new minimally invasive option for the treatment of gluteal

ORTHOPEDICS | Healio.com/Orthopedics

n Feature Article

abstractFull article available online at Healio.com/Orthopedics. Search: 20121120-11

Gluteal muscle contracture is a clinical syndrome that involves contracture and distor-tion of the gluteal muscles and fascia fibers due to multiple causes. Physical examina-tion demonstrates a characteristic gait due to hip adduction and internal thigh rotation. This study introduces a new minimally invasive method for surgical release of gluteal muscle contracture. Patients with gluteal muscle contracture were assigned to 4 cat-egories: type A, contracture occurred mainly in the iliotibial tract; type B, contracture occurred in the Iliotibial tract and gluteus maximus; type C1, movement of the con-traction band was palpable and a snapping sound was audible during squatting; and type C2, movement of the contraction band was not palpable or almost absent and a snapping sound was audible during squatting. This classification method allowed pre-diction of the anatomic location of these pathological contractures and determination of the type of surgery required. Four critical points were used to define the operative field and served as points to mark a surgical incision smaller than 4 mm. The con-tracture was easily released in this carefully marked operative field without causing significant neurovascular damage. Over a period of 5 years, between March 2003 and June 2008, the authors treated 1059 patients with this method and achieved excellent outcomes. Most patients were fully active within 12 weeks, with the assistance of an early postoperative rehabilitation program. The most significant complication was a postoperative periarticular hematoma, which occurred in 3 patients within 10 days postoperatively and required surgical ligation of the bleeding vessel.

Messrs Ye and Yu are from the Department of Orthopedic Surgery, Mr Chen is from the Department of Nursing, Geriatric Hospital, Yangpu District, and Messrs Zhou and Xia and Dr Xu are from the Department of Orthopedic Surgery, Changhai Hospital, The Second Military Medical University, Shanghai, China.

Messrs Ye, Zhou, and Xia contributed equally to this study.Messrs Ye, Zhou, Xia, Chen, and Yu and Dr Xu have no relevant financial relationships to disclose.This study was supported by the Shanghai Health Bureau–sponsored research project (20100492175).Correspondence should be addressed to: Shuogui Xu, PhD, Department of Orthopedic Surgery,

Changhai Hospital, The Second Military Medical University, Shanghai, China ([email protected]). doi: 10.3928/01477447-20121120-11

New Minimally Invasive Option for the Treatment of Gluteal Muscle ContractureBin Ye, BS; PanYu Zhou, MS; Yan Xia, MS; YouYan Chen, BS; Jun Yu, MS; Shuogui Xu, PhD

Figure: Photograph showing the affected hip in the maximal possible degree of flexion, adduction, and internal rotation to place the majority of the contrac-tion bands under tension (A). Photograph showing that the contraction bands within the iliotibial tract should be divided in the position of hip extension to the neutral 0° with maximal hip adduction and internal rotation (B).

A

B

e1692

DECEMBER 2012 | Volume 35 • Number 12

Gluteal Muscle contracture | Ye et al

Gluteal muscle contracture is a clinical syndrome character-ized by fibrosis and contracture

of the gluteal muscles and their associ-ated soft tissues. These changes lead to a limitation of hip movement and, in some cases, anatomical deformity.1-4 Gluteal muscle contracture was first reported by Valderrania2 at a scientific meeting of the British Orthopedic Association in London in 1969. Intraoperatively, fibrous contrac-tion bands were found in the skin, subcu-taneous tissue, gluteus maximus, gluteus medius, gluteus minimus, other external rotators, and joint capsule.1,2,4 Pathologic examination showed that the resected tis-sue was fibrous, similar to that of a normal tendon, with all fibers essentially parallel.4 Gluteal muscle contracture, either congen-ital or acquired, is more common in Asian than Caucasian patients.3

Patients with gluteal muscle contrac-ture typically present with fixed hip ab-duction and external rotation when squat-ting from the standing position. Physical therapy involving stretching exercises is usually prescribed for nonoperative man-agement. Once contractures are estab-lished, exercises to stretch the shortened muscles do not work unless the contrac-tion bands are surgically divided.3 Surgery for gluteal muscle contracture mainly in-volves division of the fibrotic contraction bands, allowing normal relaxation of the gluteal muscles.2 A procedure to system-atically identify these fibrotic bands has not been described.

