Shobha W. Stack, MD, PhD

Clinical question: In patients with atrial fibrillation (AF) or flutter, is heparin bridging needed during interruption of warfarin therapy for surgery or invasive procedures?

Background: Bridging is intended to decrease the risk of stroke or other arterial thromboembolism by minimizing time off anticoagulation. Bridging may increase the risk of serious bleeding, offsetting any benefit. Guidelines have provided weak and inconsistent recommendations due to a lack of randomized trials.

Study design: Randomized, double blind, placebo-controlled trial.

Setting: More than 100 centers in the U.S. and Canada, from 2009-2014.

Synopsis: Investigators randomized 1,884 patients on warfarin with a CHADS² risk factor of one or higher undergoing elective surgery or procedure to dalteparin or placebo, from three days to 24 hours before the procedure and for five to 10 days after. Mean CHADS² score was 2.3; 3% of patients had scores of five to six. Approximately one-third of patients were on aspirin, and most procedures (89%) were classified as minor. Patients with mechanical heart valves, stroke/transient ischemic attack (TIA)/systemic embolization within 12 weeks, major bleeding within six weeks, renal insufficiency, thrombocytopenia or planned cardiac, intracranial, or intraspinal surgery were excluded.

Thirty-day incidence of arterial thromboembolism (stroke, TIA, systemic embolization) was 0.4% in the non-bridging group and 0.3% in the bridging group (P=0.01 for noninferiority). Patients suffering arterial thromboembolism had mean CHADS² scores of 2.6; most events occurred after minor procedures. Major bleeding was less common with no bridge (1.3% vs. 3.2%, relative risk 0.41, P=0.005 for superiority).

In this trial, most patients underwent minor procedures and few CHADS² 5-6 patients were enrolled; however, this well-designed, randomized trial adds important evidence to existing observational data.

Bottom line: Bridging is not warranted for most AF patients with CHADS2 scores of four or lower, at least for low-risk procedures.

Citation: Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373(9):823-833.

Clinical question: In patients with atrial fibrillation (AF) or flutter, is heparin bridging needed during interruption of warfarin therapy for surgery or invasive procedures?

Background: Bridging is intended to decrease the risk of stroke or other arterial thromboembolism by minimizing time off anticoagulation. Bridging may increase the risk of serious bleeding, offsetting any benefit. Guidelines have provided weak and inconsistent recommendations due to a lack of randomized trials.

Study design: Randomized, double blind, placebo-controlled trial.

Setting: More than 100 centers in the U.S. and Canada, from 2009-2014.

Synopsis: Investigators randomized 1,884 patients on warfarin with a CHADS² risk factor of one or higher undergoing elective surgery or procedure to dalteparin or placebo, from three days to 24 hours before the procedure and for five to 10 days after. Mean CHADS² score was 2.3; 3% of patients had scores of five to six. Approximately one-third of patients were on aspirin, and most procedures (89%) were classified as minor. Patients with mechanical heart valves, stroke/transient ischemic attack (TIA)/systemic embolization within 12 weeks, major bleeding within six weeks, renal insufficiency, thrombocytopenia or planned cardiac, intracranial, or intraspinal surgery were excluded.

Thirty-day incidence of arterial thromboembolism (stroke, TIA, systemic embolization) was 0.4% in the non-bridging group and 0.3% in the bridging group (P=0.01 for noninferiority). Patients suffering arterial thromboembolism had mean CHADS² scores of 2.6; most events occurred after minor procedures. Major bleeding was less common with no bridge (1.3% vs. 3.2%, relative risk 0.41, P=0.005 for superiority).

In this trial, most patients underwent minor procedures and few CHADS² 5-6 patients were enrolled; however, this well-designed, randomized trial adds important evidence to existing observational data.

Bottom line: Bridging is not warranted for most AF patients with CHADS2 scores of four or lower, at least for low-risk procedures.

Citation: Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373(9):823-833.

Clinical question: In patients with atrial fibrillation (AF) or flutter, is heparin bridging needed during interruption of warfarin therapy for surgery or invasive procedures?

Background: Bridging is intended to decrease the risk of stroke or other arterial thromboembolism by minimizing time off anticoagulation. Bridging may increase the risk of serious bleeding, offsetting any benefit. Guidelines have provided weak and inconsistent recommendations due to a lack of randomized trials.

Study design: Randomized, double blind, placebo-controlled trial.

Setting: More than 100 centers in the U.S. and Canada, from 2009-2014.

Synopsis: Investigators randomized 1,884 patients on warfarin with a CHADS² risk factor of one or higher undergoing elective surgery or procedure to dalteparin or placebo, from three days to 24 hours before the procedure and for five to 10 days after. Mean CHADS² score was 2.3; 3% of patients had scores of five to six. Approximately one-third of patients were on aspirin, and most procedures (89%) were classified as minor. Patients with mechanical heart valves, stroke/transient ischemic attack (TIA)/systemic embolization within 12 weeks, major bleeding within six weeks, renal insufficiency, thrombocytopenia or planned cardiac, intracranial, or intraspinal surgery were excluded.

Thirty-day incidence of arterial thromboembolism (stroke, TIA, systemic embolization) was 0.4% in the non-bridging group and 0.3% in the bridging group (P=0.01 for noninferiority). Patients suffering arterial thromboembolism had mean CHADS² scores of 2.6; most events occurred after minor procedures. Major bleeding was less common with no bridge (1.3% vs. 3.2%, relative risk 0.41, P=0.005 for superiority).

In this trial, most patients underwent minor procedures and few CHADS² 5-6 patients were enrolled; however, this well-designed, randomized trial adds important evidence to existing observational data.

Bottom line: Bridging is not warranted for most AF patients with CHADS2 scores of four or lower, at least for low-risk procedures.

Citation: Douketis JD, Spyropoulos AC, Kaatz S, et al. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373(9):823-833.

Clinical question: Does a postural modification to the Valsalva maneuver improve its effectiveness?

Background: The Valsalva maneuver, often used to treat supraventricular tachycardia, is rarely successful. A modification to the maneuver to increase relaxation phase venous return and vagal stimulation could improve its efficacy.

Study design: Multicenter, randomized controlled trial (RCT).

Setting: Ten emergency departments in England.

