Background/Purpose: Pediatric rheumatology has been an evolving subspecialty over the last few decades with a continued increase in the both the number and severity of rheumatic diseases requiring specialty care. These patients often end up in pediatric intensive care units (PICU), but little is known about the clinical characteristics and outcomes of those critically ill patients with rheumatology consultations in the PICU. To determine the characteristics and clinical outcomes of critically ill children requiring rheumatology consultations in a large quaternary children’s hospital.

Methods: We performed a retrospective chart review for all consecutive pediatric rheumatology consults admitted to the PICU between March 2012 -2016. We collected demographics and clinical data from admission, time of rheumatology consult, and discharge. We also collected validated morbidity prediction scales including the PCPC (Pediatric Cerebral Performance Category), and POPC (Pediatric Overall Performance Category). Patient characteristics were summarized using median or frequency. Summary statistics were stratified by diagnosis and compared using generalized estimating equations (GEE) to account for multiple admissions. GEE was used to estimate odds ratios for acquiring a cognitive disability while in the ICU.

Results: From total of 269 consecutive rheumatology consultations (250 unique patients), 56.1% (n=151) had a rheumatic diagnosis (RDx), 20% known and 80% new-onset. The RDx group was younger than non-RDx (median 9.3 vs. 11.3 years, p=0.027). Rheumatology was consulted sooner for RDx compared to non-RDx (median 1 vs. 3 days, p=0.003). For consults with a RDx, the most common diagnoses were various vasculitides (45.7%, n=69/151) and SLE (21.9%, n=33/151). A higher percentage of RDx were Hispanic (48.3% vs. 29.7%, p=0.002), had a previous history of autoimmune disease (50.3% vs. 12.7%, p< 0.001), on immunosuppressive medications (40.4% vs. 8.5%, p< 0.001), had hematologic involvement (79.5% vs. 63.6%, p=0.007), pulmonary hemorrhage (13.2% vs. 3.4%, p=0.028), and liver/GI involvement (49% vs. 33.9%, p=0.031). A smaller percentage of RDx had CNS involvement (33.1% vs. 51.7%, p=0.028), focal deficit (12.6% vs. 26.3%, p=0.007), epinephrine use (1.3% vs. 7.6%, p=0.029), and mechanical ventilation in the first 24 hours (25% vs. 42.7%, p=0.010).  More of the non-RDx required intubation (56.7% vs. 27.8%, p=0.001). There were no statistically significant differences in sex (p=0.235), race (p=0.435), or presence of shock at admission (p=0.090).

Conclusion: This review of a large cohort of critically ill children requiring rheumatology consultation characterizes distinctive clinical features and courses for those with suspicion for rheumatic disease. The majority of the consults were eventually diagnosed with a rheumatic disease and 80% of these were new-onset. The potential for significant morbidity with a large number of new rheumatic diagnoses presenting to the critical care setting underscores the need for timely collaboration between Pediatric Rheumatology and Critical Care Medicine.

« Back to 2020 Pediatric Rheumatology Symposium

ACR Meeting Abstracts - https://acrabstracts.org/abstract/clinical-characteristics-and-outcomes-of-critically-ill-children-requiring-rheumatology-consultation-in-a-large-intensive-care-unit/