Posterior cruciate ligament (PCL) reconstruction is associated with a higher complication rate (20.1%) compared to other knee procedures, along with specific risks such as popliteal artery injury1,2. In knee arthroscopy, surgeons commonly utilize a tourniquet to mitigate blood loss, enhance visibility, and decrease operative time3,4. Many surgeons assert that tourniquet usage has no discernible impact on complication rates or postoperative rehabilitation. Moreover, they believe it enhances the surgical view, resulting in reduced operative times, and leading to an increased frequency of tourniquet application4. Recently, studies reported that tourniquet use during arthroscopic ACL reconstruction is associated with adverse postoperative outcomes, including thigh muscle injury, increased postoperative pain and drain output, and tourniquet-related complications5,6,7. However, no studies have specifically evaluated tourniquet use in PCL reconstruction, and the recent expert consensus and systematic review on management of posterior cruciate ligament injuries also does not specify the effectiveness and risks of using tourniquets in PCL reconstruction8,9. Consequently, we aimed to determine the necessity of tourniquet use in PCL reconstruction procedures, hypothesizing that non-tourniquet PCL reconstruction would provide comparable surgical efficiency while reducing postoperative joint swelling and tourniquet-related complications.

Materials and methods

Patients

This study received approval from the ethics review committee of The Second Xiangya Hospital, and written informed consent was acquired from all participants. Participants were selected using the following inclusion criteria: (1) aged older than 18 years; (2) underwent unilateral isolated PCL reconstruction in the day-surgery unit under general anesthesia between March 2016 and July 2022; (3) complete clinical records available for analysis. Exclusion criteria included osteoarthritis, meniscus injury requiring repair, history of peripheral neuropathy, pregnancy, lumbar radiculopathy, or prior knee surgery on the affected or contralateral knee. Patients were systematically categorized into tourniquet and non-tourniquet groups according to their surgery dates. A flow diagram illustrating patient selection and grouping is presented as Fig.Ā 1. Nonsteroidal anti-inflammatory drugs were avoided for the initial two weeks post-surgery, and patients were typically mobilized and discharged on the fourth postoperative day.

Fig. 1
Fig. 1
Full size image

Flowchart illustrating patient selection, exclusion, and grouping process for the study.

Surgeries

All cases were isolated PCL reconstructions (PCLRs) without concomitant surgeries. All PCLRs were performed by the same experienced surgeon (Dr. Wu, >ā€‰10 years of PCL reconstruction experience), with the patients under general anesthesia. Patients utilizing a tourniquet were categorized into the tourniquet group, where a thigh tourniquet inflated to 300 mmHg was applied to the operated leg. In cases exceeding 90Ā min, the tourniquet was deflated and reinflated as necessary after a 10-minute interval. Routine arthroscopy was performed, and all patients received multi-modal analgesia, including an intra-articular cocktail injection (20Ā ml ropivacaine 0.25% and 1Ā ml Diprospan) and a femoral nerve block with 20Ā ml ropivacaine 0.25%. The procedure was done in the standard anatomical single-bundle PCL reconstruction technique, preserving the native PCL remnants, using autologous tendons (semitendinosus and gracilis tendons) and fixed with two biodegradable interference screws. The operative time was recorded for all patients. Considering that there was minimal bleeding throughout and subsequent to the operation and it would not affect the recovery of joint function, no drainage tube was inserted after the surgery. Postoperative analgesia began immediately after skin closure, consisting of a daily 50Ā mg flurbiprofen axetil injection for three days.

Clinical evaluation

Patient information, including age, BMI, sex, operative time, arthroscopy view, and complications (infection, VTE, wound complications, bruising, or blister in the mid-thigh) were assessed, along with total bleeding from suction and drainage. The arthroscopy view was classified as satisfactory (excellent) or unsatisfactory (good, fair, or poor) based on a combined score for visibility and ease of procedure throughout the operation, following criteria established by David et al. 10, which are shown in Supplementary Table 1. The same surgeon graded the operative view. Knee swelling was assessed with the change of the knee girth (measured at the mid patella) and patella floating test at 24Ā h postoperatively. Postoperative parameters included pain, range of motion, and analgesia consumption were recorded. VAS pain scores, ranging from ā€œNo pain at allā€ (0) to ā€œWorst pain imaginableā€ (10), were measured post-surgery. The assessment of VAS was conducted without knowledge of tourniquet inflation status, and additional analgesia was recorded as needed.

