| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Firestone Institute for Respiratory Health, Hamilton, Ontario, Canada
Correspondence and requests for reprints should be addressed to Neil W. Johnston, M.Sc., St. Joseph's Healthcare, 50 Charlton Avenue, East Hamilton, ON, Canada L8N 4A6. E-mail: njohnsto{at}sympatico.ca
ABSTRACT
The majority of chronic obstructive pulmonary disease (COPD) and asthma exacerbations in both children and adults are associated with respiratory viral infections and are cyclic in nature. Some variation in these cycles is associated with the timing of the appearance of respiratory viruses, particularly influenza and respiratory syncytial virus. Much more, however, is associated with signal events that are of either fixed or predictable timing. In children, asthma exacerbations reach epidemic levels following school return after the summer vacation and these are predominantly associated with rhinovirus infections. Although younger adults experience a rise in asthma exacerbations at this time, these are secondary to the epidemic in children. Older adults with either COPD or asthma experience only a slightly elevated risk of exacerbations after school return, and hospital presentations for pneumonia in any age group show only marginal increases at that time. Exacerbations of both COPD and adult asthma, with increasing risk with age, are at their highest average annual levels during the Christmas period. This effect appears to be independent of the timing of above average levels of influenza, RSV, parainfluenza, or adenovirus detections; however, hospitalization for respiratory tract infections in all age groups reaches high levels at the same time. Both the post–summer vacation asthma epidemic and the Christmas epidemic of COPD, asthma, and pneumonia are synchronous with the timing of signal events, the day of school return for the former and Christmas Day for the latter, and have been for several years. The agents responsible for the Christmas epidemic of respiratory diseases have not yet been identified. The differences between age and disease exacerbation patterns after school return and at Christmas suggest that either different agents are involved or that the response to a common agent is different between the two signal events.
Key Words: COPD • asthma • exacerbations • epidemiology
The burden of morbidity imposed by diseases of the respiratory system is probably greater than for that of any other body system. Respiratory diseases, and their contribution to population morbidity, are also unique in their seasonal nature. Although the three most common respiratory disease groupings, respiratory tract infections (RTIs), chronic obstructive pulmonary disease (COPD), and asthma, are usually treated as discrete in research studies, morbidity associated with them may share underlying mechanisms. The relation of asthma exacerbations to the common cold and its seasonal appearance has been recognized for hundreds of years (1, 2). However, in the classic study of the seasonal nature of rhinovirus infections, the term "rhinovirus illness" is used to group all types of respiratory symptoms reported (3).
Since the early 1990s, the importance of respiratory viral infections (RVIs) in the genesis of exacerbations of asthma and, more lately, COPD has been documented and widely accepted (4). The fact that seasonal differences in the risks of asthma and COPD exacerbations occur is commonly understood (5). What is not understood and has rarely been considered is the relationship of cycles of exacerbation of COPD and asthma to different respiratory viruses and to signal events that may foster their transmission. Differences between cycles of asthma and COPD exacerbations overall and between age groups and sexes may offer important opportunities to understand their etiology.
THE SIMILARITIES AND DIFFERENCES OF CYCLES OF RESPIRATORY DISEASE
The Differences between Cycles of Asthma Exacerbations in Children and Adults
In children, asthma exacerbations are at average or below average levels in the winter months, show a moderate rise in the spring, and a trough in the summer. This is followed in early fall by a period in which asthma exacerbations in children reach epidemic levels, when in Canada approximately one-quarter of all hospitalizations of children for asthma occur (6), followed by a gradual decline to winter levels (Figure 1). In young adults, more exacerbations are seen in the winter months than in children but a significant peak still occurs in late summer and early fall. In older adults, the shift to winter as the greatest period of risk is almost complete, although a detectable rise still occurs in early fall.
|
The amplitudes of the peaks by age group that occur following school return after the summer vacation and in the Christmas period are reversed (5). Substantially similar cycles as those described for Canada are observed in other countries in the Northern Hemisphere (5).
