Abstract
Purpose
Survival in Rett syndrome (RTT) remains unclear. Although early estimates were grim, more recent data suggest that survival into adulthood is typical. We aimed to define survival in RTT more clearly and identify risk factors for early death.
Methods
Participants with clinical RTT or Methyl CpG Binding Protein 2 mutations without clinical RTT were recruited through the RTT Natural History study from 2006 to 2015. Clinical details were collected, and survival was determined using the Kaplan-Meier estimator. Risk factors were assessed using Cox proportional hazards models.
Results
Among 1189 valid participants, 51 died (range 3.9–66.6 years) during the 9-year follow-up period. Those who died included 36 (3.9%) classic RTT females, 5 (5.9%) atypical severe RTT females, 1 (2.4%) non-RTT female, the single atypical severe male, 6 (30%) non-RTT males, and 2 (7.1%) DUP males. All atypical mild RTT females, DUP females and the single classic RTT male remain alive. Most deaths were due to cardiorespiratory issues. Only one died due to severe malnutrition, scoliosis, and extreme frailty. Survival for classic and atypical RTT was greater than 70% at 45 years. Overall severity and several modifiable risk factors, including ambulation, weight, and seizures, were associated with mortality in classic RTT.
Conclusions
Survival in to the 5th decade is typical in RTT, and death due to extreme frailty has become rare. While the leading cause of death remains cardiorespiratory compromise, many risk factors for early death are modifiable. Intense therapeutic approaches could further improve the prognosis for patients with RTT.
Keywords: Rett syndrome, risk factors, prognosis, survival, mortality
Introduction
The report of Hagberg et al. in 1983 provided the first widely-read English-language publication on Rett syndrome (RTT), resulting in the remarkable expansion of clinical studies to understand the specific features of this X-linked dominant disorder. The identification of mutations in methyl-CpG-binding protein 2 (MECP2) in 1999 allowed fundamental research to progress dramatically. However, increased recognition of the disorder, expansion of the scope of clinical assessments, and the active involvement of parents and other caregivers worldwide had begun to alter the extent of clinical involvement in the management of this unique neurodevelopmental disorder prior to 1999.1 Early longitudinal studies2 revealed that after the initial stagnation and regression of development, children reach a “steady state” in adolescence and adulthood. Neuropathological studies3 shifted the perception from “degenerative” to the current perspective of a neurodevelopmental disorder.4 With increasing recognition of the myriad clinical issues and the need for intense therapeutic approaches, longevity and overall quality of life improved.5 The predominant morbidity issues include growth,6 nutrition,7 scoliosis,8 seizures,9 aspiration risk, and gastrointestinal dysfunction (gastroesophageal reflux, delayed gastric emptying, and constipation).10 Clinical experience suggests that intense physical and occupational therapies reduced development of contractures and skeletal deformities and communication technologies improve engagement.
The initial report of deaths in RTT occurred prior to implementation of intense therapeutic approaches.11 Thus, of the reported deaths, half were attributed to frailty and debilitation with frequent aspiration, and 25% were attributed to an unwitnessed event assumed to be related to seizures or aspiration. More than 10 years later, the first survival study, conducted among more than 1900 participants in the US and Canada indicated 50% survival at 50 years.5 Shortly, thereafter, an analysis of the original cohort seen by Rett indicated a much reduced survival rate.12 The same report, together with a recent study providing data from Australia13 yield similar results to those from the US and Canada.
In this report, we show that survival from the US RTT Natural History study (RNHS) is somewhat better than the 2010 survival study.5 Using the detailed clinical data in the RNHS, we have analyzed the fifty-two deaths reported in this cohort; together, these represented 4.3% of those enrolled. In contrast to the 1997 study, only one of these was related to a debilitated condition. We sought to identify characteristics associated with greater likelihood of death.
Methods
Participants
Through the multicenter RNHS, individuals with clinical RTT were recruited from March 2006 until February 2015, and evaluated at one of eight US sites every six to twelve months, as described previously.6,9 The RNHS consortium is part of the Rare Diseases Clinical Research Network, an initiative of the Office of Rare Diseases Research, National Center for Advancing Translational Sciences. An RNHS neurologist or geneticist (J.L.N., W.E.K., D.G.G., S.A.S., and A.K.P.) confirmed the diagnosis based upon diagnostic criteria.14,15 Two scales were used to assess overall severity, the clinical severity scale (CSS) and motor behavioral assessment (MBA), and quality of life was assessed using the Child Health Questionnaire 50 (CHQ) as previously reported.6 Reports of EKGs were collected and QT and corrected QT (QTc) intervals were recorded. All participants had MECP2 testing by a qualified laboratory. Although >95% of those with classic RTT have a mutation, participants with clinical RTT were included even if they lacked a mutation. Deaths and related events indicative of causation were assessed from participants with classic RTT, atypical RTT, those who had mutations in MECP2 but did not meet criteria for classic or atypical RTT (non-RTT), and those with MECP2 duplication syndrome (DUP). Death certificates were reviewed and details extracted; additional information was collected from caregivers on events preceding death. The first contact with the RNHS was used as the point of entry, and the last contact or age at death was used as the study exit point.
