Abductor recovery after muscle-sparing periacetabular osteotomy using a lateral approach

ABSTRACT

To decrease hip abductor dysfunction after periacetabular osteotomy using a lateral/trochanteric approach, we aimed to modify transposition osteotomy of the acetabulum (TOA) to not cut the greater trochanter and abductor–iliac crest detachment. We subsequently compared abductor muscle strength recovery between TOAs with [conventional TOA (C-TOA)] and without [modified TOA (M-TOA)] trochanteric osteotomy. C-TOA and M-TOA were performed in 27 and 34 hips, respectively. Hip abduction, flexion and knee extension muscle strength were measured preoperatively and at 3, 5, 10, 24 and 52 weeks postoperatively. The muscle strength ratio of the affected and contralateral lower limbs was compared between the C-TOA and M-TOA groups. Neither the mean Merle d’Aubigné–Postel score at the final follow-up nor the postoperative center-edge angle showed significant differences between the M-TOA and C-TOA groups (15.7 versus 16.4 points; P = 0.25 and 38.5° versus P = 0.62 and 39.8°, respectively). The mean muscle strength ratios of hip abduction at 5, 12 and 24 weeks postoperatively were significantly higher in the M-TOA group than in the C-TOA group (0.62 versus 0.39, 0.76 versus 0.59 and 0.94 versus 0.70; P = 0.03, 0.04 and 0.01, respectively). There were no significant differences between groups at Postoperative Week 52 (P = 0.36). Discomfort at the greater trochanter was observed in 18 hips (66.7%) in the C-TOA group but only in 4 hips (11.2%) in the M-TOA group. In conclusion, M-TOA is less invasive than C-TOA and allows an earlier recovery of abductor muscle strength without significant correction loss.

INTRODUCTION

Developmental dysplasia of the hip is characterized by several morphological abnormalities, including acetabular dysplasia, decreased acetabular coverage and excessive femoral anteversion [1–3]. These morphological characteristics cause an abnormal concentration of joint stress, which successively generates premature hip osteoarthritis (OA) [4]. Periacetabular osteotomy (PAO) is an established procedure to reduce pain and prevent OA progression in young patients with acetabular dysplasia [5–14].

PAO can be categorized based on lateral and anterior approaches. Rotational acetabular osteotomy (RAO) and transposition osteotomy of the acetabulum (TOA) are performed using the lateral approach [8–14], while Ganz osteotomy and curved PAO are performed using the anterior approach [5–7]. PAO via the anterior approach is less invasive to the abductors than PAO via the lateral approach because PAO via the anterior approach does not require abductor–iliac crest detachment for an osteotomy from the inner pelvic wall. However, technical difficulties and complications unique to the anterior approach have been reported [15–17]. The lateral approach can broadly expose the osteotomy area, while a detachment of the hip abductor muscles (abductors) is necessary to achieve osteotomy from outside the iliac bone in PAO via the lateral approach. TOA is performed using the trochanteric approach [10, 13, 14, 18, 19], and RAO detaches the anterior insertion of the abductor muscles from the iliac crest [20]. Thus, postoperative abductor dysfunction over a certain period is inevitable in these lateral approaches.

To minimize post-TOA abductor dysfunction, we developed a novel method, aiming to (i) not cut the greater trochanter and (ii) avoid abductor–iliac crest detachment. A spoon-shaped Wagner-type chisel enabling the osteotomy of the iliac bone from the anterior part was developed (Fig. 1). Subsequently, this osteotomy line was connected with the osteotomy line extending from the posterior section using a conventional curved chisel, helping in achieving curved osteotomy in the same manner as in the conventional method without cutting the greater trochanter and inducing abductor–iliac crest detachment. This study aimed to introduce this modified TOA (M-TOA) and compare the recovery of muscular function around the hip joint between M-TOA and conventional TOA (C-TOA).

Fig. 1.

A spoon-shaped chisel (a and b). A spoon-shaped Wagner-type chisel was developed to be double-curved with a dent in the center, and the curves are 40 mm in radius and width. The 40-mm curvature radius matches that of the curved chisel, which is usually used in the C-TOA of the acetabulum.

