Recurrence is a common complication after surgical correction of hallux valgus deformity, and reported recurrence rates after treatment with proximal metatarsal osteotomy range from 4% to 25%1-6. A large preoperative hallux valgus angle (HVA)7; insufficient correction of the HVA5, the intermetatarsal angle (IMA)8, sesamoid position4, and the distal metatarsal articular angle (DMAA)9; severe metatarsus adductus10; and a round-shaped metatarsal head11 on postoperative radiographs have been identified as risk factors for the recurrence of hallux valgus deformity. However, only 1 or 2 factors were analyzed in each of those studies, and thus, a comprehensive analysis of risk factors is necessary to evaluate the relative importance of a specific risk factor compared with another risk factor.
Most studies of the recurrence of hallux valgus have involved an analysis of radiographic parameters on weight-bearing radiographs4,5,8,9, with the exception of a study by Okuda et al.11, who investigated the relationship between the shape of the lateral edge of the first metatarsal head and recurrence. If the recurrence of hallux valgus can be predicted using immediate postoperative non-weight-bearing radiographs, evaluating intraoperative radiographs and performing additional intraoperative procedures may be helpful to decreasing the risk of recurrence. We believe that an immediate postoperative radiograph reflects the intraoperative correction at the end of skin closure because both immediate postoperative and intraoperative radiographs are made in a non-weight-bearing state.
The aims of the current study were to identify risk factors for recurrence and to clarify whether recurrence after surgery can be predicted using radiographic parameters assessed on immediate postoperative non-weight-bearing radiographs.
Materials and Methods
The current study was approved by our institutional review board. We retrospectively reviewed 105 consecutive patients (131 feet) with symptomatic, moderate to severe hallux valgus deformity (HVA of ≥20° or IMA of ≥12°) in whom a proximal chevron osteotomy of the first metatarsal combined with a distal soft-tissue procedure was performed by a single surgeon between January 2008 and December 2009. The inclusion and exclusion criteria are listed in Table I. Ten patients (12 feet) were lost to follow-up, and 2 patients (2 feet) were excluded because of the absence of the medial sesamoid. A total of 93 patients (117 feet) were included (a follow-up rate of 89.3% of feet). All except 2 of the patients (2 feet) were female. The mean patient age was 51 years (range, 19 to 71 years), and the mean duration of follow-up was 27.5 months (range, 24 to 35 months). A closing-wedge osteotomy of the proximal phalanx was performed in 70 patients (80 feet) in whom residual hallux valgus deformity was apparent after the proximal chevron osteotomy of the first metatarsal.
Hallux valgus recurrence was defined as an HVA of ≥20°. Feet were grouped according to nonrecurrence or recurrence. No differences in terms of age (p = 1.0), duration of follow-up (p = 0.284), and sex (p = 0.99) were observed between these groups. Changes in the HVA, IMA, and sesamoid position over time were analyzed by comparing values measured preoperatively, immediately postoperatively, at 6 weeks and 3 and 6 months postoperatively, and at the time of the last follow-up in both groups.
The preoperative and immediate postoperative HVA, IMA, and sesamoid position; the preoperative metatarsus adductus angle; and the immediate postoperative DMAA were compared between the 2 groups. We did not perform a comparison of the preoperative DMAA in the current study because the medial and lateral margins of the distal articular surface could not be defined on many preoperative radiographs and the preoperative DMAA had low intraobserver and interobserver reliabilities (Table II). The shape of the lateral edge of the first metatarsal head was classified according to whether the round sign was positive and negative using the method described by Okuda et al.11 and was compared between the groups. In addition, cutoff values for recurrence were determined for each radiographic parameter, and the relative risks of recurrence as indicated by preoperative and postoperative radiographic parameters were determined.
We randomly selected the preoperative and postoperative dorsoplantar radiographs of 51 patients on the basis of a calculation of sample size according to a Bonett approximation12 to assess the intraobserver and interobserver reliabilities of radiographic measurements in this study. Two foot and ankle surgeons who were independent of the operative team and were blinded to the outcome performed measurements of the radiographic parameters. Measurements were repeated 2 weeks later. The intraobserver and interobserver reliabilities of measurements of the HVA, IMA, DMAA, and metatarsus adductus angle were analyzed using intraclass correlation coefficients. Kappa statistics were used for an analysis of the reliabilities of sesamoid position and the shape of the lateral edge of the first metatarsal head.