Data Categorization
Atypical RTT typically presents in two ways: the first scenario is an individual who achieved few early skills and, with fewer skills to lose, did not express regression in either language or hand use, generally with a more severe disease course; the second scenario is an individual who acquired language and hand use either on time or with some delay, and did not experience regression in hand use or language, generally with a less severe disease course. The clinical severity score in atypical RTT has a bimodal distribution, with nadir at 21; therefore, these two groups were divided into atypical mild (CSS ≤20), and atypical severe (CSS ≥21).16 Ambulation was categorized in a binary (able to stand or walk, unable to stand or walk) and ordinal fashion. Epilepsy was categorized in a binary (those with seizures despite medical management, and those with well controlled or no seizures) and ordinal fashion. Other characteristics and comorbidities (e.g., breathing dysregulation) were characterized based on frequency during examination, and socioeconomic status was categorized into ordinal groups based on comparison to the national median income. Interval of QTc was categorized into normal (≤450), borderline (451–<470) and abnormal (>470). Growth parameters (height, weight, BMI, head circumference) were categorized using both normative17,18 and Rett-specific z-scores. The standard cutoffs of +/− 2 SD (approximating 2nd and 98th percentiles) were used for normative charts and more liberal cutoffs of −1.28 SD (10th percentile) and −0.67 SD (25th percentile) on RTT charts were tested, in keeping with recent recommendations.7 Mutation severity was categorized into mild (R133C, R294X, R306C, and 3' truncations) and severe (T158M, R168X, R255X, and R270X) MECP2 mutations.19
Statistical Analysis
Descriptive statistics included age, MECP2 status, growth, MBA and CSS, number who died, and causation of death. Survival was calculated using the Kaplan-Meier estimator, conditional upon survival up to initial encounter. The association of categorical or ordinal variables (growth, mutation type, race, ethnicity, socioeconomic status, ability to walk, scoliosis, breathing dysregulation, sleep disturbance, hand function, frequency of stereotypies, tone, reflexes, autonomic dysfunction, epilepsy) and continuous variables (quality of life, number of hospitalizations or fractures, MBA and CSS) with survival was assessed by fitting to Cox proportional hazards models. Survival was compared among diagnostic categories using the log-rank test, and significant predictors in the Cox models were plotted as individual survival curves. Nonparametric data were summarized using the median and interquartile range (IQR). Statistical analyses were performed using SPSS.20
Human Studies Approval
Each site obtained and maintained IRB approval for the performance of this study. Parental approval for study conduct and publication of results was obtained before entry into the study. The study has been registered with ClinicalTrials.gov: NCT00299312 since March 3, 2006.
Results
Overall, 1205 individuals were enrolled in the NHS. Diagnosis could not be verified on 14. Two with CDKL5 mutation and atypical RTT were excluded from further analysis; one of these died at age 6.9y due to presumed cardiac arrest. Fifty-one deaths occurred in the remaining cohort of 1189 participants; these were followed for up to 9.0 years (median 7.0y), and included 925 with classic RTT (1 male), 80 females with atypical mild RTT, 86 with atypical severe RTT (1 male), 62 non-RTT (20 males), and 36 DUP (28 males). Of the female participants, thirty-six (3.9%) with classic RTT, five (5.9%) with atypical severe RTT, and one non-RTT female (2.4%) died during the study. Of the male participants, the single atypical severe male, six (30%) of the non-RTT males, and two (7.1%) of the DUP males died. All of the atypical mild RTT females, DUP females and the single classic RTT male remain alive. None born after 1997 lived in a group home or institution. The proportion of those older than 18 years who lived in a group home was 7.3%, and in an institution was 1.2%.
The causes of death divided by diagnostic category are listed in Table 1. According to death certificates, of those with classic RTT with an unknown cause of death, the majority was presumed to be related to cardiac arrest or respiratory compromise secondary to aspiration or pneumonia; four occurred at night and were unwitnessed. Post-operative complications were a cause of death in a minority, following scoliosis surgery, ileostomy, and tonsillectomy. Epilepsy was cited as cause of death (without respiratory complications) in four classic RTT participants and one atypical RTT participant. Additionally, the non-RTT male who died after VNS placement had experienced worsening seizures for several months, then developed cardiac dysrhythmia in the operating room. For atypical RTT, the two with unknown cause of death were presumed to be due to aspiration. One of those with atypical RTT who died with overwhelming infection was regarded as severely malnourished with severe scoliosis resulting in organ displacement and an extremely frail condition.
Table 1.
