To Tie or Not to Tie: A Systematic Review of Postaxial Polydactyly and Outcomes of Suture Ligation Versus Surgical Excision

Abstract

Background: Ulnar polydactyly is frequently encountered in the newborn nursery and is commonly treated with bedside suture ligation. However, growing concern about the complications associated with suture ligation has led some practitioners to advocate for primary surgical excision instead. Thus, we set out to compare outcomes of suture ligation and surgical excision by systematic appraisal of the literature. Methods: Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, a systematic review was performed to identify studies published between 1950 and 2017 that described outcomes of suture ligation, surgical excision, or both. Baseline characteristics, complications, and study quality were extracted for each included article. Results: A total of 900 articles were reviewed, of which 10 studies (8 case series, 2 comparative analyses) met the inclusion criteria. There was considerable heterogeneity among the studies with respect to patient characteristics and reported outcomes. There were 2 retrospective case series of suture ligation that reported no acute complications and a variable proportion of patients with residual remnants or neuromas. Studies evaluating surgical ligation reported no acute or long-term complications, with only 1 case series reporting a small percentage of residual remnants. However, in the largest cohort analysis, the difference in complication rate was reported to be as high as 23.5% for suture ligation compared with 3% for surgical excision. Conclusions: There is a paucity of literature limiting the comparison of suture ligation and surgical excision for ulnar polydactyly. Further studies are required to determine the optimal treatment.

Introduction

Ulnar polydactyly is one of the most common congenital anomalies of the upper extremity and affects nearly 4 in every 10 000 live births per year.¹ Typical presentation consists of a nonfunctional digit attached by a skin bridge or a small bump on the ulnar aspect of the hand (Figure 1).^2-4 These rudimentary phenotypes may be accompanied by an accessory digital nerve. Pediatricians often treat these minor forms in the newborn nursery by tying off the base of the skin bridge with a nonabsorbable suture^5,6; however, reports of acute and long-term complications have led some to question the traditional practice of suture ligation (Figure 2).^7-10 In particular, the development of amputation neuromas is often cited by advocates of primary surgical excision.¹¹ When transected by the suture ligation technique, the functional digital nerve may not retract into the subcutaneous tissue, and as such, the nerve stump may be incorporated into the cutaneous scar and exposed to mechanical trauma. Surgical excision instead employs an elliptical excision at the base of the supernumerary digit, and dissection and high ligation of the accessory digital nerve, followed by closure of the wound (Figure 3). The treatment of complex polydactylies (involving articulation with the fifth or sixth metacarpal) is well understood, with the majority of patients requiring surgical intervention. Yet from a multidisciplinary standpoint, the initial management of rudimentary phenotypes remains controversial.¹² The purpose of this study is to delineate between treatment modalities for common types of ulnar polydactyly. Thus, we set out to compare outcomes of suture ligation and surgical excision by systematic appraisal of the literature.

Figure 1.

(a) Soft tissue nub/remnant (Rayan-Frey type I, Temtamy-McKusick type b, Stelling type I). (b) Pedunculated nonfunctioning digit (Rayan-Frey type II, Temtamy-McKusick type b, Stelling type I). (c) Well-formed functioning digit that is articulating with a bifid fifth metacarpal head (Rayan-Frey type III, Temtamy-McKusick type a, Stelling type II). (d) Well-formed digit that is fused at a right angle to the fifth metacarpal (Rayan-Frey type III, Temtamy-McKusick type II, Stelling type II). (e) Well-formed digit that has a hypoplastic/absent proximal phalanx (Rayan-Frey type III, Tamtamy-McKusick type A, Stelling type II). (f) Complete duplication with a separate sixth metacarpal (Rayan-Frey type IV, Temtamy-McKusick type a, Stelling type III).

Figure 2.

(a) Nonabsorbable suture tied to the base of the duplicated digit. (b) Duplicated digit becomes necrotic and autoamputates at the base.

Figure 3.

(a) Elliptical incision at the base of the duplicated digit, (b) excision of the duplicated digit and high ligation/excision of the accessory nerve, and (c) closure.