The Differences between Cycles of COPD and Asthma Exacerbations
Annual cycles of COPD exacerbations and those of asthma in adults of the same age are essentially identical in form. As in the case of adult asthma, a major epidemic occurs annually during the Christmas period, and this occurs in every country studied so far in the Northern Hemisphere (Figure 2). The Christmas period epidemics of COPD and adult asthma in Canada, the United Kingdom, and Sweden apparently all occur in Week 52. Preliminary data suggest that adult asthma and COPD exacerbations in New Zealand both show annual cycles with peaks in July. These are analogous seasonally to the peaks observed during influenza periods in the Northern Hemisphere but are not of fixed timing as is the case with the Northern Hemisphere Christmas peaks. Peaks synchronous with Christmas Day do not appear to occur in New Zealand.
|
|
FACTORS ASSOCIATED WITH CYCLES OF RESPIRATORY DISEASES
The remarkable similarity of cycles of COPD, asthma, and RTIs in older adults requires explanation. One possibility is that diagnostic discrimination among the three diseases is so poor in hospital records that they appear to have identical patterns of incidence. Although no specific validation study for the assignment of diagnoses of asthma and COPD has been performed for Canadian hospitalization data, reabstraction studies over all age groups have examined ambulatory care–sensitive conditions as a group that includes both COPD and asthma and found incorrect assignment in 10.7% of records (7). For pneumonias (RTI), the error rate was 6.9%. These figures suggest that the patterns of the three diseases would be unlikely to be so conformal unless common etiologic factors were also operating. Also, the very clear distinctions observed between cycles of RTIs and asthma in children argue that diagnostic discrimination is reasonable during hospital encounters. The most plausible explanation for the conformity of COPD and asthma exacerbation cycles in older adults and RTI cycles in all age groups is that they are related to common causal factors. Thus, the differences observed between cycles of asthma in children and those seen transitionally in younger adults and cycles of RTIs and asthma and COPD in older adults may hold important clues to the etiology of asthma and COPD exacerbations.
Signal Events
School return after the summer vacation.
In children in both the Northern and Southern Hemispheres, the late summer asthma epidemic is synchronous with the date of school return after the summer vacation, occurring between 2 and 3 weeks later (8, 9), and this phenomenon is identical in countries with different dates of school return (10). The smaller peaks observed in adult asthma exacerbations and COPD after school return are also in synchrony with the date of school return but occur 1 to 2 weeks later than in children, and this temporal relationship has been consistent for many years (9). The post–summer vacation asthma epidemic occurs in both the Northern and Southern Hemispheres and is seen in tropical countries such as Australia and Trinidad as well as in northern Europe and North America (10).
Children returning to school after the summer vacation are reexposed to RVIs (6), most commonly rhinovirus (3, 11), at a time when they may not have been exposed to them for many weeks. The late summer is a period with high levels of environmental allergens (12) and sensitizing allergens are commonly present in the school environment (13, 14). School return is a period of high stress, which can worsen asthma symptoms in children (15), and levels of use of asthma control therapy may be at their lowest levels of the year immediately before school return (6). Although an RVI may be the immediate trigger for an asthma exacerbation, it is likely that the above factors act together to bring about the "September epidemic," one of the most interesting features of which is the precision of its relation to the date of school return in children and secondarily in younger adults (9). The lesser amplitude of the peaks of exacerbations of asthma and COPD in older adults observed after the summer vacation may reflect older individuals' degree of exposure to school-age children at this time.
If the late summer and early fall are the "season" for rhinovirus, one would expect some variation in the timing of the September epidemic of asthma. This does not occur, except in synchrony with the date of school return, raising the possibility that at least in as far as its relation to asthma exacerbation is concerned, the rhinovirus season is brought about by the perfect opportunity for mass viral transmission occasioned by school return and children's social behavior.
Christmas.
The epidemic peaks of COPD, adult asthma, and RTIs in all age groups observed in the Christmas period (Figures 1 and 2) bear the same precise relation as the September asthma epidemic to a signal event—in this case, Christmas Day—and preliminary data show that they occur an average of 6 days later. The exception is the peak in RTIs for school-aged children, which occurs 1 week earlier.
No information about the types or respiratory viruses prevalent during the Christmas period is presently available; however, it could be speculated that, such as following school return after the summer vacation, school-aged children are the primary vectors of RVIs to older and perhaps more vulnerable family members. The lack of an asthma peak in children of any age group in the Christmas period suggests that children's asthma symptoms are not worsened by infections with the agents prevalent at Christmas or that they have developed resistance to them through prior exposure, having just completed a school term.
The Christmas epidemics appear to be independent of levels of detection of influenza, RSV, parainfluenza, and adenovirus, which occur with variable timing (11). Peaks of influenza detections may coincide with the Christmas period (Figure 3), and when this occurs, the burden of respiratory disease morbidity may be extreme. The role of other viruses commonly detected during exacerbations of COPD and asthma, such as rhinovirus, in the Christmas epidemics is presently unknown.
Medical offices may be less accessible during the Christmas period and this may explain some of the increase in emergency room (ER) presentations for respiratory diseases. However, the same phenomenon is also seen in inpatient episodes (Figure 2), suggesting that overuse of the ER as an alternative to a family physician visit is not a substantial contributor to the peak in ER use.
