ANTIBODY LEVEL AND IMMUNITY AGAINST HEPATITIS B VIRUS INFECTION AMONG GENERAL DENTAL PRACTITIONERS

Introduction. Immunization against Hepatitis B virus (HBV) is crucial for an effective control in dental healthcare settings. Nevertheless, vaccination rates among general dental practitioners (GDPs) from developed countries range between roughly 50%, as in Italy and Japan, and >90% as in US and UK. Furthermore, vaccination does not necessarily imply immunity, as serum anti-HBs antigen (Ag) level tends to decrease and booster doses are periodically required. Aim. To investigate HBV vaccination and immunity rates


Introduction
Immunization against Hepatitis B virus (HBV) by dental healthcare workers is proba bly the most effective method to control the risk for infection among patients and dental staff.A narrative review of observational studies from US made in the pre-vaccination era, between 1975 and 1989, reported that the occupational risk for HBV infection among the general den tal practitioners (GDPs) was between two to three times greater than in the general popula tion.Such a risk was even doubled among oral surgeons (reviewed by 1).Using these data it was possible to estimate that the risk for HBV infection in non-immunized GDPs was 3% af ter 100 visits of HBV carriers 2 .
Several studies, also published in the prevaccination era, reported HBV transmission to patients from fourteen oral surgeons and nine GDPs, including an oral surgeon who did not use gloves and transmitted HBV infection to fifty-five patients (reviewed by 3,4).Despite such a high risk for infection in dental health care settings, since 1987 no cases of transmis sion of HBV from dentist to patient has been reported, while there was only one case of pa tient-to-patient transmission 5 .The occupational risk in dental healthcare settings also improved.In US, for example, incidence in 2009 among healthcare workers was one hundredth the in cidence in 1983 6 .Nevertheless, HBV infection risk is persistently high in endemic areas among nonimmunized individuals 79 .The improved HBV control in dentistry is generally attributed to the widespread application of the guidelines released by the Centres of Disease Control and Prevention, based on the so called univer sal precautions which assume that all patients are potential carriers of bloodborne/airborne infections 4 .These guidelines are periodically updated, because an effective control of HBV transmission is unfeasible without an evidence based assessment of the risk for infection in dental healthcare settings 10 .This assessment is essential to understand that, for example, HBV transmission is unlikely through dental unit wa terlines, although oral fluid retraction is com mon during turbine use and microorganisms can be transmitted between patients through this route, despite turbine change or steriliza tion 11 ; or that effective dental healthcare work ers' immunization requires that guidelines are not imposed, but proposed, because the mere knowledge of the risk of HBV infection does not necessarily result in high compliance to ward guidelines 12,13 .
Barriers and immunization are, therefore, the cornerstones of hepatitis B prevention in dental healthcare settings.However, despite these measures are considered mandatory for dental healthcare workers by many professional organizations, they are not unanimously adopt ed by dental healthcare workers probably for the aforementioned reasons.This explains why some dentists are not immunized against HBV.For example, only 56% of interviewed dentists from Italy reported to be vaccinated against HBV 14 , while another study reported that only about one half of the Italian GDPs interviewed tested their serum anti-HBs antigen (Ag) level during the last ten years 13 .Vaccination rates as low as 48% and 68% also are reported from Ja pan 9 and Mexico 15 , respectively.Conversely, high immunization rates, higher than 90%, are reported from UK 16,17 .

Aim
The aim of this study was to investigate HBV vaccination rate and the level of immu nity in a sample of Italian GDPs.

Material and Methods
GDPs working in private and public of fices in Rome, Italy, were considered.Details regarding their recruitment were previously described 18 .GDPs were invited to participate before registering to Continuing Medical Edu cation (CME) courses, by signing an informed consent to interview and serological analysis.Participation was on a voluntary basis and there were no incentives.Data protection and anonymity were guaranteed.The study proto col was approved by the Review Board of the Medical and Dental Association of Rome.
GDPs were asked whether they underwent HBV vaccination during the last ten years, whether their main affiliation was public or private service and the duration of their prac tice.A blood sample was collected from each participant in a public health clinical labora tory by specialist healthcare workers and the level of antibody to hepatitis B surface antigen (anti-HBs Ag) was assessed with the Enzyme Immunoassay (EIA) method and was expressed in mIU/mL.A fee for service (called "ticket" in Italy) was charged to GDPs for this analysis of approximately 10 euro.GDPs who participated to the study provided the results of the analysis before the end of the CME course.
Prevalence and 95% confidence interval (95CI) of GDPs who reported to be vaccinated was estimated, as well as prevalence of HBV immune GDPs, that is, with serum level of an ti-HBs Ag ≥10 mIU/mL, the level thought to provide protection 19 .In order to estimate the power of vaccination to predict actual immuni ty against HBV, the methodology to assess the predictive power of screening tests was used 20 .Namely, true positive rate (TPR, i.e., Sensitiv ity) was the proportion of immune individuals who declared to be vaccinated; false negative rate (FNR, i.e., "1-Sensitivity") was the propor tion of immune individuals who declared not to be vaccinated; true negative rate (TNR, i.e., Specificity) was the proportion of non-immune individuals who declared not to be vaccinated; false positive rate (FPR, i.e., "1-Specificity") was the proportion of non-immune individuals who declared to be vaccinated; Accuracy (i.e., the proportion of correctly predicted as pro tected or non-protected).The Discriminatory Power, that is, the likelihood of vaccinated in dividuals to be protected against HBV relative to the likelihood of non-vaccinated individuals to be protected against HBV.The difference in mean serum level of anti-HBs Ag in the group of those who were vaccinated and in those who were not vaccinated was assessed using the Student's t-test for unpaired samples and nor malizing the antibody levels by log transfor mation.Undetected values were treated as log of the mean distance between 0 and the lowest detected value.
Finally, the associations between age, gen der, public-private affiliation, years of practice and vaccination or HBV immunity were as sessed.Unadjusted and adjusted odds ratios (ORs) were estimated with logistic regression analysis.Potential collinearity between covari ates was investigated with pairwise Pearson's correlation coefficient r.If two variables yield ed values higher than 0.6 they were not used together in the same regression model.

