  |
|
|
|||||||||||||||||||||||||
|
|||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
19-May-2006, Vol 119 No 1234
|
|||||||||||||||||||||||||||
|
Māori
have a much higher incidence of community-acquired pneumonia and pneumococcal
pneumonia than non-Māori: findings from two New Zealand
hospitals
Stephen Chambers, Richard Laing, David Murdoch, Christopher
Frampton, Lance Jennings, Noel Karalus, Graham Mills, Ian Town
Community-acquired pneumonia (CAP) is the most common cause
of admission to hospital for adults in New Zealand; it has a reported mortality
of between 6.5% and 8%.1–3 Streptococcus
pneumoniae is the most frequently identified pathogen in CAP in New
Zealand1,2,4 and worldwide.3,5,6
Invasive forms of pneumococcal disease are associated with a
high mortality, thus immunisation with the polysaccharide vaccine is recommended
for the elderly and those with chronic disease or impaired immunity in New
Zealand and elsewhere.7
Despite these recommendations, pneumococcal immunisation is
very uncommon in New Zealand.4 Some ethnic groups are also at increased
risk from invasive
pneumococcal infection. For instance, a population-based survey in Auckland
found that Māori and Pacific Island adults in New Zealand had increased
rates of invasive pneumococcal disease, and Māori children have a higher
rate of invasive pneumococcal disease than Caucasian (New Zealand
European) children.8,9
High rates of invasive pneumococcal disease have been
reported in native Americans, African Americans, and indigenous
Australians.10–14 In response, the Australian guidelines now recommends
immunisation of indigenous peoples and Torres Strait Islanders from age of 50
years.15
There is no convincing evidence that pneumonia can be
prevented by the polysaccharide vaccine in older Western populations,16 but the
spread of penicillin-resistant strains of S.
pneumoniae has renewed interest in the prevention of pneumococcal
infections in order to reduce antibiotic pressure as well as reduce morbidity,
mortality, and hospitalisation.17,18
Possible strategies include smoking reduction and the use of
the newer conjugate vaccines.19,20 Because of the potential importance of such
strategies for prevention of pneumonia and pneumococcal infection in New
Zealand, in this study we determined the age-specific rates of CAP and
pneumococcal pneumonia in
Māori and
non-Māori populations in two major regional centres.
MethodsParticipants—All
patients over 18 years of age admitted to Christchurch and Waikato Hospitals
between 27 July 1999 and 27 July 2000 with a diagnosis of community-acquired
pneumonia were screened for inclusion into the study. Christchurch Hospital and
Waikato Hospital each have approximately 600 beds.
Both hospitals are the only hospitals in their
respective regions that admit patients with CAP, although both act as tertiary
referral centres for larger populations. This study is a further evaluation of
patients described previously,4 and the inclusion and exclusion criteria for
this study were those used in a previous CAP study.1
Pneumonia was
defined as an acute illness with radiographic pulmonary shadowing (at least
segmental or present in one lobe), which was neither pre-existing or of another
known cause. Patients were excluded from the study when pneumonia was not the
principle reason for their admission, when they were moribund at presentation
(as relevant history microbiological samples and ethnicity could not be
obtained), and when the pneumonia was associated with bronchial obstruction or
bronchiectasis (as underlying abnormalites alter host susceptibility). Patients
with known tuberculosis were also excluded.
Patients with severe
immunosuppression—neutropaenia, individuals with AIDS, or those currently
receiving cancer chemotherapy were excluded. Patient characteristics and
admission clinical and physical findings were recorded on a standardised
proforma. Patients identified their own ethnicity by answering the question.
“ How do you identify ethnically? You may identify more than
one—Pakeha/European, New Zealand
Māori, Pacific
Islander, Asian, European, Other.”
Recent antibiotic use and pneumococcal vaccination
status were self reported by the patient, and comorbidites were reported by the
patient and verified by reference to medical records. At the time of enrolment,
blood was drawn for haematological, biochemical, and microbiological analysis.
