Preoperative incidence and risk factors of deep vein thrombosis in patients with an isolated patella fracture

The purpose of this study was to investigate the incidence, location, and related factors of preoperative deep venous thrombosis (DVT) in patients with isolated patellar fractures. Methods Patients with an isolated patellar fracture, admitted between January 2013 and December 2019 at our institution, were retrospectively analyzed. Upon admission, patients underwent routine Doppler ultrasound scanning (DUS) of the bilateral lower extremities to detect DVT; those with DVT were assigned to the case group and those without DVT to the control group. Data on demographics, comorbidities, and laboratory test results upon admission were extracted. Differences between the two groups were evaluated using univariate analyses, and independent risk factors associated with DVT were identied by logistic regression analysis.


Background
DVT is a common complication and cause of death in hospitalized patients [1]. According to reports, the incidence of DVT in hospitalized patients is 30.4% overall, and the related mortality rate is approximately 1.3% [2]. DVT is also common in patients with lower limb fractures. Studies have reported that the incidence of DVT in the perioperative period is as high as 10.1% to 29.1% [3,4], and that of pulmonary embolism was approximately 2.73% [3]. DVT has a serious impact on postoperative rehabilitation and quality of life in patients with fractures, which necessitates early diagnosis and appropriate treatment [5]. Therefore, in the management of patients with fractures, it is vital to distinguish preoperative DVT and postoperative DVT for timely diagnosis and treatment.
The literature reported that the incidence of lower extremity DVT in patients with fracture upon admission is 20.1-29.8% [6,7]. Patients with lower extremity fractures often require surgical treatment to restore lower extremity function. Preoperative DVT seriously affects the surgery, prolong the treatment period, and hinders the healing process. Current research on the incidence and location of preoperative DVT in fractures mainly focuses on the lower extremity, such as pelvic fractures, femoral fractures, and tibial fractures [3,8,9] or the entire lower extremity fracture [10][11][12]. Patella is an important structure in the process of knee extension, with an incidence rate in the population being 0.13‰ to 0.61‰ [13,14], accounting for 0.5% to 1.5% of the total incidence of adult fractures [15,16]. It has been reported in the literature that the incidence of perioperative DVT for patellar fractures is 0.3-9% [10][11][12]. However, these studies did not distinguish between the preoperative and postoperative incidence of DVT. In addition, the reports on risk factors for DVT have been inconsistent across studies. For example, Zhang et al. found that preoperative time and plasma D-dimer are independent risk factors for perioperative DVT [10]. Li et al. found that age > 50 years, arthroplasty, and surgery time longer than 3 hours are independent risk factors [12]. However, in these studies, the risk factors for DVT in patients with fracture were only considered for the entire lower extremity fracture.
To the best of our knowledge, there are no studies about the epidemiologic characteristics and risk factors associated with preoperative DVT in patients with isolated patellar fractures. The purpose of this study was to retrospectively analyze the data of patients with an isolated patellar fracture in our institution and to evaluate the characteristics of preoperative DVT in these patients.

Patients
This study retrospectively collected information on patients with patellar fracture treated in our hospital from January 2013 to December 2019. This study was approved by the Ethics Committee of the Third Hospital of Hebei Medical University. The inclusion criteria were age > 18 years, isolated patellar fracture, and surgical treatment. The exclusion criteria were bilateral patellar fracture, multiple fractures, open fracture, old fracture and pathological fracture, use of blood circulation pumps, autoimmune diseases, anticoagulant use within 3 months of admission, and incomplete medical records.

Diagnosis criteria of thrombosis
After admission and before surgery, the patients were subjected to routine DUS of the lower extremity. The According to the thrombotic test criteria, the ultrasound physician examined and reported the ndings of the femoral vein trunk and the femoral deep, super cial, popliteal, tibial, and bular veins of both lower extremities. Patients found to be present with DVT were consulted with an experienced vascular surgeon for appropriate treatment based on their medical conditions. The clinical signi cance of the intermuscular vein, small saphenous vein, and great saphenous vein is relatively small; therefore, they were excluded from this study [17].
Statistical analysis SPSS 25.0 software (IBM, Armonk, New York, USA) was used for statistical analysis. The measurement data were rst explored using the Shapiro-Wilk test for their distribution status (normal or non-normal). Normal distribution data were expressed as mean ±standard deviation (Sd), and an independent sample t-test was used to compare the differences between groups. The Mann-Whitney U test was used for nonnormally distributed data. Categorical variables were evaluated using the chi-square or Fisher's exact tests. P values < 0.10 in the univariate analyses were further analyzed by multivariate logistic regression. P values < 0.05 were considered statistically signi cant for all analyses.

Results
A total of 1,049 patients with patellar fractures were admitted during the study period. Among them, 73 patients were excluded because they were younger than 18 years; 39 due to bilateral patellar fracture and non-surgical treatment; 41 due to multiple fractures, open fracture, old fracture, and pathological fractures; 16 due to use of blood circulation pumps, 25 due to autoimmune diseases, and anticoagulant drugs were used in the last 3 months and 28 due to incomplete medical records (Fig 1).
Among the DVT patients, 22 had left-sided fractures and 26 had right-sided fractures. Of the 22 patients with left-sided fracture, 21 had thrombus in the injured extremity and 1 had thrombus in the bilateral extremity. Among the 26 patients with right-sided fractures, the thrombus was located in the injured extremity, non-injured extremity, and bilateral extremity in 20, 4, and 2 patients, respectively ( Table 2). The distribution of thrombus in blood vessels was as follows: 1 in the femoral super cial vein, 7 in the popliteal veins, 20 in the peroneal veins, and 20 in the posterior tibial veins in patients with left-sided fractures; and 1 in the femoral super cial vein, 2 in the popliteal veins, 15 in the peroneal veins, and 9 in the posterior tibial veins in patients with right-sided fractures ( Table 3). None of the patients developed preoperative pulmonary embolism.
Comparison of the variables between the DVT and non-DVT groups showed statistically signi cant differences in sex, age, residential area, time from injury to DUS, ALB lower limit, ALT upper limit, CREA upper limit, NEUT upper limit, and D-dimer upper limit (P < 0.05, Table 1).

