Ramiro Jervis, MD

The terms “hospital medicine” and “evidence-based medicine” (EBM) are both recent arrivals in the history of medicine. Both have spread through medicine at a rapid pace, highlighting the attraction and fundamental soundness of their core ideas. Much has been written about the benefits of the hospitalist movement regarding quality, patient throughput and financial indicators. The next phase in the revolution of patient care is the confluence of technology, EBM, and hospital medicine. One of the pillars for the continued success of the hospital medicine movement will be EBM. EBM must become an integral part of the skill set for all hospitalists.

EBM is an analytical approach with a fundamental knowledge base and a set of tools. The exponential growth of clinical information requires that physicians use an analytical approach for answering clinical questions and keeping up-to-date. This may be easier if you work at an academic center rather than a non-teaching hospital, although this is not guaranteed. Regardless of the working environment, an analytical approach will be needed if we are to build on initial success and unrealized potential to improve quality and patient safety.

The term “EBM” was introduced by a group of clinician researchers and educators at McMaster University during the early 1990s. It was initially defined as “a systemic approach to analyzed published research as the basis of clinical decision making.” Subsequently, as the EBM movement matured as a discipline, the early proponents and developers provided a more complete definition: “Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research (1).”

click for large version

Of course, the concept of practicing medicine based on the scientific method has been around for years. The days of bloodletting with leeches are behind us, but rigorous scientific evaluation of medicine reached critical mass only in the last century. The first double-blind randomized controlled trial (RCT) was conducted in 1931; the study tested the use of sanocrysin for treatment of tuberculosis (2). Since then, there has been an exponential growth in clinical trials. This information explosion requires new approaches to integrating the ever-increasing knowledge with patient care; the concurrent revolution in information technology provides opportunities limited only by our own imaginations.

EBM is not without its critics: “It is cookbook medicine,” “It focuses on cost efficiency,” “I can’t find an RCT that fits my patient,” and “It doesn’t take into account the clinician’s experience” are often argued points. If EBM is not used effectively, all the criticisms are appropriate. Sacket, one of the earliest proponents and likely EBM’s most eloquent champion, refuted the claim that EBM is “cookbook medicine.” He argued that EBM requires a bottom-up approach that integrates the best external evidence with individual clinical expertise and patients’ choice. External clinical evidence informs, but does not replace, individual clinical expertise. The physician must decide whether the external evidence applies to the individual patient at all and, if so, how it should be integrated into a clinical decision (1).

Neither is EBM strictly about RCTs and meta-analysis. It’s about tracking down the best available evidence for your question and, thus, your patient. Sometimes a cohort study is best when you want to find the prognosis of a certain illness. Similarly, a cross-sectional study may be most appropriate when you’re trying to determine the sensitivity and specificity of a test. If a disease once thought universally fatal is proven otherwise in a case report, then a randomized control trial is hardly necessary. Finally, an RCT or meta-analysis is not always going to be available for the disease process you are dealing with, but EBM gives us the skill set to look down the evidence pyramid and find the next best thing (Figure 1).

It is also clear that EBM is not about cost cutting, although many hospital medicine programs were started with this as the primary goal, given the current healthcare environment. Certainly fears exist that EBM is being used by healthcare managers, organizations, and administrators as a cost-efficiency tool. It may be that good evidence is cost-efficient in certain situations, while in others it may require the healthcare system to invest more in itself if available evidence supports doing so. Thus it is imperative that hospitalists accept the challenge of incorporating EBM into their daily practice and become leaders in its application, with patient safety and quality of care as primary goals. If we don’t, others will define the role of EBM for us, with a potential for poor outcomes for the patient and the profession.

The EBM skill set and its tools are being continuously refined, with the evidence pyramid as one of the most basic principles (3). This evidence pyramid is a model for grading the evidence. It puts in perspective the different grades of evidence or study designs. For example, a systematic review of randomized controlled trials that show consistent results provides the highest quality evidence and is ranked accordingly in the pyramid. In contrast, a case report or case series of a treatment would be ranked much lower.

The first step in incorporating EBM into one’s daily practice requires an understanding of its analytical approach and access to the necessary tools. The process begins with asking a question that is answerable. A well-built clinical question is one that benefits the patient and clinician. Such questions are directly relevant to patient problems and phrased in ways that direct your search to relevant and precise answers.

click for large version

In forming the question the following process, referred to as the PICO method, is helpful (4).

With the question formed, consider what type of question you have. This is often referred to as the typology of the question.

1. Clinical Findings: Gathering and interpreting findings from the history, clinical examination, and test results.
2. Etiology: Identifying causes for disease.
3. Differential Diagnosis: Ranking by likelihood, seriousness, and treatability of the patients problem.
4. Prognosis: Figuring out how to estimate the likely clinical course and complications over time of the disease
5. Therapy: Selecting treatments to offer that do more good than harm and that are worth the effort and cost of using them.
6. Prevention: Reducing the chance of disease by identifying and modifying risk factors and how to diagnose disease early by screening.
7. Self-improvement: Keeping up-to-date, improve your clinical skills, and run a better, more efficient clinical practice.

The types of questions can next be matched to the type of research that may provide the answer:

• Diagnosis: prospective cohort study with good quality validation against “gold standard.”
• Prognosis: prospective cohort study.
• Therapy or prevention: prospective, randomized controlled clinical trial (RCT).
• Harm/Etiology: RCT, cohort or case-control study (probably retrospective).
• Economic: analysis of sensible costs against evidence-based outcome.

Once the question has been formed, the following steps lie ahead: finding the evidence, critically appraising the evidence, acting on the evidence, and, finally, evaluating one’s performance. Very much like the formation of the question, each of the subsequent steps involves an analytical approach that can be mastered. Technology—particularly personal computers, the Internet and PDAs—has made the task of mastering EBM easier in many ways. The additional steps in using EBM effectively will be addressed in future articles. A list of useful links is provided below.

http://www.cebm.utoronto.ca/

http://library.downstate.edu/EBM2/contents.htm

http://healthsystem.virginia.edu/internet/library/collections/ebm/index.cfm

http://www.ebmny.org

http://www.cebm.net/

Dr. Kathuria may be reached at Navneet.kathuria@mssm.edu.

Endnotes

1. Guyatt GH, Haynes RB, Jaeschke RZ, et al. Users’ guides to the medical literature: XXV. Evidence-based medicine: principles for applying the users’ guides to patient care. Evidence-Based Medicine Working Group. JAMA. 2000;284:1290-6.
2. Claridge, J, Fabian, T. History and Development of Evidence Based Medicine. World Journal of Surgery 2005
3. Guyatt GH, Haynes RB, et. al. Users’ guides to the medical literature: XXV. Evidence-Based Medicine: Principles for Applying the Users’ Guides to Patient Care. 2000;284:1290-1296
4. Sackett DL, Richardson WS, Rosenberg W, Haynes RB (1997). Evidence-based medicine: How to practice and teach EBM. New York: Churchill Livingston

The terms “hospital medicine” and “evidence-based medicine” (EBM) are both recent arrivals in the history of medicine. Both have spread through medicine at a rapid pace, highlighting the attraction and fundamental soundness of their core ideas. Much has been written about the benefits of the hospitalist movement regarding quality, patient throughput and financial indicators. The next phase in the revolution of patient care is the confluence of technology, EBM, and hospital medicine. One of the pillars for the continued success of the hospital medicine movement will be EBM. EBM must become an integral part of the skill set for all hospitalists.

EBM is an analytical approach with a fundamental knowledge base and a set of tools. The exponential growth of clinical information requires that physicians use an analytical approach for answering clinical questions and keeping up-to-date. This may be easier if you work at an academic center rather than a non-teaching hospital, although this is not guaranteed. Regardless of the working environment, an analytical approach will be needed if we are to build on initial success and unrealized potential to improve quality and patient safety.

The term “EBM” was introduced by a group of clinician researchers and educators at McMaster University during the early 1990s. It was initially defined as “a systemic approach to analyzed published research as the basis of clinical decision making.” Subsequently, as the EBM movement matured as a discipline, the early proponents and developers provided a more complete definition: “Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research (1).”

click for large version

Of course, the concept of practicing medicine based on the scientific method has been around for years. The days of bloodletting with leeches are behind us, but rigorous scientific evaluation of medicine reached critical mass only in the last century. The first double-blind randomized controlled trial (RCT) was conducted in 1931; the study tested the use of sanocrysin for treatment of tuberculosis (2). Since then, there has been an exponential growth in clinical trials. This information explosion requires new approaches to integrating the ever-increasing knowledge with patient care; the concurrent revolution in information technology provides opportunities limited only by our own imaginations.

EBM is not without its critics: “It is cookbook medicine,” “It focuses on cost efficiency,” “I can’t find an RCT that fits my patient,” and “It doesn’t take into account the clinician’s experience” are often argued points. If EBM is not used effectively, all the criticisms are appropriate. Sacket, one of the earliest proponents and likely EBM’s most eloquent champion, refuted the claim that EBM is “cookbook medicine.” He argued that EBM requires a bottom-up approach that integrates the best external evidence with individual clinical expertise and patients’ choice. External clinical evidence informs, but does not replace, individual clinical expertise. The physician must decide whether the external evidence applies to the individual patient at all and, if so, how it should be integrated into a clinical decision (1).

