Owing to the venous congestion, vascular injury and immobilization and other medical problems, patients with hip fractures had an increased risk of developing DVT[5]. Previous studies had investigated the prevalence and risk factors of preoperative DVT, but fewer studies assessed the prevalence of DVT on admission in elderly Chinese patients with intertrochanteric fractures[13–15]. In fact, many hip fracture patients might have already had DVT on admission. Therefore, early DVT identification is essential to limit late complications of DVT and prevent clot extension, acute PE, and recurrent thrombosis. Additionally, most of previous studies evaluating risk factors for DVT consisted of both femoral neck and trochanteric fractures. Unlike femoral neck fractures, trochanteric fractures tended to have a greater risk of preoperative hemoglobin (Hgb) drop, which is a risk factor for preoperative DVT[10, 11]. To our knowledge, this study provides the first evidence of risk factors for pre-admission DVT in elderly Chinese patients with intertrochanteric fractures.
In this study, the overall prevalence of pre-admission DVT was 20.81% in patients with intertrochanteric fractures, which was a relative high incidence of DVT compared with previous studies[5, 6]. A longer waiting time was one of the most important factors contributing to the high prevalence of preoperative DVT[6, 13]. The time from injury to surgery consisted of time waiting for admission and the period awaiting surgery for preoperative evaluation. The main reason for longer time from injury to admission was transfer from community hospitals to our trauma center and the poor comorbidity of elderly patients, which needed more time for preoperative preparation[16]. Ideally, surgery should be performed as early as possible for early mobilization and relieve of pain[17]. In present study, the mean time from injury to admission was significantly longer in patients who developed DVT compared with patients who didn’t develop DVT. Furthermore, an increasing linear association was found between the occurrence of DVT and the time from injury to admission on cumulative hazard plotting. One possible reason might be that prolonged immobilization could result in venous congestion. Another reason might be that the fracture could lead to vascular injury, which might activate the coagulation system. The third reason might that fracture was frequently coupled with dominant and hidden blood loss, especially hidden blood loss for intertrochanteric fractures. Therefore, earlier admission was necessary for intertrochanteric fractures.
In present study, an increased BMI was an independent risk factor for pre-admission DVT in intertrochanteric fractures. This result was consistent with one published study, which revealed that obesity was associated with DVT[18]. An increased BMI was associated with venous thromboembolism, with multiple mechanisms and pathways contributing to this effect. The possible mechanism might be that an increased BMI could not only alter the expression of proteins of the coagulation and fibrinolytic cascade, but also change the platelet biology and function, which could promote the increased thrombotic risk[19]. In addition, some studies reported that female patients had a higher risk of DVT than male patients, while others had different opinions[5, 15, 20]. Our study revealed that female sex was a risk factor for DVT in elderly intertrochanteric fractures. Moreover, this study demonstrated that the incidence of DVT on admission in patients with high-energy injury was significantly higher than in patients with low-energy injury (34.72% versus 19.41%, P = 0.002). The adjusted multivariate logistic regression analysis showed that high energy injury was an independent risk factor for DVT on admission in patients with intertrochanteric fractures, which was supported by a previous study[21].
Various studies had demonstrated that the occurrence of DVT in patients with hip fractures was strongly associated with medical comorbidities, especially in elderly patients[14]. However, which specific comorbidity related to the occurrence of DVT remained controversial in previous studies[5, 22]. Shin et al[5] investigated the prevalence and risk factors of preoperative VTE in 208 patients with hip fractures and identified that pulmonary disease and VTE history were independent predictive factors for preoperative VTE. Another study reported that coronary heart disease was independent risk factor of DVT in patients with hip fractures[22]. The possible reason might be that coronary heart disease was associated with hypercoagulability state[23]. In present study, we identified that atrial fibrillation was independent risk factor for DVT on admission in patients with intertrochanteric fractures. Furthermore, we found that COPD was independent risk factor for DVT on admission. The possible reason might be that COPD was associated with an increased atherosclerotic disease burden derived from a chronic inflammation[24].
The CCI was first reported in 1987 to estimate the probability of death within 1 year[8]. The CCI also correlated with the probability of death for patients with breast cancer, and adverse events after spine surgery[25]. However, rare studies evaluated the association between CCI and the occurrence of DVT in hip fracture patients. In present study, the mean value of age-adjusted CCI in patients with DVT was significantly higher than in patients without DVT (4.63 versus 4.17, P < 0.001). However, the multivariate logistic regression analysis showed that the age-adjusted CCI was no longer significant predictor of DVT on admission after adjusting for potential confounders. The possible reason might be that the age-adjusted CCI was highly associated with other risk factors, especially age and medical comorbidities. This result was supported by one previous study, which reported that CCI was not related with the occurrence of preoperative DVT[14]. Till now, there was not scientific evaluation method to assess the relationship between preoperative comorbidities and the occurrence of DVT in patients with hip fractures. A larger, prospective and multi-center study was necessary to develop a scientific evaluation system.
In this study, we demonstrated that A3 type intertrochanteric fracture was independent risk factor for the occurrence of DVT on admission in patients with intertrochanteric fractures. Shin et al[5] demonstrated that subtrochanteric fracture was independent predictive factor for preoperative VTE in patients with hip fracture whose surgery was delayed by > 24 hours. For anatomic reasons, extracapsular fractures tended to have a greater blood loss than intracapsular fractures. It was known that the anemia and low hemoglobin concentrations were significantly associated with frailty[26]. Moreover, frailty had been demonstrated to predict adverse outcomes in older surgical patients[27]. Furthermore, one study demonstrated that preoperative anemia is a risk factor for preoperative DVT in hip fracture patients[11], and preoperative anemia was very common in patients with intertrochanteric fractures due to the dominant and hidden blood loss.
The major strength of present study was that this was the first study to evaluate risk factors for DVT on admission in elderly Chinese patients with intertrochanteric fractures. However, several limitations existed in this study. First, this study was a retrospective study. Second, all patients in this study came from one trauma center. Therefore, a multi-center large sample study would be required to validate our findings. Third, laboratory tests were not obtained in this study. Previous studies showed that increased D-dimer and hemoglobin drop were independent risk factors for the occurrence of DVT[11, 15]. Future study should enrolled laboratory tests, especially the hemoglobin in patients with intertrochanteric fractures. Fourth, patients were diagnosed with DVT by color Doppler ultrasonography, which might have a lower accuracy than venography.