Rapid whole-genome sequencing in the pediatric intensive care unit: from etiologic diagnosis to precision management—a narrative review

Abstract

Background and Objective

Children admitted to the pediatric intensive care unit (PICU) with critical illness may have underlying genetic diseases that are difficult to distinguish from sepsis, hypoxia, shock, postoperative complications, medication effects, or secondary organ injury. Rapid whole-genome sequencing (rWGS) provides a broad genomic framework within clinically actionable timeframes. This narrative review aims to summarize the diagnostic scope, implementation workflow, and clinical utility of rWGS in critically ill children.

Methods

A focused literature search was conducted in PubMed using terms related to rWGS, rapid genome sequencing, rapid exome sequencing (rES), PICU, critical illness, and critically ill children. Original studies, multicenter studies, and relevant consensus documents were considered, and the primary literature search was conducted on June 25, 2026, and the final update was performed on August 20, 2026.

Key Content and Findings

Evidence from critical care cohorts indicates that rapid genomic sequencing can increase etiologic diagnosis and influence clinical management through disease-specific therapy, metabolic crisis management, adjustment of antimicrobial or immunomodulatory treatment, avoidance of contraindicated interventions, reduction of low-yield procedures, prognostication, and family risk counseling. Clinical utility depends on appropriate patient selection, rapid turnaround time, trio analysis when feasible, standardized phenotyping, and multidisciplinary interpretation.

Conclusions

rWGS can support precision management in critically ill children when integrated into a structured PICU genomic pathway. Nondiagnostic results and variants of uncertain significance require continued reassessment and complementary testing.

Introduction

Critically ill children admitted to the pediatric intensive care unit (PICU) may harbor underlying genetic diseases that are difficult to recognize during acute deterioration. Inherited metabolic disorders, mitochondrial diseases, neurogenetic disorders, inborn errors of immunity, inherited cardiomyopathies, arrhythmia syndromes, and syndromic genetic conditions can all present with overlapping critical-care phenotypes, including seizures, encephalopathy, metabolic acidosis, hyperammonemia, hypoglycemia, hyperlactatemia, cardiomyopathy, arrhythmia, severe infection, congenital anomalies, and multisystem organ dysfunction (1,2). In the PICU, these manifestations are frequently attributed to sepsis, hypoxia, shock, postoperative complications, drug effects, or secondary organ injury, especially when the phenotype is incomplete or modified by resuscitation, organ support, antimicrobial treatment, nutritional intervention, or renal replacement therapy (3).

Conventional diagnostic strategies remain important but are often poorly aligned with the tempo and complexity of PICU practice. Targeted biochemical tests, enzyme assays, immunologic testing, chromosomal microarray analysis (CMA), and gene panels can provide valuable information, but some depend on a prespecified clinical suspicion or interrogate a restricted range of abnormalities. Whole-exome sequencing (WES) provides broad analysis of protein-coding regions and remains an important genomic diagnostic approach. However, when these investigations are performed sequentially, the overall diagnostic process may extend beyond the available clinical decision window (3,4). Rapid genomic sequencing approaches, including rapid whole-genome sequencing (rWGS), ultra-rapid genome sequencing (urGS), and rapid exome sequencing (rES), have therefore emerged as clinically relevant diagnostic approaches for critically ill infants and children. Genome sequencing provides a broad single-test framework for the analysis of single-nucleotide variants (SNVs), insertions and deletions (indels), copy-number variants (CNVs), selected structural variants, uniparental disomy (UPD), mosaic aneuploidy, and mitochondrial DNA (mtDNA)-related findings, while supporting phenotype-driven interpretation and multidisciplinary decision-making (2,5).

This article is a narrative review focusing on the clinical application of rWGS in the PICU. This review summarizes the role of rWGS in critically ill children with suspected genetic disease in the PICU. First, it outlines the major disease groups and diagnostic challenges that make genetic etiologies difficult to identify in acute critical illness. Secondly, the discussion covers the technical scope, workflow, trigger criteria and interpretive requirements of rWGS. It emphasizes trio sequencing, standardised Human Phenotype Ontology (HPO)-based phenotyping, dynamic phenotype updating and multidisciplinary review. Finally, it assesses the clinical utility of rWGS by considering diagnostic yield, turnaround time, management change, disease-specific therapy, avoidance of inappropriate interventions, prognostication, genetic counseling, and future implementation challenges (2,5). We present this article in accordance with the Narrative Review reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0625/rc).

Methods

A focused literature search was performed to identify relevant studies evaluating rapid genomic sequencing in critically ill pediatric populations. PubMed was searched using combinations of the following terms: “rapid whole-genome sequencing”, “rapid genome sequencing”, “rapid exome sequencing”, “pediatric intensive care unit”, “PICU”, “critical illness”, and “critically ill children”. Studies were selected based on relevance to PICU practice, including diagnostic yield, turnaround time, management impact, implementation strategies, and genetic counseling. Original research articles, multicenter studies, and consensus documents were considered. Neonatal studies were included when they provided important evidence applicable to pediatric critical care, particularly regarding rapid sequencing implementation and clinical utility. Additional relevant articles were identified through reference screening. The primary literature search was conducted on June 25, 2026, and the final update was performed on August 20, 2026. The literature search strategy is summarized in Table 1, and the detailed PubMed search strategy is provided in Table S1.

Table 1. The search strategy summary.

