• classification;
  • congenital anomalies;
  • limb reduction defect


  1. Top of page
  2. Abstract
  7. Acknowledgements

Limb deficiencies, the congenital absence or hypoplasia of a long bone and/or digits, vary greatly in their anatomy and etiology. Previous attempts to classify the range of possible phenotypes have not included all types of deficiencies. We present a new classification system, which includes all potential phenotypes. Infants with limb deficiencies were identified in the hospital-based Active Malformations Surveillance Program at Brigham and Women's Hospital in Boston, MA from the years 1972 to 1974 and 1979 to 2000. Affected infants were classified based on the anatomy and apparent cause of their deficiencies. The prevalence rate of all types of limb deficiency was 0.79/1,000. Upper limb deficiencies were significantly more common than lower limb deficiencies. There was no significant difference in frequencies between deficiencies on the left and right sides of the body. Longitudinal defects were more common than terminal transverse defects; intercalary defects were uncommon. Longitudinal defects were most likely to occur on the preaxial side of the limb. Almost half of affected infants had affected digits, with normal long bones. The most common apparent cause of limb deficiencies was vascular disruption defects (0.22/1,000), such as amniotic band-related limb deficiency. This new classification system includes deficiency of each long bone, as well as absence of any finger or toe. This system will make it possible to establish the prevalence of each specific phenotype. The large number of distinct apparent causes illustrates the marked etiologic heterogeneity of limb deficiencies. © 2011 Wiley-Liss, Inc.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Many different causes of limb deficiency or limb reduction defects have been identified. These include genetic disorders, such as Fanconi anemia [Giampietro et al., 1993], chromosomal abnormalities, such as trisomy 18, and environmental exposures, such as the medications thalidomide [Smithells, 1973] and misoprostol [Gonzalez et al., 1998] and the prenatal diagnosis procedure chorionic villus sampling (CVS) [Golden et al., 2003].

The range of severity of limb deficiencies varies greatly, with certain phenotypes classically associated with particular etiologies. For example, Fanconi anemia is associated with the absence of the thumb [Fanconi, 1967], while CVS is associated with the absence of the distal phalanx of the third finger [Golden et al., 2003]. Holt-Oram syndrome, an autosomal dominant malformation [Hurst et al., 1991; Brassington et al., 2003], is associated with the hypoplasia of the radius and thumb of one or both arms.

Many anatomic classification systems have been described [O'Rahilly, 1951; Frantz and O'Rahilly, 1961; Hall et al., 1962; Swanson et al., 1968; Kay et al., 1975; Swanson, 1976; Kallen et al., 1984; Stoll et al., 1992, 1998; Froster, 1996]. These have included deficiencies defined by their relationship to the central axis of the arm and leg: preaxial defects (radius/tibia), central deficiencies, and postaxial defects (ulna/fibula). However, some tabulations of limb deficiencies, such as those associated with CVS [Froster and Baird, 1993], did not include the category of amniotic band limb related limb deficiency in which there is often a deficiency of one or two digits [Golden et al., 2003].

We present a system of anatomic classification of all limb deficiencies and demonstrate its use in a consecutive population of newborn infants. This report reflects an expansion of an earlier classification scheme based on the findings in a smaller number of affected infants identified in the same hospital [McGuirk et al., 2001].


  1. Top of page
  2. Abstract
  7. Acknowledgements

Each infant with a limb deficiency was identified in the database of the Active Malformations Surveillance Program at Brigham and Women's Hospital in Boston, MA. In this program, the medical records of each liveborn and stillborn infant, and each elective termination because of prenatally identified fetal anomalies, were reviewed 6 days a week [Nelson and Holmes, 1989; Peller et al., 2004]. The surveillance was conducted initially from February 16, 1972 through February 15, 1975 (interrupted because of lack of office space), resumed on January 1, 1979, and is ongoing. For the present analysis, ascertainment ended on December 31, 2000, when there was a significant decrease in the occurrence of elective terminations because of fetal anomalies at this hospital. This analysis was also limited to infants born to mothers who had always planned to deliver at BWH, referred to as maternal “non-transfers.” In the previous analysis [McGuirk et al., 2001], the search for missed infants in the records of the hand surgeons in the Boston area showed that 92.7% (102 of 110) of all infants with a limb deficiency had been identified at birth by the Surveillance Program.

