Gastroenteritis in Marines during Operation Iraqi Freedom • CID 2005:40 (15 February) • 519
M A J O R A R T I C L E
Gastroenteritis in US Marines
during Operation Iraqi Freedom
Scott A. Thornton,1Sterling S. Sherman,2Tibor Farkas,3Weiming Zhong,3Pete Torres,2and Xi Jiang3
1Navy Environmental and Preventive Medicine Unit No. 6, Pearl Harbor, Hawaii;
San Diego, California; and
2Navy Environmental and Preventive Medicine Unit No. 5,
3Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine, Cincinnati, Ohio
(See the editorial commentary by Matson et al. on pages 526–7 and the article by Crum et al. on pages 511–8)
Approximately 83,000 US Marines participated in the opening phase of OperationIraqiFreedom
in Spring 2003. A Navy Preventive Medicine laboratory was set up in Ad Diwaniyah, Iraq, to provide clinical
diagnostic support for Marine medical units during a period of repositioning in south-central Iraq.
Specimen collection boxes were sent to 130 primary care medical stations handling 500–900 per-
sonnel each. The laboratory had capability to detect many different disease agents, especially those causing febrile
illness. Diarrheal stool diagnostic evaluation included plating and biochemical identification, antigen serologic
testing, fluorescent antibody antigen detection, disk diffusion antimicrobial susceptibility testing, enzyme immu-
noassay, and reverse-transcriptase polymerase chain reaction for norovirus (NV). Confirmation and sequencing
work for NV was done at Cincinnati Children’s Hospital Medical Center (Ohio).
By far the most common reason for infectious disease sick call visits was gastrointestinal illness; no
other symptoms had equivalent impact. An enteropathogen was detected in 57 (44%) of 129 stool samples, with
NV detected in 30 stool samples (23%) obtained from 14 different battalion or similar-sized units;nextinfrequency
were Shigella flexneri and Shigella sonnei, which were isolated from 26 stool samples (20%) obtained from 15 units.
Sequencing the NV RNA polymerase gene demonstrated that NV strains represented 7 genetic clusters, including
2 strains from genogroup I and 5 from genogroup II. Ciprofloxacin was effective in vitro against most bacterial
agents, but neither doxycyline (which was taken daily as the antimalarial prophylaxis dose) nor trimethoprim-
sulfamethoxazole were effective.
Multiple strains of Shigella species and NV predominated, probably because they do not require
a large inoculum to cause infection. Otherwise, personnel remained free of infectious illness during this phase of
the conflict, because other infectious agents were rare or absent.
Infectious diseases have long been problematic to de-
ployed US military ground forces. During recent op-
erations in theMiddleEast andSomalia,foodandwater
sources were tightly controlled, and medical interven-
tion measures, such as administration of vaccines and
antimicrobial prophylaxis regimens, have been success-
Received 24 June 2004; accepted 14 October 2004; electronically published 20
The views expressed in this article are those of the authors and do not
necessarily reflect the official policy or position of the Department of the Navy,
Department of Defense, or the US Government.
Presented in part: 43rd Navy Occupational Health and Preventive Medicine
Workshop, Chesapeake, Virginia, 18–24 March 2004; and 23rd Annual Meeting
of the American Society for Virology, Montreal, Quebec, Canada, 10–14 July 2004
Reprints or correspondence: LCDR Scott A. Thornton, Head, Microbiology Dept.,
Navy Environmental and Preventive Medicine Unit No. 6, Pearl Harbor, HI 96860
Clinical Infectious Diseases2005;40:519–25
This article is in the public domain, and no copyright is claimed.
fully implemented to reduce disease incidence. When
such measures are not followed, even the preventable
diseases again become an issue [1–3]. Iraq has many
of the diarrheal disease agents typically found in the
Middle East, including Vibrio cholerae in some areas.
