Comparative assessment of graft maturity after anterior cruciate ligament reconstruction using different graft types: a systematic review

Abstract

Background

The selection of graft remains a subject of ongoing debate in anterior cruciate ligament (ACL) reconstruction, with distinct maturation processes having been observed among different graft types. A thorough understanding of these differences in graft maturation is crucial for optimizing rehabilitation protocols and ensuring a safe return to sports. This study aimed to systematically review the differences in graft maturation among different graft types following ACL reconstruction.

Methods

A comprehensive literature search was conducted in PubMed, Embase, and the Cochrane Library in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies were included only if they compared intra-articular ACL graft maturity across different graft types.

Results

Twenty-one studies met the inclusion criteria. Graft maturity was assessed using magnetic resonance imaging (MRI) in 15 studies, second-look arthroscopy in 4 studies, and histological biopsy in 1 study; one additional study utilized both MRI and second-look arthroscopy. Hamstring tendon (HT) and bone-patellar tendon-bone (BPTB) autografts showed similar maturity, as assessed by MRI signal intensity (SI) and histological findings, after ACL reconstruction. However, results from second-look arthroscopy were inconclusive. HT autografts exhibited MRI SI comparable to soft-tissue allografts within the first postoperative year, but demonstrated superior maturity and graft appearances at approximately 2 years postoperatively. Quadriceps tendon (QT) autografts, both with and without a patellar bone block, revealed lower MRI SI compared to HT autografts, suggesting better graft maturity. HT autografts with preserved tibial insertion maintained relatively lower SI during the early maturation phase (6 and 12 months) than free HT autografts, though no significant differences were observed at later stages (24 and 60 months).

Conclusion

MRI, second-look arthroscopy, and histological biopsy analysis indicated distinct graft maturation levels following ACL reconstruction. No conclusive evidence established whether HT or BPTB autografts are superior in terms of graft maturity. Compared to free HT autografts and soft-tissue allografts, QT autografts and HT autografts with preserved tibial insertion may mature earlier, which may allow for consideration of an earlier return to sports in clinical decision-making. These grafts may therefore represent viable alternatives to HT and BPTB grafts, particularly in young and active patients.

Level of Evidence

III, systematic review of level Ⅰ-Ⅲ investigation.

Introduction

Anterior cruciate ligament (ACL) reconstruction is currently the standard surgical treatment for ACL rupture [1–4]. However, the selection of the optimal graft type for ACL reconstruction remains a subject of ongoing debate. Commonly used grafts include bone-patellar tendon-bone (BPTB), hamstring tendon (HT), quadriceps tendon (QT) without a bone block, and quadriceps tendon with a patellar bone block (QTB) [5, 6]. Additionally, the choice between autografts and allografts involves weighing factors such as cost, donor site morbidity, patient satisfaction, and clinical outcomes [7–11].

After ACL reconstruction, the tendon graft undergoes a process of functional adaptation and continuous remodeling within the intra-articular environment, transforming into a structure biomechanically similar to the native ACL. This process is termed “ligamentization” [12]. Both animal and human studies have delineated three sequential stages of graft ligamentization: an initial healing phase, followed by a remodeling phase, and finally a maturation phase [13]. Although histological analysis of human biopsy samples has offered valuable insights into this process, the invasive nature and associated costs of second-look arthroscopy make routine clinical biopsy impractical [13, 14]. Therefore, a growing number of studies have employed magnetic resonance imaging (MRI) as a non-invasive diagnostic tool to evaluate graft maturity and morphological changes during graft maturation [15–17].

For long-term graft survival, favorable ligamentization with good biological quality and mechanical properties would be essential [18, 19]. Previous animal studies have demonstrated that graft signal intensity (SI) measured with MRI could predict the graft maturity and mechanical strength, where elevated SI corresponds to hypercellular and hypervascular tissue with inferior biomechanical properties, and reduced SI suggests greater graft strength and structural integrity [20–22]. Clinically, patients who successfully returned to preinjury sports level exhibited superior graft maturity on MRI [23], and those with better graft tension during second-look arthroscopy showed enhanced knee stability [24], collectively indicating a direct relationship between graft quality and functional outcomes. Therefore, a thorough understanding of graft maturation is crucial for designing optimized rehabilitation protocols and facilitating safe return to sports.

