Sport and Venous Thromboembolism

Abstract

Background

The occurrence of venous thromboembolisms (VTE) in association with sporting activity has been described but not yet systematically explored. The aim of this study was to determine the sites at which VTE occur in athletes, the accompanying features, and the special features of the symptoms and diagnosis, so that physicians can take the findings into consideration.

Methods

A search of the literature in the databases PubMed, Web of Science, and Cochrane in accordance with the PRISMA criteria, together with a search of Google Scholar up to 29 February 2020.

Results

No observational studies were identified. A total of 154 case descriptions were evaluated: 89 on upper-extremity deep vein thrombosis (DVT), 53 on lower-extremity DVT, and 12 on pulmonary embolisms with no evidence of thrombosis. Ninety-five percent of the upper-extremity DVT involved the region of the subclavian/axillary vein. Thoracic outlet syndrome (38%), hereditary thrombophilia/a family history of VTE (16%), intensive training (12%), and the use of oral contraceptives (7%) were identified as accompanying features. The upper-extremity DVT occurred mainly in male strength athletes and ball sports players. The lower-extremity DVT were located in the lower leg/knee (30%), the thigh (19 %), or occurred in combination in the lower leg–to-pelvis region (30 %). The features accompanying lower-extremity DVT were hereditary thrombophilia/a family history of VTE (30%), trauma (25%), immobilization (21%), and the use of oral contraceptives (11%). The lower-extremity DVT were found in endurance sports and ball sports. The symptoms may be obscured by sport-specific symptoms/trauma, and diagnosis is often delayed. Early D-dimer determination is useful and is complemented by diagnostic imaging.

Conclusion

VTE are found in association with sports. The background factors, the sites of VTE, the types of sports involved, and the accompanying features are all important to know. The symptoms may be obscured, and it may be difficult to reach the correct diagnosis. The possible presence of DVT must be borne in mind.

Venous thromboembolism (VTE) is a vascular occlusion of a blood clot in deep veins (deep vein thrombosis; DVT) or of an embolism (usually) in the pulmonary artery (pulmonary embolism; PE) and has an average incidence rate of around 100 to 200 per 100 000 person-years in the general population (1).

In the development of a VTE, hereditary risk factors, age, anatomical causes, the sex of the person, and use of oral contraceptives play a role independent of sports (2). Trauma, intense physical stress, dehydration, and prolonged periods of immobilization due to sports injuries/long-haul travel can be induced by, or associated with, sports (3). Since immobilization, physical activity, and sports can influence the development of VTE, the site of VTE / accompanying factors and symptoms / diagnosis of a VTE in the context of sports and types of exercise are examined here.

Immobilization is a well-known cause of VTE (4). In the study by Samama et al. (4), the odds ratio was increased by about 5- to 6-fold with immobilization due to bed confinement, and in Roach et al. (5), it was increased by 7- to 9-fold within a narrow time around hospitalization. Notably, it is often assumed that physical activity always has antithrombotic properties. However, whether this is really the case is still a matter of debate. While some studies have shown that physical activity is associated with a reduced risk of VTE (6, 7), others have showed it to be associated with an increased risk (8, 9). Borch et al. (10) examined 26 490 people between the ages of 25 and 97 as part of the Tromsø study and could not demonstrate that moderate physical activity had any influence on VTE risk. Van Stralen et al. showed a reduced risk of VTE for participants in sports activities (11), but an increased risk for >65-year-olds who do strenuous exercise (such as jogging) in 2008 (8). The positive influences of physical activity on the venous system are determined by activity and overall situation. Depending on duration / intensity, physical activity leads to an increases in factor VIII activity (12) and the von Willebrand factor (vWF) (13), to shortening of the activated partial thromboplastin time (aPTT) (14, 15), and to increases in platelet count and platelet activity / reactivity (16– 18). For fibrinolysis, the tissue-specific plasminogen activator (tPA) activity is increased as a compensatory measure, while the plasminogen activator inhibitor-1 (PAI-1) activity is decreased (14, 19). With respect to duration, changes in plasmatic blood coagulation can be detected for up to one day and last much longer than changes in fibrinolysis (20, 21). These changes are influenced by additional alterations, for example hereditary thrombophilia. These occur in a comparable prevalence among high-performance athletes (22). Activated protein C (APC) resistance does not lead to an increased coagulation reaction; however, there are indications that a protein C deficiency or an antiphospholipid syndrome increases the pro-coagulatory potential (23, 24), and that the protein C system can be influenced by physical activity (25). Endurance training modifies the response of the hemostasis system, and comparable changes in coagulation in participants of endurance training are only achieved at greater levels of energy expenditure (26). In summary, the positive effects of physical activity seem to outweigh the negative effects on a net basis, but not in a direct dose relationship, especially with regard to the intensity of exercise (27). Although there is no clear database, a comparable prevalence of thrombosis in active and sedentary people is assumed (3). VTEs associated with sporting activities are often described as case studies.

