Autograft maturation assessed by sequential quantitative MR T2 mapping and its correlation with patient-reported outcomes and return to sports during the first year after anterior cruciate ligament reconstruction

Abstract

Background

Although quantitative T2 mapping magnetic resonance imaging (MRI) is recognized as a non-invasive method for assessing graft maturity after anterior cruciate ligament (ACL) reconstruction, its longitudinal changes and association with clinical outcomes within the first postoperative year are not well established. The purpose of this study was to prospectively investigate hamstring autograft maturity via sequential quantitative T2 mapping and to examine its relationship with patient-reported outcomes and return to sports during the first year after ACL reconstruction.

Methods

Twenty-seven patients undergoing primary ACL reconstruction with hamstring tendon autografts were enrolled for MRI scans at 3, 6, and 12 months after surgery, and 15 patients with healthy ACLs served as controls. The quantitative MRI-based T2 relaxation time and conventional MRI-based signal/noise quotient (SNQ) of the graft were calculated and correlated with the Lysholm score, International Knee Documentation Committee (IKDC) subjective score, Knee Injury and Osteoarthritis Outcome Score (KOOS), and return to preinjury sports levels.

Results

T2 relaxation times and SNQ values of the grafts increased from 3 to 6 months (p < 0.01) and then decreased from 6 to 12 months postoperatively (p < 0.05). Graft T2 relaxation times were significantly lower than those of native ACLs at 3, 6, and 12 months postoperatively (p < 0.01). A significant negative correlation was observed between graft T2 relaxation time and the Lysholm score at 6 months postoperatively (p = 0.028), as well as with the KOOS pain subscale at 12 months (p = 0.023). Patients returning to preinjury sports levels at 12 months postoperatively had relatively lower graft T2 relaxation times (p = 0.045) and SNQ values (p = 0.043) compared to those who did not.

Conclusions

The graft T2 relaxation time and SNQ value increased during the first 6 months, followed by a subsequent reduction from 6 to 12 months after ACL reconstruction, and graft T2 relaxation time showed significant negative correlations with the Lysholm score at 6 months and with the KOOS pain subscale at 12 months postoperatively. Successful return to preinjury sports levels at 12 months was associated with lower graft T2 relaxation times and SNQ values. Quantitative MRI T2 mapping may provide an important assessment technique for monitoring graft maturation and guiding return to sports.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-025-06628-9.

Introduction

Anterior cruciate ligament (ACL) tears are among the most common sports-related injuries, and ACL reconstruction is the standard surgical intervention for these injuries [1–3]. A major unresolved challenge in postoperative management is the prevention of graft re-rupture following return to sports (RTS). Despite advancements in surgical techniques and standardized rehabilitation protocols, the reported rate of graft re-rupture remains high, ranging from 6 to 31% [4, 5]. Premature RTS during the graft remodeling and healing phase is widely considered a key risk factor for these failures [6]. As biological failure has been identified as a potential cause of re-rupture, adhering to biologically appropriate timelines is crucial for reducing the graft failure rate [7].

After ACL reconstruction, the graft undergoes functional adaptation and a continuous maturation process known as “ligamentization”, during which it transforms from a tendinous structure into a tissue resembling the native ACL [8, 9]. This biological process significantly affects the structural and mechanical properties of the graft [10]. Theoretically, completion of ligamentization process implies that the graft becomes histologically indistinguishable from the native ACL under light microscopy, which may extend over two years or more [11, 12]. However, most current rehabilitation protocols allow for return to sports between 9 and 12 months postoperatively [13–15]. This discrepancy between the prolonged biological timeline and earlier clinical decision-making point highlights the clinical importance of establishing a reliable and objective method to assess graft maturity before allowing patients to return to sports.

Although conventional magnetic resonance imaging (MRI) signal intensity (SI) is widely used to assess graft maturation after ACL reconstruction, it is highly susceptible to scanner manufacturers and MRI acquisition parameters, and fails to correlate consistently with clinical and functional outcomes [16–19]. In contrast, quantitative MRI T2 mapping has been reported to be strongly related to the water content, collagen content, and fibre structural alignment in soft tissue, while being largely independent of scanner and acquisition parameters [20, 21]. Previous studies have shown the feasibility of T2 mapping for objectively monitoring graft maturity [22–24]. However, the longitudinal changes in T2 relaxation time of ACL autografts relative to native ACLs, as well as their correlation with clinical outcomes and return to sports, remain incompletely elucidated [25–27].