Norovirus (NV), a common cause of viral gastroen-
teritis outbreaks in military units, was undocumented
in Iraq but was presumably present, because it was a
cause of outbreaks of infection in the military in nearby
Saudi Arabia during the 1991 Gulf War [4, 5]. Other
disease threats endemic in Iraq include malaria(mostly
due to Plasmodium vivax), arboviruses, brucellosis,
leishmaniasis, leptospirosis, infection with rickettsial
agents, schistosomiasis, Q fever, hepatitis A and B, and
typhoid fever [6, 7].
On 20 March 2003, Operation Iraqi Freedom com-
menced, with US and allied military entering Iraq. The
ground component included ∼83,000 Marines of the
First Marine Expeditionary Force. The Marines quickly
by guest on November 5, 2015
520 • CID 2005:40 (15 February) • Thornton et al.
moved north through the Tigris-Euphrates floodplain to Bagh-
dad and Tikrit, but during the time period covered in this arti-
cle (24 April through 1 June 2003), they were redeploying
south of the capital. Marines generally chose to sleep in open
buildings, without electricity or windows and/or screens. The
weather was rapidly warming up to 40?C, with occasional dust
storms. Sanitation conditions were poor, because camps were
not well established and were often overcrowded. Unscreened
burn-barrel shelters were the most common latrine models,
and male urination was handled by use of pipes inverted into
sand pits. Hand washing was encouraged with placement of
water bags (Lister bags) and waterless alcoholic gel dispensers
near food facilities and latrines. Food came in prepackaged
individual meals, later replaced by precooked bulk trays, which
were all brought from the United States. Filth flies were abun-
dant both in the latrines and dining facilities. Drinking water
was from reverse-osmosis– and chlorine-treated river water or
regionally imported bottled water only. Personnel were not
allowed to acquire food in town, although this rule was broken
more than once, usually for grilled chickens.
Three Navy Preventive Medicine–Mobile Medical Augmen-
tation Readiness Teams (PM-MMARTs) were deployed to sup-
port Operation Iraqi Freedom. Part of their mission was to
determine the risks and causes of infectious diseases, using
surveillance and a clinical diagnostic laboratory to directly de-
tect which agents were causing illness. PM-MMART 5 was set
up in the central Iraqi city of Ad Diwaniyah, which is ∼175
km south of Baghdad. This location was ideal to support the
∼30 First Marine Expeditionary Force battalions, each with
1900 personnel, which were located in cities of central Iraq.
Although the laboratory had the capabilityandmissiontoiden-
tify many infectious disease pathogens, including traditional
agents of biological warfare, this article focuses on cases of
gastroenteritis, because those represented by far the major dis-
ease impact on the troops.
MATERIALS AND METHODS
on 21 April2003andsetupatanabandoneduniversitycampus.
Within a few days, it began dispersing boxes containing col-
lection and transport supplies for stool, blood, and other clin-
ical specimens to 30 First Marine ExpeditionaryForceBattalion
Aid Stations or other medical stations in the area. (In this
article, “units” are defined as battalions or similar groups.)This
included all units located from north of Nasiriyah to the south-
ern outskirts of Baghdad. Specimens were collected by First
Marine Expeditionary Force Battalion Aid Station personnel
and stored at room temperature until transportation was ar-
ranged to get them to PM-MMART 5, usually ?24 h later.
Acute-phase blood specimens were to be collected from all
febrile patients, with fever defined as a temperature of ?38.3?C
(?101?F) measured sublingually; convalescent-phase blood
specimens were to be collected from febrile patients for whom
no enteric pathogens were identified. Stool samples were col-
lected from any patient with diarrhea, regardless of whether
fever was present. If the First Marine Expeditionary Force Bat-
talion Aid Station location was 11 h distant from Diwaniyah,
then stool samples were divided into 3 transport vials (Ecofix,
Enteric Plus, and a clean vial [Meridian Diagnostics]), whereas
local First Marine Expeditionary Force Battalion Aid Station
specimens were brought directly to the laboratory in the orig-
inal collection cup. No other medical data were collected from
the patients, although PM-MMART 5 did collect Disease and
Non-Battle Injury reports from Diwaniyah-area units. All First
Marine Expeditionary Force personnel were ordered to take
doxycyline (100 mg q.d.) as prophylaxis against malaria.
