Incidence, Risk Factors, Risk Assessment Model and Compliance of Patients on Anticoagulants for Asymptomatic Venous Thromboembolism in Nononcological Urological Inpatients Kaixuan Li1, Quan Zhu1, Haozhen Li1, Ziqiang Wu1, Feng Han1, Zhengyan Tang1,2** Zhao Wang1* Purpose: To study the incidence, risk factors for developing asymptomatic venous thromboembolism and the compliance of patients on anticoagulants for asymptomatic venous thromboembolism (VTE) in nononcological urological medium-high risk inpatients, and build a risk assessment model (RAM) for early screening for asymp- tomatic VTE. Materials and Methods: We conducted a retrospective analysis of 573 inpatients admitted to a nononcological urological ward of a tertiary hospital in China from January 1, 2017, to June 30, 2019. Data were collected using the electronic medical record system, and patients underwent a follow-up by phone 6 months after discharge. Results: Among the 573 medium-high risk inpatients, 73 (15.4%) were diagnosed with VTE, including 20 (4.2%) symptomatic and 53 (11.2%) asymptomatic. Prior history of VTE, a history of anticoagulants or antiplatelet agents before admission, and D-dimer ≥ 1 were the potential risk factors identified for asymptomatic VTE. Patients with poor awareness of VTE and its dangers, and patients who lived more than 1 hour away from the hospital had a high probability of poor compliance with anticoagulation therapy after discharge. Using D-dimer (1.785 μg/ml), we built a RAM for the early diagnosis of asymptomatic VTE. Conclusion: We found that patients with urinary nontumor VTE had low compliance with anticoagulation ther- apy after discharge. The key factors for determining asymptomatic VTE in nononcological urological inpatients included prior history of VTE, a history of taking anticoagulants or anti-platelet agents before admission, and D-dimer ≥ 1. Furthermore, we found that the threshold of D-dimer should be elevated to 1.785 μg/ml to predict asymptomatic VTE. Keywords: asymptomatic; nononcological; compliance; risk assessment model; urology; venous thromboembo- lism 1Department of Urology, Xiangya Hospital Central South University, Changsha 410008, China. 2Provincial Laboratory for Diagnosis and Treatment of Genitourinary System Disease, Changsha 410000, China. Kaixuan Li and Quan Zhu contributed equally to the article. *Correspondence: Department of Urology, Xiangya Hospital Central South University; Changsha, 410008, China. Tel: +86-15116358241, E-mail: xywangz07@163.com. **Department of Urology, Xiangya Hospital Central South University; Changsha, 410008, China. Tel: +86-13507318268, E-mail: xytzyan@163.com. Received July 2021 & Accepted November 2021 UNCLASSIFIED INTRODUCTION Venous thromboembolism (VTE), which denotes both pulmonary embolism (PE) and deep vein thrombosis (DVT), is a common cause of morbidity and mortality after urological surgeries. VTE was a particularly common problem in the past, with an es- timated 22% incidence rate due to no prophylaxis for patients who underwent pelvic surgeries(1). Although the occurrence of VTE has decreased in recent years, including for both uro-oncologic and nononcological surgeries(2,3), VTE is still a serious perioperative ad- verse event. Furthermore, it places a heavy burden on the healthcare systems and brings about higher mortali- ty with potentially fatal PE. VTE can be either symptomatic or asymptomatic, with the former often being much easier for clinicians to iden- tify. Asymptomatic VTE can also cause fatal PE and it is associated with an increased risk in all-cause mor- tality(4). The European Association of Urology (EAU) and the Canadian Urological Association (CUA) rec- ommend providing perioperative thromboprophylaxis for both symptomatic and asymptomatic VTE(5,6). How- ever, the evidence grades in these guidelines are rela- tively weak, for example, in EAU guidelines, the rec- ommended levels of prophylaxis for many urological nontumor surgeries are weak, including transurethral resection of the prostate (TURP), prostatectomy, lapa- roscopic without pelvic lymph node dissection (PLND), nephrectomy, and reconstructive pelvic surgery et al(6). And whether thromboprophylaxis for asymptomatic VTE can balance thrombosis prevention and bleeding remains uncertain(7). Although the Caprini risk assess- ment tool is recommended to evaluate the occurrence of postoperative VTE(8), a high proportion of patients after urological surgeries are still classified into the high-risk group. This suggests that the validity of this evaluation model is open to question. One way we felt this issue could be overcome was by building a risk assessment model (RAM) for VTE. In recent years, urologists have paid increasing atten- tion to the occurrence and prevention of perioperative Urology Journal/Vol 20 No. 1/ January-February 2023/ pp. 56-65. [DOI:10.22037/uj.v18i.6893] VTE. However, noncancer urological patients have still not received