Items	Specification
Date of search	June 25, 2026 (primary search); August 20, 2026 (final update)
Databases and other sources searched	PubMed, additional relevant articles were identified through reference screening
Search terms used	“rapid whole-genome sequencing”, “rapid genome sequencing”, “rapid exome sequencing”, “pediatric intensive care unit”, “PICU”, “critical illness”, and “critically ill children”
Timeframe	January 2020 to June 25, 2026 (primary search); January 2020 to August 20, 2026 (final update)
Inclusion and exclusion criteria	Original research articles, multicenter studies, and relevant consensus documents related to rapid genomic sequencing in critically ill pediatric populations were considered. Neonatal studies were included when they provided evidence applicable to pediatric critical care
Selection process	X.Y. and D.L. independently screened the retrieved literature for relevance to the review topic. Disagreements were resolved through discussion and consensus

Disease spectrum and diagnostic challenges in critically ill children with suspected genetic disease in the PICU

For critically ill children with suspected genetic disease in the PICU, the main diagnostic difficulty lies not only in the wide range of genetic disorders, but also in the way diverse molecular diseases converge on a restricted group of acute critical-care phenotypes. Inherited metabolic disorders, mitochondrial disorders, neurogenetic diseases, primary immunodeficiencies and immune dysregulation disorders, inherited cardiomyopathies, and syndromic genetic conditions can each appear in the PICU as hypoglycemia, hyperlactatemia, metabolic acidosis, seizures, encephalopathy, hypotonia, arrhythmia, cardiomyopathy, uncontrolled infection, congenital anomalies, or multisystem dysfunction (1,6). Because these presentations rarely point to a single disease entity, they are readily subsumed under infection, hypoxia, shock, drug effects, or secondary organ injury, thereby delaying consideration of an underlying genetic etiology (7).

Acute metabolic decompensation is the principal route by which inherited metabolic diseases enter the PICU. Infection, fasting, dehydration, surgery, fever, or accelerated catabolism may convert a previously compensated defect, such as an organic acidemia, urea cycle disorder, fatty acid oxidation disorder, or glycogen storage disease, into hyperammonemia, hypoglycemia, metabolic acidosis, hyperlactatemia, hepatic injury, encephalopathy, seizures, or coma (8,9). These presentations are often interpreted as septic shock, hypoxic-ischemic injury, acute liver failure, or secondary metabolic derangement. Once fluid resuscitation, glucose infusion, nutritional support, renal replacement therapy, or antimicrobial treatment has begun, glucose, lactate, ammonia, ketones, amino acids, and organic acid profiles may be rapidly altered, further reducing the diagnostic specificity of conventional metabolic markers (10-12).

For mitochondrial disease, the critical-care entry point is multisystem bioenergetic failure. Impaired oxidative phosphorylation can simultaneously affect tissues with high energy demand, including the brain, myocardium, liver, skeletal muscle, renal tubules, hematopoietic system, and respiratory system, resulting in hyperlactatemia, hypotonia, seizures, encephalopathy, liver failure, hypoglycemia, rhabdomyolysis, cardiomyopathy, arrhythmia, respiratory failure, and growth impairment (13,14). In the PICU, mitochondrial disease frequently mimics infectious encephalopathy, hypoperfusion-associated lactate elevation, drug-related liver injury, primary cardiomyopathy, or respiratory failure. Even when suggestive clues such as stroke-like episodes, lesion migration, or neuroimaging abnormalities that do not conform to a vascular territory are present, interpretation often requires integration of nuclear genes, mtDNA variants, CNVs, deep intronic variants, selected structural variants, and, increasingly, long-read or multi-omic approaches in unresolved cases (15-17).

Neurogenetic disorders usually present to the PICU as early-onset, persistent, or refractory seizures, often accompanied by abnormal brain development or neuronal excitability. Neurogenetic disorders may present with seizures, encephalopathy, developmental abnormalities, or progressive neurological deterioration. Genetic diagnosis is particularly valuable when clinical manifestations are atypical or when conventional evaluations fail to explain the phenotype (15,18,19). Early in the course, such children are commonly managed within pathways for hypoxic-ischemic encephalopathy, stroke, intracranial hemorrhage, infection, hypoglycemia, hypocalcemia, or electrolyte disturbance. A genetic epilepsy becomes more likely when magnetic resonance imaging (MRI) does not show a sufficient acute lesion, electroencephalogram (EEG) findings are discordant with the presumed acute trigger, seizures recur, or standard antiseizure medications are ineffective (20).

Primary immunodeficiencies and immune dysregulation disorders are particularly difficult to recognize in the PICU because infection susceptibility and inflammatory dysregulation frequently coexist. Inborn errors of immunity may present with severe infection, immune dysregulation, cytopenia, or multisystem inflammation. Their variable manifestations and overlap with acquired critical illness can delay recognition of genetic causes (21-23). However, prematurity, corticosteroid exposure, malnutrition, and critical illness itself may be associated with transient lymphopenia, altered immunoglobulin levels, or nonspecific inflammatory biomarker abnormalities, which can complicate interpretation of immune testing in the PICU. As a result, single time-point measurements of IgG, T-cell receptor excision circles (TREC), lymphocyte subsets, or pathogen testing rarely provide a definitive diagnosis in isolation (14,24,25).

Inherited cardiomyopathies, inherited arrhythmia syndromes, and syndromic genetic disorders must be interpreted within a systemic genetic framework. Heart failure, malignant arrhythmia, QT prolongation, hypoperfusion, hyperlactatemia, hepatic or renal dysfunction, and respiratory failure may arise from defects in sarcomeric proteins, ion channels, mitochondrial energy metabolism, fatty acid oxidation, glycogen metabolism, or the cytoskeleton, but they may also reflect secondary injury from infection, hypoxia, shock, acidosis, electrolyte disturbance, or medication exposure (26-29). Syndromic and chromosomal disorders are similarly obscured because congenital anomalies, dysmorphic features, developmental delay, feeding difficulties, cardiopulmonary abnormalities, hematologic abnormalities, immune dysfunction, and endocrine findings may be fragmented across organ systems and deprioritized during acute resuscitation (30-33).