MATERIALS AND METHODS

Participants

The local Institutional Review Board approved this retrospective study. Between 2011 and 2014, TOAs were performed on 64 hips in 60 patients. Three hips of three patients were excluded owing to the lack of follow-up data or the presence of any condition due to physical or mental disorders unrelated to the hip joint such as cerebral palsy or severe depression. Finally, the 61 hips of 57 patients (11 male and 50 female patients; mean age, 37.6 years) were included in the study (Table I). According to the Kellgren–Lawrence (KL) classification [21] of preoperative radiographs, 27 hips had Grade I (KL-I), 28 had Grade II (KL-II) and 6 had Grade III (KL-III) hip OA. Twenty-seven hips of 25 patients were subjected to C-TOA, while 34 hips of 32 patients were subjected to M-TOA. Patient demographics showed no significant differences between groups (Table I). The mean follow-up period spanned 6.8 years (range, 1.0–11.4). Because 2011–14 served as a period of refinement for the surgical technique, C-TOA and M-TOA were chosen according to the surgeon’s discretion.

Table I.

Patients’ demographics

	Total	C-TOA	M-TOA	P-value
Hips/patients	61/57	27/25	34/32
Sex (female/male)	50/11	22/5	28/6	1.00
Age at operation	37.6 ± 11.7	36.1 ± 11.7	38.7 ± 11.8	0.39
Follow-up duration (years)	6.8 ± 3.2	6.9 ± 3.5	6.8 ± 3.0	0.58
BMI (kg/m2)	22.9 ± 3.7	22.5 ± 3.2	23.5 ± 4.3	0.45
Preoperative KL Grade (I/II/III)	27/28/6	10/14/3	17/14/3	0.60
Postoperative KL Grade (I/II/III/IV)	22/29/8/2	6/17/4/0	16/12/4/2	0.08
Preoperative Merle d’Aubigné–Postel score	13.0 ± 1.1	12.9 ± 1.5	13.0 ± 0.9	1.00
Merle d’Aubigné–Postel score at postoperative 1 year	15.6 ± 3.3	16.1 ± 1.5	15.1± 4.1	0.67
Merle d’Aubigné–Postel score at the final follow-up	15.4 ± 3.6	16.1 ± 1.8	14.9 ± 4.4	0.66
Conversion to THA	2	0	2	0.50
Preoperative lateral CE angle (°)	9.2 ± 8.2	8.6 ± 9.3	9.8 ± 7.2	0.56
Postoperative lateral CE angle (°)	39.1 ± 9.5	39.8 ± 9.5	38.5 ± 9.6	0.62
Preoperative ARO (°)	21.9 ± 7.6	22.8 ± 7.8	21.1 ± 7.5	0.41
Postoperative ARO (°)	0.3 ± 8.3	−0.6 ± 7.6	1.0 ± 8.8	0.45
Metal wire and screw removal	61	18	4	<0.001

Values are expressed as mean ± SD. BMI, body mass index.

Surgical technique

Figure 1 shows a spoon-shaped Wagner-type chisel. The chisel was double-curved with a dent in the center and had a 40-mm radius of curvature. The 40-mm radius matched that of the curved chisel used in C-TOA.

TOA was performed as previously described [10]. Briefly, surgery was performed in the lateral decubitus position with an Ollier skin incision (Fig. 2a). The short external rotator muscles were detached to expose the posterior joint capsule and iliac and ischial bones. The gluteus medius and tensor fasciae latae muscles were separated in the anterior section. Instead of trochanteric osteotomy, exposure of the iliac bone proceeded with hip abduction and internal rotation to loosen the abductor muscles in M-TOA (Fig. 2b). Attention was paid to avoid abductor–iliac crest detachment. After the circumferential exposure of the articular capsule, spherical osteotomy was performed using an ordinary curved chisel, starting 20 mm proximal to the superior acetabular edge, which was considered the first osteotomy portion. Posterior osteotomy commenced from the innominate sulcus of the ischium, passing through the mid-point between the greater sacral fossa and acetabulum’s posterior edge using a conventional curved chisel. An additional posterior osteotomy was performed to the proximal apex of the osteotomy line. During anterior/posterior osteotomy in M-TOA, hip abduction and external/internal rotation to loosen the abductor muscle were important to maintain exposing the osteotomy area using a retractor for the abductor muscle, respectively (Fig. 2b and c), while hip abduction and external/internal rotation were not necessary during anterior/posterior osteotomy in C-TOA (Fig. 2d). A spoon-shaped chisel was subsequently placed in the iliac bone’s anterior section, and osteotomy was advanced toward the first osteotomy portion along the osteotomy line (Fig. 2c). Pubic osteotomy was performed at the lateral portion of the iliopubic tubercle. After pubic osteotomy, the acetabular fragment was adducted to cover the femoral head and fixed using two or three screws. In C-TOA, the iliac bone was widely exposed with trochanteric osteotomy (Fig. 2d), and metal wires were used for greater trochanter reattachment. Wheelchair transfer was allowed on Postoperative Day 1. The range of hip motion training was started 2 days postoperatively, and partial-weight bearing was started 2 weeks postoperatively. Full-weight bearing was allowed at 8–10 weeks postoperatively. All patients underwent the same walking training program, and the postoperative rehabilitation program did not differ between both groups.

