1 March 2008 Volume 34
Number 03
F Reyes, MHSc (1), S Aziz, MSc (1), Y Li, PhD (2), JF Macey, MSc (1), B Winchester, MSc (1), M Garner, MSc (1), P Huston, MD, MPH (1), A King, MD, MHSc, FRCPC (1)
The Centre for Immunization and Respiratory Infectious Diseases (CIRID), Public Health Agency of Canada (PHAC), coordinates a national influenza surveillance network through the FluWatch program. The primary objectives of the FluWatch program are early detection of influenza activity in Canada and abroad; monitoring of circulating strains of influenza virus, including antigenic characterization, antiviral resistance and identification of new sub-types; and provision of virologic surveillance information to the World Health Organization (WHO) global influenza surveillance network to contribute to decision-making for the following season's vaccine components. This report provides an epidemiologic and virologic summary of influenza activity in Canada during the 2006-2007 season.
The FluWatch program consists of a network of sentinel laboratories, sentinel primary care practices, provincial and territorial ministries of health, and sentinel paediatric tertiary care hospitals. Five main indicators of influenza activity are reported by the network on a weekly basis across Canada and throughout the season: (1) sentinel laboratory-based influenza and other respiratory virus detections; (2) strain identification and antiviral resistance for circulating influenza viruses; (3) sentinel primary care consultation rates of influenza-like illness (ILI); (4) regional influenza activity levels as assigned by provincial and territorial FluWatch representatives; and (5) paediatric influenza-associated hospital admissions and mortality data. In addition, the FluWatch program conducts an assessment of international influenza activity by monitoring reports from other influenza surveillance programs worldwide. Detailed methodology are described in a previous report(1).
1) Sentinel laboratory-based influenza and other respiratory virus detections
Influenza and other respiratory virus detections from the Respiratory Virus Detections Surveillance System (RVDSS)(aggregate data)
Laboratory-based respiratory virus detections are reported through the sentinel laboratory Respiratory Virus Detections Surveillance System (RVDSS), which operates year-round. Between 27 August 2006, and 25 August 2007, 33 laboratories across Canada reported a total of 100,864 influenza tests, 8.1% (8,133/100,864) of which were positive for influenza viruses. Of these, 87.5% (7,116/8,133) were positive for influenza A viruses, and 12.5% (1,017/8,133) were positive for influenza B viruses.
During the same time period, a total of 90,472 respiratory syncytial virus (RSV) tests were conducted of which 10,401 or 11.5% were positive for RSV viruses; 78,861 parainfluenza virus tests were conducted of which 2,696 or 3.4% were positive for parainfluenza viruses; and 72,048 adenovirus tests were conducted of which 1,400 or 1.9% were positive for adenoviruses.
Influenza virus detections from the RVDSS (case-by-case data)
Twenty-five of the 33 RVDSS participating laboratories reported a total of 7,023 detailed case-by-case records to CIRID. Total influenza detection reports (aggregate weekly data) relative to total case-by-case records are presented in Table 1 by reporting province (total detections) and province or territory from which the specimen originated (case-by-case records). Since specimens from the territories are sent to nearby provincial laboratories for testing, total detections from the territories are rolled into the provincial aggregate counts but are separated out for the case-by-case record totals, based on geographic information available in the case-by-case data. The largest proportion of cases were reported by Quebec (35.9% or 2,518/7,023), Ontario (24.0% or 1,668/7,023) and Alberta (16.6% or 1,164/7,023).
Table 1. Aggregate and case-by-case influenza data, by province/territory or region, Canada, 2006-2007
Province/territory or region |
Aggregate detections |
Case-by-case data |
||
Number of cases |
% of total |
Number of cases |
% of total |
|
Nfld. |
169 |
2.1 |
169 |
2.6 |
P.E.I. |
29 |
0.4 |
29 |
0.4 |
N.S. |
125 |
1.5 |
151 |
2.3 |
N.B. |
313 |
3.8 |
344 |
5.2 |
Atlantic |
636 |
7.8 |
693 |
10.5 |
Que. |
2,514 |
30.9 |
2,518 |
38.2 |
Ont. |
2,720 |
33.4 |
1,688 |
25.6 |
Man. |
76 |
0.9 |
68 |
1.0 |
Sask. |
318 |
3.9 |
330 |
5.0 |
Alta. |
1,216 |
15.0 |
1,164 |
17.7 |
Prairies |
1,610 |
19.8 |
1,562 |
23.7 |
B.C. |
653 |
8.0 |
481 |
7.3 |
Y.T. |
31 |
0.5 |
||
N.W.T. |
15 |
0.2 |
||
Nun. |
35 |
0.5 |
||
Territories |
81 |
1.2 |
||
Total |
8,133 |
|
7,023 |
Of the 7,023 total cases identified during 2006-2007, 86.2% (6,051/7,023) were infected with influenza A and 13.8% (972/7,023) with influenza B. Influenza A cases were reported across Canada during the 2006-2007 season, with the greatest proportion reported from Quebec (27.8% or 1,684/6,051), Ontario (27.5% or 1,664/6,051) and Alberta (18.4% or 1,114/6,051). Of the influenza B cases identified, the greatest proportion was reported from Quebec (85.8% or 834/972) (Table 2).
