Magnetic Resonance Imaging Assessment of Hamstring Graft Healing and Integration 1 and Minimum 2 Years after ACL Reconstruction

Abstract

Background:

An increase has been seen in the number of studies of anterior cruciate ligament reconstruction (ACLR) that use magnetic resonance imaging (MRI) as an outcome measure and proxy for healing and integration of the reconstruction graft. Despite this, the MRI appearance of a steady-state graft and how long it takes to achieve such an appearance have not yet been established.

Purpose:

To establish whether a hamstring tendon autograft for ACLR changes in appearance on MRI scans between 1 and 2 years and whether this change affects a patient’s ability to return to sports.

Study Design:

Case series; Level of evidence, 4.

Methods:

Patients with hamstring tendon autograft ACLR underwent MRI and clinical outcome measures at 1 year and at a final follow-up of at least 2 years. MRI graft signal was measured at multiple regions of interest using oblique reconstructions both parallel and perpendicular to the graft, with lower signal indicative of better healing and expressed as the signal intensity ratio (SIR). Changes in tunnel aperture areas were also measured. Clinical outcomes were side-to-side anterior laxity and patient-reported outcome measures (PROMs).

Results:

A total of 42 patients were included. At 1 year, the mean SIR for the graft was 2.7 ± 1.2. Graft SIR of the femoral aperture was significantly higher than that of the tibial aperture (3.4 ± 1.3 vs 2.6 ± 1.8, respectively; P = .022). Overall, no significant change was seen on MRI scans after 2 years; a proximal graft SIR of 1.9 provided a sensitivity of 96% to remain unchanged. However, in the 6 patients with the highest proximal graft SIR (>4) at 1 year, a significant reduction in signal was seen at final follow-up (P = .026), alongside an improvement in sporting level. A significant reduction in aperture area was also seen between 1 and 2 years (tibial, –6.3 mm², P < .001; femoral, –13.3 mm², P < .001), which was more marked in the group with proximal graft SIR >4 at 1 year and correlated with a reduction in graft signal. The patients had a high sporting level; the median Tegner activity score was 6 (range, 5-10), and a third of patients scored either 9 or 10. Overall, PROMs and knee laxity were not associated with MRI appearance.

Conclusion:

In the majority of patients, graft SIR on MRI did not change significantly after 1 year, and a proximal graft SIR <2 was a sensitive indicator for a stable graft signal, implying healing. Monitoring is proposed for patients who have a high signal at 1 year (proximal graft SIR >4), because a significant reduction in signal was seen in the second year, indicative of ongoing healing, alongside an improvement in sporting level. A reduction in tunnel aperture area correlated with a reduction in graft SIR, suggesting this could also be a useful measure of graft integration.

Keywords

ACL ACL reconstruction hamstring tendon autograft MRI graft healing graft integration

Increasing interest has arisen regarding whether magnetic resonance imaging (MRI) can be used as a tool to assess graft healing after anterior cruciate ligament (ACL) reconstruction (ACLR) and in turn help guide patients in a return to functional activity and sports.

Correlating MRI graft signal with ACLR graft strength was initially developed in animal models.^2,34 High graft signal on MRI correlated with the histological presence of new hypervascular and hypercellular reparative and weaker tissue. This graft signal was seen to gradually reduce, suggesting remodeling, until by 2 years the graft had taken on a signal similar to that of the native ACL.³⁴ Human histopathological studies have demonstrated that ACLR graft remodeling appears to be an ongoing process beyond 1 year³¹; however, given the morbidity and costs associated with a second-look arthroscopy, it is unsurprising that MRI appearance of the graft has become a proxy for healing and integration.

The literature differs in both the descriptions and the time frame taken for a hamstring tendon ACLR autograft to heal and integrate. Evidence indicates ongoing metabolic activity at 1 year¹² and shows that healing can continue up to, and beyond, 2 years.^5,37 Quantitative MRI using relaxation times has even revealed a progressive reduction in activity up to 3 years.¹³ Despite this, it is common for patients to have a target-based recovery with return to sport-specific training permitted at 9 to 12 months, with ongoing progression to pivoting sports without MRI evidence of healing.³²

In a 2019 systematic review assessing ACLR graft maturity using conventional MRI scans, 34 studies were found with very heterogeneous results reported.³⁰ Only a few studies had looked at MRI scans beyond 1 year, and no consensus was found regarding the best way to analyze the images. Thus, it is unsurprising that there is a lack of understanding on the progressive appearances of graft healing and integration over time.

