Secondary Logo

Journal Logo

Original Studies

Invasive Infections Caused by Haemophilus influenzae Serotypes in Twelve Canadian IMPACT Centers, 1996–2001

McConnell, Athena MD*; Tan, Ben MD; Scheifele, David MD; Halperin, Scott MD; Vaudry, Wendy MD§; Law, Barbara MD∥¶; Embree, Joanne MD#∥ of The Canadian Immunization Monitoring Program, ACTive (IMPACT)

Author Information
The Pediatric Infectious Disease Journal: November 2007 - Volume 26 - Issue 11 - p 1025-1031
doi: 10.1097/INF.0b013e31812f4f5b

Abstract

ArticlePlus

Click on the links below to access all the ArticlePlus for this article.

Please note that ArticlePlus files may launch a viewer application outside of your web browser.

Before the availability of Haemophilus influenzae type b (Hib) vaccine, Hib was responsible for the majority of invasive infections caused by this bacterial species.1,2 The incidence of Hib invasive disease has declined significantly in countries with routine infant Hib immunization.3–14 A study from Brazil reported an increase in incidence of H. influenzae type a (Hia) meningitis 1 year after Hib immunization was introduced.15 However, surveillance in the United States, England, and Switzerland did not show an increase, but suggested that there is an ongoing background rate of disease caused by other serotypes, which has gained prominence as a result of the decline in Hib disease.5,14,16–18 In Canada, Hib polysaccharide [polyribosylribitol phosphate (PRP)] vaccine was approved in 1986 for 2-year-old children, and Hib conjugate vaccine later recommended for 18-month-old children in 1988. By 1992, all provinces and territories implemented Hib conjugate immunization in infancy, with 3 primary doses at 2, 4, and 6 months of age, followed by a booster at 18 months.

The Immunization Monitoring Program, ACTive (IMPACT) consists of 12 Canadian pediatric tertiary care centers, which draw referrals from all 10 Canadian provinces and 3 northern territories.19 The IMPACT centers manage about 75,000 inpatient admissions annually and encompass nearly 90% of tertiary care pediatric beds in Canada, serving a population base of 3 million children. Hib case reporting was initiated at 5 centers in 1992, expanding to 10 centers in 1993. The 11th center contributed cases from January 1994 and the 12th center from April 1999.4 This retrospective study was conducted in 2004 to describe the epidemiology of invasive disease caused by Hib and other H. influenzae serotypes in the 12 IMPACT centers from 1996 to 2001, after the introduction of infant Hib immunization in Canada in 1992.

MATERIALS AND METHODS

Children who had H. influenzae isolated from normally sterile body fluids (ie, blood, cerebrospinal fluid (CSF), pleural fluid, and joint aspirate) between January 1, 1996 and December 31, 2001 were identified from laboratory records at 12 IMPACT centers. The hospital charts were retrospectively reviewed for the following information: age at presentation, sex, ethnic origin of the parents, place of residence, underlying medical conditions, prior Hib vaccination, site(s) of infection, duration of hospitalization, duration in a pediatric intensive care unit (PICU), H. influenzae serotype and antibiotic susceptibility, outcome at hospital discharge, and whether there was any epidemiological link with another household or community case caused by the same serotype within 60 days before admission. Follow-up beyond the hospital stay was not performed.

Isolates were identified as H. influenzae on the basis of colony morphology, appearance on a Gram stain, and growth on chocolate agar supplemented with factors X and V.20 With the exception of screening for Hib, other H. influenzae serotypes were not routinely identified in the IMPACT centers before 1996. From 1996 onward, blood and CSF isolates were fully serotyped by the IMPACT hospital laboratory or the provincial laboratory, initially using latex agglutination tests to screen for Hib species (3 centers), and/or seroagglutination, coagglutination, or polymerase chain reaction tests (all centers) to identify the non-b serotypes (Table 1).21–24 Complete serotyping of isolates recovered from specimens other than blood or CSF was routinely performed at 2 centers, and only upon physician request at the remaining 10 centers.

TABLE 1
TABLE 1:
Haemophilus influenzae Serotypes Identified in 12 IMPACT Centers, 1996–2001

The following case definitions were used in this study: pneumonia was defined as a respiratory illness with pulmonary infiltrate(s) on a chest radiograph in association with a positive culture from either blood or pleural fluid. Positive sputum or endotracheal cultures alone were not accepted as evidence of pneumonia. A diagnosis of meningitis required a positive CSF culture, or positive blood culture in association with CSF pleocytosis (elevated CSF white blood cell count beyond the normal range specified by the reporting laboratory). Cellulitis was defined as an extending area of skin erythema associated with bacteremia. A septic joint was diagnosed on the basis of joint pain and swelling, with positive cultures of synovial fluid and/or blood. A diagnosis of osteomyelitis was accepted if there was localized pain or associated systemic symptoms with positive imaging studies (radiograph, bone scan, or magnetic resonance imaging) and positive blood or bone specimen cultures. Epiglottitis was defined as the characteristic inflammation and enlargement of the epiglottis (on lateral radiograph of the neck or laryngoscopic examination) with associated bacteremia. H. influenzae species was accepted as the etiological agent causing otitis media if there were compatible tympanic membrane findings associated with a positive blood culture.

