Quality of life in children with tuberous sclerosis complex: A pediatric cohort study

Abstract Aims To evaluate the quality‐of‐life (QOL) impairment and identify the possible risk factors in patients with tuberous sclerosis complex (TSC) in China. Methods The parent proxy‐report PedsQL 4.0 Generic Core Scales were administered to 124 caregivers of children with TSC (aged 2‐18 years). For comparison, the survey was also conducted in a demographically group‐matched sample of healthy controls (HCs) (aged 2‐18 years). Results A total of 124 children with TSC and 206 HCs were recruited. The mean parent proxy‐report total scale score, physical health summary score, and psychosocial health summary score for children with TSC were 65.0 (SD 19.7), 77.6 (SD 22.9), and 58.0 (SD 21.3), respectively, compared with the HC values of 83.6 (SD 14.3), 87.2 (SD 16.9), and 82.8 (SD 15.9). There were statistically significant differences between the two groups (P < .0001). TSC2 mutation (P = .033), epilepsy (P = .011), seizure before 2 years old (P = .001), course of epilepsy (more than 2 years) (P = .001), high reported seizure frequency (more than once a month) (HRSF) (P = .007), multiple antiepileptic drugs (≥2) (P = .002), intellectual disability (ID) (mild and moderate ID, P < .0001, and severe and profound ID, P < .0001), and TANDs (P < .0001) (ADHD, P = .004; agoraphobia, P = .007; and social anxiety disorder, P < .0001) were closely related to lower QOL scores. Conclusion This study is the first large cohort study on QOL in children with TSC in China. The results of the PedsQL 4.0 indicated that the QOL of children with TSC is significantly lower than that of HCs. TSC2 mutation, epilepsy, early onset, long disease course and HRSF, ID, and TANDs are risk factors for poor QOL.


| INTRODUC TI ON
Tuberous sclerosis complex (TSC, OMIM #191100, #613254), a genetic disease with an autosomal dominant inheritance, affects those who have it across their life span and occurs in approximately 1 in 5000-10 000 live births. [1][2][3] In our previous study, 4  repeatable, and suitable for general and disease-specific populations. [8][9][10][11][12] This cross-sectional study was designed to (a) investigate the QOL of Chinese patients with TSC (aged 2-18 years); (b) compare their QOL with that of healthy controls (HCs); and (c) more precisely explore risk factors for lower QOL scores.

| Subjects
During the inclusion period (June 2019-December 2019), all children with TSC aged 2-18 years and their parents received an invitation from Children's Hospital of Fudan University. A diagnosis of TSC was based on the latest diagnostic criteria for TSC. 13 The only exclusion criterion was an insufficient understanding of the Chinese language or unwillingness to participate in the study. A total of 206 healthy individuals aged 2-18 years were recruited from schools to create an age-and sex-matched control group.
This study was subject to approval by the ethics committee of the Children's Hospital of Fudan University (2018-No.26). Written informed consent and verbal assent were provided by all parents and participants (aged ≥ 12 years).

| Standardized questionnaire
Demographic characteristics (sex, birth date, ethnicity, years of education of the parents, monthly household income, and urban/rural residence) and clinical data (age at onset of epilepsy, course of epilepsy, antiepileptic drugs, seizure frequency, family history, multisystem clinical manifestations of TSC, etc) were assessed by a standard structured questionnaire.

| PedsQL 4.0 generic core scales
The Mandarin Chinese version of the PedsQL 4.0 parent proxy report was used to assess the QOL of Chinese children with TSC.
There are four age-group scales specific to children aged 2-4, 5-7, 8-12, and 13-18 years, and these are used to assess parents' perceptions of the QOL of children with TSC. The 21-item (2-4 years)/23-item  [14][15][16][17][18][19] Emotional, school, and social functioning can also be united into a psychosocial health domain. Each item is rated using a 5-point response scale from 0 = "never is a problem" to 4 = "almost always a problem." The weights of the items are equal, the score is reversed, and the linear conversion is 0-100 points (0 = 100, 1 = 75, 2 = 50, 3 = 25, 4 = 0). Thus, the higher the function score is, the better the QOL is. Face-to-face interviews were conducted by trained neurologists to collect the data.