Synopsis: Four hundred thirty-three patients with stable supraventricular tachycardia (excluding atrial fibrillation or flutter) were randomized to use the Valsalva maneuver (control) or modified Valsalva maneuver (intervention). In the control group, strain was standardized using a manometer (40 mm Hg for 15 seconds). In the intervention group, patients underwent the same maneuver, followed by lying supine with passive leg raise to 45 degrees for 15 seconds. Participants could repeat the maneuver if it was initially unsuccessful. Randomization was stratified by center.

Using an intention-to-treat analysis, 43% of the intervention group achieved the primary outcome of sinus rhythm one minute after straining, compared with 17% of the control group (P<0.0001). The intervention group was less likely to receive adenosine (50% vs. 69%, P=0.0002) or any emergency, anti-arrhythmic treatment (80% vs. 57%, P<0.0001).

No significant differences were seen in hospital admissions, length of ED stay, or adverse events between groups.

Bottom line: In patients with stable supraventricular tachycardia, modifying the Valsalva maneuver is significantly more effective in restoring sinus rhythm.

Citation: Appelboam A, Reuben A, Mann C, et al. Postural modification to the standard Valsalva manoeuvre for emergency treatment of supraventricular tachycardias (REVERT): a randomised controlled trial [published online ahead of print August 24, 2015]. Lancet. doi: 10.1016/S0140-6736(15)61485-4.

Clinical question: Does a postural modification to the Valsalva maneuver improve its effectiveness?

Background: The Valsalva maneuver, often used to treat supraventricular tachycardia, is rarely successful. A modification to the maneuver to increase relaxation phase venous return and vagal stimulation could improve its efficacy.

Study design: Multicenter, randomized controlled trial (RCT).

Setting: Ten emergency departments in England.

Synopsis: Four hundred thirty-three patients with stable supraventricular tachycardia (excluding atrial fibrillation or flutter) were randomized to use the Valsalva maneuver (control) or modified Valsalva maneuver (intervention). In the control group, strain was standardized using a manometer (40 mm Hg for 15 seconds). In the intervention group, patients underwent the same maneuver, followed by lying supine with passive leg raise to 45 degrees for 15 seconds. Participants could repeat the maneuver if it was initially unsuccessful. Randomization was stratified by center.

Using an intention-to-treat analysis, 43% of the intervention group achieved the primary outcome of sinus rhythm one minute after straining, compared with 17% of the control group (P<0.0001). The intervention group was less likely to receive adenosine (50% vs. 69%, P=0.0002) or any emergency, anti-arrhythmic treatment (80% vs. 57%, P<0.0001).

No significant differences were seen in hospital admissions, length of ED stay, or adverse events between groups.

Bottom line: In patients with stable supraventricular tachycardia, modifying the Valsalva maneuver is significantly more effective in restoring sinus rhythm.

Citation: Appelboam A, Reuben A, Mann C, et al. Postural modification to the standard Valsalva manoeuvre for emergency treatment of supraventricular tachycardias (REVERT): a randomised controlled trial [published online ahead of print August 24, 2015]. Lancet. doi: 10.1016/S0140-6736(15)61485-4.

Clinical question: Does a postural modification to the Valsalva maneuver improve its effectiveness?

Background: The Valsalva maneuver, often used to treat supraventricular tachycardia, is rarely successful. A modification to the maneuver to increase relaxation phase venous return and vagal stimulation could improve its efficacy.

Study design: Multicenter, randomized controlled trial (RCT).

Setting: Ten emergency departments in England.

Synopsis: Four hundred thirty-three patients with stable supraventricular tachycardia (excluding atrial fibrillation or flutter) were randomized to use the Valsalva maneuver (control) or modified Valsalva maneuver (intervention). In the control group, strain was standardized using a manometer (40 mm Hg for 15 seconds). In the intervention group, patients underwent the same maneuver, followed by lying supine with passive leg raise to 45 degrees for 15 seconds. Participants could repeat the maneuver if it was initially unsuccessful. Randomization was stratified by center.

Using an intention-to-treat analysis, 43% of the intervention group achieved the primary outcome of sinus rhythm one minute after straining, compared with 17% of the control group (P<0.0001). The intervention group was less likely to receive adenosine (50% vs. 69%, P=0.0002) or any emergency, anti-arrhythmic treatment (80% vs. 57%, P<0.0001).

No significant differences were seen in hospital admissions, length of ED stay, or adverse events between groups.

Bottom line: In patients with stable supraventricular tachycardia, modifying the Valsalva maneuver is significantly more effective in restoring sinus rhythm.

Citation: Appelboam A, Reuben A, Mann C, et al. Postural modification to the standard Valsalva manoeuvre for emergency treatment of supraventricular tachycardias (REVERT): a randomised controlled trial [published online ahead of print August 24, 2015]. Lancet. doi: 10.1016/S0140-6736(15)61485-4.

Clinical question: For patients with heart failure (HF), with and without concurrent Afib (AF), does the CHA²DS²-VASc score predict ischemic stroke, thromboembolism, and death?

Background: Factors in the CHA²DS²-VASc score predict increased risk of stroke, thromboembolism, and death, regardless of whether AF is present. It is unknown if this score can identify subgroups of patients with HF, with and without AF, at particularly high or low risk of these events.

Study design: Prospective, cohort study.

Setting: Three Danish registries, 2000-2012.

Synopsis: Among 42,987 patients 50 years and older with incident HF not on anticoagulation, the absolute risk of stroke among patients without AF was 1.5% per year or higher with a CHA²DS²-VASc score of two or higher. The absolute risk of stroke was 4% or higher at five years. Risks were higher in the 21.9% of patients with AF. The CHA²DS²-VASc score modestly predicted endpoints and had an approximately 90% negative predictive value for stroke, thromboembolism, and death at one-year follow-up, regardless of whether or not AF was present.

In this large study, HF patients in a non-diverse population were studied, and some patients may have had undiagnosed AF. Functional status and ejection fraction in patients with HF could not be categorized; however, reported five-year results may be generalizable to patients with chronic HF. Select patients with HF without AF, who have two or more factors of the score besides HF, might benefit from anticoagulation.

Bottom line: The CHA²DS²-VASc score modestly predicts stroke, thromboembolism, and death among patients with HF, but further studies are needed to determine its clinical usefulness.

Citation: Melgaard L, Gorst-Rasmussen A, Lane DA, Rasmussen LH, Larsen TB, Lip GY. Assessment of the CHA²DS²-VASc Score in predicting ischemic stroke, thromboembolism, and death in patients with heart failure with and without atrial fibrillation. JAMA. 2015;314(10):1030-1038.

