Free and Total Protein S Antigen Levels on the Risk of Venous Thrombosis: Results From the MEGA Study

Conflicting data have been reported on the risk for venous thrombosis in individuals with low total protein S levels (i.e. type I protein S deficiency) and low free protein S levels with normal total protein S levels (i.e. type III protein S deficiency). This may be due to small numbers, wrong cut off level or inclusion of individuals with mild transitory decrements in protein S levels. Most studies that showed that type I and type III protein S deficiency were related with an increased risk for venous thrombosis, have been performed in thrombophilic families, suggesting that these deficiencies are inherited. As the prevalence of inherited type I or type III protein S deficiency is not known, the relevance of these findings within normal populations remain to be established.

To assess the risk of first venous thrombosis in persons with low levels of free protein S or total protein S in a large population-based case–control study.

MEGA study, 4956 consecutive patients aged 18 to 70 years with a first episode of venous thrombosis were included. Age- and sex-matched controls were partners of patients (n=3297) or individuals recruited by random digit dialing (n=3000). DNA was obtained by standard methods and was available for 4485 patients and 4889 control subjects. Citrated plasma was available for 2471 patients and 2940 controls. Molecular basis for protein S deficiency was investigated by analysis of copy number variation of PROS1 and sequencing of individuals with the lowest levels of protein S in attempt to explain the different findings in risk estimates between families and population studies.

Odds ratios were adjusted for age and sex (matching factors) for levels of free/total protein S and their 95% confidence levels (95% CIs) with the use of logistic regression. The 2.5^th-97.5^th percentile of both total and free protein S in control subjects that did not use vitamin K antagonists (VKA) at time of blood draw were considered as the reference range. Individuals that used VKA at time of blood draw were excluded when calculating relative risk estimates. Furthermore, a preplanned sensitivity analysis was performed where we excluded estrogen users and pregnant women at time of blood draw.

Individuals with low free protein S levels or low total protein S levels (<2.5^th percentile) were not at increased risk of venous thrombosis as compared to individuals with protein S levels in the 2.5^th-97.5^th percentile; odds ratio 0.82 (95% CI, 0.56–1.21) and 0.90 (95% CI, 0.62–1.31) respectively. Excluding all women who used estrogens or were pregnant/puerperic at time of venous thrombosis or at time of blood sampling increased the odds ratios slightly to 1.55 (95% CI, 0.84–2.88) for individuals with low free protein S levels and to 1.28 (95% CI, 0.70–2.35) in individuals with low total protein S levels. We subsequently compared decreasing cut off values of free and total protein S levels on the risk of venous thrombosis as compared to the same reference group. Although numbers became small, it appeared that a free protein S cut off level of < 0.20^th or < 0.10^th percentile could identify individuals at high risk of venous thrombosis (odds ratios 2.01; 95% CI, 0.57–7.15, and 5.44; 95% CI, 0.61–48.78, respectively). Even extremely low (<0.10^th percentile) total protein S levels were not associated with venous thrombosis. Only one patient had a copy number variation of PROS1 in 2270 consecutive samples tested. Currently, we are sequencing the PROS1 gene in all individuals with protein S levels <1^st percentile of which results will be available before the ASH conference of 2011.

Low free protein S and low total protein S levels could not identify individuals at risk for venous thrombosis in a population based study. Although extremely low free protein S levels were associated with an increased risk for venous thrombosis, numbers were too small to support testing on free protein S in an unselected group of venous thrombosis patients.

No relevant conflicts of interest to declare.