The NeoSep Severity and Recovery scores to predict mortality in hospitalized neonates and young infants with sepsis derived from the global NeoOBS observational cohort study

Abstract

Background Sepsis severity scores are used in clinical practice and trials to define risk groups. There are limited data to derive hospital-based sepsis severity scores for neonates and young infants in high-burden low- and middle-income country (LMIC) settings where trials are urgently required. We aimed to create linked sepsis severity and recovery scores applicable to hospitalized neonates and young infants in LMIC which could be used to inform antibiotic trials. Methods & Findings A prospective observational cohort study was conducted across 19 hospitals in 11 countries in sub-Saharan Africa, Asia, Latin America and Europe. Infants aged <60 days with clinical sepsis fulfilling at least two clinical or laboratory criteria (≥1 clinical) were enrolled. Primary outcome was 28-day mortality. Two prediction models were developed for 1) 28-day mortality from factors at sepsis presentation (baseline NeoSep Severity Score), and 2) daily risk of death on IV antibiotics from daily updated assessments (NeoSep Recovery Score). Multivariable Cox regression models included a randomly selected 85% of infants, with 15% for validation. 3204 infants were enrolled between 2018-2020. Median age was 5 days (IQR 2-15), 90.4% (n=2,895) were <28 days. Median birth weight was 2500g (1400-3000g), and a median of 4 clinical (IQR 2-5) and 1 laboratory (0-2) signs were present. Overall mortality was 11.3% (95%CI 10.2-12.5%; n=350). A baseline NeoSep Severity Score from infants characteristics, respiratory support, and clinical signs (no laboratory tests) at presentation had a C-index 0.77 (95%CI: 0.75-0.80) and 0.76 (0.69-0.82) in derivation and validation samples, respectively. Mortality in the validation sample was 1.6% (3/189; 95%CI: 0.5-4.6%), 11.0% (27/245; 7.7-15.6%), and 27.3% (12/44; 16.3-41.8%) in low (score 0-4), medium (5-8) and high (9-16) risk groups, respectively, with similar performance across subgroups. A related NeoSep Recovery Score based on evolving post-baseline clinical signs and supportive care discriminated well between infants who died or survived the following day or subsequent few days. The area under the ROC curve for score on day 2 and death in the following 5 days was 0.82 (95%CI 0.78-0.85) and 0.85 (95%CI 0.78-0.93) in the derivation and validation data, respectively. Conclusion The baseline NeoSep Severity Score predicted 28-day mortality and could identify infants with high risk of mortality for inclusion in hospital-based sepsis trials. The NeoSep Recovery Score predicts day-by-day inpatient mortality and could, with further validation, help to identify poor response to antibiotics.

Patterns of antibiotic use, pathogens and clinical outcomes in hospitalised neonates and young infants with sepsis in the NeoOBS global neonatal sepsis observational cohort study

“…206 different empiric antibiotic combinations were used, which were structured into 5 groups that were developed from the World Health Organisation (WHO) AWaRe classification. 25.9% (n=814) of infants started a WHO first line regimen (Group 1 -Access, penicillin-based regimen) and 13.8% (n=432) started WHO second-line cephalosporins (cefotaxime/ceftriaxone) (Group 2-'Low' Watch).…”

“…Time was measured from the initial blood culture sample, censoring at the earliest of day 28, withdrawal or last contact if lost post-discharge. To analyse whether time from baseline culture to the start of new antibiotic regimen or type of pathogen were associated with mortality, respectively, we adjusted for previously established baseline predictors of mortality (Submitted manuscript 25 ). Multivariable logistic regression with backwards elimination (exit p=0.05) adjusted for site was used to analyse factors associated with starting group 3-5 versus group 1-2 regimens in sites with at least 10% of infants in either.…”

“…Multivariable logistic regression with backwards elimination (exit p=0.05) adjusted for site was used to analyse factors associated with starting group 3-5 versus group 1-2 regimens in sites with at least 10% of infants in either. Candidate factors were known at sepsis presentation and included birth weight, gestational age, postnatal age, time in hospital, central line or indwelling catheter, Intravenous (IV) antimicrobials in previous 24 hours, signs of meningitis, previous positive culture, previous surgery, and a score reflecting sepsis severity (Submitted manuscript 25 ). Cumulative incidence antibiotic escalation or stop of all iv antibiotics was estimated with death as competing risk.…”

Challenges in the Implementation of the NeoOBS Study, a Global Pragmatic Observational Cohort Study, to Investigate the Aetiology and Management of Neonatal Sepsis in the Hospital Setting

“…In 2018, the global Neonatal Sepsis Observational Study (NeoOBS) was set up by GARDP, as the sponsor, in collaboration with St George's, the University of London, the Medical Research Council Clinical Trials Unit at UCL, Laboratory of Medical Microbiology at the University of Antwerp (UA), the Penta Foundation, and a global collaborative network of hospitals, to assess the etiology and management of neonates with significant sepsis (clinicaltrials.gov reference number: NCT03721302). Significant sepsis was defined as presenting with two or more of the signs/symptoms listed in the inclusion and exclusion criteria (Appendix A, Table A1) [16,17]. This study was a multi-centre, prospective, observational cohort study of inpatient management of neonatal sepsis in 19 sites across 11 countries (Bangladesh, Brazil, China, Greece, India, Italy, Kenya, South Africa, Thailand, Uganda, and Vietnam).…”

“…Data on the clinical presentation of neonates with sepsis to inform trial inclusion criteria (the NeoSep score) and clinical progression (NeoSep Recovery score) [17] • Understand the differences between sites and the feasibility of collecting high-quality data in different settings…”