No announcement yet.

Review of the 2012 winter influenza season, southern hemisphere (WHO, November 2 2012): Variable activity seen with an increased but no exceptional mortality

  • Filter
  • Time
  • Show
Clear All
new posts

  • Review of the 2012 winter influenza season, southern hemisphere (WHO, November 2 2012): Variable activity seen with an increased but no exceptional mortality

    [Source: World Health Organization, full PDF document: (LINK). Edited.]

    Weekly epidemiological record / Relevé épidémiologique hebdomadaire

    2 NOVEMBER 2012, 87th year / 2 NOVEMbre 2012, 87e année / No. 44, 2012, 87, 421–436 -

    Review of the 2012 winter influenza season, southern hemisphere

    This report summarizes the chronology, epidemiology and virology of the winter influenza season in the temperate regions of the southern hemisphere, from January 2012 to the end of September 2012. The data presented have been derived primarily from reports published by national ministries of health or other official bodies reporting on their behalf, or reported to the WHO through FluNet and FluID.(1, 2)

    Southern cone of South America


    There were notable differences in the timing of transmission and the viruses circulating in the 4 countries of the southern cone of South America in the 2012 influenza season. Active influenza transmission in the area was first noted in late April to early May in both Chile and Paraguay, whereas Argentina and Uruguay did not report active transmission until late May or early June. Similarly, peak activity occurred about 8 weeks later in mid-June in Chile and Paraguay but not until early August in Argentina and Uruguay. In Paraguay, influenza virus had become practically undetectable by mid-September 2012 whereas Chile, Argentina and Uruguay were still reporting lower numbers of virus detections through the end of September. The 2012 season peaked approximately 10 weeks later than the historical 9 years average for Argentina (Figure 1).

    The predominant influenza virus circulating in Chile and Uruguay was A(H3N2), whereas in both Paraguay and Argentina transmission was primarily associated with A(H1N1)pdm09, and very little A(H3N2) was seen. Influenza type B virus was detected in only small numbers in all 4 countries (Map 1). In Chile, during the period from 1 January 2012 to 20 September 2012, influenza A accounted for 87% (1230/1420) of influenza viruses reported and influenza B for 13% (190/1420). Influenza A(H3N2) accounted for >99% of the 1066 influenza A viruses with subtype information; A(H1N1)pdm09 was detected in only 9 samples. In Argentina, during the same period, influenza A accounted for 83% (273/374) of influenza viruses detected and influenza B for 27% (101/374). In marked contrast to the season in Chile, influenza A(H1N1)pdm09 accounted for 89% (173/195) of the influenza A viruses with subtype information.

    The distribution of virus types and subtypes in Paraguay was nearly identical to that in Argentina. Of note, the virus distribution pattern observed in Argentina and Paraguay was very similar to that seen in most countries of the northern tropical areas of South America where influenza A(H1N1)pdm09 was the most commonly detected virus for most of the year.


    Illness and mortality

    The overall severity and impact of influenza during the 2012 season in the southern cone appeared to be less than in the previous 2 years. A large portion of the influenza-like illness (ILI) and severe acute respiratory infection (SARI) cases occurring in the area were likely associated with respiratory syncytial virus (RSV), which also circulated during much of the winter. In Chile, ILI peaked in the first week of July, with a rate of 19.4 consultations per 100 000 population. This peak was greater than that seen in 2011 (11.2 per 100 000) but lower than seen in 2007 through 2010. The highest peak in ILI consultations during the past 5 years was during the pandemic of 2009 when ILI consultations peaked at 206.5 per 100 000. Emergency department visits for a respiratory cause were less for most of the season than those recorded in 2010 for both adults and children but greater than in 2011. At the peak of weekly SARI cases in early July, 7.2% of all hospitalizations were due to SARI; during the peak, influenza viruses were found in 27% (43/157) of all SARI cases.(3) Of the SARI cases positive for influenza A(H3N2), 12% required admission to intensive-care units (ICU).The number of influenza deaths reported thus far in 2012 (n=16) is similar to the total reported in 2011 (n=20) but significantly lower than the 2010 total (n=81).(4) Of the 16 reported deaths associated with influenza in 2012, 69% (11/16) were found to have influenza A(H3N2), 3 influenza A(not subtyped) and 2 influenza B.

    The median age of fatal influenza cases was 81 years; 63% (n=10) were female; and 80% had a history of a comorbid medical condition.

