Abstract
Background:
Congenital myotonic dystrophy (CDM) is the neonatal onset and most severe presentation of Myotonic Dystrophy type 1. Since it first description, perinatal complications have been detailed including prolonged hospital stay, respiratory and feeding therapy during the neonatal period, although long-term complications are less documented.
Objective:
Present a prospective cohort of CDM and compare it to the literature of other CDM case series, to adequately describe and contrast the prenatal, neonatal and infancy features of CDM.
Methods:
A 5-year cohort of CDM eligible cases was conducted via the Canadian Pediatric Surveillance Program. 38 patients met the inclusion criteria. Comparison to other CDM case series published in the literature between 1992 and 2016 about perinatal and infancy morbidity.
Result:
From a total of 118 cases, the most frequent features were Polyhydramnios (58%), feeding therapy (77%), intubation and ventilation (58%); neonatal death was reported in 16% of the cases; the most frequent long-term morbidity were respiratory tract infections.
Conclusions:
We performed a detailed description of the main perinatal features of CDM and precise documentation of the mortality and morbidity during the first five years of life. This is an essential step in the knowledge of the natural history of CDM.
Keywords
Abbreviations
Myotonic Dystrophy type 1
Congenital Myotonic Dystrophy
CTG repeat size
Canadian Paediatric Surveillance Program
INTRODUCTION
Congenital myotonic dystrophy (CDM) is the neonatal onset and most severe presentation of Myotonic Dystrophy type 1 (DM1, OMIM 160900). Resulting from an unstable trinucleotide (CTG) repeat expansion on the 3’UTR of the DMPK gene (19q13.3), DM1 is inherited in autosomal dominant fashion trait and has a worldwide prevalence of 1:8000, [1] and an incidence of 2.1/100,000 live births in Canada [2]. In the majority of the cases, CDM is transmitted from a maternal origin, and recently it was proposed that this be caused due methylation patterns specific to the germ cells, occurring an aberrant CpG methylation pattern in the oocytes; on the contrary, it is proposed that the sperms of affected fathers may escape this aberrant methylation [3].
The clinical presentation of CDM is characterized by neonatal hypotonia, profound weakness, feeding difficulties and mechanical respiratory failure that may require intubation and ventilation immediately after birth [2].
Since it was first described by Vanier in 1960 [4], there have been efforts to detail the clinical presentation of CDM delineating the main clinical features, laboratory findings, and prognosis. First publications described polyhydramnios and decreased fetal movements as the most common features of the prenatal period; hypotonia and respiratory distress during the neonatal period; intellectual disability during infancy; and respiratory failure as the principal cause of death [5–7]. Historically, the diagnosis was base on clinical presentation and EMG criteria, until 1992 when different research groups introduced a genetic diagnosis [8–10].
In the post-genetic confirmation era, publications are primarily focused on prenatal and neonatal complications. Prolonged hospital stay, respiratory and feeding therapy during the neonatal period, and long-term complications are less widely documented.
The purpose of this study is to present a comprehensive review of CDM natural history data from studies of different methodologies, describing the prenatal, neonatal, and infancy features of CDM contrasting the prevalence of morbidities between the various methods of data collection cohorts. Specifically, offering detailed natural history from a population based, prospectively collected study with reference to largely retrospective, single centre studies provides a comprehensive delineation of the proportion of morbidities experienced by CDM patients. Further, this is the first review of case reports done in the genetic era (>1992), which allows confidence in the clinical and genetic homogeneity of the reported cases. The knowledge of the natural history of the disease will help to have an accurate understanding of the various morbidities that children and families face and allow clinicians to prognosticate better. Moreover, it is crucial, as we head into clinical trials for novel CDM therapeutics, that we have a comprehensive overview of the natural history.
