https://orcid.org/0000-0002-6160-9915
Figures
Abstract
Background
Hip fracture surgery carries the highest postoperative risk of venous thromboembolism (VTE) among surgical populations. Guidelines recommend low molecular weight heparin (LMWH) over direct oral anticoagulants (DOACs) due to limited evidence. However, oral administration and cost considerations have led to widespread real-world use of DOACs. This study aimed to compare the effectiveness and safety of extended thromboprophylaxis with DOACs versus LMWH or unfractionated heparin (UFH) after hip fracture surgery.
Methods
We conducted a retrospective cohort study of adults undergoing hip fracture surgery at Hospital Alemán, Buenos Aires, Argentina (January 2011–June 2025), covered by the hospital’s healthcare insurance. Eligible patients were discharged with extended pharmacologic thromboprophylaxis (LMWH, UFH, or DOACs). Outcomes within 3 months included VTE, major or clinically relevant non-major bleeding (MB/CRNMB), and all-cause mortality. Multivariable Cox regression, Fine-Gray models, and propensity score adjustments were applied.
Results
Of 425 hip fractures, 340 cases (301 patients) met eligibility criteria. Extended prophylaxis used was LMWH/UFH in 102 cases and DOACs in 238. VTE occurred in 2.5% of DOAC and 3.9% of LMWH/UFH cases. MB/CRNMB occurred in 2.1% vs 5.9%, and mortality in 3.0% vs 3.9%, respectively. Adjusted hazard ratios for DOACs versus LMWH/UFH were 0.64 (95% CI, 0.16–2.49) for VTE, 0.69 (0.15–2.16) for bleeding, and 0.73 (0.22–2.37) for mortality.
Citation: Clavijo MM, Ruiz JI, Brindicci MR, Rodriguez P, Wauters V, Reparaz MdlAV, et al. (2026) Extended thromboprophylaxis after hip fracture surgery: Real-world evidence of direct oral anticoagulants versus low molecular weight heparin of unfractionated heparin. PLoS One 21(3): e0343020. https://doi.org/10.1371/journal.pone.0343020
Editor: Ufuk Demirci, Celal Bayar University: Manisa Celal Bayar Universitesi, TÜRKIYE
Received: November 9, 2025; Accepted: February 1, 2026; Published: March 12, 2026
Copyright: © 2026 Clavijo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The dataset supporting this study is available at the following repository: https://doi.org/10.5281/zenodo.18711092.
Funding: Financial support for this study was provided by Raffo. The funder had no role in the design, data collection, data analysis, and reporting of this study. This work was supported by academic financial assistance. No grant number or funding URL was provided.
Competing interests: he authors have declared that no competing interests exist.
Introduction
Hip fractures are a common and growing health problem worldwide, particularly among older adults, and are associated with substantial morbidity and mortality [1–3]. Among the complications that follow hip fracture surgery, venous thromboembolism (VTE) remains one of the most serious and potentially life-threatening events [4].
Patients undergoing orthopedic surgery face the highest risk of VTE compared with other surgical populations [5]. Before the routine use of pharmacologic thromboprophylaxis, clinical trials reported postoperative (POP) VTE rates of 15–30% [5].The introduction of systematic anticoagulant prophylaxis has reduced this risk to approximately 1–5% [5]. Nevertheless, in the specific context of hip fracture surgery, postoperative VTE incidence remains both higher and markedly heterogeneous, with studies reporting rates as high as 20% [6–8] and, in some cases, exceeding 50% [1,9].
Since the advent of direct oral anticoagulants (DOACs), numerous observational studies have explored their use after hip fracture surgery [10–16]. Collectively, these investigations suggest that DOACs may offer a feasible therapeutic alternative; however, the evidence is limited by small sample sizes and considerable imprecision. Current international guidelines, including those of the American Society of Hematology and the American College of Chest Physicians, recommend low molecular weight heparin (LMWH) over DOACs in this setting, citing the lack of robust supporting data [5,17]. Despite these recommendations, DOACs are increasingly prescribed in everyday clinical practice, driven by their ease of oral administration, lower costs, and the extrapolation of efficacy data from elective orthopedic surgery.
