Quadriceps Tendon Graft Leads to Stronger Flexion and Weaker Extension of the Knee Than Hamstring Graft Post-ACLR: Systematic Review and Meta-Analysis

Abstract

Purpose

The optimal autograft for restoring knee strength and expediting return-to-sport following anterior cruciate ligament reconstruction (ACLR) remains debated. This systematic review and meta-analysis compared isokinetic strength and patient-reported outcomes after ACLR using quadriceps tendon (QT), hamstring tendon (HT), or bone–patellar tendon–bone (BTB) autografts to determine graft-specific recovery patterns.

Methods

PubMed, Cochrane Library, EMBASE, and Google Scholar were systematically searched for comparative studies evaluating isokinetic strength following primary ACLR using QT, HT, or BTB autografts. Non-English, unavailable full-text, animal/cadaveric, and physeal-sparing studies were excluded. MINORS and Detsky evaluated bias risk. Key findings were quantified, and subgroup meta-analyses were performed on limb symmetry index (LSI) values (significance: p < 0.05).

Results

Seventeen studies (1,705 patients: QT = 765, HT = 725, BTB = 215) with follow-ups through 48 months were included. Due to data heterogeneity, four to six studies qualified for each meta-analysis. Among 13 QT vs. HT studies, QT grafts demonstrated weaker extensor strength in nine (56.3%) and greater flexor strength in six (37.5%). No differences were reported in extensor or flexor strength in four (25.0%) and seven (43.8%) studies, respectively. For QT vs. BTB, two studies (12.5%) showed similar extensor strength, whereas two reported conflicting results. QT had greater flexor strength in one (6.3%) and similar strength in three (18.8%) studies. Postoperative pain was comparable between QT and HT.

Conclusions

QT autografts resulted in stronger flexion and weaker extension than HT autografts after ACLR. Tailored rehabilitation targeting graft-specific deficits may accelerate recovery. Further research is required to clarify outcomes between QT and BTB autografts.

Level of evidence: IV.

Introduction

Anterior cruciate ligament (ACL) injuries are among the most common sports-related injuries and account for almost 50% of all knee injuries [1]. Recent global data estimates an annual incidence of ACL tears at 68.6 per 100,000 person-years, with approximately 1 in 3,500 affected in the United States [1, 2]. Surgical ACL reconstruction (ACLR) is the standard operative approach to restore knee stability, improve patient outcomes, and reduce the risk of osteoarthritis in ACL-deficient knees [3]. Postoperative quadriceps and hamstring weakness are well-documented concerns after ACLR [4]. A prolonged strength imbalance is reported to delay functional recovery, hinder return-to-sport readiness and increase the risk of reinjury following ACLR [5–7]. Various tendon graft choices, including quadriceps tendon (QT), hamstring tendon (HT), and bone–patellar tendon–bone (BTB), are commonly used for ACLR [8].

Previous studies have reported diminished muscle strength immediately following ACLR [9]. Isokinetic strength testing measures the maximal strength performance of isolated muscle groups [10, 11]. Therefore, isokinetic testing can provide insights into quadriceps and hamstring recovery after ACLR, which is crucial for knee stability, return-to-sport, and long-term joint health [12]. Although prior research has compared strength deficits among QT, HT, and BTB graft types [8], the ideal choice remains a topic of debate in the orthopedic literature [2]. This study aimed to systematically review and compare isokinetic knee extensor and flexor strength testing in patients undergoing ACLR with different autografts to assist in future graft selection based on outcomes. In addition, this review examined postoperative pain and other patient-reported outcome measures (PROMs) as secondary outcomes to provide a comprehensive assessment of graft performance and patient recovery. We hypothesized that patients who received QT grafts would demonstrate weaker postoperative isokinetic knee extensor strength and stronger isokinetic knee flexor strength than those who received HT or BTB grafts.

Methods

Literature Search and Eligibility

This study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines [13, 14]. Two authors (J.C.D, E.L.M.) independently conducted comprehensive searches of PubMed, Embase, Cochrane Library, and Google Scholar with the following search terms: “Anterior Cruciate Ligament Reconstruction” AND (“Quadriceps Tendon Autograft” OR "Quadriceps Tendon Graft”) OR (“Bone–Patellar Tendon–Bone Graft” OR “Bone–Patellar Tendon–Bone Autograft” OR “Patellar Tendon”) AND (“Hamstring Tendon Graft” OR “Hamstring Tendon Autograft” OR “Semitendinosus and Gracilis Tendon Autograft”) AND “Isokinetic” (Clarivate, Philadelphia, PA). References were reviewed to ensure that no articles were overlooked for inclusion.