High signal proximally in a hamstring tendon autograft at 1 year after ACLR has been correlated with subsequent graft rupture.²⁶ To best apply this finding clinically, it is important to determine whether such MRI results at 1 year represent a permanent state with a permanently higher risk of failure or whether this condition improves over time; this determination would have significant implications about the best way to manage a return to full activity and sports. A recent study concluded that 2-year MRI results had no correlation with functional outcome, anterior laxity, or patient age, but the study did not report graft rupture.³

A link has also been demonstrated between graft signal and bone tunnel appearance; an MRI study of 22 patients over the course of 2 years demonstrated a correlation between femoral tunnel widening and higher graft signal. Detailed analysis of the entire tunnel shape found the strongest correlation at the femoral aperture of the tunnel.³⁷

The current study aimed to establish whether the MRI appearance of a hamstring tendon autograft used in ACLR changes after 1 year and, if so, where this change occurs and whether it affects a patient’s return to sports. The primary hypothesis was that a high graft signal and tunnel aperture widening at 1 year would be reduced after 2 years if the patient had avoided a graft rupture and that these patients would see an improvement in sporting level.

Methods

Participants

The study used a cohort of 250 patients who had previously undergone a hamstring tendon autograft ACLR after an acute single-ligament ACL rupture and had undergone an MRI examination at 1-year follow-up. During the recruitment period between November 2018 and June 2019, all patients reaching 2-year follow-up were contacted and invited to participate in the study; this resulted in 60 sequential patients, of whom were 42 recruited (Figure 1).

All patients needed to provide approved written consent for ongoing institutional data collection, and although no age restriction was imposed, those younger than 16 years needed additional parental or guardian consent. The index procedure was performed by 1 of 3 senior fellowship-trained surgeons. Patients with meniscal or chondral procedures that were performed concurrently with ACLR were also included. Patients with a graft rupture or any further knee surgery were recorded but excluded from final analysis.

This study received ethical approval from the institutional review board Northern Sydney Local Health District reference HREC/17/ HAWKE/140.

Surgical Technique

All patients underwent an ipsilateral hamstring tendon autograft reconstruction. Two senior surgeons (D.A.P. and B.A.F.) performed adjustable suspensory graft fixation for both tunnels using a TightRope RT (Arthrex) for femoral fixation and a TightRope ABS (Arthrex) suture on the tibial side, where tensioning was performed against a cortical button. A 4-strand semitendinosus tendon graft was used, and if the initial graft diameter was <8 mm in male patients and <7 mm in female patients, a doubled gracilis tendon was also included to create a 6-strand graft. The third senior surgeon (MC) performed a 4-strand graft with suspensory femoral fixation using an Endobutton (Smith & Nephew) and screw and sheath tibial fixation using Intrafix (Smith & Nephew). All femoral sockets were created using inside-out drilling through the anteromedial portal, creating a proximally based tunnel predominately over the anteromedial ACL origin, in line with the previously described “IDEAL” (Isometric, in the Direct fibers, Equidistant and Eccentric, Anatomic, and Low in tension) position.²³ All tibial tunnels were created with outside-in drilling from the tibial cortex. Final tensioning was performed with the knee in full extension.

Postoperative Management

Patients were discharged on the day of surgery and allowed full weightbearing using crutches for comfort and balance until walking confidently without a limp (generally <2 weeks). A standardized postoperative rehabilitation protocol was initiated that aimed for controlled restoration of range of motion, muscle strength, and proprioception. Benchmarks were straight-line running at 3 months and when selected by patients, results of objective return-to-sport testing. All patients had a goal of returning to sport-specific training between 9 and 12 months and commencing competitive sports after 1 year.

Clinical Evaluation

At the 1-year evaluation, KT-1000 arthrometer measurements of knee laxity were taken using 134 N of force to allow side-to-side comparison. Patient-reported outcome measures (PROMs) were collected at 1 year and at a final follow-up of at least 2 years and consisted of the International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form,¹ Lysholm knee scoring scale,²⁹ and Tegner activity scale.²⁹

MRI Evaluation

An examination using high-resolution 3.0-T MRI was performed at 1 year and minimum 2 years using a single dedicated unit (Magnetom Skyra, with a 15-channel phased-array send/receive knee coil; Siemens AG Healthcare). This examination entailed a validated protocol that included volumetric and near-isotropic (0.5 × 0.5 × 0.65 mm) turbo spin echo sequences allowing multiplanar reconstructions parallel with the center of the intra-articular segment of the graft (Figure 2).

An image perpendicular to each tunnel aperture was reconstructed using the same technique, where complete circumferential tunnel walls were seen, in order to analyze the mean signal within a cross section of the aperture.

