2 Ocak 2013 Çarşamba

Revisiting the Source of Candidemia: Skin or Gut?

Marcio Nucci1 andElias Anaissie2

1UniversityHospital,UniversidadeFederaldoRiodeJaneiro,Brazil;and 2UniversityofArkansasforMedicalSciences,LittleRock

Received 5 March 2001; revised 30 May 2001; electronically published 8 November 2001.

The source of candidemiahas been the subjectof considerable debate, withsome suggesting a originin the gastrointestinal tractand others suggesting askin origin. To evaluatethe potential sources ofcandidemia, we performed acomputerized search of theMEDLINE database for studiespublished from January 1966through September 2000 andwe identified relevant abstractspresented at national meetings.We reviewed the literaturewith special emphasis onstudies that used appropriatedefinitions, evaluated both gutand skin as sources,and conducted molecular-relatedness studies.Among 203 candidemia studiespublished, we identified 21that evaluated a specificsource for candidemia andonly 5 that performedmolecular typing. Those studiesand additional experimental, epidemiologic,and molecular-relatedness studies stronglysuggested that the gutis an important sourceof candidemia, and studiesthat supported the skinas a source forthis infection were surprisinglyincomplete.

Reprintsorcorrespondence:Dr.EliasJ.Anaissie,ClinicalAffairs,MyelomaandTransplantationResearchCenter,UniversityofArkansasforMedicalSciences,4301WestMarkham,Slot776,LittleRock,AR72205(anaissieeliasj@exchange.uams.edu).


The incidence of candidemiahas increased substantially duringthe past 20 years[1], and Candida speciesnow rank among the4 pathogens most frequentlyisolated in blood cultures.The source of thisinfection has been thesubject of considerable debate,with some suggesting thegastrointestinal tract (endogenous acquisition)[2] and others favoringthe skin (exogenous acquisition)as a source [3,4], the latter beingimplicated in the pathogenesisof catheter-related candidemia. Identificationof the source ofcandidemia is important becauseof the implications forpreventive strategies. Other sourcesof candidemia have alsobeen hypothesized, including thegenitourinary tract [5] andcontaminated iv solutions [6].In this article, wereview studies that evaluatedpotential sources of candidemiawith special emphasis onstudies that used appropriatedefinitions, evaluated both gutand skin sources, andperformed molecular-relatedness studies.

METHODS

Data source. We performeda computerized search ofthe MEDLINE database forstudies published from January1966 through November 2000in any language. Thekeywords used were “candidemia,”“Candida,” “gut,” “gastrointestinal tract,”"skin,” “molecular typing,” “source,”"origin,” and “animal model.”All abstracts published from1987 through 2000 atthe yearly meetings ofthe American Society forMicrobiology, the Infectious DiseasesSociety of America, andthe Society of HealthcareEpidemiology of America werealso reviewed.

Study selection. We focused onthe papers that fulfilledthe following criteria: (1)evaluation of the sourceof candidemia by investigationof prior colonization ofthe skin and/or thegastrointestinal tract and (2)use of molecular methodsto support relatedness betweencolonizing and infecting strains.We required molecular-relatedness studiesbecause of the genotypicdiversity of Candida species[7]. Therefore, finding thesame species at acolonizing site does notnecessarily imply that thesame organism is responsiblefor the bloodstream infection.We also evaluated thedensity and sequence ofcandida colonization and analyzedall of our findingsaccording to patient populationto find out whetherthe pathogenesis of candidemiadiffered among various patientpopulations.

Data extraction. We reviewed methodology, definitionof candidemia, source ofcandidemia, patient characteristics, andthe results of surveillancecultures and molecular-relatedness studies.

RESULTS

Limitations of the Current Literature

Weidentified 203 candidemia studies(figure 1). Of those studies,only 21 had investigatedspecific sources of candidemia:4 evaluated the skin[3, 4, 8, 9],11 evaluated the gut[1020], and 6 evaluatedboth sources [2126]. Amongthese 21 studies, only5 performed DNA typing:4 studies that evaluatedthe gut [1720] and1 evaluated both theskin and gut [26].

Figure 1. Selectionof studies that evaluatedthe sources of candidemia.PFGE, pulsed-field gel electrophoresis;REA, restriction enzyme analysis;RFLP, restriction-fragment length polymorphism.

Findings: The Skin Hypothesis

Experimental evidence. Theoretically,Candida species could enterthe bloodstream from theskin via 2 routes:invasion from a skinlesion (e.g., burn wounds)or contamination of avascular catheter from colonizedskin (table 1). Therefore, anexperimental model obtained byinoculating Candida organisms intoa wound or theskin, followed by theinsertion of a venouscatheter at the inoculatedsite, should lead tohematogenous candidiasis. We couldnot identify such amodel, which indicates thatno experimental data supportskin as the sourceof candidemia.