much attention from urologists. The incidence of VTE after urological nononcological sur- geries is considered to be relatively low, but the results of our prior study and other research have contradicted this idea(9,10). The main urinary system nononcological diseases are urinary stones, benign prostate hyperplasia, ureteral stricture, varicocele, renal cyst, urethral stric- ture, stress urinary incontinence, and others. Although physical and drug therapies to prevent thrombosis are conducted across the globe, VTE-related mortality after urological nononcological surgery remains a concern. Moreover, few studies have focused on asymptomatic VTE in urological nononcological hospitalized patients, and thus this issue remains unresolved for nononcolog- ical urologists. At present, most research on VTE compliance focus- es on the prevention of postoperative thrombosis in non-urologic patients. For example, a prior compliance study compared the benefits of using either preventive anticoagulants with those from using mechanical meth- ods to prevent VTE after gynecological tumor surgery and total hip and knee replacement surgery(11-13). How- ever, there are few relevant studies on the compliance of VTE anticoagulation therapy after discharge, espe- cially in patients with non-tumor VTE in the urology department. When evaluating the effect of anticoagu- lant therapy, we feel it is important to understand the patient's compliance with anticoagulant therapy. For this reason, the main aim of our preliminary study was to retrospectively investigate the incidence rates, risk factors, and compliance of nononcological inpa- tients in urology. Using these data, we then established a RAM for the screening of asymptomatic VTE by us- ing the appropriate threshold of D-dimer. MATERIALS AND METHODS Study Population This was a retrospective, single-center, cohort study ap- proved by the Ethics Committee of Xiangya Hospital (no. 2019030078). Consecutive inpatients who were admitted to the nononcological urological unit and who underwent diagnostic imaging (including computed to- mography, pulmonary angiography and ultrasound) be- Characteristics Asymptomatic VTE Symptomatic VTE P-value Age (years) 65.53 ± 10.08 59.00 ± 12.61 .024* ≤ 65 28 (52.8%) 12 (60.0%) ≥ 66 25 (47.2%) 8 (40.0%) Gender .977 Male 32 (60.4%) 12 (60.0%) Female 21 (39.6%) 8 (40.0%) BMI(kg/m2) 22.64 ± 2.79 23.25 ± 3.18 .462 ≤ 18.5 3 (5.7%) 1 (5.0%) 18.5-23.9 33 (62.3%) 8 (40.0%) ≥ 23.9 17 (32.1%) 11 (55.0%) D-Dimer (μg/mL, max) 2.34 (1.23-3.53)# 6.68 ± 9.81 .829 Caprini score(max) 5.87 ± 2.21 6.15 ± 4.74 .800 ASA score 3.00 (2.00-3.00)# 2.77 ± 0.93 .886 Operation time(min) 87.88 ± 49.49 81.38 ± 59.52 .721 Intraoperative blood loss (ml) 20.00 (5.00-100.00)# 44.46 ± 81.96 .266 Table1. Comparison between inpatients with asymptomatic and symptomatic VTE Abbreviations: VTE, venous thromboembolism; BMI, body mass index; ASA, American society of Aneshesiologists physical status classification system #Median and interquartile range *Statistically significant (α = 0.05) Characteristics No. of Sample Size Asymptomatic VTE Univariate Analysis Unadjustd OR (95%CI) N (%) Incidence (%) Total 553 (100.0%) 53 (100%) 9.6% Emergency Surgery No 520 (94.07%) 49 (92.4%) 9.4% P = .545 # 1.0 Yes 33 (5.97%) 4 (7.6%) 12.1% 1.3 (0.4-3.9) Gender Male 350 (65.67%) 32 (60.38%) 9.1% P = .643 1.0 Female 203 (34.33%) 21 (39.62%) 10.3% 1.1 (0.6-2.0) Age (years) ≤ 65 375 (67.81%) 28 (52.83%) 7.5% P = .014 1.0 ≥ 66 178 (32.19% ) 25 (47.17%) 14.0% 2.0 (1.1-3.6) * BMI (kg/m2) 18.5-23.9 304 (54.97%) 37 (69.81%) 10.5% P = .182# 1.0 ≥24 38 (6.87%) 4 (7.55%) 12.2% 0.8 (0.3-2.5) 24-27.9 133 (24.05%) 9 (16.98%) 6.8% 0.5 (0.2-1.1) ≥28 64 (11.57%) 3 (5.66%) 4.7% 0.4 (0.1-1.2) 35-39.9 14 (2.53%) 0 0% Abbreviations: VTE, venous thromboembolism; OR, odds ratios; BMI, body mass index. *Statistically significant (α = 0.05) # Fisher’s exact test Table 2. Baseline characteristics associated with asymptomatic VTE. Asymptomatic venous thromboembolism-Li et al. Vol 20 No 1 January-February 2023 57 tween January 1, 2017, and June 30, 2019, at Xiangya Hospital were enrolled in our study. Inclusion and exclusion criteria Patients were regarded as medium-high risk inpatients and advised to finish diagnostic imaging if they had one or more of the following situations: over 75 years old, had prior VTE, a body mass index over 35 kg/m2, a first degree relative (parent, full sibling, or child) with VTE, or a D-Dimer value over 0.5 μg/mL. For patients as- sessed as medium-high risk inpatients, we will perform D-Dimer examinations after admission and after sur- gery. Patients were excluded if they were younger than 18 years old, had postoperative pathological examina- tion results that showed malignancy, or had incomplete clinical medical records. In total, 573 eligible inpatients were selected for the study (Figure 1). Procedures Asymptomatic VTE was defined when a hospitalized VTE patient displayed no