|
Introduction
Salmonella enterica Paratyphi B variation Java, or simply Java, began
to increase alarmingly among chickens and in chicken products in the
Netherlands in 2000. This serovar causes gastro-enteritis in humans
through the consumption of contaminated food, but it can also be invasive,
producing typhus-like clinical symptoms, and lead to outbreaks (1-4).
In this article, we try to assess the potential threat to public health
associated with the increase of Java among poultry in the Netherlands.
Materials and methods
The National Salmonella Centre (NSC) and the National and European Reference
Laboratory (NRL) for Salmonella at RIVM identify all the isolates taken
from humans (mostly sent by regional public health laboratories covering
64% of the Dutch population) and animals, from food, animal food, and
from the environment. The Food Inspection Department East (KvW, Zutphen)
analyses chicken products from butchers, supermarkets, and poultry farmers.
Sampling is done in a representative and statistically sound manner
(regarding products, points of sale, regions, and seasons) (5). Antibiograms
were carried out at NSC until 2001, using the agar diffusion method
(ROSCO® tablets). Since 1999, the sensitivity of isolates to various
antibiotics has been quantitatively determined by the minimal inhibitory
concentration (MIC) at CIDC-Lelystad. Pulsed-field gel electrophoresis
(PFGE) typing was performed on 27 isolates derived from poultry, and
22 from humans, collected between 1998 and 2002, following the Salm-gene
scheme (see article pp46-50).
Trends in S. Java in The Netherlands and abroad
Results from the monitoring of chicken products by the KvW show that
positive Java samples increased drastically, even though the total percentage
of salmonella-positive samples more than halved since 1995 (table 1).
In poultry, the proportion of isolates with Java rose from 3% in 1995
to 15% in 1997, and then to 33% in 2000 and 61% in 2002 (figure). The
NSC also noted the same increase in products derived from chickens.
Although rarely, Java has also been isolated from humans : 51 isolates
were received at the NSC between 1996 and 2002 (0.3% of all human isolates).
In 1998 and 2000, information via Enter-net (the European surveillance
network for human infections involving salmonella and Verocytotoxin-producing
E. coli) revealed a similar low level of salmonellosis due to Java in
humans in other European countries. At meetings of the European NRLs
in 1999 and 2000, NRL-Berlin and NRL-The Netherlands reported the same
development of Java in poultry since 1995, and the explosive increase
in 2000 (figure). The NRLs from other European countries considered
that this was uniquely a German and Dutch problem. In December 2002,
however, Scotland sent out a request for information via Enter-net regarding
what seems to be an emerging problem in humans: the infection with multiresistant
Java (see article pp35-40).
Table 1
Results of the monitoring of chicken products by the Food Inspection
Department, 1995-2002*
|
|
1995 |
1996 |
1997 |
1998 |
1999 |
2000 |
2001 |
2002* |
| Prélèvements
analysés / Samples tested |
1359 |
1325 |
1314 |
1077 |
859 |
1454 |
1578 |
1196 |
| Salmonella spp. positive
(%) |
34.2 |
32.6 |
29,1 |
20.2 |
17.6 |
21 |
16.3 |
13.2 |
| Hadar positive (%) |
10 |
5.2 |
2.9 |
1.2 |
0.8 |
0.7 |
0.7 |
0.1 |
| Enteritidis positive (%) |
6.8 |
12.1 |
8,8 |
nd |
4.8 |
1.4 |
1.4 |
0.5 |
| Paratyphi B var. Java positive
(%) |
0.8 |
3.4 |
4,4 |
2.3 |
2.4 |
7.0 |
7.0 |
8.1 |
Antimicrobial resistance
The poultry sector in the Netherlands first encountered problems with
Java in five chicken farms in 1996. Currently 50-100 farms are battling
this infection. Java clones persist despite perfectly adequate quality
control scores (8), whereas S. Enteritidis, Hadar, Infantis, Virchow
etc, can be successfully eliminated with the standard cleaning and disinfection
procedures. Still, Java was no more resistant to disinfectants than
other serotypes found in the farms at the same time.
Java isolates found in poultry in the Netherlands are generally multiresistant,
in contrast to the other serotypes (table 2). According to the resistance
pattern, the human clones isolated from the 1980s until the mid-1990s
most closely resemble the genetically diverse and antibiotic-sensitive
strains prevalent in 1960-90s (9-11). The poultry isolates until 1995
most likely belong to a group deriving from a few clones multiresistant
to chloramphenicol, sulphonamide, tetracycline, trimethoprim and often
also kanamycin, neomycin and nalidixic acid (9). The isolates from 1996
onwards most likely derive from a recent clone, which had practically
replaced all the others in poultry, and is resistant to trimethoprim
in combination, to sulphonamide, streptomycin, nalidixic acid and ampicillin
(9). This clone was also found in Germany by molecular typing for two
isolates from poultry sent to the NRL-Berlin and, in December 2002 in
the Netherlands, for all the 27 preserved Java poultry isolates in 1998-2002.
