P2B144: The Clinical, Obstetric, Parent, and Sociocultural Factors Associated with Antimicrobial Resistance for Newborns in Amhara, Ethiopia

Background: Globally, 200,000 cases of neonatal mortality due to sepsis per year are a result of antimicrobial resistance. Sepsis is a systemic blood stream infection more frequent in LBW infants; it drives 30% of newborn deaths globally. Antimicrobial resistance (AMR) occurs when a microorganism is able to survive despite antibiotic treatment AMR to first, second and third-line sepsis treatments is becoming more common in sub-Saharan Africa. Gaps in effective and affordable antibiotic therapy drives morbidity and mortality which is potentially preventable. The goal of this analysis is to identify the clinical, obstetric, parent, and sociocultural factors that predict
antimicrobial resistance among newborns with sepsis in Amhara in order to more effectively stratify risk and treatment.

Methods: This prospective cohort study was conducted at 2 hospitals in the Amhara region of Ethiopia: a
large referral hospital and smaller general hospital. We recruited normal and low birth weight (LBW < 2000 gram) newborns from the two hospitals in Amhara, Ethiopia from 2018-2019 in a 1:4 ratio. Blood cultures were collected when infants developed signs concerning for sepsis. Sepsis predictors were assessed in categories including clinical, obstetric, social, parent and place-time exposure using univariate logistic regression. Then, univariate and multivariable logistic regression (MLR) to predict antimicrobial resistant sepsis. The AUC and Hoshmer-Lemsow test were used to assess the final MLR.

Results: There were 187 neonates with available data to assess AMR following a sepsis diagnosis. When assessing LBW categorically, babies who weigh < 2000 grams have the highest odds of AMR at an odds ratio of 2.01. When considering obstetric history, neonates born to mothers who had a previous pre-term birth of two or more have a 3.14 higher odds of AMR than the odds of resistance among neonates born to mothers with fewer pre-term births. Similarly, neonates born to mothers who previously had 2 more children who had low birth weight had a 3.48 higher odds of AMR compared to neonates whose mothers did not previously have 2 or more children who had low birth weight. Under the parent factors, parental age gap was statistically significant in predicting increased odds of AMR among neonates.

Conclusion: 59.4% (n=111) of septic infants had AMR. This alarmingly high AMR incidence may increasingly contribute to newborn mortality in Amhara and globally. In global government hospitals, clinicians need mechanisms to identify the newborns who are at the highest risk of developing antimicrobial resistance as a strong driver of morbidity and mortality. For Amhara newborns, the age gap between parents, maternal literacy, and number of antenatal care visits were major drivers of AMR. Knowing these factors may help clinicians to more readily identify sepsis
and create system-level interventions to prevent AMR sepsis.