Streptococcus pneumoniae, pneumococcus, causes a variety of diseases in children, including invasive pneumococcal diseases as well as lower and upper respiratory tract diseases such as pneumonia, otitis media and sinusitis. Otitis media in particular causes a considerable disease burden in young children.1,2 Otitis media is also the most common indication for antimicrobial treatment in children.3 Furthermore, tympanostomy (ventilation) tube placements are the most common surgical procedures in children.4,5
The pneumococcal vaccines have been demonstrated to prevent pneumococcal diseases early in infancy. In the Finnish Otitis Media (FinOM) vaccine trial1 in 1662 subjects, the efficacy of the 7-valent PncCRM conjugate vaccine (Prevenar; Pfizer, Philadelphia, PA) against vaccine-type culture-confirmed pneumococcal acute otitis media was 57% (95% confidence interval [CI]: 44 to 67), but only 6% (95% CI: –4 to 16) against all episodes of otitis media regardless of etiology.1 The vaccine efficacy (VE) against tympanostomy tube placements was 4% (–19 to 23) during the FinOM vaccine trial follow-up up to 24 months of age with high cumulative incidence of 19% in the control group, but from 2 to 4–5 years of age the VE against tympanostomy tube placements was 39% (95% CI: 4 to 61) with cumulative incidence of 9.5% in the control group.6
In the Northern California Kaiser Permanente trial7 nearly 38,000 children were randomized to receive either the 7-valent PncCRM or control vaccine. The VE against all otitis media episodes was 7% (95% CI: 4 to 10) and against tympanostomy tube placements 24% (95% CI: 12 to 35).8 The cumulative incidence of tube placements was 3.9% by 3.5 years of age in the control group.8 Another pneumococcal conjugate vaccine (PCV) the 11-valent PnPD (GlaxoSmithKline Vaccines, Rixensart, Belgium) reduced tympanostomy tube placement by 60% (–27 to 88), but the number of procedures was low resulting in wide CIs.9
The objective of this study was to assess the effects of 2 7-valent PCVs PncCRM and PncOMPC from 2 years up to 13 years of age vaccinated in the FinOM trial setting. No long-term evaluation of effects of PCVs on tympanostomy tube procedures have been published so far.
Altogether 2497 children having participated in the randomized, double-blind FinOM vaccine trial in 1995 to 1999 were identified from the population information system for this study. These children had been vaccinated at 2, 4, 6 and 12 months of age with PncCRM (831 children) or PncOMPC (835 children) or hepatitis B vaccine (831 children) as controls.1,10 In the PncOMPC group, 645 received the same conjugate vaccine as a booster at the age of 12 months and 190 children enrolled later during the trial received a Pneumovax booster dose instead of the conjugate vaccine.
Data on tympanostomy tube placement procedures for these children were collected from the Finnish Care register including hospital operations and private office operations from 1998 through end of 2008. The Finnish Social Insurance Institution data were also collected to obtain complete data for private office operations for the same time period. The current register study protocol was approved by the Ethics committee of the National Institute for Health and Welfare.
Tympanostomy tube placement was defined as a placement of a tube in at least 1 ear or replacement of an extruded or occluded tube by a new one. Bilateral tube placements were considered as 1 event if the tubes were placed at the same operation. Recurrent events were included in the primary analyses. The follow-up period started when the children turned 24 months and ended on December 31, 2008.
VE was estimated by comparing the risk of tympanostomy tube placement in PncCRM and PncOMPC groups to the control vaccine group using the Andersen-Gill generalization of the Cox proportional hazards model,11 allowing for inclusion of recurrent tube placement episodes (when applicable) into the analysis. The dependence due to recurrence was accounted for by using the standard error for the treatment parameter based on the sandwich estimator.12 VE was also estimated separately for the 2 PncOMPC subgroups with respect to the booster received at the age of 12 months. Due to different time periods of enrolment for these subgroups, only data on parallel control group subjects were used for these analyses.
Because the treatment effects proved to be different for children <6 years compared with those older (Fig. 1), the follow-up period was split into 2 to fulfill the assumption of proportionality of the hazards for the Cox regression model. Results are reported as estimates of VE with 95% CIs.
Register data were available for altogether 2474 (99%) of FinOM trial participants, 322 of which went through altogether 535 tube placements. Figure 1 shows the numbers of these procedures for each vaccine group by 12-month age intervals.
The vaccine effectiveness point estimates between the age interval from 24 months through 5 years of age were positive for both conjugate vaccines, but considerably higher and statistically significant for the PncCRM vaccine (Table 1). The absolute reduction of tympanostomy tube placements in the PncCRM group compared with the control was 1.77 per 100 person-years.
