Uncertainty continues over aspirin in peripheral arterial disease

US guidelines recommend low dose aspirin for people with peripheral arterial disease, although the evidence supporting this recommendation is patchy and unconvincing. The latest meta-analysis was ultimately unsatisfying too, reporting a non-significant decrease in cardiovascular events of 12% among patients taking aspirin relative to controls (relative risk 0.88, 95% CI 0.76 to 1.04). The new analysis included 18 randomised controlled trials and 5269 patients with asymptomatic or symptomatic peripheral arterial disease. More than a quarter of the patients came from two trials that recruited only individuals with diabetes. Aspirin appeared to reduce the risk of non-fatal stroke (0.66; 0.47 to 0.94), but had no effect on any other outcome, including death.

These trials had the usual shortcomings: they were relatively small; many were old; and several tested higher doses of aspirin than doctors prescribe today. Some trials combined aspirin and dipyridamole. When all the trials were assessed together, the analysis had just enough power to exclude a 25% reduction in cardiovascular events. Aspirin could, however, be associated with a clinically useful reduction of less than 25%. Doctors should, therefore, continue to prescribe low dose aspirin while waiting for better evidence, says an accompanying editorial (pp 1927-8). At least one large trial is on the way and others must be done. Peripheral arterial disease is an important but long neglected aspect of cardiovascular medicine.

Standard chemotherapy prolongs survival in older women with breast cancer

Older women are more likely than younger women to have breast cancer, but much less likely to be included in clinical trials of chemotherapy. To redress the imbalance, researchers from the US recruited only women over 65 years of age for a trial comparing standard chemotherapy with the oral prodrug capecitabine. Standard chemotherapy was clearly better overall, despite being more toxic. Women given capecitabine were more likely to relapse (hazard ratio 2.09, 95% CI 1.38 to 3.17) …