Signal intensity ratio (SIR) (the mean graft signal divided by the posterior cruciate ligament [PCL] signal) was measured and calculated using the mean average signal across a region of interest (ROI); this process was then repeated in multiple ROIs and used as a proxy for intra-articular graft healing and aperture integration. As described in previous studies,^4,20 the intra-articular graft was divided into 3 zones equally spaced apart: The distal graft ROI was set just proximal to the tibial aperture, and its distal edge was aligned with the tibial joint surface; the midsubstance graft ROI was aligned at the anterodistal intercondylar notch; and the proximal graft ROI was adjacent to the femoral aperture. The PCL signal was recorded from an ROI central to its broad distal tibial attachment (Figure 2).

Changes in femoral and tibial tunnel aperture area were evaluated using the reconstructed image perpendicular to the aperture, a measurement previously validated against computed tomography.⁷ First, changes at 1 year were determined as the difference between the 1-year MRI-derived measurement and the tunnel diameter drilled during surgery; second, changes after 2 years were determined as the difference between areas taken from the same level on 1- and 2-year MRI scans. The circular aperture ROI was used to calculate the cross-sectional area and was taken using the first reconstructed image slice placed perpendicular to the tunnel; the same ROI was used to calculate the aperture graft SIR (Figure 3).

A cutoff SIR of 4 on MRI scans of the proximal graft at 1 year after ACLR was previously identified as an appearance that gave a sensitivity and specificity of 66% and 77%, respectively, for subsequent graft rupture during pivoting sports.²⁶ Therefore, 2 cohorts were established after 1-year MRI analysis; those with a SIR of >4 in the proximal graft at 1 year, and those with SIR <4.

The study cohort with 2 complete MRI scans per patient was reanalyzed by the senior investigator (S.E.P.) for intraobserver reliability after a period of 1 month had passed. A random sample of 10 MRI scans from each time frame was analyzed independently by 2 further investigators (T.O. and S.G.) for interobserver validation.

Statistical Evaluation

Normally distributed variables are reported as mean ± SD, and nonnormally distributed variables are reported as median (range). Inter- and intraobserver reliabilities were analyzed using intraclass correlation coefficient with absolute agreement using a 2-way mixed model. Reliability was considered good if the intraclass correlation coefficient was ≥0.75 and excellent if it was ≥0.85.²⁸ Correlation between parameters was calculated using Pearson correlation. Graft signal between the 5 ROIs was compared using 1-way analysis of variance. For each ROI, the Youden index ([Sensitivity + Specificity] − 1) was used to detect an MRI appearance at 1 year that had the maximum effectiveness as a marker that the graft had reached a steady state, without significant change in signal, suggesting that the graft had healed.²⁷ Changes in tunnel aperture area were compared using the Mann-Whitney U test. The observed power in the regression analysis demonstrated sufficient size of the study sample and the observed effect.⁸ Statistical analysis was performed using SPSS Version 24.0 (IBM). Statistical significance was set at P < .05.

Results

Participants and Clinical Outcomes

Mean patient age at the time of surgery was 31.6 ± 13.6 years, and the mean body mass index was 24.3 ± 3.7. The population included 3 patients younger than 16 years: one 13-year-old male patient with open distal femoral and proximal tibial physes and 2 female patients aged 13 and 14 years who were at the final stages of epiphyseal fusion; these patients had the same operative approach as the adult population. Soon after the 1-year postoperative assessment, 1 graft rupture occurred during pivoting sports, and this patient was excluded from analysis after 1-year MRI review. At 1 year postoperatively, all 43 patients (100%) had fully completed PROMs and 37 (88%) had undergone KT-1000 arthrometer testing. Soon after the 1-year postoperative assessment, 1 graft rupture occurred during pivoting sports, and this patient was excluded from subsequent analysis. At final follow-up (mean, 25.5 ± 3.4 months; range, 24-36 months), all 42 patients (100%) had fully completed PROMs.

At 1-year assessment, the mean IKDC score was 80 ± 12, the mean Lysholm score was 88 ± 12, and the median Tegner score was 6 (range, 2-10). A total of 2 patients reported low PROMs at 1 year, with IKDC scores <60; both of them went on to show improvements with ongoing physical therapy by final follow-up (IKDC improvement from 46 to 61 and from 52 to 74, respectively). KT-1000 arthrometry at 134 N conducted at 1 year postoperatively revealed a mean stability of 6.4 ± 2.5 mm, a difference of 1.0 ± 1.8 mm when compared with the contralateral control knee. We found no significant change in Tegner or Lysholm score between 1-year assessment and final follow-up and a clinically small but statistically significant improvement in IKDC score (Table 1). We noted no correlation between KT-1000 arthrometer measurements and PROMs at any time point and no correlation between the appearance on MRI and PROMs.