Traditionally, contracture release was performed through a 6- to 8-cm incision in the affected hip. The resulting surgical trauma contributed to hematoma forma-tion, wound complications, slow postop-erative recovery,3,4 and unsightly scarring. In a series of 187 patients treated with the open excision method, severe cicatricial band formation occurred in 62 patients, hematoma in 6, wound infection in 3, and wound dehiscence in 1.5

The traditional open excision method involved a long incision or skin grafting

that could restrict mobility of the skin. Postoperative pain from the sutures and long incision and patient fear of opening the wound with movement can impede postoperative rehabilitation. Patients re-quiring Z-lengthening of the fibrous con-tracture of the iliotibial tract have been reported to need 1 to 3 months to achieve full range of active hip flexion postop-eratively.6 Hypertrophic scarring has also been frequently reported after traditional releasing incisions.7,8 Zhao et al7 reported that 64 of 129 surgical patients developed hypertrophic scarring averaging 7 mm in width. Such hypertrophic or keloid scars in the buttock or thigh have a negative psychological effect on patients.

To avoid these problems, Liu et al9 used an arthroscope introduced through a key-hole incision. Although arthros-copy may avoid a large skin wound and facilitate earlier rehabilitation after gluteal muscle contracture release, the technique itself has limitations.

The key to gluteal muscle contracture release surgery is to divide the contraction bands that limit the normal range of move-ment of the affected hip. The procedure usually involves intraoperative assess-ment of the limited hip movement while the patient is under anesthesia. Once the fibrotic contraction bands responsible for the limitation of hip motion are identi-fied, they are divided as the affected hip is gradually passively flexed, adducted, and internally rotated. However, as this happens, manipulation of the arthroscope within the artificial space created next to the greater trochanter becomes difficult. Gluteal muscle contracture (except type A) involves the external rotators of the hip or joint capsule, which are deep and difficult for the arthroscope to access ef-fectively. The efficacy of arthroscopic sur-gery is limited by “deep contractures that are not reachable with an arthroscopic in-strument” and “a small incision should be used.”8 The arthroscopic method requires a significant volume of normal saline to be infused into the operative field to cre-

ate a space for arthroscopy. This can have a negative effect on the healthy muscle for 45 to 90 minutes postoperatively.

This article introduces a new minimal-ly invasive method. Surgery is performed through a number of incisions smaller than 4 mm within 10 to 15 minutes (mean, 12 minutes) for bilateral gluteal muscle contracture. Between March 2003 and June 2008, a total of 1059 consecutive patients with gluteal muscle contracture were treated using this method.

Materials and MethodsThis study was approved by the

Institutional Review Board of Changhai Hospital. All patients signed consent forms to participate in the study. The study conformed to the Declaration of Helsinki.

Participants were 487 men and 572 women with a mean age of 23 years (range, 8-43 years) at surgery. The pa-tients included 1 brother and sister, 2 pairs of sisters, and 23 pairs of cousins. Gluteal muscle contracture involved both hips in 849 patients and 1 hip in 210 patients.

The contracted tissues included the skin, subcutaneous tissue, iliotibial tract, gluteus maximus and other external rota-tors, and joint capsule.5 To achieve a sat-isfactory correction of the deformity, all contraction bands were divided. Location of the contractures was predicted by the clinical findings and the following gluteal muscle contracture classification, guiding the type of surgery required:

Type A: Contracture occurred mainly in the iliotibial tract. The patient could squat with knees closed together throughout the en-tire process of squatting (Figures 1A-C). A snapping sound was heard, and the movement of a fi-brotic band (part of the iliotibial band) could be felt over the great-er trochanter during squatting.

○ Type A1: The lower extremi-ties could be fully crossed and overlapped on sitting (Figure 1D).

e1693


n Feature Article

○ Type A2: The lower extremi-ties could not be fully crossed and overlapped on sitting (Figure 1E).

Type B: Contracture occurred in 2 of the following structures: the ilio-tibial tract and superficial and deep fasciae of the gluteus maximus. A snapping sound was heard, and the movement of a fibrotic band could be felt over the greater trochanter during squatting. In addition, the patient was unable to pull both knees together during squatting (Figures 1A, F). The affected hip abducted and rotated externally as the hip was flexed close to 90° (Fig-ure 1F). The knees could be pulled together after passing this point and the affected hip fully flexed (Figure 1C). The lower extremities could not be crossed or overlapped after sitting (Figure 1G).

