Comparison of total blood loss between limited tourniquet use and conventional tourniquet use in total knee arthroplasty: a randomized controlled trial

Abstract

Background

This study examined the differences in total blood loss, the need for blood transfusions, length of hospital stay, wound grading scores, incidence of venous thromboembolism (VTE), and reoperation rates between conventional and limited tourniquet use during unilateral primary total knee arthroplasty (TKA).

Methods

This double-blind, randomized controlled trial included 90 patients undergoing unilateral primary TKA. Forty-five patients were allocated to the limited tourniquet use group (LIM-TKA), and 45 to the conventional tourniquet use group (CON-TKA). The study analyzed differences in total blood loss, the need for blood transfusions, wound grading scores, incidence of VTE, length of hospital stay, and reoperation rates.

Results

The mean total blood loss in the LIM-TKA group was 589.55 ± 238.2 ml, significantly lower than the 692.31 ± 276.15 ml observed in the CON-TKA group (P = 0.031). Significantly poorer wound grades were seen in the CON-TKA group compared to the LIM-TKA group, with 23 vs. 34, 15 vs. 11, 5 vs. 0, 1 vs. 0 and 1 vs. 0 patients, respectively, having grade 1, 2 A, 2B, 2 C and 3 A wounds according to the Southampton Scoring System (P = 0.032). Patients in the LIM-TKA group also had a shorter hospital stay versus the CON-TKA group (5.6 ± 1.28 vs. 7.2 ± 4.06 days, P = 0.006). The rates of blood transfusion, VTE complications, and reoperation were similar between both groups.

Conclusion

LIM-TKA results in significantly lower total blood loss and improved wound grading scores, as well as a decreased length of hospital stay compared to CON-TKA. LIM-TKA could be a suitable option for surgeons aiming to minimize tourniquet-related adverse outcomes while maintaining a dry bone surface during TKA cementation.

Clinical trial number

This study was retrospectively registered at Thai Clinical Trials Registry (thaiclinicaltrials.org) on July 28, 2024. Clinical trial number: TCTR20240728002.

Background

Tourniquets are widely used in extremity surgeries (e.g., lower extremity fracture fixation and joint reconstruction), enabling the surgeon to create a bloodless surgical field with improved visualization and minimal intraoperative blood loss [1]. Dry and clean bone surfaces are essential especially during the cementing of components in total knee arthroplasty (TKA). However, several studies have associated prolonged tourniquet use with increased complication rates, caused by direct pressure underneath the cuff and tissue hypoxia distal to the tourniquet cuff [2, 3]. Tissue hypoxia in TKA can lead to higher levels of postoperative pain, delayed recovery of knee function, increased thromboembolic complications, and other wound complications [4,5,6]. Therefore, the use of tourniquets during TKA is debatable, with no consensus reached [7, 8]. However, these previous studies [7, 8] compared the outcomes of conventional tourniquet use in TKA (i.e., an inflated tourniquet throughout the entire operation) versus no use of the tourniquet at all. The risks and benefits of using a tourniquet in TKA must be weighed, since growing evidence favors the decreased use of a tourniquet, which can confer short-term benefits due to better early postoperative patient recovery. In conventional tourniquet use, the effects of prolonged tissue hypoxia can be decreased by using the tourniquet only when it is crucially needed. One example is using the tourniquet during the cementing of TKA components, because a dry and clean bone surface is essential for its success. This technique can reduce tourniquet time by an average of approximately 50 min compared to conventional tourniquet use in TKA [9]. Thus, surgeons can opt for the limited use of a tourniquet in TKA along with contemporary modalities to decrease intraoperative blood loss in order to decrease tourniquet time while improving cementation. However, the best of our knowledge, the results of trials comparing limited and conventional tourniquet use in TKA are conflicting in terms of blood loss [9,10,11].

The primary objective of this randomized controlled trial was to compare the total blood loss between patients who underwent TKA with conventional versus limited tourniquet use. The secondary objectives were to compare the amount of blood transfusions required, rates of tourniquet-related complications (i.e., wound and thromboembolic complications), length of hospital stay, and reoperation for any cause. We hypothesized that limiting tourniquet use in TKA would reduce total blood loss, minimize post-operative complications, and shorten the hospital stay.