Articles were selected for inclusion if they met the following criteria: (1) comparative study; (2) patients who underwent primary ACLR using QT, HT, or BTB autografts; and (3) assessment of isokinetic knee extension or flexion strength. Articles were excluded if they were not published in English, retracted, published without an abstract, described physeal-sparing techniques, included cadaveric or nonhuman subjects, or had no full text available. In addition, case reports, case series, editorials, commentaries, opinion pieces, letters to the editor, gray literature, and conference abstracts were excluded.

Data Extraction

Data extraction was performed by three researchers (B.A.Y, J.C.D, and E.L.M III) for the following: author, journal, publication year, level of evidence, study country, study design, graft types, total number of patients, individual study group sizes with sex and age breakdowns, postoperative evaluation periods, and study findings. All isokinetic strength tests compared the strength of the injured limb to that of the contralateral non-injured limb using the limb symmetry index (LSI).

Data Analysis

Due to heterogeneous reporting, key findings of isokinetic strength and PROMs were quantified using descriptive statistics. Seventeen studies were included in this systematic review, but only 4 to 6 met the inclusion criteria for each specific meta-analysis due to variability in testing parameters and outcome reporting. A subgroup meta-analysis was conducted on isokinetic strength, pooling estimates based on graft type, postoperative evaluation period, and isokinetic testing velocity. Statistical significance was set at p < 0.05 for all analyses.

Risk-of-Bias Assessment

The risk of bias assessment for primary studies was conducted by two authors (J.C.D, E.L.M.III). Comparative studies were graded according to The Methodological Index for Non-Randomized Studies (MINORS) tool [15], and graded as very low quality (0–6), low quality (7–12), moderate quality (13–18), or high quality (≥ 19) [16]. This instrument has shown reliable, consistent, and valid results in characterizing the methodological and scientific values of articles [15]. The risk of bias for randomized controlled trials was graded according to the Detsky quality index [17, 18]. According to this methodology, each randomized controlled trial was graded based on components of the following six categories: randomization, outcome measurement and blinding, inclusion and exclusion criteria, description of interventions, statistical analysis, and follow-up rates. Five of the broad categories were weighed equally, each worth four points. The final category was worth one additional point. For negative trials, an additional question was added regarding confidence intervals and post-hoc power calculations. Positive and negative trials had maximum scores of 21 and 22, respectively. The 18 final Detsky scores were adjusted to a 100-point scale for consistency. For instance, a positive trial scoring 16 of 21 was equivalent to 76 of 100, or 76%. Randomized controlled trials scoring greater than 75% were considered high quality [19].

Results

Study Selection and Characteristics

Database searches identified 4,981 articles, 1,591 of which were duplicates. During screening, 3,279 articles did not meet inclusion criteria. An additional 94 articles were excluded, because they did not have sufficient data or full texts available. In total, 17 studies met inclusion criteria, ranging from 2004 to 2024. Figure 1 illustrates the detailed screening process according to the PRISMA guidelines.

Fig. 1.

PRISMA screening process

After removing individuals lost to follow-up, there were a total of 1,705 patients. Of these, 765 patients (44.9%) received QT grafts (455 males, 310 females), 725 patients (42.5%) received HT grafts (435 males, 290 females), and 215 patients (12.6%) received BTB grafts (154 males, 61 females). Of the included studies, 12 compared postoperative isokinetic strength results between QT and HT grafts, three compared QT and BTB grafts, one compared all three types, and one compared small versus large stature females following QT grafting. Regarding study design, eight articles were cohort studies (8 of 17 [47.1%]), six were randomized controlled trials (6 of 17 [35.3%]), two were cross-sectional studies (2 of 17 [11.8%]), and one was a comparative study (1 of 17 [5.9%]). Table 1 displays the general study and specific intervention characteristics.

Table 1 .

