0002-921 0/91 /921 l - I 962$03.00^)THli A\ l tRlc, \ \ Jot Hnlt- op GAsTROF:NrrrRolocY

Copyright O 1997 by Am. Col l . of Castrocnterology

Practice suidelines

Vol 92. No. I l . 1997Printed in U.S.A.

AJ

Urmidobogemi

Guidelines on Acute Infectious Diarrhea in Adults

Herbert L. DuPont, M.D., and The Practice Parameters Committee of the American College of Gastroenterology

Tlrc Amerit'un College rlt' Gastroenterology, ACG Practice Porarneters Guidelines Commiltee, Arlington, Virginia

spantirSE

arel

Guidelines for clinical practice are intended to suggestpreferable approaches to particular medical problems asestablished by the interpretation and collation of scien-tifically valid research, derived from an extensive review

of published literature. When data are not available that

will withstand objective scrutiny, a recommendationmay be made based on a consensus of experts' Guide-lines are intended to apply to the clinical situation for allphysicians without regard to specialty. Guidelines areintended to be flexible, not necessarily indicating theonly acceptable approach, and should be distinguishedfrom standards of care that are inflexible and rarelyviolated. Given the wide range of choices in any health

care problem, the physician should select the course bestsuited to the individual patient and the clinical situationpresented. These guidelines are developed under theauspices of the American College of Gastroenterologyand its Practice Parameters Committee. These guide-

lines are also approved by the governing boards of theAmerican Gastroenterology Association and the Amer-ican Society of Gastrointestinal Endoscopy. Expert opin-ion is solicited from the outset for the document. Guide-lines are reviewed in depth by the Committee, withparticipation from experienced clinicians and others in

related fields. The final recommendations are based onthe data available at the time of the production of the

document and may be updated with pertinent scientificdevelopments at a later time. The following guidelines

are intended for adults and not for pediatric patients.

INTRODUCTION

In developing the guidelines for evaluation and manage-ment of the patient with acute diarrhea, the fbllowing ma-terial is divided into eight subheadings: importance of di-arrhea, patient evaluation, noninf'ectious and extraintestinalcauses. empiric therapy in selected patients, laboratory eval-

uation, management, the international traveler, and the im-

munocompromised patient.

IMPORTANCE OF DIARRHEA

In the United States acute diarrhea is one of the most

common diagnoses in general practice. Acute diarrhea can

be defined as the passage of a greater number of stools ofdecreased form from the normal lasting less than 14 days. It

is generally associated with other signs or symptoms sug-gesting enteric involvement including nausea, vomiting, ab-

dominal pain and cramps, increase in intestinal gas-related

complaints, fever, passage of bloody stools (dysentery),

tenesmus (constant sensation of urge to move bowels), and

fecal urgency. When diarrhea lasts as long as l4 days, it canbe considered persistent. Many restrict the term chronicdianhea to indicate illness lasting at least I month. In thisreview, diagnosis and treatment of acute and persistent

diarrhea will be considered because both tend to be short-

term illnesses, infectious agents characteristically produce

both, and the tests procedures and treatments of both fbrms

of d iarrhea show s imi lar i t ies.The annual rate of diarrheal illness in the United States

and western Europe among adults averages about one epi-

sode per person per year (1,2). In one study, it was esti-

mated that for a single year, 99 million cases of gastroen-

teritis or acute diarrhea occurred among adults in the United

States (1). In this study, half of the persons with gastroen-

teritis or diarrhea had restriction of their activities for more

than a full day, a physician was consulted in 8.2 mill ion of

the cases, 250,000 persons were hospitalized, 1.9 millionpersons saw a physician yet were not hospitalized, and more

than 90 million experienced illness without seeking medical

attention. In a second study, it was fbund that gastroenteritis

and acute diarrhea accounted tor 1.57o ofhospitalizations of

adults >20 yr of age in the United States (3). Sixty-twopercent of the admissions for diarrhea were in adults greater

than 20 yr of age. In this study and others most of the deathsassociated with diarrheal illness in the United States oc-

curred in the elderly (3-5).

ttt

al

tindsl

3dnY

(t

a

C

stiiIIlI

l

Receivetl Mar. 11, 1997: uccepted Juh' 16, 1997

t962

AJG - November 1997 GUIDELINES ON

Although statistically the rate of diarrhea in adults in theUnited States and other industrialized regions is approxi-mately one episode per person per year, diarrhea actuallydoes not occur in all persons annually. Food- and water-borne outbreaks involving a relatively small subset of thegeneral population and recunent bouts of illness in othersmake up the bulk of the cases of the illness. Diarrhea is aspecial problem among adults who are exposed to childrenand nontoilet-trained infants particularly in a day care set-ting, travelers to tropical and semitropical regions, homo-sexual males, persons with underlying immunosuppression.and those living in an unhygienic environment and havingexposure to contaminated water or fbods.

PATIENT EVALUATION

Most cases of diarrhea are managed by the affected pa-tient or by a family member without need for medicalattention.

Recommendation

l. Medical evaluation should occur for a subset of pa-tients with more severe illness. Specific indications formedical evaluation include: profuse watery diarrhea withdehydration; dysentery, passage of many small volumestools containing blood and mucus; fever (temperature >

38.5'C, l0l.3"F); passage of > 6 unformed stools/24 h or aduration of illness > 48 h; diarrhea with severe abdominalpain in a patient above the age of 50 yr; diarrhea in theelderly (>70 yr of age) or the immunocompromised patient(AIDS, after transplantation, or receipt of cancer chemo-therapy).

Indications.for medical evaluation. Dehydration, definedas dry mucous membranes, decreased urination, and tachy-cardia, is the most common complication of a small bowelsecretory diarrhea and should be promptly evaluated andtreated (6). Osmotic diarrhea seen in patients with smallintestinal injury due to an infectious agent who attempt toingest carbohydrates or other substances (magnesium, salts,fiber) may present with watery diarrhea and dehydration.Patients with dysentery will have more intense and pro-longed illness without antimicrobial therapy (7). Fever isusually a finding associated with an invasive pathogen thatproduces intestinal inflammation. These cases optimallywill be studied for etiologic agents and many will benefitfrom antimicrobial therapy (8). Similarly more intense di-arrhea (>6 unformed stools/24 h) and that lasting more than48 h should be evaluated for cause of illness or treatedempirically (8, 9). Patients with severe abdominal painparticularly if above the age of 50 yr may have a compli-cating i l lness such as ischemic bowel disease (10). Diarrheain the elderly is more likely to be severe and possibly fatal(4, l1), and patients who are immunocompromised usuallyhave complicated and diff icult to manage dianhea (12).

ACUTE INFECTIOUS DIARRHEA IN ADULTS 1963

Recommendation

2. The clinical and epidemiologic history is central topatient medical evaluation and management.

The history. From the standpoint of functional impair-ment from the illness, diarrhea may be categorized as mild(no change in normal activities), moderate (fbrced change inactivities), or severe (disability generally with confinementto bed). In Table 1 the clinical syndromes seen in entericinfections are detailed along with suspected cause and an-atomic location of the disease process. Assessment of theseverity of illness, presence of dehydration, character ofstool pattern, presence of fever, vomiting, or dysentery oftenwill help to focus the evaluation to determine the likelycause of the illness. When diarrhea lasts as long as 2 wk, adifferent list of etiologic agents and conditions should beconsidered when compared with the person with acute di-arrhea. When fever (oral temperature > 38.5"C, or l0l.3"F)is present, the patient has intestinal inflammation character-istically due to invasive bacteria (Shigella, Salmonella,Campylobacter), one of the enteric viruses. or a cytotoxicorganism resulting in mucosal histologic damage and in-flammation (C. dfficile or Entamoeba histolytica). Finally,in taking a history epidemiologic factors and associationsshould be considered (Table 2).

In the homosexual male with diarrhea, three disease trans-mission or clinical scenarios may explain the enteric disease(13). First, because of the sexual practices of many homo-sexual men, there is an increased rate of fecal oral trans-mission of all agents showing this route of spread. Thisincludes Shigella, Salmonella, Campt'ktbacter, and the in-testinal protozoa. Second, in homosexual males who havebeen the recipients of anal intercourse, direct rectal inocu-lation of a pathogen (see Table 1) may lead to proctitis ( l3)associated with rectal pain and tenesmus and passage ofsmall volume stools olten containing blood and mucus. Thethird setting of diarrhea in a homosexual male is when theindividual has developed AIDS (12) (see Management ofthe Immunocompromised Patient with Diarrhea section).

Foodborne or waterborne outbreaks of diarrhea are be-coming more common (14-16). The incubation period ofresultant illness and the predominant symptoms ofien willhelp the clinician to determine the cause of the outbreakbased on clinical grounds. When diarrhea and vomitingoccurs within 6 h of exposure to a food item, food poisoningsecondary to the ingestion of preformed toxin of Staphylo-coccus aureus or Bacillus cereus should be suspected. Whenthe incubation period of an outbreak of diarrheal disease is8-14 h, Clostridium perfringens enteric infection should besuspected. When the incubation period is greater than l4 hand vomiting is the prominent feature of the diarrheal dis-ease, viral agents should be suspected. When fever and ordysentery is present in a majority of cases during an out-break, the invasive pathogens should be considered such asShigella, Salmonella, or Catnpl,lobacter. Other pathogenscan oroduce the same clinical features of an invasive oatho-

1964 DUPONT AJG Vol. 92. No. I l. 1997 AJ(

T,qst-n I

C I ini r: a I S y-nd rome s As s o c i at e d w it h Di a r rhe u

Syndrome Anatomic Location Clinical Features Suspected Etiology

Gastroenteritis

Acute watery (often secretory)diarrhea

Colitis and proctitis

Persistent, >l.l days, diarrhea

Stomach and small intestine

Small intestine, colon may beinvolved

Col i t is : colonic in l lammat iondocumented beyond l5 cm.

