Journal of Hepatology 35 (2001) 284±289

www.elsevier.com/locate/jhep

Risk of parenterally transmitted hepatitis following exposure to

surgery or other invasive procedures:
results from the hepatitis surveillance system in Italy
Alfonso Mele 1, Enea Spada 1, Luciano Sagliocca 2, Pietro Ragni 3, Maria Elena Tosti 1,
Giovanni Gallo 4, Angela Moiraghi 5, Emanuela Balocchini 6, Massimo Sangalli 7,
Pier Luigi Lopalco 8, Tommaso Stroffolini 1,9,*
1
Laboratory of Epidemiology, Istituto Superiore di SanitaÁ, Viale Regina Elena 299, 00161 Rome, Italy
2
Santobono Hospital, Naples, Italy
3
Emilia Romagna Regional Health Authority, Bologna, Italy
4
Veneto Regional Health Authority, Venice, Italy
5
Department of Public Health and Microbiology, University of Turin, Turin, Italy
6
Toscana Regional Health Authority, Florence, Italy
7
Agency for Public Health, Lazio Region, Rome, Italy
8
Institute of Hygiene, University of Bari, Bari, Italy
9
San Giacomo Hospital, Liver Unit, Rome, Italy

Background/Aims: To evaluate the strength of association between parenterally transmitted viral hepatitis and
speci®c types of invasive procedures.
Methods: Data from the surveillance system for type-speci®c acute viral hepatitis (SEIEVA) during the period 1994±
1999 were used. The association of acute hepatitis B virus (HBV) and hepatitis C virus (HCV) infection with the
potential risk factors (odds ratios (OR)) was estimated comparing 3120 hepatitis B and 1023 hepatitis C cases with
7158 hepatitis A cases, used as controls, by multiple logistic regression analysis.
Results: Most procedures resulted in being associated with the risk of acquiring acute HBV or HCV. The strongest
associations were: for HBV infection, abdominal surgery (adjusted OR ˆ 3.9; 95% con®dence intervals (CI) ˆ 2.0±7.5),
oral surgery (OR ˆ 2.7; 95% CI ˆ 1.6±4.5) and gynaecological surgery (OR ˆ 2.6; 95% CI ˆ 1.2±5.5); for HCV infec-
tion, obstetric/gynaecological interventions (OR ˆ 12.1; 95% CI ˆ 5.6±26.3), abdominal surgery (OR ˆ 7.0; 95%
CI ˆ 3.2±14.9) and ophthalmological surgery (OR ˆ 5.2; 95% CI ˆ 1.1±23.2). Biopsy and/or endoscopy were associated
with HCV, but not with HBV infection.
Conclusions: Invasive procedures represent an important mode of HBV and HCV transmission. Since a large
proportion of the adult general population is exposed to these procedures and an effective HCV vaccine is not yet
available, non-immunological means of controlling iatrogenic modes of transmission are extremely important.
q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved.
Keywords: Hepatitis B; Hepatitis C; Surgery; Italy

1. Introduction Italy, owing to the thousands of deaths for cirrhosis [1]

Infections due to hepatitis B virus (HBV) and hepatitis C
virus (HCV) represent a major public health problem in
Received 17 November 2000; received in revised form 8 March 2001;
accepted 24 April 2001
* Corresponding author. Tel.: 139-6-49903182; fax: 139-6-49387173.
E-mail address: amele@iss.it (T. Stroffolini).
and hepatocellular carcinoma [2] occurring each year.
Furthermore, although the incidence of both infections has
decreased in the past 15 years [3], new infections still occur.
The prevalence of chronic carriers of hepatitis viruses
among the Italian general population is estimated to be
about 1% for hepatitis B surface antigen (HBsAg) [4±6]
and from 3.2 (in northern Italy) to 12.6% (in southern

0168-8278/01/$20.00 q 2001 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved.
PII: S01 68-8278(01)0011 1-8
 A. Mele et al. / Journal of Hepatology 35 (2001) 284±289 285