Cause of death based on diagnostic category (total n = 1189).
| Diagnosis | Number with diagnosis | Cause of death | Number who died | % within category |
|---|---|---|---|---|
| Classic female | 924 | 36 | ||
| Epilepsy | 4 | 11.1% | ||
| Respiratory (+/− aspiration/pneumonia) | 5 | 13.9% | ||
| Infection* | 4 | 11.1% | ||
| Post-operative complications** | 4 | 11.1% | ||
| Unknown (unspecified) | 3 | 8.3% | ||
| Unknown nocturnal | 4 | 11.1% | ||
| Unknown presumed cardiorespiratory | 12 | 33.3% | ||
|
| ||||
| Atypical mild female | 80 | 0 | ||
|
| ||||
| Atypical severe | 85 | 5 | ||
| female | Unknown nocturnal (presumed cardiorespiratory/Epilepsy) | 1 | 20.0% | |
| Unknown presumed cardiorespiratory | 1 | 20.0% | ||
| Respiratory (pneumonia, restrictive lung disease) | 2 | 40.0% | ||
| Malnutrition, scoliosis, pneumonia | 1 | 20.0% | ||
|
| ||||
| Non-Rett female | 42 | 1 | ||
| Hyperkalemia | 1 | 100.0% | ||
|
| ||||
| Duplication female | 8 | 0 | ||
|
| ||||
| Classic male | 1 | 0 | ||
|
| ||||
| Atypical severe male | 1 | 1 | ||
| Respiratory (pneumonia) | 1 | 100.0% | ||
|
| ||||
| Non-Rett mutation | 20 | 6 | ||
| male | Respiratory (aspiration) | 1 | 16.7% | |
| Post-operative complications*** | 1 | 16.7% | ||
| Unknown (unspecified) | 1 | 16.7% | ||
| Unknown presumed cardiorespiratory | 3 | 50.0% | ||
|
| ||||
| Non-Rett | 28 | 2 | ||
| duplication male | Unknown (unspecified) | 1 | 50.0% | |
| Unknown presumed cardiorespiratory | 1 | 50.0% | ||
chronic UTI, cystic-fibrosis-related pneumonia, chronic infection leading to sepsis and renal failure
ileostomy, spinal fusion, tonsillectomy
VNS placement
Considering age of death among the 1089 participants with classic and atypical RTT, the youngest died at age 3.85 years and the oldest at age 66.6 years. The proportion of mortality in classic and atypical severe RTT overall was similar (Figure 1). The latter contrasts with zero mortality among the eighty atypical mild participants. Kaplan-Meier curves revealed that the proportion of participants remaining alive in both classic and atypical severe RTT, including the 95% confidence intervals, was greater than 70% at 45 years. However, too few participants were recruited after age 45 years for accurate estimation. Although many male participants with MECP2 mutation died before age 10, none with DUP died during childhood (Figure 2).
Figure 1.
Survival in Classic and Atypical Severe Rett syndrome
Figure 2.
Survival Non-RTT, MECP2 Positive Males and Duplication Males
Of the 613 participants with classic and atypical RTT who had EKG data, 18% had a borderline and 10% had an abnormal QTc interval. However, the proportion with a borderline or abnormal EKG was similar between those who died (21%) and those who survived to the end of the study (29%, p = 0.47). Ninety-seven percent of those with classic RTT had a mutation in MECP2; thirty-three (3.7%) of those with a mutation died, whereas three (9.7%) of those without a mutation died. Only 68% of those with atypical severe RTT had a MECP2 mutation, but all atypical severe RTT participants who died had a mutation. Specific MECP2 mutations were not significantly associated with mortality. However, in those with classic RTT, large deletions (7.4%) and R294X (7.0%) had the highest frequency of deaths, nearly double the average of all mutations groups combined (3.9%) and three times the frequency with R133C (2.4%, Table 2). Nine (3.7%) in the mild and nineteen (4.3%) in the severe mutation categories died during the study. For other diagnostic categories, the numbers were too small to consider a comparison.
Table 2.
Deaths in classic RTT among specific MECP2 mutations (n = 885)
| Mutation Category | MECP2 mutation | Total n | Number who died | Percent |
|---|---|---|---|---|
| Mild mutations | R133C | 41 | 2.4% | |
| R294X | 53 | 4 | 7.0% | |
| R306C | 60 | 3 | 4.8% | |
| 3' Truncation | 78 | 1 | 1.3% | |
| Severe mutations | R106W | 25 | 1 | 3.8% |
| T158M | 95 | 5 | 5.0% | |
| R168X | 95 | 3 | 3.1% | |
| R255X | 86 | 1 | 1.1% | |
| R270X | 51 | 3 | 5.6% | |
| Large Deletion | 75 | 6 | 7.4% | |
| Miscellaneous mutations | Other Point Mutation | 97 | 3 | 3.0% |
| Insertion | 20 | 0 | 0.0% | |
| Deletion | 51 | 2 | 3.8% | |
| Exon 1 | 4 | 0 | 0.0% | |
| Splice Site | 9 | 0 | 0.0% | |
| Multiple mutations | 12 | 0 | 0.0% |
Missing or unclear data in 8.