Sputum and urine samples were sought from all patients. All chest radiographs
were reviewed by a designated radiologist in each centre to confirm radiological
entry criteria. Severity of pneumonia was determined by the method of Fine et
al.21
Microbiological
methods—Blood cultures were incubated aerobically and anaerobically
using the BacT/Alert Microbial Detection System (Organon Teknika, Durham, NC,
USA). Respiratory samples were cultured on sheep-blood agar, chocolate agar,
buffered charcoal yeast extract agar supplemented with (-ketoglutarate, and
modified Wadowsky-Yee medium. Urine samples were tested using the NOW™
Streptococcus pneumoniae Urinary
Antigen Test (Binax, Portland, ME) according to the manufacturer's
recommendations.
Criteria for diagnosis as pneumococcal
pneumonia—This diagnosis was made if the clinical and radiological
criteria for pneumonia were fulfilled and S.
pneumoniae was isolated from a sterile sample (such as blood, pleural
fluid, or lung aspirate sample), or when S.
pneumoniae antigen was detected in the urine.
Statistical
analysis—Subjects were excluded from statistical analysis of the
rates of pneumococcal disease if a urine antigen test had not been done, unless
a blood culture was positive. Rates were calculated from the 2001 primary self
declared ethnicity census data for the catchment areas of the two hospitals. A
standardised morbidity ratio (SMR) was calculated for the
Māori/non-Māori
comparison. The expected values for the Māori group were calculated from
the observed rates within each age-sex group for the non-Māori group.
Hypothesis testing and 95% confidence intervals for
SMRs were derived from the standard Poisson approximation. All data was entered
into a specifically designed Microsoft Access database. The SPSS for Windows
10.0 statistical package was used for the analysis (SPSS Inc., Chicago, USA).
The level of significance was set at p<0.05.
ResultsPatient
characteristics—During the 12-month study period, 545 patients were
eligible for the study of whom 474 (87%) participants were enrolled. Of these
474 patients, 304 were from Christchurch Hospital and 170 from Waikato Hospital.
Of the 71 unenrolled patients, 37 individuals declined study enrollment, 18 were
missed for logistic reasons, 10 were unable to give consent and had no available
next of kin, and 6 died prior to consent being obtained.
The mean age of those enrolled in the study was 63.7 years
(range 18–99 years) and 53% were men. 274 participants (58%) were recorded
as having significant comorbidity at time of presentation: 100 (21%) were
smokers, 123 (26%) had chronic obstructive pulmonary disease (COPD), 66 (14%)
asthma, 52 (11%) diabetes, 95 (20%) heart failure, 28 (6%) renal disease, and 5
(1%) liver disease. Twenty-four patients (5%) were immunosuppressed, 128 (27%)
had received an antibiotic prior to admission, 237 (50%) had received influenza
vaccine prior to admission, and (19) 4% received pneumococcal vaccine in the
previous 5 years. Fifty-seven study participants (12%) identified as being
Māori, of whom 41
(72%) were admitted to Waikato Hospital.
Compared
with non-Māori, the Māori population were significantly younger (mean
age of 50 vs 66 years, p<0.001, difference=15 years, 95%CI 10–20 years)
and they had a significantly higher rate of smoking (35% vs 19%, p=0.004,
difference 16%, 95%CI 5%–28%). The mean pneumonia severity index score
(PSI)21 for CAP was
similar among Māori and non-Māori (56 vs 49, difference=7.5, CI
-6%–21%).
The
mean PSI for pneumococcal pneumonia was less among Māori (80 v 95, p=0.042,
difference = 15, 95% CI 4-25) but this was dependent on the lower age among
Māori. There was no statistical difference between the rates of comorbidity
other than asthma, antecedent antibiotic use, morbidity, or mortality at 6 weeks
follow-up between these two groups (Table 1).
Incidence of community
acquired
pneumonia—The
2001 Census populations for the Christchurch and Waikato regions for Māori
were 27,000 and 39,000, respectively; for non-Māori, they were 327,000 and
198,000 respectively. The population age-specific rates of CAP are shown
by ethnicity in Figure 1. The age-specific rates of CAP were statistically
significant for each of the three 10-year age groups from 45–54,
55–64, and 65–74 years. The pneumonia rate was 3.03 times higher
among Māori than
non-Māori in the whole population (Table 2). There was no significant
difference in incidence rates by gender or centre in any of the decade groups.