Table 2: Case-by-case influenza data, by province/territory, influenza type, Canada, 2006-2007
Influenza type |
Nfld. |
P.E.I. |
N.S. |
N.B. |
Que. |
Ont. |
Man. |
Sask. |
Alta. |
B.C. |
Y.T. |
N.W.T. |
Nun |
Total |
Type A |
169 |
29 |
151 |
319 |
1,684 |
1,664 |
68 |
323 |
1,114 |
450 |
30 |
15 |
35 |
6,051 |
Type B |
25 |
834 |
24 |
7 |
50 |
31 |
1 |
972 |
||||||
Total |
169 |
29 |
151 |
344 |
2,518 |
1,688 |
68 |
330 |
1164 |
481 |
31 |
15 |
35 |
7,023 |
% of influenza A among case-by-case data |
100.0 |
100.0 |
100.0 |
92.7 |
66.9 |
98.6 |
100.0 |
97.9 |
95.7 |
93.6 |
96.8 |
100.0 |
100.0 |
86.2 |
% of influenza B among case-by-case data |
0.0 |
0.0 |
0.0 |
7.3 |
33.1 |
1.4 |
0.0 |
2.1 |
4.3 |
6.4 |
3.2 |
0.0 |
0.0 |
13.8 |
Overall, influenza A predominated during the 2006-2007 season (86% of laboratory-confirmed cases). Regionally, influenza A accounted for 90% or more of the laboratory-confirmed cases in all provinces and territories except for Quebec where influenza B accounted for 33% of the cases in the province. Nationally, 65% of all influenza cases had onset dates in the 8-week period between week 5 and week 12 (late January to mid-March 2007), with roughly 40% having onset dates during the peak period from week 7 to week 10 (mid-February to early March 2007), and the number per week peaking in week 9 (n =790) (Figure 1). Regional peaks in onset dates for cases were also evident in all provinces and territories. The first peak reflected in the case-by-case data occurred in week 50 in Alberta, followed by a peak in the rest of the Prairie Provinces in week 4, British Columbia in week 5, Ontario in week 7 and the Atlantic Provinces in week 8. Quebec and the Territories were the last to reach peak activity (both due to influenza A) in week 9. A second peak was observed in Alberta in week 9 due to influenza A as well. A second, yet smaller peak, due to influenza B was observed in Quebec in week 14. Influenza A cases predominated for most of the season with onset dates peaking in late February/early March 2007 (week 9, n = 736). Onset dates for influenza B cases peaked in early April (week 14, n = 111).
During the 2006-2007 season, the highest proportion of cases were reported among the following age groups: 25 to 44 years (22.3% or 1,563/7,023), 0 to 4 years (22.0% or 1,547/7,023), ≥ 65 years (20.6% or 1,448/7,023). It was notable that the age groups 15 to 24 and 10 to 14 years represented the lowest proportion of laboratory-confirmed cases at 4.4% (312/7,023) and 6.0% (420/7,023), respectively. Of the influenza A cases, the highest proportion were among those 0 to 4 years (22.6% or 1,366/6,051), 25 to 44 years (22.0% or 1,329/6,051) and ≥ 65 years (20.2% or 1,221/6,051). The majority of influenza B cases were among those > 25 years of age (68.2% or 663/972) and 0 to 4 years (18.6% or 181/972) (Figure 2).