Fig. 2.

The surgical procedure of M-TOA (a–c) and C-TOA (d). (a) Surgery is performed in the lateral decubitus position with an Ollier skin incision. (b) Osteotomy commences from the posterior section using an ordinary curved chisel (R = 40 mm) and advances to the superior portion. (c) A spoon-shaped chisel is placed in the anterior section of the iliac bone, and osteotomy is advanced along the marked osteotomy line. (d) Broad exposure of the articular capsule and osteotomy area is performed with trochanteric osteotomy (C-TOA).

Clinical and radiographic assessments

The Merle d’Aubigné–Postel score was used for hip joint clinical assessment [22]. The lateral center-edge angle of Wiberg (CE angle) [23] and acetabular roof obliquity (ARO) [24] were measured as radiographic indices of acetabular dysplasia pre- and postoperatively.

Muscle strength was measured using a myodynamometer (Myutas F-1; Anima Corp., Tokyo, Japan). Abduction was measured in the supine position at 0° hip flexion. Flexion and knee extension were measured in the sitting position and at 60° hip flexion. After each trial, practical measurements were performed twice, and the maximum values were used. Measurements were conducted immediately preoperatively and at 3, 5, 10, 24 and 52 weeks postoperatively. Values were expressed as a ratio calculated by dividing by the muscle strength of the contralateral supporting side (operated side/contralateral side) to correct for individual differences. Regardless of the surgical technique, muscle strength measurements were routinely performed before and after the procedure up to 1 year postoperatively.

Statistical analyses

Values are expressed as mean ± standard deviation (SD). All statistical analyses were performed using JMP software version 11.0 (SAS Institute, Cary, NC, USA). Chi-squared or Fisher’s exact test was used to compare categorical data between groups. All parameters were tested for normality using the Shapiro–Wilk test for continuous data. Normally distributed variables were first assessed using Levene’s test to determine whether probability values (P) should be reported based on the equality of variances with Student’s t-test or non-equality with Welch’s t-test. Non-normally distributed variables were evaluated using the independent Wilcoxon signed-rank test. These tests were used to compare demographic data, radiographic parameters and muscle strength between the C-TOA and M-TOA groups. Paired t-tests were used to compare pre- and postoperative demographic data and radiographic parameters. A P-value of <0.05 was considered statistically significant. An initial power analysis showed that a sample size of 25 participants per group would provide 80.0% statistical power for detecting a 0.2 difference in the muscle strength ratio. Thus, a P-value of <0.05 and an SD of 0.3 were assumed.

RESULTS

The mean Merle d’Aubigné–Postel score significantly improved from 13.0 ± 1.1 to 15.6 ± 3.3 points (P < 0.001) at postoperative 1 year, without any difference between the C-TOA and M-TOA groups (16.1 ± 1.5 versus 15.1 ± 4.1 points, respectively, P = 0.67) (Table I). The mean Merle d’Aubigné–Postel score was retained at 15.4 ± 3.6 points at a mean 6.8-year follow-up, without any difference between the C-TOA and M-TOA groups (16.1 ± 1.8 versus 14.9 ± 4.4 points, respectively, P = 0.66) (Table I). The intended corrections were achieved in both groups. The mean postoperative CE angle and ARO were 39.1 ± 9.5° and 0.3 ± 8.3°, respectively, and there was no significant difference between groups (postoperative CE angle, P = 0.62 and postoperative ARO, P = 0.45) (Table I). The postoperative KL grade was I in 22 hips, II in 29 hips, III in 8 hips and IV in 2 hips. One hip with KL Grade III and one hip with KL Grade IV were converted to total hip arthroplasty (THA). No significant differences were found in the postoperative KL grade and the THA conversion rate between groups. No severe complications such as infection, venous thromboembolism or non-union were noted in either group. Discomfort or tenderness at the greater trochanter was observed in 18 hips (66.7%) in the C-TOA group, but only in 4 hips (11.2%) in the M-TOA group. Due to trochanteric discomfort, metal wire and screw removal were performed in 18 hips in the C-TOA group, but screw removal was performed only in 4 hips in the M-TOA group.

