Observations on the Development of the Acetabulum Following Chiari Osteotomy

Abstract

A detailed analysis of acetabular development following Chiari osteotomy in ten patients with neuromuscular disease revealed an anti-Chiari effect in only one patient.

SUMMARY

A detailed analysis was conducted of acetabular development in ten consecutive patients with neuromuscular disease (nine with spastic cerebral palsy, one with Charcot-Marie-Tooth disease) after Chiari osteotomy. Indications for surgery included painful hip subluxation or dislocation, and disturbance of sitting balance or ambulation. Patients were five to nineteen years old at the time of operation (average 11 years), and were followed for six to eleven years post-operatively (average 10 years). Patients were interviewed with their families, were physically examined, and radiographs were obtained. All results were submitted to independent statistical analysis. This study showed that for nine of the ten patients, radiographic improvements were obtained and maintained for femoral head coverage, the CE angle, the Sharp angle, overhang of the lateral acetabular margin, and acetabular depth. We observed postoperative worsening of these parameters (anti-Chiari effect) in only one patient.

Chiari osteotomy can be confidently applied to these neuromuscular patients.

INTRODUCTION

Neuromuscular diseases, such as cerebral palsy and Charcot-Marie-Tooth disease, can present with progressive hip subluxation.^8,18,21,36 Chiari osteotomy has been applied to hip joint subluxation and dislocation in patients with spastic neuromuscular disorders.^{1,2,3,5,6,7,9,10,11,12,13,22,25,26,28,29,31,32,33, 35}

Niethard and colleagues, with their careful observations, noted the ability of the Chiari osteotomy to improve femoral head coverage.^22,23 Pre-operatively, their patients had an average CE angle of 0°. In their neuromuscular patients, the final CE angle measured between 20° and 25° (average follow-up period 5.5 years, range 2-10 years). However, they observed a disturbance of the development and growth of the lateral edge of the acetabulum due to the Chiari osteotomy, and on this basis they decided that the observed improvements in femoral head coverage were dependent mainly on the extent of the medialization of the femoral head and distal pelvic fragment obtained at the time of the operation.

Thompson and colleagues³⁵ used the Chiari osteotomy in the management of 33 abnormal hips in 30 patients (including 19 with spastic cerebral palsy). The average age at the time of surgery was 12 years (range 2-25 years). Patients were followed for an average of four years (ranging from 3 months to 7 years). The average pre-operative femoral head coverage was 30%, CE angle -19°, and the Sharp angle 55°. At the last postoperative evaluation, average femoral head coverage was 100%, CE angle 36° and Sharp angle 41°.

Dietz and colleagues⁶ found that at the last postoperative evaluation for 24 spastic hips after Chiari osteotomy the femoral head coverage had increased from 26% pre-operative to 81%, the CE angle improved from -19° to 29°, and the Sharp angle improved from 50° to 39°. They noted that any loss of improvement of femoral head coverage and of the CE angle occurred mainly during the first post-operative year.

We report the long-term radiographic results after Chiari osteotomy in ten successive neuromuscular patients (nine with spastic cerebral palsy, one with Charcot-Marie-Tooth disease). In all cases we placed particular attention to the starting point of the osteotomy at the exact edge of the acetabular cartilage, and on the angle of the osteotomy (10°-15° cephalad). Our goal in this study was to analyze the post-operative course of a small number of neuromuscular patients. We compared the improvements in our patients with those in other series, and determined whether the initial improvements after surgery were gradually lost over the course of time. In the case of any loss of correction, we noted its temporal relationship to the surgery and whether the loss of correction stabilized or evolved throughout the post-operative observation period.

MATERIALS AND METHODS

Ten successive patients with neuromuscular disease (nine with spastic cerebral palsy, one with Charcot-Marie-Tooth disease) were managed in our institution from early childhood onwards (Table 1).

TABLE 1. Patients.