After 5 years of age, the rate of tympanostomy tube placements was 0.56 in the control group, 0.63 in the PncCRM group and 0.57 per 100 person-years in the PncOMPC group (with conjugate and Pneumovax boosters the rates were 0.49 and 0.89 per 100 person-years, respectively), thus giving a –13% (–137 to 46) VE for PncCRM and –2% (–123 to 54) for the PncOMPC vaccine.
The cumulative hazard of all tympanostomy tube placements in Figure 2 shows the impact of the PncCRM vaccine before 6 years of age, which sustains up to 13 years of age.
The cumulative incidence of the tympanostomy tube placements from 2 through 5 years of age was 14.4% (14.6% until 12–13 years of age) in the control group compared with 10.3% (11.6%) in the PncCRM and 12.0% (12.9%) in the PncOMPC group. In the analysis of first events only for the period from 2 through 5 years of age, the VE estimates were 26% (6 to 46) and 17% (–8 to 37) for the PncCRM and PncOMPC vaccines, respectively.
Receiving the PncCRM vaccine in infancy reduced the risk of insertion of tympanostomy tubes by 34% after the age of 2 through 5 years of age. The cumulative incidence reduction was 4 per 100 subjects in our study with high utilization of tympanostomy tube procedures. From 6 years onwards the incidence of tube surgery fell dramatically and we were not able to demonstrate further vaccine impact during this age period. The effectiveness estimates for the PncOMPC vaccine were lower and did not reach statistical significance for any age period.
Our findings are compatible with the long-term immunogenicity data assessed 3 to 4 years after vaccination.13 The pneumococcal antipolysaccharide antibodies were higher in PncCRM-vaccinated children of the current FinOM trial at the age of 4 to 5 years compared with controls. Despite increased levels of antipolysaccharide antibodies in control children due to natural exposure, there was a clear difference to the PncCRM-vaccinated children, who also showed increasing levels of antibodies with advancing age.
These data confirm our previous findings of the effectiveness of the PncCRM vaccine in the prevention of tympanostomy tube placements between the age interval from 2 to 4–5 years of age.6 We consider the current results more reliable as the previous data included only a subset of subjects; the current study included nearly all subjects of the original FinOM trial. Furthermore, the surgical procedure data in the current study were collected from registers including all the hospitals in the country while the previous study included only the local hospitals. Despite these differences, the incidence estimates in the control and PncCRM groups were nearly identical with similar vaccine effectiveness estimates; 34% in the current analysis compared with 39% in the previous one. We are not aware of any other long-term studies of PCV on tympanostomy tube placement although ecological short-term studies after PCV7 introduction have been reported from United States14 and Australia.15 They report effectiveness of the vaccination programs with slightly lower point estimates for children <2 years of age compared with our current results.
No long-term data on PncOMPC have been published before. This vaccine did not enter into licensure in any country. The effectiveness estimates were lower for the PncOMPC group compared with PncCRM. One fourth of the PncOMPC group children received a Pneumovax booster dose instead of the PncOMPC at the age of 12 months, but we were not able to show any impact of the different booster.
As otitis media is especially a disease of young children, the risk for otitis-related surgery also decreases rapidly with advancing age. In this study, the risk of tympanostomy tube placement declined sharply after 5 years of age. In these older children, 80% of the tube insertions were recurrent episodes on children with previous tube insertion after 24 months of age. Thus, these subjects were a special otitis-prone subpopulation. Additionally, the low number of events in this age period resulted in inadequate statistical power to make firm conclusions.
In 2005, 123 of 10,000 children younger than the age of 8 had tympanostomy tube placement in Finland.5 The proportion of children in each year of age with tympanostomy tube placement was roughly 2 times higher in our control group compared with the national data and international data reported.16 There is huge variation in the procedure rates both internationally and nationally. Furthermore, the active use of tympanostomy tube surgery during the trial phase may have lowered the threshold also for later surgery in the current study population.
The data for the tympanostomy tube placements were obtained from registers but were not verified from medical records. We expect that the hospital records are valid as all hospitals have to report all performed operations to registers and thus are expected to have made it systematically and correctly. Even if there would be some gaps in collecting all procedures, all data were obtained consistently and symmetrically for both original randomized treatments groups.
