Rising prevalence of venous thromboembolism in gastrointestinal cancers: a nationwide analysis of 205 million hospitalizations, 2016–2022

Introduction

Venous thromboembolism (VTE) is a chronic illness consisting of both deep vein thrombosis (DVT) and pulmonary embolism (PE), which is a major contributor to worldwide morbidity and mortality, affecting nearly 10 million people yearly (1). It can recur and lead to complications, long-term disability, and death.

Most VTE episodes are unprovoked, but patients often have multiple risk factors that can have a synergistic effect, increasing the VTE risk. Transient risk factors such as trauma, surgery, and periods of immobilization are important provoking factors. Active malignancy is another important risk factor for VTE, accounting for 20% of all cases (2). The incidence differs depending on the type of cancer, with gastrointestinal cancers (GICs) having a high incidence of VTE; 4–7% in colorectal, 10–12% in gastric, and 15% in pancreatic cancers as reported in a previous study (3). Another recently published abstract which analyzed an electronic health record cohort from 2018 to 2023 reported the 12-month VTE risk at 12.0% for pancreatic cancer, 10.8% for bile/gallbladder, and 9.5% for upper gastrointestinal cancers (4).

Prior Nationwide Inpatient Sample (NIS) analyses have examined the incidence of VTE in selected GI malignancies, specifically esophageal, gastric, colorectal, and pancreatic from 2016–2019 which analyzed 999,559 hospitalizations with VTE, and calculated the odds ratios for mortality in these malignancies (5). However, these studies were limited in scope due to fewer subtypes of cancer and a shorter timeframe. This study did not evaluate the annual trends of VTE incidence.

Our study expands upon this previous literature with a comprehensive analysis of all GICs as listed under methods across seven years from 2016–2022.

Our primary objective is to calculate the prevalence of lower extremity DVT, PE, and VTE in GIC hospitalizations and assess trends of prevalence from 2016–2022. We also calculated the incidence of inpatient deaths (IPD) for all hospitalizations, GICs, and hospitalizations with a diagnosis of VTE. Our secondary objective was to calculate the relative risk (RR) of VTE in GICs compared to all hospitalizations, RR of IPD with GICs compared to all hospitalizations, RR of IPD with VTE in GICs compared to IPD with VTE in all hospitalizations, and lastly, RR of IPD with VTE in GICs compared to the baseline risk of IPD with GICs. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-1-999/rc).

Methods

This is a retrospective cross-sectional observational study, which was performed using de-identified patient data from the NIS database, which is a part of the Healthcare Utilization Project (HCUP). NIS is the largest publicly available all-payer inpatient healthcare database designed to generate regional and national estimates of healthcare utilization, access, cost, quality, and outcomes for hospitalizations in the United States (U.S.). Unweighted, it accounts for approximately 7,000,000 hospitalizations, and weighted, it estimates 35,000,000 hospitalizations annually. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Statistical analysis

All statistical analyses were performed using Statistical Package for Social Sciences version 28.0 (SPSS, IBM Corporation, Armonk, NY, USA) and Google Sheets (Alphabet Inc., Mountain View, CA, USA). Line graphs were generated using Python (version 3.x) with the matplotlib library. An artificial intelligence (AI) assisted tool (Claude Opus 4.6, Anthropic) was used to generate the plotting code. All figures were reviewed and verified by the authors for accuracy against the source data. All hospitalizations involving patients aged ≥18 years were included. Presence of a diagnosis of MN of esophagus, stomach, small intestine, colon, rectosigmoid junction, rectum, anus and anal canal, hepatocellular carcinoma, intrahepatic biliary ductal carcinoma, MN gallbladder, biliary tract, pancreas, and neuroendocrine tumors (NETs), DVT and PE were determined using International Classification of Disease, 10th Revision (ICD-10) diagnostic codes. VTE included hospitalizations that had a diagnosis of either DVT or PE. IPD was calculated for all the hospitalizations included in the analysis using the information provided in the database. Chi-squared test was used to check the association between IPD and VTE. Significance of trend was calculated using Cochran-Armitage trend test. Results are statistically weighted. RR and confidence intervals (CI) were calculated for situations described above in “Introduction” section.