Figure 1. Case-by-case data by influenza type, region and week of onset, Canada, 2006-2007
Influenza in Canada, 2006-2007
Influenza in the Atlantic Provinces, 2006-2007
Influenza in Quebec, 2006-2007
Influenza in Ontario, 2006-2007
Influenza in the Prairies, 2006-2007
Influenza in British Columbia, 2006-2007
Influenza in the Territories, 2006-2007
2) Strain identification and antiviral resistance for circulating influenza viruses
The National Microbiology Laboratory (NML) virologic surveillance detects and describes antigenic changes in the circulating influenza viruses and conducts tests for antiviral resistance among circulating strains. During the period from 15 September 2006 to 17 September 2007, the NML antigenically characterized 1,023 influenza viruses received from provincial and hospital laboratories across Canada (Table 3): 61.4% (628/1,023) were influenza A (H3N2), 27.0% (276/1,023) were influenza A (H1N1), and 11.6% (119/1,023) were influenza B viruses. Of the 628 influenza A (H3N2) isolates characterized, 85.4% (536/628) were antigenically similar to A/Wisconsin/67/2005 (H3N2) and 14.6% (92/628) showed reduced titres to A/Wisconsin/67/2005 (H3N2). A/Wisconsin/67/2005 (H3N2) was the influenza A (H3N2) component in the 2006-2007 influenza vaccine. All 276 influenza A (H1N1) viruses that were antigenically characterized were similar to A/New Caledonia/20/1999 (H1N1), which was the influenza A (H1N1) component in the 2006-2007 influenza vaccine. Influenza B viruses currently circulating can be divided into two antigenically distinct lineages represented by B/Yamagata/16/1988-like and B/Victoria/2/1987-like viruses. Of the 119 influenza B viruses characterized, 10.1% (12/119) belonged to the B/Victoria lineage and were antigenically similar to the 2006-2007 vaccine strain B/Malaysia/2506/2004 and 89.9% (107/119) belonged to the B/Yamagata lineage and were antigenically similar to B/Shanghai/361/2002, the influenza B component of the 2005-2006 season influenza vaccine. Predominant strains for the 10 seasons from 1997 to 2007 in Canada are presented in Figure 3.
Figure 2. Proportionate distributions of case-by-case data, by influenza type and by age group, Canada, 2006-2007
A rapid assay has been established at the NML to test influenza A viruses for resistance to amantadine. Polymerase chain reaction (PCR) and sequence analysis were used to detect drug resistance markers among 1,027 influenza A viruses (including 273 A (H1N1) and 754 A (H3N2) viruses) during the 2006-2007 season. Of the total 1,027 influenza A isolates tested, 27.4% were resistant to amantadine, including 1.1% (3/273) resistance among the A (H1N1) and 36.9% (278/754) among the A (H3N2) viruses. The remaining 76.2% (270/273) A (H1N1) and (476/754) A (H3N2) viruses were sensitive to amantadine (Table 4).
Table 3. Distribution of influenza strains, by virus type and province/territory, Canada, 2006-2007
Influenza type |
Nfld. |
P.E.I. |
N.S. |
N.B. |
Que. |
Ont. |
Man. |
Sask. |
Alta. |
B.C. |
Y.T. |
N.W.T. |
Nun |
Total |
A/New Caledonia/20/ 1999(H1N1)-like |
12 |
5 |
2 |
4 |
31 |
3 |
55 |
68 |
94 |
1 |
1 |
276 |
||
A/Wisconsin/67/ 2005(H3N2)-like |
11 |
2 |
4 |
9 |
33 |
327 |
15 |
39 |
32 |
153 |
1 |
2 |
628 |
|
Total (Type A) |
23 |
2 |
9 |
11 |
37 |
358 |
18 |
94 |
100 |
247 |
2 |
1 |
2 |
904 |
B/Malaysia/2506/ 2004-like* |
3 |
4 |
1 |
3 |
1 |
12 |
||||||||
B/Shanghai/361/ 2002-like |
1 |
6 |
47 |
20 |
7 |
5 |
21 |
107 |
||||||
Total (Type B) |
0 |
0 |
1 |
6 |
50 |
24 |
0 |
8 |
5 |
24 |
1 |
0 |
0 |
119 |
Total |
23 |
2 |
10 |
17 |
87 |
382 |
18 |
102 |
105 |
271 |
3 |
1 |
2 |
1,023 |
*B/Malaysia/2506/2004-like belongs to the B/Victoria/02/87 lineage and is the recommended influenza B component for the Northern Hemisphere 2006-2007 influenza vaccine.
Figure 3: Seasonal distribution of case-by-case data by influenza type and week of onset, Canada, 1997-2007
Number of Influenza Virus Detections
The influenza virus neuraminidase (NA) inhibitors have recently been licensed for the prophylaxis and treatment of influenza virus infection in humans. In order to monitor the potential development of resistance, a new chemiluminescent (CL) neuraminidase assay was established at NML to test influenza field isolates for susceptibility to oseltamivir (Tamiflu®). A total of 1,048 influenza isolates were screened for oseltamivir resistance, (including 275 A (H1N1), 648 A (H3N2) and 125 influenza B viruses), of which only one A (H3N2) virus was resistant to oseltamivir (0.15% of A (H3N2) viruses tested and 0.10% of all influenza viruses tested).