Stool specimen diagnostic evaluation.
were cultured on MacConkey and XLD (xylose lactose de-
on thiosulfate-citrate–bile salts–sucrose agar, and samples of
stool with gross blood or fecal leukocytes (as determined by
direct methylene blue stain microscopy) were plated on
Sorbitol-MacConkey agar to screen for enterohemorrhagic
Escherichia coli and on sheep blood agar to screen for Cam-
pylobacter species with use of a membrane-filter swim-through
procedure . Lactose nonfermenters were screened with tri-
ple-sugar iron agar and motility-indole-lysine biochemicalagar
tubes and were speciated using API 20E (bioMe ´rieux). Sal-
monella and Shigella isolates were confirmed with Wellcolex
latex serology kits (Oxoid). For 4 patients, suspected E. coli
colonies were tested for heat-labile enterotoxin using the Ox-
oid VET-RPLA kit and for heat-stable enterotoxin using Oxoid
ST-EIA. Giardia and Cryptosporidium species were identified
with a Merifluor fluorescent antibody kit (Meridian) anddirect
merthiolate-iodine-formalin stain (Meridian). NV was identi-
fied by use of an RT-PCR assay (Robust RT-PCR kit; MJ Re-
search) and 289/290 primers, as well as by use of an antigen-
capture EIA [9, 10]. Both assays include specific reagents
developed by the reference laboratory at Cincinnati Children’s
Hospital Medical Center (Ohio). Bacterial isolates were tested
against commonly deployed antimicrobial agents with use of
the disk diffusion assay on Mueller-Hinton Agar.
NV strain characterization.
San Diego, stool samples were shipped on dry ice to one of
the authors (X.J.) at Cincinnati Children’s Hospital Medical
Center for confirmation and further strain characterization. A
modified degenerate primer set in the same region of the 2
original primers that contained 2 reverse primers (289H and
289I) and 4 forward primers (290H, 290I, 290J, and 290K) was
used, which also produced a 319-bp product . Sequences
of individual NV strains were determined by direct sequencing
of the RT-PCR products followingpurificationoftheamplicons
All stool samples
After they were returned to
by guest on November 5, 2015
Gastroenteritis in Marines during Operation Iraqi Freedom • CID 2005:40 (15 February) • 521
Primers amplify RNA polymerase gene fragment of 319 bp. Arrow indi-
cates 300-bp band. Lane 1, 100-bp DNA marker; lane 2, Cincinnati Chil-
dren’s Hospital Medical Center (CCHMC) number 031926; lane 3, CCHMC
031931; lane 4, CCHMC 031932; lane 5, CCHMC 031944; lane 6, CCHMC
031952; lane 7, CCHMC 031957; lane 8, CCHMC 031975; lane 9, CCHMC
031981; lane 10, CCHMC 031983; lane 11, CCHMC 031998; and lane 12,
Agarose gel of typical RT-PCR run in Ad Diwaniyah, Iraq.
from agarose gels after electrophoresis analysis. Sequences of
some strains also were determined after cloning of the RT-PCR
products if the DNA bands were weak in the gels. In brief, the
in accordance with the manufacturer’s protocol. Positiveclones
were identified by PCR screening. The cloned cDNA was se-
quenced using M13 forward and reverse primers by the chain
(Applied Biosystems). All clones were sequenced at least twice.
The following sequences, published in GenBank, were used
in phylogenic analysis based on that of Vinje et al. : geno-
group I, Chiba (AB042808) GI/4, Desert Storm GI/3, and Nor-
dale (X86557) GII/4, Melksham (X81879) GII/2, Mex7076
(AY579431), Saitama U25 (AB039780) GII/8, and VA207
(AY038599) GII/9. Saitama U25 was used for GII/8 because
RNA polymerase data for Amsterdam are not available from
public databases. Mex7076 has been characterizedatCincinnati
Children’s Hospital Medical Center, and according to the phy-
logenetic analysis of the whole capsid sequences, it represents
a new genetic cluster within genogroup II NVs (unpublished
using Omiga software, version 2.0 (Oxford Molecular).Aligned
sequences were edited in GeneDoc, version 2.5, obtaining a
210-bp consensus length . Distance trees were constructed
by the UPGMA clustering method of Molecular Evolutionary
Genetics Analysis, version 2.1, with Jukes-Cantor distance cal-
culations . The confidence values of the internalnodeswere
obtained by performing 125 bootstrap analyses.
