© Wichtig Editore, 2009
Hip International / Vol. 19 no. 4, 2009 / pp. 00-00
Lateral sided hip pain is a common problem encountered
in the orthopaedic clinic. However trying to elucidate and
treat its cause can be difﬁcult. Potential causes include;
tumours, infection, osteoarthritis of the hip, osteonecrosis
of the femoral head, spinal disease and entrapment neu-
ropathies. Trochanteric bursitis however is the most fre-
quent cause of lateral sided hip pain. This entity usually
presents with dull aching lateral sided pain over the greater
trochanter that is exacerbated by lateral pressure over the
area, weight bearing and resisted abduction. This diagno-
sis can usually be conﬁrmed if symptomatic relief is gained
with a local anaesthetic injection into the area.
Surgical repair of chronic tears of the hip abductor
HYWEL DAVIES1, SOHELIA ZHAEENTAN2, ADEL TAVAKKOLIZADEH3, GREGORY JANES4
1 Avon Orthopaedic Centre, Bristol - UK
2 Kings College Hospital, London - UK
3 Täby Närsjukhus and Danderyd University Hospital, Stockholm - Sweden
4 Perth Orthopaedics and Sports Medicine, Perth - Australia
ABSTRACT. Lateral sided hip pain frequently presents to the orthopaedic clinic. The most common
cause of this pain is trochanteric bursitis. This usually improves with conservative treatment. In a few
cases it doesn’t settle and warrants further investigation and treatment. We present a series of 28
patients who underwent MRI scanning for such pain, 16 were found to have a tear of their abductors.
All 16 underwent surgical repair using multiple soft tissue anchors inserted into the greater trochanter
of the hip to reattach the abductors.
There were 15 females and 1 male. All patients completed a self-administered questionnaire pre-
operatively and 1 year post-operatively. Data collected included: A visual analogue score for hip pain,
Charnley modiﬁcation of the Merle D’Aubigne and Postel hip score, Oxford hip score, Kuhfuss score
of Trendelenburg and SF36 scores.
Of the 16 patients who underwent surgery 5 had a failure of surgical treatment. There were 4 re
ruptures, 3 of which were revised and 1 deep infection which required debridement. In the remain-
ing 11 patients there were statistically signiﬁcant improvements in hip symptoms. The mean change
in visual analogue score was 5 out of 10 (p=0.0024) The mean change of Oxford hip score was 20.5
(p=0.00085). The mean improvement in SF-36 PCS was 8.5 (P=0.0020) and MCS 13.7 (P=0.134). 6
patients who had a Trendelenburg gait pre-surgery had normal gait 1 year following surgery.
We conclude that hip abductor mechanism tear is a frequent cause of recalcitrant trochanteric pain
that should be further investigated with MRI scanning. Surgical repair is a successful operation for
reduction of pain and improvement of function. However there is a relatively high failure rate. (Hip
International 2009; 19: 00-000)
KEY WORDS. Avulsion, Hip Abductor, Tear, Trochanteric pain
Repair of hip abductor tears
Patients and Methods
From July 2006 to February 2008, 28 patients with lateral
hip pain who had had little or no response to non-surgical
treatment were further investigated with MRI scanning of
their affected hip. Usually the patients had been treated
elsewhere initially but all reported having received at least
1 steroid injection to the trochanteric region. 22 of the 28
patients who were investigated with MRI had had 2 or
more injections. 26 of 28 patients had received a course of
physiotherapy aimed at rehabilitating their abductors. Pa-
tients with previous arthroplasty or surgery to the affected
hip were excluded from the study. All patients who were
identiﬁed as having a tear on MRI were offered surgery to
repair the tendon, or tendons. Figure 1 demonstrates the
typical radiographic ﬁndings on MRI scanning.
16 patients were identiﬁed as having a tear of the hip ab-
ductors and all of these underwent surgery following con-
sultation with the senior author. Of the 16 patients who
underwent surgery there were 15 females and 1 male. The
mean age of the patients was 63 (range 47 to 82). There
was no history of signiﬁcant trauma to the hip in any of the
patients. The mean duration of symptoms prior to surgery
was 23 months (range 6 to 48 months).
Under general anaesthesia the patient was positioned in
the lateral decubitus position. A lateral incision was made
Treatment of this entity is non-surgical in the ﬁrst instance.
This entails the use of Non-steroidal anti-inﬂammatory,
physical therapy and steroid injection into the area of the
bursa. A number of cases however remain recalcitrant and
may require surgical treatment. Tearing of the common in-
sertion of the tendons of gluteus medius and minimus on to
the greater trochanter of the proximal femur have relatively
recently been described as a potential cause of this pain.
