JCM Accepted Manuscript Posted Online 11 February 2015

J. Clin. Microbiol. doi:10.1128/JCM.03239-14

Copyright © 2015, American Society for Microbiology. All Rights Reserved.

1 Candida Colonization as a Risk Marker for Invasive Candidiasis in Mixed Medical-

2 Surgical ICUs: Development and Evaluation of a Simple, Standard Protocol.

4 Anna F. Lau1,a,*, Masrura Kabir1, Sharon C-A. Chen1,2, E. Geoffrey Playford3, Deborah J.

5 Marriott4, Michael Jones5, Jeffrey Lipman6, Emma McBryde7, Thomas Gottlieb8, Winston

6 Cheung9, Ian Seppelt10, Jonathan Iredell1,11, Tania C. Sorrell1,11

7
1
8 Centre for Infectious Diseases and Microbiology, Westmead Millennium Institute, University

9 of Sydney, Westmead Hospital, Westmead, New South Wales, 2145; 2 Centre for Infectious

10 Diseases and Microbiology Laboratory Services, ICPMR – Pathology West, Westmead

11 Hospital, New South Wales, 2145; 3Infection Management Services, Princess Alexandra

12 Hospital, Brisbane, Queensland, 4102; 4Department of Microbiology and Infectious Diseases,

13 St Vincent’s Hospital, Darlinghurst, New South Wales, 2010; 5Psychology Department,

14 Macquarie University, North Ryde, New South Wales, 2109; 6Department of Intensive Care

15 Medicine, Royal Brisbane and Women’s Hospital, The University of Queensland, Herston,

16 Queensland, 4029; 7Victorian Infectious Diseases Service, Royal Melbourne Hospital,

17 University of Melbourne, Parkville, Victoria, 3010; 8Department of Microbiology and

18 Infectious Diseases, Concord Hospital, Concord, New South Wales, 2137; 9Department of

19 Intensive Care Medicine, Concord Repatriation General Hospital, Concord, New South Wales,

20 2139; 10Department of Intensive Care Medicine, Nepean Hospital, University of Sydney,

21 Penrith, New South Wales, 2751; 11Marie Bashir Institute for Infectious Diseases and

22 Biosecurity, University of Sydney, Westmead, New South Wales, 2145, Australia.

23

Page 1 of 21
 a
24 Current address: Department of Laboratory Medicine, Clinical Center, National Institutes of

25 Health, Bethesda, MD, 20892;

26

27 Keywords

28 Candida colonization

29 Intensive Care Units

30 Determinants of risk for Invasive Candidiasis

31 Protocol evaluation

32

33 Running title: Candida colonization risk model for disease

34

35 *Corresponding author:

36 Anna F. Lau, Ph.D., D(ABMM)

37 Department of Laboratory Medicine

38 Clinical Center

39 National Institutes of Health

40 Bethesda, MD, 20814

41 Email: Anna.Lau@nih.gov

42

Page 2 of 21
43 ABSTRACT

44 Colonization with Candida species is an independent risk factor for invasive candidiasis (IC)

45 but the minimum and most practicable parameters for prediction of IC have not been

46 optimized. We evaluated Candida colonization in a prospective cohort of 6,015 non-

47 neutropenic, critically ill patients. Throat, perineum, and urine were sampled 72 h post-ICU

48 admission and twice weekly until discharge or death. Specimens were cultured onto

49 chromogenic agar and a subset underwent molecular characterization. Sixty-three patients

50 (86%) who developed IC were colonized prior to infection; 61 (97%) were positive within the

51 first two time points. The median time from colonization to IC was seven days (range 0-35).

52 Colonization at any site was predictive of IC, with the risk of infection highest for urine

53 colonization (RR=2.25) but most sensitive (98%) for throat and/or perineum colonization.

54 Colonization of ≥2 sites and heavy colonization of ≥1 site were significant independent risk

55 factors for IC (RR=2.25 and RR=3.7, respectively); increasing specificity to 71-74% but

56 decreasing sensitivity to 48-58%. Molecular testing would have prompted a resistance-driven

57 decision to switch away from fluconazole in only 11% of patients infected with C. glabrata,

58 based upon species-level identification alone. PPVs were low (2-4%) and NPVs high (99-

59 100%) regardless of which parameters were applied. In the Australian ICU setting, culture of

60 throat and perineum within the first two time points after ICU admission will capture 84%

61 (61/73 patients) of subsequent IC cases. These optimized parameters, in combination with

62 clinical risk factors, should strengthen development of a setting-specific risk predictive model

63 for IC.