Proctitis: inflammation confinedto distal 15 cm of colon

Small intestine, colon may beinvolved

Nausea. vomit ing and ul ter l

diarrhea

Voluminous stools, upperabdominal pain and cramps

Many small volume stools, fecalurgency, tenesmus, and dysentery

Clinical f-eatures will depend uponintestinal location of diseaseprocess

Viral agents or preformed toxrnproduced by S. auretts or B. cereus.any enteric pathogen

Any enteric pathogen

Shigella, Camp-vlobacter most common,also consider Salmonello, Shigatoxinproducing E. coli (.e.g. 0:157 H:1)inr asive E. col i . E. h i .s tu lvt icu.Aeromonas spp, noncholera ViDrlos,

C hlo n,- dia t rac homat i s, infl ammatorybowel diseaset in a recipient of analintercourse consider Ne i t s t rirtgono rrhoeae, herpes simplex.C h lamyclict t ra(homat is, Tre ponenrupallidum

Parasitic inf'ection (Giardia.

C n pt o s po ri diunt, C v" c kt s po ru,Microspnridium), bacterial infection,small bowel bacterial overgrowthsyndrome, lactase deficiency, Brainardsyndrome, malabsorption syndrome

St((

St<

St((

C.

ViI

DrI

Fl(

r l (

Travel to a developing region, tomountainous areas orrecreational bodies of water ofNorth America or to Russia

Recent antimicrobial therapy(past 2 months)

Male homosexual

Day care center contact

Foodborne or waterborneoutbreak

Bacterial agents are consideredamong those who travel todeveloping regions, Giarulia fbrtravel to mountainous areas andrecreational waters associatedwith wildlife andCryptosporidium or Giardict for

those visiting Russia

ClostridiLtnt dfficile

All enteric pathogens showingl 'ecal oral route ol ' t ransmission;those spread by receptive analintercourse or a variety of

agents if AIDS is present (see

Table 1 and text)Any agent spread by fecal oral

route, commonly Shigella spp,G iardia, or C mptos pnridium

The incubation period and clinicalfeatures of the illness often willgive clues as to etiology, thelaboratory will be important toidentification of entericpathogen (see text)

Trer-n 2Epidemiobgic Factors Intportant in the Evaluation of the Patient -'ith

Acute Diarrhea

Epidemiologic Consideration Etiologic Agent to Consider

NONINFECTIOUS AND EXTRAINTESTINALCAUSES

Noninfectious and extraintestinal processes may presentas acute diarrhea. In all patients with dianhea, particularlypersistent or chronic diarrhea or when there is severe ab-dominal pain or underlying disease processes, a noninfec-tious cause of the i l lness should be considered. Some of theimportant diagnoses to consider include irritable bowel syn-drome, inflammatory bowel disease, ischemic bowel disease(particularly in someone older than 50 yr of age with otherevidence of peripheral vascular disease), partial bowel ob-struction, and pelvic abscess in the area of the rectosigmoidcolon. Pernicious anemia, pellagra, malaria, Whipple's dis-ease, diabetes mellitus, small bowel scleroderma, smallbowel diverticulosis, and various malabsorption syndromesmay present as acute or persistent diarrhea resembling in-fectious diarrhea.

EMPIRIC THERAPY OF SELECTED PATIENTS

There are two types of patients who might be consideredfor empiric antimicrobial therapy without additional evalu-ation. They include patients in whom bacterial diarhea issuggested by clinical f'eatures and/or by the presence ofoccult blood or fecal leukocytes in stool samples. or patientsin whom diarrhea has persisted for 2 wk or longer andGiardia is suspected. Patients not treated empirically shouldbe studied to determine the presence of etiologic agents byusing laboratory tests and procedures (Table 3).

Re,

1 0pot

pal

mi,

tlel

her

pal

shr

nul

mi

sto

s1g

Thoul

acl

dia

are

the

Regen such as invasive Escherichia coli, Aeromorzas spp, andnoncholera Vibrios. Enterohemorrhagic E coli infectionofien is associated with the passage of bloody stools, butfever is generally absent or low grade. ter

AJG - November 1997 GUIDELINES ON ACUTE INFECTIOUS DIARRHEA IN ADULTS 1965

Taer-E 3

Selected Luboratory- Tests and Procedures Used in the Etioktgic Diagnosis oJ Putient.s vvith Diarrhea

Test or Procedure lndications Likely Etiologic Agents

Stool culture fbr enterohemorrhagic E. coli(e.g. Shigatoxin producing O157:H7)

Stool culture for Vibrios

Stool culture tor Yersinio including cold

enrichment

C. df f ic i le toxin assay

Virus examination(see likely etiologic agents)

Duodenal intubation or l4-c-d-xylose breath

test

Flexible sigmoidoscopy with biopsy of

abnormalities

Flexible sigmoidoscopy and upper

endoscopy with biopsy of abnormalities

Epidemic foodborne dysentery especially

hamburger-associated diarrhea or bloody

dianhea with few or no fecal leukocytes or

diarrhea in a case(s) of hemolytic uremic

syndromeSevere and profuse watery diarrhea in a cholera-

endemic area or seafood associated diarrhea or

diarrhea after consumption of inadequately

cooked seafood, including sushi or ceviche

Febrile dysentery or diarrhea and severe

abdominal pain or pseudoappendicitis-like

symptomsIn a patient with diarrhea who has received an

antimicrobial in the past 2 months

Waterborne or fbodborne illness with incubation

period >12 h in which vomiting is the

predominant symptomIn a patient with persistent dianhea not

responding to empiric therapy

Homosexual male with moderate to severe

diarrhea, any patient with persistent diarrhea

and clinical colitis (see Table l) not

responding to antimicrobial therapy or without

diagnosis after laboratory evaluation

Any patient with persistent dianhea without

evidence of colitis (see Table l) and without

response to empiric therapy

Cytotoxigenic E. coli (.O157:H7, 026:Hl l. etc.)

V. cholerae and noncholera Vlbrlos

Y. enterocolitica

C. diJJicile

Rotavirus, enteric adenovirus (type 40, 41),

Norwalk virus,* and other small round structured

viruses

G i a rd i a, C ry p to s p o r i di um, rarely St rongy ktide s

stercoralis, small bowel bacterial overgrowth

syndromeSee causes of colitis and proctitis (Table I above)

Causes of colitis and proctitis (Table 1 above) as

well as upper intestinal int'ection by Giardiu,

C rt'pxtspo ridium, C.'-c lospora, M icrospori di um,

cytomegalovirus, HlY, M 1" c oba ct e ri um av i um

int race llulare complex

* Norwalk virus detection is presently done only in a research laboratory

Recommendation

3. In patients with fever (oral temperature > 38.5'C, or101.3'F) plus either leukocyte-, lactoferrin-, or hemoccult-positive stools or in patients with acute dysentery or inpatients with moderate and severe travelers' diarrhea, anti-microbial therapy may be given empirically (Table 4).

Empiric therapy for presumed bacterictl diathea. In pa-tients with fever and either fecal leukocyte-, lactoferrin-, orhemoccult-positive stools, inf-ection with invasive bacterialpathogens such as Shigella, Salmonella, and Campylobactershould be considered (17, l8). A majority of patients withnumerous fecal leukocytes will respond favorably to anti-microbial therapy (19-21). Finding hemoccult-positivestool in patients with acute diarrhea has the same clinicalsignificance as finding numerous fecal leukocytes (18, 22).The same pathogens associated with the presence of numer-ous l'ecal leukocytes will generally be found in patients withacute dysenteric disease (23, 24). Patients with travelers'diarrhea, particularly those with moderate to severe illnessare characteristically infected with bacterial pathogens andtheir illness is shortened by antimicrobial therapy (25*21).

Recommendation

4. Patients with diarrhea lasting 2 to 4 wk without sys-temic symptoms or dysentery may be studied for cause of

illness or may be treated empirically with anti-Gra rdia ther-

apy (Table 4).Empiric treatment of presumed giardiasis. While many

clinicians would pref'er to evaluate the patient with per-

sistent diarrhea for cause of i l lness some elect to treat

empirically for presumed giardiasis. This approach is

reasonable in view of the importance of Giardia in this

syndrome (28) and because at least half of studied stools

of patients with giardiasis wil l be negative for the parasite(29, 30). Work-up of those failing to respond to empiric

therapy for Giardict (usually stool enzyme immunoassay)may then be indicated. The most frequently used empiric

therapy of presumed giardiasis is metronidazole (see Ta-

ble 4), which also may be eff 'ective against a small bowelbacterial overgrowth syndrome associated with persistent

diarrhea, a problem occasionally seen after an entericinfect ion (31) .

