Comparison between arthroscopic surgery and open reduction internal fixation in patients with posterior cruciate ligament tibial avulsion fracture: a systematic review with meta-analysis Authors: Bo Yu, Yanan Liao Published: 2026 ABSTRACT AbstractBackgroundStudies comparing arthroscopic surgery with open reduction internal fixation (ORIF) for posterior cruciate ligament (PCL) tibial avulsion fracture have exploded in recent years; however, the results are not consistent. This meta-analysis aimed to summarize the efficacy and safety profile of arthroscopic surgery and ORIF in patients with PCL tibial avulsion fracture.MethodsRelevant studies were searched across the Web of Science, PubMed, EMBASE, Wan Fang, CNKI, and VIP databases. The standardized mean difference (SMD) or odds ratio (OR) with 95% confidence interval (CI) was pooled for analyses.ResultsTwenty-seven studies were included in this meta-analysis. Compared with those in the ORIF group, the incision length (SMD: -4.701, 95%CI: -5.717~-3.686), perioperative bleeding (SMD: -3.304, 95%CI: -3.834~-2.774), postoperative drainage volume (SMD: -1.071, 95%CI: -1.870~-0.273), time to first ambulation (SMD: -2.294, 95%CI: -4.127~-0.460), and hospital stays (SMD: -1.918, 95%CI: -2.594~-1.242) were lower, but the operation duration (SMD: 1.998, 95%CI: 0.595 ~ 3.401) was greater in the arthroscopy group. After treatment, the Lysholm score (SMD: 1.424, 95%CI: 0.858 ~ 2.837), International Knee Documentation Committee score (SMD: 1.848, 95%CI: 0.858 ~ 2.837), range of motion (SMD: 0.527, 95%CI: 0.001 ~ 1.053) were greater in the arthroscopy group than in the ORIF group, but no statistically significant difference was observed in the pain score (SMD: -0.250, 95%CI: -0.664 ~ 0.165), and posterior drawer test negative rate (OR: 1.040, 95%CI: 0.460 ~ 2.351). Moreover, the incidence of postoperative complications (OR: 0.458, 95%CI: 0.149 ~ 1.411) was not significantly different between the arthroscopy group and the ORIF group.ConclusionArthroscopic surgery appears to provide better functional outcomes and a comparable risk of complications than ORIF does to some extent, although longer operative time and study heterogeneity warrant caution, for PCL tibial avulsion fracture.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12891-026-09631-1.Keywords: Posterior cruciate ligament, Tibial avulsion fracture, Arthroscopic surgery, Open reduction internal fixation, Meta-analysis Introduction Posterior cruciate ligament (PCL) injury is a relatively rare type of knee injury that accounts for approximately 3%-23% of all knee injuries [1, 2]. PCL injury usually results in tearing of the ligament’s collagen fibers and, in some cases, a tibial avulsion fracture [3, 4]. PCL tibial avulsion fracture is commonly caused by high-energy trauma, such as traffic accidents and sports collisions, leading to decreased knee stability [5]. Nonoperative treatment can be applied for PCL tibial avulsion fracture when knee joints are stable and there is no apparent fracture displacement; however, most cases involve unstable knee joints or apparent fracture displacement, and open reduction internal fixation (ORIF) and arthroscopic surgery are preferred treatment options [5, 6]. Traditionally, ORIF is the preferred gold-standard treatment for PCL tibial avulsion fracture [7]. Nevertheless, progress in arthroscopic methods and tools has led to a growing preference for arthroscopic procedures, offering enhanced visualization of the avulsion fracture and the ability to address associated injuries [8, 9]. The number of studies comparing arthroscopic surgery with ORIF for the treatment of PCL tibial avulsion fracture has increased over the past three years [10–24]. However, the findings are not consistent. Most studies reported that arthroscopic surgery is superior to ORIF in terms of a shorter incision length, less perioperative bleeding and complications, fewer hospital stays, and better knee function recovery [10, 11, 18, 22, 24], but some studies revealed that arthroscopic surgery had equal or lower efficacy than ORIF [19, 