Hassanain Toma, MD

Arregui M, Buijsse B, Fritsche A, di Giuseppe R, Schulze MB, Sabine Westphal, et al. Adiponectin and Risk of Stroke: Prospective Study and Meta-analysis. Stroke. 2013

Adiponectin is a hormone that is derived from adipose tissue. It is believed to have anti-inflammatory, anti-atherogenic and insulin-sensitizing properties. Due to these favorable cardiovascular effects, Arregui et al. sought to study the relationship between plasma adiponectin and stroke, through a prospective case-cohort design, and meta-analysis of 9 prospective studies.



The authors found that in their study population, adiponectin was associated with higher HDL-cholesterol and lower C-reactive protein, triglyceride levels, and diabetes, interestingly however, they found adiponectin to be directly associated with stroke, with a hazard ratio of 1.31 (95% CI 1.04-1.64).

The association between adiponectin and stroke was lost when combining their data with 9 other studies through a meta-analysis, with a pooled relative risk was 1.08 (1.01-1.15).

I totally expected adiponectin to be inversely related to stroke risk because of its ability to provide overall favorable cardiovascular effects; improved lipid profile, anti-inflammatory and anti-diabetic effects. However, the results of the cohort study showed the complete opposite. The authors did a good job in providing possible explanation for this discrepancy. On the other hand, the pooled meta-analysis did not show an association between adiponectin and stroke. I am not a statistician, but on reviewing the pooled studies, I feel more confident in the results of their cohort population than in the meta-analysis. The previous studies were adjusted for different variables, and one study was unadjusted for any variables. This may have resulted in the loss of association between adiponectin and stroke. Perhaps an individual patient-level data meta-analysis would be a more meaningful study in this case.

Furthermore, adiponectin exists as globular-adiponectin and as full-length fractions of low-, medium-, and high-molecular weight. The authors noted that they only assessed the total adiponectin, and provided partial rationalization for it. This would be analogous to assessing your patients’ cardiovascular risk based on total cholesterol and omitting the value of LDL-cholesterol. As far as I am concerned, the race for understanding the relationship between adiponectin and stroke continues.