Vitamin supplementation in the secondary prevention of venous thromboembolism: about the VITRO study

To the editor:

We read with great interest and expectation the article by den Heijer et al¹  on a randomized, placebo-controlled double-blind trial in the secondary prevention of venous thromboembolism. We point out some issues to be clarified by the authors, and a general comment.

(1) The authors do not give any information on the prevalence of thrombophilic polymorphisms, such as factor V Leiden and prothrombin G20 210A, which are expected to be highly prevalent in these patients with no major acquired risk factors. (2) Were patients with venous thromboembolism on hormonal therapy equally distributed? The risk of recurrence in these patients is, in fact, different from those with ‘true’ idiopathic venous thrombosis. (3) In den Heijer,^{1(Table 1)} the range of geometric mean baseline homocysteine (Hcy) levels in the hyperhomocysteinemic group is reported to be 6.3 to 84.8 in the multivitamin group and 7.4 to 108.3 in the placebo group. Does this mean that patients with Hcy levels below the cut-off of hyperHcy were considered as hyperhomocysteinemic? On the other hand, the range of geometric mean baseline Hcy levels in the normohomocysteinemic group treated with multivitamins is reported to be 4.0 to 23.0. Does this mean that patients with Hcy levels above the reported cut-off were considered as normohomocysteinemic? (4) What was the renal function of patients enrolled? Also documented by Spence et al,²  patients with significant renal impairment do not respond to vitamin therapy, in particular to vitamin B12.

A limitation of the trial, as of other published trials (eg, VISP³ ), is the level of Hcy chosen to diagnose and treat hyperHcy. Indeed, the 75th percentile of Hcy distribution in controls identifies a quite low level of Hcy in the population of the Netherlands of 12.6 μM/L, which is even lower in the population enrolled in Milan (10.6 μM/L) and Wien. Lowering Hcy concentrations has been shown to have a significant effect on reducing the cardiovascular disease risk in patients with homocystinuria.^4,5  However, because of this cut off level, we would probably miss, once again, the opportunity to know if the lowering of intermediate Hcy levels rather than levels of 8 to 12 μM/L may be useful for reducing the recurrence rate.

The debate continues, and we still look forward to an adequately-sized study addressing patients not homocystinuric but with intermediate hyperHcy.