Review Article Clinics in Orthopedic Surgery 2011;3:172-177 • http://dx.doi.org/10.4055/cios.2011.3.3.172

Congenital Differences of the Upper Extremity: Classifi cation and Treatment Principles Moon Sang Chung, MD

Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul, Korea

For hand surgeons, the treatment of children with congenital differences of the upper extremity is challenging because of the diverse spectrum of conditions encountered, but the task is also rewarding because it provides surgeons with the opportunity to impact a child’s growth and development. An ideal classifi cation of congenital differences of the upper extremity would refl ect the full spectrum of morphologic abnormalities and encompass etiology, a guide to treatment, and provide prognoses. In this report, I review current classifi cation systems and discuss their contradictions and limitations. In addition, I present a modifi ed classifi ca- tion system and provide treatment principles. As our understanding of the etiology of congenital differences of the upper extrem- ity increases and as experience of treating diffi cult cases accumulates, even an ideal classifi cation system and optimal treatment strategies will undoubtedly continue to evolve. Keywords: Upper extremity, Congenital differences, Classifi cation, Treatment principles

One to two percent of newborns are born with congenital In this article, I review the current classifi cation sys- defects, and 10% of them have congenital differences of tems for congenital diff erences of the upper extremity, dis- the upper extremity.1,2) As congenital diff erences of the up- cuss the limitations of obtaining the goals mentioned, and per extremity are a significant challenge, the reconstruc- suggest a classifi cation system that was modifi ed aft er the tive surgeon has a unique opportunity to positively aff ect Swanson/International Federation of Societies for Surgery the child’s growth and development.3) of the Hand (IFSSH) classifi cation.6,7) I also present prin- Th e primary purpose of a classifi cation system is to ciples of treatment of congenital difference of the upper increase communication about the specific features of a extremity according to the suggested system. condition between physicians, and provide the basis for discussion and comparison of information regarding epi- CURRENT CLASSIFICATION SYSTEMS demiology and treatment results. Th erefore, an ideal clas- sifi cation should refl ect the full spectrum of morphologic Various classification systems are divided according to abnormalities and should be simple and logical for physi- their basis of foundation; descriptive, anatomic or topo- cians to remember and to use, and would also incorporate graphical, embryologic, teratologic sequencing, genetic, etiology, guide treatment, and provide prognosis,4,5) which and diff erent combinations of these. is still unavailable in congenital differences of the upper Descriptive classifi cation is based on fi ndings of de- extremity. formity, such as radial clubhand, which describes a radially deviated hand looking like a golf club, or camptodactyly, meaning a fl exed fi nger. Although this system is intuitive and commonly used as a diagnosis in clinical practices, it Received July 26, 2010; Accepted October 5, 2010 depends on the confusing Greek and Latin terminology Correspondence to: Moon Sang Chung, MD and has little scientifi c value. S Seoul Orthopedic Specialty Hospital, Metan-dong, Suwon 443-370, Korea A more developed system is the anatomic or topo- Tel: +82-31-225-7000, Fax: +82-31-213-2600 E-mail: [email protected] Copyright © 2011 by Th e Korean Orthopaedic Association Th is is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Clinics in Orthopedic Surgery • pISSN 2005-291X eISSN 2005-4408 173