20]; however, a confirmative conclusion was lacking. This meta-analysis aimed to summarize the efficacy and safety profile of arthroscopic surgery versus ORIF in treating patients with PCL tibial avulsion fracture. Methods Search strategy Relevant studies were searched across the Web of Science, PubMed, EMBASE, Wan Fang, CNKI, and VIP databases. The search was conducted in October 2025. The search strategy included the following terms: ‘posterior cruciate ligament’, ‘PCL’, ‘tibia’, ‘tibial’, ‘avulsion fracture’, ‘arthroscopy’, ‘arthroscopic surgery’, ‘arthroscopic suture’, ‘arthroscopic reduction’, ‘arthroscopic reduction and internal fixation’, ‘ARIF’, ‘open reduction’, ‘open surgery’, ‘open reduction internal fixation’, and ‘ORIF’. “AND” and “OR” were used to optimize the search results. In addition, the references of the included studies were manually screened to identify potential additional studies. Eligibility criteria Studies were included if they (1) reported patients with PCL tibial avulsion fracture; (2) reported outcomes on the efficacy or safety of comparisons between arthroscopy and ORIF in treating PCL tibial avulsion fracture; (3) were published in English or Chinese; or (4) reported relevant target outcomes. Studies were excluded if they were (1) duplicate publications; (2) reviews, meta-analyses, case reports, study protocols, or animal studies; or (3) reported patients with knee injuries other than PCL tibial avulsion fracture. Two researchers independently screened the studies according to the eligibility criteria outlined above. Any discrepancies were resolved through discussion until a consensus was reached. Quality assessment and data extraction Two researchers independently completed the quality assessment and data extraction of the included studies. The Cochrane Risk of Bias (ROB) Tool 2.0 was utilized to evaluate the quality of randomized controlled trial (RCT), assessing bias across domains such as randomization, deviations from intended interventions, missing outcome data, measurement of outcomes, and selection of reported results [25]. For the cohort study, the Newcastle‒Ottawa Scale (NOS) was used, with a focus on selection, comparability, and outcome assessment [26]. The two researchers used the same template for data extraction, which included the study characteristics (first author, publication year, study type, sample size), patient demographics (age and sex ratio), injury details (mechanism of injury and side of injury), fixation method, and follow-up period. In addition, outcomes concerning the efficacy or safety of arthroscopy and ORIF were extracted. After all relevant data were extracted, indicators reported in ≥ 3 studies were included in the pooled meta-analysis, whereas those reported in < 3 studies were qualitatively synthesized. For studies that reported multiple independent subgroups, we calculated an overall incidence rate by combining the number of events and the total number of individuals in each subgroup for meta-analysis. Continuous variables were combined via the following formula: ; and then used for meta-analysis (Ni represents the sample size in subgroup i, and SDi represents the standard deviation of the mean in subgroup i). The final indicators that could be used for meta-analysis included the perioperative period, postoperative function, recovery evaluation, and postoperative complications. Any divergences between researchers were resolved by consensus. Statistical analysis R (ver.4.3.2) software was used for data analysis. The standardized (Std) mean difference (SMD) or odds ratio (OR) with the corresponding 95% confidence interval (CI) was used to pool continuous or categorical variables, respectively. Heterogeneity was assessed using the I² statistic; values > 50% indicated heterogeneity, warranting a random-effects model. Conversely, a fixed-effects model was employed when the I² value was 50% or lower. Subgroup analyses were performed to explore potential sources of heterogeneity. Publication bias was assessed via Peter’s test and Egger’s test for continuous and categorical