Table 1

Clinical Outcomes at 1-Year Assessment and Final Follow-up of at Least 2 Years^a

Outcome Measure	1 Year	≥2 Years	P Value
IKDC	80 ± 12	85 ± 9	.034
Tegner score	6 (2-10)	6 (3-10)	.136
Lysholm score	88 ± 12	89 ± 9	.667
Side-to-side difference in KT-1000 arthrometer measurement, mm	1.0 ± 1.8	NR

Values are expressed as mean ± SD or median (range). IKDC, International Knee Documentation Committee Subjective Knee Score; NR, not recorded.

Sporting level at the final follow-up was high; 12 patients (29%) had a Tegner score of 9 or 10, indicating participation in professional pivoting sports (most commonly rugby, football, soccer, basketball, and volleyball).

MRI Outcomes

Graft signal calculation after selection of each ROI demonstrated excellent intra- and interobserver reliabilities (0.90 and 0.92, respectively).

At 1 year, the mean SIR for the full intra-articular graft was 2.7 ± 1.2. After 2 years, there was an overall slight reduction to 2.5 ± 0.8 (P = .150). Reviewing the change in appearance of each intra-articular graft ROI between MRI scans at 1 year and ≥2 years showed that the optimal sensitivity and specificity for the graft to remain unchanged occurred when a SIR of 1.9 was seen proximally (sensitivity, 96%; specificity, 69%; area under the receiver operating characteristic curve [AUC], 0.875). Other graft ROIs were less effective at predicting a steady state, with 2.3 centrally (sensitivity, 59%; specificity, 60%; AUC, 0.536), and 2.0 distally (sensitivity, 71%; specificity, 72%; AUC, 0.762). If the combined ROI for the whole intra-articular graft was used, a SIR of 2.4 had a sensitivity of 75% and specificity of 77% (AUC, 0.805) to remain unchanged (Figure 4).

The mean SIR at the tibial tunnel aperture at 1 year (2.6 ± 1.8) was significantly lower than the mean SIR at the femoral tunnel aperture at 1 year (3.4 ± 1.3) (P = .022), and this significant difference continued after 2 years (2.5 ± 1.1 for the tibial tunnel aperture vs 3.6 ± 1.1 for the femoral tunnel aperture; P < .001). Tunnel aperture areas at 1 year were compared with the original drill diameter, showing a mean increase of 7.3 ± 12.8 mm² at the tibial aperture and 19.5 ± 15.9 mm² at the femoral aperture. The mean aperture areas then significantly decreased between 1-year and minimum 2-year MRI measurements: the tibial tunnel aperture by −6.3 ± 15.1 mm² (P < .001) and the femoral tunnel aperture by −13.3 ± 14.0 mm² (P < .001) (Table 2; Figure 5). A significant decrease in tibial aperture area between 1 and 2 years was correlated with a significantly lower signal throughout the intra-articular graft.

Table 2

Changes in MRI Parameters at Specific Regions of Interest Between 1-Year Assessment and Final Follow-up of at Least 2 Years in the Full Cohort^a

	1 Year	≥2 Years	P Value
Tibial aperture
SIR	2.6 ± 1.8	2.5 ± 1.1	.756
Δ area, mm²	↑7.3 ± 12.8	↓6.3 ± 15.1	<.001
Intra-articular graft SIR
Distal	2.3 ± 1.3	2.0 ± 1.0	.239
Middle	2.7 ± 1.4	2.7 ± 1.2	.980
Proximal	2.5 ± 1.2	2.2 ± 0.9	.197
Full graft	2.7 ± 1.2	2.5 ± 0.8	.371
Femoral aperture
SIR	3.4 ± 1.3	3.6 ± 1.1	.449
Δ area, mm²	↑19.5 ± 15.9	↓13.3 ± 14.0	<.001

Values are expressed as mean ± SD. MRI, magnetic resonance imaging; SIR, signal intensity ratio; Δ area, change in area from original drill size or between MRI examinations (↑ = increased; ↓= reduced).

The group with a proximal graft SIR >4 on the 1-year MRI scan entailed 6 patients (mean SIR, 4.9 ± 0.9; range, 4.1-6.4; for the remaining patients, the mean SIR was 2.1 ± 0.7; range, 0.5-3.5). The patient who experienced a graft rupture after the 1-year MRI examination had a proximal graft SIR of 4.4 (whole graft SIR, 3.2). At final follow-up MRI examination, all 6 patients demonstrated a significant reduction in SIR (mean reduction, 2.9 ± 1.0; range, 1.7-4.7), alongside a significant reduction in tibial aperture area (P = .004) and femoral aperture area (P = .009) (Table 3; Figure 5).

