We present a large study of HPV prevalence in women of different ages in the area of Athens, from which interesting conclusions can be drawn. As HPV DNA typing is gradually incorporated in national screening programmes knowledge of HPV type variations in different geographical regions is useful information. Similarly, as HPV vaccination is introduced as a means of primary cervical cancer prevention, better prediction models of its efficacy specific for each area can be made .
In our study and in concordance with what is seen in the general population, normal cytology dominated all age groups. The highest LgSIL rates appeared among women aged 14–25 years. Only a minority (1.5%) of young girls had a high grade lesion and this is again consistent with what is seen in the literature. It also justifies the current American and British screening programme that suggest smear testing after the age of 21 and 25 respectively [14, 15].
In our study, the HPV prevalence in normal samples was 15.7%. For Eastern Europe, HPV prevalence in a meta-analysis of 4053 samples tested with normal cytology was 21.4% . Studies from different regions of Greece have reported an overall HPV prevalence ranging from 22.7 to 50.7% [11, 12] which is in accordance to our HPV prevalence of 33.1%. Our findings are also in agreement with other studies in Greece that have suggested a prevalence of hrHPV ranging between one in five and one in three. It does however come in stark difference with what has been reported in another Greek study conducted by the University of Thessaloniki , where a prevalence of only 2.5% was reported which is among the lowest ever reported in the world. This difference might be attributed to the fact that this latter study used a different HPV identification method.
In an attempt to compare our results with other countries’ epidemiological data, we presented an age-related prevalence curve of HPV infection. This curve showed a peak at the 14–19 age group, a second one at 30–34 y.o. and then dropped. Although the study by Coupe et al conducted in the Netherlands demonstrated an association between age and HPV prevalence , others have not identified a significant relation . When we looked specifically at the age related prevalence of hrHPV, we observed a peak in the 14–25 years age group, in accordance with other European studies [18, 19]. Younger women are more prone to develop an HPV infection as they tend to have multiple partners  and are also less likely to have developed immunity to HPV given their recent exposure to the virus.
As expected, HPV prevalence, increased as lesions progressed to higher grade ones. The same trend was observed in hrHPV positivity in relation to cytological status. Investigating the results per age group, we observed that hrHPV infection dominated in 14-25 year old women irrespective of cytology. However young women are less likely to present cancerous lesions as in the majority of cases the lesion regresses after 2–3 years [21, 22].
Previously, in a 1636 women cohort study conducted by this group, where the prevalence of subtypes 6, 11, 16, 18, 31 and 33 was tested, HPV 11 was found to be the most frequent . In the current study, where a wider range of HPV subtypes were investigated, HPV 16 and 42 were the most frequent (6.7% and 6.8% respectively) in total samples followed by HPV 51 (5.7%). HPV 51 was constantly identified as a common HPV type in all age groups however its ranking dropped significantly behind HPV 16 in ages 26–46. A possible explanation for this may be that HPV 51, although highly prevalent, is cleared more quickly and possibly causes a more consistent immunologic response with a longer immune protection conferred to those that have cleared the virus, making them no longer susceptible to new infections.
HrHPV distribution in Mediterranean countries like Italy, Portugal, Spain [23–26] is similar to our data with HPV 16 detected first and HPV 51 ranking second or third. Although HPV 18 prevalence was overall low, its presence was high (12.8%) among women with high grade lesions. Generally, the predominance of HPV 16 and 18 in high grade lesions strengthens the importance of vaccination in prevention of cervical cancer as 50% of the HSIL lesions in our study harboured infection by these two types.
Multiple infections were found in 12.9% of Greek population. Multiple HPV positivity was more common in younger women. In addition, women with cervical lesions had a higher rate of multiple infections compared to those who had normal cytology. Multiple infections might be a risk factor for development of cytological abnormalities. The majority of multi-HPV infected women harboured at least one hrHPV type and the majority of those carried two types, which is consistent with other epidemiological studies .
Although this study presented here included a large number of women of a broad age range, it has some limitations. Our sample is representative of women presenting for smear testing. However, as smear taking in Greece is done on an ad hoc basis rather through an organised screening system, our sample cannot be considered representative of the Athenian population. These data may give us important information regarding regional HPV prevalence but large epidemiological studies from different regions of our country are needed.