Emily Tarvin, MD

Clinical question: Does uncontrolled diabetes mellitus increase risk for post-operative complications after elective joint replacement surgery?

Background: Several previous studies suggested that patients with uncontrolled diabetes could be at higher risk of postoperative complications and have worse functional outcomes after joint replacement surgery than patients without diabetes. Preoperative glycemic control is a potentially modifiable risk factor in patients undergoing elective joint replacement surgery. Demand for elective joint replacement is expected to increase over time, and reducing the risk of postoperative complications is essential in order to optimize functional outcomes and reduce healthcare costs.

Study design: Retrospective cohort.

Setting: Five regions of the Kaiser Permanente healthcare system.

Synopsis: The study included 40,491 patients aged 18 years and older who underwent primary knee replacement between January 2001 and December 2009 in five regions of the Kaiser Permanente system. Patients were identified using the Kaiser Permanente Total Joint Replacement Registry. Clinical information on each patient was collected from two years before the procedure to one year after the procedure using Kaiser Permanente electronic health records. Subjects were classified as nondiabetic (81.3%), diabetic with good glycemic control (12.5%), or diabetic with poor glycemic control (6.2%). Glycemic control status was assessed using the latest hemoglobin A1c (HbA1c) value measured prior to the date of the index surgery, with HbA1c <7.0% defined as good glycemic control. Outcomes included revision arthroplasty, deep infection, DVT or PE, incident myocardial infarction, and rehospitalization.

There was no significant association identified between uncontrolled diabetes and any of the five outcomes.

Limitations of the study include retrospective design, rarity of all outcomes except all-cause rehospitalization, and the small number of patients with uncontrolled diabetes in the cohort. In addition, functional outcomes were not assessed in this study.

Bottom line: The effect of uncontrolled diabetes on the risk of adverse surgical outcomes following elective joint replacement remains unclear based on currently published data; more studies are needed.

Citation: Adams AL, Paxton EW, Wang JQ, et al. Surgical outcomes of total knee replacement according to diabetes status and glycemic control, 2001-2009. J Bone Joint Surg Am. 2013;95:481-487.

Clinical question: Does uncontrolled diabetes mellitus increase risk for post-operative complications after elective joint replacement surgery?

Background: Several previous studies suggested that patients with uncontrolled diabetes could be at higher risk of postoperative complications and have worse functional outcomes after joint replacement surgery than patients without diabetes. Preoperative glycemic control is a potentially modifiable risk factor in patients undergoing elective joint replacement surgery. Demand for elective joint replacement is expected to increase over time, and reducing the risk of postoperative complications is essential in order to optimize functional outcomes and reduce healthcare costs.

Study design: Retrospective cohort.

Setting: Five regions of the Kaiser Permanente healthcare system.

Synopsis: The study included 40,491 patients aged 18 years and older who underwent primary knee replacement between January 2001 and December 2009 in five regions of the Kaiser Permanente system. Patients were identified using the Kaiser Permanente Total Joint Replacement Registry. Clinical information on each patient was collected from two years before the procedure to one year after the procedure using Kaiser Permanente electronic health records. Subjects were classified as nondiabetic (81.3%), diabetic with good glycemic control (12.5%), or diabetic with poor glycemic control (6.2%). Glycemic control status was assessed using the latest hemoglobin A1c (HbA1c) value measured prior to the date of the index surgery, with HbA1c <7.0% defined as good glycemic control. Outcomes included revision arthroplasty, deep infection, DVT or PE, incident myocardial infarction, and rehospitalization.

There was no significant association identified between uncontrolled diabetes and any of the five outcomes.

Limitations of the study include retrospective design, rarity of all outcomes except all-cause rehospitalization, and the small number of patients with uncontrolled diabetes in the cohort. In addition, functional outcomes were not assessed in this study.

Bottom line: The effect of uncontrolled diabetes on the risk of adverse surgical outcomes following elective joint replacement remains unclear based on currently published data; more studies are needed.

Citation: Adams AL, Paxton EW, Wang JQ, et al. Surgical outcomes of total knee replacement according to diabetes status and glycemic control, 2001-2009. J Bone Joint Surg Am. 2013;95:481-487.

Clinical question: Does uncontrolled diabetes mellitus increase risk for post-operative complications after elective joint replacement surgery?

Background: Several previous studies suggested that patients with uncontrolled diabetes could be at higher risk of postoperative complications and have worse functional outcomes after joint replacement surgery than patients without diabetes. Preoperative glycemic control is a potentially modifiable risk factor in patients undergoing elective joint replacement surgery. Demand for elective joint replacement is expected to increase over time, and reducing the risk of postoperative complications is essential in order to optimize functional outcomes and reduce healthcare costs.

Study design: Retrospective cohort.

Setting: Five regions of the Kaiser Permanente healthcare system.

Synopsis: The study included 40,491 patients aged 18 years and older who underwent primary knee replacement between January 2001 and December 2009 in five regions of the Kaiser Permanente system. Patients were identified using the Kaiser Permanente Total Joint Replacement Registry. Clinical information on each patient was collected from two years before the procedure to one year after the procedure using Kaiser Permanente electronic health records. Subjects were classified as nondiabetic (81.3%), diabetic with good glycemic control (12.5%), or diabetic with poor glycemic control (6.2%). Glycemic control status was assessed using the latest hemoglobin A1c (HbA1c) value measured prior to the date of the index surgery, with HbA1c <7.0% defined as good glycemic control. Outcomes included revision arthroplasty, deep infection, DVT or PE, incident myocardial infarction, and rehospitalization.

There was no significant association identified between uncontrolled diabetes and any of the five outcomes.

Limitations of the study include retrospective design, rarity of all outcomes except all-cause rehospitalization, and the small number of patients with uncontrolled diabetes in the cohort. In addition, functional outcomes were not assessed in this study.

Bottom line: The effect of uncontrolled diabetes on the risk of adverse surgical outcomes following elective joint replacement remains unclear based on currently published data; more studies are needed.

Citation: Adams AL, Paxton EW, Wang JQ, et al. Surgical outcomes of total knee replacement according to diabetes status and glycemic control, 2001-2009. J Bone Joint Surg Am. 2013;95:481-487.

Clinical question: What are the effects of the 2011 resident duty-hour requirements on first-year residents’ well-being and patient safety?

Background: In an effort to reduce adverse consequences associated with extended shift length and sleep deprivation, the Accreditation Council for Graduate Medical Education (ACGME) released a new set of duty-hour requirements effective July 2011. To date, little is known about the effects of the 2011 reforms on resident and patient outcomes.

Study design: Prospective, longitudinal cohort with pre-post analysis.

Setting: Residency programs from university- and community-based medical centers.

Synopsis: Fifty-one residency programs from 10 university-based and four community-based GME institutions were included. Incoming interns during the 2009, 2010, and 2011 academic years were invited to participate, and 58% (n=2,323) agreed to take part. Participants completed online surveys two months before starting their first residency (intern) year and at three, six, nine, and 12 months of internship. Questions addressed work hours, sleep, medical errors, depressive symptoms, and subjective well-being.

No significant differences in baseline findings were found between the pre-implementation cohort (interns entering in 2009 and 2010) and the post-implementation cohort (interns starting in 2011, following the new duty-hour requirements). Interns in the post-implementation cohort worked fewer hours than those in the pre-implementation cohort (mean hours per week 64.3 vs. 67.0, P<0.001). There were no significant changes in reported hours of sleep, depressive symptom score, or well-being score between the pre- and post-implementation cohorts. The percentage of respondents who reported committing a serious medical error increased in the post-implementation group (23.3% vs. 19.9%, P=0.007).

Limitations include the self-reported nature of the responses and the modest participation rate. The authors concluded that although the 2011 reforms decreased the total number of hours worked, additional strategies could be needed to achieve the desired effects on intern well-being and patient safety.

Bottom line: Following implementation of the 2011 ACGME duty-hour reforms, interns from diverse specialties and institutions experienced reductions in self-reported work hours, without any associated improvements in sleep, depressive symptoms, or well-being and with an increase in reported medical errors.

Citation: Sen S, Kranzler HR, Didwania AK, et al. Effects of the 2011 duty hour reforms on interns and their patients: a prospective longitudinal cohort study. JAMA Intern Med. 2013;173(8):657-662.

Clinical question: What are the effects of the 2011 resident duty-hour requirements on first-year residents’ well-being and patient safety?

Background: In an effort to reduce adverse consequences associated with extended shift length and sleep deprivation, the Accreditation Council for Graduate Medical Education (ACGME) released a new set of duty-hour requirements effective July 2011. To date, little is known about the effects of the 2011 reforms on resident and patient outcomes.

Study design: Prospective, longitudinal cohort with pre-post analysis.

Setting: Residency programs from university- and community-based medical centers.

Synopsis: Fifty-one residency programs from 10 university-based and four community-based GME institutions were included. Incoming interns during the 2009, 2010, and 2011 academic years were invited to participate, and 58% (n=2,323) agreed to take part. Participants completed online surveys two months before starting their first residency (intern) year and at three, six, nine, and 12 months of internship. Questions addressed work hours, sleep, medical errors, depressive symptoms, and subjective well-being.

No significant differences in baseline findings were found between the pre-implementation cohort (interns entering in 2009 and 2010) and the post-implementation cohort (interns starting in 2011, following the new duty-hour requirements). Interns in the post-implementation cohort worked fewer hours than those in the pre-implementation cohort (mean hours per week 64.3 vs. 67.0, P<0.001). There were no significant changes in reported hours of sleep, depressive symptom score, or well-being score between the pre- and post-implementation cohorts. The percentage of respondents who reported committing a serious medical error increased in the post-implementation group (23.3% vs. 19.9%, P=0.007).

Limitations include the self-reported nature of the responses and the modest participation rate. The authors concluded that although the 2011 reforms decreased the total number of hours worked, additional strategies could be needed to achieve the desired effects on intern well-being and patient safety.

Bottom line: Following implementation of the 2011 ACGME duty-hour reforms, interns from diverse specialties and institutions experienced reductions in self-reported work hours, without any associated improvements in sleep, depressive symptoms, or well-being and with an increase in reported medical errors.

Citation: Sen S, Kranzler HR, Didwania AK, et al. Effects of the 2011 duty hour reforms on interns and their patients: a prospective longitudinal cohort study. JAMA Intern Med. 2013;173(8):657-662.

Clinical question: What are the effects of the 2011 resident duty-hour requirements on first-year residents’ well-being and patient safety?

Background: In an effort to reduce adverse consequences associated with extended shift length and sleep deprivation, the Accreditation Council for Graduate Medical Education (ACGME) released a new set of duty-hour requirements effective July 2011. To date, little is known about the effects of the 2011 reforms on resident and patient outcomes.

Study design: Prospective, longitudinal cohort with pre-post analysis.

Setting: Residency programs from university- and community-based medical centers.

Synopsis: Fifty-one residency programs from 10 university-based and four community-based GME institutions were included. Incoming interns during the 2009, 2010, and 2011 academic years were invited to participate, and 58% (n=2,323) agreed to take part. Participants completed online surveys two months before starting their first residency (intern) year and at three, six, nine, and 12 months of internship. Questions addressed work hours, sleep, medical errors, depressive symptoms, and subjective well-being.

No significant differences in baseline findings were found between the pre-implementation cohort (interns entering in 2009 and 2010) and the post-implementation cohort (interns starting in 2011, following the new duty-hour requirements). Interns in the post-implementation cohort worked fewer hours than those in the pre-implementation cohort (mean hours per week 64.3 vs. 67.0, P<0.001). There were no significant changes in reported hours of sleep, depressive symptom score, or well-being score between the pre- and post-implementation cohorts. The percentage of respondents who reported committing a serious medical error increased in the post-implementation group (23.3% vs. 19.9%, P=0.007).

Limitations include the self-reported nature of the responses and the modest participation rate. The authors concluded that although the 2011 reforms decreased the total number of hours worked, additional strategies could be needed to achieve the desired effects on intern well-being and patient safety.

Bottom line: Following implementation of the 2011 ACGME duty-hour reforms, interns from diverse specialties and institutions experienced reductions in self-reported work hours, without any associated improvements in sleep, depressive symptoms, or well-being and with an increase in reported medical errors.

Citation: Sen S, Kranzler HR, Didwania AK, et al. Effects of the 2011 duty hour reforms on interns and their patients: a prospective longitudinal cohort study. JAMA Intern Med. 2013;173(8):657-662.

Clinical question: Are high plasma lactate levels associated with mortality and clinical deterioration among patients with acute pulmonary embolism (PE)?

Background: Prognostic clinical markers are limited in patients presenting with acute PE, especially among normotensive individuals. Plasma lactate concentration is a marker of tissue hypoperfusion that has been used to risk-stratify patients with sepsis and trauma. It is unknown whether elevated lactate levels predict poor outcomes in acute PE.

Study design: Prospective cohort.

Setting: ED in a large teaching hospital in Italy.

Synopsis: Consecutive adult patients with acute PE diagnosed by spiral computed tomography or lung scan were included. Plasma lactate levels were tested in all patients at presentation, and levels of ≥2 mmol/L were considered abnormal. The primary endpoint was all-cause death within 30 days, and the secondary endpoint was the composite of all-cause death and PE-related clinical deterioration and death.