Methods

This double-blind randomized controlled trial was conducted with the approval of the Institutional Review Board Committee of Navamindradhiraj University (COA 102/2567) and was registered at thaiclinicaltrials.org (TCTR20240728002). Informed consent was obtained from all patients prior to participation. Ninety patients undergoing TKA between April 2018 and February 2019 at faculty of medicine Vajira hospital agreed to participate in the study. We included patients aged between 50 and 80 years undergoing unilateral TKA. Patients with secondary osteoarthritis of the knee, those undergoing simultaneous bilateral TKA, and those with a history of venous thromboembolism (VTE) were excluded. This study adheres to the CONSORT guidelines, and the CONSORT flow diagram is shown in Fig. 1.

Fig. 1

CONSORT Flow Diagram of participants enrolled in this study

Participants were evenly allocated to one of two groups: the limited tourniquet use during TKA (LIM-TKA) group and the conventional tourniquet use during TKA (CON-TKA) group. Allocation was randomized by an independent investigator using a computer-generated block of four randomizations. The randomization groups were placed in sealed, opaque envelopes and kept in the patient’s chart. All patients received spinal anesthesia followed by an intraoperative tranexamic acid protocol, which included 500 mg intravenously before the incision and another 500 mg intravenously after the tourniquet was released. The envelope containing the allocated treatment was opened by the operating room nurse after spinal anesthesia was administered. A tourniquet cuff was applied to the thigh of the patient’s operated leg and connected with electronic tourniquet machine (Zimmer Biomet A.T.S.^® 4000). The tourniquet pressure was set to the patient’s systolic blood pressure plus 100 mm Hg for both groups. This was inflated after using an Esmarch bandage to exsanguinate the blood, while the knee was in full flexion with the ankle dorsiflexed. In the LIM-TKA group, the tourniquet was inflated only during the cementing phase of TKA, which occurs after all bone cuts are finished, then deflated after cementing the prosthesis. In the CON-TKA group, the tourniquet was inflated just before the incision is made, maintained throughout the surgery, then deflated after cementing the prosthesis.

All patients underwent surgery performed by three arthroplasty surgeons who each handled at least 100 TKA cases annually. All patients received cemented cruciate-retaining knee prostheses. The surgical technique was as followed. A midline skin incision and midvastus arthrotomy approach were utilized, followed by the removal of anterior cruciate ligament and osteophytes. The proximal tibia was resected using an extramedullary guide aimed perpendicular to the anatomical axis of the tibia. The distal femur was resected using an intramedullary guide rod with the valgus angle set to 5°. A medial soft tissue release was performed until rectangular extension gap was obtained [12]. Femoral sizing and rotation were determined using soft-tissue tension, as described by the gap-balancing technique [13]. Anterior, posterior, and chamfer cuts of the femur were performed. The bone plug was inserted through the IM guide hole, followed by final tibial preparation. Finally, the patella was resurfaced. After the bone surfaces were cleaned and dried, all components were cemented, and the tibial polyethylene insert was placed. The use of tourniquet was different in each group as describe above. We carefully check for bleeding after deflation of tourniquet in all cases. A suction drain was inserted into the joint capsule, and then the capsule, subcutaneous tissue, and skin were closed layer by layer. Patients in both groups followed the same VTE prophylaxis protocol, which included using an intermittent pneumatic compression device during hospital and aspirin 81 mg orally twice a day for 14 days [14]. The suction drain was removed on the second postoperative day in all patients. Patients received appropriate transfusion of packed red blood cells (PRC) if their hemoglobin (Hb) level dropped below 7 g/dl or less than 10 g/dl in patients with preexisting cardiovascular disease or with symptomatic anemia postoperatively. The pain and rehabilitation protocols were the same for both groups. The postoperative pain management protocol began with intravenous ketorolac, which was switched to oral NSAIDs after the second postoperative day. Opioids were given as needed for any additional pain. The rehabilitation protocol involved gentle range-of-motion exercises and encouraged walking with a walker under the supervision of a physiotherapist. Patients were discharged once they could walk independently with or without gait aid, had no wound complications, and adequate pain control with oral analgesics. Follow-up clinic visits were done at 2 weeks, 4 weeks, 3 months, and 6 months postoperatively, then annually thereafter.

