Symptom-based Scoring for Acute Human Immunodeficiency Virus.

To the Editor—We read with interest the report by Lin et al showing that a simple symptom score consisting of fever, myalgia, and weight loss accurately predicted acute human immunodeficiency virus (HIV) infection (AHI) [1]. We agree with the authors that symptom-based assessment is less prone to limitations inherent to risk-based scores, as “symptoms may be less subject to stigma, and therefore individuals may be more comfortable disclosing symptoms than sexual behaviors” [1]. Our concern is that the symptom score was developed in the United States, and therefore may not optimally identify AHI in resource-limited countries, as recommended by the authors.

We have previously developed a symptom-based score with data from at-risk and general populations from Kenya, Malawi, and South Africa [2]. This score assigns 1 point each for age 18–29 years or reported fever, fatigue, body pains, diarrhea, or sore throat, and 3 points for reported genital ulcer disease; individuals scoring ≥2 should be tested for AHI [2]. We are using this score to detect AHI with the Xpert HIV-1 Qual assay (Cepheid, Sunnyvale, California) among adults aged 18–39 years seeking urgent care in coastal Kenya (R01AI124968, ongoing). While HIV-1 RNA testing for AHI diagnosis is not supported by policy in sub-Saharan Africa, an exclusive focus on identifying chronic HIV in seropositive adults leads to missed opportunities [3]. This is especially important as preexposure prophylaxis (PrEP) is being scaled up in African settings.

The following case history from a voluntary testing and counseling center affiliated with our research clinic in coastal Kenya illustrates this: A 24-year-old heterosexual man tested negative on 2 HIV rapid antibody tests, whereas his female partner of 3 months tested antibody positive in the same session. He reported diarrhea and fatigue in the preceding 4 days, but no fever, weight loss, or myalgia. He was eligible for PrEP per Kenyan guidelines [4], as he was in a serodiscordant relationship. The patient met 3 of the criteria (young age, fatigue, and diarrhea) from our symptom-based score and was therefore tested with the Xpert HIV-1 Qual assay [2]. He tested positive, as confirmed by a viral load of 5500 copies/mL by Xpert HIV-1 Quant assay. He enrolled in an AHI cohort and started antiretroviral therapy shortly thereafter.

Per Kenyan guidelines, healthcare providers should assess for AHI symptoms prior to PrEP initiation when a recent high-risk exposure is reported [4]. While most front-line healthcare providers in sub-Saharan Africa received no specific training about AHI diagnosis [5, 6], PrEP guidelines offer a glimmer of hope that AHI symptoms will now be assessed in at-risk clients under evaluation for PrEP eligibility. We propose that this symptom screening should be done with our symptom-based score in African settings [2], as limiting AHI screening to those with fever, myalgia, and weight loss will lead to missed opportunities according to our data. As AHI testing should be targeted in resource-limited settings, we applaud the efforts of Lin and colleagues to promote the concept of targeted testing and encourage further research into this important area.