MATERIAL AND METHODS
A five-year cohort study from 2005 to 2010 of eligible cases of CDM was conducted via the Canadian Pediatric Surveillance Program (CPSP). With the previous study approval from the ethics committee from the CPSP and the Research Ethics Board of Western University (London, Ontario, Canada), physicians were sent an initial report form to complete and return to the CPSP. The inclusion criteria for participants were: 1) Participants must have had a confirmed diagnosis of DM1 2) The symptoms of DM1 were evident in the neonatal period (first month of life) as recognized by hypotonia, respiratory or feeding difficulties requiring admission to the NICU or another hospital ward for greater than 72 hours and 3) The diagnosis was confirmed by genetic testing demonstrating an expanded trinucleotide CTGrs >200 in the DMPK gene in the child or mother [2, 11–13].
During the five-year period, enrollment packages containing letters of information and consent forms were sent to the reporting physician of patients meeting all inclusion criteria. Guardians of all participants provided informed consent. After the initial enrolment, families of participants were contacted by phone every three months for the first year, with subsequent calls every six months until the age of five. For every phone call, a follow-up report form was completed by the Study Coordinator, who registered any pathological changes or complications. The primary physician and/or pediatrician were contacted annually to collect clinical information about the patient, including Hospitalizations, discharge summaries, oximetry, and blood work. These cases were from across the country and neither the Study Coordinator nor Physicians from our team had any direct clinical contact with them over the 5 years. Throughout the CPSP study, there were 121 cases reported. Thirty-eight met the inclusion criteria, including 18 males (47.3%) and 20 females (52.6%), these patients were enrolled in the initial phase of the study, the surveillance of incident cases of CDM (previously reported by Campbell, C et al. [2013]). The second phase of the study consisted of a 5-year cohort study, where 19 families consented to participate. From this cohort, one patient died, and three families where lost over the course of the follow-up. The study has been extended for 10 more years and is currently ongoing, but it won’t be discussed in this report.
We reviewed the medical literature about prenatal, perinatal and infancy findings of cohorts and case series of CDM. The primary search was performed in the PubMed, Medline-OVID and EBSCO databases and was limited to articles in English, Spanish and French language literature published between 1992 and 2016. Keywords searched included: “Congenital Myotonic Dystrophy,” “prenatal diagnosis,” “perinatal,” “neonatal,” “cohort” and “case series.” To include the articles, minimum requirements were: 1) Diagnosis of CDM confirmed by genetic studies; 2) Case report with minimum two cases, case series and cohorts; and 3) The article should include clinical description from at least one period of life: prenatal, neonatal, infantile, and/or childhood. A secondary search was performed by identifying pertinent articles cited in articles identified in the primary search. A total of 80 cases in nine different publications met the inclusion criteria: Esplin et al. [14]; Keller et al. [15]; Schild et al. [16]; Martinello et al. [17]; Sumi et al. [18]; Kroksmark et al. [19]; Zaki et al. [20]; Echenne et al. [21]; and Domingues et al. [22]. The analysis of each one of these articles is represented in Table 1. It is important to mention that not all subjects had information available about each clinical finding. No other study was a prospective active surveillance study of a whole country.
Description of articles from the Medical Literature Review
RESULTS
We present results from the systematic evaluation of findings in our cohort with those from a comprehensive literature review. In our cohort, we recruited a total of 38 patients and found 80 cases in the literature. Of the total 118 patients with confirmed CDM, the gender breakdown is as follows: 56 (57%) male and 42 (42%) female. Gender was not specified in 20 cases. The 38 cases from our cohort where maternal-transmitted origin. Even though not all the cases from the literature review mentioned the parental-origin of transmission, none of them report a paternal pattern of inheritance.
In our cohort, the average mother’s age was 28.5 years old (range 27–35). The diagnosis of DM1 is challenging during pregnancy because most of the patients are unaware they have the disease. This is represented in our cohort, where only 10/37 cases knew they were affected before the pregnancy. Of these cases, eight underwent a prenatal genetic diagnosis either with chorionic villus sampling or amniocentesis. Zaki et al. [20] suggest the presence of the ultrasonographical features such as “myopathic face” (represented as “tent-shaped mouth),” polyhydramnios, and congenital talipes equinovarus, should signal a suspected CDM diagnosis. We agree with these criteria, although these three signs are not present in all the patients, the presence of the three signs strongly suggests the disease. Polyhydramnios is the most frequent sign during pregnancy (58%), followed by hypotonia (represented as decreased fetal movements) present in 38% of the cases.
