Risk factors for postoperative deep vein thrombosis during hospitalization in patients with hip fractures: a prospective nested case-control study Authors: Zheng Chen, Xiaoyan Wang, Jiao Zhang, Jing Zhang, Hui Xue, Wei Chen, Luqin Di Published: 2026 ABSTRACT AbstractThis study aimed to investigate the risk factors for postoperative deep vein thrombosis (DVT) in patients with hip fractures during hospitalization to establish a foundation for early prevention and management of DVT. We conducted a prospective nested case-control study involving patients with hip fractures admitted to our hospital from January 2023 to December 2024. Patients with postoperative DVT formed the case group, while patients without DVT were matched in a 1:2 ratio according to the same department, identical fracture type, and comparable surgery duration (± 5 days) for the control group. The demographics, clinical data, and laboratory indicators were statistically evaluated in both groups. A total of 72 DVT cases and 144 controls were matched. Conditional logistic regression analysis showed that age (P<0.001, OR = 1.147), time from injury to surgery (P = 0.045, OR = 1.128), and D-dimer levels (P<0.001, OR = 1.100) were independent risk factors for postoperative DVT. The ROC curve analysis indicated that an age of 63 years, a duration of 5 days from injury to surgery, and a D-dimer level of 4.675 mg/L were the predictive thresholds for evaluating the risk of DVT. Besides, excessive sweating following antipyretic administration(meaning that sweat soaks through an entire set of hospital gowns) was associated with DVT in univariate analysis but did not reach statistical significance in the conditional logistic regression analysis (P = 0.307) and calls for additional research as an exploratory clinical indicator. Patients with hip fractures exhibit an increased prevalence of DVT. Our findings could help identify high-risk patients and enable more targeted interventions at an early stage.Subject terms: Diseases, Health care, Medical research, Risk factors Introduction Hip fractures, including femoral neck fractures and intertrochanteric fractures, are common orthopedic conditions, accounting for approximately 12% of adult systemic fractures1. In recent years, hip fractures have become a progressively significant public health issue. It is estimated that 4.5 million people globally will get hip fractures by 2050, predominantly in Asia, especially in China2. It is speculated that the incidence of hip fractures in China will exceed 1.3 million by 2050, with direct medical costs potentially reaching 5 billion dollars, bringing a heavy economic burden to the country3. The one-year mortality rate of hip fracture might attain 20% to 40%, and only 30% -40% of survivors can recover to their previous functional state4. The unfavorable prognosis of patients is closely associated with the occurrence of complications. Lower limb deep vein thrombosis (DVT) is the most common complication of hip fractures. Despite prophylactic medication therapy, 11.1% to 31.2% of hip fracture patients still experience DVT5,6, and the incidence postoperatively is higher than preoperatively7. DVT can result in chronic pain, skin ulcers, and significantly affect the quality of life, perhaps culminating in fatal pulmonary embolism8. According to reports, 1.99% of hip fractures can lead to pulmonary embolism9, and pulmonary embolism accounts for 13% of hip fracture deaths10. Early identification of DVT risk factors and standardized treatment can reduce the occurrence of fatal pulmonary embolism. Currently, there is no consensus on the risk factors for postoperative DVT in patients with hip fractures, and most studies are retrospective research11,12.These studies inevitably suffer from limitations such as recall bias and incomplete recording of certain perioperative variables. Prospective cohort studies are effective for identifying the risk factors for postoperative DVT, but they are time-consuming and require a large amount of human and material resources13. Prospective nested case-control studies can provide preliminary evidence while increasing efficiency in the absence of cohort studies by conducting follow-up observations based on a predefined cohort. Therefore, the