Impairments and comorbidities in adults with cerebral palsy and spina bifida: a meta-analysis

Introduction Aging with a childhood-onset disability, such as cerebral palsy (CP), spina bifida (SB), and muscular diseases (MD), comes along with significant impairments and comorbidities. Despite the increasing evidence an overall picture is lacking. This study aimed to review the literature about adults with CP/SB/MD and impairments and comorbidities to perform a meta-analysis. Materials and methods Embase, PubMed, Cinahl, and Google Scholar were searched (2000–2020). Search terms included adults with one of the aforementioned disabilities combined with impairments and comorbidities. If specific impairments or comorbidities were reported by at least four studies, these were included in the study. Pooled prevalence (95% Confidence Interval) of impairments/comorbidities were calculated. Results The search yielded 7,054 studies of which 95 were included in the meta-analysis (64 CP, 31 SB, 0 MD). In total estimates were calculated for 26 (CP) and 11 (SB) outcomes. In adults with CP, pain [56.4% (95%CI 48.8–63.8)], deformities [44.2% (95%CI 12.9–78.4)], intellectual disability [37.2% (95%CI 26.7–48.3)], and fatigue [36.9% (95%CI 24.6–50.1)] were most prevalent; renal disease [3.0% (95%CI 2.1–4.2)] and stroke/rheumatic diseases {4.8% (95%CI 3.4–6.5; 4.8% (95%CI 1.5–9.9)] respectively} were least prevalent. For adults with SB, bladder incontinence [60.0% (95%CI 50.5–69.2)], bowel incontinence [49.2% (95%CI 34.5–64.0)], pain [44.1% (95%CI 27.4–61.5)], and sleeping problems [30.3% (95%CI 4.7–65.8)] were most prevalent; diabetes [4.8% (95%CI 2.8–7.3)] and renal disease [8.7% (95%CI 2.0–19.9)] were least prevalent. The included studies showed large heterogeneity. Conclusions More research is needed to study health issues in adults with MD. Adults with CP or SB deal with a variety of health issues. More attention for the mental health of these adults is needed. There also is a need for accessible and adequate screening, preventive measures and clinical follow-up.

Pain, fatigue, epilepsy and asthma are prevalent in adults with CP (11). In addition, these adults are at risk of several health complaints, including hypertension, depressive symptoms, osteoarthritis, cardiovascular diseases, type 2 diabetes (6, 12, 13). Adults with SB often experience bladder and bowel problems and fatigue (14, 15); fecal incontinence is more often observed with increasing age. Moreover these adults are at risk of renal failure (16). Adult men with DMD/BMD report urine incontinence (17), as well as psychiatric problems such as depressive and stress symptoms (18) and cardiac and renal dysfunction (19). Pain and fatigue are also common (20).
Recently three systematic reviews were published on adults with CP (11,21,22). These studies focused on specific health issues (pain and hypertension) or aimed at the most studied outcomes (including participation). As shown, adults with SB or MD develop significant health issues. However, they experience many barriers to healthcare services and screening (23,24), hampering timely detection and secondary preventive measures. To inform both healthcare professionals as well as adults with CP, SB, or MD, we aim to estimate the prevalence of a broader scope of impairments and comorbidities in these adults. The present study goes beyond focusing on one diagnosis group and had more strict criteria to include outcomes to have more robust estimates. As such, it provides a broader overview of comorbidities that people with CP, SB, or MD often have to deal with than current literature does.

. . Study design and participants
We conducted a systematic review of the literature including meta-analysis to estimate the prevalence of impairments and comorbidities in adults with CP, SB, or MD. No review protocol was prepared, and the review was not registered in any register.

. . Search strategy
A search strategy was formulated and used in four databases: Embase, Pubmed, Cinahl, and Google Scholar. Search terms included the conditions "cerebral palsy", "spina bifida", "spinal muscular atrophy", and "Duchenne muscular dystrophy" in combination with possible impairments and comorbidities such as "fatigue", "pain", and "diabetes. Some impairments and comorbidities were not included as search terms, but were still picked up, because they were often included as one among other outcomes in studies. This was the case for osteoporosis, obesity (as reflected by BMI) and gastroenterological problems. The full search strategy for Pubmed is presented in Supplementary material 1. After removing duplicates, publications were screened on title and abstract to check for eligibility (by both reviewers). Subsequently, full texts were screened, and disagreements were discussed and resolved.