Conventional investigations remain clinically important, but the PICU amplifies their limitations. Metabolic screening, enzyme assays, biochemical testing, immunologic evaluation, CMA, and gene panels each provide valuable information, but selection of these tests may depend on the suspected disease category or expected abnormality. WES, by contrast, provides broad analysis of protein-coding regions and does not require a narrowly defined pre-test differential diagnosis (34,35). Early test selection is difficult because critically ill children frequently show incompletely evolved phenotypes, rapid clinical deterioration, multisystem involvement, and biomarkers altered by treatment (11,36,37). Conventional diagnostic pathways may fail to provide timely etiologic confirmation in a substantial proportion of critically ill children. In a 2022 matched cohort study of critically ill pediatric patients, rWGS achieved a diagnostic yield of 39.8%, compared with 13.5% with standard diagnostic genetic testing (38). In addition, a randomized time-delayed trial of 354 acutely ill infants showed that WGS increased the proportion of patients receiving a molecular diagnosis when results were returned earlier, with WGS assigned at 15 versus 60 days after enrollment (19).

Technical advantages and implementation workflow of rWGS in the PICU

In the PICU, rWGS is best understood as a rapid etiologic diagnostic pathway integrated into resuscitation and organ support. For critically ill children with suspected genetic disease, the practical aim is to perform sample acquisition, standardized phenotyping, rapid sequencing, bioinformatic analysis, variant interpretation, and result communication during ongoing acute care, thereby generating the diagnostic substrate for downstream clinical action (39,40).

Diagnostic scope of rWGS

The main technical strength of rWGS is its ability to evaluate diverse variant classes within a single genomic framework. Compared with targeted approaches, CMA, and WES, rWGS provides broader detection of genomic alterations, which is particularly valuable for critically ill children with heterogeneous or evolving phenotypes. When parental samples are available, trio-based analysis also enables assessment of inheritance patterns and facilitates the detection of UPD. This breadth is particularly relevant for PICU patients with unexplained disease, multisystem involvement, or variant classes that cannot be predicted at presentation (39,41). In phenotypically overlapping scenarios such as mitochondrial disease, complex developmental disorders, cardiomyopathy, immunodeficiency, and metabolic disease, broader coverage may reduce delays caused by selecting an overly narrow panel or pursuing sequential testing (42,43).

Short-read WGS has recognized limitations for balanced chromosomal rearrangements, complex repeat expansions, methylation abnormalities, pseudogene-rich regions, low-complexity regions, transcript-level splicing abnormalities, and some forms of tissue-specific mosaicism. RNA analysis, long-read sequencing, methylation testing, enzyme assays, or functional validation may therefore be required in selected cases (17,42,43). The appropriate role of rWGS is to provide a broad and rapid analytic scaffold while maintaining a clear understanding of its technical boundaries and the need for orthogonal confirmation when indicated (Figure 1).

Figure 1.

Overview of the clinical role of rWGS in the PICU. This figure summarizes the rationale for applying rWGS in critically ill children, highlighting broad genomic detection, rapid etiologic diagnosis, and integration of molecular findings. CMA, chromosomal microarray analysis; CNV, copy-number variant; mtDNA, mitochondrial DNA; PICU, pediatric intensive care unit; SNV, single-nucleotide variant; SV, structural variant; rWGS, rapid whole-genome sequencing; UPD, uniparental disomy; WES, whole-exome sequencing.

PICU implementation workflow and trigger criteria

The first step in a PICU rWGS pathway is front-end case identification. Testing may be deferred when the clinical course is fully explained by isolated prematurity, a clear infection with expected treatment response, trauma, malignancy, or an already established syndrome. However, overly restrictive criteria risk missing genetic disease masked by infection, hypoxia, postoperative status, or organ support (44,45) (Table 2).

Table 2. Key indications for initiating rWGS in the PICU.

Indication	Typical clues	Clinical relevance
Early severe disease	Neonatal or infantile onset; rapid deterioration	Suggests possible monogenic disease
Multisystem involvement	Neurologic, cardiac, metabolic, immune, hepatic, or developmental abnormalities	Supports broad genomic testing
Metabolic crisis	Hypoglycemia, hyperlactatemia, acidosis, hyperammonemia	Requires urgent dietary, cofactor, or dialysis-related treatment
Neurologic critical illness	Refractory seizures, encephalopathy, severe hypotonia	Guides antiseizure or cofactor therapy
Cardiac critical illness	Cardiomyopathy, malignant arrhythmia, QT prolongation	Suggests channelopathy, mitochondrial disease, or metabolic disease
Immune/inflammatory phenotype	Severe infection, opportunistic infection, HLH-like inflammation	Guides IVIG, antimicrobial therapy, immunomodulation, or HSCT evaluation
Syndromic or familial clues	Congenital anomalies, dysmorphism, developmental delay, family history	Supports trio testing and genetic counseling

HLH, hemophagocytic lymphohistiocytosis; HSCT, hematopoietic stem cell transplantation; IVIG, intravenous immunoglobulin; PICU, pediatric intensive care unit; QT, QT interval; rWGS, rapid whole-genome sequencing.

The second step is pretest preparation and sampling strategy. Once indications are confirmed, informed consent, choices regarding secondary findings, counseling about variants of uncertain significance (VUS), explanation of the implications of a negative result, and sample collection should proceed without delay so that procedural bottlenecks do not erode the time advantage of rapid sequencing (41,46-48). Trio sequencing involving the proband and both parents should be obtained whenever feasible. When parental samples are not immediately available, singleton rWGS may be started first, with parental samples incorporated later for segregation analysis, inheritance assessment, and phasing of compound heterozygous variants (49,50).

