The two study outcomes are (1) a numeric visual analogue scale score and (2) a categorical grading scale rating. To examine the interrater reliability among the 18 raters, we used kappa statistics for categorical grading. We divided the 18 raters into four groups—laypersons, international hand surgery fellows, occupational therapists, and surgeons—and calculated interrater reliability within each group and all 18 raters using Fleiss’ kappa. We then examined interrater reliability between two raters among the 18 raters using the Cohen kappa. We calculated the intraclass correlation coefficient for visual analogue scale among each group for the interrater reliability. For the intrarater reliability, we calculated kappa statistics for grading and the intraclass correlation coefficient for visual analogue scale for each of the 18 raters.
The primary predictor of the study cohort was the type of flap reconstruction (i.e., dorsal rectangular or dorsal pentagonal flap). We performed descriptive analyses on our study cohort for the distribution of each patient characteristic by type of flap received. We performed Fisher’s exact test for the association between flap type and categorical patient variables (i.e., type of syndactyly, patient age, patient sex, right versus left hand, type of rater, web space, and revision) and Wilcoxon rank sum test for continuous patient variables (i.e., patient age and number of years of follow-up).
To examine the correlation between visual analogue scale and each patient characteristic, we performed simple linear regression models for each patient characteristic as the predictor. Given that we had grading by 18 raters, we added a random intercept for each type of flap in each model to control for the intercorrelation of repeated measures. We had a total of 810 data points because each of our 18 raters reviewed 45 patients total (n = 18 × 45 = 810).
To control for patient characteristics that significantly influenced the visual analogue scale score, we then performed a multivariable linear regression model with a random intercept for each surgical reconstruction using the type of flap as the primary predictor. We used all characteristics with significant associations with the visual analogue scale score selected from our simple linear regression models as the covariates into the multivariable linear regression model; the final multivariable models were established by maximum likelihood. We reported adjusted mean visual analogue scale scores and 95 percent confidence intervals for each patient group. Using the same strategy, we performed univariate analysis on patient characteristics and the binary outcome excellent versus not excellent (which includes very good, good, or poor) using a simple logistic regression model with random effect on each type of flap reconstruction. We used binary outcomes to ensure less stratification of our results and more power. We then performed multivariable logistic regression with random effect using type of flap as the primary predictor while controlling for those patient characteristics that had significant associations with each grade. We reported odds ratios of being excellent for each patient characteristic (compared to the reference group) and 95 percent confidence intervals.
A total of 45 patients were reviewed: 13 female and 32 male patients (Table 1). Patients who had dorsal pentagonal flap reconstruction were evaluated at a longer duration of follow-up compared to those patients who had dorsal rectangular flaps (1.6 years versus 2.3 years; p = 0.03). There was no significant difference in the distribution of other patient characteristics between the two different types of flap reconstruction. Sixteen patients underwent dorsal pentagonal flap reconstruction and 29 patients underwent dorsal rectangular flap reconstruction with full-thickness skin grafting. The majority of patients across both flap types had simple complete syndactyly. The third web space was the most commonly affected web space between both surgical groups. Twenty-five percent of patients required revision surgery after pentagonal flap surgery, whereas only 10 percent required revision surgery after rectangular flap reconstruction.
For grading between excellent and not excellent, the Fleiss kappa value among the four groups of raters ranged from 0.40 to 0.42; the Fleiss kappa value for all 18 raters was 0.38 (p < 0.001). The Cohen kappa value when comparing two raters’ grades ranged from 0.07 to 0.69, with significance ranging from p < 0.001 to p = 0.433. The intraclass correlation coefficient among our four groups of raters ranged from 0.52 to 0.65; the intraclass correlation coefficient for all 18 raters was 0.56 (p < 0.001). For grading between excellent and not excellent, kappa values among all of raters ranged from 0.16 to 0.76. The intraclass correlation coefficient among all of our raters ranged from 0.40 to 0.91. These values indicate that the raters generally agreed across all ratings.
Overall, the affected web space, patient age at follow-up, and length of follow-up were significantly associated with mean visual analogue scale values (Table 2). The mean visual analogue scale score among patients who underwent reconstruction of the fourth web space was 1.64 compared with 3.36 among those who underwent reconstruction of the second web space (p = 0.012), regardless of flap type. With 1 more year of follow-up, the average visual analogue scale score decreased by 0.36 (p = 0.003) across both flap types. Four patients with pentagonal flaps required revision surgery; three patients had web creep and one patient had a scar contracture. Three patients with rectangular flaps required revision surgery; two patients had scar contractures and one patient had web creep.
After controlling for the significant variables found in our univariate analysis (i.e., affected web space, patient age, and length of follow-up) (Table 2), multivariable analysis revealed that patients who underwent dorsal rectangular flap reconstruction had significantly lower visual analogue scale values, indicating better results, than patients who underwent dorsal pentagonal flap reconstruction (3.12 versus 4.24; p = 0.025) (Table 3).