The parents of each index case included in the present study were, when interviewed, asked to provide written permission to include their child in the Surveillance Program. There is an annual internal review board (IRB) review and approval of the study.

A congenital limb deficiency was defined as the absence or hypoplasia of a long bone, metacarpal, metatarsal, or phalanx of one or more limbs, which was significant enough in appearance to be detected by an examining physician within the first 5 days of life. Excluded from this definition were mild shortening of the digits due to isolated brachydactyly, shortening of digits in skeletal dysplasias, and curvature of digits (clinodactyly).

The surveillance process to identify limb deficiencies included reading the findings of the pediatricians' examinations, the reports from evaluations by consultants and the reports from diagnostic studies, including radiographs, ultrasound, magnetic resonance imaging, CT scans, chromosome analyses, chromosome microarrays, and mutation analysis. Information on the sex of an infant, birth status, multiple gestations, family and pregnancie's histories, mother's medication use, infertility treatments, and non-limb anomalies was obtained from the mother's medical record and often confirmed with a personal interview by a research assistant.

Two classifications were made: an anatomic classification and an apparent etiology of the limb deficiency.

Anatomic Classification

The anatomic classification (Figs. 1 and 2) began with the anatomic subdivisions used by the International Classification of Diseases, Ninth Revision (ICD-9) codes 755.2 to 755.4 (for upper, lower, and unspecified limb defects) and 658.8 for amniotic band syndrome. The extension of this classification to six digits, developed by the British Pediatric Association, was used. The separate listing of deficiencies involving one or more fingers or toes, used previously by Golden et al. [2003], was included.

thumbnail image

Figure 1. Anatomic classification of upper limb deficiencies.

Download figure to PowerPoint

thumbnail image

Figure 2. Major groups of upper and lower congenital limb deficiencies. Absent or hypoplastic structures are shaded. A: Complete absence of limb; (B) Intercalary defect; (C) Terminal transverse defect; (D) Longitudinal defect, preaxial; (E) Longitudinal defect, central; (F) Longitudinal defect, postaxial; (G) Longitudinal, pre- and postaxial.

Download figure to PowerPoint

The initial subdivision of limb deficiencies was between the complete absence of the limb and the partial absence of a limb. The absence of part of a limb was divided into three secondary groups: intercalary, terminal transverse, or longitudinal defects.

Intercalary defects were defined as the absence or hypoplasia of a middle section of a long bone such as the femur or radius, with normal distal structures such as the hand, foot, or digits.

Terminal transverse defects had absence of all distal structures beyond a specific point perpendicular to the limb, such as absence of the lower half of the forearm and hand. In the arms, the specific points were the metacarpal-phalangeal (MCP) joints, mid-hand, wrist, forearm, or upper arm. In the legs, the points included the: metatarsal-phalangeal (MTP) joints, mid-foot, ankle, lower leg, or upper leg. Terminal transverse limb defects at the level of the MCP or MTP joints affected either digits 1, 2, 3, and 4; or 2, 3, 4, and 5; or all five digits.

Some limbs with terminal transverse deficiencies had a smooth or dimpled distal end. Others had small soft-tissue “nubbins,” which resemble rudimentary digits in the proximal forearm [Drapkin et al., 2003] or at the level of the carpal bones.

Longitudinal defects were defined as the absence or hypoplasia of a bone parallel to the long axis of the limb and included preaxial, central, postaxial, and mixed pre- and postaxial longitudinal defects.

Preaxial defects occurred on the medial side of the limb. Upper preaxial limb deficiencies included the absence or hypoplasia of the: radius; radius and thumb; radius, thumb, and second digit; thumb, second, and third digits; thumb and second digit; or thumb alone. Lower preaxial defects affected the analogous structures of the legs. Defects of the first toe were substituted for those affecting the thumb and defects of the tibia were substituted for defects of the radius.

Central longitudinal defects affected the portion of the limb nearest to the central axis: absence or hypoplasia of the second through fourth digits; third and fourth digits; second and third digits; or the second, third, or fourth digits only.

Postaxial defects occurred on the lateral side of the limb, and included the absence or hypoplasia of the: ulna; ulna and fifth digit; third through fifth digits; the individual fourth and fifth digits; and in the legs, absence of fibula with or without absence of the fifth and fourth toes. Sirenomelia, with the fusion of the legs, was also considered a lower postaxial defect [Stevenson et al., 1986].

