Abstract
Importance
Rapid whole-genome sequencing (WGS) has advanced inpatient genetic diagnosis, although most evidence comes from neonatal and pediatric intensive care unit (ICU) populations. Its utility for pediatric neurology inpatients across broader acute settings remains insufficiently established.
Objective
To assess the diagnostic yield, clinical impact, and factors associated with phenotype-concordant rapid WGS diagnoses.
Design, Setting, and Participants
This was a retrospective cohort study conducted from June 1, 2022, through January 17, 2026, at a tertiary medical center. Pediatric inpatients in the neonatal ICU, pediatric ICU, or general floor with an unexplained neurologic presentation were included. Those with established nongenetic etiology were excluded from analyses.
Exposures
Trio (77.8%), duo (9.1%), or singleton (13.1%) rapid WGS.
Main Outcomes and Measures
The primary outcome was the initial phenotype-concordant diagnostic yield at the point of care. Secondary outcomes were reanalysis yield (new diagnoses from reinterpreting nondiagnostic cases prompted by a new unexplained phenotype or family request), cumulative yield, and management change. Associations were assessed by logistic regression (odds ratios [ORs] and 95% CIs).
Results
Among 175 children (median [IQR] age, 4 months [1 day-4 years]; 93 male [53.1%]) who met inclusion criteria (82 neonates, 93 nonneonates), initial rapid WGS established a phenotype-concordant diagnosis in 79 (45.1%), higher in nonneonates (49 of 93 [52.7%]) than neonates (30 of 82 [36.6%]). Preliminary results returned at mean (SD) 4.0 (1.6) days and final reports at 10.4 (4.5) days. Positive nonconcordant findings occurred in 33 children (18.9%). Reanalysis of 14 nondiagnostic cases yielded 3 additional diagnoses, for a cumulative yield of 82 of 175 children (46.9%). Most diagnoses were monogenic (63 [76.8%]). Management changed in 54 of 82 phenotype-concordant cases (65.9%; 53 at the point of care, 1 after reanalysis) and 8 of 33 (24.2%) positive nonconcordant cases. Family history was most strongly associated with diagnosis (adjusted OR [aOR], 6.46; 95% CI, 3.23-12.93;
P
< .001), followed by multisystem involvement (aOR, 2.85; 95% CI, 1.24-6.54;
P
= .01) and congenital anomaly or dysmorphism (aOR, 2.47; 95% CI, 1.07-5.68;
P
= .03). Hypotonia was most strongly associated in neonates (OR, 20.63; 95% CI, 5.46-77.87;
P
< .001); ICU admission (OR, 10.25; 95% CI, 2.79-37.67;
P
< .001) predominated in nonneonates.
Conclusions and Relevance
Results of this cohort study suggest that rapid WGS diagnosed nearly half of pediatric neurology inpatients and changed management in two-thirds of diagnosed cases, supporting its potential use as early diagnostic testing. Associated features may guide patient selection, and reanalysis extends yield longitudinally.