Type C: Contracture occurred not only in the abovementioned super-ficial structures, but also extended to the deep structures, including the gluteus medius, gluteus mini-mus, piriformis, and joint capsule. The knees turned outward when the patient was walking and could not be pulled together throughout the entire process of squatting (Figures 1A, F, H). The lower ex-tremities could not be crossed or overlapped after sitting, similar to type B (Figure 1G). ○ Type C1: The movement of

the contraction band was pal-pable, and a snapping sound was audible during squatting.

○ Type C2: The movement of the contraction band was not palpable or almost absent, and a snapping sound was audible during squatting, in-dicating significant adhesions between the contracted tis-sues. These adhesions more frequently occurred as a re-

sult of previous unsuccessful or only partially successful open contracture release.

Of the 1059 patients, 35 were classified as type A (including 12 type A1 and 23 type A2), 216 as type B, 779 as type C1, and 29 as type C2. Of the 1059 patients, 978 patients had a history of repeated gluteal intramuscular injection of various drugs (mainly antibiotics) during child-hood. In patients with bilateral asymmet-ric gluteal muscle contracture, the pelvis would tilt to the more severely involved side, which would lead to an apparent leg-length discrepancy and lumbar scoliosis. Ninety percent of these patients also had knee crepitus, which may have been the result of chronic stress of rotational mis-alignment of the knee joint as the patients attempted to adjust the outturned knee.

surgical techniqueAlthough contrac-

ture release can be per-formed using general anesthesia, the current authors preferred lum-bar spinal or epidural anesthesia. Epidural anesthesia had the least effect on the pa-tients’ general health.

Each patient was laid supine on the op-erating table. The sur-gical area of the subcu-taneous tunnel, where the contraction bands were located, was de-fined before washing the patient. The affect-ed hip was flexed, ad-ducted, and internally rotated to the maxi-mal possible degree. Using the palpation method (or a portable radiograph machine in obese patients), a hori-zontal line 3 cm inferi-

or to the tip of the greater trochanter where it meets the posterior margin of the greater trochanter was defined as point A (Figure 2A). The tip of the greater trochanter was defined as point B (Figure 2A). The hip was then extended to the neutral 0° posi-tion. Point C was located on the lateral hip 5 cm from the line joining the anterior su-perior iliac spine (Figure 2B). The lateral edge of the patella was defined as point D (Figure 2B), which was 5 cm away from point C toward the knee. The surgical area was defined by the quadrilateral of the points A, B, C, and D (Figure 2C).

According to the anatomic locations, the contraction bands were classified in the fol-lowing order: gluteus maximus, tensor fas-cia lata, iliac reinforcement beam, gluteus maximus muscle, gluteus minimus muscle, joint capsule, superficial and deep fascia, and skin. The assistant maintained the af-fected hip in the maximal possible degree of

Figure 1: Gluteal muscle contracture classification. Standing (A). Type A: In the process of squatting, a snap sound can be heard, and the glide of a fibrotic band (part of the iliotibial tract) can be felt over the greater tro-chanter while squatting (B). Squatting (C). Type A1: Lower extremities can be fully crossed and overlapped (D). Type A2: Lower extremities cannot be fully crossed and overlapped (E). Type B: In the process of squatting, the affected hip abducts and externally rotates as the hip is flexed close to 90° (F), and the lower extremities cannot be crossed and overlapped on sitting (G). Type C: The knees cannot be kept together when squatting (H).

1A 1B 1C 1D

1E 1F 1G 1H

e1694



flexion, adduction, and internal rotation to place the majority of the contraction bands under tension (Figure 3A). This was the ideal position for division of most gluteal muscle contractures, except for contraction bands within the iliotibial tract. These were divided with the hip extended to the neutral 0° position and maximal hip adduction and internal rotation (Figure 3B).

Point A was the entry point for division of contraction bands of the gluteus maxi-mus and its associated subcutaneous tissue (Figure 3A). Point B was the entry point for division of contraction bands of the gluteus medius and lateral hip joint capsule (Figure 3A). Point C was the entry point for division of contraction bands of the ten-sor fascia lata, iliotibial tract, and their as-sociated subcutaneous tissues (Figure 3B).