Discussion
DVT is a common complication in patients with fracture, potentially affecting their prognosis. In addition, there are many uncertainties about the risk factors associated with thrombus. It is important to understand the signi cance of preoperative DVT in patients with isolated patellar fractures. In this study, we found that the preoperative incidence of DVT after patellar fracture was 5.8% (48/827), and 85.4% (41/48) occurred in the injured extremity. We also found that the associated risk factors were age, residence, time from injury to DUS, and plasma D-dimer level.
DVT is a common problem in patients with fractures and has been widely discussed in clinical practice [18]. We found a 5.8% (48/827) preoperative incidence of DVT after an isolated patellar fracture. However, Wang et al. [9] reported a 15.3% incidence of 59 patellar fractures. Li et al. [12] also reported an overall 8.2% rate in lower extremity fractures, with 9% of DVT in 177 knee fractures. The use of DUS, compared to the use of computed tomography (CT) or magnetic resonance imaging (MRI) venography for the detection of DVT may be partly explained by the relatively large gap in incidence rates. We also found that DVT occurred not only in the injured extremity but also in the non-injured extremity, which is consistent with the ndings of Wang et al. [19]. This suggests the equal importance of screening for DVT in both the non-injured and injured extremities.
In this study, we found that for each additional increment of 1 year in age, the risk of DVT increased by 2%. Similarly, Li et al. [12] found that the risk of DVT in patients over 50 years of age increased by 43% in 829 patients with lower limb fractures. Auer et al. [20] also found that patients in the DVT group had a higher age than those in the non-DVT group. According to Virchow's principles, age-related decreases in physiological function, vascular elasticity, increased blood viscosity, and poorer venous valve function account for the propensity to develop DVT [21,22]. Therefore, elderly patients with trauma remain the focused population with a higher risk of DVT and should be emphatically monitored or suspected.
Our results showed that the preoperative days were signi cantly longer in the DVT group than in the non-DVT group, with a 33% increased risk of preoperative DVT for delay time from injury to DUS (in each day).
Similarly, Zuo et al. [6] found that the days before admission to the hospital in the DVT group were longer than in the non-DVT group after intertrochanteric fractures, and the risk of DVT was increased by 37% every day delay to admission. The hypercoagulable state of blood after trauma is the pathophysiological basis of DVT. In particular, the blood coagulation dynamic value was the highest in the rst 24 hours after the trauma and remained hypercoagulable during the rst 4 days [23]. Similar conclusions were reached in the study conducted by Decker et al. [24].
It is known from current reports that D-dimer is a highly sensitive laboratory marker for DVT [25,26]. Our results showed that D-dimer levels above 0.5 ug/ml at admission were associated with a 47% increased risk of DVT, which was comparable to the ndings of Zhang et al. [10] study of lower extremity fractures. D-dimer is a brin degradation marker that represents secondary brinolytic activity, hypercoagulability, and brinolytic activity in the blood, which has clinical value in the diagnosis of thrombus events [27]. Yamasaki et al. [28] retrospectively collected 588 patients undergoing lumbar spine surgery and found that the risk of DVT increased by 9% with D-dimer greater than 19.2 ug/ml one week after surgery. However, the cut-off values used in various researchers are highly variable, which may be due to the heterogeneity of the subjects and the study designs. Therefore, the age-adjusted D-dimer levels should be more appropriately applied.
Our study has several limitations. First, since it was a retrospective study, and information might have been missing in the data collected. Second, we might have underreport DVTs, primarily because of the use of DUS due to its relatively lower sensitivity compared to CT or MRI angiography. Considering that angiography is an invasive examination, routine DUS use for DVT screening is acceptable and is generally used in most medical centers. Third, we conducted a cross-sectional rather than a cohort study with long-term follow-up. Therefore, the causative relationship of variables with DVT cannot be established. Instead, there is an association which should be cautiously treated during interpretation.

Conclusions
In summary, the incidence of preoperative DVT following an isolated patellar fracture was 5.8%. Age (each increase 1 year), residential area (urban or rural area), fracture detected by DUS, and elevated plasma D-dimer levels were independent risk factors for preoperative thrombosis. Despite a low prevalence of DVT after an isolated patellar fracture, this study underscores the importance of identifying those with a high risk of DVT, especially those with multiple identi able factors, as well as the early targeted use of thromboembolic agents, to reduce DVT occurrence.

Availability of data and materials
All the data will be available upon motivated request to the corresponding author of the present paper Ethics approval and consent to participate This study was approved by the ethics committee of the 3rd Hospital of Hebei Medical University.

Consent for publication
Written informed consent was obtained from each patient to authorize the publication of their data.

Con icts of Interest
All authors have read and contributed to the submitted manuscript and have no con ict of interest to declare.      Figure 1 Flow chart of patient inclusion and exclusion in this study.