Geographic Variation in Invasive Pneumococcal Disease Following Pneumococcal Conjugate Vaccine Introduction in the United States

Transcription

1 MAJOR ARTICLE Geographic Variation in Invasive Pneumococcal Disease Following Pneumococcal Conjugate Vaccine Introduction in the United States Jennifer B. Rosen, 1,2 Ann R. Thomas, 3 Catherine A. Lexau, 4 Art Reingold, Jim L. Hadler, 6 Lee H. Harrison, 7 Nancy M. Bennett, 8 William Schaffner, 9 Monica M. Farley, Bernard W. Beall, 1 and Matt R. Moore 1 ; CDC Emerging Infections Program Network 1 Centers for Disease Control and Prevention, Atlanta, Georgia; 2 Epidemic Intelligence Service, Office of Workforce and Career Development, Atlanta, GA; 3 Oregon Public Health Division, Portland, Oregon; 4 Minnesota Department of Health, Saint Paul, Minnesota; California Emerging Infections Program, Oakland, California; 6 Connecticut Emerging Infections Program, Yale School of Public Health, New Haven, Connecticut; 7 Johns Hopkins Bloomberg School of Public Health, Department of International Health, Baltimore, Maryland; 8 University of Rochester School of Medicine and Dentistry, Center for Community Health, Rochester, New York; 9 Vanderbilt University School of Medicine, Department of Preventive Medicine, Nashville, Tennessee; and Georgia Emerging Infections Program, Department of Medicine, Emory University School of Medicine and the VA Medical Center, Atlanta, Georgia Background. Rates of invasive pneumococcal disease (IPD) varied among the United States before pneumococcal conjugate vaccine (PCV7) introduction. We compared trends in IPD rates among diverse US sites over years since PCV7 introduction. Methods. Patients with IPD of all ages were identified through active population and laboratory-based surveillance in 8 geographic areas under continuous surveillance during Isolates were serotyped. IPD incidence rates and percent changes were calculated by site, serotype group, age, and year. Results. Reductions in rates of IPD ranged, by site, from 19 to 29.9 cases per, population during to cases per, population during 29 (rate reduction, cases per, population). Reductions in IPD rates among children aged, years ranged from 3.7 to cases per, population across the sites. Reductions in rates of IPD due to PCV7 serotypes were seen in all age groups at all sites, ranging from 12 to 21.4 cases per, population during to,2 cases per, population during 29 (92% 98% reductions). Serotype 19A rates ranged from.4 to 1. cases per, population during to 1.3 to 3.4 cases per, population during 29 (rate difference, cases per, population); modest increases were observed for most age groups across the sites. Rates of IPD due to all other serotypes ranged from 6.3 to.3 cases per, population during to cases per, population during 29 (rate difference, 2.4 to.7 cases per, population). Across the sites, the greatest rate increases were seen in the 64 and.6 year age groups. Conclusions. Reductions in IPD due to vaccine serotypes were consistent across sites. Changes in serotype 19A and all other serotypes were variable. Although relative increases in non vaccine type serotypes were large in some sites, absolute rate increases were small. Streptococcus pneumoniae (pneumococcus) is a major cause of bacteremia, pneumonia, meningitis, and otitis Received 21 January 211; accepted 4 April 211. Correspondence: Jennifer Rosen, MD, NYC Department of Health and Mental Hygiene, 2 Lafayette St, New York, NY 7 (jrosen4@health.nyc.gov). Clinical Infectious Diseases 211;3(2): Published by Oxford University Press on behalf of the Infectious Diseases Society of America /211/32-$14. DOI:.93/cid/cir326 media in the United States. A 7-valent pneumococcal conjugate vaccine (PCV7) was licensed in the United States during 2 for routine use in children,2 years of age and for use in children aged 2 years with particular high-risk conditions. Before vaccine introduction, serotypes included in PCV7 accounted for 8% of cases of invasive IPD in US children aged, years and for % of cases in persons aged $6years [1]. PCV7 has been shown to be 96% effective against IPD due to vaccine serotypes in healthy children [2]. Geographic Variation in IPD d CID 211:3 (1 July) d 137