Clinical question: For patients with heart failure (HF), with and without concurrent Afib (AF), does the CHA²DS²-VASc score predict ischemic stroke, thromboembolism, and death?

Background: Factors in the CHA²DS²-VASc score predict increased risk of stroke, thromboembolism, and death, regardless of whether AF is present. It is unknown if this score can identify subgroups of patients with HF, with and without AF, at particularly high or low risk of these events.

Study design: Prospective, cohort study.

Setting: Three Danish registries, 2000-2012.

Synopsis: Among 42,987 patients 50 years and older with incident HF not on anticoagulation, the absolute risk of stroke among patients without AF was 1.5% per year or higher with a CHA²DS²-VASc score of two or higher. The absolute risk of stroke was 4% or higher at five years. Risks were higher in the 21.9% of patients with AF. The CHA²DS²-VASc score modestly predicted endpoints and had an approximately 90% negative predictive value for stroke, thromboembolism, and death at one-year follow-up, regardless of whether or not AF was present.

In this large study, HF patients in a non-diverse population were studied, and some patients may have had undiagnosed AF. Functional status and ejection fraction in patients with HF could not be categorized; however, reported five-year results may be generalizable to patients with chronic HF. Select patients with HF without AF, who have two or more factors of the score besides HF, might benefit from anticoagulation.

Bottom line: The CHA²DS²-VASc score modestly predicts stroke, thromboembolism, and death among patients with HF, but further studies are needed to determine its clinical usefulness.

Citation: Melgaard L, Gorst-Rasmussen A, Lane DA, Rasmussen LH, Larsen TB, Lip GY. Assessment of the CHA²DS²-VASc Score in predicting ischemic stroke, thromboembolism, and death in patients with heart failure with and without atrial fibrillation. JAMA. 2015;314(10):1030-1038.

Clinical question: For patients with heart failure (HF), with and without concurrent Afib (AF), does the CHA²DS²-VASc score predict ischemic stroke, thromboembolism, and death?

Background: Factors in the CHA²DS²-VASc score predict increased risk of stroke, thromboembolism, and death, regardless of whether AF is present. It is unknown if this score can identify subgroups of patients with HF, with and without AF, at particularly high or low risk of these events.

Study design: Prospective, cohort study.

Setting: Three Danish registries, 2000-2012.

Synopsis: Among 42,987 patients 50 years and older with incident HF not on anticoagulation, the absolute risk of stroke among patients without AF was 1.5% per year or higher with a CHA²DS²-VASc score of two or higher. The absolute risk of stroke was 4% or higher at five years. Risks were higher in the 21.9% of patients with AF. The CHA²DS²-VASc score modestly predicted endpoints and had an approximately 90% negative predictive value for stroke, thromboembolism, and death at one-year follow-up, regardless of whether or not AF was present.

In this large study, HF patients in a non-diverse population were studied, and some patients may have had undiagnosed AF. Functional status and ejection fraction in patients with HF could not be categorized; however, reported five-year results may be generalizable to patients with chronic HF. Select patients with HF without AF, who have two or more factors of the score besides HF, might benefit from anticoagulation.

Bottom line: The CHA²DS²-VASc score modestly predicts stroke, thromboembolism, and death among patients with HF, but further studies are needed to determine its clinical usefulness.

Citation: Melgaard L, Gorst-Rasmussen A, Lane DA, Rasmussen LH, Larsen TB, Lip GY. Assessment of the CHA²DS²-VASc Score in predicting ischemic stroke, thromboembolism, and death in patients with heart failure with and without atrial fibrillation. JAMA. 2015;314(10):1030-1038.

Clinical question: What blood pressure deviations during surgery are predictive of mortality?

Background: Despite the widely assumed importance of blood pressure (BP) management on postoperative outcomes, there are no accepted thresholds requiring intervention.

Study design: Retrospective cohort.

Setting: Six Veterans’ Affairs hospitals, 2001-2008.

Synopsis: Intraoperative BP data from 18,756 patients undergoing major noncardiac surgery were linked with procedure, patient-related risk factors, and 30-day mortality data from the VA Surgical Quality Improvement Program database. Overall 30-day mortality was 1.8%. Using three different methods for defining hyper- or hypotension (based on standard deviations from the mean in this population, absolute thresholds suggested by medical literature, or by changes from baseline BP), no measure of hypertension predicted mortality; however, after adjusting for 10 preoperative patient-related risk factors, extremely low BP for five minutes or more (whether defined as systolic BP <70 mmHg, mean arterial pressure <49 mmHg, or diastolic BP <30 mmHg) was associated with 30-day mortality, with statistically significant odds ratios in the 2.4-3.2 range.

Because this is an observational study, no causal relationship can be established from these data. Low BPs could be markers for sicker patients with increased mortality, despite researchers’ efforts to adjust for known preoperative risks.

Bottom line: Intraoperative hypotension lasting five minutes or more, but not intraoperative hypertension, predicts 30-day mortality.

Citation: Monk TG, Bronsert MR, Henderson WG, et al. Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery. Anesthesiology. 2015;123(2):307-319.

Clinical question: What blood pressure deviations during surgery are predictive of mortality?

Background: Despite the widely assumed importance of blood pressure (BP) management on postoperative outcomes, there are no accepted thresholds requiring intervention.

Study design: Retrospective cohort.

Setting: Six Veterans’ Affairs hospitals, 2001-2008.

Synopsis: Intraoperative BP data from 18,756 patients undergoing major noncardiac surgery were linked with procedure, patient-related risk factors, and 30-day mortality data from the VA Surgical Quality Improvement Program database. Overall 30-day mortality was 1.8%. Using three different methods for defining hyper- or hypotension (based on standard deviations from the mean in this population, absolute thresholds suggested by medical literature, or by changes from baseline BP), no measure of hypertension predicted mortality; however, after adjusting for 10 preoperative patient-related risk factors, extremely low BP for five minutes or more (whether defined as systolic BP <70 mmHg, mean arterial pressure <49 mmHg, or diastolic BP <30 mmHg) was associated with 30-day mortality, with statistically significant odds ratios in the 2.4-3.2 range.

Because this is an observational study, no causal relationship can be established from these data. Low BPs could be markers for sicker patients with increased mortality, despite researchers’ efforts to adjust for known preoperative risks.

Bottom line: Intraoperative hypotension lasting five minutes or more, but not intraoperative hypertension, predicts 30-day mortality.

Citation: Monk TG, Bronsert MR, Henderson WG, et al. Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery. Anesthesiology. 2015;123(2):307-319.

