Elevated testosterone levels linked to excess stroke in children

February 1, 2010

A recent study suggests that a child's risk for arterial ischemic stroke (AIS) or cerebral sinovenous thromboembolism (CSVT) is related to endogenous testosterone concentrations.

A recent study suggests that a child's risk for arterial ischemic stroke (AIS) or cerebral sinovenous thromboembolism (CSVT) is related to endogenous testosterone concentrations. Children 18 years or younger with the highest levels of testosterone had the greatest risk for AIS/CVST, found Sandra Normann, MD, and colleagues.

Previous studies have found a gender imbalance in AIS/CVST, with boys affected at a greater rate than girls.

Included in the present study (Ann Neurol. 2009;66:754-758) were 124 unselected children 18 years or younger (57 females and 67 males) with AIS/CVST confirmed by neurovascular imaging. All had their plasma testosterone concentrations determined by immunoassay. This subset of children were single-center participants (Münster, Germany) of the larger International Pediatric Stroke Study, conducted in 30 centers in 10 countries, which enrolled 1,187 children with confirmed ischemic stroke.

Ten children with AIS (4 female, 6 male) and 10 with CVST (2 female, 8 male) had testosterone levels above the 90th percentile for age and gender, or 16.7% of the total cases, compared to only 2 of the 109 controls (1.8%; P=.002), found Dr Normann, who was with the department of pediatric hematology/oncology, University Children's Hospital, University of Münster, Münster, Germany, at the time of this analysis.

Univariate analysis revealed a 4- to 5-fold increase in the risk of overall thromboembolism for children with testosterone levels above the 90th age- and gender-dependent percentiles. The association between elevated testosterone level and increased risk of thromboembolism was statistically significant for boys but not for girls on univariate analysis.

When adjusting for pubertal status, hematocrit, and total cholesterol level, testosterone levels above the 90th percentile remained significantly associated with both AIS and CVST.

In boys, each 1 nmoL/L increase in testosterone was associated with a 32% increase in the odds of thromboembolism, and a 48% increase in the odds of CVST in particular.

Androgen interaction with platelets, coagulation and fibrinolytic proteins, and the vasculature have been proposed to explain the mechanism behind increased testosterone levels and increased risk of cerebral thromboembolism, note the authors. Testosterone may act as a mediator between the coagulation system and blood viscosity, they write, adding that interactions between estrogen and the intrinsic anticoagulant system have already been well established.

In an editorial to accompany the study, Susan Vannucci, PhD (department of pediatrics/newborn medicine at Weill Cornell Medical College in New York City) and Patricia D Hurn, PhD (departments of anesthesiology and perioperative medicine, physiology, and neurology at Oregon Health Sciences University, Portland), point out that testosterone levels were measured 6 to 12 months after the cerebral ischemic event in the cases, and these levels were assumed to reflect the circulating levels at the time of the event. Despite this limitation, they write that the study is "the first to address androgens in pediatric stroke and is consistent with the rather sparse literature that androgens impact ischemic outcomes and mechanisms of brain damage."