Etiology of Bacterial Sepsis and Isolate Resistance Patterns in Hospitalized Neonates in Zambia.

BACKGROUND: The Sepsis Prevention in Neonates in Zambia study is a prospective cohort study that evaluated an infection prevention and control (IPC) bundle in the University Teaching Hospital neonatal intensive care unit (NICU) in Lusaka, Zambia. We present here the etiologies, antimicrobial resistance profiles, and associated mortality of bloodstream infections (BSI) in this cohort. METHODS: Venous blood was collected from neonates with clinically suspected sepsis and cultured with an automated blood culture system. Organism identification and susceptibility testing were done using the Vitek II system. We used the CDC National Health Safety Network criteria to define pathogens and commensals. RESULTS: There were 1120 blood cultures performed for 1060 neonates with suspected sepsis. Overall, 38% (424/1120) of cultures were positive of which 72% (306/424) grew pathogens. Blood cultures obtained after, as compared to before, 2 days of hospitalization were more likely to yield a pathogen (77% vs. 65%; P < 0.001). Klebsiella pneumoniae was the most prevalent organism, accounting for 74% (225/306) of all pathogens . K. pneumoniae isolates were highly resistant: 98% (221/225) were extended-spectrum beta-lactamase (ESBL)-positive, while 81% were resistant to gentamicin (182/225) and fluoroquinolones (177/219). Only one isolate was carbapenem resistant. Observed mortality rate was 32% (122/380); 61% (75/122) of the deaths was related to Klebsiella BSI. CONCLUSIONS: Multidrug-resistant ESBL-producing Klebsiella species were the main organisms responsible for BSI and were associated with increased mortality. BSI risk increased with prolonged hospitalization, underscoring the importance of IPC measures in the NICU.

Bridging the gap in neonatal resuscitation in Zambia.

Neonatal resuscitation has been poorly instituted in many parts of Africa and most neonatal resuscitation algorithms are adapted from environments with abundant resources. Helping Babies Breathe (HBB) is an algorithm designed for resource-limited situations and most other algorithms are designed for resource-rich countries. However, there are neonatal referral centers in resource-limited countries who may provide more advanced resuscitation. Thus, we developed a neonatal resuscitation algorithm for a resource-limited country (Zambia) which considers more advanced interventions in situations where they can be provided. The algorithm described in this paper is based on the Newborn Life Support algorithm from the UK as well as the HBB algorithm and accounts for all situations in a resource-limited country. Most importantly, it focuses on non-invasive ventilation but includes advice on more advanced resuscitation including intravenous access, fluid management, chest compressions and adrenaline for resuscitation. Although intubation skills are included in neonatal training workshops, it is not the main focus of the algorithm as respiratory support equipment is scarce or lacking in most health facilities in Zambia. A home-grown neonatal resuscitation algorithm for a resource-limited country such as Zambia is likely to bridge the gap between limited situations requiring only bag and mask ventilation and better equipped institutions where more advanced resuscitation is possible. This algorithm will be rolled out in all training institutions and delivery facilities across Zambia over the next months.

Healthcare worker perceptions of the implementation context surrounding an infection prevention intervention in a Zambian neonatal intensive care unit.

BACKGROUND: Infants in the neonatal intensive care unit (NICU) are particularly susceptible to healthcare-associated infections (HAIs). NICUs in low- and middle income countries face additional challenges to HAI prevention. There is a need to better understand the role of the implementation context surrounding infection prevention interventions in low- and middle income countries. AIM: The aim of this study was to identify NICU healthcare worker perceptions of an intervention to reduce bloodstream infections in a large Zambian NICU. METHODS: Semi-structured interviews were conducted with NICU staff during a prospective cohort study examining the impact of an infection prevention bundle on bloodstream infections. Interviews were analyzed using an integrated approach, combining inductive theme generation with an application of the Consolidated Framework for Implementation Research (CFIR). RESULTS: Interviews were conducted with 17 NICU staff (5 physicians and 12 nurses). Respondents believed the bundle elements were easy to use, well-designed and facilitated improved performance. Four organizational characteristics that facilitated HAI transmission were identified - (1) lack of NICU admission protocols; (2) physical crowding; (3) understaffing; and (4) equipment shortages. Respondents suggested that NICU resource constraints reflected a societal ethos that devalued the medical care of infants. Despite the challenges, respondents were highly motivated to prevent HAIs and believed this was an achievable goal. They enthusiastically welcomed the bundle but expressed serious concern about sustainability following the study. CONCLUSIONS: By eliciting healthcare worker perceptions about the context surrounding an infection prevention intervention, our study identified key organizational and societal factors to inform implementation strategies to achieve sustained improvement.

Preventing Bloodstream Infections and Death in Zambian Neonates: Impact of a Low-cost Infection Control Bundle.

BACKGROUND: Sepsis is a leading cause of neonatal mortality in low-resource settings. As facility-based births become more common, the proportion of neonatal deaths due to hospital-onset sepsis has increased. METHODS: We conducted a prospective cohort study in a neonatal intensive care unit in Zambia where we implemented a multifaceted infection prevention and control (IPC) bundle consisting of IPC training, text message reminders, alcohol hand rub, enhanced environmental cleaning, and weekly bathing of babies ≥1.5 kg with 2% chlorhexidine gluconate. Hospital-associated sepsis, bloodstream infection (BSI), and mortality (>3 days after admission) outcome data were collected for 6 months prior to and 11 months after bundle implementation. RESULTS: Most enrolled neonates had a birth weight ≥1.5 kg (2131/2669 [79.8%]). Hospital-associated mortality was lower during the intervention than baseline period (18.0% vs 23.6%, respectively). Total mortality was lower in the intervention than prior periods. Half of enrolled neonates (50.4%) had suspected sepsis; 40.8% of cultures were positive. Most positive blood cultures yielded a pathogen (409/549 [74.5%]), predominantly Klebsiella pneumoniae (289/409 [70.1%]). The monthly rate and incidence density rate of suspected sepsis were lower in the intervention period for all birth weight categories, except babies weighing <1.0 kg. The rate of BSI with pathogen was also lower in the intervention than baseline period. CONCLUSIONS: A simple IPC bundle can reduce sepsis and death in neonates hospitalized in high-risk, low-resource settings. Further research is needed to validate these findings in similar settings and to identify optimal implementation strategies for improvement and sustainability. CLINICAL TRIALS REGISTRATION: NCT02386592.