Patient	Diagnosis	Age at Chiari Osteotomy (Yrs + Mos.)	Length of Follow-Up (Yrs + Mos.)	Age at Last Follow-Up (Yrs + Mos.)
1	spastic quadriparesis	13 + 3	11 + 9	25 + 0
2	spastic quadriparesis	11 + 2	11 + 10	23 + 0
3	spastic quadriplegia	10 + 5	11 + 10	21 + 3
4	spasticity due to prematurity	7 + 2	10 + 8	17 +10
5	severe spastic quadriplegia	5 + 10	6 + 4	12 +2
6	spastic quadriplegia	12 + 10	10 + 2	23 + 0
7	Charcot-Marie-Tooth disease	13 + 3	9 +10	23 + 1
8	spastic quadriplegia	10 + 11	10 + 9	20 + 9
9	spastic quadriplegia	14 + 6	10 + 11	25 + 5
10	athetosis and spastic quadriplegia	19 + 6	10 + 6	30 +0
	Average:	11 + 6	10 + 8	22 + 2

Patients were 5-19 years old at the time of surgery (average 11 years old). The length of the follow-up period was between 6 and 11 years (average 10 years), and at the time of last evaluation the patients ranged in age from 12-30 years old (average 22 years old). The authors performed each patient's surgery and managed both their pre- and postoperative course.

We studied the following aspects of acetabular development:

1. Percentage of femoral head coverage by the acetabulum was measured by the method of Heyman and Herndon (acetabular-femoral head quotient)^{16, 24, 30}). The horizontal distance from the medial edge of the femoral head to a line perpendicular to the lateral edge of the acetabulum is divided by the width of the femoral head.

2. The center edge angle (CE angle) of Wiberg is the angle subtended by a vertical line through the center of the femoral head, and a line connecting the center of the femoral head to the lateral-most edge of the acetabulum³⁷).

3. Both the Sharp angle³⁴ and the acetabular index were measured. Because the triradiate cartilage ossified during the follow-up period for all patients (except patient 5) it was only possible to follow the Sharp angle throughout the entire post-operative period.

4. AP radiographs of the pelvis were measured for the overhang of the lateral edge of the acetabulum as an indicator of acetabular development. The overhang is the perpendicular distance between a horizontal line through the dome of the acetabulum and a horizontal line through the lateral most point of the acetabular margin²⁴. The absence of such an overhang implies femoral head subluxation and acetabular dysplasia. Increasing overhang implies progressive development of the depth of the acetabulum and increasing stability.

5. A direct measure of the depth of the acetabulum is the horizontal distance from the medial edge of the teardrop to Perkins' line²⁴.

Analysis of variations for repeated measurements was used, to statistically evaluate the importance of the changes in the ten patients. This was done in a series of planned comparisons of hip joint measures against time post-operatively. A p value of less than 0.05 is significant.

OPERATIVE TECHNIQUE

A smooth guide pin is introduced under fluoroscopic control into the exact cartilage-bone interface at the lateral-most, cephalad edge of the acetabulum. This entry point is critical and must be exact.

When the entry point is precisely determined, a plain film AP radiograph of the pelvis is taken, to ensure the mildly cephalad angle of the guide pin (10-15° ). The fluoroscope cannot provide the necessary image for this angle, which is as critical as the entry point.

After the entry point and the 10-15° cephalad angle of the pin are precisely determined, a 7mm straight Lambotte osteotome is placed on either the cephalad or caudal side of the pin (whichever is closest to the lateral edge of the acetabular cartilage, without entering it). This osteotome is left in position as a guide, and a 4mm straight Lambotte osteotome is used for the anterior bow of the osteotomy, which ends just distal to the anterior inferior iliac spine.

The osteotomy proceeds posteriorly and distally from the guide osteotome, following the posterior edge of the acetabulum, always osteotomizing parallel to the guide osteotome. The posterior-distal, then proximal, bow of the cut protects against posterior displacement of the completed osteotomy. The osteotomy ends just distal to the apex of the sciatic notch. After medial displacement of the femoral head and distal fragment of the pelvis, two threaded pins are drilled medially and caudally from the proximal pelvic fragment into the distal fragment. A 1.5 hip spica cast is applied for six weeks. The two threaded pins are removed under general anesthesia at eight weeks following surgery.