RVIs
Rhinovirus has been shown to be the virus most commonly coinciding with exacerbations of COPD and asthma (4). The rhinovirus "season" is reported to be maximal in the month of September, with a smaller peak occurring in the spring (3, 11). In Canada, the majority of children presenting to ERs with asthma during the post–school return epidemic have been shown to have RVIs, two-thirds of which are rhinovirus infections (6). In the same study, 40% of children with asthma not requiring emergent care were also found to have predominantly rhinovirus infections during the September epidemic at a single encounter.
The day of school return after the summer vacation is followed with precision 2.5 weeks later by epidemic peaks of asthma exacerbations in children. These, in turn, are followed by peaks in exacerbations of adult asthma and COPD that are smaller but clinically important. It is possible that the rhinovirus season may not occur independently but may be created by optimal conditions favorable to transmission of infection by index cases, particularly to naive or vulnerable subjects, after school return and by changes in family routines. If this is the case, rhinovirus behaves differently than respiratory viruses for which public health surveillance commonly occurs (influenza, RSV, adenovirus, and parainfluenza), and which appear at higher levels at different times from year to year (11). Preliminary data show that higher levels of influenza, RSV, parainfluenza, and adenovirus significantly increase the risk of RTIs in all age groups and COPD and asthma exacerbations in older adults. However, they are not associated with any increased risk of asthma exacerbations in children of any age or in young adults.
Age and Sex Differences between COPD, Asthma, and RTIs
The form of annual cycles of asthma are very similar between the sexes; however, in children over the whole annual cycle, the risk of an asthma exacerbation requiring hospitalization is approximately double in males compared with females (5). In adults, the situation is reversed, with the risk over the whole year being double in females compared with males. The crossover in relative risk occurs very suddenly, during the years of puberty (Figure 4) (5). A similar phenomenon occurs with RTIs, based on ER presentation data (Figure 4). Here, however, the first crossover in risk from boys to girls occurs at age 9. After that, the risk of an RTI requiring emergent care is significantly greater in females than males throughout adulthood. In the mid-60s, a second crossover occurs and for the balance of the life cycle the risk of an RTI requiring emergency care is significantly greater in males. It could be speculated that smoking patterns are responsible for the latter change, but no data are available to support this.
|
|
In the Formoterol And Corticosteroids Establishing Therapy (FACET) study (17) comparing treatment of asthma with budesonide at two different doses alone compared with budesonide combined with formoterol, women were observed to have 1.9 times higher risk of having an exacerbation than men. It is possible that this was because of the higher overall risk of asthma exacerbations in women than men. It is also possible that this was related at least partially to a differential effect of the therapy between the sexes.
These results suggest that the crossovers in the incidence of COPD and asthma exacerbations and RTIs should be considered in the evaluation of therapeutic interventions.
DISCUSSION
The distinctions between annual cycles of asthma, COPD, and RTIs offer significant opportunities for further research. If the Christmas period is truly one of high risk for all three diseases in adults and RTIs in children, because it presents an ideal opportunity for transmission of infection, one would expect a similar phenomenon to occur at Thanksgiving in the United States. If this were to be the case, comparison of the agents responsible would be valuable.
The effects of school return after the summer vacation on children with asthma and their association primarily with rhinovirus infection, the progressively lesser effects in this period with age, and the contrasts by age group in this period with the epidemics observed at Christmas suggest that COPD and asthma may respond differently to different respiratory viruses over the life cycle. Given the primacy of RVIs among causes of exacerbations, differences observed between the sexes in the risks of disease exacerbation also raise the possibility that the sexes may respond differently to respiratory viruses.
The signal events described above represent an opportunity for the evaluation of respiratory disease therapies and management strategies when higher levels of important outcomes may occur, permitting shorter periods of evaluation. Their existence also raises the warning that clinical trials of therapies for respiratory diseases that enroll patients for periods of less than a year at either fixed or variable times should consider the period for which a given patient was providing data. Infectious agents that precipitate exacerbations of asthma and COPD may show annual variation in the timing of their appearance as may the degree of response of patients with asthma and COPD to them.
FOOTNOTES
Supported by unrestricted grants from AstraZeneca and GlaxoSmithKline Canada.
Conflict of Interest Statement: N.W.J. has received speaking fees and research funding from AstraZeneca, GlaxoSmithKline, and Merck Frosst Canada.
(Received in original form June 1, 2007; accepted in final form July 5, 2007)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