The term “unilateral limb deficiency” was used for infants with one limb deficiency or two ipsilateral limbs deficiencies affecting one arm and one leg. “Bilateral limb deficiency” was used to refer to the occurrence of deficiency of the same structures in both arms or both legs. Infants with anatomically different limb deficiencies on two or more limbs were classified as “mixed” and were listed separately.

The term “isolated” was used to refer to the limb deficiency of an infant who had no major anomalies in non-limb structures. A child with an “isolated” limb deficiency could have malformations of two or more limbs. The term “multiple congenital anomalies” was used to distinguish infants who had one or more limb deficiencies, as well as at least one associated anomaly in a non-limb structure. Limb deficiencies were listed as not classified if the original descriptions in the medical record did not provide sufficient detail.

Etiological and Pathogenic Classification

A classification system of apparent etiology or pathogenesis included: chromosomal abnormalities (aneuplodies); dominant or recessive genes (“Mendelian inheritance”); familial inheritance in the absence of a Mendelian syndrome; known syndromes, sequences, associations, and related anomalies (i.e., VACTERL association, prune belly syndrome); teratogenic exposures; presumed vascular disruption defects; and unknown causes. The apparent etiology was established after a review of the infant's phenotype and the descriptions in each infant's medical record, and was not based on the ICD-9 codes.


  1. Top of page
  2. Abstract
  7. Acknowledgements

There were 206,224 liveborn infants, stillborn infants, and elective terminations born to women who had always planned to deliver at this hospital. There were 162 infants with congenital limb deficiencies. The prevalence of infants born with limb deficiencies in this sample was 0.79/1,000.

Of those analyzed, 101 (62.3%) were liveborn infants who survived the neonatal period, defined as the first month of life. The remainder of the sample included: 10 (6.2%) liveborn infants who did not survive the neonatal period, 17 (10.5%) stillborn infants, and 34 (21.0%) elective terminations. Eighty-seven infants (53.7%) were male, 64 (39.5%) were female, and in 11 (6.8%) elective terminations, sex was not determined.

There was no significant difference between the number of infants with isolated limb deficiencies (78 [48.1%]) and those who had multiple congenital anomalies (84 [51.8%]) (χ2 (1, N = 162) = .22, P = 1.27). Fifteen infants (9.3%) were twins or triplets.

Anatomic Classification

Each infant was represented once in all tabulations, regardless of the number of limb deficiencies that he or she had. Some anatomical tabulations include the 135 infants with unilateral and symmetrical limb deficiencies (Table I), but not the 25 infants with mixed deficiencies (Tables II and III) or 2 infants with unspecified deficiencies. Other tabulations include all 162 affected infants.

Table I. Anatomic Classification of Infants' Unilateral and Bilateral Limb Deficiencies (n = 135)
Anatomical classificationTotal numberArmsLegsArms and legs
  1. The side of two additional infants' unilateral arm defects were not specified. The 25 infants with mixed limb deficiencies are listed in Tables II and III.

Complete absence of the limb2    2    
Partial absence of the limb126         
   Radius and digit 1/tibia and digit 114616  1   
   Radius and digits 1–2/tibia and digits 1–222        
   Digit 113362     2
   Digits 1–2         
   Digits 1–3         
   Ulna and digit 5/fibula and digit 51   1     
   Digit 5943 1    1
   Digits 4–51 1       
   Digits 3–52 1      1
   Sirenomelia1     1   
  Pre- and postaxial13         
   Ulna, radius and digit 1/tibia, fibula and digit 16321      
   Digit 1 and digit 511        
   Digit 2 and digit 5         
   Digit 1 and digits 3–5         
   Digits 1–2 and digits 4–5         
   Digits 2–54 3  1    
   Radius and digit 5/tibia and digit 5         
   Radius, digit 1 and digit 5/tibia, digit 1 and digit 51 1       
   Radius and digits 1–4/tibia and digits 1–41  1      
   Digit 23 2 1     
   Digit 32    2    
   Digit 4         
   Digits 2–32   1 1   
   Digits 3–41  1      
   Digits 2–49511 11   
 Terminal transverse39         
  MCP/MTP joints16         
   Digits 2–543   1    
   Digits 1–4211       
   Digits 1–510332 11   
  Distal humerus/distal femur2         
   No nubbins1  1      
   Nubbins1    1    
  Proximal forearm/lower leg12         
   No nubbins32   1    
   Nubbins962 1     
   No nubbins         
   Nubbins962  1    
   No nubbins         
 Side known72 1112   
Table II. Mixed Limb Deficiencies in Patients With Isolated Anomalies (n = 14)
 L armR armL legR legEtiology
  1. IDM, infant of a diabetic mother.