main objective of this study was to investigate the risk factors for DVT formation in hip fracture patients after surgery through a prospective nested case-control design. Materials and methods Study population We chose 825 hip fracture patients who were hospitalized in our hospital from January 2023 to December 2024 as research subjects for the study cohort. The endpoint is whether DVT occurs at discharge. Inclusion criteria: age ≥ 18 years; hip fracture diagnosed by imaging (femoral neck fracture, intertrochanteric fracture); underwent surgical treatment; the preoperative Doppler Ultrasound (DUS) results were negative.Exclusion criteria: pathological fractures or old fractures (time from injury to surgery > 21 days); history of thrombotic diseases; combined with coagulation disorders or severe liver and kidney diseases. Patients with DVT in this cohort were included in the case group, while patients without DVT were matched in a 1:2 ratio based on the same department, same fracture type, and similar surgical period (± 5 days) for the control group. A total of 216 patients were enrolled in this study, including 72 patients with DVT and 144 matched controls. All patients received conventional anticoagulation therapy (injection of low molecular weight heparin, once daily) during their hospitalization. Age and sex were deliberately not matched to allow evaluation of their potential association with DVT, as these were variables of interest. However, this method might result in residual confounding, which we attempted to address through multivariable adjustment using conditional logistic regression. The study was approved by the hospital ethics committee (approval number: W2023-013-1), and all methods were carried out in accordance with relevant guidelines and regulations. Sample size consideration A post-hoc power analysis was performed based on the actual sample size (72 DVT cases and 144 matched controls). Assuming a two-sided significance level of α = 0.05 and a statistical power of 1-β = 0.80, the minimum detectable odds ratio for a binary risk factor with a 20% exposure prevalence in the control group was 2.2. This indicates that the study had adequate statistical power to identify clinically meaningful effect sizes. Diagnosis of DVT Doppler ultrasound was used to detect the emergence of DVT, with a positive diagnosis based on deep vein obstruction or persistent intraluminal filling defect14. All patients were tested for DUS since admission, postoperatively, and before discharge. Additional scans were mandated if DVT was suspected due to symptoms such as limb edema and lower extremity pain. Investigation tools We performed a comprehensive analysis of current studies, consulted expert opinions, and finally developed a table of indicators. The table mainly includes the following content. ① General patient information: gender, age, BMI, fracture type, mechanism of injury, time from injury to admission, manner of admission; ② Past history: smoking, diabetes, hypertension, cardiovascular and cerebrovascular diseases, etc.; ③ Clinical data: time between injury and surgery, history of blood transfusion during hospitalization, use of dehydrating drugs, fever, use of antipyretic medications, excessive sweating following antipyretic use, standardized ankle pump exercises, limb elevation, adequate intake, mechanical prophylaxis, ASA scores of anesthesia, anesthesia duration, operation time, blood loss, use of hemostatic during surgery; ④ Laboratory indicators: blood type, white blood cells(WBC), hemoglobin(HGB), hematocrit(HCT), platelets(PLT), C-reactive protein(CRP), homocysteine(HCY), total cholesterol(TC), triglycerides(TG), high-density lipoprotein cholesterol(HDL-C), low-density lipoprotein cholesterol(LDL-C), sodium(NA), potassium(K), creatinine(CREA), osmotic pressure(OSM), prothrombin time(PT), prothrombin time activity(PTA), international normalized ratio(INR), fibrinogen(FIB), activated partial thromboplastin time(APTT), activated partial thromboplastin time ratio(APTT-R), thrombin time(TT), thrombin time ratio(TT-R), D-Dimer, fibrin/fibrinogen degradation products(FDP), Antithrombin III(AT III). Data collection Data collection is carried out by three fixed researchers using