LABORATORY EVALUATION

Laboratory tests and procedures may be used to evaluatepatients with illness calling for medical evaluation (see

above) and fbr other patients in whom a definable pathogen

is suggested by the history (Tables I and 2).

l 966 DUPONT AJG - Vol. 92, No. I I, 1997

TaeLL, 4

Indications for Empiric and Specific Antimicrobial Therapt in InJectious Diarrheu

Indication for Antimicrobial Therapv

Fever (oral temperature >38.5 'C or 101.3 'F) together

with one ol the following: dysentery (grossly bloody

stools) or those with leukocyte-, lactof'errin-, or

hemoccul t -posi t ive stools

Moderate to severe travelers' diarrhea

Persistent diarrhea (possible Grarulia inf'ection)

Shigel los is

Intestinal salmonellosis

CampylobacteriosisEnteropathogenic E. coLi diarrhea (EPEC)

Enterotoxigenic E colr diarrhea (ETEC)

Enteroinvasive E. colr diarrhea (EIEC)

Enterohemorrhagtc E. coLi diarrhea (EHEC)

Ae romonas diarrheaNoncholera Vibrb dianheaYersin iosis

Giardiasis

Intestinal anrebiirsis

C n pto s po rid i urn diarrhea

Isospora diarrheaCtt:losporu diarrhea

Suggested Antimicrobial Therapy

Quinolone:'k NF 4(X) mg, CF 500 m, OF 300 rng b.i.d. for

3-5 days (see text)

Quinolone:+ NF 400 mg. CF 500 mg, OF 300 rng b.i.d. for

1-5 days (see text)Metronidazole 250 mg g.i.d. fcrr 7 days

If acquired in the U.S. give TMP/SMX 160/tt(X) mg b.i.d.

for 3 days, if acquired during international travel treat as

f'ebrile dysentery (above)l check to be certain of

susceptibility to drug employed

If healthy host with mild or moderate symptoms, no

therapyi fbr severe disease or that associated with f'ever

and systemic toxicity or other important underlying

condition (see text) use TMP/SMX 160 mg/t3(X) mg or

quinolone:'k NF 400 rng, CF 500 mg, OF mg bid lbr 5 to

7 days depending on speed of response

Erythrornycin stearate 500 mg h.i.d. for 5 days

Treat as f'ebrile dysenteryTreat as moderate to severe travelers' diarrhea

Treat as shigel los isAntin'ricrobials are generally withheld except in particularly

severe cases in which usefulness of these drugs is

uncertaln (see text)Treat as febrile dysenteryTreat as febrile dysentery

For most cases, treat as febrile dysentery, for severe cases

give ceftriaxone 1 g q.d. lY fbr 5 days

Metronidazole 250 mg q.i.d. for 7 days or (if available)

tinidazole 2 gr in a single dose or quinacine 100 mg t.i.11.

fbr 7 daysMetronidazole 750 mg t.i.d. fbr 5 l0 days plus a drug to

treat cysts to prevent relapses: diiodohydroxyquin 650

mg t.i.d. for 20 days, or paromomycin 500 mg r.l.d. fbr

l0 days or diloxanide furoate 500 mg t.i.ri. fbr l0 d

None, fbr severe cases, consider paromomycin 500 mg

r.1.21. fbr 7 daysTMP/SMX 160 rng/t100 mg b.i.d. fbr 7 days

TMP/SMX 160 rng/tt00 mg D.i.rl fbr 7 days

AJG

proknons

Stare tmedrouttivit$9ottureappl

Rec,'7

shothavRusexpoccdiarblor

Istu(obtracupatldiarEntC4whporGi(tfar

NoGi(ofthe

i 'Fluoroquinolones include nortloxacin (NF), ciprofloxacin (CF). and ofloxacin (OF).

Rec'ommendilion

5. The f-ecal leukocyte, lactof'errin, or hemoccult blood

test is a useful screening test in patients with moderate to

severe acute infectious diarrhea because they support the

use of empiric therapy in the febrile patient (see above) and

when negative may eliminate the need for stool culture in

some cases of diarrhea.Laboraton screening. Fecal leukocytes, lactoferrin, or

occult blood are fbund in diarrhea patients with difluse

colonic inflammation (17, l8). The most commonly identi-

fied pathogens in patients with a positive test result include:

Shi ge lla, Salmone Lla, Cump,,-lobacter, Ae romonas, Ye rs inia,

noncholera Vibrios (17, l8), and Clostridium dfficile (32).

Low numbers of fecal leukocvtes are found in patients with

intestinal amebiasis.

Recommendcttion

6. A stool culture should be obtained in a patient with one

of the following: severe diarhea; a temperature > 38.5'C,

or 101.3'F, (orally); passage of bloody stools; stools contain

leukocytes, lactoferrin, or hemoccult blood; or, the patient

with persistent diarrhea has not been treated with antibac-

terial agents empirically.Routine stool culture. The bacterial enteropathogens

identified by normal stool culture are Shigella, Salmonella,

C ampy I obac t e r, Ae r o mona s, and Y e r s inia. Severe (i ntense)

diarrhea, moderate to high f'ever, dysentery, finding f'ecal

leukocytes, lactoferrin, or hemoccult blood positive stools

all are predictive of finding an identifiable bacterial entero-pathogens when fecal samples are submitted to the labora-

tory for culture. The bacterial pathogens may cause more

Hosrt(trelmasidan(waforberukpr(wilerstrterofabenth

AJG - November 1997 GUIDELINES ON

prolonged diarrhea when compared with pathogen-negativenonspecific diarrhea (25, 33, 34).

Studies in the United States have found that stool studies

are frequently inappropriately ordered resulting in excessivemedical costs (35). In two studies, stool cultures as obtainedroutinely in selected locations were found to have a posi-

tivity rate of 2Vo or less, making the cost of the test between

$900 and $1200 per pathogen detected (35, 36). Stool cul-

tures should not be ordered routinely but reserved for the

appropr iate pat ient and set t ing.

Recommendation

7. Patients not treated with empiric antiparasitic therapy

should be studied for parasitic causes of diarrhea if they

have persistent diarrhea; diarrhea has followed travel to

Russia, Nepal, or mountainous regions; they have beenexposed to infants attending day care centers; diarrhea has

occurred in a homosexual male or a patient with AIDS;

diarrhea is part of a community waterborne outbreak; or has

bloody diarrhea with few or no fecal leukocytes.Laboratory evaluation for parasites. Although less well

studied than the value of routine stool culture, the common

obtaining of "O and P's" (ova and parasites) in patients withacute diarrhea is not cost effective (37). As indicated above,patients with giardiasis often have persistent diarrhea. Thediarrheal illness associated with Cryptosporidium (28) and

Entamoeba histolytica (38) may be protracted. Infection by

Cryptosporidium, Giardia, or both, should be suspectedwhenever diarrhea follows trips to Russia (28,39); Cyclos-pora should be considered in travelers to Nepal (40); and

Giardia should be suspected in persons who have recentlytraveled to mountainous areas or to recreational waters of

North America (41). Among infants in day care centers,

Giardia (42) and Cryptosporidium (43) are common causesof diarrhea and may be spread to adult contacts because of

the low inoculum required for human infection (44, 45).

Homosexual males may be infected with a variety of para-

sites including Giardia and Entttmoeba hi.stolytica. In pa-

tients with AIDS-associated diarrhea, parasites represent the

major definable pathogens (46). The specific agents to con-

sider in this setting are mentioned below. Both Giardia (41)

and Cryptosporidium (16) may cause extensive communitywaterbome outbreaks. Numerous leukocytes are not usuallyfound in the stools of patients with intestinal amebiasisbecause of the presence of uninflamed mucosa between

ulcerations and because of the lytic effect of exotoxinsproduced by the organism (48). Therefore when a patient

with diarrhea is passing bloody stools and there are few

leukocytes, amebiasis should be considered. Stools may be

studied by routine microscopic techniques for parasitic en-

teropathogens by a well-trained parasitologist or in the case

of Giardia and Cryptosporidiwn, by a commercially avail-

able immunofluorescent antibody tests (49) or by diagnostic

enzyme immunoassays (50), which may be more sensitive

than microscopic studies (5 I, 52).

ACUTE INFECTIOUS DIARRHEA IN ADULTS 1961

Recommendation

8. In patients with certain epidemiologic findings, fecalsamples should be collected and sent to the laboratory forspecific enteropathogens including enterohemorrhagic E.

coli. Vibrio cholerae, noncholera Vibrios, and Yersinia.

Selected patients may be studied for the presence of C.

dfficile toxin, or viruses or they may be studied by endos-copy (see Table 3).

Special bacterial enteropathogens. A number of bacte-rial enteropathogens will not be detected by routine stoolculture. These pathogens include enterohemorrhagic colitisproducing E. coli 0157:H1 and other Shigatoxin producing

E. coli (53), y. cholerae, other noncholera Vibrios (54), andYersinia (55), although routine stool cultures will identifymost of the strains of Yersinia. Except for E. coli 0157:Hi ,which is readily detected by stool culture using specializedmedia, the other diarrheagenic E. coli are presently onlydetected by research laboratories. Stool assays for C. dffi'cile toxin by tissue culture assay or enzyme immunoassayshould be carried out in the patient who is currently receiv-ing antimicrobials or who has received antimicrobials in thelast 2 wk. In the case of food- or waterborne outbreaks, inwhich the illness is associated with vomiting as the majorclinical feature and the incubation period is > 12 h, viralagents should be considered (see Table 3). Homosexualmales with diarrhea and any patients with persistent diarrheanot responding to empiric therapy may be evaluated byendoscopy.

MANAGEMENT

Fluid and Electrolyte Treatment

Recommendation

9. In all patients with diarrhea requiring medical evalua-tion, fluid and electrolyte therapy and alteration of the dietshould be part of the management.

