POEMs

				Using the search form

If you prefer to view this article in PDF form, click here.

Little information is available on racial response differences between antihypertensives

Clinical question Are there racial differences in response to antihypertensive medications?

Bottom line There are few high-quality studies of different antihypertensive medications that compare clinical outcomes of patients across different racial and ethnic groups. Most of the existing studies find no difference in outcomes between white and nonwhite patients. (Level of evidence = 1a–)

Synopsis These authors systematically reviewed multiple databases looking for clinical trials of antihypertensive medications that compared treatment response among patients of different racial groups and ethnicities. The searches were supplemented by hand-searching reference lists of other systematic reviews and practice guidelines, and by speaking with experts, but the authors don’t describe looking for unpublished data. Studies that did not report clinically important outcomes, such as mortality or cardiovascular events, were excluded. Two authors independently determined if studies were to be included in this review and they resolved disagreements by consensus. The authors then extracted data on race, ethnicity, and outcomes from each trial. Finally, they contacted the original study authors to obtain missing information. Of the 28 studies making the final cut, only eight studies described outcomes in nonwhite patients and only five compared outcomes across groups. The quality scores of the studies ranged from 3 to 5, using the Jadad score (range: 0-5). Given the variability in inclusion criteria, follow-up duration, treatments, and so forth, the authors appropriately refrained from pooling the results. Of the five studies comparing outcomes across ethnic and racial groups, four found no difference in outcomes. Only ALLHAT (JAMA. 2002;288[23]:2981-2997) found blacks experienced greater benefits from diuretics than nonblack patients.

Park IU, Taylor AL. Race and ethnicity in trials of antihypertensive therapy to prevent cardiovascular outcomes: a systematic review. Ann Fam Med. 2007;5(5):444-452.

Trial compares yields of CRC screening methods

Clinical question What are the yields for different approaches to colorectal cancer (CRC) screening?

Bottom line In order to have the true POEM on this topic, a randomized trial comparing colonoscopy, flexible sigmoidoscopy (FS), and fecal occult blood testing (FOBT) using an outcome of cancer specific or all-cause mortality needs to be conducted. In the meantime, this study found that (1) fewer patients accept an invitation for colonoscopy than for FS and FOBT, (2) colonoscopy and FS have a similar yield in patients aged 55 to 59 years, and (3) the yield of colonoscopy is better for patients aged 60 to 64 years. Thus, FS should remain a screening option, particularly given its greater acceptability, lower cost, and the larger number of clinicians who can perform it. (Level of evidence = 1b)

Synopsis This is the first large-scale trial comparing the yield of colonoscopy, FS, and FOBT in average-risk patients in the community. The researchers invited 20,042 patients aged 55 to 64 years in six Italian towns to participate, excluding 1,595 patients (largely because of previous diagnosis of polyps or colorectal cancer) and randomizing the remaining 18,447 patients to receive FOBT, FS, or colonoscopy. These patients were sent an invitation to participate in the study and have the test to which they were randomized. The attendance rate was 32.3% for FS and FOBT, but only 26.5% for colonoscopy (adjusted odds ratio = 0.74; 95% CI, 0.68-0.80). Attendance at colonoscopy was lower for older patients and for women, a pattern also seen for FS. The yield of advanced adenomas and colorectal cancer was 1.1% and 0.1% for FOBT, 4.6% and 0.6% for FS, and 6.3% and 0.8% for colonoscopy, but if you compare FS and colonoscopy in patients aged 55 to 59 years, there is no clinically meaningful difference: 5.2% and 0.7% for FS versus 5.8% and 0.5% for colonoscopy. In the older patients (60 to 64 years), however, more advanced adenomas (6.9% vs 3.8%) and colorectal cancers (1.1% vs 0.5%) were found with colonoscopy than with FS. The authors estimate that if you screen 10,000 people, FS would detect 168 patients with advanced neoplasm compared with 191 using colonoscopy.

Segnan N, Senore C, Andreoni B, et al; for the SCORE3 Working Group - Italy. Comparing attendance and detection rate of colonoscopy with sigmoidoscopy and FIT for colorectal cancer screening. Gastroenterology. 2007;132(7):2304-2312.