Because the postoperative HVA can be decreased by a proximal phalangeal osteotomy, an additional analysis was performed to determine the effect of a proximal phalangeal osteotomy. Thus, the feet were grouped on the basis of whether or not a proximal phalangeal osteotomy was performed.
All radiographs were made at a single facility and according to the same radiographic protocol. Weight-bearing dorsoplantar radiographs were made preoperatively, at 3 and 6 months postoperatively, and at the time of the last follow-up. Non-weight-bearing dorsoplantar radiographs were made immediately postoperatively and at 6 weeks after surgery. The radiographs were retrieved by a picture archiving and communication system (PACS) (IMPAX; Agfa HealthCare). Radiographic measurements were conducted using PACS software by 1 observer who was independent of the operative team and was blinded to the outcome.
The HVA was defined as the angle between the longitudinal axis of the first metatarsal and that of the proximal phalanx. The IMA was defined as the angle between the longitudinal axis of the first and that of the second metatarsal. The longitudinal axis of the first metatarsal was defined as the line connecting the center of the proximal articular surface of the first metatarsal to the center of the first metatarsal head13. The longitudinal axis of the proximal phalanx and that of the second metatarsal was defined as the line connecting the centers of the proximal and distal ends of the diaphysis13. Sesamoid position was defined as the position of the medial sesamoid in relation to the longitudinal axis of the first metatarsal and was graded from 1 to 714.
The DMAA was defined as the angle between a line perpendicular to the longitudinal axis of the first metatarsal and a line delineating the orientation of the articular surface of the metatarsal head15. The metatarsus adductus angle was measured as the position of the lesser metatarsus relative to the midfoot as described by Engel et al.16.
A 7-cm medial incision was made along the inferior margin of the first metatarsal. A medial capsule was vertically resected just proximal to the base of the proximal phalanx, and excision of the medial eminence was performed at 2 mm medial to the sagittal sulcus of the metatarsal head in a line parallel with the longitudinal axis of the first metatarsal. A full-thickness dorsal flap including skin and subcutaneous tissue was elevated to the first web space. Further dissection progressed laterally, just superficial to the extensor hallucis longus tendon. The distal end of the adductor tendon and deep transverse metatarsal ligament were released. A capsular incision was then made along the dorsal margin of the lateral sesamoid.
A proximal chevron osteotomy was performed at an angle of 60°. The chevron apex was placed 7 mm distal to the first metatarsocuneiform joint. The correction was then accomplished by translation and angulation at the osteotomy site. Three Kirschner wires, 1.6 mm in diameter, were then inserted proximally to distally into the metatarsal head and buried under the skin. Correction was checked using C-arm fluoroscopy. The medial protruding bone of the proximal fragment was removed flush with the distal fragment, and this bone was placed to fill the osteotomy gap. If required, a closing-wedge osteotomy was performed at 5 mm distal to the proximal phalanx base. The medial capsule was repaired using absorbable sutures without tension.
A short leg splint was applied for 1 week postoperatively. Weight-bearing on the heel was allowed on the day after surgery. Full weight-bearing on the first ray was not allowed until the seventh postoperative week, and Kirschner wires were removed during the ninth postoperative week.
All dependent variables were tested for normality of the data distribution using a Kolmogorov-Smirnov test. A Mann-Whitney U test and chi-square test were used to assess the difference in age, duration of follow-up, and sex. A Mann-Whitney U test was used for the comparison of preoperative and postoperative radiographic results between the nonrecurrence and recurrence groups. A Wilcoxon signed-rank test was used for the comparison of radiographic results at each postoperative period. Cutoff values for radiographic parameters that allowed the prediction of hallux valgus recurrence were determined by receiver operating characteristic (ROC) curve analysis. Cutoff values were selected on the basis of the maximal sensitivity and specificity sum. Using determined cutoff values, radiographic results were dichotomized. Binary logistic regression analysis was then used to determine the risk factors for recurrence and performed using preoperative and postoperative variables separately. Odds ratios (ORs) were calculated with 95% confidence intervals (CIs). A chi-square test was used for an analysis of the association between the shape of the lateral edge of the first metatarsal head and recurrence. For all tests, a p value of <0.05 was considered significant.
