Search

Background Little is known about the prevalence or incidence of Zika virus (ZIKV) infection in settings affected by the 2015–2016 Zika pandemic and associated risk factors. We assessed these factors among household contacts of patients with ZIKV disease enrolled in a cohort study in Puerto Rico during 2016–2017. Methods Household contacts of index case patients completed a questionnaire and gave specimens for real-time polymerase chain reaction (RT-PCR) and immunoglobulin M enzyme-linked immunosorbent assay testing to detect ZIKV infection. We measured the prevalence of ZIKV infection among contacts and associated individual and household factors, examined sexual transmission using a sexual-networks approach, and assessed incident infection among initially uninfected household contacts 2–4 months later. Results Of 366 contacts, 34.4% had evidence of ZIKV infection at enrollment, including 11.2% by RT-PCR. Having open doors and windows that were either screened (prevalence ratio [PR], 2.1 [95% confidence interval {CI}, 1.2–3.6]) or unscreened (PR, 2.5 [95% CI, 1.5–4.1]) was associated with increased prevalence. Sexual partners were more likely to both be RT-PCR positive relative to other relationships (odds ratio, 2.2 [95% CI, 1.1–4.5]). At follow-up, 6.1% of contacts had evidence of incident infection. Conclusions This study identified sexual contact as a risk factor for ZIKV infection. Persons living with ZIKV-infected individuals should be a focus of public health efforts.

Background Identifying factors associated with time-to-loss of Zika virus (ZIKV) RNA in serum and semen is important to inform diagnostic testing and prevention recommendations. CDC currently recommends RT-PCR testing of serum up to two weeks after symptom onset. We evaluated such associations among participants of the Zika virus Persistence (ZiPer) study in Puerto Rico. Methods Patients presenting for care with Zika-like illness and ZIKV RNA detected by RT-PCR in serum or urine (index cases) were offered study participation. Index cases’ household members were offered study participation, and those with detectable ZIKV RNA were eligible for the prospective cohort. Serum and semen were collected weekly for the first month, and biweekly thereafter for participants with detectable ZIKV RNA in any fluid and at 2, 4, and 6 months post-enrollment for all others. We used chi-squared and Fischer’s exact tests to assess if detecting ZIKV RNA in specific specimens at any point was associated with sex, age, Zika-like symptoms (rash, fever, arthralgia, or conjunctivitis), or pregnancy. We performed Weibull regression models to estimate time-to-loss of ZIKV RNA in days post symptom onset (DPO) and evaluated associations between covariates and duration of detection. Results Among 295 participants, 260 (88.1%) had ZIKV RNA detected in serum at any point. Participants aged ≥18 years (n = 244) had a significantly longer median time-to-loss of ZIKV RNA in serum than participants aged < 18 years (n = 50) (13.1 vs. 7.8 DPO, respectively; P = 0.003) (Figure 1). Among women aged 18–39 years (n = 60), pregnant women (n = 9) had a significantly longer median time-to-loss of ZIKV RNA in serum than non-pregnant women (n = 51) (37.4 vs. 15.5 DPO, respectively; P = 0.0005) (Figure 2). The proportion of men who had detectable ZIKV RNA in semen at any point was significantly higher among men with conjunctivitis (47 of 82) than among men without conjunctivitis (3 of 14) (P = 0.01). No other associations were significant. Conclusion Time-to-loss of ZIKV RNA in serum was longer among adults than children, and conjunctivitis was associated with detecting ZIKV RNA in semen. This study provides evidence that time-to-loss of ZIKV RNA is longer among pregnant women than non-pregnant women. Findings may inform the recommended period to test pregnant women for ZIKV using RT-PCR. Disclosures All authors: No reported disclosures.