Functional Morbidity Following PICU Discharge

Because mortality rates are low in critically ill children, evaluations of long-term morbidities are increasingly appreciated as important outcome measures. Pediatric survivors of critical illness frequently have long-term consequences such as cognitive disturbances, impaired physical function, mental illness, and reduced health-related quality of life, all culminating in post-intensive care syndrome in pediatrics (PICS-p). The incidence of PICS-p in the literature varies widely, from 5% to 50%, largely because of heterogeneity in inclusion criteria, the assessment tools used, and definitions of dysfunction.¹ A recent study by Rosenbaum et al sought first to determine the prevalence of an acquired dysfunction (i.e., PICS-p) in children up to a year following critical illness in a PICU (compared with a pre-illness assessment) and to track the trajectory of this dysfunction.² A secondary aim was to determine whether any clinical or epidemiologic factors were associated with acquired dysfunction.

The authors of this single-center study prospectively enrolled children aged 1 month to 17 years who were admitted to the PICU between May 2019 and October 2022, including patients with PICU stays of more than 24 hours with expected survival. Patients were excluded if they were readmitted to the PICU after being enrolled in the study during their previous PICU stay. Caregiver recall was used to complete the Ages and Stages Questionnaire, 3rd edition (ASQ-3), if the patient was younger than 3 years; the Pediatric Evaluation of Disability Inventory Computer Adaptive Test (PEDI-CAT) if the subject was 5 years or older; and both of these assessments for patients between the ages of three and five years. 

The ASQ-3 evaluates five development domains: communication, gross motor, fine motor, problem solving, and social skills. The PEDI-CAT measures performance in three development domains: daily activities, mobility, and social/cognitive skills. The PEDI-CAT consists of dynamic questions and must be completed in person. Following PICU discharge, caregivers were contacted by phone after one month, six months, and 12 months to reevaluate the patients using the same tools. 

Only patients who completed the one-month assessment were included in the primary analysis, although patients who completed only the six- and/or 12-month assessments were included for the secondary aim. Patient demographics and clinical information were obtained for analysis and the Child Opportunity Index was used to evaluate local resources, which have been shown to be associated with PICS-p.³ 

For the purposes of this evaluation, a decline in an assessment was defined as a decrease in the score by at least one standard deviation in the one-month assessment compared with the initial assessment. Acquired dysfunction was defined as a decline in scores in three out of the five domains in the ASQ-3 or two out of the three domains in the PEDI-CAT. Persistence of dysfunction was defined as acquired dysfunction that remained for six to 12 months after discharge. Resolution of dysfunction was defined as having scores less than one standard deviation below baseline. Incomplete resolution was defined as having scores that met neither persistence nor resolution categories.

The authors enrolled 502 patients, with 362 having at least one post-PICU assessment and 286 completing a one-month post-PICU assessment. Of the 286 patients, 36 (approximately 12%) met the criteria for acquired dysfunction. In a univariate analysis, female sex, use of intermittent mandatory ventilation (IMV), and median PICU length of stay (LOS) were associated with acquired dysfunction. Other variables such as age, race, Child Opportunity Index score, and primary diagnosis were not associated with acquired dysfunction. 

Using a multivariate regression model, the authors demonstrated that boys were less likely to have acquired dysfunction (OR, 0.41; 95% CI, 0.19-0.85; P = 0.016) and that use of IMV remained associated with dysfunction (OR, 2.58; 95% CI, 1.14-5.85; P = 0.024). PICU LOS was not associated with dysfunction in this model. Boys in this study had lower baseline developmental scores than girls, so the authors added an interaction term between male sex and preexisting neurologic comorbidity and subsequently found that only the use of IMV was associated with decline in this new model (OR, 2.72; 95% CI, 1.18-6.30; P = 0.019). The domains most affected were the problem-solving domain in the ASQ-3 and the mobility domain in the PEDI-CAT.

Of the 36 children with acquired dysfunction diagnosed at one month, 26 completed subsequent assessments at six and/or 12 months. Of these, 21 of the final assessments occurred at 12 months. The authors then determined the number of patients who had complete resolution (n=4, 15%), persistent dysfunction (n=9, 35%), or incomplete resolution (n=13, 50%). The sample size was too small to determine which characteristics were associated with final outcomes, but the authors claim that, numerically, PICU LOS was highest with persistent dysfunction and lowest with complete resolution. The domains most often affected in the persistent dysfunction group were the gross motor domain in the ASQ-3 and the mobility domain in the PEDI-CAT. 

This study reports on the largest cohort of outcomes following PICU admission from a single center. The authors report an approximate 12% incidence in the development of an acquired dysfunction in children who survived a PICU admission. Of these, only 15% had complete resolution at their final assessment. The use of IMV was consistently found to be a risk factor under different analyses, and gross motor morbidities were the most commonly seen abnormalities.

Limitations of this study are consistent with others examining PICS-p, such as the inclusion criteria, the assessment tools used, and the self-reported conservative definition of functional decline. The authors also cite the modest numbers of patients who had a decline and the fact that the assessments were completed by caregivers rather than by certified examiners as other potential limitations of the study. 


References

1. Biagas KV, Heneghan JA, Abu-Sultaneh S, et al. Scoping review: neurocognitive outcome assessments after critical illness in children. J Intensive Care Med. 2023 Apr;38(4):358-367.
2. Rosenbaum DJ, Minich NM, Rasal R, Senthil A, Marlow CE, Shein SL. Functional morbidity within 12 months following PICU discharge using validated screening tools. Hosp Pediatr. 2025 Dec 1;15(12):986-994.
3. Klapp JM, Smith MB, Turner AD, Williams CN. Area-level socioeconomic disadvantage and post-PICU outcomes in children with an acquired brain injury: single-center retrospective cohort, 2016-2023. Pediatr Crit Care Med. 2025 May 1;26(5):e633-e646.