTL]) and more than 100,000 nearby (within 1 Mb) CpG islands thought to be involved in changes in DNA methylation (methylation quantitative trait loci [mQTL]). Each functionally connected GWAS-eQTL-mQTL association is referred to as a triplet.

The next arm of the study investigatedthe connections and pathways within each triplet. Three possible scenarios were explored for birthweight GWAS SNPs using a causal interference test (CIT):

1. the SNP alters placental DNA methylation, which then influences gene expression
2. the SNP first alters placental DNA expression, which then influences methylation
3. the SNP influences placental DNA expression and methylation independently, with no notable crossover between their pathways.

Triplets were investigated using the Mendelian randomization (MR) Steiger directionality test to validate the directionality of the pathways found by CIT. Lastly, the possibility of linkage disequilibrium was also studied using the moloc test. 

Results

Using CIT, a causal relationship was predicted in 88 of 197 triplets, in which 84 (95.5%) indicated DNA methylation influences gene expression, and 4 (4.5%) indicated gene expression influences DNA methylation. The authors also used the MR Steiger test to investigate triplets to identify possible causal pathways. Using the MR Steiger test, only 3 of 45 (7%) triplets were found to have independent gene expression and methylation pathways. Thirty-eight of 45 (84%) triplets indicated that gene expression influences DNA methylation, and 7 (15%)triplets demonstrated that DNA methylation influences gene expression. Consistent predictions between CIT and the MR Steiger test revealed 3 triplets in which DNA methylation influences gene expression for the following genes: WNT3A, CTDNEP1, and RANBP2. Additionally, a strong colocalization signal was found among birthweight, DNA methylation, and gene expression for the following genes: PLEKHA1, FES, PRMT7, and CTDNEP1. Gene set enrichment analysis was performed as well and found that low birthweight is associated in substantial upregulation of genes associated with oxidative stress, immune response, adipogenesis, myogenesis, and the production of pancreatic ß cells.  

Study strengths and limitations

The study is one of the first to identify regulatory targets for placental DNA methylation and gene expression in previously identified GWAS loci associated with low birthweight. For example, DNA methylation was found to influence gene expression of WNT3A, CTDNEP1, and RANBP2, which have previously been shown in animal studies to impact the vascularization and development of the placenta, embryogenesis, and fetal growth. The study also identified 4 genes (PLEKHA1, FES, PRMT7, and CTDNEP1) thought to have direct regulatory influence on placental DNA methylation and gene expression.

A limitation of the study is that it could not distinguish between whether the epigenetic changes we outlined have a maternal or fetal origin. Another limitation is that tissue used by the authors for analysis was a small placental biopsy, which does not accurately reflect the genetic heterogeneity of the placenta. Finally, this study does not establish causality between the identified epigenetic pathways and low birthweight. ●

Illustration: Kimberly Martens for OBG Management

Intrauterine devices (IUDs) are now used by more than 15% of US contraceptors. The majority of these IUDs release the progestin levonorgestrel, and with now longer extended use of the IUDs approved by the US Food and Drug Administration (FDA),^1-3 they become even more attractive for use for contraception,control of menorrhagia or heavy menstrual bleeding (HMB) during reproductive years and perimenopause, and potentially, although not FDA approved for this purpose, postmenopause for endometrial protection in estrogen users. In this roundtable discussion, we will look at some of the benefits of the IUD for contraception effectiveness and control of bleeding, as well as the potential risks if used for postmenopausal women.

Progestin IUDs and contraception

JoAnn V. Pinkerton, MD, NCMP: Dr. Kaunitz, what are the contraceptive benefits of progestin IUDs during perimenopause?
 

Andrew M. Kaunitz, MD, NCMP: We know fertility declines as women approach menopause. However, when pregnancy occurs in older reproductive-age women, the pregnancies are often unintended, as reflected by high rates of induced abortion in this population. In addition, the prevalence of maternal comorbidities (during pregnancy and delivery) is higher in older reproductive-age women, with the maternal mortality rate more than 5 times higher compared with that of younger women.⁴ Two recently published clinical trials assessed the extended use of full-size IUDs containing 52 mg of levonor-gestrel (LNG), with the brand names Mirena and Liletta.^1,2 The data from these trials confirmed that both IUDs remain highly effective for up to 8 years of use, and currently, both devices are approved for up to 8 years of use. One caveat is that, in the unusual occurrence of a pregnancy being diagnosed in a woman using an IUD, we as clinicians, must be alert to the high prevalence of ectopic pregnancies in this setting.

Progestin IUDs and HMB

Dr. Pinkerton: Dr. Goldstein, can you comment on how well progestin IUDs work for HMB?

Steven R. Goldstein, MD, NCMP, CCD: Many women who need contraception will use these devices for suppressing HMB, and they can be quite effective, if the diagnosis truly is HMB, at reducing bleeding.⁵ But that efficacy in bleeding reduction may not be quite as long as the efficacy in pregnancy prevention.⁶ In my experience, among women using IUDs specifically for their HMB, good bleeding control may require changing the IUD at 3 to 5 years.

Barbara S. Levy, MD: When inserting a LNG-IUD for menorrhagia in the perimenopausal time frame, sometimes I will do a progestin withdrawal first, which will thin the endometrium and induce withdrawal bleeding because, in my experience, if you place an IUD in someone with perimenopausal bleeding, you may end up with a lot of breakthrough bleeding.

Perimenopause and hot flashes

Dr. Pinkerton: Dr. Kaunitz, we have learned that hot flashes often occur and become bothersome to women during perimenopause. Many women have IUDs placed during perimenopause for bleeding. Can you comment about IUD use during perimenopause and postmenopause?
 

Dr. Kaunitz: In older reproductive-age women who already have a progestin-releasing IUD placed, as they get closer to menopause when vasomotor symptoms (VMS) might occur, if these symptoms are bothersome, the presence or placement of a progestin-releasing IUD can facilitate treatment of perimenopausal VMS with estrogen therapy.

Progestin IUDs cause profound endometrial suppression, reduce bleeding and often, over time, cause users to become amenorrheic.⁷

The Mirena package insert states, “Amenorrhea develops in about 20% of users by one year.”² By year 3 and continuing through year 8, the prevalence of amenorrhea with the 52-mg LNG-IUD is 35% to 40%.⁸ From a study by Nanette Santoro, MD, and colleagues, we know that, in perimenopausal women with a progestin-releasing IUD in place, who are experiencing bothersome VMS, adding transdermal estrogen is very effective in treating and suppressing those hot flashes. In her small clinical trial, among participants with perimenopausal bothersome VMS with an IUD in place, half were randomized to use of transdermal estradiol and then compared with women who did not get the estradiol patch. There was excellent relief of perimenopausal hot flashes with the combination of the progestin IUD for endometrial suppression and transdermal estrogen to relieve hot flashes.⁹

Dr. Pinkerton: Which women would not be good candidates for the use of this combination?