Chung. Congenital Diff erences of the Upper Extremity Clinics in Orthopedic Surgery • Vol. 3, No. 3, 2011 • www.ecios.org graphical classifi cation based on anatomic fi ndings or the the separation of morphologically similar congenital dif- extent of involvement. is a kind of descriptive ferences and creating a major contradiction to accept intu- classifi cation for a fi nger deformity, but complete, incom- itively.5,14) Tonkin 14) points out that the diff erence between plete, simple or complex syndactyly is an anatomic clas- abnormal formation and abnormal induction appears to sification and indicates the extent of involvement. This be one of semantics. can guide treatment recommendations and sometimes Congenital differences can be classified according prognosis. Similarly, most sub-classifications for various to their severity of expression, which is called teratologic diagnostic conditions are based on anatomy, such as the sequence classifi cation. Although this classifi cation is not Wassel classifi cation for thumb which is based comprehensive of all anomalies, this system is oft en used on the level of thumb bifurcation.8) Frantz and O’Rahilly9,10) for classification for specific conditions, as the extent of further developed the anatomic or topographic classifi ca- pathology can determine function and guide treatment. tion and expanded the concept of intercalary defi ciencies. For example, teratologic sequencing of thumb hypoplasia Th ey divided limb defi ciencies into terminal and interca- as used in Blauth and Manske’s 5 category system classifi es lary, and each into transverse and paraxial. thumb hypoplasia according to the severity of hypoplasia The anatomic classification system was further de- and is helpful for directing the treatment (Table 2).15,16) veloped into embryologic classifi cation by Swanson et al.11) Finally, there can be a classification system based in 1964, which was based on the concept that anomalies on particular genetic or molecular abnormalities. For should be grouped according to the part that were aff ected instance, defects in the HOXD13 have been implicated during development. In the Swanson classification, each in several common congenital hand differences such as limb malformation is classifi ed according to the most pre- syndactyly and polydactyly.17) However, this type of classi- dominant anomaly and is placed into one of seven catego- fi cation may be cumbersome as the genetics are extremely ries (Table 1). Th is system was accepted by the American complex and involve multiple steps and interactions be- Society for Surgery of the Hand (ASSH), the IFSSH, and tween many genes and proteins, in addition to the interac- the International Society for Prosthetics and Orthotics tion with environmental factors.5) Progress in understand- (ISPO), and is now termed as the IFSSH classifi cation.6) ing of these complex genetic and molecular interactions Recently, the Japanese Society for Surgery of the may contribute to a better classifi cation scheme and guide Hand (JSSH) suggested a modifi cation of the IFSSH sys- potential genetic treatment options. tem, adding two groups; “Abnormal induction of rays” and 12) “Unclassifi able cases.” Th e category of “Abnormal induc- LIMITATIONS OF THE IFSSH CLASSIFICATION tion of rays” includes syndactyly, the central polydactyly- cleft hand-osseous syndactyly complex, and triphalangeal At present, the most widely accepted classification is the thumb. This concept was based on recent embryologic IFSSH classification. However, it has been criticized for studies supporting a common etiology for central poly- its inherent limitations because it attempts to incorporate actyly, syndactyly, and typical cleft hand.13) Although this etiology into morphologically-based classifi ciation,4,14) and makes sense from an embryologic standpoint, the new cat- is difficult to classify complex cases, especially with the egory separates central polydactyly from radial and ulnar polydactyly that are classifi ed as “Duplication,” resulting in Table 2. Classifi cation of the Hypoplastic Thumb15,16)

Table 1. The Classification System Developed by the International Type 1 Minimally shortened and narrowed structures Federation of Societies for Surgery of the Hand6,7) Type 2 Mild underdevelopment of all structures; short bones; small diameters; mild thenar muscles hypoplasia; unstable thumb MCP I. Failure of formation joint; narrow fi rst web space II. Failure of differentiation Type 3A Stable CMC joint; signifi cant decrease in the thumb size; severe intrinsic and extrinsic muscle hypoplasia; unstable MCP joint; III. Duplication narrow fi rst web space IV. Overgrowth Type 3B Type 3A with an unstable CMC joint V. Undergrowth Type 4 Pouce fl ottant; rudimentary thumb VI. Congenital constriction band syndrome Type 5 Complete aplasia of the thumb VII. Generalized skeletal abnormalities MCP: metacarpophalangeal, CMC: carpometacarpal. 174

Chung. Congenital Diff erences of the Upper Extremity Clinics in Orthopedic Surgery • Vol. 3, No. 3, 2011 • www.ecios.org

Table 3. The Author’s Classifi cation of Congenital Differences of the Upper Extremity

