The outcome of arthroscopic repair of acetabular labral tears using the iHOT-33

The acetabular labrum has a role in shock absorption, joint lubrication, pressure distribution and improves stability by deepening the acetabulum [1, 2]. A torn labrum compromises these functions, with acetabular labral lesions being associated with early degenerative joint disease [3]. Arthroscopic management of a labral tear has been shown to relieve pain and improve function with a low morbidity and complication rate [4, 5].

Outcome studies have utilized a variety of patient reported outcome measures (PROMS) including; the Hip Outcome Score (HOS), Non-Arthritic Hip Score (NAHS), Harris Hip Score (HHS) and Western Ontario and McMaster Universities Osteoarthritis index (WOMAC). Many of which have ceiling effects and comparatively poor psychometric properties in regards to newer validated scores such as the International Hip Outcome Tool (i-HOT-33) [6, 7]. Critical appraisal of development, measurement properties and comparison studies suggest the iHOT-33 is currently the most validated tool for hip preservation surgery in young and active patients [6].

The purpose of the study was to investigate the outcome of primary repairs of labral tears in patients without an underlying osseous abnormality using the validated iHOT-33 outcome score with minimum 1-year post-operative follow-up.

A retrospective review of our hip arthroscopy database was performed. There were 2541 patients included in the database from hip arthroscopies performed by two surgeons between September 2012 and September 2015. Of these patients 459 had labral repairs. Patients were excluded if they were younger than 18 years, had less than one-year follow-up, had prior ipsilateral hip surgery and had radiological or arthroscopic evidence of femoroacetabular impingement (FAI), hip dysplasia or other bony dysmorphism.

After exclusion criteria was applied there were 71 procedures in 69 patients who received primary repair of a torn acetabular labrum. Of these there were 26 patients failed to respond postoperatively via email or post. Thus the final sample size was 43 patients undergoing 45 procedures.

Indication for surgery in all patients was recalcitrant hip pain and associated mechanical symptoms that were not responsive to conservative treatment for at least 6 months. On clinical examination, all patients had positive pain provocation tests (flexion, adduction and internal rotation) and were investigated using plain radiographs (AP pelvis, lateral, Dunn view), magnetic resonance imaging scans (MRI) and CT scans to exclude osseous abnormality including femoroacetabular impingement and hip dysplasia.

Of the 43 patients undergoing arthroscopy there were 29 right and 16 left hips repaired. The majority of the patients were female (34 females, 9 males). The mean age at surgery was 37.4 years (95% CI 19–63 years) with a mean follow up of 1.7 years (95% CI 1.0–2.6 years). Only 26.7% of patients could recall an obvious event or trauma causing their pain. Patients were involved in common sports included dancing, running, triathlons and weightlifting (Table 1).

The majority of labral tears were anterior in 40 hips (88.9%) and superior in 5 hips (11.1%) (Table 2). Labral tears were identified intraoperatively using the clock face method of localization and were between the 10 o’clock and 2 o’clock position. The mean labral tear size was 1.78 clock face spacing’s (1–5).

Associated chondral damage was present in 37 hips (82.2%). They were classified according to the Outerbridge classification and The international cartilage research society (ICRS) grading system [10, 11]. 35 hips (77.8%) had partial tears of the LT, which underwent radiofrequency ablation. LT tears were classified using the arthroscopic classification proposed by Sales & O’Donnell. [12] No correlation between change in i-HOT33 score and ligaemtnum teres was shown (p = 0.52). Similarly, i-HOT33 score change was not affected in those patients who required chondroplasty(p = 0.15).

At follow up the mean total iHOT-33 score improved from 34.1 (95% CI 16.8–51.4) to 67.3 (95% CI 44.7–89.9), p < 0.02). The mean improvement was 33.2 (p = < 0.02). In addition, significant improvements were described in all 4 iHOT-33 sub sections (Table 3). The minimum clinically important difference for the i-HOT33 is 6.1. [7]

The majority of labral tears are associated with other bony pathology such as FAI. Isolated labral tears occur less frequently and often result from a significant traumatic event or repetitive trauma or iliopsoas impingement [14, 15]. This is generally secondary to the labrum’s weight bearing role at extremes of joint range [16]. Regardless of the cause impaired labral function can result in significant pain, reduced function and can impair a patient’s quality of life.