VTE related symptoms (such as swelling and painful in the lower limbs, pain in low- er limbs, decreased SpO 2 , dyspnea, chest pain, or an electrocardiogram (ECG) performance with unstable circulation)(14). Patients who were assessed with medi- um-high thrombosis risk inpatients underwent Doppler ultrasound examination of lower extremity blood ves- sels after admission and after surgery, and 25 postop- erative patients underwent pulmonary artery CTA for the occurrence of suspicious symptoms of pulmonary embolism like difficulty breathing, chest pain, etc. All the data were retrospectively collected from the Electronic Medical Record System (EMRS) and anes- thetic records in Xiangya Hospital through each inpa- tient’s unique ID number. In addition, the Caprini RAM was also collected and it was invented by Caprini et al. Caprini RAM is based on the risk of the risk factors to stratify the patients for VTE risk. For patients with a very low risk of VTE (Caprini score of 0), additional prevention is not recommended. For patients with low risk of VTE (Caprini score of 1-2), mechanical or drug prevention is recommended. Patients with intermediate risk of VTE (Caprini score of 3-4) to high risk (Caprini score of ≥ 5) are recommended for anticoagulant alone or combined with mechanical prevention(15). The occur- rence of VTE was detected by the imaging procedure reports, diagnosed by board-certified radiologists as well as the course records written by attending doctors. Preoperative bleeding was defined as hematuria, blood in the stool, and bleeding in other parts of the body. Regardless of the size (massive or small) and location (proximal or distal) of the thrombus, once the VTE oc- curred in the deep veins of the lower extremities or the pulmonary artery and its branches, we considered that a VTE event had occured. The occurrence of sepsis was defined according to the Sepsis - 3 criteria and the calculation formula for creatinine clearance (Cock- Characteristics No. of Sample Size Asymptomatic VTE Univariate Analysis Unadjustd OR (95% CI) N (%) Incidence (%) Hypertension No 368 (66.55%) 18 (33.96%) 9.5% P = .934 1.0 Yes 185 (33.45%) 18 (33.96%) 9.7% 1.0 (0.6-1.9) Diabetes No 493 (89.15%) 4 (7.55%) 9.9% P = .416 1.0 Yes 60 (10.85%) 4 (7.55%) 6.7% 0.6 (0.2-1.9) Prior VTE No 524 (95.30%) 13 (24.53%) 7.6% P < .001# 1.0 Yes 29 (4.70%) 13 (24.53%) 44.8% 9.8(4.4-21.9) * Family history of VTE No 522 (94.39%) 3 (5.66%) 9.6% P = 1.000# 1.0 Yes 31 (5.61%) 3 (5.66%) 9.7% 1.0 (0.3-3.4) History of varicose veins in the lower extremities No 534 (96.56%) 4 (7.55%) 9.2% P = .099# 1.0 Yes 19 (3.44%) 4 (7.55%) 21.1% 2.6 (0.8-8.3) History of surgery within a month No 520 (94.03%) 6 (11.32%) 9.0% P = .117 # 1.000 Yes 33 (5.97%) 6 (11.32%) 18.2% 2.2 (0.9-5.7) Anticoagulants or anti-platelet agents prescription before admission No 534 (96.56%) 6 (11.32%) 8.8% P = .006# 1.0 Yes 19 (3.44%) 6 (11.32%) 31.6% 4.8(1.7-13.2) * COPD No 544 (98.37%) 1 (1.89%) 9.6% P = .599 # 1.0 Yes 9 (1.63%) 1 (1.89%) 11.1% 1.2 (0.1-9.6) Coronary atherosclerotic heart disease No 517 (93.49%) 1 (1.89%) 10.1% P = .238# 1.0 Yes 36 (6.51%) 1 (1.89%) 2.8% 0.3 (0.0-1.9) Other cardiovascular diseases No 529 (95.66%) 3 (5.66%) 9.5% P = .494 # 1.0 Yes 24 (4.34%) 3 (5.66%) 12.5% 1.4 (0.4-4.8) Cancer history No 533 (896.38%) 2 (3.77%) 9.6% P = 1.000 # 1.0 Yes 20 (3.62%) 2 (3.77%) 10.0% 1.1 (0.2-4.7) Cerebral stroke history No 525 (94.94%) 4 (7.55%) 9.3% P = .331# 1.0 Yes 28 (5.06%) 1.6 (0.5-4.9) Table 3. Characteristics of medical history associated with asymptomatic VTE. Abbreviations: VTE, venous thromboembolism; COPD, chronic obstructive pulmonary disease; OR, odds ratios. *Statistically significant (α = 0.05) # Fisher’s exact test Asymptomatic venous thromboembolism-Li et al. Unclassified 58 croft-Gault equation) was conducted based on creati- nine levels(15,16). At 6 months after discharge, we checked on each pa- tients’ thrombosis progression through a phone fol- low-up. If we were unable to follow-up with a patient after 3 consecutive days of attempts, they were exclud- ed from our study. Statistical Analysis All statistical analyses were performed using SAS ver- sion 9.3 software (SAS Institute Inc., Cary, NC, USA). Categorical variables were described using frequency and percentage while means and standard deviations were applied to the continuous variables. A chi-square testing and independent t test were used to assess the risk factors for asymptomatic VTE to compare asymp- tomatic and symptomatic VTE. The minimum expected cell frequency was accessed, and when the data did not meet Pearson's χ 2 test conditions (E ≥ 5 and n ≥ 40), Fisher exact test is used. The normality and homogenei- ty of variance were assessed, and median and interquar- tile range of variables with skewed distribution were reported. When the P-values were under 0.1 in the uni- variate analyses, the related factors were chosen to be evaluated by multivariable