Neither is EBM strictly about RCTs and meta-analysis. It’s about tracking down the best available evidence for your question and, thus, your patient. Sometimes a cohort study is best when you want to find the prognosis of a certain illness. Similarly, a cross-sectional study may be most appropriate when you’re trying to determine the sensitivity and specificity of a test. If a disease once thought universally fatal is proven otherwise in a case report, then a randomized control trial is hardly necessary. Finally, an RCT or meta-analysis is not always going to be available for the disease process you are dealing with, but EBM gives us the skill set to look down the evidence pyramid and find the next best thing (Figure 1).

It is also clear that EBM is not about cost cutting, although many hospital medicine programs were started with this as the primary goal, given the current healthcare environment. Certainly fears exist that EBM is being used by healthcare managers, organizations, and administrators as a cost-efficiency tool. It may be that good evidence is cost-efficient in certain situations, while in others it may require the healthcare system to invest more in itself if available evidence supports doing so. Thus it is imperative that hospitalists accept the challenge of incorporating EBM into their daily practice and become leaders in its application, with patient safety and quality of care as primary goals. If we don’t, others will define the role of EBM for us, with a potential for poor outcomes for the patient and the profession.

The EBM skill set and its tools are being continuously refined, with the evidence pyramid as one of the most basic principles (3). This evidence pyramid is a model for grading the evidence. It puts in perspective the different grades of evidence or study designs. For example, a systematic review of randomized controlled trials that show consistent results provides the highest quality evidence and is ranked accordingly in the pyramid. In contrast, a case report or case series of a treatment would be ranked much lower.

The first step in incorporating EBM into one’s daily practice requires an understanding of its analytical approach and access to the necessary tools. The process begins with asking a question that is answerable. A well-built clinical question is one that benefits the patient and clinician. Such questions are directly relevant to patient problems and phrased in ways that direct your search to relevant and precise answers.

click for large version

In forming the question the following process, referred to as the PICO method, is helpful (4).

With the question formed, consider what type of question you have. This is often referred to as the typology of the question.

1. Clinical Findings: Gathering and interpreting findings from the history, clinical examination, and test results.
2. Etiology: Identifying causes for disease.
3. Differential Diagnosis: Ranking by likelihood, seriousness, and treatability of the patients problem.
4. Prognosis: Figuring out how to estimate the likely clinical course and complications over time of the disease
5. Therapy: Selecting treatments to offer that do more good than harm and that are worth the effort and cost of using them.
6. Prevention: Reducing the chance of disease by identifying and modifying risk factors and how to diagnose disease early by screening.
7. Self-improvement: Keeping up-to-date, improve your clinical skills, and run a better, more efficient clinical practice.

The types of questions can next be matched to the type of research that may provide the answer:

• Diagnosis: prospective cohort study with good quality validation against “gold standard.”
• Prognosis: prospective cohort study.
• Therapy or prevention: prospective, randomized controlled clinical trial (RCT).
• Harm/Etiology: RCT, cohort or case-control study (probably retrospective).
• Economic: analysis of sensible costs against evidence-based outcome.

Once the question has been formed, the following steps lie ahead: finding the evidence, critically appraising the evidence, acting on the evidence, and, finally, evaluating one’s performance. Very much like the formation of the question, each of the subsequent steps involves an analytical approach that can be mastered. Technology—particularly personal computers, the Internet and PDAs—has made the task of mastering EBM easier in many ways. The additional steps in using EBM effectively will be addressed in future articles. A list of useful links is provided below.

http://www.cebm.utoronto.ca/

http://library.downstate.edu/EBM2/contents.htm

http://healthsystem.virginia.edu/internet/library/collections/ebm/index.cfm

http://www.ebmny.org

http://www.cebm.net/

Dr. Kathuria may be reached at Navneet.kathuria@mssm.edu.

Endnotes

1. Guyatt GH, Haynes RB, Jaeschke RZ, et al. Users’ guides to the medical literature: XXV. Evidence-based medicine: principles for applying the users’ guides to patient care. Evidence-Based Medicine Working Group. JAMA. 2000;284:1290-6.
2. Claridge, J, Fabian, T. History and Development of Evidence Based Medicine. World Journal of Surgery 2005
3. Guyatt GH, Haynes RB, et. al. Users’ guides to the medical literature: XXV. Evidence-Based Medicine: Principles for Applying the Users’ Guides to Patient Care. 2000;284:1290-1296
4. Sackett DL, Richardson WS, Rosenberg W, Haynes RB (1997). Evidence-based medicine: How to practice and teach EBM. New York: Churchill Livingston

The terms “hospital medicine” and “evidence-based medicine” (EBM) are both recent arrivals in the history of medicine. Both have spread through medicine at a rapid pace, highlighting the attraction and fundamental soundness of their core ideas. Much has been written about the benefits of the hospitalist movement regarding quality, patient throughput and financial indicators. The next phase in the revolution of patient care is the confluence of technology, EBM, and hospital medicine. One of the pillars for the continued success of the hospital medicine movement will be EBM. EBM must become an integral part of the skill set for all hospitalists.

EBM is an analytical approach with a fundamental knowledge base and a set of tools. The exponential growth of clinical information requires that physicians use an analytical approach for answering clinical questions and keeping up-to-date. This may be easier if you work at an academic center rather than a non-teaching hospital, although this is not guaranteed. Regardless of the working environment, an analytical approach will be needed if we are to build on initial success and unrealized potential to improve quality and patient safety.

The term “EBM” was introduced by a group of clinician researchers and educators at McMaster University during the early 1990s. It was initially defined as “a systemic approach to analyzed published research as the basis of clinical decision making.” Subsequently, as the EBM movement matured as a discipline, the early proponents and developers provided a more complete definition: “Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. The practice of evidence-based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research (1).”

click for large version

Of course, the concept of practicing medicine based on the scientific method has been around for years. The days of bloodletting with leeches are behind us, but rigorous scientific evaluation of medicine reached critical mass only in the last century. The first double-blind randomized controlled trial (RCT) was conducted in 1931; the study tested the use of sanocrysin for treatment of tuberculosis (2). Since then, there has been an exponential growth in clinical trials. This information explosion requires new approaches to integrating the ever-increasing knowledge with patient care; the concurrent revolution in information technology provides opportunities limited only by our own imaginations.

EBM is not without its critics: “It is cookbook medicine,” “It focuses on cost efficiency,” “I can’t find an RCT that fits my patient,” and “It doesn’t take into account the clinician’s experience” are often argued points. If EBM is not used effectively, all the criticisms are appropriate. Sacket, one of the earliest proponents and likely EBM’s most eloquent champion, refuted the claim that EBM is “cookbook medicine.” He argued that EBM requires a bottom-up approach that integrates the best external evidence with individual clinical expertise and patients’ choice. External clinical evidence informs, but does not replace, individual clinical expertise. The physician must decide whether the external evidence applies to the individual patient at all and, if so, how it should be integrated into a clinical decision (1).

Neither is EBM strictly about RCTs and meta-analysis. It’s about tracking down the best available evidence for your question and, thus, your patient. Sometimes a cohort study is best when you want to find the prognosis of a certain illness. Similarly, a cross-sectional study may be most appropriate when you’re trying to determine the sensitivity and specificity of a test. If a disease once thought universally fatal is proven otherwise in a case report, then a randomized control trial is hardly necessary. Finally, an RCT or meta-analysis is not always going to be available for the disease process you are dealing with, but EBM gives us the skill set to look down the evidence pyramid and find the next best thing (Figure 1).

It is also clear that EBM is not about cost cutting, although many hospital medicine programs were started with this as the primary goal, given the current healthcare environment. Certainly fears exist that EBM is being used by healthcare managers, organizations, and administrators as a cost-efficiency tool. It may be that good evidence is cost-efficient in certain situations, while in others it may require the healthcare system to invest more in itself if available evidence supports doing so. Thus it is imperative that hospitalists accept the challenge of incorporating EBM into their daily practice and become leaders in its application, with patient safety and quality of care as primary goals. If we don’t, others will define the role of EBM for us, with a potential for poor outcomes for the patient and the profession.

The EBM skill set and its tools are being continuously refined, with the evidence pyramid as one of the most basic principles (3). This evidence pyramid is a model for grading the evidence. It puts in perspective the different grades of evidence or study designs. For example, a systematic review of randomized controlled trials that show consistent results provides the highest quality evidence and is ranked accordingly in the pyramid. In contrast, a case report or case series of a treatment would be ranked much lower.

The first step in incorporating EBM into one’s daily practice requires an understanding of its analytical approach and access to the necessary tools. The process begins with asking a question that is answerable. A well-built clinical question is one that benefits the patient and clinician. Such questions are directly relevant to patient problems and phrased in ways that direct your search to relevant and precise answers.

click for large version

In forming the question the following process, referred to as the PICO method, is helpful (4).

With the question formed, consider what type of question you have. This is often referred to as the typology of the question.

