A Hospital-Based Case-Control Study: Analysis of Risk Factors for Venous Thromboembolism at High Altitudes
Pharmacy Department of the People’s Hospital of the Xizang Autonomous Region, Lhasa, China
Department of Pharmacy, The Affiliated Taizhou People’s Hospital of Nanjing Medical University, Taizhou, China
Department of Pharmacy, Peking Union Medical College Hospital, No. 1 Shuaifuyuan, Dongcheng District, Beijing, China
Outpatient Department of the People’s Hospital of the Xizang Autonomous Region, Lhasa, China
Xizang Autonomous Region Maternity and Children’s Hospital, Lhasa, China
Zuogong County People’s Hospital, Chamdo, China
*Correspondence: Yuting Wu, 417881492@163.comAbstract
Introduction
The incidence of venous thromboembolism (VTE), a leading cause of morbidity and mortality worldwide, exhibits considerable geographical variability. Although VTE is well- documented in low-altitude settings, the role of high-altitude (HA) environments on the occurrence of VTE remains unclear. Here, we aimed to address this gap by examining the prevalence of VTE at the HAs of the Tibet Autonomous Region, China, offering a unique perspective on how altitude influences VTE risk.
Methods
In this retrospective analysis, we reviewed the medical records from the Tibet Autonomous Region People’s Hospital between January 2018 and December 2023. We included 1,084 patients diagnosed with VTE in this quantitative synthesis and comprehensively analysed demographic data, clinical outcomes, and anticoagulant treatment patterns. We used statistical methods, including logistic regression, to identify potential risk factors and explore the relationship between HA and VTE incidence.
Results
The prevalence of VTE in the Tibet Autonomous Region People’s Hospital between 2013 and 2023 increased annually from 0.74% to 1.12%. Additionally, VTE was mainly associated with the 18–64-year age group, low haemoglobin levels, and hypertension at HA regions. The occurrence of VTE among residents of HA areas was more common than that reported for lower-altitude populations.
Conclusion
This study highlights the critical influence of HA environments on VTE prevalence and the need for tailored healthcare approaches. The unique patterns of anticoagulant therapy after discharge highlight the importance of personalized treatment plans. Overall, our findings contribute to the global understanding of VTE at HA regions and the development of tailored treatment for patients with VTE from HA areas.
Article notes
Competing Interest Statement
he authors have declared that no competing interests exist.
Funding Statement
The author(s) received no specific funding for this work.
1Introduction
Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), represents a substantial public health concern owing to its widespread prevalence (Prabhakar et al., 2019), potential for causing severe complications, and risk of recurrence. Specifically, VTE is the third leading cause of vascular mortality worldwide (Srivastava et al., 2020). Epidemiological data reveal a geographic variation in the incidence of VTE, with estimated rates typically between 1–2 per 1,000 person-years in western countries and less than 1 per 1,000 person-years in eastern countries (Tritschler et al., 2018).
Numerous risk factors, including prolonged immobility (such as during lengthy flights or extended bed rest), surgical interventions, cancer, hormonal therapy, pregnancy, and familial predisposition to blood clots, have been implicated in the development of VTE (Lowe et al., 1999; Dicks et al., 2024). However, the specific contribution of each factor to thrombotic risk and the pathogenetic mechanisms underlying their contributions remain unclear. Lifestyle factors, such as smoking and obesity (Cheng et al., 2013), are also associated with the increased risk of developing VTE. High altitudes (HAs) are associated with a hypercoagulable state (Jones et al., 2022; Broggi et al., 2021), potentially increasing the risk of thrombotic events. An increased prevalence of VTE is associated with populations residing at elevated altitudes (Pelouch et al., 1997; Trunk et al., 2019); some studies have suggested that reduced oxygen availability at high altitudes leads to hypercoagulability (Trunk et al., 2019; Agostoni et al., 2000). However, the effects of HA exposure remain poorly understood, potentially increasing the risk of thrombosis. These inconsistencies highlight the need for real-world data to clarify the impact of HA exposure on VTE and determine whether specific populations, such as older adults or individuals with cardiovascular conditions, are more affected by VTE or more susceptible to developing this condition.
