Reemergence of Pertussis in the Highly Vaccinated Population of The Netherlands: Observations on Surveillance Data

Hester E. de Melker, J.F.P. Schellekens, S.E. Neppelenbroek, F.R. Mooi, H.C. Rümke, M.A.E. Conyn-van Spaendonck, National Institute of Public Health and the Environment, Bilthoven, the Netherlands

Disclosures

Emerging Infectious Diseases. 2000;6(4) 

In This Article

Results

In the first years of mandatory reporting of pertussis cases, the case count was the lowest (Table 1). From 1983 to 1987, after immunoassays for pertussis serology became available, the number of reported cases increased yearly. In 1988, the year in which a case definition was introduced and positive serology was restricted to an increase in titer in paired sera, the number of reported cases declined sharply. Somewhat greater numbers of cases were reported in 1989-90 and 1993-94. In 1996, the number of cases was 12 times higher than in 1995, while a twofold decrease from the 1996 number was observed in 1997.

The number of reported cases from January to September 1998 (n = 1,582) was lower than that of the same periods in 1996 (n = 2,171) and 1997 (n = 2,004) but approximately six times higher than the average number of reported cases in the same months of 1989 to 1995 (n = 269).

The trend of hospitalizations was similar to that of case reports; however, the ratios varied by period (Table 1). This ratio was below one from 1976 to 1984, increased to 5.7 in 1987, and decreased sharply to 1.2 in 1988; it remained relatively stable from 1989 to 1995 but increased to 8.2 in 1996 and 6.1 in 1997.

From 1976 to 1997, seven deaths caused by pertussis were reported: one in 1981, two in 1993, two in 1996, and two in 1997. They occurred among children <1 year of age, except for one death in 1993, which occurred in the 5- to 9-year age-group. According to the number of hospitalizations among children <1 year of age, the case-fatality rates amounted to 0.1% on average in 1989 to 1995, 0.6% in 1996, and 0.7% in 1997.

The cases with positive one-point serology followed a trend similar to that of cases reported in 1986 to 1987 and 1989 to 1998 (Table 1). In 1986 and 1987, the ratio of reported cases to cases with positive one-point serology was highest (0.8); it decreased in 1989, and remained relatively stable from 1989 to 1996 (0.2 to 0.4). In 1996 to 1998, this proportion increased from 0.5 to 0.7.

The trends of cases with positive two-point serology and reported cases were similar (Table 1). From 1989 to 1995, the number of cases in each database was similar (ratio 0.7 to 1.5); in 1996 and 1997 (ratio 2.2 and 2.9) and particularly in 1998 (ratio 5.4), the number of reported cases was higher than the number of cases with positive two-point serology.

In 1989 to 1995, the average annual incidence from case reporting was highest for infants <1 year of age. Such data for 1993 to 1995 show that the age-specific peak incidence occurred among <=5-month-old infants. For infants <1 year of age, the incidence in 1996 was four times higher than the average incidence from 1989 to 1995 and 13 times higher than the incidence for older age-groups (Figure 1). The age-specific peak incidence shifted to 4-year-old children in 1996 and 1997.

Figure 1. Average annual age-specific incidence (number per 100,000) in 1989 through 1995, in 1996, and in 1997.

From 1993 to 1997, when the method of laboratory diagnosis was available for reported cases, similar shifts in age distribution were observed for cases confirmed by microbiologic method (i.e., culture or PCR), cases with positive two-point serology, and cases with positive one-point serology. However, reported cases confirmed microbiologically were those of the youngest patients; cases confirmed with one-point serology were those of the oldest patients. In contrast to reported cases, the greatest age-specific incidence of hospitalizations occurred among infants <1 year of age from 1989 to 1997 (Figure 1). The increase in incidence in 1996 was more stable for the various age-groups.

As in case reports, because of a relatively greater increase of pertussis among older age-groups, the peak incidence among cases with positive two-point serology occurred among <=5-month-old infants from 1989 to 1995 and among 4-year-old children in 1996 and 1997. The incidence of cases with positive one-point serology was greatest among 7-year-old children in 1989 and shifted towards 4-year-old children from 1992 to 1997. As with case reports, the increase for infants <1 year old was smaller (fourfold) than that for older age-groups (fivefold to sevenfold) (Figure 1).

In March and April 1996, reported cases started to increase (Figure 2). The largest monthly number of cases occurred later in the year (October 1996) than in 1989-1995 and 1997-1998 (mostly in August, sometimes in July or September). The seasonal trend of positive two-point serology and positive one-point serology was similar to that of case reports (Figure 2).

Figure 2. Reported pertussis cases per month, 1989-1998, cases with positive two-point serology, and cases with positive one-point serology.

According to case reports, vaccine effectiveness was high in 1981 to 1984 (1 to 4 years of age: 94% to 99%; 5 to 9 years of age: 87% to 100%) and in 1988 to 1993 (1 to 4 years of age: 89% to 95%; 5 to 9 years of age: 78% to 89%). The estimates were somewhat lower in 1985 to 1987 (72% to 85% for children 1 to 4 years of age and 56% to 77% for children 5 to 9 years of age). The estimates decreased after 1993 (Table 2), were lowest in 1996, and could not be determined in 1997 since the proportion of vaccinated patients exceeded 96%.

Vaccine effectiveness for almost all methods of diagnosis was greater in 1993 than in 1994 to 1997 (Table 2). The decreasing trend was not consistent with all methods of diagnosis. Estimates for cases diagnosed microbiologically tended to be the highest; estimates for cases confirmed by one-point serology tended to be the lowest.

Linkage of the case-report and serodiagnosis databases for 1993 to 1997 showed that the proportion of reported cases confirmed microbiologically, epidemiologically, or by two-point serology, decreased, while the proportion of cases confirmed with positive one-point serology increased (Figure 3). The proportion of cases confirmed serologically but not matched with the serologic database also increased. Differentiating these cases by positive one-point and positive two-point serology was not possible. During 1996 and 1997, the proportion of cases confirmed microbiologically by quarter year was similar (8.7% to 10.2%), except for a smaller proportion (6.4%) in the last quarter of 1997. The proportion of cases confirmed by two-point serology was 12% to 19%. In the first two quarters of 1996 (34.9% to 35.6%) and in the fourth quarter of 1997 (34.3%), the proportion of cases confirmed by positive one-point serology was similar to the proportions for 1994 and 1995. The numbers increased to >50% in the fourth quarter of 1996 and the first quarter of 1997. The proportion of cases confirmed serologically that could not be matched with the serodiagnostic database was highest (35.3%) in the fourth quarter of 1997.

Figure 3. Method of diagnosis for reported pertussis cases from 1993 through 1997.

The reported proportion of cases with positive two-point serology increased from 24% in 1993 to 26% in 1994, 28% in 1995, and 42%-43% in 1996 and 1997. A similar trend was observed for the various age-groups. For positive one-point serology, the reported proportion increased from 6% in 1993, to 10% in 1994, 12% in 1995, and 27% to 29% in 1996 and 1997. The discrepancy was somewhat greater with increasing age. The increase in the reported proportions of cases with positive two-point and one-point serology was probably underestimated because the database of reported cases contained serologically confirmed cases that could not be matched with the serodiagnosis database. This proportion was highest in 1997 (Figure 3).

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....