1. Clinical Findings: Gathering and interpreting findings from the history, clinical examination, and test results.
2. Etiology: Identifying causes for disease.
3. Differential Diagnosis: Ranking by likelihood, seriousness, and treatability of the patients problem.
4. Prognosis: Figuring out how to estimate the likely clinical course and complications over time of the disease
5. Therapy: Selecting treatments to offer that do more good than harm and that are worth the effort and cost of using them.
6. Prevention: Reducing the chance of disease by identifying and modifying risk factors and how to diagnose disease early by screening.
7. Self-improvement: Keeping up-to-date, improve your clinical skills, and run a better, more efficient clinical practice.

The types of questions can next be matched to the type of research that may provide the answer:

• Diagnosis: prospective cohort study with good quality validation against “gold standard.”
• Prognosis: prospective cohort study.
• Therapy or prevention: prospective, randomized controlled clinical trial (RCT).
• Harm/Etiology: RCT, cohort or case-control study (probably retrospective).
• Economic: analysis of sensible costs against evidence-based outcome.

Once the question has been formed, the following steps lie ahead: finding the evidence, critically appraising the evidence, acting on the evidence, and, finally, evaluating one’s performance. Very much like the formation of the question, each of the subsequent steps involves an analytical approach that can be mastered. Technology—particularly personal computers, the Internet and PDAs—has made the task of mastering EBM easier in many ways. The additional steps in using EBM effectively will be addressed in future articles. A list of useful links is provided below.

http://www.cebm.utoronto.ca/

http://library.downstate.edu/EBM2/contents.htm

http://healthsystem.virginia.edu/internet/library/collections/ebm/index.cfm

http://www.ebmny.org

http://www.cebm.net/

Dr. Kathuria may be reached at Navneet.kathuria@mssm.edu.

Endnotes

1. Guyatt GH, Haynes RB, Jaeschke RZ, et al. Users’ guides to the medical literature: XXV. Evidence-based medicine: principles for applying the users’ guides to patient care. Evidence-Based Medicine Working Group. JAMA. 2000;284:1290-6.
2. Claridge, J, Fabian, T. History and Development of Evidence Based Medicine. World Journal of Surgery 2005
3. Guyatt GH, Haynes RB, et. al. Users’ guides to the medical literature: XXV. Evidence-Based Medicine: Principles for Applying the Users’ Guides to Patient Care. 2000;284:1290-1296
4. Sackett DL, Richardson WS, Rosenberg W, Haynes RB (1997). Evidence-based medicine: How to practice and teach EBM. New York: Churchill Livingston

Literature at a Glance

• Drug-eluting stents decrease the need for revascularization.
• Case volume is related to hospital performance assessment.
• Prolonged QRS duration in patients with CHF is associated with increased morbidity and mortality.
• For out-of-hospital ACLS, vasopressin plus epinephrine is not better than vasopressin alone.
• Oral rivaroxaban is more efficacious than enoxaparin for VTE prophylaxis after total hip replacement.
• LMWH and UFH offer similar perioperative VTE prophylaxis benefit in patients with cancer.
• Salmeterol added to inhaled corticosteroids decreases severe asthma exacerbations.
• Early invasive strategy has unclear benefit in low-risk women with unstable angina or NSTEMI.
• Strategies to prevent contrast-induced acute kidney injury are not uniform.
• Hyperglycemia in hospitalized children is common and associated with ICU admission.

Do drug-eluting stents improve outcomes after ST-elevation myocardial infarction (STEMI)?

Background: Drug-eluting stents reduce restenosis rates compared to bare-metal stents. However, there is concern drug-eluting stents increase the risk of stent thrombosis leading to MI and death. Prior studies compared patients who received bare-metal versus those who received drug-eluting stents. Outcomes on a population level might provide new insight.

Study design: Observational study.

Setting: 100% national sample of patients 65 and older who received a coronary stent from 2002-05 enrolled in the traditional fee-for-service Medicare program.

Synopsis: 38,917 patients in the pre-drug-eluting-stent era from October 2002 to March 2003 received bare-metal stents. Nearly 62% of 28,086 patients studied from September to December 2003 received drug-eluting stents. The remaining 38.5% received bare-metal stents. Outcomes of percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), STEMI, and death were observed through December 31, 2005.

Patients in the drug-eluting-stent era had a lower two-year risk for repeat revascularization compared to patients in the bare-metal-stent era. In the drug-eluting versus bare-metal eras, repeat PCI was 17.1% versus 20.0% (p<0.001) and need for CABG was 2.7% versus 4.2% (p<0.01). Comparing adjusted outcomes for death, or STEMI, at two years, the two groups appeared similar.

The study did have limitations: the data only reflect sirolimus stents, the authors could not assess dual-antiplatelet therapy or obtain information on coronary anatomy or procedure details to account for selection bias in stent utilization, and the patients were all Medicare beneficiaries.

Bottom line: Drug-eluting stents are associated with fewer repeat revascularization procedures than bare-metal stents, but have not shown a significant improvement in the subsequent risk of STEMI or death.

Citation: Malenka DJ, Kaplan AV, Lucas FL, Sharp SM, Skinner JA. Outcomes following coronary stenting in the era of bare-metal vs. the era of drug-eluting stents. JAMA 2008;299(24):2868-2877.

Does case volume affect hospital performance for publicly reported process measures?

Background: Hospitals are increasingly graded and compared to one another. “Top medical centers” are defined as those within the top 10% of hospitals in specified performance measures. Hospitals with large and small case volumes might not be compared evenly and fairly.

Study design: Eight publicly reported process measures for acute myocardial infarction (AMI) were compared to hospital case volume, process performance, and label as “top hospital.”

Setting: Data were analyzed from the Hospital Quality Alliance for 3,761 hospitals from January to December 2005.

Synopsis: Hospitals with large case volume overall had better process performance. For example, looking at use of beta-blockers in patients with AMI on arrival to a hospital, small-volume hospitals (<10 AMI cases) averaged 72% while large volume (>100 AMI cases) averaged 80% (p<0.001). However, hospitals with small case volumes were more likely to receive “top hospital” rating even when hospitals with very low case volumes were excluded.

Hospital quality reporting that does not account for case volume is misleading to hospitals and consumers. In this study, larger-volume hospitals appeared to perform better in process measures, but were less likely to receive “top hospital” rating.

Bottom line: Hospitals with large and small case volumes can easily be compared to one another for process measures in AMI.

Citation: O’Brien SM, DeLong ER, Peterson ED. Impact of case volume on hospital performance assessment. Arch Intern Med. 2008;168(12):1277-1284.

What is the predictive value of QRS duration in patients hospitalized with worsening CHF?

Background: In outpatients, a prolonged QRS duration (greater than 120 ms) is associated with increased mortality. Its value in the inpatient setting is unclear. For patients hospitalized with CHF exacerbations, establishing the value of QRS duration may allow for tailored management.

Study design: Retrospective post hoc analysis from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST).

Setting: 4,133 patients were enrolled from North American, South American, and European sites.

Synopsis: Of 2,962 patients included in the final post hoc analysis, 1,321 (44.6%) had a prolonged QRS duration. During a median follow up of 9.9 months, the all-cause mortality rate was 18.7% for patients with a normal baseline QRS duration and 28.1% for patients with a prolonged baseline QRS.

After adjusting for confounding variables, patients with a prolonged baseline QRS had a 24% increased risk of all-cause mortality and a 28% increased risk for a composite endpoint of cardiac mortality or hospitalization for heart failure exacerbation.

The retrospective nature of the analysis represents the major limitation of this study. In addition, most of the enrolled patients were white, which limits the studies generalizability to other ethnic groups.

Bottom Line: A prolonged QRS duration for patients admitted with decompensated left ventricular heart failure is common and may be associated with increased morbidity and mortality.

Citation: Wang NC, Maggioni AP, Konstam MA, Zannad F, Drasa HB, Burnett JC, et al. Clinical implications of QRS duration in patients hospitalized with worsening heart failure and reduced left ventricular ejection fraction. JAMA. 2008;299(22):2656-2666.

For patients with out-of-hospital cardiac arrest, does the addition of vasopressin to epinephrine in a protocol for ACLS improve outcomes?

Background: The outcome for patients experiencing cardiac arrest who require vasopressors remains extremely poor. Despite disappointing data on vasopressin as an alternative treatment during cardiac arrest, a recent subgroup analysis suggested patients who received epinephrine and vasopressin together had superior clinical outcomes.

Study Design: Prospective multicenter randomized double-blind controlled trial.

Setting: 31 emergency medical service organizations in France.

Synopsis: Of the 2,894 patients, 20.7% of those receiving combination treatment (vasopressin plus epinephrine) survived to hospital admission versus 21.3% of those in the epinephrine-only group. For those same groups, 1.7% of combination and 2.3% of epinephrine-only patients survived to hospital discharge. No significant outcome differences were found in any group or subgroup analysis.

The study had lower-than-expected overall survival to hospital discharge, which may have handicapped its effort to find a true difference in treatment arms.

Bottom line: The addition of vasopressin to epinephrine in the treatment of out-of-hospital cardiac arrest does not improve outcomes.