Dr. Kaunitz: We know that, as women age, the prevalence of conditions that are contraindications to combination contraceptives (estrogen-progestin pills, patches, or rings) starts to increase. Specifically, we see more: hypertension, diabetes, and high body mass index (BMI), or obesity. We also know that migraine headaches in women older than age 35 years is another condition in which ACOG and the Centers for Disease Control and Prevention (CDC) would not recommend use of combination contraceptives.^10,11 These older perimenopausal women may be excellent candidates for a progestin-only releasing IUD combined with use of transdermal menopausal doses of estradiol if needed for VMS.

Dr. Goldstein: I do want to add that, in those patients who don’t have these comorbidities, combination estrogen-progestin contraceptives do a very nice job of ovarian suppression and will prevent the erratic production of estradiol, which, in my experience, often results in not only irregular bleeding but also possible exacerbation of perimenopausal mood symptoms.

Dr. Kaunitz: I agree, Steve. The ideal older reproductive-age candidate for combination pills, patch, or ring would be a slender, healthy, nonsmoking woman with normal blood pressure. Such women would be a fairly small subgroup of my practice, but they can safely continue combination contraceptives right through menopause. Consistent with CDC and ACOG guidance, rather than checking gonadotropins to “determine when menopause has occurred,” (which is, in fact, not an evidence-based approach to diagnosing menopause in this setting), such women can continue the combination contraceptive right up until age 55—the likelihood that women are still going to be ovulating or at risk for pregnancy becomes vanishingly small at that age.^11,12 Women in their mid-50s can either seamlessly transition to use of systemic estrogen-progestin menopausal therapy or go off hormones completely.

Continue to: The IUD and HMB...

The IUD and HMB

Dr. Pinkerton: Dr. Goldstein, there’s been some good literature on the best management options for women with HMB. What is the most current evidence?

Dr. Goldstein: I think that the retiring of the terms menorrhagia and metrorrhagia may have been premature because HMB implies cyclical bleeding, and this population of women with HMB will typically do quite well. Women who have what we used to call metrorrhagia or irregular bleeding, by definition, need endometrial evaluation to be sure they don’t have some sort of organic pathology. It would be a mistake for clinicians to use an LNG-IUD in patients with abnormal uterine bleeding (AUB) that has not been appropriately evaluated.

If we understand that we are discussing HMB, a Cochrane Review from 2022¹³ suggests that an LNG intrauterine system is the best first-line treatment for reducing menstrual blood loss in perimenopausal women with HMB. Antifibrinolytics appeared second best, while long-cycle progestogens came in third place. Evidence on perception of improvement in satisfaction was ranked as low certainty. That same review found that hysterectomy was the best treatment for reducing bleeding, obviously, followed by resectoscopic endometrial ablation or a nonresectoscopic global endometrial ablation.

The evidence rating was low certainty regarding the likelihood that placing an LNG-IUD in women with HMB will result in amenorrhea, and I think that’s a very important point. The expectation of patients should be reduced or a significantly reduced amount of their HMB, not necessarily amenorrhea. Certainly, minimally invasive hysterectomy will result in total amenorrhea and may have a larger increase in satisfaction, but it has its own set of other kinds of possible complications.

Dr. Kaunitz: In an industry-funded, international multicenter trial,¹⁴ women with documented HMB (hemoglobin was eluted from soiled sanitary products), with menstrual blood loss of 80 mL or more per cycle, were randomized to placement of an LNG 52-mg IUD (Mirena) or cyclical medroxyprogesterone acetate (MPA)—oral progestin use.

Although menstrual blood loss declined in both groups, it declined dramatically more in women with an IUD placed, and specifically with the IUD, menstrual blood loss declined by 129 mL on average, whereas the decline in menstrual blood loss with cyclical MPA was 18 mL. This data, along with earlier European data,¹⁵ which showed similar findings in women with HMB led to the approval of the Mirena progestin IUD for a second indication to treat HMB in 2009.

I also want to point out that, in the May 2023 issue of Obstetrics & Gynecology, Creinin and colleagues published a similar trial in women with HMB showing, once again, that progestin IUDs (52-mg LNG-IUD, Liletta) are extremely effective in reducing HMB.¹⁶ There is crystal clear evidence from randomized trials that both 52-mg LNG-IUDs, Mirena and Liletta, are very effective in reducing HMB and, in fact, are contributing to many women who in the past would have proceeded with surgery, such as ablation or hysterectomy, to control their HMB.

Oral contraception

Dr. Pinkerton: What about using low-dose continuous oral contraceptives noncyclically for women with HMB?

Dr. Goldstein: I do that all the time. It is interesting that Dr. Kaunitz mentions his patient population. It’s why we understand that one size does not fit all. You need to see patients one at a time, and if they are good candidates for a combined estrogen-progestin contraception, whether it’s pills, patches, or rings, giving that continuously does a very nice job in reducing HMB and straightening out some of the other symptoms that these perimenopausal women will have.

IUD risks

Dr. Pinkerton: We all know about use of low-dose oral contraceptives for management of AUB, and we use them, although we worry a little bit about breast cancer risk. Dr. Levy, please comment on the risks with IUDs of expulsions and perforations. What are the downsides of IUDs?

Dr. Levy: Beyond the cost, although it is a minimally invasive procedure, IUD insertion can be an invasive procedure for a patient to undergo; expulsions can occur.¹⁷ We know that a substantial percentage of perimenopausal women will have fibroids. Although many fibroids are not located in the uterine cavity, the expulsion rate with HMB for an LNG-IUD can be higher,^13,16,18,19 perhaps because of local prostaglandin release with an increase in uterine contractility. There is a low incidence of perforations, but they do happen, particularly among women with scars in the uterus or who have a severely anteflexed or retroflexed uterus, and women with cervical stenosis, for example, if they have had a LEEP procedure, etc. Even though progestin IUDs are outstanding tools in our toolbox, they are invasive to some extent, and they do have the possibility of complications.

Dr. Kaunitz: As Dr. Levy points out, although placement of an IUD may be considered an invasive procedure, it is also an office-based procedure, so women can drive home or drive back to work afterwards without the disruption in their life and the potential complications associated with surgery and anesthesia.

Continue to: Concerns with malpositioning...

Dr. Pinkerton: After placement of an IUD, during a follow-up visit, sometimes you can’t visualize the string. The ultrasonography report may reveal, “IUD appears to be in the right place within the endometrium.” Dr. Goldstein, can you comment on how we should use ultrasound when we can’t visualize or find the IUD string, or if the patient complains of abdominal pain, lower abdominal discomfort, or irregular bleeding or spotting and we become concerned about IUD malposition?