Persistent problems requiring regular medical attention formed the basis for defining underlying conditions by systems. Neurologic conditions also included genetic disorders with a predominant or exclusively neurologic component, as well as benign febrile seizures and developmental delay. Effects of injury captured the longer-term sequelae of trauma needing special attention, including CSF leakage after head injury, burns, chronic infection after fractures, mobility impairment, and organ loss (lung or spleen removal). Acute concurrent infection was used to code for situations in which the onset of H. influenzae infection coincided with a concurrent acute infection (such as varicella or respiratory syncytial virus).

With respect to Hib immunization, vaccine failures were defined as cases in which patients had received at least 3 age-appropriate doses of the vaccine in those at least 6 months of age, and/or 1 dose at 15–59 months of age. Incomplete immunization referred to patients who did not satisfy these criteria.

Data entry was centralized at the Data Center, British Columbia (BC) Children’s Hospital, Vancouver, British Columbia. Data analysis was performed using SAS version 8. Disease incidence for Hia infection was calculated using population figures obtained from Statistics Canada, 2001 census. This study was approved by the research ethics boards of each IMPACT center.

RESULTS

For the study period, there were 166 cases with at least 1 positive culture for H. influenzae. These cases represented 166 individuals as no recurrences were identified. The serotypes identified at each IMPACT center are shown in Table 1. The isolates from 19 (11%) cases were not serotyped. Of the remaining 147 cases, 58 (39%) were caused by Hib, 25 (23%) by Hia, 11 (10%) by H. influenzae type f (Hif), and 47 (44%) by nontypable H. influenzae. There were 4 cases caused by H. influenzae type d (Hid) (4%), 2 by H. influenzae type e (Hie) (1.9%), and none attributed to H. influenzae type c during the study period.

Figure 1 shows the year-to-year trend for Hib cases, non-b cases, as well as the cases caused by unknown serotypes. For each year, the non-b isolates outnumbered Hib isolates, with an average of 14.8 non-b cases/yr (range, 12–19 cases/yr) compared with 9.7 Hib cases/yr (range, 5–16 cases/yr). The number of unknown isolates also increased from 1997 to 2001, with 2 centers contributing 12 of the 19 unknown isolates (Table 1). The reasons for the unknowns included: serotyping of isolates from specimens other than blood or CSF was not performed, missing isolates, and serotype reports not received from the provincial (reference) laboratory.

FIGURE 1.
FIGURE 1.:
Graph showing year-to-year trend for Hib cases, non-b cases, as well as cases caused by unknown serotypes.

Table 2 (online only) compares the demographic and clinical information for cases caused by the Hib, Hia, Hif, and nontypable isolates. Of the H. influenzae non-b cases that were fully typed, 38 (42.7%) occurred in patients that were healthy before admission. Preexisting conditions included neurologic (n = 9, 10.1%), gastrointestinal (n = 8, 9.0%), neoplasm either in the past or present (n = 7, 7.9%), prematurity (n = 7, 7.9%), concurrent acute infection (n = 6, 6.7%), and cardiovascular disorders (n = 5, 5.6%). All 7 premature infants were born at gestational ages between 24 and 32 weeks of age.

H. influenzae Type b.

Of the 58 Hib cases, there were 34 with meningitis, 11 with epiglottitis, 10 with pneumonia, 5 with empyema or pleural effusion, 4 with septic shock, and 4 with isolated bacteremia. The median age of Hib cases was 1.2 years, with 24 cases younger than 12 months of age, 22 cases between 1 and 4 years of age, and 12 who were 5–12 years of age. Of the 12 patients older than 5 years of age, 8 had epiglotittis, 9 had received age-appropriate Hib vaccination, and only 1 was considered immunocompromised (nephrotic syndrome). No epidemiologic link to any other Hib case in the previous 60 days to hospitalization were identified in any of the individuals, and there were no temporal or spatial clusters.