Study characteristics

Authors (year)	Level of evidence	Study design	Patient	Graft type	Age, years (SD)	Male/female	Postoperative evaluation periods
Ebert et al. [21]	I	Prospective randomized control trial	N = 112	QT: 57 HT: 55 BTB: 0	QT: 28.1 (± 8.2) HT: 29.4 (± 7.7)	QT: 28/29 HT: 28/27	6 weeks, 3, 6, 12, 24 months
Ebert et al. [34]	IV	Prospective randomized control trial subanalysis	N = 97	QT: 48 HT: 49 BTB: 0	QT: 28.7 (16–47)b HT: 29.2 (16–43)b	QT: 23/25 HT: 24/25	12, 24 months
Holmgren et al. [28]	III	Cross-sectional study	N = 312	QT: 104 HT: 104 BTB: 104	QT: 27.0 (± 9.5) HT: 26.0 (± 9.7) BTB: 26.0 (± 9.4)	QT: 60/44 BTB: 60/44 HT: 60/44	7 months
Horstmann et al. [22]	II	Prospective randomized control trial	N = 51	QT: 24 HT: 27 BTB: 0	QT: 24.1 (± 3.6) HT: 32.7 (± 11.4)	QT: 21/3 HT: 12/15	3, 6, 12, 24 months
Johnston et al. [24]	III	Retrospective cohort study	N = 105	QT: 35 HT: 70 BTB: 0	QT: 20.0 (± 8) HT: 20.0 (± 9)	QT: 28/7 HT: 54/16	6, 12 months
Guney-Deniz et al. [32]	III	Cross-sectional, case–control study	N = 65	QT: 22 HT: 22 BTB: 0 TA: 21	QT: 27.8 (± 2.8) HT: 26.7 (± 4.6) TA: 26.4 (± 5.5)	QT: 17/5 HT: 18/4 TA: 17/4	15.6 (12–18) monthsb
Sinding et al. [30]	II	Prospective randomized control trial	N = 135	QT: 42 HT: 43 BTB: 0 CON: 50	QT: 28.7 (± 6.4) HT: 28.3 (± 6.2) CON: 28.3 (± 6.2)	QT: 25/17 HT: 23/20 CON: 27/23	12.5 (± 0.7) monthsa
Csapo et al. (2019)	III	Retrospective cohort study	N = 46	QT: 21 HT: 25 BTB: 0	QT: 20.5 (± 2.8) HT: 21.0 (± 3.0)	QT: 6/15 HT: 15/10	5.3 (± 0.8) monthsa
Hunnicutt et al. (2019)	III	Cohort study	N = 30	QT: 15 HT: 0 BTB: 15	QT: 25 (14–41)b BTB: 18 (15–32)b	QT: 12/3 BTB: 7/8	8 (6–23) monthsc
Fischer et al. [27]	III	Retrospective cohort study	N = 124	QT: 61 HT: 63 BTB: 0	QT: 21.7 (± 7.4) HT: 21.5 (± 6.9)	QT: 34/27 HT: 47/16	5.5 (± 1.2), 7.6 (± 1.6) monthsa
Martín-Alguacil et al. [23]	II	Randomized control trial	N = 51	QT: 26 HT: 25 BTB: 0	QT: 18.7 (± 3.6) HT: 19.2 (± 3.6)	QT: 23/3 HT: 16/9	3, 6, 12 months
Cavaignac et al. [36]	III	Cohort study	N = 86	QT: 45 HT: 41 BTB: 0	QT: 32.1 (± 8) HT: 30.9 (± 9)	QT: 25/20 HT: 24/17	43.2 (± 4.8) monthsa
Lee et al. [33]	III	Cohort study	N = 96	QT: 48 HT: 48 BTB: 0	QT: 31.1 (17–57)b HT: 29.9 (17–58)b	QT: 44/4 HT: 44/4	12, 24 months
Han et al. [20]	III	Retrospective comparative study	N = 144	QT: 72 HT: 0 BTB: 72	QT: 27.8 (15–51)b BTB: 27.8 (15–51)b	QT: 68/4 BTB: 68/4	3, 6, 9, 12, 24, 36, 48 months
Pigozzi et al. [25]	III	Prospective randomized control trial	N = 48	QT: 24 HT: 0 BTB: 24	QT: 33.0 (21–47)b BTB: 35.0 (23–41)b	QT: 17/7 BTB: 19/5	6 months
Weaver et al. [31]	III	Retrospective cohort study	N = 201	QT: 48 HT: 153 BTB: 0	QT: 15.6 (± 1.53) HT: 15.7 (± 1.48)	QT: 24/24 HT: 70/83	6–9 months
Goto et al. [35]	IV	Retrospective cohort study	N = 73	QT: 73 HT: 0 BTB: 0	QT: 33.8 (± 11.5)	QT: 0/73	S: 9.1 (± 2.3) L: 8.9 (± 2.6) monthsa