A sample-size calculation for the logistic regression was based on the guideline of Peduzzi et al.17: n = 10 × k/p (where n is the minimum number of cases needed, k is the number of predictor variables, and p is the proportion of cases with recurrence). For our regression model, there were 4 predictor variables and the recurrence rate was 0.171; therefore, a minimum sample size of 234 would have been needed (see Discussion).
The intraobserver and interobserver reliabilities of the radiographic measurements are shown in Table II. The preoperative DMAA had the lowest reliabilities.
Twenty (17.1%) of the 117 feet showed hallux valgus recurrence at the last follow-up: in 10 (50%) of the feet, the HVA was 20° to 25°; in 4 (20%), it was 26° to 30°; and in 6 (30%), it was ≥31°. Changes in the HVA, the IMA, and sesamoid position in both groups are shown in Figures 1, 2, and 3.
The mean preoperative HVA was significantly larger in the recurrence group (p < 0.001). The mean preoperative IMA and sesamoid position did not differ between the nonrecurrence and recurrence groups. The mean immediate postoperative HVA and grade of sesamoid position were significantly greater in the recurrence group (p < 0.001 and p = 0.008, respectively). The mean immediate postoperative IMA did not differ between the 2 groups. The HVA and the IMA stabilized at 6 months after surgery in the nonrecurrence group. The mean preoperative metatarsus adductus angle and immediate postoperative DMAA were significantly larger in the recurrence group (p < 0.001 and p = 0.002, respectively). No significant difference in the shape of the lateral edge of the first metatarsal head was observed between the groups (Table III).
Cutoff values for recurrence, which were calculated using ROC curve analysis, were 40° (area under the curve, 0.801) for the preoperative HVA, 8° (0.795) for the immediate postoperative HVA, 3° (0.507) for the immediate postoperative IMA, grade 4 (0.682) for the immediate postoperative sesamoid position, 17° (0.724) for the immediate postoperative DMAA, and 23° (0.750) for the preoperative metatarsus adductus angle. The immediate postoperative IMA, DMAA, and round sign of the first metatarsal head did not show a significant association with hallux valgus recurrence. However, an immediate postoperative HVA of ≥8°, an immediate postoperative sesamoid position of grade 4 or greater, a preoperative metatarsus adductus angle of ≥23°, and a preoperative HVA of ≥40° showed a significant association with recurrence (p < 0.001 to p = 0.007) (Table IV).
No significant difference with respect to the HVA at all time points after surgery was observed between the group that underwent proximal phalangeal osteotomy and the group that did not (Fig. 4). However, the mean correction of HVA immediately postoperatively was significantly larger in the group that underwent proximal phalangeal osteotomy (p < 0.001). At the last follow-up, no significant difference in the HVA was observed between the 2 groups (p = 0.918).
The recurrence of hallux valgus has been variously defined2,4,5,8,18. Veri et al.5 defined recurrence as an increase in the HVA of ≥10°. Okuda et al.4,8 and Coughlin and Jones2 defined recurrence as an HVA of >20°. The lower limit of moderate to severe hallux valgus is 20°, and several articles have defined recurrence after hallux valgus correction as an HVA of ≥20°2,4,8,11,19-21, and thus, recurrence in the current study was also defined as an HVA of ≥20°.
The recurrence of hallux valgus is one of the most important complications after surgery because it is closely related to patient satisfaction22. However, some previous studies either did not mention the recurrence rate23,24 or provided recurrence rates without a definition of recurrence3,6,25-27. Reported recurrence rates after a proximal metatarsal osteotomy have ranged from 4% to 25%1-6. In the current study, recurrence was found in 20 (17.1%) of the 117 feet at the time of the last follow-up. Although a recurrence rate of 17.1% falls within the range previously reported, it is higher than that reported in several studies2,3,5,6. Patients in the current study had a larger mean preoperative metatarsus adductus angle and postoperative DMAA than typically observed7,28-30, which might explain the high rate of recurrence encountered. In addition, the mean postoperative HVA in this study was also comparable with that in previous reports, and 10 (50%) of all feet with recurrence had an HVA of 20° to 25°.
