Personalized tourniquet pressure may be a better choice than uniform tourniquet pressure during total knee arthroplasty: a meta-analysis of randomized- controlled trials.

Background: Pneumatic tourniquets are widely used in total knee arthroplasty (TKA). Some surgeons prefer a uniform tourniquet ination pressure (UTIP)for all patients; others use personalized tourniquet ination pressures (PTIP) based on systolic blood pressure (SBP) and limb occlusion pressure(LOP). However, no consensus exists regarding the optimal mode of ination pressure during TKA. This review aimed to appraise if personalized tourniquet ination pressures are better than uniform tourniquet ination. Methods: The databases (Web of Science (cid:0) Embase (cid:0) PubMed (cid:0) Cochrane Controlled Trials Register, (cid:0) Cochrane Library (cid:0) Highwire (cid:0) CBM (cid:0) CNKI (cid:0) VIP (cid:0) Wanfang) were searched on March 2021 to systematically identify and screen the literature for randomized controlled trials (RCTs) involving PTIP and UTIP during total knee arthroplasty. Results: Thirteen randomized controlled trials, involving 1204 TKAs (1201 patients) were included in the systematic review. The meta-analysis identied a trend toward less Visual Analogue Scale (VAS) score at rest with PTIP group at one day (cid:0) P=0.002 (cid:0)(cid:0) 2-3day (cid:0) P=0.01 (cid:0) , and less VAS score at activity 1day (P (cid:0) 0.0001 ),2-3 days after the operation (P (cid:0) 0.00001 )and discharge (P (cid:0) 0.0001). No signicant difference was found between the groups in terms of VAS score at rest when discharge (P=1.0). We also found no signicant difference in terms of intraoperative blood loss (cid:0) P=0.48 (cid:0) , total blood loss (cid:0) P=0.15 (cid:0) , lower limb vein thrombosis (cid:0) P=0.42 (cid:0) and thigh bullae (cid:0) P=0.17 (cid:0) . However, in the PTIP group, we found a signicant higher Hospital for Special Surgery (HSS) score (cid:0) P=0.007 (cid:0) ,broader knee Range of motion(ROM) (cid:0) P=0.02 (cid:0) ,less rate of thigh ecchymosis (cid:0) P=0.00001 (cid:0) and shorter thigh circumference at one day(P=0.006), 2-3day (P=0.0005), and discharge(P=0.02). meta-analysis to compare the impact of PTIP with conventional UTIP during TKA


Background
Pneumatic tourniquets that are used in total knee arthroplasty (TKA) may lead to soft tissue damage, including the skin, vessels, muscles, nerves, and brinolytic activity due to unnecessarily excessive in ation pressure [1][2][3][4]. However, many orthopedic surgeons use it. A study of the American Association of Hip and Knee Surgeons found that approximately 95% of surgeons used tourniquets during TKA [5].
The tourniquet can provide a clear bloodless eld, which potentially reduces intraoperative blood loss, operative time and better prepares the cement-bone interface, despite the possible adverse effects associated with its use during total knee arthroplasty (TKA) [6]. The tourniquet use is almost indispensable in orthopedic practice. Although a lot of procedures employ the use of a tourniquet, there is still a lack of evidence-based guidelines of standard practice regarding optimal in ation pressures [7][8][9]. While some prefer a uniform tourniquet in ation pressure(UTIP) for all patients [10][11][12], others use personalized tourniquet in ation pressures (PTIP), which based on systolic blood pressure (SBP) [3,13,14] or limb occlusion pressure (LOP) . This study aimed to compare the effects of the PTIP with conventional UTIP on rehabilitation outcomes in TKA patients.

Methods
Our meta-analysis was registered on PROSPERO (International prospective register of systematic reviews), and the registration number was CRD42020168432. We assessed the quality of the included studies according to the items recommended in Cochrane Collaboration (Revman 5.3; http://handbook.cochrane.org/), and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines

Search strategy
We identi ed relevant randomized controlled trials involving PTIP or conventional UTIP in total knee arthroplasty in electronic databases, including PubMed, Web of Science Embase Cochrane Controlled Trials Register, Cochrane Library Highwire CBM CNKI VIP Wanfang database, up to March 2021. The keywords included "total knee arthroplasty," "total knee replacement," " tourniquet," "pressure" in conjunction with Boolean operators "AND" or "OR." Review Manager Software was used to perform our meta-analysis.

Inclusion criteria
The inclusion criteria were: 1. The intervention was PTIP, based on SBP or LOP in TKA; 2. The comparator was the UTIP based on surgeon experience;3. Randomized controlled trial studies; 4. The outcomes are intraoperative blood loss, total blood loss, Visual Analogue Scale (VAS) score, Hospital for Special Surgery (HSS) score, knee range of motion(ROM), thigh circumference, complication rates including lower limb vein thrombosis, thigh bullae, and thigh ecchymosis; 5. The follow-up rate was at least 80%. 6. At least one outcome was included in the study; The exclusion criteria were as follows:1. Observational studies;2. non-RCTs; 3. The included studies have insu cient outcome data.

