Introduction
Pork, horsemeat and game may be infected with muscle larvae of the zoonotic
nematode Trichinella, which can cause severe disease in humans.
Consequently, all countries in the EU perform mandatory official inspection
of slaughtered animals intended for export to prevent distribution of
infected meat to consumers. Trichinella infections have worldwide
socioeconomic significance, and are of medical and veterinary concern
in France, Germany, Italy, and Spain, but foremost in the east and central
European countries [1, 2] where human trichinellosis is reported to be
a very important zoonosis. Some of the new EU member states (Latvia,
Lithuania and Estonia) as well as some candidate countries (Bulgaria,
Romania, Turkey and Croatia) have outbreaks every year (reported by the
International Commission on Trichinellosis (ICT) [3]. The costs for inspection
of pork in the EU is estimated to €570 million annually [1, 4].
All procedures for Trichinella inspection are based
on direct detection of the parasite larval stages in muscle tissue,
initially (from around 1860s) by direct microscopy of compressed
muscle tissue, termed trichinoscopy [5.6.7.8.9,10], later (in
the 1970s) by pooled digestion of 1g muscle tissue from up to
100 pigs, which allowed for significant improvements in sensitivity
of the inspection test and were less labour intensive, hence
allowing larger numbers of animals to be examined.
In the present EU legislation (Directive 77/96/EEC), seven methods
are accepted [TABLE 1]: six digestion methods and trichinoscopy.
In the anticipated future EU legislation (SANCO/1900/2002 Rev.
8 draft, in force 01-2006), the number of inspection methods
has been reduced to four with magnetic stirrer digestion as the
reference method to be preferred before three alternative (termed ‘equivalent’)
methods. Trichinoscopy is only allowed as a transitional measure,
and meat inspected by this method should be clearly marked. Furthermore,
such meat is limited to be sold on the national market and is
not acceptable for products where the production process does
not kill Trichinella. The digestion methods have a theoretical
detection limit of down to 1 larva per gram muscle tissue (lpg).
However, there are several critical steps, which may compromise
the sensitivity of the techniques [11, 12] and many of these
are not adequately addressed in the new EU legislation.
Trichinella inspection methods after the present
legislation
Below are brief descriptions of the methods allowed for meat inspection
according to directive 77/96/EEC, Annex 1, (amended by Council Regulation
(EC) No. 807/2003) [TABLE 1], along with some critical points and suggestions
for improvements.
Method I: Trichinoscopy
The classical method for detection of Trichinella in pork is trichinoscopy
(also termed the compressorium technique). Muscle samples from each
of the two diaphragm pillars are cut into 7 very small (oat kernel
sized) pieces, which are subsequently squeezed between two glass plates
and examined under a microscope at 30-40 X magnification for the presence
of capsules containing Trichinella larvae. The microscopic
examination must last at least 3 minutes to ensure adequate time for
the finding of larvae. For routine inspection, trichinoscopy is labour
intensive, it is not as sensitive as the digestion methods (examines
less tissue, 14 oat kernel sized pieces ~0.5 g) and finally does not
detect larvae of T. pseudospiralis as this species lacks the
physical structure (a surrounding collagen capsule) that is detected
for other Trichinella species 2-4 weeks after infection. Due
to the inherent errors of trichinoscopy, this method should no longer
be used, hence there are no suggestions for improvements.
Methods II and III: Digestion
of single or pooled samples with either no mechanical intervention
or manual shaking of digestion fluid
These manual methods allow for artificial digestion of pools of minced
meat samples (10 grams from each of 10 pigs (method II) and 1 gram from
each of 100 pigs (method III). Any Trichinella larvae present
in the pooled sample are released into the artificial digestion fluid
and settle at the bottom of the beaker. For method II, the digestion
fluid is left undisturbed for 18-20 h, whereas for method III the fluid
must be shaken twice per hour for 4 h. The sediment from the digest is
examined for larvae under a stereomicroscope at 20-40 X magnification.
Although the methods work on a pool of samples, they are too time consuming
and with the inbuilt risk of dead or young larvae being digested along
with the muscle tissue resulting in a false negative outcome. Since other
pooled digestion methods are superior, methods II and III have been omitted
from future legislation.