Intensive lipid lowering reduces mortality following acute coronary syndrome

Clinical question Does intensive statin therapy reduce mortality in patients with acute coronary syndrome (ACS)?

Bottom line Intensive statin therapy following ACS reduces cardiovascular and overall mortality. (Level of evidence = 1a)

Synopsis This study pooled results from two randomized trials of intensive lipid lowering versus moderate lipid lowering following ACS to attain a large enough sample to detect a mortality difference. The PROVE-IT-TIMI 22 trial enrolled patients within 10 days of ACS and randomized them to receive either atorvastatin 80 mg or pravastatin 40 mg. The A to Z trial enrolled patients within 5 days of ACS and compared simvastatin 40 mg for 1 month followed by simvastatin 80 mg with placebo for 4 months followed by simvastatin 20 mg. Mortality rates for this evaluation were calculated by pooled patient-level data. The two trials provided data from a total of 8,658 patients. At 4 months and 8 months, the reduction in LDL levels was greatest in the intensive statin group. At 2 years, all-cause mortality was reduced in the intensive statin group as compared with the moderate statin group. This result was primarily due to a reduction in cardiovascular deaths, which were significantly reduced with intensive therapy.

Murphy SA, Cannon CP, Wiviott SD, et al. Effect of intensive lipid-lowering therapy on mortality after acute coronary syndrome (a patient-level analysis of the Aggrastat to Zocor and Pravastatin or Atorvastatin Evaluation and Infection Therapy-Thrombolysis in Myocardial Infarction 22 trials). Am J Cardiol. 2007;100(7):1047-1051.

Epoetin alfa does not reduce transfusions in critically ill patients

Clinical question Does epoetin alfa reduce the need for transfusions and improve clinical outcomes in critically ill patients?

Bottom line In this trial, epoetin alfa did not reduce the need for transfusions in critically ill patients, but increased thrombotic events. Mortality was reduced among the trauma patients. (Level of evidence = 1b)

Synopsis All patients admitted to a medical or surgical ICU who remained there for at least 2 days were eligible for this study. Exclusion criteria included acute ischemic heart disease, a left ventricular assist device, history of arterial or venous thromboembolism or a hypercoagulable disorder, dialysis, BP greater than 200/100 mm Hg after therapy, seizures, extensive burns, and GI bleeding. A total of 1,460 patients were randomized to receive a subcutaneous injection of epoetin alfa 40,000 U on days 1, 8, and 15, or placebo. If the hemoglobin level was then higher than 12 g/dL, subsequent injections were withheld. All patients received iron supplementation. Transfusion was targeted to maintain hemoglobin between 7 g/dL and 9 g/dL, but was at the discretion of the individual physician. The primary end point was percentage of patients receiving red-cell transfusion by day 29. Other end points were need for transfusion by day 140, adverse events, and death. Analyses were done by intention to treat, and were reviewed by an independent consultant. Follow up was complete for 94% of patients at 29 days, and for 83% of patients at 140 days. Patients were matched for baseline characteristics. Approximately half of those enrolled were trauma patients. There was no difference in the percentage of patients who received a red-cell transfusion (epoetin 46% vs placebo 48%) or in the number of units transfused. When adjusted for clinical characteristics, including severity of illness and hemoglobin at baseline, the mortality reduction in the overall population did not reach statistical significance (hazard ratio [HR] for adjusted mortality at 29 days = 0.79; 95% CI, 0.56-1.10; and HR for adjusted mortality at 140 days = 0.86; 0.65-1.13). Adjusted mortality was significantly reduced at 29 and 140 days among trauma patients who received epoetin (number needed to treat [NNT] = 21; 6-1,133 for unadjusted mortality at 29 days). Patients who received epoetin had more thrombotic vascular events, primarily deep venous thrombosis (NNT to harm = 20; 12-70). There was also a trend toward an increased risk of MI among patients treated with epoetin.

Corwin HL, Gettinger A, Fabian TC, et al; for the EPO Critical Care Trials Group. Efficacy and safety of epoetin alfa in critically ill patients. N Engl J Med. 2007;357(10):965-976.

Pneumonia can be treated with antibiotic therapy for 3 to 5 days

Clinical question Can community-acquired pneumonia be treated with 3 to 5 days of antibiotic therapy?