Table 1. Studies that evaluatedthe source of candidemia.

Clinical evidence from epidemiological studies. Amongthe 4 studies thatinvestigated only skin colonization,2 were conducted usingpatients with burns. In1 study, 50% ofpatients with candidemia hadpositive burn wound cultures[4]. However, no correlationbetween burn wound culturesand candidemia was made.In the other study,positive burn wound cultureswere not predictors ofcandidemia, according to multivariateanalysis [7].

Another study [9]performed skin surveillance cultures(umbilicus and groin) on57 lowbirth-weight infants admittedto a neonatal unit.All 3 patients withcandidemia had the sameCandida species recovered fromskin cultures.

In the fourthstudy [3], 53 patientswho received parenteral nutritionwere prospectively evaluated. Atthe time of catheterremoval, cultures of skinsamples obtained from thesite of catheter insertionand of the intravascular(first 3 cm fromthe tip of thecatheter) and the subcutaneous(1 cm distal tothe site of dermalentry) segments of thecatheter were performed. Sixpatients had candidemia, andin 2 patients, thesame species was isolatedfrom the blood, skin,and subcutaneous and intravascularsegments of the catheter.

Twoother studies evaluated skinand other sites aspotential sources for candidemia.Sheridan et al. [21]retrospectively reviewed Candida colonizationand infection in 962children with burns byreviewing surveillance cultures ofsamples of the respiratorytract, wound, stool, andurine. All 17 patientswith candidemia had thesame species isolated fromthe wound samples, whereasonly 35% of stoolcultures yielded the sameCandida species that wasrecovered from blood samples.

Huanget al. [23] prospectivelyevaluated 116 lowbirth-weight infantswith weekly cultures fromspecimens of oropharynx, rectum,skin (groin and axilla),urine, and endotracheal aspirates.Three patients developed candidemia,and 1 of thesepatients had prior colonizationat 4 sites bythe same species. However,the sites of positivecultures were not mentioned.

Clinical evidence from molecular-relatedness studies. Onereport evaluated skin (andgut) colonization and performedmolecular-relatedness studies with restrictionenzyme analysis and restriction-fragmentlength polymorphism [26]. Inthis study, 64 neonateshad weekly cultures ofspecimens of the oropharynx,umbilicus, groin, and ivsite(s), starting in thefirst 24 h afterbirth. One of theseneonates developed candidemia, andDNA typing showed anidentical pattern of thestrains isolated from blood,mouth, groin, and umbilicussamples. However, it ispractically impossible to identifythe primary source ofinfection in this singlecase, because, as shownby El-Mohandes et al.[25], the groin mayhave became secondarily colonizedby strains that originatedin the gastrointestinal tract,particularly given that thegut mucosa (mouth) harboredthe same organism andthat the sequence ofcolonization and infection inthis patient was notevaluated.

Density and sequence of colonization and infection. No studies have beenpublished that support theskin hypothesis on thebasis of density andsequence of candidal colonization.

Findings: The Gut Hypothesis

Experimental evidence: laboratory. Alarge number of experimentalanimal models of gastrointestinalcolonization and disseminated infectionhave been published elsewhere[27]. In these models,candidal gut colonization isestablished through feeding theanimals with a dietthat contained Candida species.Immunosuppression of these colonizedanimals usually leads todisseminated candidiasis [2834].

Experimental evidence: clinical. A healthyhuman subject volunteered toreceive an oral suspensionof 1012 cells ofCandida albicans and developed hematogenouscandidiasis [35].

Clinical evidence from epidemiological studies. El-Mohandes et al.[25] performed cultures ofstool, gastric aspirate, andskin samples obtained from82 neonates admitted toan intensive care unit.Candida parapsilosis was recovered in75% of stool culturesand 80% of culturesof gastric aspirates. Bycontrast, only 50% ofpositive skin cultures yieldedC. parapsilosis. Furthermore, positive stoolcultures always preceded positiveskin cultures, and thebaseline stool cultures ofall 4 patients withcandidemia yielded C. parapsilosis, whereasbaseline skin cultures werenegative in these samepatients.

In a prospective study,Bow et al. [10]showed that colonization byCandida species of sitesother than the skin(e.g., nose, oropharynx, anus,and urine) was arisk factor for hematogenouscandidiasis. Similar findings werereported in 2 studiesby Martino et al.[12, 13]. Chryssanthou etal. [11] reported that3 patients with candidemia(out of 277 patientswith neutropenia) had thesame Candida species isolatedfrom stool and bloodcultures.

MacFie et al. [14]prospectively studied 279 patientswho underwent laparotomy andobtained cultures of gastricaspirates. Candida species werethe organisms most frequentlyrecovered from gastric aspirates(54%), and the sameCandida species was recoveredfrom blood and gastricaspirate in 1 patientwith candidemia.