Cesarean sections were performed in many of the patients (48%) compared to vaginal delivery, in only 38% of the cases. In the cases from our cohort, it is unknown the cause of the cesarean, but it was documented the evidence of fetal distress in the 30% of the cases which may have elevated this procedure. The Apgar score was less than or equal to 6 at minute 1 in 54%, and at minute 5 in 51% of thecases.
Polyhydramnios
Polyhydramnios is described as a typical consequence of different neuromuscular disorders due to reduced or absent fetal swallowing secondary to impaired fetal deglutition [23]. In our cohort, this complication was present in 47% of the cases, compared to the literature, where it was present in 65% of the cases. Zaki et al. [20] reported retrospectively 17 cases of ongoing pregnancies affected by CDM, with 82% marked by some degree of polyhydramnios at some point during the pregnancy, predominantly between 28 and 34 weeks’ gestation. It is important to point out that the primary objective of this study was to describe prenatal ultrasound findings in patients with DM1. On the other hand, on the retrospective study done by Kroksmark et al. [19], only 43% of the cases presented polyhydramnios, which is more consistent with what we found in ourcohort.
Gestational age
In numerous publications, preterm delivery is acknowledged as a frequent CDM complication. In 1998, Rudnik-Schöneborn et al. [24] reported an association between CDM and preterm deliveries present in 55% of their cases, this is not consistent with our cohort and literature review, where we found only 32% of the patients were of moderate to late prematurity with the majority (57%) full term with adequate birth weight (53%).
Fifty-eight percent of the patients required intubation and ventilation, 28% of them for more than 30 days. In our cohort four patients ventilation needed for more than 30 days, three patients died at 30 days of age, three of them because of withdrawal of life support. As previously published by one of the authors [25] and by Keller et al. [15], prolonged ventilation should not be a used as a primary criterion for a prognosis of fatal outcome.
The length of hospital stay tends to be prolonged for patients requiring intubation and ventilation. In our review we found an average of 58 days hospital stay, with the necessity of multidisciplinary support, mainly for ventilation and feedingtherapies.
The neonatal complications, as represented on Table 3, were divided into respiratory, gastrointestinal, neurological, musculoskeletal, cardiac, and “other” complications. The most common complications included: raised diaphragm, congenital talipes equinovarus, and cryptorchidism, with a description of each tofollow.
Other less frequent features found in our cohort were: Dysrhythmia (2/37), Sepsis (3/37), Nephrocalcinosis (1/37), Intraventricular hemorrhage (1/37), Vesicoureteral reflux (1/37), and Seizures (1/37).
Respiratory complications
The respiratory failure in CDM results from a combination of factors: “diaphragmatic and intercostal muscle weakness, pulmonary hypoplasia attributable to insufficient respiratory movement in utero, failure of central respiratory control, and aspiration pneumonia caused by weak, uncoordinated swallowing, and the inability to tolerate oral feeding or even their secretions” [15]. The most frequent respiratory complication reported in these patients was raised-hemidiaphragm. In our review, it was present in 30% of the cases, consistent with previous publications. Other less common respiratory complications present in our study, but not included in Table 2, were: Pulmonary hypoplasia (4/43), pneumothorax (3/41), bilateral perihilar atelectasis (1/40), and bronchopulmonary dysplasia (3/41).
Pregnancy and birth
Gastrointestinal complications
We found that the swallowing disorder is, in general, the main complication during the neonatal period in CDM patients (58/74). It is the consequence of the severe facial and oropharyngeal weakness and profound hypotonia. Other less frequent GI complications are delayed gastric emptying (14%), Gastroesophageal reflux disease (GERD [11%]), and Constipation (11%). Table 2 does not include the latter. Seventy-seven percent of the patients required feeding therapy, and in 69% of them, it exceeded 14 days in duration. Usually, the treatment initiates with a nasogastric tube, which was required in 46/66 of the cases with an average period of 20 days (±3.59). Due to the delayed intestinal transport and reflux saw in these patients, the use of total parenteral nutrition is common and was reported in 14/42 cases.