variables, respectively. If there was significant publication bias (P value of Peter’s or Egger’s test less than 0.05), a trim-and-fill method was employed for bias correction. Sensitivity analysis was performed to assess the robustness of the findings by omitting studies one by one. P < 0.05 indicated statistical significance. Results Study flow A total of 257 studies were initially identified through database searching, 98 duplicate publications were excluded, and 159 studies remained. After being screened through title and abstract reading, 115 studies were excluded, leaving 44 studies. Seventeen studies were excluded after full-text screening. Finally, a total of 27 studies comparing the efficacy or safety of arthroscopy versus ORIF in the treatment of PCL tibial avulsion fracture were included in this meta-analysis [10–24, 27–38] (Fig. 1). Study flow for literature searching and screening Study information and quality assessment Among the 27 included studies, 8 were RCTs, and the other 19 were cohort studies; the basic information for the included studies is shown in Table 1, and the fixation methods and follow-up periods are shown in Supplementary Table 1. Table 1Information of included studiesStudy ID #Study typeCountrySample size (n)Age range (years) and Sex ratio (Male/Female)Mechanism of injury (number of cases)Side of injury (Left/Right)Arthroscopy groupORIF groupArthroscopy groupORIF groupArthroscopy groupORIF groupArthroscopy groupORIF groupZhao ZL (2014) [27]RCTChina2323Age: All: 15.0–65.0; Sex: All: 25/21All: Traffic accident: 32; Falling objects: 10; Sports accident: 4NRNRSu JB (2017) [28]Cohort studyChina4040Age: 19.0–55.0; Sex: 23/17Age: 19.0–56.0; Sex: 24/16NRNRNRNRLai XY (2018) [29]Cohort studyChina1314Age: 15.0–46.0; Sex: 9/4Age: 14.0–52.0; Sex: 9/5All: Traffic accident: 28; Sports accident: 7; Others: 3NRNRZhang YP (2018) [30]Cohort studyChina2828Age: 21.0–59.0; Sex: 18/10Age: 22.0–59.0; Sex: 19/9Traffic accident: 14; Sports accident: 5; Fell from a height: 5; Falling objects: 4Traffic accident: 13; Sports accident: 5; Fell from a height: 6; Falling objects: 415/1316/12Lin YC (2019) [31]RCTChina1822Mean age: 29.4; Sex: 10/8Mean age: 31.7; Sex: 12/10Traffic accident: 10; Sports accident: 6; Others: 2Traffic accident: 12; Sports accident: 7; Others: 38/1013/9Wang GF (2019) [32]RCTChina3637Age: 17.0–48.0; Sex: 20/16Age: 18.0–50.0; Sex: 22/15Falling objects: 3; Fall: 7; Traffic accident: 11; Sports accident: 15Falling objects: 4; Fall: 6; Traffic accident: 11; Sports accident: 1617/1921/16Wu L (2019) [33]RCTChina2120Age: All: 16.0–56.0; Sex: All: 26/15All: Fall: 37; Traffic accident: 4NRNRFeng YS (2020) [34]RCTChina2630Age: All: 18.0–55.0; Sex: All༚35/21All: Sports accident: 39; Traffic accident: 17NRNRLu DL (2020) [35]Cohort studyChina2727Age: 16.0–66.0; Sex: 15/12Age: 15.0–65.0; Sex: 14/13NRNRNRNRXu HY (2021) [36]Cohort studyChina3232Age: 22.0–49.0; Sex: 20/12Age: 21.0–56.0; Sex: 22/10Traffic accident: 14; Sports accident: 10; Others: 8Traffic accident: 16; Sports accident: 10; Others: 618/1416/16Yuan YF (2021) [37]Cohort studyChina2222Age: 19.0–63.0; Sex: 11/11Age: 20.0–64.0; Sex: 12/10NRNR10/1211/11Deng SC (2022) [10]RCTChina2626Age: 22.0–57.0; Sex: 15/11Age: 23.0–50.0; Sex: 14/12Traffic accident: 11; Sports accident: 9; Others: 6Traffic accident: 13; Sports accident: 8; Others: 5NRNRZhang MT (2022) [11]Cohort studyChina2725Age: 19.0–52.0; Sex: 16/11Age: 18.0–54.0; Sex: 14/11Traffic accident: 13; Sports accident: 9; Falling objects: 5Traffic accident: 11; Sports accident: 8; Falling objects: 69/1810/15Zhang Y (2022) [12]RCTChina100100Age: 22.0–58.0; Sex: 56/44Age: 21.0–57.0; Sex: 55/45NRNRNRNRDu RZ (2023) [13]Cohort studyChina3030Age: NR; Sex: 19/11Age: NR; Sex: 18/12Traffic accident: 12; Sports accident: 10; Fall: 8Traffic accident: 10; Sports accident: 12; Fall: 8NRNRHe J (2023) [14]Cohort studyChina2828Age: 23.0–67.0; Sex: 13/15Age: 24.0–67.0; Sex: 14/14Sprain: 9; Traffic accident: 9; Sports accident: 10Sprain: 10; Traffic accident:9; Sports accident: 9NRNRLi JL (2023) [15]RCTChina3333Age: 22.0–48.0; Sex: 20/13Age: 