Pierro et al.[15] prospectively evaluated 94infants who received parenteralnutrition with twice-weekly oraland anal swabs. Only1 patient developed candidemia,and the same species(C. albicans) was isolated fromthe anal swab surveillancecultures and blood samples.

Pittetet al. [16] studied650 critically ill surgicaland neonatal patients withsequential cultures of samplesof the oropharynx, trachea,and stomach. Among 29patients with significant colonization(positive cultures in 3samples obtained from thesame or different bodysites on 2 consecutivescreening days), density ofcolonization was a goodpredictor of subsequent candidemiaby multivariate analysis. Theassociation between density ofcolonization and invasive candidiasiswas also demonstrated inanother study of surgicalpatients [36].

Two retrospective studiesevaluated patients with cancerwho had candidemia inwhich the researchers obtainedsamples of the skinand gut. In onestudy [22], gut (butnot skin) colonization wasa risk factor forcandidemia by univariate analysis,whereas colonization at anysite (skin, urine, oropharynx,or feces) was arisk factor for candidemiaby multivariate analysis. Theother study [24] alsoidentified colonization at anysite (nose, pharynx, rectum,urine, or skin) asa risk factor forcandidemia by multivariate analysis.

Molecular-relatedness studies. Saimanet al. [20] prospectivelyevaluated 2847 neonates withweekly swabs of theperianal and rectal areas.Among 35 patients withcandidemia, gut colonization precededhematogenous infection in 15patients. By use ofpulsed-field gel electrophoresis (PFGE),all blood and gutstrain pairs exhibited identicalDNA patterns.

Reagan et al.[18] prospectively surveyed patientsadmitted to the bonemarrow transplant and hematologicalmalignancy units with weeklycultures of samples ofthe pharynx, urine, andstool. Sixteen patients hadpositive surveillance cultures andsubsequently developed candidemia. In5 patients, the gutwas the colonizing site,and in all 5cases, both colonizing andinfecting strains had thesame DNA pattern whentested by restriction endonucleasedigestion of chromosomal DNAwith Southern hybridization.

Klempp-Selb etal. [19] identified 2patients with candidemia inwhom stool cultures wereobtained. In 1 ofthese 2 patients, theDNA patterning by useof PFGE was similarfor blood and stoolisolates.

Colombo et al. [17]reported an HIV-positive patientwho developed a C. albicansfungemia. Oral, anal, andurine samples were obtainedfor culture, which yieldedthe same Candida species.By use of PFGE,those authors observed identicalDNA profiles of thecolonizing and infecting strains.

Density and sequence of colonization and infection. Thegut origin of candidemiais supported by thefindings that the densityof colonization from gutsources is a goodpredictor of subsequent candidemia[10, 12, 13, 16,22] and that gutcolonization precedes hematogenous candidiasis[1416, 18, 20, 25,35].

Evaluation of Specific Patient Populations

The patient populations ofthe 21 studies reviewedwere clustered in 4major categories: patients withneutropenia who had cancer[1013, 22, 24], newborns[9, 15, 16, 20,23, 25, 26], surgicalpatients [14, 16], andpatients with burns [4,8, 21]. We analyzed3 factors that mayhelp identify the sourceof candidemia in thesepatient populations: density ofcolonization, sequence of colonizationand subsequent infection, andmolecular-relatedness of colonizing andinfecting strains. Data inpatients with neutropenia andneonates strongly suggest, onthe basis of densityand sequence of colonizationand supporting molecular-relatedness studies,that the gut isthe likely source ofinfection [10, 12, 13,16, 18, 20, 22,25]. On the otherhand, the paucity ofdata in the surgicaland burn population precludesany conclusion regarding theprimary source of theinfection in these settings(table 2).

Table 2. Evaluation of density andsequence of colonization andmolecular-relatedness studies in specificpatient populations.

Application of Koch’s Postulates to the Skin and Gut Hypotheses

We applied theKoch’s postulates to furtherdetermine the strength ofthe relationship between candidemiaand its source (table 3).All postulates were fulfilledfor the gastrointestinal tractorigin of candidemia, butnot for the cutaneousorigin. The postulates included(1) development of candidemiaafter the inoculation ofthe organism at thelikely source of infection(animal models and experimentin a human volunteer)[2835], (2) recovery ofCandida species from thesource in almost everycase of candidemia [1013,1622, 25], (3) eradicationof Candida species fromthe likely source resultsin reduction of theincidence of candidemia (antifungalprophylaxis studies) [37], (4)demonstration of the sequenceof colonization at thelikely source and subsequentdevelopment of candidemia [16,18, 20, 25], (5)demonstration of the influenceof the density ofcolonization at the likelysource and the developmentof candidemia [12, 13,16, 22], and (6)supporting molecular-relatedness studies [1720,26].

