All extremities had an abnormal hand. Two had a relatively normal thumb with two additional, abnormal digits. Three extremities had a hypoplastic thumb and either two or three abnormal digits. The other fifteen extremities had an absent thumb with an array of hand abnormalities. Syndactyly was common, and motion of the digits was abnormal.
Eleven of the seventeen patients were born after an uneventful gestation and with an uneventful delivery. The mothers of the other six patients had major intrauterine events. One had gynecological surgery at five weeks of gestation (when it was not known that she was pregnant), one had an unknown illness requiring hospitalization during the first trimester, one had an upper respiratory viral infection in the first trimester and a history of marijuana use during pregnancy, one had toxemia of pregnancy, one used multiple medications (prescription and illegal) in the first trimester, and one had a “significant illness” in the first trimester. One infant was born prematurely at thirty-two weeks, but there were no delivery-related or adverse perinatal events.
Associated musculoskeletal abnormalities were common. Four patients had scoliosis, three had bilateral proximal femoral focal deficiency, three had bilateral fibular hemimelia, one had associated palate abnormalities, and one had Klippel-Feil syndrome. The sibling of one patient had amelia bilaterally.
Severe Combined Dysplasia
Seven extremities in seven patients were classified as having type-A severe combined dysplasia. There were four female and three male patients. Four of the seven patients had an abnormality of the contralateral extremity: two patients had a hypoplastic extremity with all segments present, one had amelia, and one had a type-B severe combined dysplasia.
The patients commonly had a relatively normal humeral segment with an essentially absent forearm segment and an abnormal hand (Fig. 5). The shoulder was normal in five patients, and two patients had a hypoplastic glenoid. All seven hands had both thumb and finger abnormalities: one had four fingers and a hypoplastic thumb, two had three fingers and an absent thumb, and four had two fingers and an absent thumb. All patients had shoulder and wrist motion.
One patient was adopted (family history unknown). One mother had a urinary tract infection treated with antibiotics during pregnancy, and another mother had preeclampsia. Delivery was uneventful for all patients. Three patients had congenital cardiac abnormalities, and two patients had an associated musculoskeletal abnormality: one had proximal femoral focal deficiency and the other, amelia. There was no family history of limb abnormalities.
Four extremities in four patients (two female and two male) were classified as having type-B severe combined dysplasia. All had complete absence of the humerus and forearm segments as well as major hand abnormalities, with either one or two digits (Fig. 6). The contralateral extremity was abnormal in all patients: two contralateral extremities had a proximal ulnar longitudinal dysplasia, one had a Bayne7 type-II ulnar longitudinal dysplasia, and one had a type-A severe combined dysplasia.
Only one patient was known to have been born after an abnormal gestation: the mother was exposed to a variety of legal and illegal medications during the first trimester. All four deliveries were uneventful. One patient had a congenital cardiac anomaly, one had a cleft palate, one had bilateral proximal femoral focal deficiency, and one had bilateral fibular hemimelia. There was no family history of limb abnormalities.
Traditionally, major limb deficiencies have been classified as either transverse or longitudinal and as either terminal or intercalary according to a schematic rather than a developmental concept5,9. Genetic and developmental biological research has provided us with a better understanding of early limb development and serves as the basis for an alternative approach to classification5,9-19.
There are three longitudinal axes of formation: proximodistal, anteroposterior, and dorsoventral. The apical ectodermal ridge controls proximodistal development through fibroblast growth factors. Proximodistal limb arrests are probably caused by abnormalities of the apical ectodermal ridge. The zone of polarizing activity controls the anteroposterior axis, subsequently known as the pre/post axis or the radial/ulnar axis, with progressive fetal development through a morphogen, sonic hedgehog. The apical ectodermal ridge and the zone of polarizing activity are closely linked by diffusion through a feedback loop13,14,18-21.