Table 4 shows that, overall, the odds of getting an excellent rating were significantly associated with web space and number of years of follow-up. For example, the odds ratio of getting an excellent rating among patients who had second-web-space reconstruction compared to patients who had fourth-web-space reconstruction was 0.06 (95 percent CI, 0.01 to 0.59). With 1 more year of follow-up, the odds of getting an excellent rating increased 1.85 times (95 percent CI, 1.21 to 2.84 times).
Again, after controlling for significant variables (i.e., affected web space and length of follow-up) (Table 4), patients who underwent dorsal pentagonal flap reconstruction had significantly lower odds of getting an excellent rating compared with patients who underwent dorsal rectangular flap surgery (OR, 0.20; 95 percent CI, 0.05 to 0.83; p = 0.032) (Table 5).
To date, no studies have compared long-term outcomes between skin graft and non–skin graft techniques in syndactyly reconstruction.6 , 12 Both proximally based dorsal rectangular flaps and dorsal pentagonal advancement flaps have been successfully used at our institution. In our study, we compared the long-term aesthetic outcomes between these two forms of web-space reconstruction. Our study shows that rectangular flaps with groin skin grafting result in better long-term aesthetic outcomes compared with pentagonal flaps.
Numerous methods of syndactyly reconstruction have been proposed. All methods use a common set of techniques that any surgeon must be comfortable with before attempting a repair: use of full-thickness flaps to resurface the web space, preservation of vascular supply to the digit, and a meticulous approach to reconstruction with attention to preserving anatomical proportions and details.2 , 3 Achieving aesthetically pleasing fingers is likely just as important to a child as achieving fully functional digits. Traditional methods of syndactyly reconstruction use full-thickness skin grafts, often from the groin region, to facilitate a tension-free closure between separated digits.2 , 3 , 5 However, these techniques require postoperative immobilization and bandaging, which can be cumbersome in a young active child, and can lead to scar contracture or graft failure.2 , 3 , 5 , 16 Moreover, case series have reported high incidences of hair growth in the grafted areas, resulting in revision of the grafted area, epilation, and trimming of the excessive skin. Recently, there has been a trend toward reconstruction without the use of skin grafts. However, extensive defatting of the interdigital space is often required, which can lead to nerve injury, venous congestion, and a withered finger appearance.2 , 3 , 5 , 16 , 17
From our review of the patients, the main reason patients underwent revision surgery after pentagonal flap reconstruction was web creep, although scar contracture was the impetus for revision surgery after rectangular flap reconstruction. A recent systematic review by Sullivan and Adkinson revealed that techniques using skin grafts resulted in higher rates of web creep and revision surgery compared with those using only dorsal advancement flaps. In their review, they mention that revision rates may not be a valid comparative measure, however, for several reasons, such as surgeon threshold, patient preference, or even change of surgeon.12 Thus, our revision results may be underestimated if patients were not willing to undergo reoperation (e.g., they were satisfied with their results even with a certain level of web creep or scar contracture).
This investigation has several limitations. First, the use of photographs as the only means of evaluating aesthetic outcomes is limited by the quality of the photographs. The distinction between different grades was made primarily based on color mismatch and skin distortion; matte was found to be difficult to assess in photographs. However, we did maintain consistency across all photographs used, as they were all taken with standardized views, backgrounds, camera settings, and lighting. In addition, even though our classification systems were based on distinct criteria, the evaluation of these criteria (e.g., skin distortion and matte) is completely subjective. Second, although our total sample size was sufficiently powered, we were not able to stratify our patients into syndactyly type because of small respective sample sizes. Third, our study focused only on aesthetic outcomes, as we felt that the majority of our patients would be too young to fully participate in functional testing (e.g., strength tests). In addition, future studies should incorporate patient-reported outcomes if possible, which would provide further insight into patient satisfaction and effects on daily life activities. Lastly, we did not use a standardized follow-up period at the time of photographic assessment. Patients with longer follow-up periods at the time of their photograph may have better aesthetic outcomes solely because their scars have had a longer time to heal and remodel.
Our study is unique in that it offers long-term comparative outcomes between syndactyly reconstruction with skin grafts and dorsal advancement flaps without skin grafts. The majority of previous studies are case series that only report outcomes of particular flap techniques, with the majority among patients with simple syndactyly.6 , 12 We found that our classification system was easy to use even by nonsurgeon observers, who are usually unaware of the details of different syndactyly release procedures. Interestingly, our results also showed that fourth-web-space reconstructions fared better in both classification systems, particularly when compared against second-web-space reconstructions, regardless of flap technique. These findings may be attributable to the smaller surface area of the fourth web space, thus requiring less graft for coverage, less manipulation of the graft (e.g., defatting), and a more tension-free closure. Moreover, our classification system properly captures the principle that as scars continue to heal, their appearance becomes more normalized compared with surrounding tissue.
The results from our study can serve as important discussion points when counseling patients on the types of syndactyly reconstruction available and adequately managing their postoperative expectations. Specifically, despite the fact that rectangular flaps use skin grafting with some donor site morbidity, overall outcomes may still be better than with advancement flaps. In addition, further attention should be made toward the technical aspects of these operations, particularly the liberal use of skin grafts to accomplish tension-free closure and minimize the risk of web creep. Future prospective comparative studies are needed that incorporate both functional and subjective criteria, with collection of both preoperative and postoperative data.