Table 3. Koch’s postulates adapted tosource(s) of candidemia.

DISCUSSION

Our reviewled us to 2conclusions. First, the dataon the source ofcandidemia is limited toa few studies. Second,and most important, ourreview suggests an endogenous,gastrointestinal origin for candidemia,as evidenced by experimental[2835], clinical [1016, 18,20, 22, 25, 35],and molecular-relatedness studies [1720],and this is furthersupported by fulfillment ofKoch’s postulates (table 3). Additionalarguments that support thegut origin include thepresence of extensive gastrointestinaltract involvement in disseminatedcandidiasis [38], the identificationof gastrointestinal colonization asa risk factor forcandidemia [16], and theefficacy of antifungal prophylaxisin patients with cancerin preventing hematogenous candidiasisby eliminating or reducingcolonization by Candida species[37].

The cutaneous origin ofcandidemia is suggested inC. parapsilosis infections. A centralvenous catheter is arisk factor for candidemia[22, 39, 40]. BecauseC. parapsilosis is frequently recoveredfrom skin samples [41]and candidemia with thisorganism occurs more frequentlyin patients with centralvenous catheters [42], itis tempting to speculatethat contamination of skinnear the insertion siteof a venous catheterleads to venous cathetercolonization, biofilm production [43],and subsequent fungemia. However,the gut and notthe skin appears tobe the primary sourcefor this infection, asshown in the studyby El-Mohandes et al.[25]. In that study,stool colonization by C. parapsilosiswas present in allpatients, and a clear-cutsequence of colonization (stoolsfollowed by skin) wasobserved.

The data used tosupport skin as thesource of candidemia aresurprisingly incomplete. However, andalthough it appears thatskin may not bean important source ofinfection in neonates andin patients with neutropenia,it is possible thatthis site plays animportant role as aprimary source in otherpatient populations, such aspatients with burns. Indeed,extensive burn wounds maybe colonized and infectedwith Candida species, whichincreases the risk ofcandidemia [8]. In addition,the limited power ofa literature review andthe paucity of thedata available to supporta cutaneous source ofcandidemia cannot exclude thepotential role of skinas a source forcandidemia. Indeed, septic thrombophlebitisand candidemia acquired fromtotal parenteral nutrition orfrom contamination of skin(by stools colonized byCandida species) at vascularinsertion sites do suggesta skin origin. Alternatively,these infections may resultfrom contamination of adeep vein thrombus bycandidemia that originated fromthe gut.

If confirmed bycarefully conducted prospective clinicalstudies, our findings haveimplications for the clinicalmanagement of candidemia. Becausethe gut and notthe skin may bethe primary source ofcandidemia, attempts aimed atreducing gut colonization byCandida species may havea favorable impact inreducing the incidence ofcandidemia. Indeed, studies inpatients with neutropenia whoare at high riskfor candidal infections haveproved the efficacy ofthis strategy [37]. Bycontrast, and although formallyrecommended by experts inthis field [44], theautomatic removal of allvascular catheters to treatall cases of candidemiashould be reevaluated, particularlygiven that successful therapyof some cases ofcandidemia without vascular catheterremoval is possible [45].

Finally,the most rigorous approachto confirm or refuteour hypothesis is theconduct of carefully plannedprospective clinical trials. Thesetrials need to useappropriate definitions of candidemia,obtain surveillance cultures fromvarious sites, including skinand gut, test densityand sequence of colonization,and perform molecular-relatedness studies.In addition, prospective randomizedstudies should be performedto address the questionof vascular catheter removalin the setting ofcandidemia.

In conclusion, the availabledata suggest that thesource of candidemia hasnot been well studiedand that the gutmay be an importantsource. Additional studies areneeded to confirm orrefute our findings andto further evaluate therole of the skinas a potential sourceof this infection.

References

1.