Clinical question: What blood pressure deviations during surgery are predictive of mortality?

Background: Despite the widely assumed importance of blood pressure (BP) management on postoperative outcomes, there are no accepted thresholds requiring intervention.

Study design: Retrospective cohort.

Setting: Six Veterans’ Affairs hospitals, 2001-2008.

Synopsis: Intraoperative BP data from 18,756 patients undergoing major noncardiac surgery were linked with procedure, patient-related risk factors, and 30-day mortality data from the VA Surgical Quality Improvement Program database. Overall 30-day mortality was 1.8%. Using three different methods for defining hyper- or hypotension (based on standard deviations from the mean in this population, absolute thresholds suggested by medical literature, or by changes from baseline BP), no measure of hypertension predicted mortality; however, after adjusting for 10 preoperative patient-related risk factors, extremely low BP for five minutes or more (whether defined as systolic BP <70 mmHg, mean arterial pressure <49 mmHg, or diastolic BP <30 mmHg) was associated with 30-day mortality, with statistically significant odds ratios in the 2.4-3.2 range.

Because this is an observational study, no causal relationship can be established from these data. Low BPs could be markers for sicker patients with increased mortality, despite researchers’ efforts to adjust for known preoperative risks.

Bottom line: Intraoperative hypotension lasting five minutes or more, but not intraoperative hypertension, predicts 30-day mortality.

Citation: Monk TG, Bronsert MR, Henderson WG, et al. Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery. Anesthesiology. 2015;123(2):307-319.

Clinical question: Is the Caprini Risk Assessment Model for VTE risk valid in critically ill surgical patients?

Background: Critically ill surgical patients are at increased risk of developing VTE. Chemoprophylaxis decreases VTE risk, but benefits must be balanced against bleeding risk. Rapid and accurate risk stratification supports decisions about prophylaxis; however, data regarding appropriate risk stratification are limited.

Study design: Retrospective, cohort study.

Setting: Surgical ICU (SICU) at a single, U.S. academic medical center, 2007-2013.

Synopsis: Among 4,844 consecutive admissions, the in-hospital VTE rate was 7.5% (364). Using a previously validated, computer-generated, retrospective risk score based on the 2005 Caprini model, patients were most commonly at moderate risk for VTE upon ICU admission (32%). Fifteen percent (723) were extremely high risk. VTE incidence increased linearly with increasing Caprini scores. Data were abstracted from multiple electronic sources.

Younger age, recent sepsis or pneumonia, central venous access on ICU admission, personal VTE history, and operative procedure were significantly associated with inpatient VTE events. The proportion of patients who received chemoprophylaxis postoperatively was similar regardless of VTE risk. Patients at higher risk were more likely to receive chemoprophylaxis preoperatively.

Results from this retrospective, single-center study suggest that Caprini is a valid tool to predict inpatient VTE risk in this population. Inclusion of multiple risk factors may make calculation of this score prohibitive in other settings unless it can be computer generated.

Bottom line: Caprini risk scores accurately distinguish critically ill surgical patients at high risk of VTE from those at lower risk.

Citation: Obi AT, Pannucci CJ, Nackashi A, et al. Validation of the Caprini venous thromboembolism risk assessment model in critically ill surgical patients. JAMA Surg. 2015;150(10):941-948. doi:10.1001/jamasurg.2015.1841.

Clinical question: Is the Caprini Risk Assessment Model for VTE risk valid in critically ill surgical patients?

Background: Critically ill surgical patients are at increased risk of developing VTE. Chemoprophylaxis decreases VTE risk, but benefits must be balanced against bleeding risk. Rapid and accurate risk stratification supports decisions about prophylaxis; however, data regarding appropriate risk stratification are limited.

Study design: Retrospective, cohort study.

Setting: Surgical ICU (SICU) at a single, U.S. academic medical center, 2007-2013.

Synopsis: Among 4,844 consecutive admissions, the in-hospital VTE rate was 7.5% (364). Using a previously validated, computer-generated, retrospective risk score based on the 2005 Caprini model, patients were most commonly at moderate risk for VTE upon ICU admission (32%). Fifteen percent (723) were extremely high risk. VTE incidence increased linearly with increasing Caprini scores. Data were abstracted from multiple electronic sources.

Younger age, recent sepsis or pneumonia, central venous access on ICU admission, personal VTE history, and operative procedure were significantly associated with inpatient VTE events. The proportion of patients who received chemoprophylaxis postoperatively was similar regardless of VTE risk. Patients at higher risk were more likely to receive chemoprophylaxis preoperatively.

Results from this retrospective, single-center study suggest that Caprini is a valid tool to predict inpatient VTE risk in this population. Inclusion of multiple risk factors may make calculation of this score prohibitive in other settings unless it can be computer generated.

Bottom line: Caprini risk scores accurately distinguish critically ill surgical patients at high risk of VTE from those at lower risk.

Citation: Obi AT, Pannucci CJ, Nackashi A, et al. Validation of the Caprini venous thromboembolism risk assessment model in critically ill surgical patients. JAMA Surg. 2015;150(10):941-948. doi:10.1001/jamasurg.2015.1841.

Clinical question: Is the Caprini Risk Assessment Model for VTE risk valid in critically ill surgical patients?

Background: Critically ill surgical patients are at increased risk of developing VTE. Chemoprophylaxis decreases VTE risk, but benefits must be balanced against bleeding risk. Rapid and accurate risk stratification supports decisions about prophylaxis; however, data regarding appropriate risk stratification are limited.

Study design: Retrospective, cohort study.

Setting: Surgical ICU (SICU) at a single, U.S. academic medical center, 2007-2013.

Synopsis: Among 4,844 consecutive admissions, the in-hospital VTE rate was 7.5% (364). Using a previously validated, computer-generated, retrospective risk score based on the 2005 Caprini model, patients were most commonly at moderate risk for VTE upon ICU admission (32%). Fifteen percent (723) were extremely high risk. VTE incidence increased linearly with increasing Caprini scores. Data were abstracted from multiple electronic sources.

Younger age, recent sepsis or pneumonia, central venous access on ICU admission, personal VTE history, and operative procedure were significantly associated with inpatient VTE events. The proportion of patients who received chemoprophylaxis postoperatively was similar regardless of VTE risk. Patients at higher risk were more likely to receive chemoprophylaxis preoperatively.