Results/Discussion

All the results are reported in chronological order from 2016–2022. Total 205,075,197 hospitalizations were included in the analysis. Prevalence of DVT for all hospitalizations was 1.1% in 2016, 1.1% in 2017, 1.1% in 2018, 1.1% in 2019, 1.3% in 2020, 1.4% in 2021, 1.3% in 2022 (P<0.001). Prevalence of DVT in MN pancreas was 4.7% in 2016, 4.7% in 2017, 5.2% in 2018, 4.9% in 2019, 5.1% in 2020, 5.0% in 2021, 5.1% in 2022 (P<0.001). Prevalence of DVT in MN stomach was 3.5% in 2016, 3.6% in 2017, 3.8% in 2018, 3.9% in 2019, 3.6% in 2020, 4.3% in 2021, 3.7% in 2022 (P<0.001). Prevalence of DVT in MN colon was 2.4% in 2016, 2.5% in 2017, 2.6% in 2018, 2.5% in 2019, 2.4% in 2020, 2.7% in 2021, 2.5% in 2022 (P=0.01). Prevalence of DVT with individual GICs can be reviewed in Table 1.

Prevalence of PE for all hospitalizations was 1.2% in 2016, 1.2% in 2017, 1.3% in 2018, 1.3% in 2019, 1.5% in 2020, 1.8% in 2021, 1.6% in 2022 (P<0.001). Prevalence of PE in MN pancreas was 4.9% in 2016, 5.1% in 2017, 5.4% in 2018, 5.3% in 2019, 5.9% in 2020, 6.1% in 2021, 6.3% in 2022 (P<0.001). Prevalence of PE in MN stomach was 4.3% in 2016, 3.7% in 2017, 4.3% in 2018, 4.5% in 2019, 4.7% in 2020, 5.3% in 2021, 4.9% in 2022 (P<0.001). Prevalence of PE in MN colon was 2.4% in 2016, 2.6% in 2017, 2.6% in 2018, 2.7% in 2019, 2.8% in 2020, 2.9% in 2021, 2.8% in 2022 (P<0.001). Prevalence of PE with individual GICs can be reviewed in Table 2.

Prevalence of VTE for all hospitalizations was 2.0% in 2016, 2.0% in 2017, 2.0% in 2018, 2.0% in 2019, 2.4% in 2020, 2.7% in 2021, 2.5% in 2022 (P<0.001. Prevalence of VTE in MN pancreas was 7.8% in 2016, 8.0% in 2017, 8.4% in 2018, 8.2% in 2019, 8.8% in 2020, 8.8% in 2021, 8.9% in 2022 (P<0.001). Prevalence of VTE in MN stomach was 6.6% in 2016, 6.0% in 2017, 6.6% in 2018, 6.8% in 2019, 6.8% in 2020, 7.8% in 2021, 7.0% in 2022 (P<0.001). Prevalence of VTE in MN colon was 4.0% in 2016, 4.2% in 2017, 4.3% in 2018, 4.3% in 2019, 4.5% in 2020, 4.6% in 2021, 4.4% in 2022 (P<0.001). Prevalence of VTE with individual GICs can be reviewed in Table 3.

Prevalence of coronavirus disease 2019 (COVID-19) in all hospitalizations was 0.2% in 2019, 6.2% in 2020, 8.4% in 2021, and 6.8% in 2022. Prevalence of COVID-19 in GICs was 0.1% in 2019, 1.8% in 2020, 2.5% in 2021, and 4.3% in 2022. Among all hospitalizations with COVID-19, the prevalence of VTE was 2.1% in 2019, 4% in 2020, 6.2% in 2021, and 4.6% in 2022. Among GIC hospitalizations with COVID-19, the prevalence of VTE was 12% in 2019, 7.1% in 2020, 8.6% in 2021, and 7.8% in 2022. The prevalence of COVID-19 and VTE among individual GICs can be reviewed in Table 4.