Italy) for HCV, with the highest prevalence rates reported The BMDP statistical software package, version 1990 [16], was used for
among the elderly in southern Italy (.30%) [7±9]. statistical analysis.
The Italian National Hepatitis Surveillance System
(SEIEVA) and case±control and cohort studies have
shown that, in Italy, surgery and diagnostic or therapeutic 3. Results
invasive procedures are important modes of transmission for
both HBV and HCV infections [10±14]. Using SEIEVA In the period 1994±1999, 17,441 cases of acute HAV,
data, we investigated the association, if any, between speci- HBV, and HCV infection were noti®ed to SEIEVA. A
®c types of surgery or other invasive procedures and the total of 6140 individuals were excluded from the analysis
incidence of acute HBV and HCV infections. because they reported intravenous drug use and/or blood
transfusion in the 6 months before the onset of the acute
illness or because they were less than 15 years of age. A
2. Materials and methods total of 11,301 individuals were included in the analysis:
3120 with acute HBV infection, 1023 with acute HCV
SEIEVA was established in 1985 [15]; it is co-ordinated by the Istituto infection and 7158 with acute HAV infection.
Superiore di SanitaÁ (the National Health Institute of Italy). The system
relies on the participation of the local health units (LHUs), which during
Table 1 shows the demographic features of the study
the study period (i.e. 1994±1999) covered from approximately 40 (1994) to population by type of hepatitis infection. As expected, indi-
57% (1999) of the Italian population. viduals with acute HAV infection tended to be younger than
Individuals diagnosed with acute viral hepatitis, hospitalized or not, are those with parenterally transmitted acute infections and
interviewed by either a public health inspector or a physician of the LHU. A were more likely to be from southern Italy. Tables 2 and 3
standardized two-page questionnaire collecting information on the socio-
demographic characteristics of the subjects, exposure to parenteral risk
show the proportions of individuals reporting surgery or
factors within 6 months of disease onset and exposure to faecal/oral risk other invasive procedures by type of procedure and the
factors within 6 weeks of onset is used. Furthermore, since 1994, informa- adjusted ORs and their 95% CIs derived by multiple logistic
tion has been collected on the type of surgery and other invasive proce- regression analysis.
dures, including biopsy and endoscopy. The interviewer is blinded with Overall, 14.6% of the individuals with acute HBV infec-
respect to the type of hepatitis to avoid bias in the identi®cation of risk
factors. After the interview, the results of assays for hepatitis markers are
tion (457/3120) and 25.5% with acute HCV infection (261/
recorded. Copies of both serological results and the completed question- 1023) reported that they had undergone at least one invasive
naires are forwarded once a week to the national co-ordinating centre at the procedure, compared with 4.9% of the individuals with
Istituto Superiore di SanitaÁ. acute HAV infection (349/7158). The most frequent inva-
Case de®nition for acute hepatitis is based on clinical, biochemical and sive procedures for individuals with acute HBV infection
serological criteria. The clinical and biochemical criteria are acute illness
compatible with hepatitis and serum alanine aminotransferase (ALT) levels
were oral surgery, biopsy/endoscopy and obstetric/gynaeco-
greater than 2.5 times the normal upper limit. The serological criteria used
Table 1
to distinguish the different types of hepatitis are as follows: hepatitis A is
Study population by demographic features a,b
de®ned as immunoglobulin M (IgM) anti-HAV positive, regardless of
HBsAg status; hepatitis B as HBsAg positive, IgM anti-HBc positive and
Hepatitis B Hepatitis C Hepatitis A
IgM anti-HAV negative or not performed; hepatitis C as IgM anti-HAV
negative, IgM anti-HBc negative and anti-HCV positive, regardless of n % n % n %
HBsAg status. Assays of hepatitis markers are performed in different
laboratories in various parts of the country using standardized methodol- Age
ogy. The data analyzed in this paper refer to the period from 1994 to 1999. 15±24 699 22.4 173 16.9 4140 57.8
In this period, no changes were made in the noti®cation system. Most 25±34 1071 34.3 280 27.4 2051 28.7
individuals with acute hepatitis are hospitalized and virtually all hospita- 35±44 592 19.0 142 13.9 588 8.2
lized patients are noti®ed to LHUs. 45±54 364 11.7 118 11.5 251 3.5
The study protocol conforms to the ethical guidelines of the 1975 55±64 203 6.5 137 13.4 70 1.0
Declaration of Helsinki. 65±74 122 3.9 117 11.4 28 0.4
75±84 53 1.7 48 4.7 16 0.2
2.1. Statistical analysis $ 85 16 0.5 8 0.8 14 0.2
Sex
To evaluate the strength of associations between parenterally transmitted Male 2253 72.6 602 59.1 4573 64.1
viral hepatitis and different types of surgery or invasive procedures, indi- Female 851 27.4 416 40.9 2566 35.9
viduals with acute HBV or HCV infection were compared with those with Educational level
acute HAV infection (identi®ed by the same surveillance system), using a $ 9 years 1473 47.2 398 38.9 3429 47.9
case±control method after excluding blood-transfusion recipients and intra- # 8 years 1647 52.8 625 61.1 3729 52.1
venous drug users. Multiple logistic regression analysis was used to esti- Area of residence
mate the adjusted odds ratios (OR) and their 95% con®dence intervals (CI). North ± centre 2635 84.5 787 76.9 2554 35.7
Age, sex, educational level, area of residence, other parenteral exposures South ± islands 485 15.5 236 23.1 4604 64.3
(ear piercing, tattooing, acupuncture, attendance at a chiropodist or mani- Number of cases 3120 1023 7158
curist) and intercourse with multiple sexual partners, were adjusted for in
a
the analysis. Individuals less than 15 years of age were excluded from the SEIEVA (National Hepatitis Surveillance System) 1994±1999, Italy.
b
analysis because few of them were exposed to surgery or other invasive Subjects ,15, intravenous drug users, and transfused patients were
procedures. excluded from the analysis.
286 A. Mele et al. / Journal of Hepatology 35 (2001) 284±289