Results from this retrospective, single-center study suggest that Caprini is a valid tool to predict inpatient VTE risk in this population. Inclusion of multiple risk factors may make calculation of this score prohibitive in other settings unless it can be computer generated.

Bottom line: Caprini risk scores accurately distinguish critically ill surgical patients at high risk of VTE from those at lower risk.

Citation: Obi AT, Pannucci CJ, Nackashi A, et al. Validation of the Caprini venous thromboembolism risk assessment model in critically ill surgical patients. JAMA Surg. 2015;150(10):941-948. doi:10.1001/jamasurg.2015.1841.

Clinical question: Can a multifaceted approach involving clinician education, peer email feedback, and monthly audit data improve adherence to restrictive red blood cell (RBC) transfusion guidelines?

Background: Randomized controlled trials and professional society guidelines support adoption of RBC transfusion strategies in stable, low-risk patients. Studies suggest that education and feedback from specialists may decrease inappropriate transfusion practices, but peer-to-peer feedback has not yet been explored.

Study design: Prospective, interventional study.

Setting: Tertiary care center SICU, single U.S. academic center.

Synopsis: All stable, low-risk SICU patients receiving RBC transfusions were included in this study. Intervention consisted of educational lectures to clinicians, dissemination of monthly aggregate audit transfusion data, and direct email feedback from a colleague to clinicians ordering transfusions outside of guidelines. Six-month intervention data were compared with six months of pre-intervention data.

During the intervention, total transfusions decreased by 36%, from 284 units to 181 units, and percentage of transfusions outside restrictive guidelines decreased to 2% from 25% (P<0.001). Six months after the end of the intervention period, transfusions outside restrictive guidelines increased back to 17%, suggesting a lack of permanent change in transfusion practices.

Bottom line: A multifaceted approach involving education, peer-to-peer feedback, and monthly audits improved adherence to restrictive RBC transfusion guidelines; however, changes were not sustained.

Citation: Yeh DD, Naraghi L, Larentzakis A, et al. Peer-to-peer physician feedback improves adherence to blood transfusion guidelines in the surgical intensive care unit. J Trauma Acute Care Surg. 2015;79(1):65-70.

Clinical question: Can a multifaceted approach involving clinician education, peer email feedback, and monthly audit data improve adherence to restrictive red blood cell (RBC) transfusion guidelines?

Background: Randomized controlled trials and professional society guidelines support adoption of RBC transfusion strategies in stable, low-risk patients. Studies suggest that education and feedback from specialists may decrease inappropriate transfusion practices, but peer-to-peer feedback has not yet been explored.

Study design: Prospective, interventional study.

Setting: Tertiary care center SICU, single U.S. academic center.

Synopsis: All stable, low-risk SICU patients receiving RBC transfusions were included in this study. Intervention consisted of educational lectures to clinicians, dissemination of monthly aggregate audit transfusion data, and direct email feedback from a colleague to clinicians ordering transfusions outside of guidelines. Six-month intervention data were compared with six months of pre-intervention data.

During the intervention, total transfusions decreased by 36%, from 284 units to 181 units, and percentage of transfusions outside restrictive guidelines decreased to 2% from 25% (P<0.001). Six months after the end of the intervention period, transfusions outside restrictive guidelines increased back to 17%, suggesting a lack of permanent change in transfusion practices.

Bottom line: A multifaceted approach involving education, peer-to-peer feedback, and monthly audits improved adherence to restrictive RBC transfusion guidelines; however, changes were not sustained.

Citation: Yeh DD, Naraghi L, Larentzakis A, et al. Peer-to-peer physician feedback improves adherence to blood transfusion guidelines in the surgical intensive care unit. J Trauma Acute Care Surg. 2015;79(1):65-70.

Clinical question: Can a multifaceted approach involving clinician education, peer email feedback, and monthly audit data improve adherence to restrictive red blood cell (RBC) transfusion guidelines?

Background: Randomized controlled trials and professional society guidelines support adoption of RBC transfusion strategies in stable, low-risk patients. Studies suggest that education and feedback from specialists may decrease inappropriate transfusion practices, but peer-to-peer feedback has not yet been explored.

Study design: Prospective, interventional study.

Setting: Tertiary care center SICU, single U.S. academic center.

Synopsis: All stable, low-risk SICU patients receiving RBC transfusions were included in this study. Intervention consisted of educational lectures to clinicians, dissemination of monthly aggregate audit transfusion data, and direct email feedback from a colleague to clinicians ordering transfusions outside of guidelines. Six-month intervention data were compared with six months of pre-intervention data.

During the intervention, total transfusions decreased by 36%, from 284 units to 181 units, and percentage of transfusions outside restrictive guidelines decreased to 2% from 25% (P<0.001). Six months after the end of the intervention period, transfusions outside restrictive guidelines increased back to 17%, suggesting a lack of permanent change in transfusion practices.

Bottom line: A multifaceted approach involving education, peer-to-peer feedback, and monthly audits improved adherence to restrictive RBC transfusion guidelines; however, changes were not sustained.

Citation: Yeh DD, Naraghi L, Larentzakis A, et al. Peer-to-peer physician feedback improves adherence to blood transfusion guidelines in the surgical intensive care unit. J Trauma Acute Care Surg. 2015;79(1):65-70.

Clinical question: Are adjunctive corticosteroids beneficial for patients hospitalized with community-acquired pneumonia (CAP)?

Background: Numerous studies have tried to determine whether or not adjunctive corticosteroids for CAP treatment in hospitalized patients improve outcomes. Although recent trials have suggested that corticosteroids may improve morbidity and mortality, prior meta-analyses have failed to show a benefit, and steroids are not currently routinely recommended for this population.

Study design: Systematic review and meta-analysis of 13 RCTs, predominantly from Europe.

Synopsis: Analysis of 1,974 patients suggested a decrease in all-cause mortality (relative risk (RR) 0.67, 95% CI 0.45-1.01) with adjunctive corticosteroids. Subgroup analysis for severe CAP (six studies, n=388) suggested a greater mortality benefit (RR 0.39, 95% CI 0.2-0.77). There was a decrease in the risk of mechanical ventilation (five studies, n=1060, RR 0.45, CI, 0.26-0.79), ICU admission (three studies, n=950, RR 0.69, 95% CI, 0.46-1.03), and development of acute respiratory distress syndrome (four studies, n=945, RR 0.24, 95% CI 0.10-0.56).