The known risk factors for hallux valgus recurrence are a large preoperative HVA7; insufficient correction of the HVA5, the IMA8, sesamoid position4, and the DMAA9; severe metatarsus adductus10, and a round-shaped metatarsal head11. Our analysis showed that insufficient correction of the HVA and sesamoid position as assessed immediately postoperatively, severe preoperative metatarsus adductus, and severe hallux valgus deformity as assessed by preoperative HVA are risk factors for recurrence. Hallux valgus with severe metatarsus adductus is difficult to treat. Therefore, surgeons should explain the risk of recurrence when counseling patients with severe metatarsus adductus prior to surgery to correct hallux valgus deformity. Of the risk factors examined, an immediate postoperative HVA of ≥8° had the strongest association with recurrence (an OR of 28 times that of feet with an immediate postoperative HVA of <8°). The IMA was sufficiently corrected to an average of 3.1° immediately postoperatively in both the nonrecurrence and recurrence groups. We think that this is the reason why insufficient correction of the IMA was not found to be a risk factor for recurrence in this study. In addition, a round-shaped first metatarsal head was also not found to be a risk factor. In fact, 12 (18.2%) of the 66 feet with a negative round sign on the immediate postoperative radiograph showed recurrence at the last follow-up. Although the DMAA was also previously identified as a risk factor for recurrence9, in the logistic regression analysis in the current study, the immediate postoperative DMAA was not significantly associated with recurrence. However, this parameter was significantly larger in the recurrence group than in the nonrecurrence group. Therefore, we believe that in a future study with a larger number of cases, a large DMAA might be demonstrated to be a risk factor for recurrence.
The method of determining the axis of the first metatarsal by bisecting the shaft of the metatarsal is widely used to assess hallux valgus deformity14,31. Some authors reported that this method showed high reliability of measurement31. However, Schneider et al.32 reported that this method showed high preoperative reliability but low postoperative reliability. Shima et al.13 recommended the method of determining the axis of the first metatarsal by connecting the centers of the first metatarsal head and the proximal articular surface because that had the highest preoperative and postoperative reliabilities. Therefore, we used the latter method in the present study.
This study demonstrates that recurrence can be predicted from non-weight-bearing radiographs made immediately postoperatively. We believe that the results from the analysis of immediate postoperative radiographs can be used to suggest intraoperative guidelines for satisfactory correction of radiographic parameters because both intraoperative and immediate postoperative radiographs are made in a non-weight-bearing state. Therefore, we suggest that further correction should be performed if an HVA of ≥8° or sesamoid position of grade 4 or greater is observed on intraoperative radiographs.
One limitation of this study was that a closing-wedge osteotomy was performed at the base of the proximal phalanx in conjunction with a proximal chevron osteotomy in 80 of the 117 feet. Although a proximal phalangeal osteotomy alone is known to be ineffective in the prevention of hallux valgus recurrence33-35, we know of no previous studies of the effect when proximal phalangeal osteotomy is performed in conjunction with metatarsal osteotomy. In this study, the HVA showed a tendency to increase over time in patients who underwent proximal phalangeal osteotomy, although the mean correction of HVA was significantly larger after surgery in patients who underwent proximal phalangeal osteotomy (Fig. 4). Thus, we think that proximal phalangeal osteotomy might not have an effect on reducing the recurrence of hallux valgus. However, a prospective case-controlled trial may be needed to assess the effect of a proximal phalangeal osteotomy on the recurrence of hallux valgus.
The retrospective nature of this study and the relatively small sample size were also limitations of this study. A power analysis of the logistic regression revealed that a minimum sample size of 234 would have been required. Because this study did not have a large enough sample size, we performed logistic regression using preoperative and postoperative variables separately to enhance statistical power. Therefore, we think that a randomized controlled trial with a large sample is needed. Another limitation was that immediate postoperative radiographs were used instead of intraoperative radiographs to assess the state of intraoperative correction at the end of surgery. Although both immediate postoperative radiographs and intraoperative radiographs are made in a non-weight-bearing state, the surgeon can make radiographs intraoperatively that are similar to those of a weight-bearing state because patients are under anesthesia. In addition, there is a possibility that a radiograph made after dressing the wound and applying a splint may not be the same as the intraoperative radiograph. However, we did not attempt to spread the big toe from the second toe using thick dressing material. Therefore, we believe that the immediate postoperative radiograph reflects the state of intraoperative correction.