Table 2
Adjusted OR and 95% CI for different types of invasive procedure among hepatitis B cases a,b

Intervention type Hepatitis B (3120 cases) Hepatitis A (7158 cases) OR adjusted c (95% CI)

n % n %

Minor surgery 56 1.8 41 0.6 1.9 (1.1±3.2)

Gynaecological d 24 2.8 23 0.9 2.6 (1.2±5.5)
Orthopaedic 20 0.6 34 0.5 1.3 (0.6±2.7)
Abdominal 54 1.7 31 0.4 3.9 (2.0±7.5)
Cardiovascular 25 0.8 8 0.1 1.7 (0.6±4.8)
Dermatological 31 1.0 18 0.3 1.8 (0.9±3.6)
Oral surgery 77 2.5 40 0.6 2.7 (1.6±4.5)
Ophthalmological 11 0.4 5 0.1 2.4 (0.6±9.2)
Urological 14 0.4 7 0.1 2.0 (0.6±6.0)
Other intervention 73 2.3 74 1.0 1.3 (0.8±2.1)
Biopsy/endoscopy 72 2.3 67 0.9 1.4 (0.9±2.2)
a
SEIEVA (National Hepatitis Surveillance System) 1994±1999, Italy.
b
Subjects ,15, intravenous drug users, and transfused patients were excluded from the analysis.
c
Adjusted for sex, age, educational level, area of residence, other parenteral exposures and multiple sexual partners in multiple logistic regression analysis.
d
For females.