There were no significant differences in preoperative muscle strength between the operated and contralateral hips in both groups. Generally, the muscle strength of operated hips decreased to less than half at 3 weeks postoperatively compared to that preoperatively, subsequently recovering slowly with time (Figs 3–5). At 24 weeks postoperatively, hip flexion and knee extension muscle strength in the operated hips were equal to those in the contralateral hips (Figs 4 and 5). When comparing the muscle strength ratios between groups (Table II), there were no significant differences in hip abduction preoperatively and at 3 weeks postoperatively. At 5 and 10 weeks postoperatively, the hip abduction muscle strength ratios in the M-TOA group were significantly higher than those in the C-TOA group (0.62 ± 0.2 versus 0.39 ± 0.2, P < 0.001 and 0.76 ± 0.2 versus 0.59 ± 0.2, P = 0.04, respectively) (Fig. 3). At 24 weeks postoperatively, the M-TOA group showed better recovery in the hip abduction muscle strength ratio than the C-TOA group (0.94 ± 0.2 versus 0.70 ± 0.2, P = 0.01). There was no significant difference in the hip abduction muscle strength ratio between the M-TOA and C-TOA groups at 52 weeks postoperatively (0.92 ± 0.1 versus 0.86 ± 0.2, P = 0.36). The M-TOA group showed a significantly higher knee extension muscle strength ratio than the C-TOA group at 3 and 5 weeks postoperatively (Fig. 5). There were no significant differences at 10 weeks postoperatively.

Fig. 3.

The time course of the muscle strength of hip abduction after TOA. Values (mean ± SD) are expressed as the ratio divided by the muscle strength of the contralateral supporting side (affected side/contralateral side).

Fig. 4.

The time course of the muscle strength of hip flexion after TOA. The values (mean ± SD) are expressed as the ratio divided by the muscle strength of the contralateral supporting side (affected side/contralateral side).

Fig. 5.

The time course of the muscle strength of knee extension after TOA. The values (mean ± SD) are expressed as the ratio divided by the muscle strength of the contralateral supporting side (affected side/contralateral side).

Table II.

The changes in the muscle strength ratio between operative and non-operative limbs

		Preoperative	3 weeks	5 weeks	10 weeks	24 weeks	52 weeks
Hip abduction	C-TOA	0.88 ± 0.3	0.43 ± 0.0	0.39 ± 0.2*	0.59 ± 0.2*	0.70 ± 0.2*	0.86 ± 0.2
	M-TOA	0.89 ± 0.3	0.48 ± 0.2	0.62 ± 0.2	0.76 ± 0.2	0.94 ± 0.2	0.92 ± 0.1
Hip flexion	C-TOA	0.90 ± 0.3	0.44 ± 0.2	0.55 ± 0.2	0.72 ± 0.2	0.78 ± 0.3	1.00 ± 0.1*
	M-TOA	0.90 ± 0.2	0.61 ± 0.3	0.67 ± 0.1	0.78 ± 0.2	0.90 ± 0.2	0.85 ± 0.1
Knee extension	C-TOA	0.90 ± 0.2	0.46 ± 0.2*	0.57 ± 0.2*	0.78 ± 0.2	0.81 ± 0.2	1.01 ± 0.2
	M-TOA	0.90 ± 0.2	0.69 ± 0.2	0.78 ± 0.2	0.83 ± 0.3	0.93 ± 0.1	0.95 ± 0.2

Values are expressed as mean ± SD.

^*Significantly different between C-TOA and M-TOA (P < 0.05).

When comparing preoperative KL grades in each group, hips with KL-I consistently had better abductor muscle strength recovery than hips with KL-II or KL-III (Fig. 6a) in the M-TOA group. A significant difference was found in the hip abduction muscle strength ratio between hips with KL-I and those with KL-II or KL-III at 10 weeks post-M-TOA (0.86 ± 0.2 versus 0.66 ± 0.2, P = 0.02). However, this trend was not significant in the C-TOA group (Fig. 6b).

Fig. 6.

The time course of the muscle strength of hip abduction after C-TOA (a) and M-TOA (b). The values (mean ± SD) are expressed as the ratio divided by the muscle strength of the contralateral supporting side (affected side/contralateral side).

