User login
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago. She is editor of the Journal of Hospital Medicine POEMs.
New-Onset Perioperative Atrial Fibrillation Associated With Increased Risk of Ischemic Stroke
Clinical question
For patients undergoing any inpatient surgery, is the presence of new-onset perioperative atrial fibrillation associated with a greater long-term risk of ischemic stroke?
Bottom line
Perioperative atrial fibrillation (AF) is associated with an increased risk of ischemic stroke for patients who are hospitalized for surgery. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Setting
Inpatient (any location)
Synopsis
New-onset perioperative AF is very common but its long-term association with ischemic stroke is unknown. Using administrative data, these investigators identified all adult patients who underwent inpatient surgery during a 4-year period. Patients with preexisiting AF and those with documented cerebrovascular disease were excluded. More than 1.7 million patients were included in the study, with a mean follow-up of 2 years.
Overall, perioperative AF was found in 1.4% of this population, more frequently following cardiac surgery than any other type of surgery (16% vs 0.78%; P < .001). Patients who experienced perioperative AF were also more likely to have high vascular comorbidities, such as hypertension, diabetes, and coronary artery disease. The incidence of ischemic stroke after discharge in the overall cohort was 0.81%.
After adjusting for potential confounders, including age, sex, race, and cardiovascular comorbidities, perioperative AF was independently associated with ischemic stroke both following noncardiac surgery (hazard ratio [HR] 2.0, 95% CI 1.7 - 2.3) and cardiac surgery (HR 1.3, 95% CI 1.1 - 1.6). A further analysis using a specific diagnostic code for cardioembolic stroke showed an even greater association between perioperative AF and this subset of stroke (noncardiac surgery: HR 4.9, 95% CI 3.5 - 6.7; cardiac surgery: HR 2.1, 95% CI 1.4 - 3.1).
Of note, sicker patients in this study may have had more intense cardiac monitoring following their surgeries, leading to an ascertainment bias that could overestimate the association between perioperative AF and stroke. However, a sensitivity analysis using a comorbidity index did not change the findings of the primary analysis.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients undergoing any inpatient surgery, is the presence of new-onset perioperative atrial fibrillation associated with a greater long-term risk of ischemic stroke?
Bottom line
Perioperative atrial fibrillation (AF) is associated with an increased risk of ischemic stroke for patients who are hospitalized for surgery. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Setting
Inpatient (any location)
Synopsis
New-onset perioperative AF is very common but its long-term association with ischemic stroke is unknown. Using administrative data, these investigators identified all adult patients who underwent inpatient surgery during a 4-year period. Patients with preexisiting AF and those with documented cerebrovascular disease were excluded. More than 1.7 million patients were included in the study, with a mean follow-up of 2 years.
Overall, perioperative AF was found in 1.4% of this population, more frequently following cardiac surgery than any other type of surgery (16% vs 0.78%; P < .001). Patients who experienced perioperative AF were also more likely to have high vascular comorbidities, such as hypertension, diabetes, and coronary artery disease. The incidence of ischemic stroke after discharge in the overall cohort was 0.81%.
After adjusting for potential confounders, including age, sex, race, and cardiovascular comorbidities, perioperative AF was independently associated with ischemic stroke both following noncardiac surgery (hazard ratio [HR] 2.0, 95% CI 1.7 - 2.3) and cardiac surgery (HR 1.3, 95% CI 1.1 - 1.6). A further analysis using a specific diagnostic code for cardioembolic stroke showed an even greater association between perioperative AF and this subset of stroke (noncardiac surgery: HR 4.9, 95% CI 3.5 - 6.7; cardiac surgery: HR 2.1, 95% CI 1.4 - 3.1).
Of note, sicker patients in this study may have had more intense cardiac monitoring following their surgeries, leading to an ascertainment bias that could overestimate the association between perioperative AF and stroke. However, a sensitivity analysis using a comorbidity index did not change the findings of the primary analysis.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients undergoing any inpatient surgery, is the presence of new-onset perioperative atrial fibrillation associated with a greater long-term risk of ischemic stroke?
Bottom line
Perioperative atrial fibrillation (AF) is associated with an increased risk of ischemic stroke for patients who are hospitalized for surgery. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Setting
Inpatient (any location)
Synopsis
New-onset perioperative AF is very common but its long-term association with ischemic stroke is unknown. Using administrative data, these investigators identified all adult patients who underwent inpatient surgery during a 4-year period. Patients with preexisiting AF and those with documented cerebrovascular disease were excluded. More than 1.7 million patients were included in the study, with a mean follow-up of 2 years.
Overall, perioperative AF was found in 1.4% of this population, more frequently following cardiac surgery than any other type of surgery (16% vs 0.78%; P < .001). Patients who experienced perioperative AF were also more likely to have high vascular comorbidities, such as hypertension, diabetes, and coronary artery disease. The incidence of ischemic stroke after discharge in the overall cohort was 0.81%.
After adjusting for potential confounders, including age, sex, race, and cardiovascular comorbidities, perioperative AF was independently associated with ischemic stroke both following noncardiac surgery (hazard ratio [HR] 2.0, 95% CI 1.7 - 2.3) and cardiac surgery (HR 1.3, 95% CI 1.1 - 1.6). A further analysis using a specific diagnostic code for cardioembolic stroke showed an even greater association between perioperative AF and this subset of stroke (noncardiac surgery: HR 4.9, 95% CI 3.5 - 6.7; cardiac surgery: HR 2.1, 95% CI 1.4 - 3.1).
Of note, sicker patients in this study may have had more intense cardiac monitoring following their surgeries, leading to an ascertainment bias that could overestimate the association between perioperative AF and stroke. However, a sensitivity analysis using a comorbidity index did not change the findings of the primary analysis.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Colchicine Prevents Postpericardiotomy Syndrome Following Cardiac Surgery
Clinical question
Does perioperative colchicine reduce postpericardiotomy syndrome following cardiac surgery?
Bottom line
The use of colchicine in the perioperative period decreases the incidence of postpericardiotomy syndrome after cardiac surgery with a number needed to treat (NNT) of 10. However, colchicine leads to adverse effects—specifically, gastrointestinal intolerance—and may not be tolerated during the vulnerable postoperative period. Findings in this study also suggest a role for colchicine in the prevention of postoperative atrial fibrillation, but this was not a primary outcome of this study and requires further investigation. (LOE = 1b)
Reference
Study design
Randomized controlled trial (double-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Postpericardiotomy syndrome is a common complication after cardiac surgery and is defined by the presence of 2 of the following 5 criteria: fever without another cause, pleuritic chest pain, friction rub, new or worsening pericardial effusion, or pleural effusion with elevation in C-reactive protein levels.
To test the theory that perioperative colchicine can prevent postpericardiotomy syndrome, these investigators randomized 360 patients undergoing cardiac surgery to receive either colchicine at 0.5 mg twice daily (once daily for those who weighed less than 70 kg) or matching placebo. The study medication was started at 48 hours to 72 hours prior to surgery and continued for 1 month following surgery. Baseline characteristics of the 2 groups were similar, with two thirds of the patients undergoing either heart valve surgery or coronary artery bypass graft surgery.
The primary analysis was by intention to treat, but a prespecified on-treatment analysis was also performed. The overall study drug discontinuation rate for this trial was high (20%). Postpericardiotomy syndrome occurred less frequently in the colchicine group than in the placebo group at 3-month follow-up (19% vs 29%; absolute difference 10%; NNT = 10).
Additionally, although no significant difference was detected in the primary analysis, the on-treatment analysis did show a decrease in postoperative atrial fibrillation in the colchicine group (27% vs 41%; absolute difference 14%; NNT = 7). Adverse events were significantly greater in the colchicine group (20% vs 12%; number needed to treat to harm = 12), mainly due to increased gastrointestinal intolerance, but there was no difference between the 2 groups in the rate of study drug discontinuation.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does perioperative colchicine reduce postpericardiotomy syndrome following cardiac surgery?