unified survey forms. To ensure the reliability and consistency of the data, researchers received unified training before data collection. Data collection is conducted in layers according to the following nodes: Within 24 h after surgery, baseline data collection is finalized through case review, on-site inquiries, and physical examination, covering patient general information, medical history, medication history, laboratory test results, and intraoperative data; Daily dynamic monitoring indicators after surgery: whether there is fever (≥ 38.0 ℃), whether there is excessive sweating after using antipyretic drugs (excessive sweating was defined as visible diaphoresis documented by nursing staff or reported by the patient after administration of antipyretic medication for fever, requiring a change of hospital gowns or bedclothes), whether ankle pump exercise is performed regularly (3–4 times a day, 20–30 groups each time), whether the affected limb is elevated (the affected limb is elevated by 30 °, using a professional elevation pillow), whether the intake is sufficient (daily intake of 1500 ml-2500 ml, providing patients with a graduated water cup); On the day of DVT diagnosis, gather thrombosis-related data and correlate instances to concurrently extract exposure data. Statistical analysis SPSS 25.0 (IBM, New York, USA) was used for statistical analysis. The enumeration data are presented in the form of frequency and percentage (%), and group comparison is performed using the chi-square test or Fisher’s exact probability method. For the measurement data, the distribution status was first explored. Normal distribution data were presented as mean ± standard deviation, and analyzed with the independent sample t-test; The rank sum test is used for non-normal distribution data. P < 0.05 was considered statistically significant. And conditional logistic regression analysis on the significantly different factors was conducted to evaluate the independent risk factors for DVT after hip fracture surgery. To avoid overfitting, we adhered to the events-per-variable (EPV) principle, which recommends at least 10 outcome events per predictor variable. Given the number of DVT events (n = 72), the primary model was limited to a maximum of 7 predictors. Clinical relevance, statistical significance, and exploratory interest were taken into consideration when choosing the variables. Receiver operating characteristic curve was used to evaluate the diagnostic performance of each independent risk factor based on the area under the curve.The larger area under the curve means the better diagnostic performance. Results We adopted a prospective nested case-control study design by matching 72 hip fracture patients who experienced DVT after surgery with 144 hip fracture patients who had similar fracture types and surgical dates from the same department but did not suffer from DVT (ratio of 1:2). Therefore, this resulted in a cohort of 216 patients (Fig. 1). Flow chart of patients inclusion and exclusion. Table 1 presents the demographic features and parameter results of patients in the DVT group compared to the non-DVT group. Univariate analysis was conducted on each parameter, and it was found that there was a significant difference (P < 0.05) in gender, age, fracture type, time from injury to surgery, and smoking history between the two groups of patients; There was no significant difference between the two groups in BMI, blood type, fracture side, time from injury to admission, hypertension, diabetes, cerebrovascular disease, and cardiovascular disease history (P > 0.05). Table 1Univariate analysis of demographic characteristics (n/%).VariablesNon-DVT group (n = 144)DVT group (n = 72)Statistical value P Gender (n/%)12.8090.000*Male77/53.4720/27.78Female67/46.5352/72.22Age (, years)61.22 ± 13.0374.61 ± 11.54−7.3940.000*Fracture type (n/%)10.8610.001*femoral neck fracture92/63.8929/40.28intertrochanteric fractures52/36.1143/59.72BMI (, kg/m2)24.05 ± 3.6622.98 ± 4.511.8680.063Blood type (n/%)2.5600.464A31/21.5319/26.39B52/36.1126/36.11O47/32.6417/23.61AB14/9.7210/13.89Fracture side (n/%)0.0370.847Left74/51.3936/50.00Right70/48.6136/50.00Time from injury to admission (, h)34.59 ± 57.9432.67 ± 68.140.2050.838Time from injury