Materials and Methods

Study Selection

In accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, 3 reviewers (M.C., L.S., P.J.B.) conducted a systematic literature search within the MEDLINE database. All articles published between 1950 and 2017, with English as the primary language, were included in the search query. A combination of Boolean operators with the following MeSH terms were used to conduct the search: “polydactyly,” “supernumerary digit,” “excision,” “ligation,” “complication,” “management,” “treatment,” and “approach.” The 3 reviewers (M.C., L.S., P.J.B.) independently performed each step in the study selection process. Articles were screened based on title and abstract for ulnar polydactyly or supernumerary digits (upper extremity) as the topic of primary concern. Articles were selected for inclusion if they met the following criteria: (1) subjects with minor forms of ulnar polydactyly (upper extremity), such as Temtamy-McKusick type b, Stelling type I, and Rayan-Frey type I and II, or subjects with pedunculated supernumerary digits in which no subtype was specified; and (2) reported outcomes for either ligation or surgical excision of accessory digits. Furthermore, studies were included if greater than 50% of the subjects were identified with one or more of the aforementioned subtypes of ulnar polydactyly. Studies in which reconstruction was performed at the time of surgical excision were omitted. Manuscripts containing case reports, review articles, book chapters, editorials, or commentaries were also excluded.

Data Extraction

The 3 reviewers (M.C., L.S., P.J.B.) independently assessed the quality and methodology of each study meeting the inclusion criteria. These manuscripts were further analyzed with respect to patient sample size, interventions, and associated outcome/complications. Baseline data including age, sample size, type of polydactyly, number of digits involved, history of ligation, and follow-up period were collected. The study type and design were assigned to each article. The level of evidence for each article was assessed using the standardized reporting scheme provided by the Oxford Centre for Evidence-based Medicine (OCEBM). Apart from the biases inherent to study designs, additional sources of bias were identified independently by each reviewer. Outcomes of interest were restricted to acute or long-term complications. Acute complications included hematoma, infection, and incomplete detachment of digit. Long-term complications consisted of residual remnant (incomplete amputation/excision), neuromas, sensory or functional deficits, and the need for secondary interventions. Neuromas were defined as tender or painful scars/remnants that were reported by the parents of treated children. Satisfaction scores were also sought as an outcome of interest. All statistical analyses and graphical representations were conducted in R (R Core Team, New Zealand).¹³

Results

Systematic search of the MEDLINE database using the aforementioned keywords yielded 900 articles. After screening by title and abstract, 27 articles were assessed in full-text and 10 articles were eligible for inclusion in our analysis. A flow diagram is depicted in Figure 4. These 10 studies comprised 8 case series and 2 comparative analyses (1 prospective and 1 retrospective cohort analyses). Tables 1 and 2 summarize baseline characteristics and major outcomes for each included article. Table 3 outlines their study design, OCEBM level of evidence, and alternative sources of bias.

Figure 4.

Flow diagram showing the inclusion and exclusion of articles reviewed.

Table 1.

Baseline Characteristics of Included Articles.

Study	Polydactyly type	Study size	Digits	Age (range)	History of ligation (no. of subjects)	Follow-up time
Ligation
Mills et al14	Temtamy-McKusick type b	132	231	2-40 wk		2 wk, 14-142 mo
Watson and Hennrikus15	Temtamy-McKusick type b	21	37	Newborns	0	12-37 mo
Excision
Carpenter et al16	Temtamy-McKusick type b	26	38	9 d to 4.2 y	5
Katz and Linder17	Temtamy-McKusick type b	11	15	2-3 d	0	12 mo
Leber and Gosain11	Supernumerary	9	16	3 wk to 13 y	3	2.5-6 y
Mullick and Borschel18	Ulnar polydactyly	10	14	1 wk to 13 y	10	1-3 mo
Patillo and Rayan19	Ulnar polydactyly	2	3	2 wk to 3 y	2	1 wk
Singer et al20	Stelling and Turek type I (77%), type II (15%), type III (8%)	32	53	0.1-10 y		2.1-10 y
Comparative analyses
Rayan and Frey21	Rayan-Frey type I (0.7%), type II (81.7%), type III (15.5%), type IV (0.7%), type V (1.4%)	148			3
Samra et al22	Temtamy-McKusick type b	14	25	Newborns	0	1 and 3 mo

Table 2.

Major Outcomes of Included Articles.

Study	Outcomes
Ligation
Mills et al14	Painful, residual nubbins (7%, n = 16 extremities) Residual nubbins, no operative intervention (3.7%, n = 5 patients)
Watson and Hennrikus15	Small scar, barely noticeable (57%) Small bump (32%) Large bump (11%) Operative intervention required (3%)
Excision
Carpenter et al16	No complications
Katz and Linder17	No complications
Leber and Gosain11	No complications
Mullick and Borschel18	No complications
Patillo and Rayan19	No complications
Singer et al20	Flat scar (90%) Residual nubbins (10%)
Comparative analyses
Rayan and Frey21	Complication rate: 3% surgical excision vs 23.5% suture ligation Ligation group, total complications (n = 25):  Neuromas or residual bump (n = 17)  Infection (n = 7)  Operative intervention (n = 3)  Emergency department visits (n = 5)
Samra et al22	Higher levels of pain with suture ligation than surgical excision Local > general anesthesia Waiting 4 mo for excision (anesthesia) caused emotional distress Satisfaction scores (out of 10) to be 9.8 or higher in all groups

Table 3.