Bottom line
The use of colchicine in the perioperative period decreases the incidence of postpericardiotomy syndrome after cardiac surgery with a number needed to treat (NNT) of 10. However, colchicine leads to adverse effects—specifically, gastrointestinal intolerance—and may not be tolerated during the vulnerable postoperative period. Findings in this study also suggest a role for colchicine in the prevention of postoperative atrial fibrillation, but this was not a primary outcome of this study and requires further investigation. (LOE = 1b)
Reference
Study design
Randomized controlled trial (double-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Postpericardiotomy syndrome is a common complication after cardiac surgery and is defined by the presence of 2 of the following 5 criteria: fever without another cause, pleuritic chest pain, friction rub, new or worsening pericardial effusion, or pleural effusion with elevation in C-reactive protein levels.
To test the theory that perioperative colchicine can prevent postpericardiotomy syndrome, these investigators randomized 360 patients undergoing cardiac surgery to receive either colchicine at 0.5 mg twice daily (once daily for those who weighed less than 70 kg) or matching placebo. The study medication was started at 48 hours to 72 hours prior to surgery and continued for 1 month following surgery. Baseline characteristics of the 2 groups were similar, with two thirds of the patients undergoing either heart valve surgery or coronary artery bypass graft surgery.
The primary analysis was by intention to treat, but a prespecified on-treatment analysis was also performed. The overall study drug discontinuation rate for this trial was high (20%). Postpericardiotomy syndrome occurred less frequently in the colchicine group than in the placebo group at 3-month follow-up (19% vs 29%; absolute difference 10%; NNT = 10).
Additionally, although no significant difference was detected in the primary analysis, the on-treatment analysis did show a decrease in postoperative atrial fibrillation in the colchicine group (27% vs 41%; absolute difference 14%; NNT = 7). Adverse events were significantly greater in the colchicine group (20% vs 12%; number needed to treat to harm = 12), mainly due to increased gastrointestinal intolerance, but there was no difference between the 2 groups in the rate of study drug discontinuation.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does perioperative colchicine reduce postpericardiotomy syndrome following cardiac surgery?
Bottom line
The use of colchicine in the perioperative period decreases the incidence of postpericardiotomy syndrome after cardiac surgery with a number needed to treat (NNT) of 10. However, colchicine leads to adverse effects—specifically, gastrointestinal intolerance—and may not be tolerated during the vulnerable postoperative period. Findings in this study also suggest a role for colchicine in the prevention of postoperative atrial fibrillation, but this was not a primary outcome of this study and requires further investigation. (LOE = 1b)
Reference
Study design
Randomized controlled trial (double-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Postpericardiotomy syndrome is a common complication after cardiac surgery and is defined by the presence of 2 of the following 5 criteria: fever without another cause, pleuritic chest pain, friction rub, new or worsening pericardial effusion, or pleural effusion with elevation in C-reactive protein levels.
To test the theory that perioperative colchicine can prevent postpericardiotomy syndrome, these investigators randomized 360 patients undergoing cardiac surgery to receive either colchicine at 0.5 mg twice daily (once daily for those who weighed less than 70 kg) or matching placebo. The study medication was started at 48 hours to 72 hours prior to surgery and continued for 1 month following surgery. Baseline characteristics of the 2 groups were similar, with two thirds of the patients undergoing either heart valve surgery or coronary artery bypass graft surgery.
The primary analysis was by intention to treat, but a prespecified on-treatment analysis was also performed. The overall study drug discontinuation rate for this trial was high (20%). Postpericardiotomy syndrome occurred less frequently in the colchicine group than in the placebo group at 3-month follow-up (19% vs 29%; absolute difference 10%; NNT = 10).
Additionally, although no significant difference was detected in the primary analysis, the on-treatment analysis did show a decrease in postoperative atrial fibrillation in the colchicine group (27% vs 41%; absolute difference 14%; NNT = 7). Adverse events were significantly greater in the colchicine group (20% vs 12%; number needed to treat to harm = 12), mainly due to increased gastrointestinal intolerance, but there was no difference between the 2 groups in the rate of study drug discontinuation.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Shorter Stay, Fewer Interventions With Initial Cholecystectomy for Choledocholithiasis
Clinical question
Which strategy — initial cholecystectomy or common duct endoscopic investigation followed by cholecystectomy — leads to better outcomes in the treatment of suspected common duct stones?
Bottom line
As compared with a strategy of sequential common duct endoscopy followed by surgery, the use of initial cholecystectomy with intraoperative cholangiogram for patients at intermediate risk of choledocholithiasis leads to shorter hospital length of stay and fewer interventions with otherwise comparable outcomes. (LOE = 1b)
Reference
Iranmanesh P, Frossard J, Mugnier-Konrad B, et al. Initial cholecystectomy vs. sequential common duct endoscopic assessment and subsequent cholecystectomy for suspected gallstone migration. JAMA 2014;312(2):137-144.
Study design
Randomized controlled trial (nonblinded)
Funding source
Unknown/not stated
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
The role of common duct endoscopic investigation prior to cholecystectomy for patients at intermediate risk of choledocholithiasis is questionable as stones often migrate into the duodenum without such intervention. Using concealed allocation, investigators randomized 100 patients who presented to an emergency department with clinical suspicion of choledocholithiasis (right upper quadrant or epigastric abdominal pain with elevated liver function test results and ultrasound-confirmed gallstone) to 1 of 2 groups. The study group underwent primary cholecystectomy with intraoperative cholangiogram, while the classical treatment group first had common duct exploration with magnetic resonance cholangiopancreatography (MRCP), endoscopic ultrasound (EUS), and endoscopic retrograde cholangiopancreatography (ERCP), if needed, followed by cholecystectomy. The authors excluded patients with severe sepsis, pancreatitis, bilirubinemia greater than 4 mg/dL, or proven common duct stones on admission. The 2 groups had similar baseline characteristics, including similar preoperative health status scores. Overall, the number of confirmed common duct stones in each group was approximately 20%. Patients in the cholecystectomy-first group had a significantly shorter hospital length of stay than the classical treatment group (5 days vs 8 days; P < .001) and also had fewer common duct investigations (MRCP, EUS, or ERCP; 25 vs 71; P < .001). There were no significant differences detected between the 2 groups in quality-of-life assessments or in surgical or medical complications within 6 months.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Which strategy — initial cholecystectomy or common duct endoscopic investigation followed by cholecystectomy — leads to better outcomes in the treatment of suspected common duct stones?
Bottom line
As compared with a strategy of sequential common duct endoscopy followed by surgery, the use of initial cholecystectomy with intraoperative cholangiogram for patients at intermediate risk of choledocholithiasis leads to shorter hospital length of stay and fewer interventions with otherwise comparable outcomes. (LOE = 1b)
Reference
Iranmanesh P, Frossard J, Mugnier-Konrad B, et al. Initial cholecystectomy vs. sequential common duct endoscopic assessment and subsequent cholecystectomy for suspected gallstone migration. JAMA 2014;312(2):137-144.