Control failure Formation failure Under-formation Longitudinal: radial/central/ulnar defi ciency, madelung deformity Transverse: congenital , constriction band syndrome Mixed: symbrachydactyly Over-formation Longitudinal: hyperphalangism Transverse: polydactyly, mirror hand Mixed: polydactyly with hyperphalangism, polydactyly with triphalangeal thumb Separation failure Under-separation Longitudinal: symphalangism Transverse: syndactyly, synostosis (radioulnar/metacarpal) Mixed: carpal coalition, camptodactyly, delta bone, Over-separation Longitudinal: congenital dislocation, congenital laxity Transverse: congenital radial head dislocation Growth failure Under-growth Longitudinal: Transverse: microdactyly Mixed: small hand Over-growth Longitudinal: macrodactyly (long) Transverse: macrodactyly (wide) Mixed: macrodactyly (typical), hemihypertrophy Structural failure Generalized Osteogenesis imperfecta, achondroplasia Localized Melorheostosis, congenital tumorous conditions complex spectrum of cleft hand and symbrachydactyly.18) suggesting that Swanson initially considered the etiologies Th e IFSSH classifi cation system separates “Failure of the two conditions as diff erent. However, in the current of formation” from “Failure of diff erentiation” on the ba- modification of the IFSSH classification, symbrachydac- sis of timing of the causative insult, but creates separate tyly is brought back to “Failure of formation” category,19,20) groups for “Duplication,” “Overgrowth,” and “Under- highlighting the fact that the IFSSH classifi cation is truly growth,” although duplication is a kind of failure of forma- morphological and not etiologically based.5) Concerning tion and “Overgrowth” and “Undergrowth” are examples the typical cleft hand, which is usually associated with of diff erentiation or development failure (failure of diff er- central polydactyly and syndactyly, the JSSH modifi cation entiation).14) tried to overcome the problem by introducing another cat- In addition, the “cleft hand” is difficult to explain egory of “Abnormal induction of rays,” but also has its own by the IFSSH system. Cleft hand can be divided into typi- contradictions as we noted before. cal and atypical. A typical cleft hand (central deficiency) is not associated with forearm anomalies unlike radial or THE AUTHOR’S CLASSIFICATION SYSTEM ulnar deficiencies, and is usually bilateral, familiar, and associated with polydactyly, syndactyly and cleft ing of the I suggest a new classifi cation for congenital diff erences of feet, while atypical cleft hand or symbrachydactyly is usu- the upper extremity, which is modified after the current ally unilateral, not hereditary, and not associated with foot IFSSH classification system (Table 3). This classification anomalies, suggesting diff erent etiologies for similar con- is also an embryologic classifi cation, and is based on the ditions. In the original Swanson/IFSSH classifi cation, typi- concept that an organ develops and matures through cal cleft hand was classifi ed under “Failure of formation” three distinct stages, which are formation, separation, and symbrachydactyly was classified as “Undergrowth,” and growth, and each stage can have two types of control 175

Chung. Congenital Diff erences of the Upper Extremity Clinics in Orthopedic Surgery • Vol. 3, No. 3, 2011 • www.ecios.org failure, which are early control failure (resulting in under- concepts. In my experience, most of the congenital diff er- formation, under-separation, or under-growth) and late ences could be categorized reliably with some reasonable control failure (resulting in over-formation, over-separa- explanations. tion, or over-growth). Th ese 6 control failures have three In addition, my classifi cation system can categorize morphological types, which are longitudinal, transverse, some disease entities, which have not been included in any and mixed. By having these morphological types for each classifi cation schemes previously, such as congenital laxity control failure category, my classifi cation system makes it of joint, which is not included in the IFSSH system, can be easy to intuitively classify complex deformities. classifi ed as “Transverse type over-separation.” Also, delta Concerning the “Generalized skeletal abnormali- bone is not classifi ed into any category in the IFSSH sys- ties” in the IFSSH system, my system classifi es the condi- tem, but it can be classifi ed as “Longitudinal type under- tions as “Structural failure” as opposed to “Control fail- separation,” because the anomalous orientation of the ure,” because those anomalies have abnormal tissue from growth plate causing coronal deviation is thought to be structural or cellular failures while conditions of “Control caused by incomplete separation of the epiphysis from the failure” have normal tissue quality. Therefore, congenital primary ossifi cation center. diff erences of the upper extremity can be fi rst divided into Lastly, my classification system can logically and “Control failure” (with normal tissue) and “Structural fail- intuitively guide the treatment principle for each category, ure” (with abnormal tissue), and then the “Control failure” such as “Under-formation” needs lengthening (or trans- can be further classified into “Formation failure,” “Sepa- plantation), and “Over-formation” needs resection (or ration failure,” and “Growth failure,” and the “Structural amputation); “Under-separation” requires separation (or failure” into “Generalized structural failure” and “Localized division), and “Over-separation” requires reduction (or structural failure.” arthroplasty or fusion); “Under-growth” needs augmenta- Th e most signifi cant diff erence of my classifi cation tion and “Over-growth” needs resection (or debulking). from the IFSSH system is that it divides the concept of “Differentiation” of the IFSSH system into “Separation” TREATMENT PRINCIPLES ACCORDING TO and “Growth.” Th is system has three large categories under SUGGESTED CLASSIFICATION SYSTEM “Control failure” and can avoid placing separate categories of “Overgrowth” and “Undergrowth” (embryologically Treatment of upper extremity congenital differences failure of differentiation) at the same level of hierarchy. should be initiated as soon as possible, because most Furthermore, by placing “over-formation” and “under- anomalies tend to change with time and the human body formation” under the large category of “Formation failure,” is considered to maintain skeletal remodeling power until this system removes the redundant category of “Duplica- two to three years of age. tion,” which is also embryologically a kind of formation failure. In addition, this classifi cation system removes con- Under-formation striction band syndrome from the 7 categories of the IF- Cases of under-formation require ‘lengthening or trans- SSH system and places it under the category of “Formation plantation,’ because even the most advanced medical tech- failure” (“transverse type under-formation”), because the nologies cannot create a simple nail plate. Accordingly, main pathology of distal tissue agenesis or coalescence is a even simple cases of under-formation cannot be cured vascular compromise at some point of intra-uterine organ satisfactorily. formation. For longitudinal under-formations, such as, radial Th e advantage of my classifi cation is that physicians deficiency (radial clubhand), realignment procedures can intuitively classify most of the congenital anomalies including osteotomy or tendon transfer are necessary to because the 3 stages (formation, separation, growth), the 3 improve function. Hypoplastic or absent thumbs can be types (longitudinal, transverse, and mixed) and the 2 pre- treated by pollicization (a type of transplantation from the fi xes (under-, over-) are anatomically or morphologically index finger), and recently bone lengthening and living straightforward, although the system is based on embryol- tissue transfer have been used to replace an absent bone, ogy. For example, the complex symbrachydactyly, which joint, or muscle. has mixed longitudinal and transverse deficiency by the In cases of transverse under-formation, such as, anatomical classifi cation, and can be simply classifi ed into congenital amputation, digit creation using an autograft “Mixed type under-formation” by my classifi cation system, or allograft can be attempted. Having created a small successfully including both anatomical and embryological digit, lengthening procedures can improve function and 176