Surgery was performed with a scalpel designed specifically for contracture re-lease. A 333-mm incision was made. The cutting edge of the scalpel was directed toward the front of the patient when the incision was made from point A or B. For this approach, the incision did not extend past the line joining points C and D. This avoided injury to the femoral neurovascu-lar bundle. When entry was made at point C, the cutting edge of the scalpel was di-rected to the back of the patient. For this approach, the incision did not extend past the line joining points A and B in the verti-cal plane. In this way, the scalpel was natu-rally stopped by the greater trochanter me-dially, avoiding injury to the sciatic nerve.

When no audible snapping sound or pal-pable movement of the contraction bands

was detected, a volume of 60 to 100 mL of 1:250,000 iced adrenaline–saline solution was used to irrigate the wound 3 times, with a few mL left inside the wound deliberately. After release of the contralateral hip, an ex-amination was performed on the operating table to see whether the hip could be fully flexed and extended and the knees pulled to-gether (Figure 4A) and crossed over (Figure 4B) without audible snapping sounds or palpable movement of contraction bands over the greater trochanter. Release of a tight gluteal skin contracture in patients with type C2 gluteal muscle contracture was completed with numerous 3-mm skin incisions perpendicular to the longitudinal axis of the femur. The incisions were placed more than 3 mm apart (Figure 4C).

It is important to emphasize that an-other reason to maintain the intraoperative posture of maximal hip flexion, adduction, and internal rotation was to displace the contraction bands limiting the movement of the greater trochanter anterior to the posterior margin of the greater trochanter to ensure the safety of the surgery (Figure 5). The sciatic nerve could not move ante-rior to the posterior margin of the greater trochanter because it was located behind the piriformis muscle. When the contrac-ture extended to involve the structures posterior to the line joining points A and B, a postoperative rehabilitation program was required to achieve full range of mo-tion. Structures like the piriformis muscle are usually flexible and do not exert sig-nificant resistance to movement.

Figure 2: Photographs showing confirmation of points A and B (A) and points C and D (B) and the surgical target area (C).

2A 2B 2C

Figure 3: Photograph showing the affected hip in the maximal possible degree of flexion, adduction, and internal rotation to place the majority of the contraction bands under tension (A). Photograph showing that the contraction bands within the iliotibial tract should be divided in the position of hip extension to the neutral 0° with maximal hip adduction and internal rotation (B).

3A 3B

e1695


n Feature Article

Postoperative RehabilitationThe patient was laid flat on the bed

with the head slightly lowered for the first 3 hours postoperatively. The patient was allowed to flex the hip and knee joint and cross legs (either positively or passively) at 20-minute intervals as tolerated. The legs had to be crossed as much as possible to stretch out the released contracted mus-cles and minimize the possibility of post-operative hematoma formation. No further rehabilitation exercises were attempted on postoperative day 1. A rehabilitation pro-gram involving gradual full range of hip movement and stretching the previously contracted muscle was started on post-operative day 2. The patient was advised to avoid any injury to the surgical site for the first 3 weeks. Rehabilitation exercises, including strenuous rapid movement of the hip joint and greater stretching of the gluteal muscles, were introduced after the third week. The rehabilitation program was maintained for at least 6 months to achieve rapid recovery and a long-term optimal therapeutic outcome.

resultsThe patients were followed up for be-

tween 6 months and 5 years (mean, 2.5 years). The outcome of hip contracture correction was good in all patients.

According to the assessment crite-ria (Table 1), all 1059 patients achieved excellent results between 1 week and 9 months, with a mean of 2.6 weeks for

type A, 3.2 weeks for type B, 3.5 weeks for type C1, and 11.5 weeks for type C2 (Table 2).

No mortality or other long-term com-plications, such as permanent gluteal muscle weakness, or secondary neuro-vascular injury, such as sciatic nerve or femoral neurovasculature, occurred in the series. Short-term postoperative problems included pain, swelling, keloid formation of the surgical side, shuffling gait, mus-cular weakness around the hip joint, and asymmetric recovery in patients with bi-lateral hip involvement. All were usually resolved within 6 weeks to 6 months post-operatively.