The third step consists of standardized phenotyping, rapid sequencing, and tiered reporting. Detailed clinical phenotypes should be documented by the PICU team and subsequently mapped to standardized HPO terms by trained laboratory or genetics personnel, with computational tools used to assist phenotype extraction and standardization when available. Relevant information should include clinical presentation, neurological findings, congenital anomalies, metabolic abnormalities, cardiac and immune features, treatment response, and family history (39,49,51). The laboratory subsequently follows an urgent workflow comprising DNA extraction, quality control, library preparation, sequencing, alignment, SNV/indel calling, CNV and selected structural-variant analysis, mtDNA-related analysis, variant annotation, and candidate prioritization. Laboratory reports identify pathogenic or likely pathogenic variants and relevant variants of uncertain significance according to established reporting criteria. Cases with no reportable findings should be clearly distinguished from those with reported variants. The extent to which a reported finding explains the patient’s phenotype should then be determined through clinical correlation and multidisciplinary interpretation. Clinicians should initiate reanalysis when new phenotypic information, family data, functional evidence, or updated genomic knowledge becomes available (52) (Figure 2).

Figure 2.

Clinical workflow of rWGS implementation in the PICU. This figure illustrates the major steps from patient selection and phenotype assessment to sequencing, variant interpretation, multidisciplinary review, and clinical application of genomic findings. HPO, Human Phenotype Ontology; PICU, pediatric intensive care unit; rWGS, rapid whole-genome sequencing; VUS, variants of uncertain significance.

Determinants of interpretive efficiency

Four elements are particularly important for interpretive efficiency. First, trio sequencing rapidly distinguishes de novo variants, inherited variants, recessive inheritance, X-linked inheritance, and familial variants, thereby reducing uncertainty around VUS (49,50). Second, standardized HPO-based phenotyping strengthens candidate ranking and reverse phenotyping, whereas nonspecific inputs such as critical illness, infection, or encephalopathy alone markedly reduce interpretive strength (39,49). Third, dynamic phenotype updating is necessary because clinically relevant features can become apparent as sedation is reduced, infection is controlled, organ support changes, or the disease evolves; such new phenotypic information should be returned to variant interpretation (53). Fourth, multidisciplinary review should involve clinical geneticists, laboratory specialists, and relevant clinical specialists according to the patient’s phenotype. This collaboration supports clinical interpretation of reported variants, assessment of phenotype-genotype concordance, and translation of genomic findings into patient management (44,45,54).

AI-assisted tools may support phenotype extraction, HPO mapping, and patient prioritization. A mature PICU rWGS pathway should integrate computational support with clinical review throughout the diagnostic process (49,50,54).

Clinical utility of rWGS for critically ill children with genetic disease in the PICU

Diagnostic utility: increasing etiologic confirmation

The primary clinical utility of rapid genomic sequencing is to increase molecular diagnostic yield in critically ill children with unexplained disease, although reported yields vary across patient populations and clinical settings. In a multicenter PICU/cardiac intensive care unit (CICU) rWGS cohort, 79 of 133 children received a molecular diagnosis, corresponding to a diagnostic yield of 59% (2). In Project Baby Bear, rWGS identified a genetic diagnosis explaining intensive care unit (ICU) admission in 74 of 184 critically ill infants, corresponding to a yield of 40% (5). Randomized evidence also demonstrates downstream clinical utility. In the NSIGHT2 trial, rapid genomic sequencing led clinicians to report changes in management in 57 of 201 infants (28%) and perceived changes in clinical outcomes in 32 of 201 infants (15%) (50). These findings indicate that rapid genomic sequencing can provide substantial diagnostic and clinical utility, which depends on patient selection, testing strategy, and clinical context.

Variation in diagnostic yield is driven mainly by four factors. Firstly, the selection of patients is of significance: children exhibiting a high pretest probability, such as those with suspected genetic disease, refractory seizures, complex metabolic abnormalities, severe hypotonia, arrhythmia, multiple congenital anomalies, or multisystem involvement, are more likely to receive a diagnosis, whereas cohorts that include all ICU patients or broader critical-care phenotypes have lower yields (5,42). Secondly, it is imperative to consider the temporal aspect of genetic testing. When utilised as a primary screening modality, the implementation of urGS or rWGS can facilitate the identification of genetic diseases prior to the initiation of conventional diagnostic pathways. Conversely, when rWGS or rES is postponed until after negative CMA, panel, single-gene, or metabolic testing, there is a possibility of diminished diagnostic yield and clinical impact (33). Thirdly, the scope of the test is pertinent: Whilst WES remains a significant approach for the detection of coding variants, rWGS provides augmented coverage of non-coding regions and other variant classes, including selected structural variants and repeat expansions, within a unified analytic framework (33,55). Fourthly, the interpretation strategy is influential because trio sequencing, HPO phenotyping, secondary analysis, inclusion of VUS, reporting of secondary findings and multidisciplinary review all shape the final diagnostic rate (56,57).

These data emphasize that the diagnostic value of rWGS is not merely the detection of additional variants. Rather, in the PICU context of unexplained disease, multisystem involvement, and unstable conventional diagnostic trajectories, rWGS increases the probability of an interpretable molecular diagnosis. Results with uncertain or incomplete phenotype correlation may provide clues for further evaluation and longitudinal reassessment. Negative results can also be clinically useful by lowering the likelihood of selected genetic disorders and redirecting attention toward infection, hypoxia, postoperative status, malignancy, immune therapy-related complications, or other nongenetic causes (56,57).

Although much of the early evidence for rapid genomic sequencing was generated in neonatal intensive care unit (NICU) cohorts, PICU patients represent a partially distinct population. Compared with neonates, PICU patients often have a broader age distribution, more heterogeneous acquired triggers, more prior diagnostic testing, evolving or partially treated phenotypes, medication exposure, postoperative status, malignancy-related complications, and chronic disease backgrounds (30,42). NICU-derived diagnostic yields should therefore not be generalized directly to the PICU without considering patient selection, timing of testing, and pretest probability. PICU-specific cohorts remain particularly important because they better capture the interaction between genetic disease and infection, hypoxia, shock, organ support, and prior treatment (2,58).