    In Paraguay, ILI consultation rates coincided with a peak in RSV in mid to late July with influenza detection peaking more than 4 weeks earlier. The total number of ILI consultations in this period was approximately double the number of ILI cases in the same period in the previous 2 years. The maximum impact of influenza on ILI and SARI occurred in late May, when 36% (38/105) of weekly ILI and 32% (23/72) of SARI respiratory samples were positive for an influenza virus. The number of SARI deaths reported through September 2012 was 226. Of the 32 deaths with a laboratory confirmed virus, 18 were found to have influenza A(H1N1) pdm09 and 10 had RSV.(5)

    In Argentina, the cumulative number of cases of ILI and pneumonia reported by the end of August was lower than for the same period in 2010. Argentina also experienced a significant RSV season in 2012, with weekly virus detections peaking higher than the previous 2 years.

    Between January and August 2012, RSV accounted for approximately 90% of all laboratoryconfirmed respiratory viruses, whereas influenza accounted for only 4% (748/17734). The predominance of RSV is also represented in the age distribution of ILI; 89% of ILI cases were seen in the age group .4 years.(6) In Uruguay, SARI cases in 2012 were primarily reported in the .5 years and .65 years age groups. Influenza A(H3N2) accounted for 52% (13/25) of influenza-associated SARI cases in 2012, with influenza B accounting for the remaining 48% (12/25). The majority of SARI cases requiring ICU admission were in the .65 years age group; of these influenza A(H3N2) was found in 71% (5/7) and influenza type B in the remaining 2.

    South Africa


    In South Africa, active influenza virus transmission was first noted in late May, peaking in mid-August. At the peak of the season, 68% of all ILI samples from sentinel sites of the Viral Watch Program tested positive for influenza.(7) This is a similar rate to that seen in the previous 2 years but occurred about 9 weeks later than median for the past 7 years (Figure 1).

    The SARI cases that were positive for influenza virus reached a peak 1 to 2 weeks later than for ILI and were associated with a notable increase in influenza type B virus detections. In the period 1 January 2012 to 30 September 2012, influenza A accounted for 68% of all influenza viruses detected (543/853) and influenza type B 36% (310/853) among ILI cases.

    Of the influenza A viruses with subtype information, A(H3N2) accounted for 99% (423/427).

    Influenza A transmission peaked about 4 weeks earlier than influenza B resulting in a marked predominance of A(H3N2) in the first half of the season transitioning to a marked predominance of influenza type B in the latter half. The proportions of virus types differed somewhat among SARI cases compared with ILI; among influenza-associated SARI in 2012, influenza A was found in 51% (121/236) and influenza type B in 49% (115/236) of cases. However, it should be noted that the provinces included in the ILI and SARI systems do not completely overlap.

    Australia and New Zealand


    The timing of the 2012 influenza season was similar in Australia and New Zealand. Active transmission of influenza in both countries was first noted in early June and peaked in mid-July. Australia experienced sustained virus circulation until late August before steadily decreasing whereas New Zealand reported declines in virus detections 2 weeks earlier, in early August. Virus detections in 2012 for the 2 countries peaked 5 weeks earlier than the median week of peak transmission for the period of 2003 through 2011.

    The most commonly detected influenza virus in the 2 countries was influenza A(H3N2). Australia, however, reported much more co-circulation of influenza type B virus, which differed markedly between jurisdictions within the country. Influenza type B accounted for approximately one-third of all positive virus detections in the Northern Territory, Queensland and Western Australia, whereas in Victoria influenza type B accounted for only 10% of all virus detections. Nationwide, between 1 January to 20 September 2012, influenza A viruses were 70% (3043/4323) of all viruses detected and influenza B, 30% (1280/4323). Of influenza A viruses with subtype information, 97% (2878/2965) were A(H3N2) and influenza A(H1N1)pdm09 only 3% (87/2965).(1)

    In New Zealand, influenza B was detected much less frequently, accounting for only 8% (156/1860) of all influenza viruses nationally. New Zealand also found a higher proportion of influenza viruses to be A(H1N1) pdm09; of the influenza A viruses with subtype information, 86% (1171/1369) were A(H3N2) and 14% (198/1369) A(H1N1)pdm09.(8) As was observed in Australia, regional variation of strain predominance was also reported in New Zealand.

    Illness and mortality

    In 2012, the overall impact of the influenza season appeared to be somewhat higher than average in Australia according to some indicators, though not exceptionally. At the seasonal peak, the number of influenza cases reported per week and ILI consultation rates were higher than in any previous season since 2005, except for the season in which the new pandemic virus appeared in 2009. In the New South Wales Registered Death Certificates data, the rate of deaths classified as influenza and pneumonia met or exceeded the epidemic threshold for most of July, which was higher than in the previous 2 years but lower than in 2007 and 2008.