One woman and 2 men developed a sudden painful hematoma 7 to 10 days postoperatively. The predisposing fac-

Figure 4: Photographs showing full hip flexion with knees kept together (A) and knees crossed over (B) and release of the tight gluteal skin contracture with numerous 3-mm skin incisions (C).

4A 4B 4C

Figure 5: Schematic of the posture of maximal hip flexion, adduction, and internal rotation. The contraction bands limiting the movement of the greater trochanter are displaced to the anterior of the posterior margin of the greater trochanter to ensure the safety of the operation.

5

Table 1

Evaluation Criteria forGluteal Muscle Contracture

I. Closing knees together while squatting and standing

35Squatting and standing freely

25Squatting and standing partly with help

15Squatting and standing wholly with help

05Cannot squat or stand

II. Crossing and overlapping the legs with 90° flexion of the hip and knee joints

35Crossing and overlapping the legs freely

25Crossing and overlapping the legs partly with help

15Crossing and overlapping the legs wholly with help

05Can not cross and overlap the legs

III. Ambulation

25No waddling gait involuntarily

15No waddling gait consciously

05Waddling gait consciously

IV. Gliding of fibrotic band in the iliotibial tract

25No gliding of fibrotic band and no resistance

15Gliding of fibrotic band, and resistance can be felt

05No gliding of fibrotic band, but resistance can be felt

Total clinical grade

9-105Excellent

7-85Good

0-65Poor

e1696



tor was sudden and rapid hip flexion and extension with the knees closed (dur-ing standing from the toilet in 2 cases and sitting on a chair in 1 case), leading to rupture of a circumflex femoral arte-rial branch at the neck of the femur. All 3 patients were treated successfully with explorative emergency surgery, during which a 3-cm longitudinal incision was made at the site of the tip of the greater trochanter. The bleeding arterial branch was identified and ligated. The 3 patients were rehabilitated as per the guidelines and achieved excellent outcomes within 1, 3, and 6 weeks, respectively. No such complication occurred beyond 10 days postoperatively. This uncommon but possible complication was disclosed to the patients, and the importance of grad-ual squatting and sitting during the first 3 weeks of postoperative rehabilitation was emphasized (Tables 1, 2).

discussionGluteal muscle contracture appears

to be caused by multiple factors. A his-tory of repeated intramuscular injections has been recognized as the most com-mon factor in gluteal muscle contrac-ture.1 Intramuscular penicillin therapy for febrile conditions has been a common medical practice in China since 1970.10,11

Gluteal muscle contracture is not uncom-mon in patients treated with intramuscular penicillin.12,13 Genetics may also play a role in the pathogenesis of gluteal muscle contracture because relatives of patients with gluteal muscle contracture are also at a higher risk of developing gluteal muscle contracture.5 Wolbrink et al4 reported a 9-year-old girl with an abduction contrac-ture of the shoulders and hips secondary to fibrous bands and observed that the girl’s mother had a less severe form of the same condition (a minimal abduction con-tracture in the arms).

The current authors postulated the most likely causative event in their series. First, either a large volume of agents or an irritating agent, such as benzyl penicil-lin, was administered to the gluteal region of the patient, commonly a febrile infant or child, in a repeated manner. Injections were usually twice a day over a period of 1 week. The resulting increased tissue pres-sure after injection and muscle inflamma-tion could lead to compartment syndrome, resulting in tissue necrosis and fibrosis. Second, children who later developed glu-teal muscle contracture were subjected to frequent courses of intramuscular penicil-lin due to frequent fevers. Fever is the re-flection of a systemic inflammatory state in the body. Such a patient would be at

higher risk of developing fibrous contrac-tures and not exercising due to muscular pain and swelling.

Because gluteal muscle contracture usually begins in early childhood, the fi-brotic contraction bands would not grow with other normal skeletal elements. The gluteal muscle contracture would become progressively worse and the child may move from type A to type C with age. The authors’ clinical experience, in addition to other reports,14 has shown that treatment with physical therapy7 has limited efficacy once contraction bands have formed.