Dr. Goldstein: Ultrasound is not really required after an uncomplicated placement of an IUD or during routine management of women who have no problems who are using an IUD. In patients who present with pain or some abnormal bleeding, however, sometimes it is the IUD being malpositioned. A very interesting study by the late great Beryl Benacerraf²⁰ showed that there was a statistically significant higher incidence of the IUD being poorly positioned when patients have pain or bleeding (FIGURE 1). It was not always apparent on 2D ultrasonography. Using a standard transvaginal ultrasound of the long access plane, the IUD may appear to be very centrally located. However, if you do a 3D coronal section, not infrequently in these patients with any pain or bleeding, one of the arms has pierced the myometrium (FIGURE 2). This can actually be a source of pain and bleeding.

It’s also very interesting when you talk about perforation. I became aware of a big to-do in the medical/legal world about the possibility of the IUD migrating through the uterine cavity.²¹ This just does not exist, as was already pointed out. If the IUD is really going to go anywhere, if it’s properly placed, it’s going to be expelled through an open cervix. I do believe that, if you have pierced the myometrium through uterine contractility over time, some of these IUDs could work their way through the myometrium and somehow come out of the uterus either totally or partially. I think ultrasound is invaluable in patients with pain and bleeding, but I think you need to have an ultrasound lab capable of doing a 3D coronal section.

Progestin IUDs for HT replacement: Benefits/risks

Dr. Pinkerton: Many clinicians are excited that they can use essentially estrogen alone for women who have a progestin IUD in place. What about the possible off-label use of the progestin IUD to replace oral progestogen for hormone therapy (HT)? Dr. Kaunitz, are there any studies using this for postmenopausal HT (with a reminder that the IUD is not FDA approved for this purpose)?

Dr. Kaunitz: We have data from Europe indicating that, in menopausal women using systemic estrogen, the full-size LNG 52 IUD—Mirena or Liletta—provides excellent endometrial suppression.²² Where we don’t have data is with the smaller IUDs, which would be Kyleena and Skyla, which release smaller amounts of progestin each day into the endometrial cavity.

I have a number of patients, most of them women who started use of a progestin IUD as older reproductive-age women and then started systemic estrogen for treatment of perimenopausal hot flashes and then continued the use of their IUD plus systemic estrogen in treating postmenopausal hot flashes. The IUD is very useful in this setting, but as you pointed out, Dr. Pinkerton, this does represent off-label use.

Dr. Pinkerton: I know this use does not affect plasma lipids or cardiovascular risk markers, although users seem to report that the IUD has improved their quality of life. The question comes up, what are the benefits on cancer risk for using an IUD?

Dr. Levy: It’s such a great question because, as we talk about the balance of risks and benefits for anything that we are offering to our patients, it is really important to focus on some of the benefits. For both the copper and the LNG-IUD, there is a reduction in endometrial cancer,²² as well as pretty good data with the copper IUD about a reduction in cervical cancer.²³ Those data are a little bit less clear for the LNG-IUD.

Interestingly, at least one meta-analysis published in 2020 shows about a 30% reduction in ovarian cancer risk with the LNG-IUD.²⁴ We need to focus our patients on these other benefits. They tend to focus on the risks, and, of course, the media blows up the risks, but the benefits are quite substantial beyond just reducing HMB and providing contraception.

Dr. Pinkerton: As Dr. Kaunitz said, when you use this IUD, with its primarily local uterine progestin effects, it’s more like using estrogen alone without as much systemic progestin. Recently I wrote an editorial on the benefits of estrogen alone on the risk of breast cancer, primarily based on the Women’s Health Initiative (WHI) observational long-term 18-year cumulative follow-up. When estrogen alone was prescribed to women after a hysterectomy, estrogen therapy used at menopause did not increase the risk of invasive breast cancer, and was associated with decreased mortality.²⁵ However, the nurse’s health study has suggested that longer-term use may be increased with estrogen alone.²⁶ For women in the WHI with an intact uterus who used estrogen, oral MPA slightly increased the risk for breast cancer, and this elevated risk persisted even after discontinuation. This leads us to the question, what are the risks of breast cancer with progestin IUD use?

I recently reviewed the literature, and the answer is, it’s mixed. The FDA has put language into the package label that acknowledges a potential breast cancer risk for women who use a progestin IUD,²⁷ and that warning states, “Women who currently have or have had breast cancer or suspect breast cancer should not use hormonal contraception because some breast cancers are hormone sensitive.” The label goes on to say, “Observational studies of the risk of breast cancer with the use of a levonorgestrel-releasing IUS don’t provide conclusive evidence of increased risk.” Thus, there is no conclusive answer as to whether there is a possible link of progestin IUDs to breast cancer.

What I tell my patients is that research is inconclusive. However, it’s unlikely for a 52-mg LNG-IUD to significantly increase a woman’s breast cancer risk, except possibly in those already at high risk from other risk factors. I tell them that breast cancer is listed in the package insert as a potential risk. I could not find any data on whether adding a low-dose estradiol patch would further increase that risk. So I counsel women about potential risk, but tell them that I don’t have any strong evidence of risk.

Continue to: Dr. Goldstein...

Dr. Goldstein: If you look in the package insert for Mirena,² similar to Liletta, certainly the serum levels of LNG are lower than that for combination oral contraceptives. For the IUD progestins, they are not localized only to the uterus, and LNG levels range from about 150 to 200 µg/mL up to 60 months. It’s greater at 12 months, at about 180 µg/mL,at 24 months it was 192 µg/mL, and by 60 months it was 159 µg/mL. It’s important to realize that there is some systemic absorption of progestin with progestin IUDs, and it is not completely a local effect.

JoAnn, you mentioned the WHI data,²⁵ and just to specify, it was not the estrogen-only arm, it was the conjugated equine estrogen-only arm of the WHI. I don’t think that estradiol alone increases breast cancer risk (although there are no good prospective, follow-through, 18-year study data, like the WHI), but I think readers need to understand the difference in the estrogen type.

Endometrial evaluation. My question for the panel is as follows. I agree that the use of the progestin-releasing IUD is very nice for that transition to menopause. I do believe it provides endometrial protection, but we know from other studies that, when we give continuous combined HT, about 21% to 26% of patients will experience some bleeding/staining, responding in the first 4-week cycles, and it can be as high as 9% at 1 year. If I have a patient who bleeds on continuous combined HT, I will evaluate her endometrium, usually just with a simple transvaginal ultrasound. If an IUD is in place, and the patient now begins to have some irregular bleeding, how do you evaluate her with the IUD in place?