21.0-50.1; Sex: 22/11NRNRNRNRPan JB (2023) [16]Cohort studyChina2020Age: 26.0–46.0; Sex: 12/8Age: 25.0–48.0; Sex: 13/7NRNR6/147/13Pei YR (2023) [17]Cohort studyChina2020Age: 16.0–56.0; Sex: 11/9Age: 19.0–51.0; Sex: 10/10Traffic accident: 11; Falling objects: 3; Sports accident: 6Traffic accident: 9; Falling objects: 6; Sports accident: 5NRNRSun Y (2023) [18]RCTChina3030Age: 28.0–59.0; Sex: 16/14Age: 31.0–57.0; Sex: 18/12NRNR15/1516/14Wang HJ (2023) [19]Cohort studyChina1731Age: 10.0–51.0; Sex: 8/9Age: 12.0–55.0; Sex: 16/15Sports accident: 5; Fall: 5; Traffic accident: 7Sports accident: 12; Fall: 10: Traffic accident: 9NRNRDeng Y (2024) [20]Cohort studyChina1923Age: All: 19.0–57.0; Sex: All:24/18All: Sports accident: 12; Fall: 22; Traffic accident: 8All: 16/26Guo SY (2024) [21]Cohort studyChina4642Mean age: 41.6; Sex: 26/20Mean age: 43.1; Sex: 22/20Traffic accident: 16; Sports accident: 10; Fell from a height: 9; Others: 11Traffic accident: 12; Sports accident: 11; Fell from a height: 8; Others: 11NRNRLi J (2024) [22]Cohort studyChina3636Age: 24.0–66.0; Sex: 20/16Age: 23.0–65.0; Sex: 19/17Sports accident: 10; Traffic accident: 21; Fall: 5Sports accident: 9; Traffic accident: 23; Fall: 421/1520/16Wu R (2024) [23]Cohort studyChina1820Age: 24.0–70.0; Sex: 11/7Age: 26.0–74.0; Sex: 16/4Traffic accident: 9; Sports accident: 7; Others: 2Traffic accident: 11; Sports accident: 6; Others: 311/713/7Zhang L (2024) [24]Cohort studyChina3313Mean age: 48.0; Sex: 18/15Mean age: 43.6; Sex: 10/3NRNRNRNRHu Z (2025) [38]Cohort studyChina4033Mean age: 51.2; Sex: 17/23Mean age: 50.5; Sex: 15/18NRNR17/2317/16ORIF open reduction internal fixation, RCT random controlled trial, NR not reportSpecial statement of superscript #: the “Study ID” comprised the first author’s name and publication year Information of included studies ORIF open reduction internal fixation, RCT random controlled trial, NR not report Special statement of superscript #: the “Study ID” comprised the first author’s name and publication year The risk of bias regarding included RCTs (n = 8) was assessed according to the Cochrane ROB tool 2.0 in terms of 5 domains (Domain 1: bias arising from the randomization process; Domain 2: bias due to deviations from intended interventions; Domain 3: bias due to missing outcome data; Domain 4: bias in measurement of the outcome; Domain 5: bias in selection of the reported result), which disclosed that the majority of domains were low risk or some concerns in RCT studies. Generally speaking, 2 studies were low risk overall, 4 studies were high risk overall, and 2 studies were high risk overall (Table 2). Regarding the included cohort studies (n = 19), the risk of bias was assessed according to the Newcastle‒Ottawa scale involving 3 domains (selection, comparison, and outcome), which revealed that 16 studies had a total score ≥ 7. In contrast, another 3 studies had a total score = 6 (Table 3). Table 2Quality assessment of RCT by Cochrane ROB tool 2.0Study IDDomain 1Domain 2Domain 3Domain 4Domain 5OverallZhao ZL (2014) [27]Some concernsLow riskLow riskLow riskLow riskSome concernsLin YC (2019) [31]High riskLow riskLow riskSome concernsLow riskHigh riskWang GF (2019) [32]Some concernsLow riskHigh riskLow riskSome concernsHigh riskWu L (2019) [33]Some concernsLow riskLow riskSome concernsLow riskSome concernsFeng YS (2020) [34]Some concernsLow riskLow riskSome concernsLow riskSome concernsDeng SC (2022) [10]Low riskLow riskLow riskLow riskLow riskLow riskZhang Y (2022) [12]Some concernsLow riskLow riskSome concernsLow riskSome concernsLi JL (2023) [15]Some concernsLow riskLow riskLow riskLow riskSome concernsSun Y (2023) [18]Low riskLow riskLow riskLow riskLow riskLow riskRCT random controlled trial, ROB risk of biasDomain 1: bias arising from the randomization process;Domain 2: bias due to deviations from intended interventions;Domain 3: bias due to missing outcome data;Domain 4: bias in measurement of the outcome;Domain 5: bias in selection of the reported result Quality assessment of RCT by Cochrane ROB tool 2.0 RCT random controlled trial, ROB risk of bias Domain 