Postoperative rehabilitation

Both groups adhered to an identical rehabilitation program, focusing on quadriceps function restoration and complete extension. Accelerated rehabilitation commenced immediately post-operation, with full weight-bearing encouraged after two weeks. Compression and cooling systems were employed until visible swelling reduction. A brace facilitated full extension in the initial two weeks, followed by gradual flexion to 90 degrees in four weeks. Unrestricted physical activity, including pivotal sports like skiing or football, was recommended after at least six months.

Statistical methodology

Statistical analyses were conducted with SPSS version 24 software (IBM Corp., Armonk, NY, USA). Descriptive statistics for clinical and demographic features were presented as the mean and standard deviation for continuous variables. The Studentā€™s t-test was used to compare continuous variables that followed a normal distribution, whereas the Wilcoxon rank-sum test evaluated differences in non-normally distributed variables. Fisherā€™s exact test and Pearsonā€™s Chi-Squared test determined correlations between categorical variables. P-valuesā€‰<ā€‰0.05 were considered significant.

Ethics approval

The study was conducted in compliance with the principles outlined in the Declaration of Helsinki and received approval from the Human Ethics Committee for Medical Research at The Second Xiangya Hospital of Central South University. The clinical trial number is not applicable. Due to the retrospective nature of this study, the formal ethics approval ID was not required by our institutional policy at the time of data collection. However, all data were anonymized and handled in compliance with the Declaration of Helsinki. All procedures adhered to pertinent guidelines and regulations. Informed consent was obtained from all participants.

Results

Out of the 108 patients, 55 patients underwent tourniquet inflation between March 2016 and October 2019, while the remaining 53 patients did not between November 2019 and July 2022. The baseline characteristics, duration of operation, and bleeding for these patients are listed in TableĀ 1. Demographic profiles revealed no significant differences between the two groups. The operative time averaged 73ā€‰Ā±ā€‰18Ā min in the tourniquet group and 74ā€‰Ā±ā€‰20Ā min in the non-tourniquet group, with no statistically significant difference observed (pā€‰=ā€‰0.87). In terms of bleeding volume, the tourniquet group showed 10ā€‰Ā±ā€‰4Ā ml, compared to 15ā€‰Ā±ā€‰10Ā ml in the non-tourniquet group. However, no significant difference in total bleeding between the two groups was found (pā€‰=ā€‰0.57) The detailed characteristics of the patients are shown in TableĀ 1.

Table 1 Clinical characteristics of the patients with or without tourniquet.
Full size table

Regarding the operative view, the tourniquet group had 54 cases rated as excellent and 1 as good, while the non-tourniquet group had 49 excellent and 4 good cases. No significant difference was found between the two groups in this regard (pā€‰=ā€‰0.66). The detailed results are shown in TableĀ 2. VAS pain scores, measured on three occasions, did not reveal any significant differences between the groups (Fig.Ā 2). Postoperative analgesic consumption analysis indicated no analgesic requirements other than the routine 50Ā mg flurbiprofen axetil injection (data not presented).

Table 2 Assessment of arthroscopic visibility.
Full size table
Fig. 2
Fig. 2
Full size image

Pain score of tourniquet and non-tourniquet group postoperatively.

Concerning the range of motion, the average postoperative flexion in the second week was 30 degrees in the tourniquet group and 28 degrees in the non-tourniquet group, with no significant difference between the two groups (Fig.Ā 3). Both groups experienced no wound complications, VTE, or postoperative infections. However, the rate of joint swelling was increased in the tourniquet group (9.1%) compared with the non-tourniquet group (0%) (pā€‰=ā€‰0.01). The increase of the knee girth at 24Ā h postoperatively averaged 2.24ā€‰Ā±ā€‰0.60Ā cm in the tourniquet group and 1.33ā€‰Ā±ā€‰0.39Ā cm in the non-tourniquet group, with a statistically significant difference observed (pā€‰<ā€‰0.001). We conducted post hoc power analyses to confirm ample statistical power. For increase in the knee girth, powerā€‰>ā€‰99.9% (dā€‰=ā€‰1.79). For knee swelling incidence, Powerā€‰=ā€‰85.7%. This exceeds the 80% threshold recommended by Cohen, ensuring reliable detection of clinically meaningful effects. We also estimate the required sample size. Increase in the knee girth was used as the primary outcome measure. With a significance level of Ī±ā€‰=ā€‰0.05 and power (1ā€‰āˆ’ā€‰Ī²)ā€‰=ā€‰0.90, the results indicated that a minimum of 6 patients per group (12 total) would be sufficient to achieve statistical power greater than 0.90.

