High incidence of classic Kaposi's sarcoma in Mantua, Po Valley, Northern Italy (1989–1998)

The incidence of classic Kaposi's sarcoma was estimated in the province of Mantua, Po Valley, Northern Italy, yielding age-standardized rates of 2.5/100 000 men and 0.7/100 000 women (1989–98). Elevated rates in the rural zone of Viadana/Sabbioneta (5.0/100 000 men and 2.8/100 000 women) are among the highest so far reported for Italian communities. © 2001 Cancer Research Campaign http://www.bjcancer.com

analysis to a subset of 13 variables, including population density and proportion of active population employed in agriculture (ISTAT, 1986) ); main rivers flowing through the territory; whether (or not) it was included among the endemic areas for malaria both in 1902-1903(Biancorosso, 1935 and 1908 (Soliani, 1909). Cases were coded accordingly to residence at birth, during the first 15 years of life, and main residence. The 3 different incidence rates so computed for the 70 municipalities were treated as outcome variables in a multiple linear regression model, in which the previously mentioned characteristics were present as predictive variables.
As expected, rates increase with age and appear substantially higher among males though in the oldest age group (≥90), cases were observed only among females, perhaps reflecting the higher mean ages of women.
The multivariate analysis showed that the best-fit model included the independent variables listed in Table 3. Incidence according to residence during the first 15 years of life was the best outcome variable. Zones 4 and 9, which were selected by the model as being significantly associated with the incidence of KS, include 13 municipalities; of these, 9 were formerly endemic for malaria (69%). The corresponding proportion in the other zones is 39% (P = 0.045).

DISCUSSION
Elevated rates of classic KS were found far above those reported for Italy as a whole and for most European countries (Geddes et al, 1994;Zanetti et al, 1997). The estimated rates vary greatly among   (Zanetti et al, 1997). In order to make the figures comparable, age standardized rates are computed considering only cases and populations aged≥50, under the assumption that HIV-related KS cases should be below the age of 50.

Po River
Venice nearby municipalities, and are particularly high both in men and women in a rural area lying within the confluence triangle of the Oglio and Po rivers ( Figure 2). This may reflect geographical variations in the prevalence of HHV-8 infection, but may also be dependent on genetic homogeneity and/or as-yet-unknown environmental agents. None of the variables considered in the multivariate analysis appears to have a major role in predicting the incidence within the province, except 'belonging to certain geographic areas' where the proportion of municipalities formerly endemic for malaria is significantly higher than in others. The presence of 2 or more rivers and the rural/semi-rural environment may also be relevant.

MANTUA
No data are available regarding HHV-8 seroprevalence in Mantua though several of the co-factors claimed to be associated with an increase in the risk of KS are present such as (i) areas formerly endemic for malaria; (ii) rural lifestyle and cereal farming (Cottoni et al, 1997); (iii) presence of iron oxide-rich clay in soils (Ziegler, 1993); (iv) plenty of blood-sucking insects (McHardy et al, 1984).
There is some evidence that certain zones in the Po Valley are 'endemic' for HHV-8 infection and KS. Incidence rates in Ferrara (delta of the Po River) are among the highest in the world (Zanetti et al, 1997). Classic KS has been documented among Po Valley people from Lombardy (Rabbiosi, 1959;Zanca et al, 1973;Brambilla et al, 1988), Emilia-Romagna (Martinotti, 1938) and Piedmont (Vineis et al, 1987). More recently, serological (Calabrò et al, 1998;Whitby et al, 1998;Whitby et al, 2000) and molecularbased studies (Luppi et al, 1996;Monini et al, 1996) have shown that HHV-8 infection is present in healthy subjects of the Po Valley. Our findings contribute to the mapping of classic KS in Italy and, besides the south of Italy, Sardinia and Sicily, indirectly confirm the Po Valley as a high-risk area. A sero-epidemiological survey of the prevalence of HHV-8 infection in the KS-free population is in progress. Po River Po River Figure 2 Geographical distribution of standardized incidence rates by zone, for men and women together