logical interventions; for those with acute HCV infection,
the most frequent invasive procedures were obstetric/gynae-
cological interventions and biopsy/endoscopy. Among indi-
viduals who had undergone biopsy/endoscopy in the
previous 6 months, 36% of those with acute HBV infection
(26/72) and 40% of those with acute HCV infection (19/47)
had undergone gastrointestinal endoscopy. Among indivi-
duals who had undergone an ophthalmological intervention,
73% of those with acute HBV infection (8/11) and 56% of
those with acute HCV infection (9/16) had undergone this
procedure for cataracts.
Most of the invasive procedures considered were asso-
ciated with the risk of acquiring acute HBV or HCV infec-
tion, even if to a varying extent. Adjusting by multiple
logistic regression analysis for the confounding effect of
sociodemographic characteristics of subjects, other parent-
eral exposures and intercourse with multiple sexual part-
ners, an association was found to link acute HBV
infection with abdominal surgery (OR ˆ 3:9; 95%
CI ˆ 2.0±7.5), oral surgery (OR ˆ 2:7; 95% CI ˆ 1.6±
4.5). obstetric/gynaecological interventions (OR ˆ 2:6;
95% CI ˆ 1.2±5.5), and minor surgery (OR ˆ 1:9; 95%
CI ˆ 1.1±3.2; Table 2). Acute HCV infection resulted asso-
ciated with several types of surgical intervention, the stron-
gest associations being with obstetric/gynaecological,
abdominal, cardiovascular, and ophthalmological interven-
tions (Table 3).
Bioptic and/or endoscopic procedures resulted in being
associated with HCV infection, whereas no association was
found with HBV infection. An additional analysis, compar-
ing gastroscopy versus no intervention, showed a strong
association (OR ˆ 3:2; 95% CI ˆ 1.3±7.9) with acute hepa-

Table 3
Adjusted OR and 95% CI for different types of invasive procedure among hepatitis C cases a,b

Intervention type Hepatitis C (1023 cases) Hepatitis A (7158 cases) OR adjusted c (95% CI)

n % n %

Minor surgery 18 1.8 41 0.6 3.0 (1.5±6.1)

Gynaecological d 31 7.5 23 0.9 12.1 (5.6±26.3)
Orthopaedic 18 1.8 34 0.5 3.5 (1.6±7.5)
Abdominal 23 2.2 31 0.4 7.0 (3.2±14.9)
Cardiovascular 25 2.4 8 0.1 4.1 (1.4±11.9)
Dermatological 8 0.8 18 0.3 1.5 (0.5±4.7)
Oral surgery 28 2.7 40 0.6 2.8 (1.4±5.7)
Ophthalmological 16 1.6 5 0.1 5.2 (1.1±23.2)
Urological 5 0.5 7 0.1 0.8 (0.1±4.8)
Other intervention 42 4.1 74 1.0 3.3 (1.9±5.7)
Biopsy/endoscopy 47 4.6 67 0.9 2.1 (1.2±3.6)
a
SEIEVA (National Hepatitis Surveillance System) 1994±1999, Italy.
b
Subjects ,15, intravenous drug users, and transfused patients were excluded from the analysis.
c
Adjusted for sex, age, educational level, area of residence, other parenteral exposures and multiple sexual partners in multiple logistic regression analysis.
d
For females.
 A. Mele et al. / Journal of Hepatology 35 (2001) 284±289 287

