Labral size measured on preoperative magnetic resonance imaging not predictive of the need for labral reconstruction in patients undergoing primary hip arthroscopy

Abstract

Preoperative radiographic measurements may help predict which patients with hip labral tears ultimately undergo repair versus primary reconstruction. This study investigated if radiographic parameters: (i) preoperatively predict labral repair versus reconstruction and (ii) correlate with T2 magnetic resonance imaging (MRI) mapping values of the labrum. This retrospective comparative study included patients aged 14–50 years who underwent labral repair or reconstruction at a single institution over a 2-year period. Patients with prior open or arthroscopic hip surgery or who had inadequate preoperative computed tomography (CT) and MRI imaging were excluded. Labral size was measured at multiple positions on preoperative MRI images. A blinded reviewer used three-dimensional CT analysis to record lateral center edge angle (LCEA), acetabular version, Tonnis angle, acetabular coverage, alpha angle, femoral torsion, and neck-shaft angle (FNSA). T2 MRI mapping values of the labrum were obtained via sequencing analyses on each patient’s optimal sagittal cut. Univariate mixed linear models were used to identify associations between each radiographic measurement and decision to repair or reconstruct the labrum. Fifty-two operations were included. Labral size had no predictive effect on undergoing labral reconstruction versus repair. Likelihood for undergoing labral reconstruction was associated with LCEA (P = .003) and Tonnis angle (P = .034). There was an association (P < .05) between labral T2 mapping values and all radiographic parameters except for FNSA and combined version. Labral size was not associated with whether patients underwent labral reconstruction or repair. The data showed an association between labrum T2 mapping values and nearly all radiographic parameters.

Introduction

The labrum serves a vital biomechanical role in the hip joint. It creates a suction seal for the hip joint, deepens the acetabulum, and allows for more uniform distribution of weight-bearing forces across the articular cartilage [1–3]. Consequently, labral tears pose a significant threat to the longevity of the native hip joint. While arthroscopic labral debridement may improve the pain and symptoms associated with labral tears, labral restoration demonstrates improved long-term outcomes, thus repair has emerged as a gold-standard treatment for labral tears [4–7]. There are, however, times when the tissue quality is such that labral repair is not feasible. In these cases, labral reconstruction using donor tissue is “used” to restore the suction seal of the hip joint. While reconstruction is currently mostly employed in the revision setting, there is evidence that some patients undergoing primary labral reconstruction may have improved outcomes and longevity compared to repair [8].

The decision to repair versus reconstruct a torn labrum is typically based on the surgeon’s intraoperative assessment of the labrum. Therefore, it can be difficult to predict which technique will be employed preoperatively. Labral reconstruction is a more technically demanding surgery and generally requires more time under anesthesia [8] and availability of graft tissue, so the ability to anticipate which patients may require a primary reconstruction would be beneficial for presurgical planning. Prior studies have investigated which preoperative variables may predict which patients undergo repair or primary reconstruction. Radiographic and demographic preoperative parameters associated with need for reconstruction include increased lateral center-edge angle, anterior center-edge angle, osteoarthritis severity (Tönnis grade), age, and body mass index (BMI) [9–11]. Nevertheless, intraoperative inspection of labral tissue remains the primary criterion by which surgeons decide operative treatment of labral tears [12–17].

In this study, the authors expand upon preoperative risk factors for undergoing a primary labral reconstruction by analyzing presurgical computed tomography (CT) scans and magnetic resonance imaging (MRI) of patients undergoing hip arthroscopy with labral repair or reconstruction. T2 mapping MRI, which measures changes in chondral collagen matrix and the overall water content of the cartilage and is an objective measure of soft tissue damage [18–20], was also utilized and compared to the other radiographic parameters in order to examine the relationship between bony anatomy and labral tissue quality.

Methods

Study design and sample

This is an IRB-approved Level 3 retrospective comparative study at a single institution with two high-volume hip arthroscopy surgeons (J.W.G. and S.W.M) between March 2021 and February 2023. Eligibility for inclusion entailed patients who underwent hip arthroscopy with labral repair or reconstruction between the ages of 14- and 50-years old with preoperative CT and MRI at the UCHealth Steadman Hawkins Clinic of Denver. Exclusion criteria consisted of previous hip arthroscopy procedures, prior trauma to the hip/labral anatomical area, history of inflammatory arthritis, history of intra-articular injection of steroid or bioactive agent within 3 months of surgery, and patients who received MRIs without appropriate T2 mapping or CTs without appropriate three-dimensional CT analysis software compatibility (Stryker HipMap®).