Fluid therapy and diet alteration. For most cases ofacute diarrhea, the most important form of therapy consistsof t'luid combined with electrolytes (56). Whereas suchtreatment is life saving for young infants in the developingworld, in the United States the most important adult groups

in which special attention should be given to fluid therapyare the elderly and the immunosuppressed. For these per-

sons solutions containing sodium in the range of 45 to 75mEq/L are recommended (Pedialyte or Rehydrolyte solu-

tions). In the nondehydrated otherwise healthy person with

acute diarrhea, sport drinks, diluted fruit juices, and otherflavored soft drinks augmented with saltine crackers andbroths and soups can meet the fluid and salt needs in nearlyall cases. For dehydrating cholera-like diarrhea more ag-gressive fluid therapy will be required. Here the ideal for-mulation of oral fluids is Na 60-90 mBq[. K 20 mEq/L, Cl80 mEq/L, citrate 30 mEq/L, and glucose 20 g/L. A home-made version of this form of oral rehydration solution fbrmore severe diarrhea is to prepare two separate glasses that

r 968 DUPONT AJG Vol. 92, No. I l, 1997

Taer-c 5S\-nptomutic Treatmetlt oJ Acute Diarrhea

Indication/Perspective Dose and AdministrationPhlil"l:91.s. lqT,Loperan.ride (lmodium)

Drphenoxylate withatropine (Lomot i l )

Tincture of opium

Bismuth subsal icy late(Pepto-Bisrnol)

Octre0tide

Acute diarrhea, f'ever is absent or low grade, dysentery isnot present/minirnal central opiate effects, thepref'erred symptomatic drug when used fbr most

nonf'ebrile, nondysenteric casesAcute diarrhea, t-ever is absent or low grade, dysentery is

not presenVhas central opiate eff'ects, with overdoseliability. atropine may cause side eflects withoutof lering antidianheal efl'ects

Acute diarrhea, fever is absent or low grade, dysentery isnot present/occasionally useful in HlV-associated

diarrhea when the saf'er loperamide is notAny fbrm of acute diarrhea/in most cases is less

efl'ective than loperan'ride and cannot be cornbinedwith ant imicrobia ls l should not be used in HIV-positive patients with diarrhea

AlDS-associated diarrhea not responding to othertreatment/considered last resort therapy for thesepatients, once symptoms are controlled, the patientshould be started on other more convenientpreparations

2 tablets (4 mg) initially then 2 mg aftereach unfbrmed stool not to exceed 8 mg/day (OTC dose) or l6 mg/day (prescription

dose) <2 days2 tablets (.4 ng) q.i.d. fbr <2 days

0.5-l rnl p.o. cq 4-6 h for <2 days

30 ml or two tablets each 30 minutes fbreight doses, may repeat once again daytwo

100-500 pg subcutaneously t.i.d.

are consumed alternately. The first contains 8 ounces oforange. apple. or other fruit juice (supplying potassium), %teaspoon of honey or corn syrup, and I pinch table salt; thesecond glass contains 8 ounces clear water plus I /4 teaspoonof baking soda (56).

During a bout of acute diarrhea, calories (energy) shouldbe provided to facilitate enterocyte renewal (56). Diet fol-lows clinical course. Boiled starches/cereals (potatoes, noo-dles/rice, wheat, oat) with some salt represent ideal foodsduring episodes of watery diarrhea. Crackers, bananas, yo-gurt, soup, and boiled vegetables can also be used. Whenstools are formed, diet may return to normal as tolerated.Many authorit ies would exclude milk products early in theil lness but clinical lactose intolerance is not commonlyfound in cases of acute diarrhea.

Nonspecific Therapy

Rec'onunendcttion

10. When nonspecific therapy is desired, loperamide isthe drug of choice for most cases of diarrhea.

Svmptomatic treatment o.f acute diarrhea. In Table 5 abrief perspective on the use of symptomatically acting drugsis provided. The antimotil i ty drugs (loperamide, diphenoxy-late with atropine, and tincture of opium) are the mostetfective drugs directed to treating symptoms. They work byslowing the intraluminal f ' low of l iquid facil i tating intestinalabsorption (57). Loperamide is generally the recommendedagent fbr most cases of diarrhea when symptomatic treat-ment is used because of saf'ety and expected efficacy inwhich stool number is generally reduced by approximately807c (58. 59). Diphenoxylate may not be the ideal antimo-ti l i ty drug. although it is less expensive than the pref-erredloperamide. Diphenoxylate possesses central opiate eff'ects

and may lead to death if a child takes a parent's medication.Also, the atropine added to the drug may lead to objection-able cholinergic side effects without adding antidiarrhealeffects. The antimotility drugs should not be used in patientswith f'ebrile dysentery in which disease may be prolonged(60). This disease prolongation by antimotility agents is notcommonly seen, and most clinicians do not need to worryabout use of the drugs in nondysenteric forms of diarrheacaused by invasive colonic pathogens, providing antimicro-bial therapy is administered (61). In patients with enterohe-morrhagic E. coli (EHEC) infection, the hemolytic uremicsyndrome (HUS) may be facil i tated by administering anti-motility agents (62) or these agents may worsen neurologicsymptoms (63).

Recommendation

I l. Bismuth subsalicylate is the pref'erred agent whenvomiting is the imponant clinical manif 'estation of entericinfection.

Bismuth subsalicylate treatment of gastroenteritis. Bis-muth subsalicylate is effective in treating acute diarrhea,reducing the number of stools passed by approximately 507ocompared with a placebo (64). The antidiarrheal eff'ect ofthe drug is through its antisecretory salicylate (65, 66) Thedrug has antibacterial properties, which may explain itsvalue in prevention of travelers' diarrhea when used as aprophylactic agent (67, 68). Bismuth subsalicylate seems tobe the most eff'ective agent in improving the symptom ofvomiting associated with enteric viral infection (69). Atta-pulgite (Kaopectate or Diasorb), a clay-like material, ab-sorbs water and makes stools more fbrmed. The preparationis less efl'ective than loperamide (70). Because it is notubsorbed. it is ouite safe.

AJG

Reco

1 2nocoBisn

S,yp rotltientrt o mnot lo f dtherewlsesionBisnpronexceathywithidecaseclicWherhea(13)Theexpeatic

Antr

Irinf'er

Rect

l :ablelabo

(,

cau!strala p aandfoo<willrecc

J ,

firntrimStattant

AmicFirsorgidosrexp

AJG ltlovember 1997

Recommendation

GUIDELINES ON

12. Loperamide is the recommended treatment for immu-nocompromised patients with diarrhea of uncertain etiology.Bismuth subsalicylate should not be used in these patients.

S\tmptomatic treatment oJ'diarrhea in the immunocom-promised patient. A number of immunocompromised pa-tients with diarrhea have persistent diarrhea that is difficultto manage (12). In most of the cases a treatable pathogen isnot identified or if a potential agent is found the symptomsof diarrhea continue in the face of specific antimicrobialtherapy. Table 5 lists the drugs available. As in the other-wise healthy patient, those with underlying immunosuppres-sion can best be managed with symptomatic therapy (11).Bismuth subsalicylate should not be given to immunocom-promised patients with diarrhea to prevent the taking ofexcessive doses and the occurrence of bismuth encephalop-athy (72).In some patients tincture of opium has been usedwith success in treating patients not responding to loperam-ide (unpublished observations). In otherwise refractorycases of AIDS-induced diarrhea, octreotide, a synthetic cy-clic octapeptide analog of somatostatin, may be effective.Whereas the drug is best used in pathogen-negative diar-rhea, it may be useful in some patients with microsporidiosis(73) and possibly other otherwise nontreatable conditions.The drug must be given subcutaneously, and it is quiteexpensive. It should be considered a last resort to symptom-atlc management.

Antimicrobial Therapy

In Table 4 recommendations fbr antimicrobial therapy ininfectious diarrhea are provided.

Recomrnendation

13. Specific antimicrobial therapy is given when a treat-able pathogen is identif ied in stool samples submitted to thelaboratory.

SpeciJic antimicrobial therapy. For specific therapy, thecause of i l lness is usually determined by laboratory demon-stration of a treatable pathogen (see Table 4). Alternatively,a patient may be treated based on epidemiologic evidenceand laboratory study of other cases during a well-definedfoodborne or waterborne outbreak. A few specific infectionswill be discussed. Most are l isted in tables along withrecommendations for therapy without turther discussion.

Shigella-Recommendation 14: All patients with con-firmed shigellosis are treated with an antimicrobial agent:trimethoprim/sulfamethoxazole if acquired in the UnitedStates and a fluoroquinolone if acquired outside or if resis-tant to trimethoprim.

Antimicrobial therapy of shigellosis: In shigellosis, anti-microbial therapy is indicated in all cases fbr two reasons.First, the i l lness is shortened by antimicrobials providing theorganisms is susceptible to the drug used (74). Second, thedose of ShigeLla required to cause inf'ection is low (75)explaining the potential for int'ecting strains to be spread

ACUTE INFECTIOUS DIARRHEA IN ADULTS 1969

from person-to-person. Therefbre, there is a public healthreason fbr treatment of inf'ected patients. The treatments ofchoice fbr culture proven shigellosis are TMP/SMX or afluoroquinolone given fbr 3 to -5 days. It the Shigella inf'ec-tion is acquired in the U.S., TMP/SMX is the pref'erredinitial treatment of choice (16). If the inf'ection is acquiredduring travel, TMP resistance occurs commonly (77). Forthese cases a quinolone is indicated.

Salmonella-Recommendation I 5: Selected patients withintestinal salmonellosis should be treated with a quinolonefor possible systemic infection including those with feverand systemic toxicity, those with dysentery or with under-ly ing immunosuppression.

