Acute encephalopathy syndrome (AES) is characterized by sudden onset of seizures and altered sensorium of infectious or non-infectious origin. Seasonal outbreaks of fatal hypoglycaemic AES in children, associated with eating fruit from the Sapindaceae family (e.g., ackee, litchi), have been reported globally [1,2]. Since 1995, AES outbreaks have been reported during the litchi-harvesting season from May–July in Muzaffarpur, the largest commercial litchi-producing district of Bihar, India [3]. An AES outbreak investigation in Muzaffarpur in 2014 linked known toxins hypoglycin A and α-methylene cyclopropyl glycine (MCPG) in litchi fruit to hypoglycaemic AES in children [3]. Following the 2014 outbreak, the Government of Bihar implemented community-based interventions to prevent hypoglycemia in children. They also strengthened the clinical management of hypoglycaemic seizures in public health facilities [4]. The number of AES cases and deaths declined from 2015–18, suggesting that the interventions were effective. However, in May–June 2019, AES cases increased. We conducted a descriptive epidemiological analysis of the AES cases. | | Methods | | We identified AES cases from established hospital-based surveillance in the two tertiary referral hospitals in Muzaffarpur. We defined a suspected AES case as seizures or altered sensorium in a child aged ≤15 years admitted from 1 May to 2nd July 2019. We excluded patients aged six months to 6 years who were admitted for fever and a single generalized convulsion of <15 minutes in duration and recovered consciousness within 60 minutes of seizure. We conducted a review of medical records and abstracted data using a structured tool for socio-demographics, clinical history, duration of hospitalization, treatment, and laboratory profile. We also assembled a prospective cohort of probable cases admitted to the hospital during the investigation. We defined a probable AES case as new-onset seizures or altered sensorium of <7 days duration in a child aged ≤15 years admitted t from 1 May to 2 July 2019. For the cohort of probable cases, we interviewed the caregivers using a structured clinical-epidemiological questionnaire for socio-demographics, anthropometry, illness characteristics, treatment-seeking behavior, meal assessment, exposure to litchi fruit, and exposure to health messages. For anthropometry, we calculated Z-scores using the World Health Organization 2006 standardized growth tables [5]. | | Results | | Of the 655 suspected and probable AES cases identified, the case fatality rate (CFR) was 21% (139 deaths). The median age was four years (interquartile range: 3 months–14 years), and 58% (378) were females. The first case was reported on 5 May 2019, cases peaked on 15 June, and the last case on 2 July (Figure 1). Among cases with available data, 75% (389/518) had blood glucose levels of <70 mg/dL upon hospital admission, and 75% (476/638) were residents of Muzaffarpur district. We identified cases from 15 (94%) of 16 blocks in the Muzaffarpur district and calculated a district incidence of 22 per 100,000 children ≤15 years old. | | The prospective cohort comprised 94 probable AES cases; CFR was 26%. Among probable cases, 63% (49/78) of caregivers were wage workers, and 34% (31/91) were of low socioeconomic status. Symptoms were reported in the early morning (3 am to 8 am) for 67% (62/93) of cases, and 97% (90/93) presented with seizures. Among probable cases with anthropometry data, 62% (43/69) were underweight (i.e., weight-for-age Z score <-2), 44% (25/57) stunted (i.e., height-for-age Z score <-2), and 43% (10/23) wasted (i.e., weight-for-height Z score <-2). Primary health facilities referred 46% (43/93) of probable cases to the two tertiary hospitals for admission. Among cases referred, only 30% (13/43) received hypoglycemia and seizure management at the primary health facility. | | Eating litchis in the 24 hours and seven days before illness onset was reported by 57% (54/94) and 87% (59/68) of caregivers, respectively. Skipping any meal and skipping the evening meal in the 24 hours before illness onset was reported by 55% (48/88) and 44% (28/63) of caregivers, respectively. Among probable cases, 45% (27/60) of caregivers reported Government Supplementary Nutrition (GSN) programme enrollment. Sixty percent (50/83) of caregivers said a visit by health workers in the week before illness. Still, only 8% (7/83) reported receiving messages on AES prevention and early treatment by health workers in the past month. | | Conclusions | | The 2019 AES outbreak in Muzaffarpur district, Bihar, occurred among young children with hypoglycemia upon hospital admission and had high associated mortality. Although the Government of Bihar implemented community and clinical measures to prevent AES cases after the 2014 outbreak, a large proportion of the AES cases did not benefit from the prevention measures based on our investigation [4]. New state and district health leadership, turnover of community and facility-level healthcare workers, lack of ongoing training and focused community outreach, and competing health priorities might have been factors responsible for the resurgence. To prevent future AES cases, we recommended prompt emergency management of hypoglycemia and seizures at primary health facilities before referral. We recommend enrollment of all eligible children to GSN and enhanced community health communications to reinforce the importance of an evening meal for children and limiting the eating of litchi fruit during the harvesting season from May to July