The incidence of clinical influenza in Europe has remained around or below baseline levels (the levels that clinical influenza activity remains in throughout the summer and most of the winter) from week 40 up to the end of week 52 of 2005.

Only sporadic cases of laboratory confirmed influenza (n=166) have been reported to the European Influenza Surveillance Scheme (EISS, http//:www.eiss.org) so far. Of those, 54% were influenza type A and 46% influenza type B. More than 50% of all influenza virus isolates were reported from the United Kingdom (England: 34% A and 66% B, and Scotland: 65% A and 35% B) and France: 58% A and 42% B. This may be an indication that influenza activity is about to increase in these countries.

For the increased influenza activity to start in Europe only after the New Year during the 2005/2006 northern hemisphere influenza season seems late when compared with the two previous seasons. However, this observation is not exceptional once data gathered by EISS since 1996 is taken into account. During the 1997/1998 winter, increased influenza activity also started after New Year, and in the 2001/2002 and 2002/2003 winters, increased influenza activity started after New Year in more than 80% of the participating countries in EISS (Table).

The timing of the start of increased activity does not seem to be related to the proportion of influenza A and B viruses (Table), co-circulation or successive circulation of influenza A and B viruses (Table), or circulation of a new virus variant (data not shown). Further research is needed to determine what other factors could be important for the start of increased influenza activity, e.g. humidity, temperature, UV radiation, air traffic, etc.

Table. Summary of historical influenza data for Europe¹

Influenza season	No. of countries in EISS2	Timing of start of increased influenza activity in Europe in countries with unambiguous data3				Influenza virus detections (% of total detections)		Circulation of A and B viruses in Europe4
		N	Before New Year (%)	After New Year (%)		Influenza A	Influenza B
2005/20065	31	25	0	n.a.		54	46	Co-circulation
2004/2005	27	21	43	57		83	17	Successive
2003/2004	26	19	74	26		99	1	n.a.
2002/2003	22	16	19	81		64	36	Co-circulation
2001/2002	22	16	13	87		75	25	Co-circulation
2000/2001	17	16	31	69		70	30	Successive
1999/2000	13	12	92	8		99	1	n.a.
1998/1999	11	11	64	36		72	28	Co-circulation
1997/1998	10	8	0	100		93	7	n.a.
1996/1997	8	8	100	0		80	20	Successive

¹Based on sentinel clinical and sentinel and non-sentinel virological data available in the EISS database on 5 January 2006; n.a. = not applicable as no data was available or insufficient data was available.
²England, Northern Ireland, Scotland and Wales are treated as countries for the purpose of this paper.
³Calculated based on the week that influenza incidence clearly increased above baseline activity.
⁴Successive: the influenza A virus decrease overlapped the influenza B virus increase. Co-circulation: influenza A and B viruses circulating together during the whole season.
⁵Preliminary data week 40/2005 - week 52/2005.

EISS has collected data on typing and subtyping of influenza virus detections from sentinel and non-sentinel sources from week 40 to week 52 in 2005. Of the total 166 detections reported so far, 59 (36%) were influenza A (not subtyped), 18 (11%) were A(H3) (of which nine were A(H3N2)), 12 (7%) were A(H1) (of which three were A(H1N1)) and 77 (46%) were influenza B (Figure).

Thirty three of these isolates have been antigenically and/or genetically characterised. Six isolates were A(H3) A/California/7/2004 (H3N2)-like, 14 were A/New Caledonia/20/99 (H1N1)-like, five were B/Malaysia/2506/2004-like (a non-vaccine virus strain; see below) and eight were B/Jiangsu/10/2003-like. These are preliminary data and due to the differences in timing of the reports of typing, subtyping and strain characterisation data, there may be an artificial discrepancy between total numbers subtyped and characterised.

Figure. Graph presenting influenza virus detections reported to EISS up to week 52/2005.

High proportion of influenza B viruses
The current distribution of about 50% influenza A and 50% influenza B viruses is notable when compared to past data (Table). In the 9 years that EISS has existed, the maximum proportion of influenza B virus detections in any one season has been 36% (in 2002/2003). The higher proportion of influenza B virus detections could be caused by the introduction of a new strain of B Victoria lineage viruses, the B/Malaysia/2506/2004-like viruses, at the end of last season [1].

The currently circulating influenza B viruses in the world are divided into two distinct lineages represented by the B/Yamagata/16/88 and B/Victoria/2/87 viruses. These lineages have deviated to the extent that cross-reactive antibody protection against these viruses is considerably reduced [1, 2]. However, the trivalent influenza vaccine only contains one B virus component and the component in the northern hemisphere vaccine for the 2005/2006 season is a B/Shanghai/361/2002-like virus (currently used in the vaccine is B/Jiangsu/10/2003), which belongs to the Yamagata lineage of influenza B viruses [2]. Victoria-like viruses circulated only in low amounts during the 2003/2004 and 2004/2005 seasons in Europe [3, 4] and a Victoria-like virus was not included in the northern hemisphere vaccine for the 2004/2005 season [5].

Although only 13 type B viruses from Europe have been strain-characterised so far this season, five were B/Malaysia/2506/2004-like and eight were B/Jiangsu/10/2003-like. The real impact of the mismatch between the circulating B/Malaysia/2506/2004-like viruses and the vaccine component remains to be seen when influenza activity in Europe increases.

New Zealand has also experienced the emergence of the B/Malaysia/2506/2004-like viruses [1] during the southern hemisphere 2005 influenza season (87% of all virus detections were influenza B virus and more than 90% belonged to the Victoria lineage of B viruses) when there was normal seasonal activity, albeit of ‘high’ intensity: 150-249 consultations for influenza-like illness per 100 000 population [6]. When compared with previous seasons, clinical activity in New Zealand was similar to the 2003 season and slightly higher than the 2004 season [6].