Page 3 of 21
64 INTRODUCTION

65 Invasive candidiasis (IC), particularly candidemia, is a significant cause of mortality in

66 critically ill patients, accounting for almost a third of nosocomial infections in intensive care

67 units (ICUs) (1, 2). Early antifungal therapy reduces IC-related mortality and approximately

68 halves the incidence of IC (3-8). Untargeted prophylactic antifungal use, however, is

69 expensive, has the potential to cause adverse drug reactions, and may select for resistant fungal

70 species (9).

71 Clinical risk prediction rules that identify high-risk patients most likely to benefit from

72 prophylaxis have been developed (10-16), but these studies have used different sets of

73 predictors including (a) clinical risk factors only; or (b) clinical risk factors in combination

74 with colonization indices. These rules have been applied in various settings, and have included

75 assessment at different times post-admission to the ICU and different types of ICU (surgical

76 ICUs only or mixed medical/surgical ICUs – defined as units that house both medical and

77 surgical populations). These differences may explain why predictive models and algorithms

78 have performed poorly outside their derivative populations (17).

79 Since IC is preceded by colonization of mucosal surfaces with the infecting strain (14,

80 18-20), and colonization is an independent risk factor for IC (14, 19, 21, 22), it is logical that it

81 be incorporated into predictive models (15, 16). Using data from mixed medical/surgical ICUs

82 in Australia, we demonstrated that the post-hoc addition of colonization parameters to two

83 published clinical risk factor-only predictive models improved their performance

84 characteristics (17).

85 Although several studies examining colonization as a risk factor for IC have been

86 published, the methodology has not been standardized and most were confined to surgical

Page 4 of 21
87 ICUs (14, 23-25). Inter-study differences existed between sites of sampling, use of

88 convenience and/or specified sample sites, timing of first samples and frequency of sampling,

89 quantification of colonization density, culture media and methods used for species

90 identification. Furthermore, nucleic acid based methods of fungal identification were not

91 included.

92 A systematic evaluation of Candida colonization in a prospective multi-centre cohort

93 study of Australian ICU patients was conducted between June 2007 and January 2012. The

94 overall project aimed to develop and validate a clinical/colonization-based risk predictive

95 model suitable for use in mixed Australian medical and surgical ICUs. In this study, we

96 determined the utility of culture on CHROMagar for semi-quantitative estimates of Candida

97 colonization (with parallel molecular testing performed on a subset of patients), the optimal

98 timing of sample collection, and the minimum number of sites to be sampled. Our overall aim

99 was to develop a protocol to optimize detection of colonization status for inclusion into risk

100 predictive models of IC in critically ill patients.

101

102 MATERIALS & METHODS

103 Study Design

104 A total of 6,015 non-neutropenic critically-ill patients admitted consecutively to mixed medical

105 and surgical ICUs in seven Australian hospitals [Westmead, Sydney (n = 1,163); Princess

106 Alexandra, Brisbane (n = 1,231); Royal Brisbane Women’s and Children’s, Brisbane (n =

107 896); St Vincent’s Hospital, Sydney (n = 611); Royal Melbourne Hospital, Melbourne (n =

108 1,227); Concord Hospital, Sydney (n = 371); and Nepean Hospital, Sydney (n = 516)] for at

109 least 72 h were studied prospectively from June 15, 2007 to January 1, 2012. Patients with

Page 5 of 21
110 neutropenia (absolute neutrophil count <0.5 x 109/L) within the first 72 h of ICU admission

111 were excluded. Approval for the study was obtained from each respective Human Research

112 Ethics Committee.

113

114 Surveillance Specimen Collection

115 Published literature (23) and a preliminary study at the Westmead Hospital site suggested that

116 the throat, perineum (rather than groin or rectum) and urine were the most accessible, feasible

117 and highest yield sites for the assessment of Candida colonization. Sampling techniques were

118 standardized across sites. In brief, standard sterile red top culturette swabs containing liquid

119 Amies media (Copan, Murrieta, CA, USA) were used to sample the oropharynx and perineum

120 (just anterior to the anus), and catheter (or mid-stream) urine specimens were obtained.

121 Specimens were obtained independently of any clinically indicated samples. Preliminary

122 experiments (performed prior to the commencement of this study) showed that sampling the

123 throat and perineum with dry swabs followed by immersion in 2 mL nutrient broth containing

124 10% glycerol at the bedside, was a suitable but less practical alternative (data not shown).

125 Eswabs (Copan Diagnostics; and Becton, Dickinson and Co., Franklin Lakes, NJ) that allow

126 for automated specimen processing and improved microbial transport and recovery (26, 27)

127 were not evaluated in this study due to their lack of availability in Australia during the study

128 period. All three surveillance sites were sampled at 72 h following ICU admission and twice

129 weekly thereafter until ICU discharge or death.