This clinical entity has been described as a rotator cuff tear
of the hip (1, 2) as it appears to show some similarities with
tears of the rotator cuff musculature in the shoulder. It has
been postulated that pelvic morphology, leg-length discrep-
ancy and a high valgus angle of the knee may produce a
biomechanical environment that predispose patients to in-
jury as the greater trochanter impinges on a tight iliotibial
band. Tension within the iliotibial band may result in frictional
trauma to the gluteal tendons, just as the acromial process
causes trauma to the rotator cuff in the shoulder (3-7).
Tears of the hip abductors have been detected in 22% of
patients undergoing ﬁxation of fractures of the neck of fe-
mur1 and in 20% of patients undergoing total hip replace-
ment (8). Undoubtedly therefore there are large numbers of
such tears which remain asymptomatic.
We present a series of patients who were investigated for
recalcitrant trochanteric pain syndrome with MRI. All pa-
tients who were shown to have torn abductors underwent
surgical repair and have been prospectively followed up.
To our knowledge this is the ﬁrst prospective study review-
ing the results of treatment of such lesions.
Fig. 1 - T2 weighted MRI image demonstrating tear of hip abductor
complex at insertion onto the greater trochanter.
Fig. 2 - Intra-operative ﬁndings post debridement demonstrating
complete detachment of gluteus minimus and maximus.
Davies et al
weeks. They commence active range of movement exer-
cises at 6 weeks and resisted exercises at 12 weeks.
Functional outcome analysis
All patients completed a self-administered questionnaire
pre-operatively and at 1 year post-operatively. Data collect-
ed included: A visual analogue score for hip pain, the Merle
D’Aubigne and Postel hip score9, Oxford hip score (10),
Kuhfuss score of Trendelenburg (11) and SF36 (12) scores.
All information was gathered onto a Microsoft excel
spreadsheet (Microsoft, Richmond, Washington) for analy-
sis. Comparison between pre- and post-operative scores
was performed with a paired t-test. Signiﬁcance was ac-
cepted with a p value of less than 0.05.
At surgery 4 of the tears were deemed to involve both the
gluteus medius and minimus tendons with at least 2/3 de-
tachment of the medius tendon. We called this a severe
tear. In 2 of these, part of the gluteus medius muscle was
retracted signiﬁcantly. 8 of the tears involved gluteus me-
dius only and had between 1/3 and 2/3 detachments of
the medius tendon. We called this a moderate tear. 4 tears
involved less than 1/3 of the gluteus medius tendon. We
called this a mild tear.
Of the 16 patients who underwent surgery 11 completed
full follow up. Of the 5 remaining patients, 4 had re-tears
of the repair during the follow up period. These re-tears
were all proven on MRI scanning. The indication for repeat
scanning was poor function and return of Trendelenburg at
follow up. 3 of these have been revised with repeat repair.
1 of these patients with a partial re-tear has decided not to
undergo further surgery but remains under review. 1 Pa-
tient sustained a deep wound infection and had to undergo
debridement. These 5 patients have not been included in
the follow up group. 2 of these failures had large tears, 1
had a moderate tear and 1 had a small tear as found at
the initial operation. 1 of the failures with a large initial tear
was an insulin dependant diabetic otherwise there were
no discernable co morbidities that may have contributed
to failure. Length of symptoms prior to presentation didn’t
appear to have any affect on the likelihood of repair fail-
centred over the trochanter of the hip. The incision was
deepened through the fascia lata exposing the greater
trochanter and hip abductors. The trochanteric bursa was
excised and any defect in the insertion of the hip abduc-
tors was identiﬁed. An attempt was made to document
the extent of the defect. Figure 2 demonstrates the intra-
operative appearance of the torn abductors. Frequently
the Abductors had reattached to the trochanter with poor
quality scar tissue in an elongated position. This scar tis-
sue was excised. Both Gluteus Medius and Minimus were
identiﬁed as separate layers. The insertion points of the
abductors on the trochanter were identiﬁed and debrided.
A high-speed burr was then used to produce a roughened
surface on the trochanteric attachment area to improve the
healing potential. In some cases osteophytes were found
on the femoral neck that would have prevented ﬁrm reat-
tachment of the abductors, these were removed. Suture
anchors were inserted into the abductor footprint and the
sutures used to suture the free edge of the avulsed abduc-
tors back onto the anchors. A double row conﬁguration of
sutures was used to produce a ‘watertight’ seal akin to a
rotator cuff repair in the shoulder (Fig. 3). The free edge of
the abductors was then over sewn with a heavy absorb-
able suture. The hip is ﬁnally put through a range of move-
ment to test the quality of the repair. The surgical wound
was closed in standard fashion.
