Perinatal features of Prader-Willi syndrome: a Chinese cohort of 134 patients

Background Prader-Willi syndrome (PWS) is a rare and complex genetic disorder caused by lacking expression of imprinted genes on the paternally derived chromosome 15q11-q13 region. This study aimed to characterize the perinatal features of 134 Chinese individuals with PWS. Methods This study included the patients of a PWS registry in China. Anonymous data of 134 patients were abstracted. Perinatal and neonatal presentations were analyzed, and compared between the two PWS genetic subtypes. We also compared the perinatal features of PWS patients with the general population and other previous reported large cohorts from France, UK and USA. Results This study included 134 patients with PWS (115 patients with 15q11-q13 deletion and 19 with maternal uniparental disomy). Higher mean maternal age was found in this cohort (30.5 vs. 26.7), particularly in the maternal uniparental disomy (UPD) group (36.0 vs. 26.7) comparing with the general population. 88.6% of mothers reported a decrease of fetal movements. 42.5 and 18.7% of mothers had polyhydramnios and oligohydramnios during pregnancy, respectively. 82.8% of the patients were born by caesarean section. 32.1% of neonates had birth asphyxia, 98.5% had hypotonia and 97.8% had weak cry or even no cry at neonatal period. Feeding difficulty existed in 99.3% of the infants, 94.8% of whom had failure to thrive. 69.4% of the infants ever used feeding tube during hospitalization, however, 97.8% of them discontinued tube feeding after discharge. Maternal age and pre-pregnancy weight were significantly higher in the UPD group (both P < 0.05). Conclusions Differential diagnosis of PWS should be highlighted if infants having following perinatal factors including polyhydramnios, decreased intrauterine fetal movements, caesarean section, low birth weight, feeding difficulty, hypotonia and failure to thrive. Higher maternal age may be a risk factor of PWS, especially for UPD. Further studies are needed for elucidating the mechanism of PWS.


Introduction
Prader-Willi syndrome (PWS) is a rare and complex genetic disorder characterized by severe hypotonia and feeding difficulties in early infancy, followed by excessive eating and gradual development of morbid obesity in early childhood, together with a series of comorbidities including short stature, typical facial dysmorphism, psychomotor delay, behavioral abnormalities and cognitive impairment [1]. It is caused by an absence of imprinted genes expression on the paternally derived chromosome 15q11-q13 region. PWS was divided into several genetic subtypes: deletion of the paternal copy of 15q11-13 in about 65% of the cases, maternal uniparental disomy (UPD) of chromosome 15 in about 30%, imprinting center defect in less than 5% and very rare cases of translocation involving the chromosome 15q11-q13 region [2,3]. Genetic diagnosis for PWS is now easily available, in which diagnosis can be made in most patients within the first months of life even in prenatal period if obstetricians or neonatologists can recognize the perinatal features of PWS well. Increased awareness among obstetricians and healthcare providers would promote earlier diagnosis and treatment of PWS by pediatricians/neonatologists [4]. Early diagnosis and tailored multidisciplinary treatment are extremely important for better life quality of the infants with PWS as they ensure comprehensive advice to prevent obesity and to stimulate cognitive and adaptive skills [4,5].
It is imperative to make more obstetricians, pediatricians/neonatologists be aware of the perinatal features of PWS. Our study aimed to characterize the perinatal features in a cohort of Chinese individuals with PWS. We also compared the perinatal features of PWS patients with the general population and other previously reported large cohorts from France, UK and USA. This is the first large study on patients with PWS in China.

Subjects
This study is part of a project started by PWS Research Group from Children's Hospital, Zhejiang University School of Medicine. The PWS Research Group established a PWS Registry for patients in China. The study was approved by Institutional Review Board of Zhejiang University School of Medicine (no. 2018-IRB-055). Written informed consents were obtained from all parents or the patients (above 8 years) registered in the PWS Registry.
We included 134 patients from the PWS Registry into this study. All of the patients were confirmed by molecular genetic diagnosis. MS-PCR or MS-MLPA was used for initial diagnosis. For those with PWS methylation pattern but non-deletion type after initial diagnosis, microsatellite linkage analysis was used to differentiate UPD or an imprinting defect. Perinatal and neonatal presentations were analyzed and compared between the two PWS genetic subtypes as well as general population [6][7][8][9][10] and data reported from other countries [5,[11][12][13][14]. The analysis excluded one female and six males diagnosed only by methylation test.