The majority of patients with Hib infection, 50 (86%), had no underlying medical illnesses before the infection. Only 2 patients were considered immunocompromised at the time of the infection. Thirty (52%) patients were Caucasian, 11 (19%) were aboriginal, and ethnicity was not documented for 13 (22%) patients. The mean duration of hospitalization for Hib cases was 11 days. Thirty-one (53%) required admission to the PICU for a mean duration of 4.5 days. The case fatality rate for Hib cases was 3.2%. At hospital discharge, 78% of the patients had fully recovered or were clinically improved. Among survivors of invasive Hib disease, hearing impairment occurred in 6 (11%), neurologic deficits in 4 (7%), and 3 (5.2%) continued to need anticonvulsant therapy at discharge.

Twenty-one (36%) Hib cases were considered vaccine failures (14 had received PRP-tetanus toxoid conjugate vaccine and 7 received PRP-diphtheria toxoid or PRP polysaccharide vaccines). Twenty-one children (36%) had not been vaccinated, 9 because the parents specifically declined immunization and 3 were too young for vaccination. The remaining 16 (28%) of 58 patients were incompletely immunized against Hib. Thirty-one of 53 (58%) Hib isolates tested were susceptible to ampicillin.

H. influenzae Type a.

Hia was the second most common typable strain after Hib, with 25 cases. There were 13 patients with meningitis, 5 with pneumonia, 2 with septic shock, 2 with isolated bacteremia, and none with epiglottitis or empyema. The median age of Hia cases was 8.5 months, with 20 (80%) less than 1 year of age. Twenty-one (84%) had received Hib immunization. Sixteen (67%) cases had no underlying medical illnesses before the infection. Patients with Hia infection were hospitalized for a mean of 11.7 days, and 10 (40%) required PICU care for a mean duration of 2.1 day. The case fatality rate for Hia cases was 4 of 25 (16%). Two of the 4 patients who died of Hia infection had Down’s syndrome. At hospital discharge, 60% of the patients had fully recovered or were expected to return to baseline status. Among the 21 survivors, 3 (12%) patients had hearing impairment, and 3 (12%) patients had neurologic impairment at hospital discharge. Twenty-one of 23 (91%) Hia isolates tested were ampicillin-susceptible, and all 11 isolates tested were susceptible to cefotaxime and ceftriaxone.

Twenty-four patients (96%) were admitted to IMPACT hospitals in the 4 western Canadian provinces, ie, Manitoba (12 cases), Saskatchewan (6 cases), Alberta (5 cases), and BC (1 case). Only 1 Hia case was reported from an eastern province (Quebec). Nineteen of the Hia cases (76%) were aboriginal children, defined as having at least 1 parent of First Nations, Metis, or Inuit heritage. Fourteen were First Nations or Metis children who were residents of Manitoba, Saskatchewan, Alberta, and BC. There were 5 Inuit children, one of whom was a resident of the Northwest Territories who was transferred to the IMPACT center in Edmonton, Alberta. Four Inuit patients from the Keewatin Region of Nunavut Territory were transferred to the IMPACT center in Winnipeg, Manitoba, all occurring in 2001. Two of the 4 cases originated from the same community and were admitted within 60 days of each other, in June and August 2001. Other than being immunized in the same clinic in April 2001, no epidemiological link between the cases could be ascertained through review of the chart.

To enable comparison with other reported rates, the Hia disease incidence in aboriginal children was estimated using 2001 census data for Keewatin Region and the 4 western provinces. The population of Inuit children less than 5 years of age in Keewatin Region in 2001 was approximately 955, and the population of First Nations and Metis children less than 5 years of age in the 4 western provinces was approximately 60, 400 (making up roughly 13.8% of the population of less than 5 years in the 4 provinces). As no Hia cases occurred in 1996–2000, the Hia disease incidence for Inuit children in Keewatin Region was only calculated for 2001 and was 418.8 of 100,000 less than 5 years of age. The mean incidence for children of First Nations and Metis heritage residing in the 4 western provinces from 1996 to 2001 was 3.7 of 100,000 children less than 5 years of age (95% CI: 1.1–8.51). In contrast, there was 1 Caucasian patient with Hia infection in the 4 western provinces, giving a possible maximum Hia disease incidence of 2.3 of 1,000,000 children less than 5 years of age in nonaboriginal children.

H. influenzae Type f.

Of the 11 Hif cases, there were 4 with meningitis, 4 with pneumonia, 2 with isolated bacteremia, 1 with septic shock, and 1 with epiglottitis. The median age of cases was 1.8 years of age, with 8 (73%) cases more than 1 year of age. Seven (64%) had received Hib immunization. The ethnicity of 5 cases (45%) was not documented and of the remainder, 3 were Caucasian and 1 was an aboriginal child. Mean stay in hospital was 7.6 days, with 4 (36%) patients requiring admission to the PICU for a mean of 2 days. All patients had recovered fully or were clinically improved by hospital discharge. There were no fatalities attributed to Hif. All tested isolates were susceptible to ampicillin, cefotaxime, or ceftriaxone.