Ventriculomegaly
It is described that ventricle enlargement is one of the common neuroradiology features for DM1, and when it is diagnosed in the neonatal period usually is after a brain ultrasound; other characteristic findings in the literature are: Atrophy of the cortex, midline, and cerebellum, and neuronal migration abnormalities [26, 27]. Even so, the reports of ventriculomegaly in congenital myotonic dystrophy after the genetic era are few; and the development of obstructive hydrocephalus that requires surgical intervention is uncommon, and the cause of this complication is still unknown [27]. From the five patients of our cohort with reported ventriculomegaly, one (female) developed hydrocephalus who needed surgical intervention when she was two years old.
Cryptorchidism
Cryptorchidism is likely multifactorial when non-syndromic [28], and is a frequent feature in different neuromuscular disorders, well documented among male CDM newborns [5] In our review it was one of the most frequent findings, observed in 42% of the male cases.
Musculoskeletal deformities
Skeletal anomalies in DM1 can be divided into spinal deformities (scoliosis and kyphoscoliosis), contractures (Hip or knee) or foot deformities (equinus deformity, congenital talipes equinovarus, pes cavus, pes planovalgus, pes cavovarus and claw toes) [29]. The most frequent musculoskeletal (MSK) deformity found in these patients was congenital talipes equinovarus (37%), one of the leading causes of surgical interventions, frequently requiring more than one surgery. Other MSK deformities less frequent were: Arthrogryposis (6/68), Scoliosis (4/56), pes planus (1/37), and congenital hip dysplasia (1/37). Although we found cases with scoliosis as a neonatal complication on the literature review, it is more common to see this entity with the progression of the disease, predominantly after ten years of age [30]. Arthrogryposis is defined as “multiple congenital contractures with 2 or more areas in different body parts with the limitation of the movement present at birth” [30]. In the articles where this sign was reported, it was not specified which joints were affected or if they met the definition criteria.
Neonatal period
Neonatal mortality was reported in 16% of the cases, with respiratory failure as the primary cause of death. Zaki et al. [20] reported five neonatal fatalities and mentioned that the principal cause of death was a respiratory failure without specifying the existence of any other cause of death. Esplin et al. [14] reported one case of neonatal death at eleven days after birth with multisystem complications including hyaline membrane disease, patent ductus arteriosus, and intraventricular hemorrhage. In our cohort, one patient died at 30 days of age secondary to Sepsis; the rest of the reported deaths were caused by Respiratory failure (5/38) in three of them, secondary to withdrawal of life support as we mentioned before, and two of them at 24 hrs. of birth.
Neonatal death
n.d. The cause of death was not described in the literature.
To describe the long-term morbidities in CDM patients (Table 5) is challenging because of the lack of information in the literature, and as previously mentioned, from the 38 patients that met the inclusion criteria, 15 families agreed to continue in the cohort. The majority of the results obtained are from our cohort, where the leading causes of morbidity were respiratory tract infections (tonsillitis, pharyngitis, bronchiolitis, bronchitis, pneumonia, and croup). Sixty percent of the patients experienced from 1 to 4 episodes of these diseases a year, requiring multiple visits to the emergency room and in some cases hospitalizations, with an average of 4.4 admissions over the 5-year follow-up, and a mean hospital stayof 8.2 days.
The principal cause of surgical intervention was tympanostomy tube insertion secondary to recurrent otitis media (7/15). Three cases of those required two or more interventions. Other surgical procedures reported were: Orchidopexy (4/8), myringotomy (3/15), Achilles tendon lengthening (3/15), and tonsillectomy (1/15).
