Trisomy 18—when the diagnosis is compatible with life

Trisomy 18 is an autosomal chromosomal disorder characterized by the presence of an extra 18 chromosome. In the last decades, and as novel therapeutic options emerged, a paradigm shift on the treatments available to these children occurred, establishing the need to deepen the knowledge regarding the management/treatment of children diagnosed with trisomy 18. This retrospective cohort study sought to characterize the clinical path and survival of the children with the diagnosis of trisomy 18 followed in a tertiary pediatric hospital between 1995 and 2020. Medical records were reviewed, and epidemiological and clinical features and follow-up data were collected. Six patients were identified, two with mosaicism (33.3%) and four were female (66.7%). All had cardiovascular, cognitive, and physical development anomalies or minor congenital anomalies. Most presented neurological anomalies (n = 4, 66.7%) and feeding difficulties (n = 4, 66.7%). Four children (66.7%) required medical devices or equipment and all required chronic medication. Two children (33.3%) underwent surgical interventions. Four children (66.7%) were hospitalized in the last year of life. Three patients had a do not resuscitate order (50%) but only one child was referred to a pediatric palliative care team (16.7%). One-month, 1-year, and 10-year survival were 66.7% (n = 4), 33.3% (n = 2, both with mosaicism), and 16.7% (n = 1, with mosaicism) respectively. Conclusions: Knowledge of the multiple comorbidities and complex care needs of children with this syndrome is crucial. Every-day care and decisions about invasive treatments may raise ethical issues. Early referral to pediatric palliative care teams is essential to promote a holistic advanced care plan for both the patient and his family. What is Known: • The increase in survival and the high morbimortality that trisomy 18 still entails demands a careful deliberation on the use of invasive treatment. What is New: • Recent studies show that the labels of “incompatible with life”/“lethal” are not adequate, establishing a need to change this mindset. • The development of pediatric palliative care teams in the last decade and early referral allow for an optimal individualized advanced care plan. Under-referral to pediatric palliative care teams persists and efforts must be made to increase awareness of their existence and role in patient care. What is Known: • The increase in survival and the high morbimortality that trisomy 18 still entails demands a careful deliberation on the use of invasive treatment. What is New: • Recent studies show that the labels of “incompatible with life”/“lethal” are not adequate, establishing a need to change this mindset. • The development of pediatric palliative care teams in the last decade and early referral allow for an optimal individualized advanced care plan. Under-referral to pediatric palliative care teams persists and efforts must be made to increase awareness of their existence and role in patient care.


Introduction
Trisomy 18, also known as Edwards syndrome, was first described in 1960 as an autosomal chromosomal disorder resulting from an extra 18 chromosome [1,2]. In reality, this syndrome is a highly heterogeneous entity that encompasses patients with a full extra chromosome, a partial trisomy (18q), or even mosaicism.
A 2019 worldwide study estimated the total prevalence of trisomy 18 as 4.08 per 10,000 births and the prevalence in live births as 1.07 per 10,000 births. These findings reflect the reality that most parents opt for medical termination of the pregnancy when faced with this diagnosis and that those who instead choose to continue the pregnancy will often experience a stillbirth [3,4].
Survival outcomes and clinical findings vary greatly [2]. Cardiovascular anomalies are the most frequent findings (present in 80-90% of patients), followed by musculoskeletal and nervous system anomalies. Other important manifestations are digestive system defects as their identification is paramount for a correct treatment plan after birth [2,[4][5][6].
In the past, trisomy 18 was considered to be "incompatible with life" and palliative care was offered as the only possibility for these children and their families [7,8]. Over the past decades, medical centers started to offer corrective treatment of congenital anomalies (mainly cardiac) in cases that might benefit from it, contributing to an increase in the survival of these children [6,7,9,10].
This increase in survival and the high mortality and morbidity that persists demands careful deliberation on which patients may benefit from invasive treatment. Additionally, timely referral to pediatric palliative care teams and development of an anticipated care plan that takes into consideration the clinical evolution of the child throughout time, the patient's best interests, and treatment goals, as well as the family's expectations and beliefs, are crucial for optimal management [2,6,7].
This study aims to characterize the clinical path of the children with the diagnosis of trisomy 18 who were followed in a tertiary pediatric hospital from 1995 to 2020, including comorbidities and survival, invasive procedures performed, presence of advanced care planning, and referral to a pediatric palliative care team. A brief more detailed description of one of the patients that better illustrates the current paradigm is also provided.