Therefore, this study aimed to investigate the quantitative MRI-based T2 relaxation times and conventional MRI-based SI in hamstring autografts during the first year after ACL reconstruction, comparing them with native ACL values, and to examine the correlations between MRI maturity indicators and clinical outcome scores as well as return to preinjury sports levels. It was hypothesized that both the T2 relaxation time and SI would increase from baseline and then gradually decline as the grafts underwent the maturation process, approximating the native ACL values at final follow-up, and that higher graft maturity (lower T2 relaxation time and SI) would correlate with improved clinical scores and higher rates of return to preinjury sports levels.

Materials and methods

This was a prospective observational study, and ethical approval was provided by the Institutional Review Board of Peking University People′s Hospital. Written informed consent regarding data collection and publication was obtained from the participants.

Participant selection

Between January 2021 and November 2023, 35 patients who underwent ACL reconstruction were initially enrolled. Inclusion criteria comprised: (1) age between 18 and 50 years at the time of surgery, (2) unilateral ACL injury, (3) primary single-bundle ACL reconstruction using an ipsilateral four-stranded hamstring autograft. Exclusion criteria included: (1) history of previous ipsilateral knee surgery, (2) multi-ligament injury, (3) history of inflammatory arthritis, and (4) inability or unwillingness to comply with the scheduled follow-up assessments at 3, 6, and 12 months postoperatively.

Among the initial 35 patients screened, one was excluded due to age < 18 years, and five declined to participate. Twenty-nine patients consented to participate and were enrolled in the study. Two patients were further excluded for undergoing secondary arthroscopic surgery due to meniscus injury during follow-up. Finally, 27 patients completed all follow-up assessments and were included in the final analysis. The flowchart is summarized in Fig. 1. Additionally, 15 individuals with no prior knee ligament injuries were enrolled as a sex-, age-, and BMI-matched control group. They underwent a single MRI scan to establish the baseline MRI characteristics of native ACLs.

Fig. 1.

Flowchart of the study enrollment process. ACL, anterior cruciate ligament

Surgical procedure

All patients underwent ACL reconstruction with hamstring tendon autografts performed by a senior surgeon (TS). After diagnostic arthroscopy, associated meniscal injuries were treated as needed before ACL reconstruction. However, meniscal suturing was performed after ACL reconstruction to avoid suture failure caused by knee flexion and extension during the procedure. In this study, 7 patients presented with an intact meniscus, 9 underwent partial meniscectomy, and 11 underwent meniscal repair. Grafts were prepared by four-stranded double-looped semitendinosus and gracilis tendons, with diameters ranging from 7 to 9 mm (7 mm, n = 6; 8 mm, n = 20; 9 mm, n = 1). The femoral socket was drilled inside-out through the anteromedial portal in an anatomic position close to the anteromedial bundle footprint. The tibial tunnel was drilled in an outside-in manner from the anteromedial cortex to the center of the tibial ACL footprint. The graft was pulled through the tibial tunnel into the femoral tunnel. On the femoral side, the graft was fixed with a cortical suspension device (Endobutton CL Ultra, Smith & Nephew, USA). The knee was then cycled several times to tension the graft. The graft within the tibial tunnel was fixed using an interference screw (18 knees with BioSure from Smith & Nephew, USA, and 9 knees with Bio-Intrafix with a sheath from DePuy Mitek, USA).

Postoperative rehabilitation

All patients followed the same rehabilitation protocol. Partial weight-bearing immediately after surgery was allowed with crutches and a knee brace. Range-of-motion exercises for the knee joint were initiated at two weeks postoperatively, with full range of motion achieved by 6 weeks postoperatively. Swimming and cycling were permitted at 3 months. Return to sport-specific training and restoration of preinjury activity levels were allowed at least 9 months after surgery.