logistic regression analysis. A receiver operating characteristic (ROC) curve was plotted by referring to the sensitivity vs. 1 – specificity of D-dimer level. The areas under the curve (AUCs), cutoff value sensitivity, specificity, negative predic- tive value (NPV), positive predictive value (PPV) and Youden index were used to assess the prediction model in this study. A P-value of < .05 was considered to be statistically significant. RESULTS Of the 592 patients selected in the initial sample, 19 (3.2%) patients were excluded based on the exclusion criteria, leaving a total of 573 patients for our study. Among these patients, 73 (15.4%) were diagnosed with VTE, including 20 (4.2%) symptomatic VTE patients and 53(11.2%) asymptomatic patients. The diagnoses of the patients include urinary stones (305, 53.2%), be- nign prostate hyperplasia (58, 10.1%), benign adrenal Characteristics No. of Sample Size Asymptomatic VTE Univariate Analysis Unadjustd OR (95% CI) N (%) Incidence (%) Preoperative bleeding No 521 (94.21%) 48 (90.57%) 9.2% P = .218# 1.0 Yes 52 (5.79%) 5 (9.43%) 15.6% 1.8 (0.7-5.0) Preoperative sepsis No 526 (95.12%) 48 (90.57%) 9.1% P = .167 # 1.0 Yes 27 (4.88%) 5 (9.43%) 18.5% 2.3 (0.8-6.2) Creatinine clearance levels (mL/min)a 0.6(0.5-0.8) <15 45 (8.14%) 8 (15.09%) 17.8% p = .001# 1.0 15-29 37 (6.69%) 5 (9.43%) 13.5% 0.7 (0.2-2.4) 30-59 179 (32.37%) 24 (45.28%) 13.4% 0.7 (0.3-1.7) 60-90 227 (41.05%) 16 (30.19%) 7.0% 0.4 (0.1-0.9) * D-Dimer (μg/mL,max) 2.1 (1.3-3.3) * < 0.5 135 (24.41%) 6 (11.32%) 4.4% p=.001# 1.0 0.5-1 104 (18.81% ) 4 (7.55%) 3.8% 0.9 (0.2-3.1) ≥1 314 (56.78%) 43 (81.13%) 13.7% 3.4 (1.4-8.2) * Caprini score, max 1.6 (0.9-2.7) ≤2 19 (3.44%) 2 (3.77%) 10.5% p = .125 # 1.0 3-4 191 (34.54%) 12 (22.64%) 6.3% 0.6 (0.1-2.8) ≥5 343 (62.02%) 39 (73.58%) 11.4% 1.1 (0.2-4.9) Re-admission surgery No 548 (99.10%) 52 (98.11%) 9.5% p =.397# 1.0 Yes 5 (0.90%) 1 (1.89%) 20.0% 2.4 (0.3-21.7) Complications No 536 (96.93%) 50 (94.34%) 9.3% p = .217# 1.0 Yes 17 (3.07%) 3 (5.66%) 17.6% 2.1 (0.6-7.5) Table 4. Other related characteristics associated with asymptomatic VTE. Abbreviations: VTE, venous thromboembolism; OR, odds ratios. a.Creatinine clearance levels were calculated with the use of the Cockcroft–Gault equation *Statistically significant (α = 0.05) # Fisher’s exact test Variable Characteristics Adjusted OR p-value aOR 95% CI Age 58.69 ± 12.81 1.0 .9-1.0 .112 Prior VTE 10.0* 3.9-25.9 < .001 Anticoagulants or anti-platelet agents prescription before admission 4.2* 1.3-13.9 .019 Creatinine clearance levels (mL/min)a 63.91 ± 29.06 1.0* 1.0-1.1 .023 D-Dimer, (μg/mL,max) 2.65 ± 4.74 0.2* .1-.4 < .001 History of varicose veins in the lower extremities 1.5 .4-5.7 .552 Abbreviations: VTE, venous thromboembolism; OR, odds ratios. *Statistically significant (α = 0.05) baseline level(p < 0.1) Table 5. Clinical factors associated with asymptomatic VTE based on multivariable logistic regression analysis Asymptomatic venous thromboembolism-Li et al. Vol 20 No 1 January-February 2023 59 tumor (39, 6.8%), ureteral stricture (24, 4.2%), varico- cele (12, 2.1%), renal cyst (19, 3.3%), urethral stricture (10, 1.7%), stress urinary incontinence (15, 2.6%), oth- ers (91, 15.9%). When comparing the baseline charac- teristics between asymptomatic and symptomatic VTE, only age showed a significant difference, which sug- gested that asymptomatic VTE was more common in older inpatients (Table1). Table 2 shows the patient characteristics associated with asymptomatic VTE. There is a statistically signif- icant difference in age (adjusted odds ratio (OR) = 2.2, 95%CI: 1.1-3.6), with age ≥ 66 having a higher risk of asymptomatic VTE compared to those aged ≤ 65. Table 3 shows the characteristics of medical history associated with asymptomatic with patients who had prior VTE (aOR = 9.8, 95%CI: 4.4-21.9) and who had received anticoagulant or antiplatelet agent use before admission (aOR = 4.8, 95%CI: 1.7-13.2) being more vulnerable to asymptomatic VTE. Other characteristics which may contribute to asymp- tomatic VTE were also analyzed. Table 4 shows cre- atinine clearance (aOR = 0.6, 95%CI: 0.5-0.9) to be a protective factor for asymptomatic VTE while D-Dimer (aOR = 2.1, 95%CI: 1.3-3.3) appears as a potential risk factor, with patients with D-Dimer ≥ 1 being signifi- cantly associated with a susceptibility toward asymp- tomatic VTE. Multivariable logistic regression analysis was carried out to combine all the potential risk factors for asymp- tomatic VTE mentioned above (Table 5). The linearity for quantitative predictors was assessed, and the multi- collinearity problem was avoided (.794, 95%CI: .741- .848). The AUC for the logistic regression model was calculated (Figure 2). Hosmer-Lemeshow goodness of fit test was used to evaluate the calibration ability of the prediction model. The results show that Hos- mer-Lemeshow χ 2 =10.725, P = .218 > .05, suggest- ing that the difference between the predicted