Citation: Gueugniaud PY, David JS, Chanzy E, Hubert H, Dubien P, Mauriaucourt P, et al. Vasopressin and epinephrine vs. epinephrine alone in cardiopulmonary resuscitation. N Engl J Med. 2008;359(1):21-30.

Is oral rivaroxaban more efficacious than subcutaneous enoxaparin in preventing VTE after hip-replacement surgery?

Background: Venous thromboembolism (VTE) prophylaxis after total hip replacement (THR) is important but can be cumbersome because the most commonly used anticoagulants are either subcutaneous or require frequent monitoring. Rivaroxaban, an oral direct inhibitor of factor Xa may provide more convenient anticoagulation postoperatively. However, its efficacy and safety are unknown.

Study design: Randomized double-blind trial.

Setting: Multicenter study performed in 27 countries.

Synopsis: Patients undergoing THR surgery were randomized to oral rivaroxaban (10mg once daily without monitoring, started six to eight hours after surgery) or subcutaneous enoxaparin (40mg once daily, started 12 hours prior to surgery). After surgery, prophylaxis was administered for 35 days. The primary outcome was a composite of asymptomatic deep venous thrombosis (DVT), symptomatic DVT or pulmonary embolism (PE), or death from any cause at 36 days after surgery.

In the enoxaparin group, 3.7% of patients experienced the primary outcome. This decreased to 1.1% in the rivaroxaban group. Approximately one-third of events were symptomatic. Major bleeding occurred in 0.1% and 0.3% (p=NS) of patients in the enoxaparin and rivaroxaban groups, respectively.

The study is limited by the exclusion of 1,388 of the 4,541 patients (30.6%) randomized, primarily due to having inadequate venography. Also, because the majority of thromboembolic events were asymptomatic, the primary outcome overemphasizes the clinical difference.

Bottom line: Oral rivaroxaban without monitoring is more efficacious than, and as safe as, subcutaneous enoxaparin when used for VTE prophylaxis for THR.

Citation: Eriksson B, Borris LC, Friedman RJ, Hass S, Huisman MV, Kakkar AK, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med. 2008;358:2765-75.

Is LMWH more efficacious than UFH in preventing postoperative VTE in cancer patients?

Background: Patients with cancer are at increased risk for VTE and require prophylaxis to prevent this complication postoperatively. Low molecular weight heparin (LMWH) has proven more efficacious than subcutaneous unfractionated heparin (UFH) in other settings (e.g., DVT treatment). However, it is still unknown whether LMWH offers better prophylaxis compared to UFH for cancer patients undergoing surgery.

Study design: Systematic review and meta-analysis.

Setting: 14 randomized controlled trials.

Synopsis: Eleven trials exclusively examined patients with cancer (n=4006) and three trials reported data for cancer patients as subgroups (n=1816). There were in differences in mortality, pulmonary embolism, and symptomatic DVT rates between the two groups.

LMWH was associated with a decrease in total (asymptomatic or symptomatic) DVT (RR, 0.72; 95% CI, 0.55-0.94). Rates of major bleeding, minor bleeding, and intraoperative blood loss were similar between the two treatments.

This meta-analysis is limited because 12 remaining trials (n=3185) also enrolled cancer patients but did not provide specific data for the cancer patient subgroup. The study also is limited by the heterogeneity of the original trials, including utilizing varying LMWHs and dosing regimens, numerous types of surgeries, and a wide range of neoplasms.

Bottom line: LMWH does not decrease mortality, pulmonary embolism, or symptomatic DVT compared to UFH in cancer patients undergoing surgery.

Citation: Akl EA, Terrenato I, Barba M, Sperati F, Sempos EV, Muti P, et al. Low-molecular-weight heparin vs. unfractionated heparin for perioperative thromboprophylaxis in patients with cancer. Arch Intern Med. 2008;168:1261-9.

Does salmeterol added to inhaled corticosteroids improve severe asthma-related events?

Background: Asthma is a chronic disease causing major morbidity and mortality worldwide. Disease guidelines recommend all patients with persistent asthma be treated with inhaled corticosteroids. These same guidelines recommend adding a long-acting beta-agonist for patients whose symptoms persist. However, the safety of this practice has come under scrutiny.

Study design: Meta-analysis.

Setting: Sixty-six randomized, controlled trials conducted worldwide.

Synopsis: Analysis included 66 GlaxoSmithKline trials with a total of 20,966 patients with persistent asthma. Patients used either salmeterol (50mcg twice daily) plus inhaled corticosteroid (10,400 patients) or inhaled corticosteroid alone (10,566 patients).

Results showed no differences in asthma-related hospitalizations, asthma-related intubations, or deaths between the two groups. However, due to the low number of events, definitive conclusions are difficult to make. Severe asthma exacerbations requiring systemic corticosteroids significantly decreased in the inhaled corticosteroid plus salmeterol group.

The study is limited by it inclusion of only those trials sponsored by GlaxoSmithKline and by the short duration of most of the studies. Additionally, the studies included in the analysis used clinical outcomes as secondary endpoints.

Bottom line: Adding salmeterol to inhaled corticosteroid decreases severe asthma exacerbations and is likely safe, but does not have an effect on asthma-related hospitalization or death.

Citation: Bateman E, Nelson H, Bousquet J, Kral K, Sutton L, Ortega H, et.al. Meta-analysis: Effects of adding salmeterol to inhaled corticosteroids on serious asthma-related events. Annals Intern Med. 2008;149:33-42.

Is an early invasive strategy effective in women with unstable angina or NSTEMI?

Background: Despite many trials showing the value of an early invasive strategy for patients with non-ST-segment elevation acute coronary syndrome (NSTE ACS), data from several trials question this benefit in women. Some trials show higher risk of death and myocardial infarction (MI) in subgroup analysis of women.

Study Design: Meta-analysis.

Setting: Eight randomized, controlled trials conducted worldwide.

Synopsis: Analysis included eight trials with 10,412 patients (3,075 women) with NSTE ACS. The invasive group (5,083 patients) was defined as those referred for coronary angiography with subsequent intervention as needed. The composite endpoint of death, MI, or rehospitalization within 12 months with ACS occurred in 21.1% of the invasive group and 25.9% of the medically managed group (OR, 0.78; CI, 0.61-0.98).

The subgroup, including only women, had a non-statistically significant OR of 0.81 (CI, 0.65-1.01), including no effect on all-cause mortality, nonfatal MI, or the composite of death and MI. However, women with high-risk features (elevated biomarkers) undergoing the invasive strategy had a significant reduction in the composite endpoint (OR, 0.67; CI, 0.50-0.88).

The study is limited by the use of subgroup analysis, secondary endpoints, heterogeneity between trials, and possible publication bias.

Bottom line: Early invasive strategy is effective in men and high-risk women with NSTE ACS, but not in low-risk women.

Citation: O’Donoghue M, Boden W, Braunwald E, Cannon CP, Clayton TC, Winter RJ, et.al. Early invasive vs. conservative treatment strategies in women and men with unstable angina and non-ST-segment elevation myocardial infarction. JAMA. 2008;300:71-80.

What strategies are used to prevent contrast-induced acute kidney injury?

Background: Contrast-induced acute kidney injury (CIAKI) is a condition potentially amenable to preventive care. Several trials have identified intravenous hydration, N-acetylcysteine, and withdrawal of NSAIDS as interventions that reduce the possibility of CIAKI in high-risk patients. Little is known about whether healthcare providers routinely use these strategies.

Study design: Prospective observational cohort study.

Setting: Veterans Affairs (VA) Pittsburgh Healthcare System.

Synopsis: 11,410 patients scheduled for radiographic procedures were screened. After exclusion criteria and eligibility, 660 patients with an estimated glomerular filtration rate less than 60ml/min/1.73m2 were identified. Usage of intravenous fluids, N-acetylcysteine, and discontinuation of NSAIDS were recorded. Serum creatinine (SCr) was measured 48 to 96 hours post-procedure. CIAKI was defined as relative increase in SCr from baseline (≥25%, ≥50% and ≥100%) and absolute increase in SCr levels from baseline (≥0.25, ≥0.5, and ≥1.0). CIAKI association with adverse outcomes was evaluated by tracking 30-day mortality, need for dialysis, and hospitalization.

The incidence of CIAKI was less common in patients undergoing CT scans versus those having angiograms. Adverse 30-day outcomes were uncommon. Pre- and post-procedure intravenous hydration was administered to 40% of study patients, more commonly with coronary angiogram than with computed tomography (91.2% vs. 16%, p<0.0001). N-acetylcysteine was administered to 39.2%. Only 6.8% of those taking NSAIDS reported being told to discontinue the medication.

Study limitations include the small sample size and the single site location, both limiting generalizability.

Bottom line: Clinically significant CIAKI is uncommon, and preventive care is not uniformly implemented in patients undergoing contrast-enhanced radiographic procedures.

Citation: Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Sonel AF, Fine MJ, et al. Prevention, incidence, and outcomes of contrast-induced acute kidney injury. Arch Intern Med. 2008;168(12):1325-1332.