Collection of specimens and clinical data.
laboratory received its first specimen on 24 April 2003, and a
total of 129 stool samples were collected, representing 33 units
scattered across south-central Iraq. Clinical information was
sparse, because no units made log entries during the major
combat phase; however, discussions with medical care givers
suggested that a viral etiology for gastroenteritis predominated
during the early part of the conflict, with large outbreaks of
nausea, vomiting, and diarrheathatlasted24–48h.Thesebegan
shortly after the fall of Baghdad on 9 April and continued
through the first weeks of May, when cases of febrile dysentery
prevailed. Most units were in little or no contact with each
other during this period. Some were front-line fighting units,
whereas others were support units that arrived in Iraq at ap-
proximately the same time as PM-MMART 5.
Disease etiology and location.
tected in 57 (44%) of 129 stool samples. NV was detected by
RT-PCR in 30 stool samples (23%) obtained from patients in
14 units (figure 1), and either Shigella sonnei or Shigellaflexneri
was isolated from 26 stool samples (20%) from 15 units; in
An enteropathogen was de-
only 1 unit were both Shigella species found (table 1). All 5
stool samples that yielded Campylobacter isolates were from
different units. A total of 43 of 109 stool samples had fecal
leukocytes present, indicating inflammatory diarrhea. Among
the stool specimens with an agent identified, 24 were positive
for NV only, 22 were positive for Shigella only, and 4 were
positive for another single agent (including a case of Crypto-
sporidium infection). The other cases included 6 mixed infec-
tions with NV and either Shigella or Campylobacter species and
1 case in which a single Salmonella arizonae colony mixed with
abundant Campylobacter species. Although NV-positive stool
samples were collected up through the last week of May, the
number of such samples had decreased from theearlierperiods.
Antimicrobial susceptibility testing.
was found in all but 1 Shigella isolate, although the number of
doxycyline-susceptible Campylobacter strains may indicate that
compliance with antimalarial therapy prevented many more
cases of Campylobacter infection. Resistance to trimethoprim-
sulfamethoxazole was similarly common. On the other hand,
by guest on November 5, 2015
522 • CID 2005:40 (15 February) • Thornton et al.
isolates resistant to antimicrobial agents, as determined by disk diffusion assay.
Major findings from 129 stool specimens obtained from soldiers in Operation Iraqi Freedom and percentage of bacterial
Percentage of drug-resistant isolates, by drug
Doxycycline CiprofloxacinTMP-SMZ Cefoxitin CefazolinCeftriaxone
and ceftriaxone, 30 mg. NA, not applicable.
aFifty-six stool samples were tested for Campylobacter species.
bPutative enteroinvasive Escherichia coli, 2 cases; Salmonella arizonae, 1 case; and Plesiomonas shigelloides, 1 case.
Doxycycline, 30 mg in disk; ciprofloxacin, 5 mg; trimethoprim-sulfamethoxazole (TMP-SMZ), 1.25 mg trimethoprim; cefoxitin, 30 mg; cefazolin, 30 mg;
all Shigella species were susceptible to ciprofloxacin, whereas
all Campylobacter species were resistant to this agent. All 3
cephalosporins—especially ceftriaxone—inhibited the bacterial
enteric pathogens isolated, except for Campylobacter species.
NV strain comparison.