Fig. 3
Fig. 3
Full size image

Range of motion for tourniquet and non-tourniquet group postoperatively.

There were three cases of bruising in the mid-thigh and two cases of blistering in the mid-thigh in the tourniquet group, compared to none in the non-tourniquet group. The detailed results are shown in TableĀ 3. At a mean follow-up time of 1 year, no patients in either group required repeat arthroscopy or further imaging.

Table 3 Complications of the patients with or without tourniquet.
Full size table

Discussion

Our study demonstrates that PCL reconstruction without tourniquet use significantly reduces postoperative joint swelling and thigh complications while maintaining equivalent operative times, visibility, and functional outcomes. To our knowledge, this is the first study specifically evaluating tourniquet use in isolated PCL reconstruction. Our findings challenge the routine use of tourniquets in complex knee procedures and provide evidence that omitting tourniquets reduces joint swelling without compromising surgery. This directly impacts clinical practice by improving early rehabilitation comfort.

As arthroscopic techniques have evolved into the preferred approach for PCLR, the ongoing debate regarding the necessity of tourniquet use in these intricate and more time-consuming procedures persists. The tourniquet has been commonly used in knee arthroscopic surgery to minimize intraoperative bleeding within the joint, thereby creating a bloodless surgical field, enhancing visualization during the operation, and reducing the overall duration of the surgery3,4. However, using tourniquets is related to possible complications, including vascular damage, neuropathies, muscle weakness and atrophy, delayed recovery of function postoperatively, abnormalities in electromyogram (EMG), and an elevated risk of deep vein thrombosis10,11,12. Additionally, alterations in acid-base balance, acute pulmonary edema, and cardiac arrest have been reported13,14. Despite longer operation times and an increased demand for tourniquet application, the complication rate increases15. Studies report a 21.9% incidence of deep vein thrombosis (DVT) after arthroscopic PCL reconstruction without the use of anticoagulant drugs, a deviation from the standard practice in arthroscopic surgery2. In China, there seems to be a prevailing reluctance to perform certain orthopedic procedures, including arthroscopy, without a tourniquet. This enduring practice is fueled by the desire for a clear operative view, reduced bleeding, and the misconception that tourniquets are harmless. Many previous studies concerning tourniquet use in arthroscopic knee surgery have evaluated its effects following meniscectomy and ACL reconstruction. However, there are few studies focusing on the effects of tourniquet use during PCL reconstruction16. Recent meta-analyses discourage tourniquet use in knee arthroscopic surgery, although not specifically addressing PCL reconstruction17. Consequently, our study was to assess the efficacy and safety of the employment of a tourniquet in PCL reconstruction surgery.

Results from our study reveal no significant differences between the groups in terms of operation time, intraoperative bleeding, visibility, pain scores, range of motion, and wound complications. This indicates that PCL arthroscopy can be performed with equal effectiveness whether or not a tourniquet is used. Similar to other studies, Johnson18 found that operative visualization was not significantly compromised without the use of a tourniquet in the majority of arthroscopic surgeries (mostly meniscal surgeries). Reda19 indicated that the operative view was rated as excellent for all participants, regardless of whether they were in the tourniquet or non-tourniquet group, and highlighted that there was no significant difference in the average operative time between the two groups. Additionally, a previous meta-analysis indicated that there was no notable difference in both visualization quality and operative time between the two groups17. Interestingly, certain procedures, such as synovial biopsy and meniscal repair, are reported to be better executed without a tourniquet, indicating potential advantages for non-tourniquet surgery, especially in cases involving concurrent PCL injury and meniscal tear20.