Clinical evaluation

Patient-reported outcome measures (PROMs), including the Lysholm score [28], International Knee Documentation Committee (IKDC) subjective score [29], and Knee Injury and Osteoarthritis Outcome Score (KOOS) [27, 30], were assessed preoperatively and at 3, 6, and 12 months postoperatively. The Tegner activity score (TAS) [31] was used to evaluate postoperative sports participation, and the patient was considered to have returned to the same sports level when the postoperative and preinjury TAS values were identical.

MRI evaluation

At each postoperative time point, all MRI examinations were performed on a 1.5-T MRI scanner (Magnetom Tim Trio; Siemens Medical Solutions, Erlangen, Germany) using a 15-channel knee coil at a single institution. Oblique sagittal images were acquired using a fat-saturated proton density-weighted imaging (FS-PDWI) sequence with the following parameters: repetition time (TR)/echo time (TE) = 2500 ms/40 ms; flip angle = 150°; matrix = 320 × 320; field of view (FOV) = 18 × 18 cm; slice thickness = 3 mm; bandwidth = 191 Hz; and scan time = 2 min. Additional oblique sagittal T2 mapping images were acquired using a multiple-echo spin-echo sequence with the following acquisition parameters: TR/TE = 2000 ms/13.8, 27.6, 41.4, 55.2, and 69 ms; flip angle = 180°, matrix = 256 × 256; FOV = 16 × 16 cm; slice thickness = 3 mm; bandwidth = 227 Hz; and scan time = 4 min 36 s. T2 maps were generated directly using a mono-exponential fitting algorithm based on pixel-by-pixel calculation in the inline software. To evaluate graft maturity, the SI and T2 relaxation time of the graft were calculated. The same sequences were used to assess native ACLs in the control group.

After image acquisition, all DICOM data were imported into RadiAnt DICOM Viewer 2024.1 software (Medixant, Poland). The sagittal slice with the clearest graft visualization was selected. The region of interest (ROI) outlining the intra-articular graft portion was manually delineated while excluding surrounding tissues. The SI represents the mean value across the ROI in the FS-PDWI images, and the mean SI of the entire intra-articular ACL graft was derived from three regions: distal, midsubstance, and proximal. To normalize the SI of the graft, the background signal (measured approximately 2 cm anterior to the tibial tuberosity) and posterior cruciate ligament (PCL) signal (measured near the tibial insertion) were acquired. The signal-to-noise quotient (SNQ) was calculated via the following equation: SNQ = (ACL graft SI - PCL SI)/background SI (Fig. 2a) [15, 32]. T2 relaxation times were measured directly on T2 maps referring to the same PD-weighted slice with clear ligament anatomy. Since T2 relaxation time reflects intrinsic tissue characteristics and requires no normalization, only the entire intra-articular graft T2 relaxation time was recorded (Fig. 2b). All measurements were performed twice at 2-week intervals by one orthopaedic surgeon (YP) for test–retest reliability. Interobserver reliability was assessed through independent measurements performed by a second orthopaedic surgeon (SX) using identical methodology.

Fig. 2.

Evaluation of intra-articular graft maturity on magnetic resonance imaging. a Sagittal proton density-weighted imaging showing the positions of the five regions of interest (area of the circle = 0.20 cm²), which include distal, midsubstance, proximal, posterior cruciate ligament, and background site (approximately 2 cm anterior to the tibial tuberosity). b Sagittal T2 mapping image showing the T2 relaxation time color map, where the T2 value (milliseconds) corresponds to the color bar

Statistical analysis

All statistical analysis was performed using SPSS 22.0 (IBM, Chicago, IL, USA) and GraphPad Prism 8.0.2 (GraphPad Software, La Jolla, CA, USA). Continuous variables conforming to normal distribution are reported as the mean ± standard deviation (SD) to one decimal place, whereas the median (range) was used for nonnormally distributed variables. Categorical variables are presented as frequencies (n, %). One-way repeated-measures analysis of variance (ANOVA) was used to compare the T2 relaxation time, SNQ, and PROMs across the three follow-up time points. The χ² test or independent samples t-test was used to compare differences between the ACL graft group and the native ACL group. One-way ANOVA was used to compare T2 relaxation time, SNQ values, and PROMs among the intact meniscus, partial meniscectomy, and meniscal repair groups. To control for potential confounding factors, including sex, age, body mass index (BMI), and meniscal lesions, partial correlation analysis was performed to assess the relationship between MRI-based maturity indicators and PROMs at the indicated time points. Intraclass correlation coefficient (ICC) was used to assess intra- and interobserver reliability for all measurements. ICC reliability was categorized as good (ICC > 0.75), marginal (0.4 ≤ ICC ≤ 0.75), or poor (ICC < 0.4).