Time utility: bringing diagnosis into the therapeutic window

In the PICU, the value of genetic diagnosis is tightly linked to turnaround time. Standard genetic testing often requires weeks or longer, by which time the child may have died, been discharged, transitioned to long-term care, or sustained irreversible organ injury. The key advantage of rapid genomic sequencing approaches, including urGS, rWGS, and rES, is that they compress genetic diagnosis into a period during which ICU decisions are still active (59). urGS can return results within 3 days of sample receipt, and some clinical workflows have reported provisional results in an average of approximately 36 hours. The median time to preliminary results was approximately 3 days in both the multicenter PICU/CICU rWGS cohort and Project Baby Bear. In the PICU RES study, mean time to verbal and written results was approximately 5 and 10 days, respectively (5,60).

Time utility depends on whether results arrive while ICU decisions are still modifiable. For metabolic crisis, status epilepticus, arrhythmia, immune dysregulation, severe infection susceptibility, or neuromuscular disease, etiologic confirmation within days may alter medication selection, diet, respiratory support, renal replacement therapy, immunotherapy, or the use of invasive procedures (40,61). Children who received genomic results within 15 days had management-change rates approximately twice those of children whose results returned within 60 days (19). Project Baby Bear modeling similarly showed that hospital days and costs were saved most when results returned within 3 days, with diminishing benefit when turnaround extended to 7 or 14 days (62,63). The central issue is therefore not sequencing speed alone, but whether the diagnostic result reaches the bedside during the therapeutic window.

Clinical management change: the core measure of utility

Clinical management change is the most important measure of rWGS utility in the PICU. Diagnostic yield indicates whether testing can identify an etiology, whereas management change indicates whether the molecular diagnosis has entered the decision-making process of critical care. Across 39 studies including 2,858 children, the weighted mean rate of management change was 26%, with a range of 7% to 63%; urGS was associated with a management-change rate of approximately 37%, higher than the approximately 25% reported for rWGS or rES (1). In the multicenter PICU/cardiac ICU rWGS cohort, 19 of 79 diagnosed children had an ICU management change (24%). When post-ICU management, palliative care, resuscitation status, long-term monitoring, and family screening were included, 63 diagnosed children had at least one management change (80%) (2). In Project Baby Bear, 58 of 184 children (32%) had at least one medical management change attributable to rWGS, including changes in surgery, medication, diet, and other aspects of care (5).

The first category of change is disease-specific therapy. In refractory epilepsy, channelopathies, congenital myasthenic syndromes, arrhythmia syndromes, metabolic disorders, immunodeficiencies, and inflammatory dysregulation disorders, molecular diagnosis can rapidly redirect treatment. ALDH7A1-related epilepsy supports pyridoxine therapy, PNPO deficiency supports pyridoxal 5'-phosphate therapy, KCNQ2-related and selected SCN2A-related epilepsies may support more appropriate antiseizure drug selection, and SCN1A-related disease requires avoidance of sodium-channel blockers. Congenital myasthenic syndromes, mTOR pathway-related epilepsy, channelopathies, and inborn errors of immunity may similarly become eligible for mechanism-directed therapy after diagnosis (15,40,42,61). Reported examples include eflornithine for ODC1-related Bachmann-Bupp syndrome, pyridoxal phosphate for PNPO deficiency, bisphosphonate therapy for COL1A1-related osteogenesis imperfecta, and IL-1 receptor antagonism with consideration of bone marrow transplantation in mevalonate kinase deficiency, illustrating how rapid diagnosis can directly enable precision treatment (42,64).

Pharmacogenomic information can further refine medication selection and dosing in critically ill children. CYP2D6 genotype influences the metabolism and clinical response to selected opioids, and Clinical Pharmacogenetics Implementation Consortium (CPIC) provides genotype-guided prescribing recommendations for codeine and tramadol, including alternative analgesic strategies for patients at risk of inadequate analgesia or increased toxicity (65). Pediatric implementation studies have also demonstrated the feasibility of incorporating CYP2D6-guided opioid prescribing into clinical care (66). As pharmacogenomic interpretation becomes increasingly integrated into genomic medicine, validated gene-drug associations may provide an additional basis for individualized analgesic management in the PICU.

The second category is management of metabolic crisis. rWGS can identify organic acidemias, urea cycle disorders, fatty acid oxidation disorders, glycogen storage diseases, mitochondrial disorders, and other treatable metabolic diseases, thereby supporting special diets, substrate restriction, avoidance of fasting, glucose provision, vitamins or cofactors, carnitine, nitrogen scavengers, and renal replacement therapy in the acute setting (8,10,11,40). In urea cycle disorders and hyperammonemia, rapid etiologic confirmation can support protein restriction, nitrogen scavenger use, and urgent ammonia-lowering strategies. In organic acidemias, methylmalonic acidemia, propionic acidemia, and fatty acid oxidation disorders, early diagnosis can prevent inappropriate nutritional regimens and reduce recurrent hypoglycemia, acidosis, liver injury, encephalopathy, or myocardial involvement (5). For mitochondrial disorders, diagnosis may also affect antiseizure medication selection, anesthetic risk assessment, nutritional support, and avoidance of mitochondrially toxic drugs (13,40).

The third category is adjustment of antimicrobial and immunomodulatory therapy. Although rWGS primarily identifies host genetic disease, molecular diagnosis in children with recurrent infection, severe viral or fungal infection, HLH-like inflammation, persistent diarrhea, cytopenias, or unusual pathogens can determine whether a monogenic inborn error of immunity or immune dysregulation disorder is present. This information can influence intravenous immunoglobulin (IVIG), antimicrobial strategy, immunosuppression, targeted anti-inflammatory therapy, hematopoietic stem cell transplantation evaluation, and infection surveillance (21,24,25). In a RES study, a PIK3CD-related diagnosis prompted immunology consultation and initiation of IVIG, demonstrating how rapid genomic diagnosis can convert refractory infection or chronic inflammation into immune etiology-directed care (60).