titis C, but no association (OR ˆ 1:6; 95% CI ˆ 0.8±3.4) the association between parenterally transmitted hepatitis
with acute hepatitis B (data not shown). and speci®c types of surgery or invasive procedures. Never-
As control-group individuals over 65 years of age were theless, the low incidence of both acute HBV and HCV
poorly represented in our study, a further analysis limited to infection in the general population and their often asympto-
individuals aged less than 65 years was also performed. The matic course would entail using a very large sample size and
strength of the risk associated with each type of invasive a serological follow-up to detect the occurrence of these
procedure remained essentially unchanged, except for infections. For these reasons, a prospective cohort study is
ophthalmological surgery, which resulted not yet associated not feasible. Thus, case±control studies represent the most
with acute HCV infection (OR ˆ 3:0; 95% CI ˆ 0.4±20.2; effective means of evaluating the role of surgery or invasive
data not shown). procedures in the spread of parenterally transmitted hepati-
The scoring of the other variables included in the logistic tis in the general population.
regression analysis model is shown in Table 4. In the present study, we have estimated the risk associated
with each major type of surgery or invasive procedure in the
Italian general population during a relatively long period of
4. Discussion time (6 years). Almost all the invasive procedures showed
an association with both HBV or HCV acute infection. The
Almost all studies focusing on the risk of acquiring strongest associations were found with abdominal, oral and
parenterally transmitted hepatitis associated with surgery obstetric/gynaecological surgery for HBV infection, and
or medical invasive procedures are represented by case with obstetric/gynaecological, abdominal and ophthalmolo-
reports or descriptions of epidemic clusters [17±24]. The gical interventions for HCV infection, which showed the
carrier state of health care workers [17±21], often involved highest magnitude of the OR point estimates.
in exposure-prone procedures, the contamination of the To our knowledge, this is the ®rst study showing an asso-
instruments used [22±24], and the extensive contamination ciation between HCV infection and ophthalmological
with blood of particular health care environments [25] have surgery, mostly performed for cataracts. Bacterial and
been identi®ed or suspected as the most frequent reasons for fungal contamination of the automated surgical equipment
virus transmission. Prospective cohort studies would repre- used during routine cataract surgery has been reported
sent the most adequate means to obtain valid estimates of [26,27]. Even if during cataract surgery, a serious blood
contamination usually does not take place, the inadequate
Table 4 sterilization of certain instruments used could probably
Scoring of the other variables included in the logistic regression analy-
favour patient-to-patient virus transmission.
sis model
Our data also highlight the importance of minor invasive
Variable Hepatitis B Hepatitis C procedures in the transmission of both HBV and HCV, such
OR (95% CI) OR (95% CI) as oral surgery, minor surgery, biopsy/endoscopy (asso-
Age
ciated only with HCV infection). The different risk esti-
15±24 1.0 1.0 mates associated with the various types of invasive
25±34 1.6 (1.4±1.8) 2.1 (1.7±2.7) procedure between HBV and HCV acute infections could
35±44 2.3 (1.9±2.8) 3.2 (2.4±4.4) be explained by the much higher prevalence rates of HCV
45±54 3.3 (2.6±4.2) 5.4 (3.8±7.6) infection in the general Italian population, particularly
55±64 8.1 (5.6±11.7) 24.5 (15.8±37.9)
65±74 18.5 (10.1±34.4) 106.0 (55.6±203)
among the elderly [4±9]. In fact, admitting a person-to-
75±84 10.6 (5.4±20.9) 28.1 (13.5±58.5) person mode of viral transmission by improperly sterilized
$85 3.0 (1.1±8.0) 8.2 (2.5±27.0) medical instruments, the larger pool of subjects infected