While the important roles of the apical ectodermal ridge and the zone of polarizing activity in limb development have become better understood, there remain many unanswered questions concerning the teratologic events leading to limb deficiencies. Ogino and Kato16 demonstrated that radial and ulnar longitudinal deficiencies can be induced in rats with administration of Myleran (busulfan). The timing of the Myleran administration determined the type of longitudinal dysplasia: administration at nine to ten days of gestation produced ulnar deficiency, and administration at ten to eleven days of gestation, radial deficiency. The severity of the deficiency was related to the dosage of the Myleran15,16. In another study, exposure of rats to the cytotoxic agent Adriamycin (doxorubicin) at approximately eight days of gestation induced esophageal, tracheal, cardiovascular, vertebral, and limb abnormalities10. Most notably for this discussion, Adriamycin led to limb malformation in 61% of animals with esophageal atresias and involved the humerus in addition to the radius and ulna. Bilateral deficiencies were common.
Our investigation demonstrated that so-called intercalated deficiencies, which are difficult to explain from a developmental biology standpoint, may represent forms of longitudinal deficiencies. The severity of the longitudinal dysplasia is likely related to both the timing and the severity of an insult to the zone of polarizing activity and/or the apical ectodermal ridge.
In 1961, Frantz and O'Rahilly classified phocomelia according to several different patterns of transverse, intercalary segmental dysplasia1. In 1976, Swanson summarized the classification adopted by the American Society for Surgery of the Hand and the International Federation of Societies for Surgery of the Hand5 and questioned the existence of intercalary deformities; the report concluded that “true intercalary deficiencies rarely, if ever, existed” as “all `phocomelias'... have some terminal manifestations.” It was suggested that the intercalary deficiencies might be a part of the spectrum of longitudinal deficiencies. Kelikian agreed that true segmental absence was uncommon22. Nevertheless, the concept of transverse deficiencies persisted, and the diagnosis of phocomelia has continued to be utilized.
Recently, in 2003, Tytherleigh-Strong and Hooper23 evaluated the Frantz and O'Rahilly classification of phocomelia1 and presented their review of the findings in forty-four involved upper extremities. These authors questioned the existence of intercalary defects. The clavicle and scapula were present in all limbs. The lateral aspect of the scapula and the glenoid were always abnormal, and the humeral head, when present, was also abnormal. None of the hands were normal and most were missing at least one digit. Twelve of twenty patients who had abnormalities of both upper extremities had an identical dysplasia bilaterally. Only eleven of the forty-four limbs were classifiable with the Frantz and O'Rahilly system. The other thirty-three limbs had been classified as having phocomelia, but, on review of the findings, Tytherleigh-Strong and Hooper found these limbs to be non-classifiable. They divided these thirty-three extremities into three distinct groups. The limbs in Group A had an abnormal humerus with an abnormal single forearm bone; those in Group B had an abnormal humerus, radius, and ulna; and those in Group C had an abnormal humerus fused to a forearm bone or bones. Groups A and C appear to represent the deficiencies that we would classify as proximal radial longitudinal dysplasia and proximal ulnar longitudinal dysplasia, respectively.
Our findings, which were similar to those of Tyther-leigh-Strong and Hooper23, suggest that phocomelia as a classification term is overused and may be completely incorrect. The deformities in forty-nine of the sixty extremities in our study could be considered a continuum of radial or ulnar longitudinal dysplasia. In those cases, the abnormalities were demonstrated throughout the upper extremity without evidence of an intercalary deficit.
While the deformities in the other eleven extremities were not as clearly classifiable as either a radial or an ulnar longitudinal dysplasia, we think that the term phocomelia is inappropriate for them for three primary reasons. First, all eleven extremities had hand abnormalities in addition to the severe forearm and/or arm dysplasia and none had a true intercalary deficiency. Second, three of the four extremities that were classified as having type-B severe combined dysplasia (formerly known as complete phocomelia) had ulnar longitudinal dysplasia on the contralateral side; this strongly suggests that the deformities in the involved extremities represent severe manifestations in the continuum of ulnar longitudinal dysplasia. Lenz and Feldmann previously noted the relationship between ulnar defects in one extremity and more severe abnormalities (amelia, ulnar defects, humeroradial synostosis, and ectrodactyly) in the other, and they suggested that the association of “such diverse malformations with ulnar defects is suggestive of common etiology”—i.e., bilateral ulnar longitudinal dysplasia24. Third, the eleven extremities in our series potentially could be considered as having a combined longitudinal radial and ulnar dysplasia manifested by an absence of both forearm bones (as well as absence of the humerus in type B) with associated hand abnormalities. On the basis of these observations, we propose that upper-extremity disorders previously labeled as phocomelia should be considered to be longitudinal dysplasias.