Postoperatively patients are allowed to mobilise touch
weight bearing with the aid of crutches for a period of 6
Fig. 3 - same hip as ﬁgure 2 demonstrating positioning of sutures
leading from suture anchors with double row conﬁguration
Repair of hip abductor tears
failure post total hip replacement (16). In 9 patients followed
up at a mean of 4.5 years 4 were deemed to be signiﬁcantly
better, 3 slightly better and 2 had no improvement.
Tendinitis, tears, and ruptures of the gluteal tendons are
most accurately diagnosed by magnetic resonance im-
aging, with coronal T1-weighted images with fat satura-
tion and axial fast-spin-echo T2-weighted images. These
techniques may reveal calciﬁcation within the tendons and
oedema within the muscle and adjacent compartments3.
The number of patients presenting with intractable greater
trochanteric pain who were found to have an abductor
tear on MRI scanning in our series is similar to a previous
study performed by Bird (17) In their series of 24 patients
presenting with intractable trochanteric pain 11 patients
(45.8%) were found to have a tear on MRI scanning. In our
study 16 0f 28 patients (57%) were found to have a tear
on scanning. This suggests that MRI scanning should be
advocated in all patients in whom Trochanteric pain fails to
resolve with conservative measures, as hip abductor avul-
sion is probably an under-diagnosed entity.
We feel that tears of the hip abductor mechanism represent
one end of a spectrum of the pathological process with
bursitis being at the other end, this probably progresses
in some patients to a tendonitis that may ﬁnally result in
partial or complete tearing of the hip abductors.
In our series there is a relatively high rate of repair failure.
3 of 16 patients having a complete re-tear requiring revi-
sion surgery and 1 sustaining a partial re-tear, which was
managed conservatively. On further analysis of the clinical
notes of the patients that failed we were unable to estab-
lish any differences between them and the patients that had
successful surgery. The failure rate may reﬂect poor qual-
ity tissues that are frequently found in these patients. The
numbers involved in our study were not sufﬁcient to draw
any signiﬁcant conclusions with regard to initial size of tear
and likelihood of recurrence following repair however 2 of
the failures did occur in the severe tear group. It would seem
likely that a similar phenomenon would occur in the hip ab-
ure and the failures were not in any particular age range.
2 of the failures had received 2 steroid injection at various
points prior to surgery 1 had received 1 injection and 1,
3 injections. Numbers were not sufﬁcient to demonstrate
any correlation between number of injections of steroid
and failure of repair.
Table I demonstrates the outcome scores pre and post-
surgery. There was signiﬁcant improvement in all measure-
ments post surgery apart from the Sf-36 MCS score, which
showed a trend towards improvement but was not statisti-
Evaluation of Trendelenburg according to Kuhfuss pre-oper-
atively demonstrated 4 patients with mild Trendelenburg, 4
patients with moderate and 3 patients with severe Trendelen-
burg. Post surgery all patients improved. There were 6 patients
rated as no Trendelenburg, 4 with mild and 1 moderate.
The 4 re-tears had a mean oxford hip scores of 26.2 at the
time of detection as opposed to the mean hip score of 38.4
at follow up of the successful repairs. The revision proce-
dures have not been included in this review.
We have presented a prospectively followed series of pa-
tients undergoing hip abductor repair for recalcitrant tro-
chanteric pain. This procedure provides for good relief of
pain and improvement in hip function. However there is a
relatively high failure rate.
There have been a number of previous isolated case reports
(7, 13) and a small case series (14) reporting on operative
repair of primary hip abductor avulsion. Kagan (14) reported
on 7 repairs with a mean follow up of 45 months. All re-
ported good relief of pain, all bar 1 patient returned to nor-
mal function. There has also been a report of the surgical
technique of arthroscopic repair of the hip abductors (15)
although this didn’t present any follow up. Surgical repair
of the hip abductors has also been reported on following
TABLE I - OUTCOME SCORES (MEAN VALUES)
ScorePre-operative Post-operative Improvement p-value
MDP 10.5 15 4.5 0.0003
Oxford Hip 21.4 38.9 17. 5 0.0008
Sf-36 PCS 28.4 40.2 11. 8 0.0020
Sf-36 MCS 54.9 59.4 4.5 0.134
VAS, Visual analogue scale; MDP, Merle D’Aubigne Postel,: PCS, Physical component score; MCS, Mental component score.