Growth in Classic RTT
Differences in mortality were more profound when the RTT-specific charts were used, and these are reported here, although differences in height and weight were also seen on normative charts. Mortality was similar for those with normal and low BMI. However, the majority of those who died (54%, or 19/35) weighed less than the 25th percentile on the Rett-specific charts, compared to 22% of those who survived (p < .001). Additionally, the majority of those who died (57%, or 20/35) measured less than the 25th percentile for height on the Rett-specific charts, compared to 23% of those who survived (p < .001). The proportion of those with a smaller head (<25th percentile) on the Rett-specific charts was 40% for those who died (14/35) compared to 22% (187/862) for those who survived (p = .01).
Risk Factors for Mortality
In classic RTT, median scores on one or both of the two severity scales were higher in those who died compared to those who were living at the end of the study (Table 3). In atypical severe RTT and male DUP patients, severity was higher on one of the two scales but not the other in those who died. For classic RTT, several variables, such as ambulation, frequent seizures, frequent hospitalizations and illness, microcephaly and low weight, were independently associated with higher odds of mortality using a Cox proportional hazard model (Table 4). Many of the categorical variables associated with increased odds of mortality also revealed significant differences when Kaplan Meier curves were compared (Figure 3). For atypical severe RTT, only number of hospitalizations was associated with higher mortality (hazard ratio 2.6, 95% CI 1.1–6.3, p = .03). No significant associations were found between risk factors and mortality for other diagnostic categories, notably scoliosis, sleep disturbance, autonomic dysfunction, frequent fractures, hyperventilation or breath-holding.
Table 3.
Median severity scores and mortality
| Diagnosis | Severity Scale | Survival | N | Median | Interquartile Range | Minimum | Maximum | Mean Severity Rank | Mann-Whitney p-value |
|---|---|---|---|---|---|---|---|---|---|
| Classic RTT | CSS | Alive at end | 858 | 25 | 20–31 | 5 | 45 | 435 | |
| Died before end | 35 | 35 | 29–37 | 20 | 43 | 730 | <.001 | ||
| MBA | Alive at end | 861 | 54 | 44–64 | 12 | 96 | 439 | ||
| Died before end | 35 | 68 | 61–73 | 49 | 95 | 689 | <.001 | ||
| Atypical Severe RTT | CSS | Alive at end | 76 | 31 | 27–34 | 22 | 47 | 41 | |
| Died before end | 5 | 31 | 29–36 | 27 | 38 | 45 | .68 | ||
| MBA | Alive at end | 76 | 59 | 51–65 | 23 | 88 | 40 | ||
| Died before end | 5 | 81 | 57–86 | 46 | 87 | 61 | .04 | ||
| Non-RTT female | CSS | Alive at end | 39 | 7 | 2–10 | 0 | 41 | 21 | |
| Died before end | 1 | 1 | N/A | N/A | N/A | 6 | .30 | ||
| MBA | Alive at end | 39 | 13 | 7–25 | 0 | 69 | 21 | ||
| Died before end | 1 | 3 | N/A | N/A | N/A | 7 | .35 | ||
| Non-RTT male | CSS | Alive at end | 13 | 25 | 15–36 | 9 | 39 | 14 | |
| Died before end | 6 | 37 | 26–43 | 23 | 44 | 8 | .06 | ||
| MBA | Alive at end | 14 | 50 | 42–77 | 22 | 78 | 13 | ||
| Died before end | 6 | 71 | 57–77 | 45 | 81 | 9 | .24 | ||
| DUP male | CSS | Alive at end | 26 | 14 | 10–21 | 6 | 32 | 14 | |
| Died before end | 2 | 27 | N/A | 27 | 27 | 26 | .05 | ||
| MBA | Alive at end | 26 | 42 | 22–52 | 5 | 80 | 14 | ||
| Died before end | 2 | 64 | N/A | 61 | 66 | 25 | .06 |
Clinical severity scale (CSS) is scored from 0 to 58, and motor behavioral assessment (MBA) is scored from 0 to 148, and higher scores on both indicate greater severity.
Table 4.