Both hospital length of stay (LOS) and time to clinical stability (hemodynamically stable with no hypoxia) were significantly decreased (mean decrease LOS one day; time to clinical stability 1.22 days). Adverse effects were rare but included increased rates of hyperglycemia requiring treatment (RR 1.49, 95% CI 1.01-2.19). There was no increased frequency of gastrointestinal hemorrhage, neuropsychiatric complications, or rehospitalization.

Bottom line: Adjunctive corticosteroids for inpatient CAP treatment decrease morbidity and mortality, particularly in severe disease, and decrease LOS and time to clinical stability with few adverse reactions.

Citation: Siemieniuk RA, Meade MO, Alonso-Coello P, et al. Corticosteroid therapy for patients hospitalized with community-acquired pneumonia: a systematic review and meta-analysis. Ann Intern Med. 2015;163(7):519-528.

Clinical question: Are adjunctive corticosteroids beneficial for patients hospitalized with community-acquired pneumonia (CAP)?

Background: Numerous studies have tried to determine whether or not adjunctive corticosteroids for CAP treatment in hospitalized patients improve outcomes. Although recent trials have suggested that corticosteroids may improve morbidity and mortality, prior meta-analyses have failed to show a benefit, and steroids are not currently routinely recommended for this population.

Study design: Systematic review and meta-analysis of 13 RCTs, predominantly from Europe.

Synopsis: Analysis of 1,974 patients suggested a decrease in all-cause mortality (relative risk (RR) 0.67, 95% CI 0.45-1.01) with adjunctive corticosteroids. Subgroup analysis for severe CAP (six studies, n=388) suggested a greater mortality benefit (RR 0.39, 95% CI 0.2-0.77). There was a decrease in the risk of mechanical ventilation (five studies, n=1060, RR 0.45, CI, 0.26-0.79), ICU admission (three studies, n=950, RR 0.69, 95% CI, 0.46-1.03), and development of acute respiratory distress syndrome (four studies, n=945, RR 0.24, 95% CI 0.10-0.56).

Both hospital length of stay (LOS) and time to clinical stability (hemodynamically stable with no hypoxia) were significantly decreased (mean decrease LOS one day; time to clinical stability 1.22 days). Adverse effects were rare but included increased rates of hyperglycemia requiring treatment (RR 1.49, 95% CI 1.01-2.19). There was no increased frequency of gastrointestinal hemorrhage, neuropsychiatric complications, or rehospitalization.

Bottom line: Adjunctive corticosteroids for inpatient CAP treatment decrease morbidity and mortality, particularly in severe disease, and decrease LOS and time to clinical stability with few adverse reactions.

Citation: Siemieniuk RA, Meade MO, Alonso-Coello P, et al. Corticosteroid therapy for patients hospitalized with community-acquired pneumonia: a systematic review and meta-analysis. Ann Intern Med. 2015;163(7):519-528.

Clinical question: Are adjunctive corticosteroids beneficial for patients hospitalized with community-acquired pneumonia (CAP)?

Background: Numerous studies have tried to determine whether or not adjunctive corticosteroids for CAP treatment in hospitalized patients improve outcomes. Although recent trials have suggested that corticosteroids may improve morbidity and mortality, prior meta-analyses have failed to show a benefit, and steroids are not currently routinely recommended for this population.

Study design: Systematic review and meta-analysis of 13 RCTs, predominantly from Europe.

Synopsis: Analysis of 1,974 patients suggested a decrease in all-cause mortality (relative risk (RR) 0.67, 95% CI 0.45-1.01) with adjunctive corticosteroids. Subgroup analysis for severe CAP (six studies, n=388) suggested a greater mortality benefit (RR 0.39, 95% CI 0.2-0.77). There was a decrease in the risk of mechanical ventilation (five studies, n=1060, RR 0.45, CI, 0.26-0.79), ICU admission (three studies, n=950, RR 0.69, 95% CI, 0.46-1.03), and development of acute respiratory distress syndrome (four studies, n=945, RR 0.24, 95% CI 0.10-0.56).

Both hospital length of stay (LOS) and time to clinical stability (hemodynamically stable with no hypoxia) were significantly decreased (mean decrease LOS one day; time to clinical stability 1.22 days). Adverse effects were rare but included increased rates of hyperglycemia requiring treatment (RR 1.49, 95% CI 1.01-2.19). There was no increased frequency of gastrointestinal hemorrhage, neuropsychiatric complications, or rehospitalization.

Bottom line: Adjunctive corticosteroids for inpatient CAP treatment decrease morbidity and mortality, particularly in severe disease, and decrease LOS and time to clinical stability with few adverse reactions.

Citation: Siemieniuk RA, Meade MO, Alonso-Coello P, et al. Corticosteroid therapy for patients hospitalized with community-acquired pneumonia: a systematic review and meta-analysis. Ann Intern Med. 2015;163(7):519-528.

Clinical question: How do current Helicobacter pylori treatments compare in efficacy and tolerance?

Background: Efficacy of standard “triple therapy” for H. pylori (proton pump inhibitor plus clarithromycin and amoxicillin or metronidazole) is declining due to the development of antibiotic resistance. Different medication combinations and/or time courses are currently used, but comparative effectiveness of these treatments has not been evaluated comprehensively.

Study design: Systematic review and network meta-analysis.

Setting: Cochrane Library, PubMed, and Embase databases.

Synopsis: One hundred forty-three RCTs evaluating a total of 14 treatments for H. pylori were identified. Network meta-analysis was performed to rank order treatments for efficacy (eradication rate of H. pylori) and tolerance (adverse event occurrence rate). Seven days of “concomitant treatment” (proton pump inhibitor plus three antibiotics) ranked the highest in efficacy (eradication rate 0.94; 95% confidence interval [CI] 0.89-0.98), though this treatment group comprised a very small proportion of overall participants and studies and may be subject to bias. Seven days of standard “triple therapy” ranked the lowest in efficacy (eradication rate 0.73; 95% CI 0.71-0.75).

Of note, subgroup analysis showed variation in efficacy rankings by geographic location, suggesting that findings may not be universally applicable. Only two treatments showed significantly different adverse event occurrence rates compared to standard “triple therapy,” indicating overall similar tolerance for most treatments.

Bottom line: Several treatment regimens may be more effective than standard H. pylori “triple therapy” and equally well tolerated.

Citation: Li BZ, Threapleton DE, Wang JY, et al. Comparative effectiveness and tolerance of treatments for Helicobacter pylori: systematic review and network meta-analysis. BMJ. 2015;351:h4052.