Graft selection represents a critical aspect of individualized ACL reconstruction. The timeline and biological characteristics for achieving full graft maturity vary significantly among different graft types [13, 25]. While previous reviews focused mainly on clinical outcomes and graft failure rates [26–28], comprehensive research characterizing the maturation process across different grafts remains relatively lacking. Therefore, the purpose of this review is to systematically review and compare the maturity differences among various graft types following ACL reconstruction. We hypothesized that bone-based autografts (BPTB and QTB) and HT autografts with preserved tibial insertion would exhibit earlier maturation and superior histological characteristics compared to soft-tissue grafts.

Methods

Search strategy

This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [29]. A comprehensive literature search was performed using the electronic databases PubMed, EMBASE, and the Cochrane Library from their inception until April 2024, including articles published online as “Epub ahead of print”. The search strategy incorporated key concepts related to anterior cruciate ligament reconstruction, graft types (autograft, allograft), and biological processes (ligamentization, maturity, maturation, remodeling, and healing), along with relevant abbreviations and synonym terms. The detailed search strategies for each database are provided in Supplementary Table 1. Additionally, the reference lists of identified articles were manually screened to supplement further eligible studies.

Study selection and criteria

Two authors (YP and SX) independently screened and assessed all retrieved articles for eligibility. All authors concurred on the final study selection, and any discrepancies were resolved through consensus. The inclusion criteria were as follows: (1) studies comparing graft maturity among different ACL graft types using MRI, second-look arthroscopy, or histological analysis of biopsy specimens after primary arthroscopic ACL reconstruction; (2) studies focusing on the intra-articular portion of the ACL graft; (3) studies that included at least one assessment ≥ 6 months; and (4) original studies published in English and performed on human subjects. Exclusion criteria comprised: (1) animal or cadaveric studies; (2) case reports, case series, reviews, commentaries, or conference abstracts; and (3) studies evaluating only a single graft type or lacking comparative analysis across graft types.

Data extraction

Data extraction was independently performed by one reviewer (YP) and verified by another (SX), with all data recorded in a standardized Microsoft Excel spreadsheet. The extracted information included general study characteristics (first author, publication year, level of evidence, and study design), patient demographics, graft types, number of participants evaluated at follow-up, time points of assessment, methods used to evaluate graft maturity, and key outcomes. Among the studies that conducted serial MRI assessments, data at early postoperative time points (e.g., 3 months) were extracted for descriptive analysis of maturation trajectories. Authors were not contacted to obtain unpublished data.

Quality assessment

Two authors (YP and SX) independently assessed the level of evidence and methodological quality of the included studies, with any conflict resolved by discussion and consensus among all authors. The level of evidence for each study was determined based on the Oxford Centre for Evidence-Based Medicine 2011 (if not available from the study) [30]. The methodological quality of observational studies (cohort and case–control studies) was assessed using the Newcastle–Ottawa Quality Assessment Scale [31]. Studies with scores of 0–3, 4–6, and 7–9 were classified as low, moderate, and high quality, respectively. Randomized controlled trials were assessed based on the Cochrane Risk of Bias tool.

Results

Literature search

A flowchart detailing the study selection process is presented in Fig. 1. The initial database search yielded 4054 potential articles. After removing duplicates and screening titles, abstracts, and full texts, 21 articles [24, 32–51] met the eligibility criteria and were included in this systematic review. Publication years ranged from 2005 to 2024. The included studies comprised 7 randomized controlled trials [33, 34, 36, 37, 47, 50, 51], 7 prospective cohort studies [35, 39, 40, 42, 45, 46, 48], 6 retrospective cohort studies [24, 32, 38, 41, 43, 44], and 1 retrospective case–control study [49]. Methodological quality assessments are summarized in Supplementary Fig. 1 (for randomized controlled trials) and Supplementary Table 2 (for observational studies).

Fig. 1.

PRISMA flowchart showing the selection of studies for review. ACLR, anterior cruciate ligament reconstruction

Study identification and characteristics

Among the included studies, graft maturity was assessed using conventional, contrast-enhanced, or quantitative MRI in 15 studies [32–34, 36, 37, 39–41, 44–49, 51], histological biopsy in 1 study [43], second-look arthroscopy in 4 studies [24, 38, 42, 50], and both MRI and second-look arthroscopy in 1 study [35]. The main characteristics of the studies are summarized in Table 1.

Table 1.