1: bias arising from the randomization process; Domain 2: bias due to deviations from intended interventions; Domain 3: bias due to missing outcome data; Domain 4: bias in measurement of the outcome; Domain 5: bias in selection of the reported result Table 3Quality assessment of cohort study by Newcastle-Ottawa scaleStudy IDSelectionComparisonOutcomeTotalSu JB (2017) [28]4116Lai XY (2018) [29]4127Zhang YP (2018) [30]3126Lu DL (2020) [35]4127Xu HY (2021) [36]4138Yuan YF (2021) [37]4116Zhang MT (2022) [11]4239Du RZ (2023) [13]3238He J (2023) [14]4127Pan JB (2023) [16]4138Pei YR (2023) [17]4138Wang HJ (2023) [19]4138Deng Y (2024) [20]3137Guo SY (2024) [21]4239Li J (2024) [22]4138Wu R (2024) [23]4138Zhang L (2024) [24]4228Hu Z (2025) [38]4217 Quality assessment of cohort study by Newcastle-Ottawa scale Comparison of perioperative indices and hospital stays The incision length (SMD: -4.701, 95%CI: -5.717 to -3.686, P < 0.001), perioperative bleeding volume (SMD: -3.304, 95%CI: -3.834 to -2.774, P < 0.001), postoperative drainage volume (SMD: -1.071, 95%CI: -1.870 to -0.273, P = 0.009), time to first ambulation (SMD: -2.294, 95%CI: -4.127 to -0.460, P = 0.014), and length of hospital stay (SMD: -1.918, 95%CI: -2.594 to -1.242, P < 0.001) were all lower, but the operation duration (SMD: 1.998, 95%CI: 0.595 to 3.401, P = 0.005) was longer, in the arthroscopy group than in the ORIF group (Fig. 2A-F). Pooled analyses for perioperative indices and hospital stays. Pooled analyses for incision length (A), perioperative bleeding (B), operation duration (C), postoperative drainage volume (D), time to first ambulation (E), and hospital stays (F) between arthroscopic surgery and ORIF in treating patients with PCL tibial avulsion fracture Comparison of postoperative functional recovery Twenty-four, 11, 12, 4, and 7 studies were pooled to analyze the differences in the Lysholm score, International Knee Documentation Committee (IKDC) score, range of motion (ROM), postoperative pain visual analog scale (VAS) score, and postoperative posterior drawer test (PDT) negative rate between arthroscopy and ORIF. After treatment, the Lysholm score (SMD: 1.424, 95%CI: 0.858 to 2.837, P < 0.001), IKDC score (SMD: 1.848, 95%CI: 0.858 to 2.837, P < 0.001), and ROM (SMD: 0.527, 95%CI: 0.001 to 1.053, P = 0.050) were all greater in the arthroscopy group than in the ORIF group (Fig. 3A-C). However, the postoperative pain VAS score (SMD: -0.250, 95%CI: -0.664 to 0.165, P = 0.238) and postoperative PDT negative rate (OR: 1.040, 95%CI: 0.460 to 2.351, P = 0.925) were not significantly different between the arthroscopy group and the ORIF group (Fig. 3D-E). Moreover, two studies reported Tegner active scale score after treatment, since the number of studies was < 3, a qualitative presentation was provided. Guo SY’s study (2024) reported no difference in Tegner active scale score (SMD: 0.132, 95%CI: -0.286 to 0.551), whereas Li J’s study (2024) reported a higher Tegner active scale score (SMD: 0.551, 95%CI: 0.080 to 1.023) in the arthroscopy group than in the ORIF group. These data suggested that arthroscopic surgery appeared to provide better functional outcomes compared to ORIF did to some extent. Pooled analyses for postoperative function recovery. Pooled analyses for Lysholm score (A), IKDC (B), ROM (C), postoperative pain VAS score (D), and postoperative PDT negative rate (E) between arthroscopic surgery and ORIF in treating patients with PCL tibial avulsion fracture Comparison of postoperative complications Nine studies were pooled to analyze the difference in postoperative complication incidence between arthroscopy and ORIF, revealing that the incidence of postoperative complications (OR: 0.458, 95%CI: 0.149 to 1.411, P = 0.174) was not significantly different between the arthroscopy and ORIF groups (Fig. 4). These data implied that arthroscopic surgery did not increase the risk of complications compared with ORIF. Pooled analysis for postoperative complications Publication bias and sensitivity analysis Among all the assessed indicators, the incision length (P = 0.004), perioperative