value of the model and the actual observed value is not statis- tically significant, and the prediction model has good calibration ability. Patients who experienced prior VTE (aOR = 11.3, 95%CI: 4.4-29.0), received anticoagulant D-Dimer Prediction of Asymptomatic VTE(95% CI) Observed Asymptomatic VTE (95% CI) Total No Yes No n 96 1 97 % of predicted asymptomatic VTE 99.0% (99.5%-99.7%) 1.0% (0.3%-4.5%) 100.0% % of observed asymptomatic VTE 63.6% (57.8%-76.9%) 6.7% (3.0%-30.4%) % of total patients 57.8% (52.6%-70.0%) 0.6% (0.3%-2.7%) Yes n 55 14 69 % of predicted asymptomatic VTE 79.7% (77.1%-85.0%) 20.3% (15.0%-22.9%) 100.0% % of observed asymptomatic VTE 36.4% (25.7%-40.6%) 93.3% (50.8%-99.7%) % of total patients 33.1% (23.3%-36.9%) 8.4% (4.6%-9.6%) Total n 151 15 166 % of observed asymptomatic VTE 100.0% 100.0% 100.0% % of total patients 91.0% 9.0% 100.0% Table 6. Cross tabulation of D-Dimer to predict the occurrence of asymptomatic VTE Abbreviations: VTE, venous thromboembolism. Variable OverallPopulation(n = 56) Good Compliance (n =27) Poor Compliance(n = 24) P -value OR Value Level of education Grade school and illiteracy 17 (33.3%) 8 9 P = .91 1.0 Junior high school and technical secondary school 15 (29.4%) 8 7 1.9 (0.3-10.5) High school 11 (21.6%) 6 5 1.5 (0.3-8.4) Undergraduate and junior college 8 (15.7%) 5 3 1.4 (0.2-8.9) Admission time 2017/01/01-2017/12/31 8 (15.7%) 4 4 P =.235 1.0 2018/01/01-2018/12/31 22 (43.1%) 9 13 2.0 (0.4-10.5) 2019/01/01-2019/06/30 21 (41.2) 14 7 2.9 (0.8-10.0) Understanding of VTE and its dangers Understand 37 (72.5%) 25 12 P = .001 1.0 Don't understand 14 (27.5%) 2 12 12.5 (2.4-64.9)* Patient satisfaction with hospital treatment Satisfied 43 (84.3%) 24 19 P = .571 1.0 Not satisfied 8 (15.7%) 3 5 2.1 (0.5-10.0) Self-rated general health Lower than average 4 (7.8%) 1 3 P = .256 1.0 Average 42 (82.4%) 22 20 0.3 (0.0-3.2) Higher than average 5 (9.8%) 4 1 0.1 (0.0-2.0) Time spending on the journey to the hospital < 1h 16 (31.4%) 12 4 P = .033 1.0 ≥ 1h 35 (68.6%) 15 20 4.0 (1.1-14.9)* Transportation to the hospital Car 31 (60.8%) 17 14 P = .622 1.0 Train 12 (23.5%) 5 7 1.4 (0.3-6.8) Bus 8 (15.7%) 5 3 2.3 (0.4-14.6) Abbreviations: VTE, venous thromboembolism; OR, odds ratios; PE, pulmonary embolism. *Statistically significant (α = 0.05) Table 7. Univariate analysis of compliance with anticoagulation therapy after discharge, good compliant group vs poor compliant group Asymptomatic venous thromboembolism-Li et al. Unclassified 60 or antiplatelet agent use before admission (aOR = 4.5, 95%CI: 1.3-14.9) and had a D-Dimer ≥ 1 (aOR = 4.6, 95%CI: 1.55-13.5) had a positive association with a susceptibility towards asymptomatic VTE. Futhermore, a ROC curve was created to evaluate the ability of the D-Dimer value to discriminate between the symptomat- ic VTE and non-VTE patients (Figure 3). The AUCs for D-dimer diagnosis of thrombosis is .715 (95%CI: .634-.797). And the cut-off point of D-Dimer for the diagnosis of asymptomatic VTE (the threshold value of D-Dimer was determined when the sum of sensitivity and specificity was at its maximum) was 1.785. This means that patient who had a D-Dimer value ≥ 1.785 were asymptomatic VTE. We further assessed the RAM by 4-fold cross-vali- dation, and found its sensitivity was 93.3% (50.8%- 99.7%), specificity is 63.6% (57.8%-76.9%), misdiag- nosis rate is 36.4% (25.7%-40.6%); missed diagnosis Figure 1. Flowchart of patient selection. Figure 2. ROC curve analysis for the logistic regression model. Asymptomatic venous thromboembolism-Li et al. Vol 20 No 1 January-February 2023 61 rate is 6.7% (3.0%-30.4%), PPV is 20.3% (15.0%- 22.9%), NPV is 99.0% (99.5%-99.7%), correct rate is 69.3% and Youden index is 40.7% (Table 6). As for the compliance of anticoagulant medication for asymptomatic VTE, among the 53 patients, 2 were ex- cluded due to contact being lost. After discharge from the hospital, a further 2 of these patients developed low- er extremity pain, 1 patient developed lower extremity swelling, 1 patient had worsening hematuria (which improved 10 days after stopping rivaroxaban), and no patients developed dyspnea, chest pain, or other PE symptoms. The compliance of anticoagulant therapy after discharge was mainly evaluated as either “good” or “poor”. Good compliance was defined if a patient took the anticoagulant drugs regularly and underwent re-examination at the vascular surgery clinic within the time specified by the doctor. Among 51 patients, 46 (90%) took rivaroxaban after discharge from hospital, 4 (8%) warfarin, and 1 (2%) aspirin. 