Of the 270 patients, 81 (30%) had abnormal lactate levels, though only 12 (4.4%) had shock or hypotension. Patients with elevated lactate had higher mortality compared with patients with lower levels (17.3% vs. 1.6%, OR 12.95, 95% CI 3.43-58.73). Plasma lactate ≥2 mmol/L was associated with higher all-cause mortality (HR 11.67, 95% CI 3.32-41.03) and the composite endpoint (HR 8.14, 95% CI 3.83-17.34). This association was independent of the presence of hypotension, right ventricular dysfunction, or elevated troponin.

Limitations include the single study site (which limits generalizability of the findings) and that lactate levels were only checked once (which might not have fully reflected each patient’s clinical picture). The authors suggest that plasma lactate levels might have utility in determining which patients should be treated more aggressively for PE.

Bottom line: In patients presenting with acute PE, elevated plasma lactate levels are associated with increased risk of short-term mortality and morbidity, independent of the presence of hypotension or markers of right ventricular injury.

Citation: Vanni S, Viviani G, Baioni M, et al. Prognostic value of plasma lactate levels among patients with acute pulmonary embolism: the thrombo-embolism lactate outcome study. Ann Emerg Med. 2013;61:330-338.

Clinical question: Are high plasma lactate levels associated with mortality and clinical deterioration among patients with acute pulmonary embolism (PE)?

Background: Prognostic clinical markers are limited in patients presenting with acute PE, especially among normotensive individuals. Plasma lactate concentration is a marker of tissue hypoperfusion that has been used to risk-stratify patients with sepsis and trauma. It is unknown whether elevated lactate levels predict poor outcomes in acute PE.

Study design: Prospective cohort.

Setting: ED in a large teaching hospital in Italy.

Synopsis: Consecutive adult patients with acute PE diagnosed by spiral computed tomography or lung scan were included. Plasma lactate levels were tested in all patients at presentation, and levels of ≥2 mmol/L were considered abnormal. The primary endpoint was all-cause death within 30 days, and the secondary endpoint was the composite of all-cause death and PE-related clinical deterioration and death.

Of the 270 patients, 81 (30%) had abnormal lactate levels, though only 12 (4.4%) had shock or hypotension. Patients with elevated lactate had higher mortality compared with patients with lower levels (17.3% vs. 1.6%, OR 12.95, 95% CI 3.43-58.73). Plasma lactate ≥2 mmol/L was associated with higher all-cause mortality (HR 11.67, 95% CI 3.32-41.03) and the composite endpoint (HR 8.14, 95% CI 3.83-17.34). This association was independent of the presence of hypotension, right ventricular dysfunction, or elevated troponin.

Limitations include the single study site (which limits generalizability of the findings) and that lactate levels were only checked once (which might not have fully reflected each patient’s clinical picture). The authors suggest that plasma lactate levels might have utility in determining which patients should be treated more aggressively for PE.

Bottom line: In patients presenting with acute PE, elevated plasma lactate levels are associated with increased risk of short-term mortality and morbidity, independent of the presence of hypotension or markers of right ventricular injury.

Citation: Vanni S, Viviani G, Baioni M, et al. Prognostic value of plasma lactate levels among patients with acute pulmonary embolism: the thrombo-embolism lactate outcome study. Ann Emerg Med. 2013;61:330-338.

Clinical question: Are high plasma lactate levels associated with mortality and clinical deterioration among patients with acute pulmonary embolism (PE)?

Background: Prognostic clinical markers are limited in patients presenting with acute PE, especially among normotensive individuals. Plasma lactate concentration is a marker of tissue hypoperfusion that has been used to risk-stratify patients with sepsis and trauma. It is unknown whether elevated lactate levels predict poor outcomes in acute PE.

Study design: Prospective cohort.

Setting: ED in a large teaching hospital in Italy.

Synopsis: Consecutive adult patients with acute PE diagnosed by spiral computed tomography or lung scan were included. Plasma lactate levels were tested in all patients at presentation, and levels of ≥2 mmol/L were considered abnormal. The primary endpoint was all-cause death within 30 days, and the secondary endpoint was the composite of all-cause death and PE-related clinical deterioration and death.

Of the 270 patients, 81 (30%) had abnormal lactate levels, though only 12 (4.4%) had shock or hypotension. Patients with elevated lactate had higher mortality compared with patients with lower levels (17.3% vs. 1.6%, OR 12.95, 95% CI 3.43-58.73). Plasma lactate ≥2 mmol/L was associated with higher all-cause mortality (HR 11.67, 95% CI 3.32-41.03) and the composite endpoint (HR 8.14, 95% CI 3.83-17.34). This association was independent of the presence of hypotension, right ventricular dysfunction, or elevated troponin.

Limitations include the single study site (which limits generalizability of the findings) and that lactate levels were only checked once (which might not have fully reflected each patient’s clinical picture). The authors suggest that plasma lactate levels might have utility in determining which patients should be treated more aggressively for PE.

Bottom line: In patients presenting with acute PE, elevated plasma lactate levels are associated with increased risk of short-term mortality and morbidity, independent of the presence of hypotension or markers of right ventricular injury.

Citation: Vanni S, Viviani G, Baioni M, et al. Prognostic value of plasma lactate levels among patients with acute pulmonary embolism: the thrombo-embolism lactate outcome study. Ann Emerg Med. 2013;61:330-338.

Clinical question: How often do inpatient providers report direct communication with outpatient providers, and how is direct communication associated with 30-day readmissions?

Background: Studies have demonstrated that adverse events and errors occurring after hospital discharge can result from poor provider communication between the inpatient and outpatient setting.

Study design: Prospective cohort.

Setting: Johns Hopkins Hospital, Baltimore.

Synopsis: The presence or absence of direct communication between inpatient and outpatient healthcare providers was captured from a required field in an electronic discharge worksheet completed by the discharging physician. Of 6,635 hospitalizations studied, successful direct communication was reported in 36.7% of cases. Predictors of successful direct communication included patients cared for by hospitalists without house staff (OR 1.81, 95% CI 1.59-2.08), high expected 30-day readmission rate (OR 1.18, 95% CI 1.10-1.28), and insurance by Medicare (OR 1.35, 95% CI 1.16-1.56) and private insurance companies (OR 1.35, 95% CI 1.18-1.56). In adjusted analyses, direct communication between the inpatient and outpatient providers was not associated with 30-day readmissions (OR 1.08, 95% CI 0.92-1.26).

There were several limitations in this study. Only the primary team was surveyed; thus, it is not known if consulting providers might have contacted the outpatient providers. Only readmissions to the same medical center were studied, and therefore it is not known if patients were readmitted to other facilities. Additionally, information regarding discharge communication was self-reported, which might have introduced bias.

Bottom line: Self-reported direct communication between inpatient and outpatient providers occurred infrequently and was not associated with 30-day same-hospital readmission.

Citation: Oduyebo I, Lehmann C, Pollack C, et al. Association of self-reported hospital discharge handoffs with 30-day readmissions. JAMA Intern Med. 2013;173:624-629.

Clinical question: How often do inpatient providers report direct communication with outpatient providers, and how is direct communication associated with 30-day readmissions?

Background: Studies have demonstrated that adverse events and errors occurring after hospital discharge can result from poor provider communication between the inpatient and outpatient setting.

Study design: Prospective cohort.

Setting: Johns Hopkins Hospital, Baltimore.

Synopsis: The presence or absence of direct communication between inpatient and outpatient healthcare providers was captured from a required field in an electronic discharge worksheet completed by the discharging physician. Of 6,635 hospitalizations studied, successful direct communication was reported in 36.7% of cases. Predictors of successful direct communication included patients cared for by hospitalists without house staff (OR 1.81, 95% CI 1.59-2.08), high expected 30-day readmission rate (OR 1.18, 95% CI 1.10-1.28), and insurance by Medicare (OR 1.35, 95% CI 1.16-1.56) and private insurance companies (OR 1.35, 95% CI 1.18-1.56). In adjusted analyses, direct communication between the inpatient and outpatient providers was not associated with 30-day readmissions (OR 1.08, 95% CI 0.92-1.26).

There were several limitations in this study. Only the primary team was surveyed; thus, it is not known if consulting providers might have contacted the outpatient providers. Only readmissions to the same medical center were studied, and therefore it is not known if patients were readmitted to other facilities. Additionally, information regarding discharge communication was self-reported, which might have introduced bias.

Bottom line: Self-reported direct communication between inpatient and outpatient providers occurred infrequently and was not associated with 30-day same-hospital readmission.

Citation: Oduyebo I, Lehmann C, Pollack C, et al. Association of self-reported hospital discharge handoffs with 30-day readmissions. JAMA Intern Med. 2013;173:624-629.

Clinical question: How often do inpatient providers report direct communication with outpatient providers, and how is direct communication associated with 30-day readmissions?

Background: Studies have demonstrated that adverse events and errors occurring after hospital discharge can result from poor provider communication between the inpatient and outpatient setting.

Study design: Prospective cohort.

Setting: Johns Hopkins Hospital, Baltimore.

Synopsis: The presence or absence of direct communication between inpatient and outpatient healthcare providers was captured from a required field in an electronic discharge worksheet completed by the discharging physician. Of 6,635 hospitalizations studied, successful direct communication was reported in 36.7% of cases. Predictors of successful direct communication included patients cared for by hospitalists without house staff (OR 1.81, 95% CI 1.59-2.08), high expected 30-day readmission rate (OR 1.18, 95% CI 1.10-1.28), and insurance by Medicare (OR 1.35, 95% CI 1.16-1.56) and private insurance companies (OR 1.35, 95% CI 1.18-1.56). In adjusted analyses, direct communication between the inpatient and outpatient providers was not associated with 30-day readmissions (OR 1.08, 95% CI 0.92-1.26).

There were several limitations in this study. Only the primary team was surveyed; thus, it is not known if consulting providers might have contacted the outpatient providers. Only readmissions to the same medical center were studied, and therefore it is not known if patients were readmitted to other facilities. Additionally, information regarding discharge communication was self-reported, which might have introduced bias.

Bottom line: Self-reported direct communication between inpatient and outpatient providers occurred infrequently and was not associated with 30-day same-hospital readmission.

Citation: Oduyebo I, Lehmann C, Pollack C, et al. Association of self-reported hospital discharge handoffs with 30-day readmissions. JAMA Intern Med. 2013;173:624-629.

Clinical question: In patients who undergo inferior vena cava (IVC) filter placement for venous thromboembolism (VTE) prophylaxis or treatment, what are the associated patient characteristics, indications for placement, complications, retrieval date, and use of concomitant anticoagulant therapy?

Background: Retrievable IVC filters were designed to provide short-term protection from pulmonary embolism but are often left in place indefinitely. Retrievable IVC filters that are not removed can carry significant long-term risks. Further, the use of filters for VTE prophylaxis is controversial, and there are multiple sets of conflicting guidelines for filter insertion provided by various professional groups.

Study design: Retrospective review of IVC filter use over an eight-year period.

Setting: Boston Medical Center.

Synopsis: Medical records from all patients at Boston Medical Center who had a billing code for placement of an IVC filter between August 2003 and February 2011 were manually reviewed. Nine hundred fifty-two medical records were evaluated, of which 679 (71.3%) patients had retrievable IVC filters placed. The most common indications for filter placement were trauma (50.2%), malignancy (15.9%), and bleeding during anticoagulation (11.8%).

In total, 448 patients (47.1%) had filters placed for prophylactic purposes in the absence of documented VTE. Seventy-four patients developed VTE after filter placement; 48.2% of post-filter insertion VTEs occurred in patients who had no VTE prior to the filter; and 89.4% occurred in patients not receiving anticoagulants. An attempt was made to remove 71 of 679 (10.5%) retrievable filters, and 58 (8.5%) attempts were successful. There were 10 serious complications related to mechanical filter failure, including migration or fracture of the filter.

In this study, there was a high volume of filter use by the trauma service; thus, the patient population might be different from the hospital medicine patient population. The study also lacked systematic imaging and follow-up data. Further studies are needed to analyze the risks associated with IVC filter placement.

Bottom line: Use of IVC filters for VTE treatment and prophylaxis, in the context of low filter retrieval rates and lack of appropriate anticoagulant therapy, results in suboptimal outcomes.

Citation: Sarosiek S, Crowther M, Sloan M. Indications, complications, and management of inferior vena cava filters: the experience in 952 patients at an academic hospital with a level I trauma center. JAMA Intern Med. 2013;173:513-517.

Clinical question: In patients who undergo inferior vena cava (IVC) filter placement for venous thromboembolism (VTE) prophylaxis or treatment, what are the associated patient characteristics, indications for placement, complications, retrieval date, and use of concomitant anticoagulant therapy?

Background: Retrievable IVC filters were designed to provide short-term protection from pulmonary embolism but are often left in place indefinitely. Retrievable IVC filters that are not removed can carry significant long-term risks. Further, the use of filters for VTE prophylaxis is controversial, and there are multiple sets of conflicting guidelines for filter insertion provided by various professional groups.

Study design: Retrospective review of IVC filter use over an eight-year period.

Setting: Boston Medical Center.

Synopsis: Medical records from all patients at Boston Medical Center who had a billing code for placement of an IVC filter between August 2003 and February 2011 were manually reviewed. Nine hundred fifty-two medical records were evaluated, of which 679 (71.3%) patients had retrievable IVC filters placed. The most common indications for filter placement were trauma (50.2%), malignancy (15.9%), and bleeding during anticoagulation (11.8%).