| Statistical analysis
The statistical analyses were conducted with JMP Pro software, version 14.3.0 (SAS Institute, USA). The Shapiro-Wilk normality test was used to test the normality of the data distribution. The unpaired t test for continuous data and the chi-square test or Fisher's exact test for categorical data were used to test for sociodemographic differences between TSC patients and HCs. The unpaired t test was used for the comparison of the parent proxy-report PedsQL 4.0 scores between the two groups. ANOVA F tests were used to compare the PedsQL 4.0 subscores among all of the age-groups and different genotype groups (TSC1, TSC2, and NMI). We used multivariate linear regression models to identify risk factors for low QOL scores.
Some potential confounding factors were adjusted in Models 1b, 2b, and 3b, such as sex, parents' years of education, monthly household income, and residence. A P-value < .05 was considered significant.

| Characteristics of the study population
Altogether, 124 children with TSC (median age ± standard deviation:

| Comparison between the TSC group and healthy controls
The mean total scale score, physical health summary score, and psy- There were statistically significant differences between the two groups (P < .0001) ( Figure 1A). The QOL of children with TSC was significantly worse than that of HCs. As shown in Figure 1B, there were also significant differences in the social, emotional, and school functioning scale scores between the two groups.
There were no significant differences among any of the agegroups for any of the QOL domains (P > .05). On average, the parent-reported psychosocial health scores were significantly

| Analysis of risk factors for lower QOL scores
Multiple linear regression models adjusted for potential confounders were used to explore the associations between QOL scores and clinical manifestations of TSC (including TANDs).
As shown in Table 3, high reported HSF (P = .015), multiple antiepileptic drugs (≥2) (P < .0001), ID (mild and moderate ID, P = .001, and severe and profound ID, P < .0001), and TANDs (P < .0001) (social anxiety disorder, P = .002) were closely related to lower physical health summary scores after adjusting for sex (male/female), mother's years of education, father's years of education, monthly household income, and residence in Model 2b.
As shown in Table 4

| D ISCUSS I ON
As far as we know, this study was the first cohort study of QOL in children with TSC in China. A total of 65.3% of patients presented with ID, which is why we chose the parent proxy version. The mean total scale score, psychosocial score, and physical health score for the children with TSC were significantly lower than those in the healthy population. QOL in patients with TSC was significantly compromised compared with that in the healthy group, as might be expected.
Notably, the psychosocial health scores were significantly lower than the physical health scores, which was consistent with previous studies. 20,21 As in previous studies, TANDs can exist at all ages. 5 The lifetime cumulative incidence of TANDs is approximately 90%. 22 The incidence of neuropsychiatric comorbidities was significantly higher in TSC patients than in HCs. 7 .007 * Note: Abbreviations: ADHD, attention-deficit/hyperactivity disorder; AED, antiepileptic drugs; HRSF, high reported seizure frequency (more than once per month); ID, intellectual disability; IQ, intelligence quotient; NMI, no mutation identified; TANDs, tuberous sclerosis-associated neuropsychiatric disorders.
Data are presented as n (%).

| CON CLUS IONS
Data are presented as n (%).

TA B L E 4
The associations between the psychosocial health summary scores and clinical manifestations of TSC (including TANDs) .003 * Note: Abbreviations: ADHD, attention-deficit/hyperactivity disorder; AED, antiepileptic drugs; HRSF, high reported seizure frequency (more than once per month); ID, intellectual disability; IQ, intelligence quotient; NMI, no mutation identified; TANDs, tuberous sclerosis-associated neuropsychiatric disorders.
Data are presented as n (%).

ACK N OWLED G M ENTS
We thank the participants and Dr X Wang for help with statistical support.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used during and/or analyzed during the current study are available from the corresponding author on request.