Table 4. Amantadine susceptibility results, by province/territory and influenza A sub-type, Canada, 2006-2007
Influenza sub-type |
S/R* |
Nfld. |
P.E.I. |
N.S. |
N.B. |
Que. |
Ont. |
Man. |
Sask. |
Alta. |
B.C. |
Y.T. |
N.W.T. |
Nun |
Total |
Type A (H1N1) |
S |
12 |
5 |
4 |
4 |
29 |
2 |
55 |
69 |
88 |
1 |
1 |
270 |
||
R |
2 |
1 |
3 |
||||||||||||
Type A (H3N2) |
S |
4 |
5 |
40 |
315 |
18 |
32 |
25 |
34 |
3 |
476 |
||||
R |
8 |
2 |
4 |
13 |
17 |
68 |
7 |
10 |
145 |
1 |
2 |
1 |
278 |
||
Total (type A) |
S |
16 |
0 |
5 |
9 |
44 |
344 |
20 |
87 |
94 |
122 |
1 |
1 |
3 |
746 |
R |
8 |
2 |
4 |
13 |
17 |
70 |
0 |
7 |
10 |
146 |
1 |
2 |
1 |
281 |
*S = sensitive; R = resistant
3) Sentinel primary care consultation rates of ILI
ILI consultations reported by sentinel practitioners
Overall, FluWatch sentinel ILI site recruitment represented most of the well-populated urban and rural regions across Canada, meeting or exceeding the FluWatch objective of one sentinel per 250,000 population. However, sentinel recruitment for the national system was less than representative in Quebec where a total of eight sentinels (approximately 1/950,000 population) were recruited for the FluWatch program in five (28%) of the 18 health regions. In the remaining provinces and territories, 263 sentinels (approximately 1/95,000 population) were recruited for the FluWatch program and/or collaborating provincial sentinel programs, representing 72% of census divisions outside of Quebec (136/189). On average, CIRID received weekly ILI data from 60% of FluWatch sentinels across Canada (162/271) for the entire 2006-2007 season and 68% (185/271) during the active season only (from October 2006 to mid-May 2007). During the entire 2006-2007 season, 66% of the sentinels (179/271) provided ILI data for at least 50% of the reporting weeks (77% in the active season), and 4% (12/271) provided ILI data for at least 90% of the reporting weeks (8% in the active season).
During the 2006-2007 influenza season, ILI consultation rates increased around week 48 (late November to early December) and peaked in week 9 (late February to early March) with a rate of 50 per 1,000 patient visits. Between week 40 and week 20 (early October 2006 to mid-May 2007), ILI rates remained within or below the expected range, with the exception of week 18 when the ILI rate was higher than expected, based on mean observation rates for the 10 previous seasons (Figure 4). During the active season, an average of 20 ILI consultations were reported per 1,000 patients seen. The highest ILI consultation rates were reported in children, averaging 40/1,000 patients seen in the 0 to 4 year age group and 30/1,000 in those aged 5 to 19 years.
Figure 4. ILI consultation rates, by week, Canada, 2006-2007, compared to baseline (1996-1997 to 2005-2006 seasons)
Note: No data available for mean rate in previous years for weeks 19 to 40 (1996-1997 through 2001-2002).
4) Regional influenza activity levels as assigned by provincial and territorial FluWatch representatives
Sporadic influenza activity was first reported in British Columbia, Alberta and Ontario in early September. Localized influenza activity was first reported in the central eastern region of Ontario during week 44 (late October/early November), followed by the central region in Alberta in week 46 (mid-November). The number of provincial/territorial designated influenza surveillance regions across Canada reporting localized activity increased gradually over the following weeks, with widespread activity first reported during week 51 in Alberta. The proportion of influenza surveillance regions reporting localized or widespread activity peaked during week 9 at 56%. Most reporting of influenza activity levels (65% of widespread activity reported and 53% of localized activity reported) occurred over an 8-week period from week 7 to week 14 (mid-February to early April of 2007) (Figure 5). All 10 provinces and the Northwest Territories reported either widespread or localized activity in at least one region during weeks 7 to 14.
Provincial and territorial weekly participation in reporting of activity levels remained high throughout the season, with timely activity level assessments reported to CIRID for all provincial/territorial designated influenza surveillance regions in 77% of the reporting weeks (40/52) during the 2006-2007 season.
During the 2006-2007 influenza season, nine provinces and one territory reported a total of 180 laboratory-confirmed influenza outbreaks in long-term care facilities (LTCF). Most of these (128 or 71%) were reported between week 5 and week 12 (late-January to mid-March of 2007). The peak in outbreak reports occurred in week 9, when 38 new LTCF outbreaks were reported in eight provinces (Figure 6). Outbreaks of influenza or ILI were also reported in hospitals (n = 11, in three provinces), schools (n = 268, in seven provinces) and other types of facilities (n = 32, in five provinces).
