J.-C. Manuguerra¹ , A. Mosnier ² et W.-J. Paget ³ on behalf of EISS (European Influenza Surveillance Scheme)

1 WHO Collaborating Centre for the reference and research on influenza virus and other respiratory viruses, National Influenza Centre for Northern France, Institut Pasteur, Paris, France;

2 National Coordination of the Regional Groups for the Observation of Influenza, OPEN/ROME, Paris, France

3 European Influenza Surveillance Scheme coordination centre, Netherlands Institute for Health Services Research (Nivel), Utrecht, Netherlands

In countries covered by the European Influenza Surveillance Scheme (EISS), the 2000-2001 winter was marked mainly by the spread of influenza A(H1N1) viruses. Influenza B, which globally represented a minority of cases, was common later in the season and predo-minant in Great Britain, Ireland, and Portugal. Influenza activity was at its maximum during the period of January and February/March 2001 with little time lag between countries (maximum four weeks). Overall, the morbidity rates reported were much lower than for the previous season, illustrating a moderate level of influenza activity.

Introduction

The early warning system for influenza in Europe, called the ‘European Influenza Surveillance Scheme’ (EISS), has been operational in its current form since 1996. The objectives of EISS are as follows: 1/ facilitate the rapid exchange of data concerning influenza activity obtained from sentinel physicians and virology laboratories; 2/ combine medical and virological data originating from a given population; 3/ provide national and European authorities and the World Health Organization with an ongoing description of the influenza situation in Europe; 4/ contribute to the determination of vaccine content and 5/ help national networks provide high quality information based on indicators which are standardised and comparable at the European level. EISS has benefited from financial support from the European Union since 1999. In this paper, we will describe the results recorded during the winter of 2000-2001, which were marked by a majority of cases of influenza A virus (H1N1), followed by influenza B virus which was the most prevalent form in certain countries. In most countries covered by EISS, the peaks in recorded morbidity rates were much lower than those for the previous season 1999/2000, illustrating a moderate level of influenza activity.

Method

From week 2000/40 (from 2 to October 2000) to week 2001/15 (from 9 to 15 April 2001), 14 networks in 12 countries actively participated in the EISS system: Germany, Belgium, Denmark, Spain, France, Great Britain (English, Scottish and Welsh networks), Italy, the Netherlands, Portugal, the Czech Republic, Slovenia, and Switzerland. Two networks, one in Ireland and the other in Sweden, joined the EISS programme as associate members this year.

In each of these countries, one or several networks of sentinel physicians collected clinical activity indicators (for example, number of consultations per week for influenza, number of ARI cases, number of influenza-like illness cases) and obtained nasal, pharyngeal, or nasopharyngeal specimens for influenza research purposes (1). Case definitions – when available – and outbreak alert thresholds vary according to the networks and are currently subject to standardisation. The main characteristics of these networks are presented in the table. Virological monitoring derives essentially from the National Reference Centres in each country. The virological data collected results from rapid diagnostic tests (immuno-enzymological or immunofluorescence) and from cell cultures with specific identification. Certain centres also use reverse transcriptase polymerase chain reaction (RT-PCR) routinely.

Table. Summary characteristics of the sentinel surveillance networks in EISS

(1) De nombreux réseaux/pays étaient membres de projets ayant précédé EISS (créé en 1996)– Eurosentinel (1987-91) et Système d’alerte précoce ENS-CARE Influenza (1992-95) / Many of the networks/countries were members of pre-EISS surveillance projects in Europe – the Eurosentinel (1987-91) and ENS-CARE Influenza Early Warning System (1992-95) projects.
(2) Nombre de praticiens au cours de la saison 2000-2001/ Number of physicians during the 2000-2001 influenza season
(3) ARI: infection respiratoire aiguë/ acute respiratory infection; ILI: syndrome grippal/ influenza-like illness
(4) Praticiens travaillant au sein d’écoles (enfants) et en services de soins pour jeunes/ Physicians working in community schools (children) and youth health services
(5) Praticiens spécialisés en médecine interne/ physicians specialised in internal medicine.

During the monitoring season, weekly data on influenza activity is centralised nationally. The analysis of the epidemiological situation is mainly based on incidence rates (ARI and influenza-like illness / 100 consultations or influenza-like illnes / 100 000 inhabitants) and on virological results: number of isolates / influenza virus detections and percentage of influenza positive specimens. In the EISS system, influenza activity is described according to five different levels: no influenza activity, sporadic activity, local centres of infection, regional centres of infection, and extended activity. After processing and analysis by national experts, the data gathered are transmitted electronically, on the Thursday of the same week at the latest, at 10 am, to the other EISS member countries via the internet. The EISS IT organisation has been described previously (2). For the first time, weekly summary bulletins were drafted by four experts, covering week 2000/41 to week 2001/16, and published each Friday on the EISS site (www.eiss.org). In addition, these weekly bulletins include a section with comments by each network when available, a map indicating the level of influenza activity, and the type or subtype of the dominant influenza virus. Lastly, a table is provided summarising virological data and medical activity for each network or region.