Surgical techniques

All surgeries were performed in the supine position on a traction table using standard arthroscopic portals. The labral tear was then confirmed and assessed. If the tissue was of too poor quality or too diminutive to form a gasket seal, a primary reconstruction was performed.

Preoperative imaging technique

CT images were performed supine with helically acquired axial images obtained through the bilateral hips at 2-mm intervals in bone algorithm, and reformatted in coronal, sagittal, and axial oblique planes.

Regarding MRI, each patient was imaged with a 3.0 T scanner prior to hip arthroscopy. The imaging protocol included standard clinical morphologic sequences followed immediately by a sagittal multi-echo spine-echo (MESE) T2-mapping sequence. The T2 mapping sequence (TR/TE 1530.0/13.80–69.00 ms; VS, 0.5 × 0.5 × 3.0 mm³; slices, 20; slice thickness 3.0 mm; FOV, 160 mm; AT, 4:51 min; FOV read 150 mm; flip angle 180°) was acquired in the sagittal plane for optimal mapping of the anterolateral joint, the area of most common pathology for labral and cartilage lesions in femoroacetabular impingement (FAI). The T2 mapping sequence was acquired after the standard MRI images were obtained and the patient had been recumbent for a prolonged period of time. The T2 mapping images were derived from the Siemens MapIT software algorithm (Seimens Healthineers, Erlangen, Germany).

Data collection

The three individual reviewers were blinded to patients’ surgical information. For labral size measurements, the reviewers followed the process defined by Comfort et al., taking three measurements for each hip: one from the coronal view (12 o’clock) and two from the axial view (3 and 9 o’clock) using a mid-sagittal plane, centered on the femoral head, as a standard reference [21]. An example is shown in Fig. 1. The reviewers made measurements under the supervision of a fellowship-trained musculoskeletal radiologist attending with over 30 years of experience.

T2 mapping values of the labrum were obtained by three individual reviewers, who were blinded to patients’ surgical information, with each reviewer performing three separate sequencing analyses using the T2 mapping software on an optimal sagittal cut of each patient’s preoperative MRI. The analyzed cuts chosen from each patient’s MRI were those which subjectively provided the highest definition for the labrum morphology at the roughly mid-sagittal level. Figure 2 demonstrates an example of these measurements.

Additional radiographic parameters were obtained by analyzing each patient’s preoperative CT scan using three-dimensional analysis software (Stryker HipMap®). Radiographic measurements collected include alpha angle, lateral central edge angle (LCEA), acetabular coverage (as a percentage), femoral version, acetabular version at 12 o’clock, 2 o’clock, and 3 o’clock, femoral neck-shaft angle (FNSA), and Tönnis angle. Demographic information was obtained from electronic health records.

Statistical analyses

Intraclass correlation models for the labral size at the 12, 3, and 9 o’clock positions were constructed to ensure overall agreement between raters. The Average Raters (Random) profile (ICC2k) was used to determine overall agreement with estimates generalized as per Table 1. Three Bayesian generalized linear mixed models were constructed to estimate the odds of needing a reconstruction based on labral size. Data were split into position measurements (12, 3, and 9 o’clock) and modeled separately.

Summary statistics for background characteristics and radiographic variables were also calculated. To quantify differences between the Reconstruction and Repair groups, P-values correspond to independent t-tests and chi-squared tests for continuous and categorical variables, respectively. Statistical significance is set at 0.05. Lastly, Univariate Mixed Linear Models (Univariate MLMs) were created to evaluate the effect of various radiographic measurements on T2 mapping values. As with the Bayesian models, these models calculate a regression coefficient (Beta) that represents the linear association of a radiographic parameter and the mapping value. A supplement to this article is included providing further detail on the statistical models.