Chung. Congenital Diff erences of the Upper Extremity Clinics in Orthopedic Surgery • Vol. 3, No. 3, 2011 • www.ecios.org cosmesis. Autogenous bone grafting has the associated When reduction is impossible or too late, arthroplasty or limitations of donor site morbidities and of a lack of avail- fusion can improve function. able bone. Toe transplantation or proximal fi bular trans- plantation can create a digit with growth potential, but the Under-growth techniques involved are demanding and have a measurable Under-growth is when a portion of an extremity is under- risk of failure. Allograft s introduce problems of rejection, sized but near normal functionally, and requires ‘augmen- and these patients may require lifelong immunosuppres- tation.’ However, it is not possible to increase the size of the sive therapy. aff ected hand to the normal level, because we are limited to increasing only its length. When the required length is Over-formation within 1-2 cm, osteotomy and intraoperative distraction Over-formation can oft en be treated satisfactorily by sim- with interpositional strut bone graft ing and internal fi xa- ply removing the redundant tissue (‘resection or amputa- tion provides a straightforward option, and when the re- tion’). Sometimes additional procedures are necessary, quired length exceeds 2 cm, external fi xation and gradual such as, tendon realignment, corrective osteotomy to lengthening can be performed. realign residual digit, resection arthrodesis to reduce an extra-bone and joint in cases of hyperphalangism, or com- Over-growth bining two small thumbs to create a balanced, normally Overgrowth needs ‘reduction or debulking.’ The affected sized thumb. regions include both normal and abnormal tissues, such as, those affected by neurofibromatosis, hemangioma, Under-separation lymphangioma, or arteriovenous malformation. Some- Conditions of under-separation, such as symphalangism times bones and joints are deformed, and muscles and (longitudinal), syndactyly (transverse), and carpal coali- tendons are abnormal. The treatment goal is to decrease tion (mixed) require ‘separation,’ but these conditions are length and volume to obtain levels of function and appear- usually combined with incomplete or deformed bones, ance that are similar to those of normal side. However, joints, and muscles that need additional procedures. decreasing size eff ectively is diffi cult and postoperative tis- Coalition can be defined as a condition whereby sue necrosis is common, and thus, in most cases results are an under-developed joint becomes a syndesmotic or syn- less than satisfactory. chondrosic joint rather than a synovial joint. Because the creation of a normal synovial joint is not possible, sepa- Structural Failure ration means making a pseudarthrosis to gain mobility. The primary goals of the treatment of generalized struc- However, such newly created pseudarthroses have prob- tural failure are to extend life expectancy and to prevent lems of reunion, instability, and inadequate motion, since progression of the anomaly. In cases with a hormone or soft tissues around these joints are also inadequate. Some- enzyme deficiency, hormone or enzyme replacement times, osteotomy is necessary to obtain a more functional therapy can achieve the treatment goal. However, in most position. cases, the causes are unknown and medical treatment is In cases with under-separations of soft tissues, such rarely eff ective. If the anomaly is severe enough to impair as, those encountered in camptodactyly, causative tissues walking or daily activities, a child’s quality of life can be (abnormal skin, pulley, fascia, or tendon) that are not criti- improved by treatments, such as, bracing, cast immobili- cally required for function must be removed. zation, or surgery, and in cases with a localized structural failure, correction of the localized anomaly alone may Over-separation achieve its long-term correction. However, the molecular Conditions of over-separation, such as congenital disloca- events that underlie such abnormalities have not been elu- tion or laxity, are treated by ‘reduction or stabilization.’ cidated. Th e general treatment principle for congenital dislocation requires joint reduction by a closed or open method and CONCLUSIONS the maintenance of reduction until the remodeled joint becomes stable. When a joint has dysplasia or subluxation, I have reviewed the current classifi cation systems for con- attempts are made to make a more stable joint by intra-ar- genital differences of the upper extremity, discussed the ticular or per-articular osteotomy, as the joint tends to de- contradictions and limitations of the IFSSH classifi cation, velop early osteoarthritis or become completely dislocated. and presented a modified classification system. In addi- 177