In total, 448 patients (47.1%) had filters placed for prophylactic purposes in the absence of documented VTE. Seventy-four patients developed VTE after filter placement; 48.2% of post-filter insertion VTEs occurred in patients who had no VTE prior to the filter; and 89.4% occurred in patients not receiving anticoagulants. An attempt was made to remove 71 of 679 (10.5%) retrievable filters, and 58 (8.5%) attempts were successful. There were 10 serious complications related to mechanical filter failure, including migration or fracture of the filter.

In this study, there was a high volume of filter use by the trauma service; thus, the patient population might be different from the hospital medicine patient population. The study also lacked systematic imaging and follow-up data. Further studies are needed to analyze the risks associated with IVC filter placement.

Bottom line: Use of IVC filters for VTE treatment and prophylaxis, in the context of low filter retrieval rates and lack of appropriate anticoagulant therapy, results in suboptimal outcomes.

Citation: Sarosiek S, Crowther M, Sloan M. Indications, complications, and management of inferior vena cava filters: the experience in 952 patients at an academic hospital with a level I trauma center. JAMA Intern Med. 2013;173:513-517.

Clinical question: In patients who undergo inferior vena cava (IVC) filter placement for venous thromboembolism (VTE) prophylaxis or treatment, what are the associated patient characteristics, indications for placement, complications, retrieval date, and use of concomitant anticoagulant therapy?

Background: Retrievable IVC filters were designed to provide short-term protection from pulmonary embolism but are often left in place indefinitely. Retrievable IVC filters that are not removed can carry significant long-term risks. Further, the use of filters for VTE prophylaxis is controversial, and there are multiple sets of conflicting guidelines for filter insertion provided by various professional groups.

Study design: Retrospective review of IVC filter use over an eight-year period.

Setting: Boston Medical Center.

Synopsis: Medical records from all patients at Boston Medical Center who had a billing code for placement of an IVC filter between August 2003 and February 2011 were manually reviewed. Nine hundred fifty-two medical records were evaluated, of which 679 (71.3%) patients had retrievable IVC filters placed. The most common indications for filter placement were trauma (50.2%), malignancy (15.9%), and bleeding during anticoagulation (11.8%).

In total, 448 patients (47.1%) had filters placed for prophylactic purposes in the absence of documented VTE. Seventy-four patients developed VTE after filter placement; 48.2% of post-filter insertion VTEs occurred in patients who had no VTE prior to the filter; and 89.4% occurred in patients not receiving anticoagulants. An attempt was made to remove 71 of 679 (10.5%) retrievable filters, and 58 (8.5%) attempts were successful. There were 10 serious complications related to mechanical filter failure, including migration or fracture of the filter.

In this study, there was a high volume of filter use by the trauma service; thus, the patient population might be different from the hospital medicine patient population. The study also lacked systematic imaging and follow-up data. Further studies are needed to analyze the risks associated with IVC filter placement.

Bottom line: Use of IVC filters for VTE treatment and prophylaxis, in the context of low filter retrieval rates and lack of appropriate anticoagulant therapy, results in suboptimal outcomes.

Citation: Sarosiek S, Crowther M, Sloan M. Indications, complications, and management of inferior vena cava filters: the experience in 952 patients at an academic hospital with a level I trauma center. JAMA Intern Med. 2013;173:513-517.

Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?

Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.

Study design: Retrospective cohort.

Setting: Academic medical center in Boston.

Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.

This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.

Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.

Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.

Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?

Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.

Study design: Retrospective cohort.

Setting: Academic medical center in Boston.

Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.

This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.

Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.

Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.

Clinical question: Can a prediction model based on administrative and clinical data identify potentially avoidable 30-day readmissions in medical patients prior to discharge?

Background: An estimated 18% of Medicare beneficiaries are readmitted to the hospital within 30 days of discharge, costing nearly $17 billion per year. Interventions to reduce readmission rates are costly and should be focused on high-risk patients. To date, using models to predict 30-day readmission has been problematic and unreliable.

Study design: Retrospective cohort.

Setting: Academic medical center in Boston.

Synopsis: Using consecutive discharges from all medical services of Brigham and Women’s Hospital occurring over one year, this study derived and internally validated a prediction model for potentially avoidable 30-day readmissions. Of 10,731 discharges, there were 2,399 (22%) 30-day readmissions, and 879 (8.5%) were deemed potentially avoidable. Seven independent predictors for readmission were identified and used to create a predictor score referred to as the HOSPITAL score. Predictors included hemoglobin and sodium levels at discharge, number of hospitalizations in the past year, and four features of the index hospitalization, including type, discharge from an oncology service, presence of procedures, and length of stay. The score was internally validated and found to predict potentially avoidable 30-day readmission in medical patients with fair discriminatory power and good calibration.

This study is unique in that none of the classic comorbidities (e.g. congestive heart failure) were associated with a higher risk of 30-day readmission. Previously unrecognized predictors, including hemoglobin, sodium, and number of procedures performed, were incorporated. This suggests that comorbidities are not as important as illness severity or clinical instability. Hospitalists should await studies that externally validate the HOSPITAL score before incorporating it into practice.

Bottom line: A unique and simple seven-item prediction model identifies potentially avoidable 30-day readmissions but needs to be externally validated before being widely utilized.

Citation: Donze J, Drahomir A, Williams D, Schnipper JL. Potentially avoidable 30-day hospital readmissions in medical patients. JAMA Intern Med. 2013;137(8):632-638.

Clinical question: How does timing of surgical antibiotic prophylaxis affect risk of postoperative surgical-site infections (SSIs)?

Background: Antibiotic prophylaxis for major surgical procedures has been proven in clinical trials to reduce rates of SSI. The Centers for Medicare & Medicaid Services’ (CMS) Surgical Care Improvement Project (SCIP) has implemented quality metrics to ensure antibiotics are administered within 60 minutes of incision; however, studies have failed to show that a 60-minute pre-incision window is advantageous.

Study design: Retrospective cohort.

Setting: Veterans Affairs hospitals.

Synopsis: Using SCIP and VA Surgical Quality Improvement Program data from 112 VA hospitals, 32,459 cases of hip or knee arthroplasty, colorectal surgery, arterial vascular surgery, and hysterectomy from 2005-2009 were reviewed. A postoperative SSI occurred in 1,497 cases (4.6%). Using several statistical methods, the relationship between timing of prophylactic antibiotic administration and postoperative SSI within 30 days was evaluated. In unadjusted models, higher SSI rates were observed with antibiotic administration more than 60 minutes prior to incision (OR 1.34, 95% CI 1.08-1.66) but not after incision (OR 1.26, 95% CI 0.92-1.72), compared with procedures with antibiotics administered within 60 minutes pre-incision. However, after adjustment for patient, procedure, and antibiotic variables, no significant relationship between timing and SSI was observed (P=0.50 for all specialties).

The study sample was comprised primarily of older men and did not include patients who underwent cardiac procedures, limiting the generalizability of the findings. Nonetheless, the study is the largest of its kind and confirms previous studies that suggest there is no significant relationship between timing of antibiotics and SSI. Prophylactic antibiotics should still be used when indicated; however, using timing of prophylactic antibiotics as a quality measure is unlikely to improve outcomes.

Bottom line: Adherence to the empiric 60-minute window metric for timing of prophylactic antibiotics is not significantly associated with risk of SSI.

Citation: Hawn MT, Richman JS, Vick CC, et al. Timing of surgical antibiotic prophylaxis and the risk of surgical site infection. JAMA Surg. 2013 March 20:1-8. doi: 10.1001/jamasurg.2013.134 [Epub ahead of print].

Clinical question: How does timing of surgical antibiotic prophylaxis affect risk of postoperative surgical-site infections (SSIs)?

Background: Antibiotic prophylaxis for major surgical procedures has been proven in clinical trials to reduce rates of SSI. The Centers for Medicare & Medicaid Services’ (CMS) Surgical Care Improvement Project (SCIP) has implemented quality metrics to ensure antibiotics are administered within 60 minutes of incision; however, studies have failed to show that a 60-minute pre-incision window is advantageous.

Study design: Retrospective cohort.

Setting: Veterans Affairs hospitals.

Synopsis: Using SCIP and VA Surgical Quality Improvement Program data from 112 VA hospitals, 32,459 cases of hip or knee arthroplasty, colorectal surgery, arterial vascular surgery, and hysterectomy from 2005-2009 were reviewed. A postoperative SSI occurred in 1,497 cases (4.6%). Using several statistical methods, the relationship between timing of prophylactic antibiotic administration and postoperative SSI within 30 days was evaluated. In unadjusted models, higher SSI rates were observed with antibiotic administration more than 60 minutes prior to incision (OR 1.34, 95% CI 1.08-1.66) but not after incision (OR 1.26, 95% CI 0.92-1.72), compared with procedures with antibiotics administered within 60 minutes pre-incision. However, after adjustment for patient, procedure, and antibiotic variables, no significant relationship between timing and SSI was observed (P=0.50 for all specialties).

The study sample was comprised primarily of older men and did not include patients who underwent cardiac procedures, limiting the generalizability of the findings. Nonetheless, the study is the largest of its kind and confirms previous studies that suggest there is no significant relationship between timing of antibiotics and SSI. Prophylactic antibiotics should still be used when indicated; however, using timing of prophylactic antibiotics as a quality measure is unlikely to improve outcomes.

Bottom line: Adherence to the empiric 60-minute window metric for timing of prophylactic antibiotics is not significantly associated with risk of SSI.

Citation: Hawn MT, Richman JS, Vick CC, et al. Timing of surgical antibiotic prophylaxis and the risk of surgical site infection. JAMA Surg. 2013 March 20:1-8. doi: 10.1001/jamasurg.2013.134 [Epub ahead of print].

Clinical question: How does timing of surgical antibiotic prophylaxis affect risk of postoperative surgical-site infections (SSIs)?

Background: Antibiotic prophylaxis for major surgical procedures has been proven in clinical trials to reduce rates of SSI. The Centers for Medicare & Medicaid Services’ (CMS) Surgical Care Improvement Project (SCIP) has implemented quality metrics to ensure antibiotics are administered within 60 minutes of incision; however, studies have failed to show that a 60-minute pre-incision window is advantageous.

Study design: Retrospective cohort.

Setting: Veterans Affairs hospitals.

Synopsis: Using SCIP and VA Surgical Quality Improvement Program data from 112 VA hospitals, 32,459 cases of hip or knee arthroplasty, colorectal surgery, arterial vascular surgery, and hysterectomy from 2005-2009 were reviewed. A postoperative SSI occurred in 1,497 cases (4.6%). Using several statistical methods, the relationship between timing of prophylactic antibiotic administration and postoperative SSI within 30 days was evaluated. In unadjusted models, higher SSI rates were observed with antibiotic administration more than 60 minutes prior to incision (OR 1.34, 95% CI 1.08-1.66) but not after incision (OR 1.26, 95% CI 0.92-1.72), compared with procedures with antibiotics administered within 60 minutes pre-incision. However, after adjustment for patient, procedure, and antibiotic variables, no significant relationship between timing and SSI was observed (P=0.50 for all specialties).

The study sample was comprised primarily of older men and did not include patients who underwent cardiac procedures, limiting the generalizability of the findings. Nonetheless, the study is the largest of its kind and confirms previous studies that suggest there is no significant relationship between timing of antibiotics and SSI. Prophylactic antibiotics should still be used when indicated; however, using timing of prophylactic antibiotics as a quality measure is unlikely to improve outcomes.

Bottom line: Adherence to the empiric 60-minute window metric for timing of prophylactic antibiotics is not significantly associated with risk of SSI.

Citation: Hawn MT, Richman JS, Vick CC, et al. Timing of surgical antibiotic prophylaxis and the risk of surgical site infection. JAMA Surg. 2013 March 20:1-8. doi: 10.1001/jamasurg.2013.134 [Epub ahead of print].

Clinical question: What is the long-term outcome of elderly survivors of in-hospital cardiac arrest?

Background: Previous studies have examined in-hospital survival from in-hospital cardiac arrest but have not looked at long-term outcomes and readmission of in-hospital cardiac arrest survivors.

Study design: Retrospective cohort.

Setting: Acute-care hospitals that submitted data to the Get with the Guidelines—Resuscitation registry between 2000 and 2008.

Synopsis: Using the Get with the Guidelines—Resuscitation registry from 401 acute-care hospitals, data from 6,972 Medicare patients aged 65 years or older who had a pulseless in-hospital cardiac arrest and survived to discharge were analyzed. Survival rates were 82% at 30 days, 72% at three months, 58.5% at one year, and 49.6% at two years. Survival at three years was 43.5%, similar to patients discharged with heart failure.

One-year survival decreased with increasing age. Survival also decreased with black race (52.5% vs. 60.4% for white patients, P=0.001) and male sex (58.6% vs. 60.9% for women, P=0.03). Patients with mild or no neurologic disability at discharge had a higher survival rate at one year than patients with moderate, severe, or coma state. Readmission rates at one year after discharge were 65.6% and 76.2% at two years. Black patients, women, and patients with neurologic disability at discharge were more likely to be readmitted.

Because this is an observational study looking at a quality database of Medicare patients, it excludes patients at VA hospitals and non-Medicare facilities. This data excludes assessments of quality of life after discharge and health status among those with long-term survival, and does not include cause of death.

Bottom line: One-year survival following in-hospital cardiac arrest for patients over age 65 approaches 60% and decreases with increasing age, male sex, and black race.