This study aimed to answer the following questions:

• Which types of sporting activities are associated with VTE? At which sites, and with which accompanying factors, do VTE occur?

• What special features should be taken into account with respect to symptoms or diagnostics in VTEs associated to sporting activities?

Methods

A literature search was conducted following the criteria of the PRISMA statement (e1) (figure). Using the language filter “English/German”, the PubMed / Web-of-Science and Cochrane databases were searched (from the start of database records to 29 February 2020) using the key words: (“venous thrombosis” OR “venous thromboembolism” OR “vein thrombosis” OR “phlebothrombosis” OR “pulmonary embolism”) AND (“exercise” OR “physical activity” OR “sports” OR “athlete”). The search was supplemented by Google Scholar. Thrombophlebitis was not taken into account.

Figure.

A flow diagram for selection of literature in accordance with the PRISMA statement

All authors first screened the titles and abstracts and then subsequently the full texts. Hits were evaluated according to the PICo scheme (e2) (participants/population: patients with thrombosis; interest: conditions under which thrombosis occurs; context: active sporting activity) for qualitative studies. A total of 1282 literature references were detected, of which 1042 were excluded as they did not meet the inclusion criteria with respect to title/abstract or were duplicates. Sixty-three papers were analyzed again by the authors, of which 34 were excluded. The resulting 206 literature references were checked again with respect to bibliography, and a further 17 relevant literature references were detected. Finally, 37 full texts did not meet the inclusion criteria, so that 186 full texts were included in the study.

Results

VTE and sport—site and accompanying factors

Based on the systematic literature search, we detected 154 case studies. Of these, 89 DVT affected the upper extremities, and 53, the lower extremities; in twelve cases, pulmonary embolism (PE) occurred with no evidence of thrombosis. As the thromboses presented differently with respect to type of sport, accompanying factors, etc., the studies on upper-extremity DVT, lower-extremity DVT, and PE with no evidence of thrombosis were analyzed separately. Individual cases in the literature, such as sinus-, ophthalmic-, or portal vein thrombosis, were not thematically followed up (e3 – e12).

Upper-extremity DVT (N = 89)

An overview of upper-extremity DVT is shown in Table 1. Significant sport types come from strength / ball sports, with a total of 63% of cases. DVT was particularly common in weightlifters / baseball players. The age range was between 14 and 29 years, but cases were also detected in the fourth and fifth decades of life. Males are affected more often. In more than 95% of the cases, the axillary–subclavian veins were detected as the central region. A major cause of upper-extremity DVT of the was anatomical constrictions of a thoracic outlet syndrome (TOS) (38%). These structural changes can cause Paget-Schroetter syndrome. Anatomically, this problem can be explained by constrictions between the scalene muscles and the first rib (e.g., the cervical rib or hypertrophy of the anterior scalene muscle in weight lifters), between the first rib and the clavicle (e.g., after a clavicle fracture with callus formation), or in the subcoracoid space between the coracoid process and tendon of the pectoralis minor muscle (e.g., hypertrophy in swimmers) (28, 29, e22). Although the cause of the constriction was not always proven, surgical decompression was still carried out in some cases. In individual cases, the structural narrowness in the costoclavicular space and adjacent to the subclavian muscle was also demonstrated by imaging (e21, e24, e47, e78). In the case of excessive muscle hypertrophy, the use of anabolic agents should also be considered (e82).

Table 1. Upper-extremity deep vein thrombosis (N = 89) (e13 - e81).

