Total knee arthroplasty patients in our institution from 1 January 2010 to 31 December 2016 were reviewed retrospectively. Inclusion criteria: adult patients (≥ 18 years), first time of TKA, inexistence of DVT preoperative (determined by lower extremity deep venous of ultrasonography). Exclude criteria: haemophiliac, revision surgery, limb activity limitation for stroke, sym-DVT occurred at health extremity, and loss to follow-up. The follow-up flowchart was showed in figure 1.
Medical data was included: age, gender, BMI (body mass index), cause of surgery, blood test (CRP (C-reactive protein), albumin, D-dimer, lactic acid), medical history of (diabetes mellitus, hypertension, coronary heart disease, myocardium infarction MI (Myocardial infarction), DVT), drug history of antiplatelet drug (aspirin or clopidogrel), operation and anesthesia data (ASA (American society of anesthesiologists) grade, anesthesia type, procedure time, tourniquet use and time), bleeding during operation, drainage postoperative, postoperative data (low limb swelling (calf perimeter of 10 cm below tibia tuberosity), time of weight bearing. According to WHO (World health organization) described, BMI was divided as: low weight (BMI < 18.5 kg/m2), normal (BMI = 18.5 - 25 kg/m2), over weight (BMI = 25 - 30 kg/m2), obesity I grade (BMI = 30 - 35 kg/m2), obesity II grade (BMI = 35 - 40 kg/m2), obesity III grade (BMI > 40 kg/m2). CCI (Charlson comorbidity index) was used and divide into three level: low level (0 score) means no comorbidity, middle level (1 - 2 scores) and high level (over 3 scores) (11).
Thrombo-prophylaxis was included chemoprophylaxis and physical prophylaxis. LMWH (low molecular weight heparin) 4000 U was used 12 hours after surgery and once daily last hospitalization. LMWH was used as bridging therapy if patient was used antiplatelet drug. In the meanwhile, pump treatment was used one day after surgery until full weight bearing.
Fellow-up and diagnosis of symptomatic DVT
If the patient occurred low leg pain and tenderness with swelling, or calf circumference (10 cm under tibia tuberosity) was increased over 2 cm than health limb, or swelling was severing. (12) And deep vein Doppler ultrasound would be performed if patient had positive of Homans sign. Then the patient would be diagnosed had sym-DVT. Pulmonary arterial CTA (computed tomography angiography) would be performed immediately if patient had chest pain, short of breath or dyspnea. All patients had education of DVT and PE, and they would be required go to hospital immediately if the symptoms appeared.
Statistical method
Student t-tests or Wilcoxon rank sum tests were used to compare numerical variables between groups, and chi-square tests were used in categorical variables. Repeated Measures were used to analyze BP parameter at different period. One-sample t-tests were used to analyze outlier from base line. Multivariate Covariance analyses were used in variation of BP parameter during procedure. Pearson Correlation analyses were used in two numerical variables. Variables with P ≤ 0.1 were considered as covariates (potential risk factors) and K-Independent Samples tests were used to analyze correlation between groups. ROC analyses were used in numerical variables (if P ≤ 0.1) to assess fitness cut-off value, then rounded to the nearest whole number and transform numerical variables to binary variables. Variables with P ≤ 0.1 enter in multiple logistic regression analysis (choose Forward: Conditional method). The independent risk factors were showed as OR (odds ratio, 95% CI). P values < 0.05 were considered statistically significant. The statistical analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, Illinois).
The ethics committee approved the study and each participant provided verbal consent.