^aValues are mean (standard deviation); ^bvalues are mean (range); ^cvalues are median (range)

Quality Assessment Risk-of-Bias Results

The MINORS criteria scores of the included comparative studies ranged from 15 to 21, with an average score of 18 (± 1.64), indicating a low overall risk of bias. MINORS criterion for the included studies are summarized in Fig. 2. Detsky quality index scores for the randomized controlled trials ranged from 86 to 96%, with an average score of 89% (± 3.65%), indicating a low overall risk of bias. A summary of Detsky quality analysis for each randomized controlled trial is shown in Fig. 3. The full grading for both assessments is provided in the Appendix.

Fig. 2.

MINORS score for comparative studies. Green: 100% category score = low bias. Red: 0% category score = high risk of bias. Yellow: all other scores

Fig. 3.

Detsky score for randomized controlled trials. Green: 100% category score = low bias. Red: 0% category score = high risk of bias. Yellow: all other scores

Isokinetic Strength Outcomes

Three Months

Four studies (23.5%) tested isokinetic strength within the first 3 postoperative months [20–23]. One study reported significantly stronger knee extensor strength in the HT group than in the QT group at velocities of 60°/s, 180°/s, and 300°/s [23]. Two other studies reported no significant extensor strength differences between QT and HT groups at 60°/s or 90°/s [21, 22]. These studies reported no differences in knee flexor strength results between QT and HT groups at any of the previously mentioned velocities. [21–23]. One study compared QT and BTB grafts at 60°/s and 180°/s but reported no differences in knee extensor or flexor strength [20]. These findings indicated that QT and HT grafts influence extensor strength at higher velocities during early recovery; however, flexor strength appears to be less affected by graft type early on.

Six Months

Eight studies (47.1%) assessed isokinetic strength between 4 and 6 months postoperatively. [20–27] Four studies reported significantly better knee extensor strength in the HT group than in the QT group at a velocity of 60°/s, [23, 26, 27] 90°/s, [21] 180°/s, and 300°/s [23]. Another study reported no significant difference between QT and HT at 60°/s [22]. Pooled meta-analysis showed a statistically significant advantage in extension LSI for HT grafts at 60°/s (LSI = 0.88; 95% CI 0.62 to 1.14; p = 0.00, I [2] = 0%; Fig. 4). Figure 4 demonstrates that HT grafts are associated with a moderate-to-large advantage in knee extension LSI compared to QT grafts at 6 months, suggesting superior quadriceps strength recovery, which may impact return-to-sport readiness.

Fig. 4.

Comparison of knee extension limb symmetry index (LSI) at 60°/s between QT and HT autografts at 6 months postoperatively

Regarding flexor strength, two studies reported significantly stronger results for the QT group than the HT group at 60°/s and 90°/s [21, 27], whereas three others reported no difference at 60°/s, [22, 23, 26] 180°/s, or 300°/s. [23] Pooled analysis revealed a small-to-moderate advantage in flexion LSI for QT grafts at 60°/s (LSI = − 0.45; 95% CI − 0.69 to − 0.20; p = 0.00, I [2] = 0%), as shown in Fig. 5. Patients who received QT autografts demonstrated greater hamstring strength recovery than those who received HT grafts at 6 months postoperatively.

Fig. 5.

Comparison of knee flexion limb symmetry index (LSI) at 60°/s between QT and HT autografts at 6 months postoperatively

The findings of the comparison between QT and BTB grafts were inconsistent. One study reported similar isokinetic knee extensor and flexor results at 60°/s and 180°/s [20], whereas another reported QT to have significantly greater extensor and flexor strengths at 80°/s and 160°/s. [25]