Citation: Chan PS, Nallamothu BK, Krumholz HM, et al. Long-term outcomes in elderly survivors of in-hospital cardiac arrest. N Engl J Med. 2013;368:1019-1026.

Clinical question: What is the long-term outcome of elderly survivors of in-hospital cardiac arrest?

Background: Previous studies have examined in-hospital survival from in-hospital cardiac arrest but have not looked at long-term outcomes and readmission of in-hospital cardiac arrest survivors.

Study design: Retrospective cohort.

Setting: Acute-care hospitals that submitted data to the Get with the Guidelines—Resuscitation registry between 2000 and 2008.

Synopsis: Using the Get with the Guidelines—Resuscitation registry from 401 acute-care hospitals, data from 6,972 Medicare patients aged 65 years or older who had a pulseless in-hospital cardiac arrest and survived to discharge were analyzed. Survival rates were 82% at 30 days, 72% at three months, 58.5% at one year, and 49.6% at two years. Survival at three years was 43.5%, similar to patients discharged with heart failure.

One-year survival decreased with increasing age. Survival also decreased with black race (52.5% vs. 60.4% for white patients, P=0.001) and male sex (58.6% vs. 60.9% for women, P=0.03). Patients with mild or no neurologic disability at discharge had a higher survival rate at one year than patients with moderate, severe, or coma state. Readmission rates at one year after discharge were 65.6% and 76.2% at two years. Black patients, women, and patients with neurologic disability at discharge were more likely to be readmitted.

Because this is an observational study looking at a quality database of Medicare patients, it excludes patients at VA hospitals and non-Medicare facilities. This data excludes assessments of quality of life after discharge and health status among those with long-term survival, and does not include cause of death.

Bottom line: One-year survival following in-hospital cardiac arrest for patients over age 65 approaches 60% and decreases with increasing age, male sex, and black race.

Citation: Chan PS, Nallamothu BK, Krumholz HM, et al. Long-term outcomes in elderly survivors of in-hospital cardiac arrest. N Engl J Med. 2013;368:1019-1026.

Clinical question: What is the long-term outcome of elderly survivors of in-hospital cardiac arrest?

Background: Previous studies have examined in-hospital survival from in-hospital cardiac arrest but have not looked at long-term outcomes and readmission of in-hospital cardiac arrest survivors.

Study design: Retrospective cohort.

Setting: Acute-care hospitals that submitted data to the Get with the Guidelines—Resuscitation registry between 2000 and 2008.

Synopsis: Using the Get with the Guidelines—Resuscitation registry from 401 acute-care hospitals, data from 6,972 Medicare patients aged 65 years or older who had a pulseless in-hospital cardiac arrest and survived to discharge were analyzed. Survival rates were 82% at 30 days, 72% at three months, 58.5% at one year, and 49.6% at two years. Survival at three years was 43.5%, similar to patients discharged with heart failure.

One-year survival decreased with increasing age. Survival also decreased with black race (52.5% vs. 60.4% for white patients, P=0.001) and male sex (58.6% vs. 60.9% for women, P=0.03). Patients with mild or no neurologic disability at discharge had a higher survival rate at one year than patients with moderate, severe, or coma state. Readmission rates at one year after discharge were 65.6% and 76.2% at two years. Black patients, women, and patients with neurologic disability at discharge were more likely to be readmitted.

Because this is an observational study looking at a quality database of Medicare patients, it excludes patients at VA hospitals and non-Medicare facilities. This data excludes assessments of quality of life after discharge and health status among those with long-term survival, and does not include cause of death.

Bottom line: One-year survival following in-hospital cardiac arrest for patients over age 65 approaches 60% and decreases with increasing age, male sex, and black race.

Citation: Chan PS, Nallamothu BK, Krumholz HM, et al. Long-term outcomes in elderly survivors of in-hospital cardiac arrest. N Engl J Med. 2013;368:1019-1026.

Clinical question: What are the changes in the updated Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines?

Background: Chronic obstructive pulmonary disease (COPD) remains a leading cause of death in the U.S. and worldwide. The GOLD guidelines are an international consensus report on COPD diagnosis, management, and prevention, first released in 2001. The 2011 revision to the guidelines was recently published and outlines substantial changes based on updated literature and expert opinion.

Study design: Guidelines based on studies with varying designs.

Setting: Expert panel review of multiple studies from different settings.

Synopsis: While the diagnosis of COPD remains based on a post-bronchodilator fixed ratio of FEV1/FVC <0.70, there is more emphasis on global clinical assessment in the new guidelines. The updated approach describes classifying COPD severity based on risk/symptom frequency using established symptom assessment and the frequency of acute exacerbations of COPD. Instead of five “stages” based on FEV1 measures alone, there are now four "grades" of A through D (A: low risk/fewer symptoms; B: low risk/more symptoms; C: high risk/fewer symptoms; D: high risk/more symptoms) to more easily guide treatment options.

Treatment strategies are also updated, focusing not only on reduction of current symptoms, but also risk of future events. Pharmacologic treatment recommendations include using bronchodilator monotherapy in Group A patients, favoring long-acting over short-acting bronchodilators in Group B patients, prescribing inhaled corticosteroids only in combination with long-acting bronchodilators in Groups C and D patients, and considering newer agents such as phosphodiesterase-4 inhibitors in Group D patients.

Non-pharmacologic interventions include ongoing smoking cessation strategies, exercise promotion, treatment of comorbidities, and even public health strategies in pollution control.

Bottom line: The GOLD guidelines have undergone major revisions that provide a more practical approach to classification of COPD based on symptom severity and risk assessment in order to direct providers in evidence-based treatment that addresses both short-term and long-term impact of the disease.

Citation: Global Initiative for Chronic Obstructive Lung Disease. Global strategy for diagnosis, management, and prevention of COPD. Global Initiative for Chronic Obstructive Lung Disease website. Accessed Oct. 29, 2012.

For more physician reviews of recent HM-relevant literature, visit our website.

Clinical question: What are the changes in the updated Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines?

Background: Chronic obstructive pulmonary disease (COPD) remains a leading cause of death in the U.S. and worldwide. The GOLD guidelines are an international consensus report on COPD diagnosis, management, and prevention, first released in 2001. The 2011 revision to the guidelines was recently published and outlines substantial changes based on updated literature and expert opinion.

Study design: Guidelines based on studies with varying designs.

Setting: Expert panel review of multiple studies from different settings.

Synopsis: While the diagnosis of COPD remains based on a post-bronchodilator fixed ratio of FEV1/FVC <0.70, there is more emphasis on global clinical assessment in the new guidelines. The updated approach describes classifying COPD severity based on risk/symptom frequency using established symptom assessment and the frequency of acute exacerbations of COPD. Instead of five “stages” based on FEV1 measures alone, there are now four "grades" of A through D (A: low risk/fewer symptoms; B: low risk/more symptoms; C: high risk/fewer symptoms; D: high risk/more symptoms) to more easily guide treatment options.

Treatment strategies are also updated, focusing not only on reduction of current symptoms, but also risk of future events. Pharmacologic treatment recommendations include using bronchodilator monotherapy in Group A patients, favoring long-acting over short-acting bronchodilators in Group B patients, prescribing inhaled corticosteroids only in combination with long-acting bronchodilators in Groups C and D patients, and considering newer agents such as phosphodiesterase-4 inhibitors in Group D patients.

Non-pharmacologic interventions include ongoing smoking cessation strategies, exercise promotion, treatment of comorbidities, and even public health strategies in pollution control.

Bottom line: The GOLD guidelines have undergone major revisions that provide a more practical approach to classification of COPD based on symptom severity and risk assessment in order to direct providers in evidence-based treatment that addresses both short-term and long-term impact of the disease.

Citation: Global Initiative for Chronic Obstructive Lung Disease. Global strategy for diagnosis, management, and prevention of COPD. Global Initiative for Chronic Obstructive Lung Disease website. Accessed Oct. 29, 2012.

For more physician reviews of recent HM-relevant literature, visit our website.

Clinical question: What are the changes in the updated Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines?

Background: Chronic obstructive pulmonary disease (COPD) remains a leading cause of death in the U.S. and worldwide. The GOLD guidelines are an international consensus report on COPD diagnosis, management, and prevention, first released in 2001. The 2011 revision to the guidelines was recently published and outlines substantial changes based on updated literature and expert opinion.

Study design: Guidelines based on studies with varying designs.

Setting: Expert panel review of multiple studies from different settings.

Synopsis: While the diagnosis of COPD remains based on a post-bronchodilator fixed ratio of FEV1/FVC <0.70, there is more emphasis on global clinical assessment in the new guidelines. The updated approach describes classifying COPD severity based on risk/symptom frequency using established symptom assessment and the frequency of acute exacerbations of COPD. Instead of five “stages” based on FEV1 measures alone, there are now four "grades" of A through D (A: low risk/fewer symptoms; B: low risk/more symptoms; C: high risk/fewer symptoms; D: high risk/more symptoms) to more easily guide treatment options.

Treatment strategies are also updated, focusing not only on reduction of current symptoms, but also risk of future events. Pharmacologic treatment recommendations include using bronchodilator monotherapy in Group A patients, favoring long-acting over short-acting bronchodilators in Group B patients, prescribing inhaled corticosteroids only in combination with long-acting bronchodilators in Groups C and D patients, and considering newer agents such as phosphodiesterase-4 inhibitors in Group D patients.

Non-pharmacologic interventions include ongoing smoking cessation strategies, exercise promotion, treatment of comorbidities, and even public health strategies in pollution control.

Bottom line: The GOLD guidelines have undergone major revisions that provide a more practical approach to classification of COPD based on symptom severity and risk assessment in order to direct providers in evidence-based treatment that addresses both short-term and long-term impact of the disease.

Citation: Global Initiative for Chronic Obstructive Lung Disease. Global strategy for diagnosis, management, and prevention of COPD. Global Initiative for Chronic Obstructive Lung Disease website. Accessed Oct. 29, 2012.

For more physician reviews of recent HM-relevant literature, visit our website.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Interventions that improve discharge handovers reviewed
2. Duration of in-hospital cardiac resuscitation and survival rates
3. Early sepsis intervention strategies to decrease mortality risk
4. Hypoglycemia linked to increased mortality in critically ill
5. Increased bleeding risk for cardiac patients
6. Hospital-run vs. anesthesiologist-run preoperative clinics
7. Postoperative delirium and cognitive impairment in cardiac patients
8. Benefits of resuming anticoagulants after GI bleeding
9. Preoperative hyponatremia and risk of perioperative mortality

Systematic Review Highlights Several Interventions That Improve Discharge Handovers

Clinical question: Do interventions to improve patient handovers at discharge have positive effects on patient care?

Background: The transition from hospital to primary care is often suboptimal and has been associated with unfavorable outcomes, including hospital readmission, increased healthcare utilization, and adverse drug events post-discharge. This review sought to characterize different types of interventions aimed at improving discharge handovers and to evaluate their effects.

Study design: Systematic review of randomized controlled trials.

Setting: Studies published from January 1990 to March 2011.

Synopsis: Review of published databases identified 36 randomized controlled studies on interventions to improve discharge handovers. Studies were blindly evaluated by two reviewers on quality, interventions, and outcomes. There was significant heterogeneity in interventions and outcomes; thus, statistical analysis was not possible. Most studies evaluated multicomponent interventions and had more than one outcome measure.

Of the 36 studies reviewed, 25 reported statistically significant improvements in outcomes, including reduced hospital utilization and improved continuity of care. Effective interventions included medication reconciliation; structured discharge information (facilitated by electronic resources); multidisciplinary discharge planning; shared involvement in arranging care between inpatient and outpatient physicians; and Web-based access to discharge information by the outpatient provider.

The complexity of the interventions and the heterogeneity of reported results did not allow for firm conclusions to be drawn regarding which specific interventions had the strongest effects.

Bottom line: Interventions that target the quality and safety of handovers between hospital and outpatient providers at discharge can significantly reduce hospital utilization and improve continuity of care.

Citation: Hesselink G, Schoonhoven L, Barach P, et al. Improving patient handovers from hospital to primary care: a systematic review. Ann Intern Med. 2012;157:417-428.

Longer Duration of In-Hospital Cardiac Resuscitation Associated with Increased Survival

Clinical question: Are prolonged cardiac resuscitation efforts associated with improved outcomes?

Background: There is little evidence or guidelines on how long to maintain resuscitative efforts during in-hospital cardiac arrest, leading to variation in practice. This study characterized patterns of resuscitation duration and relationship to the return of spontaneous circulation and survival to discharge.

Study design: Retrospective observational study.

Setting: Four hundred thirty-five U.S. hospitals reporting data to the American Heart Association’s Get With The Guidelines: Resuscitation registry.

Synopsis: Using duration of resuscitation in nonsurvivors as a surrogate for the tendency of a facility to perform prolonged efforts, hospitals were divided into quartiles. Overall, of 64,339 patients in the registry, 31,198 (48.5%) had return of circulation and 9,912 (15.4%) survived to discharge. Resuscitative efforts in nonsurvivors ranged from a median of 16 to 25 minutes between the lowest and highest quartiles.

There was a stepwise increase in the likelihood for patients to have return of spontaneous circulation and to survive to discharge between each quartile. Specifically, comparing shortest to longest, there was a significant adjusted odds ratio of 1.12 for both the return of circulation (P<0.0001) and survival to discharge (P=0.021). The survival benefit was most apparent for those with pulseless electrical activity or asystole as initial rhythms, as compared to ventricular tachycardia or fibrillation.