Results

Fifty-two operations qualified for the study. Of these, 40 were primary labral repairs and 12 were reconstructions (Table 2). There were 31 females and 21 males. Eleven patients received arthroscopic procedures on bilateral extremities. Of these patients, nine received bilateral repairs while two received bilateral reconstructions. The average age of the participants was 29.33 years for those who underwent repair, and 34.50 years for those who underwent reconstruction. The youngest patient was 14-years old and the eldest was 50-years old. Neither group was statistically significantly different than the other with regards to age (P = .11), sex (P = .7), and BMI (P = .3).

There was excellent agreement between the raters at the 12 o’clock position [ICC2k = 0.905, 95% CI (0.0.832, 0.944), P < .001], and moderate agreement at both the 3 o’clock position [ICC2k = 0.723, 95% CI (0.0.574, 0.825), P < 0.001] and 9 o’clock position [ICC2k = 0.594, 95% CI (0.372, 0.743), P < 0.001]. Overall, inter-rater reliability was moderate (ICC = 0.74) for measuring labral size.

Regression analyses were performed for each labral measurement position (12, 3, and 9 o’clock) to assess each of their predictive value on odds of undergoing a primary labral reconstruction while controlling for patient age and sex. Full results for each regression analysis can be found in Tables 3–5. None of the labral measurement locations were found to predict the likelihood of reconstruction, with regression coefficients of 0.160 at 12 o’clock [95% CI (−1.31, 0.497)], 0.185 at 3 o’clock [95% CI (−1.36, 0.608)], and 0.158 at 9 o’clock [95% CI (−0.594, 0.028)]. Our regression models did show a statistically significant positive trend of age on the odds of needing reconstruction. For each 1-year increase in age, the odds of reconstruction in the three groups increased by 7.9% [95% CI (0.026, 0.15)], 8.7% [95% CI (0.028, 0.167)] and 9.1% [95% CI (0.034, 0.169)].

Summary statistics for radiographic and demographic characteristics investigated using univariate mixed linear model are shown in Table 6. The results of the t-tests show that the mean LCEA was greater (P = .004) and the mean Tönnis angle was lower in those who received a reconstruction (P = .033). There was no evidence of an interaction effect between LCEA and Tönnis angle despite their established radiographic relationship. None of the additional demographic and radiographic parameters investigated were found to be associated with increased odds of undergoing primary labral reconstruction including sex, BMI, acetabular version, acetabular coverage, alpha angle, femoral torsion, FNSA, or combined version, though there was a trend for increased acetabular coverage being associated with increased odds of undergoing primary reconstruction (P = .057).

Table 7 demonstrates results from multiple different univariate mixed linear models where each model tests whether a linear relationship exists between a given radiographic parameter and T2 labral mapping values. Increased acetabular version at 12, 2, and 3 o’clock, and Tönnis angle were associated with increased T2 mapping values (P < .001). Increased LCEA, acetabular coverage, femoral torsion, and max alpha angle were associated with decreased T2 mapping values (P < .001). FNSA and combined version did not show an association with labral measurements (P = .7, P = .10).

Discussion

The data did not show an association between labral size measurements at the 3, 12, and 9 o’clock positions and whether a patient underwent primary labral reconstruction versus repair. While there are multiple described methods to measure the labrum on preoperative MRI [21–23], the authors employed the methodology described by Comfort et al. [21]. While a strong ICC of 0.905 indicates reproducibility of this method, given the method reduces the labrum to three measurements on two MRI slices, it may not represent the labrum as a whole and could miss focal atrophic changes in the labrum not visualized on the selected MRI cuts. Consideration of different labral measurement strategies compared to the one employed in this study should be undertaken in the future.

The results of the present study demonstrate that increased lateral center-edge angle and decreased Tönnis angle on a patient’s preoperative CT scan are associated with increased odds of undergoing hip labral reconstruction. This is consistent with the published literature. Maldonado et al. similarly found increased LCEA and decreased Tönnis angle to be associated with increased odds of undergoing labral reconstruction [9]. Nakashima et al. identified a different marker of acetabular coverage (anterior center edge angle) to be positively associated with need for labral reconstruction [10]. The data of this study as well as published literature suggest that increased acetabular coverage is associated with increased likelihood of undergoing primary labral reconstruction, suggesting that patients with pincer FAI, as opposed to dysplasia, may be more likely to need reconstruction in the setting of a tear. Repetitive trauma to the labrum from the bony morphology in FAI may cause atrophic changes to the labrum, leaving less viable tissue for repair. I contrast, the microinstability often seen in hip dysplasia leads to a more hypertrophic labrum [24, 25], which may more consistently leave tissue available for repair. Previous literature has shown both increased age and male sex to be associated with increased odds of undergoing primary labral reconstruction. In this study, when controlling for sex and labral size at each position, increased age was also found to be associated with increased odds of reconstruction. Sex alone, however, was not found to have an association.