Risk Factors for Mortality in Classic RTT
| Variable | N | Hazard Ratio | 95% Confidence Interval | p value |
|---|---|---|---|---|
| Inability to walk* | 882 | 3.2 | 1.6–6.5 | 0.00 |
| Number of hospitalizations | 892 | 1.1 | 1.0–1.2 | 0.00 |
| Microcephalic | 819 | 9.9 | 1.7–58.0 | 0.01 |
| Poor global health** | 813 | 1.8 | 1.1–3.1 | 0.02 |
| Seizure severity*** | 892 | 2.4 | 1.0–6.8 | 0.03 |
| Unable to babble or use some words | 819 | 2.6 | 1.1–6.5 | 0.03 |
| Degree of rigidity*** | 882 | 1.3 | 1.0–1.7 | 0.04 |
| Poor hand use*** | 882 | 2.6 | 1.0–6.8 | 0.05 |
| Low weight Z-score based on RTT references | 854 | 2.9 | 1.0–8.8 | 0.05 |
| Poverty**** | 819 | 7.4 | 0.9–60.5 | 0.06 |
| Unable to sit independently* | 882 | 1.3 | ||
| Immunity** | 813 | 1.4 | 1.0–1.9 | 0.07 |
| Degree of dystonia*** | 882 | 1.4 |
CSS
CHQ
MBA
Categories based on national median
Figure 3a–j.
Categorical Risk Factors for Mortality in classic RTT
Discussion
Compared to the initial report of Kerr et al. in 1997, in which nearly 50% of deaths were attributed to a frail or debilitated state with recurrent aspiration and pneumonia, our study has revealed a striking difference in the general health of participants.11 Only one of the individuals in our cohort died as the result of a frail condition. Fifty of the fifty-seven reported causes of death in the 2014 Australian study of survival in RTT were similar to those reported here and specifically did not mention frailty or debilitation.13 The cause of death in the majority of both ours and the Australian cohorts was presumed or confirmed cardio-respiratory issues, often due to aspiration. Prolonged QTc has received much attention in RTT, and, indeed, we found a striking prevalence of borderline and abnormal QTc intervals.21 However, the proportion of those with an abnormal QTc interval was nonsignificantly lower in those who died. Therefore, although many died of unknown, “presumed” cardio-respiratory events, QTc prolongation was not a strong contributor.
We have demonstrated that several modifiable risk factors are independently associated with risk for mortality. These results strongly support the need for physicians to provide close overall supervision for nutrition, gastrointestinal issues (gastroesophageal reflux, constipation, or even gallbladder dysfunction), scoliosis monitoring, aspiration risk, and epilepsy. Additionally, physicians and parents should continue demanding intense therapeutic approaches not only during the school age years but also throughout life. Attention to proper and continued implementation of strong therapeutic practices to prevent or manage contractures, dystonic postures, and proper positioning will minimize risk factors for mortality such as rigidity and dystonia. Moreover, such therapy can help to maintain a proper level of engagement with family and peers, improving quality of life as patients age. Studies from both Australia and the US provide evidence of the benefit of proper surveillance which should continue to improve with the broadening awareness of this neurodevelopmental disorder.
Despite the grim picture painted by Rett's original cohort, both ours and the Australian cohorts indicate that survival into adulthood is typical.12 Additionally, remarkable cases of older patients with RTT exist. One woman with RTT survived until age 79, dying of postoperative complications.22,23 Another woman diagnosed at age 52 years who had not walked or used her hands in over 20 years regained the ability to walk independently and feed herself with her hands through intense therapy.24 Cross-sectional studies suggest that women with RTT are generally healthy, and may improve with respect to function and comorbidities during adolescence and adulthood.25,26 Yet, few longitudinal studies have examined survival or risk factors for mortality in RTT. In a cohort of 53 Dutch adult women with RTT, only 37 were successfully followed for 5 years, and 7 died.27 Within this cohort, cognitive function appeared to be preserved, overall health was good, and comorbidities such as epilepsy and autonomic dysfunction stabilized or decreased over time in most. Only slight motor decline occurred over 5 years on average. Most hospitalizations were for pneumonia, but cause of death was often sudden and unclear. A recent population-based study of the 102 patients diagnosed with RTT in Serbia also found the leading cause of death among the 19 patients who died to be pneumonia.28 This cohort was relatively young, the oldest being 31 years old, and their incidence of 0.59:10,000 female live births is lower than that of larger studies, suggesting that many patients remain undiagnosed in that country.29,30 With up to 20 years of follow-up data, the Australian cohort remains the most robust sample from a longitudinal standpoint. However, this sample of almost 400 participants is substantially smaller than ours, and rate of death was higher overall. The Australian study examined the associations between MECP2 mutation and various comorbidities, but not the mortality risk conferred by specific features or comorbidities.
Ours is the first study of survival with assessment of all participants directly examined by a Rett syndrome specialist, and the first to examine mortality risk associated with RTT phenotype. The 2010 US study of survival included a subpopulation of those in the current study (less than 1/3 of the current cohort) followed for less than 2 years, but found somewhat lower survival overall in classic RTT (approximately 60% at 40 years and 45% at 50 years).5 Survival in the atypical RTT group was found to be significantly better than in classic RTT; however, this was likely due to the inclusion of both mild atypical and severe atypical individuals in the same group. Although categorizing mild and severe individuals together as “atypical” has been done since the 1995 diagnostic criteria defined atypical RTT, atypical mild participants rarely die prematurely. This disparity lends more support to the separation of these groups suggested in a recent publication.16
Overall disease severity was strongly associated with mortality in classic RTT, and weakly associated with mortality in atypical severe RTT and DUP males. Classic RTT is a heterogeneous disorder in many respects, and the association with mortality likely reflects the presence or absence of specific phenotypic characteristics. The lack of association between mortality and severity in the other groups may be due to lower heterogeneity, perhaps because of selection bias or the absence of the effect of lyonization (in the case of males). However, the small number of participants in these groups would require a profound effect to be statistically significant.