Antimicrobial therapy of intestinal salmonellosis: Nonty-phoid salmonellosis is an important form of diarrheal dis-ease occurring in up to 2 mill ion persons in the United Stateseach year (78). The decision to treat patients with intestinalsalmonellosis with antimicrobial agents should depend uponthe age, underlying health of the host, and severity ofclinical i l lness. In healthy persons with mild symptomstreatment should not be given because therapy may encour-age the prolif'eration of the int'ecting organism, leading toprolongation of excretion of the strain (79). However, in-testinal salmonellosis is associated with bacteremia in be-tween 2 and 14Va of cases (80) and systemic complicationsof the bacteremia may occur. Certain underlying conditionsincrease the fiequency of bacteremia. Bacteremia is com-mon in infants under 3 months of age (8 I ) and in the elderly(>65 yr of age) (l l). Other risk factors for bacteremiainclude HIV infection and AIDS (82); uremia (83); malig-nancy, including hematologic, lymphatic, and solid tumors(84); after renal transplantation (85); and congenital andacquired immunodeficiencies including corticosteroid use(86). There are other conditions that may predispose patientsto localized extraintestinal infection when Salmonella gas-troenterit is develops. This includes patients with an aorticaneurysm, prosthetic heart valve, vascular graft, or ortho-pedic prosthesis. Antimicrobials are indicated fbr cases ofintestinal salmonellosis complicated by any one of theseconditions as well as fbr otherwise healthy persons withf'ebrile i l lness, systemic toxicity or dysenteric disease.

For all practical purposes, all patients with intestinalsalmonellosis illness severe enough to lead to hospitaliza-tion should be given antimicrobial therapy (Table 4).

Camltylobacter-Recommendation l6: Patients with cul-ture-proven Campylobacter inf'ection are treated with anantimicrobial agent to shorten illness, although developmentof antimicrobial resistance is becoming a problem.

Treatment of Campylobacter diarrhea'. Cunpylobac'terresembles Salmonella in many ways. First they both showan animal, often poultry, reservoir fbr human inf-ection.Second, antimicrobial resistance occurs commonly duringtherapy or over time in a population (87). Erythromycin wil lshorten the duration of Camp,vbbacter dianhea (88). lfsusceptible to the drugs, the fluoroquinolones (89) are ef-fective against campylobacteriosis. Unfbrtunately, quin-

t970 DUPONT

TABLE 6

Prevention of Travelers' Diarrhea

All TravelersDiet and Beverage Precautions

Consume only safe items while in high risk area, including airplane

leaving area:l) Steaming hot fbods and beverages (e.g. cooked foods, coffee, tea)

2) Acidic foods (e.g. citrus)3) Dry foods (e.g. bread)4) Foods with high sugar content (e.g. syrups, jellies)

5) Carbonated drinks (e.g. bottled soft drinks and beer); bottled,

uncarbonated water may not be safe

Selected TravelersThose who wish prophylaxis

BSS* 2 tablets with meals and at bedtime (8 tablets/day)(62Vc effective in eliminating diarrhea)

Those who might be encouraged to take prophylaxis include those with

underlying illness: AIDS, prior gastric surgery and those taking proton

pump inhibitors (omeprazole), or those who cannot afford an 8 hour

illness (e.g. politician, honeymoon couple, or weekend scuba diver)

Antimicrobial agent (same drug used normally for therapy(Table 4) in single daily dose during time at risk(907c effective in eliminating dianhea)

Use of prophylactic antimicrobials is controversial

* BSS, bismuth subsalicylate.

AJG - Vol. 92, No. 11, 1997

olone resistance is increasing worldwide among Campr--lobacter isolates (87, 90, 9l). It is necessary to confirm invllro susceptibility of infecting strains of Campylobacter.Inaddition to erythromycin, azithromycin may be used inquinolone-re sistant C ampy I o b ac t e r infections ( 8 7 ).

Diarrheagenic E. coli: There are four major diarrheagenicE. coli: enteropathogenic E coli (EPEC), enterotoxigenic E.coli (ETEC), enteroinvasive E. coli (EIEC), and enterohe-monhagic E. coli (EHEC). EPEC diarrhea is primarily aproblem of infant populations. The related HEp-2 cell ad-herent E coli are pathogens ofboth children and adults (33,

92). The diarrhea caused by HEp-2 cell adherent E. colitends to be persistent (33, 93). These strains have beenshown to be causes of prolonged diarrhea in patients withAIDS in Zambia (94). EPEC and HEp-2 cell adherent E. coliare highly resistant to most antimicrobials except the newerquinolones (95). It is not known whether antimicrobials willshorten the illness.

ETEC diarrhea will respond to antimicrobial therapy (21,

25,96). This form is the major cause of travelers' diarrhea.TMP/SMX or the quinolones remain standard therapy forinfection depending upon susceptibility.

AJC

Fpatir

Rec

lforshi

I

dia,

con

f'ect

cok

crei

synmortheropnrec(thethet

I

\ / n m i f i n r - r i s M : i 6 1 '

M a n i f e s t a t i o n o fI l - l -nes s

Y e sJ

Bi - smu thQ r r h < a l i a r ; l r l - a

( s e e T a b l e 5f o r d o s e )

A l - I Take ORT anda r r a r r r r a r r : o a t l o v l - \

\ v v v r v r \ e /

I

D i a r r h e a i s M a i o r

M a n i f e s t a t i o no f I l l - n e s s

i

M i l dI l - f n e s s

I

N o o t h e r' l - r a r l - m o n I

C o n s i d e rT , n n o r : m i r l a

or b i smu t .hq r r l ' r q a l i c r r l a J - a

( s e e T a b l - e 5 )

M o d e r a t e t oSeve re I . l - l - ness

I

I s Feve r and , /o rf ) r r s e n f e r r r P r c s e n t ?

N o -

N o

O r r i n n l n n a *Y s + r r v + v r f v

/ e o a ' f : h l a t r \

n l r r q

T . n n o r r m i d a

/ ^ ^ ^ f r - L l ^ t r \

Y e s

I

Q u i n o l o n e */ q a a T : h l a 1 \

A l o n e

Frc;. 1. Self-treatment of gastroenteritis and diarrhea in the international traveler. *From the interior of Mexico during the summertime, TMP/SMX maybe used instead of a quinolone.

AJG - November 1997 GUIDELINES ON ACUTE INFECTIOUS DIARRHEA IN ADULTS

A l l P a t i e n t s F u r n i s h T w o E r e s h l yP a s s e d S t o o l s F o r R o u t i n e B a c t e r j - o l o g y ,P : r a s i t o l o c v : n d C . l i f f i c i L e T o x i n a n d

A B l - o o d C u i t u r e i s O n - [ E l n E O f o r B a c t e r i aA n d M y c o b a c t e r i a

IT r o : t a l r l e P a t h n c q n l d e n t i f i e d

{N o

t91 |

IY e s

IP : f h n a o n T r o : t o c i / ( a a

T a b l e 7 )

1 ^ n ^ , , ^ ^ r ^ ^ , , r r o l _ o n eL W u O y - V r O 9 U ! r .

Admi n i s te red*1 J

S r u m n t o m s D i s : n n c a r S r r m n f n m s R e m a i n

lN o E u r t h e r S y m p t o m a t i cT r e a t m e n t T r e a t m e n t ( S e e T e x t

a n d T a b l e 5 )

l

S y m p t o m s R e c u r o r a r e u n c o n t r o f l e d

.t

E v j - d e n c e o f C o l i t i s i s P r e s e n t ( P a s s a g e o f M a n y S m a - L l V o l u m e S t o o l sW i t h F e c a l U r g e n c y W i t h a t L e a s t O n e o f t h e F o l l o w i n g : T e n e s m u s ,n U c 6 - f 6 r t / E - a a : I T , e r r k o c r r f c q n r H e m o c C u ] L - P O S i C i V e S t O O I - Sv - L I t

F f e x i b l e S i g m o i - d o s c o p y w i t h G a s t r o d u o d e n o s c o p y w i t h b i o p s yB i o p s y o f L e s l o n s W l t h B i o p s y , S m e a r s a n d C u l t u r e s F o r S p e c i a lA t t e n t i o n L o C M V , C M V , M y c o b a c t e r i a , F u n g i a n dM y c o b a c t e r i a , A d e n o v i r u s , P a r a s i t e s P l u s F l e x i b l eF u n g i , a n d H e r p e s S i m p l e x S i g m o i d o s c o p y ( S e e D e t a i f s i n

A d j a c e n t B o x )

Frc;. 2. Approach to the work-up of the immunocompromised patient with acute or persistent diarrhea. Imrnunocolnpromised patients include thosepatients with AIDS. afier organ transplantation, and those receiving cancer chemotherapy. 'r'Quinolones include norfloxacin, ciprofloxacin, or ofloxactn.

I

Y e s

Rec'ommendation

17. Antimicrobial therapy is currently not recommendedtbr patients with diarrhea due to E. coli 015'7:51 and otherShigatoxin producing E. coli.

Antimicrobial treatment oJ' enterohemorrhagic E. colidiarrhea. The antimicrobial therapy of EHEC infection iscontroversial. Administration of antimicrobials early in in-f'ection may enhance the release of toxin from killed intra-colonic organisms leading to greater absorption and in-creased propensity to develop the hemolytic uremicsyndrome (HUS) (97). Retrospective analyses have further-more oft-ered suggestive infbrmation that prior antimicrobialtherapy may in fact be a predisposing tactor in the devel-opment of HUS (98, 99). In a single prospective study. priorreceipt of TMP/SMX late in the i l lness did not predispose tothe development of HUS (100), making the whole issue oftheraov uncertain.