Table 3

Changes in MRI Parameters Between 1-Year Assessment and Final Follow-up of at Least 2 Years in Those With Graft SIR >4 at 1 Year^a

	1 Year	≥2 Years	P Value
Tibial aperture
SIR	4.7 ± 3.7	3.2 ± 1.0	.013
Δ area, mm²	↑17.2 ± 13.9	↓21.3 ± 20.3	.004
Intra-articular graft SIR
Distal	4.0 ± 1.4	2.5 ± 1.2	.074
Middle	4.8 ± 1.5	2.9 ± 1.1	.031
Proximal	4.9 ± 0.9	2.9 ± 1.0	.005
Full graft	4.7 ± 1.2	3.2 ± 0.7	.026
Femoral aperture
SIR	5.5 ± 1.5	4.8 ± 0.9	.350
Δ area, mm²	↑19.6 ± 16.4	↓22.8 ± 19.2	.009

Values are expressed as mean ± SD. MRI, magnetic resonance imaging; SIR, signal intensity ratio; Δ area, change in area from original drill size or between MRI examinations (↑ = enlarged; ↓ = reduced).

An improvement was seen in PROM scores in this group of 6 patients, with the median Tegner significantly increased by 2 (median 5, range 2-7 at 1 year; median 7, range 5-9 at final follow-up).

Of the 42 patients in the final cohort, 38 patients had adjustable suspensory fixation on both the femur and the tibia and 4 patients had hybrid fixation hamstring tendon grafts. Cyst formation was seen in 5 of the suspensory tibial tunnels (13%) and 2 of the screw and sheath fixations (50%), and 1 of these hybrid cases also had femoral tunnel cyst formation.

Age had a positive correlation with MRI signal change between 1 and 2 years, but only at the proximal graft (P = .018; r = 0.364), with a greater reduction in proximal graft signal seen in older patients. No correlation was found between KT-1000 arthrometer readings and MRI graft signal at 1 year or change in graft signal at final follow-up.

Discussion

This study demonstrated that the majority of hamstring tendon autografts had reached their final MRI signal appearance at 1 year after hamstring tendon autograft ACLR, without significant change after 2 years. This was particularly the case if the intra-articular graft SIR was <2, and even more so if it was low in the proximal aspect.

In this study, we found no correlation between any of the measured MRI parameters at each time point and the corresponding PROMs; therefore, graft rupture may be the most important outcome measure.²⁶ For those patients who had high MRI signal at 1 year, however, we noted ongoing activity with a significant reduction in graft signal after 2 years, implying ongoing graft maturation. This finding was demonstrated by analysis of those patients with a proximal graft SIR >4. This group also contained the single patient with a graft rupture after 1 year, but the remaining patients saw significantly improved Tegner scores during this time (from median 5 to median 7). These findings may have significant implications for timing of safe return to sports in this group. MRI is therefore useful in this instance to provide serial information on the rate of graft healing and integration.

Identifying those patients who may take longer for graft healing and integration and are therefore at a higher risk when returning to pivoting sports is clearly important in order to reduce graft rupture rates.¹ This process may take longer than a year; MRI studies using ultrashort echo time enhanced T2* to evaluate ACLR graft maturation demonstrated that patients showed a significant reduction in relaxation times between 1 and 2 years, suggestive of ongoing graft maturation.⁵ Similar findings over a 3-year period have also been demonstrated.¹³

A secondary finding of this study is an overall reduction in tibial and femoral tunnel aperture area occurring after the first postoperative year, which was correlated with a decrease in graft signal. Tunnel or aperture widening has been recorded extensively in other studies with summary evidence that femoral and tibial tunnels increase in area primarily at the apertures during the first 6 months, with gradual progressive reduction over the next 18 months.³³ Although it seems logical that excessive tunnel aperture widening cannot be compatible with graft bone integration and subsequent graft healing, no correlation with tunnel narrowing and overall intra-articular graft signal has been previously found. This could be an important aspect of graft integration with bone ingrowth at the tendon-bone interface as the aperture reduces in size. It is noted, however, that our results from the cross-sectional femoral aperture ROI recorded a mean SIR that remained high after 2 years, and more precise measurements of the MRI aperture appearances are likely to be required.