Incidence of pneumococcal community acquired
pneumonia—The population age-specific rates of pneumococcal CAP are shown
by ethnicity in Figure 2. There was no statistically significant difference in
pneumococcal CAP for any of the 10-year age groups, but the rate was higher in
the population overall (Table 2). There was no significant difference in
incidence rates by gender or centre.
Figure 1. Population age-specific rate of
community-acquired pneumonia
for Maori and non-Maori (p<0.05)
 Figure 2. Population age-specific rates of pneumococcal
community-acquired pneumonia for Maori and non-Maori
 Table 1. Study population patient characteristics and
outcome at 6 weeks
*p=0.004;
**p=0.002.
Table 2. Standardised incidence rates of pneumonia and
pneumococcal pneumonia:
Māori and
non-Māori compared
DiscussionIn
the catchment areas studied, we have shown that Māori have an increased
(3–4 fold) incidence rate of CAP and pneumococcal pneumonia compared with
non-Māori. Moreover, the increase in CAP incidence rates seen in
non-Māori at aged 65 occurred 20 years earlier among Māori, and
pneumococcal pneumonia incidence rates showed a similar pattern.
These incidence rates results cannot be generalised to the
general population as the catchment areas of
Christchurch and
Waikato Hospitals is not representative of the New Zealand population. However,
the clear discrepancy in the rates of CAP and pneumococcal pneumonia between
Māori and non-Māori (as demonstrated by the disease ratios) probably
represents a disease
discrepancy
present in the general population. These findings are consistent with previous
studies that have shown Māori (and other Polynesian) children in South
Auckland have increased rates of admission to hospital for pneumonia, and adult
Māori have increased rates of invasive pneumococcal disease in the
Auckland area compared to Europeans.8,9,22
Identification of ethnicity is important for interpretation
of the results of this study. To cross-check that the ethnicity data had been
collected correctly, we
approached 50% of
those who identified as Māori at the conclusion of the study, and found
complete correlation with the ethnicity as originally recorded. This reassured
us that the ethnicity recorded originally in response to the study ethnicity
question was correct.
The denominator for calculation incidence rates was derived
from the 2001 Census ethnicity data. Although the 2001 Census does not cover the
period of the study, we chose data from this Census rather than an interpolation
of the ethnicity data between 1996 and 2001 because the question asked was
different in those years.
The ethnicity question asked in this study was slightly
different from the question used in both the 1996 and the 2001 census. The 2001
census question reads “Which ethnic group do you belong to? Mark the space
or spaces which apply to you.”
This
instruction is followed by a list of ethnicities as follows: New Zealand
European, Māori, Samoan, Cook Island Māori, Tongan, Nuiean, Chinese,
Indian, Other. In the Census analysis, people who have recorded more than
one ethnic group have been counted in each applicable group.
People answering the Census questionnaire may have been more
likely to include more than one ethnicity than those answering the study
questionnaire. If so, this
will have introduced
some systematic bias into the results as Māori ethnicity would have been
under-reported in the present study compared with the census data. Such an
effect is likely to have reduced the difference, however, and thus would
strengthen our study conclusions.
Eighty-seven percent of patient eligible for the study
participated, but 71 subjects could not be enrolled. Of these, 34 were unlikely
to introduce any ethnic bias, as they could not be enrolled for logistic
reasons, death, or an inability to obtain consent. The other 37 declined to be
enrolled. This could in part be ethnicity related, but we doubt this was
sufficiently strong an influence to compromise the results.
There is also a potential bias in that the
non-Māori group
included non-Māori Polynesians who have increased rates of childhood
pneumonia and invasive pneumococcal disease compared with European New
Zealanders).8,22 This would tend to decrease observed differences between
the groups and thus increase the robustness of the observations, however.
The determination of incidence rates for pneumococcal
pneumonia is difficult, as isolation of S.
pneumoniae from blood or another sterile fluid is insensitive although
highly specific, and isolation for sputum has low specificity because of
potential contamination from organisms colonising the upper respiratory tract.
