A nomogram model for predicting lower extremity deep vein thrombosis after gynecologic laparoscopic surgery: a retrospective cohort study

Introduction

Deep vein thrombosis (DVT) is a common complication that typically occurs within one week after surgery. If venous thrombosis dislodges, it can cause acute pulmonary embolism, which poses a serious threat to the patient’s life (Gutzeit et al., 2020; Hui et al., 2020; Kaya et al., 2021). Recently, with the advancements in surgical technology and equipment, laparoscopic surgery has gained popularity in gynecology due to its advantages such as lesser physical damage, reduced stress response, lower blood loss, and faster recovery (Vedantham, 2020). However, the incidence of lower extremity deep vein thrombosis (LEDVT) after gynecological laparoscopic surgery is 4.0%, which is not lower than that of open surgery due to factors like surgical posture and pneumoperitoneum (Chan & Weitz, 2019; Chong, Bui & Menhaji, 2020; Qu et al., 2015). Gynecological disease is a high risk factor for developing venous thromboembolism (Abu Saadeh et al., 2013). The occurrence of LEDVT can negatively affect the limb motor function of patients and decrease their quality of life, making it crucial to accurately identify high-risk patients for LEDVT (Han et al., 2021; Liu et al., 2006; Moragon-Ledesma et al., 2020).

In current practice, the diagnosis and prediction of LEDVT primarily rely on the Caprini score and Padua score, alongside laboratory examination and ultrasonic imaging examination (Willan, Katz & Keeling, 2019). These scoring systems were initially published in the 1990s and early 21st century and were found to be valuable in predicting the onset of LEDVT in open surgery. However, their predictive efficacy in laparoscopic surgery remains uncertain (Lu et al., 2021; Yang et al., 2019b).

Many previous studies have reported that their model has considerable predictive value for LEDVT (Hu et al., 2022; Tian & Li, 2021; Wu & Cheng, 2020). However, they did not take into account the relevant preoperative biochemical indicators. Some studies have revealed that D-dimer is an independent risk factor for LEDVT (Yago et al., 2020; Zhao, Tian & Zhang, 2021). However, the influence of other biochemical indicators on LEDVT remains unclear. Gynecological patients undergo blood biochemical tests before laparoscopic surgery, which makes it convenient to investigate the relationships between biochemical indicators and LEDVT.

The aim of this study was to identify independent risk factors and develop a predictive model for the occurrence of LEDVT in patients who underwent gynecological laparoscopic surgery.

Methods

Results

A nomogram model for predicting postoperative LEDVT

A nomogram predictive model was established for postoperative LEDVT in patients who underwent gynecological laparoscopic surgery according to independent risk factors (Fig. 1). Additionally, a receiver operating characteristic (ROC) curve of the predictive model was established and is shown in Fig. 2. The area under the curve (AUC) was 0.927 (95% CI [0.893–0.961]; P < 0.05), and the value of the Youden index was 0.720, with corresponding sensitivity of 96.1% and specificity of 75.9%. The calibration curve of the nomogram predictive model demonstrated that the C-index was 92.4% (Fig. 3).

Discussion

This research has shown that patients who have undergone gynecological laparoscopic surgery are at risk of LEDVT due to several factors. These risk factors include older age, longer operation duration, shorter activated partial thromboplastin time (APTT), higher D-dimer, higher HE4, and a history of hypertension. To prevent the occurrence of LEDVT in clinics, a nomogram model was established with a sensitivity of 96.1% and a specificity of 75.9%.