bleeding (P = 0.016), length of hospital stay (P = 0.010), Lysholm score (P = 0.021), ROM (P = 0.033), and postoperative complications (P = 0.020) presented with publication bias (Table 4) and were therefore adjusted by the trim-and-fill method. After adjustment, the trends and significance of these 6 indicators between the arthroscopy and ORIF groups remained unchanged. In addition, sensitivity analyses were also performed, which revealed that the trend and significance/nonsignificance of all indicators remained stable after each study was omitted (Supplementary Fig. 1A-L). Table 4Publication biasItemsP valueBias estimate (SE)Adjusted SMD or OR [95% CI] #Incision length0.004-8.84 (1.72)-3.54 [-4.58, -2.51]Perioperative bleeding0.016-7.57 (2.11)-2.94 [-3.47, -2.41]Operation duration0.1777.48 (5.28)Postoperative drainage volume0.064-17.63 (1.76)Time to first ambulation0.182-16.22 (4.78)Hospital stays0.010-12.85 (4.02)-1.30 [-2.04, -0.55]Lysholm score0.0219.10 (3.64)0.60 [-0.12, 1.32]IKDC score0.3136.10 (5.70)ROM0.0338.59 (3.47)0.53 [ 0.00, 1.05]Postoperative pain score (VAS)0.144-7.46 (3.18)Postoperative PDT negative rate0.9299.56 (99.17)Postoperative complications0.020-617.72 (182.99)0.59 [0.20, 1.75] Publication bias Subgroup analyses Subgroup analyses were performed on the basis of the stratification of study type (RCT or cohort study), follow-up period (≤ 6 months or > 6 months), and fixation method (arthroscopy: suture, Endo-Button plate, or others; ORIF: screw, anchor, or others). Most trends in the pooled subgroup analyses between arthroscopy and ORIF remained the same as in the full-group analyses (Supplementary Tables 2 and Supplementary Table 3). Discussion Arthroscopy is a focus of orthopedic surgery, including PCL tibial avulsion fracture [39–41]. A previous study reported that arthroscopic surgery reduced the incision length, time to first ambulation, and hospital stay while increasing the operation duration compared with ORIF in patients with PCL tibial avulsion fracture [10]. Another study revealed that arthroscopic surgery decreased perioperative bleeding, postoperative drainage volume, time to first ambulation, and hospital stay but increased operation duration compared with ORIF in patients with PCL tibial avulsion fracture [14]. However, another study reported that arthroscopic surgery did not shorten hospital stays compared with ORIF in patients with PCL tibial avulsion fracture [19]. These findings indicate that the findings regarding perioperative indices and hospital stays between arthroscopic surgery and ORIF in patients with PCL tibial avulsion fracture are not consistent, making this meta-analysis necessary. This meta-analysis pooled existing data and revealed that arthroscopic surgery appeared to decrease the incision length, perioperative bleeding volume, postoperative drainage volume, time to first ambulation, and hospital stay but increased the operation duration compared with ORIF in patients with PCL tibial avulsion to some degree. Two reasons may explain the above pooled analysis findings: (1) Arthroscopic surgery is a minimally invasive surgery with less invasion than ORIF; therefore, reduced levels of incision length, perioperative bleeding, postoperative drainage volume, time to first ambulation, and hospital stays are discovered. (2) Arthroscopic surgery takes more time than ORIF does because of its technical complexity, requiring precise manipulation in a confined space, longer setup for specialized equipment, and steep learning curve. With respect to postoperative functional recovery, arthroscopic surgery may be superior to ORIF in patients with PCL tibial avulsion fracture. A previous study revealed that the Lysholm score and ROM were both greater after arthroscopic surgery than after ORIF treatment in patients with PCL tibial avulsion fracture [18]. Another study revealed that the Lysholm score was higher but the postoperative pain VAS score was lower after arthroscopic surgery than after ORIF treatment in patients with PCL tibial avulsion fracture [24]. Nevertheless, a previous study