Clinical question: How do current Helicobacter pylori treatments compare in efficacy and tolerance?

Background: Efficacy of standard “triple therapy” for H. pylori (proton pump inhibitor plus clarithromycin and amoxicillin or metronidazole) is declining due to the development of antibiotic resistance. Different medication combinations and/or time courses are currently used, but comparative effectiveness of these treatments has not been evaluated comprehensively.

Study design: Systematic review and network meta-analysis.

Setting: Cochrane Library, PubMed, and Embase databases.

Synopsis: One hundred forty-three RCTs evaluating a total of 14 treatments for H. pylori were identified. Network meta-analysis was performed to rank order treatments for efficacy (eradication rate of H. pylori) and tolerance (adverse event occurrence rate). Seven days of “concomitant treatment” (proton pump inhibitor plus three antibiotics) ranked the highest in efficacy (eradication rate 0.94; 95% confidence interval [CI] 0.89-0.98), though this treatment group comprised a very small proportion of overall participants and studies and may be subject to bias. Seven days of standard “triple therapy” ranked the lowest in efficacy (eradication rate 0.73; 95% CI 0.71-0.75).

Of note, subgroup analysis showed variation in efficacy rankings by geographic location, suggesting that findings may not be universally applicable. Only two treatments showed significantly different adverse event occurrence rates compared to standard “triple therapy,” indicating overall similar tolerance for most treatments.

Bottom line: Several treatment regimens may be more effective than standard H. pylori “triple therapy” and equally well tolerated.

Citation: Li BZ, Threapleton DE, Wang JY, et al. Comparative effectiveness and tolerance of treatments for Helicobacter pylori: systematic review and network meta-analysis. BMJ. 2015;351:h4052.

Clinical question: How do current Helicobacter pylori treatments compare in efficacy and tolerance?

Background: Efficacy of standard “triple therapy” for H. pylori (proton pump inhibitor plus clarithromycin and amoxicillin or metronidazole) is declining due to the development of antibiotic resistance. Different medication combinations and/or time courses are currently used, but comparative effectiveness of these treatments has not been evaluated comprehensively.

Study design: Systematic review and network meta-analysis.

Setting: Cochrane Library, PubMed, and Embase databases.

Synopsis: One hundred forty-three RCTs evaluating a total of 14 treatments for H. pylori were identified. Network meta-analysis was performed to rank order treatments for efficacy (eradication rate of H. pylori) and tolerance (adverse event occurrence rate). Seven days of “concomitant treatment” (proton pump inhibitor plus three antibiotics) ranked the highest in efficacy (eradication rate 0.94; 95% confidence interval [CI] 0.89-0.98), though this treatment group comprised a very small proportion of overall participants and studies and may be subject to bias. Seven days of standard “triple therapy” ranked the lowest in efficacy (eradication rate 0.73; 95% CI 0.71-0.75).

Of note, subgroup analysis showed variation in efficacy rankings by geographic location, suggesting that findings may not be universally applicable. Only two treatments showed significantly different adverse event occurrence rates compared to standard “triple therapy,” indicating overall similar tolerance for most treatments.

Bottom line: Several treatment regimens may be more effective than standard H. pylori “triple therapy” and equally well tolerated.

Citation: Li BZ, Threapleton DE, Wang JY, et al. Comparative effectiveness and tolerance of treatments for Helicobacter pylori: systematic review and network meta-analysis. BMJ. 2015;351:h4052.

Clinical question: What are predictors of primary medication nonadherence after discharge?

Background: Primary nonadherence occurs when a patient receives a prescription at hospital discharge but does not fill it. Predictors of post-discharge primary nonadherence could serve as useful targets to guide adherence interventions.

Study design: RCT, secondary analysis.

Setting: Two tertiary care, U.S. academic hospitals.

Synopsis: Using the Pharmacist Intervention for Low Literacy in Cardiovascular Disease (PILL-CVD) study database, investigators conducted a secondary analysis of adults hospitalized for acute coronary syndrome or acute decompensated heart failure who received pharmacist-assisted medication reconciliation, discharge counseling, low-literacy adherence aids, and a follow-up phone call. The prevalence of primary nonadherence one to four days post-discharge was 9.4% among 341 patients. In subsequent multivariate analysis, significant factors for noncompliance were living alone (odds ratio 2.2, 95% CI 1.01-4.8, P=0.047) and more than 10 total discharge medications (odds ratio 2.3, 95% CI 1.05-4.98, P=0.036).

Limitations to this study include biases from patient-reported outcomes, lack of patient copayment data, and limited characterization of discharge medication type.

Bottom line: Among patients hospitalized for cardiac events, social isolation and polypharmacy predict primary medication nonadherence to discharge medications despite intensive pharmacist counseling.

Citation: Wooldridge K, Schnipper JL, Goggins K, Dittus RS, Kripalani S. Refractory primary medication nonadherence: prevalence and predictors after pharmacist counseling at hospital discharge [published online ahead of print August 21, 2015]. J Hosp Med. doi: 10.1002/jhm.2446.

Clinical question: What are predictors of primary medication nonadherence after discharge?

Background: Primary nonadherence occurs when a patient receives a prescription at hospital discharge but does not fill it. Predictors of post-discharge primary nonadherence could serve as useful targets to guide adherence interventions.

Study design: RCT, secondary analysis.

Setting: Two tertiary care, U.S. academic hospitals.

Synopsis: Using the Pharmacist Intervention for Low Literacy in Cardiovascular Disease (PILL-CVD) study database, investigators conducted a secondary analysis of adults hospitalized for acute coronary syndrome or acute decompensated heart failure who received pharmacist-assisted medication reconciliation, discharge counseling, low-literacy adherence aids, and a follow-up phone call. The prevalence of primary nonadherence one to four days post-discharge was 9.4% among 341 patients. In subsequent multivariate analysis, significant factors for noncompliance were living alone (odds ratio 2.2, 95% CI 1.01-4.8, P=0.047) and more than 10 total discharge medications (odds ratio 2.3, 95% CI 1.05-4.98, P=0.036).

Limitations to this study include biases from patient-reported outcomes, lack of patient copayment data, and limited characterization of discharge medication type.

Bottom line: Among patients hospitalized for cardiac events, social isolation and polypharmacy predict primary medication nonadherence to discharge medications despite intensive pharmacist counseling.