Figure 5. Number of surveillance regions reporting localized or widespread influenza activity, by week, Canada, 2006-2007
Figure 6. Number of outbreaks reported in long-term care facilities (LTCF), by week, Canada, 2006-2007
5) Paediatric influenza-associated hospital admissions and mortality data
During the 2006-2007 influenza season, 12 IMPACT hospitals in eight provinces reported 370 influenza-associated paediatric hospitalizations. Influenza A represented 84.1% of the total number of hospitalized paediatric cases (311/370), and influenza B represented 15.1% of cases (56/370). Three cases (0.8% or 3/370) were infected with both influenza A and B. Influenza A infections accounted for the majority of hospitalizations for most of the season; however, the last few hospitalizations reported for the season (in late May/early June) were for influenza B infections. Hospitalizations due to influenza A infection peaked during week 9 (late February/early March; n = 41), whereas those due to influenza B infection peaked in week 14 (early April; n = 7) (Figure 7). Of the 370 paediatric hospitalizations, two resulted in deaths attributable to influenza or influenza-related complications (case fatality = 0.5%) and were reported by two different centres in two provinces. One of the deaths was attributed to influenza A infection (H1N1 subtype) and the other death was attributed to influenza B infection.
Figure 7. Weekly paediatric admissions to IMPACT hospitals, by influenza type, Canada, 2006-2007
Reporting of influenza hospitalizations in children began at different times in the 12 IMPACT hospitals. The first hospital to report a case was in Edmonton in week 37 (mid-September), from influenza A followed by the first cases in Toronto (influenza B) and Montreal (influenza A) in weeks 46 and 47. All of the 12 centres had reported at least one case by week 6 (early February). The number of cases reported per week peaked in week 9 (late February to early March) with 43 cases: 41 influenza A, one influenza B and one influenza A and B co-infection. The last cases of the season were reported by hospitals in Quebec and Alberta in weeks 22 and 23 (late May to early June) respectively; both cases were infected with influenza B.
The distribution of hospitalized cases by age group was as follows: 19.7% of cases (73/370) were aged 0 to 5 months; 29.7% (110/370) were 6 to 23 months; 25.9% (96/370) were 2 to 4 years; and 24.6% (91/370) were ≥ 5 years. Full or partial vaccination details were available for 88.4% of cases (327/370). Although overall, only a small proportion of children were immunized, the proportion of children ≥ 6 months of age immunized for influenza increased by age group (Table 5). The distribution of hospitalized cases by influenza type and by age group is presented in Figure 8.
Table 5: Influenza-associated paediatric admissions to IMPACT hospitals, by age and immunization status, Canada, 2006-2007
Age group |
Number (%) of laboratory-confirmed influenza admissions |
Number (% by age group) of admissions with at least partial vaccination details |
Number (% by age group) immunized against influenza prior to admission |
0-5 months |
73 (19.7%) |
73 (100%) - n/a* |
0 (0%) n/a* |
6-23 months |
110 (29.7%) |
97 (88.2%) |
15 (15.5%) |
2-4 years |
96 (25.9%) |
84 (87.5%) |
17 (20.2%) |
5 + years |
91 (24.6%) |
73 (80.2%) |
20 (27.4%) |
Total |
370 |
327 |
52 |
* According to NACI recommendations, immunization with currently available influenza vaccines is not recommended for infants < 6 months of age.
Figure 8. Distribution of paediatric admissions to IMPACT hospitals, by type and age group, Canada, 2006-2007
United States: (Centers for Disease Control and Prevention) (CDC)(2)
During the 2006-2007 season, influenza A (H1) viruses were most commonly isolated, but influenza A (H3) viruses were more frequently identified than influenza A (H1) viruses from early March through May. A small number of influenza B viruses were also identified. Nationally, low levels of influenza activity were reported during October through mid-December, increased during January, and peaked in mid-February.
Between week 40/2006 and week 20/2007 (1 October 2006 to 19 May 2007), 13.2% of the specimens tested (23,753/179,268) for influenza viruses by the WHO and the National Respiratory and Enteric Virus Surveillance System laboratories in the United States (US) were positive. Of the 23,753 influenza viruses tested, 79.2% (18,817/23,753) were influenza A viruses, and 20.8% (4,936/23,753) were influenza B viruses. Influenza activity in the US peaked in mid-February.