Outcomes

The primary outcome of this study is the total blood loss, which is determined by the Hb balance formula [15]. First, the amount of total Hb loss after TKA is calculated using the formula:

$$\:{Hb}_{loss}\:=\:BV\times\:\left({Hb}_{i}-{Hb}_{e}\right)\times\:0.001+\:{Hb}_{t}$$

BV represents the blood volume in milliliters, calculated based on sex, weight, and height using the Nadler et al. method [16]. Hb_i is the patient’s Hb concentration before surgery, Hb_e is the patient’s Hb concentration on the third post-operative day, and Hb_t is the total amount of Hb concentration that has been transfused to the patient. A unit of banked blood generally contains approximately 52 ± 5.4 g of Hb [17]. After the total Hb loss has been calculated, the total blood loss after surgery is calculated using the formula:

$$\:Blood\:loss\:=\:1000\times\:\:{Hb}_{loss}/{Hb}_{i}$$

An independent investigator collected the Hb levels preoperatively and on the third postoperative day, assessed the need for blood transfusion, calculated total blood loss, and recorded the length of hospital stay after TKA. During the second postoperative week, the investigator graded the operative wound using the Southampton Scoring System [18] and evaluated for VTE complications. Patients with suspected VTE were referred for Doppler ultrasonography for confirmation. Any TKA-related complications will be recorded throughout the follow-up period.

Statistical analysis

A sample size of 45 patients per group was calculated based on the total blood loss reported in a previous study by Tarwala et al. [11] to achieve a power of 80% and a Type I error rate of 5%. Continuous data were reported as means and standard deviations (SD). The Kolmogorov-Smirnov test was used to assess data normality, and differences between the two groups were evaluated using an independent t-test. Categorical data were presented as absolute numbers and percentages, and differences between groups were tested using the Chi-square test. IBM SPSS Statistics for Windows (Version 29.0, Armonk, NY: IBM Corp.) was used for statistical analysis, with p < 0.05 considered statistically significant.

Results

This study included 90 patients who were equally divided into the LIM-TKA and CON-TKA groups (n = 45 each). Demographic data such as gender, age, BMI, and American Society of Anesthesiologists (ASA) classification were comparable between both groups (Table 1). Patients in the LIM-TKA group had a greater mean varus mechanical axis deviation compared to those in the CON-TKA group (11.13° ± 7.40° vs. 7.90° ± 6.49°, P = 0.015; Table 2). However, the postoperative mechanical axis was similar between both groups. Table 2 presents the radiographic data.

Table 1 Patient demographics and characteristics

Table 2 Radiographic outcomes data

The outcomes related to blood loss and blood transfusion are shown in Table 3. The LIM-TKA group had significantly less total blood loss versus the CON-TKA group (589.55 ± 238.20 vs. 692.31 ± 276.15 mL, P = 0.031; Table 3). The Hb levels were similar between the groups preoperatively and on the third postoperative day, although the CON-TKA group had slightly lower Hb on the third postoperative day. More patients in the CON-TKA group required blood transfusions after surgery versus the LIM-TKA group, but this difference was not statistically significant (4 vs. 2 patients).

Table 3 Blood loss and blood transfusion outcome data

Patients in the CON-TKA group had significantly worse Southampton wound scores versus those in the LIM-TKA group (P = 0.032; Table 4). Specifically, the LIM-TKA and CON-TKA groups, respectively, had 34 and 23 patients with grade 1 wounds, 11 and 15 patients with grade 2 A wounds, and 0 and 7 patients with grade 2B or higher wounds. The total length of stay was significantly lower in the LIM-TKA group versus the CON-TKA group (5.6 ± 1.28 vs. 7.2 ± 4.06 days, P = 0.006; Table 4). There were no cases of suspected VTE symptoms requiring further evaluation for VTE complications. A summary of wound grading and the length of hospital stay is shown in Table 4. During the follow-up period, three patients in the LIM-TKA group experienced TKA-related complications. One patient required revision due to prosthesis malalignment noted 1 month postoperatively. Two patients developed acute periprosthetic joint infections at 4 and 5 years post-surgery, but both were successfully treated with debridement, antibiotics, and implant retention. Meanwhile, three patients in the CON-TKA group developed postoperative complications. One patient required a revision for prosthesis malalignment 2 months postsurgery, while 2 patients had wound complications that required surgical management during their hospital stay after the initial arthroplasty. The average WOMAC score at the last follow-up was 6 and 3.58, respectively, for the CON-TKA and LIM-TKA groups, respectively.