An a priori power analysis was performed using G*Power 3.1 software (Heinrich-Heine University, Dusseldorf, Germany) to determine the sample size required for assessing longitudinal T2 relaxation time changes. Based on the F-statistic reported by Lansdown et al. (F (3, 114) = 11.0, p < 0.00001) [23], an effect size f of 0.54 was derived. With an alpha level of 0.05 and 95% power, a minimum sample size of 13 patients was calculated to be required. To mitigate potential attrition and strengthen the robustness of the statistical model, a larger cohort was enrolled.

Results

A total of 42 participants were enrolled, including 27 participants who underwent ACL reconstruction (patient group) and 15 participants with an intact native ACL (control group). Demographic characteristics of the participants are presented in Table 1. No significant differences were observed in age, gender, BMI, or involved side between the two groups.

Table 1.

Participant demographic characteristics

	Patient group (n = 27)	Control group (n = 15)	P value
Age	31.6 ± 8.9	34.6 ± 8.6	0.289
Gender (male/female)	19/8	10/5	1.0
BMI (kg/m2)	25.0 ± 3.2	24.3 ± 4.1	0.569
Affected side (left/right)	18/9	7/8	0.206
Meniscus treatment	Intact, n = 7 MS, n = 9 MR, n = 11

Data are expressed as mean ± standard deviations or patient numbers. BMI, body mass index. MR, meniscus repair; MS, partial meniscectomy

MRI evaluation

The T2 relaxation time and SNQ value of the intra-articular ACL graft portion during the postoperative period are shown in Fig. 3. T2 relaxation times increased significantly from 3 to 6 months (p < 0.001), and then decreased from 6 to 12 months (p = 0.01). Mean graft T2 relaxation times were significantly lower than those of native ACLs in uninjured controls at all time points (p < 0.001). Similarly, SNQ values increased from 3 to 6 months (p = 0.001) and then decreased from 6 to 12 months (p = 0.012). Mean graft SNQ values were significantly lower than those of native ACLs in controls at 3 months (p < 0.001) and 12 months (p = 0.001), but were not significantly different at 6 months (p = 0.084).

Fig. 3.

Temporal changes in a the mean T2 relaxation time and b the mean SNQ value of the ACL graft during 12 months after ACL reconstruction, and differences from those of the native intact ACL

When the cohort was stratified into intact meniscus (n = 7), partial meniscectomy (n = 9), and meniscal repair (n = 11) groups, all three groups demonstrated similar longitudinal trends in both T2 relaxation times and SNQ values. Specifically, in the meniscal repair group, T2 relaxation times increased significantly from 3 to 6 months, followed by a subsequent decrease from 6 to 12 months. There were no significant differences in graft T2 relaxation times and SNQ values among the three groups at 3, 6, and 12 months, respectively. Detailed MRI outcome data for all subgroups are presented in Supplementary Table 1. The ICCs for graft T2 relaxation time were 0.97 (intraobserver reliability) and 0.95 (interobserver reliability), whereas the ICCs for graft SNQ value were 0.97 (intraobserver reliability) and 0.96 (interobserver reliability).

Clinical outcome

As shown in Table 2, the Lysholm score, IKDC score, and all KOOS subscales increased significantly from preoperatively to 3, 6, and 12 months postoperatively. Significant improvements were also observed in all clinical scores between 3 and 6 months postoperatively and in Lysholm score, IKDC score, and KOOS subscales (sports/recreation and quality of life) between 6 and 12 months. In the subgroup analysis, the intact meniscus, partial meniscectomy, and meniscal repair groups all demonstrated similar trends in each clinical score. There was no significant difference between the groups at any of the assessed time points. Detailed clinical outcome data for the subgroups are presented in Supplementary Table 2.

Table 2.

