Comparison of Vickers’ Physiolysis With Osteotomy for Primary Correction of Clinodactyly

Abstract

Background: The purpose of this study was to compare the Vickers physiolysis procedure with osteotomy for correction of digital clinodactyly and determine which method provides better correction at final follow-up or whether the patient’s age, preoperative angulation, or presence of syndactyly affects final outcomes. Methods: All patients of skeletal immaturity who underwent surgical correction of clinodactyly were evaluated with clinical examination and radiographs to determine the percentage and absolute change in the degree of clinodactyly pre- versus postoperatively, in addition to stratification based on the degree of deformity, age, and presence of syndactyly. Results: Vickers’ physiolysis and osteotomy were undertaken in 30 and 11 digits, respectively. The angulation significantly improved from 43.0° to 23.9°, with a 46.2% correction of deformity in the Vickers group at 46.3 months. The angulation decreased from 39.2° to 22.4° in the osteotomy group, with a 55.3% correction of deformity at 55.3 months. There was better correction in those with isolated clinodactyly compared with those with concomitant syndactyly and better percentage of correction in patients with lesser deformity in the Vickers group. There were more reoperations in the osteotomy group. Conclusions: The use of osteotomy may lead to more revision cases, whereas the Vickers procedure has minimal complications and need for revision. The Vickers physiolysis procedure is more effective in those with angulation <55°.

Introduction

Clinodactyly is a common congenital hand abnormality, occurring in approximately 1% of normal newborns and 12% of newborns with major congenital abnormalities.¹ Clinically, it is a radial or ulnar angular abnormality seen in pediatric patients. Clinodactyly may also occur sporadically or after physeal trauma.¹ Clinodactyly can occur in isolation, usually involving the small finger, or as part of a spectrum of greater congenital abnormalities.² Functionally, clinodactyly can result in limitation of pinch if the thumb or index finger is involved and can be a cosmetic concern for the patient and parents. Despite its prevalence, its management is a topic of debate and controversy.

The most common radiographic appearance is a longitudinal bracketed epiphysis of the middle phalanx of the involved digit with radial deviation due to abnormal, unbalanced longitudinal growth (Figure 1).^3,4 Cooney classified clinodactyly into simple, complicated, and complex types based on the degrees of angulation and presence of soft tissue abnormalities.⁵ The type and timing of surgical intervention are still debated. Numerous surgical options have been described, leading to functional and aesthetic improvement, including observation, physiolysis,^6–8 distraction,⁹ and multiple types of osteotomies (opening wedge,¹⁰ closing wedge,^11,12 and reverse wedge¹³).

Figure 1.

Different phenotypic appearances of clinodactyly: (a) isolated small finger, (b) ring finger with metacarpal synostosis, (c) index finger with type A postaxial polydactyly, and (d) small finger with complete syndactyly of long, ring, and small digits.

Physiolysis with fat graft interposition removes the abnormal physis and bony bracket, allowing longitudinal finger growth and self-correction of the angular deformity over time. Physiolysis has been shown to have a role in the early correction of clinodactyly; however, no comparisons have been drawn to performing an osteotomy for the correction of clinodactyly in the current literature.^6,8,14 The purpose of this investigation was to critically review the outcomes of primary clinodactyly correction in a matched cohort of children who underwent either a Vickers physiolysis with fat graft interposition or a corrective osteotomy. The primary outcome measure was the change in angular deformity of the involved digit. As a secondary outcome, we wished to determine whether age, presence of syndactyly, or preoperative angulation had an effect on the outcome after clinodactyly correction.

Materials and Methods

After institutional review board approval was obtained, patients treated for congenital clinodactyly at 2 institutions by 2 surgeons were retrospectively reviewed from 1996 to 2016. The pertinent medical records, operative reports, and radiographs of all patients treated surgically with physiolysis or osteotomy for the correction of clinodactyly were evaluated. The 2 groups of patients were matched for age at surgery, sex, and preoperative angulation deformity. For the purpose of this investigation, we defined clinodactyly as digital curvature greater than 10° in a radioulnar plane on an anteroposterior (AP) radiograph. Inclusion criteria consisted of children having radiographic evidence of clinodactyly who underwent a physiolysis or osteotomy procedure to improve clinical appearance and function. Exclusion criteria included children with less than 9 months of documented follow-up, thumb clinodactyly, traumatic clinodactyly, and the absence of adequate pre- or postoperative radiographs to measure angular changes.