Study characteristics of the included studies

Lead author (year)	Level of evidence; Study design	Graft type; Patients, n	Follow-up	Method of examination
Ahn (2007) [24]	Ⅲ; Retrospective cohort study	HT autograft; n = 129 BPTB autograft; n = 80	21.2 (range, 14–70) mo	Second-look arthroscopy
Aitchison (2021) [32]	Ⅲ; Retrospective cohort study	HT autograft; n = 37 QT autograft; n = 33	6, 12 mo	Conventional MRI
Covey (2018) [33]	Ⅰ; Randomized controlled trial	HT autograft; n = 41 BPTB autograft; n = 31	3, 6, 9, 12 mo	Contrast-enhanced MRI
Cusumano (2022) [34]	Ⅰ; Randomized controlled trial	HT autograft; n = 25 PT or PL allograft; n = 25	6, 12 mo	Conventional MRI
Fukuda (2022) [35]*	Ⅱ; Prospective cohort study	HT autograft; n = 45 BPTB autograft; n = 30	2 yr	Conventional MRI
Fukuda (2022) [35]*	Ⅱ; Prospective cohort study	HT autograft; n = 40 BPTB autograft; n = 22	1 yr	Second-look arthroscopy
Gupta (2024) [36]	Ⅰ; Randomized controlled trial	HT-PI autograft; n = 23 BPTB autograft; n = 24	8, 14 mo	Conventional and contrast-enhanced MRI
Kim (2018) [38]	Ⅲ; Retrospective cohort study	HT autograft; n = 26 TA allograft; n = 30	22.5 ± 7.8 mo	Second-look arthroscopy
Kim (2021) [37]	Ⅱ; Randomized controlled trial	HT autograft; n = 15 TA allograft; n = 38	1 yr	Contrast-enhanced MRI
Lansdown (2020) [39]	Ⅱ; Prospective cohort study	HT autograft; n = 27 Soft-tissue allograft; n = 12	6, 12, 24, 36 mo	Quantitative MRI
Li (2017) [40]	Ⅲ; Prospective cohort study	HT autograft; n = 21 TA allograft; n = 17	3, 6, 12 mo	Conventional MRI
Ma (2015) [41]	Ⅲ; Retrospective cohort study	HT autograft; n = 14 QTB autograft; n = 12	6 mo	Conventional MRI
Mae (2019) [42]	Ⅲ; Prospective cohort study	HT autograft; n = 57 BPTB autograft; n = 56	10 (range, 6–23) mo	Second-look arthroscopy
Marumo (2005) [43]	Ⅱ; Retrospective cohort study	HT autograft; n = 20 BPTB autograft; n = 30	11–13 mo	Histological analysis
Muramatsu (2008) [44]	Ⅲ; Retrospective cohort study	BPTB autograft; n = 20 BPTB allograft; n = 24	1, 4, 6, 12 mo	Contrast-enhanced MRI
Panos (2021) [45]	Ⅱ; Prospective cohort study	HT autograft; n = 55 QT autograft; n = 23	3, 9 mo	Conventional MRI
Rose (2017) [46]	Ⅱ; Prospective cohort study	HT allograft; n = 16 TA allograft; n = 16	6 mo	Conventional MRI
Ruffilli (2016) [47]	Ⅰ; Randomized controlled trial	HT-PI autograft; n = 20 Free HT autograft; n = 20	6 mo	Conventional MRI
Vari (2023) [48]	Ⅲ; Prospective cohort study	HT-PI autograft; n = 90 Free HT autograft; n = 90	12.24 ± 2.02 mo	Conventional MRI
Yamasaki (2024) [49]	Ⅲ; Retrospective case–control	HT autograft; n = 42 QTB autograft; n = 21	2.1 (IQR, 2.0–3.5) yr	Conventional MRI
Yoo (2017) [50]	Ⅰ; Randomized controlled trial	HT autograft; n = 26 TA allograft; n = 25	18.1 (range, 12.5–24.2) mo	Second-look arthroscopy
Zhang (2020) [51]	Ⅱ; Randomized controlled trial	HT-PI autograft; n = 18 Free HT autograft; n = 19	6, 12, 24, 60 mo	Conventional MRI

^*The study contributed data from the same patient cohort assessed with different modalities and at different postoperative time points. BPTB, bone-patellar tendon-bone; HT, hamstring tendon; HT-PI, hamstring tendon with preserved tibial insertion; IQR, interquartile range; MRI, magnetic resonance imaging; PL, peroneus longus; QT, quadriceps tendon; QTB, quadriceps tendon with patellar bone block; TA, tibialis anterior tendon; TP, tibialis posterior tendon

