Disclosure: The authors declared no conflict of interest.
Primary antiphospholipid antibody syndrome—one further aspect of thrombophilia in overweight and obese patients with venous thromboembolism
Article first published online: 13 MAY 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 9, pages E463–E466, September 2013
How to Cite
Gary, T., Belaj, K., Bruckenberger, R., Hackl, G., Hafner, F., Froehlich, H., Zebisch, A., Pilger, E. and Brodmann, M. (2013), Primary antiphospholipid antibody syndrome—one further aspect of thrombophilia in overweight and obese patients with venous thromboembolism. Obesity, 21: E463–E466. doi: 10.1002/oby.20188
- Issue published online: 23 SEP 2013
- Article first published online: 13 MAY 2013
- Accepted manuscript online: 29 NOV 2012 04:10PM EST
- Manuscript Accepted: 12 NOV 2012
- Manuscript Received: 26 JUN 2012
Objective: Overweight and obesity are established risk factors for venous thromboembolism (VTE). We examined the difference in the frequency of primary antiphospholipid antibody syndrome (PAPS) in VTE patients according to their BMI.
Design and Methods: We included 998 VTE patients treated at our institution between 2009 and 2011 in a retrospective data analysis. Thrombophilia screening including evaluation for APS (lupus anticoagulant, anti-cardiolipin, and anti-B2-glycoprotein-I IgG and IgM antibodies) was performed in all patients.
Results: PAPS was diagnosed in 6.8% (24/355) of normal weight (BMI < 24 kg/m2) VTE patients, in 11.1% (50/452) of overweight (BMI 25–30 kg/m2) VTE patients, and in 15.7% (30/191) of obese (BMI > 31 kg/m2) VTE patients. The difference of PAPS occurrence between these groups was statistically significant (P = 0.001). PAPS patients demonstrated higher fibrinogen levels as compared to non-PAPS patients (median 416.0 md/dl vs. 352.0 mg/dl, P = 0.001). Furthermore, fibrinogen levels increased significantly according to the body weight of patients (median normal weight patients 330.0 mg/dl vs. overweight patients 359.0 mg/dl vs. obese patients 415.0 mg/dl, P = 0.001).
Conclusion: PAPS seems to be more frequent in overweight and obese patients. As PAPS patients showed significantly higher fibrinogen levels and as fibrinogen levels increased significantly according to the body weight of patients, an elevated inflammatory state in overweight and obese patients as a reason for the increased PAPS occurrence can be assumed.
Overweight and obesity are established risk factors for development of venous thromboembolism (VTE). Reasons therefore are various. In a recent study, the combination of obesity and a tall stature was associated with increased risk of VTE, particularly in men. The authors therefore assumed synergistic effects between obesity and height in VTE development .
The antiphospholipid antibody syndrome (APS) is a thrombophilic diathesis leading to an increased rate of thrombotic events, both in the arterial and venous system, as well as to recurrent fetal loss. Diagnosis is based on the serologic presence of antiphospholipid antibodies, including anti-cardiolipin antibodies, anti-B2-glycoprotein-I antibodies, and lupus anticoagulant (LAC), respectively. To justify the diagnosis of APS, antibodies have to be detectable on two or more occasions at least 12 weeks apart . Secondary APS with an underlying autoimmunological condition can be distinguished from primary APS (PAPS). The reasons causing thrombophilia in patients with APS have not been studied in detail. The imbalance of pro- and antithrombotic mechanisms toward a hypercoagulable state was investigated in the last decades. High induced tissue factor expression was described as well as resistance to annexin A5 binding and anticoagulant activity in APS patients [3, 4].
A possible link between overweight and APS has been described previously in APS patients with thrombotic events in the arterial system . In a study by Medina et al., patients with PAPS had a higher prevalence of elevated carotid artery intima media thickness (IMT) and a decreased lumen diameter. Furthermore, patients differed statistically significant concerning their BMI. In another study by the same group, PAPS patients demonstrated a high prevalence of metabolic syndrome, although it has to be noted that the authors failed to prove a statistically significant difference when they compared PAPS patients to the general population .
As overweight and obesity are established VTE risk factors and as PAPS patients show a worse outcome in case of coexistence of obesity , we investigated the distribution of PAPS in VTE patients in different weight groups in a retrospective data analysis.