27 patients (52.9%) had good compliance with anticoagulation therapy and 24 patients (47.1%) had poor compliance. From our univariate analysis, we found that patients with a poor awareness of VTE and its dangers (P = .001; 95%CI: 2.406-64.932), and patients who lived more than 1 hour away from the hospital (P = .033; 95%CI: 1.074- 14.896), were more likely to have a poor compliance with anticoagulant therapy after discharge, the differ- ence being statistically significant. (Table 7). DISCUSSION VTE is a serious complication during hospitalization, and its incidence of VTE in the nononcological unit of urology has still not received much attention, especially for asymptomatic VTE. We estimated the incidence and risk factors of asymptomatic VTE, and built an RAM for asymptomatic VTE in urological nononcological inpatients. Our results cast new light on the incidence of VTE in urological nononcological medium-high risk inpatients, with the RAM performing well and yielding good results. Among the 573 inpatients who took part in our study, 73 (15.4%) were diagnosed with VTE, including 20 (4.2%) symptomatic and 53(11.2%) asymptomatic pa- tients. We found that among asymptomatic VTE and symptomatic VTE patients, older patients were more likely to be asymptomatic (65.53 ± 10.08 vs 59.00 ± 12.61, P = .024). This result might be due to the low responsiveness and sensitivity of the older patients, or because the clinical manifestations were atypica,l hid- den, or absent, and therefore could not truly reflect the condition. A large number of clinical studies have shown that asymptomatic VTE remains common despite antico- agulant therapy(16). For this reason, it is important to identify the risk factors associated with asymptomatic VTE to improve its early diagnosis. According to our multivariable logistic regression analysis, we found that prior VTE, administering anticoagulants or antiplatelet agents to patients before admission, or a D-dimer ≥ 1 Figure 3. ROC curve analysis to evaluate the ability of D-Dimer ROC = Receiver Operating Characteristic Asymptomatic venous thromboembolism-Li et al. Unclassified 62 were potential risk factors. Interestingly, older age was not a risk factor for asymptomatic VTE based on mul- tivariable analysis even though age was independent- ly associated with asymptomatic VTE in other stud- ies(17). Prior history of VTE is widely perceived to be a risk factor, regardless of whether it is symptomatic or asymptomati(18,19). It is also easy to understand why administering anticoagulant or antiplatelet agent to paeitnes before admission is also a risk factor, as pa- tients who need to take anticoagulants or antiplatelet agents are more vulnerable to VTE due to their hy- percoagulable state of the blood. As for the D-dimer value, our results were similar to those of other stud- ies, which found it to be significantly associated with asymptomatic VTE(20,21). Interestingly, when D-dimer is considered to be a risk factor for asymptomatic VTE, its value is often higher than the prescribed abnormal value (0.5 μg/mL) (something to be discussed in more detail below). Furthermore, creatinine clearance levels between 60 and 90 mL/min were considered to be a pro- tective factor for asymptomatic VTE patients, which is consistent with other research(22). However, some OR estimates and confidence limits are inflated which sug- gesting sparse-data bias.The majority of data with miss- ing or zero values are included in the data set. Quartile division or cutting off deletion value are used to modify proper data sets. As far as we are concerned, there are no specific anti- coagulation strategy in the current guidelines for uro- logical nononcological asymptomatic VTE, and there are few randomized clinical trials evaluating the effec- tiveness and safety of anticoagulation for asymptomatic DVT. Yugo et al retrospectively evaluated 300 patients with asymptomatic lower extremity DVT and found that most asymptomatic DVT patients had undergone long-term anticoagulation therapy, due to the risk of major bleeding(23). Our RAM assessed the risk of asymptomatic VTE was based on the D-dimer value, which was recommend- ed to predict VTE in the guidelines of the American College of Chest Physicians(15). D-dimer is produced during the endogenous fibrinolysis of blood clots and plays an important role in the diagnosis algorithm to rule out VTE. It is considered to be the best biomarker for the early screening of VTE due to its high sensitivity but does have a poor specificity, and thus false-positive D-dimer results may occasionally occur(24). Balogun et al found that, a D-dimer cut-off point of 1660 ng/ mL in the 48 hours following a stroke could effectively distinguish the asymptomatic VTE patients, with a di- agnosis rate of 72% (13/18)(25). This suggests D-dimer might be a sensitive predictor for asymptomatic VTE. In our study, we noticed that D-dimer ≥ 1 μg/mL was a risk factor for asymptomatic VTE, and that it might increase the likelihood of asymptomatic VTE by about 4.6 times. However, a D-dimer level < 1 μg/mL had no statistical difference when compared with a level < 0.5 μg/mL. In our RAM, the threshold value of D-Dimer was determined when the sum of sensitivity and spec- ificity was at its maximum and the cut-off point was 1.785. The sensitivity of this RAM was 