Maturity assessment methods

As summarized in Table 2, the graft maturity assessment methods varied considerably across studies. Among the 16 studies utilizing MRI to evaluate graft maturity, conventional MRI was used in 11 studies [32, 34, 35, 40, 41, 45–49, 51], contrast-enhanced MRI in 3 studies [33, 37, 44], quantitative MRI in 1 study [39], and a combination of conventional and contrast-enhanced MRI in 1 study [36]. Second-look arthroscopy was used in 5 studies to assess graft maturity. Evaluated parameters included graft integrity (n = 4 [24, 38, 42, 50]), graft tension (n = 3 [24, 38, 42]), synovial coverage (n = 5 [24, 35, 38, 42, 50]), and surface neovascularization (n = 1 [38]). Finally, one study performed histological biopsy to compare total collagen content and the ratio of nonreducible to reducible collagen fibers between BPTB and HT autografts [43]. Furthermore, considerable heterogeneity was observed in MRI protocols, including magnetic field strength, coil design, sequence types, and acquisition parameters (Supplementary Table 3). Together with diverse maturity assessment methods and inconsistent measurement intervals of the included studies, direct comparison of the maturity of different graft types across studies is complicated and statistically questionable. Consequently, no meta-analysis was performed in this systematic review.

Table 2.

Maturity assessments methods of the included studies

Methods	Maturity indicators
MRI
Conventional MRI	Subjective (n = 3) [36, 47, 49] SNQ = (SIgraft - SIPCL) / SIbackground (n = 3) [35, 41, 48] SNQ = (SIgraft - SIQT) / SIbackground (n = 4) [34, 36, 40, 46] SIR = (SIgraft / SIPCL) (n = 1) [32] SNR = SIgraft / SIbackground (n = 1) [51]
Contrast-enhanced MRI	Subjective (n = 1) [33] SNQpost-contrast = (SIgraft - SIQT) / SIbackground (n = 2) [36, 44] SIRpost-contrast = SIgraft / SIgastrocnemius (n = 1) [37]
Quantitative MRI	T1rho and T2 mapping (n = 1) [39]
Second-look arthroscopy	Graft integrity (n = 4) [24, 38, 42, 50] Graft tension (n = 3) [24, 38, 42] Synovial coverage (n = 5) [24, 35, 38, 42, 50] Neovascularization (n = 1) [38]
Histological analysis	Total collagen content (n = 1) [43]

MRI, magnetic resonance imaging; PCL, posterior cruciate ligament; QT, quadriceps tendon; SI, signal intensity; SIR, signal intensity ratio; SNQ, signal-to-noise quotient; SNR, signal-to-noise ratio

Graft type

Regarding graft types, six studies compared HT autografts versus BPTB autografts [24, 33, 35, 36, 42, 43], six studies compared HT autografts versus soft-tissue allografts [34, 37–40, 50], four studies compared HT autografts versus QT or QTB autografts [32, 41, 45, 49], and three studies compared HT autografts with preserved tibial insertion (HT-PI) versus free HT autografts (FHT) [47, 48, 51]. The remaining two studies compared: BPTB autografts versus BPTB allografts [44], and HT allografts versus tibialis anterior tendon (TA) allografts [46].

HT autograft versus BPTB autograft

HT and BPTB autografts are the most commonly used grafts in ACL reconstruction. Six studies compared between HT autografts and BPTB autografts. As shown in Table 3, at short-term follow-up (≤ 12 months), HT and BPTB autografts demonstrated comparable SI and enhancement extent on both conventional and contrast-enhanced MRI [33, 36]. Concurrently, histological analysis showed no significant differences in total collagen content and the ratio of nonreducible to reducible cross-links [43]. However, second-look arthroscopy demonstrated superior graft integrity and synovial coverage in BPTB autografts relative to HT autografts [35, 42]. At medium- to long-term follow-up (> 12 months), there was no definitive consensus regarding maturity comparison between HT and BPTB autografts. Although HT autografts showed higher SI on conventional MRI compared to BPTB autografts [35], they were observed to have slightly better synovial coverage as assessed by second-look arthroscopy [24, 35].

Table 3.