Sex with HCV than with HBV in the general population causes
Male 1.0 1.0 a higher likelihood of exposure to hepatitis C. This is parti-
Female 1.0 (0.9±1.2) 1.3 (1.0±1.5)
Educational level
cularly true for the risk associated with ophthalmological
$9 years 1.0 1.0 surgery, and even more in the case of cataract surgery, a
#8 years 1.7 (1.5±1.9) 2.4 (2.0±2.9) procedure mostly performed among elderly people who
Area of residence have very high prevalence rates of HCV infection [7±9].
North ± centre 1.0 1.0 The present results are consistent with those of two case±
South ± islands 0.14 (0.12±0.16) 0.26 (0.20±0.32)
Other parenteral exposure a
control studies carried out in Naples, Italy [12,13] in the
No 1.0 1.0 earlier 1990s. In these studies, an independent association
Yes 1.8 (1.6±2.0) 1.6 (1.3±2.0) was found to link acute HCV and HBV infection with surgi-
Number of sexual partners cal interventions and dental therapy; a very high risk of
#1 1.0 1.0 acute HCV infection (OR ˆ 32; 95% CI ˆ 7.5±147.0) was
$2 2.9 (2.6±3.4) 2.1 (1.7±2.7)
found for females exposed to obstetric and gynaecological
a
Ear piercing, tattooing, acupuncture, attendance at a chiropodist, or interventions. Furthermore, in a cohort study of blood
manicurist. donors, anti-HCV seroconversion was more frequently
288 A. Mele et al. / Journal of Hepatology 35 (2001) 284±289
observed among individuals undergoing surgery or invasive
procedures [14]. Thus, the high risk of parenterally trans-
mitted hepatitis associated with invasive procedures seems
to be constant over time, suggesting that this problem is not
limited to certain types of hospitals or health care providers.
It is probably dif®cult to control the transmission of blood-
borne viruses in some medical settings, particularly during
invasive procedures.
Considering the large proportion of the general popula-
tion undergoing surgery or other invasive procedures, the
present results stress the importance of complying with
universal precautions [28] and implementing ef®cient ster-
ilization and maintenance methods for medical instruments.
Ideally, disposable materials should be used, especially
when there is an increased risk of infection due to a high
turnover of patients, combined with emergency surgical
interventions, such as in gynaecology or emergency rescue.
The use of multidose vials should also be limited. Similarly,
protocols and effective sterilization procedures for dental
therapy and biopsy/endoscopy need to be improved. Exten-
sive HBV vaccination of health care workers, particularly
those involved in exposure-prone procedures [29] should be
implemented.
In the ®eld of invasive procedures for diagnostic or ther-
apeutic purposes, the above mentioned non-immunological
measures are the only effective tools in preventing the trans-
mission of HCV. Furthermore, until HBV vaccination
results in the accumulation of immune cohorts, these
measures also represent the only effective measures against
HBV transmission in adulthood; these measures are also
useful in preventing other bloodborne viral infections (e.g.
HIV).
We would like to comment on the potential biases of this
study. This is one of the very few studies that have focused
on patients with acute HBV and HCV infection for whom a
limited period of previous exposure (6 months) was inves-
tigated. Thus, any potential recall bias has been minimized.
Individuals with acute HAV infection were used as controls
to estimate the strength of the association between acute
HBV and HCV infection and various types of invasive
procedures. HAV has different modes of transmission
compared with HBV and HCV, although they can occasion-
ally share the same risk factors, such as intravenous drug use
or blood transfusion. However, patients reporting such
exposures were excluded from the study. In case±control
studies, the comparability between cases and controls is a
crucial point. All of the individuals with acute hepatitis were
noti®ed cases, coming from the same geographical area,
