Associated anomalies in Pierre Robin sequence

Pierre Robin sequence (PRS) is frequently co‐occurring with other non‐PRS congenital anomalies. The types and the prevalence of anomalies co‐occurring with PRS vary in the reported studies. The aims of this report was to study the types and the prevalence of the anomalies co‐occurring with PRS in a well‐studied population northeastern France. The types and the prevalence of anomalies co‐occurring in cases with PRS were ascertained in all terminations of pregnancy, stillbirths and live births in 387,067 births occurring consecutively during the period 1979–2007 in the area covered by our registry of congenital anomalies which is population‐based, 89 cases of PRS were registered during the study period with a prevalence of 2.29 per 10,000 births, 69.7% of the cases had associated non‐PRS anomalies. Chromosomal abnormalities were present in 10 (11.2%) cases including three 22 q11.2 deletion. Non‐chromosomal recognizable conditions were diagnosed in 27 cases (30.3%) including 10 Stickler syndrome, 8 Treacher Collins syndrome, 3 cases with short stature and 6 other syndromes. Multiple congenital anomalies (MCA) were present in 25 cases (28.1%). The most frequent MCA were in the ear, face and neck (35 out of 98 anomalies, 35.7%), cardiovascular (18 anomalies, 18.4%), musculoskeletal (11 anomalies, 11.2%), central nervous (7 anomalies, 7.1%), urinary (6 anomalies, 6.1%), and eye (6 anomalies, 6.1%) system. The high prevalence of associated anomalies justifies a thorough screening for other congenital anomalies in cases with PRS.

. PRS may occur in isolation or in the context of other, associated anomalies. Among the various reported studies the proportion of isolated versus nonisolated PRS is highly variable as shown in Table 1. Mortality associated with Robin sequence is reported to be between 1.7% and 65% (Costa et al., 2014). Surgical intervention has been reserved for patients with severe airway obstruction in which conservative treatment has been unsuccessful. The prognosis of patients with idiopathic PRS without associated anomalies is usually good or with slight handicap but outcome may be bad in patients with associated anomalies (Scott & Mader, 2014).
The aim of this study was to determine the prevalence, the frequency and the type of congenital anomalies co-occurring in cases with PRS in a well-studied population.

| CASES AND METHODS
Our study included all cases of PRS collected from 387,067 consecutive pregnancies of known outcome recorded through our registry of congenital malformations (RMCBR), previously described (Stoll, 1985).
The newborns of 11 maternity hospitals were examined from January 1, 19791, , to December 31, 2007. The region of investigation was the area defined by the "Departement du Bas-Rhin" which includes the urban area of Strasbourg and surrounding rural areas. This research project was reviewed and approved by the Ethics Committee of the Medical Faculty of Strasbourg following the World Medical Association Declaration of Helsinki. Parental informed consent is mandatory for RCMBR. All newborns including live births and stillbirths of at least 22 weeks of gestation were registered within the first 8 days postpartum, as well as all TOPFA (Terminations Of Pregnancy for Fetal Anomaly), regardless of gestational ages. In this area all deliveries take place in maternity hospitals resulting in a complete ascertainment. A clinical geneticist examined every case and interviewed pediatricians/neonatologists. Surveillance for anomalies was carried out until 2 years of age. Every case was reported to RMCBR.
Analysis of the data was done retrospectively. In the region under study, in the mid-trimester of pregnancy, a prenatal ultrasound examination for congenital anomalies was proposed to every pregnant woman. When a suspected or confirmed case was reported, information was obtained from all available records. PRS was defined by the triad of micrognathia, glossoptosis and concomitant airway obstruction with or without cleft palate. Cases with anomalies were subdivided into "isolated," when only PRS was present, and "associated," when one or more additional non-PRS major anomalies were recognized. The cases with non-chromosomal syndromic associated anomalies were classified as having either a syndrome, a sequence, an association, or a spectrum (Hennekam et al., 2013). The nonsyndromic multiple congenital anomalies (MCA), were sub-classified according to the main organ system affected. For each case with associated anomalies, a complete description was obtained, including photographs, radiographs, karyotype, and autopsy. However, at the period of the study, chromosomal microarray (aCGH) was not available in our region as well as molecular testing, WGS and whole exome sequencing. A major anomaly was defined as a structural defect of the body and/or organs that impair viability and require intervention, T A B L E 1 Isolated, syndromic, associated and non-isolated congenital anomalies in the studies with Pierre Robin sequence recently reported. need to be treated (Queißer-Luft & Spranger, 2006). Major non-PRS anomalies within a system were counted as one defect. For example, a case with hydronephrosis and kidney agenesis was counted once as kidney agenesis and a case with omphalocele, intestinal atresia, and malrotation was counted once as omphalocele (Lowry et al., 2013). A case with a Mendelian disorder that includes multiple anomalies e.g. Stickler syndrome (MIM#609508) was classified as having a recognizable non-chromosomal condition. Minor congenital anomalies such as cryptorchidism were excluded. Intellectual disability was not included because it is difficult to assess in infancy.
The prevalence of the anomalies most frequently associated with cases of PRS with MCA as for example ventricular septal defect, was compared with the prevalence of these anomalies (here ventricular septal defect) in the population under study.
For the calculation of prevalence rates the total number of births including livebirths, stillborn and termination of pregnancy for fetal anomalies from 1979 to 2007 was used: 387,067.
SAS Version 8 (SAS Institute, Inc, Cary, NC) was used for statistical analysis. The chi-squared test was used for comparisons.

