Background: Diabetic nephropathy resulting in end-stage renal disease is a reality for more than 44 percent of patients with diabetes mellitus. In addition to the medical challenges, treatment of nephropathy causes a serious financial burden, costing the United States more than $10 billion annually. Observational studies have suggested a relationship between elevated plasma homocysteine levels and the risk of developing diabetic nephropathy. House and colleagues hypothesized that B-vitamin therapy, which has been shown to lower homocysteine levels, would slow diabetic nephropathy and prevent subsequent vascular events, such as myocardial infarction and stroke.

The Study: The DIVINe (Diabetic Intervention with Vitamins to Improve Nephropathy) study, a multicenter, randomized, double-blind, placebo-controlled trial of patients with types 1 or 2 diabetes, was conducted between May 2001 and July 2007. Adults with diabetes and known renal disease were eligible, although those with stages 4 or 5 renal failure, expected survival of less than three years, creatinine clearance of less than 30 mL per minute per 1.73 m² (0.50 mL per second per m²), or those already on dialysis were excluded. In all, 238 patients were randomized to receive a daily B-vitamin supplement tablet that included folic acid (2.5 mg), vitamin B₆ (25 mg), and vitamin B₁₂ (1 mg) or a matching placebo. Follow-up occurred at six-month intervals for up to 36 months. Patients were permitted to take all other vitamin supplements but not additional doses of the B vitamins studied in the trial. A baseline glomerular filtration rate (GFR) and plasma homocysteine level were established for each patient, and levels were checked periodically throughout the study. The primary end point was progression of nephropathy, as measured by change in GFR. Secondary end points included dialysis, occurrence of vascular events, all-cause mortality, cognitive decline, and amputation.

Results: Although the initial trial size of 286 patients was designed to provide 80 percent power to detect a 25 percent reduction in GFR, the authors stopped enrollment after 252 patients when it was noted that the intervention group's GFR was falling faster than the predicted rate of the placebo group. The data and safety monitoring committee determined that continuing the study would not likely yield a significant benefit in the primary end point. Although the intervention group had significantly lower homocysteine levels than the placebo group at 36 months, the B-vitamin group had a statistically significant decrease in mean GFR (16.5 mL per minute per 1.73 m² [0.28 mL per second per m²]) compared with placebo (10.7 mL per minute per m² [0.18 mL per second per m²]; P = .02). Equal numbers of patients progressed to dialysis in each group, but patients in the B-vitamin group had higher rates of myocardial infarction, stroke, revascularization, and all-cause mortality. None of the individual events reached statistical significance, but the overall rate of secondary outcomes was statistically significant (P = .04).

Conclusion: The authors conclude that supplementation with high doses of B vitamins lowers plasma homocysteine levels but worsens diabetic nephropathy and increases the risk of cerebrovascular and cardiovascular events in patients with diabetes. Other nonvitamin methods of reducing plasma homocysteine levels should be investigated.