to surgery (, d)6.41 ± 3.447.61 ± 4.38−2.2020.029*Smoking history (n/%)7.7140.005*No104/72.2264/88.89Yes40/27.788/11.11Hypertension (n/%)No89/61.8142/58.330.2420.622Yes55/38.1930/41.67Diabetes (n/%)0.7990.371No117/81.2562/86.11Yes27/18.7510/13.89Cerebrovascular disease (n/%)2.1860.139No130/90.2860/83.33Yes14/9.7212/16.67Cardiovascular disease (n/%)0.1900.663No127/88.1962/86.11Yes17/11.8110/13.89Note: * P < 0.05, statistically significant. Univariate analysis of demographic characteristics (n/%). Note: * P < 0.05, statistically significant. Clinical data differentiating the DVT and non-DVT groups are presented in Table 2. Significant differences between two groups were found in terms of the history of blood transfusion during hospitalization, use of dehydrating drugs, excessive sweating after using antipyretic drugs, ankle pump movement, mechanical prophylaxis, ASA classification, blood lose, and use of hemostatics during surgery (all P<0.05 )。. Table 2Univariate analysis of clinical characteristics (n/%).VariablesNon-DVT group (n = 144)DVT group (n = 72)Statistical value P History of blood transfusion during hospitalization (n/%)26.6700.000*No101/70.1424/33.33Yes43/29.8648/66.67Use of hemostatic drugs (n/%)22.1860.000*No125/86.8142/58.33Yes19/13.1930/41.67Use of analgesic (n/%)0.0150.904No29/20.1414/19.44Yes115/79.8658/80.56Fever (n/%)0.0430.836No98/68.0650/69.44Yes46/31.9422/30.56Use of antipyretics (n/%)3.0480.081No130/90.2859/81.94Yes14/9.7213/18.06Excessive sweating after using antipyretic drugs (n/%)4.0210.045*No130/90.2858/80.56Yes14/9.7214/19.44Ankle pump movement (n/%)21.0940.000*No33/22.9239/54.17Yes111/77.0833/45.83Elevate the affected limb (n/%)4.4830.034*No66/45.8344/61.11Yes78/54.1728/38.89Sufficient intake (n/%)3.3010.069No35/24.3126/36.11Yes109/75.6946/63.89Mechanical prophylaxis (n/%)24.7710.000*No43/29.8647/65.28Yes101/70.1425/34.72Preoperative fluid infusion (n/%)1.9730.160No123/85.4256/77.78Yes21/14.5816/22.22ASA classification (n/%)9.4020.002*I, II99/68.7534/47.22III, IV45/31.2538/52.78Type of anesthesia (n/%)1.0400.308general anesthesia113/78.4752/72.22combined spinal and epidural anesthesia31/21.5320/27.78Anesthesia time (, h)2.57 ± 0.822.71 ± 1.06−1.0110.313Time of surgery (, h)1.44 ± 0.661.53 ± 0.91−0.7450.457Blood loss (, ml)183.47 ± 174.37241.67 ± 184.81−2.2660.024*Use of hemostatics during surgery (n/%)30.4170.000*No106/73.6125/34.72Yes38/26.3947/65.28Note: * P < 0.05, statistically significant. Univariate analysis of clinical characteristics (n/%). Note: * P < 0.05, statistically significant. The laboratory results showed that compared with the DVT group, patients without DVT had significantly higher levels of hemoglobin, hematocrit, albumin, and antithrombin III, while D-dimer levels were lower (all P < 0.05). There was no statistically significant difference in other laboratory indicators between the two groups (Table 3). Table 3Univariate analysis of laboratory results ().Laboratory resultsNon-DVT group (n = 144)DVT group (n = 72)t P WBC (×109/L)8.76 ± 2.698.38 ± 2.870.9680.334HGB (g/L)126.33 ± 18.17111.20 ± 17.355.8520.000*HCT (%)37.72 ± 5.4433.84 ± 8.264.1230.000*PLT (×109/L)211.22 ± 65.09208.01 ± 74.220.3260.745ALB (g/L)40.79 ± 5.0638.70 ± 5.852.7130.007*CRP (mg/L)30.13 ± 30.2738.11 ± 36.05−1.6470.101HCY19.25 ± 18.0716.66 ± 7.991.1150.159TC (mmol/L)4.12 ± 1.014.21 ± 0.94−0.6180.537TG (mmol/L)1.19 ± 0.741.15 ± 0.890.3080.758HDL-C (mmol/L)1.25 ± 0.341.30 ± 0.27−1.0330.303LCL-C (mmol/L)2.54 ± 0.842.33 ± 0.731.8100.072NA (mmol/L)137.99 ± 2.97137.62 ± 3.260.8350.405K (mmol/L)3.82 ± 0.383.82 ± 0.530.0440.965CERA (umol/L)63.75 ± 16.1669.98 ± 25.98−1.8640.065OSM (mOsml/L)296.08 ± 6.89296.75 ± 7.91−0.6370.525PT (s)12.14 ± 0.9813.51 ± 13.24−1.2420.216PTA (%)88.62 ± 9.5690.56 ± 10.90−1.3390.182INR1.09 ± 0.081.07 ± 0.071.7420.083FIB (g/L)3.51 ± 0.753.48 ± 0.790.2340.815APTT (s)31.39 ± 3.0930.82 ± 3.211.2670.207APTT-R0.89 ± 0.090.87 ± 0.131.6250.106TT (s)14.08 ± 1.1614.13 ± 1.19−0.2680.789TTR0.97 ± 0.070.97 ± 0.080.1440.886D-DIMER (mg/L)7.23 ± 8.6115.47 ± 31.73−2.1660.033*FDP (mg/L)18.67 ± 36.6927.08 ± 32.83−1.6340.104ATIII (%)93.84 ± 14.2488.86 ± 12.792.4900.014*WBC white blood cells; HGB hemoglobin; HCT hematocrit; PLT platelets; ALB albumin; CRP C-reactive protein; HCY