NV strains were compared by se-
quencing the RNA polymerase genes of 21 strains, which could
be grouped into 2 genogroup I (GI) and 5 genogroup II (GII)
genetic clusters on the basis of sequences of known strains
(figure 2). This is not a definitive classification of strains and
clusters, because it is based on a fairly short sequence. Three
strains (031952, 031967, and 031973) had nucleotide sequence
identities of 89%–100% with each other and 89%–97% with
Saitama U25. Three strains (031931, 032040, and 032041) that
had a 99% identity with each other had a 90% identity with
Mex7076, which is a new cluster representative within GII.Two
strains (031933 and 031966) that had a 98% identity with each
other hada90% identitywithVA207;3strains(031948,032000,
and 032046) that had a 96%–98% identity with each other had
91%–93% identity with Melksham. Three strains (031945,
031962, and 031998) that had 96%–100% identity with each
other had an equal 83% identity with the Hawaii andLordsdale
viruses and 91%–94% identity with a Spanish GGIIb strain
(AJ487789). One strain (032036) had 80% identity with the
Desert Shield virus (GI/3), and 6 strains (031944, 031951,
031955, 031957, 031980, and 032047) that had 95%–100%
identity with each other had 87%–93% homology to the Chiba
The occurrence of NV strains, by unit and timing.
tary units were arbitrarily designated 1–14; units 12–14 did not
have a strain sequenced from their single specimen. One strain
(032040) did not have a unit identifier with the specimen.
Cluster GI/4 contained 5 strains (4 of which were from the
same battalion) collected in late April 2003 and 1 strain (from
a different unit) collected in late May. Units 2 and 3 had 2
different strains detected at the same time, but most units had
only 1 strain detected during the same 2- or 3-week period.
On the other hand, all but 1 strain were found in multi-
ple units, with GI/3 being detected from only 1 specimen.
According to pairwise distance analysis, the intracluster dis-
tances for all of the strains on the distance tree ranged between
0 and 0.088. The intercluster distance was referred by 2 well-
characterized genetic cluster representative strains, Hawaii and
Lordsdale, and was 0.112. Strains 031967 and 032036 separated
from any of the strains with greater distances than thatbetween
Hawaii and Lordsdale (10.118 and 10.195, respectively).Future
study is necessary on the capsid sequences to decide whether
these 2 strains represent new genetic clusters.
The results for First Marine Expeditionary Force reinforce ex-
perience elsewhere; despite strict control of food and water
sources, certain enteropathogens that have very low inoculum
requirements (e.g., Shigella species and NV) causedasignificant
number of cases in a highly unsanitary environment. Filth flies
may account for part of their spread, along with person-to-
person contact. Shigella species was the most common enter-
opathogen in First Marine Expeditionary Force personnel in
Somalia in 1992–1993, where lack of access to local sources of
food and water kept exposure to other causes of traveler’s di-
arrhea low, and NV testing was not applied there. For those
agents that require a larger inoculum, heat-labile and heat-
stable enterotoxins were not found at all, whereas Salmonella
and Campylobacter were rare and probably linked to roast
chicken meals acquired in town. As expected, most of the iso-
lates encountered were resistant to doxycycline, so the number
of enteric cases prevented by daily doxycycline antimalarial
prophylaxis is not known.Infectionwithotherdrug-susceptible
pathogens, such as Leptospira species, may also have been pre-
vented by doxycycline therapy.Ciprofloxacin,anothercommon
by guest on November 5, 2015
Gastroenteritis in Marines during Operation Iraqi Freedom • CID 2005:40 (15 February) • 523
2003. Strains are listed by identification numbers, followed by arbitrary numbers for military units of the strains isolated, with the question mark
indicating that the unit was not recorded. The reference strains included in the comparison are in bold.
Distance tree of norovirus polymerase gene sequences detected in the US First Marine Expeditionary Force in Iraq, April through May
antimicrobial used empirically by First Marine Expeditionary
Force Battalion Aid Station personnel, was highly effective in
vitro against all but the Campylobacter isolates. Enteropatho-
gens were identified in dozens of separate First Marine Expe-
a few strains of S. sonnei caused large outbreaks of infection
in military dining facilities, in Iraq, there was no area-wide
occurrence of infection with any 1 strain,althoughtherelatively
small units described in this article tended to have the same
species of Shigella strain isolated. Characterization of the Shi-
gella isolates is currently underway elsewhere.
