Neonatal invasive candidiasis in low- and middle-income countries: Data from the NeoOBS study

Abstract

Neonatal invasive candidiasis (NIC) has significant morbidity and mortality. Reports have shown a different profile of those neonates affected with NIC and of fluconazole-resistant Candida spp. isolates in low- and middle-income countries (LMICs) compared to high-income countries (HICs). We describe the epidemiology, Candida spp. distribution, treatment, and outcomes of neonates with NIC from LMICs enrolled in a global, prospective, longitudinal, observational cohort study (NeoOBS) of hospitalized infants <60 days postnatal age with sepsis (August 2018-February 2021). A total of 127 neonates from 14 hospitals in 8 countries with Candida spp. isolated from blood culture were included. Median gestational age of affected neonates was 30 weeks (IQR: 28-34), and median birth weight was 1270 gr (interquartile range [IQR]: 990-1692). Only a minority had high-risk criteria, such as being born <28 weeks, 19% (24/127), or birth weight <1000 gr, 27% (34/127). The most common Candida species were C. albicans (n = 45, 35%), C. parapsilosis (n = 38, 30%), and Candida auris (n = 18, 14%). The majority of C. albicans isolates were fluconazole susceptible, whereas 59% of C. parapsilosis isolates were fluconazole-resistant. Amphotericin B was the most common antifungal used [74% (78/105)], followed by fluconazole [22% (23/105)]. Death by day 28 post-enrollment was 22% (28/127). To our knowledge, this is the largest multi-country cohort of NIC in LMICs. Most of the neonates would not have been considered at high risk for NIC in HICs. A substantial proportion of isolates was resistant to first choice fluconazole. Understanding the burden of NIC in LMIC is essential to guide future research and treatment guidelines.

Keywords: Candida auris; Candida parapsilosis; candidiasis; low- and middle-income countries; neonatal candidemia.

Figure 1.

Candida spp. sub-study population derived from the overall NeoOBS study. See Supplementary Figure 1 for detailed schematic of the study population indicating the two enrollment cohorts. Note: Overall NeoOBS enrollment: cohort 1 was 3204 babies from 19 hospitals in 11 countries; cohort 2 was 169 babies from 14 hospitals in 10 countries. Candida sub-analysis cohort: cohort 1 includes 67 babies from 12 hospitals in 7 countries; cohort 2 was 60 babies from 12 hospitals in 7 countries.

Figure 2.

Reported susceptibility profiles to amphotericin B, fluconazole, and micafungin for the most common Candida species.

Figure 3.

Patient-based antifungal treatment choice by country indicating the day the blood culture was taken (day 0), the day the fungal organism was identified (open squares), and mortality (solid dots). White space indicates calendar days that antifungal treatment was not given. *Other countries are comprised of five countries, each contributing < 8 participants (range: 1–7 per country).

Figure 4.

Patient-based antifungal treatment choice by causative Candida spp. Day 0 is the day the blood culture was taken that grew Candida spp. Day that the Candida spp. was identified is indicated with open squares and mortality with solid dots.

Figure 5.

Kaplan–Meier curve for mortality from day of culture for each Candida species.