Limitations included the study’s observational design, which meant causality could not be determined. Additionally, the study did not account for the quality of resuscitative efforts (e.g. depth of chest compressions, adherence with guidelines), which might have influenced outcomes. Importantly, the study looked at survival to discharge, but it did not evaluate long-term survival or functional status post-discharge, which might better reflect the success of resuscitation.

Bottom line: Prolonged resuscitative efforts were observed to be associated with increased likelihood of return of spontaneous circulation and survival to discharge; this data might provide clinical guidance in determining when to stop resuscitation efforts.

Citation: Goldberger ZD, Chan PS, Berg RA, et al. Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study. Lancet. 2012;380:1473-1481.

Early Sepsis Intervention Strategies Decrease Mortality, Length of Stay, and Cost

Clinical question: For patients with sepsis or septic shock, what is the impact of a real-time comprehensive continuous quality-improvement (QI) initiative on in-hospital mortality, morbidity, and healthcare resource utilization in community and tertiary-care hospitals in the U.S.?

Background: Multiple single-center trials have demonstrated that early sepsis intervention strategies (early goal-directed therapy, resuscitation bundles) improve in-hospital mortality. Little is known about the effectiveness of incorporating these strategies into a real-time continuous QI initiative and implementing interventions across multiple sites simultaneously.

Study design: Pre-post at some sites and concurrent implementation design at other sites.

Setting: Five community and six-tertiary care U.S. hospitals.

Synopsis: The GENeralized Early Sepsis Intervention Strategies (GENESIS) project was a CQI initiative that combined several QI concepts with validated early sepsis interventions known as resuscitation bundles (RB). Continuous QI was implemented on patients with severe sepsis or septic shock in both before and after designs (eight hospitals), and in concurrent designs (three hospitals). The control group was comprised of historical controls treated before GENESIS and patients with incomplete implementation of RB, totaling 1,554 patients. The treatment group included patients treated after GENESIS and those with complete RB compliance, totaling 4,801 patients.

Compared with the control group, patients in the treatment group had a 33% decreased risk of in-hospital mortality (RR 0.67, 95% CI 0.63-0.72), an absolute decrease in hospital length of stay (LOS) by 5.1 days (20.7 days vs. 15.6 days, P<0.001) and a $47,923 reduction in total hospital charges (P<0.001). Limitations included the study design (not a prospective randomized trial), and the possibility of other concurrent unmeasured quality initiatives taking place at the study sites, which might have contributed to improved outcomes.

Bottom line: Early sepsis intervention strategies in the form of a comprehensive continuous QI initiative can decrease mortality, hospital LOS, and cost in both tertiary-care and community hospitals.

Citation: Cannon CM, Holthaus CV, Zubrow MT, et al. The GENESIS project (GENeralized Early Sepsis Intervention Strategies): a multicenter quality improvement collaborative. J Intensive Care Med. 2012; Aug 17. doi:10.1177/0885066612453025.

Hypoglycemia Associated with Increased Mortality in the Critically Ill

Clinical question: Is hypoglycemia associated with mortality in critically ill patients?

Background: Initial studies suggested that intensive glucose control reduces mortality in surgical ICU patients and reduces morbidity in medical ICU patients, but further studies have not supported these findings. Recent literature shows conflicting results on the effects of intense glucose control in critically ill patients.

Study design: Post-hoc analysis of the NICE-SUGAR study database.

Setting: ICUs in 42 hospitals in Australia, New Zealand, and Canada.

Synopsis: The NICE-SUGAR study was a multicenter trial that randomized 6,104 ICU patients to intensive (glucose 80 to 108 mg/dL) or conventional glucose control (≤180 mg/dL). Patients were followed for 90 days or until death, with death being the primary end point. Severe hypoglycemia (<40 mg/dL) was recorded in 6.8% of patients in the intensive glucose control group versus 0.5% in the conventional group. The study showed that intensive glucose control was associated with increased mortality among adult ICU patients.

Using the NICE-SUGAR database, the authors conducted a Cox regression analysis to examine the associations between hypoglycemia and death. A total of 2,714 patients had moderate hypoglycemia (glucose 41 to 70 mg/dL), and 223 patients had severe hypoglycemia. The hazard ratio for mortality was 1.41 (95% CI 1.21-1.62, P<0.001) for patients with moderate hypoglycemia and 2.10 (95% CI 1.59-2.77, P<0.001) for severe hypoglycemia, compared to patients without hypoglycemia.

These findings show a strong association between mortality and hypoglycemia but do not prove causality. Hospitalists caring for ICU patients must be aware that hypoglycemia is associated with mortality and focus on avoiding hypoglycemia. The American Diabetes Association currently recommends a target blood glucose level of 140 to 180 mg/dL for most critically ill patients.

Bottom line: Hypoglycemia (glucose <70 mg/dL) is associated with increased risk of mortality in ICU patients.

Citation: Finfer S, Chittock DR, Su SY, et al. Hypoglycemia and risk of death in critically ill patients. New Engl J Med. 2012;367(12):1108-1118.

Increased Bleeding Risk for Cardiac Patients on Multiple Antithrombotic Drugs

Clinical question: Is there an increased risk of bleeding in atrial fibrillation patients treated with multiple antithrombotic agents following acute myocardial infarction (MI) or percutaneous coronary intervention (PCI)?

Background: Current treatment for atrial fibrillation patients with MI or PCI includes vitamin K antagonist (VKA) therapy to prevent stroke and antiplatelet agents to prevent further coronary events. There are inconsistent findings on the safety and efficacy of combined therapy with VKA, aspirin, and clopidogrel, specifically with regard to bleeding risk.

Study design: Retrospective cohort study.

Setting: Nationwide registry in Denmark.

Synopsis: Using the National Patient Registry in Denmark, 11,480 patients with atrial fibrillation who were admitted for MI or PCI were identified. Patients were grouped by medication regimen, including monotherapy (aspirin, clopidogrel, or VKA), dual therapy (dual antiplatelet or VKA+antiplatelet), or triple therapy (VKA+aspirin+clopidogrel). The primary outcome was nonfatal or fatal bleeding within one year.

Patients receiving triple therapy had the highest rate of bleeding, with a crude incidence of 14.2 events per 100 person-years. Patients treated with VKA+aspirin+clopidogel were significantly more likely than those on VKA+antiplatelet (HR 1.47, 95% CI 1.04-2.08) or dual antiplatelet (HR 2.20, 95% CI 1.58-3.08) treatment to have a bleeding event within 90 days, and similar trends were seen at 90 to 360 days. There was no significant difference in thromboembolic events among patients on VKA+aspirin+clopidogrel versus VKA+antiplatelet therapy.

Only bleeding events that required hospitalization were recorded, which might underestimate the bleeding risks of these regimens. Additionally, INR levels were not determined, which could impact both bleeding and thromboembolic outcomes.

However, this study does suggest that there are significant bleeding risks among patients treated with triple therapy. Hospitalists should weigh the risks of thromboembolic events with bleeding risks in patients with atrial fibrillation and MI/PCI, and only prescribe VKA+aspirin+clopidogrel with these risks in mind.

Bottom line: Immediate and continued bleeding risk is increased in patients with atrial fibrillation admitted with PCI or MI who are placed on triple antithrombotic therapy with VKA+aspirin+clopidogrel.

Citation: Lamberts M, Olesen JB, Ruwald MH, et al. Bleeding after initiation of multiple antithrombotic drugs, including triple therapy, in atrial fibrillation patients following myocardial infarction and coronary intervention. Circulation. 2012;126:1185-1193.

Hospitalist-Run Preoperative Clinic Improves Outcomes in Complex Surgical Patients

Clinical question: Do hospitalist-run preoperative clinics improve outcomes for medically complex patients undergoing noncardiac surgery compared to traditional anesthesiologist-run preoperative clinics?

Background: Studies of perioperative medical consultation have shown inconsistent effects on quality of care, but preoperative medical consultation has only been evaluated in the immediate preoperative period (one day prior to surgery or less). Little is known about the impact of involving hospitalists earlier in the preoperative period.

Study design: Retrospective, pre-post study.

Setting: Veterans Affairs Greater Los Angeles Healthcare System (VAGLAHS).

Synopsis: In July 2004, the VAGLAHS Preoperative Clinic transitioned from being anesthesiologist-run to hospitalist-run. Mid-level providers were trained on preoperative medical assessment, and patients were only evaluated by anesthesia staff on the day of surgery, after they had been deemed medically acceptable for surgery. All patients seen in the clinic from July 2003 to July 2005 were included in the study. Period A included patients evaluated when anesthesia staff supervised the clinic. Period B included patients evaluated during the first year of the hospitalist-run system.

There were 1,101 patients with inpatient surgeries in Period A, and 1,126 patients in Period B. Mean length of stay (LOS) decreased to 5.28 days during the hospitalist-run model from 9.87 days during the anesthesiologist-run model. LOS reductions were most notable in patients with ASA scores of 3 or higher; LOS reductions were not seen in an internal control of surgical patients who were not evaluated in the preoperative clinic. Inpatient mortality was also reduced in Period B compared with Period A, to 4 cases (0.36%) from 14 cases (1.27%) (P=0.0158). Specific processes that led to improved outcomes for patients in Period B could not be identified. The VA study setting might limit the generalizability of the results.

Bottom line: A hospitalist-run preoperative clinic was associated with decreased LOS and inpatient mortality compared with a traditional anesthesiologist-run clinic.

Citation: Vazirani S, Lankarani-Fard A, Lian LJ, Stelzner M, Asch SM. Preoperative processes and outcomes after implementation of a hospitalist-run preoperative clinic. J Hosp Med. 2012 Sep 7. doi:10.1002/jhm.1968.

Delirium after Cardiac Surgery Associated with Prolonged Cognitive Impairment

Clinical question: Is postoperative delirium associated with decreased cognitive function in the first year after cardiac surgery?

Background: In general populations, delirium has been associated with long-term decline in cognitive ability. Delirium and cognitive dysfunction are both common following cardiac surgery, but the effects of postoperative delirium on the trajectory of cognitive function over time is unclear.

Study design: Prospective cohort study.

Setting: Two academic medical centers and a Veterans Administration hospital.

Synopsis: Two hundred twenty-five patients aged 60 or older who were scheduled to undergo coronary artery bypass grafting or valve replacement surgery were included. Patients underwent preoperative assessment cognitive function with the use of the Mini-Mental State Examination (MMSE). Starting on postoperative Day 2, patients underwent daily assessment for delirium. After discharge, cognitive function was reassessed at months one, six, and 12.

Postoperative delirium occurred in 103 patients (46%). Patients with delirium were older, had higher comorbidity scores, and had lower MMSE scores at baseline. Among the overall study population, adjusted MMSE scores dropped 4.6 points from baseline to postoperative Day 2, then were observed to increase by approximately one point per day during postoperative days 3 to 5 with minimal change thereafter. Patients with delirium had greater decrease in cognitive function in the immediate postoperative period compared to patients without delirium (7.7 points vs. 2.1, P<0.001).

Patients without delirium returned to their baseline cognitive ability by one month postoperatively, while patients who had delirium were still making gains up to six months post-operatively, never returning to baseline level of function by one year. Unmeasured confounders and uncertain sensitivity of the MMSE to detect mild cognitive impairment might limit these findings.

Bottom line: Cognitive function decreases in the immediate postoperative period following cardiac surgery. Compared to patients without delirium, patients with delirium experience more dramatic and prolonged cognitive impairment postoperatively, without returning to their preoperative level of cognitive function at one year.

Citation: Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories after postoperative delirium. New Engl J Med. 2012;367:30-39.

Benefits of Resuming Anticoagulation after GI Bleed Outweigh Risks for Most Patients

Clinical question: In warfarin-treated patients who have experienced gastrointestinal (GI) bleeding, what are the patterns of restarting warfarin therapy and the incidence of thrombosis, recurrent GI bleed, and death in the 90 days following index bleed?

Background: In warfarin-treated patients who experience GI bleeding, warfarin is often temporarily held or permanently discontinued, placing patients at increased risk for developing thromboembolism. Little is known about the risks, benefits, and timing of restarting warfarin in this patient population.

Study design: Retrospective cohort study.

Setting: Kaiser Permanente Colorado.

Synopsis: Using clinical and administrative databases, 442 patients who presented with GI bleeding while receiving warfarin therapy were identified. Patients were grouped by whether they resumed warfarin (n=260, including 41 patients in whom anticoagulation was never interrupted), or did not resume warfarin therapy (n=182) in the 90 days following index GI bleed. Patients with prosthetic heart valves or GI bleeding localized to the rectum/anus were more commonly restarted on warfarin, whereas older patients and those in whom the source of bleeding was not identified were less likely to be restarted on warfarin therapy.

Restarting warfarin therapy after index GI bleed was associated with lower risk of thrombosis (HR 0.05, 95% CI 0.01-0.58) and death from any cause (HR 0.31, 95% CI 0.15-0.62), and it was not associated with a significant increase in risk for recurrent GI bleed (HR 1.32, 95% CI 0.50-3.57).

The authors concluded that for many patients who experience a warfarin-associated GI bleed, the benefits of restarting warfarin therapy outweigh the risks. No conclusions were made regarding the optimal timing of resuming therapy. Limitations included the use of administrative data and inability to determine the potential influence of aspirin use on outcomes.