For the 2006-2007 season, the CDC antigenically characterized 1,107 influenza viruses: 486 influenza A (H1), 289 influenza A (H3) and 332 influenza B viruses. Of the 486 influenza A (H1) viruses: 90.3% (439/486) were characterized as A/New Caledonia/20/1999 (H1N1)-like. Of the 47 low reacting H1N1 viruses, 9.3% (45/486) showed reduced titres with antisera produced against A/New Caledonia/20/1999 (H1N1) and are similar to A/Solomon Islands/3/2006 (H1N1), which is a recent antigenic variant of A/New Caledonia/20/1999 (H1N1) and 0.4% (2/486) showed reduced titres with antisera produced against A/New Caledonia/20/1999 (H1N1) and A/Solomon Islands/3/2006 (H1N1). Of the 289 influenza A (H3) viruses, 23.9% (69/289) were characterized as A/Wisconsin/67/2005 (H3N2)-like and 76.1% (220/289) showed reduced titres with antisera produced against A/Wisconsin/67/2005 (H3N2). Of the 332 influenza B viruses that have been characterized, 23.5% (78/332) belonged to the B/Yamagata lineage. The majority of influenza B viruses characterized (76.5% or 254/332) belonged to the B/Victoria lineage: 128 (50.4%) were similar to B/Ohio/1/2005 and 126 (49.6%) showed reduced titres with antisera produced against B/Ohio/1/2005.
From 1 October 2006 through 30 September 2007, 73 deaths from influenza in children were reported to CDC. Data on the presence (or absence) of bacterial co-infections were recorded for 69 of these cases of which 30 (44%) had a bacterial co-infection and 22 (73%) of these were infected with Staphylococcus aureus (15 of the 22 or 68% had infections with methicillin-resistant S. aureus (MRSA)). The number of pediatric influenza-associated deaths reported during 2006-2007 was moderately higher than the number reported during the 2 previous surveillance years; the number of these deaths in which pneumonia or bacteremia due to S. aureus was noted represents a five-fold increase. Children with S. aureus co-infection were more likely to have pneumonia and Acute Respiratory Distress Syndrome and had a median age that was higher than children without S. aureus co-infection. The influenza strains isolated from these children were not different from common strains circulating in the community and the MRSA strains were similar to those associated with MRSA skin infection outbreaks in the US(3).
International: WHO(4, 5)
Between September 2006 and August 2007, the level of influenza activity was generally mild to low. In North America, influenza activity began in November 2006, which was late compared with previous years, and increased in December. In Asia and Europe, influenza activity began in December 2006 and increased in January 2007. Overall influenza activity in the northern hemisphere declined in April and May 2007. In the southern hemisphere, influenza activity began in April 2007 in South America, increased in May through July, and declined in August. In Oceania and South Africa, influenza activity started in June 2007, peaked in July to August, and declined in September.
Influenza A (H1N1) viruses circulated and caused outbreaks in some countries of Asia, eastern Europe, North America and Oceania. Viruses characterized earlier in the season were antigenically closely related to A/New Caledonia/20/1999; however, the majority of later season isolates characterized were antigenically similar to A/Solomon Islands/3/2006.
Influenza A (H3N2) viruses circulated and caused outbreaks in many countries of Asia, Europe, South America and Oceania. Although some of the viruses characterized were antigenically similar to the vaccine virus A/Wisconsin/67/2005, the proportion of viruses antigenically similar to A/Brisbane/10/2007 increased.
Influenza B viruses circulated and caused outbreaks in some countries in Asia, eastern Europe and South America. Both B/Victoria/2/1987 and B/Yamagata/16/1988 lineage viruses were detected in many countries but occurred in varying proportions.
Avian Influenza(6)
From 1 September 2006 to 30 November 2007, the WHO reported a total of 93 cases (65 fatal) of human infection with avian influenza A (H5N1) from eight countries including Nigeria and Lao People's Democratic Republic where the cases reported were the first cases identified in the country. Since the first cases were reported in 1997 to the end of November 2007, a total of 335 cases have been reported (of whom 206 have died: 61% fatality) from the following countries: Viet Nam, Thailand, Cambodia, Indonesia, Azerbaijan, China, Djibouti, Egypt, Iraq, Turkey, Nigeria and Lao People's Democratic Republic.
More detailed epidemiologic information was reported on identified cases up until 24 November 2006 (n = 256)(7). Fifty-two percent (132/256) of the cases were < 20 years of age and 89% (227/256) were < 40 years. The ratio of males (n = 129) to females (n = 127) was 1.0. The overall case fatality rate (CFR) was 60%, with the highest CFR (76%) found among those aged 10 to 19 years and the lowest CFR (40%) found among those aged ≥ 50 years. The number of cases peaked during the cooler months of the northern hemisphere.