Statistical analysis
All of the data were analyzed by the SPSS (version 16.0) and R3.3.1 software. The categorical variables were summarized using frequencies and continuous variables were expressed as mean ± SD scores. The statistical differences between groups were compared by student t-test for continuous variables and Pearson χ2 test for categorical variables. Fisher's exact test was performed to analyze differences of categorical variables when the expected cell count was less than 5. We applied one way ANCOVA to compare maternal pre-pregnancy weight and BMI between UPD and Deletion groups adjusted by maternal age. P value or FDR P value values were all two-sided in this study with statistically significantly level of 0.05.

Results
This study included 134 patients with PWS including 115 (85.8%) with 15q11-q13 deletion and 19 with maternal UPD (14.2%). The studied patients' birth dated from October 1997 to August 2018; among them, 117 patients' birth dated between January, 2009 and August 2018. Among these patients, 73 (54.5%) were boys (62 with deletions and 11 with maternal UPD) and 61 (45.5%) girls (53 with deletions and 8 with maternal UPD). The mean age of diagnosis was 31.75 ± 4.72 months with a range from 10 days to 17 years.
Higher maternal age was found in this cohort (30.5 vs. 26.7), particularly in the UPD group (36.0 vs. 26.7), when comparing with general population. 87.9% (109/124) of the mothers reported a decrease of fetal movements during pregnancy. 42.5 and 18.7% of mothers had polyhydramnios and oligohydramnios during pregnancy, respectively. 82.8% of the patients were born by caesarean section, significantly higher than 34.9% in 2014 [7]; only 17.2% of the patients were born by vaginal delivery (23/ 134), and 30% of them by forceps. 2.2% of mothers had hypertension during pregnancy, 5.2% had gestational diabetes and 9.7% had premature rupture of membranes (Table 1). This cohort of patients had a higher rate of prematurity (16.4%) and low birth weight (34.3%) compared with 7.2 and 7.1%, respectively in China. The patients revealed a higher rate of birth asphyxia (32.1%) than the general population of neonates in 2014 reported by Ministry of Health of China (6.3%) [6]. 94.8% of the patients were hospitalized after birth with a median duration of 17 days (range: 0-90 days). 94.8% of the patients had failure to thrive which was greatly higher than the general population (8.1%) [9]. Only 11.2% of the patients were purely breastfed at the first three months of life, which is lower than the general population (58.5%) [9]. Feeding difficulty existed in 99.3% of the patients. 69.4% of the patients used feeding tube, however, 97.8% of them discontinued feeding tube and used silicone bottle, spoon and even a syringe for feeding at home. Only two of them continued using feeding tube after hospitalization. 98.5% of the patients had hypotonia and 97.8% had weak cry during neonatal period.