NV was shown to be a cause of morbidity in US personnel
in Operation Iraqi Freedom, just as it is in US sailors and
marines worldwide. Since 1999, we have documented 24 out-
breaks of viral gastroenteritis on the largest US Navy ships and
a dozen more on numerous smaller vessels and in shore units,
and we have confirmed that NV was responsible in almost ev-
ery incident in which specimens were submitted for analysis
(S.A.T., unpublished data) . Even with weak or anecdotal
data, a conservative estimate would be that thousands of cases
of NV illness occurred in the First Marine Expeditionary Force
in April through May 2004. Although the symptoms are short-
lived, the effects of an outbreak can affect a unit undergoing
the physical demands of near-continuous combat. Confirma-
tion of NV in these outbreaks can ease psychological worries
about chemical and biological agent use, which was expected
on a daily basis during this time. An outbreak in a coalition
military hospital in Bagram, Afghanistan, in 2002 caused con-
by guest on November 5, 2015
524 • CID 2005:40 (15 February) • Thornton et al.
cern about such an attack, because laboratory confirmation of
NV was not possible in theater . The present report prob-
ably represents the first confirmation of NV gastroenteritis in
a wartime theater using a field laboratory. Even when an out-
break of nausea, vomiting, and/or diarrhea is presumed to be
due to NV, there are problems in containing spreadofthevirus,
especially in the environment faced by the troops in Iraq.
It is not surprising that several distinct NV strains were de-
tected, because the units had little contact witheachotherduring
their movements. Units were often affected by multiple strains
during this period, and multiple strains coexisted in a few units.
Three amphibious ships reported outbreaks of viral gastroenter-
itis shortly after re-embarking Marines to return to the United
States in late May 2003 (S.A.T., unpublished data). Although
none of the subjects had specimens collected, and although the
units were not the same as those reported above, it wouldappear
that the Marines brought the strains aboard the ships. Viral
gastroenteritis in Iraq has not gone away: an outbreak affected
Dutch troops patrolling the southern border of Iraq in July
2003 , as well as US Marines in central Iraq in June 2004
(S.S.S., personal communication). Although this is probably
the first recorded confirmation of NV in Iraq, that virus would
seem to be ubiquitous in the country, with several strains that
included both major genogroups. No specimens were collected
from Iraqis to characterize strains from the civilian population.
In the United States, GII strains predominate, and one such
strain, Farmington Hills, was detected in 41% of 27 outbreaks
nationwide and on cruise ships . One of the geneticclusters
detected in this article, GGIIb, has been reported as the emerg-
ing genotype in Europe . In our experience, US Navy ships
that experience outbreaks of NV gastroenteritis after visiting
Asian ports are more likely to have GI strains.
Although the PM-MMARTs were deployed to OperationIraqi
Freedom primarily because of the perceived threat of chemical
and biological warfare agents, infectious gastroenteritis proved
to be a larger health concern. PM-MMART 5 demonstrated the
value of having a clinical infectious disease/public health sur-
veillance laboratory in close proximity to the echelon of care
normally given to these cases, so results of etiology and drug
susceptibility studies could be useful in implementing counter-
measures in a timely manner. The laboratory was equipped to
identify etiologic agents of many diseases besides gastroenteritis;
however, there were very few patients who presented to the aid
stations whose cases did not involve enteric symptoms as the
major complaint. The lack of solid denominator data and attack
rates during the major maneuver campaign were the result of
the inability to be in constant touchwiththevariousunits,which
also had priorities other than patient data entry. Once the units
they indicated overall illness attack rates were higher than back-
ground levels previously recorded in their US home garrisons.
Aid stations are not supplied for collection of clinical specimens,
so the laboratory must be further equipped to send supplyboxes
to them. The mission of the PM-MMART is to act as a public
but not to provide a clinical service to every patient in the theater.