Data extraction process
Two reviewers independently extracted the available data from each study. The primary data were based on the following: rst author, year of publication, country, number of TKAs and participants, age, gender, BMI, the primary indication for TKA, prosthesis, anesthesia, operation time, mean tourniquet time, mean in ation pressure, practices of tourniquet pressure, the time for loosening the tourniquet. The primary outcome consisted of Intraoperative blood loss, total blood loss, VAS score, HSS score, complications such as lower limb vein thrombosis, thigh bullae, and thigh ecchymosis. Secondary outcomes included knee ROM and thigh circumference.We resolved the disagreements by discussion to reach a consensus.

Quality Assessment
We used the Cochrane risk of bias tool to assess the risk of bias in the RCTs and determine whether biases might have affected the results.

Statistical Analysis
Review Manager Software for MAC (version 5.3) was used to perform the meta-analysis. The Q test and I 2 were used to evaluate the heterogeneity between studies. The random-effects model was in the place of the xed effects model for heterogeneity test, P values ≤.1 or I 2 ≥ 50%. The mean difference (MD) or standard mean difference (SMD) was used to assess continuous outcomes such as VAS, blood loss, HSS, knee ROM, and thigh circumference with a 95% con dence interval (CI). We used Relative risks with a 95% CI to assess dichotomous outcomes such as rate of lower limb vein thrombosis, thigh bullae, and thigh ecchymosis. We considered the results as a statistically signi cant difference when P values were less than 0.05

Search results
The detailed literature screening process is shown as the PRISMA ow diagram in Fig. 1. The literature search identi ed 489 citations. Of these, we removed 330 duplicates. Upon reviewing the titles and abstracts of the 159 remaining articles, we excluded 144 papers according to the inclusion criteria and retrieved the full text of 13 articles. Finally, we identi ed 1204 TKAs (1201 patients) assessed in 13 randomized controlled trials [15][16][17][18][19][20][21][22][23][24][25][26][27] We presented the detailed baseline characteristics in table 1 and tourniquet intervention information in Table 2. All the papers were published in English and Chinese between the years 2005 and 2021.

Risk of Bias assessment
The risk of bias summary and bias graph for RCTs is shown in Figs. 2 and 3. The correct randomization and su cient allocation concealment were adequately described in ten studies. The blinding of outcome assessment was described in thirteen studies, and the blinding of participants and personnel was described in three studies. Each study retained complete outcome data and avoided selective reporting. Other potential biases of all studies can't be ignored. Therefore, we rated them as having an unclear risk of other bias. As a result, the included studies' overall quality was considered adequate ( Fig. 2 and 3).