130

131 Candida Surveillance Cultures

Page 6 of 21
132 Swabs of the throat and perineum and 10 μL of urine were plated onto Candida chromogenic

133 agar (CHROMagar™, Paris, France) using a half-plate quantitative urine streaking method.

134 Plates were incubated at 35C for 48 h. Candida albicans, Candida tropicalis and Candida

135 krusei were differentiated by color as described by the manufacturer. All other yeast isolates

136 (nearly all were Candida glabrata or Candida parapsilosis as confirmed by MT-PCR) were

137 collectively designated as ‘other Candida spp.’. Growth of each morphologically distinct yeast

138 was enumerated semi-quantitatively and the extent of growth was classified as none, light (<10

139 colonies), moderate (10-100 colonies), or heavy (>100 colonies).

140

141 Multiplex-tandem PCR (MT-PCR)

142 Swabs of the throat and perineum obtained from the first 205 patients sampled at Westmead

143 Hospital underwent MT-PCR analysis in addition to culture. Following culture, the swabs were

144 placed into 3 mL saline containing 10% nutrient broth (Oxoid, Adelaide, South Australia) and

145 stored at 4C until nucleic acid was extracted. DNA was isolated from 1 mL of the suspension

146 using the nucliSENS easyMAG instrument (bioMerieux, Baulkham Hills, New South Wales,

147 Australia) according to the manufacturer’s instructions. The elution volume was 110 μL. As

148 previously described (28-30), the MT-PCR method enabled identification of seven Candida

149 species including C. albicans, Candida dubliniensis, C. glabrata, Candida guilliermondii, C.

150 krusei, C. parapsilosis complex and C. tropicalis, as well as Saccharomyces cerevisiae. An

151 internal positive control (artificial DNA) was included with each specimen to test for PCR

152 inhibition, and a negative water control was included in each run to monitor contamination.

153 Urine specimens were not tested by MT-PCR as the application of this technology to urine

154 samples had not been validated during the study period.

Page 7 of 21
155

156 Definitions

157 Colonization was defined as the isolation of a Candida species from at least one surveillance

158 site. Positive samples collected as part of routine clinical management were not included in the

159 analysis. The colonization index (CI) was defined as the ratio of the number of sites colonized

160 by Candida spp. to the number of body sites surveyed (14). The corrected colonization index

161 (CCI) was defined as the ratio of the number of sites heavily colonized by Candida spp. to the

162 number of body sites surveyed (14). In this present study, the CCI threshold was lowered from

163 ≥0.5 (14) to ≥0.3 to accommodate the reduced number of sites surveyed (three as opposed to

164 five).

165 Surveillance specimens were collected at 72 h following ICU admission and twice

166 weekly thereafter until ICU discharge or death. Time points 1, 2, and 3 were defined as three to

167 four days, six to eight days, and nine to 11 days post-ICU admission, respectively.

168 IC episodes were defined using modified European Organisation for Research and

169 Treatment of Cancer/Mycoses Study Group (EORTC/MSG) criteria (12, 17). Patients with

170 candiduria alone were excluded because this was not considered an invasive infection. IC

171 episodes detected at least 72 h following ICU admission and within 72 h after ICU discharge

172 were considered ICU-acquired.

173

174 Statistical Analysis

175 Surveillance culture results from ICU patients who did or did not develop IC were compared

176 with respect to categorical variables that included the anatomical site of colonization, the

177 colonization indices (CI and CCI), density of colonization, number of colonizing species and

Page 8 of 21
178 the time of surveillance. The relative risk of developing IC based on colonization status was

179 calculated using Fisher’s exact test. A p-value <0.05 was considered significant. The predictive

180 performance characteristics (including sensitivity, specificity, positive predictive value (PPV)

181 and negative predictive value (NPV)) were calculated for each variable studied. MT-PCR and

182 culture were compared for the detection of colonization status and the time to detect

183 colonization status using Fisher’s exact test. All statistical calculations were performed using

184 MedCalc (www.medcalc.org).

185

186 RESULTS

187 Patient Characteristics

188 Of the 6,015 patients studied, 73 (1%) developed IC in the ICU. Fifty-eight (79%) of these had

189 proven infection - candidemia (n = 43) or non-candidemic IC (n = 15); the remaining 15 (21%)

190 patients were classified as probable cases (12, 17). Infections were caused by C. albicans (n =

191 45, 62%), C. glabrata (n = 8, 11%), C. tropicalis (n = 8, 11%), C. parapsilosis (n = 6, 8%),

192 Candida spp. (n = 5, 7%), and C. krusei (n = 1, 1%). The median time from ICU admission to

193 the development of ICU-acquired IC was 11 days (range 4-41, average 14 days).

