Differentiating Kawasaki Syndrome From Microbial Infection

July 2008 INFECTIONS in MEDICINE 311

DIFFERENTIAL DIAGNOSIS

Kawasaki syndrome (KS), alsoknown as mucocutaneouslymph node syndrome, the

cause of which is unknown, is a com-mon vasculitis seen in the pediatricpopulation. Epidemiologically, it issimilar to an infectious disease inthat it has a seasonal occurrence andhas been implicated in epidemics.Clinically, it is a vasculitis that is un-responsive to antibiotics. Two im-portant aspects of KS are that it de-velops in healthy children and is themost common cause of acquired car-diac disease in the developed world.In fact, chronic cardiac complicationsdevelop in 20% to 25% of children inwhom KS goes untreated. The mostcommon and dangerous cardiaccomplication is coronary artery di-

latation, which may result in ruptureand exsanguination. Children alsomay experience a prolonged courseof arthritis, arthralgias, and cramp-ing abdominal pain. Therefore, it isvery important to diagnose KS withcertainty, treat it appropriately, andfollow up patients meticulously.

DIFFERENTIAL DIAGNOSESSince there is no specific diagnostictest for KS, the diagnosis hinges onmeeting 4 of the 5 criteria as well asthe presence of fever for 5 days. Be-cause the diagnostic features oftenhave different presentations, the di-agnosis of KS may be difficult evenfor the experienced clinician. Thus, itis always important to consider thedifferential diagnosis when con-

fronted with a child in whom KS issuspected. KS should be consideredin any child with fever for more than5 days, especially if the child has arash and nonpurulent conjunctivitis.The differential diagnosis of KS is ex-tensive and includes bacterial andviral infections and rheumatologicaldiseases, among other causes.

Bacterial infectionsBacterial infections that play into the differential diagnosis includescarlet fever; staphylococcal scaldedskin syndrome (SSSS); toxic shocksyndrome (TSS); Rocky Mountainspotted fever (RMSF) and otherforms of rickettsial infection, such as typhus; leptospirosis; rat-bitefever; and Yersinia pseudotuberculosisinfection.1-10

Scarlet fever is a syndrome thatresults from erythrogenic exotoxin Aproduction by group AStreptococcus.It is similar to KS in that it causesdesquamation with time, includingperiungual desquamation. Infectionalso can cause cervical adenopathy,exudative tonsillitis, and strawberrytongue.

A few distinguishing featureshelp differentiate scarlet fever fromKS. Although a desquamating rash isa characteristic of both diseases, therash associated with scarlet fevermay become blanched. It is diffuselyerythematous, resembling a sun-burn, and is rough with a sandpaper

Dr Ryan is chairman of pediatrics for the Geisinger Health System in Danville, Pa. Dr Keck andDr Musso are residents and were medical students at Geisinger at the time the manuscript wasprepared. Dr Capp is a medical/pediatric resident at Geisinger.

Differentiating Kawasaki SyndromeFrom Microbial InfectionMichael E. Ryan, DO, Terrah Keck, DO, Mary Frances Musso, DO, and Kimberly C. Capp, DO

Kawasaki syndrome (KS) is a serious disorder affecting children aged 1 to 8 years. It mimics a range of other diseases of childhood. Diagnosis is based on physical examination findings coupled with the exclusion of other causes. To provideoptimal care for patients, it is important to be aware of the differential diagnoses of KS, which include bacterial and viralinfections; rheumatological diseases, which may be secondaryto infectious diseases; and other causes, such as antimicrobialdrug reactions. [Infect Med. 2008;25:311-316]

Key words: Kawasaki syndrome ■ Vasculitis ■ Hypersensitivity syndrome■ Febrile rash

IIM_07012008_00311.ps 6/16/08 10:56 AM Page 311

312 INFECTIONS in MEDICINE July 2008

KAWASAKI SYNDROME OR INFECTION? continued

texture. The rash is most intense onthe axillae and on the groin, ab-domen, and trunk. It generally ap-pears about 24 hours after the onsetof fever. It is first seen on flexor sur-faces of the extremities and becomesgeneralized in 24 to 48 hours. Pastiasign (Figure 1), nonblanching skinfolds, and circumoral pallor alsomay be noted. Cephalic to caudaldesquamation occurs about a weekafter the onset of rash. Finally, ifphysical findings are not sufficient todetermine a diagnosis, checking an-tistreptolysin O titers can be helpfulbecause levels may be elevated inpatients with scarlet fever.

SSSS is another toxin-mediateddisorder. It shares with KS the char-acteristics of a desquamating truncalrash and an erythematous, peeling,fissured “sunburst” rash around themouth. Unlike KS, SSSS is preceded

by an initial infection of the upperrespiratory tract. Another distin-guishing characteristic is that therash of SSSS usually spares thepalms, soles, and mucous mem-branes. The peeling is confined toareas around body orifices.

One to 2 days after the rash mani-fests, bullae may appear and exfoli-ate in sheets, which is referred to as apositive Nikolsky sign. Isolation ofstaphylococci from a site other thanthe blisters (eg, conjunctivae or na-sopharynx) or from the blood willaid in the diagnosis.

TSS is generally caused by Staphy-lococcus aureus. Similarities betweenTSS and KS include edema of theface, palms, and soles. In addition,desquamation of the skin 1 to 2weeks after illness onset, strawberrytongue, and bulbar conjuctival hy-peremia are present in both illnesses.

Distinguishing characteristics of TSSinclude shock, a widespread blanch-ing erythroderma eruption that ismost prominent on the trunk and ex-tremities, and possible subconjuncti-val hemorrhage (Figure 2). Approxi-mately 85% of TSS patients have Saureus isolated from their mucosa orwound sites, but isolation of the or-ganism is not required to make thediagnosis.

RMSF, a rickettsial infection, iscaused by the spirochete Rickettsiarickettsii. It is transmitted through thebite of a tick and is most prevalent inthe southeastern and central Missis-sippi valley regions of the UnitedStates, with North Carolina and Oklahoma having the highest in-cidence. Similarities to KS includefever, maculopapular rash with in-volvement of palms and soles, andconjunctival hyperemia. In contrastto KS, RMSF causes a peripherallydistributed eruption beginning onthe ankles, wrists, and forehead, inwhich the initial lesions may blanchand appear as small red macules thatrapidly progress to maculopapulesand finally to petechiae. The onset ofrash is preceded by a 3- to 7-day pro-drome of chills, fever, and severefrontal headache, malaise, andanorexia. Thrombocytopenia, hypo-natremia, elevated aminotransferaselevels, hyperbilirubinemia, leuko-penia, and coagulopathies mightemerge. Serum antibodies reactive toR rickettsii may be detected by indi-rect immunofluorescence assay, butdiagnostic levels might be unde-tectable until the second week aftersyndrome onset.

Febrile rickettsial infection in-cludes epidemic and murine typhus,and it is transmitted by fleas or liceharboring Rickettsia species. Similar-ities to KS include high fever, a mac-ulopapular petechial eruption, andcervical adenopathy. Some distin-guishing characteristics include a 4-to 6-day prodrome with high fever,

Figure 1 – The appear-ance of Pastia sign

helps distinguish scarlet fever from

other causes of rash.

Figure 2 – Subconjuncti-val hemorrhage may appear as a symptom

of toxic shock syndrome(TSS). Its appearance

helps distinguish TSS from Kawasaki

syndrome.

IIM_07012008_00312.ps 6/16/08 10:56 AM Page 312


KAWASAKI SYNDROME OR INFECTION?

chills, headache, and generalizedaches and pains. In addition, themaculopapular rash is often central-ly distributed, and neuroretinitismay be found. The indirect fluores-cent antibody test is often used toconfirm the diagnosis.

Leptospirosis is caused by thespirochete Leptospira interrogans,which is transmitted by dogs, swine,rodents, and contaminated water.Similarities to KS include a macu-lopapular rash with peripheraldesquamation, conjunctivitis, cervi-cal lymphadenopathy, and pharyn-gitis. One of the distinguishing char-acteristics is conjunctivitis with epi-scleral injection and uveitis that maybe unilateral or bilateral and usuallyinvolves the entire uveal tract. Therash is maculopapular to general-ized and may be petechial or pur-puric. Erythema nodosa also may benoted.

The anicteric form of leptospirosisis the most common form and is as-sociated with biphasic fever, myal-gias, and chills. Acalculous chole-cystitis or intense jaundice is occa-sionally seen in children. Laboratorystudies that may aid in the diagnosisinclude those that evaluate for leuko-cytosis, hematuria, proteinuria, azo-temia, and hyperbilirubinemia.

Rat-bite fever is a very rare syn-drome caused by either Spirillumminus or Streptobacillus moniliformis.Similarities to KS include intermit-tent fever, rash on the palms andsoles, and lymphadenopathy. It hasmany distinguishing characteristics,including a waxing and waning pat-tern of fever of 3 to 4 days’ durationalternating with afebrile periods last-ing 3 to 9 days; this cycle may persistfor weeks. The rash often develops 1to 8 days after fever onset. Laborato-ry tests may indicate leukocytosisand may yield false-positive resultsfor venereal diseases.

Y pseudotuberculosis is transmittedby the ingestion of incompletely

cooked pork, unpasteurized milk, orcontaminated well water or by indi-rect contact with infected animals.This bacterium causes a fever, rash,lymphadenitis, and conjunctivitissimilar to those seen with KS. Somedistinguishing characteristics of Ypseudotuberculosis infection includevarying degrees of fever, a scarlatini-form rash, and mesenteric adenitis(which may mimic acute appendici-tis). It is also associated with Pari-naud oculoglandular syndrome,which includes unilateral conjunc-tivitis with conjunctival granulomas,ptosis, preauricular adenopathy,photophobia, and external signs ofinflammation. Of interest, 75% of patients with clinically apparent Ypseudotuberculosis infection are chil-dren younger than 15 years.

Viral infectionsViral infections that are symp-tomatically similar to KS includeadenovirus infections as well asmeasles, German measles, roseolainfantum, erythema infectiosum,and mononucleosis.1,2,4-7

Adenovirus infections, like KS,are characterized by a persistenthigh fever, pharyngitis (Figure 3),conjunctivitis, cervical lymphade-nopathy, and rash.

Distinguishing characteristics ofadenovirus infections include sorethroat, rhinitis, and unilateral con-junctivitis that can include serousdischarge, subconjunctival hemor-rhages, and the formation of a gray-ish pink friable membrane on thepalpebral conjunctiva. The conjunc-tivitis also is associated with an itch-ing, burning, foreign-body sensationthat is not seen in KS.

The discrete generalized erythem-atous maculopapular rash of adeno-virus infections often appears whilethe child is febrile. Adenovirusesalso can cause right iliac fossa ab-dominal pain. Direct antigen testingor viral culture can be used to detectadenoviruses.

Measles, caused by the rubeolavirus, shares similarities with KS inthat it is characterized by swelling ofthe hands and feet, a maculopapularrash with desquamation, conjunc-

Figure 3 – Pharyngeal manifestations, including erythema, tonsillitis, and exudates, also areindicative of adenovirus infection.

IIM_07012008_00313.ps 6/16/08 10:56 AM Page 313



tivitis, and fever that persists for 5 to7 days. Some unique characteristicsof measles include a 3- to 4-day pro-drome of fever, conjunctivitis, cory-za, and severe cough. In contrast toKS, the conjunctivitis of measles isexudative. The brick-red rash ofrubeola starts on the face, the neck,and behind the ears; it then extendsdown the trunk and onto the ex-tremities. The rash is initially macu-lopapular and becomes more conflu-ent before it begins to fade after 3days, leaving behind a distinctivebrownish hue. This is often followedby a branny desquamation that doesnot involve the hands and feet. In ad-dition, Koplik spots on the buccalmucosa and a central, white coatingof the tongue with an erythematoustip and margins may be seen.

Similarities between KS and Ger-man measles, which is caused by therubella virus, include a maculopapu-lar rash, adenopathy, and fever. Ger-man measles is distinguished fromKS by a nonspecific prodrome offever, coryza, sore throat, arthralgias,and adenopathy that occurs 1 to 5days before exanthem onset, but thisis more common in adolescents andadults than in infants and youngchildren.

The rash is characterized by anondesquamating, pale pink, mor-billiform maculopapular eruptionthat begins on the face and neck andprogresses down the trunk to the ex-tremities. The rash is generalized in24 to 48 hours, lasts 1 day in eacharea, and fades rapidly. In additionto cervical adenopathy, postauricu-lar and occipital lymphadenopathyand arthralgias also may be noted.

Roseola infantum is caused byhuman herpesvirus 6 and usually oc-curs in children aged 6 to 36 months.It is characterized by persistent feverof 3 to 5 days’ duration, followed byrash and lymphadenopathy. Distin-guishing features of roseola infan-tum include an erythematous and

morbilliform rash that consists ofrose-colored macules appearing onthe neck, trunk, and buttocks andless frequently on the face and ex-tremities that begins as the feverabates. The mucous membranes areoften spared, and the rash resolves in1 to 2 days. Patients with roseola in-fantum also are at increased risk forfebrile convulsions. Laboratory stud-ies often show leukopenia.

