Surveillance of HCV infection in the Netherlands

Les Pays-Bas ont deux systèmes de surveillance des infections dues au virus de l’hépatite C (VHC). Les données agrégées de laboratoire sont disponibles depuis 1990, et en 1999, la déclaration de l'hépatite C est devenue obligatoire. Les résultats montrent que le nombre de cas déclarés est stable, avec une prédominance des cas de sexe masculin (66 %), âgés de 15 à 54 ans. L’utilisation de drogues injectables était le principal mode de transmission (64 %). Le système de déclaration devrait inclure plus de données cliniques pour permettre de mieux dépister les changements des modes de transmission.

En avril 1999, l'infection à VHC est devenue une maladie à déclarer aux Pays-Bas.L'infection à VHC est déterminée par détection 1. Antibody to hepatitis C virus positive verified by a supplemental test (Recombinant Immunoblot Assay, Polymerase Chain Reaction). 2. Representatives of the laboratories of the sentinel system that gather all data on viral diagnosis made by those laboratories.

H
epatitis C is a viral infection mainly parenterally transmitted, discovered in a context of transfusion associated hepatitis in 1989 (1).Since then, hepatitis C virus (HCV) infection has proved to be a major public health problem worldwide.The introduction of blood screening programmes has dramatically reduced infection in blood and blood products recipients (2,3).However, HCV continues to be transmitted nosocomially, especially in dialysis settings (4,5).In many countries, the most important risk factor for HCV transmission is injecting drug use (6)(7)(8).Preventive measures such as needle and syringe exchange programmes have not reduced HCV transmission among injecting drug users (IDU) as expected (9)(10)(11).

The burden of HCV infection and its epidemiology in the
Netherlands is yet to be well documented.Since 1990 there has been voluntary reporting of weekly aggregated data of laboratory confirmed HCV diagnoses in the country.Though useful for monitoring trends, these data lack epidemiological information.In order to estimate the morbidity due to HCV and the contributions of the various routes of transmission, HCV infection was made a notifiable disease from April 1999.In this article we use data from both surveillance systems currently in place in the Netherlands to describe the epidemiology of newly diagnosed HCV infections and highlight the added value of the new notification system.

Materials and Methods
Since 1990 a sentinel system of virological laboratories has reported newly diagnosed HCV infections 1 to the Working Group 2 for Clinical Virology as aggregated data.A newly diagnosed case is a person who has a laboratory confirmed diagnosis of HCV infection for the first time.This data is shared every week with the national authorities.The number of laboratories participating in the system increased from 8 to 17 between 1990 and 1992.The number of laboratories and the population coverage (40% of the Dutch population) remained the same since then.There is no zero reporting system in place.
The aggregated laboratory data of HCV infection diagnoses from 1990 to 2000 inclusive was analysed in Excel.Data from the national notification system from April 1999 to February 2001 was imported and analysed in Stata.The data were described by sex and age, nationality, type of infection, information on origin of infection (country where infection was acquired), and by suspected route of transmission.Chi squared values were used to test for difference in proportions and analysis of variance for difference in means.Data from the notification system were also used to calculate cumulative incidence rates in each Municipal Health Authority (MHA) since April 1999, using the Dutch population census from the year 2000.

➤ Notification data
Since the introduction of notification in April 1999 to February 2001, 875 cases of HCV infection were reported over 94 weeks.The mean number of notifications per week was 9.31 (SD=5.84),ranging from 1 to 32.
The overall ratio of males to females was 2.2:1.The mean age of cases reported among males was 41 years (SD=12.5)and among females 43 years (SD=16.5),the difference being not significant (F-test=3.42,p=0.065).The difference in the gender distribution was greatest between age 15 to 54 (Figure 2).Of all the cases reported, 6% were described as acute HCV infections, and 94% as chronic hepatitis at diagnosis.The mean age of infection of acute and chronic cases was 42 years.There was no difference in sex, nationality, origin of infection and route of transmission between chronic and acute cases.
Among the 863 cases where the origin of infection was recorded, 44% were reported as acquiring the infection in the Netherlands and 16% abroad, with a remaining 40% unknown.Nationality was recorded in only 389 cases, of which 32% were non-Dutch nationals.Cross tabulation of nationality and origin of infection for 383 cases where information was complete, showed that 67% of non-Dutch nationals reported having acquired the infection abroad, compared to only 3% of Dutch nationals (χ 2 =198.2,p<0.0001).
Data on suspected transmission route was recorded for 690 cases ( The cumulative incidence of HCV infections notified by 42 MHAs showed wide variation (Figure 3).Only four MHAs did not report a newly diagnosed case of HCV infection in this period.The mean national cumulative incidence was 5.9 per 100 000, varying from 0.18 to 47.27 cases per 100 000.Twenty three MHAs had a cumulative incidence under the national mean and 13 were between 5 and 10 per 100 000.Four MHAs had a cumulative incidence double the national mean and two were four to eight times higher.