Types of sports			Age (in years)		Hereditary thrombophilia*		Acquired risk factors*
Weight training N = 30			14–29	N = 55	Factor V Leiden mutation(N = 1 homozygous; all + other mutation)	N = 4	Contraceptives (oral)	N = 6
Weight lifting		N = 17	30–39	N = 15			Tobacco use	N = 5
	+ Baseball/softball	N = 3	40–49	N = 10			Immobilization	N = 0
	+ Football/rugby	N = 4	50–65	N = 5	MTHFR mutation*7 (C677T) (homozygous, N = 1; heterozygous, N = 5)	N = 6	Other accompanying factors*
	+ Other*1	N = 4	> 65	N = 1			Intensive training	N = 11
Bodybuilding		N = 1	Not specified	N = 3			Trauma	N = 4
Not specified		N = 1	Sex		AT-III deficiency	N = 1	Local compression	N = 3
Ball sports N = 26			Male	N = 67	Family history		Infection	N = 3
Baseball	Pitcher	N = 4	Female	N = 22	Thromboembolism in first-degree relatives	N = 3	Other*8	N = 3
	Catcher	N = 1	Anatomical changes				Complications*
	Not specified	N = 4	Thoracic ‧outlet syndrome ‧assumed	N = 34			Pulmonary embolism	N = 10
Basketball + other types of sports*2		N = 4					– With a fatal outcome	N = 1
Volleyball		N = 3	Operative decompression	N = 31
Rugby/football		N = 3	Site localization				Post-thrombotic syndrome/ complaints	N = 14
Other*3		N = 7	Subclavian vein / axillary vein involved	N = 85
Endurance sports N = 15			Other*6	N = 4			Recurrence	N = 13
Swimming		N = 7
Other*4		N = 8
Contact sports N = 7
Wrestling/judo/pro wrestling		N = 6
+ Football		N = 1
Other*5 N = 11

* Multiple answers possible

*¹ Alpine skiing (N = 1), soccer (N = 1), swimming (N = 1), basketball (N = 1)

*² Softball (N = 1), tennis (N = 1), volleyball (N = 1), wrestling (N = 1)

*³ Water polo (N = 2), handball (N = 2), softball (N = 1), tennis and Badminton (N = 1), lacrosse (N = 1)

*⁴ Rowing (N = 2), running/marathon (N = 2), triathlon (N = 2), Nordic skiing (N = 1), spinning (N = 1)

*⁵ Dance (N = 2), climbing (N = 2), recreational sports (N = 2), gymnastics (N = 1), aerobics (N = 1), track and field athletics (N = 1), bowling (N = 1), Kaatsu (N = 1)

*⁶ Brachiocephalic vein (N = 1), brachial vein (N = 2), ulnar vein (N = 1)

*⁷ With exclusion of hyperhomocysteinemia (N = 3), no further information (N = 3)

*⁸ Pregnancy (N = 1), high-altitude exposure (N = 1), Hashimoto‘s thyroiditis (N = 1)

Hereditary thrombophilia / a family history of VTE (16%) can occur, but less frequently than cases with anatomical changes. In 12% of the cases, intensive training was an influencing factor. The combination of weight training, TOS, and intensive training intensifies the individual effect. Additional acquired risk factors, such as oral contraceptives (7% total/27% of women), were also evaluated. According to the literature, however, women are less affected.

PEs have also been described, some of which were fatal. The symptoms cited as post-thrombotic syndrome were pain, swelling, heaviness, reduced resilience, and sensitivity disorders of the arm (e15, e26, e39, e69, e72, e80, e81). However, no uniform classification of the post-thrombotic syndrome is currently available (30). Return-to-sport at a potentially high-performance level is also rarely discussed. In the few cases for which information was available, the original physical activity was only resumed after weeks or months —if at all (e33, e41, e79).

Lower-extremity DVT (N = 53)

Fifty-three case studies of lower-extremity DVT were included (table 2). These related primarily to endurance sports (45%) and ball sports (40%). Weight training plays a subordinate role here; the patients in most of the case studies often practiced selected sports, such as running and marathons (25%). The age range was between 15 and 29 years, and males (79%) were more severely affected. The sites of occurrence were divided roughly into the regions of lower leg / knee and thigh / pelvis; in 30% of the cases, combinations of the lower leg up to the pelvis could be detected. Hereditary thrombophilia/a family history of VTE (30%) were detected as accompanying factors. Trauma was often given as an accompanying circumstance (25%).