Study design and patient selection
A retrospective cohort study was conducted that compiled all cases of children diagnosed with trisomy 18 who were followed in a tertiary pediatric hospital from 1995 to 2020. All children who had a diagnostic code (International Classification of Diseases, Ninth Revision or Tenth Revision [ICD-9 or ICD-10] and Orphanet nomenclature of rare diseases) for trisomy 18 (ICD-9, 758.2 or ICD-10, Q91.0-Q91.3, and ORPHA:3380) were included. All patients included were diagnosed through karyotyping and fluorescence in situ hybridization (FISH) as per local protocol. Minimum number of cells counted to study mosaicism was 50. Children with irreconcilable data errors, including uncertain genetic diagnosis, were excluded. Neonates who died within 24 h of birth and whose diagnosis was only confirmed post-mortem were also not included. Clinical data were obtained through review of the medical and surgical records. In the cases where children were mainly followed in other hospitals, information was complemented whenever possible through review of the respective hospital's clinical file.

Data collection
Demographics (gender, year of birth), prenatal factors (fetal abnormalities, presence of prenatal diagnosis), neonatal factors (gestational age at birth, birth weight), cytogenetic status, age of cytogenetic diagnosis, diagnoses, comorbidities, and date and circumstances of death were included in the analysis. Treatments given such as surgeries, chronic medication, devices/equipment, specialties involved in treatment, and hospital admissions in the last year of life, as well as treatment decisions (advanced care planning and reason for withdrawal of therapeutic procedures), were also retrieved. Referral to the pediatric palliative care team in the hospital in which the study was conducted was also a variable in our study (although it should be noted this team was only created in 2016).
Fetal growth restriction was defined as an estimated fetal weight inferior to the 10th percentile for the gestational age using the 2013 Fenton growth chart for preterm infants.

Data analysis
Descriptive analysis was performed. Categorical variables are presented as frequencies and percentages; continuous variables as means and standard deviations if normally distributed; or medians and interquartile ranges for variables with skewed distributions. Survival curves were created to assess 1-month, 1-year, and 10-year survival probability. Analysis was performed with the use of IBM® SPSS® 27 (NY, USA).
Approval was obtained from the local ethics committee (process number 128-20).

Results
Six children with trisomy 18 were followed between 1995 and 2020, and by the time of conclusion of this study, only one child remained alive. Detailed individual patient description with most of the variables studied can be found in Table 2 in the Appendix. The most common cytogenetic abnormality was full trisomy (n = 4, 66.7%), with mosaicism detected in the remainder of patients (n = 2). Four children were female (66.7%) and half were term infants (n = 3, 50%). Despite being diagnosed with fetal growth restriction, none had prenatal diagnosis. One child (16.7%) was diagnosed with trisomy 18 in the first week of life, four children (33.3%) were diagnosed between the first week and first month of life, and one child was diagnosed at 15 months of age. Further description of all baseline and prenatal characteristics, as well as all primary and secondary diagnoses, can be assessed in Table 1.