71.7%, spec- ificity 69%, and NPV 95.8%. Other studies have used D-dimer to assess the risk of VTE. For instance, Shi et al. reported in a study of gynecologic malignancy in- patients that the D-dimer threshold needed to be raised to 1.5 μg/ml, with a sensitivity of 87.5%, a specificity 93.8%, and NPV 99.2% in patients with gynecologic malignancies(26). Another study indicated that D-dimer ≥ 0.89 μg/mL might be more suitable for urological on- cological patients, reporting a sensitivity of 83.9%, and a specificity of 80.0%(27). Different groups of people have their own suitable cut- off values. Nevine et al stratified the D-dimer cutoff according to age, and found that sensitivity was com- promised in patients older than 80 years(28). Douma et al. defined a new D-dimer cutoff value as patient's age x 10 in patients aged > 50 years and greatly increased the proportion of older patients in whom PE could be safe- ly excluded(29). Our own aim was to build a RAM for the early detection of asymptomatic VTE in urological nononcological inpatients. Our study found that only 27 (52.9%) patients with non- tumor VTE in the urology department received antico- agulation therapy according to the requirements of the medical advice after discharge. This ratio was far lower than the ratio of good compliance with postoperative preventive anticoagulant therapy reported in other lit- erature(12,30). Among our 51 VTE patients, 14 (27.0%) patients did not understand what VTE was and what its dangers were. Combining the experience of clinical work and the analysis of data, we found that this was mainly due to 2 reasons.(1) Urinary nontumor patients, all were hospitalized because of urological diseases. During the hospitalization period, they were found to have VTE by accident. VTE was not the main reason for their visit.(2) Apart from 15.7% of patients had an un- dergraduate and junior college degree, with the educa- tion level of the remaining patients being low, limiting their understanding of the disease. Our research also found that patients who spend more than 1 hour on the journey to hospital had poor compli- ance with re-examination at vascular surgery within the time specified by the doctor. This made the length of the journey an important indicator affecting patient compli- ance(12,30). Long distances increase the cost of travel and require more time, reducing the willingness of patients to undergo re-examination. Furthermore, after visiting a higher-level hospital, patients were also unwilling to visit a lower-level hospital closer to their homes. It is also important to mention that our study had certain limitations. First, it was a single-center, retrospective analysis, and retrospective bias might be present. Sec- ond, the study only enrolled urological nononcological inpatients admitted to hospital from January 1, 2017, to June 30, 2019, restricting the sample size through an in- sufficient time span. Third, VTE in some asymptomatic patients might have gone undetected due to the patients’ normal D-dimer value or due to their being evaluated as low risk by the in-hospital VTE risk assessment team. Despite the above limitations, we were still able to use some risk factors, such as D-dimer, to build a RAM for the early screening of asymptomatic VTE. CONCLUSIONS We found that prior VTE, use of anticoagulants or an- tiplatelet agents before admission, or a D-dimer ≥ 1 were potential risk factors for the patients in our study. We also found that the more appropriate threshold of D-dimer (at least in Chinese urological nononcological inpatients) for asymptomatic VTE should be elevated to 1.785. In addition, patients with asymptomatic VTE Asymptomatic venous thromboembolism-Li et al. Vol 20 No 1 January-February 2023 63 have low compliance with anticoagulation therapy af- ter discharge. Urologists should strengthen hospital- ization education, carry out targeted instructions, and follow-up regularly after discharge. CONFLICT ON INTEREST There are no conflicts of interest. REFERENCES 1. Allgood R, Cook J, Weedn R, Speed H, Whitcomb W, Greenfield L. Prospective analysis of pulmonary embolism in the postoperative patient. Surgery. 1970;68:116- 22. 2. Chen E, Papa N, Lawrentschuk N, Bolton D, Sengupta S. Incidence and risk factors of venous thromboembolism after pelvic uro- oncologic surgery--a single center experience. BJU int. 2016;null:50-3. 3. Tyson M, Castle E, Humphreys M, Andrews P. Venous thromboembolism after urological surgery. J Urol. 2014;192:793-7. 4. Kalayci A, Gibson C, Chi G, et al. Asymptomatic Deep Vein Thrombosis is Associated with an Increased Risk of Death: Insights from the APEX Trial. Thromb Haemost. 2018;118:2046-52. 5. Violette P, Lavallée L, Kassouf W, Gross P, Shayegan B. Canadian Urological Association guideline: Perioperative thromboprophylaxis and management of anticoagulation. Can Urol Assoc J. 2019;13:105-14. 