While the data suggest only a handful of parameters are statistically significantly associated with whether a patient underwent primary labral reconstruction, the authors found an association between T2 mapping values of the labrum and nearly all radiographic parameters, suggesting nearly all aspects of bony anatomy may affect labral tissue quality. The authors showed that increased T2 mapping values (indicative of poorer tissue quality) were associated with decreased LCEA, increased acetabular version, increased Tönnis angle, decreased acetabular coverage, and decreased alpha angle. On the spectrum of FAI versus hip dysplasia, each of these trends in measurements is indicative of more significant dysplasia. As such, the data suggest that patients with more severe radiographic hip dysplasia, as opposed to FAI, tend to have poorer labral tissue quality. These findings initially appear in contrast to the previously stated finding that FAI patients are more likely to need reconstruction, presumably secondary to worse labral quality; however, these two findings are not mutually exclusive. Based on the method of T2 mapping value labral measurement, the labral mapping values represent a global measurement of labral tissue quality. Hip dysplasia and related microinstability causes diffuse trauma to the labrum, likely leading to globally decreased tissue quality captured by increased mapping values. In contrast, FAI may result in a more focal wear on the labrum such that the same increase in T2 mapping value is not seen despite severe atrophic changes that necessitate reconstruction. In summary, the authors suggest that while FAI may cause more focal atrophic changes to the labrum, predisposing it to a higher likelihood of reconstruction, the microinstability seen in dysplasia may cause more significant diffuse damage to the labrum shown through increased T2 mapping values.

This study is not without its limitations. First, the sample size of 52 is relatively small, secondary to the requirement of having a preoperative T2 mapping MRI. Consequently, this study may have been underpowered to detect labral reconstruction associations seen in previous literature, such as BMI, and associations with variables not previously investigated, such as femoral torsion and acetabular version. The results of this study will help guide future prospective analyses in determining appropriate sample size to identify statically significant associations with sufficient power. The way the labral measurements were obtained must also be considered. Variable image quality and tissue damage made labral identification difficult for measurement of labral size and T2 mapping. While this limitation was addressed through having three independent reviewers and using a regression modeling system that accounts for this variability through designation as random effects, there is potential for inaccurate labral size measurement and T2 mapping which may impact study results. Finally, this was a Level III, retrospective study, allowing the possibility for selection bias, particularly regarding repair versus reconstruction operative decisions. This was based on consistent intraoperative criteria, however, there remains some subjectivity based on surgeon judgement and is therefore a limitation of this and any retrospective study on this subject. Future work should emphasize prospectively collected data with intraoperative decision for repair versus reconstruction based on strict objective criteria.

In summary, there was no association between the labral size measured on preoperative MRI and the intraoperative decision to repair versus reconstruct a torn labrum. Patients with increased LCEA and decreased Tönnis angle were more likely to undergo primary reconstruction of their labral tears, consistent with established literature. Nearly all radiographic parameters of patients’ bony anatomy on preoperative CT were associated with T2 mapping values of the labrum, such that bony anatomy consistent with hip dysplasia was associated with poorer labral tissue quality. This study suggests that use of radiographic parameters taken from preoperative advanced imaging may be a viable means of helping to predict whether a patient with a labral tear undergoes primary reconstruction versus repair, and these results should guide future prospective studies to determine which variables are predictive.

Supplementary data

Supplementary data is available at Journal of Hip Preservation Surgery online.

Conflict of interest

The authors of this manuscript have no financial, consultant, institutional, or other relationships that might lead to bias or a conflict of interest. UCHealth Steadman Hawkins Clinic Denver is the source of these data; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

Funding

None declared.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References