Maturity assessment outcomes between HT autografts and BPTB autografts

Lead author (year)	Maturity assessment method	Essential outcomes
Short-term follow-up (≤ 12 months)
Gupta (2024) [36]	Subjective SNQ = (SIgraft - SIQT) / SIbackground EI = SNQpost-contrast / SNQpre-contrast	No significant differences in MRI SI between BPTB autografts and HT-PI grafts at 8 and 14 months postoperatively. Gadolinium-enhanced MRI revealed significantly greater revascularization in BPTB autografts at both time points
Covey (2018) [33]	Subjective	No significant differences in gadopentetate dimeglumine enhancement grades were observed between HT and BPTB autografts at 3, 6, 9, and 12 months postoperatively
Marumo (2005) [43]	Total collagen content	No significant differences in total collagen content and nonreducible/reducible crosslink ratios between BPTB and HT autografts at 11–13 months postoperatively
Mae (2019) [42]	Graft integrity, graft tension, synovial coverage	Graft integrity and synovial coverage were significantly better in BPTB autografts at a mean 10-month follow-up
Medium- to long-term follow-up (> 12 months)
Fukuda (2022) [35]	SNQ = (SIgraft - SIPCL) / SIbackground Synovial coverage	At 1 year postoperatively, synovial coverage was significantly better with BPTB autografts. At the 2-year follow-up, mean SI was significantly higher in HT autografts compared with BPTB autografts
Ahn (2007) [24]	Graft integrity, graft tension, synovial coverage	Synovial coverage was superior in HT autografts compared with BPTB autografts at a mean follow-up of 21.2 months. No significant differences were observed in graft integrity and tension

BPTB, bone-patellar tendon-bone; HT, hamstring tendon; HT-PI, hamstring tendon with preserved tibial insertion; MRI, magnetic resonance imaging; PCL, posterior cruciate ligament; QT, quadriceps tendon; SI, signal intensity; SNQ, signal-to-noise quotient

HT autograft versus soft-tissue allograft

The soft-tissue allografts used in the six included studies primarily consisted of TA, tibialis posterior tendon (TP), and peroneus longus tendon (PL). Specifically, four studies compared HT autografts with TA allografts [37, 38, 40, 50], one study compared HT autografts with TP or PL allografts [34], and one study did not specify the allograft type [39]. Comparative analysis of the six studies revealed a distinct temporal pattern in maturation between HT autografts and soft-tissue allografts. At short-term follow-up (≤ 12 months), four studies demonstrated comparable maturity and revascularization extent between HT autograft and soft-tissue allografts based on conventional, contrast-enhanced, and quantitative MRI [34, 37, 39, 40]. However, at medium- to long-term follow-up (> 12 months), second-look arthroscopy revealed superior graft integrity, tension, and synovial coverage in HT autografts [38, 50]. This finding was further corroborated by quantitative MRI, which showed significantly lower T1rho and T2 relaxation times in HT autografts, indicating superior graft maturity [39]. Detailed results are summarized in Table 4.

Table 4.

Maturity assessment outcomes between HT autografts and soft-tissue allografts

Lead author (year)	Method of maturity assessment	Essential outcomes
Short-term follow-up (≤ 12 months)
Cusumano (2022) [34]	SNQ = (SIgraft - SIQT) / SIbackground	Mean SNQ was significantly higher in HT autografts compared with PL or TP allografts at 6 months postoperatively; No significant difference was observed between the two groups at 12 months
Kim (2021) [37]	SIRpost-contrast = SIgraft / SIgastrocnemius	No significant difference in revascularization based on dynamic contrast-enhanced MRI analysis between HT autografts and TA allografts at 1 year postoperatively
Li (2017) [40]	SNQ = (SIgraft - SIQT) / SIbackground	No significant difference in SNQ between HT autografts and TA allografts at 3, 6, and 12 months postoperatively
Medium- to long-term follow-up (> 12 months)
Kim (2018) [38]	Graft integrity, graft tension, synovial coverage, vascularization	Graft integrity, tension, and vascularization were significantly better in HT autografts compared with TA allografts at a mean 22.5-month follow-up
Yoo (2017) [50]	Graft integrity, synovial coverage	Synovial coverage was significantly better in HT autografts compared with TA allografts at a mean 18.1-month follow-up
Lansdown (2020) [39]	T1rho and T2 relaxation times	No significant differences in T1rho and T2 relaxation time between HT autografts and soft-tissue allografts at 6 and 12 months after surgery. T1rho and T2 relaxation times were significantly lower in HT autografts at 24 and 36 months postoperatively

HT, hamstring tendon; MRI, magnetic resonance imaging; PL, peroneus longus tendon; QT, quadriceps tendon; SI, signal intensity; SIR, signal intensity ratio; SNQ, signal-to-noise quotient; TA, tibialis anterior tendon; TP, tibialis posterior tendon

HT autograft versus QTB/QT autograft

Two studies compared HT autografts with QTB autografts [41, 49], while two others compared HT autografts with QT autografts [32, 45]. As shown in Table 5, QTB autografts consistently demonstrated lower SI between 6 and 24 months postoperatively, suggesting superior graft maturity compared to HT autografts [41, 49]. In studies evaluating QT autografts without bone block, a distinct temporal pattern in SI was observed: QT autografts showed higher SI than HT autografts at 3 months, comparable SI at 6 and 9 months, and significantly lower SI by 12 months. Notably, QT autografts showed a more pronounced SI decline over the period from 3 to 12 months [32, 45]. These studies indicate that QT autografts, especially when incorporating a bone block, may undergo a more rapid maturation process.