I

N o

I

THE INTERNATIONAL TRAVELER

Diarrhea occurs during travel with rates varying accord-ing to levels of microbial contamination in the host countryand the region of origin. For persons from the United States,diarrhea occurs in 2 to 4o/o of travelers during travel toanother low-risk region [United States, Canada, westernEurope, Japan (if raw fish, which may be contaminated withnoncholera Vibrios, is not consumed), South Africa, Aus-tralia, and New Zealandl. For travel fiom the United Statesto high-risk areas (most parts of Latin America, Africa, andsouthern Asia. the rate of diarrhea occurrence is 40olo. Fortravel to intermediate risk regions (northern Mediterraneancountries, the Middle East, China, and Russia) the rate ofdiarrhea is l0 to l5c/o.The remaining discussion wil l be fortravelers fiom the United States during travel to high-riskregions. This illness has two unique f'eatures. First, morethan 80o/c of the i l lness is caused by bacterial pathogens. In

1972 DUPONT

Tasr-B 7Indications for Specific Antimicrobial Therapy in Infectious Diarrhea in

I mmuno c omp romi s e da P at i e nt s

Indication forAntimicrobial

TherapySuggested Antimicrobial Therapy

Shigellosis If acquired in the U.S. give TMP/SMX160/800 ng b.i.d. for 7-10 days, ifacquired during international travel treat asfebrile dysentery (above) Check to becertain of susceptibility to drug employed

TMP/SMX 160 mg/800 mg b.i.d. orquinolone:i NF 400 mg, CF 500 mg, OF300-400 mg b.i.d. for 14 days, repeat stool

cultures 1 week after treatmentParomomycin 500 mg p.a. 4.i.d. with food

for 14-28 days, then 500 mg b.l.dindefinitely, with treatment failures, maytry azithromycin 2.4 g p.o. day 1, then1.2 glday for 27 days, then 600 mg/day formaintenance treatment given indefinitely

320 mg TMP/I600 mg p.o. SMX b.i.d. fbr2*4 wk then 160-320 mg TMP/800-1600mg SMX q.d.(p.o.)

TMP/SMX 160 mg/800 mg p.o. q.i.d. for l0days then 160 mg/800 mg once three timesa week indefinitely

Albendazole 400 mg p.o. b.i.d. >4 wk ormetronidazole 500 mg p.o. t.i.d. oratovaquone 150 mg p.o. r.i.d. continuedindefinitely

Ganciclovir 5 mg/kg i.v. q 12 h or q 8 h for14-21 days and foscarnet 60 mg/kg i.v. q

8 h or 90 mg/kg i.v. q 12 h for 14-21 daysClarithromycin 500 mg b.i.d. ethambutol l5

mg/kg/day plus one of the lbllowing CF500-750 mg b.i.d. clofazimine 100 mg/day, rifampin 6O0 mg q.cl., or rifabutin 300mg/day p.o. indefinitely

Intestinalsalmonellosis

Cryptosporidiumdiarrhea

Isospora diarrhea

Cyclosporadianhea

Microsporidiosis

Cytomegalovirusdiarrhea

Mycobacteriumavrumintrace llularecomplex

* Patients with AIDS, after organ transplantation, and those receivingcancer chemotherapy.

t Fluoroquinolones include norfloxacin (NG), ciprofloxacin (CF), orofloxacin (OF).

contrast, less than one-third of diarrhea occurring in theUnited States is caused by bacterial enteropathogens. Sec-ond, the traveler typically becomes ill while away fromhome and customary medical help.

Recommendation

18. Only a small proportion of travelers venturing intohigh-risk areas should use chemoprophylaxis. All travelersshould bring along with them loperamide and an antimicro-bial agent and instructed on the self-therapy of diarrhea thatmay occur.

Prophylaxis and self-therapy for travelers' diaruhea.Chemoprophylaxis is not generally recommended for mostotherwise healthy persons during travel to high-risk areasunless the traveler has an important predisposing illness orif the purpose of the trip is particularly important (27). Ifpreventive medication is requested by the traveler, bismuth

AJG - VoL 92, No. I I, 1997

subsalicylate is recommended (67). For patients with certainserious underlying illnesses (e.9., AIDS, inflammatorybowel disease, hypochlorhydria induced by prior gastricsurgery, or treatment with a proton inhibitor such as ome-prazole, diabetes mellitus on insulin, and systemic malig-nancy) prophylactic antimicrobials may be used duringtravel to high-risk areas. Intermittent use of H, antagonistsfor gastroesophageal disease does not appear to predisposeto travelers' diarrhea and is not an indication for use ofprophylaxis. In the settings in which antimicrobial prophy-laxis is considered, the uncommon risk of developing anadverse reaction to therapy including a superinfection maybe outweighed by beneficial effect of the drug in preventingdiarrhea. For a subset of persons, a trip is so important thata short (6-10 h) illness would not be tolerated. For thesepersons an antimicrobial would be more advisable in viewof its greater protective efficacy (Table 6). For all travelersexercising care about what one eats or drinks is the comer-stone of prevention of i l lness (101). Moist foods served atroom temperature are generally unsafe including many buf-fet items and lealy green vegetables. In Table 6 the safefoods are identified.

While TMP/SMX is currently the most cost effectivetreatment for diarrhea among U.S. persons visiting the in-terior of Mexico as studied in Guadalajara (102) during therainy season (our warmer months), for other regions and drywintertime in the interior of Mexico, TMP-resistant bacteriacan be expected (103, 104). The quinolones are the recom-mended treatment in these settings (27). In Figure l, asuggested management strategy for travelers' diarrhea isprovided. When dianhea persists after quinolone therapy,the illness should be treated as persistent diarrhea in non-travelers (see Tables I and 3).

THE IMMUNOCOMPROMISED PATIENT

The patient with HIV infection and reduced CD 4 count(<200/mm3) and other persons with altered immunity (e.g.,patients after organ transplantation or when cancer chemo-therapy is being given) are at special risk of developingopportunistic enteric infection. The causes of diarrhea inpatients with AIDS have been evaluated and found to bevaried. The most common definable agents are the parasiticorganisms including C. parvum, L belli, Cyclospora, Mi-crosporidia, the bacterial enteropathogens S. enteritidis,Campylobacter, Shigella spp, and Ml,cobacterium-avium-intracellulare, and the viral pathogens cytomegalovirus,herpes simplex, adenovirus, and HIV itself.

Recommendation

19. Immunocompromised patients with diarrhea shouldreceive a limited evaluation followed by full evaluation onlyfor patients failing to respond to specific or symptomatictreatment.

AJG

Mtacutepromdetaifor trundewhermentmanltime.shouthe ein vithe cfor Idian

PrAC(Sout

RrAver

6

l 0

l l

t2

l 3

AJG - November 1997

Mmtagement of the immunocomltromised patient withacute diarrltea. The initial evaluation of the immunocom-promised host with diarrhea is given in Figure 2. A moredetailed evaluation is not indicated in view of the low yieldfor treatable etiologic agents when an expensive work-up isundertaken. As in infections outside the gut in these patients,when a treatable enteropathogen is identified, curative treat-ment characteristically requires prolonged treatment, and inmany of the cases suppressive therapy is required fbr life-time. When curative therapy is being attempted, stoolsshould be obtained I wk after stopping medication to verifythe eradication of the organism. If the organism is present,in vitro susceptibility should be perfbrmed if f'easible, andthe drug begun again for a longer period of time, possiblyfor lif'etime. In Table 7 specific management of inf-ectiousdiarrhea in immunocompromised patients is outl ined.

ACKNOWLEDGMENT

Prepared for The American College of Gastroenterology,ACG Practice Parameters Guidelines Committee. 4900 BSouth 31st Street, Arlingron. Virginia 22206-1656

Reprint requests and correspondence: Dr. H.L. DuPont, 6720 BertnerAvenue. MC I 164, Houston. TX 77030.

REFERENCES

l. Carthwright W, Archer D, Kvenberg J. Estimates of incidence andcost of intestinal inf'ectious diseases in the United States. publicHea l t h Rep 1988 ; 103 : 107 -15 .

2. Feldman R, Banatvala N. The fiequency of culturing stools fiomadul ts wi th d iarrhoea in Great Br i ta in. Epidemiol Inf 'ect 1994: l l3:4 1 4 .

3. Gangarosa R, Glass R. Lew J, et a l . Hospi ta l izat ions involv ing gas-troenteritis in the United States 1985: The special burden of thedisease among the elder ly. Arn J Epidemiol 1992;135:281-90.

4. Lew J. Glass R, Gangarosa R, et al. Diarrheal deaths in the UnitedStates 1979 through l9tl7: A special problern fbr the elderly. JAMA1991 :265 :3280 4 .

5. Class R. Lew J, Gangarosa R, et al. Estimates of morbidity andniortality rates fbr diarrheal diseases in American children. J pediatrl 9 9 l r l l 8 : S 2 7 - 3 3 .

6. Faruque A. Mahalanabis D, Islam A, et al. Common diarrhea patho-gens and the risk ofdehydration in young children with acute waterydiarrhea: A case-control study. Arn J Trop Med Hyg 1993149:93 100.

7. Oldf ie ld E. Bourgeois A, Omar A-K, et a l . Empir ical r reatment ofShigella dysentery with trimethoprirn: Five-day course vs. singledose. Arn J Trop Med Hyg 1987;37:616-23.

ll. DuPont H. Review article. Inf'ectious diarrhoea. Alirnent PharmacolThe r 1994 : t i : 3 -13 .

