Date: Fri, 26 Apr 1996 17:13:34 -0400 From: "Flynn Mclean" Subject: MMWR 04/26/96 MORBIDITY AND MORTALITY WEEKLY REPORT ****************************************** Centers for Disease Control and Prevention April 26, 1996 Vol. 45, No. 16 * Update: Influenza Activity _ United States and Worldwide, 1995_96 Season, and Composition of the 1996_97 Influenza Vaccine * Multidrug-Resistant Tuberculosis Outbreak on an HIV Ward _ Madrid, Spain, 1991_1995 Update: Influenza Activity _ United States and Worldwide, 1995_96 Season, and Composition of the 1996_97 Influenza Vaccine To monitor influenza activity and to detect antigenic changes in the circulating strains of influenza viruses, CDC conducts surveillance in collaboration with the World Health Organization (WHO) and its international network of collaborating laboratories and with state and local health departments in the United States. This report summarizes surveillance for influenza in the United States and worldwide during the 1995_96 season and describes the composition of the 1996_97 influenza vaccine. United States Influenza activity began in November 1995 and peaked during late December 1995 and early January 1996. In many parts of the country, influenza activity declined steadily during January and February; of the 34 states that reported levels of influenza-like illness for the week ending April 13, a total of 16 states reported sporadic* levels of influenza-like illness, and 18 states reported no activity. Of the 4132 influenza virus isolates reported to CDC from WHO collaborating laboratories in the United States from October 1, 1995, through March 30, 1996, a total of 3786 (92%) were influenza type A and 346 (8%) influenza type B. Of the 2416 type A isolates that were subtyped, 1427 (59%) were type A(H1N1), and 989 (41%) were type A(H3N2). Influenza type A(H3N2) predominated in the Mountain, New England, and Pacific regions, accounting for 70%, 56%, and 55% of subtyped influenza A isolates, respectively. Influenza type A(H1N1) predominated in the other six regions, accounting for 55%_82% of subtyped influenza A isolates. During February, although the total number of isolates decreased, the number and proportion of influenza type B isolates began to increase, and during March 1996, 50%_72% of all isolates reported were type B. The proportion of all deaths reported by the vital statistics offices of 121 U.S. cities that were attributed to pneumonia and influenza (P&I) only slightly exceeded the epidemic threshold ** during 3 of the 8 weeks from October 29 through December 23, 1995. During the 6 weeks from December 24, 1995, through February 3, 1996, the proportion of P&I deaths remained above the epidemic threshold, peaking at 8.2% of all deaths during the week ending January 20. However, since February 10, percentages of P&I deaths have been below the epidemic threshold. Worldwide Influenza activity occurred at moderate to severe levels during October 1995_ March 1996. Epidemics associated with influenza A(H3N2) and A(H1N1) viruses were reported in countries in Europe, Asia, and North America, while influenza B viruses circulated at low levels. School outbreaks caused by influenza A(H3N2) viruses were reported in England beginning in September and October. During November and December, epidemic activity associated primarily with A(H3N2) viruses was reported by countries throughout Europe, including Belarus, Bulgaria, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Latvia, Netherlands, Norway, Slovakia, Spain, Sweden, and the United Kingdom. During January, influenza A(H3N2) viruses were associated with outbreaks in Beijing and with high levels of influenza-like illness in six northern provinces of China. Isolation of influenza A(H3N2) viruses also was reported in North America (Canada and the United States), Europe (Belgium, Iceland, Ireland, Italy, Poland, Portugal, the Russian Federation, and Switzerland), Asia (Guam, Hong Kong, Japan, and Singapore), and Oceania (Australia and New Zealand). For the first time sincethe 1991_92 influenza season, A(H1N1) viruses were associated with epidemics in several regions of the world. Epidemic activity associated with influenza A(H1N1) viruses was reported and predominated in Belgium, Canada, Japan, southern France, Switzerland, and the United States. Influenza A(H1N1) viruses were isolated in association with sporadic activity in Europe (Finland, Germany, Italy, Latvia, Netherlands, Poland, Romania, Russian Federation, Spain, Sweden, and the United Kingdom) and Asia (China, Hong Kong, Israel, and Thailand) In comparison to type A influenza viruses, type B viruses have been isolated later in the season and less frequently worldwide. Influenza B viruses were isolated primarily in association with sporadic activity in North America (Canada and the United States), Asia (China, Hong Kong, Israel, Japan, and Singapore), Europe (Belarus, Finland, France, Germany, Greece, Hungary, Netherlands, Poland, Romania, Russian Federation, Sweden, Switzerland, and the United Kingdom), and Oceania (Australia and New Zealand). Composition of the 1996_97 Vaccine The Food and Drug Administration Vaccines and Related Biological Products Advisory Committee (VRBPAC) has recommended that the 1996_97 trivalent influenza vaccine for the United States contain A/Wuhan/359/95-like(H3N2), A/Texas/36/91-like (H1N1), and B/Beijing/184/93-like viruses. This recommendation was based on the antigenic analysis of recently isolated influenza viruses and the antibody responses of persons vaccinated with the 1995_96 vaccine. Although most of the influenza type A(H3N2) viruses that have been antigenically characterized are similar to the A/Johannesburg/33/95 strain, increasing numbers of recently isolated A(H3N2) strains from Asia, Europe, and North America are more similar to the antigenic variant A/Wuhan/359/95 (Table 1). Vaccines containing the A/Johannesburg/33/94 (H3N2)-like virus induced a good antibody response to the vaccine strain but induced lower and less frequent antibody responses to recent type A(H3N2) strains such as A/Wuhan/359/95 (1). Therefore, VRBPAC recommended changing the influenza type A(H3N2) vaccine component to an A/Wuhan/359/95-likestrain for the 1996_97 season. The strain that will be used by U.S. vaccine manufacturers because of its growth properties will be the antigenically equivalent A/Nanchang/ 933/95 virus. Virtually all (98%) influenza A(H1N1) viruses that have been antigenically characterized are similar to the reference strains A/Taiwan/01/86 and A/Texas/36/91. Because vaccines containing the A/Texas/36/91 strain induced antibodies with similar frequency and titer to both the vaccine virus and to recent type A(H1N1) strains (1), VRBPAC recommended retaining an A/Texas/36/91-like strain in the 1996_97 vaccine. Antigenically characterized influenza B viruses isolated recently in Asia, Europe, and the United States have been similar to the reference strains B/Beijing/184/93 and B/Harbin/07/94. Vaccines containing the B/Harbin/07/94 strain induced antibodies with similar frequency and titer to the vaccine virus and to recently isolated influenza B strains (1). Therefore, VRBPAC recommended retaining B/Harbin/07/94-like strain in the 1996_97 vaccine. Reported by: Participating state and territorial health dept epidemiologists and state public health laboratory directors. M Zambon, PhD, Central Public Health Laboratory, A Hay, PhD, National Institute for Medical Research, London; G Schild, DSc, J Wood, PhD, National Institute for Biological Standards and Control, Hertfordshire, England. I Gust, MD, A Hampson, Commonwealth Serum Laboratories, Parkville, Australia. L Canas, Armstrong Laboratory, Brooks Air Force Base, Texas. Y Guo, Institute of Virology, National Center for Preventive Medicine, Beijing, People's Republic of China. World Health Organization National Influenza Centers, Div of Emerging and other Communicable Diseases Surveillance and Control, Geneva. Div of Virology, Center for Biologics Evaluation and Research, Food and Drug Administration. Influenza Br, Div of Viral and Rickettsial Diseases, National Center for Infectious Diseases, CDC. Editorial Note: During the 1995_96 season, the impact of influenza in many parts of the United States and in some other countries in the Northern Hemisphere was more severe than during the previous season (2). In the United States, influenza type A(H1N1) viruses predominated for the first time since the 1986_87 season; although this subtype has not been associated with excess mortality in recent decades, the incidence of infection with type A(H1N1) has been high, especially among school-aged children. Influenza type A(H3N2) was not the predominant strain but circulated throughout the season and was associated with outbreaks among all age groups. Continued circulation of influenza type A(H3N2) and type A(H1N1) is anticipated during the 1996_97 season. Influenza B activity increased late in the 1995_96 influenza season, suggesting that type B viruses may circulate more widely next winter. Strains to be included in the influenza vaccine usually are selected during the preceding January through March because of scheduling requirements for production, quality control, packaging, and distribution of vaccine for administration before onset of the next influenza season. Recommendations of the Advisory Committee on Immunization Practices for the use of vaccine and antiviral agents for prevention and control of influenza will be published in an MMWR Recommendations and Reports on May 3, 1996. References 1. World Health Organization. Recommended composition of influenza virus vaccines for use in the 1996_1997 flu season. Wkly Epidemiol Rec 1996;71:57_61. 