Our study explored the complex relationship between tourniquet application and pain. Although no postoperative pain score differences were observed between the groups, the tourniquet group experienced slightly higher scores at 6, 12, and 24 h postoperatively, with a notable increase in both groups at 48 h. The mechanism of tourniquet pain is still unclear, possibly resulting from local pressure, ischemic or reperfusion injuries to muscular and neural tissues21,22, as well as the activation of small, unmyelinated C-fibers23,24. In contrast to our results, several studies25,26,27 found that individuals who underwent surgery without the tourniquet experienced less postoperative pain and less use of analgesics, which may be attributed to the relatively short duration of tourniquet application in our study (73 Ā± 18 min) and the multimodal analgesic approach, including femoral nerve block and intra-articular injection of ropivacaine and diprospan. Interestingly, the pain in the tourniquet group was localized in the upper thigh, possibly attributable to tourniquet application. To address postoperative pain, we routinely administered flurbiprofen axetil for three days.

The knee swelling rate was 9.1% in the tourniquet group as we observed and it is higher than the tourniquet group. The increase in knee circumference at 24 h postoperatively in the tourniquet group was significantly greater than in the non-tourniquet group. Such results are consistent with a previous study which indicates that the extent of joint swelling in the non-tourniquet group was decreased compared with the tourniquet group in knee surgery17. Knee swelling might be attributed to blood loss postoperatively, reactive hyperemia, and postanoxic edema. Some studies indicated that using a tourniquet may increase the postoperative hidden blood loss and drainage volume28,29,30.

In the tourniquet group, due to squeezing the thigh, a few cases exhibited bruising and blisters in the mid-thigh, coinciding with reported pain cases. While both groups demonstrated pain relief within 96Ā h, skin injuries can adversely affect the patient experience and increase the risk of infection. Despite no significant difference in the range of motion, intraoperative bleeding in the non-tourniquet group appeared slightly more, primarily from hamstring harvest procedures. However, this minimal bleeding is unlikely to impede patient recovery.

This study has some limitations. First, there is an inaccuracy in the evaluation of total perioperative blood loss and postoperative blood loss in our study. In the study of tourniquets in ACL reconstructions (ACLRs), Masaki Nagashima calculated the total perioperative blood loss from the change in hemoglobin between that preoperatively and on postoperative day (POD) 1. Masaki Nagashima found total blood loss was significantly higher in the non-tourniquet group(339 Ā± 216 mL)than in the tourniquet group (258 Ā± 199 mL)31. Since the patient was discharged early after surgery and hemoglobin was not reexamined timely, we could not accurately assess the total perioperative blood loss. Hiroshi Nakayama15 assessed the total postoperative blood loss by the fluid collected via the suction drain and the total postoperative blood loss averaged 133.6 Ā± 62.4 ml (range60-250 ml) and 85.3 Ā± 47.3 ml (range20-190 ml) in the tourniquet and non-tourniquet groups. There are few studies on whether using drains after PCLRs is necessary. However, there is still debate over the use of drains following other knee surgery. Some previous studies suggested not using a drain following anterior cruciate ligament reconstruction32. Kęska R showed that primary ACLRs without using drainage did not adversely affect the recovery of patients33. Also, some studies reported increased pain in the drained group after ACLRs and removal of the drain is uncomfortable and carries theoretical and avoidable risks34. Some studies have shown that it is feasible to forgo wound drainage in simple, primary, tourniquet-free total knee arthroplasty (TKA)35,36. Primary TKA without using drainage could reduce local inflammation, make recovery faster, and improve early knee function37. With the increased cost, time, and need for drain removal, using the drain after TKA is likely an unnecessary intervention38. Considering the above research results, economic cost, and clinical experience, we decided not to use drains following PCLRs.

Second, due to its retrospective design and a relatively low level of evidence resulting from the limited number of PCL reconstruction cases. Third, the single-center design may affect generalizability, and long-term functional outcomes (>ā€‰1 year) were not assessed. While we posit that PCL reconstruction without a tourniquet is advantageous, larger-scale, multicenter, and high-quality randomized controlled studies are warranted to confirm our results.

In conclusion, considering equivalent operation times, bleeding, arthroscopic visibility, pain scores, range of motion, and less knee swelling, we advocate discontinuing routine tourniquet use in PCL reconstruction. This approach significantly reduces swelling and local complications while maintaining surgical efficacy, aligning with modern minimally invasive principles.