We were able to identify altogether 2474 of the 2497 subjects who were initially enrolled to the FinOM vaccine trial at the age of 2 months. The original FinOM trial study population comprised 57% of the total source population living in the study area at the time of enrolment. Thus, we consider the study population representative of the source population of young infants in western countries with no major selection bias in the context of this study. Confounding factors should not introduce bias into the results because the FinOM trial was a double-blinded randomized clinical trial. The vaccination status was unblinded both to parents and the investigators in fall 1999, but in our earlier analysis, the effectiveness estimate until unblinding in September 1999 was highly similar to the estimate obtained for the whole follow-up period until June 2001.6
Our study results estimate the direct effectiveness of the vaccines because the PCVs were only used in the trial context during the original FinOM study period. The 7-valent PncCRM was the first PCV licensed in Europe in 2001, but PCV use remained low until the 10-valent PCV was introduced into the Finnish national vaccination program for infants in fall 2010. The public health impact of the PCV in prevention of tympanostomy tube placement may be higher in the context of high-coverage vaccination programs with ensuing indirect protection.
In conclusion, the PncCRM vaccine given in infancy conferred long-term protection against new tympanostomy tube placements that lasted up to the age of 5 years and the benefit achieved by that age persisted up to 13 years of age. Since most of the disease burden of complicated otitis media occurs in young children under the school-age, significant impact can be accomplished with pneumococcal vaccination.
1. Eskola J, Kilpi T, Palmu A, et al.Finnish Otitis Media Study Group. Efficacy of a pneumococcal conjugate vaccine against acute otitis media. N Engl J Med. 2001;344:403–409
2. Paradise JL, Rockette HE, Colborn DK, et al. Otitis media in 2253 Pittsburgh-area infants: prevalence and risk factors during the first two years of life. Pediatrics. 1997;99:318–333
3. Centers for Disease Control and Prevention.. Office-related antibiotic prescribing for persons aged < 14 years–United States, 1993–1994 to 2007–2008. MMWR Morb Mortal Wkly Rep. 2011;34:1153–1156
4. Owings MF, Kozak LJ. Ambulatory and impatient procedures in the United States, 1996. Vital Health Stat 13. 1998;139:1–119
5. Haapkylä J, Karevold G, Kvaerner KJ, et al. Trends in otitis media surgery: a decrease in adenoidectomy. Int J Pediatr Otorhinolaryngol. 2008;72:1207–1213
6. Palmu AA, Verho J, Jokinen J, et al. The seven-valent pneumococcal conjugate vaccine reduces tympanostomy tube placement in children. Pediatr Infect Dis J. 2004;23:732–738
7. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity of heptavalent pneumococcal conjugate vaccine in children. Northern California Kaiser Permanente Vaccine Study Center Group. Pediatr Infect Dis J. 2000;19:187–195
8. Fireman B, Black SB, Shinefield HR, et al. Impact of the pneumococcal conjugate vaccine on otitis media. Pediatr Infect Dis J. 2003;22:10–16
9. Prymula R, Peeters P, Chrobok V, et al. Pneumococcal capsular polysaccharides conjugated to protein D for prevention of acute otitis media caused by both Streptococcus pneumoniae and non-typable Haemophilus influenzae: a randomised double-blind efficacy study. Lancet. 2006;367:740–748
10. Kilpi T, Ahman H, Jokinen J, et al.Finnish Otitis Media Study Group. Protective efficacy of a second pneumococcal conjugate vaccine against pneumococcal acute otitis media in infants and children: randomized, controlled trial of a 7-valent pneumococcal polysaccharide-meningococcal outer membrane protein complex conjugate vaccine in 1666 children. Clin Infect Dis. 2003;37:1155–1164
11. Andersen PK, Gill RD. Cox’s regression model for counting processes. A large sample study. Ann Stat. 1982;10:1100–1120
12. Lin DY, Wei LJ. The robust inference for the Cox proportional hazards model. J Am Stat Ass. 1989;82:1075–1078
13. Åhman H, Palmu A, Kilpi T, et al.Finnish Otitis Media (FinOM) Study Group. . Persistence of anti-pneumococcal antibodies 3–4 years after immunization with 7-valent pneumococcal conjugate vaccine PncCRM in infancy. ISPPD-2002, the 3rd International Symposium on Pneumococci and Pneumococcal Diseases. 2002 Anchorage, Alaska
14. Poehling KA, Szilagyi PG, Grijalva CG, et al. Reduction of frequent otitis media and pressure-equalizing tube insertions in children after introduction of pneumococcal conjugate vaccine. Pediatrics. 2007;119:707–715
15. Jardine A, Menzies RI, Deeks SL, et al. The impact of pneumococcal conjugate vaccine on rates of myringotomy with ventilation tube insertion in Australia. Pediatr Infect Dis J. 2009;28:761–765
16. Desai SN, Kellner JD, Drummond D. Population-based, age-specific myringotomy with tympanostomy tube insertion rates in Calgary, Canada. Pediatr Infect Dis J. 2002;21:348–350