Methods IV and V: Mechanical
digestion (with a stomacher blender) of pooled samples followed
by either sedimentation or filtration
The pooled sample (100 x 1 g pieces) and the digestion fluid are mixed
in a plastic bag placed in the stomacher chamber where it is mechanically
agitated for 25 min at 41°C. The fluid is then passed into a sedimentation
funnel through a 177 µm sieve with addition of ice and either left
to sediment under 1 minute vibration cycles every second minute for 30
min total (method IV), or poured through a 35 µm filter, which
will hold back larvae (method V). Larvae have a tendency to adhere to
the plastic bag causing a risk for false negative readings. Thus, the
sensitivity of this method is less than the theoretical 1 lpg. Because
the pooled sample consists of 1 g pieces, there is a risk of undigested
residue after digestion for the recommended time (authors’ own
observations). For the improvement of these methods, the pooled sample
may be subjected to blending or mincing prior to digestion, the initial
filtration to retain undigested particles could be done with a larger
mesh size (355µm) that allows all larvae to pass (see Method VI
below), and finally, the adhesion of larvae to the plastic bag could
be lessened by flushing the plastic bag twice.
Method VI: Mechanical digestion of pooled samples
with magnetic stirrer
Minced or blended meat samples are placed in the digestion fluid for
30 min at 46-48 C° under constant stirring by the use of a magnetic
stirrer, and subsequently poured through a sieve into a sedimentation
funnel. After a sedimentation period of 30 min, the sediment is removed
from below the funnel, and the volume further reduced through more sedimentation
steps. The digestion is more complete with this method because the 1
g meat samples are minced. Improved larval recovery can be obtained by
changing the filter size from 177µm (180µm) to 355µm
(11; authors’ own observations).
Method VII: Mechanical digestion with the Trichomatic35
The Trichomatic35 apparatus blends, digests and filters a maximum of
35 pooled 1 g samples in one short process (5-8 min). Digested material
is filtered under high pressure and the resulting filter is examined
under a stereomicroscope as above. This method is fast with a high
sensitivity [13] but a disadvantage might be that the filter requires
extra washing procedures to prevent cross contamination between samples
[14]. It is therefore recommended to use a new filter for every sample
tested. The Trichomatic35 is no longer on the market and once the existing
spare parts have been distributed, no more will be available from the
manufacturer.
Future legislation and performance of the future recommended
techniques
In the future EU legislation (SANCO/1900/2002 Rev. 8 draft, in force
01-2006), the magnetic stirrer method is identified as the reference
method and the two versions of the stomacher method and the Trichomatic35
method may be considered equivalent methods if the magnetic stirrer method
is not accessible. For routine inspection, trichinoscopy will only be
allowed as a national transitional measure, as it does not detect the
non-encapsulating species, T. pseudospiralis, or young larvae
of encapsulated species with incomplete capsule development. Thus, meat
inspected by trichinoscopy cannot be sold to other EU countries or exported
out of the EU.
Related to the four digestion methods, which remain in the future
legislation, there are inherent critical aspects that compromise
the sensitivity of the methods and therefore need to be dealt
with. These aspects are, for example, related to washing and
sieving procedures, the nature of employed materials (plastic
versus glass), incubation times, contamination problems, and
the condition of the meat to be inspected [11, 15, 16]. Other
problems are related, for example, to the technical equipment
failure, enzyme failure, and human errors, which all lead to
a lack of compliance with protocols [17], reducing the efficiency
of the methods.
At least two published studies have demonstrated that the sensitivity
of the recommended methods is lower than stated [11, 12]. Forbes
and Gajadhar [18] documented a higher sensitivity of the magnetic
stirrer method when compared with trichinoscopy. In early studies
forming the basis for recommendation of the stomacher method,
larval recovery as low as 79% was reported [9]. A recent comparative
testing at the Danish Institute for Food and Veterinary Research
(Maddox-Hyttel et al, unpublished data) indicates that the sensitivity
of the magnetic stirrer is lower than required by the legislation,
and importantly, both the sensitivity and reproducibility of
the stomacher methods are considerably lower as compared to the
magnetic stirrer method. Thus in the test, the recovery of larvae
spiked into ground meat, varied from as little as 34-40% using
the stomacher method (V) to an average of 63% (range 18-86%)
or 85% (range 74-100%) using the magnetic stirrer method with
filter mesh size of 177µm (recommended in the present and
future EU legislation) or 355µm (recommended by Gamble
[11]), respectively. Trichinella larvae from the meat
were obviously lost at various steps of the procedures and these
steps need to be identified and corrected through optimisation
measures to ensure reliable detection methods.