Beck-Sagué CM, Jarvis WR. Secular trends in the epidemiology of nosocomial fungal infections in the United States, 19801990. The National Nosocomial Infections Surveillance System. J Infect Dis 1993; 167:124751. First citation in article | PubMed
2. Cole GT, Halawa AA, Anaissie EJ. The role of the gastrointestinal tract in hematogenous candidiasis: from the laboratory to the bedside. Clin Infect Dis 1996; 22(Suppl 2):S7388. First citation in article | PubMed
3. Bjornson HS, Colley R, Bower RH, Duty VP, Schwartz-Fulton JT, Fischer JE. Association between microorganism growth at the catheter insertion site and colonization of the catheter in patients receiving total parenteral nutrition. Surgery 1982; 92:7207. First citation in article | PubMed
4. Spebar MJ, Pruitt BA. Candidiasis in the burned patient. J Trauma 1981; 21:2379. First citation in article | PubMed
5. Ang BS, Telenti A, King B, Steckelberg JM, Wilson WR. Candidemia from a urinary tract source: microbiological aspects and clinical significance. Clin Infect Dis 1993; 17:6626. First citation in article | PubMed
6. McNeil MM, Lasker BA, Lott TJ, Jarvis WR. Postsurgical Candida albicans infections associated with an extrinsically contaminated intravenous anesthetic agent. J Clin Microbiol 1999; 37:1398403. First citation in article | PubMed
7. Bretagne S, Costa JM, Besmond C, Carsique R, Calderone R. Microsatellite polymorphism the promotor sequence of the elongation factor 3 gene of Candida albicans as the basis for a typing system. J Clin Microbiol 1997; 35:177780. First citation in article | PubMed
8. Ekenna O, Scherertz RJ, Bingham H. Natural history of bloodstream infections in a burn patient population: the importance of candidemia. Am J Infect Control 1993; 21:18995. First citation in article | PubMed
9. Shattuck KE, Cochran CK, Zabransky RJ, Pasarell L, Davis JC, Malloy MH. Colonization and infection associated with Malassezia and Candida species in a neonatal unit. J Hosp Infect 1996; 34:1239. First citation in article | PubMed
10. Bow E, Loewen R, Cheang MS, Schacter B. Invasive fungal disease in adults undergoing remission-induction therapy for acute myeloid leukemia: the pathogenetic role of antileukemic regimen. Clin Infect Dis 1995; 21:3619. First citation in article | PubMed
11. Chryssanthou E, Kalin M, Engervall P, Petrini B, Bjorkholm M. Low incidence of candidemia among neutropenic patients treated for hematological diseases. Scand J Infect Dis 1998; 30:48993. First citation in article | PubMed
12. Martino P, Girmenia C, Venditti M, et al. Candida colonization and systemic infection in neutropenic patients. Cancer 1989; 64:20304. First citation in article | PubMed
13. Martino P, Girmenia C, Micozzi A, De Bernardis F, Boccanera M, Cassone A. Prospective study of Candida colonization, use of empiric amphotericin B and development of invasive mycosis in neutropenic patients. Eur J Clin Microbiol Infect Dis 1994; 13:797804. First citation in article | PubMed
14. MacFie J, O’Boyle C, Mitchell CJ, Buckley PM, Johnstone D, Sudworth P. Gut origin of sepsis: a prospective study investigating associations between bacterial translocation, gastric microflora, and septic morbidity. Gut 1999; 45:2238. First citation in article | PubMed
15. Pierro A, van Saene HKF, Donnell SC, et al. Microbial translocation in neonates and infants receiving long-term parenteral nutrition. Arch Surg 1996; 131:1769. First citation in article | PubMed
16. Pittet D, Monod M, Suter PM, Frenk E, Auckenthaler R. Candida colonization and subsequent infections in critically ill surgical patients. Ann Surg 1994; 220:7518. First citation in article | PubMed
17. Colombo AL, Branchini ML, Geiger D, Schmidt AL, Pignatari ACC, Fischman O. Gastrointestinal translocation as a possible source of candidemia in an AIDS patient. Rev Inst Med Trop Sao Paulo 1996; 38:197200. First citation in article | PubMed
18. Reagan DR, Pfaller MA, Hollis RJ, Wenzel RP. Characterization of the sequence of colonization and nosocomial candidemia using DNA fingerprinting and a DNA probe. J Clin Microbiol 1990; 28:27338. First citation in article | PubMed
19. Klempp-Selb B, Rimek D, Kappe R. Kariotyping of Candida albicans and Candida glabrata from patients with Candida species. Mycoses 2000; 43:15963. First citation in article | PubMed
20. Saiman L, Ludington E, Pfaller M, et al. Risk factors for candidemia in neonatal intensive care unit patients. Pediatr Infect Dis J 2000; 19:31924. First citation in article | PubMed
21. Sheridan RL, Weber JM, Budkevich LG, Tompkins RG. Candidemia in the pediatric patient with burns. J Burn Care Rehabil 1995; 16:4403. First citation in article | PubMed
22. Karabinis A, Hill C, Leclercq B, Tancrede C, Baume D, Andremont A. Risk factors for candidemia in cancer patients: a case-control study. J Clin Microbiol 1988; 26:42932. First citation in article | PubMed