Evidence to date indicates that close contact with dead or sick birds is the principal source of human infection with the H5N1 virus, and avian influenza H5N1 infection in humans remains a rare disease(8). To date, the WHO influenza pandemic preparedness level remains at Phase 3 where there is a new influenza virus subtype that is causing disease in humans, but not yet spreading efficiently and sustainably among humans(9).
Overall, the 2006-2007 influenza season was relatively mild (in terms of a lower proportion of positive laboratory tests for influenza, fewer LTCF outbreaks and fewer influenza surveillance regions reporting localized and widespread activity), and, similar to the 2005-2006 season, influenza activity presented later in the season. Influenza A predominated and had a greater impact on the very young and elderly compared to the 2005-2006 season.
Influenza viruses began circulating late in Canada in the 2006-2007 season, as demonstrated by the late peak in laboratory-confirmed cases (week 10 or early March) compared with the median peak from previous seasons (weeks 5 to 6 or early to mid-February). The first peak in laboratory-confirmed cases was observed in the Prairies in week 50 (mid-December) followed by peaks in British Columbia (week 5), Ontario (week 9) and in Quebec and the Atlantic region (week 10). Influenza viruses were detected over a longer period of time in British Columbia and Ontario (48 of 52 weeks) and in Alberta (45 of 52) than the rest of the country.
The total number of influenza tests performed in Canada for the 2006-2007 season (n = 100,864) was higher than observed for the previous season (n = 87,303). The per capita influenza testing rate in Canada for the 2006-2007 season was 313 influenza tests per 100,000 population. Regionally, the highest per capita influenza testing rates were in Alberta and Saskatchewan (796 and 602 tests per 100,000 respectively) and lowest in Prince Edward Island, Nova Scotia and British Columbia (100, 107 and 115 per 100,000 respectively). An increase in per capita influenza testing rates have been observed in Canada over the last several years and may be due to the increased awareness of the importance of influenza, present concerns over emerging viruses (e.g. human infection with avian influenza) and pandemic preparedness, as well as the increasing ease and use of rapid tests for influenza diagnosis.
The overall percentage of positive tests in the 2006-2007 season (8.1%) was similar to the 2005-2006 season (8.5%) but was lower than the 2004-2005 and 2003-2004 seasons (12.7% and 12.3% respectively). This is consistent with the milder seasons. Over the 11 seasons from 1996-1997 to 2006-2007, the percentage of positive influenza tests has ranged from 8.1% and 12.7% for predominantly influenza A seasons (n = 7). During the 2006-2007 season, the peak in percentage of positive influenza tests was 21.3% (942/4,418 in week 10) compared with 20.4% (649/3,181 in week 11) in 2005-2006.
Of the total positive influenza identifications, 87.5% were influenza A viruses and 12.5% influenza B. Influenza A virus detections predominated for the majority of the season except from weeks 15 to 21 (early April to end of May) when the proportion of influenza B detections was slightly higher than the proportion of influenza A detections. In contrast, during the 2005-2006 season, both influenza A and B detections predominated up until week 8 (mid-February) after which influenza A detections predominated.
The peak in RSV detections occurred earlier in the season (30% in week 52) compared to the peak in influenza detections (21% in week 10). The percent positive for RSV detections for the entire 2006-2007 (11.5%) was similar to previous seasons (7.1% in 2005-2006 and 9.0% in 2004-2005). There were no evident peaks observed for percent positive detections for parainfluenza viruses and adenoviruses in 2006-2007. Weekly percent positive detections remained below 7% for parainfluenza viruses and below 4% for adenoviruses throughout the season. The percent positive for parainfluenza and adenovirus detections for the entire 2006-2007 (3.4% and 1.9% respectively) were similar to previous seasons (4.0% in 2005-2006 and 2.7% in 2004-2005 for parainfluenza viruses and 2.4% in 2005-2006 and 2.6% in 2004-2005 for adenoviruses).
The NML characterized 1,023 influenza viruses representing 12.6% of the total number of influenza virus detections in Canada for the 2006-2007 season. Of these, influenza A (H3N2) accounted for 61.4% (628/1,023) and influenza A (H1N1) for 27.0% (276/1,023). A small number of influenza B viruses were also characterized (119 or 11.6%), the majority of which were isolated in Quebec, Ontario and British Columbia. The majority of influenza A (H3N2) and A (H1N1) viruses characterized during the 2006-2007 season were identified as A/Wisconsin/67/2005 (H3N2)-like and A/New Caledonia/20/1999 (H1N1)-like viruses respectively, which were the recommended influenza A components included in the 2006-2007 influenza vaccine. Only 10.1 % of all influenza B isolates characterized (12/119) matched the B/Malaysia/2506/2004-like strain, which was the recommended influenza B component included in the 2006-2007 vaccine. The remaining 89.9% (107/119) were B/Shanghai/361/2002-like viruses belonging to the B/Yamagata lineage. While influenza A (H3N2) viruses predominated in Canada and Europe, influenza A (H1N1) viruses accounted for the majority of influenza detections in the US.