The People’s Hospital of the Tibet Autonomous Region in Lhasa, China, located at an altitude of 3,650 m, serves as the largest and most comprehensive healthcare institution in the region. The residents of this region are genetically adapted to live in hypoxic conditions due to the thin air at HAs. This unique demographic factor can influence the occurrence of a wide range of health issues, including cardiovascular health and blood clotting disorders, both of which are closely related to VTE (Fung et al., 2019; Jha et al., 2018). Considering these aspects, in the present study, we aimed to analyse the prevalence of VTE in HA areas of the Tibetan Autonomous Region by statistically comparing the occurrence and risk factors of VTE between residents of HA and plain areas. Through this comparison, we sought to gain a comprehensive understanding of how HA environments influence the risk of VTE. Such information will facilitate the development of targeted prevention and treatment strategies against these risk factors. Additionally, we aimed to explore VTE treatment and management strategies tailored for HA areas, providing specific clinical guidance. The unique geographical, cultural, and healthcare context of Tibet enhances the relevance of the present study and provides insights not typically available from other regions.
2Methods
2.1Study Design
This retrospective cross-sectional study analysed VTE cases using electronic medical records from the People’s Hospital of the Tibet Autonomous Region, China, from January 1, 2018, to December 31, 2023. The study aimed to describe the prevalence, risk factors, and treatment patterns of VTE in HA populations.
2.2Study Setting and Participants
Setting: The study was conducted at the People’s Hospital of the Tibet Autonomous Region (altitude: 3,650 m; catchment population: ∼3.2 million).
Participants: Inclusion Criteria: Hospitalized patients aged ≥18 years. Primary discharge diagnosis of VTE (ICD-10 codes: I80.2 [deep vein thrombosis], I26.9 [pulmonary embolism]). Patients with concurrent diagnoses (e.g., lower limb venous thrombosis and pulmonary embolism) were categorized hierarchically based on clinical severity. Pulmonary embolism (PE) was prioritized over isolated venous thrombosis, as PE represents a more acute and life-threatening manifestation of VTE. For example, a patient with both lower limb thrombosis and low-risk PE was classified under the Pulmonary Embolism Low Risk category. Patients with multiple PE risk strata (e.g., concurrent low- and intermediate-risk PE) were assigned to the highest applicable risk category.
Exclusion Criteria: (1) Short hospitalization duration: Hospital stay <3 days. (2) Specific thrombotic or vascular diagnoses: Arterial thrombosis (ICD-10: I74.9), Thrombotic microangiopathy (ICD-10: M31.1), Acute cerebral infarction (ICD-10: I63.9), Thrombotic haemorrhoids (ICD-10: K64.8), Thrombotic thrombocytopenic purpura (ICD-10: M31.1), Tethered spinal cord syndrome (ICD-10: Q06.8), Portal vein embolism (ICD-10: I81), Mesenteric venous embolism (ICD-10: K55.0), Other venous thrombosis (ICD-10: I82.90) (3) Systemic complications: Multiple organ failure (ICD-10: R65.2). (4) Data limitations: Missing critical clinical data (e.g., incomplete laboratory results or imaging reports).
Data Source: Structured data were extracted from the Hospital’s Information System Information System (HIS) system, including demographics, laboratory results, comorbidities, and medication records.
2.3Variables
Primary outcome: Incidence of VTE. Secondary outcomes: Clinical improvement, mortality. Covariates included age, sex, and ethnicity, haemoglobin level (g/L), D-dimer (mg/L), fibrin degradation products (FDP, mg/L), and creatinine clearance (CrCL, mL/min) estimated via the Cockcroft-Gault equation; hypertension (defined as systolic blood pressure ≥140 mmHg, diastolic ≥90 mmHg, or documented use of antihypertensive medications); diabetes (fasting glucose ≥7.0 mmol/L or use of glucose-lowering agents), history of VTE; and recent surgical history (within 30 days prior to diagnosis). Covariates: Demographics: age, sex, ethnicity (Han/Tibetan). Laboratory Parameters: D-dimer (mg/L), activated partial thromboplastin time (APTT) (s), and fibrin degradation products (FDP, mg/L).