We reviewed the preoperative radiographs in all patients to assess for location of involved phalanx, geometric shape of involved phalanx, and degree of angulation of the involved digit (Figure 2).¹⁵ Postoperative AP radiographs at last follow-up were utilized to calculate the degree of angulation and to determine the absolute degree of correction. Percentage of correction was based on the ratio of the preoperative angulation minus the postoperative angulation divided by the preoperative angulation and multiplied by 100.

Figure 2.

Sample radiographs demonstrating measurement of clinodactyly angulation divergence: (a) preoperative appearance and (b) postoperative appearance after Vickers’ physiolysis.

Our primary outcome measure was the difference in angular correction between the physiolysis and osteotomy groups at final follow-up. This was done by comparing the mean absolute difference between pre- and postoperative angulation and percentage correction between the 2 groups. Outcomes were also compared by sex, age group, preoperative degree of angulation, presence of syndactyly, and family history of clinodactyly.

Surgical Technique

Vickers’ physiolysis with fat grafting

A mid-axial incision on the radial border of the affected digit was made. The radial neurovascular bundle was retracted and the involved phalanx was exposed. The periosteum was elevated in palmar and dorsal directions. A curved Beaver blade was used to remove a segment of epiphysis and radial physis in a layered fashion, resulting in the removal of bracketed epiphysis. To ensure complete removal of the physeal cartilage, resection proceeded until cancellous bone was encountered. Caution was taken to maintain proximal physis to prevent disruption of subsequent longitudinal growth. A small intercalary fat graft was harvested from various donor sites for placement into the physiolysis defect. The fat graft was secured into the defect with a single absorbable suture in the overlying periosteum. The skin was then closed using interrupted absorbable sutures. All patients were casted for 2 weeks to allow for incorporation of the fat graft. After 2 weeks, the cast was removed, and the digit was mobilized.

Osteotomy

If an opening osteotomy was to be performed, the phalanx was approached similar to the physiolysis procedure. A 25-gauge needle was then inserted transversely into the joint above and below. A unicortical transverse osteotomy was performed from the radial to the ulnar direction with a sagittal saw, and the phalanx was straightened until the 2 needles were parallel, thus correcting the deformity to 0°. A single 0.035-inch Kirschner wire (K-wire) was then inserted retrograde across the osteotomy site. The patients were casted for 4 to 6 weeks, after which the K-wire was removed and the digit mobilized.

Statistical analysis

The data were summarized using means and standard deviations for continuous variables and counts and percentages for categorical variables. Outcomes comprising continuous variables were compared between the groups using Wilcoxon signed rank test, whereas categorical variables were evaluated using the Mann-Whitney U test. The Spearman correlation coefficient was used for correlation analysis. Because 9 of the 21 physiolysis patients had undergone bilateral surgery resulting in multiple records per patient, the analysis was conducted using generalized estimating equations in a generalized linear models framework to properly account for within-patient correlation. The groups were arbitrarily stratified based on preoperative angulation by determining the range of values and dividing the patients into thirds based on the range. All statistical tests were 2-sided and values of P < .05 were considered significant.

Results

We identified 26 patients (36 digits) who underwent the physiolysis procedure, of which 9 had undergone bilateral procedures. However, 5 of these patients (6 digits) did not have adequate radiographs and were excluded. There were 12 patients (19 digits) who underwent osteotomy for correction of clinodactyly, with 3 patients (4 digits) without adequate follow-up and 3 patients (4 digits) without adequate radiographs. Thus, included in this study were 21 patients (30 digits) who underwent the Vickers procedure and 6 patients (11 digits) who underwent osteotomy. Of the physiolysis patients, 14 (66.6%) had a documented diagnosis of an additional upper limb congenital abnormality, including syndactyly, polydactyly, brachydactyly, metacarpal synostosis, and radial and ulnar longitudinal deficiency (Table 1). There were 3 (50.0%) osteotomy patients with concomitant syndactyly (Table 1).

Table 1.

Summary Statistics of Vickers’ Physiolysis Versus Osteotomy for Clinodactyly Correction.