This publication was supported in part by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (2 K24-AR053120-06). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
1. Tonkin MA. Failure of differentiation part I: Syndactyly. Hand Clin. 2009;25:171–193.
2. Oda T, Pushman AG, Chung KC. Treatment of common congenital hand conditions. Plast Reconstr Surg. 2010;126:121e–133e.
3. Goldfarb CA, Steffen JA, Stutz CM. Complex syndactyly: Aesthetic and objective outcomes. J Hand Surg Am. 2012;37:2068–2073.
4. Malik S. Syndactyly: Phenotypes, genetics and current classification. Eur J Hum Genet. 2012;20:817–824.
5. Lumenta DB, Kitzinger HB, Beck H, Frey M. Long-term outcomes of web creep, scar quality, and function after simple syndactyly surgical treatment. J Hand Surg Am. 2010;35:1323–1329.
6. Goldfarb CA. Congenital hand anomalies: A review of the literature, 2009-2012. J Hand Surg Am. 2013;38:1854–1859.
7. Tuma P Jr, Arrunategui G, Wada A, Friedhofer H, Ferreira MC. Rectangular flaps technique for treatment of congenital hand syndactyly. Rev Hosp Clin Fac Med Sao Paulo 1999;54:107–110.
8. Yildirim C, Sentürk S, Keklikçi K, Akmaz I. Correction of syndactyly using a dorsal separated V-Y advancement flap and a volar triangular flap in adults. Ann Plast Surg. 2011;67:357–363.
9. Jose RM, Timoney N, Vidyadharan R, Lester R. Syndactyly correction: An aesthetic reconstruction. J Hand Surg Eur Vol. 2010;35:446–450.
10. Wafa AM. Hourglass dorsal metacarpal island flap: A new design for syndactylized web reconstruction. J Hand Surg Am. 2008;33:905–908.
11. Gao W, Yan H, Zhang F, et al. Dorsal pentagonal local flap: A new technique of web reconstruction for syndactyly without skin graft. Aesthetic Plast Surg. 2011;35:530–537.
12. Sullivan MA, Adkinson JM. A systematic review and comparison of outcomes following simple syndactyly reconstruction with skin grafts or a dorsal metacarpal advancement flap. J Hand Surg Am. 2017;42:34–40.e6.
13. Beausang E, Floyd H, Dunn KW, Orton CI, Ferguson MW. A new quantitative scale for clinical scar assessment. Plast Reconstr Surg. 1998;102:1954–1961.
14. Durani P, McGrouther DA, Ferguson MW. Current scales for assessing human scarring: A review. J Plast Reconstr Aesthet Surg. 2009;62:713–720.
15. Vercelli S, Ferriero G, Sartorio F, Stissi V, Franchignoni F. How to assess postsurgical scars: A review of outcome measures. Disabil Rehabil. 2009;31:2055–2063.
16. Dao KD, Shin AY, Billings A, Oberg KC, Wood VE. Surgical treatment of congenital syndactyly of the hand. J Am Acad Orthop Surg. 2004;12:39–48.
17. Bates SJ, Hansen SL, Jones NF. Reconstruction of congenital differences of the hand. Plast Reconstr Surg. 2009;124(Suppl):128e–143e.
Coding perspective provided by Dr. Raymond Janevicius is intended to provide coding guidance.
- 26560 Repair of syndactyly (web finger) each web space; with skin flaps
- 26561 Repair of syndactyly (web finger) each web space; with skin flaps and grafts
- 26562 Repair of syndactyly (web finger) each web space; complex (e.g., involving bone, nails)
- Repair of a finger syndactyly using skin flaps only is reported with code 26560. This includes elevation and transposition of the flaps, as well as dressings, initial splints, and 90 days of uncomplicated postoperative care.
- If skin flaps and skin grafts are used for syndactyly repair, code 26561 is reported. This code includes elevation and transposition of the flaps, harvest and placement of skin grafts, as well as dressings, initial splints, and 90 days of uncomplicated postoperative care.
- Each web space reconstructed is separately reported. Thus, the release and reconstruction of syndactyly of the index-middle web space and syndactyly of the ring-small web space, each with skin flaps and skin grafts, is reported with two codes:
- 26561 Index-middle web syndactyly reconstruction
- 26561-59 Ring-small web syndactyly reconstruction
- Complex syndactyly involves bone and/or nails. The reconstruction of complex syndactyly is reported with code 26562.
CODING PRINCIPLE: The syndactyly reconstruction codes are global and include elevation and transposition of skin flaps as well as skin grafts, including closure of donor sites. It is not appropriate to report codes 14040 (adjacent tissue transfer) or 15240 (full thickness skin graft) in addition to codes 26560, 26561, and 26562.
Disclosure: Dr. Janevicius (email@example.com) is the president of JCC, a firm specializing in coding consulting services for surgeons, government agencies, attorneys, and other entities.©2018American Society of Plastic Surgeons