Older age seems to correlate with greater SIR changes between 1 and 2 years in the proximal graft, which might indicate a delayed healing response with increasing age. We found no correlations between MRI appearances at 1 or 2 years and any of the PROMs used in this analysis. For this study, PROMs were used to demonstrate the relatively high sporting level of the study cohort; as a secondary outcome measure, PROMs were underpowered to demonstrate a correlation with MRI appearance. Factors influencing PROMs are multifactorial, and they have been shown to be significantly influenced by preoperative PROMs and associated pathology such as medial arthrosis and lateral meniscal damage.³⁵ In the present study, 1 patient was excluded for graft rupture, the 1-year MRI scan demonstrated a high proximal graft signal, which has previously been shown to correlate with risk of rupture.²⁶

In this series, KT-1000 arthrometer measurements were not correlated with graft MRI signal at 1 year or signal change thereafter and had no correlation with PROMs at any time point. Currently, KT-1000 arthrometry is still being used to demonstrate a difference between laxity of healthy and operated legs, with <3 mm difference historically described as an indicator for a stable knee.¹⁸ Recent studies have suggested that KT-1000 arthrometry is highly user dependent, showing both an investigator and a device effect,¹⁰ and has significant variation when compared with a device that exerts reproducible mechanical anterior knee translation.¹¹

Increasing evidence shows that the femoral side of the graft is the “weak link” in ACLR hamstring tendon autograft healing. A number of studies have demonstrated that the highest graft signal at 1-year MRI examination is in the proximal graft and within the femoral tunnel, where we also see greater tunnel widening than in the tibia.^9,17,24,26 Several theories for this finding have been advanced: 1) that a superior gap or “attic” forms in the femoral tunnel where the graft bends into the joint through a circular aperture drilled obliquely into a vertical wall,¹⁹ with femoral tunnel expansion also correlated with higher signal on MRI with the expansion predominantly occurring on the anteroinferior wall that contains the graft³⁷; 2) that the acute graft bending angle during knee movement is far higher at the femoral aperture than the tibial aperture, causing more graft movement and less chance of integration¹⁴; and 3) that the ability to preserve the ACL stump is better on the tibial side.³⁶ A study that used a complex analysis of the transition zone between bone and tendon in the femoral tunnel using 5 grades demonstrated a correlation between the appearance of more complete bone-to-tendon graft healing and improved PROMs at 1 year; additionally, a fibrous interzone was seen as a sign of delayed healing, which was observed significantly more at the femoral side than the tibial side.¹⁵

The concept that the blood supply at the tibial aperture is important for overall graft healing is supported by evidence that hamstring tendon grafts show less high-signal fluid on MRI at 4 and 18 months when the tibial pes anserinus insertion is preserved and that by 18 months the tibial tunnel area is reduced alongside lower graft signal.⁶ In a further study directly comparing distal hamstring attachment sparing or detachment, serial MRI measurements revealed a significant increase in graft signal at 6 and 12 months for detached grafts. Interestingly and importantly, this effect was not evident on the 24-month MRI scan in either cohort, suggesting significant changes in the graft throughout this time period.¹⁶ In the present study, all hamstring tendon grafts were detached distally, and yet significantly better healing was seen at the tibial aperture compared with the femoral aperture.

Future efforts to improve the rate of healing and integration within the tunnels should focus on how we achieve a faster reduction in tunnel area with bone ingrowth around the tendon. This could come in the form of tunnel shapes that conform better to the graft shape¹⁹ or novel biologic additions.²² Further studies are required to identify whether different rehabilitation strategies can influence graft healing and integration.

A previous study concluded that serial MRI examination during the first postoperative year was potentially clinically useful to identify biologically or mechanically deficient ACL grafts at risk of failure.²¹ With the findings from the present study, it can now be proposed that if high signal is seen at 1 year, particularly proximally, then this should be monitored on MRI for improvement.

Some limitations of this study need to be noted. To ensure that second MRI examinations were conducted as close as possible to 2 years postoperatively during the study period, only 42 patients from an original cohort of 250 could be included. The rate of graft healing and integration is multifactorial, with the potential to be influenced by patient factors such as age and the type of rehabilitation²⁶ and surgical factors such as the graft type, graft source,²¹ or fixation method.²⁵ This study included only patients with a hamstring tendon autograft. The number of patients with hybrid fixation in the final cohort was small and therefore underpowered for statistical comparison; however, a previous study demonstrated that the 2 fixation methods provide comparable MRI results at 1 year.²⁵

Conclusion

In the majority of patients, graft SIR on MRI examination did not significantly change after 1 year, and a proximal graft SIR <2 was a sensitive indicator for a stable graft signal, implying healing. Monitoring is proposed for patients who have high signal at 1 year (proximal graft SIR >4), given that we saw a significant reduction in signal in the second year, indicative of ongoing healing, alongside an improvement in sporting level. A reduction in tunnel aperture area correlated with a reduction in graft SIR, suggesting this could also be a useful measure of graft integration.

Footnotes

Acknowledgements

The authors acknowledge Dr Darli Myat for contribution to data collection.

Submitted November 1, 2021; accepted March 25, 2022.