Study Quality of Included Articles.

Article	Type of study	Study design	Level of evidence	Additional sources of bias
Carpenter et al16	Case series	Retrospective chart review	4	Follow-up period not specified
Katz and Linder17	Case series	Prospective	4	—
Leber and Gosain11	Case series	Case reports provided	4	—
Mullick and Borschel18	Case series	Retrospective chart review	4	—
Patillo and Rayan19	Case series	Case reports provided	4	Follow-up period not specified for all patients
Singer et al20	Case series	Retrospective chart review	4	Mixed morphologies included in outcome analysis, questionnaire to assess long-term outcomes, incomplete patient follow-up
Mills et al14	Case series	Retrospective chart review	4	—
Watson and Hennrikus15	Case series	Prospective	4	Incomplete patient follow-up
Rayan and Frey21	Cohort analysis	Retrospective chart review	4	Mixed morphologies within treatment arms
Samra et al22	Cohort analysis	Prospective	4	Survey to asses long-term outcomes, incomplete patient follow-up

Note: Dash means no additional sources of bias were found.

Suture Ligation

One prospective case series exists within the current literature looking at the use of suture ligation as the primary treatment of ulnar polydactyly.¹⁵ Watson and Hennrikus identified 37 ulnar polydactylies (Temtamy-McKusick type b) in 21 newborns during a screening program. Digits were treated at birth with suture ligation and examined 1 to 2 weeks postprocedurally. A subset of patients (15 children with 28 digits) were reexamined at an average of 12 months. The authors found no functional deficits as the range of motion of all digits was equal to their untreated counterparts. One patient underwent a second procedure in the operating room to remove a blackened digit that had remained firmly attached. Nine fingers (32%) had a small residual bump, and 3 fingers (11%) had a large residual bump. Despite these findings, parents declined revisions and were content with the appearance of their children’s fingers.

Considered mechanistically similar to suture ligation, surgical clip application was also studied. One case series by Mills et al analyzed 132 newborns treated with surgical clip ligation for Temtamy-McKusick type b digits with a base width less than or equal to 6 mm.¹⁴ The average age of the patient undergoing clip application was 8 weeks and patients were followed up for 2 years. Although no acute wound complications were observed, parents of 9 patients (6.8%) requested secondary interventions of scar revision for 16 extremities (6.9%) due to painful, residual remnants. An additional 5 (3.7%) patients had residual remnants that did not solicit surgical intervention.

Surgical Excision

Six studies (case series) measured clinical outcomes of surgical excision only.^11,16-20 Five studies reported no acute or long-term complications following surgical excision.^11,16-19 Of these 5, the largest case series consisted of 26 patients (38 digits), and the longest follow-up time was 6 years. In a popular study, Leber and Gosain retrospectively analyzed surgical excision.¹¹ All patients were observed to have acceptable outcomes with no acute or long-term complications at 2-to 6-year follow-up. In yet another series, surgical excision under local anesthesia was utilized in 11 patients (15 digits) at 2 to 3 days of age. There were also no acute complications such as hematoma or infection, and at 1-year follow-up, all infants had full range of motion in their fingers with no residual remnants, and parents were satisfied with the results.

The remaining study by Singer et al reported few complications following excision.²⁰ They analyzed outcomes in 32 patients who underwent excision under general anesthesia. Postoperative survey (63.80% responded) revealed that 10% of patients had a residual remnant. While no functional deficits were reported, only 1 patient indicated neuroma formation as occasional pain associated with scarring. Nonetheless, the mean satisfaction score was 89 (range 10-100), which corresponded to good postoperative outcomes.