Twelve Months

Thirteen studies (76.5%) reported isokinetic strength between 7 and 12 months postoperatively [20–24, 27–34]. Seven studies reported significantly greater knee extensor strength in the HT group when compared to the QT group at 60°/s, 90°/s, 180°/s, 240°/s, or 300°/s. [21, 23, 24, 27, 28, 31, 34]. Sinding et al. also reported significantly greater extensor strength in the HT group at 60°/s but no significant difference between groups at 180°/s [30]. Ebert et al. reported significantly greater knee extensor strength in females following HT graft in comparison with QT graft at 90°/s while reporting similar extensor strength in males following both QT and HT grafts [34]. Three studies reported no extensor strength differences between QT and HT grafts at 60°/s or 180°/s [22, 32, 33]. Meta-analysis showed continued HT advantage in extension LSI at 60°/s (LSI = 0.55; 95% CI, 0.02 to 0.90; p = 0.00, I² = 57%; Fig. 6). Figure 6 indicates that while HT retains an extension advantage at 12 months, the effect size is smaller than at 6 months, and heterogeneity increases (I² = 57%), which may reflect variability in testing intervals and participant demographics.

Fig. 6.

Comparison of knee extension limb symmetry index (LSI) at 60°/s between QT and HT autografts at 12 months postoperatively

Flexor strength results at 12 months postoperatively showed significant improvement with QT graft in six studies when compared to HT graft at 60°/s, 90°/s, or 180°/s [21, 24, 27, 30, 33, 34]. Five studies reported no differences between QT and HT groups at 60°/s, 90°/s, 180°/s, 240°/s, and 300°/s [22, 23, 28, 31, 32]. One study compared isokinetic extensor and flexor strength results between short- and normal-stature females who underwent ACLR with only QT grafts. This study observed no significant differences between the two female groups at either 60°/s or 240°/s [35]. The pooled results again favored QT in flexion LSI at 60°/s (LSI = − 0.67; 95% CI − 0.92 to − 0.41; p = 0.00; I² = 22%; Fig. 7). Figure 7 shows that hamstring recovery continued to favor the QT grafts at 12 months, with more pronounced asymmetry than at 6 months.

Fig. 7.

Comparison of knee flexion limb symmetry index (LSI) at 60°/s between QT and HT autografts at 12 months postoperatively

When comparing BTB to QT grafts, the QT group had significantly weaker quadriceps strength at 90°/s and 240°/s [28], whereas the other knee extensor and flexor evaluations, performed at 60°/s and 180°/s, returned similar outcomes between the two graft types. [20, 28, 29]

Greater Than 12 Months

Most studies reported follow-up timelines of up to 2 years, with Cavaignac et al. and Han et al. reporting follow-up periods of 36 months and 48 months, respectively [20, 36]. Seven studies (41.2%) reported isokinetic strength at greater than 1-year follow-up after ACLR [20–22, 32–34, 36]. The testing velocities for evaluative periods greater than 12 postoperative months were 60°/s, 90°/s, and 180°/s. No significant differences were reported in isokinetic knee extensor strength between QT, HT, or BTB grafts at any of the velocities tested following 12 postoperative months [20–22, 32–34, 36]. These results demonstrated that quadriceps strength recovery was similar in patients who received QT and HT grafts 12 months postoperatively. However, three studies found significantly greater knee flexor strength with QT grafts than with HT grafts at the 2-year follow-up [21, 33, 34], demonstrating that QT graft recipients had greater hamstring strength than HT recipients at 24 months. The remaining studies reported similar flexor strength between QT, HT, and BTB grafts [20, 22, 32, 36].

Patient-Reported Outcomes

Patient-reported outcomes varied in type, timing, and reporting methodology across the included studies. To enhance clarity, Table 2 summarizes PROM findings by graft type, PROM metric, evaluation timepoint, and associated outcome.

Table 2.

Summary of patient-reported outcome measures (PROM) by graft type, timepoint, and outcome

Graft type	PROM metric	Timepoint	Outcome	Reference(s)
QT versus HT	Pain (VAS, NRS)	3–12 months	No significant difference	[20, 22, 25, 34]
QT versus BTB	Anterior knee pain	≥ 12 months	QT had lower incidence of anterior knee pain	[21]
QT versus BTB	KOOS (pain/function)	12 months	Similar scores	[30]
QT versus HT	KOOS—symptoms/sport	≥ 36 months	QT showed significantly better scores	[20]
QT versus HT	Return-to-sport	3–12 months	HT showed higher return-to-sport rates	[35]
QT versus HT	Health status (males)	12–24 months	HT had greater improvement in postoperative health scores	[35]
QT versus HT	Lysholm	≥ 36 months	QT demonstrated better scores	[20]
QT versus HT	Activity/satisfaction	Multiple	No significant differences	[23–25, 31–34]