2 After vaccine introduction, rates of IPD due to vaccine serotypes decreased significantly among children in the vaccine target age group and among nonvaccinated children and adults. These indirect effects of vaccination have been attributed to reduced nasopharyngeal colonization among vaccinated children and reduced transmission from children to adults [3 ]. Studies conducted in such geographically diverse locations as Alaska, Utah, Texas, Massachusetts, California, and Arizona have consistently shown a decrease in rates of IPD due to PCV7 serotypes [6 12]. Although PCV7 has been highly successful, several studies have documented an increase in the rate of nasopharyngeal carriage of and infections with serotypes not included in the vaccine [6 8]; in the United States, serotype 19A has been largely responsible for the increase [7, 13]. Rates of IPD due to non-pcv7 serotypes have increased to a much greater extent among Alaska Natives and in Texas, compared with rates in Utah and Massachusetts [7, 12]. In contrast, rates in California and Arizona have not changed [6, 11]. The objective of our evaluation was to determine whether widespread use of PCV7 had different effects on rates of IPD and serotype distribution in 8 geographic areas under continuous surveillance since 1998 and, if so, to hypothesize why these differences exist. METHODS IPD Case Identification Cases of IPD were identified through the Centers for Disease Control and Prevention (CDC) Active Bacterial Core surveillance (ABCs), an active population-based, laboratory-based surveillance network. IPD was defined as the isolation of S. pneumoniae from a normally sterile body site (eg, cerebrospinal fluid or blood) in a surveillance area resident during 1 January 1998 through 31 December 29. Surveillance officers conducted laboratory audits periodically to ensure the completeness of reporting. Population Surveillance areas were under continuous surveillance during and included San Francisco County, California; Connecticut; 2 counties in Atlanta, Georgia; 6 counties comprising the Baltimore metropolitan area, Maryland; 7 counties in Minneapolis and Saint Paul, Minnesota; 7 counties in Rochester, New York; 3 counties in Portland, Oregon; and 4 metropolitan counties in Tennessee. The estimated combined population under surveillance in 29 was 19,319,3, including 1,328,7 aged, years, 3,349,194 aged 17 years, 8,873,29 aged years, 3,628,12 aged 64 years, and 2,14,477 aged $6 years. Population denominators were obtained from US Census population data. Laboratory Methods Serotyping of pneumococcal isolates using type-specific antiserum samples and the Quelling reaction was conducted at the CDC (Atlanta, GA) and at the Minnesota Department of Health (Saint Paul, MN). We grouped individual serotypes into the following mutually exclusive groups: PCV7-types (4, 6A, 6B, 9V, 14, 18C, 19F, and 23F), 19A, and all others. Serotype 6A was included as a PCV7 serotype on the basis of data demonstrating PCV7 cross-protection against serotype 6A [2]. Serotype 19A was analyzed separately, because we knew that there was no activity of PCV7 against serotype 19A and that this serotype has emerged as the most common nonvaccine serotype in the postvaccine era [2]. Statistical Analysis We calculated IPD incidence rates by surveillance site, serotype group, age, and year. Because isolates were not available for all cases, we assumed the same serotype distribution among cases for which serotype information was available for each site, age group, and year. To summarize the trends in IPD over time, the mean prevaccination rates in 1998 and 1999 were subtracted from the postvaccination rate in 29 for each age-site stratum to calculate the rate difference; relative risks and percent changes ([relative risk21] 3 ) with 9% confidence intervals (CIs) were also calculated for each of the comparisons in our analysis, including stratifications by serotype group, age, and site. Calculations and statistical analyses were performed using SAS, version (SAS Institute). RESULTS The total population under surveillance increased from a mean of 16,6,98 persons during to 19,319, in 29. Among residents from the ABCs sites in 29, 4% were from California, 19% were from Connecticut, 27% were from Georgia, 14% were from Maryland, 1% were