Bottom line: Resuming warfarin in the 90 days following a warfarin-associated GI bleed is associated with decreased risk of thrombosis and death without increased risk for recurrent GI bleed.

Citation: Witt DM, Delate T, Garcia DA, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding. Arch Intern Med. 2012 Sep 12. doi:10.1001/archinternmed.2012.4261.

Preoperative Hyponatremia Associated with Increased Risk for Perioperative Complications and Mortality

Clinical question: Is preoperative hyponatremia an indicator of perioperative morbidity and mortality?

Background: Hyponatremia is a common diagnosis in the hospital setting and is associated with adverse outcomes, even in mild cases. However, it is unclear if this association exists in surgical patients when detected preoperatively.

Study design: Retrospective cohort study.

Setting: Academic and community hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP).

Synopsis: A total of 75,423 adult patients with hyponatremia (sodium <135 mEq/L) who were undergoing major surgery were compared to 888,840 patients with normal preoperative sodium levels over a six-year period. The primary outcome was 30-day mortality. Secondary outcomes included postoperative major coronary events, stroke, wound infection, pneumonia, and length of stay (LOS).

Compared to patients with normal sodium levels, those with preoperative hyponatremia had higher rates of perioperative mortality (5.2% vs. 1.3%; adjusted odds ratio 1.44, 95% CI 1.38-1.50), with increased risk that correlated with increasing severity of hyponatremia. Association with postoperative mortality was particularly strong among hyponatremic patients with ASA scores of 1 or 2 and those undergoing nonemergency surgery.

Patients with preoperative hyponatremia were also found to have increased risk for all postoperative complications evaluated, with the exception of stroke. Limitations included the potential for unmeasured confounders and not being able to account for the role of medications used perioperatively. Research is needed to determine whether correcting preoperative hyponatremia lessens the risk of mortality and other postoperative complications.

Bottom line: Among patients undergoing major surgery, preoperative hyponatremia is a predictor of postoperative 30-day mortality and morbidity.

Citation: Leung AA, McAlister FA, Rogers SO, et al. Preoperative hyponatremia and perioperative complications. Arch Intern Med. 2012;172:1-8.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Interventions that improve discharge handovers reviewed
2. Duration of in-hospital cardiac resuscitation and survival rates
3. Early sepsis intervention strategies to decrease mortality risk
4. Hypoglycemia linked to increased mortality in critically ill
5. Increased bleeding risk for cardiac patients
6. Hospital-run vs. anesthesiologist-run preoperative clinics
7. Postoperative delirium and cognitive impairment in cardiac patients
8. Benefits of resuming anticoagulants after GI bleeding
9. Preoperative hyponatremia and risk of perioperative mortality

Systematic Review Highlights Several Interventions That Improve Discharge Handovers

Clinical question: Do interventions to improve patient handovers at discharge have positive effects on patient care?

Background: The transition from hospital to primary care is often suboptimal and has been associated with unfavorable outcomes, including hospital readmission, increased healthcare utilization, and adverse drug events post-discharge. This review sought to characterize different types of interventions aimed at improving discharge handovers and to evaluate their effects.

Study design: Systematic review of randomized controlled trials.

Setting: Studies published from January 1990 to March 2011.

Synopsis: Review of published databases identified 36 randomized controlled studies on interventions to improve discharge handovers. Studies were blindly evaluated by two reviewers on quality, interventions, and outcomes. There was significant heterogeneity in interventions and outcomes; thus, statistical analysis was not possible. Most studies evaluated multicomponent interventions and had more than one outcome measure.

Of the 36 studies reviewed, 25 reported statistically significant improvements in outcomes, including reduced hospital utilization and improved continuity of care. Effective interventions included medication reconciliation; structured discharge information (facilitated by electronic resources); multidisciplinary discharge planning; shared involvement in arranging care between inpatient and outpatient physicians; and Web-based access to discharge information by the outpatient provider.

The complexity of the interventions and the heterogeneity of reported results did not allow for firm conclusions to be drawn regarding which specific interventions had the strongest effects.

Bottom line: Interventions that target the quality and safety of handovers between hospital and outpatient providers at discharge can significantly reduce hospital utilization and improve continuity of care.

Citation: Hesselink G, Schoonhoven L, Barach P, et al. Improving patient handovers from hospital to primary care: a systematic review. Ann Intern Med. 2012;157:417-428.

Longer Duration of In-Hospital Cardiac Resuscitation Associated with Increased Survival

Clinical question: Are prolonged cardiac resuscitation efforts associated with improved outcomes?

Background: There is little evidence or guidelines on how long to maintain resuscitative efforts during in-hospital cardiac arrest, leading to variation in practice. This study characterized patterns of resuscitation duration and relationship to the return of spontaneous circulation and survival to discharge.

Study design: Retrospective observational study.

Setting: Four hundred thirty-five U.S. hospitals reporting data to the American Heart Association’s Get With The Guidelines: Resuscitation registry.

Synopsis: Using duration of resuscitation in nonsurvivors as a surrogate for the tendency of a facility to perform prolonged efforts, hospitals were divided into quartiles. Overall, of 64,339 patients in the registry, 31,198 (48.5%) had return of circulation and 9,912 (15.4%) survived to discharge. Resuscitative efforts in nonsurvivors ranged from a median of 16 to 25 minutes between the lowest and highest quartiles.

There was a stepwise increase in the likelihood for patients to have return of spontaneous circulation and to survive to discharge between each quartile. Specifically, comparing shortest to longest, there was a significant adjusted odds ratio of 1.12 for both the return of circulation (P<0.0001) and survival to discharge (P=0.021). The survival benefit was most apparent for those with pulseless electrical activity or asystole as initial rhythms, as compared to ventricular tachycardia or fibrillation.

Limitations included the study’s observational design, which meant causality could not be determined. Additionally, the study did not account for the quality of resuscitative efforts (e.g. depth of chest compressions, adherence with guidelines), which might have influenced outcomes. Importantly, the study looked at survival to discharge, but it did not evaluate long-term survival or functional status post-discharge, which might better reflect the success of resuscitation.

Bottom line: Prolonged resuscitative efforts were observed to be associated with increased likelihood of return of spontaneous circulation and survival to discharge; this data might provide clinical guidance in determining when to stop resuscitation efforts.

Citation: Goldberger ZD, Chan PS, Berg RA, et al. Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study. Lancet. 2012;380:1473-1481.

Early Sepsis Intervention Strategies Decrease Mortality, Length of Stay, and Cost

Clinical question: For patients with sepsis or septic shock, what is the impact of a real-time comprehensive continuous quality-improvement (QI) initiative on in-hospital mortality, morbidity, and healthcare resource utilization in community and tertiary-care hospitals in the U.S.?

Background: Multiple single-center trials have demonstrated that early sepsis intervention strategies (early goal-directed therapy, resuscitation bundles) improve in-hospital mortality. Little is known about the effectiveness of incorporating these strategies into a real-time continuous QI initiative and implementing interventions across multiple sites simultaneously.

Study design: Pre-post at some sites and concurrent implementation design at other sites.

Setting: Five community and six-tertiary care U.S. hospitals.

Synopsis: The GENeralized Early Sepsis Intervention Strategies (GENESIS) project was a CQI initiative that combined several QI concepts with validated early sepsis interventions known as resuscitation bundles (RB). Continuous QI was implemented on patients with severe sepsis or septic shock in both before and after designs (eight hospitals), and in concurrent designs (three hospitals). The control group was comprised of historical controls treated before GENESIS and patients with incomplete implementation of RB, totaling 1,554 patients. The treatment group included patients treated after GENESIS and those with complete RB compliance, totaling 4,801 patients.

Compared with the control group, patients in the treatment group had a 33% decreased risk of in-hospital mortality (RR 0.67, 95% CI 0.63-0.72), an absolute decrease in hospital length of stay (LOS) by 5.1 days (20.7 days vs. 15.6 days, P<0.001) and a $47,923 reduction in total hospital charges (P<0.001). Limitations included the study design (not a prospective randomized trial), and the possibility of other concurrent unmeasured quality initiatives taking place at the study sites, which might have contributed to improved outcomes.

Bottom line: Early sepsis intervention strategies in the form of a comprehensive continuous QI initiative can decrease mortality, hospital LOS, and cost in both tertiary-care and community hospitals.

Citation: Cannon CM, Holthaus CV, Zubrow MT, et al. The GENESIS project (GENeralized Early Sepsis Intervention Strategies): a multicenter quality improvement collaborative. J Intensive Care Med. 2012; Aug 17. doi:10.1177/0885066612453025.

Hypoglycemia Associated with Increased Mortality in the Critically Ill

Clinical question: Is hypoglycemia associated with mortality in critically ill patients?

Background: Initial studies suggested that intensive glucose control reduces mortality in surgical ICU patients and reduces morbidity in medical ICU patients, but further studies have not supported these findings. Recent literature shows conflicting results on the effects of intense glucose control in critically ill patients.

Study design: Post-hoc analysis of the NICE-SUGAR study database.

Setting: ICUs in 42 hospitals in Australia, New Zealand, and Canada.

Synopsis: The NICE-SUGAR study was a multicenter trial that randomized 6,104 ICU patients to intensive (glucose 80 to 108 mg/dL) or conventional glucose control (≤180 mg/dL). Patients were followed for 90 days or until death, with death being the primary end point. Severe hypoglycemia (<40 mg/dL) was recorded in 6.8% of patients in the intensive glucose control group versus 0.5% in the conventional group. The study showed that intensive glucose control was associated with increased mortality among adult ICU patients.

Using the NICE-SUGAR database, the authors conducted a Cox regression analysis to examine the associations between hypoglycemia and death. A total of 2,714 patients had moderate hypoglycemia (glucose 41 to 70 mg/dL), and 223 patients had severe hypoglycemia. The hazard ratio for mortality was 1.41 (95% CI 1.21-1.62, P<0.001) for patients with moderate hypoglycemia and 2.10 (95% CI 1.59-2.77, P<0.001) for severe hypoglycemia, compared to patients without hypoglycemia.

These findings show a strong association between mortality and hypoglycemia but do not prove causality. Hospitalists caring for ICU patients must be aware that hypoglycemia is associated with mortality and focus on avoiding hypoglycemia. The American Diabetes Association currently recommends a target blood glucose level of 140 to 180 mg/dL for most critically ill patients.

Bottom line: Hypoglycemia (glucose <70 mg/dL) is associated with increased risk of mortality in ICU patients.

Citation: Finfer S, Chittock DR, Su SY, et al. Hypoglycemia and risk of death in critically ill patients. New Engl J Med. 2012;367(12):1108-1118.

Increased Bleeding Risk for Cardiac Patients on Multiple Antithrombotic Drugs

Clinical question: Is there an increased risk of bleeding in atrial fibrillation patients treated with multiple antithrombotic agents following acute myocardial infarction (MI) or percutaneous coronary intervention (PCI)?

Background: Current treatment for atrial fibrillation patients with MI or PCI includes vitamin K antagonist (VKA) therapy to prevent stroke and antiplatelet agents to prevent further coronary events. There are inconsistent findings on the safety and efficacy of combined therapy with VKA, aspirin, and clopidogrel, specifically with regard to bleeding risk.

Study design: Retrospective cohort study.

Setting: Nationwide registry in Denmark.

Synopsis: Using the National Patient Registry in Denmark, 11,480 patients with atrial fibrillation who were admitted for MI or PCI were identified. Patients were grouped by medication regimen, including monotherapy (aspirin, clopidogrel, or VKA), dual therapy (dual antiplatelet or VKA+antiplatelet), or triple therapy (VKA+aspirin+clopidogrel). The primary outcome was nonfatal or fatal bleeding within one year.

Patients receiving triple therapy had the highest rate of bleeding, with a crude incidence of 14.2 events per 100 person-years. Patients treated with VKA+aspirin+clopidogel were significantly more likely than those on VKA+antiplatelet (HR 1.47, 95% CI 1.04-2.08) or dual antiplatelet (HR 2.20, 95% CI 1.58-3.08) treatment to have a bleeding event within 90 days, and similar trends were seen at 90 to 360 days. There was no significant difference in thromboembolic events among patients on VKA+aspirin+clopidogrel versus VKA+antiplatelet therapy.

Only bleeding events that required hospitalization were recorded, which might underestimate the bleeding risks of these regimens. Additionally, INR levels were not determined, which could impact both bleeding and thromboembolic outcomes.

However, this study does suggest that there are significant bleeding risks among patients treated with triple therapy. Hospitalists should weigh the risks of thromboembolic events with bleeding risks in patients with atrial fibrillation and MI/PCI, and only prescribe VKA+aspirin+clopidogrel with these risks in mind.

Bottom line: Immediate and continued bleeding risk is increased in patients with atrial fibrillation admitted with PCI or MI who are placed on triple antithrombotic therapy with VKA+aspirin+clopidogrel.

Citation: Lamberts M, Olesen JB, Ruwald MH, et al. Bleeding after initiation of multiple antithrombotic drugs, including triple therapy, in atrial fibrillation patients following myocardial infarction and coronary intervention. Circulation. 2012;126:1185-1193.

Hospitalist-Run Preoperative Clinic Improves Outcomes in Complex Surgical Patients

Clinical question: Do hospitalist-run preoperative clinics improve outcomes for medically complex patients undergoing noncardiac surgery compared to traditional anesthesiologist-run preoperative clinics?