How does hyperglycemia affect morbidity and mortality in children admitted to a community pediatric hospital?

Background: Inpatient hyperglycemia in adult patients is a predictor of poor clinical outcomes. The association of hyperglycemia and clinical outcomes in children admitted to a general community hospital has not been studied.

Study design: Retrospective observational cohort study.

Setting: A community pediatric hospital in Atlanta, Ga.

Synopsis: Review of medical records of 903 consecutive pediatric patients admitted to critical and non-critical areas took place. Of these, 542 patients constituted the study population. The study excluded 342 patients who didn’t have a blood glucose measurement. Hyperglycemia was defined as an admission or in-hospital blood glucose greater than 120mg/dl.

One-fourth of the children admitted to the hospital had hyperglycemia, most without a prior history of diabetes. The presence of hyperglycemia on admission was not associated with increased length of stay (LOS) or increased mortality. Children with hyperglycemia were more likely to be admitted to the ICU and had longer ICU LOS.

This was a retrospective study conducted at a single site whose demographics and disease spectrum may differ from those of other institutions. There were an insufficient number of deaths to make any conclusions regarding the impact of hyperglycemia on mortality. Prospective, randomized, multicenter trials are needed to better elucidate the effects of in-patient hyperglycemia.

Bottom line: Hyperglycemia is common in children with or without diabetes admitted to the hospital, and is associated with increased ICU admissions and ICU length of stay. Its connection to mortality is inconclusive.

Citation: Palaio A, Smiley D, Ceron M, Klein R, Cho IS, Mejia R, et al. Prevalence and clinical outcome of inpatient hyperglycemia in a community pediatric hospital. J Hosp Med.2008;3(3):212-217.

Literature at a Glance

• Drug-eluting stents decrease the need for revascularization.
• Case volume is related to hospital performance assessment.
• Prolonged QRS duration in patients with CHF is associated with increased morbidity and mortality.
• For out-of-hospital ACLS, vasopressin plus epinephrine is not better than vasopressin alone.
• Oral rivaroxaban is more efficacious than enoxaparin for VTE prophylaxis after total hip replacement.
• LMWH and UFH offer similar perioperative VTE prophylaxis benefit in patients with cancer.
• Salmeterol added to inhaled corticosteroids decreases severe asthma exacerbations.
• Early invasive strategy has unclear benefit in low-risk women with unstable angina or NSTEMI.
• Strategies to prevent contrast-induced acute kidney injury are not uniform.
• Hyperglycemia in hospitalized children is common and associated with ICU admission.

Do drug-eluting stents improve outcomes after ST-elevation myocardial infarction (STEMI)?

Background: Drug-eluting stents reduce restenosis rates compared to bare-metal stents. However, there is concern drug-eluting stents increase the risk of stent thrombosis leading to MI and death. Prior studies compared patients who received bare-metal versus those who received drug-eluting stents. Outcomes on a population level might provide new insight.

Study design: Observational study.

Setting: 100% national sample of patients 65 and older who received a coronary stent from 2002-05 enrolled in the traditional fee-for-service Medicare program.

Synopsis: 38,917 patients in the pre-drug-eluting-stent era from October 2002 to March 2003 received bare-metal stents. Nearly 62% of 28,086 patients studied from September to December 2003 received drug-eluting stents. The remaining 38.5% received bare-metal stents. Outcomes of percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), STEMI, and death were observed through December 31, 2005.

Patients in the drug-eluting-stent era had a lower two-year risk for repeat revascularization compared to patients in the bare-metal-stent era. In the drug-eluting versus bare-metal eras, repeat PCI was 17.1% versus 20.0% (p<0.001) and need for CABG was 2.7% versus 4.2% (p<0.01). Comparing adjusted outcomes for death, or STEMI, at two years, the two groups appeared similar.

The study did have limitations: the data only reflect sirolimus stents, the authors could not assess dual-antiplatelet therapy or obtain information on coronary anatomy or procedure details to account for selection bias in stent utilization, and the patients were all Medicare beneficiaries.

Bottom line: Drug-eluting stents are associated with fewer repeat revascularization procedures than bare-metal stents, but have not shown a significant improvement in the subsequent risk of STEMI or death.

Citation: Malenka DJ, Kaplan AV, Lucas FL, Sharp SM, Skinner JA. Outcomes following coronary stenting in the era of bare-metal vs. the era of drug-eluting stents. JAMA 2008;299(24):2868-2877.

Does case volume affect hospital performance for publicly reported process measures?

Background: Hospitals are increasingly graded and compared to one another. “Top medical centers” are defined as those within the top 10% of hospitals in specified performance measures. Hospitals with large and small case volumes might not be compared evenly and fairly.

Study design: Eight publicly reported process measures for acute myocardial infarction (AMI) were compared to hospital case volume, process performance, and label as “top hospital.”

Setting: Data were analyzed from the Hospital Quality Alliance for 3,761 hospitals from January to December 2005.

Synopsis: Hospitals with large case volume overall had better process performance. For example, looking at use of beta-blockers in patients with AMI on arrival to a hospital, small-volume hospitals (<10 AMI cases) averaged 72% while large volume (>100 AMI cases) averaged 80% (p<0.001). However, hospitals with small case volumes were more likely to receive “top hospital” rating even when hospitals with very low case volumes were excluded.

Hospital quality reporting that does not account for case volume is misleading to hospitals and consumers. In this study, larger-volume hospitals appeared to perform better in process measures, but were less likely to receive “top hospital” rating.

Bottom line: Hospitals with large and small case volumes can easily be compared to one another for process measures in AMI.

Citation: O’Brien SM, DeLong ER, Peterson ED. Impact of case volume on hospital performance assessment. Arch Intern Med. 2008;168(12):1277-1284.

What is the predictive value of QRS duration in patients hospitalized with worsening CHF?

Background: In outpatients, a prolonged QRS duration (greater than 120 ms) is associated with increased mortality. Its value in the inpatient setting is unclear. For patients hospitalized with CHF exacerbations, establishing the value of QRS duration may allow for tailored management.

Study design: Retrospective post hoc analysis from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST).

Setting: 4,133 patients were enrolled from North American, South American, and European sites.

Synopsis: Of 2,962 patients included in the final post hoc analysis, 1,321 (44.6%) had a prolonged QRS duration. During a median follow up of 9.9 months, the all-cause mortality rate was 18.7% for patients with a normal baseline QRS duration and 28.1% for patients with a prolonged baseline QRS.

After adjusting for confounding variables, patients with a prolonged baseline QRS had a 24% increased risk of all-cause mortality and a 28% increased risk for a composite endpoint of cardiac mortality or hospitalization for heart failure exacerbation.

The retrospective nature of the analysis represents the major limitation of this study. In addition, most of the enrolled patients were white, which limits the studies generalizability to other ethnic groups.

Bottom Line: A prolonged QRS duration for patients admitted with decompensated left ventricular heart failure is common and may be associated with increased morbidity and mortality.

Citation: Wang NC, Maggioni AP, Konstam MA, Zannad F, Drasa HB, Burnett JC, et al. Clinical implications of QRS duration in patients hospitalized with worsening heart failure and reduced left ventricular ejection fraction. JAMA. 2008;299(22):2656-2666.

For patients with out-of-hospital cardiac arrest, does the addition of vasopressin to epinephrine in a protocol for ACLS improve outcomes?

Background: The outcome for patients experiencing cardiac arrest who require vasopressors remains extremely poor. Despite disappointing data on vasopressin as an alternative treatment during cardiac arrest, a recent subgroup analysis suggested patients who received epinephrine and vasopressin together had superior clinical outcomes.

Study Design: Prospective multicenter randomized double-blind controlled trial.

Setting: 31 emergency medical service organizations in France.

Synopsis: Of the 2,894 patients, 20.7% of those receiving combination treatment (vasopressin plus epinephrine) survived to hospital admission versus 21.3% of those in the epinephrine-only group. For those same groups, 1.7% of combination and 2.3% of epinephrine-only patients survived to hospital discharge. No significant outcome differences were found in any group or subgroup analysis.

The study had lower-than-expected overall survival to hospital discharge, which may have handicapped its effort to find a true difference in treatment arms.

Bottom line: The addition of vasopressin to epinephrine in the treatment of out-of-hospital cardiac arrest does not improve outcomes.

Citation: Gueugniaud PY, David JS, Chanzy E, Hubert H, Dubien P, Mauriaucourt P, et al. Vasopressin and epinephrine vs. epinephrine alone in cardiopulmonary resuscitation. N Engl J Med. 2008;359(1):21-30.

Is oral rivaroxaban more efficacious than subcutaneous enoxaparin in preventing VTE after hip-replacement surgery?

Background: Venous thromboembolism (VTE) prophylaxis after total hip replacement (THR) is important but can be cumbersome because the most commonly used anticoagulants are either subcutaneous or require frequent monitoring. Rivaroxaban, an oral direct inhibitor of factor Xa may provide more convenient anticoagulation postoperatively. However, its efficacy and safety are unknown.

Study design: Randomized double-blind trial.

Setting: Multicenter study performed in 27 countries.