Erythema infectiosum (also calledfifth disease) is caused by ParvovirusB19. Like KS, it is characterized byfever, adenopathy, and rash. UnlikeKS, a prodrome of malaise, pharyn-gitis, coryza, and fever precedes theillness. The characteristic “slappedcheek” rash generally follows about10 days later. In the second phase ofthe illness, the rash spreads to ex-tremities and becomes symmetrical,morbilliform, and lacelike or annularwith central clearing and is oftenmildly pruritic. It spares the mucousmembranes, palms, and soles. In itsfinal phase, the rash may remit andrecur for weeks with stress, exercise,or bathing. Complications of erythe-ma infectiosum include arthritis, he-molytic anemia, aplastic crisis, andnonimmune hydrops in the fetusand newborn.

Mononucleosis is caused by theEpstein-Barr virus (EBV). Similari-ties to KS include fever, rash, andcervical lymphadenopathy. Mono-nucleosis is typically characterizedby a triad of membranous tonsillitiswith or without exudates, cervicallymphadenopathy, and splenomeg-aly. The rash of mononucleosis israre (5% to 10% of patients with EBVinfection) and may appear as 2 dif-ferent exanthems. The first type ofexanthem is erythematous, macu-lopapular, and rubella-form and ismore prominent on the trunk andproximal upper extremities (occa-sionally it is seen on the face, fore-arms, and legs). This is the classic,non–antibiotic-related EBV rash. The

second type is an erythematous orcopper-colored ampicillin-associat-ed rash that begins on the trunk andspreads to the face and extremities.EBV infection can be diagnosed inthe clinic with the monospot test orwith specific EBV antibody tests.

Rheumatological diseasesA few rheumatological diseasescause symptoms similar to those ofKS, including Gianotti-Crosti syn-drome, Henoch-Schönlein purpura(HSP), and juvenile rheumatoidarthritis (JRA).8

Gianotti-Crosti syndrome, an in-fantile, papular acrodermatitis orig-inally associated with hepatitis Bsurface antigen that may occur afterviral infection, is caused by patho-gens such as EBV, cytomegalovirus,enteroviruses, and respiratory syn-cytial virus. Like KS, this syndromeincludes a desquamative rash andlymphadenopathy. The rash is char-acterized by a sudden eruption ofsymmetric, flat-topped, discrete,nonpuritic, skin-colored to erythem-atous papules on the malar face, ex-tremities, and buttocks (Figure 4)that spares the trunk, mucous mem-branes, and antecubital and poplitealfossae. The lesions then fade anddesquamate spontaneously within 2to 3 weeks but may remain for up to8 weeks. The lymphadenopathy isgeneralized and inguinal, and maxil-lary nodes can be enlarged for 2 to 3months after onset.

HSP is a systemic vasculitis withdeposition of IgA-containing im-mune complexes throughout thebody. Like KS, HSP is characterizedby fever; rapidly fading rash;swollen hands, feet, and periorbitalareas; arthritis; and abdominal pain.Unlike KS, HSP is classically de-scribed as intermittent purpura,arthralgias, abdominal pain, andrenal disease. HSP also may be pre-ceded by an upper respiratory tractinfection, mild fever, and headache.

IIM_07012008_00314.ps 6/16/08 10:56 AM Page 314


KAWASAKI SYNDROME OR INFECTION?

The initial lesions are symmetrical,blotchy, erythematous macules thatbecome urticarial and then purpuricwithin a day. The palpable purpuriclesions are seen on the buttocks, ex-tensor surfaces of extremities, back,scrotum and, occasionally, the face.

In a child younger than 2 years,edema of the scalp, hands, feet, andperiorbital tissues may develop be-fore the appearance of purpuric le-sions. Cutaneous hemorrhage maybe the sole manifestation of any at-tack, with arthralgia and arthritisnoted as a migratory, periarticularswelling of the knees and ankles. Pa-tients also may have colicky abdom-inal pain associated with vomitingand melena, and mild renal involve-ment with transient proteinuria,hematuria, and focal glomerular in-volvement. Abnormal findings onlaboratory tests include leukocytosis,thrombocytosis, and elevated eryth-rocyte sedimentation rate (ESR).Skin biopsy specimens show IgA,C3, and fibrin deposits.

JRA has characteristics similar tothose of KS, including lymphad-enopathy, rash, and high, spikingfevers. These fevers are dramatic,with sweats and chills, and tempera-tures often spike to 40°C (104°F) be-fore plunging to several degreesbelow normal (picket fence tempera-ture). The rash is a transient, evanes-cent, salmon-colored, nonpruriticrash that is primarily noticeable onthe chest and abdomen. It often ap-pears and disappears with the feverspikes. Many patients may initiallycomplain of mild sore throat andjoint symptoms that become a pro-gressively destructive arthritis pri-marily affecting the wrists. He-patosplenomegaly during the rash,anemia, and leukocytosis also maybe present.

Other differential diagnosesOther syndromes with characteris-tics similar to those of KS include

Stevens-Johnson syndrome (SJS),acrodynia, and convulsant hyper-sensitivity syndrome (CHS).6,9,10

SJS is a condition caused by a se-vere allergic reaction to drugs suchas sulfonamides, NSAIDs, andphenytoin and by infections such asthose caused by Mycoplasma pneumo-niae and herpes simplex virus. LikeKS, SJS is characterized by pharyngi-tis, fever, conjunctivitis, a maculo-papular rash involving the handsand feet, and hemorrhagic lips. Therash tends to be vesicular with crust-ing of edematous, erythematouseruptions involving the face, hands,and feet. Bullous erythema multifor-ma, with lesions that may slough off

as large pieces of skin, also may benoted. Stomatitis is an early and con-spicuous symptom, beginning withvesicles on the tongue, lips, and buc-cal mucosa (Figure 5). This later be-comes more severe and includespseudomembranous exudation, ex-cessive salivation, and ulcerations.Rhinitis with epistaxis and crustingof the nares also may be seen. Theconjunctivitis of SJS is bilateral andoften exudative. Laboratory testingmay indicate an increased ESR, al-though the ESR is not as high as thatseen in KS.

Acrodynia, also known as eryth-redema polyneuropathy or pink dis-ease, is caused by mercury poisoning

Figure 4 – The rash associated with Gianotti-Crosti syndrome is desquamative like that of Kawasaki syndrome. However, it is uniquely characterized by a sudden eruption of symmetrical, flat-topped,discrete, nonpuritic,skin-colored to erythema-tous papules on the malar face, extremities,and buttocks.

IIM_07012008_00315.ps 6/16/08 10:56 AM Page 315



that usually occurs in infancy. Simi-lar to KS, it is characterized bypainful swelling of the hands andfeet, a maculopapular rash, and irri-tability. Unlike KS, the erythema isblotchy and diffuse. Hands and feetalso may be cold, clammy, pink ordusky red, and pruritic. In addition,hemorrhagic puncta may be seen.Laboratory studies may demon-strate albuminuria, hematuria, andthe presence of mercury in urine.

CHS is a systemic reaction to an-ticonvulsant therapy. Like KS, it ischaracterized by fever, maculopapu-lar rash, and lymphadenopathy. Itincludes involvement of visceral or-gans, and fulminant hepatitis maydevelop. The lymphadenopathy isgeneralized. The rash usually begins2 to 8 weeks after the drug is begunand usually resolves when the drugis stopped. CHS may be followed byan eosinophilic colitis. Helpful find-ings from laboratory tests include

leukocytosis with eosinophilia and anormal ESR.

TREATMENTHigh-dose intravenous gamma glob-ulin (IVIG) (2 g/kg) as a single dosealong with high-dose aspirin (100mg/kg/d) for 10 to 14 days followedby low-dose aspirin (5 mg/kg/d)until acute inflammatory mediatorsreturn to normal is the standard ofcare for KS.1 This therapy has beenshown to reduce the incidence ofcoronary artery disease. The most se-rious complication associated with

KS is a coronary artery aneurysm.Aneurysms may lead to suddendeath, often within the first 30 daysafter the onset of KS. Usually IVIG isfollowed by a prompt end to mostsymptoms the child is having. On oc-casion, a second dose of IVIG is re-quired. Any child whose status doesnot improve after a second dose ofIVIG should be reevaluated. ❖

REFERENCES1. American Academy of Pediatrics. Summaries

of Infectious Diseases. In: Pickering LK, ed. RedBook: 2003 Report of the Committee on InfectiousDiseases. 27th ed. Elk Grove Village, IL: Ameri-can Academy of Pediatrics; 2006:412-415.

2. Darmstadt GL, Marcy SM. Erythematous mac-ules and papules. In: Long SS, Pickering LK,Prober CG, eds. Principles and Practice of Pedi-atric Infectious Diseases. 2nd ed. New York:Churchill Livingstone; 2003:432-434.

3. Gomez HF, Cleary TG. Yersinia species. In:Long SS, Pickering LK, Prober CG, eds. Princi-ples and Practice of Pediatric Infectious Diseases.2nd ed. New York: Churchill Livingstone; 2003:839-843.

4. Demmler GL. Adenoviridae. In Long SS, Pick-ering LK, Prober CG, eds. Principles and Practiceof Pediatric Infectious Diseases. 2nd ed. NewYork: Churchill Livingstone; 2003:1076-1080.

5. Maldonado YA. Rubeola virus. In: Long SS,Pickering LK, Prober CG, eds. Principles andPractice of Pediatric Infectious Diseases. 2nd ed.New York: Churchill Livingstone; 2003:1148-1155.

6. Cohen BA, ed. Pediatric Dermatology. 3rd ed.Philadelphia: Elsevier Health Sciences; 2005:161-200.

7. Davis HW, Michaels MG. Infectious disease. In:Zitelli BJ, Davis HW, eds. Atlas of Pediatric Phys-ical Diagnosis. 4th ed. St Louis: Mosby; 2002:396-454.

8. McIntire SC, Urbach AH, Londino Jr AV.Rheumatology. In: Zitelli BJ, Davis HW, eds.Atlas of Pediatric Physical Diagnosis. 4th ed. StLouis: Mosby; 2002:225-256.

9. Committee on Injury and Poison PreventionAmerican Academy of Pediatrics. Handbook ofCommon Poisonings in Children. Elk Grove Vil-lage, IL: American Academy of Pediatrics; 1994:216-219.

10. Le J, Nguyen T, Law AV, Hodding J. Adversedrug reaction among children over a 10-yearperiod. Pediatrics. 2006;118:555-562.

Figure 5 – Stomatitis isan early symptom of

Stevens-Johnson syndrome (SJS) and helps

distinguish SJS fromKawasaki syndrome.

Vesicles develop on the tongue, lips, and

buccal mucosa. Extensivestomatitis, crusting, anderythematous eruptions

are depicted here.

Therapeutic agents mentioned in this article

Ampicillin

Aspirin

Intravenous gamma globulin

Phenytoin

IIM_07012008_00316.ps 6/16/08 10:57 AM Page 316


CASE REPORT

Kawasaki syndrome (KS) is acommon vasculitis seen in the pediatric population. It

also is known as mucocutaneouslymph node syndrome, and thecause is unknown. Epidemiological-ly, it is similar to an infectious dis-ease; it has a seasonal occurrence andhas been implicated in epidemics.Clinically, it is a vasculitis unrespon-sive to antibiotics. KS develops inhealthy children and is the mostcommon cause of acquired cardiacdisease in the developed world.Chronic cardiac complications de-velop in 20% to 25% of children withuntreated KS, and the most commonand dangerous cardiac complicationis coronary artery dilatation. It may

result in rupture of the arteries andexsanguination. KS also may causechildren to experience a prolongedcourse of arthritis, arthralgias, andcrampy abdominal pain. Therefore,it is very important that KS is diag-nosed accurately and treated appro-priately and that clinicians are dili-gent about follow-up.

Case reportA4-year-old boy with a past medicalhistory of recurrent otitis media ini-tially presented with a complaint ofright ear pain and fever. Otitis mediaof the right ear was diagnosed, andamoxicillin therapy was begun. Fourdays later, the patient presented tohis primary care physician because

of persistent fever and enlargementof a left anterior cervical node. Theear pain had resolved, however. Thepatient was not taking in an ade-quate amount of fluids and thus wasmildly dehydrated.

The patient’s temperature was38.7°C (101.6°F). His right tympanicmembrane was slightly red, withfluid behind it. His tonsils were morethan 3 mm in diameter and were ery-thematous. He had an enlarged (2cm in diameter) left anterior cervicalnode. His skin turgor was decreasedand no rash was noted.

The patient was admitted withsuspected retropharyngeal abscessand for receipt of intravenous fluidsand further evaluation and therapy.Intravenous clindamycin was ad-ministered immediately. His initiallaboratory studies revealed a whiteblood cell count of 18,480/µL, with53% polymorphonuclear cells, 30%lymphocytes, 11% monocytes, and6% eosinophils. His hemoglobin lev-el was 10.8 g/dL, and the plateletcount was 370,000/µL. The C-reac-tive protein level was 27 mg/L.