Discussion
Data from the sentinel laboratory based system in the Netherlands from 1992 to 2000 show a stable number of cases Cela peut s'expliquer, en partie, par le fait qu'une proportion d'infections à VHC n o u v e l l e m e n t d i a g n o s t i q u é e s serait survenue avant l'introduction des mesures de prévention.Autre explication possible : l'augmentation d'autres modes d e t r a n s m i s s i o n , comme l'utilisation de drogues injectables.
Il est surprenant de constater que l'âge moyen des cas cliniques aigus était identique et non inférieur à celui des cas chroniques, bien que cela puisse résulter des faibles nombres.Il est également probable que les cas asymptomatiques d'infection à VHC aiguë aient été classés à tort comme cas chroniques.La différence dans la répartition par sexe chez les cas âgés de 15 à 54 ans pourrait correspondre à un schéma de transmission différent dans ces groupes d'âge.➤ reported over the years.Two peaks were observed in 1992 and 1997, probably explained by increasing capacity for performing anti-HCV testing with the introduction of the ELISA-2 test in 1992, and greater availability of PCR techniques for diagnosis around 1997 (8,(12)(13)(14).In the United Kingdom (UK), a similar laboratory based reporting system has shown a clear increase of laboratory diagnosed HCV infection from 1992 to 1996 (15).This is probably because all positive tests are reported in the UK, whereas only newly diagnosed cases are reported in the Netherlands.In the United States, however, the estimated annual incidence rate for acute infection declined by more than 80% between 1989 and 1997, mainly attributable to introduction of blood screening measures (16).The Dutch laboratory based system does not allow differentiation of recent and long acquired infections.Nonetheless, it is surprising that there is no comparable drop in the total number of newly diagnosed cases in our data.This can be only partly explained by the fact that a proportion of newly diagnosed HCV infections occurred before introduction of preventive measures.Other explanations could include an increase in other routes of transmission such as intravenous drug use.
The notification system implemented in 1999 does not yet have enough data to identify any trends in acute infections.The notification data, however, do reveal that none of the 42 cases suspected to be caused by blood or blood products transfusion were exposed after 1990.This shows that blood screening measures have had an effect on incidence, and demonstrates the added value of the notification system.More clinical information, however, would enhance differentiate between newly acquired infection, newly diagnosed chronic HCV and asymptomatic infection.
It is unexpected that the mean age of acute clinical cases was the same as -and not lower than -the mean age for chronic cases, although this may be due to low numbers.It is also probable that cases of acute asymptomatic HCV infections were misclassified as chronically infected cases.The ➤ ➤ Il y a nettement plus de cas déclarés chaque année par le réseau de laboratoires (avec une couverture de 40%) que par le système national de déclaration.Cela peut résulter de la sous-déclaration de l'infection à VHC.Le dépistage sanguin dans certains groupes, tels que les donneurs de sang, peut également contribuer à augmenter le nombre des déclarations des laboratoires.
There are clearly more yearly cases reported by the laboratory system (with 40% coverage) compared to the national notification system.This may be due to under notification of hepatitis C infections.Blood screening in special groups, such as blood donors, may also contribute to an increased number of laboratory reports.
Since 1999, 64% of cases where a suspected transmission route was known were associated with injecting drug use, confirming this as the most important risk factor for HCV infection after the decline of transfusion associated infections, in the Netherlands, as in other countries (6)(7)(8)10,11,15,16).Cases among IDUs were on average younger than in other groups (p<0.001),probably because of the higher likelihood to be exposed to the virus at a much earlier age.Drug users are also more likely to share other risk factors for liver disease, such as alcohol abuse and concurrent viral infection, particularly with HIV and hepatitis B virus, which may accelerate the presentation of clinical disease leading to hepatitis C testing.Surveys on long term IDUs in the US report seroprevalence levels of 70-90% (17), indicating that preventive measures should be comprehensive and target the young drug user.
The proportion of cases with an unknown suspected transmission route could be partly due to the delay between infection and diagnosis, but it could also be interpreted that unknown or other common exposures (such as nosocomial transmission) may play a bigger role in the transmission of hepatitis C. In a study in Amsterdam, 10% of all cases of anti-HCV positive non-IDUs did not report any exposure to known risk factors (11).The high proportion of cases with unknown suspected route of transmission may also reflect the individual's desire not to disclose sensitive information.
The wide variation of cumulative incidence by MHA is curious.It can not be explained by the differences between MHAs in case ascertainment (active versus passive), nor by the case definitions, the number of IDUs, nor by the distribution of liver specialised centres.Varying completeness of notification may play an important role in regional differences, but this would need further investigation.