Bottom line Ten to 14 days of antibiotics are no more effective in patients with community-acquired pneumonia than 3 to 5 days of treatment. Clinical failures and mortality were similar regardless of treatment length. The equivalent effectiveness was demonstrated with 3 to 5 days of oral or parenteral azithromycin, levofloxacin for 5 days, cefuroxime for 7 days, and intravenous ceftriaxone for 5 days. (Level of evidence = 1a)

Synopsis The researchers searched three databases for studies comparing short-course antibiotic monotherapy versus an extended course (more than 7 days) for the treatment of confirmed pneumonia. Most of the studies compared 3 to 5 days of treatment with 10 days of treatment. Two reviewers independently evaluated the articles for eligibility but the paper does not describe how the data were abstracted. The 15 identified studies included 2,796 total participants and compared short courses of oral macrolide antibiotics, fluoroquinolones, ketolides, or beta-lactam antibiotics in outpatients and inpatients. The authors found no difference in clinical failure rates between short-course and extended-course treatments, either when analyzed by intention-to-treat or per-protocol analysis, with failure rates in both groups of approximately 9%. Mortality was not different between the different treatments. Approximately half the studies were considered high quality (Jadad score of at least 3). Results were homogeneous and the authors found no evidence of publication bias.

Li JZ, Winston LG, Moore DH, Bent S. Efficacy of short-course antibiotic regimens for community-acquired pneumonia: a meta-analysis. Am J Med. 2007;120(9):783-790.

Controversial evidence on the value of pioglitazone for type 2 diabetes

Clinical question Is pioglitazone (Actos) useful in the management of patients with type 2 diabetes mellitus?

Bottom line Based only on studies provided by and conducted directly by the drug manufacturer, this review reports a significantly reduced incidence only of the composite outcome of death, MI, and stroke with pioglitazone therapy. None of the same outcomes were significantly reduced on an individual basis. An earlier review of pioglitazone therapy in type 2 diabetes by the Cochrane Collaboration of all available published and peer-reviewed clinical trials found no significant evidence of improved patient-oriented outcomes. Similar to the other thiazolidinedione, rosiglitazone (Avandia), pioglitazone also increases the risk of serious heart failure in patients with type 2 diabetes. (Level of evidence = 3a)

Synopsis A recently published systematic review (N Engl J Med. 2007; 356[24]:2457-2471) reported an increased risk of adverse cardiovascular events among patients with type 2 diabetes treated with rosiglitazone. In this issue of this journal, a second review of rosiglitazone also similarly reports an increased risk of adverse cardiovascular events (JAMA. 2007;298[10]:1189-1195). Thus, all eyes from clinicians and patients hoping to still receive benefit from treatment for type 2 diabetes with the thiazolidinediones are on the remaining drug, pioglitazone. A recent Cochrane review reporting on a meta-analysis of 22 peer-reviewed trials comprising approximately 6,200 patients found no evidence of improved patient-oriented outcomes with pioglitazone treatment. The same review did, however, report a significantly increased incidence of edema among treated patients. In this current review, the investigators were contracted to perform a meta-analysis of 19 trials, which enrolled a total of 16,390 patients, of pioglitazone therapy in patients with type 2 diabetes conducted by the manufacturer. The authors note that at least 20 other completed trials evaluating pioglitazone therapy not conducted directly by the manufacturer were not included in their summary analysis. No formal search of the medical literature was independently completed by the authors, including not reviewing any evidence-based resources or the Cochrane Registry of Controlled Trials. In this review the composite outcome of death, nonfatal MI, or nonfatal stroke was significantly reduced in patients treated with pioglitazone compared to placebo. However, as was also reported in the earlier Cochrane review, no significant benefit was found for any individual patient-oriented outcome, including all-cause mortality, MI, or stroke. Similarly to the two reviews of rosiglitazone, this review of pioglitazone did report a statistically significant increased incidence of serious heart failure in treated patients compared to controls.

Lincoff AM, Wolski K, Nicholls SJ, Nissen SE. Pioglitazone and risk of cardiovascular events in patients with type 2 diabetes mellitus. A meta-analysis of randomized trials. JAMA. 2007; 298(10):1180-1188.

Levels of evidence in Bottom line are explained at www.infopoems.com/levels.html.