DOI: 10.1001/jamapediatrics.2026.4471 ISSN: 2168-6203

Diagnostic Yield of Genetic Testing in Cerebral Palsy

Sun-young Joo, Eun Jae Ko, Bo Ryun Kim, Jaewon Kim, Jiyong Kim, Hye Jung Park, Jin A Yoon, You Gyoung Yi, Hyun Jung Lee, Hoo Young Lee, Ah-Ra Cho, Ja Young Choi, Seungbeen Hong, Juntaek Hong, Dae-Hyun Jang
Show PDF Cite

Importance

Genetic etiologies are increasingly recognized in cerebral palsy (CP), yet diagnostic yields across different clinical subgroups and phenotypes remain inconsistent, limiting evidence-based prioritization of genetic testing.

Objective

To determine the pooled diagnostic yield of genetic testing in CP and identify clinical and phenotypic predictors that influence diagnostic yields.

Data Sources

A systematic search of PubMed, Embase, and the Cochrane Library was conducted for studies published between January 2010 and August 25, 2025.

Study Selection

Peer-reviewed observational studies reporting genetic testing results in cohorts of at least 10 patients with CP.

Data Extraction and Synthesis

Data were extracted independently by 2 reviewers following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. A random-effects model with logit transformation was used to calculate pooled diagnostic yields with 95% CIs. Heterogeneity was evaluated using the I 2 statistic and Cochran Q test.

Main Outcomes and Measures

The primary outcome was the pooled diagnostic yield of genetic testing stratified by CP etiology (cryptogenic, noncryptogenic, and mixed/unselected cohorts). Secondary outcomes included diagnostic yields across phenotype-based subgroups defined by brain magnetic resonance imaging (MRI) findings, motor types, comorbidities, facial dysmorphism, congenital anomalies, and clinical trajectory.

Results

A total of 31 studies met inclusion criteria, of which 20 were included in the quantitative meta-analysis. The overall pooled diagnostic yield in unselected CP cohorts was 0.19 (95% CI, 0.11-0.32; I 2  = 95.1%). In patients with cryptogenic CP, the yield was 0.44 (95% CI, 0.37-0.50; I 2  = 72.2%), compared with 0.13 (95% CI, 0.07-0.21; I 2  = 85.1%) in noncryptogenic cases. Higher yields were observed in patients with normal brain MRI findings (25%-80%), facial dysmorphism (66%-71%), congenital anomalies (85%), and neurological atypical features (50%). Combined next-generation sequencing and chromosomal microarray analysis provided the highest pooled yield of 0.53 (95% CI, 0.43-0.62) in the cryptogenic subgroup.

Conclusions and Relevance

In this systematic review and meta-analysis, genetic testing demonstrated a high diagnostic yield in CP, particularly in cryptogenic cases and in patients with specific clinical phenotypes. These findings suggest the integration of genomic evaluation into the standard diagnostic workup for CP, guided by clinical indicators, to facilitate timely diagnosis and personalized management.

More from our Archive