Comorbidities: Acute myocardial infarction (AMI) (I21.9), ischemic stroke (I63.9), cancer (C00– C97).
Confounders: Adjusted for age, sex, and BMI in regression models.
2.4Data Collection and Quality Control
Data Extraction: Trained researchers used a standardized form to collect variables from the HIS. Missing Data: BMI missing in 12.4% of cases handled via multiple imputation (SPSS MVA module). Laboratory parameters with >20% missingness (e.g., homocysteine) were excluded from multivariate analysis.
Bias Control: Selection Bias: To minimize confounding bias in subgroup analyses (e.g., comparing risk factors across age or sex strata), we performed 1:1 matching based on age and sex within the cohort. This internal matching ensured balanced distributions of these covariates when evaluating associations between altitude-related variables and VTE outcomes. Measurement Bias: ICD-10 codes validated against imaging reports (ultrasound/CT) for VTE confirmation.
2.5Statistical Analysis
Software: SPSS 21.0 (IBM) and GraphPad Prism 9.0. Descriptive Statistics: Continuous variables: Mean ± SD (normal distribution) or median [IQR]. Categorical variables: Frequency (%).
Inferential Statistics: Univariate Analysis: Student’s t-test (continuous), chi-square test (categorical). Multivariate Analysis: Binary logistic regression (enter method) to calculate adjusted odds ratios (aOR) with 95% CI. Sensitivity Analysis: Stratified by age groups (<45 vs. ≥45 years). Threshold: Statistical significance set at p<0.05 (two-tailed).
2.6Ethical Considerations
Approved by the Ethics Committee of the Tibet Autonomous Region People’s Hospital(Approval No.: ME-TBHP-25-KJ-028). Waiver of informed consent granted due to retrospective anonymized data use.
2Results
3.1Annual rate of occurrence of VTE
Table 1 illustrates the annual number of diagnosed thrombosis cases relative to the total number of patients discharged from January 1, 2018 to December 31, 2023. Based on the information collected, the prevalence of VTE in the Tibet Autonomous Region People’s Hospital between 2018 and 2023 was found to have annually increased from 0.62% to 1.07%.
3.2Baseline characteristics and comorbidities
A total of 1,084 patients were analysed. The proportion of women with VTEs was marginally higher (52.40%) than that of men with VTE (47.60%). Most patients were within the 18–64-year age group (62.84%), with a notable decline in the number of cases as the age increased. Most patients (73.32%) experienced hospital stays of 14 days or fewer, with the frequency of longer stays diminishing as the duration increased. A substantial proportion of patients exhibited clinical improvement (79.02%). The cure rate was 10.76%, and the mortality rate was comparatively low at 1.81%. Among the 1,084 patients, 63 (5.8%) presented with concurrent lower limb thrombosis and pulmonary embolism. These cases were classified under their respective PE risk categories (e.g., low, intermediate, or high risk) to avoid duplication and prioritize acute clinical management. Lower-limb venous thrombosis was the most prevalent type of thrombosis (41.21%). Various subtypes of PE were observed, with low-risk PE being the predominant subtype (17.04%). These cases were classified under their respective PE risk categories (e.g., low, intermediate, or high risk) to avoid duplication and prioritize acute clinical management. Rarer manifestations included mesenteric venous, portal vein, and intracranial venous sinus thrombosis. The Respiratory Department recorded the highest number of patients with VTE (43.56%), followed by the Hepatobiliary Surgery Department (22.47%). Other departments, such as Neurology, Mountain Disease, Cardiovascular, Obstetrics, and Gynaecology departments, reported a lower prevalence of VTE. The full list of comorbidities and their differences is presented in Table 1. Our findings indicate that in HA areas, the highest prevalence was observed among individuals aged 45–64 years (34.19%) (mean age, 54.87 years; 95% CI, 53.16–55.73).