homocysteine; TC total cholesterol; TG triglycerides; HDL-C high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; NA sodium; K potassium; CREA creatinine; OSM osmotic pressure; PT prothrombin time; PTA prothrombin time activity, INR international normalized ratio, FIB fibrinogen, APTT activated partial thromboplastin time, APTT-R activated partial thromboplastin time ratio, TT thrombin time, TT-R thrombin time ratio, FDP fibrin/fibrinogen degradation products, ATIII Antithrombin III‌.Note: * P < 0.05, statistically significant. Univariate analysis of laboratory results (). WBC white blood cells; HGB hemoglobin; HCT hematocrit; PLT platelets; ALB albumin; CRP C-reactive protein; HCY homocysteine; TC total cholesterol; TG triglycerides; HDL-C high-density lipoprotein cholesterol; LDL-C low-density lipoprotein cholesterol; NA sodium; K potassium; CREA creatinine; OSM osmotic pressure; PT prothrombin time; PTA prothrombin time activity, INR international normalized ratio, FIB fibrinogen, APTT activated partial thromboplastin time, APTT-R activated partial thromboplastin time ratio, TT thrombin time, TT-R thrombin time ratio, FDP fibrin/fibrinogen degradation products, ATIII Antithrombin III‌. Note: * P < 0.05, statistically significant. To find independent risk factors for postoperative DVT, conditional logistic regression was used. In accordance with the EPV concept, the primary model was limited to seven predictors. Based on clinical relevance, statistical significance in univariate analysis, and exploratory interest, the following variables were included: age, time from injury to surgery, D-dimer level, excessive sweating, gender, fracture type, and hemoglobin level. We found that age, time from injury to surgery, and D-dimer level remained independently associated with postoperative DVT(Table 4). Excessive sweating (P = 0.307), Gender(P = 0.502), fracture type (P = 0.504), and hemoglobin level(P = 0.120) were not statistically significant. Table 4Conditional logistic regression analysis of risk factors for postoperative DVT.VariablesβSEWaldOR (95%CI) P Age0.1370.03416.4641.147(1.073–1.225)0.000Time from injury to surgery0.1200.0604.0321.128(1.003–1.268)0.045D-dimer0.0950.02613.3141.100(1.045–1.158)0.000 Conditional logistic regression analysis of risk factors for postoperative DVT. The ROC curve analysis (Figs. 2, 3 and 4) shows the optimal cut-off values for DVT risk factors, with an age of 63 years, a time from injury to surgery of 5 days, and a D-dimer level of 4.675 mg/L. The area under the age curve is 0.780 (P < 0.05), the area under the curve from injury to surgery time is 0.591 (P < 0.05), and the area under the curve for D-dimer is 0.627 (P < 0.05). These modest values suggest that these factors alone are insufficient for accurate DVT prediction but may contribute to risk stratification when combined with clinical assessment. ROC curves for age. ROC curves for time from injury to surgery. ROC curves for D-dimer levels. Discussion DVT is a common complication of orthopedic patients, and the Virchow triad reveals three primary factors contributing to its development: venous stasis, hypercoagulability, and endothelial injury15. Patients with hip fractures are at high risk of DVT due to vascular endothelial damage and heightened blood coagulation caused by trauma, surgical intervention, and other factors. In addition, prolonged bed rest impedes venous return. Notwithstanding conventional anticoagulant medication, patients with hip fractures still have an 11.1% to 32.8% chance of developing DVT16, which seriously impacts their prognosis. Therefore, identifying risk factors for DVT formation and implementing focused preventive interventions has significant importance. Currently, researchers worldwide have undertaken several investigations into the risk factors for postoperative DVT in hip fractures. However, most of them are retrospective studies, which inevitably have some confounding variables that may compromise the study’s accuracy7,12 This study employed a prospective nested case-control design to analyze the risk factors for postoperative DVT in a cohort of patients undergoing hip fracture surgery using a matched baseline feature system. This study significantly reduced