9. Ericsson C. Patterson T. DuPont H. Clinical presentation as a guideto therapy fbr t ravelers ' d ianhea. Am J Med Sci 19t37;29.1:91-6.

10. Par ish K, Chaprnan W, Wi l l iams L. Ischemic col i t is : An ever-chang-i ng spec t rum. Am Su rg l 99 l t 57 :118 .

I 1. Thuluvath P. McKendrick M. Salmonellu and complications relatedto age - ShefTreld exper ience. Q J Med l98U;67: ,197-503.

I 2. DuPont H, Marshall G. HlV-associated diarrhoea and wasting. Lancet1995:3-16:3-52- 6.

13. Quinn T. Stamnr W. Goodel l S. et a l . The polyrnicrobia l or ig in ofintest inal intect ions in homosexual men. N Engl J Med 1983;309:-576 82.

l.{. DuPont H. How saf'e is the fbod we eat? JAMA 1992:268:3240.15. Hedberg C, MacDonald K, Osterholm M. Changing epidemiology of

lbod-borne disease: A Minnesota perspect ive. Cl in Inf 'ect Dis 199,1;l 8 : 6 7 1 U 2 .

GUIDELINES ON ACUTE INFECTIOUS DIARRHEA IN ADULTS t913

16. Mac Kenzie W, Hoxie N, Proctor M. et al. A massive outbreak inMilwaukee of Cnptosporidlrzl infection transmitted through thepublic water supply. N Engl J Med 1994;331:161-1.

17. Haris J, DuPont H, Hornick R. Fecal leukocytes in diarrheal illness.Ann Intern Med 19'72l.76:696 703.

18. McNeely W, DuPont H, Mathewson J, et al. Occult blood versus fecalleukocytes in the diagnosis of bacterial diarrhea. A study of U. S.travelers to Mexico and Mexican children. Am J Trop Med Hyg 1996;55:430-3.

19. Oberhelman R, de la Cabada F, Garibay E, et al. EfTicacy of tri-methoprirn-sulfamethoxazole treatment of acute diarrhea in a Mexi-can pediatr ic populat ion. J Pediatr 1987; l l0:960-5.

20. DuPont H. Corrado M, Sabbaj J. Use of norfloxacin in the treatmentof acute diarrheal d isease. Am J Med 1987;82(suppl 68):79 83.

21. DuPont H. Ericsson C, Mathewson J, et al. Five versus three days ofofloxacin therapy fbr traveler's dianhea: A placebo-controlled study.Ant imicrob Agents Chemother 1992;36:87-91.

22. Voegtliin J. Stalder H, Hurzeler L, et al. Modified guaiac test mayreplace search for laecal leucocytes in acute infectious diarrhoea.Lancet l9f l3:2:120.1.

23. DuPont H. Nonfluid therapy and selected chemoprophylaxis of acutedianhea. Am J Med 1985;78(suppl 68):81-90.

24. DuPont H. Ericsson C, Robinson A, et al. Current problems inantimicrobial therapy fbr bacterial enteric inf'ection. Anl J Med 1987;U2(suppl 4A):324-8.

25. DuPont H, Reves R, Galindo E, et al. Treatrnent of travelers' diarrheawith trirnethoprim/sulfamethoxazole and with trimethoprim alone.N Engl J Med 1982:307:841-4.

26. Ericsson C, Johnson P, DuPont H, et al. Ciprofloxacin and tri-nrethoprim/sulfamethoxazole as initial therapy for acute travelers'diarrhea: A placebo-controlled randomized trial. Ann Intern Med1 9 8 7 r 1 0 6 : 2 1 6 2 0 .

27. DuPont H, Ericsson C. Prevention and treatment of traveler's diar-rhea. N Engl J Med 1993;32t3:1821-1 .

21l3. Jokipii L, Pohjola S, Jokipii A. Cryptosporidirrn: A frequent findingin pat ients wi th gastro intest inal symptoms. Lancet 1983;2:358-61.

29. Jokip i i A, Jokip i i L. Prepatency ofgiardiasis. Lancet 1977; l :1095 7.30. Jokipii L, Jokipii A. Timing of symptoms and oocyst excretion in

human cryptospor id iosis. N Engl J Med 1986;315:1643-7.31 . Cohen R, Kalser M, Arteaga I, et al. Microbial intestinal flora in acute

diarrheal d isease. JAMA 1967:201 :835-40.32. Manabe Y, Vinetz J, Moore R, et aI. Clostridiun diJficile colitis: An

ef l lc ient c l in ical approach to d iagnosis. Ann Intern Med 1965;123:835-40.

33. Bhan M, Rai P, Levine M, et al. Enteroaggregative Escherichia coliassociated with persistent diarrhea in a cohort of rural children inIndia. J Inf 'ect Dis 198911-59:1061-4.

34. George W, Nakata M. Thompson J, et al. Aeromonas-related dianheain adul ts. Arch Intern Med 1985:145:220'7- l l .

35. Koplan J, Fineberg H. Fenaro M, et al. Value of stool cultures. Lancet1980 :2 : . 113 6 .

36. Guenant R, Shields D, Thorson S, et al. Evaluation and diagnosis ofacute inf 'ect ious diarrhea. Am J Med 1985:78:91-8.

37. Siegel D. Edelstein P, Nachamkin I. Inappropriate testing 1br diar-rheal diseases in the hospital. JAMA 1990;263:979-82.

38. Reed S. Amebiasis: An update. Cl in Infect Dis 1992;14:385-93.39. Jokipii L, Pohjola S, Jokipii A. Cryptosporidiosis associated with

traveling and giardiasis. Gastroenterology 1985;89:838-42.40. Connor B, Shlim D, Scholes J, et al. Pathologic changes in the small

bowel in nine patients with diarrhea associated with a coccidia-likebody. Ann Intern Med 1993:119:317 82.

.11. Wright R, Spencer H, Brodsky R, et al. Giardiasis in Colorado: Anepidemiologic study. Arn J Epidemiol 1977;105:330-6.

.12. Pickering L, Woodward W, DuPont H, et al. Occurrence of Giardiuktmblia in children in day care centers. J Pediatr 1984;104:522-6.

43. Tangermann R, Gordon S. Wiesner P. et al. An outbreak of crypto-spor id iosis in a day-care center in Georgia. Am J Epidemiol l99l ;133 : , 171 -6 .

.1,1. RendtorfT R. The experirnental transmission of human intestinal pro-tozoan parasites. ll. Giardia Iumblia cysts given in capsules. Am JHyg 1954159:209 220.

45. DuPont H, Chappell C, Sterling C. et al. The infectivity of Cr,-pto-sporitlium parvurr in healthy volunteers. N Engl J Med 199-51332:tt55-9.

1914 DUPONT

.16. Kotler D. Francisco A, Clayton F, et al. Small intestinal injury andparasi t ic d iseases in AIDS. Ann Intem Med 1990;113:444-9.

17. Lopez C, Dykes A, Juranek D, et al. Waterborne giardiasis: Acornrnunity-wide outbreak of disease and a high rate of asymptomaticinf'ection. Am J Epidemiol l980ll 12:495-501 .

48. Ravdin R. Amebiasis. Cl in Infect Dis 1995;20: lz t53 66.49. Garcia L. Shum A, Bruckner D. Evaluation ol new monoclonal

antibody combination reagent fbr direct lluorescence detection ofGiardia. Cysts and Cryprcsporidium oocytes in human fecal speci-rnens. J Clin Microbiol 1992l.30:3255-1.

50. Dagan R, Fraser D, El-On J, et al. Evaluation of an enzyme-immu-noassay fbr the detection of Cryptosporidian spp in stool specimensfion.r intants and young children in field studies. Am J Trop Med Hyg1995 :52 :13 ,1 8 .

51. RosotTJ. Sanders C, Sonnad S, et a l . Stool d iagnosis ofg iardiasisusing a cornmercially available enzyme immunoassay to detect Cl-zurlia-specific antigen 65 (GSA 65). J Clin Microbiol 1989:21:199'7-2002.

52. Alles A. Waldron M, Sierra L, et al. Prospective comparison of directimmunofluorescence and conventional staining methods fbr detectiono'i Giardia and Cnptosporidium spp in human fecal specimens. J ClinMicrobio l 1995:33: 1632-4.

53. Su C. Brandt L. Escherichia coLi Ol5'7:H7 inf'ection in hurnans. AnnIntern Med 1995:123:698 -1 14.

54. Morris J Jr. Black R. Cholera and other vibrioses in the United States.N Engl J Med 1985;312:343-50.

-55. Lee L. Taylor J, Carter G, et al. Yersinia enterocolitica 0:3: Anemerging cause of pediatric gastroenteritis in the United States.J I n f ec t D i s l 99 l : 163 :660 3 .

56. The management of acute dianhea in children. oral rehydration,maintenance. and nutritional therapy. MMWR 1992;.{l(October l6):r -20.

57. Schiller L. Sannta Ana C. Morawski S, et al. Mechanism of theantidiarrheal effect of loperamide. Gastroenterology 1984;86: 147580.

58. Johnson P. Ericsson C, DuPont H, et al. Comparison of loperamidewith bisrnuth subsalicylate fbr the treatment of acute travelers' diar-rhea. JAMA 1986'.225:1 5'7 -60.

59. DuPont H, Sanchez J, Ericsson C, et al. Comparative efficacy ofloperamide hydrochloride and bismuth subsalicylate in the manage-ment of acute diarrhea. Am J Med 1990;88(suppl 6.{ ) : l5S-195.