194 Systemic (oral or parenteral) antifungal drugs had been received by five patients (7%)

195 prior to ICU admission (days -7 to 0) or time point 1 (days 3 to 4); an additional seven (10%)

196 between days 5 and 7; and an additional six (8%) between days 8 and 10. Fluconazole therapy

197 was administered in 16 cases of IC, and caspofungin in two cases. Twelve (16%) of 73 patients

198 who developed IC had received empirical antifungal therapy (none had received antifungal

199 prophylaxis) prior to diagnosis: nine (75%) of the 12 had been treated with fluconazole (seven

200 with infection due to C. albicans, one C. lusitaniae, and one C.tropicalis).

Page 9 of 21
201

202 Colonization (Culture) versus IC: Anatomical Site of Colonization

203 Table 1 summarizes the risk of developing IC by site of colonization over the first three time

204 points. A total of 3,511 (60%), 2,025 (63%) and 1,162 (64%) patients were colonized at time

205 points 1 (days three to four post-ICU admission), 2 (days six to eight post-ICU admission), and

206 3 (days nine to 11 post-ICU admission), respectively.

207 Of the 73 patients who developed IC, Candida colonization was detected in 63 patients

208 (86%) prior to infection – 56 (89%) of these at time point 1, an additional five at time point 2,

209 and a further two at time point 3. Patients with detected colonization in the throat, perineum, or

210 urine at time point 1 were all at increased risk of developing IC (RR = 2.35, p-value = 0.002,

211 sensitivity 78%) with colonization of the urine being associated with the highest relative risk of

212 IC (RR = 2.25) compared with throat and perineum (RR = 2.04 and 1.76, respectively; Table

213 1). At time points 2 and 3, patients colonized in the throat or in at least one of the three sites

214 sampled (CI ≥ 0.3) were at increased risk of developing IC. NPVs were consistently high (99-

215 100%; Table 1).

216 Of the 63 patients in whom Candida colonization was detected prior to infection, 38

217 were colonized in at least two sites, most commonly the throat and perineum (n = 32, 84%).

218 Nineteen (30%) patients with IC were colonized in the throat alone; five (8%) in the perineum

219 alone; and one (2%) in the urine alone.

220 The median time from the first positive surveillance specimen to the first positive

221 clinical culture was seven days (range 0-35, average nine days).

222 Ten patients (14%) had negative surveillance cultures prior to the development of IC.

223 Seven (70%) of the 10 patients had only one set of surveillance cultures obtained before

Page 10 of 21
224 discharge from the ICU. The remaining three had two, five or seven sets of surveillance

225 cultures obtained; the time from the last surveillance culture to the development of IC was two,

226 15, and 19 days, respectively. All but one of the 10 patients who had negative surveillance

227 cultures prior to developing IC had undergone surgery just prior to or during ICU admission. In

228 six instances where the type of surgery was specified, five had involved penetration of the

229 gastro-intestinal tract.

230

231 Colonization Index (CI), Corrected Colonization Index (CCI) and Invasive Candidasis (IC)

232 Of the 58% patients (n = 3,511) where Candida colonization was detected at the first time

233 point, 1,671 (48%) had Candida spp. isolated from at least two surveillance sites (CI ≥ 0.5;

234 Table 2). Almost half (n = 35, 48%) of the patients who developed IC had a CI ≥ 0.5 (RR =

235 2.25, p-value = 0.0005; Table 2).

236 The sensitivity, specificity, and risk of developing IC increased slightly when heavy

237 colonization was taken into account (CCI ≥ 0.3; Table 2). At time point 1, 42 (58%) patients

238 who developed IC were heavily colonized in at least one site (RR = 3.7, p-value <0.0001).

239 Specificity increased (74% to 92%) when two sites were heavily colonized, but this in turn

240 reduced sensitivity from 58% to 21% (Table 2).

241

242 Prevalence of Candida Colonization and Incidence of IC Over Time in ICU

243 Detection of Candida colonization at any site was maintained at 60% over the first three time

244 points (Figure 1). Detection of perineum colonization increased the most over the first two

245 screens while the likelihood of Candida colonization in the urine increased steadily with

246 prolonged ICU stay. Detection of Candida colonization in the throat was highest at time point

Page 11 of 21
247 1 and decreased with time in the ICU. Of the 73 infected patients, six, 17, and 15 patients

248 developed IC within time points 1, 2, and 3, respectively, leading to a cumulative incidence of

249 IC of 8%, 32%, and 52% within the first three time points (Figure 1).

250

251 Distribution of Colonizing Species

252 Most patients were colonized with C. albicans (n = 2,854, 81%), followed by “other” Candida

253 spp. (n = 1,059, 30%), C. tropicalis (n = 343, 10%), and C. krusei (n = 123, 4%). The majority

254 of patients (80%) were colonized with a single Candida species. Mixed cultures generally

255 consisted of two species (n = 662), typically C. albicans with C. glabrata or C. parapsilosis (n

256 = 467, 67%).