QT, Quadriceps tendon; HT, Hamstring tendon; BTB, bone–patellar tendon–bone; VAS, Visual Analog Scale; NRS, Numeric Rating Scale; KOOS, Knee Injury and Osteoarthritis Outcome Score

Six studies (35.3%) evaluated patient-reported pain with different graft types. Four studies compared QT and HT grafts, and all reported no significant differences in pain after ACLR. Evaluation tools for these studies included the Shelbourne–Trumper questionnaire, a numerical pain rating scale (NRS), and the visual analog scale (VAS) for knee and graft site pain [21, 24, 33, 36]. Of the two studies comparing QT and BTB grafts, one reported a significantly lower incidence of anterior knee pain in the QT group using the Shelbourne–Trumper score [20], while the other reported similar pain between the two graft types, according to the Knee injury and Osteoarthritis Outcome Score (KOOS) [29].

Other patient-reported outcomes included the following: significantly better return-to-sport measures for the HT group than the QT group at 3, 6, and 12 postoperative months, significantly greater improvements in postoperative health status for males with HT grafts than males with QT grafts at 12 and 24 months, significantly better Lysholm (function in activities of daily living), KOOS-symptoms, and KOOS-sport measures in the QT group than the HT group after 36 postoperative months [36], and significantly greater return-to-sport and activity scores for males at 12 and 24 months compared to females, regardless of graft type [34]. No significant differences were reported for activity level, daily living, or satisfaction between QT and HT grafts [22–24, 30–33] or between QT and BTB grafts [20, 29].

Discussion

The principal finding of this study was that HT grafts for ACLR were associated with earlier quadriceps strength recovery in the first 12 months postoperatively, whereas QT grafts were associated with better hamstring strength recovery in the first 24 months postoperatively. Similar postoperative pain was observed after QT and HT grafting.

Recent high-quality evidence from Raj et al., who conducted a systematic review and meta-analysis of randomized controlled trials comparing QT and HT grafts, aligns with many of our findings [2]. Their review found no significant differences in anterior knee laxity, IKDC scores, or Lysholm outcomes between the two grafts, although several trials favored QT in terms of donor-site morbidity and pain. However, similar to our meta-analysis, their pooled data also highlighted a trend toward reduced extension strength in the QT group, reinforcing concerns regarding early quadriceps weakness following QT harvest. These consistent findings across independent meta-analyses strengthen the case for individualized rehabilitation protocols and graft selection based on patient-specific needs and activity demands [2].

Previous studies have reported postoperative isokinetic strength deficits associated with specific graft types [37–39]; however, this study provides new insights into the strength recovery timeline following ACLR based on graft site. The current return-to-sport timeline is approximately 9–12 months postoperatively [40]. A prior systematic review reported a greater extensor strength deficit following QT than HT grafts at 6 postoperative months but not at 12 months [8], indicating that beyond 6 months, the influence of graft choice on quadriceps strength may diminish. Quadriceps strength improved after 6 months, but a persistent imbalance in quadriceps strength was observed in QT graft recipients compared with HT recipients at 12 months. This updated timeline supports consideration of graft type in relation to extensor strength recovery and timing of return-to-sport. The moderate heterogeneity observed in some analyses (e.g., I² = 57%) reflects methodological variability and limits the generalizability of our conclusions. Although Hofer et al. similarly reported an improvement in isokinetic quadriceps strength between 6 and 12 months following ACLR, the extensor strength deficit remaining for these patients is likely greater than that for those with HT grafts, which could contribute to delayed return-to-sport [41]. Due to varied isokinetic testing protocols across studies—including differences in angular velocity, timing, and equipment—our pooled results may be influenced by methodological heterogeneity.