60. DuPont H, Hornick R. Adverse effect of Lomotil therapy in shigel-los is. JAMA 191 3:226:1 525- 8.

61. Murphy G, Bodhidatta L, Echeverria P, et al. Ciprofloxacin andloperamide in the treatment of bacillary dysentery. Ann Intern Med1993 : l l 8 : 582 -6 .

62. Cin'rolai N, Carter J, Morrison B, et al. Risk factors for the progres-sicrn of Esr'lrerir:hia t:oli 0157:H7 enteritis to hemolytic-uremic syn-drorne. J Pediatr 1990:l 16:582-92.

63. Cirnolai N, Morrison B, Carter J. Risk factors fbr the central nervoussystem rnanif'estations of gastroenteritis-associated hemolytic-uremicsyndrome. Pediatrics 1992:90:616-21.

6.1. DuPont H, Sullivan P, Pickering L, et al. Symptomatic treatment ofdiarrhea with bismuth subsalicylate among students attending a Mex-ican university. Gastroenterology 191'1t13:7 15 8.

65. Ericsson C. Evans D. DuPont H, et al. Bismuth subsalicylate inhibitsactivity of crude toxins of Escherichia coli and Vibrio cholerae.J Int'ect Dis 19'7'7:136:693-6.

66. Powell D. Tapper E. Morris S. Aspirin-stimulated intestinal electro-lyte transport in rabbit ileum in vitro. Gastroenterology 1979;76:t 429 -31 .

67. DuPont H, Ericsson C, Johnson P, et al. Prevention of travelers'dianhea by the tablet fbrmulation of bismuth subsalicylate. JAMAl98l :251:1341 -50.

68. Graham D, Estes M, Gentry L. Double-blind comparison of bismuthsubsalicylate and placebo in the prevention and treatment of entero-toxigenrc Escher ichia colr . Gastroenterology 1983;85: l0 l l -22.

69. SteinhofT M. Douglas R Jr, Greenberg H, et al. Bismuth subsalicylatetherapy of viral gastroenteritis. Gastroenterology 1980178: 1495-9.

70. DuPont H, Ericsson C, DuPont M, et al. A randomized open-labelcomparison of nonprescription loperarnide and attapulgite in thesymptomatic treatment of acute diarrhea. Am J Med 1990;t3tt(suppl6A ) :20S-23S.

71. Johanson J. Sonnenberg A. Efllcient management of diarrhea in the

AJG - Vol. 92, No. I 1, 1997

acquired immunodeficiency syndrome (AIDS): A medical decisionanalysis. Ann Intern Med 1990; l 12:942-8.

72. Mendelowitz P, Hoffman R, Weber S. Bisrnuth absorption and myo-clonic encephalopathy during bismuth subsalicylate therapy. AnnIn te rn Med 1990 : l l 2 : 140 -1 .

73. Simon D, Weiss L, Tanowitz H, et al. Light microscopic diagnosis ofhuman microsporidiosis and variable response to octreotide. Gastro-enterology l99l ;100:271 3.

74. Haltalin K, Nelson J, Hinton L, et al. Comparison of orally absorbableand nonabsorbable antibiotics in shigellosis: A double blind studywith ampicillin and neomycin. J Pediatr 1968:12:708-20.

75. DuPont H, Levine M, Hornick R, et al. Inoculum size in shigellosisand irnplications fbr expected mode of transmission. J Infect DisI 989; I 59: I I 26- 8.

76. Nelson J, Kusmiesz H, Jackson L, et al. Trimethoprim-sulfamethox-azole therapy for shigellosis. JAMA 1976;235:1239-43.

77. Tauxe R, Puhr N, Wells J, et al. Antimicrobial resistance ot ShigeLlaisolates in the USA: The importance of intemational travelers. J InfectD i s 1 9 9 0 1 1 6 2 : 1 1 0 7 l l .

78. Chalker R, Blaser M. A review of human salmonellosis: III. Magni-tude of Salmonel1a infection in the United States. Rev Infect Dis1988 ;10 :1 I l - 24 .

79. Neill M, Opal S, Heelan J, et al. Failure of ciprofloxacin to eradicateconvalescent f'ecal excretion after acutg Salmonellosis: Experienceduring an outbreak in healthcare workers. Ann Intern Med 1991; I l4:195-9.

80. Cherubin C, Neu H, Imperato P, et al. Septicemia with nontyphoidSalmonella Medicine l9'7 4:53:365 I 6.

81. Nelson S, Granoff D. Salmonella gastroenteritis in the first threemonths of l i fe. Cl in Pediatr 1982:21: '709-12.

82. Gruenewald R, Blurn S, Chan J. Relationship between human immu-nodeficiency virus inf'ection and salmonellosis in 20- to 59-year oldresidents of New York Ci ty. Cl in Infect Dis 1994118:358 63.

83. Lockyer W, Feinfeld D, Cherubin C, et al. An outbreak of SaLmonellaenteritis and septicemia in a populaton of uremic patients. Arch InternMed 19801140:943-5.

84. Han T, Sokal J, Neter E. SalmoneLlosis in disseminated malignantdiseases. N Engl J Med 196l;276:1045-52.

85. Samra Y, Shaked Y, Maier M. Nontyphoid salmonellosis in renaltransplant recipients. Rev Infect Dis 1988;8:431-40.

86. Mandal B, Brennand J. Bactaeremia in salmonellosis: A 15 yearretrospective study fiom a regional infectious diseases unit. Br Med J1988:297:1242 3.

87. Kuschner R, Trofa A, Thomas R, et al. Use of azithromycin for thetreatment of campylobacter enteritis in travelers to Thailand, an areawhere ciprofloxacin resistance is prevalent. Clin Int'ect Dis 1995;21:5 3 6 4 1 .

88. Pai C, Gillis FE, Marks M. Erythrornycin in treatment of canrpt,'lobacter enter i t is in chi ldren. Am J Dis Chi ld 1983:137:286 8.

89. Petruccelli B, Murphy G, Sanchez J, et al. Treatment of travelers'diarrhea with ciprofloxacin and loperamide. J Inf'ect Dis 1992;165:557-60.

90. Rautelin H, Renkonen O, Kosunen T. Emergence offluoroquinoloneresistance in Campl;lobacter jejuni and Campvkfiacter coli in sub-jects from Finland. Antimicrob Agents Chemother l99l;35:2065-9.

9 I . Segreti J, Gootz T, Goodman L, et al. High-level quinolone resistancein clinical isolates of Campylobacter jejuni. J lnfect Dis 1992;165:661-10.

92. Mathewson J, Johnson P, DuPont H, et al. A newly recognized causeof travelers' diarrhea: Enteroadherent Escherichia call. J Infect Dis1 9 8 5 : l 5 l : 4 7 1 - 5 .

93. Cravioto A, Tello A, Navarro A, et al. Association of Escherichia coliHEp-2 adherence patterns with type and duration of diarrhoea. Lancet1991l.331:262-4.

9.1. Mathewson I, JiangZ, Zumla A, et al. HEp-2 cell adherent Esclie-richia coil in patients with human immunodetlciency virus-associateddiarrhea. I Int 'ect Dis 19951171:1636-9.

95. Yamamoto T, Echeverria P, Yokota T. Drug resistance and adherenceto human intestines of enteroaggregative Escherichia coli. J lnteclDis 1992:.165'.141-9.

96. DuPont H, Ericsson C, Mathewson J et al. Oral aztreonam a poorlyabsorbed yet eff'ective therapy for bacterial diarrhea in US travelers toMexico. JAMA 1992:267 :1932-5.

97. Walterspiel J. Ashkenazi S, Morrow A, et al. Effect ol subinhibitory

AJG _

colfe<

98. Padutulob

99. O:0 lsu

100. Pran1 !

AJG - November 1997

concentrations of antibiotics on extracellular Shiga-like toxin 1. In-f'ection 1992120:25-9.Pavia A, Nichols C, Green D, et al. Hemolytic-uremic syndromeduring an outbreak of Escherichia coli Ol57:H1 inf'ections in insti-tutions for mentally retarded persons: Clinical and epidemiologicobservar ions. J Pediatr 19901116:544 51.Ostroff S, Kobayashi J, Lewis J. Infections with Escherichia coli0157:'H'7 in Washington State: The first year of statewide diseasesurvei l lance. JAMA 19891262:355-9.Prouix F, Turgeon J, Delage G, et al. Randomized controlled trial ofantibiotic therapy for Escherichia coli 015'/:Hl enteritis. J Pediatr1992:121:299 -303 .

GUIDELINES ON ACUTE INFECTIOUS DIARRHEA IN ADULTS 1915

98.

99

100

101. Kozicki M, Steffen R, Schiir J. "Boil it cook it peel it or fbrget it;"does th is ru le prevent t ravelers 'd iarrhoea? Int J Epidemiol l9t i5 l l4:169

'72.

102. Bandres J, Mathewson J, Ericsson C, et al. Trimethoprim/sulfame-thoxazole remains active against enterotoxigenic E,scherichia t.oLi andShigella spp in Guadalajara, Mexico. Am J Med Sci 1992;303:289-9 1 .

103. Munay B. Resistance of Shigella, Salmonella and orher selectedenteric pathogens to antimicrobial agents. Rev Infect Dis 1986;8 ( supp l 2 ) :S172 81 .

104. Ericsson C, DuPont H. Travelers' diarrhea: Approaches ro preventionand therapy. Cl in lnfect Dis 1993;16:616-26.