Clinical outcomes preoperatively and postoperatively at each time point

	Follow-up				p value
	Preop	3 month	6 month	12 month	Preop vs 3-month	3-month vs 6 month	6-month vs 12month
Lysholm	41.4 ± 9.5	49.9 ± 4.1	71.9 ± 7.3	88.4 ± 5.3	0.001	< 0.001	< 0.001
IKDC	48.8 ± 8.9	54.2 ± 3.7	70.7 ± 4.9	83.3 ± 5.9	0.036	< 0.001	< 0.001
KOOS
Symptoms	69.3 ± 9.8	79.9 ± 4.9	90.3 ± 6.9	91.4 ± 5.9	< 0.001	< 0.001	1.000
Pain	61.5 ± 12.7	81.4 ± 6.8	93.4 ± 3.9	94.2 ± 3.3	< 0.001	< 0.001	0.640
ADL	73.8 ± 8.4	82.1 ± 5.4	95.6 ± 2.1	96.5 ± 2.1	< 0.001	< 0.001	0.291
Sports	35.4 ± 14.4	45.2 ± 6.3	62.6 ± 7.9	80 ± 10.7	0.002	< 0.001	< 0.001
QoL	30.6 ± 12.4	41.7 ± 7.9	55.8 ± 9.3	70.9 ± 11.0	< 0.001	< 0.001	< 0.001

Data are expressed as mean ± standard deviations. IKDC, International Knee Documentation Committee; KOOS, Knee Injury and Osteoarthritis Outcome Score; ADL, activities of daily living; QoL, quality of life;

Correlation with clinical outcome and return to sports

After controlling for potential confounders, significant negative correlations were observed between T2 relaxation times and Lysholm scores at 6 months postoperatively (p = 0.028), and between T2 relaxation times and KOOS pain subscale at 12 months postoperatively (p = 0.023). However, no significant correlations were observed between SNQ values and clinical outcome measures at any of the assessed time points (Supplementary Table 3).

At the 12-month follow-up, 37% (10/27) of patients achieved preinjury TAS levels. The median TAS was 7 (range 4–7) before injury, 2 (range 1–2) at 3 months, 4 (range 2–5) at 6 months, and 5 (range 3–7) at 12 months after surgery. Neither the T2 relaxation time nor the SNQ value correlated with the TAS at any follow-up time point (Supplementary Table 3). However, the ACL graft T2 relaxation time and SNQ value were significantly lower in patients who achieved preinjury TAS levels than in those who did not (Table 3).

Table 3.

Comparison of mean T2 relaxation time and SNQ value at the 12 month follow-up

	RTS (n = 10)	Non-RTS (n = 17)	p value
T2 relaxation time (ms)	35.6 ± 2.8	39.2 ± 4.8	0.045
SNQ value	10.9 ± 5.0	15.6 ± 5.7	0.043

Data are expressed as mean ± standard deviations. RTS, return to sports; SNQ, signal/noise quotient

Discussion

The most important finding of this study is that the mean T2 relaxation times of ACL grafts increased significantly from 3 to 6 months, followed by a marked decrease from 6 to 12 months after ACL reconstruction, which is consistent with the temporal trend in SNQ values. Furthermore, graft T2 relaxation time showed significant negative correlations with the Lysholm score at 6 months and with the KOOS pain subscale at 12 months postoperatively, and successful return to preinjury sports levels at 12 months was associated with lower graft T2 relaxation times and SNQ values.

MRI SI is widely employed to evaluate graft maturity following ACL reconstruction [18, 33–36]. It correlates significantly with the biomechanical properties of ACL grafts in animal models [21], as well as with the maturation process in humans [13, 37]. To minimize interscan variability, SI is commonly normalized to reference structures such as the PCL, quadriceps tendon, or background noise, yielding metrics such as the SNQ [14–16, 33], signal intensity ratio (SIR) [38], ACL/PCL signal ratio, and ACL/muscle signal ratio [13]. Although these normalized measures are widely used, they remain susceptible to variations in scanner hardware, magnetic field strength, and coil characteristics. In addition, ultrasound also represents a potential non-invasive method for ACL evaluation. Currently, the application of ultrasound following ACL reconstruction focuses primarily on assessing cartilage damage, quadriceps muscle changes, and infrapatellar fat pad thickness. However, its utilization specifically for assessing graft maturity remains relatively limited in both research and clinical practice [39]. Quantitative approaches using T2 relaxation times allow the resulting data to be independent of equipment or acquisition parameters, and solely dependent on magnetic field strength [20, 21]. However, previous comparative studies between 3 and 7 T systems using quantitative mapping protocols for evaluating native ACLs have demonstrated that 7 T MRI analysis did not show superior performance over 3 T MRI analysis [40]. In this study, the use of consistent MRI equipment and coils helped minimize such variability; nevertheless, multicenter comparisons involving different magnetic field strengths remain challenging.