1  Terminal transverse: Lower leg absentLongitudinal, pre- and postaxial: Digits 1–2 and 4–5 absentAmniotic band syndrome
2Longitudinal, central: Digit 3 and digit 4 shortenedLongitudinal, central: Digit 3 and digit 4 shortenedTerminal transverse: Digits 1–5 shortenedTerminal transverse: Digits 1–5 shortenedAmniotic band syndrome
3  Longitudinal, preaxial: Digits 1–3 shortenedLongitudinal, central: Digits 2–4 shortenedAmniotic band syndrome
4Terminal transverse: Digits 1–5 absentLongitudinal, postaxial: Digits 3–5 hypoplastic  Amniotic band syndrome
5Longitudinal, postaxial: Digits 4–5 absentTerminal transverse: Digits 1–5 absent  Amniotic band syndrome
6Longitudinal, central: Digit 4 shortenedLongitudinal, central: Digit 4 shortenedTerminal transverse: Lower leg absent, no nubbinsLongitudinal, central: Digits 2–4 absentAmniotic band syndrome
7Longitudinal, preaxial: Digits 1–3 absent Terminal transverse: Lower leg absent, no nubbinsLongitudinal, pre- and postaxial: Digits 2–3 and digit 5 shortenedAmniotic band syndrome
8Longitudinal, postaxial: Digit 5 absentLongitudinal, preaxial: Digit 1 absent  Unknown cause
9Terminal transverse: Digits 2–5 absentTerminal transverse: Digit 1–5 absentTerminal transverse: Digits 1–4 shortenedLongitudinal, preaxial: Digits 1 – 3 absentIDM with amniotic band syndrome
10 Longitudinal, preaxial: Digit 1 absentLongitudinal, central, unspecified: “Split foot”Longitudinal, central, unspecified: “Split foot”Split hand/foot syndrome
11Longitudinal, central, unspecified: Central digits absentTerminal transverse: Forearm absent, no nubbinsTerminal transverse: Lower leg absentTerminal transverse: Lower leg absentUnknown cause
12 Longitudinal, unspecified: 1 digit absentLongitudinal, pre- and postaxial: Fibula absent, femur shortLongitudinal, pre- and postaxial: Femur short, fibula and 1 digit absentUnknown cause
13Terminal transverse: Digits 2–5 absent Terminal transverse: Digits 1–5 absent Unknown cause
14  Longitudinal, central: Digits 2–3 shortenedTerminal transverse: Digits 1–5 shortenedUnknown cause
Table III. Mixed Limb Deficiencies in Patients With Multiple Congenital Anomalies (n = 11)
 L armR armL legR legOther anomaliesEtiology
  1. VSD, ventral septal defect; TOF, Tetralogy of Fallot; ASD, atrial septal defect; IDM, infant of a diabetic mother.