A better method is needed to collect accurate data on disease
incidence in military personnel during maneuver warfare.
We wish to thank the other members of PM-MMART5fortheirsupport,
as well as the sailors, marines, and soldiers serving under the First Marine
Expeditionary Force for their cooperation with this work.
Department of Defense GlobalEmergingInfections
Systems (to S.A.T.) and US Army Medical Research Acquisition Act (grant
PR033018 to X.J.).
Potential conflicts of interest.
All authors: no conflicts.
1. Hyams KC, Bourgeois AL, Merrell BR, et al. Diarrheal disease during
Operation Desert Shield. N Eng J Med 1991;325:1423–8.
2. Wallace MR, Sharp TW, Smoak B, et al. Malaria among US troops in
Somalia. Am J Med 1996;100:49–55.
3. ProMed Mail. Malaria—Liberia (USA military personnel). Accession
no. 20031018.2623. Available at: http://www.promedmail.org.Accessed
19 October 2003.
4. Hyams KC, Malone JD, Kapikian AZ, et al. Norwalk virus infection
among Desert Storm troops. J Infec Dis 1993;167:986–7.
5. McCarthy M, Estes MK, Hyams KC. Norwalk-like virus infection in
military forces: epidemic potential, sporadic disease, and the future
direction of prevention and control efforts.JInfectDis2000;181(Suppl
6. World Health Organization (WHO). WHO communicablediseasepro-
file for Iraq. Geneva: WHO, 2003.
7. Armed Forces Medical Intelligence Center. Infectious disease risk as-
sessment: Iraq. 2003.
8. Steele, TW, McDermott SN. The use of membrane filters applied di-
rectly to the surface of agar plates for the isolation of Campylobacter
jejuni from feces. Pathology 1984;16:263–5.
9. Jiang X, Huang PW, Zhong WM, Farkas T, Cubitt DW, Matson DO.
Design and evaluation of a primer pair that detects both Norwalk-
and Sapporo-like caliciviruses by RT-PCR. J Virol Methods 1999;83:
10. Jiang X, Wilton N, Zhong W, et al. Diagnosis of human caliciviruses
by use of enzyme immunoassays. J Infec Dis 2000;181(Suppl 2):
11. Farkas T, ZhongWM, JingY, et al.Geneticdiversityamongsapoviruses.
Arch Virol 2004;149:1309–23.
12. Vinje J, Hamidjaja RA, Sobsey MD. Development and application of
a capsid VP1 (region D) based reverse transcription PCR assay for
genotyping of genogroup I and II noroviruses. J Virol Methods 2004;
13. Nicholas KB, Nicholas HB, Deerfield DWI. GeneDoc: analysis and
visualization of genetic variation. EMBNEW NEWS 1997;4:14.
14. Kumar S, Tamura K, Jakobsen IB, Nei M. MEGA2: molecular evolu-
tionary genetics analysis software. Tempe, AZ: ArizonaStateUniversity,
15. Thornton S, Davies D, Chapman F, et al. Detection of Norwalk-like
virus infection aboard two US Navy ships. Mil Med 2002;167:826–30.
16. Centers for Disease Control and Prevention. Outbreak of acute gas-
by guest on November 5, 2015
Gastroenteritis in Marines during Operation Iraqi Freedom • CID 2005:40 (15 February) • 525
troenteritis associated with Norwalk-likevirusesamongBritishmilitary
personnel—Afghanistan, May 2002. MMWR Morb Mortal Wkly Rep
17. Expatica News. Virus spreads through Dutch troops. 29 July 2003.
Available at: http://www.expatica.com. Accessed 15 August 2003.
18. Centers for Disease Control and Prevention. Norovirus 2002—USA.
MMWR Morb Mortal Wkly Rep 2003;52:41–5.
19. Buesa J, Collado B, Lopez-Andujar P, et al. Molecular epidemiology
of caliciviruses causing outbreaks and sporadic cases of acute gastro-
enteritis in Spain. J Clin Microbiol 2002;40:2854–9.
by guest on November 5, 2015