In this study, we aimed to evaluate the comparative effectiveness and safety of extended thromboprophylaxis with DOACs versus LMWH or unfractionated heparin (UFH) in patients undergoing hip fracture surgery in a real-world clinical setting.
Materials and methods
We conducted a retrospective cohort study using electronic health records from the Hospital Alemán healthcare plan insurance in Buenos Aires, Argentina. Access to the data for research purposes began on February 16, 2024. All authors are physicians working at the same institution, actively involved in patient care, and therefore have authorized access to the institutional system using individual professional identification. The data were accessed and analyzed in accordance with institutional ethical standards and regulatory requirements. No information that could directly identify individual participants was accessed or disclosed at any stage during or after data collection.
The study included adults (≥18 years) who underwent hip fracture surgery between January 2011 and June 2025 and were discharged with extended pharmacologic thromboprophylaxis. Patients were excluded if they had a diagnosis of venous thromboembolism (VTE) during the in-hospital period after hip fracture, an indication for therapeutic anticoagulation, or incomplete clinical data. Extended prophylaxis consisted of low-molecular-weight heparin (LMWH), unfractionated heparin (UFH), or direct oral anticoagulants (DOACs; rivaroxaban, apixaban, or dabigatran). According to standard clinical practice at our center, enoxaparin was the LMWH used for thromboprophylaxis. Anticoagulant regimens were as follows: enoxaparin 40 mg once daily (increased to 60 mg once daily in patients with a body mass index ≥30 kg/m² and reduced to 30 mg once daily in those with a creatinine clearance [CrCl] <30 mL/min), unfractionated heparin (UFH) 5,000 IU twice daily, rivaroxaban 10 mg once daily, apixaban 2.5 mg twice daily, and dabigatran 220 mg once daily. Dabigatran was not prescribed to patients with a CrCl < 60 mL/min or to those older than 70 years. In patients with a CrCl < 30 mL/min, apixaban was the only direct oral anticoagulant used after hospital discharge. No dose adjustments were applied for anticoagulants other than enoxaparin.
The primary outcome was VTE, defined as objectively confirmed deep vein thrombosis (DVT) or pulmonary embolism (PE) within 3 months after discharge. Secondary outcomes included a composite of major bleeding (MB) or clinically relevant non-major bleeding (CRNMB), and all-cause mortality. Bleeding was defined according to the ISTH classification as: a) MB: hemoglobin drop of >2 g/ dL, transfusion of >2 units of packed red blood cells, symptomatic bleed in a critical area (intracranial, retroperitoneal, intraspinal, intraocular), or fatal bleed; b) CRNMB: one which requires or prolongs hospitalization, results in laboratory testing, imaging, compression, a procedure, interruption of the study medication, or a change in concomitant therapies [18]. Time-to-event outcomes were assessed using Kaplan-Meier survival curves with log-rank tests, while multivariable Cox proportional hazards models were applied to adjust for potential confounders. Fine–Gray competing risk models were used to account for death as a competing event, and propensity score methods were employed to adjust for baseline differences across treatment groups. Analyses also accounted for clustering since there were patients who underwent more than one hip fracture surgery during the study period. For statistical analysis, Rstudio version 2023.06.0 + 421 was used.
Statement of ethics
This study protocol was reviewed and approved by Comité de ética independiente del Hospital Alemán (CEHIA), Av Pueyrredon 1640, Buenos Aires, Argentina. Tel: + 54 11 4827-7000. Informed consent was not obtained. Resolution 1480/2011 of the National Ministry of Health exempts retrospective cohort studies from the obligation to obtain informed consent. The confidentiality of each participant’s identity was adequately preserved according to local regulations (Administración Nacional de Medicamentos, Alimentos y Tecnología [ANMAT]).