Dr. Levy: That is a huge challenge. We know from a recent paper,²⁸ that the endometrial thickness, while an excellent measure for Caucasian and European women, may be a poor marker for endometrial pathology in African-American women. What we thought we knew, which was, if the stripe is 4 mL or less, we can forget about it, I think in our more recent research that is not so true. So you bring up a great point, what do you do? The most reliable evaluation will be with an office hysteroscopy, where you can really look at the entire cavity and for tiny, little polyps and other things. But then we are off label because the use of hysteroscopy with an IUD in place is off label. So we are really in a conundrum.

Dr. Pinkerton: Also, if you do an endometrial biopsy, you might dislodge the IUD. If you think that you are going to take the IUD out, it may not matter if you dislodge it. I will often obtain a transvaginal ultrasound to help me figure out the next step, and maybe look at the dosing of the estrogen and progestin—but you can’t monitor an IUD with blood levels. You are in a vacuum of trying to figure out the best thing to do.

Dr. Kaunitz: One of the hats I wear here in Jacksonville is Director of GYN Ultrasound. I have a fair amount of experience doing endometrial biopsies in women with progestin IUDs in place under abdominal ultrasound guidance and keeping a close eye on the position of the IUD. In the first dozen or so such procedures I did, I was quite concerned about dislodging the IUD. It hasn’t happened yet, and it gives me some reassurance to be able to image the IUD and your endometrial suction curette inside the cavity as you are obtaining endometrial sampling. I have substantial experience now doing that, and so far, no problems. I do counsel all such women in advance that there is some chance I could dislodge their IUD.

Dr. Goldstein: In addition to dislodging the IUD, are you not concerned that, if the pathology is not global, that a blind endometrial sampling may be fraught with some error?

Dr. Kaunitz: The endometrium in women with a progestin-releasing IUD in place tends to be very well suppressed. Although one might occasionally find, for instance, a polyp in that setting, I have not run into, and I don’t expect to encounter going forward, endometrial hyperplasia or cancer in women with current use of a progestin IUD. It’s possible but unlikely.

Dr. Levy: The progestin IUD will counterbalance a type-1 endometrial cancer—an endometrial cancer related to hyperstimulation by estrogen. It will not do anything, to my knowledge, to counterbalance a type 2. I think the art of medicine is, you do the best you can with the first episode of bleeding, and if she persists in her bleeding, we have to persevere and continue to evaluate her.

Dr. Goldstein: I agree 100%.

Dr. Pinkerton: We all agree with you. That’s a really good point.

Continue to: Case examinations...

Case examinations

CASE 1 Woman with intramural fibroids

Dr. Pinkerton: Dr. Goldstein, you have a 48-year-old Black woman who has heavy but regular menstrual bleeding with multiple fibroids (the largest is about 4 to 5 cm, they look intramural, with some distortion of the cavity but not a submucous myoma, and the endometrial depth is 9 cm). Would you insert an IUD, and would you recommend an endometrial biopsy first?

Dr. Goldstein: I am not a huge fan of blind endometrial sampling, and I do think that we use the “biopsy” somewhat inappropriately since sampling is not a directed biopsy. This became obvious in the landmark paper by Guido et al in 1995 and was adopted by ACOG only in 2012.²⁹ Cancers that occupy less than 50% of the endometrial surface area are often missed with such blind sampling. Thus I would not perform an endometrial biopsy first, but would rather rely on properly timed and performed transvaginal ultrasound to rule out any concurrent endometrial disease. I think a lot of patients who have HMB, not only because of their fibroids but also often just due to the surface area of their uterine cavity being increased—so essentially there is more blood volume when they bleed. However, you said that in this case the patient has regular menstrual bleeding, so I am assuming that she is still ovulatory. She may have some adenomyosis. She may have a large uterine cavity. I think she is an excellent candidate for an LNG-releasing IUD to reduce menstrual blood flow significantly. It will not necessarily give her amenorrhea, and it may give her some irregular bleeding. Then at some distant point, say in 5 or 6 months if she does have some irregular staining or bleeding, I would feel much better about the fact that nothing has developed as long as I knew that the endometrium was devoid of pathology when I started.

CASE 2 Woman with family history of breast cancer

Dr. Pinkerton: Dr. Levy, a 44-year-old woman has a family history of breast cancer in her mother at age 72, but she still needs contraceptionbecause of that unintended pregnancy risk in the 40s, and she wants something that is not going to increase her risk of breast cancer. What would you use, and how would you counsel her if you decided to use a progestin IUD?

Dr. Levy: The data are mixed,^30-33 but whatever the risk, it is miniscule, and I would bring up the CDC Medical Eligibility Criteria.¹¹ For a patient with a family history of breast cancer, for use of the progestin IUD, it is a 1—no contraindications. What I tend to tell my patients is, if you are worried about breast cancer, watch how much alcohol you are drinking and maintain regular exercise. There are so many preventive things that we can do to reduce risk of breast cancer when she needs contraception. If there is any increase in risk, it is so miniscule that I would very strongly recommend a progestin IUD for her.

Dr. Pinkerton: In addition, in recognizing the different densities of breast, dense breast density could lead to supplemental screening, which also could give her some reassurance that we are adequately screening for breast cancer.

CASE 3 Woman with IUD and VMS
 

Dr. Pinkerton: Dr. Kaunitz, you have a 52-year-old overweight female. She has been using a progestin IUD for 4 years, is amenorrheic, but now she is having moderate to severe vasomotor symptoms despite the IUD in place. You have talked to her about risks and benefits of HT, and she is interested in starting it. I know we talked about the studies, but I want to know what you are going to tell her. How do you counsel her about off-label use?

Dr. Kaunitz: The most important issue related to treating vasomotor symptoms in this patient is the route of systemic estrogen. Understandably, women’s biggest concern regarding the risks of systemic estrogen-progestin therapy is breast cancer. However, statistically, by far the biggest risk associated with oral estrogen-progestogen therapy, is elevated risk of venous thrombosis and pulmonary embolism. We have seen this, with a number of studies, and the WHI made it crystal clear with risks of oral conjugated equine estrogen at the dose of 0.625 mg daily. Oral estradiol 1 mg daily is also associated with a similar elevated risk of venous thrombosis. We also know that age and BMI are both independent risk factors for thrombosis. So, for a woman in her 50s who has a BMI > 30 mg/kg², I don’t want to further elevate her risk of thrombosis by giving her oral estrogen, whether it is estradiol or conjugated equine estrogen. This is a patient in whom I would be comfortable using transdermal (patch) estradiol, perhaps starting with a standard dose of 0.05 mg weekly or twice weekly patch, keeping in mind that 0.05 mg in the setting of transdermal estrogen refer to the daily or to the 24-hour release rate. The 1.0 mg of oral estradiol and 0.625 mg of conjugated equine estrogen refers to the mg quantity of estrogen in each tablet. This is a source of great confusion for clinicians.