recollection bias by prospectively gathering exposure factors compared with retrospective studies. We found that age, time from injury to surgery, and D-dimer are independent risk factors for postoperative DVT in hip fracture patients. Our research indicates that the risk of postoperative DVT in hip fracture patients increases with age, with an 14.7% (95%CI: 1.073–1.225)rise in DVT risk for each additional year of age. The cut-off value for age is 63 years, which means patients over this age have a markedly elevated risk of DVT. This is consistent with previous research findings17. For example, Shi et al. analyzed 424 elder hip fracture patients and 81 younger hip fracture patients, concluding that age over 60 years is an independent risk factor for DVT18; Chang et al. reported that patients over 60 years old with lower limb fractures have the highest incidence of DVT19. Analyzing the reasons, elderly patients often present with various comorbidities, extended preoperative waiting time, and limited early movement, which exacerbate venous stasis. At the same time, the elderly are often accompanied by vascular sclerosis and poor venous valve function, which leads to a high incidence rate of DVT. Therefore, we should concentrate on elderly hip fracture patients, especially those exceeding 63 years of age, and conduct individualized prevention screening to reduce the occurrence of DVT. Surgical intervention is a prevalent approach for treating hip fractures, and current studies generally suggest that early operation can reduce the incidence of complications such as DVT20. This study further focused on the population of hip fracture patients without DVT before surgery, and the results also showed that the occurrence of postoperative DVT was significantly correlated with preoperative waiting time. We found that the risk of DVT increased by 12.8% (95%CI: 1.003–1.268)for each day’s delay of surgery, and the risk increased significantly when surgery was delayed for more than 5 days. The postponement of surgery could prolong the patient’s preoperative braking time. This braking state directly induces blood stasis and impedes venous return in the lower extremities, heightening the risk of the Virchow triad. Clinically, preoperative waiting time is often influenced by various factors, including limited medical resources, inadequate personnel allocation, and the need to optimize preoperative complications to an acceptable surgical condition. All these factors contribute to prolonged preoperative waiting time. Therefore, it is essential to optimize the diagnosis and treatment process, provide a green channel for hip fracture patients, and decrease preoperative waiting times to the greatest extent possible; At the same time, remain watchful and meticulously observe the incidence of DVT. As a degradation product of fibrin, elevated D-dimer levels indicate a hypercoagulable state and an increased risk of DVT. Currently, its predictive value in DVT has been widely confirmed21,22. However, D-dimer shows high sensitivity but low specificity, and it can be raised due to trauma, infection, inflammation, tumors, and other factors. Studies have shown that adjusting the critical value of D-dimer according to age has better predictive performance for DVT23. There is also research confirming that establishing the critical value of D-dimer at 3 mg/L can have better sensitivity and specificity for patients under the age of 50 with lower limb fractures24. Diverse patient categories possess varying optimum threshold values. This study further suggests that elevated D-dimer levels are associated with postoperative DVT in hip fracture patients, with a significant increase in DVT risk when D-dimer levels exceed 4.675 mg/L. In this study, excessive sweating after the use of antipyretic drugs was investigated as a possible risk factor for postoperative DVT. Although this clinical sign was strongly linked to DVT in univariate analysis, the association did not retain statistical significance in the conditional logistic regression model (OR = 1.878, 95% CI: 0.561–6.288, P = 0.307). The reason may be attributed to the limited sample size, the indicator’s subjectivity, and the lack of objective dehydration indicators. Nevertheless, we