Results
A total of 425 hip fractures were identified, of which 340 cases (301 patients) were eligible (Fig 1). Median age was 83 years (interquartile range [IQR] 77.3–88.4), with 81% female. The characteristics of the population analyzed overall and by treatment received are described in Table 1. All patients received either UFH or LMWH during hospitalization for hip fracture. In 10 of 340 cases (2.9%), the thromboprophylaxis strategy consisted of UFH, while the remaining patients received LMWH (enoxaparin). Thromboprophylaxis was initiated within 24 hours of hospital admission and was discontinued 12–24 hours before surgery. Postoperatively, the median time to treatment initiation was 15 hours (IQR12–23) in the overall population, 13 hours in the LMWH/UFH group (IQR 12–24), and 15 hours in the DOAC group (IQR 12–22), p = 0.14. Extended prophylaxis was prescribed for one month duration after discharge. It included LMWH/UFH in 102 cases (97 LMWH, 5 UFH) and DOACs in 238 cases (181 rivaroxaban, 48 apixaban, 2 dabigatran), Fig 1. DOACs use predominated from 2018 onwards (Fig 2).
During the 3-month follow-up, VTE occurred in 10 patients (2.9% overall): 6 cases in the DOAC group (2.5%) and 4 in the LMWH/UFH group (3.9%). Regarding the subtype of event, there were 6 cases of PE, 2 of distal DVT, and 2 of proximal DVT. All cases were symptomatic; the diagnosis of PE was established by CT scan, and DVT was diagnosed by Doppler ultrasonography. One PE event was fatal. It occurred in the LMWH/UFH group in an 81-year-old woman with COPD who developed bilateral pulmonary embolism involving the main pulmonary arteries. She was admitted five days after hospital discharge and died four days later due to respiratory failure. Among the 10 patients who developed VTE, the median age was 84.35 years (IQR 81.74–86.18), all were female, 7 patients had medial fractures, 7 underwent partial arthroplasty, 2 osteosynthesis, and 1 total arthroplasty. Regarding anesthesia, 9 patients received regional block with sedation, 6 patients had an ASA score ≥3. Comorbidities included chronic obstructive pulmonary disease in 3 patients, active cancer in 1, and current smoking in 2. Time to thromboprophylaxis initiation was not associated with VTE (HR 0.99, 95% CI 0.96–1.01) and was unrelated to time to surgery (p = 0.187) or anesthesia type (p = 0.710). Patient characteristics are detailed in the Supplementary Material (S1 Table).
Regarding secondary outcomes, MB/CRNMB was observed in 12 patients (3.5% overall), including 6 cases with DOACs (2.5%) and 6 with LMWH/UFH (5.9%). Among the 12 patients who experienced bleeding complications, 7 events were classified as major bleeding. These included 3 cases of gastrointestinal bleeding (2 occurring in patients receiving LMWH and 1 fatal event in a patient treated with rivaroxaban) and 4 cases of surgical-site bleeding (1 associated with apixaban, 1 with rivaroxaban, and 2 with LMWH). The fatal event occurred in a 91-year-old woman with severe aortic stenosis and a history of mucosal bleeding, including epistaxis, oral, and gastrointestinal bleeding. The remaining 5 events were classified as clinically relevant non-major bleeding (CRNMB), comprising 4 surgical-site hematomas (2 with LMWH, 1 with rivaroxaban, and 1 with apixaban) and 1 case of gastrointestinal bleeding in a patient receiving rivaroxaban. Patient characteristics are reported in the Supplementary Material (S2 Table).
A total of 11 deaths were recorded (3.2% overall): 7 patients in the DOAC group (3.0%) and 4 in the LMWH/UFH group (3.9%). Of the 11 deaths recorded, 6 were attributed to respiratory infection (4 of these patients underwent computed tomography scanning that ruled out pulmonary embolism), 1 to pulmonary embolism, 1 to gastrointestinal bleeding, and 1 to cardiac failure; in 2 cases, the cause of death was not reported. Patient characteristics are reported in the Supplementary Material (S3 Table). Kapplan-Meier curves for time to VTE and MB/CRNMB are shown in Figs 3 and 4.