Furthermore, the observation of delayed knee flexor strength recovery following HT grafts compared to QT grafts provides further insight into the significance of graft choice. Herbawi et al. reported delayed hamstring strength recovery following HT grafts through 12 months when compared to QT grafts [8]. Although we were unable to pool data after 24 months, multiple included studies reported a significant difference between QT and HT flexor strength at 24 months, indicating that delayed hamstring strength recovery in HT graft recipients may extend beyond 12 months [21, 33, 34]. Greater hamstring LSI, demonstrating greater flexor strength symmetry between injured and uninjured knees, has been reported to be associated with psychological readiness to return-to-sport and with returning to the pre-injury level of sport [6, 7]. Raj et al. similarly noted that while QT and HT grafts perform comparably on functional scales, such as Lysholm and IKDC, the potential for lower donor-site morbidity and graft harvest-site pain with QT may have implications for return-to-sport comfort and rehabilitation compliance [2]. Therefore, we recommend targeting postoperative hamstring rehabilitation for patients receiving HT grafts to improve functionality and decrease return-to-sport time.

Although this study suggests the effects of BTB graft on early isokinetic strength in comparison with QT graft are still uncertain, our observation of increased long-term anterior knee pain with BTB graft aligns with previous findings. Multiple studies have reported a significantly lower incidence of anterior knee pain in the QT groups after 12 months than in the BTB groups [42–44]. Less graft site and harvest-site pain has also been reported at 12 months following surgery in favor of QT grafts when compared to BTB grafts [43, 44]. As knee pain may further impede an athlete’s ability to perform, clinicians should be aware of higher reported pain following the use of BTB grafts rather than QT grafts and incorporate additional pain management techniques into rehabilitation regimens for these patients.

Given these findings, we suggest that physical therapy and recovery plans be designed to target strength deficits associated with graft type. Introducing secondary supportive rehabilitation techniques, such as early progressive eccentric exercises [42, 45], whole body or localized vibration [46–48], and blood flow restriction in combination with low-intensity exercises [49] have been shown to result in significantly better early isokinetic strength recovery following ACLR than standard rehabilitation regimens. Furthermore, we recommend that physicians incorporate additional pain management techniques into the recovery plans of patients undergoing ACLR using BTB grafts. Rehabilitation plans should be personalized to address the strength deficits and pain associated with each graft type, which may enhance strength recovery and facilitate earlier return-to-sport compared to conventional protocols.

This systematic review had several limitations. The main limitation of this study is the heterogeneous nature of isokinetic testing. Testing methods were not consistent across all studies, preventing a full meta-analysis. In addition, we focused on isokinetic strength to evaluate outcomes after ACLR; however, additional outcome measures beyond isokinetic strength could provide further insights in future studies. Furthermore, although rehabilitation plans were provided to patients in each study, the researchers could not enforce protocol adherence. Follow-up strength results were dependent on plan compliance. Patient compliance with rehabilitation protocols was not measured or controlled, which may have influenced isokinetic outcomes. Because most of the included studies lacked preoperative strength data, we could not definitively interpret the magnitude of postoperative recovery; instead, strength outcomes were based on inter-limb comparisons. All included studies compared injured to non-injured limbs and received moderate or high-quality grades according to the bias risk assessment.

Further research using a standardized isokinetic strength test with the same number of repetitions, length of rest, and set angular velocities is recommended to ensure accurate comparisons between graft types. Standardized evaluation timepoints, including a preoperative baseline, may be helpful in comparing outcomes. As the rehabilitation techniques listed above evaluated isokinetic strength following ACLR using only HT and BTB grafts, further research specifically involving QT grafts is needed to address safety and efficacy concerns.

Conclusion

Our data supports our hypothesis that ACLR using QT grafts results in greater extensor strength deficits than HT grafts in the first 12 postoperative months. In addition, QT grafts led to similar or better flexor strength results than HT between 4 and 24 months postoperatively, and lower anterior knee pain than BTB. Rehabilitation targeting early strength deficits following QT and HT grafting may decrease the time of return-to-sport. Further research is needed to evaluate postoperative isokinetic strength differences between QT and BTB graft types, as current comparisons are based on limited and inconsistent evidence.