If, during follow-up, the 0.05 mg estradiol patch is not sufficient to substantially reduce symptoms, we could go up, for instance, to a 0.075 mg estradiol patch. We know very clearly from a variety of observational studies, including a very large UK study,³⁴ that in contrast with oral estrogen, transdermal estradiol is safer from the perspective of thrombosis.

Insurance coverage for IUDs

Dr. Pinkerton: Dr. Levy: Can you discuss IUDs and the Affordable Care Act’s requirement to cover contraceptive services?

Dr. Levy: Unfortunately, we do not know whether this benefit will continue based on a very recent finding from a judge in Texas that ruled the preventive benefits of the ACA were illegal.³⁵ We don’t know what will happen going forward. What I will say is that, unfortunately, many insurance companies have not preserved the meaning of “cover all things,” so what we are finding is that, for example, they only have to cover one type in a class. The FDA defined 18 classes of contraceptives, and a hormonal IUD is one class, so they can decide that they are only going to cover one of the four IUDS. And then women don’t have access to the other three, some of which might be more appropriate for them than another.

The other thing very relevant to this conversation is that, if you use an ICD-10 code for menorrhagia, for HMB, it no longer lives within that ACA preventive care requirement of coverage for contraceptives, and now she is going to owe a big deductible or a copay. If you are practicing in an institution that does not allow the use of IUDs for contraception, like a Catholic institution where I used to practice, you will want to use that ICD-10 code for HMB. But if you want it offered with no out-of-pocket cost for the patient, you need to use the preventive medicine codes and the contraception code. These little nuances for us can make a huge difference for our patients.

Dr. Pinkerton: Thank you for that reminder. I want to thank our panelists, Dr. Levy, Dr. Goldstein, and Dr. Kaunitz, for providing us with such a great mix of evidence and expert opinion and also giving a benefit of their vast experience as award-winning gynecologists. Hopefully, today you have learned the benefits of the progestin IUD not only for contraception in reproductive years and perimenopause but also for treatment of HMB, and the potential benefit due to the more prolonged effectiveness of the IUDs for endometrial protection in postmenopause. This allows less progestin risk, essentially estrogen alone for postmenopausal HT. Unsolved questions remain about whether there is a risk of breast cancer with their use, but there is a clear benefit of protecting against pregnancy and endometrial cancer. ●

Illustration: Kimberly Martens for OBG Management

Intrauterine devices (IUDs) are now used by more than 15% of US contraceptors. The majority of these IUDs release the progestin levonorgestrel, and with now longer extended use of the IUDs approved by the US Food and Drug Administration (FDA),^1-3 they become even more attractive for use for contraception,control of menorrhagia or heavy menstrual bleeding (HMB) during reproductive years and perimenopause, and potentially, although not FDA approved for this purpose, postmenopause for endometrial protection in estrogen users. In this roundtable discussion, we will look at some of the benefits of the IUD for contraception effectiveness and control of bleeding, as well as the potential risks if used for postmenopausal women.

Progestin IUDs and contraception

JoAnn V. Pinkerton, MD, NCMP: Dr. Kaunitz, what are the contraceptive benefits of progestin IUDs during perimenopause?
 

Andrew M. Kaunitz, MD, NCMP: We know fertility declines as women approach menopause. However, when pregnancy occurs in older reproductive-age women, the pregnancies are often unintended, as reflected by high rates of induced abortion in this population. In addition, the prevalence of maternal comorbidities (during pregnancy and delivery) is higher in older reproductive-age women, with the maternal mortality rate more than 5 times higher compared with that of younger women.⁴ Two recently published clinical trials assessed the extended use of full-size IUDs containing 52 mg of levonor-gestrel (LNG), with the brand names Mirena and Liletta.^1,2 The data from these trials confirmed that both IUDs remain highly effective for up to 8 years of use, and currently, both devices are approved for up to 8 years of use. One caveat is that, in the unusual occurrence of a pregnancy being diagnosed in a woman using an IUD, we as clinicians, must be alert to the high prevalence of ectopic pregnancies in this setting.

Progestin IUDs and HMB

Dr. Pinkerton: Dr. Goldstein, can you comment on how well progestin IUDs work for HMB?

Steven R. Goldstein, MD, NCMP, CCD: Many women who need contraception will use these devices for suppressing HMB, and they can be quite effective, if the diagnosis truly is HMB, at reducing bleeding.⁵ But that efficacy in bleeding reduction may not be quite as long as the efficacy in pregnancy prevention.⁶ In my experience, among women using IUDs specifically for their HMB, good bleeding control may require changing the IUD at 3 to 5 years.

Barbara S. Levy, MD: When inserting a LNG-IUD for menorrhagia in the perimenopausal time frame, sometimes I will do a progestin withdrawal first, which will thin the endometrium and induce withdrawal bleeding because, in my experience, if you place an IUD in someone with perimenopausal bleeding, you may end up with a lot of breakthrough bleeding.

Perimenopause and hot flashes

Dr. Pinkerton: Dr. Kaunitz, we have learned that hot flashes often occur and become bothersome to women during perimenopause. Many women have IUDs placed during perimenopause for bleeding. Can you comment about IUD use during perimenopause and postmenopause?
 

Dr. Kaunitz: In older reproductive-age women who already have a progestin-releasing IUD placed, as they get closer to menopause when vasomotor symptoms (VMS) might occur, if these symptoms are bothersome, the presence or placement of a progestin-releasing IUD can facilitate treatment of perimenopausal VMS with estrogen therapy.

Progestin IUDs cause profound endometrial suppression, reduce bleeding and often, over time, cause users to become amenorrheic.⁷

The Mirena package insert states, “Amenorrhea develops in about 20% of users by one year.”² By year 3 and continuing through year 8, the prevalence of amenorrhea with the 52-mg LNG-IUD is 35% to 40%.⁸ From a study by Nanette Santoro, MD, and colleagues, we know that, in perimenopausal women with a progestin-releasing IUD in place, who are experiencing bothersome VMS, adding transdermal estrogen is very effective in treating and suppressing those hot flashes. In her small clinical trial, among participants with perimenopausal bothersome VMS with an IUD in place, half were randomized to use of transdermal estradiol and then compared with women who did not get the estradiol patch. There was excellent relief of perimenopausal hot flashes with the combination of the progestin IUD for endometrial suppression and transdermal estrogen to relieve hot flashes.⁹

Dr. Pinkerton: Which women would not be good candidates for the use of this combination?