H. influenzae Type d.

There were 4 cases of invasive Hid disease, 2 with pneumonia, and 1 each with empyema, shock, and meningitis. The median age of Hid cases was 1.1 year; 3 (75%) occurring in children more than 2 years of age. Aboriginal descent was recorded for 1 case, with ethnicity data not available for the remainder. Immunization against Hib was verified for 3 patients. Patients with invasive Hid disease were hospitalized for a mean of 11.8 days, with 1 patient requiring PICU admission of 7 days. There were no fatalities among Hid cases, and no impairments caused by Hid disease at the time of discharge. There was 1 case with ampicillin resistance, but tested isolates remained susceptible to cefotaxime and ceftriaxone.

H. influenzae Type e.

Two patients were admitted with Hie disease, both with meningitis. The average age was 1.2 years. One case was Caucasian, but the ethnicity for the other patient was not known. Immunization against Hib either did not occur (n = 1) or could not be verified (n = 1). Hospitalization was for a mean of 12.5 days, and 1 patient was transferred from PICU to the ward after 1 day. Both patients were fully recovered at the time of discharge. Tested isolates were susceptible to cefotaxime and ceftriaxone, but ampicillin resistance was present in both Hie isolates recovered.

Nontypable H. influenzae.

Nontypable H. influenzae accounted for 47 cases, including 20 with pneumonia, 2 with empyema or pleural effusion, 8 with meningitis, 8 with isolated bacteremia, 5 with septic shock, and 4 with associated otitis media. Nontypable serotypes were not responsible for any epiglottitis cases. The median age for nontypable cases was 2 years, with 23 (48%) between 1 and 4 years of age and 12 (26%) were 5–18 years of age. Fifteen (32%) patients had received Hib immunization, but the Hib immunization status of 23 patients (49%) was unknown. Ethnicity data was not available for 24 (51%) of these cases, with an equal number (8 each) of Caucasian and aboriginal children. Unlike patients with Hib or Hia infections, only 16 (34%) patients were considered healthy (no associated medical illnesses) before the infection.

Patients with nontypable H. influenzae infections had a mean hospital stay of 13.2 days. Fifteen (32%) patients required admission to the PICU for a mean of 6.1 days. The case fatality rate for nontypable cases was 2.1%, with another 2 (4.3%) patients dying of causes other than nontypable H. influenzae disease. At hospital discharge, 36 (76.6%) of cases were fully recovered or returning to baseline status. Of the remainder, 3 (5.7%) were on anticonvulsants, 3 (5.7%) had hearing loss, and 4 (7.5%) had neurologic abnormalities. One case was treated as an outpatient. Thirty-three of 44 (75%) nontypable serotypes were susceptible to ampicillin and all isolates tested were susceptible to cefotaxime or ceftriaxone.

DISCUSSION

In 1985, the year before the licensure of the first polysaccharide Hib vaccine in Canada, a total of 485 invasive Hib cases were reported by 10 IMPACT centers.9 A population-based survey of meningitis before the availability of Hib vaccine in Manitoba and the Keewatin Region of Nunavut (which at the time was a part of the Northwest Territories) revealed 81 H. influenzae cases, with all but one being Hib.25 Since 1985 there has been a dramatic reduction in the number of Hib invasive disease cases,4,6–10 with our present study confirming the continued low rate of occurrence of Hib disease. From 1996 to 2001, Hib invasive disease averaged 9.7 Hib cases per year in the 12 IMPACT centers. The Hib case total in 2001 (16 cases) was unusually high, with numbers of invasive Hib disease declining in subsequent years (data not presented). Examination of the 16 cases revealed that Aboriginal individuals were overrepresented as opposed to any other year since surveillance was initiated (6 cases). Of the 6 cases, 4 occurred in patients too young to have completed the primary series of vaccinations. Other than ethnicity, no other risk factors could be identified for the increase in Hib disease.

Although surveillance for Hib disease has been and continues to be a topic of interest, the epidemiology of invasive infections caused by non-b serotypes has not been well studied in Canada. As our study demonstrates, invasive disease in children caused by non-b serotypes has increased proportionately and was responsible for 60% of invasive infections caused by H. influenzae species. Tsang et al’s26 recent study of H. influenzae in Manitoba, Canada over a 7-year period has also revealed the increased relative importance of non-Hib serotypes causing invasive disease in all age groups.