At the five-year follow-up, we found that death of CDM patients is even less frequent than during the neonatal period. After the first 30 days of age, three patients died during the first year of life: From our cohort, one patient presented multiple neonatal complications (Apnea, intraventricular hemorrhage, ventriculomegaly, hypoxic-ischemic encephalopathy, and sepsis), and the cause of death was respiratory failure at 7 months of age, the other patient from our cohort died secondary to sudden death at 3 months of age. Schild et al. [16] reported one patient who died at the age of 6 months who presented pneumothoraxes and then developed bronchopulmonary dysplasia in the neonatal period; the latter was the cause of death.
Long-term Morbidity and Mortality (5 years follow-up)
n.d. Not Described in the literature.
Ophthalmological features
In the five year follow-up, we did not find any case with lens pathology. Echenne et al. [21] reported six patients with punctuate and moderate cataracts, all of them diagnosed after ten years of age. Ekstrom et al. [31] described a CDM and childhood-onset-DM1 cohort, where they did not find an actual cataract, but a condensation on the posterior lens pole in 19 patients, where the youngest was 5.8 years old. In different publications focused on CDM ophthalmological complications, other features have been reported: Refractive error (70%), Astigmatism (40%) and Strabismus (66%) [31, 32]. 4This data differs from our review, where none of the aforementioned features were found, with the exception of Strabismus (12%), in which case we believe this discrepancy might be because these complications have not been intentionally investigated.
DISCUSSION
We performed a detailed description of the main perinatal features of CDM and precise documentation of the comorbidities during the first five years of life on a sample of well-defined cases of CDM in the era of genetic testing. Clarifying the phenotype of CDM in a precise manner represents an essential step in the knowledge of the natural history of CDM, representing a crucial step forward in the CDM literature, for the developing of care guidelines, and parameters of the clinical trials. We have shown that despite the different methodological approaches the rates of complications and morbidities is reasonably congruent across these various cohorts. Most of the articles that we reviewed were retrospective, with only two [17, 19] having a prospective design. Retrospective studies appear to show higher rates of complications during pregnancy (polyhydramnios, decreased fetal movements); three of the reviewed publications were mostly focused on pregnancy: prenatal ultrasound findings [20], and the association between myotonic dystrophy and polyhydramnios [14, 16]; and lower rates of neonatal complications (gastrointestinal, cardiologic and respiratory). Because of the study design, the retrospective studies provided limited information about morbidity and mortality in patients older than 30 days of age; most of the publications were focused on the neonatal period detailing critical perinatal features.
The design of our study -population based with active surveillance and prospective follow up- was conducive to identifying more cases and providing a detailed clinical description of the different comorbidities on CDM patients for the first five years, which will continue until patients reach the age of 18 years old.
We believe it is essential to have a precise definition of CDM as suggested by Campbell et al. [2]: 1) Evident symptoms of DM1 in the neonatal period (hypotonia, respiratory or feeding difficulties) requiring hospitalization for greater than 72 hours, and 2) Diagnostic genetic confirmation of CTG trinucleotide expansion with a repeat size greater than 200 in the DMPK gene in the child or the mother. We found some differences when comparing this definition with the literature review. In five of the articles [14, 20–22] there was no precise definition of the disease and the diagnosis was made by the severity of clinical features (without an explicit specification) and genetic confirmation, without detailing the number of CTGrs. Martinello et al. [17] included muscle biopsy for the diagnostic criteria, Schild et al. [16] define the clinical presentation as severe generalized hypotonia, feeding and respiratory difficulties, arthrogryposis and facial diplegia; and Kroksmark et al. [19] divided the cases of CDM into mild or severe, defining the latter with the presence of a life-threatening disease at birth. In this article, the cutoff for CTGrs was 40. The variability in definition has implications for both research and clinical care, and ultimately it will be necessary for the DM1 community to come to a consensus on the definition.