The fourth category is avoidance of inappropriate treatment and iatrogenic harm. PICU patients frequently receive sedation, anesthesia, antiseizure medications, antimicrobials, vasoactive drugs, and multiple supportive therapies. Molecular diagnosis can therefore guide clinicians not only toward additional treatment, but also away from harmful interventions. SCN1A-related disease requires avoidance of sodium-channel blockers; mitochondrial disease raises concern for potentially mitochondrially toxic drugs; malignant hyperthermia susceptibility may prompt avoidance of triggering agents, including volatile halogenated anesthetics such as sevoflurane, desflurane, and isoflurane, as well as the depolarizing neuromuscular blocker succinylcholine, with dantrolene readily available for emergency treatment; and G6PD deficiency affects the use of oxidant medications (15,40,59,67). This form of utility is especially important in critical care because inappropriate therapy can precipitate seizure exacerbation, metabolic crisis, arrhythmia, anesthetic complications, or hemolysis.

The fifth category is reduction or redirection of low-yield testing and invasive procedures. Once rapid genomic sequencing establishes an etiology, clinicians may discontinue serial biochemical testing, repeated imaging, additional molecular testing, tissue biopsy, or procedures that are unlikely to alter management. Recent real-world evidence supports this broader effect. In a 2026 inpatient cohort that included 28 PICU patients, 49 of 95 patients had undergone genetic testing before rapid sequencing, and rES or rGS subsequently established a diagnosis in 6 of 30 patients who had already received prior next-generation sequencing. Among patients with diagnostic results who survived to result return, 9 of 22 (40.9%) had a change in surgical planning and 8 of 22 (36.4%) had a treatment change (68). In addition, implementation of first-line rES/rWGS in a 2026 PICU/CICU cohort reduced the median time to diagnosis in the PICU from 33 to 16 days, illustrating how earlier genomic testing can shorten sequential diagnostic pathways (69). These findings suggest that the value of rapid sequencing includes avoiding or redirecting investigations and procedures that become unnecessary once genomic information clarifies the diagnostic pathway.

The sixth category is immediate ICU decision-making. In the multicenter PICU/CICU rWGS cohort, 15 patients had ICU medication changes, 14 of whom were in the diagnosed group; 12 patients had ICU procedural changes, 10 of whom were in the diagnosed group. The median time from diagnosis to ICU medication change was 1 day, and the median time to ICU procedural change was 6 days (2). These changes may include starting or stopping medications, modifying nutrition, postponing or cancelling procedures, establishing long-term monitoring, arranging subspecialty follow-up, discussing resuscitation status, involving palliative care, or planning long-term device support (70-72). Thus, the clinical value of rWGS extends beyond diagnostic yield: it depends on whether the result meaningfully changes PICU decisions (Figure 3).

Figure 3.

Clinical utility of rWGS in critically ill children. This figure highlights how genomic diagnosis supports targeted treatment, avoidance of harmful interventions, reduction of unnecessary procedures, and improved family-centered management. CICU, cardiac intensive care unit; ICU, intensive care unit; NICU, neonatal intensive care unit; PICU, pediatric intensive care unit; rWGS, rapid whole-genome sequencing.

Prognostication and selection of treatment intensity

rWGS also contributes to prognostication and selection of treatment intensity. PICU patients may be receiving mechanical ventilation, vasoactive medications, extracorporeal membrane oxygenation, or other forms of multiorgan support, while clinicians and families must assess whether ongoing escalation is likely to provide meaningful benefit. After a molecular diagnosis is established, clinicians can integrate natural history, treatability, reversibility of organ injury, risk of irreversible neurologic damage, and long-term care needs when deciding whether to continue intensive therapy, initiate transplant evaluation, transition toward palliative care, modify resuscitation status, or prepare for long-term respiratory support and home care (5,62).

Prognostication should not be equated with withdrawal of treatment. For treatable genetic diseases, early molecular diagnosis may identify a reversible window and support continued intensive therapy, disease-specific medication, dietary intervention, cofactor therapy, immunotherapy, transplant evaluation, or long-term organ support. For diseases with extremely poor prognosis, no effective therapy, or irreversible injury, the diagnosis can support earlier discussions of life-sustaining treatment boundaries, resuscitation status, palliative care, and comfort-focused treatment (47,64). Long-term follow-up studies show that critically ill children who receive RGS may continue to have high health-care utilization and may develop new phenotypes years later; prognostication should therefore extend beyond ICU mortality to long-term functional outcomes, subspecialty follow-up, rehabilitation, complication surveillance, and family caregiving burden (53,73).

Genetic counseling and family risk management

The value of rapid genomic sequencing ultimately extends to family risk management. Once a molecular diagnosis is made, clinicians can determine the mode of inheritance, including autosomal recessive, autosomal dominant, X-linked, de novo, parental balanced rearrangement, or other mechanisms, thereby informing parental carrier status, sibling evaluation, and cascade testing of family members (74,75). In the Israeli national rapid trio genome sequencing study, recurrence risk could be defined for all families with an inherited disorder, representing 30 families and 46% of diagnosed children (42). In the PICU/CICU rWGS cohort, family-member screening was recommended in 43 of 79 diagnosed children (54%) (2,76).