There have been relatively few published studies on the impact of Hib vaccination on the emergence of non-b isolates as invasive pathogens in children. A prospective study in England showed that the number of invasive infections attributed to nontypable species increased from 36 to 83 cases after initiation of an infant immunization program.11 However, the incidence of nontypable invasive infections rose significantly only in those older than 65 years of age. There was no increase in invasive infections because of Hie or Hif, which were the only other non-b isolates reported.11

Ribeiro et al15 conducted a prospective study on the incidence of H. influenzae meningitis in 1 region of Brazil from March 1996 to September 2000, after Hib conjugate vaccine was introduced for routine childhood immunization in August 1999. Concurrent with a decline in incidence of Hib meningitis, the incidence of Hia meningitis in Brazil rose from a prevaccine rate of 0.02 of 100,000 person-years to 0.16 of 100,000 person-years, 1 year after the introduction of Hib vaccination. Although this was only based on a single year trend, the investigators speculated that serotype replacement may have taken place in this region of Brazil.15 The Hia cases in this region of Brazil had similar PICU admission rates (23% versus 21%, respectively) and case fatality rates (23% versus 16%, respectively) when compared with the Hib cases.15

Investigators in Utah reported 5 invasive Hia cases with a virulence pattern similar to cases caused by Hib.27 The ethnicity of patients was not specified. Four of the Hia patients had meningitis and bacteremia. Although none of these patients succumbed, the patients with Hia meningitis had a complicated and prolonged stay in hospital. Two survivors were left with hearing loss and developmental delay. Three Hia isolates were found to contain the IS1016-bexA deletion, which had been described in Hib isolates, previously postulated to be a possible mechanism for increased Hib virulence by leading to an increase in capsule production.28,29

Hia invasive infections have been reported among Native American children in 2 studies. Hammitt et al30 reported an outbreak of 5 Hia infections which occurred in 3 Alaskan native infants in 2 remote villages, whereas Millar et al14 reported 76 Hia invasive infections cases occurring among Navajo and White Mountain Apache children from 1988 to 2003. In the latter study, Hia cases outnumbered Hib cases since 1995. The mortality rate from Hia infection was 2.6%, and 13% of Hia meningitis survivors developed hearing loss. The Hia incidence in the Navajo and White Mountain Apache children was estimated at 20 of 100,000 children less than 5 years of age. In comparison, our present study estimated Hia disease incidence of 418.8 of 100,000 Inuit children less than 5 years of age in the sparsely populated Keewatin Region of Nunavut in 2001, and 3.7 of 100,000 less than 5 years of age in aboriginal (First Nations and Metis) children residing in the more densely populated areas within the 4 western provinces. We emphasize that the calculated Hia disease incidence among aboriginal children in the 4 provinces is a minimum estimate, because some children with Hia disease may not have been transferred to the IMPACT center during the study period. Hia cases were uncommon among aboriginal children in the eastern Canadian provinces, which may reflect regional differences in susceptibility to this infection. It is also possible that cases were missed because of incomplete serotyping or because patient ethnicity was not documented.

Hif invasive infections are relatively uncommon. Urwin et al31 reported an increase in Hif disease incidence from 0.5 per 1,000,000 population in 1989 to 1.9 per 1,000,000 population in 1994. However, only 25% of these cases occurred in children less than 5 years of age. The case-fatality rate among children with Hif infections was 21%.31

Nontypable H. influenzae has been recognized as an important cause of acute otitis media, sinusitis, and pneumonia. Other invasive infections associated with nontypable serotypes include sepsis, bacteremia, meningitis, and ventriculoperitoneal shunt infection.2,16,32–37 Early-onset neonatal disease caused by nontypable serotype has been described, typically presenting with respiratory distress and pneumonia, with a case-fatality of approximately 50%.34 A prospective surveillance study for invasive H. influenzae infections in the United Kingdom and Ireland revealed a mortality rate of 9.7% for nontypable serotypes, with prematurity as an underlying risk factor in greater than 40% of the cases.16 The IMPACT surveillance network only captures patients admitted to Children’s Hospitals, which may not include the majority of neonatal care beds for a region. Therefore both outpatients and neonates would be excluded from this surveillance and it is not surprising that relatively few cases of both neonatal and outpatient diseases such as otitis media were captured.

Underlying medical conditions were present in slightly more than 50% of the children with non-b infections in our study, which is similar to rates of 26-88% in previous reports. Risk factors identified in other studies included prematurity, Down’s syndrome, renal failure, sickle cell disease, congenital heart disease, lymphoproliferative disorder, and immunosuppression secondary to malignancy or acquired immunodeficiency syndrome. The presence of a CSF leak or ventriculoperitoneal shunt also predisposed to meningitis or shunt infection, whereas chronic lung disease and mechanical ventilation were risk factors for pneumonia.16,31,32,34,38

