Congenital limb malformations are rare, and the clinical presentations range from isolated malformations to complex syndromes and aneuploidies. Because there are limited data available, the aim of this study was to analyze the perinatal outcomes of infants with congenital limb malformations.
According to our data, cases with congenital limb malformations were shown to have a rather high rate of induced abortion, which is understandable, given the selected patient group and our high level of perinatal care. This assumption is supported by our data, showing that only 83.9% of cases had available follow-up data. Apart from this theory, there remains the possibility that patients were treated at outpatient departments or were only seeking a second opinion at our department, making it impossible to follow up their cases. Our study showed a higher loss-to-follow-up rate than those of studies by Zelop et al. [8] and Sharma et al. [14], who reported 96.3% and 88.5% follow-up rates, respectively. A possible explanation could be that the followed cases had lower rates of syndromes and bilateral malformations as well as a higher rate of both upper and lower limb malformations. In these cases, parents were more likely to terminate the pregnancy.
For statistical analysis, we stratified perinatal outcomes by IUFD, TOP, or live birth, and found that younger maternal age was associated with more favorable outcomes; this might be related to the higher rate of more complex syndromes in older women (data not shown) [11]. According to the literature, the mean maternal age of women with a congenital malformation of the fetal limb ranged between 22.1 and 31.5 years [9, 10, 13, 20]. Of note, positive family history of congenital limb malformations showed no association with the pregnancy outcome (i.e., IUFD versus TOP versus live birth). When comparing our results with those reported in the literature, we found a higher rate of family history compared with those of studies by Paladini et al. [9] and Kutuk et al. [10], who reported rates of 8% and 9.8%, respectively. It can be argued that our study cohort included a high number of cases with clubfeet compared with these studies that focused on the outcomes associated with ULM. Sharma et al. [14] reported a higher family history rate of 16.3%, while Sharon-Weiner et al. [11] reported rates of 9.2% and 4.6% for parental and sibling affectedness, respectively. These studies investigated the outcomes of fetuses with clubfeet and found a higher family history rate, supporting this theory.
We found that bilateral limb malformations were associated with worse outcomes and higher termination rates compared with unilateral cases; the TOP rate was significantly higher, and the live birth rate was significantly lower in BLM compared to ULM and LLM. In our study, the IUFD rate was significantly higher in the ULM group compared to the others. The live birth rate was the highest in isolated LLM, followed by ULM and BLM, which is consistent with the findings of Bakalis et al. [12] and Sharma et al. [14]; this might most likely be attributed to the high incidence of isolated clubfeet in the LLM group.
The most commonly occurring malformation was clubfoot in all three groups (IUFD, TOP and live birth), a finding that accords with the existing literature reporting that clubfeet are the most common pediatric foot malformations with a prevalence of 0.6 to 1.5 per 1,000 live births [21]. The second and third most common malformations in our study were longitudinal reduction defects of the upper limb and of the lower limb, respectively, in all groups. Our findings lead to the assumption that clubfoot, in particular, is associated with good obstetric outcomes, as we found a high proportion of clubfeet in live birth cases.
We also found higher rates of IUFD and TOP in cases with a fetal syndrome, as has previously been described [7-12, 14, 16, 17]. A majority of existing studies [8, 11, 13, 14, 17, 18] reported trisomy 18 as the most common aneuploidy associated with congenital limb malformations, and this was consistent with our data. Paladini et al. [9] and Dicke et al. [17] reported trisomy 13 to be the second most common aneuploidy associated with limb malformations. With regard to the perinatal outcome data of live-born infants, the live birth cases in our study showed a high preterm delivery rate of 42.4%, remarkably higher than the overall preterm delivery rate in Austria.
In part, the increased rate of preterm delivery in our study might be explained by contributing risk factors for preterm delivery (e.g., previous preterm delivery) that are criteria for registration for a planned delivery at our tertiary referral center [22]. Compared with the results of Sharma et al. [14], we also found a rather high preterm delivery rate in cases with isolated clubfeet, attributed to the fact that we are the largest perinatal center in the region and that cases with other comorbidities and reasons for preterm delivery were included. We are aware that many of the reported preterm deliveries might have been iatrogenic due to preeclampsia, cervical insufficiency, intrauterine growth retardation, maternal comorbidities, imminent fetal asphyxia or for various other reasons.
With regard to localization of the limb malformation, we found that the perinatal outcome was worse in cases that affected the upper or both limbs compared with isolated LLM (e.g., clubfeet) that more frequently resulted in live birth. Koskimies et al. [16] found a lower perinatal mortality rate of 13.3% in LLM compared with 14.8% in ULM. In the subgroup analysis of live birth cases, preterm delivery was found in 75% of ULM, 30.6% of LLM, and 53.3% of BLM cases, again demonstrating the superior outcomes of infants with LLM. From a clinical perspective, preterm delivery should be anticipated, including timely information of parents in cases with ULM and BLM. Evaluating the Apgar score at 5 minutes, a well-known short-term outcome parameter for both obstetricians and neonatologists, we found also found superior outcomes in LLM cases compared to ULM and BLM (Table 3), suggesting that isolated LLM is unlikely to worsen the neonatal outcome. Considering together the cases with NICU transfer and comfort terminal care, rates were again higher in the BLM and ULM groups than in the LLM group.
We are aware that our study has several limitations. First among them is the retrospective study design that could have led to selection bias and false conclusions. Furthermore, our observational report of cases could have benefited from a matched control group. The malformations that we reported were diagnosed using ultrasonography; fMRI reports were only occasionally available; postnatal confirmation of prenatally diagnosed malformations would have been beneficial as well as standardized fMRI measurements in all analyzed cases. Finally, we could provide neither long-term outcomes nor chromosomal testing results of all observed infants, possibly resulting in the underestimation of the number of associated syndromes, especially in the TOP and IUFD groups.
Despite these limitations, our study has strengths, including its implementation at a single tertiary center with a high number and various types of malformations. Treatment at our center was associated with examinations that were exclusively conducted by well-trained and certified examiners, which is of paramount importance in this context. Therefore, our data are homogeneous and reliable, which are particularly important in observational studies with relatively small sample sizes.