The sensitivity is also related both to the amount of meat and
the type of muscle tissue used for inspection [12,19,20,21],
and increasing the sample size would improve any detection method
[12,18,22]. The detection limit for the artificial digestion
is reported to be approximately 1 lpg, if at least 5g of muscle
sample per animal is digested [23]. However, according to the
legislation, the recommended amount of tissue for pork allows
sampling of down to 1g/pig and, as a consequence, the detection
may be only 3-5 lpg rather than 1 lpg as stated. The inspection
methods are intended to have a detection limit to prevent clinical
trichinellosis. There are, however, only estimates [24] and no
reliable data on the actual margins of such a limit. Consequently,
the detection limit should be as low as possible.
The efficacy of the above digestion tests when used on meat
from horses, wild boar, and other animals, is relatively unexplored
although important because digestibility varies considerably
both between muscle types and animal species. Some muscle groups
from horses are readily digested within 30 min (diaphragm, tenderloin,
fillet, and rump), whereas others need up to 2-3 times as long
(masseter, tongue and leg muscles) [25]. These requirements for
longer digestion time according to muscle type of different hosts
have not been addressed in the new legislation and hence, the
recommended digestion times may lead to an incomplete digestion
of several grams of tissue, depending on the choice of muscle.
Consequently, it is imperative that the sensitivity of each method
should be listed in detail for different muscle types and animal
species.
Thorough comparison of the efficiency of the recommended detection
methods (excluding the Trichomatic35, which is no longer produced)
is therefore required and the future legislation should include
a revised description with correct sensitivity and reproducibility
of each muscle type from target animal species. Furthermore,
guidelines for proficiency testing are urgently needed to ensure
optimal test accuracy and quality of inspection. A recent ring
trial among 33 laboratories in Germany [16] only emphasises this
need; half the laboratories participating detected false negative
or false positive results in between one and six of 10 examined
samples. Meat samples for the trial, were prepared as duplicate
samples containing between 8 and 71 T. spiralis larvae
per gram of meat (that is, a high infection level), or without
any larvae (negative controls), and were examined using the magnetic
stirrer method. The draft of the future legislation (SANCO/1900/2002
Rev. 8 draft, in force 01-2006) states that the competent authority
should ensure that all personnel, who are involved in the examination
of samples to detect Trichinella, are properly trained, participating
in proficiency testing programs and in a regular assessment of
the sensitivity and the specificity of the test involved. However,
hitherto no protocols or guidelines have been formulated for
uniform proficiency testing and quality assurance systems in
the EU.
Level of implementation of direct detection techniques
in EU
Tables 2, 3 and 4 aim to provide an overview of the rather heterogeneous
implementation levels of Trichinella inspection in the EU member states.
Especially for horsemeat and wild boar meat, data are scarce due to the
lack of registration within several countries. Although meat inspection
for Trichinella is mandatory in the EU, registration and reporting of
the number of animals inspected and the methods by which inspection was
performed is not required. Comparing the available data on the present
inspection methods for a range of EU countries, it is evident that many
countries do not have optimal Trichinella control. Most countries and
laboratories have implemented the magnetic stirrer method at the large
slaughterhouses, however all seven methods are reported to be in function
in several EU countries and according to the personal experience of the
authors, even at the national level, there can be as many variations
of the techniques as there are laboratories. Furthermore, a surprisingly
large number of countries still use trichinoscopy and although it is
likely that the method is primarily applied to detect Trichinella larvae
in muscles from wildlife or from a limited number of domestic pigs (single
animal examination), the use of this technique represents a major problem.
Because of the low sensitivity and inability to detect of T. pseudospiralis,
this method should be abolished as soon as possible.
Conclusions
In conclusion, there are several indications that the sensitivity of
the recommended methods - used in their present form - is effectively
lower and more variable than stated in the present legislation and
accordingly also in the new EU Commission legislation draft for the
future meat inspection procedures. Despite the fact that the new legislation
draft requires quality control on the actual procedures, and calls
for proficiency testing of Trichinella control laboratories,
there are presently no guidelines for proper and uniform proficiency
testing of the recommended direct detection methods. Thus, the future
challenge is to develop and implement a meat inspection system, which
is more complete, comprising a fully optimised gold standard method
for Trichinella detection with reliable sensitivity and in
addition provide guidelines for a quality assurance system to ensure
uniform meat inspection within the EU. This will ensure a high quality
of food and food safety for the consumers, and reinforce export opportunities.
Acknowledgements
The work was conducted under MedVetNet (WP 11: TRICHINET) in the EU FP6,
Quality of Life and Management of Living Resources (QoL) and the EU TRICHIPOSE
contracts QLRT-2000-01156, QLKI-CT-2002-02826. The authors thank Mr.
Ronald Dwinger, SANCO and the International Commission for Trichinellosis
for the data provided.
|