Congenital abnormalities and comorbidities
The number of organ systems with congenital abnormalities varied between 3 and 8 systems, with a mean of 6. All children presented cardiovascular, cognitive, and physical development and minor congenital anomalies. The most common cardiovascular anomalies included ventricular septal defects and patent ductus arteriosus (n = 5, 83.3%) followed by atrial septal defects (n = 4, 66.7%) and arterial hypertension and pulmonary hypertension and aortic anomalies (n = 3, 50%). The most observed minor congenital anomalies were short palpebral fissures and low-set ears (both n = 4, 66.7%). Three children (50%) had musculoskeletal anomalies. Neurological anomalies were present in 4 patients (66.7%), with mega cisterna magna and hypotonia being the most frequent (n = 3, 50%). Ophthalmological anomalies were found in half the cases (n = 3). The most common comorbidity found was feeding difficulties (n = 4; 66.7%), followed by failure to thrive (n = 2, 33.3%). All main diagnoses identified are described on Table 3 in the Appendix.
Out of the six children, four (66.7%) relied on at least one device or equipment. In one child, it was not possible to determine their need. The average number of devices or equipment needed per child was 1.8, with a maximum of 3. All four children who needed equipment assistance were started on enteral tube feeding in the first 48 h of life and had some form of respiratory support, namely oxygen therapy from birth (n = 3) or mechanical ventilation (n = 2, non-invasive ventilation from birth in one and both invasive and non-invasive in the last 5 days of life in the other).
All patients were on at least one chronic drug. The main class was diuretics (n = 5) followed by antihypertensive medication (n = 3). There were a mean number of 2 drug classes (minimum 1; maximum 5) per child.
Two of the children underwent surgery. One child had a patent ductus arteriosus ligation performed at 3 months old (corrective surgery) and survived to 9 years of age. The

Clinical follow-up
In the last year of life, four children (66.7%) had hospital admissions and had not been discharged home since birth. Of these, three (75%) had one intensive care unit (ICU) admission and one child (25%) had two admissions. These hospital admissions had a mean total duration of 49 days (14-98 days), with a total mean length of stay in the intensive care unit of 9 days (8-10 days). A do not resuscitate order (DNR) was present in half the cases (n = 3), with one of those also having a decision withdrawal of life-sustaining treatments (invasive ventilation and oxygen therapy). All three children died less than 3 months after birth (two of them before 1 month of age). Out of the three children eligible (born or alive after 2016), only one child was referred to the palliative care team for optimal symptomatic control and advance care planning 1 month before his death. This plan included DNR, discussion with the family of the preferred location for end-of-life care and psychoemotional, comfort care-centered approach, and spiritual support. DNR in the other 2 children were made 5 days before death, in the ICU.
By the conclusion of this study, only one child was still alive. This child continues to be followed by the cardiology and orthopedics teams for conservative management of aortic valve dysplasia and scoliosis respectively. Despite the diagnosis of global developmental delay, this child was able to finish the 4th school year at the age of 11 years old.

Survival analysis
Median age of death was 62 days (minimum 14 daysmaximum 9 years). This analysis also indicated a total 1-month, 1-year, and 10-year survival of 66.7%, 33.3%, and 16.7%, respectively (Fig. 1). Regarding the children with full trisomy 18 (n = 4), their life span ranged from 14 to 98 days of age, with half (n = 2) dying before 1 month of age (full trisomy 1-year survival: 0%). One mosaicism had a life span of 9 years, and the remaining child is still alive at the time of conclusion of this study. Cause of death was identified in three children, all of them due to cardiogenic shock. Concerning the place of death, 4 children died in-hospital (one in the intensive care unit), while in the remaining patient, this information was not available.

Case report
A 36-week-gestation, female newborn weighing 1820 g was born to a healthy 36-year-old multigravida. There was no fetal surveillance during pregnancy. The mother was admitted to the hospital with preterm labor and prompt delivery through caesarean section was performed due to acute fetal distress. Physical examination at birth revealed severe hypoxemia and polymalformative syndrome (overlapping fingers, claw hand, short palpebral fissures, low-set ears, short neck, labia minora hypertrophy). The patient was consequently admitted to the ICU and continuous low-flow oxygen therapy was initiated. On day 2, echocardiography detected atrial and ventricular septal defect, patent ductus arteriosus, pulmonary hypertension, and moderate aortic valve insufficiency. A nasogastric feeding tube was placed on the same day due Fig. 1 1-year and 10-year survival to feeding difficulties and the patient was started on oral furosemide and spironolactone. Early referral to cytogenetic study confirmed the diagnosis of full trisomy 18.
Observation by the physiotherapy department was requested and a full rehabilitation program was planned. Despite optimal care, the patient clinically deteriorated and remained in the ICU throughout her whole life. Subsequent cardiac arrest secondary to cardiogenic shock occurred at 98 days old with no response to advanced life support interventions.