As noted with the different definitions for the disease, one of the main difficulties during this review was the various terms used for the phenotypical description (i.e., “clubfoot,” “congenital talipes equinovarus,” “pes equinovarus,” “arthrogryposis” [?]). We believe these discrepancies are a result of errors in translation or changes in the terminology over time, and speaks to the necessity and importance of a common terminology when designing cohorts, case series or even case reports. a moreconsistent ontology system like the Human Phenotype Ontology (HPO), will facilitate the achievement of uniformity when describing the different phenotypes. The authors have previously described a different cross-section of children using the HPO system [33].
The pathogenesis of DM1 is related to the effects of accumulated “toxic” mRNA in the cell nucleus with a gain of function that leads to sequestration and accelerated degradation of different nuclear proteins that regulate splicing (MBNL and CELF1). This accumulation is higher in the cortical neurons, but the cell degeneration is more severe in the muscle [34, 35]. As it is to expect, the motor milestones are delayed at first, but with a difference from other congenital muscle diseases, patients with CDM have a significant motor delay improvement in early childhood [13]. Currently, there are ongoing efforts to approach the treatment for DM1, with the aim to reduce the sequestration and degradation of these proteins on the nucleus with genetic based therapies (i.e. Gene therapy to correct RNA splicing by using a decoy protein that bind toxic RNA [VAL-0411] [36], anti-sense oligonucleotides against CUG repeat domain) [37], or not genetic based therapies to treat the muscle symptomatology (antimyotonic agents – Mexiletine [38]), daytime sleepiness treatment (methylphenidate) [36] and management of the life-threatening cardiac complications [39]. Most of these efforts are directed to the adult population, primarily oriented to treat the severe myotonia for these patients, which does not represent a significant complication for children with the congenital form. A recent trial with a GSK-3 inhibitor (Tideglusib), with the objective to improve the myogenesis in CDM myoblasts [40, 41], is ongoing for adolescent and adult patients with Congenital and Juvenile onset DM1 (NCT02858908).
The limitations of our study are a) The size sample of our cohort: currently there are fifteen participants enrolled in the study, more than half of the patients did not continue after the first phase of the study; b) Low response rate of the primary physicians or pediatricians who are following the patients. Most of the information is given by the families of the patients; c) Data collection: the data of our cohort was collected via phone interviews, and d) Methods: It is difficult to know the significance from the comparison between our prospective cohort and other retrospective studies. During the data collection, many reports done by the primary physician or pediatrician had another diagnosis from DM1, e.g. Myotonia Congenita or other Congenital Muscle Diseases. CDM is a rare disease, and the objective of our research is to enrich the knowledge of the natural history of the disease.
In the meantime, the most significant effort relies on the adequate management of the patients with CDM and the development of pediatric care guidelines which is an ongoing effort of the Myotonic Dystrophy Foundation. Currently, we are collaborating with the University of Utah (UT, USA) and the Centro Clinico Nemo (Milan, Italy) in a prospective study of patients with CDM to determine important clinical endpoints at different stages of the disease. It is crucial for accurate research and comprehensive clinical care to have a clear understanding of the natural history of CDM, and this manuscript can be used as a guide to developing different projects in term of clinical care and research.
FUNDING SOURCE
The study was funded by the William Singeris Family Foundation via the Child Health Research Institute.
FINANCIAL DISCLOSURE
Dr. Campbell has been a site investigator for Ionis Pharmaceuticals and holds a grant from Valerian Pharmaceuticals. He has consulted for AMO.
CONFLICT OF INTEREST
The Authors declares that there is no conflict of interest.
CONTRIBUTORS’ STATEMENT
Dr. Craig Campbell conceptualized and designed the study, reviewed, and revised the manuscript.
Ms. Hicks collected data, and reviewed and revised the manuscript.
Dr. Ceballos Saenz and Dr. Zapata Aldana carried the initial analyses, drafted the initial manuscript, and reviewed and revised the manuscript.
All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
Footnotes
ACKNOWLEDGMENTS
The authors would like to thank Diane Love for her careful edits on the final draft of the manuscript, to the patients and the families who participated in the cohort, and the William Singeris National Centre for Myotonic Dystrophy.