Patient Characteristics Related to Blood Loss in High Tibial Osteotomy in Novel Multiple Linear Regression Analysis

The purpose of the study was to identify patient characteristics related to blood loss following high tibial osteotomy (HTO). We evaluated 48 patients undergoing HTO from August 2018 to August 2019. The data of 48 patients were collected, including gender, age, height, weight, body mass index (BMI), smoking, alcohol consumption, hypertension, diabetes, history of aspirin, and pre-postoperative hematocrit (Hct). Multiple linear regression analysis was used to analyze the risk factors related to blood loss in HTO. The mean age of patients was 56.6 ± 10.2 years, including 22 males and 26 females. The mean BMI was 28.5 ± 4.2 kg/m2, and the mean blood loss volume was 383.3 ± 181.3 mL, 13 patients with smoking (27.1%), 15 patients with alcohol consumption (31.3%), 23 patients with hypertension (47.9%), 10 patients with diabetes mellitus (20.8%), and 12 patients with history of aspirin (25.0%). Multiple linear regression model suggested alcohol consumption and BMI were associated with blood loss in HTO, R2 = 0.451, F(9, 38) = 3.462 (P < 0.05). Our study indicates that alcohol consumption and BMI are important risk factors related to blood loss in HTO.

Gunter et al. studied 55 patients with HTO and pointed out that the incidence of hematoma was 4.7%, infection was 4.7%, and thrombosis was 2.3% [20]. Seo et al. studied 167 patients undergoing HTO with the TomoFix plate and reported that the incidence of hematoma was 2.4% and stiffness was 1.2 [18]. Martin et al. studied 323 patients undergoing HTO with Puddu or TomoFix plate and reported that the incidence of hematoma was 3%, delayed healing was 12%, cellulitis was 10%, stiffness was 1%, and thrombosis was 1% [19]. Therefore, accurate preoperative prediction of blood loss in high tibial osteotomy is a matter of great concern to surgeons.
This study retrospectively analyzed 48 patients undergoing HTO from August 2018 to August 2019. The risk factors were studied to provide a basis for assessing blood loss before surgery.

Patient Selection.
We retrospectively reviewed patients undergoing HTO from August 2018 to August 2019. The inclusion criteria were (1) symptomatic medial osteoarthritis, (2) the first operation, and (3) clinical data is integrity. Exclusion criteria were (1) revision surgery, (2) medical disputes, (3) preoperative hemoglobin ðHbÞ < 90 g/L, and (4) history of thromboembolism. 48 patients were included finally. All operations were performed by the same surgeons.

Data
Collection. The data of 48 patients were collected, including gender, age, height, weight, BMI, smoking, alcohol consumption, hypertension, diabetes, history of aspirin, and pre-postoperative Hct. The total blood volume was calculated using height and weight by the formula of Nadler et al. [21]. The blood loss was calculated by the Gross formula [22]. Collinearity statistics was used to assess multicollinearity between independent variables, and multiple linear regression analysis was used to analyze the risk factors related to blood loss in HTO. The SPSS 22.0 software was used for statistical analysis, and P < 0:05 was considered statistically significant.
The multiple linear regression analysis showed that there is no autocorrelation detected in residuals (the Durbin-Watson value was 2.348), without the presence of multicollinearity (the tolerance value > 0:1 in all independent variables), the regression of standardized residual of blood loss volume follows a normal distribution closely, and the adjusted R 2 value was 0.451.

Discussion
Application of tourniquet, osteotomy at the metaphyseal level, and soft tissue release intraoperative may cause intraand postoperative bleeding. Kim et al. studied 150 patients undergoing OW-HTO and pointed out that the blood loss volume was 502:4 ± 294:9 mL in the group with tranexamic acid and 882:7 ± 482:0 mL in the group without tranexamic acid [32]. Jeya et al. studied 156 patients undergoing OW-HTO and pointed out that the blood loss volume was 372 ± 36 mL in the group with tranexamic acid and 635 ± 53 mL in the group without tranexamic acid [33]. Our study showed that the blood loss volume was 383:3 ± 181:3 mL without the application of tranexamic acid in all patients.
To our knowledge, this is the first study investigating the patient characteristics related to blood loss in HTO. In our multiple linear analysis, we found that BMI and alcohol consumption are risk factors related to blood loss in HTO.
Jibodh et al. pointed out that operative time is longer and blood loss is higher in patients with BMI ≥ 40 kg/m 2 in hip arthroplasty [36]. Frisch et al. studied 2399 patients undergoing total hip arthroplasty (THA) or total knee arthroplasty (TKA) and pointed out that the estimated blood loss of the normal group was 268:2 ± 313:9 mL, the overweight group was 282:0 ± 208:7 mL, and the obese group was 330:5 ± 302:4 mL in THA (P = 0:001) and the estimated blood loss of the normal group was 85:7 ± 153:8 mL, the overweight group was 90:5 ± 164:6 mL, and the obese group was 89:4 ± 72:4 mL in TKA (P = 0:001) [39]. This could explain that the obese patients own greater overall blood volume physiologically and need larger exposure, need greater soft tissue dissection, and has the possibility of bleeding [39,40].
Excessive alcohol consumption has been associated with adverse outcomes in orthopedic surgeries [41][42][43]. However,  there is rare research on the relationship between alcohol consumption and blood loss. We consider that alcohol consumption leads to excessive blood loss at multiple levels, for example, abnormally low platelet numbers in the blood, impaired platelet function, and impairment of fibrinolysis. Alcohol consumption can exert direct toxic effect on platelet production, survival, and function; previous studies have shown that alcohol has negative effects on platelet-platelet interaction and can produce thrombocytopenia which seems to be the commonly cause of platelet depletion [44][45][46]. Our study analyzes patient characteristics related to blood loss in HTO, wishing to establish application model in order to predict the risk of blood loss preoperative and guide blood management perioperative by patient characteristics.
This study has limitations: (1) It is a small retrospectively study. (2) The research is a single-center study in which all operations were performed by the same surgeons. The results may be biased and further larger, multicenter research was needed to confirm our findings.

Conclusion
Our study indicates that alcohol consumption and BMI are important risk factors related to blood loss in HTO; the surgeons can predict the risk of blood loss by evaluating patient characteristics.

Data Availability
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.