Table 5.

Maturity assessment outcomes between HT autografts and QT/QTB autografts

Lead author (year)	Method of maturity assessment	Essential outcomes
HT autograft vs. QT autograft
Aitchison (2021) [32]	SIR = SIgraft / SIPCL	No significant difference in SIR between QT and HT autografts at 6 months postoperatively. SIR of QT autografts was significantly lower than that of HT autografts by 12 months
Panos (2021) [45]	SIgraft	SI was significantly higher in QT autografts compared with HT autografts at 3 months, but no significant difference between the two graft types by 9 months postoperatively
HT autograft vs. QTB autograft
Ma (2015) [41]	SNQ = (SIgraft - SIPCL) / SIbackground	Mean SNQ was significantly lower in QTB autografts compared with HT autografts at the 6-month follow-up
Yamasaki (2024) [49]	Subjective	QT autografts with bone block exhibited significantly lower SI than HT autografts at a median 2-year follow-up

HT, hamstring tendon; PCL, posterior cruciate ligament; QT, quadriceps tendon; QTB, quadriceps tendon with patellar bone block; SI, signal intensity; SIR, signal intensity ratio; SNQ, signal-to-noise quotient

HT-PI graft versus FHT autograft

Three studies used conventional MRI to compare graft maturity between HT autografts with preserved tibial insertion and free HT autografts. As summarized in Table 6, HT-PI grafts demonstrated significantly lower SI at 6 and 12 months postoperatively [47, 48, 51], suggesting superior early maturation. This enhanced early incorporation is likely attributable to the preservation of the graft’s native vascular supply through the maintained tibial insertion, which may facilitate more rapid biological remodeling. However, this early advantage appears to diminish over time, as no significant SI differences were observed between the groups at 24 and 60 months postoperatively [51], indicating that free HT autografts eventually achieve comparable maturation levels.

Table 6.

Maturity assessment outcomes between HT-PI and free HT autografts and OTHER GRAFT COMPARISONS

Lead author (year)	Method of maturity assessment	Essential outcomes
HT-PI graft vs. Free HT autograft
Ruffilli (2016) [47]	Subjective	Graft SI and morphology were significantly better in HT-PI grafts compared with free HT autografts at the 6-month follow-up
Vari (2023) [48]	SNQ = (SIgraft - SIPCL) / SIbackground	Mean SNQ was significantly lower in HT-PI grafts compared with free HT autografts at 1 year postoperatively
Zhang (2020) [51]	SNR = SIgraft / SIbackground	Mean SNR was significantly lower in HT-PI grafts compared with free HT autografts at 6 and 12 months after surgery. No significant difference between the two graft types at 24 and 60 months postoperatively
Other grafts
Muramatsu (2008) [44]	SNQpost-contrast = (SIgraft - SIQT) / SIbackground	Contrast enhancement based on contrast-enhanced MRI was more pronounced in BPTB autografts compared with allografts during the 12 months after surgery
Rose (2017) [46]	SNQ = (SIgraft - SIQT) / SIbackground	No significant difference in mean SNQ between HT and TA allografts at 6 months after surgery

BPTB, bone-patellar tendon-bone; HT, hamstring tendon; HT-PI, hamstring tendon with preserved tibial insertion; MRI, magnetic resonance imaging; PCL, posterior cruciate ligament; QT, quadriceps tendon; SI, signal intensity; SNQ, signal-to-noise quotient; SNR, signal-to-noise ratio; TA, tibialis anterior tendon

Other graft comparisons

One contrast-enhanced MRI study compared revascularization between BPTB autografts and allografts, revealing an earlier onset and higher rate of revascularization in autografts [44]. Another study compared different allograft types and found no significant difference in SI between HT and TA allografts at 6 months postoperatively [46]. The detailed results are presented in Table 6.