We included 998 consecutive VTE patients treated at our outpatient clinic between 2009 and 2011 in a retrospective data analysis. Pulmonary embolism (PE) was diagnosed using high-resolution CT scans of the pulmonary arteries and deep vein thrombosis (DVT) was diagnosed by compression ultrasound (CUS), respectively. The study was approved by the local ethics committee.
Thrombophilia screening was performed in each of our patients during a regular visit at our outpatient clinic in a clinically stable condition at least three months after the acute VTE event. Serologic screening included analysis of fibrinogen, deficiency of antithrombin, protein C/S, LAC, and anti-cardiolipin or anti-B2-glycoprotein-I IgG and IgM antibodies. Genetic evaluation comprised screening for mutations in factor V Leiden and prothrombin (G20210A), respectively. LAC detection was performed in our laboratory according to current recommendations .
In case of pathological results, APS was diagnosed according to published guidelines . During the visit at our outpatient clinic, demographic data including body weight and height were obtained and BMI was calculated.
Continuous variables are described as median and its 25th and 75th percentile, categorical variables are described by absolute numbers and percentages. Differences of groups were calculated using Chi square test. In case of non-normal distribution of data, Mann–Whitney U (comparison between two groups) and Kruskall–Wallis (comparison between more than two groups) tests were applied. All statistical analyses were performed using SPSS 17.0.
Table 1 shows the patients' characteristics as well as differences between different weight groups. We included 998 VTE patients in our data analysis. Of the patients included, 104 were diagnosed with PAPS according to current guidelines . In all of these patients a positive LAC was observed. In three patients we were also able to observe a positive test on anti-B2-glycoprotein-I antibodies, and in seven patients a positive test on anti-cardiolipin antibodies.
|Characteristic||Normal weight (BMI ≤24 kg/m2)||Overweight (BMI 25–30 kg/m2)||Obese (BMI ≥31 kg/m2)||P-value|
|n (%)||355 (35.6)||452 (45.3)||191 (19.1)|
|Female sex, n (%)||173 (48.7)||223 (49.3)||93 (48.7)||0.8|
|BMI, mean (±SD)||22.1 (±1.8)||27.1 (±1.6)||34.7 (±3.6)||0.001|
|Age, mean (±SD)||44 (±18)||50.5 (±16.4)||49.1 (±16.5)||0.6|
|PAPS, n (%)||24 (6.8)||50 (11.1)||30 (15.7)||0.001|
|Positivity for LAC, n (%)||24 (6.8)||50 (11.1)||30 (15.7)|
|Positivity for anti-B2-glycoprotein I IgG and IgM antibodies, n (%)||1 (0.3)||2 (0.4)||0 (0)|
|Positivity for anti-cardiolipin IgG and IgM antibodies, n (%)||2 (0.6)||3 (0.7)||2 (1.0)|
|CAD, n (%)||8 (2.3)||9 (2.0)||9 (4.7)||0.02|
|CVD, n (%)||11 (3.1)||13 (2.9)||6 (3.1)||0.9|
|PAOD, n (%)||8 (2.3)||13 (2.9)||5 (2.6)||0.2|
|Estrogen therapy, n (%)||11 (3.1)||11 (2.4)||6 (3.1)||0.2|
|Smoking, n (%)||101 (28.5)||152 (33.6)||58 (30.4)||0.1|
|Heterozygous factor V Leiden Mutation, n (%)||26 (7.3)||37 (8.2)||11 (5.8)||0.1|
|Homozygous factor V Leiden Mutation, n (%)||2 (0.6)||3 (0.7)||1 (0.5)||0.9|
|Heterozygous Prothrombin G20210A Mutation, n (%)||17 (4.8)||19 (4.2)||10 (5.2)||0.5|
|Fibrinogen mg/dl, Median (25th, 75th percentile)||330 (270, 390)||359 (297, 439)||415 (322, 482)||0.001|
We divided the cohort into three groups concerning to the patient's BMI (Table 1). Normal weight was defined by a BMI of <24 kg/m2, overweight by a BMI of 25–30 kg/m2, and obesity by a BMI of >31 kg/m2, respectively. The occurrence of PAPS was strictly BMI dependent (Table 1) with higher BMI relating to a higher rate of PAPS occurrence (P = 0.001) as calculated by Chi square test.
In a second step, differences in fibrinogen, a common marker of inflammation, were calculated between PAPS and non-PAPS patients, respectively. We observed statistically significant higher fibrinogen levels in PAPS compared to non-PAPS patients (median 416.0 mg/dl, 25th percentile 338.0, 75th percentile 515.3 vs. median 351.0 mg/dl, 25th percentile 290.0, 75th percentile 426.8, P = 0.001).