Citation: Wooldridge K, Schnipper JL, Goggins K, Dittus RS, Kripalani S. Refractory primary medication nonadherence: prevalence and predictors after pharmacist counseling at hospital discharge [published online ahead of print August 21, 2015]. J Hosp Med. doi: 10.1002/jhm.2446.

Clinical question: What are predictors of primary medication nonadherence after discharge?

Background: Primary nonadherence occurs when a patient receives a prescription at hospital discharge but does not fill it. Predictors of post-discharge primary nonadherence could serve as useful targets to guide adherence interventions.

Study design: RCT, secondary analysis.

Setting: Two tertiary care, U.S. academic hospitals.

Synopsis: Using the Pharmacist Intervention for Low Literacy in Cardiovascular Disease (PILL-CVD) study database, investigators conducted a secondary analysis of adults hospitalized for acute coronary syndrome or acute decompensated heart failure who received pharmacist-assisted medication reconciliation, discharge counseling, low-literacy adherence aids, and a follow-up phone call. The prevalence of primary nonadherence one to four days post-discharge was 9.4% among 341 patients. In subsequent multivariate analysis, significant factors for noncompliance were living alone (odds ratio 2.2, 95% CI 1.01-4.8, P=0.047) and more than 10 total discharge medications (odds ratio 2.3, 95% CI 1.05-4.98, P=0.036).

Limitations to this study include biases from patient-reported outcomes, lack of patient copayment data, and limited characterization of discharge medication type.

Bottom line: Among patients hospitalized for cardiac events, social isolation and polypharmacy predict primary medication nonadherence to discharge medications despite intensive pharmacist counseling.

Citation: Wooldridge K, Schnipper JL, Goggins K, Dittus RS, Kripalani S. Refractory primary medication nonadherence: prevalence and predictors after pharmacist counseling at hospital discharge [published online ahead of print August 21, 2015]. J Hosp Med. doi: 10.1002/jhm.2446.

Clinical question: Does a patient navigator (PN) who facilitates communication between patients and providers impact hospital length of stay (LOS) and readmissions?

Background: Increasing complexity of hospitalization challenges the safety of care transitions. There are few studies about the effectiveness of innovations targeting both communication and transitional care planning.

Study design: Retrospective, cohort study.

Setting: Single academic health center in Canada, 2010-2014.

Synopsis: PNs, dedicated team-based facilitators not responsible for clinical care, served as liaisons between patients and providers on general medicine teams. They rounded with medical teams, tracked action items, expedited tests and consults, and proactively served as direct primary contacts for patients/families during and after hospitalization. PNs had no specified prior training; they underwent on-the-job training with regular feedback.

Researchers matched 7,841 hospitalizations (5,628 with PN; 2,213 without) by case mix, age, and resource intensity. LOS and 30-day readmissions were primary outcomes. Hospitalizations with PNs were 21% shorter (1.3 days; 6.2 v 7.5 days, P<0.001) than those without PNs.

There were no differences in 30-day readmission rates (13.1 v 13.8%, P=0.48). In this single center study in Canada, the impact of PN salaries (the only program cost) relative to savings is unknown.

Bottom line: Inpatient navigators streamline communication and decrease LOS without increasing readmissions. Additional cost-benefit analyses are needed.

Citation: Kwan JL, Morgan MW, Stewart TE, Bell CM. Impact of an innovative patient navigator program on length of stay and 30-day readmission [published online ahead of print August 10, 2015]. J Hosp Med. doi: 10.1002/jhm.2442.

Clinical question: Does a patient navigator (PN) who facilitates communication between patients and providers impact hospital length of stay (LOS) and readmissions?

Background: Increasing complexity of hospitalization challenges the safety of care transitions. There are few studies about the effectiveness of innovations targeting both communication and transitional care planning.

Study design: Retrospective, cohort study.

Setting: Single academic health center in Canada, 2010-2014.

Synopsis: PNs, dedicated team-based facilitators not responsible for clinical care, served as liaisons between patients and providers on general medicine teams. They rounded with medical teams, tracked action items, expedited tests and consults, and proactively served as direct primary contacts for patients/families during and after hospitalization. PNs had no specified prior training; they underwent on-the-job training with regular feedback.

Researchers matched 7,841 hospitalizations (5,628 with PN; 2,213 without) by case mix, age, and resource intensity. LOS and 30-day readmissions were primary outcomes. Hospitalizations with PNs were 21% shorter (1.3 days; 6.2 v 7.5 days, P<0.001) than those without PNs.

There were no differences in 30-day readmission rates (13.1 v 13.8%, P=0.48). In this single center study in Canada, the impact of PN salaries (the only program cost) relative to savings is unknown.

Bottom line: Inpatient navigators streamline communication and decrease LOS without increasing readmissions. Additional cost-benefit analyses are needed.

Citation: Kwan JL, Morgan MW, Stewart TE, Bell CM. Impact of an innovative patient navigator program on length of stay and 30-day readmission [published online ahead of print August 10, 2015]. J Hosp Med. doi: 10.1002/jhm.2442.

Clinical question: Does a patient navigator (PN) who facilitates communication between patients and providers impact hospital length of stay (LOS) and readmissions?

Background: Increasing complexity of hospitalization challenges the safety of care transitions. There are few studies about the effectiveness of innovations targeting both communication and transitional care planning.

Study design: Retrospective, cohort study.

Setting: Single academic health center in Canada, 2010-2014.

Synopsis: PNs, dedicated team-based facilitators not responsible for clinical care, served as liaisons between patients and providers on general medicine teams. They rounded with medical teams, tracked action items, expedited tests and consults, and proactively served as direct primary contacts for patients/families during and after hospitalization. PNs had no specified prior training; they underwent on-the-job training with regular feedback.

Researchers matched 7,841 hospitalizations (5,628 with PN; 2,213 without) by case mix, age, and resource intensity. LOS and 30-day readmissions were primary outcomes. Hospitalizations with PNs were 21% shorter (1.3 days; 6.2 v 7.5 days, P<0.001) than those without PNs.

There were no differences in 30-day readmission rates (13.1 v 13.8%, P=0.48). In this single center study in Canada, the impact of PN salaries (the only program cost) relative to savings is unknown.

Bottom line: Inpatient navigators streamline communication and decrease LOS without increasing readmissions. Additional cost-benefit analyses are needed.

Citation: Kwan JL, Morgan MW, Stewart TE, Bell CM. Impact of an innovative patient navigator program on length of stay and 30-day readmission [published online ahead of print August 10, 2015]. J Hosp Med. doi: 10.1002/jhm.2442.