RESULTS

1. Percentage of femoral head coverage (Table 2): Immediately before osteotomy, the percentage of femoral head coverage from the acetabulum ranged from frank dislocation to 68% coverage (average 47% coverage). Immediately after the osteotomy, coverage ranged from 66% to 100% (average 94% coverage). At the time of last follow-up coverage ranged from 70%- 100% (average 94% coverage) (p = 0.001). One patient (patient 6) showed a decrease in coverage from 100% immediately after surgery, to 70% at last followup. Two other patients (patients 8 and 9) showed slight decreases in their coverage (-5% and -10%)during the post-operative follow-up period.

2. CE angle of Wiberg (Table 3): Pre-operatively, the CE angle ranged from -110° (frankly dislocated) to only 16° (average -13°). Immediately after the operation the CE angle ranged from 7° to 65° (average 44°), for an increase that ranged from +31° to +117° (average +58° increase). At the time of last follow-up, the CE angle ranged from 25° to 65° (average 47°) (p =0.001). During the follow-up period, two patients (patients 6 and 9) showed decreases in the CE angle (-18° and -13°).

3. Sharp angle (Table 4): In nine hips there was an immediate decrease in the Sharp angle (improvements between -6° and -20°, average -11°). In one patient (patient 5) the Sharp angle remained unchanged immediately after the operation, but during the first four post-operative years this decreased by 18°, then remained stable over the rest of the postoperative period (another 2.5 years). At the time of last follow-up the average Sharp angle (38°) remained significantly reduced (p = 0.01).

4. Overhang of the lateral edge of the acetabulum (Table 5): Before the osteotomy, two hips had horizontal lateral acetabular margins (patients 3 and 10). In the other eight patients, the lateral edge of the acetabulum sloped cephalad. Immediately after the operation, eight hips had acquired lateral acetabular edge overhang (ranging from 2 to 5 mm). This immediate improvement is due to the 10° to 15° cephalad inclined angle of the osteotomy, which begins at the exact lateral point of the acetabular cartilage. The two patients without immediate lateral acetabular margin overhang (patients 2 and 5) had the least cephalad-inclined osteotomy angles (5° and 8°). However, at the time of last follow-up, both had developed some lateral overhang. By the last follow-up only one patient (patient 6) had a cephalad sloping lateral acetabular edge. The average overhang remained significantly improved (p = 0.01).

5. Distance from the medial teardrop to Perkins' line (Table 6): As expected, the distance between the teardrop and Perkins' line increased immediately after the osteotomy. (Range of improvement of this distance is between +5mm and +28mm, average +18mm). During the post-operative period one patient (Patient 1) showed a decrease in the depth of the acetabulum (-9mm), which occurred during the first post-operative year. However, her improvement in comparison to pre-operative (+19 mm) remained intact. Two patients (Patients 6 and 9) remained stable during the post-operative period, and maintained the increased depth of the acetabulum due to the osteotomy. The other seven patients (Patients 2, 3, 4, 5, 7, 8, and 10) showed a progressive increase in the distance between the teardrop and Perkins' line (depth of the acetabulum) throughout the post-operative period. At the time of last follow-up the average increase in the depth of the acetabulum remained very significant (p = 0.0001).

TABLE 2.