from Minnesota, 6% were from New York, 8% were from Oregon, and 8% were from Tennessee. During , 894 cases of IPD were reported. An additional 27,269 cases were identified during Among the 3,363 IPD cases, 31,34 (89%) had isolates available for serotyping. Site- and Age-Specific Changes by Serotype All Serotypes During the pre-pcv7 baseline period , the IPD incidence across the sites for all age groups combined was cases/, population. After vaccine introduction in 2, rates of IPD decreased dramatically (Figure 1A) and, in 29, were cases/, population (Table 1). The overall reduction for all sites combined was.9 cases/, population, with significant decreases across all sites (range, 27%.9%) (Table 1). California, Georgia, and Maryland had the highest rates 138 d CID 211:3 (1 July) d Rosen et al

3 A 3 B Incidence per, 2 1 Incidence per, C 3 D Incidence per, 2 1 Incidence per, Figure 1. Rates of overall invasive pneumococcal disease (IPD) (A), PCV7-type IPD (B), serotype 19A IPD (C), and all other IPD (D), by site and year. before vaccine introduction and the greatest absolute rate reductions after vaccine introduction. Tennessee had the highest absolute rate in 29, driven by increases in rates of serotype 19A and all other nonvaccine serotypes (Table 1, Figure 1A). Reductions in overall IPD rates between the pre-pcv7 period and 29 among children aged, years were cases/, population across the sites. Rate reductions also occurred among unvaccinated older children and adults at most surveillance sites (Figure 2A). Absolute rate reductions in the $6 year age group were cases/, population. However, in Tennessee, there was an increase of 12.3 cases/, population in the 64 year age group that was driven by an increase in 19A and non-pcv7 IPD rates. PCV7 Serotypes The incidence of PCV7-type IPD for all ages combined was cases/, population across the sites during the prevaccination baseline period (Table 1, Figure 1B). After vaccine introduction, geographic variation was nearly eliminated. Rates of PCV7 type disease at all sites decreased to,2 cases/, population, representing statistically significant reductions of 92% 98% (Table 1). Rates of PCV7-type IPD also decreased across all age groups at all sites (Figure 2B). The greatest PCV7-type IPD reductions were seen in the vaccine target age group, resulting in 98% % reductions in overall IPD rates among children, years of age (Figure 2B). Rate reductions in this age group were cases/, population across the sites and were most pronounced in Georgia (Figure 2B).The next greatest decrease in PCV7-type IPD rates was seen in the $6 year age group, with an overall reduction of 3.3 cases/, population, which was most pronounced in Maryland (43.6 cases/, population). Serotype 19A Trends toward increasing rates of serotype 19A IPD were observed at most sites. Rates of serotype 19A before PCV7 vaccine introduction were.4 1. cases/, population and increased significantly at all sites except for California (91.4% 29.8%) Geographic Variation in IPD d CID 211:3 (1 July) d 139

4 Table 1. Invasive Pneumococcal Disease Before ( ) and After (29) Introduction of PCV7: Number of Cases, Rates, Rate Differences, and Percent Changes by Site and Serotype Group Number of cases Rate (cases/ population) Serotype group Site (mean) (mean) 29 Rate difference % Change (9% CI) All IPD CA (261.,239.) CT (246.1,231.7) GA (26.1,21.4) MD (21.2,238.2) MN (24.3,239.8) NY (24.2,217.7) OR (238.9,212.8) TN (239.7,217.4) All (247.4,241.9) PCV7-type a CA (29.4,284.9) CT (298.6,29.6) GA (298.4,296.4) MD (296.3,292.3) MN (298.2,294.3) NY (297.9,29.2) OR (298.3,292.) TN (296.2,289.7) All (296.7,29.3) 19A CA (26.7,292.6) CT (122.4,48.8) GA (11.1,348.) MD (134.9,497.4) MN (11.9,23.2) NY (142.,127.6) OR (31.8,826.8) TN (186.4,128.3) All (191.1,329.) All others b CA (23.,3.) CT (31.6,8.) GA (19.,62.7) MD (3.8,88.) MN (27.7,3.4) NY (14.7,.7) OR (42.,132.7) TN (36.1,117.3) All (33.4,4.1) a Includes serotypes included in the 7-valent pneumococcal conjugate vaccine (PCV7): 4, 6A, 6B, 9V, 14, 18C, 19F, 23F. b Excludes serotype 19A, 6A, and PCV7-type serotypes. (Table 1). Absolute increases in incidence were cases/, population, with relatively large 9% CIs. Modest increases in rates of serotype 19A IPD were