The appropriate public health response to H. influenzae non-b serotype infections is presently unknown. Unlike Hib, which has a secondary attack rate among household contacts roughly 300-500 times higher than the general population, the secondary attack rates for Hia and other non-b serotypes are unknown. The study by Millar et al14 did not identify any who were secondary cases. We attempted to ascertain the secondary attack rate for non-b serotypes and used a wider capture period, ie, finding secondary cases within 60 days of an index case in the household or community. Two cases from the same community were identified with invasive Hia disease, and other than being immunized on the same day several months previously and residence in a small community, not enough information was available to suggest an epidemiologic link between the patients. Calculation of the secondary attack rate is this manner is limited, because contact information for 44% of the patients was not documented in the hospital charts. In our study, the absence of any confirmed secondary cases does not support providing chemoprophylaxis to contacts of patients with invasive H. influenzae non-b infections.39

There were some limitations to this retrospective study. Eleven percent of the isolates were not serotyped and serotyping was not available from the prevaccine era. Patient ethnicity was incompletely documented in many IMPACT centers during the study period. Finally, with the exception of Inuit cases from Keewatin Region and aboriginal cases from the western provinces, we were unable to calculate H. influenzae non-b disease population-based incidence rates in other regions monitored by IMPACT.

It is also possible that there may have been inaccuracies because of the diversity of methods used for serotyping the isolates. Seroagglutination appears to be an accurate test, especially if performed by experienced technologists in a centralized (national) laboratory.40 In our study, serotyping was performed by the local IMPACT hospital or by the provincial laboratories. PCR has been proposed as the most accurate method in typing H. influenzae species, because the problems with autoagglutination and cross-reactions associated with serotyping are circumvented.24 The concordance between serotying performed at different laboratories using different methodologies has been reported to be 68–80%, whereas results obtained by PCR concur with serotyping in the range of 60–88%.21,41 LaClaire et al21 reported that 19% of isolates identified as Hib by serotyping where found to be nontypable by PCR. Bokermann et al41 reported that 22% and 45% of isolates initially identified as nontypable by 2 different seroagglutination methods were later reclassified as either Hia or Hib using PCR. In our study, only 2 hospitals relied on PCR testing. It is possible that the misidentification of isolates could have altered the disease epidemiology considerably in our study.

CONCLUSIONS

In 1996–2001, 60% of H. influenzae invasive disease cases in 12 IMPACT centers were caused by non-b serotypes, which were associated with significant morbidity and mortality. There was a high incidence of Hia infection in aboriginal children, and the severity of Hia infections was comparable with that caused by Hib in the prevaccine era. IMPACT will conduct ongoing prospective surveillance for non-b serotypes, incorporating the lessons learned in this retrospective study. In particular, the use of a centralized laboratory for performing PCR and seroagglutination tests is needed to better delineate the infections caused by H. influenzae serotypes in Canada.

ACKNOWLEDGMENTS

The diligent efforts of the IMPACT monitors and staff of the IMPACT data center are gratefully acknowledged. The IMPACT project is administered by the Canadian Pediatric Society, and supported by the Public Health Agency of Canada and Alberta Health.

This surveillance activity was conducted as part of the Canadian Immunization Monitoring Program Active (IMPACT), a national surveillance initiative managed by the Canadian Paediatric Society (CPS) and conducted by the IMPACT network of paediatric investigators on behalf of the Public Health Agency of Canada’s (PHAC’s) Immunization and Respiratory Infections Division (IRID). Funding for these surveillance activities was provided by PHAC and Alberta Health.

IMPACT Investigators and Participating Centers: Scott Halperin, IWK Health Centre, Halifax, Nova Scotia; Robert Morris, Eastern Janeway Children’s Health and Rehabilitation Center, St. John’s, Newfoundland; Pierre Déry, Centre Mére-Enfant de Québec, Québec; Marc Lebel, CHU Ste-Justine pour les Enfants, Montreal, Québec; Dorothy Moore, Montreal Children’s Hospital, Montreal, Québec; Nicole Le Saux, Children’s Hospital of Eastern Ontario, Ottawa, Ontario; E. Lee Ford-Jones, The Hospital for Sick Children, Toronto, Ontario; Barbara Law, Winnipeg Children’s Hospital, Winnipeg, Manitoba; Ben Tan, Royal University Hospital, Saskatoon, Saskatchewan; Taj Jadavji, Alberta Children’s Hospital, Calgary, Alberta; Wendy Vaudry, Stollery Children’s Hospital, Edmonton, Alberta; David Scheifele, BC’s Children’s Hospital, Vancouver, British Columbia; Theresa Tam, Immunization and Respiratory Infections Division, Public Health Agency of Canada, Ottawa, Ontario; Joanne Embree, Canadian Pediatric Society liaison; Karen Grimsrud, Alberta Health liaison, Edmonton, Alberta.