    The number of laboratory confirmed influenza-associated deaths reported to the Nationally Notified Deaths Surveillance System (NNDS) was more than 3 times the number reported in 2011 (43 versus 14 respectively).(9)

    The Influenza Complications Alert Network (FluCAN), a sentinel system that includes hospitals in Victoria, South Australia, Western Australia and the Australian Capital Territory, reported that 85% of influenza-associated hospital admissions had influenza A and 15% influenza B. The median age of patients hospitalized with influenza was 61 years, with 45% of cases aged .65 years and 75% reported to have a pre-existing comorbidity. Of all influenza-associated admissions, 9% required ICU admission, somewhat less than in 2011 or 2010 (14% and 21% respectively). Between 1 July and 31 August 2012, 28 children aged .15 years were hospitalized with severe influenza complications; this number included 8 ICU admissions.

    More than 60% of paediatric influenza hospitalizations were associated with influenza A and more than one-third of these cases had an underlying chronic comorbidity. The median age of patients who died with laboratory-confirmed influenza reported to the NNDS was 80 years (compared with 45 years in 2011 when influenza A(H1N1)pdm09 was the most commonly detected virus). Almost all deaths were reported as having influenza A (not subtyped) or A(H3N2), with the A (not subtyped) infections most likely attributable to A(H3N2) as very few A(H1N1) pdm09 viruses have been reported in 2012.

    In New Zealand, the impact of the 2012 influenza season was also somewhat higher than average compared to previous years. The peak consultation rate in 2012 was 154.1/100 000 population which was the highest rate since 2005 (excluding 2009) but similar to that observed in 2010.

    From 30 April to 2 September 2012, both the national International Classification of Diseases (ICD) coded influenza hospitalizations and SARI hospitalizations for the Auckland and Counties Manukau District Health Boards recorded the highest influenza hospitalization rates in the age group .1 year, with rates of 152.3 per 100 000 and 228.7 per 100 000 age group population respectively. This was followed by age group .80 years (87.3 versus 128.6 per 100 000 respectively), children aged 1.4 years (39.7 versus 54.9 per 100 000 respectively), age group 65.79 years (32.0 versus 81.1 per 100 000 respectively) and age group 50.64 years (16.3 versus 37.7 per 100 000 respectively). The distribution of influenza hospitalizations by ethnicity in 2012 was also analysed by using the ICD coded and SARI hospitalization data. The analysis showed that Pacific Peoples had the highest hospitalization rate (68.4 versus 83.3 per 100 000 respectively), followed by Maori (22.6 versus 42.2 per 100 000 respectively), Asian (17.2 versus 16.7 per 100 000 respectively) and European or Other (15.5 versus 18.3 per 100 000 respectively).

    Between 30 April and 2 September 2012, 22% (311/1410) of all SARI specimens tested were positive for an influenza virus. Of note, among SARI cases reported to the Southern Hemisphere Influenza Vaccine Effectiveness Research and Surveillance (SHIVERS) project since April 2012, influenza B and A(H1N1)pdm09 were both proportionately higher than in ILI cases reported by general practitioners. While 78% of SARI cases were associated with influenza A viruses (versus. 92% of ILI cases), only 43% (104/244) of influenza A viruses with subtype information were A(H3N2) and 41% (99/244) A(H1N1)pdm09 (in ILI cases, the proportions were 86% and 14% respectively). It should be noted, however, that 42% of influenza A viruses were not subtyped and the SHIVERS project only includes data from the Auckland and Counties Manukau District Health Boards.(10)

    Antigenic characterization

    All influenza A(H1N1) viruses detected globally since January 2010 have been of the A(H1N1)pdm09 type; the influenza A(H1N1) that was circulating prior to 2009 has not been seen since that time. The majority of influenza A(H1N1)pdm09 viruses characterized antigenically by the WHO Global Influenza Surveillance and Response System were related to the vaccine virus A/California/7/2009 contained in the trivalent seasonal influenza vaccine for the 2012 southern hemisphere season. In contrast, the vast majority of A(H3N2) viruses were antigenically closely related to A/Victoria/361/2011, which will be in the northern hemisphere vaccine for 2013 but was not in the 2012 season southern hemisphere vaccine. The A(H3N2) component of the 2012 southern hemisphere vaccine was A/Perth/16/2009 (H3N2)-like. Prior to the mid-2011/12 influenza season of the northern hemisphere, nearly all circulating A(H3N2) viruses had been antigenically related to A/Perth/16/2009 (H3N2). Influenza type B viruses identified globally were both the Victoria and Yamagata lineages, with Victoria generally more prevalent but with regional variation. The majority of B/Victoria-lineage viruses were antigenically and genetically closely related to the vaccine virus (B/Brisbane/60/2008-like virus).