observed for most age groups (Figure 2C). The greatest increase in 19A rates was seen in the, year age group (7.8 cases/, population). The greatest increases occurred in Tennessee and Maryland (11.2 cases/, population). All Other Serotypes Changes in rates of IPD caused by all other serotypes were less consistent across sites. Rates before vaccine introduction were cases/, population. Oregon and Tennessee had the greatest rate increases in all other serotypes (range, 72% 81.8%) (Table 1), but absolute rate increases were small (Figure 2D). Overall rates remained relatively unchanged in California, with a nonsignificant reduction of.4 cases/, population (Table 1), but varied by age group (Figure 2D). Changes in rates of IPD due to all other serotypes also varied by age group at other sites (Figure 2D). For the, year age group, Minnesota experienced a small rate decrease, whereas other sites experienced rate increases. Rates of IPD 14 d CID 211:3 (1 July) d Rosen et al

5 Figure 2. Absolute changes in rates in invasive pneumococcal disease (IPD), by age group and site, for 29, compared with the mean of 2 baseline years (1998 and 1999) before vaccine introduction for overall IPD (A), PCV7-type IPD (B), serotype 19A (C), and all other type IPD (D). The scale for A and B differs from that for C and D. due to all other serotypes decreased in the 17 and the 64 year age groups in California and increased at all other sites. The greatest rate increases were seen in the 64 and.6 year age groups, with the greatest increase in persons aged 64 years in Tennessee (16.4 cases/, population). Data from Tennessee were more closely evaluated to determine why overall IPD rates have been increasing among persons aged 64 years. There was no clustering of cases reported at the county or hospital level that would suggest a change in clinical practices. There were no clusters of individual serotypes suggestive of an outbreak. Although the distribution of all other serotypes was generally similar across the sites, serotype 12F was more common in Tennessee. The significance of this finding is unclear. No changes in case ascertainment practices in Tennessee were reported. The rate reduction in overall (.9 cases/, population) and PCV7-type (1.9 cases/, population) IPD exceeded the increase in IPD rates caused by 19A (1.9 cases/, population) and all other serotypes (3.1 cases/, population) at all sites for all ages combined (Table 1). All sites experienced at least a 27% reduction in IPD rates. DISCUSSION PCV7 has been highly effective at reducing PCV7-type IPD across the 8 ABCs sites included in this analysis, consistent with previous findings [6 12]. The similar 4-dose PCV7 coverage rates across the ABCs states (range, 79% 91%) among children 19 3 months of age may help explain the consistent trends in IPD rates that we observed ( imz-coverage.htm) [14]. The reduction in PCV7-type IPD rates in the, years age group was expected because this age group both had the highest rates of disease before vaccination and was the target group for vaccination. Although reductions in PCV7-type IPD rates were quite consistent across sites, increases in rates of IPD due to non- PCV7 serotypes, including serotype 19A, were more variable, despite similar rates of overall IPD evident before PCV7 introduction (Figure 1A). These differences may be related to a variety of factors. First, the modest differences in IPD rates before PCV7 introduction may reflect important differences in the underlying populations, such as differences in racial distributions; the prevalence of AIDS, which substantially increases the risk of IPD; differences in socioeconomic status; and Geographic Variation in IPD d CID 211:3 (1 July) d 141