Discussion
Although clinical efforts and advances in tourniquet technology have resulted in the use of lower in ation pressures, there was no metaanalysis comparing the effects of PTIP with UTIP on rehabilitation outcomes and postoperative complications. Our meta-analysis is the rst meta-analysis to compare the impact of PTIP with conventional UTIP during TKA The current meta-analysis's main nding was that both PTIP and conventional UTIP ensure equal blood loss in total knee arthroplasty. No signi cant difference was observed between the groups in terms of rate of lower limb vein thrombosis, and thigh bullae. However, in patients using a tourniquet with PTIP, we found a signi cant reduction in postoperative pain, thigh circumference, rate of thigh ecchymosis, higher HSS and a better initial recovery of knee exion.
The present work analysis was not able to identify any differences between the two groups in the case of intraoperative blood loss and total blood loss. These ndings mean PTIP would provide a bloodless surgical eld comparable to conventional UTIP.
Immediate postoperative pain relief following TKA is crucial in facilitating early recovery. We were able to detect a signi cantly lower pain intensity within three days after operation both at rest and during mobilization in patients with PTIP group. We also identi ed a signi cantly lower pain intensity at the activity when patients were at discharge; however, we couldn't identify any difference of pain intensity at rest when patients left the hospital. An explanation for the increased pain in the early postoperative period with conventional uniform pressure group could be direct higher pressure on the surrounding soft tissues due to the tourniquet. In our study, the pressure of the PTIP is lower than the conventional UTIP group. Worland et al. [28] showed an essential correlation between tourniquet pressure and thigh pain in the immediate postoperative period. We thought that the PTIP lowers pain levels while increasing patients' adherence to rehabilitation, which resulted in earlier restoration in functions.
In patients using a tourniquet with PTIP, we found a signi cant reduction in thigh circumference. We think the reason may be due to less stress on the thigh muscles in the PTIP group.
Knee exion ROM is often used to evaluate short-term effectiveness. Besides, discharge from the hospital is dependent on the mobility of patients following TKA. The PTIP group documented a signi cantly higher postoperative ROM. It may be related to using a conventional UTIP with higher tourniquet pressure that causes some temporary loss of exibility in the tight thigh muscles. The PTIP group also reveal a higher HSS score. The reason may be less pain, more knee ROM in the PTIP group.
As for complications, all studies did not experience major signi cant complications such as symptomatic PE, thigh necrosis, nerve palsy, or delayed rehabilitation. We found no signi cant difference between groups regarding the rate of lower limb vein thrombosis and thigh bullae.
However, in patients using a tourniquet with personalized tourniquet in ation, we found a signi cant reduction in the quality of thigh ecchymosis. It is possible to achieve functional bene ts with decreasing some complications related to the tourniquet and to have the advantages as with the personalized tourniquet application.
The pressure for safe tourniquet use remains controversial, and no strict guidelines have been established. Most of the orthopedic surgeons routinely apply xed tourniquet pressure in TKA based on individual experiences. It was very convenient to choose the xed pressure value. However, it did not take patients' actual individual situation into account, so the selected pressure values were mostly on the high side. Some researchers suggested that upper limb pressure in an adult is 250 to 300 mm Hg, and lower limb pressure is 350 to 500 mm Hg [29]. A higher tourniquet pressure ensures the reliable function of the tourniquet; however, it may lead to a higher incidence of complications. The pressures higher than 350mmHg on the lower limbs increase neuropraxia and compression [8,13]. While a lower tourniquet pressure is safer than higher pressure, it may not provide a bloodless operative eld. Optimal tourniquet pressure should be determined to balance safety and e cacy. In recent years, some investigators proposed that the tourniquet pressure setting should be personalized. Compression pressure on a pneumatic tourniquet's limb artery wall is different due to different physiological functions, such as systolic blood pressure, age, weight, limb circumference size, and muscle tissue thickness.
Setting the tourniquet pressure based on SBP or LOP allows us to use a personalized tourniquet pressure in each patient and is useful in optimizing tourniquet cuff pressures. The rationale behind in ating the tourniquet beyond the SBP, allowing a certain amount of safety margin, which added to the SBP ranges widely, from 100 to 250 mmHg in the literature [30,31]. LOP is the term that mean the lowest tourniquet pressure is required to cease the arterial blood ow into the extremity distal to the cuff. LOP can be determined automatically or manually by slow cuff in ation to pulse cessation with diagnostic equipment such as Doppler owmeter or pulse oximeter [32][33][34][35]. Now, modern tourniquet systems permit an automated LOP estimation through a probe incorporated in the tourniquet system itself [4].
Following an analysis of the current literature, this work demonstrated a relative predominance of the advantages when a tourniquet is used with the personalized application. However, the present meta-analysis has several limitations: First, there are two methods for personalized tourniquets, including SBP and LOP. Because of the limited data, we were not able to evaluate one of them separately. We performed a sensitivity analysis on them and found that the conclusion is stable when removing one method. Second, the studies' comparability was complicated through the different measurement methods and follow-up examination time points; however, we have tried our best to evaluate results based on time points. Third, the tourniquet time, the time for loosening the tourniquet, and the cuff pressure used were also not uniform (see Tab. 1). Fourth, there are no worldwide uniform guidelines for performing total knee arthroplasty. Different surgical techniques (such as the selection of approach, methods of anesthesia, drainage patterns hemostasis, and anticoagulation regimens) were used in the individual studies.

Conclusion
In conclusion, personalized tourniquet in ation pressure provides a bloodless surgical eld comparable to that of a conventional uniformed method with less pain intensity, thigh circumference, rate of thigh ecchymosis, higher HSS and better initial recovery knee exion in total knee arthroplasty. Therefore, we recommend using personalized tourniquet in ation pressure during TKA. However, due to the limited comparability of the studies available, more longer follow-up period and overall higher quality RCTs are needed to con rm the present metaanalysis results.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request. The detailed baseline characteristics information, including the number of TKAs, age, gender, BMI, diagnosis, prothesis, anesthesia of two groups. After the application of a bandage to the limb The tourniquet intervention information including the operation time, Mean tourniquet time, mean in ation pressure, practices of tourniquet pressure, the time for loosening the tourniquet of two groups. Figure 1 The search results and selection procedure. The literature search identi ed 489 citations. Of these, we removed 330 duplicates. Upon reviewing the titles and abstracts of the 159 remaining articles, we excluded 144 papers according to the inclusion criteria and retrieved the full text of 13 articles. Finally, we identi ed 1204 TKAs (1201 patients) assessed in 13 randomized controlled trials.

Figure 2
The risk of bias summary for RCTs +:no bias; -:bias; ?:bias unknown. The correct randomization and su cient allocation concealment were adequately described in ten studies. The blinding of outcome assessment was described in 13 studies, and the blinding of participants and personnel was described in three studies. Each study retained complete outcome data and avoided selective reporting. Other potential biases of all studies can't be ignored. Therefore, we rated them as having an unclear risk of other bias. As a result, the included studies' overall quality was considered adequate Page 12/15

Figure 3
The risk of bias graph. The overall quality of the studies was considered adequate.