23. Huang YC, Li CC, Lin TY, et al. Association of fungal colonization and invasive disease in very low birth weight infants. Pediatr Infect Dis J 1998; 17:81922. First citation in article | PubMed
24. Pagano L, Antinori A, Ammassari A, et al. Retrospective study of candidemia in patients with hematological malignancies: clinical features, risk factors and outcome of 76 episodes. Eur J Haematol 1999; 63:7785. First citation in article | PubMed
25. El-Mohandes AE, Johnson-Robbins L, Keiser JF, Simmens SJ, Aure MV. Incidence of Candida parapsilosis colonization in an intensive care nursery population and its association with invasive fungal disease. Pediatr Infect Dis J 1994; 13:5204. First citation in article | PubMed
26. Reef SE, Lasker BA, Butcher DS, et al. Nonperinatal nosocomial transmission of Candida albicans in a neonatal intensive care unit: prospective study. J Clin Microbiol 1998; 36:12559. First citation in article | PubMed
27. Odds FC. Pathogenesis of candidosis. In: Odds FC, ed. Candida and candidosis. 2d ed. Philadelphia: Bailliere Tindall, 1988:25278. First citation in article
28. Myerowitz RL. Gastrointestinal and disseminated candidiasis: an experimental model in the immunosuppressed rat. Arch Pathol Lab Med 1981; 105:13843. First citation in article | PubMed
29. Sandovsky-Losica H, Barr-Nea L, Segal E. Fatal systemic candidiasis of gastrointestinal origin: an experimental model in mice compromised by anti-cancer treatment. J Med Vet Mycol 1992; 30:21931. First citation in article | PubMed
30. Field LH, Pope LM, Cole GT, Guentzel MN, Joe L. Persistence and spread of Candida albicans after intragastric inoculation of infant mice. Infect Immun 1981; 31:78391. First citation in article | PubMed
31. Umenai T, Kono S, Ishida N. Systemic candidiasis from Candida albicans colonizing the gastrointestinal tract of mice. Experientia 1979; 35:13312. First citation in article | PubMed
32. Kinsman OS, Pitblado K. Candida albicans gastrointestinal colonization and invasion in the mouse: effect of antibacterial dosing, antifungal therapy and immunosuppression. Mycoses 1989; 32:66474. First citation in article | PubMed
33. de Repentigny L, Phaneuf M, Mathieu LG. Gastrointestinal colonization and systemic dissemination by Candida albicans and Candida tropicalis in intact and immunocompromised mice. Infect Immun 1992; 60:490714. First citation in article | PubMed
34. Ekenna O, Sherertz RJ. Factors affecting colonization and dissemination of Candida albicans from the gastrointestinal tract of mice. Infect Immun 1987; 55:155863. First citation in article | PubMed
35. Krause W, Matheis H, Wulf K. Fungaemia and funguria after oral administration of Candida albicans. Lancet 1969; 1:5989. First citation in article | PubMed
36. Calandra T, Bille J, Schneider R, Mosimann F, Francioli P. Clinical significance of Candida isolated from peritoneum in surgical patients. Lancet 1989; 2:143740. First citation in article | PubMed
37. Menichetti F, Del Favero A, Martino P, et al. Preventing fungal infection in neutropenic patients with acute leukemia: fluconazole compared with oral amphotericin B. The GIMEMA Infection Program. Ann Intern Med 1994; 120:9138. First citation in article | PubMed
38. Walsh TJ, Merz WG. Pathologic features in the human alimentary tract associated with invasiveness of Candida tropicalis. Am J Clin Pathol 1986; 85:498502. First citation in article | PubMed
39. Wey SB, Mori M, Pfaller MA, Woolson RF, Wenzel RP. Risk factors for hospital-acquired candidemia: a matched case-control study. Arch Intern Med 1989; 149:234953. First citation in article | PubMed
40. Bross J, Talbot GH, Maislin G, Hurwitz S, Strom BL. Risk factors for nosocomial candidemia: a case-control study in adults without leukemia. Am J Med 1989; 87:61420. First citation in article | PubMed
41. McGinley KJ, Larson EL, Leyden JJ. Composition and density of microflora in the subungual space of the hand. J Clin Microbiol 1988; 26:9503. First citation in article | PubMed
42. Abi-Said D, Anaissie E, Uzum U, Raad I, Pinzcowski H, Vartivarian S. The epidemiology of hematogenous candidiasis caused by different Candida species. Clin Infect Dis 1997; 24:11228. First citation in article | PubMed
43. Branchini ML, Pfaller MA, Rhine-Chalberg J, Frempong T, Isenberg HD. Genotypic variation and slime production among blood and catheter isolates of Candida parapsilosis. J Clin Microbiol 1994; 32:4526. First citation in article | PubMed
44. Rex JH, Walsh TJ, Sobel JD, et al. Practice guidelines for the treatment of candidiasis. Clin Infect Dis 2000; 30:66278. First citation in article | Full Text | PubMed
45. Anaissie EJ, Rex JH, Uzun O, Vartivarian S. Predictors of adverse outcome in cancer patients with candidemia. Am J Med 1998; 104:23845. First citation in article | PubMed