6 differences in risk factors, such as smoking or prevalence of influenza infection [1 17]. The prevalence of chronic illness may vary by site, and patients with chronic illness may be more likely to have blood cultures performed. Thus, clinical practices, such as blood culture, among different patient groups may differ by site, leading to variable identification of IPD cases. Since vaccine introduction, the proportion of antibiotic-resistant serotype 19A isolates has increased significantly [18 2]; therefore, differences in rates across sites may be related to differing antimicrobial use [21, 22]. In a recent study using IMS Health National Sales Perspectives Data, antibiotic prescription rates per capita in the southern United States were significantly higher than rates in the western United States [23]. It is possible that certain clones of serotype 19A are more prevalent in some parts of the country. Increasing rates of IPD due to serotype 19A and all other serotypes may be of greater concern in particular sites. However, the significant reductions in IPD rates conferred by PCV7 at all sites continue to outweigh the relatively small rates of IPD due to other serotypes that we have observed. There are limitations to this analysis. First, we were unable to adjust for differences in age and race. Because data on risk factors (eg, underlying conditions, smoking prevalence, and influenza incidence) were not available at the level of the catchment area, we were unable to account for these factors in our analysis. Second, small case counts at several sites resulted in wide 9% CIs for percent changes in IPD rates (especially for serotype 19A). Third, the population for which vaccine coverage was determined is not necessarily the same as the ABCs population, limiting the ability to make a true comparison of vaccine coverage among surveillance sites. Fourth, the demographic characteristics of populations living in ABCs sites differ from those of the overall US population. On 24 February 2, a 13-valent pneumococcal conjugate vaccine (PCV13) was licensed by the US Food and Drug Administration for prevention of IPD caused by 7 serotypes in PCV7 plus 6 additional serotypes, including serotype 19A. Continued surveillance across different geographic locations to monitor changes in rates of disease due to non-pcv13 vaccine type will be crucial. Acknowledgments We thank Ruth Link-Gelles, Melissa M. Lewis, Elizabeth R. Zell, Cynthia G. Whitney, Chris Van Beneden, Tamara Pilishivi, Karrie-Ann Toews, Emily Weston, Carolyn Wright, Karen Stefonek, Mark Schmidt, Lori Triden, Brenda Jewell, Lindsey Lesher, Ruth Lynfield, Billie Juni, Susan Petit, Carmen Marquez, Mona Mondour, Rosemary Hollick, Kim Holmes, Kathleen Shutt, Glenda Smith, Jillian Karr, Suzanne Solghan, Gettha V. Nattanmai, Brenda G. Barnes, Terri L. McMinn, Wendy Baughman, Kathryn E. Arnold, and Paul Malpiedi for their contributions. Financial support. This work was supported by the Centers for Disease Control and Prevention Emerging Infections Program and the National Center for Immunization and Respiratory Diseases. Potential conflicts of interest. C. A. L. has received institutional grant funding from the CDC. J. L. H. has received CDC and CT Department of Public Health grant funding and has received payment for a board membership with CDC Office of Infectious Diseases Board of Scientific Counselors, for a consultancy with Council of State and Territorial Epidemiologists, and for a lecture with American Academy of Family Practitioners. L. H. H. has received payment for serving on the scientific advisory board for Sanofi Pasteur, Novartis Vaccines, GlaxoSmithKline, and Pfizer; has received an institutional grant from Sanofi Pasteur; and has received payment for lectures from Sanofi Pasteur and Novartis and for development of educational presentations by Discovery Institute of Medical Education. W. S. has received institutional grant funding from CDC; has received payment as an occasional consultant for Pfizer, Dynavax, Novartis, and GlaxoSmithKline; and has been a member of the data safety monitoring board at Merck. M. M. F. has received institutional grant funding from the CDC. N. M. B. has received payment for meeting consultancy with Wyeth. References 1. Centers for Disease Control and Prevention. Preventing pneumococcal disease among infants and young children: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR 2; 49: Whitney CG, Pilishvili T, Farley MM, et al. Effectiveness of sevenvalent pneumococcal conjugate vaccine against invasive pneumococcal disease: a matched case-control study. Lancet 26; 368: Whitney CG, Farley MM, Hadler J, et al. Decline in invasive pneumococcal disease after the introduction of protein-polysaccharide conjugate vaccine. N Engl J Med 23; 348: Centers for Disease Control and Prevention. Direct and indirect effects of routine vaccination. of children with 7-valent pneumococcal conjugate vaccine on incidence of invasive pneumococcal disease United States, 23. MMWR 2; 4: Lexau CA, Lynfield R, Danila R, et al. Changing epidemiology of invasive pneumococcal disease among older adults in the era of pediatric pneumococcal conjugate vaccine. JAMA 2; 294: Lacapa R, Bliss SJ, Larzelere-Hinton F, et al. Changing epidemiology of invasive pneumococcal disease among White Mountain Apache persons in the era of the pneumococcal conjugate vaccine. Clin Infect Dis 28; 47: Singleton RJ, Hennessy TW, Bulkow LR, et al. Invasive pneumococcal disease caused by nonvaccine serotypes among Alaska native children with high levels of 7-valent pneumococcal conjugate vaccine coverage. JAMA 27; 297: Messina AF, Katz-Gaynor K, Barton T, et al. Impact of the pneumococcal conjugate vaccine on serotype distribution and antimicrobial resistance of invasive Streptococcus pneumoniae isolates in Dallas, TX, children from 1999 through 2. Pediatr Infect Dis J 27; 26: Hsu K, Pelton S, Karumuri S, Heisey-Grove D, Klein J. Populationbased surveillance for childhood invasive pneumococcal disease in the era of conjugate vaccine. Pediatr Infect Dis J 2; 24: Byington CL, Samore MH, Stoddard GJ, et al. Temporal trends of invasive disease due to Streptococcus pneumoniae among children in the intermountain west: emergence of nonvaccine serogroups. Clin Infect Dis 2; 41: Black S, Shinefield H, Baxter R, et al. Postlicensure surveillance for pneumococcal invasive disease after use of heptavalent pneumococcal conjugate vaccine in Northern California Kaiser Permanente. Pediatr Infect Dis J 24; 23: Pelton SI, Huot H, Finkelstein JA, et al. Emergence of 19A as virulent and multidrug resistant pneumococcus in Massachusetts following universal immunization of infants with pneumococcal conjugate vaccine. Pediatr Infect Dis J 27; 26: Centers for Disease Control and Prevention. Invasive pneumococcal disease in children years after conjugate vaccine introduction eight states, 2. MMWR 28; 7: National Center for Immunization and Respiratory Diseases, CDC. National immunization survey, d CID 211:3 (1 July) d Rosen et al

7 1. Flannery B, Schrag S, Bennett NM, et al. Impact of childhood vaccination on racial disparities in invasive Streptococcus pneumoniae infections. JAMA 24; 291: Fry AM, Facklam RR, Whitney CG, Plikaytis BD, Schuchat A. Multistate evaluation of invasive pneumococcal diseases in adults with human immunodeficiency virus infection: serotype and antimicrobial resistance patterns in the United States. J Infect Dis 23; 188: Harrison L, Dwyer D, Billmann L, Kolczak M, Schuchat A. Invasive pneumococcal infection in Baltimore, MD: implications for immunization policy. Arch Intern Med 2; 16: Davies TA, Yee YC, Bush K, Sahm D, Evangelista A, Goldschmidt R. Effects of the 7-valent pneumococcal conjugate vaccine on U.S. levofloxacin-resistant Streptococcus pneumoniae. Microb Drug Resist 28; 14: Kyaw MH, Lynfield R, Schaffner W, et al. Effect of introduction of the pneumococcal conjugate vaccine on drug-resistant Streptococcus pneumoniae. N Engl J Med 26; 34: Moore MR, Gertz RE, Woodbury RL, et al. Population snapshot of emergent Streptococcus pneumoniae serotype 19A in the United States, 2. J Infect Dis 28; 197: Greenberg D, Givon-Lavi N, Sharf AZ, Vardy D, Dagan R. The association between antibiotic use in the community and nasopharyngeal carriage of antibiotic-resistant Streptococcus pneumoniae in Bedouin children. Pediatr Infect Dis J 28; 27: Dagan R. Increased importance of antibiotic-resistant (AR) S. pneumoniae serotype 19A (Pn19A) in acute otitis media (AOM) occurring before introduction of 7-valent pneumococcal conjugate vaccine (PCV7) in Southern Israel. 47th Interscience Conference on Antimicrobials and Chemotherapy. Chicago, IL: American Society for Microbiology, Hicks LA, Suda KJ, Roberts RM, et al. Antimicrobial prescription data reveal wide geographic variability in antimicrobial use in the United States, th Annual Meeting of the Infectious Diseases Society of America. Vancouver, Canada: Infectious Disease Society of America, 2. Geographic Variation in IPD d CID 211:3 (1 July) d 143