Femoral Head Coverage: Extent of Coverage

Patient	Immediately Pre-Operative (%)	Percent Displacement of Osteotomy	Immediately Post-Operative		At Last Follow-Up
			Percent Coverage	Change in Percent Coverage	Percent Coverage	Change in Percent Coverage Compared to Immediately Post-Operative	Difference in Percent Coverage Compared to Pre-Operative
1	65%	75%	100%	+35	100%	0	+35
2	57%	88%	100%	+43	100%	0	+43
3	68%	100%	100%	+32	100%	0	+32
4	frank dislocation	60%	100%	+100	100%	0	+100
5	frank dislocation	69%	66%	+66	91%	+25	+91
6	65%	62%	100%	+35	70%	-30	+5
7	45%	87%	89%	+44	100%	+11	+55
8	53%	37%	100%	+47	95%	-5	+42
9	51%	81%	90%	+39	80%	-10	+29
10	64%	46%	98%	+34	100%	+2	+36
Average	47%	70%	94%	+48	94%	+47	-7

TABLE 3.

CE Angle of Wiberg

Patient	Immediately Pre-Operative (degrees)	Immediately Post-Operative		At Last Follow-Up
		Degrees	Change in Degrees	Degrees	Change in Degrees Compared to Immediately Post-Operative	Difference in Degrees Compared to Pre-Operative
1	16°	58°	+42°	53°	-5°	+37°
2	0°	48°	+48°	50°	+2°	+50°
3	8°	65°	+57°	65°	0°	+57°
4	-53°	40°	+93°	42°	+2°	+95°
5	-110°	7°	+117°	49°	+42°	+159°
6	10°	43°	+33°	25°	-18°	+15°
7	-7°	45°	+52°	50°	+5°	+57°
8	-1°	53°	+54°	48°	-5°	+49°
9	-7°	43°	+50°	30°	-13°	+37°
10	10°	41°	+31°	59°	+18°	+49°
Average	-13°	44°	+58°	47°	+3°	+61°

TABLE 4.

Sharp Angle

Patient	Immediately Pre-Operative (degrees)	Immediately Post-Operative		At Last Follow-Up
		Degrees	Change in Degrees	Degrees	Change in Degrees Compared to Immediately Post-Operative	Difference in Degrees Compared to Pre-Operative
1	45°	27°	-18°	38°	+11°	-7°
2	58°	40°	-18°	33°	-7°	-25°
3	51°	31°	-20°	25°	-6°	-26°
4	55°	49°	-6°	47°	-2°	-8°
5	51°	52°	+1°	34°	-18°	-17°
6	58°	49°	-9°	50°	+1°	-8°
7	50°	40°	-10°	38°	-2°	-12°
8	46°	34°	-12°	30°	-4°	-16°
9	43°	33°	-10°	38°	+5°	-5°
10	55°	44°	-11°	43°	-1°	-12°
Average	51°	40°	-11°	38°	-2°	-13°

TABLE 5. Overhang of the Lateral Edge of the Acetabulum.

Patient	Immediately Pre-Operative	Cephalad Angle of the Osteotomy (degrees)	Immediately Post-Operative (mm)	At Last Follow-Up (mm)
1	C	10°	5	4
2	C	5°	C	5
3	H	11°	5	22
4	C	9°	2	4
5	C	8°	C	3
6	C	16°	4	C
7	C	13°	4	6
8	C	10°	4	4
9	C	10°	3	3
10	H	20°	3	7

C= Lateral acetabular margin slopes cephalad

H= Lateral acetabular margin is horizontal

TABLE 6.

Horizontal Distance from Medial Edge of Teardrop to Perkins' Line

Patient	Immediately Pre-Operative (mm)	Immediately Post-Operative		At Last Follow-Up
		mm	Change in mm	mm	Change in mm Compared to Immediately Post-Operative	Difference in mm Compared to Pre-Operative
1	49	77	+28	68	-9	+19
2	37	61	+24	66	+5	+29
3	46	63	+17	72	+9	+26
4	33	38	+5	54	+16	+21
5	34	47	+13	60	+13	+26
6	41	55	+14	54	-1	+13
7	52	71	+19	82	+11	+30
8	43	58	+15	65	+7	+22
9	50	77	+27	79	+2	+29
10	52	67	+15	77	+10	+25
Average	44	61	+18	68	+6	+24

DISCUSSION

For the entire group of ten patients the increase in femoral head coverage immediately after Chiari osteotomy was of great significance (Table 2) (p=0.001) Except for Patient 6, the stability of this improvement throughout the follow-up period was reassuring.

