REVIEWS A Systematic Review of Venous Thromboembolism Prophylaxis Strategies in Patients With Renal Insufficiency, Obesity, or on Antiplatelet Agents Sosena Kebede, MD, MPH1*, Kalpana R. Prakasa, MBBS, MS1, Kenneth Shermock, PharmD, PhD1,2, Hasan M. Shihab, MBChB, MPH2, Daniel J. Brotman, MD1, Ritu Sharma, BSc2, Yohalakshmi Chelladurai, MBBS, MPH2, Elliott R. Haut, MD3, Sonal Singh, MD, MPH1,2, Jodi B. Segal, MD, MPH1,2 1Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland; 2Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland; 3Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland. BACKGROUND: There is uncertainty about optimal strat- RESULTS: After a review of 30,902 citations, we identified 9 egies for venous thromboembolism (VTE) prophylaxis among controlled studies, 5 of which were trials, and the other 4 select populations such as patients with renal insufficiency, were observational studies. Five articles addressed prophy- obesity, or patients taking antiplatelet drugs including aspirin. laxis of patients with renal insufficiency, 2 addressed obese Their physiologies make prophylaxis particularly challenging. patients, and 2 addressed patients on antiplatelet agents. No study tested prophylaxis in underweight patients or PURPOSE: We performed a comparative effectiveness those with liver disease. The majority of observational review of the literature on efficacy and safety of VTE pro- studies had a high risk of bias. The strength of evidence phylaxis in these populations. ranged from low to insufficient regarding the comparative DATA SOURCES: We searched MEDLINE, EMBASE, SCO- effectiveness and safety of VTE prophylaxis among these PUS, CINAHL, International Pharmaceutical Abstracts, clini- patients. caltrial.gov, and the Cochrane Library through August 2012. CONCLUSION: The current evidence is insufficient regard- Eligible studies included controlled trials and observational ing optimal VTE prophylaxis for patients with renal insuffi- studies. ciency, obesity, or those who are on antiplatelet drugs DATA EXTRACTION: Two reviewers evaluated studies for including aspirin. High-quality studies are needed to inform eligibility, serially abstracted data, and independently eval- clinicians about the best VTE prophylaxis for these patients. uated the risk of bias and strength of evidence supporting Journal of Hospital Medicine 2013;8:394–401. VC 2013 interventions to prevent VTE in these populations. Society of Hospital Medicine Venous thromboembolism (VTE), including deep ve- also at high risk of bleeding, the most important com- nous thrombosis (DVT) and pulmonary embolism plication of VTE prophylaxis.3–6 (PE), is estimated to affect 900,000 Americans each The objectives of this systematic review were to year and is a cause of significant morbidity and mor- define the comparative effectiveness and safety of tality with associated high healthcare costs.1 Accord- pharmacologic and mechanical strategies for VTE pre- ingly, the comparative effectiveness and safety of vention in some of these select medical populations interventions for the prevention and treatment of VTE including obese patients, patients on concomitant anti- are among the national priorities for comparative platelet therapy, patients with renal insufficiency, effectiveness research.2 Whereas we have evidence- patients who are underweight, and patients with coa- based guidelines for the prophylaxis of VTE in the gulopathy due to liver disease. general population, there are no guidelines informing the care of select patient populations. Select popula- METHODS tions are those patients in whom there is decisional The methods for this comparative effectiveness review uncertainty about the optimal choice, timing, and (CER) follow the guidelines suggested in the Agency dose of VTE prophylaxis. Not only do these patients for Healthcare Research and Quality (AHRQ) have an increased risk of DVT and PE, but most are Methods Guide for Effectiveness and Comparative Effectiveness Reviews.7 The protocol was publically posted.8 *Address for correspondence and reprint requests: Sosena Kebede, MD, MPH, Department of Medicine, 600 North Wolfe Street, Nelson 215, Baltimore, MD 21287; Telephone: 443-287-4538; Fax: 410–502-0923; Search Strategy E-mail: [email protected] We searched MEDLINE, EMBASE, and SCOPUS Additional Supporting Information may be found in the online version of through August 2011, CINAHL, International Phar- this article. maceutical Abstracts, clinicaltrial.gov, and the Received: January 11, 2013; Revised: March 13, 2013; Accepted: Cochrane Library through August 2012. We devel- March 19, 2013 2013 Society of Hospital Medicine DOI 10.1002/jhm.2047 oped a search strategy based on medical subject head- Published online in Wiley Online Library (Wileyonlinelibrary.com). ings (MeSH) terms and text words of key articles that 394 An Official Publication of the Society of Hospital Medicine Journal of Hospital Medicine Vol 8 | No 7 | July 2013 Review of VTE Prophylaxis Strategies | Kebede et al we identified a priori9 (see the Appendix for search diagnoses prompting hospitalization were congestive strategy details). heart failure, respiratory failure, and infectious dis- eases. Compression ultrasound was performed in all Study Selection patients by day 21 of hospitalization. The primary We reviewed titles followed by abstracts to identify end point was the composite of VTE, fatal PE, and randomized controlled trials (RCTs) or observational sudden death, and secondary end points included studies with comparison groups reporting on the effec- DVT, bleeding, and thrombocytopenia by day 21 tiveness or safety of VTE prevention in our popula- (Table 1). In obese patients, the primary end point tions. Two investigators independently reviewed occurred in 2.8% (95% confidence interval [CI]: 1.3– abstracts, and we excluded the abstracts if both inves- 4.3) of the dalteparin group and in 4.3% (95% CI: tigators agreed that the article met 1 or more of the 2.5–6.2) of the placebo group (relative risk [RR]: exclusion criteria. We included only English-language 0.64; 95% CI: 0.32-1.28). In nonobese patients, the articles that evaluated the effectiveness of pharmaco- primary end point occurred in 2.8% (95% CI: 1.8- logical or mechanical interventions that have been 3.8) and 5.2% (95% CI: 3.9-6.6) of the dalteparin approved for clinical use in the United States. To be and placebo groups, respectively (RR: 0.53; 95% CI: eligible, the studies must have addressed relevant key 0.34-0.82). When weight was modeled as a continu- questions in the population of our interest. We ous variable, no statistically significant interaction resolved disagreements by consensus. We used Distill- between weight and dalteparin efficacy was observed erSR (Evidence Partners Inc., Ottawa, Ontario, Can- (P 5 0.97). The authors calculated the RR in prede- ada), a Web-based database management program to fined BMI subgroups and found that dalteparin was manage the review process. Two investigators assessed effective in reducing VTE in patients with BMIs up to the risk of bias in each study independently, using the 40, with RRs of <1.0 for all (approximate range, 0.2– Downs and Black instrument for observational studies 0.8). However, a fixed dose of dalteparin 5000 IU/day 10 and trials. was not better than placebo for individuals with BMI > 2 Data Synthesis 40 kg/m . There was no significant difference in mortality or major hemorrhage by day 21 between For each select population, we created detailed evi- treatment and placebo groups. dence tables containing the information abstracted Freeman and colleagues prospectively assigned 31 from the eligible studies. After synthesizing the evi- medically ill patients with extreme obesity (BMI >40 dence, we graded the quantity, quality, and consis- kg/m2) to 1 of 3 dosing regimens of enoxaparin: a tency of the best available evidence for each select fixed dose of 40 mg daily enoxaparin (control group, population by adapting an evidence-grading scheme n 5 11), enoxaparin at 0.4 mg/kg (n 5 9), or enoxa- recommended in the Methods Guide for Conducting parin at 0.5 mg/kg (n 5 11).12 The average BMI of the Comparative Effectiveness Reviews.7 entire cohort was 62.1 kg/m2 (range, 40.5–82.4). All RESULTS patients had anti-factor Xa levels drawn on the day of We identified 30,902 unique citations and included 9 enrollment and daily for 3 days (Table 2). The rela- studies (Figure 1). There were 5 RCTs with relevant tionship between anti-factor Xa levels and clinical effi- subgroups and 4 observational studies (Table 1). Two cacy of low-molecular weight heparin (LMWH) in studies reported on the risk of bleeding in patients VTE prophylaxis is still unclear; however, an anti- given pharmacologic prophylaxis while they are con- factor Xa level of 0.2 to 0.5 IU/mL, measured 4 hours comitantly taking nonsteroidal anti-inflammatory after the fourth dose of LMWH, is the target level rec- 13 drugs (NSAIDS) or antiplatelet agents/aspirin, 1 RCT ommended for VTE prophylaxis. Patients who and 1 prospective observational study reported on received weight-based enoxaparin at 0.5 mg/kg obese patients, and 5 studies described outcomes of achieved target anti-factor Xa level 86% of the time patients with renal insufficiency (see Supporting Infor- compared to 32% of the time in those receiving 0.4 mation, Table 1, in the online version of this article). mg/kg and 19% of the time for those in the fixed-dose No study tested prophylaxis in underweight patients group (P < 0.001). No clinical outcomes were or those with liver disease. reported in this study. Obese Patients Patients on Antiplatelet Drugs We found 1 subgroup analysis of an RCT (total 3706 We did not find studies that directly looked at the patients, 2563 nonobese and 1118 obese patients) that comparative effectiveness of VTE prophylaxis in reported on the comparative effectiveness and safety patients who were on antiplatelet drugs including as- of fixed low-dose dalteparin 5000 IU/day compared to pirin.