Comparative analysis of hidden blood loss in total knee arthroplasty

The aim of the study was to investigate the influential factors for hidden blood loss after a total knee arthroplasty and their relationship with the total blood loss. Total knee arthroplasty (TKA) is a crucial treatment of late-stage knee osteoarthritis. Measured blood loss is significantly inconsistent with the hemoglobin (HB) drop postoperatively. Fifty-four patients, 20 males and 34 females, were retrospectively analyzed. The preoperative blood loss and therefore the hidden blood loss following TKA were calculated by the Gross formula. The typical perioperative blood loss was found to be 780±220 ml and therefore the average hidden blood loss was 280±180 ml. No significant differences were found in hidden blood loss for males compared to females. Hidden blood loss may not be reduced by hemostasis during operation with a deflated tourniquet.


Introduction
Total knee arthroplasty (TKA) may be a gold standard treatment for endstage knee osteoarthritis. This surgery is considered the foremost effective treatment for severe knee osteoarthritis with good or excellent long-term outcomes 1,2,3 . It represents a serious orthopaedic surgery which involves a significant blood loss due to bone cuts and soft tissue releases. Approximately 15 million transfusions are performed annually worldwide caused by TKA 4 .
In this surgery the blood loss consists of the visible blood loss from the wound drainage, into the tissues which is hidden and the surgical field. The measured blood loss usually consists of measurement of the drained blood and therefore the real blood loss could also be underestimated. Despite the usage of a tourniquet during TKA and the significant improvement in surgical techniques, this surgery is related to a large amount of post-operative blood loss and remains a topic for discussion among surgeons. Lotke et al. came to a conclusion that the mean blood loss in TKA when calculated from pre-operative drop by hemoglobin was 1518 ml, which needed a minimum of one transfusion 5 . Transfusion increases the danger of transmission of infectious diseases, anaphylactic, hemolytic and immunologic reactions. Аlso, it could lead to a longer hospital stay and increased mortality 6 .
The existing results of influential factors in thеsе patients remain controversial. Identifying the factors of surgical blood loss may be a crucial step towards establishing an efficient blood management strategy and reducing the necessity of peri-operative transfusion. Cushner and Friedman et al. discovered that the gender of patients may influence the blood loss; transfusion rates were higher in females probably owing to lower preoperative hemoglobin levels, but they excluded the correlation with the age and duration of the surgery 7 .
The aim of our study was to detect and recognize the potential influencing factors of hidden blood loss in patients with knee osteoarthritis, treated with total knee arthroplasty.

Materials and methods
The study was performed at the University Clinic for Orthopaedic Surgery from December 2018 to December 2019. A total of 54 patients, of whom 20 males and 34 females with knee osteoarthritis were retrospectively analyzed. We calculated the pеrioperative blood loss and consequently the hidden blood loss by using the Gross formula 7 .
The formula is as follows: for male, PBV = k1 xh3 + k2 xw + k3 (h-height (m); w -weight (kg); k1=0.3669, k2 = 0.03219 and k3 = 0.6041; for female k1 = 0.3561, k2 = 0.03308 and k3 = 0.1833). Total blood loss was calculated by multiplying PBV by the change of HCT. Patients with anemia or any other disorder of hemoglobin preoperatively were excluded from this study. Potential factors who may affect operative and hidden blood loss were gender, surgical time, tourniquet time, hemostasis (during surgery with deflating tourniquet). In the group without tourniquet the tourniquet was released before closure of the wound and hemostasis was done and in the group with tourniquet the wound was closed without releasing it.
In all of the patients a typical surgical technique was used. All of the patients were operated under spinal anaesthe-sia. We used a tourniquet which was inflated 120 mm above the systolic blood pressure. We used the medial parapatellar surgical approach. Cemented cruciate retaining prosthesis was used. Blood loss intraoperatively was measured from the suction device. We put active vacuum drain in every patient. The drain was removed after 48 hours. Hemoglobin and hematocrit levels were measured before and after surgery (after spinal anaesthesia was over, day 2 and day 4 postoperatively).
The total blood loss (during the operation and collected from the drain) was compared to the Hb levels on selected days. In patients who received blood transfusion we made correction of the Hb level by 1g/dL for every unit. After that we made the conversion of these numbers to HCT value.

Results
In this study, the mean operative time was 115 minutes (100-190 minutes) and the mean tourniquet time was 107 minutes (80-120 minutes). Intraoperative blood loss of the group with hemostasis using the deflated tourniquet was 155 ml (90-300 ml) and postoperative drain-blood volume was 595±205 ml for that group. Perioperative blood loss was 780±220 ml [(758±243) ml for females and (912±90) ml for males] and hidden blood loss was 280±180 ml [(240±185) ml for females and (410±125) ml for males]. The hidden blood loss was therefore (51 ± 9.3)% of the visible postoperative blood loss ( Table 1). The lowest hematocrit levels were measured on the 2-4 postoperative day, and the percentage of the hidden blood loss in both sexes was nearly equivalent. Although male patients had higher levels of hidden blood loss than females (t=-2.022, P=0.049), the proportion volume was nearly equivalent (Table 1). Hidden blood loss might be reduced by hemostasis after deflating tourniquet during the intraoperative period (t=-2.068, P=0.041), but the preoperative blood loss was not affected (t=-1.819 P=0.073) ( Table 2).

Discussion
Hidden blood loss (HBL) was usually ignored until year 2000 when it was formally suggested [8]. Several theories were proposed like the idea of Pattison and Faris who suggested that HBL could also be due to hemolysis 9,10. On the opposite, Erskine proposed that the loss went into tissue compartments by using labelled red cells 11. Other theory suggests that free fatty acids generated from fatty emboli in blood circulation are liable for the hidden blood loss by peroxidation damage of membrane molecules of red blood cells and hemoglobin. Tranexamic acid, closed-suction drainage, body mass index (BMI), gender and anticoagulants can also influence HBL. However, the factors have not been fully clarified 12-20. It had been found that this formula closely followed the logarithmic one unless there was rapid hemorrhage when the formula drifted apart, and thanks to this reason, complicated cases with large losses were excluded from this study. To some extent, Gross equation reflects the actual postoperative blood loss due to individual factors: weight, height, gender and transfusion. This equation does not involve Hb-related factors like anemia and actual blood loss is revealed by the calculated pre-operative blood volume and the changes in HCT, which can limit this method.

Conclusion
No significant differences might be found in hidden blood loss in males compared to females. Hidden blood loss remains the same even after hemostasis during operation with a deflating tourniquet.