Study design
Randomized controlled trial (nonblinded)
Funding source
Unknown/not stated
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
The role of common duct endoscopic investigation prior to cholecystectomy for patients at intermediate risk of choledocholithiasis is questionable as stones often migrate into the duodenum without such intervention. Using concealed allocation, investigators randomized 100 patients who presented to an emergency department with clinical suspicion of choledocholithiasis (right upper quadrant or epigastric abdominal pain with elevated liver function test results and ultrasound-confirmed gallstone) to 1 of 2 groups. The study group underwent primary cholecystectomy with intraoperative cholangiogram, while the classical treatment group first had common duct exploration with magnetic resonance cholangiopancreatography (MRCP), endoscopic ultrasound (EUS), and endoscopic retrograde cholangiopancreatography (ERCP), if needed, followed by cholecystectomy. The authors excluded patients with severe sepsis, pancreatitis, bilirubinemia greater than 4 mg/dL, or proven common duct stones on admission. The 2 groups had similar baseline characteristics, including similar preoperative health status scores. Overall, the number of confirmed common duct stones in each group was approximately 20%. Patients in the cholecystectomy-first group had a significantly shorter hospital length of stay than the classical treatment group (5 days vs 8 days; P < .001) and also had fewer common duct investigations (MRCP, EUS, or ERCP; 25 vs 71; P < .001). There were no significant differences detected between the 2 groups in quality-of-life assessments or in surgical or medical complications within 6 months.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Which strategy — initial cholecystectomy or common duct endoscopic investigation followed by cholecystectomy — leads to better outcomes in the treatment of suspected common duct stones?
Bottom line
As compared with a strategy of sequential common duct endoscopy followed by surgery, the use of initial cholecystectomy with intraoperative cholangiogram for patients at intermediate risk of choledocholithiasis leads to shorter hospital length of stay and fewer interventions with otherwise comparable outcomes. (LOE = 1b)
Reference
Iranmanesh P, Frossard J, Mugnier-Konrad B, et al. Initial cholecystectomy vs. sequential common duct endoscopic assessment and subsequent cholecystectomy for suspected gallstone migration. JAMA 2014;312(2):137-144.
Study design
Randomized controlled trial (nonblinded)
Funding source
Unknown/not stated
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
The role of common duct endoscopic investigation prior to cholecystectomy for patients at intermediate risk of choledocholithiasis is questionable as stones often migrate into the duodenum without such intervention. Using concealed allocation, investigators randomized 100 patients who presented to an emergency department with clinical suspicion of choledocholithiasis (right upper quadrant or epigastric abdominal pain with elevated liver function test results and ultrasound-confirmed gallstone) to 1 of 2 groups. The study group underwent primary cholecystectomy with intraoperative cholangiogram, while the classical treatment group first had common duct exploration with magnetic resonance cholangiopancreatography (MRCP), endoscopic ultrasound (EUS), and endoscopic retrograde cholangiopancreatography (ERCP), if needed, followed by cholecystectomy. The authors excluded patients with severe sepsis, pancreatitis, bilirubinemia greater than 4 mg/dL, or proven common duct stones on admission. The 2 groups had similar baseline characteristics, including similar preoperative health status scores. Overall, the number of confirmed common duct stones in each group was approximately 20%. Patients in the cholecystectomy-first group had a significantly shorter hospital length of stay than the classical treatment group (5 days vs 8 days; P < .001) and also had fewer common duct investigations (MRCP, EUS, or ERCP; 25 vs 71; P < .001). There were no significant differences detected between the 2 groups in quality-of-life assessments or in surgical or medical complications within 6 months.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
No Harm in Stopping Antibiotics After Cholecystectomy for Acute Cholecystitis
Clinical question
Does stopping antibiotic treatment after cholecystectomy for mild to moderate acute calculous cholecystitis affect outcomes?
Bottom line
Stopping antibiotic treatment after cholecystectomy for mild to moderate acute cholecystitis does not increase postoperative infection rates compared with a strategy of 5 days of postoperative antibiotics. (LOE = 1b)
Reference
Regimbeau JM, Fuks D, Pautrat K, et al, for the FRENCH Study Group. Effect of postoperative antibiotic administration on postoperative infection following cholecystectomy for acute calculous cholecystitis. JAMA 2014;312(2):145-154.
Study design
Randomized controlled trial (nonblinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Using concealed allocation, these investigators randomized 414 adult patients who presented to an emergency department with mild or moderate acute calculous cholecystitis requiring cholecystectomy into 2 groups: (1) continue taking antibiotics or (2) stop taking antibiotics during the postoperative period. Those with severe cholecystitis, defined as concomitant dysfunction of other organ systems, were excluded, as were those with acute pancreatitis, cholangitis, biliary peritonitis, or cirrhosis. All study patients received amoxicillin plus clavulanic acid 3 times a day from admission to day of surgery. The treatment group continued the same antibiotic regimen for 5 days after surgery, while the nontreatment group received no further antibiotics. The 2 groups were well balanced, with a mean age of 56 years and mean duration of preoperative antibiotics of 2 days. Approximately half the patients in each group had mild cholecystitis. For the primary outcome of postoperative surgical site or distant site infections at 4 weeks, there was no significant difference detected between the 2 groups in either the intention-to-treat or per-protocol analyses (intention-to-treat: 17% for nontreatment vs 15% for antibiotic group; per-protocol: 13% for both groups). This held true when the outcomes were analyzed according to severity of cholecystitis or duration of preoperative antibiotic use.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does stopping antibiotic treatment after cholecystectomy for mild to moderate acute calculous cholecystitis affect outcomes?
Bottom line
Stopping antibiotic treatment after cholecystectomy for mild to moderate acute cholecystitis does not increase postoperative infection rates compared with a strategy of 5 days of postoperative antibiotics. (LOE = 1b)
Reference
Regimbeau JM, Fuks D, Pautrat K, et al, for the FRENCH Study Group. Effect of postoperative antibiotic administration on postoperative infection following cholecystectomy for acute calculous cholecystitis. JAMA 2014;312(2):145-154.
Study design
Randomized controlled trial (nonblinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Using concealed allocation, these investigators randomized 414 adult patients who presented to an emergency department with mild or moderate acute calculous cholecystitis requiring cholecystectomy into 2 groups: (1) continue taking antibiotics or (2) stop taking antibiotics during the postoperative period. Those with severe cholecystitis, defined as concomitant dysfunction of other organ systems, were excluded, as were those with acute pancreatitis, cholangitis, biliary peritonitis, or cirrhosis. All study patients received amoxicillin plus clavulanic acid 3 times a day from admission to day of surgery. The treatment group continued the same antibiotic regimen for 5 days after surgery, while the nontreatment group received no further antibiotics. The 2 groups were well balanced, with a mean age of 56 years and mean duration of preoperative antibiotics of 2 days. Approximately half the patients in each group had mild cholecystitis. For the primary outcome of postoperative surgical site or distant site infections at 4 weeks, there was no significant difference detected between the 2 groups in either the intention-to-treat or per-protocol analyses (intention-to-treat: 17% for nontreatment vs 15% for antibiotic group; per-protocol: 13% for both groups). This held true when the outcomes were analyzed according to severity of cholecystitis or duration of preoperative antibiotic use.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does stopping antibiotic treatment after cholecystectomy for mild to moderate acute calculous cholecystitis affect outcomes?
Bottom line
Stopping antibiotic treatment after cholecystectomy for mild to moderate acute cholecystitis does not increase postoperative infection rates compared with a strategy of 5 days of postoperative antibiotics. (LOE = 1b)
Reference
Regimbeau JM, Fuks D, Pautrat K, et al, for the FRENCH Study Group. Effect of postoperative antibiotic administration on postoperative infection following cholecystectomy for acute calculous cholecystitis. JAMA 2014;312(2):145-154.
Study design
Randomized controlled trial (nonblinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
Using concealed allocation, these investigators randomized 414 adult patients who presented to an emergency department with mild or moderate acute calculous cholecystitis requiring cholecystectomy into 2 groups: (1) continue taking antibiotics or (2) stop taking antibiotics during the postoperative period. Those with severe cholecystitis, defined as concomitant dysfunction of other organ systems, were excluded, as were those with acute pancreatitis, cholangitis, biliary peritonitis, or cirrhosis. All study patients received amoxicillin plus clavulanic acid 3 times a day from admission to day of surgery. The treatment group continued the same antibiotic regimen for 5 days after surgery, while the nontreatment group received no further antibiotics. The 2 groups were well balanced, with a mean age of 56 years and mean duration of preoperative antibiotics of 2 days. Approximately half the patients in each group had mild cholecystitis. For the primary outcome of postoperative surgical site or distant site infections at 4 weeks, there was no significant difference detected between the 2 groups in either the intention-to-treat or per-protocol analyses (intention-to-treat: 17% for nontreatment vs 15% for antibiotic group; per-protocol: 13% for both groups). This held true when the outcomes were analyzed according to severity of cholecystitis or duration of preoperative antibiotic use.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Questionable Benefit of Oral Chlorhexidine in Mechanically Ventilated Patients
Clinical question
For patients receiving mechanical ventilation, does routine oral care with chlorhexidine improve outcomes?