Table 1. CITED IN TEXT | TYPESET IMAGE |

Studies that evaluated the source of candidemia.

Source, author, year,
and reference
Patients Study design Definitions of candidemia, source of blood collection Sites of colonization sampled Results and comments
Skin
Schattuck 1996 [9] 57 lowbirth-weight neonates, 3 with candidemia Prospective cohort Definition not mentioned; blood was drawn from an iv catheter Umbilicus and groin Positive skin cultures in 9 patients (16%); 3 had positive blood cultures with the same species recovered from skin
Bjornson, 1982 [3] 53 patients receiving TPN, 6 with candidemia Prospective cohort Not mentioned Site of iv catheter insertion In 2 of 6 patients with candidemia, Candida albicans grew from the iv catheter insertion site, subcutaneous catheter segment, and blood
Spebar, 1981 [4] 452 patients with burns who had Candida species isolated from any source, 52 with candidemia Retrospective Not mentioned Burn wound 50% of patients with candidemia had positive burn wound cultures; no species identification provided
Ekenna 1993 [8] 209 patients with burns, 16 with candidemia Retrospective Not mentioned Burn wound Burn wound culture positive for Candida species was not predictor of candidemia by multivariate analysis; large burn size was a risk factor for candidemia by multivariate analysis
Gut and skin
Sheridan, 1995 [21] 962 patients with burns, 17 with candidemia Retrospective 1 positive blood culture, source of blood collection not mentioned Respiratory tract, wound, stool, and urine All patients with candidemia had the same species isolated from the wound; 35% had the same species isolated from the stool; colonization at multiple sites was associated with candidemia
Karabinis, 1988 [22] 30 patients with cancer who had candidemia and 58 controls Retrospective, case-control 2 positive blood cultures, recovered within 72 h of each other (source of blood collection not mentioned) Skin, urine, oropharynx, and feces Fecal and oral colonization were risk factors for candidemia by univariate analysis; colonization from 1 site was a risk factor by multivariate analysis.
Huang, 1998 [23] 116 lowbirth-weight infants in neonatal ICU, 3 with candidemia Prospective cohort 1 positive blood culture from peripheral vein plus clinical signs of infection Oropharynx, rectum, axilla, groin, urine, and tracheal tube aspirate Only 1 of 3 patients with candidemia (C. albicans) was colonized (4 sites, not specified) by the same species
Pagano, 1999 [24] 76 patients with candidemia who had hematological malignancies Retrospective 1 positive blood culture from peripheral vein or an iv catheter plus clinical signs of infection Nasal, pharyngeal, rectal, urinary, and skin swabs Colonization by Candida species (any site) was a risk factor for candidemia by multivariate analysis
El-Mohandes, 1994 [25] 82 neonates in NICU, 4 with candidemia due to C. parapsilosis Prospective cohort 1 positive blood culture from peripheral vein plus clinical signs of infection Stool, gastric aspirate, and skin (retroauricular) C. parapsilosis was predominant in gastric and stool cultures; positive skin cultures were always preceded by positive stool cultures; all 4 patients with C. parapsilosis fungemia had rectal colonization with the same species
Reef, 1998 [26] 64 neonates, 1 with candidemia due to C. albicans Prospective cohort Not mentioned Swabs from oropharynx, umbilicus, groin, and iv site Isolates from mouth, umbilicus, and groin showed DNA patterns that were identical to the blood isolate (RFLP and REA)
Gut
Bow, 1995 [10] 138 patients with AML, 16 with hematogenous candidemia Prospective cohort Not mentioned Nasal, oral, and anal swabs and urine Number of sites colonized by Candida species was a risk factor for invasive candidiasis by univariate but not multivariate analysis
MacFie, 1999 [14] 279 patients undergoing laparotomy, 1 patient with candidemia Prospective cohort Not mentioned Gastric aspirate and mesenteric lymph nodes Candida species was the most frequent isolate in gastric aspirates (54%); the same Candida species was recovered from blood and gastric aspirate in 1 patient.
Chryssanthou, 1998 [11] 277 patients with neutropenia, 3 with candidemia Retrospective 1 positive blood culture, source of blood collection not mentioned Throat, urine, and feces All 3 patients with candidemia had positive stool cultures for the same species
Pierro, 1996 [15] 94 infants receiving TPN, 1 with candidemia Prospective cohort 1 positive blood culture (iv catheter or peripheral vein) plus clinical signs of infection Oral and anal swabs The same Candida species was recovered from blood and anal swab
Pittet, 1994 [16] 650 patients admitted to surgical and neonatal ICUs Prospective cohort 1 positive blood culture (peripheral vein or iv catheter) plus histological documentation of invasive candidiasis or endophtalmitis, or 2 positive blood cultures from peripheral vein or 1 positive blood culture from peripheral vein and 1 from iv catheter, both with identical species Oropharynx or trachea and stomach Intensity of colonization was a risk factor for candidemia by multivariate analysis (species not mentioned)
Martino, 1989 [12] 424 patients with neutropenia Retrospective analysis 2 positive blood cultures from peripheral vein; if blood sample was taken from an iv catheter, candidemia was considered present if clinical signs of infection persisted after catheter removal Nasal, oral vaginal, and anal swabs, urine, stool, and sputum Colonization at multiple, noncontiguous sites was associated with increased risk for hematogenous candidiasis
Martino, 1994 [13] 139 patients with neutropenia Prospective cohort 1 positive blood culture from peripheral vein; if blood sample was taken from an iv catheter, candidemia was considered present if clinical signs of infection persisted after catheter removal Nasal, oral vaginal, and anal swabs, urine, stool, and sputum Invasive candidiasis occurred more frequently in patients colonized at multiple, noncontiguous sites
Colombo, 1996 [17] 1 AIDS patient with C. albicans candidemia Case report Not mentioned Oral and anal swabs and urine Blood, urine, oral cavity, and stool strains showed identical DNA profiles (PFGE)
Reagan, 1990 [18] 16 patients with positive blood and surveillance cultures (various sites) Prospective cohort Not mentioned Oral swab, urine, and stool 5 patients with blood and stool cultures positive had strains available for testing and showed identical DNA patterns (RED with Southern hybridization)
Klempp-Selb, 2000 [19] 6 patients with candidemia Prospective 1 positive blood culture, source of blood collection not mentioned Genital swabs, stool, urine, and tracheal secretion or sputum Stool culture was available for 2 patients; 1 had similar DNA pattern in blood and stool (PFGE)
Saiman, 2000 [20] 2847 patients in NICU, 35 with candidemia Prospective cohort 1 positive blood culture, source of blood collection not mentioned Perianal and rectal swabs C. albicans (63%), C. parapsilosis (29%); all 14 blood and colonization strains available for testing (species not specified) showed identical DNA pattern (PFGE); colonization preceded candidemia in 15 of 35 patients