Table 2. Deep vein thrombosis (DVT) of the lower extremities (N = 53) (e14, e82– e124).

Type of sports			Age (years)		Anatomical changes		Hereditary thrombophilia*		Acquired risk factors*
Endurance sports N = 24			15–29	N = 26	May-Thurner syndrome	N = 4	Factor V Leidenmutation (homozygous, N = 2; + other mutation, N = 2)	N = 5	Immo- bilization*3	Long-haul trip	N = 7
Running/marathon		N = 13	30–39	N = 7	Site of occurrence					Injury	N = 2
Bicycling		N = 3	40–49	N = 9	Lower leg/knee N = 16					Combination*4	N = 2
Triathlon		N = 2	50–65	N = 6	Peroneal vein	N = 3			Contraceptive (oral)		N = 6
Rowing		N = 2	> 65	N = 1	Tibial vein	N = 2	MTHFR mutation*2 (C677T) (homozygous)	N = 2	Tobacco use		N = 2
Hiking		N = 1	Not specified	N = 4	Popliteal vein	N = 4			Other accompanying factors *
Swimming		N = 1			Several	N = 5	Prothrombin G20210A mutation (heterozygous)	N = 1	Trauma*5		N = 13
Not specified		N = 2	Sex		Not specified	N = 2			Intensive training		N = 2
Ball sport N = 21			Male	N = 42	Thigh N = 10				Suspected dehydration		N = 2
Football		N = 7	Female	N = 9	Femoral vein	N = 8	Protein C deficiency (heterozygous)	N = 2	Other*6		N = 6
	+ Volleyball	N = 1	Not specified	N = 2	Not specified	N = 2			Complications*
Soccer		N = 5			Pelvis N = 2		Protein S deficiency (heterozygous)	N = 1	Pulmonary embolism		N = 17
Basketball		N = 3			Iliac vein	N = 1			– With a fatal outcome		N = 4
Racquetball		N = 2			Caval vein	N = 1	Hyperhomo- cysteinemia	N = 2	DVT with a fatal outcome		N = 2
Ice hockey/hockey		N = 2			Combination N = 16				Post-thrombotic syndrome/complaints		N = 5
Baseball (catcher)		N = 1			From lower leg to pelvis	N = 16	Not specified	N = 1
Weight training N = 2							Family history		Recurrence		N = 2
Weight lifting		N = 1			Not specified N = 9		Thromboembolism in first-degree relatives1. Grades	N = 2
Not specified		N = 1
Contact sports N = 0
Other*1 N = 6

* Multiple answers possible

*¹ Track and field athletics (N = 1), alpine skiing (N = 1), mountain sports (N = 1), motorcycle racer (N = 1), not specified (N = 2)

*² With proven hyperhomocysteinemia (N = 1), not specified (N = 1)

*³ > 4 h within a relative time frame of 8 weeks

*⁴ Recovery from an injury and additional immobilization due to travel

*⁵ Four with immobilization

*⁶ Overweight (N = 1), pregnancy abortion (N = 1), anabolic steroids (N = 1), exposure to high-altitude/cold/virus infection (N = 1), heat (N = 1), Hashimoto‘s thyroiditis (N = 1)

Immobilization was detected in 21% of the cases. A flight or car trip played a role for a total of seven patients; these may well be sports-related. Unfortunately, no assessment about either the influence of dehydration alone (3), or about a combination of dehydration/prolonged inactivity due to travel, can be drawn from the data. The use of oral contraceptives (11% in total) was listed in six of the nine affected female athletes. Anatomical changes such as May-Thurner syndrome were rarely mentioned, and „popliteal entrapment syndrome“ was only mentioned as a suspicion (e105). In lower-extremity DVT, PE occurred in 17 cases, four of which were fatal.

Pulmonary embolism without evidence of thrombosis (N = 12)

Twelve cases of PEs (table 3) with no specific evidence of thrombosis were listed. These cases involved ball / endurance athletes. The age range between 16 and 29 years was more affected. Eight out of twelve cases were female athletes; seven reported taking oral contraceptives. One fatal outcome was described.

Table 3. Pulmonary embolism (N = 39).

