Surveillance of respiratory viruses in the outpatient setting in rural coastal Kenya: baseline epidemiological observations

Abstract

Background: Endemic and seasonally recurring respiratory viruses are a major cause of disease and death globally. The burden is particularly severe in developing countries. Improved understanding of the source of infection, pathways of spread and persistence in communities would be of benefit in devising intervention strategies. Methods: We report epidemiological data obtained through surveillance of respiratory viruses at nine outpatient health facilities within the Kilifi Health and Demographic Surveillance System, Kilifi County, coastal Kenya, between January and December 2016. Nasopharyngeal swabs were collected from individuals of all ages presenting with acute respiratory infection (ARI) symptoms (up to 15 swabs per week per facility) and screened for 15 respiratory viruses using real-time PCR. Paediatric inpatient surveillance at Kilifi County Hospital for respiratory viruses provided comparative data. Results: Over the year, 5,647 participants were sampled, of which 3,029 (53.7%) were aged <5 years. At least one target respiratory virus was detected in 2,380 (42.2%) of the samples; the most common being rhinovirus 18.6% (1,050), influenza virus 6.9% (390), coronavirus 6.8% (387), parainfluenza virus 6.6% (371), respiratory syncytial virus (RSV) 3.9% (219) and adenovirus 2.7% (155). Virus detections were higher among <5-year-olds compared to older children and adults (50.3% vs 32.7%, respectively; χ ²(1) =177.3, P=0.0001). Frequency of viruses did not differ significantly by facility (χ ²(8) =13.38, P=0.072). However, prevalence was significantly higher among inpatients than outpatients in <5-year-olds for RSV (22.1% vs 6.0%; χ ²(1) = 159.4, P=0.0001), and adenovirus (12.4% vs 4.4%, χ ²(1) =56.6, P=0.0001). Conclusions: Respiratory virus infections are common amongst ARI outpatients in this coastal Kenya setting, particularly in young children. Rhinovirus predominance warrants further studies on the health and socio-economic implications. RSV and adenovirus were more commonly associated with severe disease. Further analysis will explore epidemiological transmission patterns with the addition of virus sequence data.

Keywords: Outpatient; Respiratory viruses; Surveillance; Real-time PCR; Nasopharyngeal samples; Coastal Kenya.

Figure 1.. A map of the Kilifi Health and Demographic Surveillance System (HDSS) area, coastal Kenya, expanded from a map of Kenya, showing population density (person per Km ²) and the health facilities where the study was conducted in 2016.

The red circles show the nine participating health facilities while the green markers show the other public health facilities within the KHDSS area.

Figure 2.. Percentage of nasopharyngeal swab samples positive for each of 15 respiratory virus targets for the nine health facilities (all facilities together and each individually) from ARI surveillance in the Kilifi Health and Demographic Surveillance System, Kenya, January to December 2016.

RSVA, RSV group A; RSVB. RSV group B; HRV, human rhinovirus; PIV1-PIV4, parainfluenza virus types 1–4; ADV, adenovirus; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229; FLUA-FLUC, influenza viruses types A-C; HMPV, human metapneumovirus.

Figure 3.. The relative proportion of nasopharyngeal swab samples positive across nine health facilities, by virus group, through surveillance of ARI presentations within the Kilifi Health and Demographic Surveillance System, Kenya, January to December 2016.

RSV, RSV A and B groups; Influenza, influenza A, B and C types; Coronaviruses, coronavirus OC43, NL63, 229e; Parainfluenza, parainfluenza types 1–4.

Figure 4.. Age-distribution of detections of 15 virus targets in nasopharyngeal swabs (NPS) identified from ARI surveillance at nine health facilities in the Kilifi Health and Demographic Surveillance System, Kenya, January to December 2016.

The primary Y axis shows frequency while secondary Y axis shows proportion in each age group. RSVA, RSV group A; RSVB. RSV group B; HRV, human rhinovirus; PIV1–PIV4, parainfluenza virus types 1–4; ADV, adenovirus; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229; FLUA-FLUC, influenza viruses types A–C; HMPV, human metapneumovirus.

Figure 5.. The distribution by month of the proportion of virus-positive nasopharyngeal swab samples over the period January to December 2016 for each of 15 virus targets, obtained through ARI surveillance at nine health facilities in the Kilifi Health and Demographic Surveillance System, Kenya.

Secondary Y axis records number of samples collected from recruits per month. RSVA, RSV group A; RSVB. RSV group B; HRV, human rhinovirus; PIV1-PIV4, parainfluenza virus types 1-4; ADV, adenovirus; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229; FLUA-FLUC, influenza viruses types A-C; HMPV, human metapneumovirus.

Figure 6.. Virus surveillance comparison between inpatient and outpatient facilities for children under 5 years in the Kilifi Health and Demographic Surveillance System (KHDSS), Kenya, 2016.

Panel A compares the proportion of nasopharyngeal swab samples positive for each of the 15 virus targets in samples collected from severe pneumonia admissions to Kilifi County Hospital (grey bars) and from outpatients presenting to nine health facilities (black bars). Violin plots show the distribution (median, IQR) for detectable rtPCR cycle threshold (Ct) values (i.e. Ct<=40) from respiratory samples for KHDSS outpatients (B) and for KCH inpatients (C). Threshold used for determining positive and negative samples shown by dashed line (Ct=35.0). RSVA, RSV group A; RSVB. RSV group B; HRV, human rhinovirus; PIV1-PIV4, parainfluenza virus types 1-4; ADV, adenovirus; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229; FLUA-FLUC, influenza viruses types A-C; HMPV, human metapneumovirus.