Discussion

The main finding of this systematic review is that the maturation process of ACL grafts, as evaluated via MRI, second-look arthroscopy, and histological biopsy, demonstrates a time-dependent progression with substantial variation across different graft types. Comparative analyses suggest that maturity outcomes between BPTB and HT autografts remain inconclusive. In contrast, quadriceps tendon autografts, particularly those with a bone block, and hamstring tendon autografts with preserved tibial insertion may mature earlier than conventional HT autografts and soft-tissue allografts.

Previous studies consistently describe a process wherein implanted grafts progressively shed their “tendinous” biological characteristics while simultaneously acquiring more and more “ligamentous” histological properties [13]. Histological analysis of biopsy specimens described the changes concerning cellular, extracellular matrix, and vascularization, which remains the gold standard for assessing graft maturity [13, 25]. In contrast, second-look arthroscopy provides gross findings based on graft integrity, tension, and synovial coverage. While integrity and tension mainly reflect graft mechanical properties, synovial coverage serves as a measure of biological maturity [18]. Nevertheless, the invasive nature of both biopsy procedure and second-look arthroscopy impedes their routine application in clinical practice. As a non-invasive imaging modality, MRI has been widely adopted clinically to monitor the graft quality and indirectly assess the graft maturity after ACL reconstruction [52]. Signal intensity, an MRI parameter reflective of tissue characteristics and water content, has been linked to graft vascularity [53]. Weiler et al. [21] in a sheep model investigated the value of MRI in predicting ACL graft vascularity, biomechanical properties, and in observing the ligamentization process. They found that the temporal changes in MRI SI of tendon grafts corresponded to the maturation process determined histologically. In the initial phase, graft SI continuously increased, coinciding with the maximum density of vascularity, cellularity, and extracellular matrix disorganization in histology. Subsequently, the SI gradually decreased, indicating graft maturation and restoration of structural properties resembling those of the native ACL. Furthermore, a significant negative linear correlation was observed between MRI SI and biomechanical parameters, reflecting the progression of mechanical properties during maturation. It is noteworthy that the current evidence supporting MRI-derived parameters as predictors of graft structural properties following ACL reconstruction is primarily derived from animal studies. To date, no human studies have directly correlated MRI metrics with either structural or histological outcomes. Therefore, the clinical translatability of these animal findings to human graft maturity should be interpreted more cautiously.

The graft maturation process is a continuous biological process rather than a series of distinct, time-specific biological events. Monitoring this process and accurately identifying its different phases is clinically valuable, particularly for guiding postoperative rehabilitation protocols and determining the optimal timeline for return to preinjury sports activities. Since graft biological failure has been considered as a possible cause of retear [54–57], making adherence to the maturation timeframes is vital to minimize failure rates. Researche indicates that the maturation process may persist for two years or longer [19]. Consequently, the pursuit of grafts with faster maturation profiles represents a logical approach to facilitating patients earlier return to sports. BPTB autografts are often regarded as advantageous due to their stronger graft material properties, less graft tunnel widening, shorter fixed distance, lower graft failure rates, and direct bone-to-bone integration [58, 59]. Therefore, it is plausible that these advantages of BPTB could promote more advanced graft maturity compared to the HT autografts. However, based on the evidence compiled in this systematic review, no consensus has been established concerning the relative rate and extent of maturity between these two autograft types.

Some animal studies have demonstrated that extensive necrosis occurs in the central region of the graft during the early phase following ACL reconstruction, which potentially compromises its biomechanical properties [12, 60]. To bypass the necrosis phase, some surgeons have proposed using vascularized grafts for ACL reconstruction, such as HT autografts with preserved tibial insertion [61, 62]. Indeed, stable vascular supply plays a crucial role in maintaining tendon viability, thus preventing the phase of graft necrosis. In addition, preserving the tibial insertion of the HT permits early graft vascularization through the medial inferior genicular artery, which in turn facilitates the remodeling process [63]. In this systematic review, three studies consistently demonstrated superior maturity in vascularized HT-PI grafts compared to free HT autografts during the early postoperative period based on MRI evaluation [47, 48, 51]. Therefore, preserving the tibial insertion during hamstring tendon harvest should be considered a valuable surgical strategy to enhance early graft vascularization and maturation.

