Congenital anomalies of upper extremities: prenatal ultrasound diagnosis, significance, and outcome

Objective

We sought to assess the role of ultrasound in the prenatal characterization of fetal malformations of the upper extremities (MUE).

Study Design

Ultrasound findings, associated anomalies, pregnancy, and fetoneonatal outcome were analyzed in 100 fetuses with MUE, categorized after Swanson classification of hand congenital abnormalities. Follow-up information was available in all.

Results

Twelve cases had an isolated defect, while 88 showed associated abnormalities involving other organ systems. Fetuses with multiple malformations showed a 76% syndromic risk (32% chromosomal, 41% nonchromosomal), with unfavorable outcome in almost all cases (only 4 survivors, 2 with neurodevelopmental delay). In 27% of syndromic fetuses, the sonographic evidence of the MUE represented the key finding leading to the final diagnosis. The prognosis was usually favorable in cases of isolated defects.

Conclusion

A thorough prenatal ultrasound characterization of fetal MUE can assist in the differential diagnosis of many syndromic conditions. When isolated, MUE tend to have a good outcome.

Malformations of extremities may be isolated or associated with chromosomal and nonchromosomal syndromes, deformation sequences, and skeletal dysplasias. Hence, the ultrasound evaluation of the limbs plays a critical role in the expert assessment of the fetus with congenital anomalies, as malformed upper extremities may be suggestive of several syndromic conditions. This is why, although a limb malformation may be sometimes fortuitously discovered during routine obstetric ultrasound, more often the expert sonologist will actively look for it following the detection of other major anomalies, to reach a final diagnosis.

Three-dimensional ultrasound is not mandatory to assess fetal limbs but has been demonstrated to offer a number of advantages over 2-dimensional ultrasound in characterizing limbs’ abnormalities.

The aim of this study is to assess the role of 2-dimensional and 3-dimensional ultrasound in the prenatal diagnostic evaluation and prognostic characterization of 100 fetuses with malformations of the upper extremities (MUE) managed at our fetal unit over an 8-year period.

Materials and Methods

This study represents a retrospective observational analysis of cases with MUE managed at our fetal unit since 2001. In this period, 1569 cases of fetal anomalies have been managed at our department, including 115 (7.3%) with MUE. Final confirmation of the diagnosis could be retrieved in 100 of 115 cases, which represent the study population. From our computerized database, the following variables were retrieved and analyzed: referral indication, gestational age at diagnosis, type of upper limb malformation, associated anomalies, karyotype, pregnancy, and fetoneonatal outcome.

Anomalies of the upper limbs considered in the study include structural abnormalities, contractures, and postural deformities involving the extremity either unilaterally or bilaterally. These malformations have been classified according to the embryological classification scheme proposed by Swanson, which represents one of the most widely accepted classifications of congenital limb anomalies. According to this classification system, 7 categories are identified: (1) failure of differentiation (eg, syndactyly); (2) failure of formation of parts (eg, radial hypo/aplasia with clubhand); (3) duplication defects (eg, polydactyly); (4) overgrowth (eg, macrodactyly); (5) undergrowth; (6) constriction band sequence; and (7) generalized skeletal malformations. However, generalized skeletal malformations were only included if associated with clear abnormalities of the extremities, such as polydactyly. All other types of skeletal dysplasias were excluded from the analysis together with cases showing isolated rhizo/mesomelia and focal abnormalities of upper limbs not involving the acral portion (eg, constriction band of the rhizomelic tract). As per the suggestion of Swanson, in case of multiple limb malformations, the defect was classified according to the most predominant aspect for diagnostic and prognostic purposes.

From August 2001 through January 2004, ultrasound examinations were performed with 2-dimensional ultrasound equipment (Prosound SSD-5000; Aloka, Tokyo, Japan) using 2- to 5-MHz and 4- to 7-MHz convex probes and/or a 5- to 7-MHz endovaginal transducer. From February 2004 through August 2008, all fetuses were assessed by 2- and 3-dimensional ultrasound (GE Voluson 730 Expert or GE8; General Electric Medical System, Kretztechnik, Zipf, Austria) using a 4- to 8-MHz transabdominal volumetric transducer or 5- to 9- and 6- to 12-MHz endovaginal volumetric probes. All the scans were performed and/or reviewed by the same experienced physician (D.P.). Three-dimensional volume data sets were acquired in all cases studied with this technique. The volumes were processed offline with the dedicated software (4-dimensional viewer, General Electric Medical System). Rendering modes included both the transparent maximum mode and the surface mode, used to display the bones and the outline of the limb, respectively.