One or more of the authors has declared the following potential conflict of interest or source of funding: M.R.J.C. receives institutional support from Arthrex and Smith & Nephew in the form of a funded fellowship. B.A.F. receives institutional support from Arthrex and Smith & Nephew in the form of a funded fellow and is a paid consultant for Arthrex. D.A.P. receives institutional support from Arthrex and Smith & Nephew in the form of a funded fellow. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

ORCID iDs

Sven E. Putnis

Thomas Neri

References

Anderson

Irrgang

Kocher

Mann

Harrast

; International Knee Documentation Committee. The International Knee Documentation Committee Subjective Knee Evaluation Form: normative data. Am J Sports Med. 2006;34(1):128-135.

Biercevicz

Miranda

Machan

Murray

Fleming

. In situ, noninvasive, T2*-weighted MRI-derived parameters predict ex vivo structural properties of an anterior cruciate ligament reconstruction or bioenhanced primary repair in a porcine model. Am J Sports Med. 2013;41(3):560-566.

Bouguennec

Robinson

Douiri

Graveleau

Colombet

. Two-year postoperative MRI appearances of anterior cruciate ligament hamstrings autografts are not correlated with functional outcomes, anterior laxity, or patient age. Bone Joint J Open. 2021;2(8):569-575.

Cavaignac

Marot

Faruch

, et al. Hamstring graft incorporation according to the length of the graft inside tunnels. Am J Sports Med. 2018;46(2):348-356.

Chu

Williams

. Quantitative MRI UTE-T2* and T2* show progressive and continued graft maturation over 2 years in human patients after anterior cruciate ligament reconstruction. Orthop J Sports Med. 2019;7(8):2325967119863056.

Grassi

Casali

Macchiarola

, et al. Hamstring grafts for anterior cruciate ligament reconstruction show better magnetic resonance features when tibial insertion is preserved. Knee Surg Sports Traumatol Arthrosc. 2021;29(2):507-518.

Grasso

Linklater

Parker

. Validation of an MRI protocol for routine quantitative assessment of tunnel position in anterior cruciate ligament reconstruction. Am J Sports Med. 2018;46(7):1624-1631.

Hoenig

Heisey

. The abuse of power: the pervasive fallacy of power calculations for data analysis. Am Stat. 2001;55(1):19-24.

Jansson

Harilainen

Sandelin

Karjalainen

Aronen

Tallroth

. Bone tunnel enlargement after anterior cruciate ligament reconstruction with the hamstring autograft and endobutton fixation technique: a clinical, radiographic and magnetic resonance imaging study with 2 years follow-up. Knee Surg Sports Traumatol Arthrosc. 1999;7(5):290-295.

10.

Klasan

Putnis

Kandhari

Oshima

Fritsch

Parker

. Healthy knee KT1000 measurements of anterior tibial translation have significant variation. Knee Surg Sports Traumatol Arthrosc. 2020;28(7):2177-2183.

11.

Klasan

Putnis

Kandhari

Oshima

Parker

. Anterior knee translation measurements after ACL reconstruction are influenced by the type of laximeter used. Knee Surg Sports Traumatol Arthrosc. 2020;28(11):3639-3646.

12.

Korbin

Salerno

Achonu

, et al. PET/MRI reveals ongoing metabolic activity in ACL grafts one year post-ACL reconstruction. J Exper Orthop. 2020;7(1):40.

13.

Lansdown

Xiao

Zhang

, et al. Quantitative imaging of anterior cruciate ligament (ACL) graft demonstrates longitudinal compositional changes and relationships with clinical outcomes at 2 years after ACL reconstruction. J Orthop Res. 2020;38(6):1289-1295.

14.

Liu

Sun

Chen

. Influence of graft bending angle on graft maturation, the femoral tunnel, and functional outcomes by 12 months after anterior cruciate ligament reconstruction. Orthop J Sports Med. 2019;7(11):2325967119882663.

15.

Chen

. MRI-based tendon bone healing is related to the clinical functional scores at the first year after anterior cruciate ligament reconstruction with hamstring tendon autograft. Knee Surg Sports Traumatol Arthrosc. 2018;26(2):615-621.

16.

Liu

Tao

Sun

Chen

. A randomized clinical trial to evaluate attached hamstring anterior cruciate ligament graft maturity with magnetic resonance imaging. Am J Sports Med. 2018;46(5):1143-1149.

17.

Mayr

Smekal

Koidl

, et al. Tunnel widening after ACL reconstruction with aperture screw fixation or all-inside reconstruction with suspensory cortical button fixation: volumetric measurements on CT and MRI scans. Knee. 2017;24(5):1047-1054.

18.