reported that the Lysholm score, ROM, and postoperative PDT negative rate did not differ after arthroscopic surgery versus ORIF treatment in patients with PCL tibial avulsion fracture [23]. Another study revealed that the IKDC score, ROM, postoperative pain VAS score, and postoperative PDT negative rate did not differ after arthroscopic surgery compared with ORIF treatment in patients with PCL tibial avulsion fracture [31]. After pooling the existing data, this meta-analysis revealed that the Lysholm score, IKDC score, and ROM appeared to be greater after arthroscopic surgery than after ORIF treatment in patients with PCL tibial avulsion fracture to a certain extent, whereas the postoperative pain VAS score and postoperative PDT negative rate were not significantly different. The reason may be that arthroscopy is less invasive than ORIF is, resulting in fewer surgical wounds and relatively better recovery from surgery. In the context of postoperative complications, seven existing studies reported this aspect; among them, 1 study revealed that the incidence of postoperative complications was lower after arthroscopic surgery than after ORIF treatment in patients with PCL tibial avulsion fracture [10], another 5 studies reported that the incidence of postoperative complications was not different between them [14, 15, 18, 30, 36], while 1 study disclosed that the incidence of postoperative complications was higher after arthroscopic surgery than after ORIF treatment [38]. After these existing data were pooled, this meta-analysis revealed that the incidence of postoperative complications was not significantly different after arthroscopic surgery than after ORIF in patients with PCL tibial avulsion fracture. In terms of the difference between our meta-analysis and previous ones, a previous meta-analysis compared arthroscopic and open surgery for tibial-sided bony avulsions of the PCL, which reported generally similar outcomes and risks between the two approaches, although arthroscopy appeared to yield slightly higher knee outcome scores [4]. However, that meta-analysis did not specifically focus on or conduct subgroup analyses for patients with PCL tibial avulsion fracture, which is quite different from our meta-analysis. Moreover, another meta-analysis compared arthroscopic suture fixation with an open posterior approach in patients with PCL tibial avulsion fracture and reported satisfactory clinical outcomes and knee stability between the two techniques [6], which was quite different from our meta-analysis in the following dimensions: (1) The previous meta-analysis focused specifically on arthroscopic suture fixation, whereas our meta-analysis evaluated overall arthroscopic surgery. (2) The previous meta-analysis included only studies with follow-up durations exceeding 12 months, while we imposed no such restrictions. (3) The previous meta-analysis presented descriptive results without pooling the data statistically, in contrast to our pooled quantitative analysis. (4) The previous meta-analysis included a small number of studies (n = 12), whereas our meta-analysis incorporated a substantially larger number (n = 27). In addition, a previous meta-analysis was recently published [42], which compared arthroscopic fixation with ORIF in patients with PCL tibial avulsion fracture. However, the previous meta-analysis included only studies published in English from international databases (PubMed, Embase, Scopus, and the Cochrane Library) and was limited to studies published up to July 2024; moreover, only 10 studies were included. Our present meta-analysis was quite different from the previous meta-analysis: firstly, we included Chinese databases and literature; secondly, we updated the study search period to October 2025; thirdly, a larger number of studies (n = 27) were included in our meta-analysis; fourthly, we performed a number of subgroup analyses. Assessment of heterogeneity is vital in meta-analyses, and several factors contribute to heterogeneity. Firstly, there is clinical heterogeneity, including differences in subject characteristics, intervention