The incidence of VTE, MB/CRNMB, and mortality stratified by clinical characteristics such as age, creatinine clearance, ASA score, fracture type, and type of surgery is shown in Supplementary S4 Table.
In multivariable Cox and Fine–Gray regression analyses, prior VTE was the only variable independently associated with VTE occurrence (hazard ratio [HR] 8.87, 95% CI 2.13–36.98, sub hazard ratio [SHR] 8.89, 97.5% CI 2.01–39.35), whereas thromboprophylaxis with DOACs was not significantly different from LMWH/UFH, although a nonsignificant reduction in VTE risk was observed (HR 0.64, 95% CI 0.16–2.49, SHR 0.62, 95% CI 0.16–2.40) (Table 2). Thromboprophylaxis with DOACs was not significantly associated with the risk of MB/CRNMB compared with LMWH/UFH (HR 0.69, CI 95% 0.15–2.16, SHR 0.51, 97.5% CI 0.13–2.03), whereas an ASA score ≥3 was independently associated with a significantly increased bleeding risk (HR 6.02, CI 95% 1.25–28.92, SHR 4.91, 97.5% CI 1.19–20.21), (Table 3). Age was the only variable independently associated with mortality (HR 1.11, CI 95% 1.00–1.22), and thromboprophylaxis with DOACs was not significantly associated with mortality when compared with LMWH/UFH (HR 0.73, CI 95% 0.22–2.37) (Table 4).
Discussion
This large real-world cohort provides worthy evidence on the comparative effectiveness of DOACs versus LMWH/UFH after hip fracture surgery. Although statistical power was limited by the low number of events, DOACs showed comparable rates of VTE, bleeding, and mortality. These findings are consistent with prior small randomized trials and observational studies suggesting non-inferiority of DOACs compared with LMWH.
We identified seven relevant publications, showing marked heterogeneity in study design, therapeutic strategies, diagnostic criteria, and follow-up. The largest cohort (>3000 patients) was limited to in-hospital events and found no difference between LMWH and DOACs regarding VTE or bleeding [16]. Only two studies extended follow-up beyond 35 days [13,14], while the others assessed outcomes up to 30–35 days postoperatively [10–12,15]. One RCT compared LMWH, DOACs, and sequential therapy (LMWH for 7 days followed by rivaroxaban), finding no significant differences [10]. In contrast, two retrospective studies reported lower VTE with DOACs(14, 15). Reported VTE incidence with DOACs ranged from 0% to 9.5% [10–15], and with LMWH from 1.1% to 26.6% [10–16]. Importantly, most studies included both in-hospital and post-discharge events. In our study, the 3-month post-discharge VTE incidence was 2.9% (2.5% with DOACs vs. 3.9% with LMWH/UFH). Including in-hospital events, the rate increased to 8%, consistent with prior reports.
Bleeding outcomes also showed variability across studies. We observed a 2.5% rate with DOACs and 5.9% with LMWH/UFH, which are slightly lower than most published data, likely explained by exclusion of in-hospital events. Reported rates varied widely, from 0.29% to 15.6% with DOACs and from 0.45% to 11.6% with LMWH [10–13,15]. One RCT with edoxaban found lower bleeding rates with DOACs vs. LMWH (3.4% vs. 6.9%) [11], whereas other studies suggested higher bleeding risk with DOACs, including one with significant differences [15].
Regarding mortality, Lassen et al. reported 0% mortality at 90 days with DOACs vs. 0.9% with LMWH [13]. In our study, mortality was 3.9% with LMWH/UFH and 3% with DOACs. Other studies reported variable 30-day mortality, ranging from 0% [11,15] to 7.4% with DOACs vs. 3.4% with LMWH [12], without significant differences in VTE- or bleeding-related deaths.
