ABSTRACT
Background: Multiplex lateral flow rapid diagnostic tests (LF-RDTs) may aid management of patients with acute non-malarial febrile illness (NMFI) in rural south and southeast Asia. We aimed to evaluate the cost-effectiveness in Cambodia and Bangladesh of a putative, as-yet-undeveloped LF-RDT capable of diagnosing enteric fever and dengue, as well as measuring C-reactive protein (CRP) to guide antibiotic prescription, in primary care patients with acute NMFI. Methods: A country-specific decision tree model-based cost-effectiveness analysis was conducted from a health system plus limited societal perspective considering the cost of antimicrobial resistance. Parameters were based on data from a large observational study on the regional epidemiology of acute febrile illness, published studies, and procurement price lists. Costs were expressed in US$ (value in 2022), and cost-effectiveness evaluated by comparing incremental cost-effectiveness ratios with conservative opportunity cost-based willingness-to-pay thresholds and the more widely used threshold of per capita gross domestic product (GDP). Findings: Compared to standard of care, LF-RDT-augmented clinical assessment was dominant in Cambodia, being more effective and cost-saving. The cost per disability-adjusted life year (DALY) averted in Bangladesh was US$482, slightly above the conservative opportunity cost-based willingness-to-pay threshold of US$388 and considerably lower than the GDP-based threshold of US$2687. The intervention remained dominant in Cambodia and well below the GDP-based threshold in Bangladesh when antimicrobial resistance costs were disregarded. Interpretation: These findings provide guidance for academic, industry, and policymaker stakeholders involved in acute NMFI diagnostics. While definitive conclusions cannot be made in the absence of established thresholds, our results suggest that similar results are highly likely in some target settings and possible in others. Funding: Wellcome Trust, UK Government, Royal Australasian College of Physicians, and Rotary Foundation.
ABSTRACT
BACKGROUND: Fever is a common presenting symptom in low- and middle-income countries (LMICs). It was previously assumed that malaria was the cause in such patients, but its incidence has declined rapidly. The urgent need to develop point-of-care tests for the most important causes of non-malarial acute febrile illness is hampered by the lack of robust epidemiological data. We sought to obtain expert consensus on analytes which should be prioritized for inclusion in fingerprick blood-based multiplex lateral flow rapid diagnostic tests (LF-RDTs) targeted towards four categories of patients with acute non-malarial fever in South and Southeast Asian LMICs, stratified by age (paediatric vs. adult) and care setting (primary vs. secondary care). METHODOLOGY/PRINCIPAL FINDINGS: We conducted a two-round modified e-Delphi survey. A total of 84 panellists were invited, consisting of seven each from 12 countries, divided into three regional panels (Mainland Southeast Asia, Maritime Southeast Asia, and South Asia). Panellists were asked to rank their top seven analytes for inclusion in LF-RDTs to be used in each patient category, justify their choices, and indicate whether such LF-RDTs should be incorporated into algorithm-based clinical decision support tools. Thirty-six panellists (43%) participated in the first round and 44 (52%) in the second. There was consensus that such LF-RDTs should be incorporated into clinical decision support tools. At a minimum, these LF-RDTs should be able to diagnose dengue and enteric fever in all patient categories. There was a clear preference to develop LF-RDTs for pathogens not readily detected by existing technologies, and for direct diagnosis through antigen detection. Pathogen biomarkers were prioritized over host inflammatory biomarkers, with CRP being the only one ranked consistently highly. CONCLUSIONS/SIGNIFICANCE: Our results provide guidance on prioritizing analytes for inclusion in context-specific multiplex LF-RDTs and similar platforms for non-malarial acute febrile illness, for which there is an urgent unmet need.
Subject(s)
Diagnostic Tests, Routine , Malaria , Adult , Child , Humans , Diagnostic Tests, Routine/methods , Malaria/epidemiology , Fever/diagnosis , Fever/etiology , Asia, SoutheasternABSTRACT
BACKGROUND: Magnetic resonance imaging (MRI) allows detailed study of structural and functional changes in the brain in patients with cerebral malaria. METHODS: In a prospective observational study in adult Bangladeshi patients with severe falciparum malaria, MRI findings in the brain were correlated with clinical and laboratory parameters, retinal photography and optic nerve sheath diameter (ONSD) ultrasound (a marker of intracranial pressure). RESULTS: Of 43 enrolled patients, 31 (72%) had coma and 12 (28%) died. MRI abnormalities were present in 79% overall with mostly mild changes in a wide range of anatomical sites. There were no differences in MRI findings between patients with cerebral and non-cerebral or fatal and non-fatal disease. Subtle diffuse cerebral swelling was common (n = 22/43), but mostly without vasogenic oedema or raised intracranial pressure (ONSD). Also seen were focal extracellular oedema (n = 11/43), cytotoxic oedema (n = 8/23) and mildly raised brain lactate on magnetic resonance spectroscopy (n = 5/14). Abnormalities were much less prominent than previously described in Malawian children. Retinal whitening was present in 36/43 (84%) patients and was more common and severe in patients with coma. CONCLUSION: Cerebral swelling is mild and not specific to coma or death in adult severe falciparum malaria. This differs markedly from African children. Retinal whitening, reflecting heterogeneous obstruction of the central nervous system microcirculation by sequestered parasites resulting in small patches of ischemia, is associated with coma and this process is likely important in the pathogenesis.