Results

For the 2000-2001 season, the first detected cases of influenza A virus in specimens obtained by sentinel physicians were reported in France and Ireland at the beginning of October (week 2000/40), as shown in figure 1. Subsequently, the influenza A virus started to spread successively in November to Belgium (2000/45), Germany (2000/47), Portugal, Netherlands and Great Britain (2000/48), in December to the Czech Republic (2000/49), Switzerland (2000/50), and Italy and Denmark (2000/52). Lastly, influenza A only appeared in January in Spain (2001/03) and in Slovenia (2001/04). In most EISS countries, influenza A virus was predominant overall during the season. Italy was the only country in the EISS zone where only influenza A was present. In all other countries, influenza B virus cases were reported and even represented a majority of cases in the western portion of Europe: Great Britain, Ireland, and Portugal. Influenza B virus cases were detected as early as October in the Czech Republic and Portugal (during weeks 2000/42 and 2000/43, respectively), then successively in December in Germany (2000/49) and Belgium (2000/51), and in January (2001/01) in Great Britain and Ireland, France (2001/02), Switzerland (2001/03), and the Netherlands (2001/04). Lastly, influenza B only appeared in February 2001 in Slovenia (2001/06), Spain (2001/07) and Denmark (2001/09). The strains of influenza B analysed were antigenetically similar to the vaccine variant B/Yamanashi/-166/99.

Figure 1. Clinical and virological sentinel monitoring of influenza in European member countries of EISS during the 2000 – 2001 season

Morbidity rates for influenza-like illness (ILI) or acute respiratory infections (ARI) for 2000-2001 for each member country in the EISS programme (including the two associate members) are indicated from week 40 (2000) to week 15 (2001). Isolation/detection of cases of viral infection for 2000–2001 are indicated in the bar chart. For Great Britain, morbidity indicator graphs are provided separately for each of the three British networks: England, Scotland, and Wales, whereas the bar charts on detection/isolation of influenza virus in primary care correspond to the English network.
*The numbers of isolates relate to the English sentinel network.

In all countries covered by EISS member networks, influenza activity in the general population was mainly due to influenza A virus (83% of 2995 virus cases detected or isolated reported by the EISS system); influenza B virus strains, even though a minority, circulated at significant levels since they represented 17% (n=520) of all viruses detected or isolated reported throughout all networks. Among the 2475 influenza A viruses reported by all networks, 1011 were subtyped, including 986 (or 97.5%) belonging to subtype H1N1 and the rest to subtype H3N2. Influenza A viruses (H3N2), which were antigenetically similar to the vaccine variant A/Panama/2007/ 99(H3N2), circulated very sporadically, even in Spain where they were detected at almost the same levels as the influenza A(H1N1) which is antigenetically similar to the vaccine variant A/New Caledonia/20/99(H1N1). Influenza A viruses were predominant (with respect to B viruses) from week 2000/46 (five compared to one) to week 2001/10 (44 compared to 40) with a peak level in weekly detection/isolation in week 2001/04 (528 for all EISS networks). Influenza B viruses were predominant starting in week 2001/11 (45 compared to 34) and remained so until week 15 (20 compared to 4) with a peak level in weekly detections/isolations in week 2001/12 (52 for all EISS networks).

As shown in figure 1, the morbidity rates generally reached peak levels at the same time as detection of influenza virus by sentinel clinicians. In the EISS zone, influenza activity reached its maximum level between the end of January (2001/04) and the end of February (2001/08). As compared with the maximum rate of clinical indicators from 1999-2000 (see figure 1), those observed during this season were, in the large majority of EISS zone countries (10 out of the 13 networks already participating in 1999-2000), much more moderate, illustrating a weak influenza activity in Europe during the 2000-2001 winter (3). The morbidity rates recorded this winter at the epidemic peak in Spain, Scotland and Wales only reached 10% of the levels reached during the preceding season. Additionally, peak 2000-2001 levels in both the Netherlands, and in England, Switzerland, and in Italy, amounted to only 21% and 41 to 45%, respectively, of the values recorded during the 1999-2000 season. The difference in the peak epidemic levels reached in 1999-2000 and 2000-2001 is less marked for Belgium, Portugal, and France, where this season’s values correspond to 70% or 80% of those recorded in 1999-2000. In Germany, the peak rates recorded during 2000-2001 were roughly on the same order as the preceding season (a slight 6% increase). As for the Czech Republic and Denmark, the increase between the last two seasons was a bit more marked, in the order of 11% to 18%. No comparison is possible between this season and the preceding seasons for the Irish, Slovenian and Swedish networks, which reported to EISS for the first time in 2000-2001.