. . Selection criteria
Studies were included if they met the following criteria: 1. Published in the period January 1 st 2000-December 31 th 2020; 2. Including a study sample of n ≥ 25; 3. All participants ≥18 years of age; 4. Not a follow-up intervention study; 5. No selected samples (i.e., only dyskinetic CP).
Of longitudinal designs, the most recent follow-up study with the specific comorbidity or impairment as outcome was included.

. . Data extraction
Data extraction was done by both reviewers with a standardized data extraction form in Microsoft Excel. Study sample characteristics and number of cases reported to have an impairment or a comorbidity were recorded for every study. Sample characteristics included first author and publication year, country, sex, age (mean/median), sample size and for CP the GMFCS levels (25). All impairments or comorbidities reported in the studies were recorded, but if less than four studies reported on a specific impairment or comorbidity, we did not include this outcome in the analysis.

. . Data analysis
Overall mean proportions and 95% Confidence Interval (95%CI) were estimated for each comorbidity. Random-effects meta-analysis models were used (with DerSimonian and Laird estimator). Meta-analysis modeling was done using the proportion meta-analysis function in StatsDirect. The random-effects model takes the heterogeneity of samples into account. The I 2 measure indicated heterogeneity and represents the variation attributed to heterogeneity rather than sampling error across samples.

. . Study characteristics
The full selection process is presented in Figure 1. In total 110 (of 7,054) studies met the inclusion criteria. Of these, 15 were excluded because these studies reported on impairments or comorbidities that were studied in less than three other studies. Of the 95 included studies, 64 reported on CP and 31 reported on SB. Regarding adults with MD, no outcome was reported more than three times, and therefore no results on health issues of these adults could be described. Risk of bias was assessed with a quality assessment checklist for prevalence studies that Nguyen and colleagues (26) adapted from Hoy and colleagues (27) and is presented in Table 1.
Almost half of all studies were conducted in the United States of America (USA) (n . = 47), followed by eleven studies from Sweden (12%), eight studies conducted in the United Kingdom (UK) (8%), and seven studies from The Netherlands (7%). Other studies were Sattoe and Hilberink .
/fneur. . from different countries all over the world, but the number of studies for specific countries was small (range: 1-4). Sample size of the included studies varied from 26 to 17,212 people with CP or SB. One study only included females in the study population  and one study included only males (Mezaal et al., 2009). Other studies had mixed study populations in terms of sex, the proportion of males varied from 25 to 69%. Of these, 53 studies (57%) had a more or less equal distribution of males and females in the study population (between 45 and 55%). All study characteristics are presented in Table 1.
. . Health issues in adults with CP Figure 2 shows the number of studies per impairment or comorbidity and the estimated prevalence (95%CI) of these in adults with CP. A total overview of the health issues, the studies that reported on them, and the number of cases included in the analyses is given in Supplementary material 2. For all analyses the level of heterogeneity (I 2 ) was high (>70%), indicating substantial variation in results across the studies.
The health issue that was most often included in study designs was hypertension (n = 22), followed by pain (n = 21) and epilepsy (n = 19). Least studied were stroke (n = 4), cancer (n = 4), bowel problems (n = 4), and anxiety (n = 4). The most prevalent comorbidity in adults with CP was pain, the overall prevalence was 56%. Deformities were the second most prevalent (44%). Intellectual disability, fatigue, obesity and hypertension, asthma, epilepsy, depression and anxiety were prevalent in more than 20% of adults with CP. Least common were renal diseases (3%), stroke (5%), and rheumatic disorders (5%). Figure 3 shows the number of studies per impairment or comorbidity and the overall proportion of these in adults with SB. A total overview of the health issues, the studies that reported on them and the number of cases included in the analyses is given in Supplementary material 3. For almost all of the analyses the level of heterogeneity was also high (>70%) except for diabetes (I 2 = 41%).