Although this was a single-center study with a limited sample, our cohort represents one of the largest reported to date in patients undergoing hip fracture surgery. No statistically significant differences were observed in VTE, MB/CRNMB, or mortality between patients receiving extended prophylaxis with DOACs versus LMWH/UFH. Despite the heterogeneity of published results, our findings in 340 patients are consistent with most prior reports.
This study, however, has several limitations that should be acknowledged. Its retrospective, single-center design entails an inherent risk of information bias and variability in data quality, inconsistency, and imprecision in data quality. In addition, the relatively low number of events resulted in wide confidence intervals in adjusted analyses, limiting the precision of hazard ratio estimates. Potential selection bias may have been introduced, as a larger proportion of patients received DOACs compared with LMWH/UFH. Furthermore, despite propensity score adjustments, some residual imbalance between treatment groups may persist due to unmeasured factors, such as socioeconomic status or temporal confounding. To address these limitations and enhance statistical power, we have established a collaborative project with the Hospital Italiano de Buenos Aires. This ongoing multicenter study will enable the evaluation of a larger cohort, thereby improving estimate precision and providing more robust evidence for the medical community.
Conclusions
In this retrospective cohort of patients undergoing hip fracture surgery, extended thromboprophylaxis with DOACs showed effectiveness and safety comparable to that of LMWH/UFH. The rates of VTE, MB/CRNMB and mortality were low and showed no statistically significant differences between groups, supporting the notion that DOACs represent a valid alternative for managing this high-risk population. These findings are particularly relevant in real-world practice, where oral administration and cost considerations may favor the use of DOACs.
Nevertheless, the inherent limitations of the retrospective, single-center design and the modest sample size require cautious interpretation of these results. Confirmation through multicenter, prospective studies with greater statistical power is needed to more precisely assess the benefits and risks of DOACs in this setting. The available evidence suggests that, DOACs can be considered an effective and safe option for extended thromboprophylaxis after hip fracture surgery.
Supporting information
S1 Table. Baseline characteristics of patients who developed VTE.
https://doi.org/10.1371/journal.pone.0343020.s001
(DOCX)
S2 Table. Baseline characteristics of patients who experienced bleeding.
https://doi.org/10.1371/journal.pone.0343020.s002
(DOCX)
S3 Table. Baseline characteristics of deceased patients.
https://doi.org/10.1371/journal.pone.0343020.s003
(DOCX)
S4 Table. Clinical events (VTE, bleeding, and mortality) stratified by baseline characteristics.
https://doi.org/10.1371/journal.pone.0343020.s004
(DOCX)
References
- 1. Wang T, Guo J, Long Y, Yin Y, Hou Z. Risk factors for preoperative deep venous thrombosis in hip fracture patients: a meta-analysis. J Orthop Traumatol. 2022;23(1):19. pmid:35391566
- 2. Bentler SE, Liu L, Obrizan M, Cook EA, Wright KB, Geweke JF, et al. The aftermath of hip fracture: discharge placement, functional status change, and mortality. Am J Epidemiol. 2009;170(10):1290–9. pmid:19808632
- 3. Walter N, Szymski D, Kurtz S, Alt V, Lowenberg DW, Lau E, et al. Factors associated with mortality after proximal femoral fracture. J Orthop Traumatol. 2023;24(1):31. pmid:37365418