Since severity scales in RTT are constructed based on a collection of the features and comorbidities of the disorder, we examined these in detail in classic RTT using a regression model, and found that several features were independently associated with mortality. Among these, the most profound differences in survival curves (Figure 3) were for those who were unable to stand, walk or sit independently, were low weight, had frequent hospitalizations, or had poor overall health as perceived by their caregivers. Additionally, seizure severity, dystonia, rigidity, hand use, verbal language and microcephaly were independently associated with mortality. Many of the features associated with mortality are treatable or preventable conditions. Early and constant implementation of therapy can maintain ambulation and may maintain hand use and some verbal language.23 Moreover, aggressive nutrition and gastrostomy placement results in improved weight.31 Therefore, these results suggest specific features for physicians and therapists to target that could result in improved survival in RTT. Notably, degree of parental concern about health or mortality was not associated with death.
This is also the first natural history study to report survival in MECP2 duplication syndrome, and adds to the small body of work on males with MECP2 mutations.32 The clinical phenotype of these disorders is substantially different from that of RTT, and, not surprisingly, survival is quite different from classic RTT. Although DUP can result in frequent infections, and one review suggests a high rate of early mortality, only 2 of the 28 male DUP participants in our study died, both of unknown causes, and others were followed into their third decade. Death was more common in non-RTT males, occurring in 6 of the 20 recruited, with death before age 10 in 3 of these. Again, cause of death was often unknown, but was presumed to be cardiorespiratory in most.
Neither our cohort nor the Australian cohort demonstrated a significant association between mutation type and mortality. Although participants with large deletions had the highest proportion of mortality in our study and the Australian study, neither reached statistical significance. Moreover, the R294X mutation had the second highest mortality rate in our study, but conferred a nonsignificant protective effect in the Australian cohort. While one study suggested that R270X is associated with early death, other longitudinal studies have found that individuals with R270X can survive into their fifth decade.23,33,34 These inconsistencies support the notion that the association between genotype and survival, if any exists, is weak. Nonetheless, the common finding of higher mortality in the large deletion group bears further examination.
Estimates of longevity in all longitudinal studies of RTT to date, including our study, are subject to survival bias. However, our large cohort and long follow-up period, overlapping across several decades, resulted in a narrow confidence interval for classic RTT up to age 50 years. We can estimate, with 95% confidence, that at least 95% of the classic RTT population survive until age 20, 80% survive until age 35, and over 70% survive until age 50. In contrast, in the Australian cohort, only 77.6% survived until 20 years and 59.8% survived until 37 years. Several differences between these studies bear examination. In the Australian cohort, only 18% walked independently. We found that walking, even with assistance, was associated with improved survival, and this could suggest either a selection bias in our study, or that aggressive physical therapy in the US has resulted in improved ambulation. Additionally, only 71% of those over age 18 in the Australian cohort were living at home, in contrast with the 91% over 18 years old living at home in our cohort. This may suggest that caregivers of individuals in institutions were less likely to travel to participate in our natural history study.
When examined individually, lower height, weight, and head circumference, but not low BMI, were associated with a higher proportion of mortality. This is likely because BMI is a ratio of height and weight, and does not reflect severity in those with both low weight and height. Additionally, because the majority of RTT patients are below the lower cutoffs for weight and height on normative references, the RTT-specific references were more sensitive to the association with mortality. Low weight and microcephaly were the only anthropometric variables which were significant in the regression model. Since weight is modifiable, targeting this aggressively with early supplementation, via gastrostomy tube if necessary, may improve longevity. However, this strategy is only likely to benefit those who are malnourished, not those who are small due to non-modifiable disease factors.
Conclusion
Longevity is a major concern of families after receiving the diagnosis of RTT. These data help with anticipatory guidance for families, which we believe should be addressed soon after the diagnosis is made to allay unspoken concerns. Because patients may outlive their caregivers, the issue of long-term planning needs to be addressed as children age. Additionally, the natural history data reveal associations with growth, epilepsy, and ambulation, all of which can be addressed with appropriate routine management.