A CT scan of the neck showed no retropharyngeal abscess, but pan-sinusitis, bilateral middle ear opac-ification, and bilateral cervical ad-enopathy were evident. Further laboratory studies revealed an anti-streptolysin O titer of less than 20IU/mL, alanine aminotransferaselevel of 8 U/L, and aspartate amino-

Dr Ryan is chairman of pediatrics for the Geisinger Health System in Danville, Pa. Dr Keck andDr Musso are residents and were medical students at Geisinger Health System at the time the article was prepared. Dr Capp is a medical/pediatric resident at Geisinger Health System.

Fever and Rash: Infection or Kawasaki Syndrome?Michael E. Ryan, DO, Terrah Keck, DO, Mary Frances Musso, DO, and Kimberly C. Capp, DO

Kawasaki syndrome (KS) is a common and serious disorderthat most often affects children aged 1 to 8 years but mimics a range of other diseases of childhood. Diagnosis of KS is based on physical examination findings coupled with the exclusion of other causes. To provide optimal care for patients,it is important to be aware of the differential diagnosis of KS. We report a case of a 4-year-old boy who presented with persistent fever and cervical lymphadenitis; later, mucousmembrane changes, rash, and conjunctival injection characteristic of KS developed. [Infect Med. 2008;25:320-322]

Key words: Kawasaki syndrome ■ Vasculitis ■ Hypersensitivity syndrome■ Febrile rash

IIM_07012008_000320RR.ps 6/24/08 10:36 AM Page 320


FEVER AND RASH

transferase level of 22 U/L. Test re-sults were negative for Epstein-Barrvirus and Parvovirus.

The patient remained febrile after3 days of intravenous clindamycintherapy. On the third hospital day, adiffuse macular rash developed. Thepatient’s lips were cherry red anddried, and his eyes were marked by a nonpurulent conjunctival injec-tion. Cervical adenopathy was un-changed, and no swelling of thehands or feet occurred.

Findings on echocardiographywere normal. The albumin level was2.6 g/dL, and the erythrocyte sedi-mentation rate (ESR) was 78 mm/h.Intravenous clindamycin therapywas stopped, and the patient wasgiven aspirin, 100 mg/kg/d, and in-travenous immunogammaglobulin(IVIG), 2 g/kg.

Over the next 18 hours, the pa-

tient’s condition rapidlyimproved. He became afe-brile, his rash resolved, andhis eyes and lips improved.He began to eat and drinknormally, and the cervicaladenopathy resolved. Hewas discharged home onthe fifth hospital day withinstructions to continuetaking aspirin and to fol-low up with the pediatricinfectious disease clinic in 1 week. At the follow-upvisit, peeling of the handsand feet was observed. Thepatient continued to dowell with no evidence ofcoronary abnormalities.

DiscussionThe diagnosis of KS is de-fined by the presence offever of at least 5 days’ du-ration and 4 of the 5 fol-lowing characteristics: con-junctival hyperemia, mu-cous membrane changes,distal extremity changes,

polymorphous exanthema, and cer-vical adenopathy. The fever charac-teristic of KS is high and spiking. It is not affected by antibiotics or anti-pyretics and resolves within 24 to 48 hours of IVIG therapy. The onset

of fever is considered the first day of the illness, from which all otherevents are measured.1

The conjunctival hyperemia seenin KS is characteristically nonexuda-tive and is most apparent in the bul-bar conjunctiva, with sparing of thelimbic region around the iris. Hyper-emia is noticeable a few days afterthe onset of fever and may persist for1 to 2 weeks if left untreated.

Effects of KS on mucous mem-branes are extensive and include dif-fuse erythema of the oral and pha-ryngeal mucosa without discrete ul-cerative lesions. The lips also may be injected, dried, or fissured, andthe patient may have a strawberrytongue (Figure 1).

The distal extremity changes ofKS are progressive and include ery-thema of the palms and soles alongwith indurative edema of the handsand feet. Periungual desquamation(Figure 2) of the fingers and toes fol-lows the swelling and is noticed ondays 10 to 25. Beau lines (transversegrooves across the fingernails) mayappear 2 to 3 months after diseaseonset.

The polymorphous exanthema ofKS is characterized by macules andpapules without any vesicle or bul-lae formation. It is prominent on thetrunk and extremities. In two-thirds

Figure 1 – These images illustrate the characteristic white(top) and red (bottom) strawberry tongue seen in Kawa-saki syndrome.

Figure 2 – Periungualdesquamation followsswelling in Kawasaki syndrome, as seen here on a child’s fingertips.

IIM_07012008_00321.ps 6/16/08 10:58 AM Page 321


FEVER AND RASH continued

of cases, it is accen-tuated in the perinealarea (Figure 3). Feverand sore throat mani-fest 2 to 5 days beforethe rash appears, andthe rash fades with-out residua within 10days.

The least promi-nent sign of KS is cervical adenopathy,which is present inabout 50% of childrenwith KS (other signsand symptoms dis-cussed are present inabout 90% of pa-tients). This cervical adenopathy ischaracterized by an erythematous,

indurated, nonsuppurative anteriorcervical node of at least 1.5 cm in di-ameter (Figure 4).

Additional characteristics of KSinclude constitutional symptoms (ir-ritability, fatigue, and anorexia), GIsymptoms (abdominal pain with orwithout vomiting and diarrhea, he-patic dysfunction with possible hy-drops of the gallbladder, and pancre-atitis), skeletal symptoms (arthritisor tympanitis with possible hearingloss), and lymphoid symptoms (ex-

udative tonsillitis).Laboratory abnormalities may in-

clude an elevated ESR or increases inother acute phase reactants, elevatedtransaminase levels, thrombocytosis,and hypoalbuminemia. ❖

REFERENCE1. American Academy of Pediatrics. Summaries

of infectious diseases. In: Pickering LK, ed. RedBook: 2003 Report of the Committee on InfectiousDiseases. 27th ed. Elk Grove Village, IL: Ameri-can Academy of Pediatrics; 2006:412-415.


Amoxicillin

Clindamycin

Intravenousimmunogammaglobulin

Figure 3 – Perineal accentuation of exanthema occurs in two-thirds of cases of Kawasaki syndrome.

Figure 4 – Cervical adenopathy, characterized by an erythematous, indurated, nonsuppurative anterior cervical node of at least 1.5 cm in diameter, is less prominent than other signs or symptoms of Kawasaki syndrome. Nevertheless, it occurs in up to 50% of affected children.

IIM_07012008_00322.ps 6/16/08 10:58 AM Page 322


The CDC’s Advisory Committee onImmunization Practices (ACIP) recommends that all persons older than 60 years be immunizedagainst the varicella-zoster virus that causes herpes zoster with a single dose of the live, attenuatedvirus vaccine Zostavax (Merck & Co, Inc, Whitehouse Station, NJ).Furthermore, it urges clinicians to offer the vaccine on the first available clinical encounter.

The recommendations, which willappear in Morbidity and MortalityWeekly Report (MMWR), can be accessed online.1 These recommen-dations are the first made by theACIP advocating use of a live, atten-uated virus vaccine for prevention ofherpes zoster. The hope—promisedby results of several clinical trialscited in the MMWR article—is thatroutine immunization of olderadults will significantly amelioratethe incidence of and morbidity associated with herpes zoster.

About 1 million cases of herpeszoster are diagnosed annually in theUnited States. Many of them will beassociated with postherpetic neural-gia (PHN), stated the ACIP, whichsummarized various other complica-tions in its recommendations state-ment. Included are ocular symptomsand sequelae (ie, herpes zoster ophthalmicus); varicella-zoster virusviremia; and serious, potentiallyfatal neurological conditions andviral dissemination to viscera thatcan occur in persons who are immunocompromised. Older per-sons (beginning at about age 50) areparticularly at risk for development

of herpes zoster, with 2 studies cal-culating that 50% of persons wholive to age 85 years will have experi-enced this condition and subsequentPHN.2,3

The Vaccine and VaccinationEach 0.65-mL dose of the zoster vaccine (when reconstituted andstored at room temperature for up to 30 minutes) contains a minimumof 19,400 plaque-forming units of theOka/Merck strain of varicella-zostervirus. This vaccine is appreciablymore potent than the varicella vaccine routinely used in children toprevent chickenpox (ie, Varivax, alsomanufactured by Merck & Co, Inc).It is administered subcutaneously tothe deltoid area. A single dose is allthat is required (booster doses arenot licensed for use).

Zoster vaccine should be stored ina freezer that maintains an averagetemperature of �15°C (5°F) or cold-er. Once reconstituted, the vaccineshould be used immediately: within30 minutes. After this time, the potency degrades. If unused, the reconstituted vaccine should be discarded.

The zoster vaccine is licensed foruse only in persons 60 years andolder. It is safe for use in persons receiving blood products. Personswho already have received immu-nization against varicella-zostervirus should not be re-immunized;however, the ACIP stated that concern regarding unintentional re-immunization in persons 40 yearsand older was slight because varicella vaccination did not begin in

the United States until 1995. TheACIP also noted that clinicians neednot question older patients about ahistory of chickenpox or conductserological testing for varicella immunity before administering thevaccine. Persons who have had anepisode of herpes zoster in the pastcan receive the vaccine, but it shouldnot be used to treat acute herpeszoster or PHN or be used as prophy-laxis against PHN. Precluding contraindications and precautionsrelated to health status, persons with chronic renal failure, diabetesmellitus, rheumatoid arthritis,chronic pulmonary disease, or otherchronic conditions can receive thevaccine.

Vaccine CoadministrationAlthough the zoster vaccine can be administered along with thetrivalent inactivated influenza vaccine without compromising theeffectiveness of either one, no dataare available on the effects of admin-istering the zoster vaccine with other vaccines that are routinely recommended for persons 60 yearsand older. Because simultaneous administration of most commonlyused live, attenuated and inactivatedvaccines, in general, has yet to be associated with impaired immuneresponse and has not been associat-ed with an increased rate of adverseevents,4 the zoster vaccine can be administered in the setting of otherindicated (inactivated) vaccines during the same office visit.

The ACIP reminded cliniciansthat when multiple vaccines are to

■ What You Need to Know About the ACIP’s Recommendations on Herpes Zoster Vaccination

[Infect Med. 2008;25:323-325]

IDAlert

IIM_07012008_00323.ps 6/16/08 11:00 AM Page 323


IDAlert continued

be administered during a single office visit, they should be adminis-tered to different anatomic sitesusing separate syringes. Althoughthe zoster vaccine can be adminis-tered at any time along with an inactivated virus vaccine, it shouldbe administered at least 4 weeks before or after administration of an-other live, attenuated virus vaccine.

Who Should Not Be VaccinatedBecause the risk of morbidity andmortality from herpes zoster isheightened in immunocompro-mised persons, eligible patientswho are scheduled to begin im-munosuppressive therapy shouldbe immunized at least 14 days (and preferably a month, according

to some experts5) before immuno-suppressive therapy is initiated.Otherwise, immunization is con-traindicated in immunocompro-mised persons, including those receiving immunosuppressivedrugs. However, exceptions andcaveats to these guidelines exist(Table).

Although a history of neomycin-associated contact dermatitis is nota contraindication to receiving thezoster vaccine, persons who have a history of anaphylactic reaction to any component of the vaccine, including neomycin, should not receive it.4 Pregnant women—whoare not in the target age group for herpes zoster immunizationanyway—also should not receivethe vaccine. Clinicians should be aware that the CDC and the vaccine’s manufacturer have established a registry to monitormaternal-fetal outcomes of preg-nant women who inadvertentlyhave been administered live, atten-uated varicella-zoster virus–typevaccines within a month of becom-ing pregnant. The telephone num-ber of the registry is 800-986-8999.

Persons who are receiving an antiviral medication, such as acyclovir, famciclovir, or valacy-clovir, should not be vaccinated inthe setting of active therapy. Rather,therapy should be discontinued forat least 24 hours before the zostervaccine is administered and at least14 days should elapse postvaccina-tion before antiviral therapy is resumed.4 Because antiviral agentsare active against herpesviruses,they could interfere with vaccine effectiveness.