2. CDC. Update: influenza activity_United States and worldwide, 1994_95 season, and composition of the 1995_96 influenza vaccine. MMWR 1995;44:292_5. *Levels of activity are 1) sporadic_sporadically occurring influenza-like illness (ILI) or culture-confirmed influenza with no outbreaks detected; 2) regional_outbreaks of ILI or culture- confirmed influenza in counties with a combined population of <50% of the state's total population; and 3) widespread_outbreaks of ILI or culture-confirmed influenza in counties with a combined population of =>50% of the state's total population. ** The epidemic threshold is 1.645 standard deviations above the seasonal baseline calculated using a periodic regression model applied to observed percentages since 1983. The baseline was calculated using a robust regression procedure. Multidrug-Resistant Tuberculosis Outbreak on an HIV Ward _ Madrid, Spain, 1991_1995 Beginning in 1990, outbreaks of multidrug-resistant tuberculosis (MDR-TB) have been reported in hospitals and prisons in the eastern United States (1). During June 1991_January 1995, MDR-TB was diagnosed in 47 patients and one health-care worker at a 120-bed, infectious disease referral hospital in urban Madrid; on April 19, 1995, the Spanish Field Epidemiology Training Program was asked to investigate this outbreak. This report summarizes the findings of this investigation, which suggested that nosocomial transmission of MDR-TB occurred on a hospital ward for patients with human immunodeficiency virus (HIV) infection. A case of MDR-TB was defined as culture-confirmed TB that was resistant to at least rifampin and isoniazid in a patient hospitalized on the ward for HIV-infected persons during June 1991_January 1995 and with no previous history of TB treatment. Case finding was coordinated by the mycobacteriology laboratory director and an infectious disease specialist, who reviewed medical records and laboratory results for persons with suspected MDR-TB. In addition to drug-susceptibility testing, analysis of resistant strains included DNA fingerprinting with restriction fragment-length poly-morphism (RFLP). Because the hospital did not have in place a ventilation system that recirculated or removed air, the acid-fast bacilli (AFB) isolation capacity (negative pressure and number of air interchanges per hour) could not be assessed on the HIV ward. MDR-TB was identified in 47 HIV-positive patients who had been hospitalized on the HIV ward during June 1991_January 1995. The mean age of case-patients was 34 years (range: 25_54 years); 39 (81%) were male, and 32 (67%) were injecting-drug users. The one health-care worker was HIV-positive and had worked on the HIV ward during 1990_1994. A total of 47 (98%) patients, including the health-care worker, had died at the time of the investigation; the mean interval from diagnosis of MDR-TB to death was 78 days. An analysis of isolates from TB cases throughout the hospital during 1991_June 1995 identified 104 that were drug-susceptible; 12 that were resistant to one drug; and 66 that were resistant to isoniazid, streptomycin, ethambutol, and rifampin (HSER) (Figure 1). The proportion of Mycobacterium tuberculosis strains identified that were MDR-TB increased from 10% in 1991 to 53% in 1993 to 65% in June 1995. Beginning in 1993, the resistance pattern identified consistently in isolates was HSER: of the 26 cases diagnosed during October 1993_June 1995, this pattern was present in 24 (92%). Of the 12 isolates available for DNA fingerprinting, the same band patterns were present in 11 (Figure 2). For comparison, TB isolates were obtained from the two patients with different antibiograms; their RFLP analyses were distinct from those of isolates from the other patients. A case-control study was conducted to identify potential risk factors for MDR-TB among HIV-infected patients who had been hospitalized on the