DISCUSSION

To minimize surgical invasiveness on the abductor muscles, we modified TOA to enable PAO without cutting the greater trochanter and inducing abductor–iliac crest detachment. M-TOA helped achieve earlier recovery of muscle strength in hip abduction at 5, 10 and 24 weeks postoperatively compared to C-TOA without significant correction loss.

Regardless of the surgical approach, a literature review showed that a decrease in muscle strength continued for several months post-PAO [25–27]. Sucato et al. compared the differences in muscle strength around the hip joint preoperatively, 6 months postoperatively and 1 year after Ganz osteotomy. They demonstrated that muscle weakness could be observed 6 months postoperatively and that muscle strength recovered to almost the preoperative level only 1 year postoperatively [25]. Jacobsen et al. also reported that walking and running characteristics improved 1 year following PAO in patients with symptomatic acetabular dysplasia [26]. These results are comparable to those of this study where the muscle strength of the operated hip recovered to the original level without a significant difference from that of the contralateral hip between 24 and 52 weeks postoperatively. Hip flexion and knee extension strength also decreased. The muscle strength recovered to the same level as that preoperatively in several weeks, and this may be due to postoperative pain and the effects of retraction of the rectus femoris muscle during pubic osteotomy, even though it was not detached. Accordingly, muscle strength recovery may exhaust 24–52 weeks following PAO.

RAO and TOA are widely performed in Japan, and their long-term clinical outcomes are favorable [8–11]. However, time is required to improve claudication due to abductor weakness postoperatively. An osteotomy method that did not require greater trochanter cutting or did not lead to abductor–iliac crest detachment was successfully developed to decrease abductor dysfunction. The posterior section was osteotomized using an ordinary curved chisel, whereas the anterior iliac bone was osteotomized using a spoon-shaped chisel. Osteotomy of the anterior iliac bone was connected to that of the posterior section. Compared with the C-TOA group, the M-TOA group achieved significant recovery in the abductors early postoperatively. As there were no significant differences in hip abduction and knee extension muscle strengths between both groups at 52 weeks postoperatively, the success of M-TOA was defined as the speed of muscle strength recovery within 6 months postoperatively. Furthermore, the absence of stimulus pain due to the metal wires used to reattach the greater trochanter was another advantage of M-TOA.

Ezoe et al. previously showed that the muscle strength of patients with Grade 0 disease based on the Tönnis classification at 12 months post-PAO was greater than that of patients with Tönnis Grades 1 and 2 [27]. They concluded that the preoperative OA stage was essential for postoperative muscle strength recovery. Accordingly, we also observed faster muscle recovery of the hips with KL-I than that of the hips with KL-II or KL-III in the M-TOA group, suggesting the role of OA progression in the clinical outcomes and the importance of patient selection for surgery. On the other hand, no significant difference was found in muscle strength recovery between hips with KL-I and hips with KL-II or KL-III in the C-TOA group, suggesting that the effects of cutting the greater trochanter on the muscle strength recovery were much greater than those attributed to the preoperative KL grade.

This study has some limitations. First, we examined only short-term results with M-TOA; thus, further follow-up and evaluation are needed to validate our findings. However, the clinical outcomes of TOA have previously been reported to be favorable for >10 years [10, 11]. Thus, similar results would be expected with M-TOA. Second, although the backgrounds of the C-TOA and M-TOA groups were equivalent, the participants in this study were not randomized. Therefore, the exact effect of the M-TOA procedure must be further determined in a randomized controlled study. Third, we did not compare different PAOs using the anterior approach. It would be interesting to clarify the extent to which the detachment of abductor muscles by the lateral approach affects postoperative muscle recovery. Finally, as preoperative MRI was not obtained in the present series, the presence of any pre-existing pathology of the abductor complex, such as bursitis in the greater trochanter, was unclear. However, no significant differences were found in preoperative muscle strength between the C-TOA and M-TOA groups, suggesting that the comparison of muscle strength recovery between groups was valid.

CONCLUSION

To decrease the invasiveness of the hip abductor muscles intraoperatively, an osteotomy method that did not require greater trochanter cutting and did not lead to abductor–iliac crest detachment was developed. Compared with the conventional transtrochanteric approach, our new approach induced significantly earlier recovery of abductor muscles at the early periods postoperatively without significant correction loss.

DATA AVAILABILITY

The present article’s data will be shared upon reasonable request to the corresponding author.

FUNDING

Japan Society for the Promotion of Science (19K09652).

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests.