An electronic search of PubMed and Embase databases was conducted on 12 September 2022 to explore studies reporting clinical outcomes of ACLR using ST-augmented grafts. The following key terms were used for the electronic search: (anterior cruciate ligament OR ACL) AND (tape OR augment OR reinforce OR internal brace). The full search strategies used for each of these electronic databases are available in Supplementary Appendix Tables S1, S2. A manual search of Google Scholar was then performed to identify studies not indexed by the Web of Science.

Two authors (Z.Z. and Z.F.) independently assessed all studies and cross-checked the finalized ones according to the inclusion and exclusion criteria (Table 1). The titles and abstracts were initially screened for relevance, then the full texts were critically retrieved for further selection. Reference lists of all included articles were reviewed for potentially eligible studies. Any disagreement about a study’s inclusion was resolved by discussion with a third senior author (H.Z.) involved if consensus could not be achieved.

Each finally included study was abstracted regarding the study features, patient data, surgical information, and outcome measures. Two authors (G.S. and Y.L.) independently extracted the original data, and the final decision on the disagreement was made by a third senior author (H.Z.).

Study features consisted of author name, publication year, journal, study design, level of evidence, and methodological quality. Patient data comprised number of cases, sex, age, length of follow-up, and meniscal status. Surgical information was extracted from the specific descriptions in the original studies, including graft details (choice, construct, diameter, and fixation), tape details (indication, product, and fixation), and concomitant procedures. For outcome measures, all subjective and objective results including pain and function scores, knee laxity measurements, return to sports, graft failures, and other complications were documented. Comparison analysis between pre- and postoperative conditions and between patients with and without STA was recorded. Radiographic and arthroscopic outcomes extracted by follow-up magnetic resonance imaging (MRI) and second-look arthroscopy were not reviewed in this systematic review, as there was no available data in the original studies.

Descriptive statistics were used to report study characteristics, patient data, surgical information, and outcome measures. A quantitative comparison (meta-analysis) of the data was considered inappropriate due to the heterogeneity between studies.

The methodologic quality of each included study was assessed with the Methodological Index for Non-Randomized Studies (MINORS). The items in the MINORS criteria for non-randomized studies were scored as 0 (not reported), 1 (reported but inadequate), or 2 (reported and adequate) (Slim et al., 2003). For comparative studies, the ideal MINORS score was 24, and a study was considered at low risk of bias when it scored 21–23 and at high risk of bias when it scored ≤20. For non-comparative studies, corresponding thresholds were 16, 13–15, and ≤12 (Slim et al., 2003). The MINORS score of each study was calculated independently by two authors (G.S. and Y.L.). Any disagreement was resolved by discussion until a consensus was reached.

The electronic and manual search identified 1,250 studies for initial screening, and 74 studies proceeded to full-text review. Critical application of the inclusion and exclusion criteria finally yielded nine studies (Figure 2). One was a prospective cohort study (level 2) (Shantanu et al., 2019), six were retrospective cohort studies (level 3) (Bodendorfer et al., 2019; Parkes et al., 2021; Allom et al., 2022; Kitchen et al., 2022; Szakiel et al., 2022; von Essen et al., 2022), and two were case series (level 4) (Lavender et al., 2021; Duong et al., 2022). All seven comparative studies were designed to compare patients who underwent ACLR with and without ST-augmented grafts (Table 2).

A high risk of bias was confirmed in all seven comparative studies with a mean MINORS value of 18.0 ± 1.6 (range, 16–20). The two non-comparative studies were also at high risk of bias with a MINORS value of 11 for each (Table 2). The risk of bias mainly came from an unblinded assessment of the study endpoint, followed by a lack of prospective calculation of the study size and a historical comparison between the study and control group. The full appraisal for each study is available in Supplementary Appendix Table S3.

A total of 314 knees of 314 patients undergoing ACLR with STA were enrolled in this systematic review (Table 3). There were 190 (60.5%) males and 124 females (39.5%) with a mean age of 26.6 years (range, 15.7–33.0 years). The mean clinical follow-up was 16.7 months (range, 6.0–29.0 months). The meniscal status at the time of ACLR was documented in six studies, and the incidence of meniscal tears was 49.0% (125/255) overall, 32.0% (72/225) for the medial, and 32.9% (74/225) for the lateral.

ACLRs were performed with 283 hamstring tendon auto- or allografts in eight studies, 20 quadriceps tendon autografts in one study, and 11 bone-patellar tendon-bone autografts in one study (Table 4). ACL grafts were augmented with FiberTape (Arthrex, United States) in 231 patients and InternalBrace (Arthrex, United States) in 11 patients, and the product used in the rest 72 patients was not described. Different fixation methods of the graft-ST construct were applied. On the femoral side, the suspensory device was used in eight studies, with the ST passing through the eyelets of the suspensory button in eight studies and the loop of the suspensory button in two studies. On the tibial side, the suspensory device was used in seven studies and the interference screw in two studies, with the ST fixed independently from the graft using an additional anchor before graft fixation in six studies and along with the graft using the same fixation device in two studies. Most studies applied an equal or slightly less tension on the ST than the ACL graft during fixation. For all studies, the use of ST was based on the surgeon’s preference, except for one prospective cohort study with randomized selection. Concomitant lateral extra-articular tenodesis was performed in one study (Allom et al., 2022) for 84/169 patients with risk factors for graft rupture.

Overall, eight subjective scores were reported in eight out of nine studies: the Lysholm score in four studies, International Knee Documentation Committee (IKDC) score in three studies, Knee injury and Osteoarthritis Outcome Score (KOOS) score in three studies, Single Assessment Numeric Evaluation (SANE) score in two studies, Tegner activity score in two studies, visual analogue scale (VAS) score in two studies, Western Ontario and McMaster Universities Osteoarthritis (WOMAC) score in one study, and IKDC grade in one study. The remaining one study (Allom et al., 2022) not reporting any subjective score (Table 2).

Seven studies compared subjective scores between pre- and postoperative status for ACLR with STA, and significant improvement was reported in six studies. Those used at least in two studies included the Lysholm score from 61.2 to 91.2 (Shantanu et al., 2019; Duong et al., 2022), IKDC score from 30.7 to 85.8 (Bodendorfer et al., 2019; Lavender et al., 2021), and KOOS score from 60.9 to 89.1 (Bodendorfer et al., 2019; Szakiel et al., 2022; von Essen et al., 2022). The IKDC grade was reported in one study (Duong et al., 2022) with the grade (A/B/C/D) from 0/0/19/18 to 30/7/0/0. The comparison of the other three scores was only performed in one study (Bodendorfer et al., 2019) with the SANE score from 96.0 to 90.0, VAS score from 7.1 to 1.6, and WOMAC score from 35.5 to 2.2. Only one study (Parkes et al., 2021) demonstrated no significant improvement in the Tegner score (from 7.2 to 7.1).

Six studies compared subjective scores between ACLR with or without STA at the final follow-up. The Tegner score used in two studies (Parkes et al., 2021; Kitchen et al., 2022) both yielded superior results in the augmentation group. Results of other scores were inconsistent across studies. Only one study (Bodendorfer et al., 2019) reported that the VAS (1.6 versus 3.4), IKDC (87.6 versus 73.2), KOOS (92.2 versus 87.1), SANE (90.0 versus 80.0), and WOMAC (2.2 versus 6.2) score were all significantly better in augmentation group. However, no significant superiority of any score other than the Tenger was found in the augmentation group among other studies (Table 5).

Knee laxity was evaluated by the Lachman test in five studies, pivot shift test in four studies, and KT-1000/2000 arthrometer test in two studies (Table 6). In all these studies, significant improvement in objective stability was reported in patients with ST-augmented grafts at the final follow-up. The comparative analysis of knee laxity between ACLR using ST-augmented grafts and not was performed in four studies. No studies reported superior objective stability in the augmentation group at the final follow-up, although one of which (Shantanu et al., 2019) reported better improvement in grades of the Lachman test as the preoperative laxity in the augmentation group was severer.

The ability to return to preinjury sports level was evaluated in three studies, with an overall rate of 73.8% (range, 69.2%–81.8%) in patients with ST-augmented ACL grafts (Table 7). The ability to return to sports between patients with or without STA was compared in two studies. Both demonstrated no (but a trend toward) significant difference in the rate of return to sports at a minimum 2-year follow-up, and one of them (Bodendorfer et al., 2019) reported that the use of ST was significantly correlated with earlier time to return (9.2 months versus 12.9 months).

The graft failure and other major complications were recorded in seven studies (Table 8). The overall failure rate of ST-augmented grafts was 2.3% (range, 0%–6.7%) among 266 patients. The comparison of failure between grafts with or without ST proceeded in five studies. Almost all studies reported a lower failure rate in ST-augmented grafts, but the difference did not meet statistical significance. Other complications including further meniscal tear, arthrofibrosis, cyclops lesion, and infection were also not significantly different in any of these studies.

The limitations of this systematic review were largely from the results of input literature. Firstly, the low level of evidence and high risk of bias within the majority of included studies lowered the validity of this review. Second, the heterogeneity and insufficient data across the original literature restricted the application of a formal meta-analysis. Third, variable reports of outcome measures made this review unable to draw a firm conclusion on the superiority of STA from the clinical perspective. Fourth, the surgical indication of STA lacked a specific description, leading to uncertainty that patients with what features would more benefit from the ST-augmented grafts. Fifth, no radiological or arthroscopic methods were used to evaluate the intra-articular condition after ACLR with STA. Despite the limitations, this is the first study to systematically review the clinical outcomes of ACLR with STA, with a comparison to standard ACLR. This study could be helpful to direct future studies that seek to explore the clinical effectiveness of STA in ACLR.