Table – Who should not receive the zoster vaccine becauseof immunocompromise

Persons who should not receive the vaccine Exceptions

Persons with HIV/AIDS

Persons with leukemia, Persons in whom leukemia islymphomas, or other malignant in remission and who have not receivedbone marrow or lymphatic chemotherapy or radiation for at leastneoplasms 3 months4

Persons receiving immuno- At least a month should elapse betweensuppressive therapy, including discontinuation of the immunosuppres- high-dose corticosteroids sive therapy and zoster vaccination4,a

(> 20 mg/d of prednisone or equivalent) for 2 weeks or more

Persons with evidence of any Persons with impaired humoralunspecified cellular immunity, such as hypogammaglobu-immunodeficiency linemia or dysgammaglobulinemia

Persons undergoing hema- Clinical discretion may be applied,topoietic stem cell transplant but if a decision to vaccinate is made,

the vaccine should be administered at least 24 months after the transplant procedure

Persons receiving recombinant Vaccination should either occur weekshuman immune mediators before therapy is initiated or at leastand immune modulators 1 month after therapy is discontinued(adalimumab, infliximab, and etanercept)

a Patients receiving short-term (< 14 days) or low to moderate doses of corticosteroids (< 20 mg/d of prednisone or equivalent) or topical, intra-articular, bursal, or tendon injections or long-term alternate-day low to moderate doses of short-acting systemic corticosteroids can receive the zoster vaccine. In addition, low-dose methotrexate (< 0.4 mg/kg/wk), azathioprine (< 3 mg/kg/d), or 6-mercaptopurine (< 1.5 mg/kg/d) therapy is not a contraindication for administration of zostervaccine.

IIM_07012008_00324.ps 6/16/08 11:00 AM Page 324


Additional Notes andCaveatsClinicians are asked to documentall immunizations in the patient’smedical record, per the ACIP’sGeneral Recommendations on Immunization, published in 2006.4

The type of vaccine, the vaccine’smanufacturer, anatomic site androute of delivery, the date of vaccine administration, lot numberof the vaccine, and name of the administering facility should berecorded. In addition, to help avoidinadvertent re-immunization, patients should be given a copy ofthe document containing a recordof the vaccination.

If the zoster vaccine was inadvertently administered to achild, the ACIP recommends thatthe dose be counted as a singlevalid dose of varicella vaccine. Ifthe erroneously administered doseof zoster vaccine was given insteadof the first dose of varicella vaccine,a second dose of varicella vaccine is required. The ACIP has request-ed that such errors be reported tothe Vaccine Adverse Event Report-ing System (VAERS) whether ornot an adverse event occurs.

Conversely, if a clinician mistak-enly administers varicella vaccineto a person for whom the zostervaccine was indicated, no specificsafety concerns apply; however, thedose should be considered invalid,and the patient should immediate-ly be given a dose of zoster vaccine.If a delay in recognition of the erroroccurs, the zoster vaccine should bepromptly administered 28 daysafter the varicella vaccine wasgiven.

As with other vaccines, clinicallysignificant adverse events shouldbe reported to VAERS even if acausal relation to vaccination isquestionable. Clinicians are encour-aged to enter reports electronicallyat https://secure.vaers.org/VaersDataEntryintro.htm. The Web site of the VAERS is www.vaers.hhs.gov and the telephonenumber is 800-822-7967.

REFERENCES1. Centers for Disease Control and Prevention.

Prevention of herpes zoster. Recommen-dations of the Advisory Committee on Im-munization Practices (ACIP). www.cdc.gov/mmwr/preview/mmwrhtml/rr57e0515a1.htm?s_cid=rr57e0515_e. Accessed June 10,2008.

2. Brisson M, Edmunds WJ, Law B, et al. Epi-demiology of varicella zoster virus infection inCanada and the United Kingdom. EpidemiolInfect. 2001;127:305-314.

3. Schmader K. Herpes zoster in older adults.Clin Infect Dis. 2001;32:1481-1486.

4. Centers for Disease Control and Prevention.General recommendations on immunizationof the Advisory Committee on ImmunizationPractices. MMWR. 2006;55(RR-15):1-48.

5. Ihara T, Kamiya H, Torigoe S, et al. Viremicphase in a leukemic child after live varicellavaccination. Pediatrics. 1992;89:147-149.

IIM_07012008_00325.ps 6/16/08 11:00 AM Page 325


ImagesinInfectiousDisease

A45-year-old Hispanic man who acquired HIV in-fection in April 2003 presented with a 24-hour his-tory of worsening right lower quadrant pain ac-

companied by fever, decreased appetite, nausea, andvomiting. The pain was described as sharp, constant, andnonradiating. He denied any accompanying diarrhea,constipation, urinary frequency, dysuria, dyspepsia, re-flux symptoms, or previous episodes of abdominal pain.There was no history of recent travel. His current CD4+

cell count was 239/µL. In May 2003, he had a CD4+ cellcount nadir of 133/µL. His HIV RNA level has remainedundetectable at less than 50 copies/mL since starting first-line antiretroviral therapy in June 2003. Therapy consistsof coformulated zidovudine/lamivudine/abacavir andefavirenz. He has never had opportunistic infections orother major medical illnesses.

Physical examination revealed a temperature of 38.2ºC(101°F); pulse rate, 106 beats per minute; 10 out of 10 tenderness over the right lower quadrant with muscleguarding; and positive Rovsing, obturator, and psoassigns. Rectal examination showed tenderness over theright rectal vault.

Laboratory analysis demonstrated a white blood cellcount of 13 � 106/L (76% neutrophils). Findings from uri-

nalysis were normal. Findings on a CT scan of the ab-domen with contrast were normal with no evidence ofappendicitis. Because of persistent abdominal pain withnormal findings on the CT scan of the abdomen, colo-noscopy was performed. Figures 1, 2, and 3 are imagestaken at the level of the terminal ileum. Figure 1 shows animpacted pill in the appendiceal orifice. Figure 2 showsthe removal of the pill using forceps. Figure 3 shows mildresidual inflammation of the appendiceal orifice after re-moval of the impacted pill.

DiscussionAbdominal pain is a frequent presenting symptomamong HIV-positive patients seeking care at emergencydepartments. The incidence is estimated to be 12% to45%.1 In one retrospective study conducted in 1997 at SanFrancisco General Hospital, abdominal pain was the solecomplaint in 18% of the patients.1 The majority, however,reported other accompanying symptoms. The most com-mon were nausea/vomiting (58%), diarrhea (32%), andfever (21%).

Evaluation of the cause of abdominal pain rests onthorough history taking and physical examination. Thedifferential diagnosis of right lower quadrant pain in

Pill Impaction Mimicking Appendicitis in HIV-Positive Patient

Figure 2

Figure 1

IIM_07012008_00326.ps 6/16/08 11:01 AM Page 326


ImagesinInfectiousDisease

a non-immunocompromised, HIV-negative patient in-cludes appendicitis; diverticular disease and its compli-cations; intra-abdominal abscesses; kidney stones; sexu-ally transmitted diseases; enterocolitis; and intestinal obstruction from mechanical causes, such as strictures,masses, volvulus, and intussusception.

The differential diagnosis of right lower quadrant painamong HIV-positive patients also includes malignancies,such as lymphoma and Kaposi sarcoma, and opportunis-tic infections, such as those caused by mycobacteria (ie,Mycobacterium tuberculosis and Mycobacterium avium-intra-cellulare) and cytomegalovirus. Immune reconstitutionsyndrome associated with recent introduction of antiret-roviral therapy also may produce atypical presentationsof disease. The list of differential diagnoses becomes morelengthy and complex as the CD4+ cell count declines.

Workup of abdominal pain in a patient with HIV in-fection is the same as that for the noninfected patient.2

However, because of the vast differential diagnoses, theprobability of coexisting conditions caused by multiplepathogens,3 and atypical presentations of common disor-ders,4 radiological imaging such as CT and abdominalsonography should be used early in the assessment of ab-dominal pain in HIV-positive patients.2

Normal findings on a CT scan of the abdomen maylead the clinician to consider endoscopic procedures (ie,esophagogastroduodenoscopy or colonoscopy). Colo-noscopy, in this case, served both diagnostic and thera-peutic purposes.

Pill impaction that causes inflammation in the intes-

tinal mucosa is an unusual cause of abdominal pain. Theimpacted pill was removed endoscopically. The patientwas given metoclopramide, a prokinetic agent, for aweek. Pill impaction has not recurred. To our knowledge,this is the first reported case of pill impaction presentingclinically as an appendicitis mimic in a patient who isHIV-positive.

It is unknown what causes pill impaction among HIV-positive patients. Varying degrees of enteropathy devel-op in HIV-infected patients, especially those who haveadvanced disease. AIDS patients have been shown tohave delayed gastric emptying5 with impairment of bothintestinal absorption and permeability,6 factors that maypromote pill impaction with resulting local mucosal in-flammation. None of the medications that are used totreat HIV infection have been shown to significantly af-fect intestinal function (absorption, permeability, and in-flammation) or alter intestinal transit times.5

Recurrent pill impaction should warrant manometricstudies with or without gastric or small-intestine biopsiesto determine the cause and appropriate treatment. Pro-kinetic agents, such as metoclopramide and erythromy-cin, and also octreotide provide short-term, symptomatic relief. ❖

The case and images were submitted by Mauro Torno, MD, andMichael Shallman, MD, of Long Beach Health Department andSt Mary Medical Center in Long Beach, Calif.

REFERENCES1. Yoshida D, Caruso JM. Abdominal pain in the HIV infected patient. J Emerg

Med. 2002;23:111-116.2. Wilcox CM, Friedman SL. Gastrointestinal manifestations of the acquired

immunodeficiency syndrome. In: Feldman M, Sleisenger MH, Scharschmidt BF, eds. Sleisenger & Fordtran’s Gastrointestinal and Liver Disease. 6th ed. Phila-delphia: WB Saunders; 1998:387-409.

3. Slaven EM, Lopez F, Weintraub SL, et al. The AIDS patient with abdominalpain: a new challenge for the emergency physician. Emerg Med Clin NorthAm. 2003;21:987-1015.

4. Kuhlman JE, Fishman EK. Acute abdomen in AIDS: CT diagnosis and triage.Radiographics. 1990;10:621-634.

5. Sharpstone D, Neild P, Crane R, et al. Small intestinal transit, absorption, and permeability in patients with AIDS with and without diarrhoea. Gut.1999;45:70-76.

6. Bjarnason I, Sharpstone DR, Francis N, et al. Intestinal inflammation, ilealstructure and function in HIV. AIDS. 1996;10:1385-1391.

YOUR CONTRIBUTIONSARE INVITED!

Send slides or prints with a short description of what is shown. An honorarium will be awarded for each published photograph or set of photographs. Send to: Images, Infections in Medicine, 330 Boston Post Road,PO Box 4027, Darien, CT 06820-4027.

Figure 3

IIM_07012008_00329.ps 6/16/08 11:01 AM Page 329


Since the licensure of the heptavalent pneumococcalconjugate vaccine (PCV7) in 2000, the prevalence ofinvasive pneumococcal disease (IPD) among chil-

dren in the United States has decreased significantly. Theincidence of IPD caused by pneumococcal serotypes as-sociated with PCV7 among children younger than 5 yearsdecreased from 80 cases per 100,000 population in 1998 to1999 to 4.6 cases per 100,000 population in 2003.1 Variousstudies have demonstrated that nasopharyngeal colo-nization with pneumococcal serotypes covered by thevaccine also has decreased. However, several studies sug-gest that in some settings, these bacterial populationshave been replaced with Streptococcus pneumoniae sero-types not covered by the vaccine.2,3

Among the various pneumococcal serotypes (19A, 6A,3, and 15) that have been recognized as emerging threatsduring recent years, S pneumoniae 19A is, by far, the mostprominent. This serotype has been associated not onlywith the development of IPD but also with a high levelof resistance to multiple antibiotic classes.2,4,5

IPD surveillance in Massachusetts during 2001 to 2006identified a significant increase in cases caused by sero-types not covered by the PCV7.2 The percentage of infec-tions caused by serotype 19A increased from 10% during2001 to 2003 to 41% during 2005 to 2006. In Alaska,serotype 19Awas the cause of 28.3% of cases of IPD amongchildren younger than 2 years between 2004 and 2006.3

In addition to IPD, multiresistant serotype 19Aalso hasbeen linked to the development of otitis media. As evi-denced by Pichichero and Casey5 and by Jacobs and col-leagues6 in 2 separate studies, a significant challenge re-lated to this clinical situation is the high level of resistancethat many of these isolates have to antibiotics (such asamoxicillin, trimethoprim, macrolides, clindamycin, oral

cephalosporins, and ceftriaxone) most commonly used totreat otitis media in children.

The direct effect of PCV7 in serotype selection or re-placement is not yet fully understood. Although serotypeselection induced by lack of effectiveness of PCV7 againstserotype 19A is believed to be 1 of the main factors asso-ciated with serotype 19A proliferation, it is not the onlyfactor. Moore and colleagues7 have suggested that antibi-otic resistance, clonal expansion, and capsular switchingalso have contributed to the emergence of this serotype as the predominant cause of IPD in the United States. In addition, it is important to emphasize that according to a recent study by Hwa Choi and colleagues,8 multidrug-resistant 19A serotypes began to increase in South Koreabefore the introduction of PCV7.

