Agriculture Reference
In-Depth Information
for 117 certifi ed organic, 359 noncertifi ed organic, and 129 conventional produce
items, respectively, and the average
E. coli
counts for the positive samples was
reported as 3.1 log MPN/g. The
E. coli
incidence was sixfold higher on organic versus
conventional produce and 2.4-fold higher on produce from farms using cattle manure
compared to farms using other types of manure. Noncertifi ed organic lettuce had the
highest incidence (12/39, 30.8%) for any item with more than 10 samples tested
(Mukherjee and others 2004).
The microbiological quality of ready-to-eat produce has been surveyed in other
parts of the world. In a study of leafy salads collected from retail markets in Brazil,
>
4 logs
Enterobacteriaceae
per g (Froder
and others 2007). Leafy vegetable salads collected postpreparation from 16 university
restaurants in Spain yielded 26% positive for
E. coli
(Soriano and others 2001). In
contrast, only one (lettuce) of 50 produce items collected from retail and farmers
markets in Washington, D.C. was positive for
E. coli
(Thunberg and others 2002).
These results suggest major diversity in
E. coli
incidence depending upon the size,
time, and location of the study, and possibly differences in the sensitivity of methods.
A study initiated by the USDA Agricultural Marketing Service in 2002 and coor-
dinated with state and other federal agencies to survey the microbial quality of fresh
produce items available at terminal markets and wholesale distribution centers con-
tinues as of 2008 (USDA-AMS-MDP 2008). The cumulative results over 6 years, with
approximately 65,000 samples analyzed to date, provides a signifi cant data set for
analyzing spatial, temporal, and other factors related to produce contamination using
E. coli
incidence as the measure of fecal contamination. Multiple commodities, both
domestic and imported, have been tested during the program (e.g., cantaloupe, leaf
and romaine lettuce, tomatoes, green onions, and alfalfa sprouts) for generic
E. coli
,
E. coli “
with pathogenic potential” (including
E. coli
O157:H7), and
Salmonella
. The
results from tests of
85% of 181 samples were reported to have
>
59,000 samples from 2002-2007 indicate that low levels of
generic
E. coli
are common on produce items collected at the distribution stage of the
postharvest production cycle compared to levels on produce in the fi eld (Table 1.2);
however, only 1.5 to 2.7% of the samples by year were positive for
E. coli
at concen-
trations
>
10 MPN/ml (USDA - AMS - MDP 2008 ). Moreover,
E. coli
with pathogenic
potential based on PCR results for various virulence factors, including shigatoxin 1
and 2 (Stx 1 and 2), ranged from 0.1 to 0.4% of all samples tested each year.
Examples of individual produce items having a high percentage of samples positive
>
Table 1.2.
Incidence of
E. coli
on selected fresh produce items obtained and tested in years
2002-2007, as part of the USDA, Agricultural Marketing Service, Microbial Data Program
(USDA-AMS-MDP 2008)
Categories
2002
2003
2004
a
2005
2006
2007
a
Total no. produce samples tested
10,319
10,972
11,211
11,508
7,646
5,279
No. positive for
E. coli
b
759
730
3,226
4,201
1,569
4,420
%
positive for
E. coli
7.4
6.7
28.8
a
36.5
20.5
83.8
%
E. coli
samples
with virulence trait(s)
0.6
0.4
0.4
0.4
0.4
0.1
a
Generic
E. coli
method was modifi ed in 2004 and again in 2007.
b
A sample was considered positive if
>
0.03 MPN/ml rinse solution was determined.