3.3Test indicators of the patients
Analysis of the patient-related test results indicated that most participants (68.79%) exhibited normal alanine aminotransferase levels, whereas a considerable proportion of these participants (26.56%) presented levels above the upper normal limit, possibly indicating liver damage or inflammation. Most individuals (59.37%) had normal aspartate aminotransferase levels, whereas those of 24.97% of the participants exceeded the upper limit, suggesting potential liver or muscle tissue damage. Approximately half of the patients (44.68%) exhibited CrCL below the normal range; this may potentially be associated with reduced muscle mass or physiological normalcy in certain cases (Gustafsson et al., 1981). Approximately half of the patients (50.47%) showed haemoglobin values within the normal levels. A significant portion (71.90%) of patients exhibited an activated partial thromboplastin time above the upper limit, suggesting the presence of potential bleeding disorders or the effect of anticoagulant therapy. A considerable proportion (86.89%) of patients demonstrated elevated D-dimer levels, indicative of an abnormal clotting process, although elevated D-dimer levels can also occur under various conditions, including infection, inflammation, or postoperative states (Rostami et al., 2020; Kumaran et al., 2024). Furthermore, most patients (66.03%) had fibrin degradation product levels exceeding the reference value, potentially indicating abnormal clotting or fibrinolysis. Approximately one-third (30.94%) of patients had elevated homocysteine levels, which may be linked to vitamin B12 or folate deficiencies and represents a known risk factor for cardiovascular diseases (Mallikarjun et al., 2024). Hypoxic conditions at HAs may increase blood viscosity, thereby increasing the risk of thrombosis (Burns et al., 2018; Hudson et al., 1999).
A comprehensive list of the test parameters is presented in Table 2. After excluding 192 cases of wheelchair use, 159 cases of bed rest, and 134 cases with missing BMI data, statistical analysis was conducted on the remaining 599 medical records with recorded BMI. The analysis indicated a minimum BMI of 16.02, maximum BMI of 41.00, mean BMI of 24.87 (SD 4.72), and median BMI (50% percentile) of 24.39.
3.4Comorbidities
Table 3 lists various comorbidities, along with the number of cases and their respective percentages in the study population for VTE. The statistical results indicated that at HAs, VTE was mainly associated with hypertension (33.12%), followed by acute myocardial infarction (AMI) or ischemic stroke (12.36%). The chi-square test results indicated no statistically significant association between discharge diagnosis and hypertension (p = 0.480). Patients with VTE frequently experience serious cardiovascular events, such as acute AMI or ischemic stroke, indicative of the interconnection of these conditions (Sławek-Szmyt et al., 2024).
3.5Selection of anticoagulant drugs for hospitalized patients with embolism
Upon diagnosis of VTE, immediate anticoagulation therapy was initiated to prevent clot progression and embolism. Anticoagulant selection is guided by patient-specific factors, including renal function, bleeding risks, preferences, drug interactions, and thrombus characteristics. The primary treatment goal is to stabilize the clot and minimize the risk of embolism, which often requires elevated anticoagulant dosages. Upon hospital admission, VTE risk assessment considers factors such as immobilization, surgical history, and comorbidities to determine appropriate thromboprophylaxis and treatment levels (Lutfi et al., 2024; Prandoni et al., 2007). Most patients received either monotherapy (40.68%) or dual antithrombotic therapy (37.08%), reflecting standard regimens. Acute VTE management typically involves the administration of parenteral anticoagulants, such as unfractionated heparin or low-molecular-weight heparin. Specifically, 53.06% of 441 patients who received monotherapy during hospitalization received sodium heparin or enoxaparin sodium injections.
3.6Discharge with medications
We compiled data regarding the antithrombotic medication status at discharge for 1084 patients with venous thrombosis. The details are provided in Table 4, indicating that most patients were discharged and prescribed oral antithrombotics for continued treatment. Among these, 43.54% of the patients with VTE were treated with rivaroxaban tablets alone after discharge, whereas 32.29% were not prescribed antithrombotic medication.