The introduction and widespread administration ofPCV7 to children in the United States has led to a signif-icant decrease in the incidence of IPD. However, it is important for clinicians to recognize that because of multiple factors, multidrug-resistant serotype 19A hasemerged as a significant cause of pneumococcal disease.Clinicians need to consider the presence of this serotypein situations in which IPD is observed. In addition, infec-tion with S pneumoniae 19A should be considered in chil-dren with otitis media who fail to respond to antibiotictherapy. ❖

REFERENCES1. Centers for Disease Control and Prevention. Direct and indirect effects of rou-

tine vaccination of children with 7-valent pneumococcal conjugate vaccineon incidence of invasive pneumococcal disease—United States, 1998-2003.MMWR. 2005;54:893-897.

2. Centers for Disease Control and Prevention. Emergence of antimicrobial-resistant serotype 19A Streptococcus pneumoniae—Massachusetts, 2001-2006.MMWR. 2007;56:1077-1080.

3. Singleton RJ, Hennessy TW, Bulkow LR, et al. Invasive pneumococcal dis-ease caused by nonvaccine serotypes among Alaska native children with

Dr Estrada is professor of pediatrics, division of pediatric infectious diseases, University of South Alabama, Mobile.

PediatricBulletin

Streptococcus pneumoniae 19A: An Emerging ThreatBenjamin Estrada, MDUniversity of South Alabama, Mobile

[Infect Med. 2008;25:330, 334]

Key words: Heptavalent pneumococcal conjugate vaccine (PCV7) ■ Streptococcus pneumoniae 19A■ Multidrug resistance

continued on page 334

IIM_07012008_000330R.ps 6/23/08 12:50 PM Page 330


MENINGITIS

A septic meningitis refers to anonbacterial inflammation ofthe leptomeninges.1 Viruses

are the most common cause of asep-tic meningitis, and the most commonviruses that cause aseptic meningitisare enteroviruses. Drug-inducedaseptic meningitis is rare but proba-bly more common than the literaturewould suggest; therefore, it shouldbe included in the differential diag-nosis of aseptic meningitis, par-

ticularly if aseptic meningitis de-velops in association with the use ofNSAIDs or other offending drugs(Table 1) and if clinical recovery israpid following cessation of the drugor if results of viral studies are negative.

The pathogenetic mechanisms ofdrug-induced aseptic meningitis arenot fully understood, but 2 majormechanisms have been proposed.One proposed mechanism is that the

meninges are directly irritated by theintrathecal administration of drugs.The other is that the meninges are expressing an immunological hyper-sensitivity—most often a type 3 ortype 4 hypersensitivity reaction—tothe offending drug.2

An association between hyper-sensitivity reactions and underlyingcollagen-vascular disease or rheu-matological disease has been report-ed.1-10 Typically, the cerebrospinalfluid (CSF) profile in drug-inducedaseptic meningitis is that of a neu-trophilic pleocytosis accompaniedby a normal CSF lactic acid level and a variably elevated CSF proteinlevel.1,3 Patients who have drug-induced meningitis may have eo-sinophils present in the CSF (fewerthan 5%).

THE CLINICAL PICTUREPatients who have drug-inducedaseptic meningitis typically presentwith fever, headache, and nuchalrigidity. Signs and symptoms usual-ly appear within 24 to 48 hours afterdrug ingestion, but symptoms maynot occur until 2 years post-thera-py.2,6 Drug-induced aseptic meningi-tis may develop in a patient who ini-tially was able to tolerate the caus-ative drug.1,6

In patients who have drug-in-duced aseptic meningitis, the typicalCSF profile reveals a neutrophilicpleocytosis, with several hundred toseveral thousand white blood cells

A Differential Diagnosis of Drug-Induced Aseptic MeningitisClair Cascella, MD, Sara Nausheen, MD, and Burke A. Cunha, MD

Drug-induced aseptic meningitis should be included in the differential diagnosis of viral/aseptic meningitis. Cliniciansshould use historical clues in patients presenting with signs andsymptoms of viral meningitis to aid in the differentiation ofdrug-induced aseptic meningitis from other causes of asepticmeningitis. Viruses are the most common cause of asepticmeningitis, with enteroviruses being the most common amongviruses in cases presenting as aseptic meningitis. Ibuprofen iscurrently the most common cause of drug-induced asepticmeningitis. Drug-induced aseptic meningitis is a benign con-dition without long-term sequelae. The diagnosis of drug-induced aseptic meningitis is made by establishing a causal relationship between the use of the drug and the onset of signsand symptoms, supported by negative tests for infectious causes of symptoms and rapidity of resolution after the drug is discontinued. [Infect Med. 2008;25:331-334]

Key words: Drug-induced aseptic meningitis ■ Enteroviral meningitis

Dr Cascella is a first-year medical resident in the department of medicine at Winthrop-UniversityHospital in Mineola, NY. Dr Nausheen is a second-year fellow in the infectious disease division atWinthrop-University Hospital. Dr Cunha is chief of the infectious disease division at Winthrop-University Hospital and professor of medicine at State University of New York School of Medi-cine in Stony Brook.

IIM_07012008_00331.ps 6/16/08 11:04 AM Page 331


ASEPTIC MENINGITIS continued

per microliter; normal glucose lev-els; and variably elevated proteinlevels.1,2,4-7 Results of CSF Gram stainand cultures are negative, and lym-phocytic or eosinophilic pleocytosis

may occur. Drug-induced asepticmeningitis is reversible, with mostsigns and symptoms resolving with-in 24 to 48 hours after the drug is discontinued.2,4-7

DIFFERENTIAL DIAGNOSISThe differential diagnosis of asepticmeningitis is extensive and includesinfectious and noninfectious causes(Table 2).1-10 Drug-induced aseptic

Table 1 – Medications known to cause aseptic meningitisMedications Common Uncommon Rare

NSAIDs Ibuprofen Sulindac KetoprofenNaproxen Salicylatesa

Diclofenac TolmetinRofecoxib Piroxicam

Celecoxib

Antimicrobials Trimethoprim/ Sulfonamides Cephalosporinssulfamethoxazole Penicillin

Trimethoprim AmoxicillinAmoxicillin/clavulanateIsoniazidCiprofloxacinMetronidazolePyrazinamideValacyclovirIndinavirOrnidazole

Immunomodulating agents Monoclonal Azathioprine Levamisoleantibody OKT3 Efalizumab

Intravenous IgG InfliximabSulfasalazine

Intrathecal agents Metrizamide MethotrexateCytarabine GentamicinMethylprednisolone Iophendylate

acetate IopamidolIohexolHydrocortisoneBaclofenGadoliniumDiethylenediamine

pentaacetic acidSpinal anesthesia

Other Carbamazepine Monovalent mumps andrubella vaccine

Hepatitis B vaccineRanitidineFamotidineDexchlorpheniraminePhenazopyridineRadiolabeled albuminLamotrigineAllopurinolPentoxifyllineMethotrexate

a With serum levels ≈ 70 mg/dL.

Adapted from Hopkins S, Jolles S. Expert Opin Drug Saf. 20053; Marinac J. Ann Pharmacother. 1992.4

IIM_07012008_000332R.ps 6/23/08 12:50 PM Page 332


ASEPTIC MENINGITIS

Table 2 – Causes of acute aseptic meningitisCommon Uncommon Rare

Infectious causesBacterial Lyme disease Treponema pallidum Borrelia recurrentis

Leptospirosis infection infection (relapsing fever)Mycobacterium Mycoplasma pneumoniae Spirillum minus infection

tuberculosis infection infection (rat-bite fever)Subacute bacterial Rocky Mountain Nocardiosis

endocarditis spotted fever Actinomyces infectionParameningeal infection Brucellosis

(epidural subdural Ehrlichiosisabscess, sinus or ear infection)

Partially treated bacterial meningitis

Viral Echovirus infection Epstein-Barr virus Parainfluenza virus infectionCoxsackievirus infection infection Rotavirus infectionMumps Adenovirus infection Vaccinia virus infectionSt Louis encephalitis Cytomegalovirus infection West Nile virus infectionEastern equine encephalitis Varicella Human herpesvirus 6 infectionWestern equine encephalitis Measles Japanese B encephalitisCalifornia encephalitis Rubella Murray Valley encephalitisHerpes simplex virus

type 1 infectionHerpes simplex virus

type 2 infectionHIV infection Lymphocytic

choriomeningitis Poliovirus infection

Fungal Cryptococcosis CandidiasisCoccidioidomycosis BlastomycosisHistoplasmosis Aspergillosis

Sporotrichosis

Parasitic Angiostrongylus Cysticercosiscantonensis infection Trichinella spiralis infection

Toxoplasmosis

Noninfectious causesNeoplastic diseases Intracranial tumors

LymphomaLeukemiaMetastatic carcinomas

Systemic diseases Neurosarcoidosis Vogt-Koyanagi-HaradaBehçet disease syndromeSystemic lupus Sjögren syndrome

erythematosus Rheumatoid arthritis

Neurosurgical Neurosurgery (posteriorprocedures fossa syndrome)

Intrathecal agents

Medications See Table 1

Adapted from Chaudhry HJ, Cunha BA. Postgrad Med. 19911; Connolly KJ, Hammer SM. Infect Dis Clin North Am. 1990.8

IIM_07012008_00333.ps 6/16/08 11:04 AM Page 333


meningitis is a diagnosis of exclu-sion. It is important to obtain a histo-ry of medical disorders such as sys-temic lupus erythematosus, the mostfrequent underlying condition asso-ciated with drug-induced asepticmeningitis.7 It is also important tomake inquiries about recent vaccina-tions that may be implicated in thedevelopment of aseptic meningitis.2

Patients with enteroviral menin-gitis often present with an early neu-trophilic pleocytosis, although a shiftto lymphocytic pleocytosis usuallyoccurs within the first 48 hours.1,4,7,10

In normal hosts, enteroviral menin-gitis has a benign course, usuallylasting about 2 weeks. Recovery istypically characterized by decreas-ing frequency of headaches and stiffneck within the 2-week period.7,9

The condition may be diagnosed bypolymerase chain reaction testing ofthe CSF, by viral culture of throatand rectal specimens, or by serologi-cal tests for enteroviruses. CSF lacticacid levels readily differentiate bac-terial from viral meningitis.9

Quick resolution of symptoms isan important sign that distinguishesdrug-induced aseptic meningitisfrom viral meningitis, in which re-covery usually requires 10 to 14days.7 CSF glucose levels are usual-ly normal in drug-induced asepticmeningitis, which may help in dif-ferentiating it from bacterial menin-gitis in which glucose levels usuallyare low.4,6,7,10

Analysis of C-reactive protein(CRP) levels also may be helpful in

distinguishing bacterial from a drug-induced aseptic meningitis becauseCRP levels are usually highly elevat-ed in bacterial meningitis comparedwith drug-induced aseptic meningi-tis.2,5 The diagnosis of drug-inducedaseptic meningitis is made by estab-lishing a temporal relationship withthe administration of the drug andonset of clinical symptoms and rapidresolution of the syndrome afterdrug withdrawal.2-4,7,10

The most common cause of drug-induced aseptic meningitis isNSAIDs. The list of medications thatcause drug-induced aseptic menin-gitis continues to increase and cur-rently includes a wide variety ofmedications (Table 1).11-28

REFERENCES1. Chaudhry HJ, Cunha BA. Drug-induced asep-

tic meningitis: diagnosis leads to quick resolu-tion. Postgrad Med. 1991;90:65-70.

2. Jolles S, Sewell WA, Leighton C. Drug-inducedaseptic meningitis: diagnosis and manage-ment. Drug Saf. 2000;22:215-216.

3. Hopkins S, Jolles S. Drug-induced asepticmeningitis. Expert Opin Drug Saf. 2005;4:285-297.

4. Marinac J. Drug- and chemical-induced asepticmeningitis: a review of the literature. Ann Phar-macother. 1992;26:813-822.

5. Nettis E, Calogiuri G, Colanardi MC, et al.Drug-induced aseptic meningitis. Curr DrugTargets Immune Endocr Metabol Disord. 2003;3:143-149.

6. Rodríguez SC, Olguín AM, Miralles CP, Vilad-rich PF. Characteristics of meningitis caused by Ibuprofen: report of 2 cases with recurrentepisodes and review of the literature. Medicine(Baltimore). 2006;85:214-220.

7. Moris G, Garcia-Monco JC. The challenge ofdrug-induced aseptic meningitis. Arch InternMed. 1999;159:1185-1194.

8. Connolly KJ, Hammer SM. The acute asepticmeningitis syndrome. Infect Dis Clin North Am.1990;4:599-622.

9. Cunha BA. The diagnostic usefulness of cere-brospinal fluid lactic acid levels in central ner-vous system infections. Clin Infect Dis. 2004;39:1260-1261.

10. Kepa L, Oczko-Grzesik B, Stolarz W, Sobala-Szczygiel B. Drug-induced aseptic meningitisin suspected central nervous system infections.J Clin Neurosci. 2005;12:562-564.

11. Bonnel RA, Villalba ML, Karwoski CB, Beitz J.Aseptic meningitis associated with rofecoxib.Arch Intern Med. 2002;162:713-715.