Synopsis: Patients undergoing THR surgery were randomized to oral rivaroxaban (10mg once daily without monitoring, started six to eight hours after surgery) or subcutaneous enoxaparin (40mg once daily, started 12 hours prior to surgery). After surgery, prophylaxis was administered for 35 days. The primary outcome was a composite of asymptomatic deep venous thrombosis (DVT), symptomatic DVT or pulmonary embolism (PE), or death from any cause at 36 days after surgery.

In the enoxaparin group, 3.7% of patients experienced the primary outcome. This decreased to 1.1% in the rivaroxaban group. Approximately one-third of events were symptomatic. Major bleeding occurred in 0.1% and 0.3% (p=NS) of patients in the enoxaparin and rivaroxaban groups, respectively.

The study is limited by the exclusion of 1,388 of the 4,541 patients (30.6%) randomized, primarily due to having inadequate venography. Also, because the majority of thromboembolic events were asymptomatic, the primary outcome overemphasizes the clinical difference.

Bottom line: Oral rivaroxaban without monitoring is more efficacious than, and as safe as, subcutaneous enoxaparin when used for VTE prophylaxis for THR.

Citation: Eriksson B, Borris LC, Friedman RJ, Hass S, Huisman MV, Kakkar AK, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med. 2008;358:2765-75.

Is LMWH more efficacious than UFH in preventing postoperative VTE in cancer patients?

Background: Patients with cancer are at increased risk for VTE and require prophylaxis to prevent this complication postoperatively. Low molecular weight heparin (LMWH) has proven more efficacious than subcutaneous unfractionated heparin (UFH) in other settings (e.g., DVT treatment). However, it is still unknown whether LMWH offers better prophylaxis compared to UFH for cancer patients undergoing surgery.

Study design: Systematic review and meta-analysis.

Setting: 14 randomized controlled trials.

Synopsis: Eleven trials exclusively examined patients with cancer (n=4006) and three trials reported data for cancer patients as subgroups (n=1816). There were in differences in mortality, pulmonary embolism, and symptomatic DVT rates between the two groups.

LMWH was associated with a decrease in total (asymptomatic or symptomatic) DVT (RR, 0.72; 95% CI, 0.55-0.94). Rates of major bleeding, minor bleeding, and intraoperative blood loss were similar between the two treatments.

This meta-analysis is limited because 12 remaining trials (n=3185) also enrolled cancer patients but did not provide specific data for the cancer patient subgroup. The study also is limited by the heterogeneity of the original trials, including utilizing varying LMWHs and dosing regimens, numerous types of surgeries, and a wide range of neoplasms.

Bottom line: LMWH does not decrease mortality, pulmonary embolism, or symptomatic DVT compared to UFH in cancer patients undergoing surgery.

Citation: Akl EA, Terrenato I, Barba M, Sperati F, Sempos EV, Muti P, et al. Low-molecular-weight heparin vs. unfractionated heparin for perioperative thromboprophylaxis in patients with cancer. Arch Intern Med. 2008;168:1261-9.

Does salmeterol added to inhaled corticosteroids improve severe asthma-related events?

Background: Asthma is a chronic disease causing major morbidity and mortality worldwide. Disease guidelines recommend all patients with persistent asthma be treated with inhaled corticosteroids. These same guidelines recommend adding a long-acting beta-agonist for patients whose symptoms persist. However, the safety of this practice has come under scrutiny.

Study design: Meta-analysis.

Setting: Sixty-six randomized, controlled trials conducted worldwide.

Synopsis: Analysis included 66 GlaxoSmithKline trials with a total of 20,966 patients with persistent asthma. Patients used either salmeterol (50mcg twice daily) plus inhaled corticosteroid (10,400 patients) or inhaled corticosteroid alone (10,566 patients).

Results showed no differences in asthma-related hospitalizations, asthma-related intubations, or deaths between the two groups. However, due to the low number of events, definitive conclusions are difficult to make. Severe asthma exacerbations requiring systemic corticosteroids significantly decreased in the inhaled corticosteroid plus salmeterol group.

The study is limited by it inclusion of only those trials sponsored by GlaxoSmithKline and by the short duration of most of the studies. Additionally, the studies included in the analysis used clinical outcomes as secondary endpoints.

Bottom line: Adding salmeterol to inhaled corticosteroid decreases severe asthma exacerbations and is likely safe, but does not have an effect on asthma-related hospitalization or death.

Citation: Bateman E, Nelson H, Bousquet J, Kral K, Sutton L, Ortega H, et.al. Meta-analysis: Effects of adding salmeterol to inhaled corticosteroids on serious asthma-related events. Annals Intern Med. 2008;149:33-42.

Is an early invasive strategy effective in women with unstable angina or NSTEMI?

Background: Despite many trials showing the value of an early invasive strategy for patients with non-ST-segment elevation acute coronary syndrome (NSTE ACS), data from several trials question this benefit in women. Some trials show higher risk of death and myocardial infarction (MI) in subgroup analysis of women.

Study Design: Meta-analysis.

Setting: Eight randomized, controlled trials conducted worldwide.

Synopsis: Analysis included eight trials with 10,412 patients (3,075 women) with NSTE ACS. The invasive group (5,083 patients) was defined as those referred for coronary angiography with subsequent intervention as needed. The composite endpoint of death, MI, or rehospitalization within 12 months with ACS occurred in 21.1% of the invasive group and 25.9% of the medically managed group (OR, 0.78; CI, 0.61-0.98).

The subgroup, including only women, had a non-statistically significant OR of 0.81 (CI, 0.65-1.01), including no effect on all-cause mortality, nonfatal MI, or the composite of death and MI. However, women with high-risk features (elevated biomarkers) undergoing the invasive strategy had a significant reduction in the composite endpoint (OR, 0.67; CI, 0.50-0.88).

The study is limited by the use of subgroup analysis, secondary endpoints, heterogeneity between trials, and possible publication bias.

Bottom line: Early invasive strategy is effective in men and high-risk women with NSTE ACS, but not in low-risk women.

Citation: O’Donoghue M, Boden W, Braunwald E, Cannon CP, Clayton TC, Winter RJ, et.al. Early invasive vs. conservative treatment strategies in women and men with unstable angina and non-ST-segment elevation myocardial infarction. JAMA. 2008;300:71-80.

What strategies are used to prevent contrast-induced acute kidney injury?

Background: Contrast-induced acute kidney injury (CIAKI) is a condition potentially amenable to preventive care. Several trials have identified intravenous hydration, N-acetylcysteine, and withdrawal of NSAIDS as interventions that reduce the possibility of CIAKI in high-risk patients. Little is known about whether healthcare providers routinely use these strategies.

Study design: Prospective observational cohort study.

Setting: Veterans Affairs (VA) Pittsburgh Healthcare System.

Synopsis: 11,410 patients scheduled for radiographic procedures were screened. After exclusion criteria and eligibility, 660 patients with an estimated glomerular filtration rate less than 60ml/min/1.73m2 were identified. Usage of intravenous fluids, N-acetylcysteine, and discontinuation of NSAIDS were recorded. Serum creatinine (SCr) was measured 48 to 96 hours post-procedure. CIAKI was defined as relative increase in SCr from baseline (≥25%, ≥50% and ≥100%) and absolute increase in SCr levels from baseline (≥0.25, ≥0.5, and ≥1.0). CIAKI association with adverse outcomes was evaluated by tracking 30-day mortality, need for dialysis, and hospitalization.

The incidence of CIAKI was less common in patients undergoing CT scans versus those having angiograms. Adverse 30-day outcomes were uncommon. Pre- and post-procedure intravenous hydration was administered to 40% of study patients, more commonly with coronary angiogram than with computed tomography (91.2% vs. 16%, p<0.0001). N-acetylcysteine was administered to 39.2%. Only 6.8% of those taking NSAIDS reported being told to discontinue the medication.

Study limitations include the small sample size and the single site location, both limiting generalizability.

Bottom line: Clinically significant CIAKI is uncommon, and preventive care is not uniformly implemented in patients undergoing contrast-enhanced radiographic procedures.

Citation: Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Sonel AF, Fine MJ, et al. Prevention, incidence, and outcomes of contrast-induced acute kidney injury. Arch Intern Med. 2008;168(12):1325-1332.

How does hyperglycemia affect morbidity and mortality in children admitted to a community pediatric hospital?

Background: Inpatient hyperglycemia in adult patients is a predictor of poor clinical outcomes. The association of hyperglycemia and clinical outcomes in children admitted to a general community hospital has not been studied.

Study design: Retrospective observational cohort study.

Setting: A community pediatric hospital in Atlanta, Ga.

Synopsis: Review of medical records of 903 consecutive pediatric patients admitted to critical and non-critical areas took place. Of these, 542 patients constituted the study population. The study excluded 342 patients who didn’t have a blood glucose measurement. Hyperglycemia was defined as an admission or in-hospital blood glucose greater than 120mg/dl.

One-fourth of the children admitted to the hospital had hyperglycemia, most without a prior history of diabetes. The presence of hyperglycemia on admission was not associated with increased length of stay (LOS) or increased mortality. Children with hyperglycemia were more likely to be admitted to the ICU and had longer ICU LOS.