Pulmonary embolism with no evidence of thrombosis N = 12 (e94, e100, e125– e134)
Type of sports			Age (years)		Hereditary thrombophilia
Ball sports N = 5			16–29	N = 10	Factor V Leiden mutation (heterozygous)		N = 2
Basketball		N = 1	30–39	N = 0
	+ Volleyball and track and field	N = 1	40–49	N = 1	Family history
			50–65	N = 0	Thromboembolism in first-degree relatives 1. Grades		N = 0
Soccer		N = 1	> 65	N = 0
	+ Wrestling	N = 1	Not specified	N = 1	Acquired risk factors*1
Softball	Pitcher	N = 1			Contraceptive (oral)		N = 7
Endurance sports N = 4			Sex		Immobilization	Flight or car trip*2	N = 2
Rowing		N = 1	Male	N = 3
Bicycling + jogging		N = 1	Female	N = 8	Tobacco use		N = 0
Cross-country running		N = 1	Not specified	N = 1	Other accompanying factors
Not specified		N = 1			Trauma		N = 1
Contact sports N = 1					Infection		N = 1
Wrestling		N = 1			Complications
Weight training N = 0					Fatal outcome		N = 1
Other N = 2
Gymnastics		N = 2
Pulmonary embolism with proven thrombosis N = 27 (see Tables 1 and 2)

*¹ Multiple answers possible

*² > 4 h within a relative time frame of 8 weeks

VTE and sport—special features of symptoms and diagnostics

In most cases, typical symptoms of thrombosis were detected. Swelling, pain, cyanosis and increased vascular markings were reported on all extremities. However, athletic individuals often had musculoskeletal problems, which masked the symptoms. Table 4 compares possible symptoms of thromboembolism and symptoms of sport-specific diagnoses (31, e100, e102). Localized pain that is typical for sport-specific trauma and musculoskeletal symptoms, or even intense muscle soreness, can be detected. Tissue hardening in the form of myogelosis can occur, and localized swelling and overheating can occur in a sport-specific manner. PE was not initially recognized either, despite severe shortness of breath and coughing. If the endurance athlete‘s physiological resting heart rate is well below fifty beats per minute, reactively increased heart rates in PE can remain within the normal range or be mistaken for overtraining. Even dyspnea can be masked.

Table 4. Possible symptoms of deep vein thrombosis and pulmonary embolism with ?differential diagnosis.

	Possible symptoms of thromboembolism	Sports-specific diagnoses with similar symptoms
Deep vein thrombosis	Localized pain	Trauma*, sore muscles, tendinopathy, stress fracture
	Hardened tissue	Myogelosis
	Tense/heaviness	Muscle exhaustion/overload
	Local swelling	Hematoma, joint effusion, Baker‘s cyst, bursitis
	Overheating	Inflammatory/degenerative arthropathy
	Cyanosis	Compartment syndrome
Pulmonary embolism	Dyspnea /cough(possibly during exertion)	Exercise-induced asthma (or allergies), hyperventilation, reduced performance, infection
	Chest pain	Intercostal neuralgia
	Increased heart rate	Overtraining, cardiac cause

* e.g. Contusion, tendon rupture, muscle strain, torn muscles

Discussion

The aim of this work was to determine sites of occurrence / accompanying factors as well as peculiarities in the symptoms / diagnosis of sport-associated VTE, so that this can be taken into consideration by the treating physicians. A final assessment of an overall reduced or increased VTE risk in athletes compared to the normal population is not possible; however, that was not the aim of the present study.

Site and accompanying factors

DVT and PE (either as a complication or without evidence of a primary thrombosis) are reported in the literature in connection with sports. It can be assumed that the number of unreported cases is higher, as many cases are not reported in the literature or are diagnostically overlooked. The majority of the reported case studies were relate to the upper extremity. Interestingly, this actually occurs about six times less frequently in the general population than leg vein thrombosis (e34). A pronounced muscle development as a specific requirement in weight training and overhead ball sports could play an essential role here. The site of occurrence is mostly limited to the axillary–subclavian vein. The main accompanying factors for upper-extremity DVT, summarized according to frequency, were TOS (38%), hereditary thrombophilia / a family history of VTE (16%), intensive training (12%), and use of oral contraceptives (total 7% [27% of females]); together, these accounted for 73% of cases. Therefore, these accompanying factors should be taken into account in the diagnosis.