4.Discussion
This study exclusively addressed non-cardiogenic venous thrombosis; the implications for AF-related thromboembolism warrant separate investigation. The annual increase in VTE prevalence observed between January 1, 2018, and December 31, 2023, aligns with the rising burden of altitude-related comorbidities in the region, such as hypertension (33.12%) and low haemoglobin levels (30.24%), which were significantly associated with VTE risk in our cohort.
The incidences of DVT and PE significantly increase with age (Patel et al., 2020), particularly after the age of 55. By the age of 80, the incidence rates reach approximately 1 in 100 per year, which is approximately 1000-fold higher than for those aged 45 years or younger. Furthermore, the rates of PE rapidly increase compared to those of DVT in older adult populations (Silverstein et al., 2007). Environmental factors can also lead to heightened DVT and PE incidences in younger demographics. The altered daily activities in HA regions, including occupational and recreational pursuits, may affect blood flow and thrombosis risk. Moreover, the scarcity of medical resources in these areas may impede early diagnosis and treatment of DVT and PE, possibly resulting in a higher incidence among younger individuals.
A notable observation in the present study was that approximately 30.70% of patients with thrombosis had haemoglobin levels below the normal range, suggesting a prevalent incidence of anaemia. This contradicts the common perception that HAs lead to polycythaemia (Gatterer et al., 2024). Approximately half of the patients (49.37%) exhibited reduced eGFR (estimated glomerular filtration rate) below the normal range, suggesting potential renal insufficiency. This finding aligns with previous reports of altered renal haemodynamics in high-altitude populations (Hurtado- Arestegui, A. et al., 2018). Moreover, a significant number of the samples (78.45%) were outside the reference range for homocysteine, with a particularly high percentage (47.51%) falling below the lower limit, indicating potential metabolic anomalies associated with an increased risk of cardiovascular diseases.
The co-occurrence of hypertension was the most prevalent among patients with VTE (33.12 %). Although hypertension is a recognized risk factor for arterial thrombosis (Miraclin et al., 2024), its direct role in VTE risk remains unclear. However, patients with hypertension should be carefully monitored owing to their elevated risk for cardiovascular complications. Following AMI or stroke, immobility, particularly paralysis, inflammation, and hypercoagulability, can predispose patients to VTE (Anderson et al., 2003; Gregson et al., 2019).
In the management of VTE, anticoagulant therapies are fundamental to preventing the formation or enlargement of blood clots. This highlights significant differences in the types and conditions of medications prescribed at discharge, depending on the different discharge diagnoses (Cohen et al., 2020). Rivaroxaban plays a pivotal role in the coagulation cascade. Its benefits include predictable pharmacokinetic profiles, elimination of routine monitoring, and minimal dietary interactions. Rivaroxaban dosing is contingent upon clinical indications and treatment phase, necessitating adjustments for renal function, potential drug interactions, and bleeding risks. For example, a regimen of 15 mg once daily may be appropriate for patients with moderate renal dysfunction after the initial treatment phase. For initial DVT and PE management, 15 mg twice daily with meals is recommended for the first 21 days, followed by a maintenance dose of 20 mg once daily (Gao et al., 2021). In our study, patients administered rivaroxaban typically received conventional treatment regimens, with only one case involving an outpatient medication regimen of rivaroxaban at a dosage of 30 mg once daily and another involving a two-year-old administered a dose of 2 mg once daily for venous sinus thrombosis.
The duration of anticoagulation therapy ranges from a minimum of three months to potentially lifelong treatment, contingent on reversible risk factors, recurrence risk, and the patient’s bleeding risk. One of the key challenges associated with anticoagulant administration in populations at HAs is that standard anticoagulants, such as warfarin, heparin, or direct oral anticoagulant drugs, may display different pharmacokinetics or pharmacodynamics due to the altered vascular environment and reduced oxygen levels. Additionally, given the changes in coagulation function, possibly occurring due to the plateau environment, adjustments in anticoagulant dosing or treatment strategies may be required. These unique environmental factors suggest that standard protocols may not be applicable in HA settings.
