Previous discuss of HBL was addressed in hip surgery and TKA. The idea that TKA results in significant HBL was reported in literatures[7, 8], and there were complications such as hypotension, acute myocardial infarction caused by acute postoperative anemia[6]. Furthermore, homologous transfusion consequent on HBL could increase a potential periprosthetic infection risk[9]. On the other hand, the literatures about blood loss of UKA was limited. Schwab et al. compared 105 UKA cases with 105 TKA cases and concluded that Hb drops more sharply in TKA with more HBL [10]. Yang et al. reported a similar result via a prospectively matched study of 100 patients[11]. Ann reported lower blood loss in bilateral simultaneous UKA compared with unilateral TKA[12]. However, the cause of HBL in UKA was not well explained. UKA does have some potential advantages such as smaller incision, less injury, shorter operation time compared with TKA, but invasive process is inevitable such as osteotomy, femoral canal opening, some synovectomy. Besides, the utilization of tourniquet is considered as a main risk factor of HBL[13].
Tourniquet is preferred by most of joint surgeons during arthroplasty. Theoretically, the HBL or drop of Hb are not supposed to be significant when the procedure is performed with tourniquet. However, the postoperative blood loss is much more than expected, sometimes over 2000ml[14]. Hemolysis is considered as a main factor for this scenario. The vein of involved extremity dilate rapidly after tourniquet release, which results in venous stasis and triggers the process of fibrinolytic reaction[3, 4]. The impact of tourniquet to blood loss is still controversial. In this study we removed the impact from tourniquet and observed significant less HBL in UKA compared with TKA, but it could not be ignored by any means. This finding may be related to the technique of controlled hypotension employed by our anesthesiologist in procedure. The blood pressure maintained at a low level for improving visualization of structures during the procedure, but it rose as a response to pain after patient’s recovery which would invite bleeding. Besides, the absence of tranexamic acid also played a role[15, 16].
Debates exist on gender being correlated to HBL. Prasad et al.reported that males suffer more HBL[1], while some other literatures demonstrated that there was no significant difference of HBL in genders [2, 6]. Our study found that HBL has no significant difference between genders in either UKA group or TKA group. It is controversial as well about the correlation between HBL and age. Yoshihara et al.took oldness as a risk factor of HBL due to the poor capacity of Hb reserve of old patients. They are prone to suffering more HBL and are more vulnerable to blood loss[17]. Durasek et al.reported an opposite result in a prospective randomized study on 184 patients, they found older patients had lower hemoglobin level before blood transfusion and received greater volumes of blood transfusion, but the difference was not statistically significant[2]. In our study, we also found the difference on HBL between different age groups was not significant. In Frisch et al.’s study[18], BMI played a role in HBL and postoperative transfusion, they suspected that vascular wall sclerosis could be a factor. The correlation between HBL and BMI was not significant in our study. However, this finding may be due to the small sample size of our study, with most patients’ BMI ranging from 18 to 26 kg/m2, and the patients with a high BMI could not be evaluated accurately.
Preoperative Hb is usually stable, while postoperative Hb varies widely. Schwab et al.found that Hb dropped to bottom around the 4th day after TKA or UKA[10]. Prasad et al. reported the reduction of Hb usually started from one week after TKA[1]. In terms of these findings, our study bracketed the first 8 days post operation as an observing period. We picked the 2nd day, the 4th day, the 6th day, and the 8th day post operation as time nodes to see the variation of Hb, and found that Hb level in UKA was higher than it was in TKA at all the four time nodes. Moreover, in UKA group the Hb level reached to the bottom at the 4th day postoperatively and begun to rebound after that, while in TKA group the Hb kept dropping during observing period. We consider that the minimal invasive procedure could benefit circulating blood volume and relieve the impact to patients.
Autologous blood transfusion was performed on all the patients of this study as the absence of tourniquet. The purpose of ABT is for the reduction of blood volume. Li et al. demonstrated that ABT could reduce the risk of allogenic blood transfusion[19]. However, ABT itself is a possible risk factor of HBL in the study of Xu et al. They found HBL was significantly less in the TKA group without ABT[20]. There was no case required allogenic blood transfusion in UKA group. Although in 2 cases the Hct dropped below 0.28L/L postoperatively but the Hb maintained beyond 80 g/L. Allogenic blood transfusion was performed on 2 female cases in TKA of our study. Brown et al. considered the very low rate of transfusion in UKA can be related to the sex distribution[21]. They explained that males took up a higer proportion in UKA than in TKA, and males’ higher level of preoperative blood volume and Hb contributed to low rate of transfusion. The difference of sex distribution between UKA group and TKA group was not significant in our study, and the transfusion rate of TKA group was even higher than 2% which was reported in Schwab et al.’s study[10], so we hold the opinion that lower transfusion rate is more affected by surgery rather than gender or ABT.
As we know, the relationship of HBL and knee function is rarely mentioned in literatures. The hypoperfusion in operation area caused by HBL would compromise soft tissue healing, and it is associated with lower limb swelling[6]. However,it is still unclear that if HBL would compromise postoperative knee function. Budde et al. found better knee function in patients with less intra- and post-operative blood loss[22]. They compared 34 patients undergoing standard TKA with 34 patients receiving a kind of hemostatic after TKA, and found a better function outcome in the hemostatic group. In Joyce et al.’s study, the relationship of blood loss and joint function was not significant[23], and they concluded that using the investigated haemostatic agent regularly was not necessary. UKA is believed having better knee function than TKA at early follow-up[21]. Our study supports this idea when we found the HSS score is significantly higher in UKA group at 3 month after procedure. However, the difference tended to disappear at 12 months postoperatively. Meanwhile, we found that the correlation of HBL and knee function has no clinical significance in UKA group. It may be due to the fact that the reduction of blood loss in UKA could be achieved by individual compensation without allogenic blood transfusion. And it also reminded us that the perioperative blood loss had limited influence on the outcome of UKA.
There were some limitations in our study. First, the correlation between BMI and HBL was not very convincing as an insufficient observation of the patients with BMI out of the interval 18ཞ26 kg/m2. Second, the continuous observation of Hb level after surgery was too short to obtain a full pattern of Hb variation. Third, all the procedures in our study were performed in one institution by one surgeon who followed a standard protocol, hence, some other risk factors of HBL such as the use of thromboprophylaxis, the use of drain were not included in the analysis.