Results
The study was between 2011 and 2012, 283 GDPs were contacted and 195 agreed to participate (participation rate, 68.9%).The majority of those who refused to participate declared that they did not want to undergo se rological analysis (51 subjects), or to pay the fee for the analysis (23 subjects).Mean age of sampled GDPs was 40 years and almost two thirds were males.They were almost equally distributed between those working in the public sector and those working in the private sector.Duration of practice was, on average, fourteen years (Table 1).
88% GDPs declared to be vaccinated during the last ten years, but the proportion of those who were actually immune were 83%.The se rum level of anti-HBs Ag ranged between un detected and 1000 mIU/mL (the highest limit of detection of the EIA) with a mean value of 11 mIU/mL (Table 1).
The power of vaccination to predict effec tive HBV protection is displayed in Table 2.With an accuracy of 88%, the large majority of those who were vaccinated were immune against HBV and vice versa.The Discrimina tory Power was highly significant (DP, 22.0; 95CI, 8.0-60.6)supporting the idea that most vaccinated GDPs were immune.The data re garding FNR suggest that 4% of immune GDPs did not undergo a complete vaccination cycle during the last 10 years.Regrettably, one half of GDPs who were not immune declared to be vaccinated (FPR, 50%; 95CI, 33%-67%).The mean serum level of anti-HBs Ag was signifi cantly higher in vaccinated individuals (11.7 vs. 6.5 mIU/mL among vaccinated and non-vacci nated GDPs, respectively p<0.0001).None of the investigated explanatory vari ables were associated with the vaccination sta tus (Table 3), thus suggesting that age, gender, years of practice and public or private practice did not affect the probability of being vaccinat ed.Conversely, the probability to be immune was partly affected by since males were less likely to be immune than females and years of practice, since the likelihood to be immune decreased progressively with the years of prac tice (Table 4).However, these associations were no longer significant after the adjustment for covariates.

Discussion
The present study is one of the papers pre sented at the workshop "Advances in Infection Epidemiology and Control in Dental Healthcare Settings", Department of Public Health and In fectious Diseases, Sapienza University, Rome, Italy on February 9th, 2013 2127 .
The most important shortcoming of this study was that almost one third of the contacted GDPs did not participate to this study and it is not possible to estimate whether non-partici pating subjects had similar or different levels of HBV immunization, thus making it impos sible to predict the overall protection rate of all the contacted GDPs.Anyway, the present vac cination rate, close to 90% (Table 1) was high enough and similar to rates reported from US and UK 16,17 and considerably higher than the rate reported from Italy in 2007 14 .Thus, the most important issue in Italian GDPs, does not seem to be whether they were vaccinated or not, but whether they periodically check their im munization level.Indeed, a previous study re ported that only one half of the sampled GDPs tested their serum level of anti-HBs Ag during the last ten years 13 .This is a serious problem associated with the results of the present study that as many as 50% of GDPs with anti-HBs Ag levels lower than the critical threshold of 10 mIU/mL were actually vaccinated against HBV during the last ten years (Table 2).In other words, these subjects believed to be im mune, because of vaccination, but they were actually not protected.Therefore, the pres ent study found that almost 10% of regularly vaccinated GDPs were not immune (data not in Table ).Other studies reported that among regularly vaccinated GDPs, those who were not immune were 25% from Japan 9 , 24.2% from South Korea 7 , 18.9% from Brazil 8 .These data globally considered, suggest that approxi mately 16% dental healthcare workers who are vaccinated, or declare to be vaccinated are not HBV immune.Therefore, although vaccination rate could be as high as 90% or greater, the ef fective rate of protection may be roughly 80%.
This high proportion of vaccinated and nonimmune dentists is a serious problem from the point of view of occupational risk among dental healthcare workers.As already noted, the risk

Conclusion
The data from the present study suggest that HBV vaccination rate among Italian GDPs from public and private sectors was high.Neverthe less, the level of immunity was lower than ex pected, probably because GDPs did not check their immunization status periodically.This result was corroborated by similar data from other developed countries.If this shortcom ing is not likely to produce any additional risk for HBV infection among patients, it could be an important occupational risk among dentists who work in areas or countries where HBV is highly endemic.
for HBV infection in non-immunized GDPs is 3% after 100 visits of HBV carrier patients 2 .However, such a risk was minimized in re cent years because of the high vaccination rate among dentists from highly developed coun tries and the decreasing prevalence of HBV carriers in the general population.The problem that protection rate among dental healthcare workers is lower than expected is particularly important in areas where HBV is endemic.For example, in Romania, 36% (200/563) dental patients resulted anti-HBc Ag positive -HBc Ag is a marker of HBV infection.Consequently and unfortunately, 43% dentists were anti-HBc Ag positive 28 .This worryingly high hepatitis B prevalence among Romanian dentists was not surprising and was not only due to high hepati tis B prevalence in the general population, since almost three fourth of dental healthcare work ers from Romania were not vaccinated against HBV 29 .Thus, low immunity rate (despite high immunization rate) among GDPs in HBV en

Table 1 .
General characteristics of the sampled GDPs.

Table 2 .
Power of vaccination to predict effective HBV immunity.

Table 3 .
Unadjusted and adjusted odds ratios (ORs) of the associations between age, gender, public-private affiliation, years of practice and HBV vaccination.