approaches, control approaches, and definitions and measurements of outcomes. Secondly, there is methodological heterogeneity, including differences in study designs, risk of bias, analysis sets, and sample sizes. Thirdly, there is statistical heterogeneity, such as variation in effect sizes. In the present meta-analysis, several substantial heterogeneities were discovered in the intervention effects of the assessed outcomes. We consider the possible reasons for the heterogeneity as follows: (1) Differences in ages, ranging from 10.0 to 74.0 years, among the studies. (2) Differences in the mechanisms of injury (traffic accidents, sports accidents, falls, and others) among studies. (3) Differences in fixation methods (arthroscopy: suture, Endo-Button plate, or others; ORIF: screw, anchor, or others) among studies. (4) Differences in follow-up periods, ranging from 1.5 to 20 months (or not specified ≥ 12 months), were noted among studies. (5) Differences in sample sizes, ranging from 27 to 200 cases. Some limitations in this meta-analysis should be considered. First, the majority of the studies are not RCTs but cohort studies, which may affect the reliability of the findings. Second, the majority of the studies are from China, which may introduce geographic or surgical-practice bias; therefore, limiting the generalizability of the study findings. Third, the number of studies reporting some indices, such as time to first ambulation and the Tegner active scale, is small, making the pooled analysis less convincing. Fourth, surgeon experience is very important in the arthroscopic group because the operative time is affected by the learning curve, which may cause performance bias in this meta-analysis. Conclusions Arthroscopic surgery appears to have improved efficacy and a comparable risk of complications to ORIF in treating patients with PCL tibial avulsion fracture to some extent, although longer operative time and study heterogeneity warrant caution. This information suggests that arthroscopic surgery may be a better choice for PCL tibial avulsion fracture with indications. However, high-quality multicenter RCTs with standardized protocols are still needed to validate these findings and broaden their generalizability. Supplementary Information Supplementary Material 1. Supplementary Material 2: Supplementary Fig. 1. Sensitivity analyses. Sensitivity analyses by omitting each study for incision length (A), perioperative bleeding (B), operation duration (C), postoperative drainage volume (D), time to first ambulation (E), and hospital stays (F), Lysholm score (G), IKDC (H), ROM (I), postoperative pain VAS score (J), and postoperative PDT negative rate (K), postoperative complications (L). Supplementary Material 3. Acknowledgements Not applicable. Registration and protocol This meta-analysis was not registered, and protocol could be access via contacting the corresponding author on reasonable request. Abbreviation open reduction internal fixation posterior cruciate ligament standardized mean difference odds ratio confidence interval risk of bias randomized controlled trial Newcastle‒Ottawa scale Authors’ contributions Bo Yu: Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Writing - original draft, Writing - review & editing. Yanan Liao: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Writing - original draft, Writing - review & editing. Funding Not applicable. Data availability The datasets used and analysed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2: Supplementary Fig. 1. Sensitivity analyses. Sensitivity analyses by omitting each study for incision length (A), perioperative bleeding (B), operation duration (C), postoperative drainage volume (D), time to first ambulation (E), and hospital stays (F), Lysholm score (G), IKDC (H), ROM (I), postoperative pain VAS score (J), and postoperative PDT negative rate (K), postoperative complications (L). Supplementary Material 3. Data Availability Statement The datasets used and analysed during the current study are available from the corresponding author on reasonable request.