Impact of fatty degeneration on the functional outcomes of 38 patients undergoing surgical repair of gluteal tendon tears

Background Gluteal tendon tears (GTT) can cause pain and weakness of the hip. We analyze the impact of gluteal muscle fatty degeneration, atrophy and tear morphology on clinical outcomes of surgical repair. Methods All sequential patients receiving surgical repair of GTTs via anchor sutures between 1/2015 and 11/2018 were retrospectively identified. MRIs were reviewed by a radiologist for tendon retraction, muscle atrophy and tear size. The Goutallier-Fuchs Classification (GFC) was used to quantify fatty degeneration as < 2° or ≥ 2°. Demographic and clinical variables were abstracted from the electronic records. The surveys HHS Section 1 and HOOS Jr. were obtained at last follow-up. The Pearson correlation and one-way ANOVA tests served for statistical analysis of clinical variance. Results 38 patients were identified, 29 (76.3%) were female. The average age was 67. Of the 11 (28.9%) patients with a prior hip arthroplasty 87.5% of primary THAs had a direct lateral approach. 29 (76.3%) patients were treated open and 9 (23.7%) arthroscopically. At an average follow-up of 20.9 months, patients reported a significant improvement in pain (97%), analgesic use (85.7%), limp (52.6%) and abduction strength (54.2%) (all: P ≤ 0.01). GFC ≥ 2° were associated with significantly worse outcomes in terms of limp (0.19/3 vs. 1.2/3, P = 0.05), HHS-S1 (58.19 vs. 71.68, P = 0.04) and complication rates (37.5% vs. 0%, P = 0.02). There was a strong correlation between tear retraction (P = 0.005), tear size (P = 0.009) and muscle atrophy (P = 0.001) with GFC ≥ 2° but not with clinical outcomes. GFC ≥ 2° was strongly related to lateral THA exposures (P < 0.001). Surgical approach had no impact on clinical outcomes. Conclusion While fatty degeneration can negatively impact functional outcomes, pain relief is reliably achieved. Tear morphology and muscle atrophy did not correlate with outcomes in this patient cohort. Patients should be counseled to expect a residual limp after surgery if they have GFC ≥ 2° on MRI.


Introduction
Gluteal tendon tears (GTT) are often unrecognized and treated inadequately [1]. Depending on their severity and etiology, symptoms may vary from a sole lateral hip pain to a persistent abductor insufficiency of the hip. Structural parameters of GTT such as tear size, retraction of the torn tendon [2][3][4], fatty degeneration and atrophy of the injured muscle [3,[5][6][7] may impact the clinical outcomes of surgical treatment. Demographic aspects of the individual patient like age [8,9], gender [10], bone mineral density (BMD) [11,12] and body mass index (BMI) as well as the presence of a preexisting total hip arthroplasty (THA) and the number of previous revision THAs (RHA) [13] pose potential risk factors for the success of the initiated therapy. In particular, postsurgical GTTs are associated with the direct lateral THA approach [14,15] or even the anterolateral THA approach [16].
Related literature up to now suffers from inhomogeneity and small sample sizes making reliable conclusions and developing transparent treatment algorithms impossible. Thus, little is known about the comparative impact of FD, MA and TM on the postoperative outcomes between open (OGR) and endoscopic gluteal repair (EGR) in minimizing pain and restoring function.
The aim of the study is to analyze the impact of fatty degeneration and the other risk factors on the mid-term results in a large cohort of patients who underwent either EGR or OGR of GTTs via suture anchors.

