Low prevalence of trochanteric bursitis in patients with refractory lateral hip pain

Trochanteric bursitis has historically been identified as a major pain generator in patients with lateral hip pain (LHP), and the majority of established treatments, such as corticosteroid injections or surgery, have focused on treatment of this anatomical structure [1-3]. However, a study in 1,000 non-selected hip MRI highlighted the potentially low prevalence of an inflamed trochanteric bursa [4]. This was supported by findings in an ultrasound study including 877 patients, which only rarely detected isolated bursitis of the greater trochanter [5].

In recent years, growing attention has been devoted to hip abductor tendon pathology (HATP) as an important cause of refractory LHP. HATP includes tendinopathies, tendon tears or ruptures of the gluteus medius (Gmed) and minimus (Gmin) tendons. Unfortunately, it appears that HATP is still under- or misdiagnosed. So far, MRI has been the cornerstone in diagnosing HATP [6-10], and it has been estimated that GMed and GMin tendon tears may be present in as many as 25% of middle-aged females and 10% of middle-aged males [11].

The purpose of this study was to evaluate the presence of isolated trochanteric bursitis on MRI in patients with refractory LHP who were referred to a public orthopaedic hip clinic.

METHODS

Study design

This was a cross-sectional study that forms part of a large prospective cohort study [12]. The study period ran from January 2017 to January 2020. The reporting of this study was done according to the STROBE guidelines.

Study setting

Our orthopaedic outpatient clinic (OOC) is located in a public teaching hospital with a local referral population of 250,000. As the clinic comprises a specialised hip preservation surgery unit, patients from other parts of the country are also referred to our OOC.

Participants

In the study period, 1,349 unique patients were registered at our OOC with one of two International Classification of Diseases – tenth edtion diagnosis code related to LHP (DM70.6 Trochanteric bursitis/DM76.0 Gluteal tendinitis) by seven consultant orthopedic surgeons. Only patients seen by one of two consultants (one hip joint preservation surgeon and one adult hip reconstruction surgeon) in the study period were evaluated for participation in this study. Patients were eligible for inclusion provided i) an MRI had been performed due to LHP in the study period, ii) the patient was 18 years or above and iii) provided informed consent.

Patients were referred to MRI by the two surgeons if distinct palpable tenderness or pain of the GMed and/or GMin tendon insertions (anterior, lateral or posterosuperior facets of the greater trochanter) were detected, and this was identical to the cause of their referral. Also, one or more of the following tests were used to underpin the suspicion of HATP and needed to be positive for the patient to be referred for MRI: Trendelenburg sign/walk, the 30-second Single Leg Stance, the FADER (Hip Flexion, Adduction, External rotation), the FADER-R (Hip FADER with resisted isometric internal rotation at end range), the Internal Resistance Test, side lying hip abduction [13].

MRI was performed in 132 patients of whom 120 (94% women) provided written informed consent to participate in this study. The median age was 54 years (range: 18-77 years) at the first consultation. Distinct bilateral complaints were reported by 29 patients (24%). The history of LHP exceeded six months in all patients referred for MRI. Due to the risk of recall bias, the exact duration of symptoms was not estimated.

MRI

The applied MRI algorithm was based on previous reports [14-17] and adapted by two of the authors (CT, JL) to a clinician-oriented concept (see Table 1). A preferred standard MRI approach to the lateral hip structures includes coronal and axial windows with both inflammation and anatomical protocols. The specific MRI protocols in the study were performed at the discretion of the individual radiology department. Because of national legislation, patients are entitled to have their MRI done within a miximum of 30 days as from their referral. As such, numerous centres performed the MRI. We had no direct influence on which exact MRI protocol was followed. All MRIs were transferred to our local database for evaluation. The evaluation of the MRI was performed by CT & JL, according to Table 1. CT is a consultant musculoskeletal radiologist with a special focus on MRI. CT has participated in selected surgeries in LHP patientsfocusing on the evaluation of the macroscopic findings during surgery and the associated MRI findings. JL is a consultant orthopaedic surgeon with expertise in hip joint replacement surgery and has performed more than 100 open hip abductor tendon surgeries and independently interpreted more than 200 hip MRIs since January 2017.

