LB 5 Healthcare-Associated Respiratory Infections in Kenyan Hospitals: Preliminary Results from a New Surveillance System

Saturday, March 20, 2010: 11:30 AM
International South (Hyatt Regency Atlanta)
Linus K. Ndegwa, MPHE , Centers for Disease Control- Kenya, Nairobi, Kenya
Mark Katz, MD , Centers for Disease Control- Kenya, Nairobi, Kenya
Lilian Mayieka, Bsc , Centers for Disease Control- Kenya, Nairobi, Kenya
Juliana A. Otieno, MD , New Nyanza Provincial General Hospital, Kisumu, Kenya
Ann G. Mungai, MD , Kenyatta National Hospital, Nairobi, Kenya
Francis M'thuranira, RN , Mbagathi District Hospital, Nairobi, Kenya
Zipporah Nganga, PhD , Jomokenyatta University of Agriculture and Technology, Nairobi, Kenya
Martin K. H. Kollmann, MD , University of Nairobi, Nairobi, Kenya
Josephine Koli, RN , Ministry of Medical Services, Nairobi, Kenya
Martha A. Atandi, DCMS , Centers for Disease Control- Kenya, Nairobi, Kenya
Kenneth M. Langakuo, DCMS , Centers for Disease Control- Kenya, Nairobi, Kenya
George M. Mtsosi, DCMS , Centers for Disease Control- Kenya, Nairobi, Kenya
Danny Feikin, MD , Centers for Disease Control- Kenya, Nairobi, Kenya
Robert F. Breiman, MD , Centers for Disease Control- Kenya, Nairobi, Kenya
Kariuki Njenga, PhD , Centers for Disease Control- Kenya, Nairobi, Kenya
Katherine Ellingson, PhD , Centers for Disease Control and Prevention, Atlanta, GA
Background:  Healthcare-associated infections (HAIs) are an important cause of morbidity and mortality worldwide. However, there are sparse data on HAIs in developing countries. As part of an initiative to build infection control capacity, CDC-Kenya and the Kenyan Ministry of Health initiated surveillance for HAIs at 3 public hospitals in late 2009: a national referral hospital, a provincial general hospital and a district hospital. Patient crowding and high prevalence of respiratory illnesses at admission have made healthcare-associated respiratory infections (HARIs) a concern for Kenyan hospitals.

Objective: To document HARIs occurring on selected wards in 3 Kenyan hospitals, and to assess the burden of viral HARIs, including pandemic (H1N1) 2009 influenza (pH1N1).

Methods: At each site, surveillance officers surveyed pediatric, adult general, surgical, and specialty wards for HAIs. Patients admitted to the hospital for >3 calendar days who developed new onset of fever or hypothermia (>38°C or <35°C) were considered suspected HAI cases and were further assessed for onset of clinical symptoms and signs by questionnaire and medical record review. Suspected HAI cases who also developed new onset of cough or sore throat >3 calendar days after admission were considered to have HARI; nasopharyngeal and oropharyngeal samples were collected from these patients. Specimens were sent to the CDC-Kenya laboratory in Nairobi, where they were tested by RT-PCR for influenza A and B, adenovirus, respiratory syncytial virus, human metapneumovirus, and parainfluenza virus 1, 2 and 3. Specimens positive for influenza A were sub typed by RT-PCR.

Results: From September 1-November 30, 2009, 54 patients met the case definition for HARI; 37 (69%) were male, and the median age was 18 years (range: 7 days to 70 years). The median time patients were in the hospital before onset of HARI was 28 days (range: 4-286 days). These infections were identified on pediatric wards [25 (46%)], specialty wards [25(46%)], and medical adult wards [4 (7%)]). Of those with HARI, 19 (35%) tested positive for a viral pathogen. The most commonly identified viruses were adenovirus [8(15%)] and influenza A [7(13%)], and 6 (11%) patients with HARI were infected with more than one pathogen. Of the 7 influenza A infections, 4 were pandemic 2009 H1N1.

Conclusions: HARIs have been documented in 3 Kenyan hospitals, and at least one-third were associated with viruses, including pH1N1. Ongoing interventions to improve hand hygiene and rapidly identify and cohort patients with febrile respiratory illnesses can help to reduce HARI. The prevalence of viral pathogens identified suggests that a substantial proportion of patients with HARI in this setting may not require empiric antibiotic treatment; this may represent an opportunity to decrease antimicrobial use.