Establishing Cutoffs for Streptococcal Antibody Titers in Uganda: Implications for Rheumatic Fever Diagnosis

Abstract

AbstractBackgroundCentral to rheumatic fever (RF) diagnosis is evidence of streptococcal exposure, specifically antistreptolysin O (ASO) and antideoxyribonuclease B (ADB) antibodies. It is unknown if these antibody titers should be adjusted to the background exposure rates of GAS or if published standards should be used. Here, we establish the normal values of ASO and ADB in Uganda and examine RF case detection using published vs. population-specific thresholds.MethodsParticipants (age 0-50 years) were recruited. ASO was measured in-country by nephelometric technique. ADB samples were sent to Australia (PathWest) for ADB determination by enzyme inhibition assay, andthe 80% upper limit values by age were established. The published standard values for ASO (200IU/ml) and ADB (375IU/ml) were compared to the Ugandan 80% upper limit of normal values (ULN) for RF case detection in children 5-15 years.FindingsOf the 428 participants, 16 were excluded from analysis (9 sore throat, 1 skin sores, 5 fever, 4 echocardiograms showing occult RHD), and 183 of the remaining were children 5-15 years. The median ASO titer in this age group was 220 IU/ml, with the 80th percentile value of 389 IU/ml. The median ADB titer in this age group was 375 IU/ml, with the 80th percentile value of 568 IU/ml. Application of new Ugandan cutoffs to 528 children enrolled in our prospective RF study, reduced the number of definite RF cases to 120/528 (22·7%), as compared to 173/528 (32·8%) using published normal values.InterpretationThe 80th percentile ULN for ASO and ADB are higher in Uganda than in other countries. Applying these higher values to RF diagnosis in Uganda results in higher diagnostic specificity, but some unknown loss in sensitivity. Implications of over-diagnosis and missed cases will be explored through a longitudinal follow-up study of children in the RF research program.FundingThis work was supported by American Heart Association Grant #17SFRN33670607 / Andrea Beaton / 2017 and DELTAS Africa Initiative.Research in contextEvidence before this studyWe searched PubMed for data on normal values of streptococcal antibody titers within diverse populations between database inception and January 1, 2019, using the search terms (rheumatic fever) OR (streptococcal antibodies). Nine studies were identified, but only one was from sub-Saharan Africa (2018, Ethiopia) and it was limited by vague exclusion criteria and lack of data on anti-DNase B. Given the high burden of rheumatic heart disease in sub-Saharan Africa, further data is needed to determine normal streptococcal antibody titers in this population and to assess the clinical impact of different cutoffs for RF diagnosis.Added value of this studyOur study utilized a rigorous approach to exclude patients with history of recent possible streptococcal exposure including skin and throat infection and employed echocardiography to exclude patients with pre-existing rheumatic heart disease. Additionally, this study was conducted in parallel to a larger epidemiological cohort study of rheumatic fever in Uganda, allowing us, for the first time, to prospectively determine how utilization of different streptococcal antibody titer cutoffs affect diagnosis of rheumatic fever.Implications of all the available evidenceRheumatic fever remains a challenging diagnosis based on a clinical decision rule with imperfect sensitivity and specificity. Improved understanding of streptococcal antibody titers in rheumatic heart disease endemic populations may improve diagnostic performance. Our study also points to the need for development of a rheumatic fever diagnostic test, in order to provide a more definitive assessment of risk.