Comparative Studies

Two cohort studies analyzed various treatment modalities for ulnar polydactyly. In a prospective nonrandomized study, Samra et al offered parents of children with polydactylous digits no treatment, suture ligation, immediate excision under local anesthesia, or delayed excision at 4 months of age with general anesthesia. Parents were surveyed 1 and 3 months after the procedure was performed. A total of 14 newborns were enrolled in the study. Surgical excision with local anesthesia was the most popular (n = 10) treatment modality. The remaining patients underwent suture ligation (n = 2) or delayed excision under general anesthesia (n = 2). Follow-up was not completed for 1 child who underwent suture ligation. No residual deformities were appreciated throughout the groups. Higher levels of perceived pain during the procedure were associated with suture ligation than surgical intervention (local > general), and emotional distress was experienced by parents who had to endure 4 months before surgery with general anesthesia could be performed. Interestingly, the authors reported satisfaction scores (out of 10) of 9.8 or higher in all groups.²²

Rayan and Frey performed a retrospective chart review of 148 patients with ulnar polydactyly. Of the 135 patients whose treatment was documented, 108 subjects underwent ligation and 27 had surgical excision. The complication rate for suture ligation was found to be higher than that of surgical excision (23.5% vs 3%, respectively). Complications following suture ligation were related to either tender or unacceptable remnant (16.1%), infections (6.6%), and need for surgical intervention (2.8%). In contrast, complications involving surgical excision only occurred in 1 patient who developed a surgical site infection.²¹ Yet, their results should be interpreted with caution, as the difference in complication rates may be confounded by the inclusion of emergency department visits.

Discussion

Optimal management of noncomplex polydactylies has yet to be determined from a multidisciplinary perspective. Here, we presented a systematic review of the literature regarding 2 modalities of treatment, suture ligation and surgical excision, for common subtypes of ulnar-based accessory digits. After meeting specific criteria for inclusion, 10 studies (8 case series and 2 comparative studies) were analyzed with respect to various baseline characteristics and outcomes. Regarding the ligation technique, there were 2 retrospective case series that reported no acute complications and a variable proportion of patients with residual remnants or neuromas. Approximately 3% of cases solicited or required a secondary intervention. Alternatively, studies evaluating surgical ligation reported no acute or long-term complications, with only 1 case series reporting a small percentage of residual remnants.²⁰ A retrospective comparative analysis performed by Rayan and Frey observed an overall complication rate of 23.5% versus 3% for suture ligation and surgical excision, respectively.²¹ Interestingly, a large cohort analysis using the National Surgical Quality Improvement Program database revealed an acute complication rate of 2.2% for surgical reconstruction of supernumerary digits.²³ Given that reconstruction is a more complex procedure than excision, the true complication rate of surgical excision may be less than that indicated in our systematic review. Also, the use of local anesthesia makes surgical excision a convenient and inexpensive procedure.^8,16,24,25

Our analysis revealed a paucity of rigorous and comparative studies of suture ligation and surgical excision for ulnar polydactyly. Small samples and a high risk of selection and specialty bias limit the results of any case series. Apart from the biases inherent to nonrandomized study designs, additional sources of bias, such as the inclusion of mixed phenotypes, may have affected the observed rates of complications for a particular procedure. Multiple different surgeons with varying techniques and skill levels could have contributed to the variation seen in surgical results. Furthermore, our analysis is limited by the considerable heterogeneity in reported baseline characteristics and outcomes among the included articles. The nonstandardized measurement of certain outcomes, such as satisfaction scores or acceptability, also makes it difficult to draw comparisons. For these reasons, the synthesis of end points was impractical. Only 1 database was utilized for this investigation, and so it is plausible that articles relating to this topic were not included for analysis. Also, interrater reliability for study selection was not statistically analyzed.

The traditional practice of suture ligation is commonly advocated in pediatric textbooks, yet its frequency in clinical practice is unknown. There are also very few studies to measure its ablative efficacy and complications. Although amputation neuromas can be distressing, the prospective series by Watson and Hennrikus also suggests the majority of residual bumps may be nonsymptomatic.¹⁵ Samra et al observed satisfaction rates to be relatively equal among various treatment modalities.²² Interestingly, Mills et al demonstrated a low complication rate with the selective application of surgical clips.¹⁴ This suggests that suture ligation may be indicated for certain morphologies of accessory digits.

The lack of systematic investigation regarding suture ligation and surgical excision may reflect the positive clinical experiences of its practitioners. However, it also may indicate a lack of cross talk between the medical and surgical specialties. If left unexamined, we are no closer to the reality of the situation and further yet from optimizing the care of patients with accessory digits. Future research should address the shortcomings highlighted in this section with comparative and prospectively designed analyses.

Conclusions

Ulnar polydactyly is commonly encountered in the newborn nursery as either a floating appendage attached by a skin bridge or a small “nubbin” on the ulnar aspect of the hand. Treatment modalities include ligation with suture material or surgical excision. Systematic appraisal of the literature revealed a deficiency in high-quality outcome studies limiting a comparison of the techniques. Further research is needed in determining the optimal initial management of noncomplex ulnar polydactylies.