Objective
We sought to assess long-term neurodevelopment of children who were treated prenatally as part of the Eurofoetus randomized controlled trial.
Study Design
The study population was composed of 128 cases of twin-to-twin transfusion syndrome (TTTS) included and followed up in France. Survivors were evaluated by standardized neurological examination and by Ages and Stages Questionnaires (ASQ). Primary outcome was a composite of death and major neurological impairment.
Results
A total of 120 children (47%) were alive at the age of 6 months and were followed up to the age of 6 years. At the time of diagnosis, only treatment and Quintero stage were predictors of a poor outcome (hazard ratio, 0.61; 95% confidence interval, 0.41–0.90; P = .01 and hazard ratio, 3.23; 95% confidence interval, 2.19–4.76; P < .001, respectively). Children treated by fetoscopic selective laser coagulation (FSLC) had higher ASQ scores at the end of follow-up ( P = .04).
Conclusion
FSLC was significantly associated with a reduction of the risk of death or long-term major neurological impairment at the time of diagnosis and treatment.
Twin-to-twin transfusion syndrome (TTTS) develops in 10-20% of monochorionic diamniotic pregnancies with a mortality rate of approximately 90% for both twins if untreated. Survivors are exposed to subsequent neurological handicap, either as a result of preterm birth or of the disease itself.
For Editors’ Commentary, see Table of Contents
In a recent randomized trial, the Eurofoetus trial, we found that fetoscopic selective laser coagulation (FSLC) of the chorionic vessels is a more effective first-line treatment than serial amniodrainage (AD) in severe TTTS diagnosed before 26 weeks of gestation. FSLC offers a treatment option that is based on the pathophysiology of the disease and interrupt placental anastomoses, resulting in a higher survival rate and fewer neurological sequelae in survivors at 6 months of age.
However, uncertainty about long-term neurological outcome remains a major concern for both parents and physicians. Cognitive and motor developmental delay are relatively common and have been related to both FSLC and AD. Therefore, we prospectively studied the neurologic and neurodevelopmental outcomes up to the age of 6 years in infants included in the Eurofoetus randomized controlled trial (RCT) in years 1999-2002.
Materials and Methods
Design and study protocol
We studied the long-term follow-up of TTTS cases included in a RCT, the Eurofetus RCT. The initial study was a multicentric, randomized, and stratified by country, AD vs FSLC controlled trial of 142 women with monochorionic twins and TTTS recruited in 6 countries. Of the total population, 90% (128 cases) were recruited in France.
The trial was conducted between January 1999 and March 2002. Briefly, women presenting between 15 and 26 weeks of gestation with severe TTTS were invited to participate and were randomly allocated to laser surgery (FSLC group) or serial amniodrainage (AD group). The Quintero system was used to stage the disease. Exclusion criteria were fetal death, major fetal anomaly, ruptured membranes, maternal conditions requiring delivery, and any previous invasive therapy directed at the syndrome.
Following treatment, the decision to deliver was based on obstetrical indications and was made by the attending perinatologist; however, all deliveries were performed no later than 37 completed weeks of gestation. Placentas were assessed after delivery to confirm chorionicity.
The protocol was approved by the institutional review board at each center. Perinatal and infant data were provided by the treating clinicians, and outcomes up to the age of 6 months were assigned by only 1 neonatologist who had no knowledge of the assigned treatment. We prospectively followed up the 128 cases that were delivered in France for neurological and neurodevelopmental evaluation up to 6 years of age.
Outcome variables
Primary outcome was a composite of pre/postnatal death and major neurological impairment (see the following text) at follow-up. In addition, survivors were subjected to neurological and neurodevelopmental follow-up up to 6 years of age.
Neurological follow-up
The children were followed up by questionnaires sent to the treating clinicians caring for patients to assess physical status as well as their neurologic development. These questionnaires were to be answered at the age of 1 and 2 years.
All children underwent the standardized physical and neurological examination developed by Amiel-Tison and Gosselin and Deschenes and colleagues at the age of 5 years. These consisted of a detailed neurological examination that explores the neurological status. A single investigator, blinded to the mode of treatment, examined all but 10 children. The treating pediatrician or general practitioner examined these 10 children living too far away from a hospital or whom parents declined follow-up.