Chung. Congenital Diff erences of the Upper Extremity Clinics in Orthopedic Surgery • Vol. 3, No. 3, 2011 • www.ecios.org tion, I provided treatment principles according to the sug- gested classifi cation system. As our understanding of the CONFLICT OF INTEREST etiology of congenital differences of the upper extremity increase and as experience of treating difficult cases ac- No potential confl ict of interest relevant to this article was cumulates, even an ideal classifi cation system and optimal reported. treatment strategies will undoubtedly continue to evolve.

REFERENCES

1. Lamb DW, Wynne-Davies R, Soto L. An estimate of the Clin North Am. 1968;48(5):1169-79. population frequency of congenital malformations of the 12. De Smet L; IFSSH. International Federation for Societies upper limb. J Hand Surg Am. 1982;7(6):557-62. for Surgery of the Hand JSSH. Japanese Society for Surgery 2. Kozin SH. Upper-extremity congenital anomalies. J Bone of the Hand. Classifi cation for congenital anomalies of the Joint Surg Am. 2003;85(8):1564-76. hand: the IFSSH classification and the JSSH modification. Genet Couns. 2002;13(3):331-8. 3. Bates SJ, Hansen SL, Jones NF. Reconstruction of congenital differences of the hand. Plast Reconstr Surg. 2009;124(1 13. Ogino T. Teratogenic relationship between polydactyly, syn- Suppl):128e-143e. and cleft hand. J Hand Surg Br. 1990;15(2):201-9.

4. Manske PR, Oberg KC. Classification and developmental 14. Tonkin MA. Description of congenital hand anomalies: a biology of congenital anomalies of the hand and upper ex- personal view. J Hand Surg Br. 2006;31(5):489-97. tremity. J Bone Joint Surg Am. 2009;91 Suppl 4:3-18. 15. Blauth W. Th e hypoplastic thumb. Arch Orthop Unfallchir. 5. Sammer DM, Chung KC. Congenital hand diff erences: em- 1967;62(3):225-46. bryology and classifi cation. Hand Clin. 2009;25(2):151-6. 16. Manske PR, McCarroll HR Jr, James M. Type III-A hypo- 6. Swanson AB. A classifi cation for congenital limb malforma- plastic thumb. J Hand Surg Am. 1995;20(2):246-53. tions. J Hand Surg Am. 1976;1(1):8-22. 17. Muragaki Y, Mundlos S, Upton J, Olsen BR. Altered growth 7. Swanson AB, Swanson GD, Tada K. A classifi cation for con- and branching patterns in synpolydactyly caused by muta- genital limb malformation. J Hand Surg Am. 1983;8(5 Pt tions in HOXD13. Science. 1996;272(5261):548-51. 2):693-702. 18. Cheng JC, Chow SK, Leung PC. Classifi cation of 578 cases 8. Wassel HD. The results of surgery for polydactyly of the of congenital upper limb anomalies with the IFSSH system: thumb: a review. Clin Orthop Relat Res. 1969;64:175-93. a 10 years' experience. J Hand Surg Am. 1987;12(6):1055- 60. 9. Frantz CH, O’Rahilly R. Congenital skeletal limb defi cien- cies. J Bone Joint Surg Am. 1961;43(8):1202-24. 19. Miura T, Nakamura R, Horii E. Th e position of symbrachy- dactyly in the classifi cation of congenital hand anomalies. J 10. O'Rahilly R. Morphological patterns in limb deficiencies Hand Surg Br. 1994;19(3):350-4. and duplications. Am J Anat. 1951;89(2):135-93. 20. Manske PR. Symbrachydactyly instead of atypical cleft hand. 11. Swanson AB, Barsky AJ, Entin MA. Classification of limb Plast Reconstr Surg. 1993;91(1):196. malformations on the basis of embryological failures. Surg