In a third step, we calculated differences in fibrinogen levels between the three different weight groups. We found a statistically significant increase in fibrinogen levels according to the body weight of patients (Table 1).
The thrombophilic situation occurring in presence of antiphospholipid antibodies was recently described as a “two-hit” situation . The “first hit” seems to be the procoagulant state induced by the presence of the antibodies. However, as there are many patients with antibodies and without thrombotic events a “second hit” has to occur, finally enabling development of the thrombotic event. Inflammation and trauma, respectively, have been discussed as such “second hit”-situations in the literature so far. Inflammation might lead to cell activation which in turn results in a further thrombophilic state.
APS is a highly thrombophilic condition leading to thrombosis in both, the arterial system and the venous system, as well as to recurrent fetal loss. Differences in BMI between APS patients with clinical manifestation in the arterial system have already been described previously . A similar phenomenon has also been described in premenopausal women with PAPS. Compared to healthy controls, patients diagnosed with PAPS were more obese and had both, a higher mass and percentage of fat, than healthy controls .
Differences in BMI between APS patients with clinical manifestation in the venous system have not been investigated so far. Sailer et al. coworkers described differences in inflammatory parameters between healthy controls and APS patients with and without venous thrombosis . Inflammatory parameters were higher in patients with antibodies as compared to healthy controls. However, the authors failed to show a significant difference in the parameters between APS patients with venous thrombosis and those without. The authors therefore concluded that development of venous thrombosis in APS patients is not caused by the inflammatory process. However, in an editorial on this paper, Bobba et al. discussed the data in a different light . They criticized that the discrepancies in the work of Sailer et al. could possibly be explained by the fact that blood samples were taken at random time points rather than at the moment the thrombotic event occurred. They imply that, by choosing this design, a possible fluctuation of inflammatory parameters in APS patients has not been taken into account, in particular as, according to Bobba et al., further prospective studies which are designed to address the issue of the correct timing—concerning measurement of exposure and outcome—will be needed.
Interestingly, neither the authors of the original report nor the authors of the editorial address the fact that LAC positivity might not be the cause but a consequence of elevated inflammatory parameters.
Overweight and obesity have been linked to low-grade inflammation earlier. Obese patients exhibit a higher concentration of inflammatory markers as compared to lean people . Our data further confirm these observations, as we report that an increase in body weight is linked to a simultaneous increase in fibrinogen levels.
Interpreting our data the following questions arise: do overweight and obese PAPS patients get symptomatic more often because the low grade inflammation (“second hit”) causes the VTE events or do overweight and obese patients have PAPS more often as the low grade inflammation might induce antibody production?
As we did not investigate a third group of patients with antiphospholipid antibodies and without clinical manifestation we cannot answer the first question stated above.
However, our findings support the second hypothesis. Fibrinogen levels were significantly higher in PAPS patients as compared to non-PAPS patients and increased with body weight. Therefore, overweight and obese patients seem to be at high risk for PAPS, which might be caused by the induction of antibody production by the low grade inflammation.
In a recent publication, Heggen et al. report a decrease of inflammatory parameters in overweight and obese patients when put on a hypocaloric diet . Whether these changes are also detectable in a possibly coexisting antiphospholipid antibody leading to a reduced thrombophilic situation should be evaluated in future studies.
Limitations of our study are the retrospective study design and the lack of a third group of patients with antiphospholipid antibodies without clinical manifestation.
However, we were able to demonstrate for the first time that occurrence of PAPS in VTE patients seems to be more frequent in overweight and obese patients and might be induced by a state of elevated inflammation. This fact could further increase the high VTE risk in these patients.
- 8Update of the guidelines for lupus anticoagulant detection. Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis. J Thromb Haemost 2009;7:1737-1740., , , et al.
- 9Pathogenesis of the antiphospholipid syndrome. Semin Thromb Hemost, in press., , .
- 11Inflammation in patients with lupus anticoagulant and implications for thrombosis. J Rheumatol 2005;32:462-468., , , et al.
- 12Thrombosis and inflammation: a question in need of an answer. J Rheumatol 2005;32:397-399., , .
- 14Effect of a low-fat versus a low-glycemic-load diet on inflammatory biomarker and adipokine concentrations. Metab Syndr Relat Disord, in press., , , , .