Background: Studies of perioperative medical consultation have shown inconsistent effects on quality of care, but preoperative medical consultation has only been evaluated in the immediate preoperative period (one day prior to surgery or less). Little is known about the impact of involving hospitalists earlier in the preoperative period.

Study design: Retrospective, pre-post study.

Setting: Veterans Affairs Greater Los Angeles Healthcare System (VAGLAHS).

Synopsis: In July 2004, the VAGLAHS Preoperative Clinic transitioned from being anesthesiologist-run to hospitalist-run. Mid-level providers were trained on preoperative medical assessment, and patients were only evaluated by anesthesia staff on the day of surgery, after they had been deemed medically acceptable for surgery. All patients seen in the clinic from July 2003 to July 2005 were included in the study. Period A included patients evaluated when anesthesia staff supervised the clinic. Period B included patients evaluated during the first year of the hospitalist-run system.

There were 1,101 patients with inpatient surgeries in Period A, and 1,126 patients in Period B. Mean length of stay (LOS) decreased to 5.28 days during the hospitalist-run model from 9.87 days during the anesthesiologist-run model. LOS reductions were most notable in patients with ASA scores of 3 or higher; LOS reductions were not seen in an internal control of surgical patients who were not evaluated in the preoperative clinic. Inpatient mortality was also reduced in Period B compared with Period A, to 4 cases (0.36%) from 14 cases (1.27%) (P=0.0158). Specific processes that led to improved outcomes for patients in Period B could not be identified. The VA study setting might limit the generalizability of the results.

Bottom line: A hospitalist-run preoperative clinic was associated with decreased LOS and inpatient mortality compared with a traditional anesthesiologist-run clinic.

Citation: Vazirani S, Lankarani-Fard A, Lian LJ, Stelzner M, Asch SM. Preoperative processes and outcomes after implementation of a hospitalist-run preoperative clinic. J Hosp Med. 2012 Sep 7. doi:10.1002/jhm.1968.

Delirium after Cardiac Surgery Associated with Prolonged Cognitive Impairment

Clinical question: Is postoperative delirium associated with decreased cognitive function in the first year after cardiac surgery?

Background: In general populations, delirium has been associated with long-term decline in cognitive ability. Delirium and cognitive dysfunction are both common following cardiac surgery, but the effects of postoperative delirium on the trajectory of cognitive function over time is unclear.

Study design: Prospective cohort study.

Setting: Two academic medical centers and a Veterans Administration hospital.

Synopsis: Two hundred twenty-five patients aged 60 or older who were scheduled to undergo coronary artery bypass grafting or valve replacement surgery were included. Patients underwent preoperative assessment cognitive function with the use of the Mini-Mental State Examination (MMSE). Starting on postoperative Day 2, patients underwent daily assessment for delirium. After discharge, cognitive function was reassessed at months one, six, and 12.

Postoperative delirium occurred in 103 patients (46%). Patients with delirium were older, had higher comorbidity scores, and had lower MMSE scores at baseline. Among the overall study population, adjusted MMSE scores dropped 4.6 points from baseline to postoperative Day 2, then were observed to increase by approximately one point per day during postoperative days 3 to 5 with minimal change thereafter. Patients with delirium had greater decrease in cognitive function in the immediate postoperative period compared to patients without delirium (7.7 points vs. 2.1, P<0.001).

Patients without delirium returned to their baseline cognitive ability by one month postoperatively, while patients who had delirium were still making gains up to six months post-operatively, never returning to baseline level of function by one year. Unmeasured confounders and uncertain sensitivity of the MMSE to detect mild cognitive impairment might limit these findings.

Bottom line: Cognitive function decreases in the immediate postoperative period following cardiac surgery. Compared to patients without delirium, patients with delirium experience more dramatic and prolonged cognitive impairment postoperatively, without returning to their preoperative level of cognitive function at one year.

Citation: Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories after postoperative delirium. New Engl J Med. 2012;367:30-39.

Benefits of Resuming Anticoagulation after GI Bleed Outweigh Risks for Most Patients

Clinical question: In warfarin-treated patients who have experienced gastrointestinal (GI) bleeding, what are the patterns of restarting warfarin therapy and the incidence of thrombosis, recurrent GI bleed, and death in the 90 days following index bleed?

Background: In warfarin-treated patients who experience GI bleeding, warfarin is often temporarily held or permanently discontinued, placing patients at increased risk for developing thromboembolism. Little is known about the risks, benefits, and timing of restarting warfarin in this patient population.

Study design: Retrospective cohort study.

Setting: Kaiser Permanente Colorado.

Synopsis: Using clinical and administrative databases, 442 patients who presented with GI bleeding while receiving warfarin therapy were identified. Patients were grouped by whether they resumed warfarin (n=260, including 41 patients in whom anticoagulation was never interrupted), or did not resume warfarin therapy (n=182) in the 90 days following index GI bleed. Patients with prosthetic heart valves or GI bleeding localized to the rectum/anus were more commonly restarted on warfarin, whereas older patients and those in whom the source of bleeding was not identified were less likely to be restarted on warfarin therapy.

Restarting warfarin therapy after index GI bleed was associated with lower risk of thrombosis (HR 0.05, 95% CI 0.01-0.58) and death from any cause (HR 0.31, 95% CI 0.15-0.62), and it was not associated with a significant increase in risk for recurrent GI bleed (HR 1.32, 95% CI 0.50-3.57).

The authors concluded that for many patients who experience a warfarin-associated GI bleed, the benefits of restarting warfarin therapy outweigh the risks. No conclusions were made regarding the optimal timing of resuming therapy. Limitations included the use of administrative data and inability to determine the potential influence of aspirin use on outcomes.

Bottom line: Resuming warfarin in the 90 days following a warfarin-associated GI bleed is associated with decreased risk of thrombosis and death without increased risk for recurrent GI bleed.

Citation: Witt DM, Delate T, Garcia DA, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding. Arch Intern Med. 2012 Sep 12. doi:10.1001/archinternmed.2012.4261.

Preoperative Hyponatremia Associated with Increased Risk for Perioperative Complications and Mortality

Clinical question: Is preoperative hyponatremia an indicator of perioperative morbidity and mortality?

Background: Hyponatremia is a common diagnosis in the hospital setting and is associated with adverse outcomes, even in mild cases. However, it is unclear if this association exists in surgical patients when detected preoperatively.

Study design: Retrospective cohort study.

Setting: Academic and community hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP).

Synopsis: A total of 75,423 adult patients with hyponatremia (sodium <135 mEq/L) who were undergoing major surgery were compared to 888,840 patients with normal preoperative sodium levels over a six-year period. The primary outcome was 30-day mortality. Secondary outcomes included postoperative major coronary events, stroke, wound infection, pneumonia, and length of stay (LOS).

Compared to patients with normal sodium levels, those with preoperative hyponatremia had higher rates of perioperative mortality (5.2% vs. 1.3%; adjusted odds ratio 1.44, 95% CI 1.38-1.50), with increased risk that correlated with increasing severity of hyponatremia. Association with postoperative mortality was particularly strong among hyponatremic patients with ASA scores of 1 or 2 and those undergoing nonemergency surgery.

Patients with preoperative hyponatremia were also found to have increased risk for all postoperative complications evaluated, with the exception of stroke. Limitations included the potential for unmeasured confounders and not being able to account for the role of medications used perioperatively. Research is needed to determine whether correcting preoperative hyponatremia lessens the risk of mortality and other postoperative complications.

Bottom line: Among patients undergoing major surgery, preoperative hyponatremia is a predictor of postoperative 30-day mortality and morbidity.

Citation: Leung AA, McAlister FA, Rogers SO, et al. Preoperative hyponatremia and perioperative complications. Arch Intern Med. 2012;172:1-8.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Interventions that improve discharge handovers reviewed
2. Duration of in-hospital cardiac resuscitation and survival rates
3. Early sepsis intervention strategies to decrease mortality risk
4. Hypoglycemia linked to increased mortality in critically ill
5. Increased bleeding risk for cardiac patients
6. Hospital-run vs. anesthesiologist-run preoperative clinics
7. Postoperative delirium and cognitive impairment in cardiac patients
8. Benefits of resuming anticoagulants after GI bleeding
9. Preoperative hyponatremia and risk of perioperative mortality

Systematic Review Highlights Several Interventions That Improve Discharge Handovers

Clinical question: Do interventions to improve patient handovers at discharge have positive effects on patient care?

Background: The transition from hospital to primary care is often suboptimal and has been associated with unfavorable outcomes, including hospital readmission, increased healthcare utilization, and adverse drug events post-discharge. This review sought to characterize different types of interventions aimed at improving discharge handovers and to evaluate their effects.

Study design: Systematic review of randomized controlled trials.

Setting: Studies published from January 1990 to March 2011.

Synopsis: Review of published databases identified 36 randomized controlled studies on interventions to improve discharge handovers. Studies were blindly evaluated by two reviewers on quality, interventions, and outcomes. There was significant heterogeneity in interventions and outcomes; thus, statistical analysis was not possible. Most studies evaluated multicomponent interventions and had more than one outcome measure.

Of the 36 studies reviewed, 25 reported statistically significant improvements in outcomes, including reduced hospital utilization and improved continuity of care. Effective interventions included medication reconciliation; structured discharge information (facilitated by electronic resources); multidisciplinary discharge planning; shared involvement in arranging care between inpatient and outpatient physicians; and Web-based access to discharge information by the outpatient provider.

The complexity of the interventions and the heterogeneity of reported results did not allow for firm conclusions to be drawn regarding which specific interventions had the strongest effects.

Bottom line: Interventions that target the quality and safety of handovers between hospital and outpatient providers at discharge can significantly reduce hospital utilization and improve continuity of care.

Citation: Hesselink G, Schoonhoven L, Barach P, et al. Improving patient handovers from hospital to primary care: a systematic review. Ann Intern Med. 2012;157:417-428.

Longer Duration of In-Hospital Cardiac Resuscitation Associated with Increased Survival

Clinical question: Are prolonged cardiac resuscitation efforts associated with improved outcomes?

Background: There is little evidence or guidelines on how long to maintain resuscitative efforts during in-hospital cardiac arrest, leading to variation in practice. This study characterized patterns of resuscitation duration and relationship to the return of spontaneous circulation and survival to discharge.

Study design: Retrospective observational study.

Setting: Four hundred thirty-five U.S. hospitals reporting data to the American Heart Association’s Get With The Guidelines: Resuscitation registry.

Synopsis: Using duration of resuscitation in nonsurvivors as a surrogate for the tendency of a facility to perform prolonged efforts, hospitals were divided into quartiles. Overall, of 64,339 patients in the registry, 31,198 (48.5%) had return of circulation and 9,912 (15.4%) survived to discharge. Resuscitative efforts in nonsurvivors ranged from a median of 16 to 25 minutes between the lowest and highest quartiles.

There was a stepwise increase in the likelihood for patients to have return of spontaneous circulation and to survive to discharge between each quartile. Specifically, comparing shortest to longest, there was a significant adjusted odds ratio of 1.12 for both the return of circulation (P<0.0001) and survival to discharge (P=0.021). The survival benefit was most apparent for those with pulseless electrical activity or asystole as initial rhythms, as compared to ventricular tachycardia or fibrillation.

Limitations included the study’s observational design, which meant causality could not be determined. Additionally, the study did not account for the quality of resuscitative efforts (e.g. depth of chest compressions, adherence with guidelines), which might have influenced outcomes. Importantly, the study looked at survival to discharge, but it did not evaluate long-term survival or functional status post-discharge, which might better reflect the success of resuscitation.

Bottom line: Prolonged resuscitative efforts were observed to be associated with increased likelihood of return of spontaneous circulation and survival to discharge; this data might provide clinical guidance in determining when to stop resuscitation efforts.

Citation: Goldberger ZD, Chan PS, Berg RA, et al. Duration of resuscitation efforts and survival after in-hospital cardiac arrest: an observational study. Lancet. 2012;380:1473-1481.

Early Sepsis Intervention Strategies Decrease Mortality, Length of Stay, and Cost

Clinical question: For patients with sepsis or septic shock, what is the impact of a real-time comprehensive continuous quality-improvement (QI) initiative on in-hospital mortality, morbidity, and healthcare resource utilization in community and tertiary-care hospitals in the U.S.?

Background: Multiple single-center trials have demonstrated that early sepsis intervention strategies (early goal-directed therapy, resuscitation bundles) improve in-hospital mortality. Little is known about the effectiveness of incorporating these strategies into a real-time continuous QI initiative and implementing interventions across multiple sites simultaneously.

Study design: Pre-post at some sites and concurrent implementation design at other sites.

Setting: Five community and six-tertiary care U.S. hospitals.

Synopsis: The GENeralized Early Sepsis Intervention Strategies (GENESIS) project was a CQI initiative that combined several QI concepts with validated early sepsis interventions known as resuscitation bundles (RB). Continuous QI was implemented on patients with severe sepsis or septic shock in both before and after designs (eight hospitals), and in concurrent designs (three hospitals). The control group was comprised of historical controls treated before GENESIS and patients with incomplete implementation of RB, totaling 1,554 patients. The treatment group included patients treated after GENESIS and those with complete RB compliance, totaling 4,801 patients.