Bottom line
Oral care with chlorhexidine decreases the incidence of respiratory tract infections in mechanically ventilated cardiac surgery patients, but not in noncardiac surgery patients. Additionally, the use of chlorhexidine does not have a statistically significant effect on mortality, length of stay (LOS), or duration of mechanical ventilation in either population. Further research is needed to determine the benefit and safety of this common intervention in different patient populations. (LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Unknown/not stated
Allocation
Uncertain
Setting
Inpatient (ICU only)
Synopsis
Previous meta-analyses demonstrated the effectiveness of chlorhexidine in the intensive care unit (ICU) for the prevention of ventilator-associated pneumonia, but did not distinguish between cardiac surgery and noncardiac surgery populations. These investigators searched multiple databases including PubMed and Embase, as well as reviewed reference lists of retrieved articles, to find randomized controlled trials that evaluated oral care with chlorhexidine versus oral care with placebo in mechanically ventilated patients for the prevention of nosocomial pneumonia, mortality, LOS, duration of mechanical ventilation, and antibiotic exposure. Two authors independently assessed the studies for inclusion and abstraction of data. The included studies were then rated for quality based on randomization strategy, allocation concealment, blinding, and completeness of follow-up. Of the 16 studies included in the review, 7 took place in combined medical–surgical units, 3 were in cardiac surgery units, and the rest were in other surgical or medical units. The presence of double-blinding in a study correlated with a higher quality rating. Data from the 3 studies of cardiac surgery patients showed fewer respiratory infections in patients receiving chlorhexidine (relative risk [RR] = 0.56; 95% CI, 0.41-0.77), but no association with mortality. The 13 studies involving noncardiac surgery patients, all of which focused specifically on ventilator-associated pneumonia, showed no significant difference in risk, even when the data was limited to the higher quality double-blind studies. However, there was a nonsignificant trend toward higher mortality with chlorhexidine use in this group of trials (RR = 1.13; 0.99-1.29). No significant differences were noted for the duration of mechanical ventilation, ICU LOS, hospital LOS, or antibiotic exposure, though data was limited for the latter 2 outcomes. The authors note that decreased exposure to an endotracheal tube in cardiac surgery patients, as compared with noncardiac surgery patients, may account for the decreased risk of pneumonia seen in this population. The potential increase in mortality in noncardiac surgery patients is harder to explain, but the authors postulate that this may be due to aspiration of chlorhexidine leading to acute lung injury.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients receiving mechanical ventilation, does routine oral care with chlorhexidine improve outcomes?
Bottom line
Oral care with chlorhexidine decreases the incidence of respiratory tract infections in mechanically ventilated cardiac surgery patients, but not in noncardiac surgery patients. Additionally, the use of chlorhexidine does not have a statistically significant effect on mortality, length of stay (LOS), or duration of mechanical ventilation in either population. Further research is needed to determine the benefit and safety of this common intervention in different patient populations. (LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Unknown/not stated
Allocation
Uncertain
Setting
Inpatient (ICU only)
Synopsis
Previous meta-analyses demonstrated the effectiveness of chlorhexidine in the intensive care unit (ICU) for the prevention of ventilator-associated pneumonia, but did not distinguish between cardiac surgery and noncardiac surgery populations. These investigators searched multiple databases including PubMed and Embase, as well as reviewed reference lists of retrieved articles, to find randomized controlled trials that evaluated oral care with chlorhexidine versus oral care with placebo in mechanically ventilated patients for the prevention of nosocomial pneumonia, mortality, LOS, duration of mechanical ventilation, and antibiotic exposure. Two authors independently assessed the studies for inclusion and abstraction of data. The included studies were then rated for quality based on randomization strategy, allocation concealment, blinding, and completeness of follow-up. Of the 16 studies included in the review, 7 took place in combined medical–surgical units, 3 were in cardiac surgery units, and the rest were in other surgical or medical units. The presence of double-blinding in a study correlated with a higher quality rating. Data from the 3 studies of cardiac surgery patients showed fewer respiratory infections in patients receiving chlorhexidine (relative risk [RR] = 0.56; 95% CI, 0.41-0.77), but no association with mortality. The 13 studies involving noncardiac surgery patients, all of which focused specifically on ventilator-associated pneumonia, showed no significant difference in risk, even when the data was limited to the higher quality double-blind studies. However, there was a nonsignificant trend toward higher mortality with chlorhexidine use in this group of trials (RR = 1.13; 0.99-1.29). No significant differences were noted for the duration of mechanical ventilation, ICU LOS, hospital LOS, or antibiotic exposure, though data was limited for the latter 2 outcomes. The authors note that decreased exposure to an endotracheal tube in cardiac surgery patients, as compared with noncardiac surgery patients, may account for the decreased risk of pneumonia seen in this population. The potential increase in mortality in noncardiac surgery patients is harder to explain, but the authors postulate that this may be due to aspiration of chlorhexidine leading to acute lung injury.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients receiving mechanical ventilation, does routine oral care with chlorhexidine improve outcomes?
Bottom line
Oral care with chlorhexidine decreases the incidence of respiratory tract infections in mechanically ventilated cardiac surgery patients, but not in noncardiac surgery patients. Additionally, the use of chlorhexidine does not have a statistically significant effect on mortality, length of stay (LOS), or duration of mechanical ventilation in either population. Further research is needed to determine the benefit and safety of this common intervention in different patient populations. (LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Unknown/not stated
Allocation
Uncertain
Setting
Inpatient (ICU only)
Synopsis
Previous meta-analyses demonstrated the effectiveness of chlorhexidine in the intensive care unit (ICU) for the prevention of ventilator-associated pneumonia, but did not distinguish between cardiac surgery and noncardiac surgery populations. These investigators searched multiple databases including PubMed and Embase, as well as reviewed reference lists of retrieved articles, to find randomized controlled trials that evaluated oral care with chlorhexidine versus oral care with placebo in mechanically ventilated patients for the prevention of nosocomial pneumonia, mortality, LOS, duration of mechanical ventilation, and antibiotic exposure. Two authors independently assessed the studies for inclusion and abstraction of data. The included studies were then rated for quality based on randomization strategy, allocation concealment, blinding, and completeness of follow-up. Of the 16 studies included in the review, 7 took place in combined medical–surgical units, 3 were in cardiac surgery units, and the rest were in other surgical or medical units. The presence of double-blinding in a study correlated with a higher quality rating. Data from the 3 studies of cardiac surgery patients showed fewer respiratory infections in patients receiving chlorhexidine (relative risk [RR] = 0.56; 95% CI, 0.41-0.77), but no association with mortality. The 13 studies involving noncardiac surgery patients, all of which focused specifically on ventilator-associated pneumonia, showed no significant difference in risk, even when the data was limited to the higher quality double-blind studies. However, there was a nonsignificant trend toward higher mortality with chlorhexidine use in this group of trials (RR = 1.13; 0.99-1.29). No significant differences were noted for the duration of mechanical ventilation, ICU LOS, hospital LOS, or antibiotic exposure, though data was limited for the latter 2 outcomes. The authors note that decreased exposure to an endotracheal tube in cardiac surgery patients, as compared with noncardiac surgery patients, may account for the decreased risk of pneumonia seen in this population. The potential increase in mortality in noncardiac surgery patients is harder to explain, but the authors postulate that this may be due to aspiration of chlorhexidine leading to acute lung injury.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Net Benefit With Azithromycin Use in Elderly Hospitalized Patients With Pneumonia
Clinical question
Is the use of azithromycin for older patients hospitalized with pneumonia associated with increased mortality or an increased risk of cardiovascular events?
Bottom line
For older patients hospitalized with pneumonia, the use of combination antibiotic therapy including azithromycin is associated with decreased mortality but increased risk of myocardial infarction (MI). You would need to treat 21 patients with azithromycin to prevent 1 death within 90 days; you would need to treat 144 patients to cause 1 MI. This results in a net benefit of 7 deaths prevented for 1 nonfatal MI induced with the use of azithromycin. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data from the Veterans Administration health care system, these authors examined the association of azithromycin with death and cardiovascular outcomes in older patients who were hospitalized with pneumonia. Patients included in the study were those aged at least 65 years who received antibiotic therapy per guidelines from the Infectious Diseases Society of America and American Thoracic Society for the treatment of community-acquired pneumonia. Primary outcomes were death at 30 days and 90 days, as well as cardiovascular events within 90 days. The cohort was divided into those who received combination therapy (which included azithromycin) and those who received other guideline-concordant antibiotics. Subsequently, propensity scores were used to match patients based on potential confounders -- such as age, intensive care unit (ICU) admission, and history of prior cardiac disease -- that could affect the severity of illness or outcomes. Almost 64,000 patients were included in the propensity-matched analysis. Patients had a mean age of 78 years, 16% were admitted to the ICU, and 5% received invasive mechanical ventilation. In this cohort, 90-day mortality was lower for azithromycin users (17% vs 22%; odds ratio [OR] = 0.76; 95% CI, 0.73-0.80). Although azithromycin users had more MIs (5.1% vs 4.4%; OR = 1.17; 1.08-1.25), there were no statistically significant differences in overall cardiac events, cardiac arrhythmias, or heart failure.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Is the use of azithromycin for older patients hospitalized with pneumonia associated with increased mortality or an increased risk of cardiovascular events?
Bottom line
For older patients hospitalized with pneumonia, the use of combination antibiotic therapy including azithromycin is associated with decreased mortality but increased risk of myocardial infarction (MI). You would need to treat 21 patients with azithromycin to prevent 1 death within 90 days; you would need to treat 144 patients to cause 1 MI. This results in a net benefit of 7 deaths prevented for 1 nonfatal MI induced with the use of azithromycin. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data from the Veterans Administration health care system, these authors examined the association of azithromycin with death and cardiovascular outcomes in older patients who were hospitalized with pneumonia. Patients included in the study were those aged at least 65 years who received antibiotic therapy per guidelines from the Infectious Diseases Society of America and American Thoracic Society for the treatment of community-acquired pneumonia. Primary outcomes were death at 30 days and 90 days, as well as cardiovascular events within 90 days. The cohort was divided into those who received combination therapy (which included azithromycin) and those who received other guideline-concordant antibiotics. Subsequently, propensity scores were used to match patients based on potential confounders -- such as age, intensive care unit (ICU) admission, and history of prior cardiac disease -- that could affect the severity of illness or outcomes. Almost 64,000 patients were included in the propensity-matched analysis. Patients had a mean age of 78 years, 16% were admitted to the ICU, and 5% received invasive mechanical ventilation. In this cohort, 90-day mortality was lower for azithromycin users (17% vs 22%; odds ratio [OR] = 0.76; 95% CI, 0.73-0.80). Although azithromycin users had more MIs (5.1% vs 4.4%; OR = 1.17; 1.08-1.25), there were no statistically significant differences in overall cardiac events, cardiac arrhythmias, or heart failure.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Is the use of azithromycin for older patients hospitalized with pneumonia associated with increased mortality or an increased risk of cardiovascular events?
Bottom line
For older patients hospitalized with pneumonia, the use of combination antibiotic therapy including azithromycin is associated with decreased mortality but increased risk of myocardial infarction (MI). You would need to treat 21 patients with azithromycin to prevent 1 death within 90 days; you would need to treat 144 patients to cause 1 MI. This results in a net benefit of 7 deaths prevented for 1 nonfatal MI induced with the use of azithromycin. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Government
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data from the Veterans Administration health care system, these authors examined the association of azithromycin with death and cardiovascular outcomes in older patients who were hospitalized with pneumonia. Patients included in the study were those aged at least 65 years who received antibiotic therapy per guidelines from the Infectious Diseases Society of America and American Thoracic Society for the treatment of community-acquired pneumonia. Primary outcomes were death at 30 days and 90 days, as well as cardiovascular events within 90 days. The cohort was divided into those who received combination therapy (which included azithromycin) and those who received other guideline-concordant antibiotics. Subsequently, propensity scores were used to match patients based on potential confounders -- such as age, intensive care unit (ICU) admission, and history of prior cardiac disease -- that could affect the severity of illness or outcomes. Almost 64,000 patients were included in the propensity-matched analysis. Patients had a mean age of 78 years, 16% were admitted to the ICU, and 5% received invasive mechanical ventilation. In this cohort, 90-day mortality was lower for azithromycin users (17% vs 22%; odds ratio [OR] = 0.76; 95% CI, 0.73-0.80). Although azithromycin users had more MIs (5.1% vs 4.4%; OR = 1.17; 1.08-1.25), there were no statistically significant differences in overall cardiac events, cardiac arrhythmias, or heart failure.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Increase in Hospitalist Workload Associated With Higher LOS and Cost
Clinical question
Does increased hospitalist workload affect efficiency and quality of care?
Bottom line
Increased hospitalist workload is associated with increased cost and length of stay (LOS). Quality metrics such as 30-day readmission rate, in-hospital mortality, and patient satisfaction were not affected by changes in workload. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Other
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data abstracted from the central data warehouse of an academic community health system, these authors examined the association between hospitalist workload and the efficiency and quality of care provided. A private hospitalist group that provided 24-hour care to patients at 2 hospitals within the system was selected for the study. Patients included were those who either had an attending of record or had admission and discharge bills submitted by a physician in this hospitalist group. Physician daily workload was measured using the total number of relative value units (RVUs) generated and the physician’s census as determined by the number of billable encounters submitted. Efficiency was measured by LOS and cost. Quality was measured by in-hospital mortality, rapid response team activation, 30-day readmission rate, and patient satisfaction. Models were adjusted for patient characteristics, including demographics; severity of illness; visit characteristics, including admission day of the week; and hospital-level characteristics, including hospital occupancy. Approximately 20,000 hospitalizations were included in the study. Hospitalists had a mean of 15.5 patient encounters and 28.6 RVUs per day. Results for LOS were stratified by hospital occupancy. For less than 75% occupancy, LOS increased linearly from 5.5 days to 7.5 days as workload increased. For greater than 85% occupancy, the change in LOS was J-shaped with a significant increase around an RVU of 30 or a census of 17. Cost also increased with higher workloads. For every unit increase in RVU, cost increased by $111; for every unit increase in census, cost increased by $205 (after adjustment for LOS). There were no significant associations with change in workload and patient satisfaction, in-hospital mortality, rapid response team activation, or 30-day readmission rate.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does increased hospitalist workload affect efficiency and quality of care?
Bottom line
Increased hospitalist workload is associated with increased cost and length of stay (LOS). Quality metrics such as 30-day readmission rate, in-hospital mortality, and patient satisfaction were not affected by changes in workload. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Other
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data abstracted from the central data warehouse of an academic community health system, these authors examined the association between hospitalist workload and the efficiency and quality of care provided. A private hospitalist group that provided 24-hour care to patients at 2 hospitals within the system was selected for the study. Patients included were those who either had an attending of record or had admission and discharge bills submitted by a physician in this hospitalist group. Physician daily workload was measured using the total number of relative value units (RVUs) generated and the physician’s census as determined by the number of billable encounters submitted. Efficiency was measured by LOS and cost. Quality was measured by in-hospital mortality, rapid response team activation, 30-day readmission rate, and patient satisfaction. Models were adjusted for patient characteristics, including demographics; severity of illness; visit characteristics, including admission day of the week; and hospital-level characteristics, including hospital occupancy. Approximately 20,000 hospitalizations were included in the study. Hospitalists had a mean of 15.5 patient encounters and 28.6 RVUs per day. Results for LOS were stratified by hospital occupancy. For less than 75% occupancy, LOS increased linearly from 5.5 days to 7.5 days as workload increased. For greater than 85% occupancy, the change in LOS was J-shaped with a significant increase around an RVU of 30 or a census of 17. Cost also increased with higher workloads. For every unit increase in RVU, cost increased by $111; for every unit increase in census, cost increased by $205 (after adjustment for LOS). There were no significant associations with change in workload and patient satisfaction, in-hospital mortality, rapid response team activation, or 30-day readmission rate.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does increased hospitalist workload affect efficiency and quality of care?
Bottom line
Increased hospitalist workload is associated with increased cost and length of stay (LOS). Quality metrics such as 30-day readmission rate, in-hospital mortality, and patient satisfaction were not affected by changes in workload. (LOE = 2b)
Reference
Study design
Cohort (retrospective)
Funding source
Other
Allocation
Uncertain
Setting
Inpatient (any location)
Synopsis
Using data abstracted from the central data warehouse of an academic community health system, these authors examined the association between hospitalist workload and the efficiency and quality of care provided. A private hospitalist group that provided 24-hour care to patients at 2 hospitals within the system was selected for the study. Patients included were those who either had an attending of record or had admission and discharge bills submitted by a physician in this hospitalist group. Physician daily workload was measured using the total number of relative value units (RVUs) generated and the physician’s census as determined by the number of billable encounters submitted. Efficiency was measured by LOS and cost. Quality was measured by in-hospital mortality, rapid response team activation, 30-day readmission rate, and patient satisfaction. Models were adjusted for patient characteristics, including demographics; severity of illness; visit characteristics, including admission day of the week; and hospital-level characteristics, including hospital occupancy. Approximately 20,000 hospitalizations were included in the study. Hospitalists had a mean of 15.5 patient encounters and 28.6 RVUs per day. Results for LOS were stratified by hospital occupancy. For less than 75% occupancy, LOS increased linearly from 5.5 days to 7.5 days as workload increased. For greater than 85% occupancy, the change in LOS was J-shaped with a significant increase around an RVU of 30 or a census of 17. Cost also increased with higher workloads. For every unit increase in RVU, cost increased by $111; for every unit increase in census, cost increased by $205 (after adjustment for LOS). There were no significant associations with change in workload and patient satisfaction, in-hospital mortality, rapid response team activation, or 30-day readmission rate.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Anticoagulant Plus NSAID or Aspirin Use Associated With Increased Bleeding Risk
Clinical question
For patients with acute venous thromboembolism, is the use of anticoagulants plus nonsteroidal anti-inflammatory drugs or aspirin associated with an increased risk of bleeding?
Bottom line
For patients treated with anticoagulants for acute venous thromboembolism (VTE), the use of concomitant nonsteroidal anti-inflammatory drugs (NSAIDs) or aspirin is associated with an increased risk of bleeding. This can occur in patients treated with rivaroxaban, as well as in those treated with enoxaparin plus a vitamin K antagonist (VKA). (LOE = 2b)
Reference
Study design
Cohort (prospective)
Funding source
Industry
Allocation
Uncertain
Setting
Outpatient (any)
Synopsis
The worldwide EINSTEIN DVT and EINSTEIN PE clinical trials compared rivaroxaban with enoxaparin plus a VKA for the treatment of acute VTE. These investigators used data from this cohort to examine the association between rivaroxaban or enoxaparin-VKA plus concomitant NSAID or aspirin use and the risk of clinically relevant and major bleeding. Clinically relevant bleeding was defined as nonmajor bleeding that resulted in either medical intervention, temporary cessation of study treatment, or patient discomfort. Major bleeding was that which resulted in death, occurred at a critical site, required tranfusion of at least 2 units of red blood cells, or was associated with a 2 g/dL drop in hemoglobin. The cohort comprised 8246 patients, with half receiving rivaroxaban and half receiving enoxaparin-VKA. The NSAID analysis was adjusted for sex, as there were more women in the NSAID group. The aspirin analysis was adjusted for age and creatinine clearance, as patients in the aspirin group tended to be older with worse kidney function than the nonaspirin group. The NSAID or aspirin exposure period included 7 days after stopping the medication to reflect an ongoing risk. Clinically relevant bleeding was more likely to occur in the NSAID group as compared with the non-NSAID group (37.5 bleeding events per 100 patient-years vs 16.6 per 100 patient-years; hazard ratio [HR] = 1.77; 95% CI, 1.46-2.14). Findings were similar for both the NSAID-rivaroxaban-treated patients and NSAID-enoxaparin-VKA-treated patients. In the aspirin group, clinically relevant bleeding was also increased as compared with the nonaspirin group (36.6 bleeding events per 100 patient-years vs 16.9 per 100 patient-years; HR = 1.70; 1.38-2.11). For both the NSAID and the aspirin groups, the bleeding events were spread evenly over the duration of the use, suggesting that longer duration of use does not increase the risk of bleeding. Major bleeding increased 2.4-fold in the NSAID group and 1.5-fold in the aspirin group.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients with acute venous thromboembolism, is the use of anticoagulants plus nonsteroidal anti-inflammatory drugs or aspirin associated with an increased risk of bleeding?
Bottom line
For patients treated with anticoagulants for acute venous thromboembolism (VTE), the use of concomitant nonsteroidal anti-inflammatory drugs (NSAIDs) or aspirin is associated with an increased risk of bleeding. This can occur in patients treated with rivaroxaban, as well as in those treated with enoxaparin plus a vitamin K antagonist (VKA). (LOE = 2b)
Reference
Study design
Cohort (prospective)
Funding source
Industry
Allocation
Uncertain
Setting
Outpatient (any)
Synopsis
The worldwide EINSTEIN DVT and EINSTEIN PE clinical trials compared rivaroxaban with enoxaparin plus a VKA for the treatment of acute VTE. These investigators used data from this cohort to examine the association between rivaroxaban or enoxaparin-VKA plus concomitant NSAID or aspirin use and the risk of clinically relevant and major bleeding. Clinically relevant bleeding was defined as nonmajor bleeding that resulted in either medical intervention, temporary cessation of study treatment, or patient discomfort. Major bleeding was that which resulted in death, occurred at a critical site, required tranfusion of at least 2 units of red blood cells, or was associated with a 2 g/dL drop in hemoglobin. The cohort comprised 8246 patients, with half receiving rivaroxaban and half receiving enoxaparin-VKA. The NSAID analysis was adjusted for sex, as there were more women in the NSAID group. The aspirin analysis was adjusted for age and creatinine clearance, as patients in the aspirin group tended to be older with worse kidney function than the nonaspirin group. The NSAID or aspirin exposure period included 7 days after stopping the medication to reflect an ongoing risk. Clinically relevant bleeding was more likely to occur in the NSAID group as compared with the non-NSAID group (37.5 bleeding events per 100 patient-years vs 16.6 per 100 patient-years; hazard ratio [HR] = 1.77; 95% CI, 1.46-2.14). Findings were similar for both the NSAID-rivaroxaban-treated patients and NSAID-enoxaparin-VKA-treated patients. In the aspirin group, clinically relevant bleeding was also increased as compared with the nonaspirin group (36.6 bleeding events per 100 patient-years vs 16.9 per 100 patient-years; HR = 1.70; 1.38-2.11). For both the NSAID and the aspirin groups, the bleeding events were spread evenly over the duration of the use, suggesting that longer duration of use does not increase the risk of bleeding. Major bleeding increased 2.4-fold in the NSAID group and 1.5-fold in the aspirin group.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
For patients with acute venous thromboembolism, is the use of anticoagulants plus nonsteroidal anti-inflammatory drugs or aspirin associated with an increased risk of bleeding?
Bottom line
For patients treated with anticoagulants for acute venous thromboembolism (VTE), the use of concomitant nonsteroidal anti-inflammatory drugs (NSAIDs) or aspirin is associated with an increased risk of bleeding. This can occur in patients treated with rivaroxaban, as well as in those treated with enoxaparin plus a vitamin K antagonist (VKA). (LOE = 2b)
Reference
Study design
Cohort (prospective)
Funding source
Industry
Allocation
Uncertain
Setting
Outpatient (any)
Synopsis
The worldwide EINSTEIN DVT and EINSTEIN PE clinical trials compared rivaroxaban with enoxaparin plus a VKA for the treatment of acute VTE. These investigators used data from this cohort to examine the association between rivaroxaban or enoxaparin-VKA plus concomitant NSAID or aspirin use and the risk of clinically relevant and major bleeding. Clinically relevant bleeding was defined as nonmajor bleeding that resulted in either medical intervention, temporary cessation of study treatment, or patient discomfort. Major bleeding was that which resulted in death, occurred at a critical site, required tranfusion of at least 2 units of red blood cells, or was associated with a 2 g/dL drop in hemoglobin. The cohort comprised 8246 patients, with half receiving rivaroxaban and half receiving enoxaparin-VKA. The NSAID analysis was adjusted for sex, as there were more women in the NSAID group. The aspirin analysis was adjusted for age and creatinine clearance, as patients in the aspirin group tended to be older with worse kidney function than the nonaspirin group. The NSAID or aspirin exposure period included 7 days after stopping the medication to reflect an ongoing risk. Clinically relevant bleeding was more likely to occur in the NSAID group as compared with the non-NSAID group (37.5 bleeding events per 100 patient-years vs 16.6 per 100 patient-years; hazard ratio [HR] = 1.77; 95% CI, 1.46-2.14). Findings were similar for both the NSAID-rivaroxaban-treated patients and NSAID-enoxaparin-VKA-treated patients. In the aspirin group, clinically relevant bleeding was also increased as compared with the nonaspirin group (36.6 bleeding events per 100 patient-years vs 16.9 per 100 patient-years; HR = 1.70; 1.38-2.11). For both the NSAID and the aspirin groups, the bleeding events were spread evenly over the duration of the use, suggesting that longer duration of use does not increase the risk of bleeding. Major bleeding increased 2.4-fold in the NSAID group and 1.5-fold in the aspirin group.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Prevention of HF Readmissions: Best Results With Home Visits and Multidisciplinary Clinics
Clinical question
Which transitional care interventions are most effective in preventing readmissions in patients hospitalized with heart failure?
Bottom line
Moderate-to-high strength of evidence supports the use of home-visiting programs and outpatient multidisciplinary heart failure (MDS-HF) clinics to reduce readmission rates and mortality for hospitalized HF patients at 3 months to 6 months. Structured telephone support may also decrease mortality and HF-specific readmissions, but does not affect all-cause readmissions.(LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
These authors searched multiple databases, including MEDLINE and the Cochrane Library, as well as reference lists of pertinent articles to find randomized controlled trials that compared transitional care interventions with either other interventions or usual care for patients hospitalized with HF with a focus on readmission rates and mortality. The interventions had to include at least one of the following components: patient/caregiver education, scheduled outpatient clinic visits, home visits, telemonitoring, structured telephone support, transition coaching, or increased provider continuity. Two authors independently reviewed the 47 selected studies for risk of bias. Studies with high or unclear risk of bias were not included in the meta-analysis. Results for readmission rates and mortality were stratified by intervention category and strength of evidence (SOE) grades were provided. For 30-day outcomes, a high-intensity home-visiting program consisting of 8 scheduled home visits starting at 24 hours after discharge decreased all-cause readmissions (number needed to treat [NNT] = 6) but the SOE was low. For 3-month to 6-month outcomes, data from 9 trials of home-visiting programs with varying intensities showed a reduction in all-cause readmissions (NNT= 9; SOE = high). Similarly, data from 2 trials of scheduled outpatient visits with MDS-HF clinics revealed lower readmission rates (NNT= 8; SOE = high). For HF-specific readmissions, 2 interventions, home-visiting programs and structured telephone support, decreased rates at 3 months to 6 months (home visits: NNT = 7; SOE = moderate; structured telephone support: NNT = 14; SOE = high). Finally, several interventions reduced mortality as compared to usual care at 3 to 6 months with moderate SOE, including home-visiting programs (NNT = 33), MDS-HF clinics (NNT = 18), and structured telephone support (NNT = 27). There were no significant benefits noted with scheduled visits to nurse-led clinics or primary care clinics or with educational interventions, although in most cases, there was insufficient evidence to determine whether these intervention were effective.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Which transitional care interventions are most effective in preventing readmissions in patients hospitalized with heart failure?
Bottom line
Moderate-to-high strength of evidence supports the use of home-visiting programs and outpatient multidisciplinary heart failure (MDS-HF) clinics to reduce readmission rates and mortality for hospitalized HF patients at 3 months to 6 months. Structured telephone support may also decrease mortality and HF-specific readmissions, but does not affect all-cause readmissions.(LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
These authors searched multiple databases, including MEDLINE and the Cochrane Library, as well as reference lists of pertinent articles to find randomized controlled trials that compared transitional care interventions with either other interventions or usual care for patients hospitalized with HF with a focus on readmission rates and mortality. The interventions had to include at least one of the following components: patient/caregiver education, scheduled outpatient clinic visits, home visits, telemonitoring, structured telephone support, transition coaching, or increased provider continuity. Two authors independently reviewed the 47 selected studies for risk of bias. Studies with high or unclear risk of bias were not included in the meta-analysis. Results for readmission rates and mortality were stratified by intervention category and strength of evidence (SOE) grades were provided. For 30-day outcomes, a high-intensity home-visiting program consisting of 8 scheduled home visits starting at 24 hours after discharge decreased all-cause readmissions (number needed to treat [NNT] = 6) but the SOE was low. For 3-month to 6-month outcomes, data from 9 trials of home-visiting programs with varying intensities showed a reduction in all-cause readmissions (NNT= 9; SOE = high). Similarly, data from 2 trials of scheduled outpatient visits with MDS-HF clinics revealed lower readmission rates (NNT= 8; SOE = high). For HF-specific readmissions, 2 interventions, home-visiting programs and structured telephone support, decreased rates at 3 months to 6 months (home visits: NNT = 7; SOE = moderate; structured telephone support: NNT = 14; SOE = high). Finally, several interventions reduced mortality as compared to usual care at 3 to 6 months with moderate SOE, including home-visiting programs (NNT = 33), MDS-HF clinics (NNT = 18), and structured telephone support (NNT = 27). There were no significant benefits noted with scheduled visits to nurse-led clinics or primary care clinics or with educational interventions, although in most cases, there was insufficient evidence to determine whether these intervention were effective.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Which transitional care interventions are most effective in preventing readmissions in patients hospitalized with heart failure?