NOTE. AML, acutemyeloid leukemia; ICU, intensivecare unit; NICU, neonatalintensive care unit; PFGE,pulsed-field gel electrophoresis; REA,restriction enzyme analysis; RED,restriction endonuclease digestion ofchromosomal DNA; RFLP, restriction-fragmentlength polymorphism; TPN, totalparenteral nutrition.

Table 2. CITED IN TEXT | TYPESET IMAGE |

Evaluation of density and sequence of colonization and molecular-relatedness studies in specific patient populations.

Study subject Patients with
neutropenia
Neonates Surgical
patients
Patients
with burns
Density of colonization
Gut Yes [10, 12, 13, 22] Yes [16] Yes [16] NAS
Skin Yes [22] NAS NAS NAS
Sequence of colonization and subsequent infection
Gut Yes [14, 18, 22] Yes [16, 20, 26a] NAS NAS
Skin NAS NAS NAS NAS
Molecular studies
Gut Yes [18] Yes [20, 26a] NAS NAS
Skin NAS [26]a NAS NAS

NOTE. NAS, no availablestudies.
a Relationship of the infectionto the site waspossible: 1 patient hadcolonization of gut andskin, but there wasnot a study ofsequence of colonization toidentify primary source.

Table 3. CITED IN TEXT | TYPESET IMAGE |

Koch’s postulates adapted to source(s) of candidemia.

Postulate Skin Gastrointestinal tract
Development of candidemia after the inoculation of the organism at the likely source of natural infection
Experimental NAS Yes [2834]
Clinical NAS Yes [35]
Eradication of Candida species from the likely source results in reduction of the incidence of candidemia NAS Yes [37]
Sequence of colonization at likely source and subsequent development of candidemia NAS Yes, with molecular studies [18, 25]; yes, but without molecular studies [1416, 20, 35]
Density of colonization at likely source and development of candidemia NAS Yes [1013, 16]
Recovery of Candida species from the source in almost every case of candidemia Yes, but without molecular studies [3, 4, 21] Yes, with molecular studies [20]; yes, but without molecular studies [1013, 1622, 25]
Supporting molecular-relatedness studies [26]a Yes [1720, 26a]

NOTE. NAS, noavailable studies.
a Relationship of theinfection to the sitewas possible: 1 patienthad colonization of gutand skin, but therewas not a studyof sequence of colonizationto identify primary source.

0 views


Revisiting the Source of Candidemia: Skin or Gut?

Hiç yorum yok:

Yorum Gönder