1Longitudinal, preaxial: Digit 1 absentLongitudinal, preaxial: Radius and digit 1 absentLongitudinal, preaxial: Digit 1 absent Encephalocele, cleft lip and palateAmniotic band syndrome
2Longitudinal, central: Digits 2–4 absentTerminal transverse: Digits 2–5 absent Terminal transverse: Digits 1–4 absentBilateral cleft lip and palateAmniotic band syndrome
3Longitudinal, preaxial: Digit 2 hypoplasticLongitudinal, postaxial: Digits 3–5 hypoplastic  Bilateral clubbed feetAmniotic band syndrome
4Longitudinal, preaxial: Digit 1 absent Longitudinal, postaxial: Digit 5 absentLongitudinal, postaxial: Digit 5 absentCleft palate, microtia (R), abnormal ear with small auditory canal (L)Nager's acrofacial dystosis
5  Longitudinal, unspecified: Fibula and 1 digit absentLongitudinal, unspecified: 1 digit absentCleft palate, micrognathia, bilateral microtiaNager's acrofacial dystosis
6Longitudinal, preaxial: Digits 1–2 absentLongitudinal, central: Digits 2–3 absentLongitudinal, central: Digits 2–4 absentLongitudinal, central: Digits 2–4 absentVSD, diaphragmatic hernia, polyvalvular diseaseSplit hand/foot syndrome and Trisomy 18
7Longitudinal, central: No. 2–4 absentTerminal transverse: Forearm and hand absent  Cleft palate, TOF, VSD, pulmonary atresia, cryptorchord testisBrachman-de Lange Syndrome
8Longitudinal, central: Digits 2–4 shortenedLongitudinal, preaxial: Radius and digit 1 absentLongitudinal, preaxial: Digit 1 absent, digit 2 hypoplasticTerminal transverse: Lower leg absentVSDs, ASD, hypoplastic ribs, inguinal herniaIDM
9 Longitudinal, postaxial: Digits 3–5 absentTerminal transverse: Lower leg absent OmphaloceleMisoprostol exposure
10Complete absence of the limb: Arm absentComplete absence of the limb: Arm absentLongitudinal, unspecified: 1 digit absentLongitudinal, unspecified: 1 digit absentAcardiac twinAcardiac twin
11Longitudinal, unspecified: 2 digits absentLongitudinal, preaxial: Digit 1 absentTerminal transverse: Digits 2–5 absentTerminal transverse: Digits 2–5 absentCystic hygroma, hydro-cephalus, cardiac defectsUnknown cause

The major findings can be summarized as follows. Most infants (104/162 [64.2%]) had only one limb deficiency. Among infants with unilateral and symmetrical defects, upper limb deficiencies (99/135 [73.3%]) were more common than lower limb deficiencies (32/135 [23.7%]). Unilateral limb deficiencies were much more common (104/162 [64.2%]) than symmetrical (33/162 [20.4%]) or mixed deficiencies (25/162 [15.4%]). There was no difference in the frequencies on the left side and the right side.

Complete absence of the limb was rare (three infants), one of which was an acardiac twin. Partial absence of the limb accounted for 93.3% (126/135) of the infants with a unilateral or symmetrical bilateral limb defect. Among these defects, longitudinal defects were the most common (73/135, 54.1%). Preaxial defects (29/73 [39.7%]) were significantly more common than central (18/73 [24.7%]), postaxial (13/73 [17.8%]), or pre- and postaxial defects (13/73 [17.8%]), (χ2 (3, N = 73) = 9.4 P < .05).

Terminal transverse defects accounted for 39/135, or 28.9% of the meromelia defects. Most infants with terminal transverse deficiencies had normal long bones proximal to the defect, and were missing distal structures. Of 39 infants with only terminal transverse deficiencies, almost half (16 [41.0%]) had deficiencies at the MCP or MTP joints. Ten (24.3%) were lacking all structures below the proximal forearm, eight (19.5%) had deficiencies below the wrist, two (4.9%) had deficiencies below the lower leg, and one each (2.4%) had deficiencies below the distal femur, distal humerus, and ankle. Of 23 patients who had terminal transverse deficiencies proximal to the digits, 19 (82.6%) had soft tissue nubbins.

Almost half of the affected infants (78/162 [48.1%]), had deficiencies involving absent or hypoplastic digits, with normally appearing long bones. Absence of the first digit (13/162 [8.0%]) was the most common, followed by absence of the fifth digit (9/162 [5.6%]).

Etiological and Pathogenic Classification

An apparent cause for infants' limb deficiencies was assigned in 105 (64.8%) cases (Table IV). Etiological/pathogenic tabulations include all infants (162), regardless of their anatomical classification.

Table IV. Apparent Etiology or Pathogenesis of Infants' Limb Deficiencies
Etiology or pathogenesis of limb deficiencyInfantsPercentage
  • a

    One infant with trisomy 18 also had split hand/foot syndrome (and an affected mother).

  • b

    Huang et al. [2002].

  • c

    Holmes et al. [1995].

  • d

    Lin et al. [1998].

  • e

    Drapkin et al. [2003].

  • f

    IDM had a unilateral absent thumb, bilateral absent thumbs, a terminal transverse deficiency at the wrist with nubbins, and two mixed deficiencies.

  • g

    Genest et al. [1999].