Line graphs for prevalence of DVT, PE, and VTE for select GICs can be reviewed in Figures 1-3, respectively.

Figure 1 DVT prevalence. DVT, deep vein thrombosis; GIC, gastrointestinal cancer; MN, malignant neoplasm.

Figure 2 PE prevalence. GIC, gastrointestinal cancer; MN, malignant neoplasm; PE, pulmonary emboli.

Figure 3 VTE prevalence. GIC, gastrointestinal cancer; MN, malignant neoplasm; VTE, venous thromboembolism.

An epidemiologic study looking at the UK Clinical Practice Research Datalink from 2001–2011 reported the incidence of first and recurrent VTE in patients with active cancers. This study reported the incidence in events per 100 person-years and found annualized incidence rates of 4–7% in colon cancer, 10–12% for stomach cancer, and 15% for pancreatic cancers (3). Our study observed the highest inpatient VTE rate of 8.4% in pancreatic cancer among all GICs, followed by MN gall bladder (7.05%) and stomach (6.05%). VTE for MN colon was 4.32% in our study.

There was a statistically significant trend of rising incidence of DVT, PE, and VTE for all hospitalizations, total GICs, and individual GICs (except DVTs in MN esophagus and PE in NETs). For all hospitalizations, the VTE incidence remained stable from 2016–2019, followed by a rise from 2020, which coincided with the COVID-19 pandemic. This is consistent with the well-established correlation between COVID-19 and increased VTE risk (6). In contrast, the incidence of DVT, PE, and VTE was steadily rising for GICs even before it rose sharply in 2020, suggesting a trend towards rising VTE in GICs before the COVID-19 pandemic. The prevalence of COVID-19 was much lower in GICs compared to all hospitalizations. COVID-19 was associated with higher prevalence of VTE for all hospitalizations, including GICs (Table 4).

A previous NIS study reported a rising trend of VTE in pancreatic cancer from 2005–2011 (7). Another study utilizing the Registro Informatizado de la Enfermedad Trombo Embólica (RIETE) calculated VTE incidence in 5-year periods and noted a rising trend of VTEs in colorectal and pancreatic cancers, but not in esophageal, stomach, liver, and biliary tract malignancies. This study had a much smaller sample size (8). Another study looking at Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (CDC-WONDER) platform noted a rise in VTE-related mortality in pancreatic cancers with a statistically significant trend from 2016–2020 (9). We hypothesize that the rise in VTE rates could be explained by improvements in overall survival in GICs with improvements in cancer-directed and supportive care treatments (10). As these patients stay alive longer, their exposure time to risk factors for VTE increases, thus resulting in increased incidence. We also noted an increase in the mean age of GIC hospitalizations as follows: 66.01 (standard deviation 13.162) years in 2016, 66.24 (13.015) years in 2017, 66.30 (12.898) years in 2018, 66.48 (12.873) years in 2019, 66.54 (12.952) years in 2020, 66.73 (12.809) years in 2021, and 66.80 (12.842) years in 2022. Increasing age is a known risk factor for VTE (11). Another CDC-WONDER study utilizing data from 1999 to 2020 assessed cancer-associated VTE mortalities among older adults (≥65 years) and reported an age-adjusted mortality rate rise from 16.8 per 100,000 to 22.8 per 100,000 in 2020, thus corroborating our findings (12).

We found that GICs have more than double the risk for VTE relative to an average hospitalization (RR 2.44; 95% CI: 2.43–2.45). IPD for all hospitalizations was 2.6%. IPD with VTE was 7.4%. IPD for all GICs was 6.0%. IPD for GICs with VTE was 11.5%. RR of IPD with VTE for all hospitalizations was 2.81 (95% CI: 2.80–2.82). RR of IPD for all GICs (compared to baseline IPD for all hospitalizations) was 2.27 (95% CI: 2.26–2.28). RR of IPD with VTE in GICs (compared with baseline IPD for GICs) was 1.91 (95% CI: 1.89–1.94). Values of RR for VTE and IPD with individual GICs, and RR of IPD with VTE in GICs can be reviewed in Table 5.