Proximal Radial Longitudinal Dysplasia
The twenty-nine extremities classified as having proximal radial longitudinal dysplasia all had an abnormal shoulder, an absent proximal part of the humerus, an identifiable distal part of the humerus and ulnohumeral joint, a forearm with a normally contoured ulna and an absent or abnormal radius, and an abnormal hand. The hand abnormality resembled that seen in radial longitudinal dysplasia (i.e., an absent thumb and abnormal radial-sided digits), as described by Bayne and Klug6.
The classification of these upper-extremity deformities as a severe form of radial longitudinal dysplasia is supported by multiple observations. First, the dysplasia of the elbow, forearm, and hand, if considered alone, would be classifiable as radial longitudinal dysplasia. Second, shoulder dysplasia was previously noted by Bayne and Klug in patients with radial longitudinal dysplasia, although the specific abnormalities were not described6. We believe that either these shoulder abnormalities have been ignored when patients have been diagnosed as having radial longitudinal dysplasia (i.e., radial clubhand) or the severity of the proximal dysplasia has led to a diagnosis of phocomelia. Third, the patients whom we diagnosed as having proximal radial longitudinal dysplasia had medical disorders typically seen in association with radial longitudinal dysplasia (such as thrombocytopenia-absent radius syndrome) and infrequently had any other musculoskeletal abnormalities. Fourth, radial longitudinal dysplasia has a high prevalence of bilaterality, and thirteen of the sixteen patients who were classified as having proximal radial longitudinal dysplasia in our study had the abnormality bilaterally. Additionally, in two of the three patients with unilateral proximal radial longitudinal dysplasia, the contralateral extremity demonstrated the findings of radial longitudinal dysplasia described by Bayne and Klug6. While it is possible that the patients had two different upper-extremity anomalies, it is more likely that both extremities were affected by the same process.
The standard and accepted classification of radial longitudinal dysplasia does not include a category for the abnormalities in the patients in our series. In 1987, Bayne and Klug6 provided a classification scheme, consisting of types I through IV, which was recently supported by James et al.12 and expanded to include a type 0. Bayne and Klug based their classification system on the findings in 103 patients, whom they noted had associated proximal upper-extremity anomalies: one patient had a hypoplastic shoulder, one had a humeroulnar synostosis, and one had a proximal phocomelia (not otherwise clarified); however, these abnormalities were not incorporated into the classification system. We think that the extremities in this group represent a severe form of radial longitudinal dysplasia that may be best categorized as a new addition (Type V) to the Bayne and Klug classification (Fig. 7).
Proximal Ulnar Longitudinal Dysplasia
The twenty extremities classified as having proximal ulnar longitudinal dysplasia in our study all had a single arm/forearm bone, eleven had a distal bifurcation of that bone, and all had hand abnormalities resembling those seen in ulnar longitudinal dysplasia as reported by Bayne7 and by Cole and Manske8. The association between humeral bifurcation and radiohumeral synostosis with ulnar aplasia is well-established11,17,24-34. Humeral bifurcation with humeroradial synostosis and ulnar aplasia with oligodactyly is extremely uncommon (less than one in 50,000 births)11. Previous studies have been limited to case reports found predominantly in the genetics literature17,26,27,31,34. While the relationship between radiohumeral synostosis with distal bifurcation and ulnar longitudinal dysplasia has not been specifically noted in the orthopaedic literature, radiographs depicting this constellation of findings can be found in many previous reports on ulnar longitudinal dysplasia5,35-37.
