ABSTRACT
Objective Thrombotic microangiopathy induced Thrombocytopenia Associated Multiple Organ Failure and hyperinflammatory Macrophage Activation Syndrome are important causes of late pediatric sepsis mortality that are often missed or have delayed diagnosis. Our objective is to derive computable 24-hour sepsis phenotypes to facilitate enrollment in early precise anti-inflammatory trials targeting mortality from these conditions.
Design Machine learning analysis using consensus k-means clustering.
Setting Nine pediatric intensive care units.
Patients 404 children with severe sepsis.
Interventions 24-hour computable phenotypes derived using 25 bedside variables including C-reactive protein and ferritin.
Measurements and Main Results Four computable phenotypes (PedSep-A, B, C, and D) are derived. Compared to the overall population mean, PedSep-A has the least inflammation (median C-reactive protein 7.3 mg/dL, ferritin 125 ng/mL), younger age, less chronic illness, and more respiratory failure (n = 135; 2% mortality); PedSep-B (median C-reactive protein 13.2 mg/dL, ferritin 225 ng/ mL) has organ failure with intubated respiratory failure, shock, and Glasgow Coma Scale score < 7 (n = 102, 12% mortality); PedSep-C (median C-reactive protein 15.2 mg/dL, ferritin 405 ng/mL) has elevated ferritin, lymphopenia, more shock, more hepatic failure and less respiratory failure (n = 110; mortality 10%); and, PedSep D (median C-reactive protein 13.1 mg/dL ferritin 610 ng/mL), has hyperferritinemic, thrombocytopenic multiple organ failure with more cardiovascular, respiratory, hepatic, renal, hematologic, and neurologic system failures (n = 56, 34% mortality). PedSep-D has highest likelihood of Thrombocytopenia Associated Multiple Organ Failure (Adj OR 47.51 95% CI [18.83-136.83], p < 0.0001) and Macrophage Activation Syndrome (Adj OR 38.63 95% CI [13.26-137.75], p <0.0001), and an observed survivor interaction with combined methylprednisolone and intravenous immunoglobulin therapies (p < 0.05).
CONCLUSIONS AND RELEVANCE Machine learning identifies four computable phenotypes (www.pedsepsis.pitt.edu). Membership in PedSep-D appears optimal for enrollment in early anti-inflammatory trials targeting Thrombocytopenia Associated Multiple Organ Failure and Macrophage Activation Syndrome.
Question Can machine learning methods derive 24-hour computable pediatric sepsis phenotypes that facilitate early identification of patients for enrollment in precise anti-inflammatory therapy trials?
Findings Four distinct phenotypes (PedSep-A, B, C, and D) were derived by assessing 25 bedside clinical variables in 404 children with sepsis. PedSep-D patients had a thrombotic microangiopathy and hyperinflammatory macrophage activation biomarker response, and improved survival odds associated with combined methylprednisolone plus intravenous immunoglobulin therapy.
Meaning Four novel computable 24-hour phenotypes are identifiable (www.pedsepsis.pitt.edu) that could potentially facilitate enrollment in early precise anti-inflammatory trials targeting thrombotic microangiopathy and macrophage activation in pediatric sepsis.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
Supported, in part, by grant R01GM108618 (to Dr Carcillo PI, Dr. Park, Dr Canna, and Dr Kellum) from the National Institutes of General Medical Sciences, by 5U01HD049934- 10S1 (to Dr Carcillo) and K12HD047349 (to Dr Kernan) from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services and the following cooperative agreements: U10HD049983, U10HD050096, U10HD049981, U10HD063108, U10HD63106, U10HD063114, U10HD050012, and U01HD049934.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
Approval was obtained from The University of Utah Institutional Review Board, Central IRB # 70976. Written informed consent was obtained from one or more parents/guardians for each child. Assent was garnered when the child was able. Patients were enrolled from 2015-2017.
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Footnotes
Supported, in part, by grant R01GM108618 (to Dr Carcillo PI, Dr. Park, Dr Canna, and Dr Kellum) from the National Institutes of General Medical Sciences, by 5U01HD049934- 10S1 (to Dr Carcillo) and K12HD047349 (to Dr Kernan) from the Eunice Kennedy Shriver National Institutes of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services and the following cooperative agreements: U10HD049983, U10HD050096, U10HD049981, U10HD063108, U10HD63106, U10HD063114, U10HD050012, and U01HD049934.
Conflicts of Interest and Source of Funding Drs. Carcillo’s, Berg’s, Wessel’s, Pollack’s, Meert’s, Hall’s, Doctor’s, Cornell’s, Harrison’s, Zuppa’s, Reeder’s, Banks’s, and Holubkov’s institutions received funding from the National Institutes of Health (NIH). Drs. Carcillo’s, Newth’s, Shanley’s, and Dean’s institutions received funding from the National Institutes of Child Health and Human Development. Drs. Carcillo, Berg, Wessel, Polack, Meert, Hall, Newth, Doctor, Shanley, Cornell, Harrison, Zuppa, Reeder, Banks, Holubkov, Notterman, and Dean received support for article research from the NIH. Dr. Carcillo’s institution also received funding from the National Institutes of General Medical Sciences. Dr. Pollack disclosed that his research is supported by philanthropy from Mallinckrodt Pharmaceuticals. Dr. Hall received funding from Bristol Myers-Squibb (for service on an advisory board) and LaJolla Pharmaceuticals (service as a consultant), both unrelated to the current submission. Dr. Newth received funding from Philips Research North America. Dr. Doctor’s institution received funding from the Department of Defense and Kalocyte. Dr. Shanley received funding from Springer publishing, International Pediatric Research Foundation, and Pediatric Academic Societies. Dr. Cornell disclosed he is co-founder of Pre-Dixon Bio. Dr. Holubkov received funding from Pfizer (Data Safety Monitoring Board [DSMB] member), Medimmune (DSMB member), Physicians Committee for Responsible Medicine (biostatistical consulting), DURECT Corporation (biostatistical consulting), Armaron Bio (DSMB past member), and St Jude Medical (DSMB past member). The remaining authors have disclosed that they do not have any potential conflicts of interest.
Data Availability
All data produced in the present work are contained in the manuscript