257 All 63 colonized patients who developed IC were colonized with the species that

258 subsequently caused infection. Twenty (32%) patients were colonized with more than one

259 Candida species; the dominant colonizing species caused subsequent infection. The remaining

260 43 patients were colonized with one species (33 C. albicans, eight ‘other’ Candida spp., and

261 one each of C. krusei and C. tropicalis). The number of colonizing species was not associated

262 with the development of IC (data not shown).

263

264 Comparison of MT-PCR and Culture for the Detection of Candida Colonization in Patients

265 with or without IC

266 In addition to culture, swabs of the throat and perineum from 205 patients underwent MT-PCR

267 analysis. The median time of IC acquisition in this patient cohort was five days (range 4-19

268 days) post-ICU admission. MT-PCR detected Candida DNA in an additional seven patients (n

269 = 173, 84%) compared with culture (n = 166, 81%). No significant difference was found

Page 12 of 21
270 between MT-PCR and culture for detecting colonization status, nor in the sampling period in

271 which colonization was detected (91% and 93% of colonized patients were positive by MT-

272 PCR and culture, respectively, at time point 1).

273 While only C. albicans, C. tropicalis and C. krusei were distinguished with certainty on

274 CHROMagar (based on manufacturer’s protocol), a further five species, including C.

275 dubliniensis, C. glabrata, C. guilliermondii, C. parapsilosis and S. cerevisiae were identified

276 by MT-PCR. In this cohort, C. albicans was cultured from 134 (65%) patients while C.

277 albicans DNA was only detected in 124 (60%) patients; C. dubliniensis DNA was detected in

278 these 10 discrepant cases. Non-albicans Candida spp. detected by MT-PCR included C.

279 dubliniensis (n = 27, 13%), C. glabrata (n = 29, 14%), C. guilliermondii (n = 3, 1%), C. krusei

280 (n = 12, 6%), C. parapsilosis (n = 17, 8%), C. tropicalis (n = 15, 7%) and S. cerevisiae (n = 35,

281 17%). MT-PCR provided species identification in 73 cases where the yeast could not be

282 speciated by CHROMagar.

283 The turn-around time was faster for MT-PCR (3 h; but in practice, 8-24 h allowing for

284 specimen transport and incorporation into routine laboratory workload) than the 48 h required

285 to obtain results from culture-based surveillance.

286

287 DISCUSSION

288 In this prospective multicentre study, we systematically determined the most effective and

289 practical protocol to measure colonization status and predict IC in non-neutropenic patients in

290 mixed-medical/surgical Australian ICUs. As expected, colonization was detected in most

291 patients (86%) prior to development of IC. Standardized cultures of the throat and perineum at

292 time points 1 and 2 captured 97% (61/63) of colonized patients who developed IC, a median of

Page 13 of 21
293 seven days prior to the diagnosis of IC (designated as the diagnostic specimen collection date).

294 Notably, 14% of patients (n = 10) had negative surveillance cultures prior to the diagnosis of

295 IC, similar to a Spanish study of mixed medical/surgical ICUs (19). Based on our prior

296 observation (17), we expect that this subgroup will be captured when clinical parameters such

297 as recent gastrointestinal surgery are built into derivation of our model.

298 Prior Candida colonization is a major risk factor for developing IC; however, the sites

299 surveyed have varied considerably (14, 19, 21, 31-33). Furthermore, proponents of “clinical

300 only” prediction rules have argued for exclusion of colonization data because the process is

301 resource intensive and adds to laboratory costs. Our study determined that collection of

302 samples from three sites alone (throat, perineum and urine) was practical and feasible, but not

303 necessary for routine surveillance since 98% (62/63) of colonized patients who subsequently

304 developed IC were captured by throat and perineum surveillance cultures. Contrary to our

305 findings, others have indicated that the mouth is not a good site for routine Candida

306 surveillance due to lower positivity rates (23); the reason for this difference is not clear but

307 may be linked to changes in the oropharyngeal niche and differences in oral hygiene practices

308 among ICUs.