Standard treatment with the arthroscop-ic method has been associated with exten-sive surgical trauma, unpleasant cosmetic results, wound dehiscence during rehabili-tation, and technical difficulties and limi-tations. The authors designed a minimally invasive surgical method for the treatment of gluteal muscle contracture to ensure the safety of the adjacent neurovascular structures. The vertical plane through the line joining points C and D should not be crossed anteriorly and medially to avoid injury to the femoral neurovascular bun-dle, the lateral cutaneous nerve of thigh, and the lateral femoral circumflex vessels. The horizontal plane through the line join-ing points A and B should not be crossed posteriorly (the blade will be stopped me-dially by the bony component of the fe-mur) to avoid injury to the sciatic nerve. The hip must be flexed, adducted, and internally rotated when the scar release is performed through the incisions made at point A or B to trap the sciatic nerve medial to the greater trochanter. The hip must be extended to the neutral 0° position when the scar release is performed through the incisions made at point C or D.

Early postoperative rehabilitation is essential for drainage of and prevention of development of wound hematoma and for rapid functional recovery. Movement must be performed gradually within the first 3 weeks to avoid injury to the surgical site during the healing period. Strenuous rehabilitation exercises are required after

Table 2

Time to Achieve Satisfactory Physical Therapy Resultsin 1059 Patients Achieving Excellent Resultsa

Postoperative Hospitalization, wk

GMC Type 1 3 6 9 12 24 36 Total

A1 2 7 3 0 0 0 0 12

A2 3 12 8 0 0 0 0 23

B 80 91 36 4 2 3 0 216

C1 118 510 135 6 3 5 2 779

C2 0 4 12 5 2 3 3 29

Total 210 625 186 15 7 11 5 1059

Abbreviation: GMC, gluteal muscle contracture. aData presented as No. of patients.

e1697


n Feature Article

the first 3 weeks and must be continued for at least 6 months for a permanent opti-mal therapeutic outcome.

references 1. Cai JH, Gan LF, Zheng HL, et al. Iliac hyper-

dense line: a new radiographic sign of gluteal muscle contracture. Pediatr Radiol. 2005; 35(10):995-997.

2. Valderrania JAF. A cause of limited flexion and adduction of the hip in children. J Bone Joint Surg Br. 1970; 52(1):179.

3. Howard RC. Iatrogenic quadriceps and gluteal fibrosis. J Bone Joint Surg Br. 1971; 53(2):354.

4. Wolbrink AJ, Hsu Z, Bianco AJ. Abduction contracture of the shoulders and hips second-ary to fibrous bands. J Bone Joint Surg Am. 1973; 55(4):844-846.

5. Hang YS. Contracture of the hip second-ary to fibrosis of the gluteus maximus muscle. J Bone Joint Surg Am. 1979; 61(1):52-55.

6. Shen YS. Abduction contracture of the hip in children. J Bone Joint Surg Br. 1975; 57(4):463-465.

7. Zhao CG, He XJ, Lu B, Li HP, Wang D, Zhu ZZ. Classification of gluteal muscle con-tracture in children and outcome of differ-ent treatments. BMC Musculoskelet Disord. 2009; 10:34.

8. Brignall CG, Stainsby OD. The snapping hip: treatment by Z-plasty. J Bone Joint Surg Br. 1991; 73(2):253-254.

9. Liu YJ, Wang Y, Xue J, et al. Arthroscopic gluteal muscle contracture release with ra-diofrequency energy. Clin Orthop Relat Res. 2009; 467(3):799-804.

10. Chung DC, Ko YC, Pai HH. A study on the prevalence and risk factors of muscular fi-brotic contracture in Jia-Dong Township, Pingtung County, Taiwan. Gaoxiong Yi Xue Ke Xue Za Zhi. 1989; 5(2):91-95.

11. Huang Y, Li J, Lei W. Gluteal muscle con-tracture: etiology, classification and treat-ment. Chin J Orthop. 1999; 19(2):106-108.

12. Sun X. An investigation on injectional glu-teal muscle contracture in childhood in Mi-anyang city. Zhonghua Liu Xing Bing Xue Za Zhi. 1990; 11(5):291-294.

13. Gao GX. Idiopathic contracture of the glu-teus maximus muscle in children. Arch Or-thop Trauma Surg. 1988; 107(5):277-279.

14. Ko JY, An KN, Yamamoto RJ. Contrac-ture of the deltoid muscle: results of dis-tal release. J Bone Joint Surg Am. 1998; 80(2):229-238.

e1698

new minimally invasive option for the treatment of gluteal

Documents