In summary, risk factors for hallux valgus recurrence were insufficient correction of the HVA and sesamoid position as assessed immediately postoperatively, severe preoperative metatarsus adductus, and severe hallux valgus deformity as assessed by preoperative HVA. Using an immediate postoperative HVA cutoff of 8°, feet with an HVA of ≥8° had an OR for recurrence of 28 times that of feet with an HVA of <8°. Accordingly, we conclude that recurrence after proximal chevron osteotomy for hallux valgus can be predicted from immediate postoperative non-weight-bearing radiographs.
Investigation performed at Inje University Seoul Paik Hospital, Seoul, Republic of Korea
A commentary by Jákup Midjord, MD, is linked to the online version of this article at jbjs.org.
Disclosure: No external funding was received for this study. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article (http://links.lww.com/JBJS/D429).
1. Tanaka Y, Takakura Y, Kumai T, Sugimoto K, Taniguchi A, Hattori K. Proximal spherical metatarsal osteotomy for the foot with severe hallux valgus. Foot Ankle Int. 2008 ;29(10):1025–30. Epub 2008 Oct 15.
2. Coughlin MJ, Jones CP. Hallux valgus and first ray mobility. A prospective study. J Bone Joint Surg Am. 2007 ;89(9):1887–98.
3. Mann RA, Rudicel S, Graves SC. Repair of hallux valgus with a distal soft-tissue procedure and proximal metatarsal osteotomy. A long-term follow-up. J Bone Joint Surg Am. 1992 ;74(1):124–9.
4. Okuda R, Kinoshita M, Yasuda T, Jotoku T, Kitano N, Shima H. Postoperative incomplete reduction of the sesamoids as a risk factor for recurrence of hallux valgus. J Bone Joint Surg Am. 2009 ;91(7):1637–45.
5. Veri JP, Pirani SP, Claridge R. Crescentic proximal metatarsal osteotomy for moderate to severe hallux valgus: a mean 12.2 year follow-up study. Foot Ankle Int. 2001 ;22(10):817–22.
6. Zettl R, Trnka HJ, Easley M, Salzer M, Ritschl P. Moderate to severe hallux valgus deformity: correction with proximal crescentic osteotomy and distal soft-tissue release. Arch Orthop Trauma Surg. 2000;120(7-8):397–402.
7. Deenik AR, de Visser E, Louwerens JW, de Waal Malefijt M, Draijer FF, de Bie RA. Hallux valgus angle as main predictor for correction of hallux valgus. BMC Musculoskelet Disord. 2008 ;9:70.
8. Okuda R, Kinoshita M, Yasuda T, Jotoku T, Shima H. Proximal metatarsal osteotomy for hallux valgus: comparison of outcome for moderate and severe deformities. Foot Ankle Int. 2008 ;29(7):664–70.
9. Bonnel F, Canovas F, Poirée G, Dusserre F, Vergnes C. [Evaluation of the Scarf osteotomy in hallux valgus related to distal metatarsal articular angle: a prospective study of 79 operated cases]. Rev Chir Orthop Reparatrice Appar Mot. 1999 ;85(4):381–6. French.
10. Pontious J, Mahan KT, Carter S. Characteristics of adolescent hallux abducto valgus. A retrospective review. J Am Podiatr Med Assoc. 1994 ;84(5):208–18.
11. Okuda R, Kinoshita M, Yasuda T, Jotoku T, Kitano N, Shima H. The shape of the lateral edge of the first metatarsal head as a risk factor for recurrence of hallux valgus. J Bone Joint Surg Am. 2007 ;89(10):2163–72.
12. Bonett DG. Sample size requirements for estimating intraclass correlations with desired precision. Stat Med. 2002 ;21(9):1331–5.
13. Shima H, Okuda R, Yasuda T, Jotoku T, Kitano N, Kinoshita M. Radiographic measurements in patients with hallux valgus before and after proximal crescentic osteotomy. J Bone Joint Surg Am. 2009 ;91(6):1369–76.