The concept of interpretation of the MRI in the study was agreed upon in consensus between CT and JL and included pre-study evaluation of five individual cases. The evaluation consisted of both “objective” findings, such as the presence of high-intensity signals, and “subjective” findings, such as the evaluators’ interpretation of the “objective” findings as the presence of HATP or bursitis. Specifically in relation to the primary outcome of this study, the presence of bursitis was evaluated on inflammatory protocols. Bursitis was defined as a clearly extra-tendinous, well-defined accumulation of high-intensity signals lateral to the greater trochanter in cases of greater trochanteric bursitis or superior to the greater trochanter in cases of subgluteal bursitis, based on the work by Cvitanic et al., Kong et al. and Dwek et al. [14, 16, 18]. Tendon discontinuity with high-intensity signals replacing the tendon structure was not defined as bursitis but as a rupture (HATP) (see Figure 1 for an example).

As the presence of isolated bursitis of the greater trochanter was the primary purpose of this study, a consensus meeting was held after finalization of the individual evaluations of the 120 MRIs. At this meeting, CT and JL evaluated 49 MRI in which one or the other had interpreted the presence of bursitis with or without the presence of concomitant HATP (see Figure 2 for an example), and a consensus on the final evaluation was registered. We registered HATP on a binary level. HATP included tendinopathies, tendon tears or tendon ruptures without distinguishing between these.

Other conceptual findings were registered according to Table 1 (see Figure 2).

Data management and statistics

REDCap was used as data extraction system and an a priori extraction report form was made comprising the information listed in Table 1. The two evaluators were blinded to each other’s interpretation of the MRI and X-ray. Continuous and categorical data are presented as medians with interquartile ranges; binary data, as proportions. All statistical analyses were conducted with Stata 16.

Ethics, registration, funding and conflicts of interest

All patients included in the study provided their written consent for participation. The study was conducted without funding. No authors had any conflicts of interest to report.

Trial registration: According to current national legislature, no formal ethical approval was required for this study. The National Data Protection Agency approved the study (1-16-02-124-19).

RESULTS

Two (2%) patients had isolated greater trochanteric bursitis with no associated HATP. Thirty (25%) patients had HATP with some elements of inflammation (high-intensity signals) in the greater trochanteric bursa (as an example of this, see Figure 2). Five (4%) patients had HATP with bursitis in the sub-gluteus minimus bursa with no bursitis in the greater trochanteric bursa. The remaining patients (69%) had no high-intensity signals indicating trochanter-related bursitis.

A total of 88(93%) patients had a Kellgren-Lawrence grade ≤ 1 on X-ray. Four patients (4%) had calcifications lateral to the greater trochanter on X-ray (see Figure 2).

DISCUSSION

We found that isolated trochanteric bursitis on MRI was very rare in patients referred with refractory LHP to our public outpatient orthopaedic clinic. In most cases, a positive finding of some degree of trochanteric bursitis was associated with HATP. The vast majority of patients with refractory LHP in our cohort did not have MRI signs of trochanteric bursitis. This finding is important as most patients with LHP have, historically, been treated for their bursitis exclusively, primarily with corticosteroid injections. We believe that our findings reveal a need for a shift in focus towards treatment of HATP, especially in refractory LHP patients.

We also found that the majority of patients with refractory LHP do not have manifest radiological coxarthrosis. Furthermore, in a few cases, the LHP could potentially be attributed to findings of tendinitis calcaria.