12. Papaioannides DH, Korantzopoulos PG, GiotisCH. Aseptic meningitis possibly associatedwith celecoxib. Ann Pharmacother. 2004;38:172.

13. Wittmann A, Wooten GF. Amoxicillin-inducedaseptic meningitis. Neurology. 2001;57:1734.

14. Mateos V, Calleja S, Jiménez L, Suárez-Moro R.Recurrent aseptic meningitis associated withamoxicillin-clavulanic acid [in Spanish]. MedClin (Barc). 2000;114:79.

15. Fobelo MJ, Corzo Delgado JE, Romero AlonsoA, Gómez-Bellver MJ. Aseptic meningitis relat-ed to valacyclovir. Ann Pharmacother. 2001;35:128-129.

16. Mondon M, Ollivier L, Daumont A. Asepticmeningitis ornidazole-induced in the course ofinfectious endocarditis [in French]. Rev Med In-terne. 2002;23:784-787.

17. Kashyap AS, Kashyap S. Infliximab-inducedaseptic meningitis. Lancet. 2002;359:1252.

18. Mäkelä A, Nuorti JP, Peltola H. Neurologicaldisorders after measles-mumps-rubella vacci-nation. Pediatrics. 2002;110:957-963.

19. Ishihara O, Omata T. A case of famotidine-induced aseptic meningitis [in Japanese]. Rin-sho Shinkeigaku. 2000;40:48-50.

20. Lafaurie M, Dixmier A, Molina JM. Asepticmeningitis associated with intravenous admin-istration of dexchlorpheniramine. Ann Med In-terne (Paris). 2003;154:179-180.

21. Greenberg LE, Nguyen T, Miller SM. Suspectedallopurinol-induced aseptic meningitis. Phar-macotherapy. 2001;21:1007-1009.

22. Mathian A, Amoura Z, Piette JC. Pentoxifyl-line-induced aseptic meningitis in a patientwith mixed connective tissue disease. Neurolo-gy. 2002;59:1468-1469.

23. Peter JB. Ibuprofen meningitis. Neurology. 1990;40:866-867.

24. Lee RZ, Hardiman O, O’Connell PG. Ibupro-fen-induced aseptic meningoencephalitis.Rheumatology (Oxford). 2002;41:353-355.

25. Hawboldt J, Bader M. Intramuscular metho-trexate-induced aseptic meningitis. Ann Phar-macother. 2007;41:1906-1911.

26. Khan S, Sharrack B, Sewell WA. Metronida-zole-induced aseptic meningitis during Heli-cobacter pylori eradication therapy. Ann InternMed. 2007;146:395-396.

27. Kluger N, Girard C, Gonzalez V, et al. Efalizu-mab-induced aseptic meningitis. Br J Dermatol.2007;156:189-191.

28. Nesseler N, Polard E, Arvieux C, et al. Asepticmeningitis associated with lamotrigine: reportof two cases. Eur J Neurol. 2007;14:e3-e4.

ASEPTIC MENINGITIS continued

high levels of 7-valent pneumococcal conjugate vaccine coverage. JAMA.2007;297:1784-1792.

4. Farrell DJ, Klugman KP, Pichichero M. Increased antimicrobial resistanceamong nonvaccine serotypes of Streptococcus pneumoniae in the pediatricpopulation after the introduction of 7-valent pneumococcal vaccine in theUnited States. Pediatr Infect Dis J. 2007;26:123-128.

5. Pichichero ME, Casey JR. Emergence of a multiresistant serotype 19A pneu-mococcal strain not included in the 7-valent conjugate vaccine as an oto-pathogen in children. JAMA. 2007;298:1772-1778.

6. Jacobs MR, Good CE, Sellner T, et al. Nasopharyngeal carriage of respirato-ry pathogens in children undergoing pressure equalization tube placementin the era of pneumococcal protein conjugate vaccine use. Laryngoscope. 2007;117:295-298.

7. Moore MR, Gertz RE Jr, Woodbury RL, et al. Population snapshot of emer-gent Streptococcus pneumoniae serotype 19A in the United States, 2005. J In-fect Dis. 2008;197:1016-1027.

8. Hwa Choi E, Hee Kim S, Wook Eun B, et al. Streptococcus pneumoniae serotype19A in children, South Korea. Emerg Infect Dis. 2008;14:275-281.

PediatricBulletin continued from page 330

IIM_07012008_00334.ps 6/16/08 11:04 AM Page 334


Clinical MycologyUpdateDepartment Editor

Duane R. Hospenthal, MD, PhD

Over the past 2 decades, there has been an alarmingincrease in opportunistic fungal infections with anassociated rise in morbidity and mortality. This

trend has been attributed to the growing number of pa-tients who are immunocompromised because of bonemarrow or solid organ transplant, immunosuppressivedrugs, AIDS, and hematological malignancies. Advancesin trauma and critical care medicine that lead to longersurvival of more patients with immunocompromisingconditions also play a role.

Historically, the most common opportunistic mycoticinfections have been those associated with Candida albi-cans, Aspergillus fumigatus, and Cryptococcus neoformans. A number of fungal pathogens, including non-fumigatus Aspergillus species and other septate moulds, as well asmembers of the Zygomycetes class are emerging as im-portant causes of fungal disease.1,2 These infections arefrequently fatal. Early recognition of these pathogens iscritical to initiating prompt, appropriate therapy.

Because all antifungal drugs, including the neweragents, have gaps in coverage, most pathogens should beidentified at the species level when devising a therapeu-tic strategy. However, conventional reliance on cultureand histopathology for diagnosis of invasive fungal in-fections is time-consuming and frequently insensitive.For this reason, nucleic acid–based assays are gaining at-tention as potentially sensitive, accurate, and rapid testsfor the diagnosis of fungal infections.

Overview of molecular diagnosticsMethods that are currently used to diagnose fungal in-fections include direct observation (smears, histopathol-ogy); culture of clinical specimens; and antigen/antibodyassays for detecting the cell wall components galac-tomannan (GM) and �-glucan. More recently, polymerasechain reaction (PCR) amplification and its variants (in-cluding multiplex PCR, nested PCR, and real-time PCR)have been used to detect fungal pathogens, such as Can-dida and Aspergillus species, by amplifying genomic se-quences unique to each organism.3-6

Multiplex PCR provides increased sensitivity overstandard PCR by using multiple primer pairs per reactionto amplify more than 1 target sequence.4 Nested PCR, inwhich the original target sequence is amplified and thenused as a template for additional amplifications with asecond set of primers, is more specific than conventionalPCR. Nested PCR has been used clinically for the detec-tion of Candida species and Histoplasma capsulatum.7-10

Real-time PCR couples the assay with an amplificationproduct detection system (typically a fluorescent label)and has been used to detect and quantify DNA from sev-eral fungal pathogens, including Aspergillus species, Can-dida species, and C neoformans.4,11-15

Fungal nuclear, mitochondrial, and ribosomal genes,as well as RNA sequences, have been used as templatesin PCR and similar assays.16 The sensitivity of these as-says is enhanced when the target sequence has multiplecopies within the genome.14,16 Ribosomal targets possessboth sequences, which are highly conserved among thefungi and species-specific variable internal transcribedspacer regions. Recent studies have focused on 5.8S, 18S,and 28S ribosomal RNAand DNAgenes for the detectionof Candida and Aspergillus species.7,8,10,11,13,15-17

Once target sequences are amplified by PCR, the amplicons can be further characterized by other molecu-

Diagnostic Nucleic Acid Testing for Invasive Fungal InfectionsRupal Mody, MD and Michael Zapor, MD, PhD

[Infect Med. 2008;25:335-338]

Key words: Nucleic acid testing ■ Polymerase chain reaction ■ Diagnosis ■ Fungal infections

Dr Mody is an infectious disease fellow at the Walter Reed Army Med-ical Center in Washington, DC, and Dr Zapor is staff physician in the in-fectious disease service at the medical center. The views expressed here-in are those of the authors and do not reflect the official policy or posi-tion of the Department of the Army, Department of Defense, or the USGovernment. The authors are employees of the US Government. Thiswork was prepared as part of their official duties and, as such, is not sub-ject to copyright.

IIM_07012008_00335.ps 6/16/08 11:05 AM Page 335


Clinical MycologyUpdatecontinued

lar biology techniques, including restriction fragmentlength polymorphism analysis, nucleic acid sequencing,Southern and Northern blot analysis, electrophoretickaryotyping, and DNA microarray genotyping.4,5,18

Clinical application of molecular diagnosticsAlthough PCR assays can be used to detect any fungi,their clinical application has mostly been applied to thedetection of Candida and Aspergillus species.13,15,19,20-26

PCR assays for Candida are very sensitive and can detectDNA from as few as 10 organisms/mL of blood. Similar-ly, PCR assays for Aspergillus can detect DNA from 10 to100 conidia/mL of blood.24

Pryce and colleagues13 suggested that real-time PCRtesting, which can detect DNA quantity over time, mightbe useful for monitoring response to antifungal therapy.Klingspor and Jalal15 also found real-time PCR assays tobe both sensitive and specific for the detection of Candidaand Aspergillus species in clinical specimens. In theirstudy, clinical samples (blood, sputum, tissue, cere-brospinal fluid, bronchoalveolar lavage fluid, pleuralfluid, ascites, bile, and urine) from transplant recipientswith suspected invasive fungal infections were assayedby PCR. Of 1650 specimens assayed, 114 (6.9%) werePCR-positive for either Candida species (n = 86) or As-pergillus species (n = 28), whereas 62 (3.8%) were culture-positive for either Candida species (n = 57) or Aspergillusspecies (n = 5). Of the PCR samples positive for Candida,72% were identifiable to species.

Ahmad and colleagues7 found semi-nested PCR to be99% specific and more sensitive than culture in diagnos-ing candidemia. White and colleagues17 found that PCRtesting, when compared with latex agglutination and en-zyme-linked immunosorbent assay, detected Candida in-fection earlier, was more sensitive, and was comparablyspecific.

It has been suggested that environmental contami-nants might cause false-positive PCR results when usedin the diagnosis of fungal infections.17 This has been sub-stantiated in a study by Ljungman and colleagues21 inwhich blood samples from patients with leukemia wereassayed weekly by PCR for Cytomegalovirus and fungi.Real-time PCR results were positive in 9 samples takenfrom 8 of 35 patients (3 samples positive for Aspergillusand 5 samples positive for Candida, with 1 sample beingpositive for both).21 However, only 3 of the 4 samples inwhich Aspergillus species were detected correspondedwith suspicion for Aspergillus infection based on the pres-ence of pulmonary infiltrates on a chest CT scan.21

In the same study, of 3 cases of proven fungemia at-tributed to Candida species, in only 1 case was the bloodPCR-positive for Candida.21 Six samples from 5 patients

were PCR-positive for Candida species.21 Two samplescame from 1 patient who had bacteremia, 1 sample camefrom an asymptomatic patient, 1 sample each came from2 patients with fever of unknown origin, and 1 samplecame from a patient with candidemia. Hence, Candidawas never recovered in culture of specimens taken from4 of the 5 patients whose blood was PCR-positive for Candida.21 The PCR test results may represent true posi-tives, although, even in the absence of growth in culture,Candida species are not always recovered from the bloodof patients with candidiasis.

Another technique useful for the diagnosis of fungalinfections is fluorescence in situ hybridization (FISH).This assay uses fluorescein-labeled peptide nucleic acid(PNA) probes specific for the ribosomal RNA sequencesof C albicans.27 The FDA approved 1 of these assays, the Calbicans PNA FISH (AdvanDx, Inc, Woburn, Mass), in2004 for use in rapidly (ie, within 2.5 hours) differentiat-ing albicans from non-albicans Candida species isolatedfrom blood. In one study, the reported sensitivity andspecificity of this technique was 99% and 100%, respec-tively.27 The same company also manufactures a dualcolor C albicans/Candida glabrata FISH assay for simulta-neous identification of both organisms from blood cul-ture. The sensitivity and specificity of this dual assay aresimilar to those of the C albicans PNA FISH.28

The value of PCR in the diagnosis of invasive as-pergillosis has been evaluated in several studies (Table).The sensitivity, specificity, negative predictive value, andpositive predictive value vary widely between studies de-pending on the pretest probability of infection and typeof PCR used (eg, real-time vs nested).22-25 The low positivepredictive value of PCR assays when bronchoalveolarlavage specimens are used (range, 38% to 83.5%) likely re-flects the difficulty in distinguishing airway colonizationfrom infection.22-25

The sensitivity of PCR assays in detecting Aspergillusin serum varies from 40% to 92.3%, with improved sensi-tivity on serial testing.19,20,26,29 The low sensitivity of theassay described in some studies (especially in early infec-tion) might be attributed to transient fungemia, low-levelfungemia (ie, below the detection limits of the assay), anda short half-life of fungal DNA (because of rapid degra-dation or clearance). PCR testing fares better in detectingAspergillus invasion of tissue, such as lung tissue, with areported sensitivity of 100% in one study.26

Comparisons of the GM assay and real-time PCR assayin detecting Aspergillus infections have shown conflictingresults. Buchheidt and colleagues29 reported that the nest-ed PCR assay is more sensitive than the GM assay, where-as both Kawazu and colleagues30 and Costa and col-leagues31 reported the GM assay to be superior. If valid,

IIM_07012008_00336.ps 6/16/08 11:05 AM Page 336


the findings of the latter 2 studies might reflect greatershedding of fungal antigen relative to the presence of Aspergillus nucleic acid in the blood of patients withfungemia.