This was a retrospective study conducted at a single site whose demographics and disease spectrum may differ from those of other institutions. There were an insufficient number of deaths to make any conclusions regarding the impact of hyperglycemia on mortality. Prospective, randomized, multicenter trials are needed to better elucidate the effects of in-patient hyperglycemia.

Bottom line: Hyperglycemia is common in children with or without diabetes admitted to the hospital, and is associated with increased ICU admissions and ICU length of stay. Its connection to mortality is inconclusive.

Citation: Palaio A, Smiley D, Ceron M, Klein R, Cho IS, Mejia R, et al. Prevalence and clinical outcome of inpatient hyperglycemia in a community pediatric hospital. J Hosp Med.2008;3(3):212-217.

Literature at a Glance

• Drug-eluting stents decrease the need for revascularization.
• Case volume is related to hospital performance assessment.
• Prolonged QRS duration in patients with CHF is associated with increased morbidity and mortality.
• For out-of-hospital ACLS, vasopressin plus epinephrine is not better than vasopressin alone.
• Oral rivaroxaban is more efficacious than enoxaparin for VTE prophylaxis after total hip replacement.
• LMWH and UFH offer similar perioperative VTE prophylaxis benefit in patients with cancer.
• Salmeterol added to inhaled corticosteroids decreases severe asthma exacerbations.
• Early invasive strategy has unclear benefit in low-risk women with unstable angina or NSTEMI.
• Strategies to prevent contrast-induced acute kidney injury are not uniform.
• Hyperglycemia in hospitalized children is common and associated with ICU admission.

Do drug-eluting stents improve outcomes after ST-elevation myocardial infarction (STEMI)?

Background: Drug-eluting stents reduce restenosis rates compared to bare-metal stents. However, there is concern drug-eluting stents increase the risk of stent thrombosis leading to MI and death. Prior studies compared patients who received bare-metal versus those who received drug-eluting stents. Outcomes on a population level might provide new insight.

Study design: Observational study.

Setting: 100% national sample of patients 65 and older who received a coronary stent from 2002-05 enrolled in the traditional fee-for-service Medicare program.

Synopsis: 38,917 patients in the pre-drug-eluting-stent era from October 2002 to March 2003 received bare-metal stents. Nearly 62% of 28,086 patients studied from September to December 2003 received drug-eluting stents. The remaining 38.5% received bare-metal stents. Outcomes of percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), STEMI, and death were observed through December 31, 2005.

Patients in the drug-eluting-stent era had a lower two-year risk for repeat revascularization compared to patients in the bare-metal-stent era. In the drug-eluting versus bare-metal eras, repeat PCI was 17.1% versus 20.0% (p<0.001) and need for CABG was 2.7% versus 4.2% (p<0.01). Comparing adjusted outcomes for death, or STEMI, at two years, the two groups appeared similar.

The study did have limitations: the data only reflect sirolimus stents, the authors could not assess dual-antiplatelet therapy or obtain information on coronary anatomy or procedure details to account for selection bias in stent utilization, and the patients were all Medicare beneficiaries.

Bottom line: Drug-eluting stents are associated with fewer repeat revascularization procedures than bare-metal stents, but have not shown a significant improvement in the subsequent risk of STEMI or death.

Citation: Malenka DJ, Kaplan AV, Lucas FL, Sharp SM, Skinner JA. Outcomes following coronary stenting in the era of bare-metal vs. the era of drug-eluting stents. JAMA 2008;299(24):2868-2877.

Does case volume affect hospital performance for publicly reported process measures?

Background: Hospitals are increasingly graded and compared to one another. “Top medical centers” are defined as those within the top 10% of hospitals in specified performance measures. Hospitals with large and small case volumes might not be compared evenly and fairly.

Study design: Eight publicly reported process measures for acute myocardial infarction (AMI) were compared to hospital case volume, process performance, and label as “top hospital.”

Setting: Data were analyzed from the Hospital Quality Alliance for 3,761 hospitals from January to December 2005.

Synopsis: Hospitals with large case volume overall had better process performance. For example, looking at use of beta-blockers in patients with AMI on arrival to a hospital, small-volume hospitals (<10 AMI cases) averaged 72% while large volume (>100 AMI cases) averaged 80% (p<0.001). However, hospitals with small case volumes were more likely to receive “top hospital” rating even when hospitals with very low case volumes were excluded.

Hospital quality reporting that does not account for case volume is misleading to hospitals and consumers. In this study, larger-volume hospitals appeared to perform better in process measures, but were less likely to receive “top hospital” rating.

Bottom line: Hospitals with large and small case volumes can easily be compared to one another for process measures in AMI.

Citation: O’Brien SM, DeLong ER, Peterson ED. Impact of case volume on hospital performance assessment. Arch Intern Med. 2008;168(12):1277-1284.

What is the predictive value of QRS duration in patients hospitalized with worsening CHF?

Background: In outpatients, a prolonged QRS duration (greater than 120 ms) is associated with increased mortality. Its value in the inpatient setting is unclear. For patients hospitalized with CHF exacerbations, establishing the value of QRS duration may allow for tailored management.

Study design: Retrospective post hoc analysis from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST).

Setting: 4,133 patients were enrolled from North American, South American, and European sites.

Synopsis: Of 2,962 patients included in the final post hoc analysis, 1,321 (44.6%) had a prolonged QRS duration. During a median follow up of 9.9 months, the all-cause mortality rate was 18.7% for patients with a normal baseline QRS duration and 28.1% for patients with a prolonged baseline QRS.

After adjusting for confounding variables, patients with a prolonged baseline QRS had a 24% increased risk of all-cause mortality and a 28% increased risk for a composite endpoint of cardiac mortality or hospitalization for heart failure exacerbation.

The retrospective nature of the analysis represents the major limitation of this study. In addition, most of the enrolled patients were white, which limits the studies generalizability to other ethnic groups.

Bottom Line: A prolonged QRS duration for patients admitted with decompensated left ventricular heart failure is common and may be associated with increased morbidity and mortality.

Citation: Wang NC, Maggioni AP, Konstam MA, Zannad F, Drasa HB, Burnett JC, et al. Clinical implications of QRS duration in patients hospitalized with worsening heart failure and reduced left ventricular ejection fraction. JAMA. 2008;299(22):2656-2666.

For patients with out-of-hospital cardiac arrest, does the addition of vasopressin to epinephrine in a protocol for ACLS improve outcomes?

Background: The outcome for patients experiencing cardiac arrest who require vasopressors remains extremely poor. Despite disappointing data on vasopressin as an alternative treatment during cardiac arrest, a recent subgroup analysis suggested patients who received epinephrine and vasopressin together had superior clinical outcomes.

Study Design: Prospective multicenter randomized double-blind controlled trial.

Setting: 31 emergency medical service organizations in France.

Synopsis: Of the 2,894 patients, 20.7% of those receiving combination treatment (vasopressin plus epinephrine) survived to hospital admission versus 21.3% of those in the epinephrine-only group. For those same groups, 1.7% of combination and 2.3% of epinephrine-only patients survived to hospital discharge. No significant outcome differences were found in any group or subgroup analysis.

The study had lower-than-expected overall survival to hospital discharge, which may have handicapped its effort to find a true difference in treatment arms.

Bottom line: The addition of vasopressin to epinephrine in the treatment of out-of-hospital cardiac arrest does not improve outcomes.

Citation: Gueugniaud PY, David JS, Chanzy E, Hubert H, Dubien P, Mauriaucourt P, et al. Vasopressin and epinephrine vs. epinephrine alone in cardiopulmonary resuscitation. N Engl J Med. 2008;359(1):21-30.

Is oral rivaroxaban more efficacious than subcutaneous enoxaparin in preventing VTE after hip-replacement surgery?

Background: Venous thromboembolism (VTE) prophylaxis after total hip replacement (THR) is important but can be cumbersome because the most commonly used anticoagulants are either subcutaneous or require frequent monitoring. Rivaroxaban, an oral direct inhibitor of factor Xa may provide more convenient anticoagulation postoperatively. However, its efficacy and safety are unknown.

Study design: Randomized double-blind trial.

Setting: Multicenter study performed in 27 countries.

Synopsis: Patients undergoing THR surgery were randomized to oral rivaroxaban (10mg once daily without monitoring, started six to eight hours after surgery) or subcutaneous enoxaparin (40mg once daily, started 12 hours prior to surgery). After surgery, prophylaxis was administered for 35 days. The primary outcome was a composite of asymptomatic deep venous thrombosis (DVT), symptomatic DVT or pulmonary embolism (PE), or death from any cause at 36 days after surgery.

In the enoxaparin group, 3.7% of patients experienced the primary outcome. This decreased to 1.1% in the rivaroxaban group. Approximately one-third of events were symptomatic. Major bleeding occurred in 0.1% and 0.3% (p=NS) of patients in the enoxaparin and rivaroxaban groups, respectively.

The study is limited by the exclusion of 1,388 of the 4,541 patients (30.6%) randomized, primarily due to having inadequate venography. Also, because the majority of thromboembolic events were asymptomatic, the primary outcome overemphasizes the clinical difference.