A previous study, which was performed using the NIS database for 2016–2019, reported high mortality with VTE among GICs, with the highest adjusted odds ratio of death (2.701; 95% CI: 1.989–3.669) in esophageal cancer. However, this study was limited in its scope, as it only looked at 4 GICs, namely esophageal, gastric, pancreatic, and colorectal cancers (5). Our analysis reported a RR of 2.13 (95% CI: 2.00–2.27) for IPD with MN esophagus with VTE, which is consistent with this previous study. By including additional malignancy subtypes and extending the timeframe through 2022, our study was able to provide a more comprehensive and up-to-date picture.

There are many strengths associated with this study, the most important being that the analysis was performed on a large nationally representative dataset. We included all subtypes of GICs and trended them over a period of 7 years. No prior studies have provided a comprehensive analysis of all GICs. Our study includes the data up to 2022, which is the latest available data at the time of the manuscript submission. We provided RR values for VTE and IPD with all GICs, which can be very informative in educating the GIC patients about their risks of VTE by assigning numeric values to this risk.

We acknowledge that this analysis does have limitations that would be expected with studies performed using the NIS database. It is a cross-sectional retrospective analysis and is subject to confounding associated with a retrospective analysis. We intend to perform a multivariate analysis of confounding variables in future studies that will address each individual GICs and variables specific to them. ICD-10 codes were used to identify the diagnoses of GICs and VTE. Clinical information, including vital signs, lab values, cancer staging, histopathology, imaging studies, and medications administered, was not available. NIS is an inpatient database that tracks hospitalizations from admission to discharge. Thus, outpatient and longitudinal follow-up data were not available. Treatment of cancers is predominantly an outpatient process, and inpatient admissions likely represent a selective sick patient population. Thus, inclusion of only inpatient hospitalizations is a potential cause of selection bias. The accuracy of the results reported is dependent on the accuracy of the data as reported in the NIS database (13,14). Information regarding the type and dose of anticoagulation administered was not available. Since the study involved data from 2016–2022, it would be reasonable to assume that these patients had access to prophylactic anticoagulation, as it was a standard and widely adopted practice in the US during this study period, based on the Surgeon General’s Call to Action to Prevent DVT and PE in 2008; and multiple studies that demonstrated efficacy and safety of pharmacologic prophylaxis (15-17). To overcome this limitation, analysis using a database where this information is provided is needed. Assessing the impact of other risk factors, including metastatic disease, and surgical or procedural interventions on the incidence of VTE in GICs is beyond the scope of this manuscript. We plan to address these limitations in future in-depth analyses of individual GICs.

Conclusions

In conclusion, we found that the VTE risk with GICs was more than double that of all hospitalizations. This increased VTE prevalence was associated with increased mortality. Pancreatic and biliary tract cancers had the highest VTE risk, which was consistent with prior literature. We report a statistically significant trend of rising VTE for all hospitalizations, including GICs, from 2016 to 2022. VTE was trending up in GICs before 2020, when COVID-19 became an important risk factor. This finding occurred during a time when anticoagulation prophylaxis for VTE in hospitalized patients was a widely adopted practice in the U.S. These results are specifically relevant to inpatient healthcare providers. As such, they should maintain a low threshold to evaluate hospitalized GIC patients for signs or symptoms of VTE. Our study reports RR values of VTE for individual GICs, which can aid physicians in decision-making, patient education, and the development of future guidelines. Further research is needed to assess the additional risk factors that may be contributing to the trend of increasing VTE in GICs.