The CE angle improved most markedly (the value at the time of last follow-up in comparison with pre-operative) for the two hips that were most dysplastic pre-operatively (Patients 4 and 5) (Table 3). Even for these two hips the CE angle was brought into a normal range, and this was maintained throughout the follow-up period. Because of the pre-existing neuromuscular disease and muscular imbalance, one could expect that for these two hips there would be a persistent tendency for loss of correction of femoral head coverage, and a decrease in the CE angle, but this did not occur. These two hips were in the two youngest patients (ages 7 and 5 at the time of surgery). If the Chiari osteotomy had caused damage to the lateral acetabular growth plate and to the secondary centers for ossification, one would expect post-operative worsening. (Patients 4 and 5 were followed for 10 and 6 years after surgery, to the ages of 17 and 12 years.)

At the time of last follow-up all ten patients had improved Sharp angles in comparison to pre-operative measurements (Table 4).

An advantage of the Chiari osteotomy is that it can create an immediate acetabular margin overhang. This requires, however, a 10° to 15° cephalad-inclined angle of the osteotomy. In two patients (Patients 2 and 5) the angle of the osteotomy was too horizontal (Table 5) and did not result in immediate lateral acetabular margin overhang; this emphasizes the importance of the 10° to 15° cephalad angle of the osteotomy.¹⁷ However, at the time of last follow-up, these two patients had 100% and 91% femoral head coverage, respectively (Table 2). The improvement in the overhang in six hips (Patients 2, 3, 4, 5, 7, and 10) during the post-operative period indicates the existence of an active, undamaged lateral acetabular growth plate and of secondary ossification centers.

Especially effective was the immediate and significant improvement in the depth of the acetabulum after the osteotomy. In only one patient (Patient 4) was this improvement less than 13 mm (Table 6). However, the acetabular depth in this patient improved markedly during the post-operative period. Seven patients showed progressive deepening of the acetabulum during the post-operative period. This indicates again that after the osteotomy there is a tendency towards further acetabular development, with increasing distance between the teardrop and Perkins' line.

Niethard and colleagues^22,23 cited Otte's²⁶ and Purath's²⁹ observations on the disturbed development of the lateral acetabulum after Chiari osteotomy. All three authors observed that the younger the patient at the time of surgery, the greater the extent of this anti-Chiari effect.

Fifteen of the 87 children in the study of Niethard and colleagues had spastic cerebral palsy. At the time of surgery, the 87 patients were between 4 and 15 years old (average 10 years), and their follow-up period was between 2 and 10 years (average 5.5 years). As in our study, they observed a great improvement in femoral head coverage due to the osteotomy. They also observed an initial improvement in the CE angle (0° preoperative, in comparison with 34° immediately post-operative). However, at the time of last follow-up, the CE angle for all 87 patients averaged 28°, with the understandable observation that there must be impaired growth of the lateral edge of the acetabulum during the follow-up period. This was noted especially in the 15 neuromuscular hips, in which the CE angle was between 20° and 25° at last follow-up. They did not discuss the lateral acetabular overhang. They found that a younger age at the time of surgery correlated with a less significant long-term improvement, and an apparent regression with further skeletal growth. Those children fared best who were nine years old or older at the time of surgery.