Laparoscopic surgery is a commonly employed minimally invasive technique within gynecology. Nonetheless, the incidence of LEDVT remains high due to the utilization of pneumoperitoneum and the patient’s position during surgery (Hu et al., 2022; Wu & Cheng, 2020). A previous study displayed the incidence of LEDVT subsequent to gynecologic laparoscopic surgery to be 7.6–11.55% (Tian & Li, 2021). Consistently, our current study discovered that the incidence of LEDVT following gynecologic laparoscopic surgery was 7.89%. A previous study found various independent risk factors for LEDVT in perinatal women, such as older age, mode of delivery, hypertension, diabetes, history of thrombosis, short activated partial thromboplastin time (APTT), high D-dimer, and fasting blood glucose (Zhao et al., 2018). Additionally, another previous study determined several distinct independent risk factors for LEDVT in patients, such as older age, extended bed rest, high blood viscosity, high D-dimer, and intraoperative blood loss (Evans et al., 2022). These two studies; however, did not examine the value of these risk factors specifically for patients undergoing gynecologic laparoscopic surgery. A previous study established a 16-point evaluation scale for venous thrombosis in gynecological surgery patients. This evaluation scale includes secondary indicators such as age, body mass index (BMI), immobilization preoperatively, specific factors for women, current high-risk disease, past and family medical history, treatment and medication history, surgery and anesthesia duration and mode, special conditions during surgery, posture during surgery, D-dimer, fibrinogen (FIB), APTT, and platelet count (Yang et al., 2019a). However, this study failed to disclose the risk factors specifically for laparoscopic surgery patients and could not predict the risk of LEDVT.

In the present study, we determined that prolonged duration of surgery is a significant risk factor for LEDVT independent of other factors. This prolonged surgical time can result in extended periods of laying down for patients, leading to the relaxation of lower limb muscles along with blood stasis in veins, thereby increasing the chance of LEDVT. Researchers have identified an elevated D-dimer level as a reliable indication of an abnormal blood coagulation system. Furthermore, previous studies have indicated that increased D-dimer levels hold immense importance in identifying LEDVT among peripheral diseases (Komatsu et al., 2020; Li et al., 2019a).

D-dimer, a primary index of coagulation function, holds great significance in the diagnosis, curative effect, and prognosis of thrombotic diseases. The elevation of D-dimer levels indicates active thrombosis or fibrinolytic activity in the blood vessels of patients. This study revealed that an increase in D-dimer is another independent risk factor for LEDVT, which is consistent with several previous studies (Jiang et al., 2021; Li et al., 2019b). However, a previous study showed that D-dimer is a risk factor for venous thromboembolism but not an independent risk factor due to the inclusion of fibrinogen degradation products, which interact with D-dimer (Shen et al., 2020). APTT, a vital indicator for screening the endogenous coagulation system, can reflect the status of coagulation factors in the body. In a hypercoagulable state or thrombosis, a decrease in APTT value indicates that the coagulation and fibrinolysis systems are unbalanced, leading to a hypercoagulable state and a higher risk of LEDVT (Zhang et al., 2019). Hypertension is another risk factor for LEDVT. Most patients with hypertension exhibit dysfunction of the renin-angiotensin-aldosterone system, leading to an increase in water and sodium retention, interstitial fluid, and decreased plasma content in blood vessels, which raises blood viscosity (Barber & Clarke-Pearson, 2016). In hypertension, vascular endothelial cell function is disordered, producing more oxygen-free substances, inactivating more vasodilators, causing vascular inflammation, activating the blood coagulation system, and promoting the onset of LEDVT (Laws et al., 2018). HE4, a member of the whey acidic protein domain-putative extracellular protease inhibitor protein family, serves as a biomarker for malignancy risk evaluation in various neoplastic diseases, especially ovarian cancer (Chhikara et al., 2012; Scaletta et al., 2017). HE4 produced by epithelial or fibroblast-derived cells can exacerbate tissue fibrosis and result in tissue damage (LeBleu et al., 2013; Zhang et al., 2020). An activity could represent an important factor in promoting thrombosis.

A nomogram model was developed in this research to predict the risk of LEDVT in gynecological patients. The model demonstrated a significant predictive capability, with a sensitivity of 96.1% and a specificity of 75.9%. It has been observed that the predictive performance of the model is higher than that of previous studies.

This study had several limitations. Firstly, due to the retrospective nature of this study, selection bias was inherent. Secondly, the model’s external validation was not performed despite the high predictive value for LEDVT, which restricted its application. Thirdly, the application of the model is restricted to patients undergoing gynecological laparoscopic surgery due to a limited number of patients.

Conclusion

The present study, for the first time to our knowledge, identified the independent risk factors for LEDVT in gynecological laparoscopic surgery patients as old age, prolonged surgical duration, shortened APTT, elevated D-dimer levels, a history of hypertension, and high HE4. Our nomogram model utilizing these factors accurately predicts the likelihood of LEDVT and may be applied to prevent its occurrence.

Supplemental Information