	Vickers’ physiolysis (n = 30 digits)	Osteotomy (n = 11 digits)	P value (2-tailed t test)
Age at operation, mo	42.2 (7 to 147, 28.9 SD)	79.0 (7 to 197, 56.5 SD)	.061a
Sex, No. (%)	Male: 18 (60.0)	Male: 5 (45.5)	.418
	Female: 12 (40.0)	Female: 6 (54.5)
Positive family history, No. (%)	8 (26.7)	2 (18.2)	.586
Side involved, No. (%)	Right: 13 (43.3)	Right: 6 (54.5)	.535
	Left: 17 (56.7)	Left: 5 (45.5)
Digit involved, No. (%)	D2: 7 (23.3)	D2: 2 (18.2)	.674
	D3: 2 (6.7)	D3: 2 (18.2)
	D4: 2 (6.7)	D4: 2 (18.2)
	D5: 19 (63.3)	D5: 5 (45.5)
Additional upper limb abnormality, No. (%)	Yes: 16 (53.3)	Yes: 5 (45)	.481
	No: 14 (46.7)	No: 6 (55)
Preoperative angulation, deg	43.0 (19.3 to 64.0, 10.7 SD)b	39.2 (14.5 to 71.8, 16.2 SD)c	.378
Postoperative angulation, deg	23.9 (4.0 to 59.0, 13.8 SD)b	22.4 (5.0 to 37.8, 11.0 SD)c	.736
Change in angulation, deg	19.0 (1.5 to 29.7, 8.6 SD)	16.7 (–6.7 to 60.8, 18.4 SD)	.696a
% correction	46.2 (2.5 to 85.7, 22.2 SD)	37.5 (–21.5 to 84.7, 33.9 SD)	.969
Duration of follow-up, mo	43.6 (14 to 114, 26.3 SD)	55.3 (10 to 125, 46.8 SD)	.450a

^aThe 2 groups did not have equal variance (Levine test P < .05); thus, significance was calculated based on unequal variances

^bWilcoxon signed rank test—significant decrease in angulation postoperatively (P < .005)

^cWilcoxon signed rank test—significant decrease in angulation postoperatively (P = .010)

The mean age at the time of clinodactyly correction was 42.2 months for the physiolysis group, with a mean duration of follow-up after physiolysis of 46.3 months. The mean age in the osteotomy group was 79.0 months, with a mean duration of follow-up of 55.3 months (Table 1). There was no statistically significant difference in age at time of surgery or follow-up between the Vickers and osteotomy groups (Table 1). The interpositional fat graft was harvested from the antecubital fossa in 18 patients (60%), the groin in 3 (10%), other digital incisions in 2 (6.7%), the plantar area in 2 (6.7%), and a multitude of other upper limb locations (hypothenar eminence, dorsal hand, defatted syndactyly flap, thenar eminence, palm, and ulnar forearm) in 1 each (3.3% each). An opening wedge osteotomy was performed in 10 of 11 digits (90.9%) and a closing wedge in 1 (9.1%) digit. Two osteotomy patients (3 digits) required a revision osteotomy at an average of 9 months after primary procedure, with 1 digit undergoing 3 revisions due to recurrent angulation. One revision was after the closing osteotomy and 2 were after opening osteotomies. During the period of follow-up, none of the patients who underwent the Vickers procedure had records indicating a secondary revision surgery for the diagnosis of clinodactyly after the primary physiolysis. One patient who underwent the Vickers procedure had a postoperative suture abscess that resolved with conservative management.

There was a significant improvement in angulation in the Vickers group, from 43.0° to 23.9°, at the last follow-up (P < .005). There was a mean absolute decrease of 19.0° and a 46.2% correction of the deformity. In the osteotomy group, there was a significant improvement in angulation from 39.2° to 22.4°. There was a mean change of 16.7° and a 37.5% correction of the deformity. There was no significant difference between the Vickers and osteotomy groups regarding absolute or percentage change of angulation (Table 1). There was a significant positive correlation between preoperative angulation and absolute change in angulation, as well as a negative correlation with preoperative angulation and age at time of surgery in all patients (Supplemental Figures S1 and S2). When separated, the osteotomy group correlated with preoperative angulation and absolute change in angulation, while the Vickers group did not.

The data were substratified to look at preoperative angulation versus total degrees of correction for each group. Three groups were formed based on preoperative angulation: ≤34°, 35° to 54°, and ≥55°. There was no effect on these groupings and absolute or percentage correction in all patients combined. Statistical significance was achieved in subgroup analysis for the Vickers patients when looking at the percentage of angular correction for the ≤34° and 35° to 54° groups compared with the ≥55° group (Table 2). Thus, a smaller percentage of correction was possible when there was more preoperative angulation using a Vickers physiolysis. There was a trend toward better correction in the ≥55° group compared with the less angulated digits in the osteotomy group, but the result was not significant.

Table 2.

Data Stratification by Preoperative Angulation for Vickers’ Physiolysis Versus Osteotomy.