Compared with the control group, patients in the treatment group had a 33% decreased risk of in-hospital mortality (RR 0.67, 95% CI 0.63-0.72), an absolute decrease in hospital length of stay (LOS) by 5.1 days (20.7 days vs. 15.6 days, P<0.001) and a $47,923 reduction in total hospital charges (P<0.001). Limitations included the study design (not a prospective randomized trial), and the possibility of other concurrent unmeasured quality initiatives taking place at the study sites, which might have contributed to improved outcomes.

Bottom line: Early sepsis intervention strategies in the form of a comprehensive continuous QI initiative can decrease mortality, hospital LOS, and cost in both tertiary-care and community hospitals.

Citation: Cannon CM, Holthaus CV, Zubrow MT, et al. The GENESIS project (GENeralized Early Sepsis Intervention Strategies): a multicenter quality improvement collaborative. J Intensive Care Med. 2012; Aug 17. doi:10.1177/0885066612453025.

Hypoglycemia Associated with Increased Mortality in the Critically Ill

Clinical question: Is hypoglycemia associated with mortality in critically ill patients?

Background: Initial studies suggested that intensive glucose control reduces mortality in surgical ICU patients and reduces morbidity in medical ICU patients, but further studies have not supported these findings. Recent literature shows conflicting results on the effects of intense glucose control in critically ill patients.

Study design: Post-hoc analysis of the NICE-SUGAR study database.

Setting: ICUs in 42 hospitals in Australia, New Zealand, and Canada.

Synopsis: The NICE-SUGAR study was a multicenter trial that randomized 6,104 ICU patients to intensive (glucose 80 to 108 mg/dL) or conventional glucose control (≤180 mg/dL). Patients were followed for 90 days or until death, with death being the primary end point. Severe hypoglycemia (<40 mg/dL) was recorded in 6.8% of patients in the intensive glucose control group versus 0.5% in the conventional group. The study showed that intensive glucose control was associated with increased mortality among adult ICU patients.

Using the NICE-SUGAR database, the authors conducted a Cox regression analysis to examine the associations between hypoglycemia and death. A total of 2,714 patients had moderate hypoglycemia (glucose 41 to 70 mg/dL), and 223 patients had severe hypoglycemia. The hazard ratio for mortality was 1.41 (95% CI 1.21-1.62, P<0.001) for patients with moderate hypoglycemia and 2.10 (95% CI 1.59-2.77, P<0.001) for severe hypoglycemia, compared to patients without hypoglycemia.

These findings show a strong association between mortality and hypoglycemia but do not prove causality. Hospitalists caring for ICU patients must be aware that hypoglycemia is associated with mortality and focus on avoiding hypoglycemia. The American Diabetes Association currently recommends a target blood glucose level of 140 to 180 mg/dL for most critically ill patients.

Bottom line: Hypoglycemia (glucose <70 mg/dL) is associated with increased risk of mortality in ICU patients.

Citation: Finfer S, Chittock DR, Su SY, et al. Hypoglycemia and risk of death in critically ill patients. New Engl J Med. 2012;367(12):1108-1118.

Increased Bleeding Risk for Cardiac Patients on Multiple Antithrombotic Drugs

Clinical question: Is there an increased risk of bleeding in atrial fibrillation patients treated with multiple antithrombotic agents following acute myocardial infarction (MI) or percutaneous coronary intervention (PCI)?

Background: Current treatment for atrial fibrillation patients with MI or PCI includes vitamin K antagonist (VKA) therapy to prevent stroke and antiplatelet agents to prevent further coronary events. There are inconsistent findings on the safety and efficacy of combined therapy with VKA, aspirin, and clopidogrel, specifically with regard to bleeding risk.

Study design: Retrospective cohort study.

Setting: Nationwide registry in Denmark.

Synopsis: Using the National Patient Registry in Denmark, 11,480 patients with atrial fibrillation who were admitted for MI or PCI were identified. Patients were grouped by medication regimen, including monotherapy (aspirin, clopidogrel, or VKA), dual therapy (dual antiplatelet or VKA+antiplatelet), or triple therapy (VKA+aspirin+clopidogrel). The primary outcome was nonfatal or fatal bleeding within one year.

Patients receiving triple therapy had the highest rate of bleeding, with a crude incidence of 14.2 events per 100 person-years. Patients treated with VKA+aspirin+clopidogel were significantly more likely than those on VKA+antiplatelet (HR 1.47, 95% CI 1.04-2.08) or dual antiplatelet (HR 2.20, 95% CI 1.58-3.08) treatment to have a bleeding event within 90 days, and similar trends were seen at 90 to 360 days. There was no significant difference in thromboembolic events among patients on VKA+aspirin+clopidogrel versus VKA+antiplatelet therapy.

Only bleeding events that required hospitalization were recorded, which might underestimate the bleeding risks of these regimens. Additionally, INR levels were not determined, which could impact both bleeding and thromboembolic outcomes.

However, this study does suggest that there are significant bleeding risks among patients treated with triple therapy. Hospitalists should weigh the risks of thromboembolic events with bleeding risks in patients with atrial fibrillation and MI/PCI, and only prescribe VKA+aspirin+clopidogrel with these risks in mind.

Bottom line: Immediate and continued bleeding risk is increased in patients with atrial fibrillation admitted with PCI or MI who are placed on triple antithrombotic therapy with VKA+aspirin+clopidogrel.

Citation: Lamberts M, Olesen JB, Ruwald MH, et al. Bleeding after initiation of multiple antithrombotic drugs, including triple therapy, in atrial fibrillation patients following myocardial infarction and coronary intervention. Circulation. 2012;126:1185-1193.

Hospitalist-Run Preoperative Clinic Improves Outcomes in Complex Surgical Patients

Clinical question: Do hospitalist-run preoperative clinics improve outcomes for medically complex patients undergoing noncardiac surgery compared to traditional anesthesiologist-run preoperative clinics?

Background: Studies of perioperative medical consultation have shown inconsistent effects on quality of care, but preoperative medical consultation has only been evaluated in the immediate preoperative period (one day prior to surgery or less). Little is known about the impact of involving hospitalists earlier in the preoperative period.

Study design: Retrospective, pre-post study.

Setting: Veterans Affairs Greater Los Angeles Healthcare System (VAGLAHS).

Synopsis: In July 2004, the VAGLAHS Preoperative Clinic transitioned from being anesthesiologist-run to hospitalist-run. Mid-level providers were trained on preoperative medical assessment, and patients were only evaluated by anesthesia staff on the day of surgery, after they had been deemed medically acceptable for surgery. All patients seen in the clinic from July 2003 to July 2005 were included in the study. Period A included patients evaluated when anesthesia staff supervised the clinic. Period B included patients evaluated during the first year of the hospitalist-run system.

There were 1,101 patients with inpatient surgeries in Period A, and 1,126 patients in Period B. Mean length of stay (LOS) decreased to 5.28 days during the hospitalist-run model from 9.87 days during the anesthesiologist-run model. LOS reductions were most notable in patients with ASA scores of 3 or higher; LOS reductions were not seen in an internal control of surgical patients who were not evaluated in the preoperative clinic. Inpatient mortality was also reduced in Period B compared with Period A, to 4 cases (0.36%) from 14 cases (1.27%) (P=0.0158). Specific processes that led to improved outcomes for patients in Period B could not be identified. The VA study setting might limit the generalizability of the results.

Bottom line: A hospitalist-run preoperative clinic was associated with decreased LOS and inpatient mortality compared with a traditional anesthesiologist-run clinic.

Citation: Vazirani S, Lankarani-Fard A, Lian LJ, Stelzner M, Asch SM. Preoperative processes and outcomes after implementation of a hospitalist-run preoperative clinic. J Hosp Med. 2012 Sep 7. doi:10.1002/jhm.1968.

Delirium after Cardiac Surgery Associated with Prolonged Cognitive Impairment

Clinical question: Is postoperative delirium associated with decreased cognitive function in the first year after cardiac surgery?

Background: In general populations, delirium has been associated with long-term decline in cognitive ability. Delirium and cognitive dysfunction are both common following cardiac surgery, but the effects of postoperative delirium on the trajectory of cognitive function over time is unclear.

Study design: Prospective cohort study.

Setting: Two academic medical centers and a Veterans Administration hospital.

Synopsis: Two hundred twenty-five patients aged 60 or older who were scheduled to undergo coronary artery bypass grafting or valve replacement surgery were included. Patients underwent preoperative assessment cognitive function with the use of the Mini-Mental State Examination (MMSE). Starting on postoperative Day 2, patients underwent daily assessment for delirium. After discharge, cognitive function was reassessed at months one, six, and 12.

Postoperative delirium occurred in 103 patients (46%). Patients with delirium were older, had higher comorbidity scores, and had lower MMSE scores at baseline. Among the overall study population, adjusted MMSE scores dropped 4.6 points from baseline to postoperative Day 2, then were observed to increase by approximately one point per day during postoperative days 3 to 5 with minimal change thereafter. Patients with delirium had greater decrease in cognitive function in the immediate postoperative period compared to patients without delirium (7.7 points vs. 2.1, P<0.001).

Patients without delirium returned to their baseline cognitive ability by one month postoperatively, while patients who had delirium were still making gains up to six months post-operatively, never returning to baseline level of function by one year. Unmeasured confounders and uncertain sensitivity of the MMSE to detect mild cognitive impairment might limit these findings.

Bottom line: Cognitive function decreases in the immediate postoperative period following cardiac surgery. Compared to patients without delirium, patients with delirium experience more dramatic and prolonged cognitive impairment postoperatively, without returning to their preoperative level of cognitive function at one year.

Citation: Saczynski JS, Marcantonio ER, Quach L, et al. Cognitive trajectories after postoperative delirium. New Engl J Med. 2012;367:30-39.

Benefits of Resuming Anticoagulation after GI Bleed Outweigh Risks for Most Patients

Clinical question: In warfarin-treated patients who have experienced gastrointestinal (GI) bleeding, what are the patterns of restarting warfarin therapy and the incidence of thrombosis, recurrent GI bleed, and death in the 90 days following index bleed?

Background: In warfarin-treated patients who experience GI bleeding, warfarin is often temporarily held or permanently discontinued, placing patients at increased risk for developing thromboembolism. Little is known about the risks, benefits, and timing of restarting warfarin in this patient population.

Study design: Retrospective cohort study.

Setting: Kaiser Permanente Colorado.

Synopsis: Using clinical and administrative databases, 442 patients who presented with GI bleeding while receiving warfarin therapy were identified. Patients were grouped by whether they resumed warfarin (n=260, including 41 patients in whom anticoagulation was never interrupted), or did not resume warfarin therapy (n=182) in the 90 days following index GI bleed. Patients with prosthetic heart valves or GI bleeding localized to the rectum/anus were more commonly restarted on warfarin, whereas older patients and those in whom the source of bleeding was not identified were less likely to be restarted on warfarin therapy.

Restarting warfarin therapy after index GI bleed was associated with lower risk of thrombosis (HR 0.05, 95% CI 0.01-0.58) and death from any cause (HR 0.31, 95% CI 0.15-0.62), and it was not associated with a significant increase in risk for recurrent GI bleed (HR 1.32, 95% CI 0.50-3.57).

The authors concluded that for many patients who experience a warfarin-associated GI bleed, the benefits of restarting warfarin therapy outweigh the risks. No conclusions were made regarding the optimal timing of resuming therapy. Limitations included the use of administrative data and inability to determine the potential influence of aspirin use on outcomes.

Bottom line: Resuming warfarin in the 90 days following a warfarin-associated GI bleed is associated with decreased risk of thrombosis and death without increased risk for recurrent GI bleed.

Citation: Witt DM, Delate T, Garcia DA, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding. Arch Intern Med. 2012 Sep 12. doi:10.1001/archinternmed.2012.4261.

Preoperative Hyponatremia Associated with Increased Risk for Perioperative Complications and Mortality

Clinical question: Is preoperative hyponatremia an indicator of perioperative morbidity and mortality?

Background: Hyponatremia is a common diagnosis in the hospital setting and is associated with adverse outcomes, even in mild cases. However, it is unclear if this association exists in surgical patients when detected preoperatively.

Study design: Retrospective cohort study.

Setting: Academic and community hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program (NSQIP).

Synopsis: A total of 75,423 adult patients with hyponatremia (sodium <135 mEq/L) who were undergoing major surgery were compared to 888,840 patients with normal preoperative sodium levels over a six-year period. The primary outcome was 30-day mortality. Secondary outcomes included postoperative major coronary events, stroke, wound infection, pneumonia, and length of stay (LOS).

Compared to patients with normal sodium levels, those with preoperative hyponatremia had higher rates of perioperative mortality (5.2% vs. 1.3%; adjusted odds ratio 1.44, 95% CI 1.38-1.50), with increased risk that correlated with increasing severity of hyponatremia. Association with postoperative mortality was particularly strong among hyponatremic patients with ASA scores of 1 or 2 and those undergoing nonemergency surgery.

Patients with preoperative hyponatremia were also found to have increased risk for all postoperative complications evaluated, with the exception of stroke. Limitations included the potential for unmeasured confounders and not being able to account for the role of medications used perioperatively. Research is needed to determine whether correcting preoperative hyponatremia lessens the risk of mortality and other postoperative complications.

Bottom line: Among patients undergoing major surgery, preoperative hyponatremia is a predictor of postoperative 30-day mortality and morbidity.

Citation: Leung AA, McAlister FA, Rogers SO, et al. Preoperative hyponatremia and perioperative complications. Arch Intern Med. 2012;172:1-8.