Chromosomal abnormalities138.0%
 Trisomy 187a4.3%
 Thrombocytopenia-absent radius syndrome21.2
 Other chromosomal abnormalities: trisomy 13; triploidy; 13q-; 46XX/47XX, +mar1 Each, 4 total0.6% each, 2.5% total
Known syndrome, association, sequence, or related anomaly148.6%
 VACTERL association42.5%
 Other known syndromes, sequences, and related anomalies: Cloacal exstrophy; encephaloceleb; persistence of common cloaca; sirenomelia; prune belly syndrome; urethral atresia1 Each, 6 total0.6% each, 3.7% total
Mendelian inheritance2314.2%
 Split hand/foot syndrome84.9%
 Fanconi anemia42.5%
 Holt-Oram syndrome21.2%
 Private syndrome, consanguineous parents: absent/hypoplastic tibia with multiple anomaliesc21.2%
 Nager acrofacial dysostosis21.2%
 Cornelia de Lange syndrome21.2
 Other Mendelian syndromes: Adams-Oliver syndromed; oro-facial-digital syndrome, type 1; Smith-Lemli-Opitz syndrome1 Each, 3 total0.6% each, 1.9% total
Familial inheritance31.9%
Presumed vascular disruption defects4628.4%
 Amniotic band syndrome2716.7%
 Terminal transverse defects with nubbinse148.6%
 Poland anomaly42.5%
 Acardiac twin10.6%
Teratogenic exposure63.7%
 Infants of diabetic mothers (IDM)f53.1%
Unknown cause5735.1%

Fifteen (9.3%) infants had chromosome abnormalities, out of 61 infants who had chromosome analysis. The most common was trisomy 18 (7 [4.3%]). TAR was listed as a presumed deletion on chromosome 1 [Klopocki et al., 2007]. None of the infants with limb deficiencies had a chromosome microarray analysis.

Among the phenotypes attributed to either dominant or recessive genes, split hand/split foot syndrome (8 [4.9%]) and Fanconi anemia (4 [2.5%]) were the most common.

Three infants (1.85%) had a family history of congenital limb deficiencies: (1) An infant with a terminal transverse deficiency with nubbins of the left wrist had a paternal uncle with a similar phenotype, (2) an infant with a terminal transverse defect of the left proximal forearm, also with nubbins, had a maternal great-aunt born without a portion of her arm, (3) an infant with absence of the fifth digit on the right hand had a maternal uncle with split hand/foot syndrome and a paternal uncle who was also missing digits on one hand.

Several syndromes were identified in affected infants (14 [8.6%]) infants' congenital limb deficiencies. The most common was the VACTERL association (4 [2.5%]). A few infants with anencephaly and encephalocele, which are known to have associated non-neural malformations, had associated limb deficiencies (4 [2.5%]).

The limb deficiencies attributed to teratogenic exposures were in infants of diabetic mothers (5 [3.1%]) and one (0.6%) infant exposed to misoprostol [Genest et al., 1999]. The limb deficiencies of infants of diabetic mothers did not occur in a consistent pattern.

Presumed vascular disruption defects accounted for 46 (28.4%) infants' malformations, with amniotic band syndrome (27 [16.7%]) and terminal transverse defects with nubbins (14 [8.6%]) the most common [Drapkin et al., 2003].

An apparent etiology was not established for 57 (35.1%) infants in the first 5 days of life or after a post-mortem examination.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We present an anatomic and etiological classification system that will make it possible to establish overall prevalence rates of all types of congenital limb deficiencies. This new system facilitates the identification of sub-groups of children with similar deficiencies, such as vascular disruption defects involving one or two digits, which have been correlated with exposure to CVS [Golden et al., 2003]. Additionally, infants with distinctive phenotypes, such as terminal transverse deficiencies with nubbins, can be distinguished from infants with similar presentations, but divergent etiologies, such as terminal transverse defects resulting from amniotic band syndrome. Straightforward terminology and a clear hierarchical order make this classification useful for tabulations of prevalence and evaluation of potential exposures.

