Introduction
In Italy, the screening of blood for hepatitis C virus (HCV) RNA became
mandatory in July 2002. Recommendations regarding the introduction of
nucleic acid amplification testing (NAT), or, as an alternative, enzyme
immunoassay (EIA) tests capable of shortening the HCV window period,
had been available from the Ministry of Health since 2001. In fact, during
the above period the HCV core antigene (HCVcoreAg) EIA assay had been
employed in blood screening, because its efficacy is similar to that
of NAT testing, and it is extremely easy to use in blood transfusion
centres.
The obligation to employ NAT technology for blood screening was limited
to HCV, because of this assay’s ability to reduce the window period
of this infection by almost 80% (from 70 to 12 days) [1]. Such reductions
are lower in the case of other important transfusion-transmissible infections.
In practice, for a number of reasons - such as the wish to guarantee higher
levels of safety in transfusion therapy, the combination of a number of
assays in a single commercial kit, and the fact that health decisions are
made at the regional government level - almost half of the blood units
collected in Italy are also screened for HIV RNA and, in a lower number
of cases, for hepatitis B virus (HBV) DNA as well.
The maintained availability of the previously employed EIA screening tests
allowed a comparison of the EIA-based residual risk projections, calculated
with mathematical models, with the effective yields of the new technology.
A national survey was therefore organised with the following aims:
1) to study the organisational aspects of the introduction of NAT testing
in Italy;
2) to assess the incidence of transfusion-transmitted infections, as detected
by new technologies;
3) to evaluate the national distribution of HIV RNA screening, which is
currently not mandatory in Italy;
4) to compare the new values of residual risk with existing data derived
from serological assays employed in Italy.
Methods
The survey was promoted by the Gruppo Italiano per lo Studio delle Malattie
Trasmissibili con la Trasfusione (Italian Group for the Study of Transfusion-Transmissible
Diseases), part of the Settore Ricerca & Sviluppo della Società Italiana
di Medicina Trasfusionale e Immunoematologia (Research & Development
Department of the Italian Society of Transfusion Medicine and Immunohaematology),
SIMTI.
From 2001 and 2003, an annual questionnaire was sent to all 308 Italian
blood transfusion centres.
Assay manufacturers (Roche Diagnostics (Roche Molecular System, Branchburg,
NJ, USA), Chiron Corporation (Chiron, Emeryville, CA, USA) and, for the
first year only when HCVcoreAg was used, Ortho Clinical Diagnostics (Ortho-Clinical
Diagnostics, Raritan, NJ, USA)) were requested to collect the questionnaires
through their commercial networks, and to pass the data on to the working
group where it was collated.
Data collection for both NAT and non-NAT procedures started in 2001,
the year of the first experimental employment of NAT testing for blood
screening prior to the date of its introduction by law (28 June 2002).
In the same period, several blood transfusion centres introduced both
HBV DNA and HIV RNA testing as part of their screening policy.
During the three-year investigation period, 219 blood transfusion centres
returned the questionnaires. During this time, 3 894 894 blood donations
(representing approximately 80% of blood donations collected in Italy
in the same period) were investigated for HCV RNA. A total of 1 798 693
units in 29 blood transfusion centres were tested by Chiron TMA HIV/1-HCV
and additional 2 096 201 units in 81 Blood transfusion centres were tested
with Cobas Ampliscreen HCV Roche, on 10-24 sample minipools.
In 2001, an additional 850 080 units were tested in 109 blood transfusion
centres by using Ortho HCVcoreAg instead of NAT.
In 2002 and 2003 HIV RNA screening data were collected as well. 2 186
468 units were examined, of which 1 640 278 units in 29 blood transfusion
centres were tested by Chiron TMA HIV/1-HCV and 546 190 units in 37 blood
transfusion centres, by Cobas Ampliscreen HIV Roche.
The blood units tested by the Roche assay were minipools ranging from
10 to 24 samples. The Chiron system was applied to single samples in
80% of blood donations and to 8-sample pools in the remaining 20%.
Before the introduction of NAT, the predicted residual risk for HCV
and HIV transmission was calculated by applying the incidence/window
period model [2,3] on the basis of serological tests performed on approximately
four million blood units collected in Lombardia (northern Italy) between
1996 and 2003 [4,5]. The blood units collected in Lombardia amount to
the 20% of the total number of blood donations in Italy each year.
Results
Table 1 illustrates the overall study data collection setting, including
the number of participating centres, number of units examined by each
test, and average number of tests performed per centre.