309 To optimize the timing of surveillance cultures, we compared results of cultures at time

310 points 1, 2, and 3 (Tables 1 and 2; Figure 1). In a trend similar to others (21, 34), the majority

311 of colonized patients who subsequently developed IC were positive on the first screen (n = 56,

312 89%) indicating that new colonizing strains are acquired in a minority of patients after 72 h in

313 the ICU (21). Detection of throat colonization unexpectedly decreased with time in the ICU

314 (Figure 1). Selective decontamination of the digestive tract (SDD) was not used in the present

315 cohort, although changes to the oropharyngeal flora upon ICU admission are expected

Page 14 of 21
316 considering variables such as patient intubation status, nutritional components, and different

317 oral hygiene methods. Since 97% (61/63) of colonized patients who developed IC had positive

318 surveillance in the throat and/or perineum within the first two time points and that the median

319 time from ICU admission to IC was 11 days, we recommend that surveillance cultures taken at

320 72 h and seven days post-ICU admission are sufficient for determining Candida colonization

321 status for inclusion into risk predictive models.

322 Multifocal colonization (at least two sites, CI > 0.5) and heavy colonization (at least

323 one site, CCI ≥ 0.3) both were significant independent risk factors in our cohort (Table 2), but

324 the decrease in sensitivity (78% to 58%) when CCI was applied indicates its poor utility for

325 predicting IC in our mixed medical/surgical ICU population. In the original study by Pittet et

326 al. (14), the CCI resulted in a 100% PPV compared with the 2-4% in our study (Table 2). This

327 large discrepancy is likely because the CCI (in the Pittet study) was derived from a small

328 cohort of critically ill surgical patients that were selected as “high risk for IC” based on

329 multiple clinical risk factors, and that colonization status was determined through a variable

330 number of “convenience” samples received in the laboratory at any time prior to the

331 development of IC (14). This contrasts to populations, such as ours, that were selected on the

332 basis of ICU length of stay, resulting in low PPV and high NPV. Defined sampling protocol

333 and the mixed medical/surgical populations in our ICUs may have also contributed to the

334 considerable difference in CCI performance.

335 As in other studies (20, 23), low PPVs were obtained when Candida colonization

336 parameters were applied as the sole criterion for predicting IC – mainly due to its low

337 incidence (1-2% as in our study). With a median time from ICU admission to infection of 11

338 days (range 4-41), initial screening at 72 h post-ICU admission was necessary to predict the

Page 15 of 21
339 25% of cases that would have been missed had we used the 7 day entry point proposed by

340 others (16, 22). Excellent NPVs were generated for all Candida colonization parameters

341 investigated, indicating that patients with negative surveillance cultures are highly unlikely to

342 develop disease and would not benefit from early antifungal intervention. This is turn would

343 reduce hospital costs, drug related toxicities and the likelihood of emerging antifungal

344 resistance.

345 In a trend similar to European and North American studies (2, 19, 21, 22, 34), C.

346 albicans was the dominant colonizing (81%) and infecting (77%) species. A rising prevalence

347 of IC due to non-albicans Candida species has been attributed to azole prophylaxis and

348 empiric treatment (6, 35-38). Others, however, have reported negligible effects of systemic

349 antifungal use on colonization status (including C. glabrata, specifically) (24, 25). In the

350 present study, only 12 of the 63 colonized IC patients had received antifungal therapy (all

351 empiric; 10 fluconazole) hence associations could not be tested.

352 MT-PCR was evaluated on a subset of patients (n = 205). Overall, detection of

353 colonization status was similar between MT-PCR and culture (81% and 84%, respectively)

354 although instances of culture misidentification were found (namely, C. dubliniensis

355 misidentified as C. albicans by CHROMagar). Other than providing species identification in

356 cases where the yeast could not be identified on CHROMagar (14% C. glabrata and 8% C.

357 parapsilosis), MT-PCR did not provide sufficient additional information to justify faster results

358 and five-fold cost increase. Furthermore, yeast speciation can now be performed rapidly and

359 inexpensively from culture using technologies such as MALDI-TOF MS (this was not

360 available in clinical laboratories in Australia during the study period).

Page 16 of 21
361 A potential limitation of this study was its multi-centre characteristic that encompassed

362 possible variation in diagnostic and therapeutic decision-making. Australian ICUs, however,

363 are comparable in case mix and management approaches, and a centre effect was not

364 demonstrated when data from individual hospitals were analysed and compared (data not

365 shown). In addition, a large sample size was employed to minimize disparity and improve the

366 generalizability of results. The possibility of systematic bias was eliminated by the inclusion of

367 patients who were admitted consecutively into the ICUs.