Materials and methods
Approval from the local Institutional Review Boards was granted. A retrospective single-center cohort study from a university hospital was conducted. All sequential patients receiving surgical repair of GTTs via anchor sutures between 1/2015 and 11/2018 were identified and retrospectively recruited for the study. Treatment was divided in: (1) EGR, (2) primary OGR (pOGR) and (3) revision total hip arthroplasties with OGR (rOGR).
All patients who were included in the study presented with persistent lateral hip pain with or without signs of hip abductor insufficiency and a positive magnetic resonance imaging (MRI) for GTT prior to surgery. Exclusion criteria were septic surgeries, active tumors or known neurological diseases that affected the hip, no preoperative MRI or inability to assess due to massive hardware artifacts, patients who refused to participate to the study and finally cases with missing or inconsistent documentation.
Demographic and clinical variables were abstracted from the electronic record using an institutional database. Age, gender, BMI and comorbidity level, as defined by the America Society of Anesthesiologist (ASA), were collected.
Preoperative x-rays were assessed to measure the femoral offset and look for any malposition or instability of preexisting implants on the hip or lumbar spine.
The preoperative MRI was used to evaluate the fatty infiltration and atrophy of the gluteal muscles and to analyze the complexity of the GTTs.
The fatty infiltration of the gluteal muscles was grouped based on the Goutallier-Fuchs classification (GFC) in: 0 = normal muscle, 1 = muscles with some fatty streaks, 2 = muscles with moderate fatty streaks (more muscle than fat), 3 = muscles with severe fatty streaks (equal amounts of fat and muscle), and 4 = muscles with more fat than muscle [55][56][57]. Based on recent studies [3,5], GFC was further divided into two clinically relevant main groups, namely into below (GFC < 2°) and above grade 2 (GFC ≥ 2°). The assessment included the gluteus minimus (Gmin) and three distinct parts of the gluteus medius muscle (Gmed): the anterior, mid and posterior portion of Gmed (Fig. 1). This technique is described by Thaunat et al. [6]. According to that study, gluteus medius and minimus muscles were evaluated on the axial T1 sequences without fat saturation on the first axial slice inferior to the sacroiliac joint.
Muscle atrophy of the hip abductors was defined as the reduction of the size of its components by ≥ 25% then that of the contralateral side [58].
The gluteus medius and minimus tendons were assessed separately for presence or absence of tear. All available fluidsensitive sequences (STIR or T2 fat saturation) were used Fig. 1 An axial T1 sequence of an MRI illustrating the Gmin and the three district parts of Gmed, that were included in the assessment of fatty degeneration as described by Thaunat [6]: Gmin (rot), anterior portion (blue), mid portion (yellow) and posterior portion of Gmed (green) 1 3 to evaluate and characterize the tear. The tears were characterized as full-thickness lesions (FT) or partial-thickness lesions (PT), which were further differentiated into lateral bursal-side (LPT) or medial joint-side partial tears (MPT). The length of tendon retraction was measured from their attachment on the greater tuberosity. The anteroposterior width of the tear was also measured (Fig. 2a-d). Any evidence of bursitis was classified in 0 = none, 1 = mild, 2 = moderate, or 3 = severe as described by Chi et al. [59]. All MRI measurements mentioned above were performed by an experienced musculoskeletal radiologist.
All patients included in the study were contacted postoperatively via e-mail or phone and invited to an anonymous survey to evaluate the postoperative results. For this purpose, following patient-reported outcome measures (PROMs) were used: The Hip disability and Osteoarthritis Outcome Score Junior (HOOS Jr.) [60], the questionnaire of the first section of Harris Hip Score (HHS-S1) [61], that concerns pain and function features, and the Visual Analog Scale of pain (VAS). Furthermore, the amount and type of analgesics been taken at the time of the survey due to still existing hip pain was verified.
Following postoperative outcome values were analyzed: The software IBM SPSS (Version 25.0) was used to perform statistical analysis. The Pearson correlation test was used to assess possible correlations between potential risk factors and outcomes. The two-tailed paired T-test was used to compare pairwise the postoperative results with the preoperative findings. The one-way analysis of variance (ANOVA) and the least significant difference (LSD) tests were performed for single or multiple intergroup comparisons between different patient subgroups.
The differences between each GFC groups are summarized as follows (Table 1).
From the 28 preoperative MRIs with possible evaluation of muscle atrophy, 24 (85.7%) cases revealed a MA on both the Gmed and Gmin whereas in 18 (64.3%) of the cases only the Gmed was affected.
Muscle atrophy was associated with the etiology of a GTT (P = 0.039), being evidenced more often in the rOGR group. Atrophy was also correlated with the Gmed retraction (P = 0.035) and tear size (P = 0.037). GTTs with a MA ≥ 25% showed significantly higher levels of Gmed retraction (0 mm vs. 15.3 ± 1.9 mm, P = 0.01) and fatty degeneration (1.48 ± 0.9 vs. 2.6 ± 0.8, P < 0.001) and were found more often in cases with preexisting THA (50% vs. 10%, P = 0.02) when compared to the GTT group without MA (Table 4). However, MA did not have any significant impact on clinical outcomes and functional PROMs at the last postoperative follow-up.