Discussion
Over the last 25 years, six children with a diagnosis of trisomy 18 were identified, the majority being female and term infants, findings that are in corroborated by numerous studies [5,6,[10][11][12]. No children had prenatal diagnosis. As such, it is impossible to evaluate what prenatal approach would have been taken given the clinical complexity, wishes, and expectations of the parents.
The great majority of anomalies found are in agreement with the usually described phenotype of trisomy 18. All patients had a diagnosis of fetal growth restriction, a frequent ultrasound finding documented in previous literature [10,[13][14][15]. As found in other studies, prevalence of cardiovascular and musculoskeletal anomalies is high [4-6, 9, 10, 16]. Infants may undergo major surgeries, not only cardiac but also gastrointestinal and neurologic [6,10,21,22]. In this study, cardiac intervention was performed in only one child. This is in accordance with other studies in which 7-26% of the children had undergone cardiac surgeries [6,10,16,22].
Furthermore, it needs to be taken into consideration that, even if corrective surgery might not be appropriate in certain cases, surgical intervention contributes to a decrease in in-hospital mortality and allows for home care more easily [22]. This paradigm shift that occurred in the last decade has shown to contribute to greater 1-year survival, ranging between 8 and 29% (with a Japanese multicenter study reporting as high as 43%), and 5-year survival, varying between 7.7 and 12.3%. A 9.8% 10-year survival rate was also reported recently [3,4,6,[10][11][12]23].
The need for ventilatory support, oxygen, and feeding assistance is common throughout children with this disease [10,17,18]. These needs and medical complexity lead to long hospital stays [19,20]. In addition, many studies show a relevant proportion of children discharged home and requiring specialized home care assistance [21]. Early detection and correct management and treatment of these children improve medical care and overall quality of life. A recent document by Kepple et al. provides a comprehensive guideline on the identification and management of postnatal complications and is a step towards a more humane and complete approach to this disease [17].
Previous studies also demonstrated longer survival in children with mosaicism in comparison to the ones presenting with full trisomy, which is in agreement with our findings and may be due to the absence of major congenital anomalies in this subset of children [23,24]. Although our study did not show survival beyond 3 months of life in children with full trisomy 18, some studies demonstrate that survival in these individuals can extend beyond 5 years, in spite of the high associated morbidity [5,12,23,25,26]. Regardless of their cytogenetic status, this is unequivocal evidence that the designations of "incompatible with life" or "lethal" are not adequate to characterize this syndrome and establishes the need to emphasize the need for improving comfort and quality of life, as well as end-of-life care.
Most parents will have difficulty in understanding the medical aspects, clinical characteristics, and treatment needs of trisomy 18 children, resulting in a need for a trustful, honest communication between parents and physicians. Validating the parents' hopes and fears will in turn potentiate informed, conscious decisions regarding the children [26][27][28][29]. Early implementation and referral to specialized pediatric palliative care teams is a crucial tool for tackling this issue as they can assist in the prenatal and birth plans and postnatal care plan and give new perspectives and insights to promote the most adequate treatment through an individualized advanced care plan. Family expectations, beliefs, and cultural background are also focal piece of advanced care plans and end-of-life care (including preferred location for the end-of-life period) [7,[29][30][31].
In this study, the only child referred to palliative care died in hospital. The complexity of the multisystem pathology and poor socioeconomic status of his family severely limited the possibility of at-home care. Nevertheless, timely referral allowed for a holistic approach, including optimized symptom control, early psychosocial and spiritual support to the family, and careful individual planning of everyday and end-of-life care.
As the paragon on management and treatment options for trisomy 18 continues to change, all new reports, studies, and revisions of current literature contribute to the foundation of future guidelines that may assist in providing personalized care to these children and their families. Only through a coordinated and early effort can these children be given the best possible care that helps to lessen their suffering and maximize their potential.

Limitations
On account of this study being restricted to a single center, these findings may be skewed and must be interpreted with caution. Complete information was impossible to obtain in certain cases as some medical records were incomplete. Due to its retrospective nature of this study, diagnoses and comorbidities may have been underreported. Neither of the children had a prenatal diagnosis. This may have resulted in a smaller number of major congenital anomalies and subsequent better prognosis.