Conclusion
This overview demonstrates the added value of the notification system which allows monitoring of trends and identifying shifts in risk factors, with additional information on geographical distribution.The value of the collection of demographic data and risk factor information will be demonstrated in a few years, when trends over time are more apparent.Prevention and control measures in the Netherlands should focus not only on reducing transmission in groups at high risk of infection (eg injecting drug users), but also on the early identification of persons with chronic infection.To further improve the system, levels of notification rates must be increased and information on clinical presentation, including the presence of chronic liver disease, should be gathered.MHAs should be encouraged to report more complete information and use these data to further investigate the incidence or notification of HCV infection in their regions.The use of capture-recapture methods would be an important step to further evaluate the two surveillance systems, and to estimate number of new and chronic infections.■ Historique L es infections dues au virus de l'hépatite C (VHC) sont pandémiques, avec une prévalence mondiale de 3% selon l'OMS (1).En Europe centrale, environ 1% de la population est infectée, principalement par le génotype 1b (85% en Autriche) (2,3).L'Autriche est considérée comme une zone de faible endémie, avec une prévalence d'environ 0,7% dans la population générale (4).Cependant, les hépatologues autrichiens ont estimé à 60 000 environ le nombre de personnes infectées, soit une prévalence de 750/100 000 (5).
En Autriche, les données de déclaration du VHC n'ont pas été analysées depuis dix ans.Les objectifs de notre étude étaient donc de : ➤ From 1993 to 2000 in Austria, the notification system registered 2232 cases of hepatitis C whereas 10 607 hospital cases were reported in the hospital discharge register (HDR).These differences can be explained by under-reporting due to lax reporting behaviour and stigma associated with this disease.The distribution of HCV infection varied geographically.The notification data showed the highest incidence rates in Tyrol, while HDR data showed highest hospitalisation rates in Vienna and Lower Austria.This study highlights the urgent need for a clear case definition at the national level and an electronic reporting system.

I
nfections with the hepatitis C virus (HCV) are pandemic, with a world-wide prevalence of 3% according to WHO (1).In Central Europe approximately 1% of the population is infected, mostly with genotype 1b (85% in Austria) (2,3).Austria is considered as a low endemic area with a prevalence of about 0.7% in the general population (4).However, Austrian hepatologists have estimated a total of about 60 000 persons to be infected which gives a prevalence of 750/100 000 (5).
There has been no comprehensive analysis of the Austrian HCV notification data in the last ten years.The objectives of our study were therefore to: ➤

Figure 1 Cas
Figure 1 Cas de diagnostics biologiques d'infection à VHC par an aux Pays-Bas Laboratory diagnoses of HCV infection by year in the Netherlands

Hepatitis C in 1
Austria 1993-2000: reporting bias distort HCV epidemiology in Austria R. Strauss 1 , G. Fülöp 2 , C. Pfeifer 3 Federal Ministry for Social Security and Generations, Public Health BMSGSection, Vienna, Austria 2 Austrian Health Institute (ÖBIG), Vienna, Austria 3 Institute for Statistics, University of Innsbrück, Austria S.S. Chaves 1 , M-A.Widdowson1, 2 , A. Bosman 1 Department of infectious disease epidemiology, National Institute of Public Health and the Environment, RIVM, The Netherlands 2 Fellow of the European Program for Intervention Epidemiology Training (EPIET), RIVM, The Netherlands