D-dimer, microparticles, and p-selectin are venous thrombotic risk markers. Elevated p-selectin is associated with increased cardiovascular events. We examined the effects of exercise and air travel on the markers of vascular risk in marathon runners. Forty-one persons participating in the 114th Boston Marathon (April 19, 2010) were divided into travel (n = 23) and nontravel “control” (n = 18) groups according to whether they lived more than a 4-hour plane flight or less than a 2-hour car trip from Boston. The subjects provided venous blood samples the day before, immediately after, and after returning home the day after the marathon. The blood was analyzed for soluble d-dimer, microparticle procoagulant activity, and p-selectin. D-dimer levels increased more before to immediately after (142 ± 83 to 387 ± 196 ng/mL) in the travel group than in the controls (85 ± 26 to 233 ± 95 ng/mL; p = 0.02). Moreover, 6 travel subjects versus 0 controls had d-dimer values >500 ng/mL after returning home the day after the marathon, the clinical threshold for excluding venous thrombosis (p = 0.03). P-selectin increased with exercise (p <0.01) regardless of travel (p = 0.09) but age was related to p-selectin (p = 0.01) such that older subjects exhibited greater p-selectin values before (r 2 = 0.14; p = 0.02) and after returning home the day after the marathon (r 2 = 0.16, p = 0.01). In conclusion, the combination of exercise and travel increases venous and arterial thrombotic risk. Moreover, the p-selectin levels at rest and after exercise were greater with age. These results might explain the reports of venous thrombosis with air travel after athletic events and the reports of cardiac events in older participants running marathons.
Several published cases and multiple anecdotes have reported deep vein thrombosis (DVT) and pulmonary embolism in otherwise healthy endurance athletes who have traveled by car or airplane to and from endurance events. Although strenuous endurance exercise such as a marathon activates both coagulation and fibrinolysis, extended car, bus, train, or air travel also activates coagulation, making it possible that the combination of endurance exercise and travel could shift the hemostatic balance in athletes toward activation of the coagulatory system. We recently tested this hypothesis by measuring the coagulatory and fibrinolytic factors in 2010 Boston Marathon participants who either flew >4 hours across country before and after the marathon (“travelers”) or drove <2 hours (“controls”). The travelers had increased thrombin antithrombin complex (TAT, a marker of coagulation) but not tissue plasminogen activator (t-PA, a marker of fibrinolysis) significantly more than in the control group. The regulation of hemostasis is complex, however, and involves many factors other than TAT and t-PA. The present report examined the effect of prolonged air travel and marathon running on d-dimer, p-selectin, and microparticles, plasma biomarkers of thrombosis, inflammation, and DVT.
Methods
A total of 23 travelers (12 men and 11 women) and 18 local controls (12 men and 6 women) running the 2010 Boston Marathon were recruited based on either residence in a geographic location that would require >4 hours of flight time to and from the Boston Marathon (California, Texas, or Colorado) or residence in a geographic location that was less than a 2-hour car trip from the greater Boston area (Massachusetts or Connecticut). The subjects were nonsmokers, aged 20 to 51 years, and free of known cardiovascular, metabolic, and coagulatory disease. The female subjects were premenopausal and did not take hormonal therapy, except for 1 woman who used norgestimate/ethinyl estradiol for contraception. No subjects used medicines known to affect coagulation, and all refrained from using medications that could affect coagulation such as aspirin or nonsteroidal anti-inflammatory drugs for 24 hours before, during, and after the marathon. The subjects provided written informed consent to participate as approved by the institutional review board at Hartford Hospital and in agreement with the Declaration of Helsinki guidelines.
The blood pressure and heart rate at rest (Welch Allen 52000 Vital Signs Monitor, Skaneateles Falls, New York), height, and body mass were measured the day before the marathon. The subjects provided a medical and exercise history, and venous blood samples were obtained. Blood was also obtained immediately after the subjects completed the marathon in the medical tent approximately 100 m from the finish line, and the day after the race at a Quest Diagnostics Laboratory in the subjects’ home cities in Texas, California, and Colorado or in the Boston/Hartford area. This last phlebotomy was obtained within 30 hours of the race finish, except for 1 traveler whose sample was obtained after 40 hours.
At each phlebotomy, blood was collected from an antecubital vein without stasis by single venipuncture. Blood from tubes containing 3.2% sodium citrate was centrifuged at 2,000 g for 10 minutes, and the plasma was transferred to cryovials and stored on dry ice until the samples could be archived. Cryovials obtained remotely were shipped overnight to Hartford Hospital on dry ice and archived in a −80°C freezer.