Preoperative angulation, deg	Vickers’ physiolysis			Osteotomy
	No. (%)	Mean ∆ angulation, dega	Mean % correctionb	No. (%)	Mean ∆ angulation, degc	Mean % correctiond
≤34	6 (20.0)	18.8	60.0	6 (54.5)	7.6	28.2
35-54	18 (60.0)	21.2	49.8	3 (27.3)	16.8	36.4
≥55	6 (20.0)	12.9	22.0	2 (18.2)	44.0	66.8

^aKruskal-Wallis—no difference between stratified degree group for Vickers’ physiolysis (P = .187)

^bKruskal-Wallis—difference between stratified degree group for Vickers’ physiolysis (P = .006)

^cKruskal-Wallis—no difference between stratified degree group for osteotomy (P = .073)

^dKruskal-Wallis—no difference between stratified degree group for osteotomy (P = .360)

Age of operation was stratified into 3 distinct groups: ≤31 months, 31 to 45 months, and operation at greater than 45 months. In the physiolysis group, there was no difference in the absolute degree of correction or percentage of correction between groups (Supplemental Table 1). There was no difference in the degree or percentage of correction based on sex and no correlation of correction with age for Vickers’ physiolysis. The same was true for the osteotomy procedure, where age range had no effect on absolute or percentage change (Supplemental Table 1). Sex had no effect on outcome in the osteotomy group (P > .05).

We compared the degrees of correction in patients with isolated clinodactyly (n = 26) versus those with concurrent syndactyly (n = 15) (Table 3). There was no significant difference in preoperative angulation in the Vickers or the osteotomy group between the isolated clinodactyly and concurrent syndactyly subgroups. In the physiolysis patients, there was better absolute and percentage correction in patients with isolated clinodactyly compared with those with concomitant syndactyly (Table 3). The presence of syndactyly did not affect outcomes in the osteotomy group or when the groups were combined.

Table 3.

Data Stratification by Presence of Syndactyly for Vickers’ Physiolysis Versus Osteotomy.

Syndactyly	Vickers’ physiolysis					Osteotomy
	No. (%)	Preoperative angulation, deg	Postoperative angulation, deg	Mean ∆ angulation, dega	Mean % correctionb	No. (%)	Preoperative angulation, deg	Postoperative angulation, deg	Mean ∆ angulation, degc	Mean % correctiond
No	20 (66.7)	42.4	20.0	22.3	53.5	6 (54.5)	33.2	23.1	10.1	30.1
Yes	10 (33.3)	44.3	31.9	12.5	31.9	5 (45.5)	46.3	21.6	24.7	46.3

^aKruskal-Wallis—difference between groups for Vickers’ physiolysis (P = .008)

^bKruskal-Wallis—difference between groups for Vickers’ physiolysis (P = .039)

^cKruskal-Wallis—no difference between groups for osteotomy (P = .273)

^dKruskal-Wallis—no difference between groups for osteotomy (P = .273)

Discussion

Clinodactyly in the pediatric hand is a result of unbalanced longitudinal growth from a bracketed epiphysis. Based on the work of Vickers and others, it is clear that resection of the physeal tether allows for spontaneous correction of the radioulnar angulation over time (Figure 3). Vickers reported a 12-digit series of clinodactyly patients treated with physiolysis and fat grafting who achieved “excellent functional and cosmetic” results, but angular measurements were not reported.⁸ Caouette-Laberge et al⁶ published a large series of 35 fingers treated with physiolysis for clinodactyly with encouraging results, especially those treated earlier in life (<6 years of age) and those with more severe angulation (>40°). Medina et al¹⁴ published a series of 27 digits treated with physiolysis and fat grafting and found that there was an average correction of 79% at a mean follow-up of 6 months. None of these patients underwent a subsequent osteotomy later in life to correct any residual deformity.

Figure 3.

Radiographic (top row) and clinical results (bottom row) after the Vickers physiolysis treatment for clinodactyly of the left index finger: (a) preoperative, (b) 11 months postoperative, and (c) 30 months postoperative.

Physiolysis as primary treatment for the primary correction of clinodactyly offers many advantages. It is a straightforward, short operation that is performed in the outpatient setting. Unlike other osteotomy-based procedures, it does not require any bone grafting and does not require any type of percutaneous fixation which has intrinsic infection risks, although not seen in our study, and a longer duration of immobilization. For correction of angulation, we did not see a difference between the Vickers and osteotomy groups. There was no difference between absolute change in angulation or percentage of correction. Both groups had significant improvement in deformity postoperatively. There was a trend toward the osteotomy group presenting later, with a mean age at operation of 79.0 versus 42.2 months, but this was not significant.