Myrer

Schulthies

Fellingham

. Relative and absolute reliability of the KT-2000 arthrometer for uninjured knees: testing at 67, 89, 134, and 178 N and manual maximum forces. Am J Sports Med. 1996;24(1):104-108.

19.

Oshima

Nakase

Numata

Takata

Tsuchiya

. The cross-sectional shape of the fourfold semitendinosus tendon is oval, not round. J Exp Orthop. 2016;3(1):28.

20.

Oshima

Putnis

Grasso

Klasan

Parker

. Graft size and orientation within the femoral notch affect graft healing at 1 year after anterior cruciate ligament reconstruction. Am J Sports Med. 2020;48(1):99-108.

21.

Panos

Webster

Hewett

. Anterior cruciate ligament grafts display differential maturation patterns on magnetic resonance imaging following reconstruction: a systematic review. Knee Surg Sports Traumatol Arthrosc. 2020;28:2124-2138.

22.

Park

Choi

Moon

, et al. Three-dimensional bio-printed scaffold sleeves with mesenchymal stem cells for enhancement of tendon-to-bone healing in anterior cruciate ligament reconstruction using soft-tissue tendon graft. Arthroscopy. 2018;34(1):166-179.

23.

Pearle

McAllister

Howell

. Rationale for strategic graft placement in anterior cruciate ligament reconstruction: I.D.E.A.L. femoral tunnel position. Am J Orthop (Belle Mead NJ). 2015; 44(6):253-258.

24.

Putnis

Neri

Grasso

Linklater

Fritsch

Parker

. ACL hamstring grafts fixed using adjustable cortical suspension in both the femur and tibia demonstrate healing and integration on MRI at one year. Knee Surg Sports Traumatol Arthrosc. 2020;28:906-914.

25.

Putnis

Oshima

Klasan

, et al. Adjustable suspension versus hybrid fixation in hamstring autograft anterior cruciate ligament reconstruction. Knee. 2021;28:1-8.

26.

Putnis

Oshima

Klasan

, et al. Magnetic resonance imaging 1 year after hamstring autograft anterior cruciate ligament reconstruction can identify those at higher risk of graft failure: an analysis of 250 cases. Am J Sports Med. 2021;49(5):1270-1278.

27.

Ruopp

Perkins

Whitcomb

Schisterman

. Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biometrical J. 2008;50(3):419-430.

28.

Smith-Crowe

Burke

Cohen

Doveh

. Statistical significance criteria for the rWG and average deviation interrater agreement indices. J Appl Psychol. 2014;99(2):239-261.

29.

Tegner

Lysholm

. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res. 1985;198:43-49.

30.

Van Dyck

Zazulia

Smekens

Heusdens

CHW

Janssens

Sijbers

. Assessment of anterior cruciate ligament graft maturity with conventional magnetic resonance imaging: a systematic literature review. Orthop J Sports Med. 2019;7(6):2325967119849012.

31.

van Groningen

van der Steen

Janssen

van Rhijn

van der Linden

Janssen

RPA

. Assessment of graft maturity after anterior cruciate ligament reconstruction using autografts: a systematic review of biopsy and magnetic resonance imaging studies. Arthrosc Sports Med Rehabil. 2020;2(4):e377-e388.

32.

van Melick

van Cingel

REH

Brooijmans

, et al. Evidence-based clinical practice update: practice guidelines for anterior cruciate ligament rehabilitation based on a systematic review and multidisciplinary consensus. Br J Sports Med. 2016;50(24):1506-1515.

33.

Weber

Delos

Oltean

, et al. Tibial and femoral tunnel changes after ACL reconstruction: a prospective 2-year longitudinal MRI study. Am J Sports Med. 2015;43(5):1147-1156.

34.

Weiler

Peters

Mäurer

Unterhauser

Südkamp

. Biomechanical properties and vascularity of an anterior cruciate ligament graft can be predicted by contrast-enhanced magnetic resonance imaging: a two-year study in sheep. Am J Sports Med. 2001;29(6):751-761.

35.

Wright

Huston

Nwosu

. Predictors of patient-reported outcomes at 2 years after revision anterior cruciate ligament reconstruction: response. Am J Sports Med. 2020;48(3):NP32.

36.

Zhao

Sun

. Preservation of remnant attachment improves graft healing in a rabbit model of anterior cruciate ligament reconstruction. Arthroscopy. 2013;29(8):1362-1371.

37.

Zhang

Liu

Yang

Chen

. Morphological changes of the femoral tunnel and their correlation with hamstring tendon autograft maturation up to 2 years after anterior cruciate ligament reconstruction using femoral cortical suspension. Am J Sports Med. 2020;48(3):554-564.