Back to past leeches: repeated phlebotomies and cardiovascular risk

A relationship between iron stores and IR has been reported, but with some inconsistencies. The risk of diabetes was significantly lower in frequent blood donors from the Health Professionals Follow-up Study when compared with non-donors after 10 years of follow-up [11]. However, the beneficial effect of frequent blood donation disappeared at 12 years [12]. In the same cohort [12] and in post menopausal women from the Iowa Women's Health Study [13], heme-iron intake from red meat sources was associated with increased risk of diabetes.

Serum ferritin, a surrogate marker of iron status as it accurately reflects body iron stores in healthy individuals [14], was associated with increased risk of T2DM [15]. Subjects with hyperferritinemia had a 2.4-fold higher risk of developing T2DM [16]. This association was further confirmed in other studies [17‐19].

None of these aforementioned studies provided information on insulin resistance, which has been explored in cross-sectional investigations. Frequent blood donors had improved insulin sensitivity and decreased insulin secretion when compared with sporadic blood donors or non-donors [20]. In an unblinded and uncontrolled study of 31 subjects with glucose intolerance, serial phlebotomy individually adjusted to induce near iron deficiency was associated with improved insulin sensitivity in response to oral glucose loading and reduced levels of glycated haemoglobin [21]. In another unblinded, randomized study in 28 patients with high-ferritin T2DM, three serial 500-ml phlebotomies reduced mean serum ferritin from 460 to 232 ng/ml and simultaneously led to increased insulin sensitivity and reduced levels of glycated haemoglobin [8].

Concerning the relationship between iron stores, MetS and IR, a number of epidemiological studies showed a relationship [22‐24]. Hence, the lack of a significant effect of phlebotomies on IR in the present series is somewhat unexpected. HOMA is a poor predictor of insulin sensitivity in patients with T2DM, but, additionally, more than two phlebotomies and steadily reduced ferritin levels might be needed to see a significant effect of iron depletion on IR in patients with long-term diabetes and on multi-drug treatment as those evaluated in the Michalsen's series [4]. Indeed, significant effects after one year of treatment in high-ferritin T2DM patients [7] and after two years in those carrying hereditary hemochromatosis [10] have been observed. Moreover, drug treatment might have confounded results and thus, it would be important to study in more depth the effects on HOMA-IR by taking into account the use of medication (that is, separating those taking metformin vs. the other groups). Alternatively, the effects of blood donation on IR may differ in subjects with normal versus abnormal glucose tolerance as observed when exploring the effect of iron depletion on brachial artery flow-mediated dilation [25] and in keeping with those of Hirai et al. [26], who demonstrated differential effects of vitamin C in subjects with normal versus abnormal glucose tolerance.

Several epidemiological studies have investigated the effect of iron depletion on atherosclerosis, cardiovascular risk, mortality and morbidity [27]. Blood pressure is one of the main factors driving cardiovascular morbidity and mortality, but no evidence was provided of any reduction in blood pressure following phlebotomies.

By stratifying men from the Health Professionals Follow-up Study according to lifetime number of donations (0, 10 to 20 and ≥30), no difference was observed in the risk for hypertension [11]. The Kuopio Study was the study which reported significant differences in mean blood pressures between donors and non-donors, but such a difference might reflect different lifestyles [16].

Thus, one of the main questions is how two phlebotomies worked to reduce blood pressure. Improvement of IR is not a nominee, at least at a first glance. Michalsen et al. [4] argue that reduced low-grade inflammation and oxidative stress may play a role as iron-dependent hydroxyl radical formation can contribute to vascular dysfunction. In this regard, high-frequency blood donation was associated with reduced serum ferritin and increased brachial artery flow-mediated dilation compared with low-frequency blood donation [22]. Serum ferritin was significantly decreased and flow-mediated dilation was increased in high-frequency donors compared with low-frequency donors with no between-group difference in insulin sensitivity. The decrease in flow-mediated dilation during oral glucose tolerance testing did not differ between high- and low-frequency blood donors [22]. Alternatively, the authors [4] speculate that changes in blood viscosity caused vasodilatation and, in turn, reduced blood pressure.

On the other hand, as blood volume was not replaced following phlebotomies, patients on multi-drug therapy or with type 2 diabetes might also have a dysfunctional endothelium and sympathetic response to relative hypovolemia. They might be unable to compensate for hypovolemia as healthy donors do. Indeed, most previous evidence were from cohort and cross-sectional studies of healthy donors [20], and from high-ferritin T2DM patients and carriers of hereditary hemocromatosis in whom blood volume was restored to normal at each procedure [7, 10].

In conclusion, findings from Michalsen's study give new perspectives for prevention and treatment of the metabolic syndrome demonstrating that repeated phlebotomies, a low-cost and minimally invasive technique, are effective in reducing blood pressure with a mechanism that is independent of insulin resistance. Routine phlebotomies in these patients may enormously reduce health care costs related to the epidemic metabolic syndrome and, importantly, also contribute to increase the rate of blood donations.