. . Health issues in adults with SB
The health issue that was most often included in study designs was bladder incontinence (n = 13), followed by epilepsy (n = 9) and renal disease (n = 8) and bowel incontinence (n = 8). Least studied were diabetes (n = 4), obesity (n = 4), and sleeping problems (n = 4). Bladder incontinence was most prevalent (60%), followed by bowel incontinence (49%) and pain (44%). Urinary tract infection, obesity and depression were present in more than 20% of adults with SB. Least common was diabetes (5%).

. . Main findings
This meta-analysis is the first to review a wide scope of impairments and comorbidities in adults with CP or SB. Impairments and comorbidities in adults with MD could not be assessed due to limited studies. The overall picture may inform health professionals and adults with CP or SB about the common prevalent health issues.
For adults with CP, the results show a lower estimated prevalence of pain (56%) compared to a previous meta-analysis [70%; (21)]. This perhaps has to do with the difference in included studies and thus in other samples concerning sex, age, GMFCS      levels, and subtypes. Nevertheless, pain is the most prevalent impairment reported, not only in our study but also in another recent study (11); it was also the second most studied health issue. Attention for pain as a health problem in adults with CP is important in clinical practice (118). This is also the case for adults with SB: They also have a high prevalence of pain (44%), but less often studied. Pain is found to be an important health issue for almost all people with childhood-onset disabilities, starting from a young age. Research also shows that pain can have a profound impact on quality of life and mental health (119). There seems to be a lack of attention for mental health. While for both depressive symptoms and anxiety, the prevalence rates in adults with CP are around 20%, these outcomes belong to the less studied ones, and this is especially true for anxiety. On the contrary, epilepsy and hypertension are not much more prevalent, but far more often studied, with hypertension being the most studied outcome among adults with CP in this review. For adults with SB, the same pattern was found. While depressive symptoms were prevalent in almost 25% of the people with SB, these were studied in <50% of the studies included. Other original studies also highlighted the risk of depression and anxiety in people with CP and SB and the need for more attention for mental health in these groups (94,120,121). Moreover, literature suggests that comprised mental health is associated with health issues such as pain and fatigue (4,20). The results call for attention for mental health.
The overall results of this meta-analysis show profound health issues that people with CP and SB have to deal with. They have increased medical needs compared to the general population. Yet, screening of people with CP or SB on health issues is not common practice yet (122)(123)(124) and access to needed healthcare is not always self-evident (125). More attention is needed for this matter of how current healthcare practice can be tailored to these increased needs of people with CP or SB. The need for prevention and clinical follow-up of health issues (including mental health) has been emphasized before (120,126). Moreover, comorbidities not only reflect medical challenges, preventive measures may positively impact social participation of adults with lifelong disabilities as well (125, 127).

. . Limitations
It is important to acknowledge the high levels of heterogeneity (I 2 ) in our analyses, indicating substantial variation in results across studies. These levels show that there is no clear pattern of comorbidities or impairments across studies. Yet, we felt it appropriate to summarize the outcomes, because the level of heterogeneity can also be influenced by the fact that outcome measures were not measured in a uniform way across studies. Also, there are differences in sex, age, disability and subtypes of conditions in the study samples. A limitation of this study is also . /fneur. . that, due to a small number of studies, other conditions than CP and SB (e.g., MD) could not be included in the analysis. Finally, it must be emphasized that most studies included in this metaanalysis were performed in high-income countries. Therefore, it is not representative for the whole world. More research in low-and middle-income countries is warranted.

. . Conclusions
Health issues in adults with MD are studied too less to perform a meta-analysis. Hence research on the impairments and comorbidities in this population is strongly recommended to inform health professionals and the adults themselves. Adults with CP or SB have to deal with a variety of health issues next to their main disability. Pain is found to be the most prevalent issue and can have profound impact on quality of life and mental health. Mental health of adults with CP or SB seems to be understudied and it is important to gain insight into useful interventions for mental wellbeing in these adults. There also is a need for accessible and adequate screening, preventive measures and clinical follow-up of health issues.

Author contributions
JS and SH contributed to the study conception and design, data collection, and interpretation of results. JS performed data analysis and drafted the manuscript. Both authors reviewed the results and approved the final version of the manuscript.

Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.