Genetic counseling should span both pretest and post-test phases. In time-sensitive PICU settings, intensivists may initiate basic pretest communication regarding the purpose, potential benefits, limitations, and possible result categories of rapid genomic testing when genetics professionals are not immediately available. Comprehensive counseling on secondary findings, familial and reproductive implications, and complex or uncertain results should involve clinical genetics or genetic counseling professionals whenever possible. Implementation of rWGS therefore requires PICU clinicians to receive focused training in test indications, informed consent, result categories, and genomic communication, supported by timely access to genetics expertise. Before testing, families should be counseled about the purpose of testing, possible result categories, VUS, uncertain or negative results, secondary findings, the significance of parental samples, and familial risk. After testing, counseling should address whether the diagnosis explains the acute presentation, disease natural history, treatment potential, prognosis, recurrence risk, and recommended follow-up testing (77,78). For families with a confirmed pathogenic variant, downstream care may include parental testing, sibling screening, carrier testing of relatives, reproductive risk assessment, and future prenatal diagnosis or preimplantation genetic testing. In instances where the results are negative or only partially explanatory, it is imperative that families are made aware that a negative result does not entirely preclude the possibility of genetic disease. It is also crucial to emphasise the continued importance of longitudinal follow-up and reanalysis (76,79).

Nondiagnostic and uncertain results in the PICU

Nondiagnostic rWGS does not exclude genetic disease. A number of potential explanations have been posited for the observed phenomenon, including the presence of regulatory or deep intronic variants, complex structural variants, repeat expansions, methylation abnormalities, tissue-specific mosaicism, transcript-level splicing defects, low-level mitochondrial heteroplasmy, incomplete gene-disease validity, and insufficient evidence in current databases (17,42,43). In the PICU, this uncertainty is clinically consequential because teams may need to decide quickly whether to intensify treatment, initiate low-risk empirical therapy, pursue additional testing, or redirect diagnostic attention.

VUS should not be treated as either diagnostic or irrelevant. Their interpretation should incorporate phenotype fit, inheritance pattern, segregation data, disease mechanism, treatability, urgency, intervention risk, and multidisciplinary consensus (47,48). Low-risk interventions may be considered when the suspected disease is treatable and the phenotype match is strong, whereas high-risk interventions should generally not be based on a VUS alone and should require additional evidence supporting pathogenicity, such as segregation or functional data. Nondiagnostic and VUS cases should have explicit reanalysis triggers, including emergence of new phenotypes, availability of parental or other family samples, database updates, functional assay results, RNA or long-read sequencing, and changes in gene-disease validity (53,79).

Overall, the clinical utility of rWGS in critically ill children with suspected genetic disease in the PICU lies in five domains: increasing etiologic diagnostic yield; shortening turnaround time so that diagnosis enters the acute therapeutic window; translating molecular diagnosis into medications, diet, cofactors, metabolic crisis management, antimicrobial or immunomodulatory adjustment, avoidance of contraindicated drugs, decisions about invasive procedures, and long-term monitoring; supporting prognostication and treatment-intensity decisions; and enabling family screening, reproductive risk assessment, and prenatal diagnosis. The central issue for this population is not whether a variant can be detected, but whether etiology can be established within a constrained therapeutic window and translated into acute treatment, management change, prognostication, and family risk management (2,78) (Table 3).

Table 3. Core clinical utility of rWGS in the PICU.

Utility	Main impact
Etiologic diagnosis	Improves molecular diagnosis in critically ill children with complex or unexplained phenotypes
Faster diagnosis	Brings genetic results into the active ICU decision window
Disease-specific treatment	Guides genotype-informed therapy, cofactors, targeted therapy, or immunotherapy
Metabolic crisis management	Supports protein restriction, glucose provision, carnitine, nitrogen scavengers, and renal replacement therapy
Avoidance of harmful treatment	Avoids contraindicated antiseizure drugs, mitochondrial toxins, anesthetic triggers, or oxidant drugs
Reduction of low-yield procedures	Reduces repeated testing, biopsies, empiric panels, or unnecessary surgery
Prognostic assessment	Helps guide treatment intensity, long-term support, and palliative care discussions
Family risk management	Clarifies inheritance, family screening, recurrence risk, and prenatal diagnosis

ICU, intensive care unit; PICU, pediatric intensive care unit; rWGS, rapid whole-genome sequencing.

Challenges and future directions

Limitations of current evidence

The current evidence base strongly supports clinical utility, but it remains heterogeneous. Studies differ in patient age, ICU type, inclusion criteria, sequencing modality, turnaround-time definitions, analytic scope, VUS handling, reporting policies, and follow-up duration (1,80). Diagnostic yield and management-change rates should therefore not be compared across studies without considering pretest probability and testing context. Moreover, the PICU-specific evidence base remains smaller and more heterogeneous than that derived from neonatal and infant populations, particularly for older children, oncology-related critical illness, real-world cost-effectiveness, and randomized comparisons of implementation strategies (58,60). Future studies should standardize clinical utility endpoints and distinguish immediate ICU management change, post-ICU management change, long-term outcomes, and family-level benefits.

Ethical and communication challenges

Pretest consent for PICU rWGS often occurs in a high-stress resuscitation environment, where families may have limited capacity to process information about VUS, secondary findings, negative results, parental samples, data storage, and future reanalysis (50,77,81). Results may simultaneously create therapeutic opportunities and reveal devastating prognostic information. Accordingly, result disclosure should be integrated with goals-of-care discussions, palliative care when appropriate, psychological support, and genetic counseling rather than treated as a purely laboratory-driven event (47,70,73).

Ethical pathways should also address secondary findings, incidental parental findings, unexpected relatedness or nonpaternity, data retention, recontact responsibilities, and the child’s future autonomy (77,81). The American College of Medical Genetics and Genomics (ACMG) secondary findings framework provides recommendations for reporting selected medically actionable secondary findings in clinical exome and genome sequencing (82); however, applying it to critically ill children requires careful pretest choice architecture, local governance, and post-test counseling resources (48). Clear follow-up processes are essential because genetic results may retain clinical relevance long after the ICU admission has ended (79) (Table 4).

Table 4. Representative medically actionable secondary findings based on ACMG recommendations.