identi®ed through the same surveillance system, and inter-
viewed by the same blinded interviewer; they were thus
exposed to the same selective factors, if any. In this survey,
hepatitis A patients tend to be younger and from the south of
Italy, whereas in contrast, hepatitis B and C patients were
older and from the north. Using the multiple logistic regres-
sion analysis model, the confounding effects of these vari-
ables, as well as those of sex, educational level, other
parenteral exposures and intercourse with multiple sexual
partners, are likely to have been removed. Moreover,
control-group individuals over 65 years of age were poorly
represented; after restriction of the analysis to individuals
aged less than 65 years of age, the risk estimates for each
type of invasive procedure did not show substantial changes,
except for ophthalmological surgery. Thus, despite the fact
that hepatitis A patients cannot be considered as absolutely
the best controls, they may represent a valid and feasible
choice in this study. Finally, we are aware that among acute
hepatitis C cases, hepatitis exacerbations in HCV chronic
carriers may have been enclosed. However, since it is unli-
kely that HCV chronic carriers with acute exacerbations had
a history of exposure different from that of the general
population, this possible misclassi®cation could have led
to an underestimate of the true risk.
In conclusion, the remarkable sociodemographic changes
observed in the last two decades, more sensitive procedures
for screening blood donors for HBV and HCV, the impact of
AIDS information campaigns, and the introduction, in 1991,
of compulsory HBV vaccination of newborns and teenagers
(at 12 years of age) have all probably contributed to the
decrease in the incidence of parenterally transmitted hepa-
titis in Italy in recent years. In spite of this decreasing inci-
dence, exposure to diagnostic or therapeutic invasive
procedures still represents an important mode of acquiring
acute HBV and HCV infections.
Acknowledgements
The authors thank Mrs Antonella Marzolini for secretar-
ial assistance and database management, and Mrs Rossana
Bernacchia and Dr Franco Santonastasi for assisting in data
collection. The authors also thank Dr Maurizio Terrana for
information regarding ophthalmological surgery proce-
dures. Grants: supported by the Viral Hepatitis Project, Isti-
tuto Superiore di SanitaÁ (D. Leg. 30/12/1992 n. 502).
References
[1] De Bac C, Stroffolini T, Gaeta GB, Taliani G, Giusti G. Pathogenic
factors in cirrhosis with and without hepatocellular carcinoma: a
multicenter Italian study. Hepatology 1994;20:1442±1449.
[2] Stroffolini T, Andreone P, Andriulli A, Ascione A, CraxõÁ A, Chiar-
amonte M, et al. Characteristic of hepatocellular carcinoma in Italy. J
Hepatol 1998;29:944±952.
[3] Mele A, Stroffolini T, Pasquini P. Integrated epidemiological system
for acute viral hepatitis. Report 1985±1994. Istisan 1996;3:1±33.
[4] Stroffolini T, Chiaramonte M, CraxõÁ A, Franco E, Rapicetta M,
Trivello R, et al. Baseline sero-epidemiology of hepatitis B virus
infection in children and teenagers in Italy. A survey before mass
hepatitis B vaccination. J Infect 1991;22:191±199.
[5] D'Amelio R, Matricardi PM, Biselli R, Stroffolini T, Mele A, Spada
E, et al. Changing epidemiology of hepatitis B in Italy: public health
implications. Am J Epidemiol 1992;135:1012±1018.
[6] Stroffolini T, Guadagnino V, Chionne P, Procopio B, Mazzuca EG,
Quintieri F, et al. A population based survey of hepatitis B virus
 A. Mele et al. / Journal of Hepatology 35 (2001) 284±289
infection in a Southern Italian town. Ital J Gastroenterol Hepatol
1997;29:415±419.
[7] Bellentani S, Tiribelli C, Saccoccio G, Sodde M, Fratti N, De Martin
C, et al. Prevalence of chronic liver disease in the general population
of northern Italy: the Dionysos study. Hepatology 1994;20:1442±
1449.
[8] Stroffolini T, Menichelli M, Taliani G, Dambruoso V, Poliandri G,
Bozza A, et al. High prevalence of hepatitis C virus infection in a