In our cohort, patients showed significant short-term improvement in all functional domains of the iHOT-33 with mean follow up of 1.7 years. These studies are consistent with other results in the literature that also show modest improvements [6, 17‐22]. Although only few examine isolated tears directly [20, 21]. In 2011 Haviv and O’Donnell noted improvements when evaluating labral repairs without bony dysmorphism using the MHHS and NAHS. In their study of 81 patients the MHHS improved by 18 points and the NAHS by 17 points [20].

There are multiple PROMS used in the literature to assess patient outcome following hip arthroscopy. Older PROMS show ceiling effects and comparatively poor psychometric properties in respect to newer and more validated scores such as the iHOT-33. Despite this the Modified Harris Hip Score (MHHS) is still the most common PROM used to assess hip arthroscopic repair of acetabular labral tear [23]. Other PROMS such as the HHS and WOMAC are also used but were originally designed for hip arthroplasty and hip osteoarthritis rather than hip arthroscopy [24, 25]. Multiple studies compare current PROMS and the iHOT-33 has shown to have superior reliability, construct validity and responsiveness for use in patients undergoing hip preservation surgery in young and active patients [26]. This is the first study that uses iHOT-33 to assess acetabular labral tear repair.

Arthroscopy allows visualisation of nearby articular structures, in particular, the articular cartilage and ligamentum teres. Acetabular labral tears are associated with articular cartilage lesions and usually occur at the watershed zone at the labrochondral junction [27]. In this study 82.2% of the patients had associated acetabular chondral rim lesions and of these 73.7% requiring chondroplasty. These results are similar to other studies that show an association with articular pathology. Mcarthy et al. found that 63% of hip arthroscopy for labral tears showed chondral damage [28]. Kamath et al. required 40% of patients to undergo concomitant chondroplasty.¹⁹ Recently Stake et al. suggested patients with articular cartilage damage have worse outcomes [18]. Other studies have reported no effect on outcomes are at least shown conflicting results. [29, 30]. Our study did not show any significant correlation with chondral damage and worse patient outcomes at 1 year. However, longitudinal follow-up is required to clarify the prognostic value of cartilage damage after arthroscopic repair of labral tears.

In our study 35 hips (77.8%) had partial tears of the ligamentum teres, which all underwent radiofrequency ablation. The ligamentum teres tightens in external rotation, and may have a secondary stabilizer role with labral deficiency [31]. If the ligament is torn it can cause impingement and be a source of disabling pain and are usually debrided to a stable remnant [32]. In this study, the presence and degree of tearing of ligamentum teres tears had no effect on outcome.

Patients with pre-existing osteoarthritis have poorer outcomes after hip arthroscopy for labral pathology [1, 22, 26, 33, 34]. Meftah et al. revealed that only 19% of patients with arthritic changes had a good or excellent outcome. Compared to 62% for the total patient cohort [22]. This study showed no statistical correlation on chondral damage and patient outcome. This could be due to the low prevalence of significant cartilage damage in our cohort. A systematic review used to grade indications of hip arthroscopy showed poor-quality conflicting evidence regarding the use of hip arthroscopy for the treatment of mild to moderate osteoarthritis of the hip [35]. It is likely that given the patient cohort size and age that a larger and older cohort would improve the assessment on labral tear repair in the setting of chondral damage.

Although the distinctive strength of this study is using a validated PROM (i-HOT33) on specific patients without bony dysmorphism; there are still limitations. The inherent limitations of retrospective studies include selection bias, loss to follow up, lack of control subjects and limited sample size. This study had significant non respondents which could also overestimate the positive effect. In addition, the study has a relatively short term follow up, as the iHOT33 is a recently available outcome score. However, we considered that 1 year minimum follow up time is a reasonable time to assess the early results of symptomatic relief and complications. Longer-term follow-up will be useful and we are conducting a longer follow up on these patients.

The study showed statistically significant favourable short-term outcomes in patients undergoing hip arthroscopy for a hip labral tear without bony dysmorphism using the validated iHOT-33.