Discussion
PWS can now be diagnosed at a very early age or even at prenatal period benefiting from the improved molecular diagnosis techniques and increased awareness of PWS features [5,[14][15][16]. Dobrescu et al. and Gold et al. reported that most patients still had a delayed diagnosis because of the early signs of PWS not recognized, inappropriate molecular diagnostic methods used and lacking of expertise in some hospitals and institutions [4,17]. Early diagnosis permits early care and treatment, which may reduce the hospital stay and tube feeding duration, hence preventing growth retardation and early onset of obesity [5,[18][19][20]. To our knowledge, this is the first large cohort study on perinatal features of PWS patients in China.
Recognizing perinatal features of PWS will be helpful for early diagnosis and multidisciplinary care of affected infants. The results of our cohort showed high rates of polyhydramnios (42.5%), decreased fetal movements (87.9%), caesarean section (82.8%), low birth weight (32.8%), feeding difficulty (99.3%), hypotonia (98.5%) and failure to thrive (94.5%). The results were similar to previously reported cohorts from the UK, France and USA. Dudley et al. reported high rates of polyhydramnios, induced labor and caesarean section, diminished fetal movement in a cohort of 86 French patients with PWS [11]. Gross et al. found decreased fetal movements, small for gestational age, asymmetrical intrauterine growth and polyhydramnios in an Israel cohort [19]. Gold et al. and Sign et al. reported significantly higher rates of caesarean section, polyhydramnios, decreased fetal movements, low birth weight, feeding difficulty, hypotonia and low Apgar scores in two cohorts of PWS patients in USA [13,14].
Hence, differential diagnosis of PWS should be highlighted if infants having following perinatal factors including polyhydramnios, intrauterine decreased fetal movements, caesarean section, low birth weight, feeding difficulty, hypotonia and failure to thrive. Besides polyhydramnios, we also found a far higher rate of oligohydramnios (18.7%) in our cohort comparing with the general population (0.4-4.0%) in China. The accurate mechanism of polyhydramnios and oligohydramnios is still unclear although reduced fetal swallowing is regarded as a cause of polyhydramnios. High rate of oligohydramnios in this cohort may be due to the high rate of premature rupture of memberanes. Further study for the mechanism of amniotic fluid disorder is required. Moreover, caesarean section rate was higher in our cohort comparing with normal populations [7]. The high caesarean section rate in this cohort was due to the high rate of obstetric complications in PWS including abnormal amnion (61.2%), decreased fetal movements (87.9%), premature rupture of membranes (9.7%) or abnormal fetal heart rate/rhythm [20][21][22] although the data of fetal heart rate was not recorded in our study. High caesarean section rates may also be associated with the concepts and selection of pregnant women in China.
Considering neonatal complications, failure to thrive, low birth weight, feeding difficulty and hypotonia were most common in our study. Moreover, rate of asphyxia was higher in our cohort than that in general population in China (32.1% vs. 6.3%) [6]. This may be associated  [11] United Kingdom (N = 46) (2008) [12] France (N = 61) (2017) [5] United States (N = 64) (2018) [13] United States (N = 355) (2018) [14] Mean with higher ratio of intrauterine abnormal fetal heart rate/rhythm and decreased fetal movements, hypotonia and weak cry after birth, which may be considered as intrauterine or intrapartum asphyxia and higher ratio of preterm in our cohort as well. More than 99% of the infants had feeding difficulty requiring tube feeding and about 70% of the infants ever used feeding tube during hospitalization. Tube feeding was discontinued in 98% of these infants by their parents. Parents refused to use tube feeding at home, and they preferred using silicone bottle, spoon or even a syringe for feeding PWS infants which could not reach the effect of tube feeding. As a result, the rate of failure to thrive (94.5%) during the first months of life in this cohort of patients was higher than that reported by Sign et al. (77%) [14]. The possible reasons for refusing tube-feeding at home may be lacking education of tube-feeding from healthcare workers, in which parents are not familiar with home tube feeding techniques. Therefore, it is very important to strengthen feeding knowledge and ability of parents of PWS patients in China. Differences between the deletion and UPD subgroups are still conflicting in literature. Gillessen-Kaesbach et al. and Whittington et al. showed that UPD patients had a significantly higher birth weight and maternal age than that of deletion patients [12,22]. Vise versa, Gunay-Aygun et al. observed a significantly lower birth weights and lengths in the UPD group than those in the deletion group [22]. Dudley and Muscatelli found a higher rate of induced labor, premature labor and higher maternal age in the UPD group, and a high rate of low birth weight in the deletion group [11]. However, Sign et al. only found higher maternal age and pre-pregnancy weight in the UPD group of a large cohort containing 355 PWS patients from USA [14]. Whittington et al. found significant difference of maternal age and birth weight between different genetic subtypes [12]. In our study, higher maternal age and pre-pregnancy weight were noted in the UPD subgroup than that in the deletion subgroup. However, the rate of hospitalization at neonatal period and feeding tube use was higher in the deletion group with marginal differences (0.05 < P < 0.15). These results above implied that patients with deletion genetic type may have more difficulty in feeding and are prone to be hospitalized at neonatal period. Large cohort study is still needed to confirm these findings.
The mechanism of PWS still remains unknown. We observed a high maternal age of PWS patients, especially in the UPD group, which was also reported in several other reports [12,14,23]. These results imply that the advanced maternal age is associated with PWS resulting from non-disjunction at meiosis 1. Moreover, we also found that mothers of PWS patients had higher prepregnancy BMI than those in the general population and higher pre-pregnancy weight in the UPD group, which was also reported in a large cohort of 355 PWS patients from USA [14]. These suggested that high maternal age and overweight may be the risk factors of PWS. Further studies for the mechanism of PWS are required.

Conclusions
In conclusion, our study highlighted the perinatal features in a large cohort of Chinese patients with PWS. We found high incidence of decreased fetal movements, polyhydramnios and delivery by caesarean section, and higher maternal age comparing with the general population. Neonatal features found in our cohort included low birth weight, birth asphyxia, failure to thrive, feeding difficulty, weak cry and hypotonia. We demonstrated that mothers of patients with UPD had higher maternal age, pre-pregnancy weight comparing with the deletion group. Most patients required tube feeding during hospitalization, however tube feeding was discontinued by their parents after discharge at home. Nutrition deficiency was a serious problem in infants with PWS.
Home tube-feeding instruction should be carried out by parents of PWS patients in China. Our study is helpful to better understand the perinatal features of PWS in China, which will benefit for early diagnosis and treatment. Moreover, our results provide valid evidence for further research and promote PWS screening into the newborn screening program in China.