14. Hardy RH, Clapham JC. Observations on hallux valgus; based on a controlled series. J Bone Joint Surg Br. 1951 ;33-B(3):376–91.
15. Richardson EG, Graves SC, McClure JT, Boone RT. First metatarsal head-shaft angle: a method of determination. Foot Ankle. 1993 ;14(4):181–5.
16. Engel E, Erlick N, Krems I. A simplified metatarsus adductus angle. J Am Podiatry Assoc. 1983 ;73(12):620–8.
17. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996 ;49(12):1373–9.
18. Okuda R, Kinoshita M, Morikawa J, Yasuda T, Abe M. Proximal metatarsal osteotomy: relation between 1- to greater than 3-years results. Clin Orthop Relat Res. 2005 ;435:191–6.
19. Park CH, Ahn JY, Kim YM, Lee WC. Plate fixation for proximal chevron osteotomy has greater risk for hallux valgus recurrence than Kirschner wire fixation. Int Orthop. 2013 ;37(6):1085–92. Epub 2013 Feb 20.
20. Park CH, Cho JH, Moon JJ, Lee WC. Can double osteotomy be a solution for adult hallux valgus deformity with an increased distal metatarsal articular angle? J Foot Ankle Surg. 2016 ;55(1):188–92. Epub 2015 Jul 26.
21. Park CH, Jang JH, Lee SH, Lee WC. A comparison of proximal and distal chevron osteotomy for the correction of moderate hallux valgus deformity. Bone Joint J. 2013 ;95-B(5):649–56. Epub 2013 May 2.
22. Schneider W, Knahr K. Surgery for hallux valgus. The expectations of patients and surgeons. Int Orthop. 2001;25(6):382–5.
23. Borton DC, Stephens MM. Basal metatarsal osteotomy for hallux valgus. J Bone Joint Surg Br. 1994 ;76(2):204–9.
24. Gallentine JW, Deorio JK, Deorio MJ. Bunion surgery using locking-plate fixation of proximal metatarsal chevron osteotomies. Foot Ankle Int. 2007 ;28(3):361–8.
25. Schneider W, Aigner N, Pinggera O, Knahr K. Chevron osteotomy in hallux valgus. Ten-year results of 112 cases. J Bone Joint Surg Br. 2004 ;86(7):1016–20.
26. Choi WJ, Yoon HK, Yoon HS, Kim BS, Lee JW. Comparison of the proximal chevron and Ludloff osteotomies for the correction of hallux valgus. Foot Ankle Int. 2009 ;30(12):1154–60.
27. Fuhrmann RA, Zollinger-Kies H, Kundert HP. Mid-term results of Scarf osteotomy in hallux valgus. Int Orthop. 2010 ;34(7):981–9. Epub 2010 Feb 16.
28. Chi TD, Davitt J, Younger A, Holt S, Sangeorzan BJ. Intra- and inter-observer reliability of the distal metatarsal articular angle in adult hallux valgus. Foot Ankle Int. 2002 ;23(8):722–6.
29. Ferrari J, Malone-Lee J. A radiographic study of the relationship between metatarsus adductus and hallux valgus. J Foot Ankle Surg. 2003 ;42(1):9–14.
30. Griffiths TA, Palladino SJ. Metatarsus adductus and selected radiographic measurements of the first ray in normal feet. J Am Podiatr Med Assoc. 1992 ;82(12):616–22.
31. Coughlin MJ, Freund E, Roger A. Roger A. Mann Award. The reliability of angular measurements in hallux valgus deformities. Foot Ankle Int. 2001 ;22(5):369–79.
32. Schneider W, Csepan R, Knahr K. Reproducibility of the radiographic metatarsophalangeal angle in hallux surgery. J Bone Joint Surg Am. 2003 ;85(3):494–9.
33. Frey C, Jahss M, Kummer FJ. The Akin procedure: an analysis of results. Foot Ankle. 1991 ;12(1):1–6.
34. Goldberg I, Bahar A, Yosipovitch Z. Late results after correction of hallux valgus deformity by basilar phalangeal osteotomy. J Bone Joint Surg Am. 1987 ;69(1):64–7.
35. Plattner PF, Van Manen JW. Results of Akin type proximal phalangeal osteotomy for correction of hallux valgus deformity. Orthopedics. 1990 ;13(9):989–96.