A worrying development is the rapid increase in resistance of the Java
clone against quinolones. Between 1996 and 2000, only 3% was flumequine-resistant,
19% in 2001 and 39% in 2002 (tables 2 and 3). Table 3 shows the shift
occurring in the past four years towards higher MIC values for flumequine
and ciprofloxacin. No evidence yet exists for clinical resistance to
ciprofloxacin, but reduced sensitivity has been demonstrated.
According to German searchers, the development of resistance could be
related to the high selection pressure due to vaccination and the intensive
use of antibiotics in poultry farming, for example, to control the previous
S. Enteritidis crisis (9). In 2000-01, 13% of flocks were treated with
a quinolone, mainly flumequine. This resistance has developed much faster
in Java than in other serotypes (table 2). The easy spreading of this
resistant clones in chickens, and the persistence in the environment
once a farm is infected are the likely reasons for the accelerated development
of resistance. Besides, the high percentages of resistance (table 2)
could be explained by therapeutic treatments of cotrimoxazole. The use
of nitrofurans (furazolidone in table 2) has, however, been forbidden
for more than 10 years. It seems that clonality of the strains is the
determining factor for the (chromosomally determined) high level of
resistance to furazolidone.
Transmission to humans
It is unclear why Java (up to now) seems innocuous to humans. However,
recent PFGE-typing results show that the multiresistant poultry clone
definitely crosses over to humans, which is consistent with the changed
resistance pattern of Java among humans (table 2). Eleven out of 22
isolates preserved from the period 1998-2002 had a PFGE pattern identical
to that of the Java poultry clone, and two unrelated isolates were associated
to foreign travel.
Table 2
Resistance patterns of S. Java isolates from humans and chickens, in
1984-1995 and 1996-2001.
|
|
Poulet / Chicken
|
Homme / Human
|
|
|
Tous les sérotypes à l’exclusion
de Java All serotypes excluding Java
|
S. Java
|
S. Java
|
|
|
1984–1995
|
1996–2001
|
1984–1995
|
1996–2001
|
1984–1995
|
1996–2001
|
|
|
N
|
N
|
N
|
N
|
N
|
N
|
|
Isolé / Isolated
|
38506
|
3718
|
52
|
1351
|
51
|
44
|
|
Testé pour la résistance / Resistance
tested
|
30418
|
991
|
37
|
517
|
43
|
13
|
|
|
%
|
%
|
%
|
%
|
%
|
%
|
| Tetracycline |
7
|
8
|
49
|
7
|
0
|
8
|
| Chloramphenicol |
1
|
2
|
27
|
1
|
0
|
8
|
| Neomycin |
1
|
0
|
30
|
0
|
0
|
0
|
| Ampicillin |
6
|
9
|
8
|
43
|
0
|
44
|
| Cotrimoxazole* |
2
|
2(16)
|
49
|
64(91)
|
0
|
30(54)
|
| Furazolidone |
9
|
10
|
27
|
99
|
0
|
44
|
| Flumequine |
0
|
7
|
0
|
17
|
0
|
18
|
Table 3
Resistance development of Salmonella Paratyphi B var. Java isolated
from poultry matériels
| Valeur de la CMI
/MIC value (µg/mL) |
0.015 |
0.03 |
0.06 |
0.125 |
0.25 |
0.5 |
1 |
2 |
4 |
8 |
16 |
32 |
64 |
Res % |
| Ciprofloxacin |
99–2000 |
15 |
33 |
11 |
1 |
1 |
|
|
|
|
|
|
|
|
0 |
| |
2001 |
|
47 |
10 |
3 |
6 |
7 |
1 |
|
|
|
|
|
|
0 |
| |
Nov 2002 |
|
39 |
23 |
4 |
14 |
21 |
6 |
|
|
|
|
|
|
0 |
| Flumequine |
1999-'00 |
|
|
|
|
|
36 |
11 |
13 |
1 |
|
|
|
|
3 |
| |
2001 |
|
|
|
|
|
40 |
10 |
9 |
1 |
8 |
3 |
3 |
|
19 |
| |
Nov 2002 |
|
|
|
|
|
38 |
21 |
5 |
2 |
11 |
15 |
15 |
1 |
39 |
Conclusions and recommendations
Until the end of 2002, the spreading of a resistant clone was considered
as a local problem. However, in December 2002, a request for information
from Scotland about an emerging Java problem in humans, revealed the
clone involved could be traced back to isolates from poultry imported
from Germany and Holland (see article pp35-40). This may cause a serious
public health threat, especially given that fluoroquinolones are the
first antibiotics of choice to treat a serious salmonellosis, and puts
the problem into an international perspective.
In the Netherlands, public health and veterinary services, research
institutes (RIVM, ID Lelystad, the Animal Health Service and KvW), and
the poultry industry have initiated coordinated initiatives to address
this emerging issue.
|