Dr. Kaunitz: We know that, as women age, the prevalence of conditions that are contraindications to combination contraceptives (estrogen-progestin pills, patches, or rings) starts to increase. Specifically, we see more: hypertension, diabetes, and high body mass index (BMI), or obesity. We also know that migraine headaches in women older than age 35 years is another condition in which ACOG and the Centers for Disease Control and Prevention (CDC) would not recommend use of combination contraceptives.^10,11 These older perimenopausal women may be excellent candidates for a progestin-only releasing IUD combined with use of transdermal menopausal doses of estradiol if needed for VMS.

Dr. Goldstein: I do want to add that, in those patients who don’t have these comorbidities, combination estrogen-progestin contraceptives do a very nice job of ovarian suppression and will prevent the erratic production of estradiol, which, in my experience, often results in not only irregular bleeding but also possible exacerbation of perimenopausal mood symptoms.

Dr. Kaunitz: I agree, Steve. The ideal older reproductive-age candidate for combination pills, patch, or ring would be a slender, healthy, nonsmoking woman with normal blood pressure. Such women would be a fairly small subgroup of my practice, but they can safely continue combination contraceptives right through menopause. Consistent with CDC and ACOG guidance, rather than checking gonadotropins to “determine when menopause has occurred,” (which is, in fact, not an evidence-based approach to diagnosing menopause in this setting), such women can continue the combination contraceptive right up until age 55—the likelihood that women are still going to be ovulating or at risk for pregnancy becomes vanishingly small at that age.^11,12 Women in their mid-50s can either seamlessly transition to use of systemic estrogen-progestin menopausal therapy or go off hormones completely.

Continue to: The IUD and HMB...

The IUD and HMB

Dr. Pinkerton: Dr. Goldstein, there’s been some good literature on the best management options for women with HMB. What is the most current evidence?

Dr. Goldstein: I think that the retiring of the terms menorrhagia and metrorrhagia may have been premature because HMB implies cyclical bleeding, and this population of women with HMB will typically do quite well. Women who have what we used to call metrorrhagia or irregular bleeding, by definition, need endometrial evaluation to be sure they don’t have some sort of organic pathology. It would be a mistake for clinicians to use an LNG-IUD in patients with abnormal uterine bleeding (AUB) that has not been appropriately evaluated.

If we understand that we are discussing HMB, a Cochrane Review from 2022¹³ suggests that an LNG intrauterine system is the best first-line treatment for reducing menstrual blood loss in perimenopausal women with HMB. Antifibrinolytics appeared second best, while long-cycle progestogens came in third place. Evidence on perception of improvement in satisfaction was ranked as low certainty. That same review found that hysterectomy was the best treatment for reducing bleeding, obviously, followed by resectoscopic endometrial ablation or a nonresectoscopic global endometrial ablation.

The evidence rating was low certainty regarding the likelihood that placing an LNG-IUD in women with HMB will result in amenorrhea, and I think that’s a very important point. The expectation of patients should be reduced or a significantly reduced amount of their HMB, not necessarily amenorrhea. Certainly, minimally invasive hysterectomy will result in total amenorrhea and may have a larger increase in satisfaction, but it has its own set of other kinds of possible complications.

Dr. Kaunitz: In an industry-funded, international multicenter trial,¹⁴ women with documented HMB (hemoglobin was eluted from soiled sanitary products), with menstrual blood loss of 80 mL or more per cycle, were randomized to placement of an LNG 52-mg IUD (Mirena) or cyclical medroxyprogesterone acetate (MPA)—oral progestin use.

Although menstrual blood loss declined in both groups, it declined dramatically more in women with an IUD placed, and specifically with the IUD, menstrual blood loss declined by 129 mL on average, whereas the decline in menstrual blood loss with cyclical MPA was 18 mL. This data, along with earlier European data,¹⁵ which showed similar findings in women with HMB led to the approval of the Mirena progestin IUD for a second indication to treat HMB in 2009.

I also want to point out that, in the May 2023 issue of Obstetrics & Gynecology, Creinin and colleagues published a similar trial in women with HMB showing, once again, that progestin IUDs (52-mg LNG-IUD, Liletta) are extremely effective in reducing HMB.¹⁶ There is crystal clear evidence from randomized trials that both 52-mg LNG-IUDs, Mirena and Liletta, are very effective in reducing HMB and, in fact, are contributing to many women who in the past would have proceeded with surgery, such as ablation or hysterectomy, to control their HMB.

Oral contraception

Dr. Pinkerton: What about using low-dose continuous oral contraceptives noncyclically for women with HMB?

Dr. Goldstein: I do that all the time. It is interesting that Dr. Kaunitz mentions his patient population. It’s why we understand that one size does not fit all. You need to see patients one at a time, and if they are good candidates for a combined estrogen-progestin contraception, whether it’s pills, patches, or rings, giving that continuously does a very nice job in reducing HMB and straightening out some of the other symptoms that these perimenopausal women will have.

IUD risks

Dr. Pinkerton: We all know about use of low-dose oral contraceptives for management of AUB, and we use them, although we worry a little bit about breast cancer risk. Dr. Levy, please comment on the risks with IUDs of expulsions and perforations. What are the downsides of IUDs?

Dr. Levy: Beyond the cost, although it is a minimally invasive procedure, IUD insertion can be an invasive procedure for a patient to undergo; expulsions can occur.¹⁷ We know that a substantial percentage of perimenopausal women will have fibroids. Although many fibroids are not located in the uterine cavity, the expulsion rate with HMB for an LNG-IUD can be higher,^13,16,18,19 perhaps because of local prostaglandin release with an increase in uterine contractility. There is a low incidence of perforations, but they do happen, particularly among women with scars in the uterus or who have a severely anteflexed or retroflexed uterus, and women with cervical stenosis, for example, if they have had a LEEP procedure, etc. Even though progestin IUDs are outstanding tools in our toolbox, they are invasive to some extent, and they do have the possibility of complications.

Dr. Kaunitz: As Dr. Levy points out, although placement of an IUD may be considered an invasive procedure, it is also an office-based procedure, so women can drive home or drive back to work afterwards without the disruption in their life and the potential complications associated with surgery and anesthesia.

Continue to: Concerns with malpositioning...

Dr. Pinkerton: After placement of an IUD, during a follow-up visit, sometimes you can’t visualize the string. The ultrasonography report may reveal, “IUD appears to be in the right place within the endometrium.” Dr. Goldstein, can you comment on how we should use ultrasound when we can’t visualize or find the IUD string, or if the patient complains of abdominal pain, lower abdominal discomfort, or irregular bleeding or spotting and we become concerned about IUD malposition?