The standard protocol for targeted ultrasound evaluation of upper extremities includes identification of the presence, number, size, morphology, position, and relationship of the following anatomical structures: the unossified hypoechogenic carpus, 5 hyperechoic and cylindric metacarpal bones, 5 independent digits of different length with 3 ossified phalanges (2 for the thumb), and normal radius and ulna. In addition, natural posture (fetal hand and wrist slightly flexed with hands closed at rest) of the limbs, joint mobility (fingers, wrists, elbows), and forearm-hand axis are assessed. In case of suspected anomalies, both hands are examined on axial and longitudinal views. A thorough anatomy check and fetal echocardiography were performed in all cases to investigate the presence of associated abnormalities. Karyotyping was performed in those cases in which the presence of an underlying chromosomal abnormality was a possibility and its presence had to be confirmed or ruled out. If the final diagnosis of a nonchromosomal condition was reached by ultrasound, no karyotyping was performed. According to the gestational age of the invasive procedure and the need for fetal blood, chorionic villus sampling, amniocentesis, or cordocentesis was performed.

Diagnostic confirmation and outcome information were obtained reviewing pediatric records and pathology reports. Additional sources of outcome data included telephone interviews with parents and referring pediatricians.

The study was approved by the institutional review board.

Statistical analysis was performed with a software package (SPSS 14.0; SPSS, Chicago, IL). Descriptive parameters are expressed as mean or median (±SD) where appropriate. Frequencies are given as number of cases or percentages. Comparison between continuous, nonparametrical variables has been conducted using Mann-Whitney test. A P value of < .05 has been considered statistically significant.

Data collection was carried out according to the principles of the Declaration of Helsinki and a standard informed consent was obtained from all patients.

Results

Mean maternal age at the time of diagnosis was 31.5 years (range, 18–46 years), while gestational age at diagnosis ranged from 13-33 weeks (median, 21.5 weeks). A total of 33 patients (33%) were specifically referred to our unit due to the suspicion of acral MUE raised at screening ultrasound. However, in most cases (67 cases; 67%) indication for referral was the presence of congenital anomalies of other organ systems. In only 1 patient, referred because of hydrocephalus, periconceptional exposure to cyclophosphamide (600 mg/m ²/mo) taken because of severe lupus nephritis was disclosed. None of the other patients referred any occupational/nonoccupational environmental risk or periconceptional exposure to teratogens. Family history was negative for consanguinity, genetic defects, chromosomal abnormalities, or other inheritable malformations in 92 of the 100 patients (92%). In 4 of the 8 remaining cases (4%), a positive family history of fetal akinesia deformation sequence (FADS) in ≥1 previous pregnancies was present; in 1 case, there was a positive family history for hands and feet polydactyly with dominant autosomal inheritance; in another, there was a positive family history for isolated clinodactyly of the fifth finger. One patient was affected with Holt-Oram syndrome (phenotypically characterized by congenital heart disease and syndactyly), while another showed the same terminal defect (absent hand and distal third of the forearm) sonographically detected in her fetus.

MUE embryologically due to a failure of differentiation were the most common, occurring in 52 fetuses (50%) with syndactyly being observed in 10 cases ( Figure 1 ) and contractural/postural defects being found in 42 cases. The latter group included arthrogryposis (17 cases), clenched hands (12 cases), camptodactyly (7 cases) ( Figure 2 , A and B), clinodactyly (4 cases) ( Figure 2 , C and D), and radial deviation of the thumb (2 cases). Abnormalities due to a failure of formation of parts were diagnosed in 31 cases (31%), comprising 16 (16%) cases of longitudinal defects and 15 (15%) cases of transverse defects. Radial deficiency accounted for all the diagnoses of longitudinal defects (16 cases; 16%), and in 4 cases it was associated with ulnar deficiency ( Figure 3 ). Transverse defects were diagnosed in 16 cases (16%) and comprised ectrodactyly (9 cases, including 1 case of lobster claw malformation) ( Figure 4 , A–D), absent fingers/phalanges or digital buds (4 cases) ( Figure 4 , E and F), and classic transverse reduction defects involving the hand or the hand and the distal forearm (2 and 1 cases, respectively) ( Figure 5 ). The difference between ectrodactyly and absence of digits or digital buds is that in the former the central (third and/or fourth) digits of the hand are longitudinally affected, whereas ulnar or radial rays, or parts of the digits are affected in the latter category. Duplication defects comprised 16 cases (16%) of polydactyly ( Figure 6 ). The distribution of upper extremity defects according to Swanson classification is summarized in Table 1 .