Merits and limitations of molecular diagnosticsWhen compared with culture and histopathology for thediagnosis of invasive fungal infections, PCR coupled withvarious hybridization techniques offers the potential ofenhanced sensitivity, specificity, and relative rapidity.Moreover, real-time PCR such as the LightCycler PCR de-tection system (Roche Applied Science, Mannheim, Ger-many) confers the advantage of quantifying fungal DNAand potentially might be used to monitor disease pro-gression as well as response to therapy.25,26 PCR testingalso permits identification of individual species and

strains as well as amplification of specific sequences forfurther study (eg, nucleic acid sequencing) and manipu-lation (eg, cloning).

However, there are drawbacks to using PCR testing inthe diagnosis of infection. The techniques for extractingand amplifying DNA are not currently standardized,and the reactions are expensive and vulnerable to false-positive results due to contamination. Most important,positive PCR results may not distinguish between con-tamination, colonization, or true infection, nor betweenDNA extracted from dead versus viable organisms.32

Lack of recovery of live organisms also removes the op-tion of performing antifungal susceptibility or retro-spective virulence or strain testing. Nevertheless, a greatpotential value of PCR derives from its negative predic-tive value.

Table – Reported performance of PCR in the detection of Aspergillus species from clinical specimens

Assay Samples (N) Source Sensitivity Specificity PPV NPV

Real-time PCR25 96 BAL, 43% NP NP NPblood

Real-time PCR29 > 1522 BAL, 63.6% 63.5% NP NPblood, other

Real-time PCR30 1251 Blood 55% 93% 40% 96%

Real-time PCR24 16 Blood 69.6% - 82.1% 80.4% - 91.1% NP NP(depending on (depending onprimers) primers)

Real-time PCR20 401 Blood 92.3% 94.6% 60% 99.3%

Real-time PCR35 1193 Blood 100% 65% NP NP

PCR-ELISA26 241 Blood, Proven infection: Proven infection: 100% 44% - 58%BAL, 40% - 100%; 100%; probabletissue probable infection: 100%

infection:44% - 66%

PCR-ELISA19 1205 Blood 63.6% 89.7% 63.6% 89.7%

Conventional 197 BAL 93.9% 94.4% 83.8% 98.1%PCR22

Conventional 68 BAL Proven infection: Proven infection: Proven ProvenPCR23 80%; probable 93%; probable infection: infection:

infection: 64% infection: 93% 38%; 99%;probable probable infection: infection:52% 96%

PCR, polymerase chain reaction; PPV, positive predictive value; NPV, negative predictive value; BAL, bronchoalveolar lavage; NP, not provided; ELISA, en-zyme-linked immunosorbent assay.

continued

IIM_07012008_00337.ps 6/16/08 11:05 AM Page 337


Clinical MycologyUpdatecontinued

ConclusionsStudies show PCR assays to be both sensitive and specif-ic in the diagnosis of infections caused by fungi such asAspergillus and Candida; sensitivity is typically greaterwith tissue than with blood. The diagnostic value of PCRtesting may be further enhanced in the appropriate clini-cal setting or when the test is done in conjunction withother tests, such as culture and the GM assay. When doneserially, quantitative PCR testing might be useful formonitoring disease progression or response to therapy,and potentially it could be used to differentiate coloniza-tion from infection. In addition, PCR testing has shownpromise in the diagnosis of infections caused by otherfungi, such as C neoformans, H capsulatum, and Pneumo-cystis jiroveci.9,33,34 However, the sensitivity of these assayspredispose them to false-positive results, and true-posi-tive results may not distinguish between contamination,colonization, and infection. Further clinical studies areneeded before PCR testing alone can be advocated for thediagnosis of fungal infection. ❖

REFERENCES1. Pfaller MA, Pappas PG, Wingard JR. Invasive fungal pathogens: current epi-

demiological trends. Clin Infect Dis. 2006;43(suppl 1):S3-S14.2. Nucci M, Anaissie E. Emerging fungi. Infect Dis Clin North Am. 2006;20:563-

579.3. O’Shaughnessy EM, Shea YM, Witebsky FG. Laboratory diagnosis of inva-

sive mycoses. Infect Dis Clin North Am. 2003;17:135-158.4. Iwen PC. Molecular detection and typing of fungal pathogens. Clin Lab Med.

2003;23:781-799.5. Alexander BD, Pfaller MA. Contemporary tools for the diagnosis and man-

agement of invasive mycoses. Clin Infect Dis. 2006;43(suppl 1):S15-S27.6. Yeo SF, Wong B. Current status of nonculture methods for diagnosis of inva-

sive fungal infections. Clin Microbiol Rev. 2002;15:465-484.7. Ahmad S, Khan Z, Mustafa AS, Khan ZU. Seminested PCR for diagnosis of

candidemia: comparison with culture, antigen detection, and biochemicalmethods for species identification. J Clin Microbiol. 2002;40:2483-2489.

8. Ahmad S, Mustafa AS, Khan Z, et al. PCR-enzyme immunoassay of rDNAinthe diagnosis of candidemia and comparison with amplicon detection byagarose gel electrophoresis. Int J Med Microbiol. 2004;294:45-51.

9. Bialek R, Feucht A, Aepinus C, et al. Evaluation of two nested PCR assays fordetection of Histoplasma capsulatum DNA in human tissue. J Clin Microbiol.2002;40:1644-1647.

10. Nazzal D, Yasin S, Abu-Elteen K. A rapid PCR-based method for identifica-tion of four important Candida species. New Microbiol. 2005;28:245-250.

11. Hsu M, Chen K, Lo H, et al. Species identification of medically importantfungi by use of real-time LightCycler PCR. J Med Microbiol. 2003;52(pt 52):1071-1076.

12. O’Sullivan CE, Kasai M, Francesconi A, et al. Development and validation ofa quantitative real-time PCR assay using fluorescence resonance energytransfer technology for detection of Aspergillus fumigatus in experimental in-vasive pulmonary aspergillosis. J Clin Microbiol. 2003;41:5676-5682.

13. Pryce TM, Kay ID, Palladino S, Heath CH. Real-time automated polymerasechain reaction (PCR) to detect Candida albicans and Aspergillus fumigatus DNAin whole blood from high-risk patients. Diagn Microbiol Infect Dis. 2003;47:487-496.

14. Bu R, Sathiapalan RK, Ibrahim MM, et al. Monochrome LightCycler PCRassay for detection and quantification of five common species of Candida andAspergillus. J Med Microbiol. 2005;54(pt 3):243-248.

15. Klingspor L, Jalal S. Molecular detection and identification of Candida and As-pergillus spp. from clinical samples using real-time PCR. Clin Microbiol Infect.2006;12:745-753.

16. Yeo SF, Wong B. Current status of nonculture methods for diagnosis of inva-sive fungal infections. Clin Microbiol Rev. 2002;15:465-484.

17. White PL, Archer AE, Barnes RA. Comparison of non-culture-based meth-ods for detection of systemic fungal infections, with an emphasis on invasiveCandida infections. J Clin Microbiol. 2005;43:2181-2187.

18. Gil-Lamaignere C, Roilides E, Hacker J, Muller FM. Molecular typing forfungi—a critical review of the possibilities and limitations of current and fu-ture methods. Clin Microbiol Infect. 2003;9:172-185.

19. Florent M, Katsahian S, Vekhoff A, et al. Prospective evaluation of a poly-merase chain reaction-ELISAtargeted to Aspergillus fumigatus and Aspergillusflavus for the early diagnosis of invasive aspergillosis in patients with hema-tological malignancies. J Infect Dis. 2006;193:741-747.

20. White PL, Linton CJ, Perry MD, et al. The evolution and evaluation of awhole blood polymerase chain reaction assay for the detection of invasive as-pergillosis in hematology patients in a routine clinical setting. Clin Infect Dis.2006;42:479-486.

21. Ljungman P, von Döbeln L, Ringholm L, et al. The value of CMV and fungalPCR for monitoring of acute leukaemia and autologous stem cell transplantpatients. Scand J Infect Dis. 2005;37:121-127.

22. Buchheidt D, Baust C, Skladny H, et al. Clinical evaluation of a polymerasechain reaction assay to detect Aspergillus species in bronchoalveolar lavagesamples of neutropenic patients. Br J Haematol. 2002;116:803-811.

23. Raad I, Hanna H, Huaringa A, et al. Diagnosis of invasive pulmonary as-pergillosis using polymerase chain reaction-based detection of Aspergillus inBAL. Chest. 2002;121:1171-1176.

24. White PL, Barton R, Guiver M, et al. Aconsensus on fungal polymerase chainreaction diagnosis? A United Kingdom-Ireland evaluation of polymerasechain reaction methods for detection of systemic fungal infections. J MolDiagn. 2006;8:376-384.

25. Spiess B, Buchheidt D, Baust C, et al. Development of a LightCycler PCRassay for detection and quantification of Aspergillus fumigatus DNA in clini-cal samples from neutropenic patients. J Clin Microbiol. 2003;41:1811-1818.

26. Lass-Flörl C, Gunsilius E, Gastl G, et al. Diagnosing invasive aspergillosisduring antifungal therapy by PCR analysis of blood samples. J Clin Micro-biol. 2004;42:4154-4157.

27. Forrest GN, Mankes K, Jabra-Rizk MA, et al. Peptide nucleic acid fluores-cence in situ hybridization-based identification of Candida albicans and its im-pact on mortality and antifungal therapy costs. J Clin Microbiol. 2006;44:3381-3383.

28. Wu FP, Della-Latta R, Addison B, et al. Dual color PNA FISH assay for si-multaneous identification of Candida albicans and Candida glabrata directlyfrom positive blood culture bottles. In: Proceedings from Infectious DiseaseSociety of America 2006 General Meeting; October 12-15, 2006; Toronto. Ab-stract 06-LB2053.

29. Buchheidt D, Hummel M, Schleiermacher D, et al. Prospective clinical eval-uation of a LightCycler-mediated polymerase chain reaction assay, a nest-ed-PCR assay and a galactomannan enzyme-linked immunosorbent assayfor detection of invasive aspergillosis in neutropenic cancer patients andhaematological stem cell transplant recipients. Br J Haematol. 2004;125:196-202.

30. Kawazu M, Kanda Y, Nannya Y, et al. Prospective comparison of the diag-nostic potential of real-time PCR, double-sandwich enzyme-linked im-munosorbent assay for galactomannan, and a (1–>3)-beta-D-glucan test inweekly screening for invasive aspergillosis in patients with hematologicaldisorders. J Clin Microbiol. 2004;42:2733-2741.

31. Costa C, Costa J, Desterke C, et al. Real-time PCR coupled with automatedDNA extraction and detection of galactomannan antigen in serum by en-zyme-linked immunosorbent assay for diagnosis of invasive aspergillosis. J Clin Microbiol. 2002;40:2224-2227.

32. Bretagne S. Molecular diagnostics in clinical parasitology and mycology: lim-its of the current polymerase chain reaction (PCR) assays and interest of thereal-time PCR assays. Clin Microbiol Infect. 2003;9:505-511.

33. Bialek R, Weiss M, Bekure-Nemariam K, et al. Detection of Cryptococcus neo-formans DNA in tissue samples by nested and real-time PCR assays. ClinDiagn Lab Immunol. 2002;9:461-469.

34. Larsen H, Masur H, Kovacs J, et al. Development and evaluation of a quan-titative, touch-down, real-time PCR assay for diagnosing Pneumocystis cariniipneumonia. J Clin Microbiol. 2002;40:490-494.

35. Hebart H, Löffler J, Meisner C, et al. Early detection of Aspergillus infectionafter allogeneic stem cell transplantation by polymerase chain reactionscreening. J Infect Dis. 2000;181:1713-1719.

IIM_07012008_00338.ps 6/16/08 11:05 AM Page 338


CASE REPORT

Acute suppurative thyroiditis(AST) is a rare inflammatorydisease. The rarity of this dis-

ease can be attributed to several fac-tors. The thyroid is well encapsulat-ed, which may hinder the transmis-sion of infection from surroundingtissue to the thyroid. In addition, a rich blood supply and lymphaticdrainage within the thyroid may beprotective against bacterial infec-tion. Furthermore, high iodine lev-els within the thyroid gland may create an environment that is unfa-vorable to bacterial growth.1 Reports of AST are uncommon in patients who have hematological malignan-cy. Only 9 cases have been reportedin the literature.2,3

Case reportA 27-year-old man presented to ourhospital with symptoms of generalweakness and fatigue. His blood testresults were positive for anemia (he-moglobin level, 2.9 g/dL). A bonemarrow biopsy specimen showedcellularity values of 0% to 10%, arange that is considered hypocellularfor the patient’s age; erythropoiesis,granulopoiesis, and megakaryocyteproduction were decreased. Aplasticanemia was diagnosed, and the pa-tient was treated with a consecutive5-day regimen of antithymoglobulin(ATG). At the start of chemotherapy,the absolute neutrophil count (ANC)was 1534/mL.