During the 2006-2007 influenza season, a lower level of resistance compared to the 2005-2006 season to the antiviral amantadine was detected among the circulating influenza A (H3N2) viruses, however the level of resistance observed was not as high as what was observed in the 2005-2006 season (91.5% or 463/506). On the basis of amantadine resistance patterns observed to date, PHAC continues to recommend against the use of amantadine for the treatment and prevention of influenza(10). During the 2006-2007 influenza season, the NML also monitored antiviral resistance to neuraminidase inhibitors among circulating influenza viruses. Only one A (H3N2) virus was found to be resistant to oseltamivir among the 1,048 influenza A and B viruses tested. The resistant isolate came from an immunocompromised individual who was treated with both oseltamivir and amantadine for influenza and pneumonia. An early isolate sent for antiviral sensitivity testing, showed resistance to amantadine but was sensitive to oseltamivir. As the patient's condition did not improve, repeat testing on a second isolate showed resistance to both amantadine and oseltamivir indicating that the development of oseltamivir resistance was treatment-associated. The development of antiviral resistance is not unexpected, particularly in immunocompromised individuals who may receive protracted and repeated treatment courses. These findings underscore the importance of influenza antiviral sensitivity testing to monitor for resistance in Canada and to inform appropriate use of antiviral medications to reduce the impact of influenza and to contribute to the global understanding of antiviral resistance patterns.
Although a large proportion of cases in the 2006-2007 season were among children < 15 years of age (39%), the proportion was slightly lower than that observed in the previous season (45%). Rather, a larger proportion of cases observed in the 2006-2007 season were among adults ≥ 25 years (56%) compared to the previous season (41%). The age distribution of cases varied by influenza type, with the majority of influenza A cases having occurred in children < 5 years (23%) followed by those 25 to 44 years (22%) and ≥ 65 years (20%) while the majority of influenza B cases (68%) occurred in adults ≥ 25 years of age.
ILI consultation rates remained below the mean rate of the previous nine seasons combined for most weeks of the 2006-2007 influenza season, except in weeks 48, 9 (peak in ILI) and 18. Peak ILI consultations also coincided with peak numbers of regions reporting widespread and localized influenza activity and peak numbers of paediatric hospitalizations due to influenza infection. The peak ILI consultation rate was 50 ILI consultations per 1,000 patient visits, similar to that of the previous season. The overall ILI rate during the active influenza season for 2006-2007 (20 cases of ILI per 1,000 patients seen) was higher than the rate for 2005-2006 (15 per 1,000 patients seen). In both seasons the highest rates of ILI were observed in children: 47 per 1,000 in the 0 to 4 year age group and 32 per 1,000 in the 5 to 19 age group in 2006-2007 compared with 36 per 1,000 and 32 per 1,000 respectively in 2005-2006.
The number of influenza surveillance regions reporting localized or widespread activity peaked during week 9, which is earlier compared with the previous season peak in week 13. The number of influenza outbreaks reported in LTCFs in 2006-2007 (n = 180) was similar to the number reported in the 2005-2006 season (n = 167); however, both were lower than the number reported in the 2004-2005 season (n = 762) and consistent with overall mild seasons.
The number of paediatric hospitalizations reported by IMPACT sites in 2006-2007 was similar to that in the previous season (370 hospitalizations in 2006-2007 compared with 374 in 2005-2006). However, there were fewer paediatric deaths reported in 2006-2007 (two deaths) than in the previous season (five deaths). Influenza A accounted for a greater percentage of hospitalizations in 2006-2007 than in the previous season (84.1% versus 61.8%). In the 2006-2007 season, the largest proportion of hospitalizations were among children < 2 years of age (49.5% or 183/370); however, the majority of hospitalizations in the 2005-2006 season were in children 2 years (61.5% or 230/374).
Results from the influenza surveillance system should be interpreted with caution for several reasons:
The authors gratefully acknowledge and thank all the FluWatch surveillance partners who participated in the FluWatch program during the 2006-2007 season. A special thank you to Peter Zabchuk, Teresa Leung, Hui Zheng, Julie Bettinger and Wendy Vaudry for their contributions to this report; the Poultry Research Facility of the University of Manitoba for providing embryonated eggs; and Estelle Arseneault for the translation of this report.