Disease category	Representative genes	Associated conditions	Medical interventions
Hereditary cancer predisposition	BRCA1, BRCA2, TP53	Hereditary breast and ovarian cancer syndrome; Li-Fraumeni syndrome	Cancer surveillance; risk-reducing strategies
Cardiovascular disorders	KCNQ1, KCNH2, MYBPC3	Long QT syndrome; hypertrophic cardiomyopathy	Cardiac monitoring; arrhythmia prevention; medication adjustment
Metabolic crisis	GAA, OTC, PAH	Pompe disease; ornithine transcarbamylase deficiency; biotinidase deficiency	Enzyme replacement therapy; metabolic management; dietary intervention
Other medically actionable conditions	RYR1, G6PD, LDLR	Malignant hyperthermia susceptibility; G6PD deficiency; familial hypercholesterolemia	Avoidance of triggering medications; prevention of hemolysis; lipid-lowering therapy

ACMG, American College of Medical Genetics and Genomics.

Health economic value and equitable access

The economic value of rWGS cannot be assessed only by the sequencing price. A comprehensive evaluation should account for ICU length of stay, repeated testing, imaging, biopsy, empiric treatment, surgical decision-making, long-term care, readmissions, and family-level consequences (26,62,63). Testing earlier in the course is more likely to generate cost offsets by preventing diagnostic odysseys and low-yield procedures before they accrue. Economic evaluations indicate that cost-effectiveness is strongly affected by rapidity, test strategy, and health-system context (54,63,83,84).

Equitable access continues to be a substantial implementation challenge. The accessibility of rWGS is subject to variation according to geographical location, hospital resources, insurance coverage, payer policy, ancestry representation in genomic databases, and the availability of clinical genetics or genetic counselling services (41,46,85). When reference datasets underrepresent genetically diverse populations, interpretive accuracy may decline, and centers without local genomic infrastructure may experience sample transport delays and limited after-hours interpretation. Inequities may be reduced through regional genomic networks, remote consent and counseling pathways, standardized referral criteria, and reimbursement models that recognize total-cost impact (46,86,87).

Technical and implementation directions

The prevailing contemporary rWGS workflows are predominantly reliant upon the utilisation of short-read sequencing, a method that is inherently constrained in its ability to adequately resolve complex structural variants, repeat expansions, methylation abnormalities, and a number of mosaic events. Conversely, long-read sequencing utilises the analysis of substantially longer contiguous DNA molecules, thereby facilitating enhanced resolution of repetitive and low-complexity regions, in addition to complex structural variants. These variants can often prove challenging to reconstruct from short reads. Consequently, short-read WGS remains constrained in its ability to address complex structural variants, repeat expansions, methylation abnormalities, transcriptomic defects, pseudogene-rich regions, and specific forms of mosaicism. Recent advances in long-read sequencing have facilitated the characterisation of genomic regions that are challenging to resolve using conventional short-read approaches. This includes repetitive sequences and complex structural variants (88). The application of long-read sequencing, in conjunction with RNA analysis, methylation profiling, metabolomics, proteomics and functional validation, has the potential to enhance the interpretation of cases that are complex or unresolved (17,42,43).

At the workflow level, AI-assisted HPO extraction, electronic health record-based screening, tiered report templates, rapid multidisciplinary feedback, and standardized PICU pathways could support the transition of rWGS from a project-based service to a sustainable precision-diagnostics infrastructure for pediatric critical care (49,54,76,78). The subsequent emphasis should be placed on the quality of implementation, incorporating such factors as appropriate patient selection, consistent turnaround time, robust interpretation of uncertainty, equitable access, and measurable effects on patient- and family-centred outcomes (Figure 4).

Figure 4.

Future perspectives of rWGS in precision medicine. This figure summarizes emerging strategies, including multi-omics integration, advanced genomic interpretation, and longitudinal management, that may further enhance the clinical impact of rWGS. ICU, intensive care unit; PICU, pediatric intensive care unit; rWGS, rapid whole-genome sequencing.

Conclusions

The utilisation of rWGS facilitates the establishment of a clinically pertinent diagnostic pathway for children in the PICU who are suspected of having genetic diseases. These patients commonly present with a variety of symptoms, including seizures, encephalopathy, metabolic instability, cardiomyopathy, arrhythmia, severe infection, congenital anomalies, or multisystem dysfunction. However, sepsis, hypoxia, shock, postoperative status, medication exposure, or organ support may obscure the underlying genetic cause. As rWGS captures a broad range of variant classes within a single rapid test, it has the capacity to reduce the time taken for diagnosis and enhance the substantiation of aetiology in children exhibiting complex, evolving, or unconventional phenotypes that are not readily explicable through conventional testing methods. In the PICU, the primary value of rWGS is its capacity to link molecular diagnosis with active bedside decision-making. Timely results can inform disease-specific therapy, cofactor supplementation, metabolic crisis management, dietary intervention, antimicrobial or immunomodulatory adjustment, and avoidance of contraindicated drugs. Such results may also decrease repeated low-yield investigations, avoid unnecessary invasive procedures, refine prognosis, guide treatment intensity, and enable family screening and reproductive counseling. For this population, clinical utility should therefore be measured by diagnostic yield, turnaround time, management change, prognostic value, and family-level benefit.

Future implementation should prioritize the development of reliable and equitable PICU genomic pathways. Standardized trigger criteria, rapid consent procedures, trio sequencing when feasible, structured HPO-based phenotyping, dynamic phenotype updates, multidisciplinary interpretation, and clear reporting frameworks are necessary to translate genomic results into clinical action. At the same time, short-read WGS has technical limitations, and unresolved cases may need RNA sequencing, long-read sequencing, methylation analysis, metabolomics, proteomics, or functional validation. Additional studies should define PICU-specific outcome measures, assess long-term clinical impact, examine cost-effectiveness, and improve access across different health care settings.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.