Five days after the administration

of ATG (day 1), sudden fever andsore throat developed. The patient’stemperature was 39.9ºC (103.8ºF).Blood pressure was 110/70 mm Hg,with a pulse rate of 96 beats perminute. Symptoms of influenzawere absent, but the patient com-plained of a sore throat and right-sided neck pain. No skin change ordiscoloration of the neck area wasobserved; however, swelling andtenderness of the neck developed.

The ANC was 252/mL. A thyroidfunction test revealed high free thy-roxine levels (2.37 ng/dL; normal,0.70 to 1.80 ng/dL), depressed thy-roid-stimulating hormone levels(0.12 mIU/L; normal, 0.4 to 4.1mIU/L), and normal total triiodo-thyronine levels (91 ng/dL; normal,87 to 184 ng/dL). A blood culturewas performed, and piperacillin and tobramycin were administeredempirically.

Radiological examination re-vealed cystic lesions of the thyroidgland with decreased enhancement;a thyroid abscess was therefore sus-pected (day 3; Figure 1). No pyri-form sinus fistula (PSF) was detectedby laryngoscopy or CT. Because theblood culture grew methicillin-sensi-tive Staphylococcus aureus (MSSA),cefazolin was added to the therapeu-tic regimen. Despite this antibiotictherapy and ultrasonography-guid-ed aspiration (day 6; Figures 2 and3), the patient’s condition did not

Dr Chung and Dr Choe are fellows and Dr Han, Dr Noh, Dr Youn, and Dr Oh are professors of med-icine in the department of general surgery at Seoul National University College of Medicine, South Korea.

Acute Suppurative Thyroiditis in aPatient With Aplastic AnemiaYoo Seung Chung, MD, Jun-Ho Choe, MD, Wonshik Han, MD, Dong-Young Noh, MD, Yeo-Kyu Youn, MD, and Seung Keun Oh, MD

Acute suppurative thyroiditis (AST) is a rare inflammatorycomplication in patients with hematological malignancy. Infection spreads to the thyroid from a distant site through the bloodstream or the lymphatics. Defects such as persistentthyroglossal duct and pyriform sinus fistula are associated with the development of AST. Ultrasonography, barium swallow testing, CT, and fine-needle aspiration are used for diagnosis. Treatment includes the administration of parenteral antibiotics, drainage, and excision. We describe a patient with aplastic anemia and bacteremic AST. [Infect Med. 2008;25:339-342]

Key words: Acute suppurative thyroiditis

IIM_07012008_00339.ps 6/16/08 11:07 AM Page 339


ACUTE SUPPURATIVE THYROIDITIS continued

improve. Surgery was performed tomanage the thyroid abscess (day 8).

During surgery, necrotic tissuewas discovered in the lower rightpole of the thyroid gland. The leftside of the gland was lesion-free. Weperformed necrosectomy of the fri-able necrotic tissue and drained yel-low pus. Ultimately, a right subtotalthyroidectomy was performed (Fig-

ure 4). We irrigated the thyroid bedwith normal saline solution and ap-plied a Jackson-Pratt closed-suctiondrain. The volume of drainage de-creased successively from 20 mL to10 mL to 5 mL over the course of 3 days postoperatively, and thewound was clean. Thyroid tissueculture revealed MSSA, as did cul-tures of blood and aspirated mucus.

The pathology laboratory reported achronic active inflammation with anabscess in the base of an adenoma-tous goiter (AG) (Figure 5).

After the operation, the patient’sfever persisted and he reported hipjoint pain, which seemed to becaused by bacteremia. A chest CTscan revealed multiple nodules withcavitation in both lungs, which wasassociated with a ground-glass opac-ity in both upper lobes and minimalbilateral pleural effusion. Thesesymptoms suggested septic pneu-monia and superinfection with Pneu-mocystis jiroveci. Nafcillin, piperacil-lin/tazobactam, and trimethoprim/sulfamethoxazole were adminis-tered. When the patient’s symptomsimproved and the bacteremia re-solved, the patient was dischargedfrom the hospital on the 51st post-operative day.

DiscussionBecause of its rich blood supply, lym-phatic drainage, abundant iodine,and protective fibrous capsule, thethyroid gland is very resistant to in-fection.1 However, AST develops inseveral settings. One setting is wheninfection spreads from a distant siteto the thyroid via the bloodstream orlymphatic system.4 Also, AST maydevelop secondary to trauma andpersistent thryoglossal duct and bydirect extension of infection from aneighboring structure to the thy-roid.4 Preexisting thyroid disease, in-cluding AG, nodular goiter, Hashi-moto thyroiditis, and thyroid cancer,can precede thyroid infection.1,5 AGcan be the indirect cause of a thyroidabscess.

In our patient, S aureus was isolat-ed from blood and thyroid tissue cul-tures. Bacteremia was attributed toimmunosuppression during chemo-therapy with ATG. It is recommend-ed that patients in whom fever de-velops following ATG administra-tion be treated with broad-spectrum

Figure 1 – This neck CT scan shows a

cystic lesion of the thyroid gland

with decreased enhancement.

Figure 2 – A sonogramat the time of aspiration

indicates diffuse heterogeneous

echogenecity in the lower right pole of the

thyroid gland, includinga multiseptated cystic portion.

Figure 3 – A microscopicphotograph of the

aspiration specimen of the thyroid gland

demonstrates the suspected abscess

with numerous neutrophils and

necrotic debris (hematoxylin-eosin

stain, original magnification �400).

IIM_07012008_00340.ps 6/16/08 11:07 AM Page 340


ACUTE SUPPURATIVE THYROIDITIS

antibiotics.6 Also, some investigatorshave recommended that patients athigh risk for infection should begiven prophylactic antibiotic and an-tifungal therapy, although there is aconcern that this strategy may ag-gravate emergence of antibiotic re-sistance.6-10 Because of this caveat,our hospital does not use prophylac-tic antibiotic therapy in neutropenicpatients during ATG chemotherapy.

AST is a rare complication ofchemotherapy in hematological ma-lignancy: Of the 9 cases reported inthe literature, 5 were attributed tofungal infection (all associated withCandida species).2,3 Of the 4 cases attributed to bacterial infection, acausative bacterial pathogen wasconfirmed in only 1 case. Blood andtissue cultures yielded Salmonella.2

Numerous imaging methods,such as ultrasonography, bariumswallow tests, and CT, are used to diagnose AST. Ultrasonography isthought to be an important methodfor diagnosing thyroid abnormali-ties.11 PSF, in particular, is consideredto be one of the most common un-derlying abnormalities in AST. Bari-um swallow is an essential diagnos-tic tool for confirming PSF.12 How-ever, Bernard and colleagues11 havesuggested that CT is extremely use-ful in diagnosing AST in its earlyphase, claiming that it provides moreaccurate mapping than ultrasonog-raphy. In our case, we used CT as thefirst-line imaging modality to evalu-ate the suspected thyroid abnormal-ity and any other pathological cervi-cal lesions. Because laryngoscopyand CT did not detect PSF and be-cause bacteremia was the suspectedcause of symptoms, we performedthe operation before ingravescence.

Treatment should include admin-istration of parenteral antibiotics,drainage of the abscess, and excisionof the affected area. There is no sig-nificant difference in the course ofdisease or survival among patients

treated with antibiotics, drainage, ora combination of both.5 Because themortality rate of AST is 8.6%5 and be-cause no improvement was seenafter administration of antibiotictherapy and aspiration of exudates,surgical management was an appro-priate decision in our case.

Few studies discuss the relation-ship between hematological malig-nancy and thyroid disease in long-term follow-up. Moskowitz and col-leagues13 reported that autoimmunedisorders such as Graves disease,Hashimoto thyroiditis, toxic multi-nodular goiter, and idiopathic hy-


Antithymoglobulin

Cefazolin

Nafcillin

Piperacillin

Piperacillin/tazobactam

Tobramycin

Trimethoprim/sulfamethoxazole

Figure 4 – This photo-graph shows the friablenecrotic tissue excisedduring a right subtotalthyroidectomy.

Figure 5 – A microscopicphotograph of the thyroidgland specimen afterright subtotal thyroid-ectomy demonstrateschronic active inflamma-tion with abscess forma-tion (hematoxylin-eosinstain, original magnifica-tion �40).

IIM_07012008_00341.ps 6/16/08 11:07 AM Page 341


ACUTE SUPPURATIVE THYROIDITIS continued

pothyroidism are closely associatedwith acute leukemia. Toubert andcolleagues14 reported that hypothy-roidism could occur after bone mar-row transplant without total body irradiation. Therefore, follow-up thy-roid function testing should be per-formed and hormonal therapy, ifneeded, should be administered. ❖

REFERENCES1. Braverman LE, Utiger RD, eds. Werner & Ing-

bar’s The Thyroid, A Fundamental and ClinicalText. 9th ed. Philadelphia: Lippincott Williams& Wilkins; 2005:541-547.

2. Dai MS, Chang H, Peng MY, et al. Suppurativesalmonella thyroiditis in a patient with chroniclymphocytic leukemia. Ann Hematol. 2003;82:646-648.

3. Imai C, Kakihara T, Watanabe A, et al. Acutesuppurative thyroiditis as a rare complicationof aggressive chemotherapy in children withacute myelogeneous leukemia. Pediatr HematolOncol. 2002;19:247-253.

4. Hazard JB. Thyroiditis: a review. Am J ClinPathol. 1955;25:289-298.

5. Berger SA, Zonszein J, Villamena P, et al. Infec-tious diseases of the thyroid gland. Rev InfectDis. 1983;5:108-122.

6. Marsh JC, Ball SE, Darbyshire P, Mittman N.

Guidelines for the diagnosis and managementof acquired aplastic anaemia. Br J Haematol.2003;123:782-801.

7. Gafter-Gvili A, Fraser A, Paul M, Leibovici L.Meta-analysis: antibiotic prophylaxis reducesmortality in neutropenic patients. Ann InternMed. 2005;142(12, pt 1):979-995.

8. Liberati A, D’Amico R, Pifferi S, et al. Antibiot-ic prophylaxis to prevent nosocomial infectionsin patients in intensive care units: evidence thatstruggle to convince practising clinicians. InternEmerg Med. 2006;1:160-162.

9. Gafter-Gvili A, Fraser A, Paul M, et al. Anti-biotic prophylaxis for bacterial infections inafebrile neutropenic patients following che-motherapy. Cochrane Database Syst Rev. 2005;(19):CD004386.

10. Lo N, Cullen M. Antibiotic prophylaxis inchemotherapy-induced neutropenia: time toreconsider. Hematol Oncol. 2006;24:120-125.

11. Bernard PJ, Som PM, Urken ML, et al. The CTfindings of acute thyroiditis and acute suppu-rative thyroiditis. Otolaryngol Head Neck Surg.1988;99:489-493.

12. Miyauchi A, Matsuzuka F, Kuma K, Takai S.Piriform sinus fistula: an underlying abnor-mality common in patients with acute suppu-rative thyroiditis. World J Surg. 1990;14:400-405.

13. Moskowitz C, Dutcher JP, Wiernik PH. Associ-ation of thyroid disease with acute leukemia.Am J Hematol. 1992;39:102-107.

14. Toubert ME, Socié G, Gluckman E, et al. Short-and long-term follow-up of thyroid dysfunc-tion after allogeneic bone marrow transplanta-tion without the use of preparative total bodyirradiation. Br J Haematol. 1997;98:453-457.

BOEHRINGER INGELHEIMPHARMACEUTICALS, INC.Viramune . . . . . . . . . . . . . . . . . . .317-319

BRISTOL-MYERS SQUIBB COMPANYReyataz . . . . . . . . . . . . . . . . . . . . .Cover 4

CUBIST PHARMACEUTICALSCubicin . . . . . . . . . . . . . . . . . . . . .327-328

ORTHO-MCNEIL, INC.Doribax . . . . . . . . . . . . . . . . . . . . .307-309

PFIZER U.S. PHARMACEUTICALSZyvox . . . . . . . . . . . . . . . . . . .Cover 2-304

Advertisers’IndexJULY 2008

Interested in placing an ad inINFECTIONS in MEDICINE?Contact Diane Carpenteri at

(203) 662-6545 or e-mail:

[email protected].

IIM_07012008_00342.ps 6/16/08 11:07 AM Page 342

Date post:	28-Jan-2017
Category:	Documents
Upload:	vutuong
View:	214 times
Download:	0 times

Differentiating Kawasaki Syndrome From Microbial Infection

Documents