Acknowledgments

During the preparation of the original submission, no Generative AI or AI-assisted Technologies were used. During the revision of this work, Claude Opus 4.6 (Anthropic) was used to assist with figure generation and editorial refinement. The authors reviewed and edited the content and take full responsibility for the content of the publication.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-1-999/rc

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-1-999/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-1-999/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Raskob GE, Angchaisuksiri P, Blanco AN, et al. Thrombosis: a major contributor to global disease burden. Arterioscler Thromb Vasc Biol 2014;34:2363-71. [Crossref] [PubMed]
2. Heit JA. Epidemiology of venous thromboembolism. Nat Rev Cardiol 2015;12:464-74. [Crossref] [PubMed]
3. Cohen AT, Katholing A, Rietbrock S, et al. Epidemiology of first and recurrent venous thromboembolism in patients with active cancer. A population-based cohort study. Thromb Haemost 2017;117:57-65.
4. Lam B, Ryu J, Jafari O, et al. Epidemiology of cancer-associated venous thromboembolism in a nationwide electronic healthrecord database. Blood 2025;146:4898.
5. Darweesh M, Haddaden M, Fatima Z, et al. The Incidence and Outcomes of Venous Thromboembolisms in Patients Admitted with Different Gastrointestinal Malignancies. Gastrointest Tumors 2023;10:67-73. [Crossref] [PubMed]
6. Zdanyte M, Rath D, Gawaz M, et al. Venous Thrombosis and SARS-CoV-2. Hamostaseologie 2022;42:240-7. [Crossref] [PubMed]
7. Atodaria KP, Cohen SJ, Dhruv S, et al. Trends in Palliative Care, Hospice Utilization, and Outcomes in Hospitalized Pancreatic Cancer Patients: A Nationwide Analysis. Cureus 2022;14:e29351. [Crossref] [PubMed]
8. Bertoletti L, Madridano O, Jiménez D, et al. Cancer-Associated Thrombosis: Trends in Clinical Features, Treatment, and Outcomes From 2001 to 2020. JACC Cardiooncology 2023;5:758-72.
9. Hassan IN, Ibrahim M, Yaqub S, et al. National trends in venous thromboembolism-related mortality among pancreatic cancer patients in the United States, 1999-2020. Thromb J 2025;23:73. [Crossref] [PubMed]
10. Siegel RL, Miller KD, Wagle NS, et al. Cancer statistics, 2023. CA Cancer J Clin 2023;73:17-48. [Crossref] [PubMed]
11. Akrivou D, Perlepe G, Kirgou P, et al. Pathophysiological Aspects of Aging in Venous Thromboembolism: An Update. Medicina (Kaunas) 2022;58:1078. [Crossref] [PubMed]
12. Saad M, Batool RM, Waqas SA, et al. Unveiling the trends: Growing cancer and venous thromboembolism mortality in older adults in the United States, 1999-2020. Thromb Res 2025;247:109259. [Crossref] [PubMed]
13. Kohli M, Atodaria KP, Patel R, et al. Incidence of deep vein thrombosis and pulmonary embolism in closed foot and ankle fractures treated with conservative versus surgical management: A nationwide analysis. Injury 2024;55:111859. [Crossref] [PubMed]
14. Atodaria KP, Dhruv S, Bruno JM, et al. Early Colonoscopy in Hospitalized Patients With Acute Lower Gastrointestinal Bleeding: A Nationwide Analysis. Gastroenterology Res 2022;15:162-72. [Crossref] [PubMed]
15. Office of the Surgeon General (US); National Heart, Lung, and Blood Institute (US). The Surgeon General’s Call to Action to Prevent Deep Vein Thrombosis and Pulmonary Embolism. Rockville, MD, USA: Office of the Surgeon General; 2008.
16. Samama MM, Cohen AT, Darmon JY, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. The New England Journal of Medicine 1999;341:793-800.
17. Cohen AT, Davidson BL, Gallus AS, et al. Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. BMJ 2006;332:325-9. [Crossref] [PubMed]