The lower extremities are more often affected in endurance sports, and especially running. The site of occurrence can affect the lower leg up to the knee, and the thigh to the pelvis; combinations of these are common. The main accompanying factors for lower-extremity DVT, summarized according to frequency, were hereditary thrombophilia / a family history of VTE (30%), trauma (25%), immobilization (21%), and use of oral contraceptives (11% [66% of females]); together, these accounted for a total of 87% of the cases. This should also be taken into account. In studies, the prevalence rate of VTE in populations with a factor V Leiden mutation was 13–25% and significantly higher in combination mutations (32). The importance of specific polymophisms for VTE in the general population can be found in the meta-analysis by Gohil et al., whereby factor XIII plays a lesser role (e135, e136). Hereditary thrombophilia also seem to be of particular importance in lower-extremity VTE in athletes.

It should be noted that some of the case reports date from years before 1994 in which, for example, no thrombophilia diagnosis could be carried out with respect to the frequent factor V Leiden mutation (33).

Trauma, including minor injuries such as torn muscle fibers or distortions of the lower extremities that do not require subsequent immobilization, lead to a three-fold increased odds ratio of a thrombosis within four weeks (34). There also appears to be an increased risk of minor trauma associated with a factor V Leiden mutation (34).

The total proportion of all cases with clinically apparent PE was 25%; however, for cases with only lower-extremity DVT, PE was apparent in about 32%. The number of cases of PE without evidence of thrombosis (12 cases) was relatively small, which limits further evaluation. However, seven of the eight female athletes of this group used of oral contraceptives.

Symptoms and diagnostics

Overall, the detection of the specifically different accompanying factors is helpful for the diagnosis. The symptoms of VTE do not differ between athletes and non-athletes, yet they can be assessed differently and can be considerably obscured by sport-specific symptoms. This complicates and delays the diagnosis.

In a review by Taylor et al. in 2019 (31), 47 cases of athletic individuals with VTE showed more than 25 misdiagnoses, with an average time to diagnosis of 56 days (unfortunately, detailed descriptions are not given). Different pain tolerances in athletes alone can result in different assessments (35). The clinical symptoms are assessed based on musculoskeletal status, which makes another diagnosis appear more likely. Due to the sport-specific characteristics, the clinical diagnosis of VTE based on the Wells score is also significantly more difficult (36). A score for athletes was therefore developed („Athlete Deep-Vein Thrombosis Risk Assessment Screening Tool“) (e103). Accompanying factors such as a positive family history, the use of contraceptives, as well as long-haul travel and immobilization periods are taken into account. All of this underscores that the symptoms in athletes can be masked, making other diagnoses of trauma / overload to be seen as more likely. As this then leads to delayed diagnoses, this article is highly relevant for the clinically active physician. Since there are limitations with regard to the informative value of the Wells score (37), it makes sense to consider D-dimer determination at an early stage if the clinical picture is not clear. It should be noted that there can be a limited increase in D-dimer up to four weeks after trauma (38), and that acute physical stress can be accompanied by an increase in D-dimer; however, the cut-off value for DVT is usually not reached (15, 39). According to the guideline, however, the D-dimer determination should only be used if the probability is low. If there is a high probability, diagnostic imaging is indicated directly (38).

Limitations/need for future research

Overall, the data available on the topic are very limited. There are no suitable observational studies with comparison groups on the subject of VTE in sports, and the information in the processed case reports is incomplete. Individual parameters, such as intensive training, could not be defined more precisely on the basis of the sources. A final evaluation based on the PRISMA criteria (risk bias, synthesis of results, effect estimates, and additional analyzes) was also not possible. This limits the meaningfulness of the results and the assessment of the probability of VTE in various sports activities. For case reports, we would recommend for instance that the Care Guideline (40) is used, which would ensure that the information required for evaluation is also available. Prospective observational studies and further data acquisition and analysis of VTE in (competitive) sports should be the subject of future research.

Conclusion

VTE are associated with sport, and the site of occurrence differs depending on the type of sports activity. Specific accompanying factors have a locally different meaning for the VTE. Classical symptoms of VTE can be masked by sport-specific symptoms. “Keep it in mind” and using an early D-dimer diagnosis and/or imaging would be helpful. A thrombophilia diagnosis in cases with high-risk constellations should be considered.