Acknowledgements
The Angelman, Rett, Prader-Willi syndrome consortium (U54HD61222) is part of the Rare Diseases Clinical Research Network (RDCRN), an initiative of the Office of Rare Diseases Research (ORDR), National Center for Advancing Translational Science (NCATS). This consortium is funded through collaboration between NCATS, and the Eunice Kennedy Shriver Child Health and Human Development Institute. This research was also supported by the International Rett Syndrome Foundation, the Civitan International Research Center, and NIH RR019478. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors thank Eric Pedrotty for statistical analysis of socioeconomic data. The authors acknowledge the gracious participation and provision of information by the families of the reported participants. Dr. Mary Lou Oster-Granite, Health Scientist Administrator at NICHD, provided invaluable guidance, support, and encouragement for this Rare Disease initiative.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The authors report no conflicts of interest.
References
- 1.Fehr S, Bebbington A, Nassar N, et al. Trends in the diagnosis of Rett syndrome in Australia. Pediatric research. 2011 Sep;70(3):313–319. doi: 10.1203/PDR.0b013e3182242461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Naidu S, Murphy M, Moser HW, Rett A, Opitz JM, Reynolds JF. Rett syndrome - natural history in 70 cases. American journal of medical genetics. 1986;25(S1):61–72. doi: 10.1002/ajmg.1320250507. [DOI] [PubMed] [Google Scholar]
- 3.Armstrong D, Dunn JK, Antalffy B, Trivedi R. Selective dendritic alterations in the cortex of Rett syndrome. Journal of neuropathology and experimental neurology. 1995 Mar;54(2):195–201. doi: 10.1097/00005072-199503000-00006. [DOI] [PubMed] [Google Scholar]
- 4.Neul JL, Zoghbi HY. Rett syndrome: a prototypical neurodevelopmental disorder. The Neuroscientist : a review journal bringing neurobiology, neurology and psychiatry. 2004 Apr;10(2):118–128. doi: 10.1177/1073858403260995. [DOI] [PubMed] [Google Scholar]
- 5.Kirby RS, Lane JB, Childers J, et al. Longevity in Rett syndrome: analysis of the North American Database. The Journal of pediatrics. 2010 Jan;156(1):135–138. e131. doi: 10.1016/j.jpeds.2009.07.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tarquinio DC, Motil KJ, Hou W, et al. Growth failure and outcome in Rett syndrome: specific growth references. Neurology. 2012 Oct 16;79(16):1653–1661. doi: 10.1212/WNL.0b013e31826e9a70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Leonard H, Ravikumara M, Baikie G, et al. Assessment and Management of Nutrition and Growth in Rett Syndrome. Journal of pediatric gastroenterology and nutrition. 2013 Oct;57(4):451–460. doi: 10.1097/MPG.0b013e31829e0b65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Percy AK, Lee HS, Neul JL, et al. Profiling scoliosis in Rett syndrome. Pediatric research. 2010 Apr;67(4):435–439. doi: 10.1203/PDR.0b013e3181d0187f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Glaze DG, Percy AK, Skinner S, et al. Epilepsy and the natural history of Rett syndrome. Neurology. 2010 Mar 16;74(11):909–912. doi: 10.1212/WNL.0b013e3181d6b852. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Motil KJ, Caeg E, Barrish JO, et al. Gastrointestinal and nutritional problems occur frequently throughout life in girls and women with Rett syndrome. Journal of pediatric gastroenterology and nutrition. 2012 Sep;55(3):292–298. doi: 10.1097/MPG.0b013e31824b6159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kerr AM, Armstrong DD, Prescott RJ, Doyle D, Kearney DL. Rett syndrome: analysis of deaths in the British survey. European child & adolescent psychiatry. 1997;6(Suppl 1):71–74. [PubMed] [Google Scholar]
- 12.Freilinger M, Bebbington A, Lanator I, et al. Survival with Rett syndrome: comparing Rett's original sample with data from the Australian Rett Syndrome Database. Developmental medicine and child neurology. 2010 Oct;52(10):962–965. doi: 10.1111/j.1469-8749.2010.03716.x. [DOI] [PubMed] [Google Scholar]
- 13.Anderson A, Wong K, Jacoby P, Downs J, Leonard H. Twenty years of surveillance in Rett syndrome: what does this tell us? Orphanet journal of rare diseases. 2014;9:87. doi: 10.1186/1750-1172-9-87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hagberg B, Hanefeld F, Percy A, Skjeldal O. An update on clinically applicable diagnostic criteria in Rett syndrome. Comments to Rett Syndrome Clinical Criteria Consensus Panel Satellite to European Paediatric Neurology Society Meeting, Baden Baden, Germany, 11 September 2001. European journal of paediatric neurology : EJPN : official journal of the European Paediatric Neurology Society. 2002;6(5):293–297. doi: 10.1053/ejpn.2002.0612. [DOI] [PubMed] [Google Scholar]