Quantitative MRI techniques such as T2 mapping directly reflect collagen content and organizational state within the graft, with lower values indicating more well-organized tissue structure and higher values reflecting greater disorganization [21, 25]. Typically, T2 relaxation times increase during the early remodeling phase after ACL reconstruction and subsequently decrease as the graft matures [23, 41]. Consistent with this pattern, the current study observed an increase in T2 relaxation times from 3 to 6 months, followed by a decline through 12 months, and this trend aligned with the changes in the SNQ value at the corresponding postoperative time points. However, existing literature presents inconsistent trends in T2 relaxation times during graft maturation. Lansdown et al. [23] investigated longitudinal changes in T2 relaxation times between 6 and 36 months following ACL reconstruction. They observed a significant decrease in T2 relaxation times from 6 to 12 months, with no further changes noted through 36 months. In a study by Niki et al. [22], T2 relaxation times were evaluated at 3, 6, and 12 months after double-bundle ACL reconstruction; a significant decrease from 3 to 6 months was reported exclusively in the anteromedial bundle. Conversely, Marchiori et al. [24] found no significant differences in graft T2 relaxation time at 4 versus 18 months postoperatively. The established influence factors for graft maturation include graft material, concomitant injury, surgical technique, and rehabilitation protocols [42–45]. Graft remodeling is a continuous biological process in which cells, vessels, and extracellular matrix dynamically change, and T2 relaxation time discrepancies might be attributed to differences in follow-up time. The associations between the graft T2 relaxation time and anatomical, surgical, and functional factors (tibial tunnel malposition, sagittal obliquity of the ACL, lateral tibial posterior slope, hamstring strength, etc.) require further investigation.

To mitigate the risk of early biological failure, recent studies have explored correlations between MRI findings and clinical outcomes. However, the relationship between MRI-based graft maturity indicators and PROMs remains inconclusive [23, 33, 37]. With respect to conventional MRI SI, Li et al. [33] observed no significant correlation between the graft SNQ value and Lysholm, IKDC, or Tegner scores within the first postoperative year. In contrast, Biercevicz et al. [37] demonstrated that a combination of graft volume and median SI predicted KOOS symptoms, pain, sports/recreation, and quality of life subscales in multiple linear regression at 5-year follow-up. In the present study, after controlling for confounding factors including sex, age, BMI, and meniscal status, no significant associations were identified between the SNQ values and PROMs (Lysholm, IKDC, or KOOS) at 3, 6, or 12 months postoperatively. However, significant negative correlations were observed between graft T2 relaxation time and the Lysholm score at 6 months, as well as with the KOOS pain subscale at 12 months postoperatively. This finding is partially consistent with that of Lansdown et al. [23], who reported significant correlations between T2 relaxation times and multiple KOOS subscales at 2 years after ACL reconstruction. T2 mapping captures subtle structural changes within the graft that correlate with clinical outcomes, offering an important adjunct method for guiding safe return to sports after ACL reconstruction.