Davies et al
Work carried out at Perth Orthopaedics and Sports Medicine,
Perth - Australia.
Institutional Review Board (IRB)/Ethics Committee approval was
Conﬂict of interest statement: No Authors have proprietry
Address for correspondence:
10 Ormerod Road
Bristol, BS9 1BB, UK
ductors as in the shoulder rotator cuff where it has been
proven that there is a reduced potential for repair as tear
size increases (18). There may be underlying biomechanical
factors that needs to be addressed or our surgical technique
may need to be reﬁned to improve the results of repair.
We are aware of the main limitation of this study, which is
the relatively small sample size. We hope to follow up fur-
ther patients in the future.
In conclusion tears of the abductor mechanism of the hip are
an underdiagnosed cause of recalcitrant greater trochanteric
pain. The use of MRI scanning can help in the diagnosis. Op-
erative repair is very successful for decreasing pain and im-
proving function however there is a relatively high failure rate.
1. Bunker TD, Esler CN, Leach WJ. Rotator-cuff tear of the hip.
J Bone Joint Surg Br 1997; 79: 618-20.
2. Kagan A, 2nd. Rotator-cuff tear of the hip. J Bone Joint Surg
Br 1998; 80: 182-3.
3. Kingzett-Taylor A, Tirman PF, Feller J, et al. Tendinosis and
tears of gluteus medius and minimus muscles as a cause of
hip pain: MR imaging ﬁndings. AJR Am J Roentgenol 1999;
4. LaBan MM, Weir SK, Taylor RS. ‘Bald trochanter’ sponta-
neous rupture of the conjoined tendons of the gluteus me-
dius and minimus presenting as a trochanteric bursitis. Am
J Phys Med Rehabil 2004; 83: 806-9.
5. Lonner JH, Van Kleunen JP. Spontaneous rupture of the glu-
teus medius and minimus tendons. Am J Orthop 2002; 31:
6. Traycoff RB. “Pseudotrochanteric bursitis”: the differen-
tial diagnosis of lateral hip pain. J Rheumatol 1991; 18:
7. Fisher DA, Almand JD, Watts MR. Operative repair of bilateral
spontaneous gluteus medius and minimus tendon ruptures. A
case report. J Bone Joint Surg Am 2007; 89: 1103-7.
8. Howell GE, Biggs RE, Bourne RB. Prevalence of abductor
mechanism tears of the hips in patients with osteoarthritis.
J Arthroplasty 2001; 16: 121-3.
9. D’Aubigne RM, Postel M. Function al results of hip arthro-
plasty with acrylic prosthesis. J Bone Joint Surg Am 1954;
10. Dawson J, Fitzpatrick R, Carr A, Murray D. Questionnaire
on the perceptions of patients about total hip replacement.
J Bone Joint Surg Br 1996; 78: 185-90.
11. Kuhfuss W, Schildhauer M, Tonnis D. [The inﬂuence of surgi-
cally introduced muscle tension on the function of total hip
prostheses (author’s transl)]. Z Orthop Ihre Grenzgeb 1975;
12. Ware JE, Jr., Sherbourne CD. The MOS 36-item short-form
health survey (SF-36). I. Conceptual framework and item se-
lection. Med Care 1992; 30: 473-83.
13. Chebil M, Mezghani M, Kooli M, Zlitni M. [From diagnosis to
treatment of gluteus medius tendon tear of the hip: a case
report]. Ann Readapt Med Phys 2007; 50: 48-54.
14. Kagan A, 2nd. Rotator cuff tears of the hip. Clin Orthop Relat
Res 1999; 368: 135-40.
15. Voos JE, Rudzki JR, Shindle MK, Martin H, Kelly BT.
Arthroscopic anatomy and surgical techniques for peritro-
chanteric space disorders in the hip. Arthroscopy 2007; 23:
16. Web er M, Berry DJ. Abductor avulsion after primary total hip
arthroplasty. Results of repair. J Arthroplasty 1997;12-2:202-6.
17. Bird PA, Oakley SP, Shnier R, Kirkham BW. Prospective
evaluation of magnetic resonance imaging and physical ex-
amination ﬁndings in patients with greater trochanteric pain
syndrome. Arthritis Rheum 2001; 44: 2138-45.
18. Matthews TJ, Hand GC, Rees JL, Athanasou NA, Carr AJ.
Pathology of the torn rotator cuff tendon. Reduction in po-
tential for repair as tear size increases. J Bone Join Surg Br
2006; 88: 489-95.