368 In conclusion, after a large-scale, multi-centred, prospective, systemic evaluation of a

369 variety of Candida colonization parameters, we propose a simplified protocol to optimize

370 detection of colonization status for inclusion into risk predictive models of IC in critically ill

371 patients. This protocol, consisting of throat and perineum surveillance sampling at 72 h post-

372 ICU admission and three to four days later, will capture most patients likely to develop IC. It

373 should be noted that colonization might not be detected prior to infection in a subset of patients

374 who will develop IC (14% in our cohort), but we expect that these patients will be captured by

375 incorporation of clinical risk factors along with these colonization data into our final risk

376 predictive model.

377

378 ACKNOWLEDGEMENTS

379 We particularly acknowledge Maureen Puhlmann, Research Nurse at the Concord Hospital

380 site, for her enthusiasm, hard work and support of this project. Maureen passed away during

381 the course of the study and is a great loss to her profession.

382 We also acknowledge the following contributors to this study: Claire Kesby, June

383 Kelly, and Catriona Halliday, Westmead Hospital, Sydney; Neil Underwood and Joan

Page 17 of 21
384 Faoagali, Princess Alexandra Hospital, Brisbane; Quoc Nguyen and Sam Rudham from St

385 Vincent’s Hospital, Sydney; Graeme Nimmo, Alex Dank, Melissa Lassig-Smith, Janine Stuart,

386 and Renae Deans from Royal Brisbane Women’s and Children’s Hospital, Brisbane; Leanne

387 Redl and Caroline Marshall from Royal Melbourne Hospital, Melbourne; Jocelynne McRae,

388 David Milliss, Evanthia Tambosis, and Charlotte Webster from Concord Hospital, Sydney;

389 James Branley and Sarah Whereat from Nepean Hospital, Sydney; staff at the Victorian

390 Infectious Diseases Reference Laboratories, Melbourne; and statistician Brian O’Toole from

391 the Centre for Infectious Diseases and Microbiology Public Health, Westmead Hospital,

392 Sydney.

393 TCS is a Sydney Medical School Foundation fellow.

394 This project was funded by grants from the National Health and Medical Research

395 Council of Australia (Project Grant #512307 and Centre of Clinical Research Excellence Grant

396 #264625) and an Australian Universities Post-graduate Award to AFL.

397

398 REFERENCES

399 1. Vazquez JA. 2010. Invasive fungal infections in the intensive care unit. Semin. Respir.
400 Crit. Care Med. 31:79-86.
401 2. Leroy O, Gangneux JP, Montravers P, Mira JP, Gouin F, Sollet JP, Carlet J,
402 Reynes J, Rosenheim M, Regnier B, and Lortholary O. 2009. Epidemiology,
403 management, and risk factors for death of invasive Candida infections in critical care: a
404 multicenter, prospective, observational study in France (2005-2006). Crit. Care Med.
405 37:1612-1618.
406 3. Garey KW, Rege M, Pai MP, Mingo DE, Suda KJ, Turpin RS, and Bearden DT.
407 2006. Time to initiation of fluconazole therapy impacts mortality in patients with
408 candidemia: a multi-institutional study. Clin. Infect. Dis. 43:25-31.
409 4. Morrell M, Fraser VJ, and Kollef MH. 2005. Delaying the empiric treatment of
410 Candida bloodstream infection until positive blood culture results are obtained: a
411 potential risk factor for hospital mortality. Antimicrob. Agents. Chemother. 49:3640-
412 3645.