| RESULTS
Eighty-nine cases of PRS were registered resulting in a prevalence at birth of 2.29 per 10,000 including 27 (30.3%) isolated, idiopathic cases of PRS (Table 2) and 62 cases (69.7%) with associated anomalies (Table 3)  There were no changes in prevalence over time.
In 37 (59.7%) out of the 62 cases with associated anomalies, the anomalies could be included into a recognizable pattern or syndrome. Table 4 shows the observed prevalence of the anomalies more commonly associated with PRS compared with the expected prevalence of these anomalies in the population under study.

| DISCUSSION
The congenital anomalies included in the triad defining the PRS are often associated with other congenital anomalies. These associated anomalies are responsible for most of the morbidity and mortality associated with PRS (Costa et al., 2014). In the population studied, the prevalence of PRS was 22.9 per 100,000 births and 69.7% of the cases had associated anomalies. However, the data used for estimating prevalence were old. The prevalence of PRS as well as the pro-  Paes et al. (2015). No duplications appear in these two reviews. We added 7 studies published in and after 2014 to the reviews of Gomez-Ospina and Bernstein (2016) and of Paes et al. (2015) and the study of Thouvenin et al. (2013) which was not included in the reviews of Gomez-Ospina and Bernstein (2016) and of Paes et al. (2015). The 16 studies on PRS reported from 1988 to 2015 by Gomez-Ospina and Bernstein (2016) were all based on hospital records, 21 of the studies reviewed by Paes et al. (2015) were based on hospital records, 21 on congenital malformations registries, 4 on social services, 3 on birth certificates, one on the National Healthcare System, one on an institute for speech disorders, and one on records of insurance companies (some authors used several data sources). Six studies only, based on well-studied population, on all types of PRS, isolated or associated, were reported (Bush & Williams, 1983;Kallen et al., 1996;Lary & Paulozzi, 2001;Paes et al., 2015;Scott & Mader, 2014;Vatlach et al., 2014). The prevalence per 100,000 births of PRS calculated by dividing the total number of cases of PRS by the total number of births ranged from 11.6 in The Netherlands  to 37.2 in East Scotland (Wright et al., 2018) (Table 5)     . However, only 17% of the cases with PRS examined by Shprintzen (1992) in New York had asso- However, in 16 out of the 42 studies reviewed by Paes et al. (2015) providing the number of males and females, there were no proportion difference between males and females.
In this series, chromosomal abnormalities were observed in 10 cases, 11.2%. The reported percentages of chromosomal abnormalities in cases with PRS varied ranging from 0.9 in Oslo (Filip et al., 2015) to 16.6 in Indianapolis (Costa et al., 2014). In this series the most common chromosomal abnormality was 22q11.2 deletion. In the previously reported series the most common chromosomal abnormalities were also 22q11.2 deletion, 30 out of 170 cases (17.6%) of PRS with chromosomal anomalies reported in the literature were 22q11.2 deletion, with a percentage ranging from 10.0 in Indianapolis (Costa et al., 2014) to 100.0 in Denmark (Printzlau & Andersen, 2004) and in Oslo (Filip et al., 2015). Trisomy 21 and trisomy 18 were less frequent with a percentage ranging for trisomy 21 from 3.4 in Cincinatti (Weaver et al., 2022) to 10 in Indianapolis (Costa et al., 2014) and for trisomy 18 from 3.4 in Cincinatti (Weaver et al., 2022) (Printzlau & Andersen, 2004). For the anomalies of the central nervous system, the percentages of reported anomalies varied from 6.2 in Denmark (Printzlau & Andersen, 2004) to 19.6 in Indianapolis (Costa et al., 2014) with variation for Arnold Chiari anomaly from 12.1 in Indianapolis (Costa et al., 2014) to 20.0 in Cincinnati (Weaver et al., 2022). For the musculoskeletal system, the percentages of anomalies reported varied from 11.2 in this series to 34.4 in Europe (Santoro et al., 2021) with variation for syndactyly from 6.8 in Europe (Santoro et al., 2021)  The studies reported before 2015 except Thouvenin et al. (2013) were reviewed by Paes et al. (2015) and by Gomez-Ospina and Bernstein (2016). b Per 100,000 births.  