In this study, the diagnosis of pneumococcal infection
depends largely on the detection of urinary pneumococcal antigen. The method
used differs from previous urinary antigen tests in that it detects a soluble
cell wall pneumococcal antigen common to all strains. There have been several
studies published on the performance of this test,23–26and it has been
licensed by the Federal Drug Administration (FDA) in the US. In an adult
population, we estimated the sensitivity to be 80% (and specificity 100%)
compared with blood culture, and found the test reliable in the presence of
previous antimicrobial therapy.23 It was highly specific in adults as no
positive were found in a large control population, however others have reported
positive results in children with nasopharyngeal carriage of
S. pneumoniae.26
Taken together these results suggest the test sufficiently
robust to be used to estimate the incidence of pneumococcal disease in adult
populations.
The reasons why there is an increased rate of CAP and
pneumococcal disease in
Māori needs to be
examined in further epidemiological studies. Indeed, indigenous peoples
worldwide have increased rates of pneumococcal disease compared with others in
the same geographic region—including Alaskan and Greenland natives,
African Americans, and Australian aborigines.14–15,26–30
It
is very likely that socioeconomic factors play an important role in this
discrepancy. Some of the increase is attributable to smoking, a powerful risk
factor for pneumococcal disease, and Māori have higher rates of
smoking than the general population.31
Other factors such as crowded living conditions, economic
status, and access to medical care may contribute to the observed discrepancy
but it was not the intention of this study to look for these specific effects
and we do not have comprehensive information on the population from which these
cases were drawn on which to base any comparisons.
It is possible that some genetic factors exist which
contribute to increased susceptibility to pneumonia. Recently Yee et al
demonstrated that a homozygous state for FcRIIa-R131 gene is associated with
increased mortality for bacteraemic pneumococcal disease, thus suggesting
inherited host factors play a role in the pathogenesis of pneumococcal
disease,32 and Roy et al observed homozygotes for mannose-binding lectin codon
variants were at increased risk of invasive pneumococcal disease.33 It is likely
that other polymorphisms in host genes influence the outcome of pneumonia.
Increased rates of CAP and pneumococcal disease demonstrated
in an ethnic group increases the potential benefit of targeted prevention
strategies. Such an intervention could include improved antismoking campaigns,19
and consideration could be given to improved influenza and pneumococcal
immunisation rates for
Māori. While
doubts remain over the efficacy of the polysaccharide vaccine for the prevention
of pneumonia,16 it is effective against bacteraemic disease (about 90% of
which is from pneumonia).
At present, this vaccine is scarcely used in any group,
presumably because of cost and access considerations, although it is recommended
by the New Zealand Ministry of Health for at-risk populations.7 There is also
evidence that the conjugate vaccine reduces pneumonia in children and has a
secondary effect in decreasing pneumococcal disease in adults.20
Careful consideration should be given to evaluating the
potential value of both pneumococcal polysaccharride vaccine among
Māori adults and
the conjugate vaccine in children.
Author information:
Stephen T Chambers, Clinical Director – Infectious Diseases;1,2 Richard T
Laing, Respiratory Physician;3 David R Murdoch, Microbiologist;1 Christopher
Frampton, Biostatistician;3,4 Lance C Jennings, Virologist;5 Noel C Karalus,
Respiratory Physician;6 Graham D Mills, Respiratory Physician;6 G Ian Town,
Respiratory Physician3,4
Acknowledgments:
We are extremely grateful to Rose Smith (Tainui Raukawa Maniapoto and Member of
the Kaumatua Kaunihera), Waikato DHB, Mairie Kipa from Ngai Tahu, Jo Baxter from
theUniversity of Otago, and Elizabeth
Cunningham (Māori
advisor to the Christchurch School of Medicine) for their help and advice during
the preparation of this manuscript.
Correspondence: Dr
ST Chambers, Department of Pathology, Christchurch School of Medicine and Health
Sciences, PO Box 4345, Christchurch. Fax: (03) 364 0952; email: Steve.Chambers@cdhb.govt.nz
References:
|
|
|||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|