Influenza was active around the months of January and February/March 2001, with a small time lag between countries (4 weeks maximum). In addition, its intensity was heterogeneous as shown in figure 2. The maximum levels reached during the 2000-2001 season were qualified as follows: 1/ extended activity in Germany, France, the Czech Republic, Denmark, and Sweden ; 2/ regional activity in Belgium, Switzerland, Italy, and Scotland ; 3/ local activity in Ireland, Slovenia, and the Netherlands ; 4/ sporadic activity in England, Spain, Portugal ; 5/ no activity in Wales.

Figure 2. Influenza activity in the countries contributing to EISS

La carte représente les niveaux d’activité grippale rapportés par chaque réseau membre de EISS ainsi que le type ou le sous-type de virus grippal dominant pour sept semaines représentatives/ The map presents levels of influenza activity as assessed by each of the networks in EISS as well as the dominant type/sub-type of influenza virus for seven selected weeks.
Colours represent an epidemiological level (see maps and text).
A = virus A dominant/ Dominant virus A
H1 = virus A(H1N1) dominant/ Dominant virus A(H1N1)
H3 = virus A(H3N2) dominant/ Dominant virus A(H3N2)
B = virus B dominant/ Dominant virus B

Discussion and conclusions

The 2000-2001 winter was marked mainly by the spread of influenza A virus (H1N1) followed by influenza B virus, which was predominant in some countries. In most of the EISS participating countries, the morbidity rates recorded during the 2000-2001 winter were much lower than those reported in the previous season, illustrating moderate influenza activity.

The data reported by the EISS members for the period of week 2000/40 to 2001/15, and which were present in the database at the time this summary report was drafted, indicate that influenza had a moderate impact in most member countries, with respect to the activity recorded during the preceding season. It is noteworthy that among the five countries where the slightest differences were found between the highest morbidity rates reported in 1999-2000 and 2000-2001, four used the ARI as numerator instead of the ILI: Germany, Belgium, France and the Czech Republic. The cooperation between European networks provided real time validation that there was no antigenic change in the circulating influenza viruses; this was important with respect to the composition of the influenza vaccine and the moderate epidemic impact experienced in Europe. It meant that no specific public health intervention was needed during the 2000-2001 season.

This year, the EISS system benefited from a data manager, and pursued work on case definitions and epidemic thresholds; new actions were implemented, such as quality control of virology laboratories. For the first season, a panel of three experts, two epidemiologists and a virologist, from three different EISS member countries, contributed with the assistance of the data manager to summaries on the European situation, and forma-lised their consensus by the publication of the weekly bulletins, commented and published on the website each Friday morning. The qualification of the various levels of activity will take on a new dimension in the season to come, since the five existing levels will be broken down into an indicator of geographical extension coupled with an intensity indicator.

To conclude, the EISS network continues to grow and Norway, Poland, Romania, and Slovakia will participate in the coming season (2001-2002). EISS is an integrated federation of national networks combining morbidity indicators and virological data. The surveillance systems in each country symbolise the parti-cipation of primary health care providers in the monitoring and control of an infectious disease; in other words, they are a practical symbol of public health.

This summary was drafted for all EISS members: Aymard M (FR), Bartelds AIM (NL), Charlier N (B), Christie P (UK), Cohen JM (FR), Falcao I (PT), Fleming DM (UK), Grauballe P (DK), Havlickova M (CZ), Heckler R (DE), Heijnen M-L (NL), de Jong JC (NL), Lina B (FR), Linde A (SW), Manuguerra J.-C. (FR), de Mateo S (ES), Mensi C (IT), Mosnier A (FR), Müller D (CH), Mullins N (IRL), Nolan D (IRL), O'Flanangan D (IRL), Paget WJ (NL), Perez-Brena P (ES), Pregliasco F (IT), Prosenc K (SL), Rebelo de Andrade H (PT), Samuelsson S (DK), Schweiger B (DE), Socan M (SL), Thomas D (UK), Thomas Y (CH), Tumova B (CZ), Uphoff H (DE), Valette M (FR), Vega T (ES), van der Velden K (NL), van der Werf S (FR), Watson J (UK), Yane F (B) et Zambon M (UK).

EISS Participants