REFERENCES

1. Peltola H. Worldwide Haemophilus influenzae type b disease at the beginning of the 21st century: global analysis of the disease burden 25 years after the use of the polysaccharide vaccine and a decade after the advent of the conjugates. Clin Microbiol Rev. 2000;13:302–317.
2. Ward J. Haemophilus influenzae. In: Feigin R, Cherry J, Demmler G, Kaplan S, eds. Textbook of Pediatric Infectious Diseases. Philadelphia, PA: Saunders; 2004:1636–1655.
3. Adams WG, Cochi SL, Plikaytis BD, et al. Decline of childhood Haemophilus influenzae type b (Hib) disease in the Hib vaccine era. JAMA. 1993;269:221–226.
4. Immunization Monitoring Program Active. Historic low Haemophilus influenzae type b case tally—Canada 2000. Can Commun Dis Rep. 2001;27:149–150.
5. Bath S, Bisgard K, Murphy T, Shutt K. Progress toward elimination of Haemophilus influenzae type b invasive disease among infants and children—United States, 1998–2000. MMWR Morb Mortal Wkly Rep. 2002;51:234–237.
6. Grewal S, Scheifele DW. Haemophilus influenzae type b disease at 11 pediatric centers, 1996–1997. Can Commun Dis Rep. 1998;24:105–108.
7. Scheifele DW, Gold R, Law B, et al. Decline in Haemophilus influenzae type b invasive infections at five Canadian pediatric centers. Can Commun Dis Rep. 1993;19:88–91.
8. Scheifele DW, Bell A, Jadavji T, et al. Population-based surveillance of Hib invasive infections in children in British Columbia, Alberta and Ontario—1995 to 1997. Can J Infect Dis. 2000;11:135–140.
9. Scheifele DW. Recent trends in pediatric Haemophilus influenzae type B infections in Canada. Immunization Monitoring Program, Active (IMPACT) of the Canadian Paediatric Society and the Laboratory Centre for Disease Control. CMAJ. 1996;154:1041–1047.
10. Scheifele D, Halperin S, Law B, et al. Invasive Haemophilus influenzae type b infections in vaccinated and unvaccinated children in Canada, 2001–2003. CMAJ. 2005;172:53–56.
11. Slack MPE, Azzopardi HJ, Hargreaves RM, Ramsay ME. Enhanced surveillance of invasive Haemophilus influenzae disease in England, 1990 to 1996: impact of conjugate vaccines. Pediatr Infect Dis J. 1998;17(suppl):S204–S209.
12. Perdue DG, Bulkow LR, Gellin BG, et al. Invasive Haemophilus influenzae disease in Alaskan residents aged 10 years and older before and after infant vaccination programs. JAMA. 2000;283:3089–3094.
13. Garpenholt O, Hugosson S, Fredlund H, Giesecke J, Olcen P. Invasive disease due to Haemophilus influenzae type b during the first six years of general vaccination of Swedish children. Acta Paediatr. 2000;89:471–474.
14. Millar EV, O’Brien KL, Watt JP, et al. Epidemiology of invasive Haemophilus influenzae type A disease among Navajo and White Mountain Apache children, 1988–2003. Clin Infect Dis. 2005;40:823–830.
15. Ribeiro GS, Reis JN, Cordeiro SM, et al. Prevention of Haemophilus influenzae type b (Hib) meningitis and emergence of serotype replacement with type a strains after introduction of Hib immunization in Brazil. J Infect Dis. 2003;187:109–116.
16. Heath PT, Booy R, Azzopardi HJ, et al. Non-type b Haemophilus influenzae disease: clinical and epidemiologic characteristics in the Haemophilus influenzae type b vaccine era. Pediatr Infect Dis J. 2001;20:300–305.
17. Rothrock G, Reingold A, Alexopoulos N, O’Malley C, Smith NJ. Haemophilus influenzae invasive disease among children aged <5 years—California, 1990–1996. MMWR Morb Mortal Wkly Rep. 1998;47:737–740.
18. Mühlemann K, Balz M, Aebi S, Schopfer K. Molecular characteristics of Haemophilus influenzae causing invasive disease during the period of vaccination in Switzerland: analysis of strains isolated between 1986 and 1993. J Clin Microbiol. 1996;34:563.
19. Scheifele D, Halperin S. Immunization Monitoring Program, Active: a model of active surveillance of vaccine safety. Semin Pediatr Infect Dis. 2003;14:213–219.
20. Popovic T, Ajello G, Facklam R. Laboratory manual for the diagnosis of meningitis caused by Neisseria meningitidis, Streptococcus pneumoniae and Haemophilus influenzae. Geneva, Switzerland: Communicable Disease Surveillance and Response, WHO; 1999.