Discussion
Independent of the operative approach used to treat gluteal tendon tears, fatty degeneration of gluteal muscles GFC ≥ 2° seems to be associated with significantly worst functional outcomes after surgical repair. GTT muscle atrophy and retraction did not reveal any association to the postoperative results in our patient cohort. Nevertheless, this should be considered cautiously since it might be due to the small subgroup sizes that met inclusion criteria within the current study design. Finally, statistically significant improvement of all main outcome measures (VAS, limp, abduction, analgesics, opioids) were achieved in all cases of GTT after surgical treatment (Tables 1 and 2) suggesting that even patients with advanced fatty degeneration may benefit from direct surgical repair.
The subgroup analysis showed detailed impact of fatty degeneration. GFC ≥ 2° was associated with considerably inferior outcomes in the functional section S1 of HHS and limp after GTT surgical repair when compared to GFC < 2° (P = 0.04 and P = 0.025, respectively). Furthermore, complication rates were significantly higher in the GFC ≥ 2° group (37.5% vs. 0%, P = 0.02). However, whilst it is known that fatty degeneration has a negative impact on postoperative functional outcomes [3,6,7], our study suggests that pain relief can be reliably accomplished after surgical treatment of GTTs regardless of the degree of GFC. Thus, our results align with the available literature, suggesting that surgical reattachment of GTT is an effective treatment option for persistent lateral hip pain [44,63].  Rates of improvement (%) Fig. 3 Percentage of the overall cases with verified improvement of the main outcome objectives after surgical treatment of GTT  [54]. However, there are some fundamental differences between these two approaches to consider when choosing the proper treatment strategy: endoscopic approach is better suited to smaller tears and tears of the gluteus minimus (cite our prior paper here) and has the benefit of also being able to address articular-side pathologies during the same procedure. Most arthroscopic procedures were used in patients with a low to moderate degree of fatty degeneration (GFC < 2°). There are no data available to our knowledge assessing the efficacy of endoscopic GTT repair in cases with severe GFC. EGR also requires specialized equipment and a considerable learning-curve with the risk of poor anchor placement [64]. Open approaches offer a greater exposure and access to the retracted edge of chronic GTT and therefore provide greater treatment flexibility with more options for fixation including, if required, the use of augmentation or reconstruction techniques in cases with more extended soft tissue damage [24-33, 65, 66] or palsy of the superior gluteal nerve [16,67,68].
In the current study, MRI findings prior to surgery revealed a strong correlation between fatty degeneration and other abnormal structural properties of the GTTs such as the retraction (P = 0.005), tear size (P = 0.009), muscle atrophy (P = 0.001) and the severity of trochanteric bursitis (P = 0.02). There was also a significant difference in the severity of retraction, tear size, muscle atrophy and bursitis of the GTTs between GFC < 2° and GFC ≥ 2° (Table 3). However, we could not find any significant impact of the above-mentioned variables on the postoperative results after GTT repair. In contrast, previous publications support the importance of tear size and retraction on the success of surgical treatment [2,3]. Amstutz and Maki using the trochanteric approach for THAs in the late 70′s found that post-operative abductor insufficiency was associated with pre-operative separation gaps > 2 cm [69]. In a recent study on 46 RHA's, Caviglia et al. drew the same conclusion indicating significantly worst hip abduction when GTT trochanteric displacement was > 2.5 cm. This finding was independent of the tear type between partial and fullthickness lesions [2].
Makridis et al. in a relative large patient cohort (70 hips) reported that muscle atrophy and not FD had a negative impact on functional outcomes (P = 0.05) after open doublerow GTT repair [5]. However, the majority of the patients included were healthy individuals with no history of THA, trauma, or any systemic inflammatory disease. 94% of these cases had partial lesions of the anterior portion of the Gmed and 20% (n = 14) a FD ≥ 2°. In 60% of their cases there was neither FD nor MA evidenced. In our cohort, FD ≥ 2° was found in 42.1% and MA in 64%, while cases with FD ≥ 2° were associated with MA in 96.4% (Tables 3 and 4). Based on known similarities between GTTs and rotator cuff tears (RCT) [70] and the available data originating from the shoulder surgery literature, duration of symptoms [71] and age of the patient [72] seem to be associated with increased fatty degeneration of the RCT muscles, whereas, inactivity and nerve injuries lead to muscle atrophy [73].
Total hip arthroplasty (THA) is considered one of the most successful and effective orthopedic procedures. However, 6% of the patients remain unsatisfied due to persistent pain after surgery [74,75]. Limited evidence exists concerning the exact etiology of pain in this critical patient group. The surgical technique chosen to perform a THA may play an essential role in pain generation if it leads to damage to the abductor tendons. A high incidence of postoperative Trendelenburg sign (27.6%) and limp (4-20%) has been described after use of the lateral THA approach [14,15,19,76].
Thus, ongoing lateral hip pain with Trendelenburg gait in a patient with a history of a direct or antero-lateral THA should always raise suspicions of a GTT or failed repair of the partial Gluteus Medius take-down. Spontaneous partial GTTs are most commonly found on the anterior portion of the Gmed due to its thinner anterolateral layer [77]. The limitations of this work include the retrospective nature of the study and thus, the potential recall bias for some subjective measures of the PROMs that were collected and the resulting small sample size of the subgroups tested here, that might have an influence on the power of the results.
However, this is one of the largest descriptive studies that analyzes the impact of some clinically relevant structural properties of GTTs on the mid-term outcomes after surgical treatment. Our thorough MRI analysis of the GTTs by specialized musculoskeletal radiologist provides a clear and well-established methodology for classifying and reporting results from GTT repairs. Lastly, the majority of the repairs described in our study were performed on "massive" GTTs and outlines the results that can be expected with these patients. We hope that the current study provides important information that could guide physicians to be more aware of the potential benefits of GTT repairs and their early referral for surgical intervention before fatty atrophy sets in.

Conclusion
We report overall excellent results in a large series of GTT repairs and a classification system to predict clinical outcomes from pre-operative imaging. While fatty degeneration can negatively impact functional outcomes, pain relief is reliably achieved. Tear morphology and muscle atrophy did not correlate PROMs in our cohort. Patients should be counseled to expect a residual limp after surgery if they have GFC ≥ 2° on MRI.