Scarf Osteotomy for Hallux Valgus Deformity: An Intermediate Followup of Clinical and Radiographic Outcomes

INTRODUCTION

Hallux valgus is a complex progressive deformity affecting the forefoot. The main pathologic anatomy concerns the first metatarsophalangeal (MTP) joint, including a varus or medial deviation of the first metatarsal and pronation deformity in the longitudinal axis. The valgus deformity of the MTP joint is classified as congruent or incongruent based on the degree of subluxation of the phalanx off the metatarsal head. ⁶ The complex deformity also may include other conditions, such as overriding of the second toe, metatarsalgia, hammer and claw deformities of the lateral toes, and bunionette deformity of the fifth metatarsal. ¹³

More than 130 procedures have been described to correct hallux valgus deformity, with bony procedures designed to correct the varus deformity of the first metatarsal. A mild deformity can be corrected by distal osteotomy, and a severe deformity can be corrected with a proximal osteotomy. These osteotomies are further divided into displacement osteotomies and angular osteotomies. While these procedures are done to correct varus deformity, other changes in the configuration of the first metatarsal can occur. The ideal osteotomy should correct the varus angle and pronation without any elevation or shortening of the first metatarsal. ^10,12,16

The scarf osteotomy is a versatile procedure for the correction of moderate to severe hallux valgus deformity. It allows lateralization of the head-shaft fragment to reduce the intermetatarsal angle, maintains joint congruity and thus motion at the MTP joint, and allows plantar or dorsal displacement, elongation or shortening, and transverse plane rotation of the first metatarsal up to 10 degrees. Originally, the scarf or z-osteotomy was described by Burutaran in 1976 ¹⁶ and was popularized by Weil ¹⁶ and Borelli ¹⁴ in the United States and Barouk in France. ² The term scarf is an architectural term defined as a joint made by notching or cutting the ends of two pieces and fastening them together so that they overlap and join firmly into one continuous piece. ^1,2

The purpose of this study was to evaluate a series of consecutive patients, over a 2-year period, after a scarf osteotomy of the first metatarsal. The clinical and radiographic outcomes were measured over a 12 to 24 month followup period. We specifically looked at preoperative and postoperative American Orthopaedic Foot and Ankle (AOFAS) scores (hallux metatarsophalangeal-interphalangeal scale); ⁹ radiographic parameters including the intermetatarsal angle (IMA), hallux valgus angle (HVA), distal metatarsal articular angle (DMAA), relative metatarsal length, and the level of the first metatarsal head, and foot pressure analysis.

MATERIALS AND METHODS

This study was approved by our institution's review board, and informed consent was obtained from all patients. From July, 2001, to January, 2003, the scarf osteotomy was done on 29 consecutive patients with moderate to severe hallux valgus deformity (IMA 12 to 20 degrees and HVA 20 to 46 degrees). All patients had idiopathic bunions. One patient had a previous bunionectomy. There were 28 women and one man, with an average age at the time of surgery of 55.4 (range 18 to 76) years. Preoperative and postoperative evaluations included standing anteroposterior and lateral radiographs, AOFAS score, physical examination, and foot pressure analysis by weight-bearing ink prints (footprints were taken using the Harris Foot ink Mat). (ACOR, Cleveland, OH) We recorded IMA, HVA, DMAA, relative metatarsal length, and the level of the first metatarsal head on preoperative and postoperative radiographs. The data are summarized in Table 1. An independent investigator measured the radiographic parameters. Patients were evaluated radiographically and clinically in the initial postoperative period (0 to 1 month), intermediate postoperative period (2 to 6 months) and final followup (12 to 24 months). A paired student-t test and repeated measures ANOVA test were used to analyze the data.

OPEN IN VIEWER

Table 1:

Patient data: preoperative and postoperative parameters

Patient	Operation(s)	HVA		IMA		AOFAS Score
	In addition to scarf osteotomy	pre-op	post-op	pre-op	post-op	pre-op	post-op
1	L, Akin, L2, L3 Weil	35	12	14	8	52	80
2	R, R2 Weil, GT	40	10	16	8	34	85
3	L	30	14	13	8	39	92
4	R, R2 PIP arthroplasty	38	20	18	9	44	81
5	R	30	14	18	9	59	92
	L	30	12	16	10	59	87
6	R, R2 Weil, R5 chevron	40	20	16	10	62	95
7	L	32	22	14	10	60	90
8	R, R2 Weil	40	20	12	20	34	80
	L	32	20	12	9	29	85
9	R	20	12	15	10	65	95
10	L, L2 Weil	35	12	17	10	53	92
11	L, Akin	40	16	20	10	55	95
12	R, R2 Weil	20	10	20	10	60	95
13	L, Akin, L2 Weil	30	5	12	6	54	92
14	R, Akin, R2 Weil	38	17	17	12	57	87
15	R, Akin, R2, R3 Weil	42	22	14	10	57	85
16	R	38	12	16	8	62	94
	L	30	12	16	10	78	94
17	R	30	22	13	9	75	95
	L	32	20	14	8	65	88
18	R	30	16	13	10	60	69
	L	30	16	12	10	70	97
19	R, R2, R3 Weil, R5 chevron	40	20	15	10	47	83
20	R, R2 Weil Neuroma excision 3/4	30	16	16	8	44	82
21	R, R2, R3 Weil	44	40	20	20	47	57
	L	46	24	20	12	50	68