Bohler, on the other hand, in his extensive clinical and experimental studies, found that Chiari osteotomy did not disturb lateral acetabular growth or blood supply.^3,4

When we placed our ten patients in age categories (Table 7 we found that at the time of last follow-up the children who were less than 8 years old at the time of surgery improved percent femoral head coverage, CE angle, and teardrop-to-Perkins' line distance throughout the post-operative period. For the 9 to 12 years age group the percent femoral head coverage and CE angle decreased slightly, and for the 13 to 19 years age group these were stable. The Sharp angle improved postoperatively for the less than 8 years and the 9 to 12 years age groups, but worsened slightly for the 13 to 19 years age group. Lateral acetabular overhang improved for the less than 8 years and the 9 to 12 years age groups during the post-operative period, and was stable for the 13 to 19 years age group. The teardrop-to-Perkins' line distance improved for all three age groups, especially for the less than 8 years age group. We observed this disturbance of lateral acetabular development noted by other authors in only one patient (Patient 6). This is not to counter the excellent studies of other authors, but simply to report that based on our data, our observations are in agreement with Bohler and colleagues.⁴ We agree with Bohler, that for nine of our patients the Chiari osteotomy disturbed neither lateral acetabular growth nor blood supply. We agree with both Niethard and Bohler that the technical details of the osteotomy are decisive for obtaining a good result.

TABLE 7. Acetabular Development Correlated with Age of Patients at Time of Chiari Osteotomy.

		Femoral Head Coverage (%)			CE Angle (Degrees)			Sharp Angle (Degrees)			Lateral Acetabular Margin Overhang (mm)			Distance from Teardrop to Perkins' Line (mm)
Age at Osteotomy (Years)	Number of Patients	Immediately Pre-Op	Immediately Post-Op	At Last Follow-Up	Immediately Pre-Op	Immediately Post-Op	At Last Follow-Up	Immediately Pre-Op	Immediately Post-Op	At Last Follow-Up	Immediately Pre-Op	Immediately Post-Op	At Last Follow-Up	Immediately Pre-Op	Immediately Post-Op	At Last Follow-Up
< 8	2	0	83	95	-82	24	46	53	50	40	0	1	4	34	43	57
9-12	4	61	100	91	8	52	48	53	39	35	0	3	8	42	59	64
13-19	4	56	94	95	3	47	48	48	36	39	0	4	5	51	73	77

The progressive increase in acetabular depth (teardrop-to-Perkins' line distance) that was observed during the follow-up period in our patients is perhaps due to the undisturbed triradiate cartilage. Lindstrom and colleagues¹⁹ and Ponseti²⁷ observed that the triradiate cartilage is normal in hip dysplasia. Ponseti also noted that the iliac portion of the acetabular growth plate is inclined more cephalad in patients with hip dysplasia. If the Chiari osteotomy begins at exactly the most cephalad edge of the acetabular cartilage, and is inclined 10° to 15° cephalad, then the acetabular growth plate and lateral acetabular ossification centers are left undamaged. With the lateral edge of the acetabulum relieved from the pressure of the subluxated femoral head, and with proper placement and direction of the osteotomy, the lateral acetabular growth plate and secondary centers of ossification are able to develop normally. Additionally, the increased acetabular depth improves the orientation of the acetabular growth plate.¹⁹ Harris^14,15 also noted that with the reduction of a developmental hip dislocation begins a progressive improvement and remodeling of the acetabular growth plate. Although certainly the Chiari osteotomy in our neuromuscular patients did not create a concentric reduction, the improved relationship of the femoral head to the acetabulum led to further improvements in hip growth and development. With respect to the younger patients, there was a longer and more effective period of acetabular remodeling. Indeed, as noted by Massie and Howorth,²⁰ the ages of 5 to 8 years are especially decisive for the development of the acetabulum. Our two youngest patients were by far the most dysplastic before surgery. Both showed great improvements in percent of femoral head coverage, the CE angle, the Sharp angle, the lateral acetabular overhang, and the teardrop-to-Perkins' line distance. The fact that we also observed favorable acetabular development in patients who were over 8 years of age at the time of surgery perhaps reflects a delay in maturation of the secondary centers of ossification. This would be consistent with a general delay in growth plate closure observed in spastic patients. Handelsman¹³ observed that acetabular development steadily improved from the time of osteotomy until the end of the period of hip joint development and remodeling.

These long-range radiographic observations lead us to the conclusion that the Chiari osteotomy can be confidently applied to neuromuscular patients with hip subluxation/dislocation.