All Content

ITL: Physician Reviews of HM-Relevant Research


In This Edition

Literature At A Glance

A guide to this month’s studies

  1. Warfarin vs. aspirin in heart-failure patients
  2. Aspirin after anticoagulation prevents VTE recurrence
  3. Novel oral anticoagulants vs. warfarin in afib patients
  4. Intravenous metronidazole for mild C. diff infection
  5. Frequency of Foley catheter complications vs. CAUTI
  6. Intensive hyperglycemia control in noncritical hospitalized patients
  7. Risk score predicts 30-day mortality after noncardiac surgery
  8. Temperature, white blood cell count are not sensitive predictors of bacteremia

Warfarin Reduces Stoke but Increases Hemorrhage Compared with Aspirin in Patients with Heart Failure and Sinus Rhythm

Clinical question: Is warfarin superior to aspirin with regard to clinical outcomes in patients with heart failure who are in sinus rhythm?

Background: Heart failure is associated with stroke and death potentially caused by atherothrombotic events. Anticoagulation is efficacious in some heart failure patients with atrial fibrillation or significant valvular disease, but the role of anticoagulation versus aspirin in patients with chronic heart failure and sinus rhythm is unclear.

Study design: Double-blind randomized controlled trial.

Setting: Multicenter, multinational trial involving outpatients.

Synopsis: This double-blind, double-dummy trial involving 2,305 patients with sinus rhythm and reduced left ventricular ejection fraction (<35%) showed no significant difference in the primary combined outcome (ischemic stroke, intracerebral hemorrhage, or death) in those treated with warfarin as compared with aspirin. Warfarin did significantly reduce the rate of ischemic stroke by 0.64 events per 100 patient-years (absolute risk reduction 2.2%, number needed to treat 45) when compared with aspirin, with no significant difference in the rate of intracerebral hemorrhage. This outcome was offset by an increased rate of major hemorrhage by 0.91 events per 100 patient-years (absolute risk increase 3.1%, number needed to harm 32).

This study included patients from all functional classes of heart failure, with a protocol to initiate treatment with other standard heart failure medications. Patients with an indication for either warfarin or aspirin were excluded. Due to recruitment difficulties, the power of this study was reduced. Other study limitations included a relatively low percentage of time that patients on warfarin were in therapeutic range and a substantial period of follow-up time in which patients did not receive the assigned study treatments.

Bottom line: The benefit of reduced stoke in patients with heart failure and sinus rhythm who take warfarin over aspirin is counteracted by an increased risk for serious bleeding outcomes.

Citation: Homma S, Thompson JLP, Pullicino PM, et al. Warfarin and aspirin in patients with heart failure and sinus rhythm. N Engl J Med. 2012;366(20):1859-1869.

Aspirin Use after Recommended Anticoagulation Treatment Prevents Recurrence in VTE Patients

Clinical question: Does aspirin prevent recurrence in patients after treatment with anticoagulation following their first episode of unprovoked VTE?

Background: VTE recurrence is common following the discontinuation of anticoagulation, particularly in patients with a history of unprovoked pulmonary embolism (PE) or DVT. Extension of treatment with vitamin K antagonists decreases recurrence but also causes more bleeding. The role of aspirin in the secondary prevention of VTE is unknown.

Study design: Double-blind randomized controlled trial.

Setting: Multicenter international trial involving outpatients.

Synopsis: This trial compared treatment with aspirin versus placebo for approximately two years in 205 patients with a history of unprovoked VTE who had completed six to 18 months of anticoagulant therapy. The relative risk reduction of recurrent VTE in the aspirin versus the placebo group was approximately 40% per year (6.6% vs. 11.2% per year; absolute risk reduction 4.6% per year; number needed to treat 22). No difference in major or clinically relevant bleeding was observed between the two groups.

This study was appropriately powered to detect the treatment effect reported by the authors. Typically, anticoagulation is discontinued in this specific patient population once the risk of bleeding and/or the patient’s perceived inconvenience of therapy outweighs the expected benefit from continuing treatment to prevent VTE recurrence. Study results suggest aspirin is effective in preventing recurrence while not conferring an increased risk for hemorrhage. It is important to note that patients with cancer and symptomatic atherosclerosis were excluded from this study.

Bottom line: Following an appropriate treatment period with standard anticoagulation, aspirin appears to be a safe and effective therapy in the secondary prevention of recurrence in patients with a history of unprovoked VTE.

Citation: Becattini C, Agnelli G, Schenone A, et al. Aspirin for preventing recurrence of venous thromboembolism. N Engl J Med. 2012;366(21):1959-1967.

Meta-Analysis Supports New Oral Anticoagulants in Patients with Atrial Fibrillation

Clinical question: Are novel oral anticoagulants both efficacious and safe in comparison to warfarin in patients with atrial fibrillation (afib)?

Background: Three large clinical trials recently evaluated novel oral anticoagulants (dabigatran, rivaroxaban, and apixaban) as alternatives to warfarin in afib patients. Although the anticoagulants all appeared efficacious for primary outcomes, results regarding secondary and safety outcomes were either inconclusive or heterogeneous.

Study design: Systematic review and meta-analysis of randomized controlled trials.

Setting: Three diverse, clinical-trial settings.

Synopsis: A systematic review and meta-analysis of randomized controlled trials comparing novel oral anticoagulants to warfarin in afib patients found three large studies examining dabigatran, rivaroxaban, and apixaban that included 44,563 patients.

Using random-effects models to pool data from these studies, the authors found a 22% relative risk (RR) reduction of stroke or systemic embolism with the use of these new anticoagulants as compared to warfarin. The risks for ischemic and unidentified stroke (RR 0.87), hemorrhagic stroke (RR 0.45), intracranial bleeding (RR 0.49), and mortality (RR 0.88) were significantly reduced in patients taking these new anticoagulants. There was no significant reduction in major bleeding.

A trend in favor of warfarin was seen for gastrointestinal bleeding, but this trend was not statistically significant (RR 1.25; 95% confidence interval, 0.91-1.72).

This meta-analysis was limited to only three randomized controlled trials, each comparing a different oral anticoagulant to warfarin and using somewhat heterogeneous study designs and patient populations.

Bottom line: Three new oral anticoagulants appear safe and more efficacious than warfarin in the prevention of stroke and systemic embolism, as well as other important clinical outcomes in patients with atrial fibrillation.

Citation: Miller CS, Grandi SM, Shimony A, Filion KB, Eisenberg MJ. Meta-analysis of efficacy and safety of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus warfarin in patients with atrial fibrillation. Am J Cardiol. 2012;110(3):453-60.

Observational Study Suggests Intravenous Metronidazole May Be Inferior for Mild Clostridium Difficile Infection

Clinical question: How do different treatment regimens (oral metronidazole vs. intravenous metronidazole vs. oral vancomycin) for mild Clostridium difficile infection (CDI) compare with regard to clinical outcomes?

Background: CDIs are increasing. Oral metronidazole or vancomycin can be used for the initial treatment of non-severe CDI. Additionally, a paucity of clinical data exists on intravenous metronidazole, which is used in some cases, typically in combination with vancomycin and in patients with severe CDI.

Study design: Prospective cohort study.

Setting: International hospital-based study.

Synopsis: This observational study assessed three different regimens (oral metronidazole, intravenous metronidazole, and oral vancomycin) for the treatment of mild CDI in 265 inpatients. Mortality at 30 days was higher in the intravenous metronidazole group (38.1%) than in the oral metronidazole (7.4%) and vancomycin (9.5%) groups. Patients receiving intravenous metronidazole also were less likely to recover from disease (52.4% versus >80% in the oral metronidazole and vancomycin groups). No statistically significant differences for other sequelae of CDI were found, except for higher risk of dehydration in the oral vancomycin group.

After controlling for age, sex, and comorbidity severity, patients in the intravenous metronidazole group were approximately four times more likely to die than patients in either the oral vancomycin group or the oral metronidazole group. There was no significant difference in mortality between the oral metronidazole and vancomycin groups.

Although this study was limited by its observational nature, no randomized controlled trial has yet to compare these three treatment regimens for mild CDI.

Bottom line: Treatment of mild C. diff infection with intravenous metronidazole appears to be associated with higher mortality and lower disease resolution than treatment with either oral metronidazole or vancomycin.

Citation: Wenisch JM, Schmid D, Kuo HW, et al. Prospective observational study comparing three different treatment regimes in patients with Clostridium difficile infection. Antimicrob Agents Chemother. 2012;56(4):1974-1978.

Traumatic Foley Catheter Complications Occur with Similar Frequency as Catheter-Associated Urinary Tract Infections

Clinical question: How does the frequency and clinical significance of urinary tract infection (UTI) compare to genitourinary trauma when associated with Foley catheterization?

Background: Foley catheter use in hospitalized patients is common and carries many associated risks. Although the risk of UTI when using catheters often is recognized, providers should consider other important complications of catheter use, such as genitourinary trauma.

Study design: Descriptive, prospective cohort study.

Setting: Single-center study in a Veterans Affairs hospital.

Synopsis: This surveillance study of 6,513 Foley catheter days compared the incidence of catheter-associated urinary tract infections (CAUTIs) to that of genitourinary trauma. Traumatic Foley events included gross hematuria, creation of false passage, ridging causing pain and/or difficulties with catheter removal, external genital trauma, catheter misplacement (ranging from prostatic to intraperitoneal), and catheter removal with the balloon inflated.

The incidence of possible UTI episodes was 1.8% of the Foley catheter days compared with 1.5% for catheter-associated trauma. Despite the fact that 72% of the UTI cases were asymptomatic, approximately 41% of these cases were treated with antibiotics, which accounted for 70% of all UTIs treated. Of the cases of Foley catheter trauma, 32% required further interventions (i.e. prolonged catheterization or cystoscopy). Trauma prompting intervention was as common as symptomatic UTIs.

The observational study design, use of a single center with a predominantly male population, and inclusion of patients undergoing urologic procedures who might have had valid indications for treatment of asymptomatic bacteriuria limit the study findings.

Bottom line: The complication of catheter-associated genitourinary trauma is just as common as CAUTIs in hospitalized patients, with each necessitating further intervention and treatment at similar rates.

Citation: Leuck AM, Wright D, Ellingson L, et al. Complications of Foley catheters—is infection the greatest risk? J Urol. 2012;187(5):1662-1666.

Intensive Control of Hyperglycemia in Noncritical, Hospitalized Patients Decreases Infection Risk, but No Significant Effect on Other Outcomes

Clinical question: What is the effect of tight glucose control in patients hospitalized in noncritical-care settings?

Background: Hyperglycemia is associated with increased in-hospital mortality and morbidity. Several trials have demonstrated the potential benefits of intensive glycemic control for patients in intensive-care settings, but this could lead to increased hypoglycemia. The effect of intensive therapy to achieve tight glycemic control in noncritically ill patients is unclear.

Study design: Systematic review and meta-analysis.

Setting: Various study sites of hospitalized patients.

Synopsis: This meta-analysis included 19 studies (nine randomized, 10 observational) published from 1995 to 2011. Intensive glycemic control (fasting blood glucose of 100-180 mg/dL) was not associated with significant effect on risk of death, myocardial infarction (MI), or stroke. There was a nonsignificant trend for increased risk of hypoglycemia (relative risk, 1.58; 95% CI, 0.97-2.57).

Intensive glycemic control was associated with a decreased risk of infection (relative risk, 0.41; 95% CI, 0.21–0.77), but this association mainly was derived from studies in surgical settings.

Subgroup analyses demonstrated an association between achieving intensive glycemic goals and an increased risk of hypoglycemia (P=0.01). Hypoglycemia also more commonly occurred in surgical patients.

There was substantial heterogeneity across studies included for all outcomes except infection. The quality of the current evidence supporting a reduction in infection is low and appears mostly derived from patients in surgical settings. The quality of evidence relating to all the other outcomes is also low and is limited by heterogeneity and imprecision.

Bottom line: Intensive control of hyperglycemia in noncritically ill hospitalized patients might reduce the risk of infection in some patients but does not appear to be significantly associated with improvement in other important clinical outcomes.

Citation: Murad MH, Coburn JA, Coto-Yglesias F, et al. Glycemic control in non-critically ill hospitalized patients: a systematic review and meta-analysis. J Clin Endocrinol Metab. 2012;97(1):49-58.

Simple Risk Score Predicts 30-Day Mortality after Noncardiac Surgery

Clinical question: Can 30-day mortality risk after noncardiac surgery be predicted by using a simple bedside-risk index?

Background: While indices exist to quantify the risk of cardiac complications in patients undergoing noncardiac surgery, there is significant perioperative mortality due to noncardiac causes. Therefore, a need exists for a simple risk index to predict all-cause mortality after noncardiac surgery.

Study design: Retrospective cohort study.

Setting: American College of Surgeons National Surgical Quality Improvement Program database of patients in more than 200 hospitals.

Synopsis: The nine-point Surgical Mortality Probability Model (S-MPM) 30-day mortality risk index was empirically created and applied to a randomly split portion of a database, which included 298,772 patients undergoing noncardiac surgery. Three risk factors were included: American Society of Anesthesiologists (ASA) physical status, surgery risk class, and emergency status. Patients with ASA physical status I, II, III, IV, or V were assigned zero, two, four, five, or six points, respectively. Patients undergoing intermediate- or high-risk procedures were assigned one or two points, respectively. Patients undergoing emergency procedures were assigned one point. The S-MPM then was applied to the validation portion of the data set.

The 30-day predicted risk of mortality was less than 0.50% for those patients with combined risk scores of less than five (S-MPM Class I), between 1.5% and 4.0% for risk scores of five to six (Class II), and more than 10% for risk scores greater than six (Class III).

The major limitation of this derived risk score is the reliance on ASA physical status, which has imprecise definitions and might lead to inconsistent ratings.

Bottom line: The S-MPM risk index is a simple bedside scoring system, which can accurately predict 30-day mortality in patients undergoing noncardiac surgery.

Citation: Glance LG, Lustik SJ, Hannan EL, et al. The Surgical Mortality Probability Model: Derivation and validation of a simple risk prediction rule for noncardiac surgery. Ann Surg. 2012;255(4):696-702.

Temperature, White Blood Cell Count Are Not Sensitive Predictors of Bacteremia

Clinical question: In ED patients with suspected infection, are temperature, white blood cell (WBC) count, and bandemia reliable predictors of bacteremia?

Background: Sepsis is a significant cause of morbidity and mortality. Early identification and treatment of sepsis improves patient outcomes. Although systemic inflammatory response syndrome criteria aids in the prompt recognition of sepsis, these markers have variable sensitivity and specificity for true infection or bacteremia, which places patients at high risk for sepsis.

Study design: Post-hoc data analysis of a prospective cohort study.

Setting: ED patients at a single tertiary-care center.

Synopsis: A retrospective analysis was performed on 3,563 patients who presented to the ED. Blood cultures obtained on patients with suspected infection revealed bacteremia in 289 patients (8.1%). Patients with bacteremia were reviewed for the presence of normal temperature (36.1° to 38° Celsius), normal WBC count (4,000 to 12,000 cells per µL), and presence of bandemia (>5% of WBC differential).

Among patients with bacteremia, 33% had a normal body temperature (67% sensitivity) and 52% had a normal WBC count (48% sensitivity). Of the 210 bacteremic patients who had a full differential performed, bandemia was present in 82% (sensitivity 82%). Bandemia was present in 80% of culture-positive patients with a normal temperature and 79% of culture-positive patients with a normal WBC count. Approximately 17% of patients with bacteremia had neither an abnormal temperature nor an abnormal WBC. This study was limited by the retrospective nature of analysis and the subjective designation of “suspected infection” by the original providers.

Bottom line: In patients presenting to the ED with suspected infection who are found to have culture-proven bacteremia, a significant percentage have normal temperature and WBC count, but the presence of bandemia could be more useful for identifying occult bacteremia.

Citation: Seigel TA, Cocchi MN, Salciccioli J, et al. Inadequacy of temperature and white blood cell count in predicting bacteremia in patients with suspected infection. J Emerg Med. 2012;42(3):254-259.

Clinical Shorts


Observational study examining patients discharged from the hospital finds approximately double 30-day mortality and 30-day readmission, while only 0.75 days less LOS for patients who left against medical advice versus those discharged conventionally.

Citation: Southern WN, Nahvia S, Arnsten JH. Increased risk of mortality and readmission among patients discharged against medical advice. Am J Med. 2012;125(6):594-602.


Meta-analysis evaluating the efficacy of probiotics finds a relative risk reduction of approximately 40% and a number needed to treat of 13 for the prevention and treatment of antibiotic-associated diarrhea.

Citation: Hempel S, Newberry SJ, Maher AR, et al. Probiotics for the prevention and treatment of antibiotic-associated diarrhea. JAMA. 2012;307(18):1959-1969.


Meta-analysis examining serious cardiovascular adverse events occurring during varenicline treatment or within 30 days of discontinuation finds no clinical or statistically significant differences when compared with placebo.

Citation: Prochaska JJ, Hilton JF. Risk of cardiovascular serious adverse events associated with varenicline use for tobacco cessation: systematic review and meta-analysis. BMJ. 2012;344:e2856.


A large, retrospective cohort study comparing patients with similar underlying medical conditions found a sevenfold increased risk of suicidal behavior and approximately double the risk for severe neuropsychiatric disorders for patients prescribed glucocorticoids.

Citation: Fardet L, Petersen I, Nazareth I. Suicidal behavior and severe neuropsychiatric disorders following glucocorticoid therapy in primary care. Am J Psychiatry. 2012;169(5):491-497.

Next Article:

   Comments ()