Relation of our findings to selected literature

MRI may aid substantially in diagnosing (and differentiating) HATP in refractory LHP. Other studies have reported on the appearance of the lateral hip structures relating to the greater trochanter [15, 16]. These studies are the foundation for the approach used in the evaluation of the MRI in patients with LHP seen at our outpatient clinic. In a study by Cvitanic et al. [14] on 45 patients, the authors concluded differentiation between pathologies would be supported by the presence of a high-intensity signal, defined as hyperintensity relative to fat on T2-weighted images or iso-intensity relative to the contents of the urinary bladder on fat-suppressed T2-weighted images, and the relative location of these to the greater trochanter. We also believe that using T2 hyperintensity signals superiorly compared to laterally to the greater trochanter as a criterion for tears may be important information in the preoperative assessment of the abductor tendons of the hips and in determining whether surgical repair is indicated. If a high-intensity area is present superior to the top of the greater trochanter, this is strongly indicative of a tear (see Figure 1).

Blankenbaker et al. [17] evaluated 131 patients for T2 hyperintensity around the greater trochanter. If hyperintensity was present, this was reported as localised above, below or both above and below the greater trochanter. The main findings of the study were that bursal fluid and peritrochanteric oedema and HATP can be seen in asymptomatic hips. This is important as the specific physiological cause(s) of pain in LHP have not been established. So, merely treating an inflamed bursa may relieve some parts of the pain but omits others. Furthermore, treating a rupture may not yield sufficient pain relief if other structures are involved in generating pain. A need exists to further evaluate the relationship between the duration and degree of the reported LHP to the MRI findings and to qualify the specific “objective“ MRI findings as described in Table 1 in LHP patients and in relevant matched patient groups.

Chi et al. [19] evaluated 185 patients and found that HATP appears to progress from tendinosis to tendon tears with advancing age and with associated progression in muscle atrophy. Furthermore, a positive association between greater trochanteric bursitis and tendon pathology with advancing age was found, suggesting that HATP may often be the true underlying cause of pain, as also found in our cohort.

Limitations

The cohort established in this study is believed to fairly represent the LHP population of the majority of orthopaedic outpatient hip clinics. However, certain limitations to the present study should be acknowledged. Like all observational studies, our study is prone to both information and selection bias and results must be interpreted accordingly. The cohort was formed during a period in which the two surgeons’ knowledge of HATP increased, and MRI was performed on a case-by-case decision at the surgeon’s discretion. We were unable to obtain completely uniform MRI protocols across the numerous radiologic centres involved. This is a major weakness, but we believe that all MRIs were suitable for evaluation of the primary endpoint – the presence of trochanteric bursitis. The interpretation of both objective and subjective findings may be questioned as our approach was designed to be used also by clinicians with no distinct expertise in musculoskeletal MRI. Tendon elongation or disruption was not registered as this may be very difficult to assess for clinicians/non-expert radiologists. Rather, tendon appearance was registered as a binary event; normal or abnormal. Neither the format nor the configuration of high-intensity signals was evaluated [17, 20] as the mere presence of such singals indicated pathologies [18], which was the relevant outcome in this study. The Goutallier/Fuchsclassification grade was determined on the image with the highest fat area in relation to the total area. As fatty infiltration or muscle atrophy appears three-dimensional, but is only evaluated on a single slide, this may interfere with the clinical utility of this information. Interestingly, a recent systematic review and meta-analysis reported that high-grade fatty infiltration of the hip abductor muscles resulted in less improvement measured by the Harris Hip Score, but this did not influence VAS pain scores after surgery.

CONCLUSION

Isolated trochanteric bursitis as the cause of refractory LHP appears to be rare. We believe that the presence of HATP in LHP has been severely underestimated, and a shift in focus towards treatment of these structures is necessary.

Correspondence Marie Bagger Bohn. E-mail: Mabohn@rm.dk
Accepted 11 May 2022
Conflicts of interest Potential conflicts of interest have been declared. Disclosure forms provided by the authors are available with the article at ugeskriftet.dk/dmj
Cite this as Dan Med J 2022;69(7):A09210714