Over the past two decades, a long series of specific
and non-specific measures have been introduced into the screening of blood
donations in order to reduce the residual risk of transmission of bloodborne
viruses. The latest specific measure has been viral nucleic acid testing
(NAT), introduced by the European plasma industry in 1995, and subsequently
introduced for blood donations in several countries in Europe and elsewhere.
NAT was implemented to reinforce the safety of the blood supply; it can
detect acute viral infections during the ‘window period’, that
are not detected by the serological screening methods. To assess the impact
of NAT on the safety of the blood supply, it is essential to estimate the
residual risk of viral transmission. In this
issue, six European countries (France, Germany, Italy, Spain, Switzerland
and the United Kingdom) that have recently implemented NAT describe their
experiences and the results of the evaluation of the residual risk of viral
transmission in their blood supply [1-6].
In these six European countries, NAT was initially introduced between 1999
and 2001 to detect hepatitis C virus (HCV), probably because the first
mandatory screening for plasma used by blood industry was HCV-NAT. In 2001,
a publication from an international forum showed that 10 out of the 25
countries that now make up the European Union had introduced HCV-NAT for
blood screening
versus two for HIV-NAT [7]. Later, HIV-NAT was progressively implemented
and, Spain is now the only country of the six reported in this issue where
this procedure has not yet been introduced. This expansion is probably
due in part to the ability to test for both viruses with one of the licensed
tests (TMA, Chiron blood testing). France is the only country where NAT
was implemented in a single stage for all blood donations collected. In
other countries, NAT was first performed on a voluntary basis, before it
was made mandatory.
In Germany, NAT is performed by ‘in-house’ assay, and the other
five countries use one or both of the commercially available nucleic acid
amplification methods (polymerase chain reaction (PCR) and transcription-mediated
amplification (TMA)), adapted for blood screening. Blood screening strategies
differ in the six countries, and there are two levels of heterogeneity
in the European practice of NAT. First, the number of blood donations included
in pools: these varied between 1 to 96 depending on the country. Second,
the variations observed in the procedures used within each country. In
France, Germany and the UK, the size of the pool is fixed for each virus,
whereas in Italy, Spain and Switzerland, the pool size varies. The variation
observed is probably due to the way in which blood donation testing is
organised locally. It should be noted that, contrary to the classical serologic
screening methods that are always used in single donation testing, current
NAT procedures usually demand pooling of blood donation samples due to
the format of the employed platforms.
The main aim of introducing NAT in blood testing was the reduction of the
residual risk of viral transmission linked to the window period. With the
exception of the UK, which has adopted a specific model ( see below), each
country bases the residual risk estimate on the mathematical model developed
by Schreiber et al [8], which takes into account the window period and
the incidence rate calculated from seroconversions observed in the repeat
blood donor population. However, due to difficulties in obtaining exhaustive
data at national level for the calculation of the national incidence rate,
most of the contributors have extrapolated from regional or partial data
that probably introduce biases. Although widely adopted, this mathematical
model has some limitations: it does not take into account the population
of first time blood donors or other parameters such as technical or human
errors or assay failures that could be implicated in the residual risk.
However, this model was validated by the observed yield of NAT [1]. The
UK has adapted the Schreiber model by using an adjustment factor in order
to evaluate the incidence rate in new donors, by calculating the risk due
to test and process errors, and by using different infectious window periods
than those currently adopted. It is therefore difficult to compare the
results obtained in the UK with those from other European countries.
All countries that analysed trends in the residual risk showed evidence
of a decrease. This trend started before the implementation of NAT, probably
due to better selection of blood donors and to preventive measures taken
in general population to avoid new infections. Before NAT implementation,
the residual risk for HCV transmission ranged from 0.64 (France) to 3.94
(Spain) per million donations, with a north-south gradient linked to HCV
epidemiology. The residual risk for HIV transmission, excluding the UK,
was estimated at between 0.59 (France) and 2.48 (Spain) per million donations.
Since NAT implementation, the residual risk for HCV transmission has ranged
between 0.1 (France) to 2.33 (Spain) per million donations and for HIV,
from 0.18 (Germany) to 1.1 (Italy) per million donations.
Yield rates observed for HIV-NAT are similar in France and Germany (about
0.3 per million donations). The higher rates observed in Italy and the
UK may reflect an increased HIV incidence in their donor populations, but
a bias due to the small number of donations screened by NAT, especially
in the UK, cannot be excluded. For HCV, the rates of NAT benefit are five
to six times lower in northern countries (from 0.5 per million donations
in Switzerland to 0.7 in the UK) than in Mediterranean countries (1.84
per million donations in Italy and 2.35 in Spain). This indicates that
the yield of HCV-NAT screening is limited in geographical areas where HCV
incidence rate is known to be very low. However, NAT has not been used
for very long, so more time and perspective are needed. Therefore, these
data should be interpreted with caution.
Despite a consensus stating that the main residual risk is currently due
to hepatitis B virus (HBV) - ranging from 10 in Spain to 1.6 per million
donations in France and Germany - only Germany reports systematically performing
HBV-NAT, a strategy which remains controversial. Indeed, it was established
that by comparison with current serological screening strategies based
on very high sensitive assays for the detection of hepatitis B surface
antigen (HBsAg), the expected benefit of the introduction of HBV-NAT screening,
especially with MP-NAT would be poor in terms of discarded donations and
clinical impact, particularly in a population that had been widely vaccinated
[9] . HBV DNA screening would be more effective in countries with high
or medium endemicity, and where anti-HBc testing is not routinely done.
Today, NAT implementation for HCV and HIV-1 is taken for granted in most
high-income countries to ensure the maximal viral safety. However, procedures
are heterogeneous and mainly adapted to the organisation of blood supply
of each country. National experiences reported in this issue of Eurosurveillance
are limited to western European countries and are not representative of
eastern Europe, or of Europe as a whole. The results of a study carried
out in 18 European countries by a European network of scientific societies
(Euronet TMS) describing the NAT situation in Europe will be published
in June 2005 in a specific report [10]. This overview will serve as a base
for further international surveillance in order to facilitate the harmonisation
of NAT in Europe. Today, the question of NAT’s cost-effectiveness
is debated. Several models have demonstrated that this measure is not cost
effective but no country has yet decided to withdraw it. Developing countries
that have not yet implemented NAT should be advised that alternatives to
NAT exist; in particular, serological assays which allow detection of viral
antigens independently or simultaneously with antibodies. These assays
offer improved safety at an affordable cost and circumvent the need to
re-organise national blood services.
|