Dr. Goldstein: Ultrasound is not really required after an uncomplicated placement of an IUD or during routine management of women who have no problems who are using an IUD. In patients who present with pain or some abnormal bleeding, however, sometimes it is the IUD being malpositioned. A very interesting study by the late great Beryl Benacerraf²⁰ showed that there was a statistically significant higher incidence of the IUD being poorly positioned when patients have pain or bleeding (FIGURE 1). It was not always apparent on 2D ultrasonography. Using a standard transvaginal ultrasound of the long access plane, the IUD may appear to be very centrally located. However, if you do a 3D coronal section, not infrequently in these patients with any pain or bleeding, one of the arms has pierced the myometrium (FIGURE 2). This can actually be a source of pain and bleeding.

It’s also very interesting when you talk about perforation. I became aware of a big to-do in the medical/legal world about the possibility of the IUD migrating through the uterine cavity.²¹ This just does not exist, as was already pointed out. If the IUD is really going to go anywhere, if it’s properly placed, it’s going to be expelled through an open cervix. I do believe that, if you have pierced the myometrium through uterine contractility over time, some of these IUDs could work their way through the myometrium and somehow come out of the uterus either totally or partially. I think ultrasound is invaluable in patients with pain and bleeding, but I think you need to have an ultrasound lab capable of doing a 3D coronal section.

Progestin IUDs for HT replacement: Benefits/risks

Dr. Pinkerton: Many clinicians are excited that they can use essentially estrogen alone for women who have a progestin IUD in place. What about the possible off-label use of the progestin IUD to replace oral progestogen for hormone therapy (HT)? Dr. Kaunitz, are there any studies using this for postmenopausal HT (with a reminder that the IUD is not FDA approved for this purpose)?

Dr. Kaunitz: We have data from Europe indicating that, in menopausal women using systemic estrogen, the full-size LNG 52 IUD—Mirena or Liletta—provides excellent endometrial suppression.²² Where we don’t have data is with the smaller IUDs, which would be Kyleena and Skyla, which release smaller amounts of progestin each day into the endometrial cavity.

I have a number of patients, most of them women who started use of a progestin IUD as older reproductive-age women and then started systemic estrogen for treatment of perimenopausal hot flashes and then continued the use of their IUD plus systemic estrogen in treating postmenopausal hot flashes. The IUD is very useful in this setting, but as you pointed out, Dr. Pinkerton, this does represent off-label use.

Dr. Pinkerton: I know this use does not affect plasma lipids or cardiovascular risk markers, although users seem to report that the IUD has improved their quality of life. The question comes up, what are the benefits on cancer risk for using an IUD?

Dr. Levy: It’s such a great question because, as we talk about the balance of risks and benefits for anything that we are offering to our patients, it is really important to focus on some of the benefits. For both the copper and the LNG-IUD, there is a reduction in endometrial cancer,²² as well as pretty good data with the copper IUD about a reduction in cervical cancer.²³ Those data are a little bit less clear for the LNG-IUD.

Interestingly, at least one meta-analysis published in 2020 shows about a 30% reduction in ovarian cancer risk with the LNG-IUD.²⁴ We need to focus our patients on these other benefits. They tend to focus on the risks, and, of course, the media blows up the risks, but the benefits are quite substantial beyond just reducing HMB and providing contraception.

Dr. Pinkerton: As Dr. Kaunitz said, when you use this IUD, with its primarily local uterine progestin effects, it’s more like using estrogen alone without as much systemic progestin. Recently I wrote an editorial on the benefits of estrogen alone on the risk of breast cancer, primarily based on the Women’s Health Initiative (WHI) observational long-term 18-year cumulative follow-up. When estrogen alone was prescribed to women after a hysterectomy, estrogen therapy used at menopause did not increase the risk of invasive breast cancer, and was associated with decreased mortality.²⁵ However, the nurse’s health study has suggested that longer-term use may be increased with estrogen alone.²⁶ For women in the WHI with an intact uterus who used estrogen, oral MPA slightly increased the risk for breast cancer, and this elevated risk persisted even after discontinuation. This leads us to the question, what are the risks of breast cancer with progestin IUD use?

I recently reviewed the literature, and the answer is, it’s mixed. The FDA has put language into the package label that acknowledges a potential breast cancer risk for women who use a progestin IUD,²⁷ and that warning states, “Women who currently have or have had breast cancer or suspect breast cancer should not use hormonal contraception because some breast cancers are hormone sensitive.” The label goes on to say, “Observational studies of the risk of breast cancer with the use of a levonorgestrel-releasing IUS don’t provide conclusive evidence of increased risk.” Thus, there is no conclusive answer as to whether there is a possible link of progestin IUDs to breast cancer.

What I tell my patients is that research is inconclusive. However, it’s unlikely for a 52-mg LNG-IUD to significantly increase a woman’s breast cancer risk, except possibly in those already at high risk from other risk factors. I tell them that breast cancer is listed in the package insert as a potential risk. I could not find any data on whether adding a low-dose estradiol patch would further increase that risk. So I counsel women about potential risk, but tell them that I don’t have any strong evidence of risk.

Continue to: Dr. Goldstein...

Dr. Goldstein: If you look in the package insert for Mirena,² similar to Liletta, certainly the serum levels of LNG are lower than that for combination oral contraceptives. For the IUD progestins, they are not localized only to the uterus, and LNG levels range from about 150 to 200 µg/mL up to 60 months. It’s greater at 12 months, at about 180 µg/mL,at 24 months it was 192 µg/mL, and by 60 months it was 159 µg/mL. It’s important to realize that there is some systemic absorption of progestin with progestin IUDs, and it is not completely a local effect.

JoAnn, you mentioned the WHI data,²⁵ and just to specify, it was not the estrogen-only arm, it was the conjugated equine estrogen-only arm of the WHI. I don’t think that estradiol alone increases breast cancer risk (although there are no good prospective, follow-through, 18-year study data, like the WHI), but I think readers need to understand the difference in the estrogen type.

Endometrial evaluation. My question for the panel is as follows. I agree that the use of the progestin-releasing IUD is very nice for that transition to menopause. I do believe it provides endometrial protection, but we know from other studies that, when we give continuous combined HT, about 21% to 26% of patients will experience some bleeding/staining, responding in the first 4-week cycles, and it can be as high as 9% at 1 year. If I have a patient who bleeds on continuous combined HT, I will evaluate her endometrium, usually just with a simple transvaginal ultrasound. If an IUD is in place, and the patient now begins to have some irregular bleeding, how do you evaluate her with the IUD in place?

Dr. Levy: That is a huge challenge. We know from a recent paper,²⁸ that the endometrial thickness, while an excellent measure for Caucasian and European women, may be a poor marker for endometrial pathology in African-American women. What we thought we knew, which was, if the stripe is 4 mL or less, we can forget about it, I think in our more recent research that is not so true. So you bring up a great point, what do you do? The most reliable evaluation will be with an office hysteroscopy, where you can really look at the entire cavity and for tiny, little polyps and other things. But then we are off label because the use of hysteroscopy with an IUD in place is off label. So we are really in a conundrum.