believe this variable retains heuristic value for clinical practice. Antipyretic medications (such as non-steroidal anti-inflammatory drugs) can target the thermoregulatory center, facilitating perspiration to enhance heat dissipation, potentially resulting in considerable fluid loss. Currently, there is limited literature directly exploring the association between sweating induced by these drugs and the risk of DVT, but multiple studies support the correlation between dehydration and DVT risk25,26. Excessive sweating causes fluid loss, reduces effective circulating blood volume, increases blood viscosity, and induces hypercoagulability, resulting in the development of DVT. Unlike uncontrollable factors like age and D-dimer levels, excessive sweating following the administration of antipyretic drugs is an important target that can be intervened clinically. Clothing saturated with sweat is a distinct and observable clinical sign. Once detected, prompt intervention may be implemented, including timely fluid replacement and physical actions to avert the onset of DVT. Simultaneously, medication indications should be strictly regulated to avoid unnecessary medication and mitigate the risk of DVT caused by fluid loss at the source. Future studies could employ objective indicators of dehydration and incorporate a substantial sample size. The advantage of this study lies in its design. Through prospective study, we intend to incorporate as many potential risk factors as possible, including objective elements such as patient basic information, treatment modalities, and laboratory indicators, as well as subjective factors such as lower limb movement and sweating after medication use. Although there are some important findings, our research still has some limitations. Firstly, this study is a single-center investigation, and patients may be influenced by geography and medical institutions, which limits the generalizability of the research results; Secondly, the research results are limited to the hospitalization period and do not include long-term follow -up; Thirdly, the indicator of excessive sweating caused by antipyretic drugs is imprecise. In the future, a standardized and replicable tool for assessing sweat loss can be developed to enhance the precision of this variable. Furthermore, matching was not performed for age and sex, which may introduce residual confounding despite multivariable adjustment. Besides, future research incorporating detailed fracture classification and surgical variables such as implant type27 are needed to develop comprehensive risk stratification models for postoperative DVT. In conclusion, our research indicates that age, preoperative waiting time, D-dimer levels are associated with increased risk of early postoperative in-hospital DVT in hip fracture patients. It is essential to focus on this group of persons and implement preventive actions to diminish the incidence of DVT. At present, the link between excessive sweating and DVT remains merely hypothetical and requires validation through objective moisture metrics in future studies. Acknowledgements We would like to thank all the team members for cooperation. Author contributions Zheng Chen: Writing original draft ; Xiaoyan Wang , Jiao Zhang , Jing Zhang : Data curation, Investigation ; Hui Xue : Formal analysis, Data curation ; Wei Chen : Writing–review and editing, Supervision ; Luqin Di : Writing original draft , Project administration. Funding This study was supported by the Health Commission of Hebei Province (Grant No. 20230075 and NO.20250624). Data availability Due to institutional and ethical restrictions, the data could not be shared, but could request from correspondence author upon reasonable request. Declarations Competing interests The authors declare no competing interests. Ethics approval and consent to participate This study was approved by the Ethics Committee of Hebei Medical University Third Hospital (approval number: W2023-013-1). All participants included in the study provided written informed consent. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Due to institutional and ethical restrictions, the data could not be shared, but could request from correspondence author upon reasonable request.