However, we acknowledge that the present study has a few limitations. First, relying on data from a single hospital, even one with a significant catchment area, may not fully represent the general population. Second, the lifestyle of Tibetan patients, including dietary habits, physical activity, and cultural practices, could markedly influence the outcomes of anticoagulant therapy. Third, traditional Tibetan medicine, such as herbal treatments or acupuncture, may also interact with conventional anticoagulant therapy. Fourth, genetic predispositions influencing clotting mechanisms in the Tibetan population could introduce bias in conclusions regarding anticoagulant effectiveness. Finally, unaccounted factors, such as climate-related stress and social determinants of health (e.g., social support and education level) may have skewed the analysis, thereby limiting the validity of our findings. Therefore, further research addressing these aspects is necessary to validate our findings and define protocols that are safe and effective for patients under similar conditions.
Further statistical analyses, including those using multivariate models, may help elucidate the relative contributions of these factors to the thrombotic risk in this patient population. Additionally, such analyses could be enhanced by incorporating qualitative data on patient history, medication regimens, and other clinical variables to gain a comprehensive understanding of the pathogenic mechanisms underlying the occurrence of VTE.
Real-world studies provide valuable insights into the impact of HA environments on the prevalence of VTE. These findings can inform the refinement of VTE risk assessment models, incorporating high-altitude environments as a distinct consideration in clinical guidelines. Based on such data, guidelines could recommend deploying more sensitive screening tools or adopting proactive imaging strategies in high-altitude regions, particularly for high-risk groups such as individuals with prolonged immobility, a history of thrombosis, or those recovering from surgery. Furthermore, these studies can help identify populations uniquely vulnerable to high-altitude conditions, such as individuals with polycythaemia, older adults, and postoperative patients, allowing for the development of targeted prevention and treatment strategies. VTE research at HAs also has the potential to garner attention to the health challenges posed by unique geographic conditions globally, paving the way for the establishment of globally standardized guidelines to support patients in similar environments, such as those characterized by hypoxia or extreme climates.
5.Conclusion
The main finding of the present cross-sectional study was that the prevalence of VTE in the Tibet Autonomous Region People’s Hospital between 2013 and 2023 increased annually from 0.74% to 1.12%. Additionally, VTE was mainly associated with the 18–64-year age group, low haemoglobin levels, and hypertension at HAs. Our findings provide a better understanding of the effects of HA environments on the occurrence and treatment of VTE, offering more effective prevention and treatment measures for patients with VTE in these regions. Our findings highlight the unique interplay of chronic hypoxia, elevated blood viscosity, and hypertension in driving VTE risk at high altitudes. Hypoxia-induced erythrocytosis, while often adaptive, may paradoxically increase thrombosis susceptibility by altering coagulation dynamics (Jha et al., 2018). These mechanisms, rather than transient external factors, likely underpin the observed trends.
Long-term observations continue to explore the influence of the HA environment on the risk factors for VTE, including hypoxemia, changes in blood viscosity, and lifestyle factors. Examining the differences in VTE incidence between residents of HA and low-altitude areas, along with the underlying physiological mechanisms underlying these differences, can aid the development of targeted strategies for preventing and treating VTE in HA regions. Longitudinal studies and databases should be established for patients with VTE in HA areas to analyse the impact of HA on VTE recurrence and treatment outcomes. Using large-data analytics, in conjunction with genetic, epidemiological, and clinical data, may also elucidate the characteristics and patterns of VTE occurrence in HA regions, providing a foundation for developing prevention strategies and improving treatment efficacy.
Data Availability
All relevant data are within the manuscript and its Supporting Information files.
Conflict of Interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This retrospective study was approved by the Medical Ethics Committee of Tibet Autonomous Region People’s Hospital(Approval No: ME-TBHP-25-KJ-028) with a waiver of informed consent.
Funding
This research receive grant natural science funding of tibet autonomous region, fund number: XZ2024ZR-ZY027(Z).
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of this manuscript.