Some limitations to the classification of limb deficiencies remain. This system does not address the severity of digital deficiencies. An infant missing the distal phalanges of the second and third fingers, for instance, is not differentiated from an infant with complete absence of these digits. Also, infants with “mixed” deficiencies, different phenotypes on two or more limbs, cannot be easily classified or analyzed. Limitations to this program's methodology exist as well. Because data is ascertained primarily during the infant's first 5 days of life, it is difficult to determine the exact number of infants who had genetic tests or what their results were. Furthermore, this paper is limited by its focus on phenotype and diagnosis. There was no consistent genomic or exomic sequencing of the affected infants. This is an important area of exploration that has the potential to identify the pathogenetic mutations and variants underlying the etiologies of limb deficiencies in infants who did not receive clinical diagnoses.

This analysis shows the significance of including elective terminations in any analysis of limb deficiencies, as 21% of the affected infants were in these pregnancies. However, the affected fetuses in terminations are less likely to have detailed descriptions of their limb deficiencies. Affected liveborn infants often do not receive comprehensive evaluations at birth; radiographs and diagnostic evaluations are completed after the infant has been discharged. Potential etiologies, such as chromosome deletions or duplications and mutations, could not be identified as the chromosome microarray testing and mutation analyses were not utilized in the evaluations of the affected infants.

Many new developments have occurred in recent years, which could have affected the etiologies recognized in this analysis. These include the search for gene deletions or duplications in infants with multiple anomalies (Table V) and the availability of mutation analysis for Fanconi anemia and Okihiro syndrome [Verlander et al., 1994; Borozdin et al., 2004], conditions associated with absent or hypoplastic thumbs, and analysis of GLI3 gene for individuals with polysyndactyly [Wild et al., 1997; Johnston et al., 2004].

Table V. Genetic Tests Available for Monogenic Causes of Limb Deficiencies
Apparent etiologyOMIM numberGene test availableaAssociated gene(s)
  • a

    Testing options as listed at

  • b

    Genzer-Nir et al. [2010].

Cornelia de Lange syndrome122,470, 300,590, 610,759NIPBL, SMC1A
Holt-Oram syndrome142,900TBX5
Mammary-digital-nail syndromebRegion of 22q12.3-13.1
Nager acrofacial dysostosis154,400AFD1
Okihiro syndrome607,323SALL4
Oro-facial-digital syndrome, type 1311,200OFD1
Smith-Lemli-Opitz syndrome270,400DHCR7
Split hand/foot syndrome183,600, 606,708, 313,350, 246,560, 605,289SHFM1, SHFM2, SHFM3, SHFM4, SHFM5

This analysis identified three separate phenotypes that are not well-known and warrant further delineation:

Terminal transverse limb defects with residual nubbins at the level of the proximal forearm or wrist. Eight affected infants were identified in this analysis. Histologic sections of one of these infants showed that the nubbins with slit-like nails contain cartilage [Drapkin et al., 2003]. Affected first- and second-degree relatives were noted in our sample of affected infants [McGuirk et al., 2001].

A more central hypoplasia of the fingers with larger nubbins and fingernails at the MCP joint (Fig. 3). Seven infants with central deficiencies of digits 2 through 4 were identified in this analysis. The thumb and fifth finger are separate and are either normal size or hypoplastic. Affected parents and children have been reported [Graham et al., 1986; Neumann et al., 1998].

thumbnail image

Figure 3. Central hypoplasia of the fingers, with nubbins in the place of the second, third, and fourth digits at the level of the MCP joint. The thumb and fifth digit are not hypoplastic.

Download figure to PowerPoint

Hypoplasia of a central metacarpal (fingers 2 through 4) and proximal phalanges with normal distal phalanges and fingernails. Only one infant with this deficiency was identified in this analysis. An affected parent and child have been reported previously [Holmes and Remensnyder, 1972].

Each of these phenotypes is separate and distinct from amniotic band related limb deficiencies, which are characterized by syndactyly of two or more fingers, constriction rings, and often deficiency of distal tissues.


  1. Top of page
  2. Abstract
  7. Acknowledgements

This study was supported by funds provided by the Massachusetts birth Defects Registry, which is part of the National Birth Defects Prevention Study, a Centers for Disease Control project. We thank the many pediatricians, geneticists, orthopedists, pathologists, and other specialists whose careful evaluations and thoughtful diagnoses contributed to the classification of the affected infants. We also thank Research Assistant Kathryn Rowan for her assistance with the classification of the affected infants born between 1994 and 2000.


  1. Top of page
  2. Abstract
  7. Acknowledgements