- 4. Pincus D, Ravi B, Wasserstein D, Huang A, Paterson JM, Nathens AB, et al. Association Between Wait Time and 30-Day Mortality in Adults Undergoing Hip Fracture Surgery. JAMA. 2017;318(20):1994–2003. pmid:29183076
- 5. Falck-Ytter Y, Francis CW, Johanson NA, Curley C, Dahl OE, Schulman S, et al. Prevention of VTE in orthopedic surgery patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 Suppl):e278S–e325S. pmid:22315265
- 6. Niu S, Li J, Zhao Y, Ding D, Jiang G, Song Z. Preoperative deep venous thrombosis (DVT) after femoral neck fracture in the elderly, the incidence, timing, location and related risk factors. BMC Musculoskelet Disord. 2021;22(1):264. pmid:33706746
- 7. Song K, Yao Y, Rong Z, Shen Y, Zheng M, Jiang Q. The preoperative incidence of deep vein thrombosis (DVT) and its correlation with postoperative DVT in patients undergoing elective surgery for femoral neck fractures. Arch Orthop Trauma Surg. 2016;136(10):1459–64. pmid:27535672
- 8. Xia Z-N, Xiao K, Zhu W, Feng B, Zhang B-Z, Lin J, et al. Risk assessment and management of preoperative venous thromboembolism following femoral neck fracture. J Orthop Surg Res. 2018;13(1):291. pmid:30458869
- 9. Handoll HH, Farrar MJ, McBirnie J, Tytherleigh-Strong G, Milne AA, Gillespie WJ. Heparin, low molecular weight heparin and physical methods for preventing deep vein thrombosis and pulmonary embolism following surgery for hip fractures. Cochrane Database Syst Rev. 2002;(4):CD000305. pmid:12519540
- 10. Tang Y, Wang K, Shi Z, Yang P, Dang X. A RCT study of Rivaroxaban, low-molecular-weight heparin, and sequential medication regimens for the prevention of venous thrombosis after internal fixation of hip fracture. Biomed Pharmacother. 2017;92:982–8. pmid:28605879
- 11. Fuji T, Fujita S, Kawai Y, Nakamura M, Kimura T, Kiuchi Y, et al. Safety and efficacy of edoxaban in patients undergoing hip fracture surgery. Thromb Res. 2014;133(6):1016–22. pmid:24680549
- 12. Goh EL, Gurung PK, Ma S, Pilpel T, Dale JH, Kannan A, et al. Direct Oral Anticoagulants in the Prevention of Venous Thromboembolism Following Surgery for Hip Fracture in Older Adults: A Population-Based Cohort Study. Geriatr Orthop Surg Rehabil. 2020;11. pmid:31976153
- 13. Lassen MR, Haas S, Kreutz R, Mantovani LG, Holberg G, Turpie AGG. Rivaroxaban for Thromboprophylaxis After Fracture-Related Orthopedic Surgery in Routine Clinical Practice. Clin Appl Thromb Hemost. 2016;22(2):138–46. pmid:26408640
- 14. Zhang C, Xu B, Liang G, Zeng X, Yang C, Zhang F, et al. Rivaroxaban versus nadroparin for preventing deep venous thrombosis after total hip arthroplasty following femoral neck fractures: A retrospective comparative study. J Int Med Res. 2018;46(5):1936–46. pmid:29560772
- 15. Zheng X, Wang L, Chen Y, Liu L, Zhang Z, Xiao C. Rivaroxaban vs. enoxaparin for preventing venous thromboembolism after hip fracture operations: a retrospective cohort study. Front Surg. 2025;12:1483611. pmid:40416725
- 16. Nederpelt CJ, Breen KA, El Hechi MW, Krijnen P, Huisman MV, Schipper IB, et al. Direct Oral Anticoagulants Are a Potential Alternative to Low-Molecular-Weight Heparin for Thromboprophylaxis in Trauma Patients Sustaining Lower Extremity Fractures. J Surg Res. 2021;258:324–31. pmid:33187673
- 17. Anderson DR, Morgano GP, Bennett C, Dentali F, Francis CW, Garcia DA, et al. American Society of Hematology 2019 guidelines for management of venous thromboembolism: prevention of venous thromboembolism in surgical hospitalized patients. Blood Adv. 2019;3(23):3898–944. pmid:31794602
- 18. Bergmark BA, Kamphuisen PW, Wiviott SD, Ruff CT, Antman EM, Nordio F, et al. Comparison of Events Across Bleeding Scales in the ENGAGE AF-TIMI 48 Trial. Circulation. 2019;140(22):1792–801. pmid:31597460