Regarding RTS, premature resumption significantly increases the graft re-rupture rate after ACL reconstruction. It has been reported that the probability of ipsilateral or contralateral ACL reinjury is 19.4% for athletes who resume their sport 9 months after the surgical procedure, and this likelihood increases to 26.4% for those who return earlier [46, 47]. Currently, there is no consensus on the optimal criteria for determining a patient’s readiness to RTS. Most studies primarily rely on the duration following surgery, along with a series of assessment measures as the critical criteria for initiating RTS training. These included reaching at least 9 months, performing over 2 functional tests (such as single leg hop test achieving more than 90% of the contralateral side, triple hop test, and knee stability assessment), completing psychological readiness evaluations, and meeting quadriceps and hamstring index strength testing more than 90% [48, 49]. Regrettably, most RTS criteria primarily emphasize muscle strength around the knee and psychological readiness, rather than focusing on the inherent characteristics of the ACL graft itself, such as its mechanical strength and structural integrity [50]. Therefore, MRI offers a non-invasive assessment method that can provide more information about the biological properties of the graft itself. Zhou et al. [51] utilized MRI T2* mapping and found that the T2* values of the graft in the RTS group were significantly lower than those in the non-RTS group at 9 months postoperatively. Furthermore, they calculated that a T2* value of 16.65 predicted patients with failed RTS with a sensitivity of 67.9% and a specificity of 88.2%. Lutz et al. [13] reported that patients who return to the preinjury sports level at 2 years after surgery exhibited significantly lower SI compared to those who did not achieve the same sports level. Similarly, in the present study, we also observed that at the 12 month postoperative follow‑up, the ACL graft T2 relaxation time and SNQ value were significantly lower in patients who achieved preinjury TAS levels than in those who did not. It should be noted that, although our study found statistically significant lower T2 values in patients who successfully returned to sports, there is currently no a validated minimal clinically important difference for graft T2 time. Future large-scale longitudinal studies that correlate T2 relaxation times with “hard” endpoints such as graft re-rupture rates are needed to establish definitive thresholds for guiding individualized rehabilitation and return-to-sport clearance.

There were several limitations to the present study. First, the duration of clinical follow-up was relatively short, with assessments confined to three time points within one year. Given that graft maturation is a prolonged biological process, graft maturity on the basis of MRI appearance should be followed up for 2 years or longer. Second, the sample size in each subgroup was limited. Although the within-group one-way repeated-measures ANOVA revealed a statistically significant main effect of time, post hoc pairwise comparisons with Bonferroni correction failed to identify significant differences between specific time points. Third, variations in postoperative RTS levels, knee stability, muscle strength, and anatomical factors (such as tibial tunnel malposition, sagittal graft obliquity, and lateral tibial posterior slope) represent potential confounding factors when investigating the correlation between graft maturity and clinical outcomes. Future studies with larger sample sizes, more stringent inclusion criteria, and rigorous control of confounding variables are required to conduct more reliable analyses of graft maturation. Fourth, second-look arthroscopy and histology evaluations were lacking. Therefore, the corresponding relationship of the time frame between MRI and histological changes could not be obtained.

Conclusion

In conclusion, both quantitative MRI-based T2 relaxation times and conventional MRI-based SNQ values of ACL grafts exhibited significant temporal changes over the first postoperative year, demonstrating similar trends throughout the follow-up period. Notably, graft T2 relaxation time showed significant negative correlations with the Lysholm score at 6 months and with the KOOS pain subscale at 12 months postoperatively. Successful return to preinjury sports levels at 12 months was associated with lower graft T2 relaxation times and SNQ values. These findings suggest that incorporating T2 mapping alongside conventional SNQ evaluation and clinical assessments could enhance the objective monitoring of graft maturation during the first year after surgery and provide valuable guidance for safe return to sports following ACL reconstruction.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ACL

Anterior cruciate ligament

BMI

Body mass index

FS-PDWI

Fat-saturated proton density-weighted imaging

FOV

Field of view

ICC

Intraclass correlation coefficient

IKDC

International Knee Documentation Committee

KOOS

Knee Injury and Osteoarthritis Outcome Score

MRI

Magnetic resonance imaging

PROMs

Patient-reported outcome measures

PCL

Posterior cruciate ligament

ROI

Region of interest

RTS

Return to sports

SD

Standard deviation

SI

Signal intensity

SIR

Signal intensity ratio

SNQ

Signal-to-noise quotient

TAS

Tegner activity score

TE

Echo time

TR

Repetition time

Author contributions

YP collected and analyzed the data and wrote the original manuscript text. SX collected and analyzed the data. GX and KZ collected the data. 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

The ethical approval was provided by the Institutional Review Board of Peking University People’s Hospital. The approval number is 2020PHB241-1.

Consent for publication

Written informed consent was obtained from all individual participants included in the study.

Competing interests

The authors declare that they have no competing interests.