- 15.Neul JL, Kaufmann WE, Glaze DG, et al. Rett syndrome: revised diagnostic criteria and nomenclature. Annals of neurology. 2010 Dec;68(6):944–950. doi: 10.1002/ana.22124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Neul JL, Lane JB, Lee HS, et al. Developmental delay in Rett syndrome: data from the natural history study. Journal of neurodevelopmental disorders. 2014;6(1):20. doi: 10.1186/1866-1955-6-20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC growth charts: United States. Adv Data. 2000 Jun 8;(314):1–27. [PubMed] [Google Scholar]
- 18.Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Stat Med. 1998 Feb 28;17(4):407–429. [PubMed] [Google Scholar]
- 19.Cuddapah VA, Pillai RB, Shekar KV, et al. Methyl-CpG-binding protein 2 (MECP2) mutation type is associated with disease severity in Rett syndrome. J Med Genet. 2014 Mar;51(3):152–158. doi: 10.1136/jmedgenet-2013-102113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.IBM SPSS Statistics for Windows. IBM Corp; Armonk, NY: 2013. [computer program]. Version 22.0. [Google Scholar]
- 21.Ellaway CJ, Sholler G, Leonard H, Christodoulou J. Prolonged QT interval in Rett syndrome. Archives of disease in childhood. 1999 May;80(5):470–472. doi: 10.1136/adc.80.5.470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nielsen JB, Ravn K, Schwartz M. A 77-year-old woman and a preserved speech variant among the Danish Rett patients with mutations in MECP2. Brain & development. 2001 Dec;23(Suppl 1):S230–232. doi: 10.1016/s0387-7604(01)00365-5. [DOI] [PubMed] [Google Scholar]
- 23.Lotan M, Merrick J, Kandel I, Morad M. Aging in persons with Rett syndrome: an updated review. TheScientificWorldJournal. 2010;10:778–787. doi: 10.1100/tsw.2010.79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Jacobsen K, Viken A, von Tetzchner S. Rett syndrome and ageing: a case study. Disability and rehabilitation. 2001;23(3–4):160–166. doi: 10.1080/09638280150504234. Feb 15–Mar 10. [DOI] [PubMed] [Google Scholar]
- 25.Halbach NS, Smeets EE, Schrander-Stumpel CT, van Schrojenstein Lantman de Valk HH, Maaskant MA, Curfs LM. Aging in people with specific genetic syndromes: Rett syndrome. American journal of medical genetics. Part A. 2008 Aug 1;146A(15):1925–1932. doi: 10.1002/ajmg.a.32361. [DOI] [PubMed] [Google Scholar]
- 26.Vignoli A, La Briola F, Peron A, et al. Medical care of adolescents and women with Rett syndrome: an Italian study. American journal of medical genetics. Part A. 2012 Jan;158A(1):13–18. doi: 10.1002/ajmg.a.34367. [DOI] [PubMed] [Google Scholar]
- 27.Halbach NS, Smeets EE, Steinbusch C, Maaskant MA, van Waardenburg D, Curfs LM. Aging in Rett syndrome: a longitudinal study. Clinical genetics. 2013 Sep;84(3):223–229. doi: 10.1111/cge.12063. [DOI] [PubMed] [Google Scholar]
- 28.Sarajlija A, Kisic-Tepavcevic D, Nikolic Z, et al. Epidemiology of rett syndrome in serbia: prevalence, incidence and survival. Neuroepidemiology. 2015;44(1):1–5. doi: 10.1159/000369494. [DOI] [PubMed] [Google Scholar]
- 29.Leonard H, Bower C, English D. The prevalence and incidence of Rett syndrome in Australia. European child & adolescent psychiatry. 1997;6(Suppl 1):8–10. [PubMed] [Google Scholar]
- 30.Bienvenu T, Philippe C, De Roux N, et al. The incidence of Rett syndrome in France. Pediatric neurology. 2006 May;34(5):372–375. doi: 10.1016/j.pediatrneurol.2005.10.013. [DOI] [PubMed] [Google Scholar]
- 31.Motil KJ, Morrissey M, Caeg E, Barrish JO, Glaze DG. Gastrostomy placement improves height and weight gain in girls with Rett syndrome. Journal of pediatric gastroenterology and nutrition. 2009 Aug;49(2):237–242. doi: 10.1097/MPG.0b013e31818f61fd. [DOI] [PubMed] [Google Scholar]
- 32.Villard L. MECP2 mutations in males. J Med Genet. 2007 Jul;44(7):417–423. doi: 10.1136/jmg.2007.049452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hagberg B. Rett syndrome: long-term clinical follow-up experiences over four decades. Journal of child neurology. 2005 Sep;20(9):722–727. doi: 10.1177/08830738050200090401. [DOI] [PubMed] [Google Scholar]
- 34.Jian L, Archer HL, Ravine D, et al. p.R270X MECP2 mutation and mortality in Rett syndrome. European journal of human genetics : EJHG. 2005 Nov;13(11):1235–1238. doi: 10.1038/sj.ejhg.5201479. [DOI] [PubMed] [Google Scholar]