21. LaClaire LL, Tondella MLC, Beall DS, et al.; and the Active Bacterial Core Surveillance Team Members. Identification of Haemophilus influenzae serotypes by standard slide agglutination serotyping and PCR-based capsule typing. J Clin Microbiol. 2003;41:396.
22. Welch DF, Hensel D. Evaluation of Bactogen and Phadebact for detection of Haemophilus influenzae type b antigen in cerebrospinal fluid. J Clin Microbiol. 1982;16:905–908.
23. Ingram DL, Pearson AW, Occhiuti AR. Detection of bacterial antigens in body fluids with the Wellcogen Haemophilus influenzae b, Streptococcus pneumoniae, and Neisseria meningitidis (ACYW135) latex agglutination tests. J Clin Microbiol. 1983;18:1119–1121.
24. Falla TJ, Crook DW, Brophy LN, Maskell D, Kroll JS, Moxon ER. PCR for capsular typing of Haemophilus influenzae. J Clin Microbiol. 1994;32:2382–2386.
25. Hammond GW, Rutherford BE, Malazdrewicz R, et al. Haemophilus influenzae meningitis in Manitoba and the Keewatin district, NWT: potential for mass vaccination. CMAJ. 1988;139:743–747.
26. Tsang RS, Sill ML, Skinner SJ, Law DK, Zhou J, Wylie J. Characterization of invasive Haemophilus influenzae disease in Manitoba, Canada, 2000–2006: invasive disease due to non-type b strains. Clin Infect Dis. 2007;44:1611–1614.
27. Adderson EE, Byington CL, Spencer L, et al. Invasive serotype a Haemophilus influenzae infections with a virulence genotype resembling Haemophilus influenzae type b: emerging pathogen in the vaccine era? Pediatrics. 2001;108:e18.
28. Kroll JS, Moxon ER, Loynds BM. An ancestral mutation enhancing the fitness and increasing the virulence of Haemophilus influenzae type b. J Infect Dis. 1993;168:172–176.
29. Kroll JS, Moxon ER, Loynds BM. Natural genetic transfer of a putative virulence-enhancing mutation to Haemophilus influenzae type a. J Infect Dis. 1994;169:676–679.
30. Hammitt LL, Block S, Hennessy TW, et al. Outbreak of invasive Haemophilus influenzae serotype a disease. Pediatr Infect Dis J. 2005;24:453–456.
31. Urwin G, Krohn JA, Deaver-Robinson K, Wenger JD, Farley MM; and the Haemophilus influenzae Study Group. Invasive disease due to Haemophilus influenzae serotype f: clinical and epidemiologic characteristics in the H. influenzae serotype b vaccine era. Clin Infect Dis. 1996;22:1076.
32. O’Neill JM, St. Geme JW, Cutter D, et al. Invasive disease due to nontypeable Haemophilus influenzae among children in Arkansas. J Clin Microbiol. 2003;41:3064–3069.
33. Cuthill SL, Farley MM, Donowitz LG. Nontypable Haemophilus influenzae meningitis. Pediatr Infect Dis J. 1999;18:660–662.
34. Campognone P, Singer DB. Neonatal sepsis due to nontypable Haemophilus influenzae. Am J Dis Child. 1986;140:117–121.
35. Lim ME, Hoffman JA, Kim KS. Recurrent ventriculoperitoneal shunt infection due to nontypeable Haemophilus influenzae. Clin Infect Dis. 1999;28:148–149.
36. Kay SE, Nack Z, Jenner BM. Meningitis and septicaemia in a child caused by non-typable Haemophilus influenzae biotype III. Med J Aust. 2001;175:485.
37. Wong GW, Oppenheimer SJ, Vaudry W. CSF shunt infection by unencapsulated Haemophilus influenzae. Clin Infect Dis. 1993;17:519–520.
38. Gilsdorf JR. Haemophilus influenzae non-type b infections in children. Am J Dis Child. 1987;141:1063–1065.
39. Murray DM, Gilsdorf JR. Chemoprophylaxis for household contacts of index cases of invasive non-type b Haemophilus influenzae disease [reply]. Pediatr Infect Dis J. 2001;20:1099–1100.
40. Centers for Disease Control and Prevention. Serotyping discrepancies in Haemophilus influenzae type b disease—United States, 1998–1999. MMWR Morb Mortal Wkly Rep. 2002;51:706–707.
41. Bokermann S, Zanella RC, Lmos AP, de Andrade AL, Brandileone MC. Evaluation of methodology for serotyping invasive and nasopharyngeal isolates of Haemophilus influenzae in the ongoing surveillance in Brazil. J Clin Microbiol. 2003;41:5546–5550.
Keywords:

Haemophilus influenzae; invasive disease; epidemiology; children

Supplemental Digital Content

© 2007 Lippincott Williams & Wilkins, Inc.