Plasma from each measurement point was analyzed for d-dimer (a derivative of crosslinked fibrin), soluble p-selectin (an adhesion glycoprotein that initiates the inflammatory cascade and regulates fibrin deposition and thrombus size), and microparticles (small fragments of cell membrane phospholipids that circulate and amplify coagulation around a developing thrombus).
The microparticle procoagulant activity (coefficient of variation 3.3%) and d-dimer levels (coefficient of variation 1.8%) were measured in triplicate using commercially available enzyme-linked immunosorbent assay kits (Aniara, Mason, Ohio). Soluble p-selectin was measured in triplicate using enzyme-linked immunosorbent assays (R&D Systems, Minneapolis, Minnesota) (coefficient of variation 3.2%). For all assays, absorbance was determined on a spectrophotometer (VersaMax Microplate Reader, Molecular Devices, Sunnyvale, California), and data were analyzed using associated software (SoftMax Pro Microplate Data Acquisition and Analysis software, version 5.3, Molecular Devices, Sunnyvale, California).
To determine the effects of travel and/or exercise on changes in d-dimer, p-selectin, and microparticles, we used a linear mixed model for repeated measurements with autoregressive variance–covariance structure, incorporating time as the within-subjects factor and group (travel vs control) as the between-subjects factor. The subjects were defined as the random factor; all other variables were fixed within the model. Potential categorical factors (e.g., gender) that could affect the relation between the main effects and outcomes were added to the model to assess significance, and the effect of continuous variables (e.g., age) was investigated using analysis of covariance. p Values for the mean difference estimates between groups at various points were adjusted using Tukey’s multiple comparison procedures to account for post hoc multiple comparison testing, ensuring that the family-wise type I error would be 5%. Univariate analysis of covariance was used to investigate the group effect on d-dimer immediately after the marathon with previously reported t-PA and TAT as covariates in the model. Statistical analyses were performed with SAS, version 9.1 (SAS Institute, Cary, North Carolina), and all data are expressed as the group mean ± SD.
Results
The baseline characteristics between travelers and controls have been previously published ( Table 1 ). Travelers flew 5.6 ± 1.4 hours to and 5.7 ± 1.4 hours from the marathon, arriving in Boston 53.3 ± 12.2 hours (approximately 2 days) before the marathon. No differences were found between the groups for time reported for sedentary, light, moderate, or vigorous physical activities before and the day after the marathon.
| Variable | Control (n = 18) | Travel (n = 23) |
|---|---|---|
| Age (years) | 32 ± 8 ⁎ | 42 ± 7 |
| Weight (kg) | 71.1 ± 10.7 | 67.2 ± 10.5 |
| Height (cm) | 171 ± 14 | 165 ± 32 |
| Systolic blood pressure (mm Hg) | 126 ± 8 | 123 ± 14 |
| Diastolic blood pressure (mm Hg) | 72 ± 7 | 73 ± 8 |
| Heart rate (beats/min) | 64 ± 9 | 62 ± 8 |
| Training mileage (km/wk) | 61 ± 23 | 63 ± 16 |
| Taper mileage (km/wk) | 37 ± 16 | 40 ± 13 |
| Official finishing Time (hr:min:s) | 3:42:24 ± 0:28:14 | 3:34:23 ± 0:24:54 |
D-dimer increased immediately after the marathon more in the travelers than in the controls (p <0.01). D-dimer remained elevated in both groups the day after the marathon but the differences between the groups only approached statistical significance (p = 0.08; Figure 1 ) . There were no age or gender main effects or interactions (all p >0.18) associated with exercise-induced increases in d-dimer. Six of the 23 travelers versus no controls had d-dimer values >500 ng/mL, the clinical threshold for excluding venous thrombosis (p = 0.03 for Fisher’s exact test).
There was an interaction between the effect of travel on the d-dimer response immediately after the marathon and t-PA measured at the same point (p <0.05) such that t-PA was directly associated with d-dimer in the controls (r = 0.24; p <0.05) but not in the travelers (r = 0.03; p = 0.42; Figure 2 ) .
Soluble p-selectin increased with exercise and remained elevated the day after the marathon (p <0.01; Figure 3 ) . There was no effect of travel on the p-selectin response to exercise (p = 0.09), but age was a significant covariate in the model (p = 0.01) such that older subjects exhibited greater p-selectin values both before (r = 0.14; p = 0.02) and immediately after the marathon (r = 0.16, p = 0.01; Figure 4 ) . Moreover, 12 runners at immediately after the marathon and 10 runners at the measurements the day after the marathon exhibited p-selectin values >45.5 ng/mL, a level thought to increase the risk of a future cardiovascular event by 28%. There were no gender main effects or interactions associated with exercise-induced increases in p-selectin (all p >0.28).