    The majority of viruses of the B/Yamagata/16/88 lineage were antigenically closely related to the vaccine virus B/Wisconsin/1/2010, which has been recommended for inclusion in the northern hemisphere 2012–2013 season vaccine. This represents a continued increase in the proportion of B viruses that are of the Yamagata lineage with a change in the antigenicity that has led to an additional recommendation for change in vaccine composition for the 2012–2013 season vaccine.(11)

    Antiviral sensitivity testing

    Only a small number of A(H1N1)pdm09 viruses were reported to be resistant to oseltamivir in 2012, although these were often not associated with exposure to oseltamivir. All of the oseltamivir-resistant influenza A(H1N1)pdm09 viruses detected had the H275Y substitution in the neuraminidase gene. All A(H3N2) viruses tested – except one – were sensitive to oseltamivir and zanamivir. All influenza A viruses tested remained resistant to adamantanes. With the exception of 1 influenza type B virus from Mongolia with reduced susceptibility to peramivir associated with H273Y substitution in the neuraminidase gene, all other B viruses tested were sensitive to neuraminidase inhibitors; however, in a few viruses sensitivity was reduced.(11)


    The 2012 influenza season in the temperate countries of the southern hemisphere began in May and was largely finished by the end of September, except in Argentina where influenza activity was still being reported. The most commonly detected virus in most areas was influenza A(H3N2). However, influenza virus types and subtypes varied considerably from country to country, even between neighbouring counties, and between domestic jurisdictions within a single country. In contrast to other temperate countries, Argentina and Paraguay had a predominance of influenza A(H1N1)pdm09 during this winter season, which was the same as neighbouring countries to the north in the tropical areas of South America.

    Influenza type B was a significant virus in South Africa and Australia, appearing as a second “wave” of influenza in the former and as the predominant influenza virus in some states of the latter. The previously circulating A(H1N1) has not been seen anywhere since January 2010, having been completely replaced by A(H1N1)pdm09 which now behaves as a seasonal virus.

    The impact or severity of the season was worse in some areas than in recent years, though not exceptionally so. Higher rates of mortality and consultations for ILI were noted in Australia, increased rates of ILI and SARI were observed in Chile, and increased ILI consultation rates in New Zealand. In some areas, notably Chile, the rates of these nonspecific indicators of respiratory disease may have been intensified by the co-circulation of RSV. The resurgence of A(H3N2) in Australia, New Zealand and Chile was associated with a shift in the age distribution of severe disease, compare to recent years when A(H1N1)pdm09 was the predominant virus circulating returning to the more typical seasonal influenza pattern that disproportionately affects the elderly and young infants. New Zealand noted an apparent increased risk of severe disease in some ethnic minority groups, which was also been noted during the 2009 pandemic, and prompted special measures in some countries to reach disadvantaged groups with vaccine.

    Resistance to neuraminidase inhibitors remained low but resistance to the older class of antiviral agents, the adamantanes, is universal.

    While nearly all of the A(H1N1)pdm09 viruses detected globally continued to be antigenically similar to the virus contained in the current trivalent vaccine, the recent antigenic drift in the A(H3N2) virus has resulted in a partial mismatch between the 2012 southern hemisphere vaccine and circulating A(H3N2) virus. This and the recent increases in influenza B viruses of the Yamagata lineage have resulted in a recommended change in the composition of the vaccine to be used in the upcoming northern hemisphere winter (2012–2013).

    Some countries may also license and use a quadrivalent vaccine containing 2 influenza B viruses in the upcoming season. The full recommendation of the consultation on vaccine composition can be found on the WHO website.(12)

    1. FluNet is available online at ; accessed October 2012.
    2. FluID is available online at ; accessed October 2012.
    3. PAHO Regional Influenza Surveillance is available at ; accessed October 2012.
    4. For additional information, see the Chile progress report. Available at ; accessed October 2012.
    5. For additional information, see the Paraguay epidemiological bulletin. Available at: 5&Itemid=129 ; accessed October 2012.
    6. For additional information, see the Argentina monitoring bulletin. Available at ; accessed October 2012.
    7. For additional information, see the National Institute of Communicable Diseases Influenza surveillance report. Available at ; accessed October 2012.
    8. For additional information, see the New Zealand public health surveillance influenzaweekly update. Available at ; accessed October 2012.
    9. For additional information, see the Australian influenza report. Available at: ; accessed October 2012.
    10. For additional information, see the Southern hemisphere influenza vaccine effectiveness research and surveillance report. Available at ; accessed September 2012.
    11. See No. 41, 2012, pp. 389–400.
    12. See No. 10, 2012, pp. 83–96.