Abbreviations

ACL

Anterior cruciate ligament

ACLR

Anterior cruciate ligament reconstruction

QT

Quadriceps tendon

HT

Hamstrings tendon

BTB

Bone–patellar tendon–bone

LSI

Limb symmetry index

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-analyses

MINORS

The Methodological Index for non-randomized studies

KOOS

Knee injury and Osteoarthritis Outcome Score

Appendix

MINORS Score

Study	A	B	C	D	E	F	G	H	I	J	K	L	Total
Holmgren et al. [28]	2	2	0	2	2	2	0	0	2	2	2	2	18/24
Johnston et al. [24]	2	2	0	2	1	2	1	0	2	2	2	2	19/24
Guney-Deniz et al. [32]	2	2	0	2	1	2	0	2	2	2	2	2	19/24
Csapo et al. [26]	2	2	0	2	1	2	0	0	1	2	1	2	15/24
Hunnicutt et al. [29]	2	2	0	2	2	2	0	2	2	2	2	2	20/24
Fischer et al. [27]	2	2	0	2	1	2	0	0	2	2	2	2	21/24
Cavaignac et al. [36]	2	2	0	2	1	2	0	0	2	2	2	2	17/24
Lee et al. [33]	2	2	0	2	1	2	0	0	2	2	2	2	17/24
Han et al. [20]	2	2	0	2	1	2	0	0	2	2	2	2	19/24
Goto et al. [35]	2	2	0	2	1	2	0	2	2	2	2	2	19/24
Weaver et al. [31]	2	2	0	2	1	2	0	0	2	2	2	2	19/24

A: Clearly state aim

B: Inclusion of consecutive patients

C: Prospective data collection

D: Endpoints appropriate to study aim

E: Unbiased assessment of study endpoint

F: Follow-up period appropriate to study aim

G: < 5% lost to follow-up

H: Prospective calculation of study size

I: Adequate control group

J: Contemporary groups

K: Baseline equivalence of groups

L: Adequate statistical analyses

Detsky Score

Section	Questions	Ebert et al. [22]	Ebert et al. [35]	Horstmannet al.	Sinding et al.	Martín-Alguacil et al.	Pigozzi et al.
1. Randomization (total out of 4)	(a) Patients randomly assigned? (1)	1	1	1	1	1	1
	(b) Method of randomization described? (2)	2	2	2	2	2	2
	(c) Randomization concealed to investigator? (1)	1	1	1	1	1	1
2. Outcome measurement and blinding (total out of 4)	(a) Outcomes described? (1)	1	1	1	1	1	1
	(b) Outcomes objective? (2)	2	2	2	2	2	2
	(c) Outcome assessors blind? (1)	0	0	1	0	0	0
3. Inclusion and Exclusion Criteria (total out of 4)	(a) Eligibility criteria well-defined? (2)	2	2	2	2	2	2
	(b) Number of excluded patients and reasons for exclusion from trial provided? (2)	0	0	2	2	2	2
4. Interventions (total out of 4)	(a) Therapeutic regimen fully described for the treatment group? (2)	2	2	2	2	2	2
	(b) Therapeutic regimen fully described for the control group? (2)	2	2	0	2	2	2
5. Statistical issues (total out of 4 if positive trial; total out of 5 if negative trial)	(a) Test stated and P values given? (1)	1	1	1	1	1	1
	(b) Statistical analysis appropriate? (2)	2	2	2	2	2	2
	(c) If trial is negative, 95% confidence intervals given or post-hoc power calculation performed? (1 if applicable)	1	1	N/A	1	1	N/A
	(d) Sample size calculated before study? (1)	1	1	1	1	0	0
6. Follow-up rate (total out of 1)	> 80% follow-up (1)	1	1	1	1	0	1
	< 80% follow-up (0)
Total Detsky Score (out of 21 or 22)		19/22	19/22	19/21	21/22	19/22	19/21

Author Contributions

B.A.Y., J.C.D., J.N.L., and A.V.S. participated in the conception of the work. B.A.Y., J.C.D., and E.L.M.III. conducted data collection. Multiple independent reviewers were necessary for data integrity; therefore, B.A.Y., J.C.D., E.L.M.III., E.L.M., C.L.F., and E.H.L. participated in data analysis, interpretation, and drafting the manuscript. J.N.L. and A.V.S. participated in critical revision and final approval of the article. All authors read and approved the final manuscript.

Funding

The Professional Student Mentored Research Fellowship (PSMRF) Project is supported by the National Center for Advancing Translational Sciences through Grant UL1TR001998, UK HealthCare and the University of Kentucky. The project was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1TR001998. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Availability of Data and Materials

All data generated or analyzed during this study are included in this published article and its appendix.

Declarations

Competing Interests

The authors declare that they have no competing interests.

Ethics Approval and Consent to Participate

Not applicable.

Consent for Publication

Not applicable.