In 2001, when HCVcoreAg was performed as an alternative to NAT testing,
167 centres examined a total of 1 219 033 units. During 2002 and 2003,
when only NAT testing was allowed, 94 centres provided results for 3
525 941 units.
Table 2 shows the numbers of units tested, as well as the number of centres,
by the method employed.
Of the 850 080 units tested by HCVcoreAg assay, none resulted positive
with negative HCV antibodies.
Of the 3 894 894 units tested for HCV RNA, 616 tested positive for both
HCV RNA and anti-HCV and 12 tested positive for HCV RNA alone.
Of the 2 186 468 units tested for HIV RNA, 59 resulted positive for HIV
RNA and anti-HIV, and four for HIV RNA alone.
Of the 12 HCV-infected donors detected during the window phase, seven
had normal ALT levels, and five had abnormal values.
All donors who had been HCV or HIV NAT positive but antibody negative
at the time of donation seroconverted during the follow-up period. The
lower levels of viral load detected during screening were approximately
100 000 IU/ml and 21 000 copies/ml for HCV and HIV, respectively.
One HCV RNA positive/anti-HCV negative donor and one HIV RNA positive/anti-HIV
negative donor were first time donors.
The risk factors reported were: a positive sexual partner (2 HCV- and
4 HIV-infected donors), drug use (n=1), surgery (n=1), and unidentified
(n=8).
Table 3 illustrates the residual risk of transmitting HCV and HIV based
on serological testing, the projected yield and the estimated residual
risk after NAT implementation, and the number of infectious units detected
by NAT in the window phase during the period 2001-2003.
Discussion
The introduction in Italy of NAT screening for blood safety determination
was complicated by several organisational difficulties, including the
large number of blood transfusion centres authorised to perform the
biological validation of donated blood. In fact, only a minority of
centres (107 by the end of 2003) was authorised to screen blood using
NAT methods, thus promoting a departmental reorganisation of blood
transfusion centres.
Since July 2002, HCV RNA testing has been routinely carried out as part
of blood screening procedures in Italy, and the numbers of units screened
for HIV RNA, and more recently for HBV DNA, are increasing.
HCVcoreAg has been completely abandoned.
The data collected in this survey, especially during its third year,
covered approximately 80% of the entire 2.5 million units collected in
Italy annually, and approximately 90% of the authorised centres provided
data.
For the overall 2001-2003 period, 46% of blood donations were tested
with Chiron assays and 54% with Roche assays: the number of centres using
Roche technology was higher than those using Chiron assay, but the number
of blood units tested was similar.
On the basis of serological data previously collected in Lombardia and
taking into account the data collected during the years 2001-2003 (corresponding
to the first period of NAT implementation), the estimated residual risk
for transfusion-transmitted infection was 2.7 x 106 for HCV,
and 2.2 x 106 for HIV infection [4]. Similar data has been
reported by others in Italy [6].
From the implementation of NAT screening until the end of 2003, 3 894
894 units of blood were tested for HCV RNA and 2 186 468 for HIV RNA.
Twelve HCV RNA-positive/anti-HCV antibody-negative, and four HIV RNA
-positive/anti-HIV antibody-negative donors were detected, with an observed
NAT versus antibody-based assay yield of 3.1 per 106 donations
for HCV and 1.8 per 106 for HIV, respectively. Significantly,
5 of the 12 HCV RNA positive/anti-HCV negative donors had abnormal ALT.
Since ALT testing is systematically performed in Italy, donations from
such donors would have been discarded even in the absence of NAT results.
Thus, the yield of NAT versus all mandatory tests for HCV is 1.79 per
106 donations.
The projected values (2.2 per 106 for HCV and 1.1 per 106 for
HIV) were calculated on the basis of epidemiological data collected in
the Lombardia region, which amounts to approximately one fifth of Italy’s
total number of blood donors and donations. Differences between the observed
and the expected yields were not significant. This data indicates the
satisfactory quality of both the surveillance system and the mathematical
model.
So far, data on transfusion-transmitted HBV infection have not been collected
at a national level, although there are plans to do so in the future.
At present, the Ministero della Salute (Ministry of Health) is not planning
to introduce HBV NAT testing for blood screening although the HBV predicted
residual risk, calculated through mathematical modelling based on incident
infections in donors screened in Lombardia during the period 1996 and
2003, is estimated to be 13.9 per 106.
Now that NAT has been implemented, the residual risk for transmitting
HCV or HIV by blood transfusion in Italy is extremely low. The surveillance
system described in this publication will be maintained to observe eventual
shifts in the epidemiology of these infections, as well as the opportunity
to introduce additional assays or to remove some of the currently performed
tests.
|