6. Tikkinen K.A.O., Cartwright R, Gould M.K., et al. EAU Guidelines on Thromboprophylaxis in Urological Surgery European Association of Urology. Eur Urol. 2017. 7. Chan N, Stehouwer A, Hirsh J, et al. Lack of consistency in the relationship between asymptomatic DVT detected by venography and symptomatic VTE in thromboprophylaxis trials. Thromb Haemost. 2015;114:1049-57. 8. Grant P, Greene M, Chopra V, Bernstein S, Hofer T, Flanders S. Assessing the Caprini Score for Risk Assessment of Venous Thromboembolism in Hospitalized Medical Patients. Am J Med. 2016;129:528-35. 9. Singh D, Lawen J, Alkhudair W. Does pretransplant obesity affect the outcome in kidney transplant recipients? Transplant Proc. 2005;37:717-20. 10. Meriwether K, Antosh D, Knoepp L, Chen C, Mete M, Gutman R. Increased morbidity in combined abdominal sacrocolpopexy and abdominoplasty procedures. Int Urogynecol J. 2013;24:385-91. 11. Wiznia DH, Swami N, Nguyen J, et al. Patient compliance with deep vein thrombosis prophylaxis after total hip and total knee arthroplasty. Hematol Rep. 2019;11:7914. 12. Marchocki Z, Norris L, O'Toole S, Gleeson N, Saadeh FA. Patients' experience and compliance with extended low molecular weight heparin prophylaxis post-surgery for gynecological cancer: a prospective observational study. Int J Gynecol Cancer. 2019. 13. Chan JCY, Roche SJ, Lenehan B, O'Sullivan M, Kaar K. Compliance and satisfaction with foot compression devices: an orthopaedic perspective. Arch Orthop Trauma Surg. 2007;127:567-71. 14. Matsuoka Y, Morimatsu H. Incidence Rates of Postoperative Pulmonary Embolisms in Symptomatic and Asymptomatic Patients, Detected by Diagnostic Images　- A Single-Center Retrospective Study. Circ J. 2019;83:432-40. 15. Gould MK, Garcia DA, Wren SM, et al. Prevention of VTE in nonorthopedic surgical patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence- Based Clinical Practice Guidelines. Chest. 2012;141:e227S-e77S. 16. Olson SR, Shatzel JJ, DeLoughery TG. Asymptomatic "breakthrough" thrombosis and anticoagulant "failure": Keep calm and carry on. Res Pract Thromb Haemost. 2019;3:498- 502. 17. Gainsbury ML, Erdrich J, Taubman D, et al. Prevalence and Predictors of Preoperative Venous Thromboembolism in Asymptomatic Patients Undergoing Major Oncologic Surgery. Ann Surg Oncol. 2018;25:1640-5. 18. Pedersen MH, Wahlsten LR, Grønborg H, Gislason GH, Petersen MM, Bonde AN. Symptomatic Venous Thromboembolism After Achilles Tendon Rupture: A Nationwide Danish Cohort Study of 28,546 Patients With Achilles Tendon Rupture. Am J Sports Med. 2019;47:3229-37. 19. Yukizawa Y, Inaba Y, Kobayashi N, Kubota S, Saito T. Current risk factors for asymptomatic venous thromboembolism in patients undergoing total hip arthroplasty. Mod Rheumatol. 2019;29:874-9. 20. Demelo-Rodríguez P, Cervilla-Muñoz E, Ordieres-Ortega L, et al. Incidence of asymptomatic deep vein thrombosis in patients with COVID-19 pneumonia and elevated D-dimer levels. Thromb Res. 2020;192:23-6. 21. Tasaka N, Minaguchi T, Hosokawa Y, et al. Prevalence of venous thromboembolism at pretreatment screening and associated risk factors in 2086 patients with gynecological cancer. The journal of obstetrics and gynaecology research. 2020;46:765-73. 22. Janus N, Mahé I, Launay-Vacher V, Laroche J, Deray G. Renal function and venous thromboembolic diseases. J Obstet Gynaecol Res. 2016;41:389-95. 23. Yamashita Y, Shiomi H, Morimoto T, et al. Asymptomatic Lower Extremity Deep Vein Thrombosis　- Clinical Characteristics, Management Strategies, and Long-Term Outcomes. Circ J. 2017;81:1936-44. 24. Kesieme, Kesieme. Deep vein thrombosis: a clinical review. J Blood Med. 2011;2:59-69. 25. Balogun I, Roberts L, Patel R, Pathansali R, Kalra L, Arya R. Clinical and laboratory predictors of deep vein thrombosis after acute Asymptomatic venous thromboembolism-Li et al. Unclassified 64 stroke. Thromb Res. 2016;142:33-9. 26. Shi J, Ye J, Zhuang X, Cheng X, Fu R, Zhao A. Application value of Caprini risk assessment model and elevated tumor-specific D-dimer level in predicting postoperative venous thromboembolism for patients undergoing surgery of gynecologic malignancies. J Obstet Gynaecol Res. 2019;45:657-64. 27. Shi A, Huang J, Wang X, et al. Postoperative D-dimer predicts venous thromboembolism in patients undergoing urologic tumor surgery. Urol Oncol. 2018;36:307.e15-.e21. 28. Kassim NA, Farid TM, Pessar SA, Shawkat SA. Performance Evaluation of Different d-Dimer Cutoffs in Bedridden Hospitalized Elderly Patients. Clin Appl Thromb Hemost. 2016;23:998-1004. 29. Douma R, le Gal G, Söhne M, et al. Potential of an age adjusted D-dimer cut-off value to improve the exclusion of pulmonary embolism in older patients: a retrospective analysis of three large cohorts. BMJ. 2010;340:c1475. 30. Hordern CE, Bircher CW, Prosser-Snelling EC, Fraser FK, Smith RP. Patient compliance with postnatal thromboprophylaxis: An observational study. J Obstet Gynaecol. 2015;35:793-6. Asymptomatic venous thromboembolism-Li et al. Vol 20 No 1 January-February 2023 65