Role of Age at Operative Intervention

Although physiolysis for the treatment of clinodactyly allows for gradual correction of the radioulnar angular deviation of the finger, it can result in incomplete correction of the initial deformity. Medina et al showed a mean correction of 82% of the preoperative angular deformity, with 11 out of 18 patients having full correction over a 6-year follow-up. In the series by both Vickers and Caouette-Laberge, 1 and 11 patients, respectively, required revision surgery after primary physiolysis.^6,8 Incomplete correction can result from 2 main issues—age at time of initial physiolysis and adequacy of bracketed epiphyseal resection. In the Medina et al. series, the average age of correction was 3.7 years, whereas in the Vickers series and the Caouette-Laberge series it was 9.2 and 6.6 years. Our mean age was 3.5 and 6.6 years in the Vickers and the osteotomy group, respectively. Caouette-Laberge showed that in patients who had undergone physiolysis under the age of 6, a greater correction of the starting deformity took place. We feel that early intervention prior to school age is of critical importance as it allows for correction of the deformity before the finger’s appearance becomes stigmatizing for the child. Also, earlier correction may lead to greater correction prior to physeal maturity. However, we did not show any correlation between age and outcomes, and there were no differences when comparing the Vickers or osteotomy correction at different age ranges in our study.

Incomplete Correction After Primary Physiolysis

The extent of resection of the bracketed epiphysis is of significant operative importance. Unlike prior descriptions of physiolysis, which removed only a small segment of physeal tether, we remove the entire physeal bar. The inherent danger in physiolysis procedures is if the dorsal and palmar extent of the bracket is not identified and erroneously removed, it results in incomplete correction and is directly correlated with poor long-term angular correction. Despite every effort to provide complete correction of clinodactyly through adequate bracket resection, there will be clinical scenarios where physiolysis does not result in complete correction of the deformity. Prior studies indicated that those cases having greater initial angular deformity had the greatest ability for angular correction after physiolysis. Our study demonstrates that the use of Vickers’ physiolysis in angulation over 55° may produce worse outcomes compared with lesser angulation deformities. In patients with significant angulation, osteotomy may be a better choice as there was a trend toward better correction in those with a preoperative angulation >55°. We do not advocate performing physiolysis for cases of incompletely corrected clinodactyly. For these patients, we recommend definitive corrective treatment with an opening or a closing wedge osteotomy once the patient reaches skeletal maturity.

Concomitant Clinodactyly and Syndactyly

The association of clinodactyly with other congenital anomalies has been well established. A particular link between clinodactyly and syndactyly was noted in this patient cohort, with 36.6% of patients having a concomitant clinodactyly and syndactyly (33.3% and 45.5% in the Vickers and the osteotomy group, respectively). Patients having clinodactyly associated with simple or complex syndactyly will often not show the true degree of angular deformity until the syndactyly is released. The syndactyly often masks the true angular deformity until the skin malformation is corrected (Figure 4). Statistically, we showed significantly greater correction in those patients with isolated clinodactyly compared with those with concomitant syndactyly in the Vickers group, but no difference in the osteotomy group. Despite this, we still perform physiolysis of digits having both deformities, as our clinical cases revealed favorable outcomes with functional and cosmetic improvement.

Figure 4.

Radiographic (top row) and clinical results (bottom row) after Vickers’ physiolysis and syndactyly release for a patient with concomitant clinodactyly of the little finger and syndactyly of the long, ring, and little finger: (a) pre–syndactyly release, (b) appearance post–syndactyly release and pre–clinodactyly correction, and (c) 34 months after Vickers’ physiolysis for correction of the little finger clinodactyly.

We recognize the limitations of this study, including its retrospective nature, small sample cohort, and lack of cosmetic outcomes. We were unable to include half of our osteotomy group due to incomplete follow-up or radiographs. Although our sample size is small, it is comparable in numbers to other seminal studies looking at outcomes of operative intervention for clinodactyly. Specifically, our study has the youngest average age of surgical physiolysis and an excellent follow-up duration. We were likely not able to show statistically that an osteotomy can achieve better correction in a more angulated digit due to the underpowered nature of the study. There was also a trend toward the osteotomy patients being older at the time of surgery, and it is possible that if an osteotomy was performed on younger patients, there may be a higher risk of recurrence without bar resection. The inability to show a statistical difference between the age difference between groups may be related to underpowering of the analysis. We do not have specific outcome data looking at patient or parental satisfaction with the cosmetic appearance of the treated digit. Other recent studies on congenital hand surgery have used a visual analog scale for rating the appearance of the correction.¹⁶ Future studies are planned to include these pertinent data.

Physiolysis and osteotomy are both effective solutions for the primary treatment of clinodactyly. The use of physiolysis may be more useful in those patients with less severe angulation due to its minimal morbidity, less need for revision, and less chance of complications. In those patients with angulation >55°, the use of an osteotomy may provide more degrees of correction. The use of physiolysis results in improved outcomes in patients with isolated single digit clinodactyly over those with concomitant syndactyly.