Bottom line
Moderate-to-high strength of evidence supports the use of home-visiting programs and outpatient multidisciplinary heart failure (MDS-HF) clinics to reduce readmission rates and mortality for hospitalized HF patients at 3 months to 6 months. Structured telephone support may also decrease mortality and HF-specific readmissions, but does not affect all-cause readmissions.(LOE = 1a)
Reference
Study design
Meta-analysis (randomized controlled trials)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location) with outpatient follow-up
Synopsis
These authors searched multiple databases, including MEDLINE and the Cochrane Library, as well as reference lists of pertinent articles to find randomized controlled trials that compared transitional care interventions with either other interventions or usual care for patients hospitalized with HF with a focus on readmission rates and mortality. The interventions had to include at least one of the following components: patient/caregiver education, scheduled outpatient clinic visits, home visits, telemonitoring, structured telephone support, transition coaching, or increased provider continuity. Two authors independently reviewed the 47 selected studies for risk of bias. Studies with high or unclear risk of bias were not included in the meta-analysis. Results for readmission rates and mortality were stratified by intervention category and strength of evidence (SOE) grades were provided. For 30-day outcomes, a high-intensity home-visiting program consisting of 8 scheduled home visits starting at 24 hours after discharge decreased all-cause readmissions (number needed to treat [NNT] = 6) but the SOE was low. For 3-month to 6-month outcomes, data from 9 trials of home-visiting programs with varying intensities showed a reduction in all-cause readmissions (NNT= 9; SOE = high). Similarly, data from 2 trials of scheduled outpatient visits with MDS-HF clinics revealed lower readmission rates (NNT= 8; SOE = high). For HF-specific readmissions, 2 interventions, home-visiting programs and structured telephone support, decreased rates at 3 months to 6 months (home visits: NNT = 7; SOE = moderate; structured telephone support: NNT = 14; SOE = high). Finally, several interventions reduced mortality as compared to usual care at 3 to 6 months with moderate SOE, including home-visiting programs (NNT = 33), MDS-HF clinics (NNT = 18), and structured telephone support (NNT = 27). There were no significant benefits noted with scheduled visits to nurse-led clinics or primary care clinics or with educational interventions, although in most cases, there was insufficient evidence to determine whether these intervention were effective.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Ramelteon Reduces Risk of Delirium Among Hospitalized Patients
Clinical question
Does ramelteon, a melatonin agonist, prevent delirium in hospitalized patients?
Bottom line
In this small, single-blinded study, ramelteon was shown to be effective in preventing delirium in elderly patients who required hospitalization for acute illness. You would have to treat 3 patients with ramelteon to prevent one episode of delirium. (LOE = 1b)
Reference
Hatta K, Kishi Y, Wada K, et al, for the DELIRIA-J Group. Preventive effects of ramelteon on delirium. JAMA Psychiatry 2014;71(4):397-403.
Study design
Randomized controlled trial (single-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location)
Synopsis
Using concealed allocation, these investigators randomized 67 hospitalized patients (24 admitted to intensive care units, 43 admitted to general wards) to receive either ramelteon 8 mg or placebo nightly up to 7 days or until the onset of delirium. Eligible patients were 65 years to 89 years old, were admitted to the hospital via the emergency department, were able to take oral medications, and had an expected length of stay of greater than 48 hours. Patients with psychiatric disorders, severe liver disease, Lewy body disease, or alcohol dependency were excluded. Nurses provided similar delirium prevention care to all patients, including frequent reorientation, adequate lighting, and noise reduction. If patients required treatment for insomnia, hydroxyzine was used with a dose limit of 25 mg per night. Baseline characteristics were similar in the 2 groups, with a mean age of 78 years. For the primary outcome of onset of delirium, experienced psychiatrists, masked to study group, assessed patients in the mornings and afternoons for up to 7 days using a delirium rating scale. Only 1 patient in the ramelteon group was diagnosed with delirium as compared with 11 patients in the placebo group (3% vs 32%; number needed to treat = 3; P = .003). Interestingly, there were no significant differences between the groups in sleep metrics such as difficulty falling asleep and poor sleep quality, although the sample was likely too small to detect such differences. Note that the patients were not masked in this study, which could have potentially affected the overall outcomes. No adverse effects attributed to the study drug were reported.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does ramelteon, a melatonin agonist, prevent delirium in hospitalized patients?
Bottom line
In this small, single-blinded study, ramelteon was shown to be effective in preventing delirium in elderly patients who required hospitalization for acute illness. You would have to treat 3 patients with ramelteon to prevent one episode of delirium. (LOE = 1b)
Reference
Hatta K, Kishi Y, Wada K, et al, for the DELIRIA-J Group. Preventive effects of ramelteon on delirium. JAMA Psychiatry 2014;71(4):397-403.
Study design
Randomized controlled trial (single-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location)
Synopsis
Using concealed allocation, these investigators randomized 67 hospitalized patients (24 admitted to intensive care units, 43 admitted to general wards) to receive either ramelteon 8 mg or placebo nightly up to 7 days or until the onset of delirium. Eligible patients were 65 years to 89 years old, were admitted to the hospital via the emergency department, were able to take oral medications, and had an expected length of stay of greater than 48 hours. Patients with psychiatric disorders, severe liver disease, Lewy body disease, or alcohol dependency were excluded. Nurses provided similar delirium prevention care to all patients, including frequent reorientation, adequate lighting, and noise reduction. If patients required treatment for insomnia, hydroxyzine was used with a dose limit of 25 mg per night. Baseline characteristics were similar in the 2 groups, with a mean age of 78 years. For the primary outcome of onset of delirium, experienced psychiatrists, masked to study group, assessed patients in the mornings and afternoons for up to 7 days using a delirium rating scale. Only 1 patient in the ramelteon group was diagnosed with delirium as compared with 11 patients in the placebo group (3% vs 32%; number needed to treat = 3; P = .003). Interestingly, there were no significant differences between the groups in sleep metrics such as difficulty falling asleep and poor sleep quality, although the sample was likely too small to detect such differences. Note that the patients were not masked in this study, which could have potentially affected the overall outcomes. No adverse effects attributed to the study drug were reported.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.
Clinical question
Does ramelteon, a melatonin agonist, prevent delirium in hospitalized patients?
Bottom line
In this small, single-blinded study, ramelteon was shown to be effective in preventing delirium in elderly patients who required hospitalization for acute illness. You would have to treat 3 patients with ramelteon to prevent one episode of delirium. (LOE = 1b)
Reference
Hatta K, Kishi Y, Wada K, et al, for the DELIRIA-J Group. Preventive effects of ramelteon on delirium. JAMA Psychiatry 2014;71(4):397-403.
Study design
Randomized controlled trial (single-blinded)
Funding source
Government
Allocation
Concealed
Setting
Inpatient (any location)
Synopsis
Using concealed allocation, these investigators randomized 67 hospitalized patients (24 admitted to intensive care units, 43 admitted to general wards) to receive either ramelteon 8 mg or placebo nightly up to 7 days or until the onset of delirium. Eligible patients were 65 years to 89 years old, were admitted to the hospital via the emergency department, were able to take oral medications, and had an expected length of stay of greater than 48 hours. Patients with psychiatric disorders, severe liver disease, Lewy body disease, or alcohol dependency were excluded. Nurses provided similar delirium prevention care to all patients, including frequent reorientation, adequate lighting, and noise reduction. If patients required treatment for insomnia, hydroxyzine was used with a dose limit of 25 mg per night. Baseline characteristics were similar in the 2 groups, with a mean age of 78 years. For the primary outcome of onset of delirium, experienced psychiatrists, masked to study group, assessed patients in the mornings and afternoons for up to 7 days using a delirium rating scale. Only 1 patient in the ramelteon group was diagnosed with delirium as compared with 11 patients in the placebo group (3% vs 32%; number needed to treat = 3; P = .003). Interestingly, there were no significant differences between the groups in sleep metrics such as difficulty falling asleep and poor sleep quality, although the sample was likely too small to detect such differences. Note that the patients were not masked in this study, which could have potentially affected the overall outcomes. No adverse effects attributed to the study drug were reported.
Dr. Kulkarni is an assistant professor of hospital medicine at Northwestern University in Chicago.