small central Italian town. Ital J Gastroenterol 1995;27:235±238.
[9] Guadagnino V, Stroffolini T, Rapicetta M, Costantino A, Kondili LA,
Menniti Ippolito F, et al. Prevalence, risk factors and genotype distri-
bution of hepatitis C virus infection in the general population: a
community-based survey in southern Italy. Hepatology
1997;26:1006±1011.
[10] SEIEVA Collaborating Group, Mele A, Stazi MA, Gill ON, Pasquini
P. Prevention of hepatitis B in Italy: lesson from surveillance of type-
speci®c viral hepatitis. Epidemiol Infect 1990;104:135±141.
[11] Mele A, Tosti ME, Marzolini A, Moiraghi A, Ragni P, Gallo G, et al.
Prevention of hepatitis C in Italy: lesson from surveillance of type-
speci®c acute viral hepatitis. J Viral Hepat 2000;7:30±35.
[12] Sagliocca L, Stroffolini T, Amoroso P, Manzillo G, Ferrigno L,
Palumbo F, et al. Risk factors for acute hepatitis B: a case±control
study. J Viral Hepat 1997;4:63±66.
[13] Mele A, Sagliocca L, Manzillo G, Converti F, Amoroso P, Stazi MA,
et al. Risk factors for acute non-A, non-B hepatitis and their relation-
ship to antibodies for hepatitis C virus: a case±control study. Am J
Public Health 1994;84:1640±1643.
[14] Prati D, Capelli C, Silvani C, De Mattei C, Bosoni P, Pappalettera M,
et al. The incidence and risk factors of community-acquired hepatitis
C in a cohort of Italian blood donors. Hepatology 1997;25:702±704.
[15] SEIEVA Collaborating Group, Mele A, Rosmini F, Zampieri A, Gill
ON. Integrated epidemiological system for acute viral hepatitis in
Italy (SEIEVA): description and preliminary results. Eur J Epidemiol
1986;2:300±304.
[16] Dixon WJ, Brown MB, Engelman L, Jennrich RI. BMDP statistical
software. Berkeley, CA: University of California Press, 1990.
[17] Esteban JI, Gomez J, Martell M, Cabot B, Quer J, Camps J, et al.
Transmission of hepatitis C virus by a cardiac surgeon. N Engl J Med
1996;334:555±560.
289
[18] Duckworth GJ, Heptonstall J, Altken C. Transmission of hepatitis C
virus from a surgeon to a patient. The Incident Control Team.
Commun Dis Public Health 1999;2:188±192.
[19] Brown P. Surgeon infects patient with hepatitis C. Br Med J
1999;319:1219.
[20] The Incident Investigation Teams and others. Transmission of hepa-
titis B to patients from four infected surgeons without hepatitis B e
antigen. N Engl J Med 1997;336:178±184.
[21] Harpaz R, Von Seidlein L, Averhoff FM, Averhoff FM, Tormey MP,
Sinha SD, et al. Transmission of hepatitis B virus to multiple patients
from a surgeon without evidence of inadequate infection control. N
Engl J Med 1996;334:549±554.
[22] Bronowicki JP, Venard V, Botte C, Monhoven N, Gastin I, ChoneÁ L,
et al. Patient-to-patient transmission of hepatitis C virus during colo-
noscopy. N Engl J Med 1997;337:237±240.
[23] Le Pogam S, Bacq Y. Nosocomial transmission of hepatitis C. Ann
Intern Med 1999;131:794.
[24] Birnie GG, Quingley EM, Clements GB, Follet EA, Watkinson G.
Endoscopic transmission of hepatitis B virus. Gut 1983;24:171±
174.
[25] Piazza M, Borgia G, Picciotto L, Nappa S, Cicciarello S, Orlando R.
Detection of hepatitis C virus RNA by polymerase chain reaction in
dental surgeries. J Med Virol 1995;45:40±42.
[26] Mino de Kaspar H, Grasbon T, Kampik A. Automated surgical
requires routine disinfection of vacuum control manifold to prevent
postoperative endophthalmitis. Ophthalmology 2000;107:685±690.
[27] Kappstein I, Schneider CM, Grundmann H, Scholz R, Janknecht P.
Long-lasting contamination of a vitrectomy apparatus with Serratia
marcescens. Infect Control Hosp Epidemiol 1999;20:192±195.
[28] Centers for Disease Control. Update: universal precautions for
prevention of transmission of human immunode®ciency virus, hepa-
titis B virus, and other bloodborne pathogens in health-care settings.
MMWR Morb Mortal Wkly Rep 1988;37:377±382 387-388.
[29] Other AN. Recommendations for preventing transmission of human
immunode®ciency virus and hepatitis B virus to patients during expo-
sure-prone invasive procedures. MMWR Morb Mortal Wkly Rep
1991;40(RR-8):1±9.