HVA = hallux valgus angle measured in degrees; IMA = intermetatarsal angle.

RESULTS

Of the 29 patients, eight were lost to followup because of time constraints, pregnancy, and relocation. Six of the remaining 21 patients had bilateral operations, making 27 feet available for analysis. Fourteen of the 27 feet had concurrent operations on the lesser toes for hammer-toe correction or preoperative metatarsalgia. The average preoperative and postoperative data are summarized in Table 2.

The paired student t-test on the 27 feet showed a statistically significant improvement (p < 0.0001) between preoperative and postoperative IMA, HVA, and AOFAS score. Two feet had recurrence of the hallux valgus deformity. The paired ANOVA analysis of the 25 feet without recurrence showed a statistically significant improvement in the preoperative and postoperative parameters (p < 0.0001). From this subset, the multiple comparison student-Newman-Keuls post-hoc test showed a statistically significant (p < 0.0001) preservation of the correction in the intermediate followup period (2 to 6 months) to final folloup at an average of 16.1 months.

The average relative lengths of the first and second metatarsals preoperatively and postoperatively, were 10.1 mm and 10.3 mm, respectively. The relative length of the metatarsals is defined as the difference between the first and second metatarsal lengths. The measurement is based on the anteroposterior standing radiographs and measured by a line intersecting the midway point at the diaphyseal-metaphyseal junction of the metatarsal and extending from the most proximal to distal aspects of the bone. ⁸ There was no attempt to change the length of first metatarsal with the index operation. Finally, there was no evidence of a transfer lesion on the preoperative and postoperative footprints in all cases.

OPEN IN VIEWER

Table 2:

Data analysis from preoperative to final postoperative measures

HV	IMA	AOFAS Score
pre-op	34.5 (20–46)	15.4 (12–20)	54.5 (34–78)
post-op	16.9 (5–40)	10.1 (8–20)	86.5 (57–97)

HV = hallux valgus; IMA = intermetatarsal angle.

DISCUSSION

The scarf osteotomy is a mechanically stable procedure that can be used to correct hallux valgus deformity with an IMA of more than 15 degrees and HVA of more than 35 degrees. Because of the stability of the construct, bilateral procedures are better tolerated by patients, with earlier weight bearing and recovery compared to proximal osteotomies that require a period of casting and nonweightbearing. ^11,14,15

With this technique, we obtained good correction of the deformity with respect to the HVA and IMA, and the correction was maintained over the 12 to 24 month followup period. The AOFAS score improved an average of 32 points, and there was no evidence of transfer lesion based on footprints. In 14 of the 27 feet additional procedures were done, which might have affected the patients’ pain ratings and had an effect on some of the radiographic parameters. This is one of the shortcomings of this study. Further investigation is needed to study the correction and pain relief obtained solely from the scarf osteotomy.

In prior studies, complications of the scarf osteotomy included superficial wound infection, stress fractures, traumatic dislocation of the distal fragment, and recurrence. ^5,7 One of our patient's postoperative period was complicated by superficial cellulitis, which resolved with a short duration of oral antibiotics. In the 27 feet, only two had recurrences. In one, the deformity recurred within 1-year after surgery. Postoperatively, this patient had increased plantarflexion of the first metatarsal that we attributed to a technical error: osteotomy was made more horizontal in the sagittal plane rather than 15 degrees dorsal to plantar angle. The other patient with recurrence of the deformity had an incongruent joint preoperatively that was not corrected at the index procedure. Both patients had large deformities with an IMA of more than 20 degrees. Therefore, we suggest that in incongruent joints the scarf osteotomy should be augmented with a soft-tissue release, and in cases with large deformities with an IMA of more than 20 degrees the scarf osteotomy will probably not reduce the IMA to normal values. However, a larger database of patients needs to be analyzed to evaluate this concern compared to proximal osteotomies that require a period of casting and nonweightbearing.