Treatment with hydroxymethyl glutaryl coenzyme A reductase inhibitors (statins) has better supporting evidence. In the Scandinavia Simvastatin Survival Study (4S),15 202 patients with diabetes, coronary artery disease, and elevated cholesterol were randomized to receive simvastatin or placebo. Major coronary events were reduced by 55% with simvastatin (NNT=22.5 patient-years). Total mortality was reduced, but not to a statistically significant extent (P=.087; RRR=43%).
The Air Force/Texas Coronary Atherosclerosis Prevention Study33 (AFCAPS/TexCAPS) compared lovastatin with placebo for patients with diabetes and healthy individuals with normal cholesterol levels. Though the overall population demonstrated a 37% reduction in first coronary events, the diabetes subgroup had insufficient power to confirm this trend independently.
The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial34 (ALLHAT-LLT) randomized 3638 patients with diabetes to receive pravastatin 40 mg or placebo. No changes were seen in total mortality or cardiovascular events. These results may have been confounded by off-study prescription for statins, given to 32% of those randomized to the placebo group.
The best independent evidence for use of statins in diabetes comes from the Heart Protection Study16 of 3982 patients with diabetes with a total cholesterol level >135 mg/dL and no evidence of coronary disease. They were randomized to receive 40 mg simvastatin or placebo for 5 years. First major vascular events (MI or stroke) were decreased in the simvastatin group by 26% over 5 years (NNT=104 patient-years).
Applying the evidence. Reducing elevated triglycerides in type 2 diabetes is not supported by clear evidence, although elevated triglycerides may be associated with the metabolic syndrome and may warrant lifestyle change (SOR: C).
However, statins—even for those with a normal cholesterol level—reduce macrovascular outcomes (SOR: A), a measure now endorsed by the American College of Physicians.
Anti-platelet therapy
Aspirin prophylaxis in diabetes has weaker support. The Early Treatment Diabetic Retinopathy Study35 (ETDRS) randomized 3711 patients with diabetes to receive aspirin 650 mg or placebo. No benefit in mortality or cardiovascular event rates was documented after 5 years. However, the results did suggest a nonsignificant trend toward reduction of MI (RRR=17%; 95% CI, –4% to 34%; NNT=333).
In the Physician’s Health Study,36 22,071 participants (most of whom did not have diabetes) were randomized to receive aspirin 325 mg every other day or placebo. MI was reduced by 44% with aspirin (NNT=500). The risk of MI in the subgroup of 533 individuals with diabetes paralleled this reduction (RRR=39%), though, independently, the reduction in the subgroup did not reach statistical significance. Bleeding problems were the most common adverse effect of the aspirin, and were increased by 32% (NNH=78 person-years).
In the HOT study,11 a subgroup of 1501 patients with diabetes was randomized to receive aspirin 75 mg or placebo daily. The rate of MI trended downward with aspirin but was not statistically significant. Among the 18,790 patients in this study (most of whom did not have diabetes), MI was 36% less likely to occur with aspirin (NNT=770).
Applying the evidence. Overall, there is some suggestion that persons with cardiac risk factors, like type 2 diabetes, benefit by taking low-dose aspirin to avoid macrovascular complications. No study has confirmed this in a population of individuals with diabetes, but the trends suggest a possible benefit (SOR: C). The decision to use aspirin should be made in consultation with an informed patient.
SCREENING FOR NEUROPATHY
One trial17 has been conducted on screening for neuropathy. It compared monofilament testing and palpation of pedal pulses with “no special care.” Patients with an original positive screen result received a calculation of their ankle-brachial index, foot x-rays, and other measurements, and were referred to a high-risk podiatry clinic if the second level testing showed abnormal results. Major amputations were decreased in the screened group by 92% over 2 years (NNT=180 patient-years). This evidence of benefit is sufficient to recommend screening patients with diabetes for peripheral neuropathy or peripheral vascular disease, and appropriate referral (SOR: B). (See the Clinical Inquiry, “What is the best treatment for diabetic neuropathy?”)
Screening for nephropathy
One systematic review37 found no randomized trials of screening for urinary microalbumin and how it might affect overt nephropathy. However, several studies have looked at treatment of gross proteinuria, and have demonstrated some benefit in patient-oriented outcomes. Improved blood pressure control has reduced progression of nephropathy to end-stage renal disease.38,39 Other studies have suggested that an ACE inhibitor or ARB might provide benefit.30,39
The ADA recommends annual screening for microalbumin based on expert consensus (SOR: C).7 Though early detection of microalbumin might optimize the treatment of nephropathy, it is also possible that screening may detect a population whose disease would have remained subclinical indefinitely. No clear evidence suggests that screening for microalbumin reduces the incidence of patient-oriented outcomes.