Recently, QT and QTB autografts have gained attention as viable alternatives. Volumetric analysis of three-dimensional MRI models showed that QT grafts possess significantly greater intra-articular volume compared to patellar tendon grafts of equivalent width [64]. Graft anatomical and biomechanical analysis further demonstrateed that the QTB graft exhibits significantly superior cross-sectional area, ultimate load to failure, and stiffness [65, 66]. These anatomical and mechanical advantages are closely linked to their unique biological characteristics. In this review, QTB autografts demonstrated significantly lower SI than HT autografts at 6 months and 2 years postoperatively. Studies by Aitchison et al. [32] and Panos et al. [45] further revealed that although QT autografts exhibited higher SI than HT autografts at 3 months, they showed significantly lower SI by 12 months, accompanied by a more pronounced SI decrease over this period. The current results suggested that the QT autografts, particularly those with a bone block, may mature earlier than the HT autografts and possess superior structural and mechanical properties.

Previous systematic reviews primarily focused on comparing remodeling time frames after ACL reconstruction across studies, or were limited to analyzing a single modality for assessing graft maturity [13, 18, 25, 67, 68]. To our knowledge, the present study represents the first attempt to comprehensively and systematically compare the maturation process of different graft types through an integrated analysis of histological findings, second-look arthroscopy, and MRI. The preliminary findings of this study suggest that QT autografts, particularly those with a bone block, and HT-PI grafts may undergo an accelerated maturation process. This implies that clinicians can develop more individualized rehabilitation protocols, particularly for young and active patients with high functionaldemand, potentially facilitating an earlier and safer return to sports [69]. However, it should be noted that MRI protocols (including scanner characteristics and acquisition parameters) and methods for quantifying graft SI varied considerably across the included studies. Coupled with inconsistencies in surgical techniques, graft dimensions, rehabilitation protocols, and activity levels between studies, these methodological heterogeneities restrict direct cross-study comparison of maturity and prevent meaningful meta-analysis. Therefore, these preliminary observations warrant confirmation through well-designed prospective cohort studies or randomized controlled trials.

Several limitations of this systematic review should be acknowledged. First, this systematic review did not include a meaningful meta-analysis due to substantial heterogeneity in MRI protocols, arthroscopic evaluation criteria, and histological methods across the included studies. Second, the lack of standardized intervals complicates the precise time frames of the continuous maturation process and affects the comparability of maturity outcomes across studies. Third, the obtained data comprised multiple sub-analyses within the included studies, potentially affecting the comprehensiveness and accuracy of the search results. Fourth, our restriction to English-language studies may have introduced a language bias, potentially overlooking relevant data published in other languages. Fifth, none of the included studies concurrently employed MRI and histological biopsy, thereby precluding definitive correlations between imaging signals and histologic stages of graft ligamentization. Consequently, further studies with standardized imaging protocols (field strength, coil design, sequence types, acquisition parameters, and normalized SI), larger sample sizes, fixed follow-up intervals, and prospective multicenter validation are necessary to correlate graft SI temporal changes with histologic maturation timelines.

Conclusion

Integrated assessment via MRI, second-look arthroscopy, and histological biopsy analysis indicated distinct graft maturation levels following ACL reconstruction. While no conclusive evidence was established the superiority of either HT or BPTB autografts in terms of maturity, QT autografts, particularly those with a bone block, and HT autografts with preserved tibial insertion may mature earlier compared to free HT autografts and other soft-tissue allografts. These findings suggest that these grafts could be regarded as potential alternatives in graft selection. Looking forward, MRI assessment of graft maturity emerges as a promising method for guiding safe functional recovery, pending validation through future prospective studies.

Abbreviations

ACL

Anterior cruciate ligament

BPTB

Bone-patellar tendon-bone

HT

Hamstring tendon

QT

Quadriceps tendon

QTB

Quadriceps tendon with a patellar bone block

MRI

Magnetic resonance imaging

SI

Signal intensity

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

ROI

Region of interest

SNQ

Signal-to-noise quotient

SIR

Signal intensity ratio

SNR

Signal-to-noise ratio

HT-PI

HT autografts with preserved tibial insertion

FHT

Free HT autografts

TA

Tibialis anterior tendon

TP

Tibialis posterior tendon

PL

Peroneus longus tendon

EI

Enhancement index

PCL

Posterior cruciate ligament

Author contributions

YP screened, extracted, and analyzed the data, and wrote the original manuscript text. SX screened and verified the data. GX and KZ reviewed the manuscript text. TS designed the study, wrote and reviewed the manuscript text. All authors reviewed the manuscript.

Funding

Peking University People’s Hospital Research Development Funds, Grant/Award Number: RDL2024-15.

Data availability

The datasets used and/or analyzed 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 that they have no competing interests.