The children were classified into 3 groups according to follow-up data. Cerebral palsy was defined clinically and classified according to the European Cerebral Palsy Network recommendations. According to the neurological questionnaires and examination, development was classified as no neurological impairment (N1, normal physical and neurological examination results), minor impairment (N2, neurological deficiencies with prospect to normalization, including strabismus, mildly retarded motor and speech development), or major neurological impairment (N3, cerebral palsy with neurological abnormalities including hemiparesis, spastic quadriplegia, and blindness).
The parents also received Ages and Stages Questionnaires (ASQ) at 12, 24, 48, and 60 months. The ASQ is a parent-completed child-monitoring system that has been validated against the Bayley Scales of Infant Development as a screening tool for abnormal development. It contains 5 domains of child development (communication, gross motor skills, fine motor skills, problem-solving skills, and personal social skills). Each domain consists of 6 questions about what the child can or cannot do. The answer to each question is “yes,” “sometimes,” or “not yet” and is graded with 10, 5, or 0 points, respectively. Domain scores are then obtained by the sum of the items and compared with established screening cutoff points according to ASQ guidelines. This evaluation tool has been used extensively and has a high sensitivity and specificity to detect children with developmental delays. All children were assessed with these questionnaires.
Neurodevelopmental follow-up
At the age of 6 years, the Wechsler Intelligence Scale for Children (WISC-IV) was used to evaluate the children’s psychological development. This is a clinical instrument administered for the evaluation of intelligence in children between 6 and 16 years. The WISC-IV addresses the intellectual functioning in specific cognitive domains (verbal comprehension [VC], perceptive reasoning [PR], work memory [WM], processing speed [PS]). The WISC-IV score also provides evaluation of the general intellectual capacity of the child through a composite score of total scale, the intelligence quotient (IQ).
The Goodenough Draw-a-Man test was also used to determine the mental age of the child according to its graphic capabilities with reference to the elements recorded on or absent from the drawing. A psychologist, blinded to the mode of treatment, specifically trained to use this method did perform this psychological evaluations in all children.
Statistical methods
Long-term survival and neurological and neurodevelopmental outcomes were studied in relation to treatment group and other potential predictors such as status of donor or recipient, gestational age (GA) at treatment (truncated at median value below or above 160 days), and Quintero stage (3 or 4 vs 1 or 2). Uni- and multivariate survival analyses were performed to identify predictors of death and of neurological impairment (N3). The relationship with outcome of infants was estimated by survival analysis, taking into account the inherent interdependence between twins by a gamma frailty Cox model.
For a composite outcome (death or major neurological impairment), cumulative incidence was described by Kaplan-Meier curves. All tests were 2 tailed and P values of less than .05 were considered significant. Analysis was performed with the use of R statistical language (R Foundation for Statistical Computing, Vienna, Austria; http://www.r-project.org ).
Results
The population initially studied included 68 women (136 fetuses) treated in the FSLC and 60 women (120 fetuses) in the AD group, respectively. The median [interquartile range] maternal age was 32 [29–36] years in both groups, GA at diagnosis was 21 [19–22.8] and 20.6 [19–22.7] weeks in the FSLC and the AD group, respectively ( P = .79). There were 24 (35%) and 18 (30%) of nulliparous in the FSLC and the AD group, respectively ( P = .5).
In the FSLC group, 37% (n = 50) and 9% (n = 13) of fetuses died in utero and in the neonatal period, respectively. In the AD group, there were 39% (n = 47) and 22% (n = 26) died in utero and in the neonatal period, respectively. Seventy-three (54%) (22 pairs, 16 recipients, and 13 donors) and 47 fetuses (39%) (16 pairs, 4 recipients, and 11 donors) were born alive and were alive at 6 months of age in the FSLC and AD groups, respectively ( Figure 1 ). As described in the initial study, the survival rate at 6 months was significantly higher in the FSLC group than the AD group ( P = .01).