Bottom line: Oral rivaroxaban without monitoring is more efficacious than, and as safe as, subcutaneous enoxaparin when used for VTE prophylaxis for THR.

Citation: Eriksson B, Borris LC, Friedman RJ, Hass S, Huisman MV, Kakkar AK, et al. Rivaroxaban versus enoxaparin for thromboprophylaxis after hip arthroplasty. N Engl J Med. 2008;358:2765-75.

Is LMWH more efficacious than UFH in preventing postoperative VTE in cancer patients?

Background: Patients with cancer are at increased risk for VTE and require prophylaxis to prevent this complication postoperatively. Low molecular weight heparin (LMWH) has proven more efficacious than subcutaneous unfractionated heparin (UFH) in other settings (e.g., DVT treatment). However, it is still unknown whether LMWH offers better prophylaxis compared to UFH for cancer patients undergoing surgery.

Study design: Systematic review and meta-analysis.

Setting: 14 randomized controlled trials.

Synopsis: Eleven trials exclusively examined patients with cancer (n=4006) and three trials reported data for cancer patients as subgroups (n=1816). There were in differences in mortality, pulmonary embolism, and symptomatic DVT rates between the two groups.

LMWH was associated with a decrease in total (asymptomatic or symptomatic) DVT (RR, 0.72; 95% CI, 0.55-0.94). Rates of major bleeding, minor bleeding, and intraoperative blood loss were similar between the two treatments.

This meta-analysis is limited because 12 remaining trials (n=3185) also enrolled cancer patients but did not provide specific data for the cancer patient subgroup. The study also is limited by the heterogeneity of the original trials, including utilizing varying LMWHs and dosing regimens, numerous types of surgeries, and a wide range of neoplasms.

Bottom line: LMWH does not decrease mortality, pulmonary embolism, or symptomatic DVT compared to UFH in cancer patients undergoing surgery.

Citation: Akl EA, Terrenato I, Barba M, Sperati F, Sempos EV, Muti P, et al. Low-molecular-weight heparin vs. unfractionated heparin for perioperative thromboprophylaxis in patients with cancer. Arch Intern Med. 2008;168:1261-9.

Does salmeterol added to inhaled corticosteroids improve severe asthma-related events?

Background: Asthma is a chronic disease causing major morbidity and mortality worldwide. Disease guidelines recommend all patients with persistent asthma be treated with inhaled corticosteroids. These same guidelines recommend adding a long-acting beta-agonist for patients whose symptoms persist. However, the safety of this practice has come under scrutiny.

Study design: Meta-analysis.

Setting: Sixty-six randomized, controlled trials conducted worldwide.

Synopsis: Analysis included 66 GlaxoSmithKline trials with a total of 20,966 patients with persistent asthma. Patients used either salmeterol (50mcg twice daily) plus inhaled corticosteroid (10,400 patients) or inhaled corticosteroid alone (10,566 patients).

Results showed no differences in asthma-related hospitalizations, asthma-related intubations, or deaths between the two groups. However, due to the low number of events, definitive conclusions are difficult to make. Severe asthma exacerbations requiring systemic corticosteroids significantly decreased in the inhaled corticosteroid plus salmeterol group.

The study is limited by it inclusion of only those trials sponsored by GlaxoSmithKline and by the short duration of most of the studies. Additionally, the studies included in the analysis used clinical outcomes as secondary endpoints.

Bottom line: Adding salmeterol to inhaled corticosteroid decreases severe asthma exacerbations and is likely safe, but does not have an effect on asthma-related hospitalization or death.

Citation: Bateman E, Nelson H, Bousquet J, Kral K, Sutton L, Ortega H, et.al. Meta-analysis: Effects of adding salmeterol to inhaled corticosteroids on serious asthma-related events. Annals Intern Med. 2008;149:33-42.

Is an early invasive strategy effective in women with unstable angina or NSTEMI?

Background: Despite many trials showing the value of an early invasive strategy for patients with non-ST-segment elevation acute coronary syndrome (NSTE ACS), data from several trials question this benefit in women. Some trials show higher risk of death and myocardial infarction (MI) in subgroup analysis of women.

Study Design: Meta-analysis.

Setting: Eight randomized, controlled trials conducted worldwide.

Synopsis: Analysis included eight trials with 10,412 patients (3,075 women) with NSTE ACS. The invasive group (5,083 patients) was defined as those referred for coronary angiography with subsequent intervention as needed. The composite endpoint of death, MI, or rehospitalization within 12 months with ACS occurred in 21.1% of the invasive group and 25.9% of the medically managed group (OR, 0.78; CI, 0.61-0.98).

The subgroup, including only women, had a non-statistically significant OR of 0.81 (CI, 0.65-1.01), including no effect on all-cause mortality, nonfatal MI, or the composite of death and MI. However, women with high-risk features (elevated biomarkers) undergoing the invasive strategy had a significant reduction in the composite endpoint (OR, 0.67; CI, 0.50-0.88).

The study is limited by the use of subgroup analysis, secondary endpoints, heterogeneity between trials, and possible publication bias.

Bottom line: Early invasive strategy is effective in men and high-risk women with NSTE ACS, but not in low-risk women.

Citation: O’Donoghue M, Boden W, Braunwald E, Cannon CP, Clayton TC, Winter RJ, et.al. Early invasive vs. conservative treatment strategies in women and men with unstable angina and non-ST-segment elevation myocardial infarction. JAMA. 2008;300:71-80.

What strategies are used to prevent contrast-induced acute kidney injury?

Background: Contrast-induced acute kidney injury (CIAKI) is a condition potentially amenable to preventive care. Several trials have identified intravenous hydration, N-acetylcysteine, and withdrawal of NSAIDS as interventions that reduce the possibility of CIAKI in high-risk patients. Little is known about whether healthcare providers routinely use these strategies.

Study design: Prospective observational cohort study.

Setting: Veterans Affairs (VA) Pittsburgh Healthcare System.

Synopsis: 11,410 patients scheduled for radiographic procedures were screened. After exclusion criteria and eligibility, 660 patients with an estimated glomerular filtration rate less than 60ml/min/1.73m2 were identified. Usage of intravenous fluids, N-acetylcysteine, and discontinuation of NSAIDS were recorded. Serum creatinine (SCr) was measured 48 to 96 hours post-procedure. CIAKI was defined as relative increase in SCr from baseline (≥25%, ≥50% and ≥100%) and absolute increase in SCr levels from baseline (≥0.25, ≥0.5, and ≥1.0). CIAKI association with adverse outcomes was evaluated by tracking 30-day mortality, need for dialysis, and hospitalization.

The incidence of CIAKI was less common in patients undergoing CT scans versus those having angiograms. Adverse 30-day outcomes were uncommon. Pre- and post-procedure intravenous hydration was administered to 40% of study patients, more commonly with coronary angiogram than with computed tomography (91.2% vs. 16%, p<0.0001). N-acetylcysteine was administered to 39.2%. Only 6.8% of those taking NSAIDS reported being told to discontinue the medication.

Study limitations include the small sample size and the single site location, both limiting generalizability.

Bottom line: Clinically significant CIAKI is uncommon, and preventive care is not uniformly implemented in patients undergoing contrast-enhanced radiographic procedures.

Citation: Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Sonel AF, Fine MJ, et al. Prevention, incidence, and outcomes of contrast-induced acute kidney injury. Arch Intern Med. 2008;168(12):1325-1332.

How does hyperglycemia affect morbidity and mortality in children admitted to a community pediatric hospital?

Background: Inpatient hyperglycemia in adult patients is a predictor of poor clinical outcomes. The association of hyperglycemia and clinical outcomes in children admitted to a general community hospital has not been studied.

Study design: Retrospective observational cohort study.

Setting: A community pediatric hospital in Atlanta, Ga.

Synopsis: Review of medical records of 903 consecutive pediatric patients admitted to critical and non-critical areas took place. Of these, 542 patients constituted the study population. The study excluded 342 patients who didn’t have a blood glucose measurement. Hyperglycemia was defined as an admission or in-hospital blood glucose greater than 120mg/dl.

One-fourth of the children admitted to the hospital had hyperglycemia, most without a prior history of diabetes. The presence of hyperglycemia on admission was not associated with increased length of stay (LOS) or increased mortality. Children with hyperglycemia were more likely to be admitted to the ICU and had longer ICU LOS.

This was a retrospective study conducted at a single site whose demographics and disease spectrum may differ from those of other institutions. There were an insufficient number of deaths to make any conclusions regarding the impact of hyperglycemia on mortality. Prospective, randomized, multicenter trials are needed to better elucidate the effects of in-patient hyperglycemia.

Bottom line: Hyperglycemia is common in children with or without diabetes admitted to the hospital, and is associated with increased ICU admissions and ICU length of stay. Its connection to mortality is inconclusive.

Citation: Palaio A, Smiley D, Ceron M, Klein R, Cho IS, Mejia R, et al. Prevalence and clinical outcome of inpatient hyperglycemia in a community pediatric hospital. J Hosp Med.2008;3(3):212-217.