Dr. Pinkerton: Also, if you do an endometrial biopsy, you might dislodge the IUD. If you think that you are going to take the IUD out, it may not matter if you dislodge it. I will often obtain a transvaginal ultrasound to help me figure out the next step, and maybe look at the dosing of the estrogen and progestin—but you can’t monitor an IUD with blood levels. You are in a vacuum of trying to figure out the best thing to do.

Dr. Kaunitz: One of the hats I wear here in Jacksonville is Director of GYN Ultrasound. I have a fair amount of experience doing endometrial biopsies in women with progestin IUDs in place under abdominal ultrasound guidance and keeping a close eye on the position of the IUD. In the first dozen or so such procedures I did, I was quite concerned about dislodging the IUD. It hasn’t happened yet, and it gives me some reassurance to be able to image the IUD and your endometrial suction curette inside the cavity as you are obtaining endometrial sampling. I have substantial experience now doing that, and so far, no problems. I do counsel all such women in advance that there is some chance I could dislodge their IUD.

Dr. Goldstein: In addition to dislodging the IUD, are you not concerned that, if the pathology is not global, that a blind endometrial sampling may be fraught with some error?

Dr. Kaunitz: The endometrium in women with a progestin-releasing IUD in place tends to be very well suppressed. Although one might occasionally find, for instance, a polyp in that setting, I have not run into, and I don’t expect to encounter going forward, endometrial hyperplasia or cancer in women with current use of a progestin IUD. It’s possible but unlikely.

Dr. Levy: The progestin IUD will counterbalance a type-1 endometrial cancer—an endometrial cancer related to hyperstimulation by estrogen. It will not do anything, to my knowledge, to counterbalance a type 2. I think the art of medicine is, you do the best you can with the first episode of bleeding, and if she persists in her bleeding, we have to persevere and continue to evaluate her.

Dr. Goldstein: I agree 100%.

Dr. Pinkerton: We all agree with you. That’s a really good point.

Continue to: Case examinations...

Case examinations

CASE 1 Woman with intramural fibroids

Dr. Pinkerton: Dr. Goldstein, you have a 48-year-old Black woman who has heavy but regular menstrual bleeding with multiple fibroids (the largest is about 4 to 5 cm, they look intramural, with some distortion of the cavity but not a submucous myoma, and the endometrial depth is 9 cm). Would you insert an IUD, and would you recommend an endometrial biopsy first?

Dr. Goldstein: I am not a huge fan of blind endometrial sampling, and I do think that we use the “biopsy” somewhat inappropriately since sampling is not a directed biopsy. This became obvious in the landmark paper by Guido et al in 1995 and was adopted by ACOG only in 2012.²⁹ Cancers that occupy less than 50% of the endometrial surface area are often missed with such blind sampling. Thus I would not perform an endometrial biopsy first, but would rather rely on properly timed and performed transvaginal ultrasound to rule out any concurrent endometrial disease. I think a lot of patients who have HMB, not only because of their fibroids but also often just due to the surface area of their uterine cavity being increased—so essentially there is more blood volume when they bleed. However, you said that in this case the patient has regular menstrual bleeding, so I am assuming that she is still ovulatory. She may have some adenomyosis. She may have a large uterine cavity. I think she is an excellent candidate for an LNG-releasing IUD to reduce menstrual blood flow significantly. It will not necessarily give her amenorrhea, and it may give her some irregular bleeding. Then at some distant point, say in 5 or 6 months if she does have some irregular staining or bleeding, I would feel much better about the fact that nothing has developed as long as I knew that the endometrium was devoid of pathology when I started.

CASE 2 Woman with family history of breast cancer

Dr. Pinkerton: Dr. Levy, a 44-year-old woman has a family history of breast cancer in her mother at age 72, but she still needs contraceptionbecause of that unintended pregnancy risk in the 40s, and she wants something that is not going to increase her risk of breast cancer. What would you use, and how would you counsel her if you decided to use a progestin IUD?

Dr. Levy: The data are mixed,^30-33 but whatever the risk, it is miniscule, and I would bring up the CDC Medical Eligibility Criteria.¹¹ For a patient with a family history of breast cancer, for use of the progestin IUD, it is a 1—no contraindications. What I tend to tell my patients is, if you are worried about breast cancer, watch how much alcohol you are drinking and maintain regular exercise. There are so many preventive things that we can do to reduce risk of breast cancer when she needs contraception. If there is any increase in risk, it is so miniscule that I would very strongly recommend a progestin IUD for her.

Dr. Pinkerton: In addition, in recognizing the different densities of breast, dense breast density could lead to supplemental screening, which also could give her some reassurance that we are adequately screening for breast cancer.

CASE 3 Woman with IUD and VMS
 

Dr. Pinkerton: Dr. Kaunitz, you have a 52-year-old overweight female. She has been using a progestin IUD for 4 years, is amenorrheic, but now she is having moderate to severe vasomotor symptoms despite the IUD in place. You have talked to her about risks and benefits of HT, and she is interested in starting it. I know we talked about the studies, but I want to know what you are going to tell her. How do you counsel her about off-label use?

Dr. Kaunitz: The most important issue related to treating vasomotor symptoms in this patient is the route of systemic estrogen. Understandably, women’s biggest concern regarding the risks of systemic estrogen-progestin therapy is breast cancer. However, statistically, by far the biggest risk associated with oral estrogen-progestogen therapy, is elevated risk of venous thrombosis and pulmonary embolism. We have seen this, with a number of studies, and the WHI made it crystal clear with risks of oral conjugated equine estrogen at the dose of 0.625 mg daily. Oral estradiol 1 mg daily is also associated with a similar elevated risk of venous thrombosis. We also know that age and BMI are both independent risk factors for thrombosis. So, for a woman in her 50s who has a BMI > 30 mg/kg², I don’t want to further elevate her risk of thrombosis by giving her oral estrogen, whether it is estradiol or conjugated equine estrogen. This is a patient in whom I would be comfortable using transdermal (patch) estradiol, perhaps starting with a standard dose of 0.05 mg weekly or twice weekly patch, keeping in mind that 0.05 mg in the setting of transdermal estrogen refer to the daily or to the 24-hour release rate. The 1.0 mg of oral estradiol and 0.625 mg of conjugated equine estrogen refers to the mg quantity of estrogen in each tablet. This is a source of great confusion for clinicians.

If, during follow-up, the 0.05 mg estradiol patch is not sufficient to substantially reduce symptoms, we could go up, for instance, to a 0.075 mg estradiol patch. We know very clearly from a variety of observational studies, including a very large UK study,³⁴ that in contrast with oral estrogen, transdermal estradiol is safer from the perspective of thrombosis.