HIV ward during September 15, 1991_December 31, 1994, and in whom TB was diagnosed in 1994. Cases included patients with isolates with the HSER resistance pattern (n=18); controls were patients with isolates sensitive to rifampin, isoniazid, streptomycin, and ethambutol (n=17). The category _potentially infective_ for TB patients was defined as the period from 2 weeks before a positive sputum smear or TB culture confirmation until sputumcultures were negative or until death. _Possibly exposed_ for patients without TB was defined as hospitalization on the HIV ward concurrent with the hospitalization of a potentially infectious patient during the period until 2 weeks before TB was diagnosed in the potentially infectious patient. Case- and control-patients were similar in age, sex, HIV risk group, interval of time between HIV diagnosis and TB diagnosis, and CD4+ T-lymphocyte count at the time of TB diagnosis. However, before the hospitalization during which MDR- TB was diagnosed, 13 (72%) of the case-patients had been hospitalized on the HIV ward, compared with five (29%) control patients (odds ratio=6.2; 95% confidence interval=1.2_36.7). Of all patients with TB diagnosed in 1994 who were hospitalized on the HIV ward, 5% had MDR-TB. Case patients were more likely to have been possibly exposed to potentially infective wardmates and to have more days of exposure (13 [72%] for a median of 26 days) than control patients (seven [41%] for a median of 8 days) (for duration of exposure, chi square for linear trend=7.0; p=0.03). Reported by: D Herrera, R Cano, P Godoy, EF Peiro, J Castell, C Ibanez, F Martinez Navarro, Field Epidemiology Training Program; V Moreno, A Ortega, L Sanchez, R Duran, F Pozo, Carlos III Health Institute, Ministry of Health and Consumer Affairs, Spain. Div of Bacterial and Mycotic Diseases, National Center for Infectious Diseases; International Br, Div of Field Epidemiology, Epidemiology Program Office, CDC. Editorial Note: The findings in this report document the first outbreak of nosocomial MDR-TB to be investigated in Spain. Characteristics of this outbreak that are similar to previously reported outbreaks include MDR-TB among patients hospitalized in an HIV-dedicated ward, a high death rate within 3 months of onset, and the role of mycobacteriology laboratory-based surveillance in recognizing similar resistance patterns with confirmation through RFLP fingerprinting (2). Measures to control this outbreak have included 1) isolating all MDR-TB patients in a separate area of the hospital and the on-site provision of all clinical and diagnostic services; 2) notifying family, community members, and wardmates of patients whose MDR-TB had been diagnosed during January_June 1995 about their exposure, scheduling follow-up evaluation, and offering isoniazid preventive therapy (although isoniazid resistance had been identified in isolates from the outbreak, this resistance was low); 3) informing all hospital staff about the outbreak, and establishing a TB screening clinic that was attended by 565 (96%) of 591 employees; 4) purchasing personal respiratory protection devices that fulfilled recommended sealage and filtering criteria (3) and distributing these devices to staff exposed to TB patients; and 5) developing plans to improve the capacity of the hospital's mycobacteriology laboratory and to install 11 AFB isolation rooms. To prevent nosocomial transmission of M. tuberculosis, hospital staff should monitor surveillance for and rapidly diagnose, isolate, and treat persons with suspected TB and ensure timely laboratory confirmation with identification of drug- susceptibility patterns. Because immunocompromised persons, such as those on HIV wards, are at increased risk for TB, surveillance and rapid confirmation are especially important to prevent M. tuberculosis transmission. In addition, hospitals and other health-care facilities should conduct regular employee TB screening clinics (graded by occupational risk category) that closely monitor tuberculin skin test conversions; such clinics can assist in surveillance for nosocomial transmission of TB. References 1. Kent JH. The epidemiology of multidrug-resistant tuberculosis in the United States. Med Clin North Am 1993;77:1391_409. 2. CDC. Nosocomial transmission of multidrug-resistant tuberculosis among HIV-infected persons_Florida and New York, 1988_1991. MMWR 1991;40:585_91. 3. CDC. Guidelines for preventing the transmission of Mycobacterium tuberculosis in health-care facilities, 1994. MMWR 1994;43(no. RR-13).