The higher frequency of additional malformations in this series may be related to the fact that our study was population based, that the ascertainment of cases was complete, that all patients were taken into consideration, including stillborn and terminations of pregnancy, and the surveillance for anomalies was continued until 2 years of age.
In order to evaluate patients and to compare studies, it is necessary to standardize the methods of case classification. Congenital anomalies and associated anomalies must be grouped into meaningful syndromes and conditions. Methods for case classification into isolated, multiple and syndrome categories were described (Rasmussen et al., 2003). Consideration of these guidelines will lead to more comparable case groups, an important element of careful studies aimed at identifying etiology of congenital anomalies. Some studies have shown an association between maternal use of methadone and increased risk of PRS (Cleary et al., 2020). A number of other environmental risk factors and medical conditions have been related to the risk of PRS in some studies, but not in others.
Maternal smoking has been associated with cleft palate with PRS (Honein et al., 2007). Fetal alcohol syndrome was reported in 10% of a US sample of 100 children consecutively diagnosed with PRS (Shprintzen, 1992) and in 3 out of 110 cases of PRS of a French case series (Holder-Espinasse et al., 2001).
A discussion on the molecular basis of PRS is beyond the scope of this paper. As mentioned above many syndromes with identified genes were reported to be associated with PRS. Our study comprised many strengths: the registration of cases was active, a clinical geneticist examined every case, a geographically well-characterized population was studied, all cases were ascertained by the Registry of Congenital Malformations including termination of pregnancies for fetal anomalies, stillbirths, and live births, and the surveillance for anomalies was extended until 2 years of age. This study had potential limitations as a small number of patients were included.
However, complete ascertainment was achieved and a homogeneous population was studied.

| CONCLUSION
In conclusion, the prevalence of PRS at birth was 2.29 per 10,000 in the population studied and the percentage of cases of PRS with associated anomalies was 69.7%, 7 in 10 cases. These figures were obtained from the study of a well-defined population of close to 400,000 births. The anomalies associated with PRS represent a large number of different conditions. An anomaly syndrome or pattern was recognizable in 59.7% of the cases of PRS with associated anomalies.
Therefore, cases with PRS should have a careful multidisciplinary checkup. A search for associated congenital anomalies, particularly of the ear, face and neck, the cardiovascular, the musculoskeletal, the urogenital, and the central nervous systems is recommended for the cases with PRS. Genetic counseling may be indicated in many of the cases of PRS with associated anomalies.

AUTHOR CONTRIBUTIONS
The authors contributed equally to data gathering, study design and analysis, revisions to the manuscript.

ACKNOWLEDMENTS
We thank Professor Nicholas Meyer for his assistance in statistical management.
T A B L E 6 (Continued) The studies reported before 2014 were reviewed by Gomez-Ospina and Bernstein (2016) and by Paes et al. (2015). b The number of cases is under the authors names. For recognizable conditions the percentage after total is the percentage of the total number of cases.
For non-recognizable patterns the percentage is the percentage of the total number of anomalies. c Septal defects. d Genitourinary.