Table 4 Wound grading according to the Southampton Scoring System and length of hospital stay

Discussion

In this study, LIM-TKA was associated with less total blood loss and better surgical wound scores when compared to CON-TKA. Conversely, a meta-analysis by Wang et al. revealed that LIM-TKA results in increased total blood loss compared to CON-TKA, but the two have a similar number of postoperative blood transfusions, while LIM-TKA had less wound complications [19]. However, some of the studies included in the meta-analysis may be outdated, since they preceded the routine use of tranexamic acid, hypotensive anesthetic techniques, and other methods of local hemostasis to decrease blood loss. A recent study by Diri et al., which utilized an intraoperative tranexamic acid regimen and spinal anesthetic technique, showed no significant difference in total blood loss between LIM-TKA and CON-TKA (1060.61 ± 252.23 vs. 1023.52 ± 190.11 mL) [20]. However, a conflicting study by Çi Nka et al. demonstrates higher total blood loss in the LIM-TKA group compared to the CON-TKA group (679.92 ± 283.48 mL vs. 478.35 ± 239.78 mL) [21]. Conversely, our study showed the LIM-TKA group had significantly less total blood loss compared with the CON-TKA group, which yielded different results versus previous studies despite using similar perioperative protocols to decrease intraoperative blood loss. Aside from the differences in surgical technique and implant usage, the timing of tourniquet release may also affect the amount of blood loss.

In our study, the tourniquet in the CON-TKA group was released before wound closure to ensure adequate hemostasis. This differs from other studies wherein the tourniquet was usually released after wound closure. Zhang et al. published a meta-analysis of RCTs which revealed that releasing the tourniquet before wound closure in TKA had higher total blood loss compared to releasing it after wound closure. Releasing the tourniquet before wound closure was believed to increase fibrinolytic activity, leading to prolonged bleeding before the wound pressure dressing was applied. However, they also concluded that releasing the tourniquet before wound closure decreased the risk of complications [22]. There are several approaches to limited tourniquet use in TKA. Some surgeons inflate the tourniquet before the incision and deflate it after cementation [23, 24], whereas others deflate it after the arthrotomy is completed [25]. Another variation involves inflating the tourniquet just before cementation and deflating it after wound closure is completed [9, 10]. These variations highlight the need for careful consideration and further review to determine the optimal timing for tourniquet inflation and deflation. Another possible reason behind the reduced blood loss with LIM-TKA is that it may help decrease the hidden blood loss associated with tourniquet use in TKA. Li et al. examined the impact of tourniquet use on hidden blood loss in TKA; using a tourniquet was associated with significantly greater hidden blood loss compared with not using one (median [range]: 701 [568–946] vs. 589 [458–775] mL) [26]. Similarly, a prospective study by Zhao et al. showed that full tourniquet use resulted in the highest mean hidden blood loss, followed by limited and no tourniquet use (1,135.53 ± 189.17 vs. 887.31 ± 171.51 vs. 654.33 ± 124.82 mL) [27].

Patients in the LIM-TKA group had better wound scores than those in the CON-TKA group. This aligns with most studies, which found higher wound complications when using a tourniquet in TKA. Olivecrona et al. found that higher tourniquet pressure and longer tourniquet time were associated with higher wound complications after TKA [2, 4]. Using a tourniquet reduces oxygen supply to the surgical site, and prolonged tourniquet time can cause further tissue ischemic injury, subsequently increasing postoperative wound complications [28]. Additionally, a meta-analysis by Wang et al. demonstrated that LIM-TKA was associated with a decreased risk of minor complications (e.g., wound complications) compared to CON-TKA [19]. None of the patients in this study had clinical findings of VTE after TKA. This may be because the overall incidence of VTE in modern TKA is low, with a reported incidence of 1.2–1.4% [29, 30]. Nevertheless, decreasing the tourniquet time during the TKA can improve patient recovery and reduce the length of hospital stay, since it has been associated with faster postoperative recovery, less pain, and improved quadriceps strength [5, 6, 27]. Our study demonstrated that the LIM-TKA group had a significantly shorter hospital stay compared with the CON-TKA group (5.6 vs. 7.2 days). Similarly, Xu et al. reported a hospital stay of 6.4 days for patients without a tourniquet during TKA, compared to 7.7 days for those with a tourniquet [31]. A meta-analysis by Han et al. also showed clear benefits of reduced tourniquet use during TKA, with hospital stays being shorter by an average of 0.59 days [7].

This study has some limitations. First, hidden blood loss was not measured, and this could have influenced the finding of reduced total blood loss in the LIM-TKA group. Instead, Hb levels were used to estimate total blood loss, since previous research suggests that the Hb-balance method is the most reliable way to estimate blood loss after TKA [32]. Second, the patients in our study were operated on by experienced surgeons, and thus our results may not be applicable to other cohorts.

Conclusions

Our findings demonstrate that the limited use of a tourniquet during TKA results in significantly lower total blood loss compared to the conventional use of a tourniquet. In addition, limited tourniquet use was also associated with fewer wound complications and a shorter hospital stay. Thus, limiting the tourniquet use in TKA is recommended to minimize tourniquet-related complications.