Multiple considerations support the categorization of the disorder in these patients as a severe form of ulnar longitudinal dysplasia. First, radiohumeral synostosis is commonly seen in patients with ulnar longitudinal dysplasia (Table II). Previous authors have recognized that the single, long upper-extremity bone with a distal bifurcation represents a radiohumeral synostosis11,17,24-34. Consequently, the absent ulna and other similarities to the more typical ulnar longitudinal dysplasia strongly suggest that these extremities had a severe, proximal manifestation of ulnar longitudinal dysplasia. Second, two of the seventeen patients had ulnar longitudinal dysplasia as classified by Bayne7 on the contralateral side. As noted in the previous discussion concerning proximal radial longitudinal dysplasia, it is possible that these patients had two separate diagnoses, but a common diagnosis is much more likely. Third, associated medical conditions were uncommon and additional musculoskeletal abnormalities were common in this group, findings similar to those noted with ulnar longitudinal dysplasia38,39. Finally, the hand abnormalities seen in our patients were similar to those commonly found in patients with ulnar longitudinal dysplasia8,16,38.
In contrast to the single well-accepted classification of radial longitudinal dysplasia, ulnar longitudinal dysplasia has been classified in a variety of ways, each emphasizing different features36,38,40. While no single classification system has been utilized by all investigators, Bayne's classification of ulnar longitudinal dysplasia, modeled after the classification of radial longitudinal dysplasia, is used frequently7 (Table II). We suggest that a new type (type V) be added to represent cases of severe radiohumeral synostosis with humeral bifurcation or a large medial condyle (Fig. 8).
Severe Combined Dysplasia
Although the data are less clear, we believe that the upper-extremity abnormalities that we classified as severe combined dysplasia should also be considered to be longitudinal deficiencies. All seven patients with type-A and all four with type-B severe combined dysplasia had hand abnormalities, thereby challenging the concept of a segmental transverse defect. These deficiencies may represent a combination of both radial and ulnar longitudinal dysplasia or they may simply be a severe manifestation of ulnar longitudinal dysplasia as suggested by the high prevalence of contralateral ulnar longitudinal dysplasia. Additional clinical studies and genetic information are needed to clarify this issue.
The primary weakness of our study is that the disorders were classified on the basis of a retrospective review of radiographs and medical records. Furthermore, some radiographs were made when the patients were of a young age, and ossification in the extremities will change the radiographic appearance, potentially altering our classification. Additionally, our patient group may not have been large enough for us to assess the reliability of a new classification system or to modify an existing system. However, we think that, given the rarity of the diagnosis of phocomelia, sixty extremities represent an unusually large group. Finally, our selection of patients for inclusion in this investigation was based on a previous diagnosis of phocomelia. The use of that search criterion may have led to the inadvertent exclusion of some patients with severe proximal limb abnormalities that cannot be classified with the proposed system.▪
The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
Investigation performed at Texas Scottish Rite Hospital, Dallas, Texas, and Shriners Hospital for Children, St. Louis, Missouri
1. , O'Rahilly R. Congenital skeletal limb deficiencies. J Bone Joint Surg Am. 1961;43: 1202-24.
2. . Thalidomide and congenital abnormalities. Lancet. 1961;2: 1358.
3. . Congenital deformities of the upper extremities. Odense: Andelsbogtrykkeriet; 1949.
4. . The care of congenital hand anomalies. St. Louis: Mosby; 1977. p 37-51.
5. . A classification for congenital limb malformations. J Hand Surg [Am]. 1976;1: 8-22.
6. , Klug MS. Long-term review of the surgical treatment of radial deficiencies. J Hand Surg [Am]. 1987;12: 169-79.
7. . Ulnar club hand. In: Green DP, editor. Operative hand surgery. 3rd ed. New York: Churchill Livingstone; 1993. p 288-303.
8. , Manske PR. Classification of ulnar deficiency according to the thumb and first web. J Hand Surg [Am]. 1997;22: 479-88.
9. , Swanson GD, Tada K. A classification for congenital limb malformation. J Hand Surg [Am]. 1983;8: 693-702.
10. , Diez Pardo J, Qi BQ, Tovar JA, Xia HM. The contribution of the adriamycin-induced rat model of the VATER association to our understanding of congenital abnormalities and their embryogenesis. Pediatr Surg Int. 2000;16: 465-72.
11. , Evans JA, Kodaj I, Lenz W. Congenital limb deficiencies in Hungary: genetic and teratologic epidemiological studies. Budapest: Akademia Kiado; 1994.
12. , McCarroll HR Jr, Manske PR. The spectrum of radial longitudinal deficiency: a modified classification. J Hand Surg [Am]. 1999;24: 1145-55.
13. , Riddle RD, Laufer E, Tabin C. Sonic hedgehog: a key mediator of anterior-posterior patterning of the limb and dorso-ventral patterning of axial embryonic structures. Biochem Soc Trans. 1994;22: 569-74.
14. , Riddle RD, Tabin CJ. Mechanisms of limb patterning. Curr Opin Genet Dev. 1994;4: 535-42.
15. , Ogino T, Minami A, Ohshio I. Experimental study of radial ray deficiency. J Hand Surg [Br]. 1990;15: 470-6.
16. , Kato H. Clinical and experimental studies on ulnar ray deficiency. Handchir Mikrochir Plast Chir. 1988;20: 330-7.
17. , Braun-Quentin C. Genetic nosology and counseling of humeroradial synostosis. Genet Couns. 1994;5: 269-74.
18. , Johnson RL, Laufer E, Tabin C. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell. 1993;75: 1401-16.
19. , Tabin C. How limbs develop. Sci Am. 1999;280: 74-9.
20. . Vertebrate limb development. Curr Opin Genet Dev. 1995;5: 478-84.
21. . Genetics and limb development. Dev Genet. 1996;19: 1-8.
22. . Congenital deformities of the hand and forearm. Philadelphia: Saunders; 1974. Phocomelia; p 891-901.
23. , Hooper G. The classification of phocomelia. J Hand Surg [Br]. 2003;28: 215-7.
24. , Feldmann U. Unilateral and asymmetric limb defects in man: delineation of the femur-fibula-ulna complex. Birth Defects Orig Artic Ser. 1977;13: 269-85.
25. , Coates V. Apparent bifurcatio of distal humerous with oligoectrosyndactyly. Am J Med Genet. 1983;14: 591-3.
26. , Reed M, Evans JA. Humeroradial synostosis, ulnar aplasia and oligodactyly, with contralateral amelia, in a child with prenatal cocaine exposure. Am J Med Genet. 2003;116: 85-9.
27. , Haan EA, Humphrey IJ. Humeroradial synostosis in a patient with lambdoid synostosis. J Med Genet. 1993;30: 81-2.
28. , O'Rahilly R. Ulnar hemimelia. Artif Limbs. 1971;15: 25-35.
29. , Speeckaert MT. Humeroradioulnar synostosis appearing as distal humeral bifurcation in a patient with distal phocomelia of the upper limb and radial ectrodactyly. Am J Med Genetics. 1984;18: 365-8.
30. , Goldberg MJ. Amelia and scoliosis. J Pediatr Orthop. 1985;5: 605-9.
31. , Joyce MR, Profumo LE. Humero-radial-ulnar synostosis: a new case and review. Am J Med Genet. 1989;33: 170-1.
32. . Congenital deformities of the hand and forearm. Philadelphia: Saunders; 1974. p 874.
33. , Farmer A. Congenital absence of ulna. Can J Surg. 1959;2: 204-7.
34. , Speeckaert MT. Humeroradioulnar synostosis appearing as distal humeral bifurcation in a patient with distal phocomelia of the upper limbs and radial ectrodactyly. Am J Med Genet. 1984;18: 365-8.
35. , Neufeld SK. Ulnar ray deficiency. Hand Clin. 1998;14: 65-76.
36. , Wenner SM, Kruger LM. Ulnar deficiency. J Hand Surg [Am]. 1986;11: 822-9.
37. , Omer GE Jr. Congenital ulnar deficiency. Natural history and therapeutic implications. Hand Clin. 1985;1: 499-510.
38. , Tada K, Yonenobu K. Ulnar ray deficiency: its various manifestations. J Hand Surg [Am]. 1984;9: 658-64.
39. , Smith RJ. Deformities of the hand and wrist with ulnar deficiency. J Hand Surg [Am]. 1979;4: 304-15.
Copyright 2005 by The Journal of Bone and Joint Surgery, Incorporated
40. . Die missbildungen der extremitaeten durch defekt, verwachsung und ueberzahl. Cassel, Germany: T.G. Fisher; 1895.