Page 18 of 21
413 5. Cruciani M, de Lalla F, and Mengoli C. 2005. Prophylaxis of Candida infections in
414 adult trauma and surgical intensive care patients: a systematic review and meta-
415 analysis. Intensive Care Med. 31:1479-1487.
416 6. Playford EG, Webster AC, Sorrell TC, and Craig JC. 2006. Antifungal agents for
417 preventing fungal infections in non-neutropenic critically ill and surgical patients:
418 systematic review and meta-analysis of randomized clinical trials. J. Antimicrob.
419 Chemother. 57:628-638.
420 7. Shorr AF, Chung K, Jackson WL, Waterman PE, and Kollef MH. 2005.
421 Fluconazole prophylaxis in critically ill surgical patients: a meta-analysis. Crit. Care
422 Med. 33:1928-1935; quiz 1936.
423 8. Vardakas KZ, Samonis G, Michalopoulos A, Soteriades ES, and Falagas ME.
424 2006. Antifungal prophylaxis with azoles in high-risk, surgical intensive care unit
425 patients: a meta-analysis of randomized, placebo-controlled trials. Crit. Care Med.
426 34:1216-1224.
427 9. Hsueh PR, Graybill JR, Playford EG, Watcharananan SP, Oh MD, Ja'alam K,
428 Huang S, Nangia V, Kurup A, and Padiglione AA. 2009. Consensus statement on the
429 management of invasive candidiasis in Intensive Care Units in the Asia-Pacific Region.
430 Int. J. Antimicrob. Agents 34:205-209.
431 10. Ostrosky-Zeichner L, Sable C, Sobel J, Alexander BD, Donowitz G, Kan V,
432 Kauffman CA, Kett D, Larsen RA, Morrison V, Nucci M, Pappas PG, Bradley
433 ME, Major S, Zimmer L, Wallace D, Dismukes WE, and Rex JH. 2007.
434 Multicenter retrospective development and validation of a clinical prediction rule for
435 nosocomial invasive candidiasis in the intensive care setting. Eur. J. Clin. Microbiol.
436 Infect. Dis. 26:271-276.
437 11. Dupont H, Bourichon A, Paugam-Burtz C, Mantz J, and Desmonts JM. 2003. Can
438 yeast isolation in peritoneal fluid be predicted in intensive care unit patients with
439 peritonitis? Crit. Care Med. 31:752-757.
440 12. Paphitou NI, Ostrosky-Zeichner L, and Rex JH. 2005. Rules for identifying patients
441 at increased risk for candidal infections in the surgical intensive care unit: approach to
442 developing practical criteria for systematic use in antifungal prophylaxis trials. Med.
443 Mycol. 43:235-243.
444 13. Ibanez-Nolla J, Nolla-Salas M, Leon MA, Garcia F, Marrugat J, Soria G, Diaz
445 RM, and Torres-Rodriguez JM. 2004. Early diagnosis of candidiasis in non-
446 neutropenic critically ill patients. J. Infect. 48:181-192.
447 14. Pittet D, Monod M, Suter PM, Frenk E, and Auckenthaler R. 1994. Candida
448 colonization and subsequent infections in critically ill surgical patients. Ann. Surg.
449 220:751-758.
450 15. Leon C, Ruiz-Santana S, Saavedra P, Almirante B, Nolla-Salas J, Alvarez-Lerma
451 F, Garnacho-Montero J, and Leon MA. 2006. A bedside scoring system ("Candida
452 score") for early antifungal treatment in nonneutropenic critically ill patients with
453 Candida colonization. Crit. Care Med. 34:730-737.
454 16. Leon C, Ruiz-Santana S, Saavedra P, Galvan B, Blanco A, Castro C, Balasini C,
455 Utande-Vazquez A, Gonzalez de Molina FJ, Blasco-Navalproto MA, Lopez MJ,
456 Charles PE, Martin E, and Hernandez-Viera MA. 2009. Usefulness of the "Candida
457 score" for discriminating between Candida colonization and invasive candidiasis in

Page 19 of 21
458 non-neutropenic critically ill patients: a prospective multicenter study. Crit. Care Med.
459 37:1624-1633.
460 17. Playford EG, Lipman J, Kabir M, McBryde ES, Nimmo GR, Lau A, and Sorrell
461 TC. 2009. Assessment of clinical risk predictive rules for invasive candidiasis in a
462 prospective multicentre cohort of ICU patients. Intensive Care Med. 35:2141-2145.
463 18. Mann PA, McNicholas PM, Chau AS, Patel R, Mendrick C, Ullmann AJ, Cornely
464 OA, Patino H, and Black TA. 2009. Impact of Antifungal Prophylaxis on
465 Colonization and Azole Susceptibility of Candida species. Antimicrob. Agents
466 Chemother.
467 19. Jorda-Marcos R, Alvarez-Lerma F, Jurado M, Palomar M, Nolla-Salas J, Leon
468 MA, and Leon C. 2007. Risk factors for candidaemia in critically ill patients: a
469 prospective surveillance study. Mycoses 50:302-310.
470 20. Pelz RK, Lipsett PA, Swoboda SM, Diener-West M, Hammond JM, and Hendrix
471 CW. 2000. The diagnostic value of fungal surveillance cultures in critically ill patients.
472 Surg. Infect. (Larchmt) 1:273-281.
473 21. Charles PE, Dalle F, Aube H, Doise JM, Quenot JP, Aho LS, Chavanet P, and
474 Blettery B. 2005. Candida spp. colonization significance in critically ill medical
475 patients: a prospective study. Intensive Care Med. 31:393-400.
476 22. Leon C, Alvarez-Lerma F, Ruiz-Santana S, Leon MA, Nolla J, Jorda R, Saavedra
477 P, and Palomar M. 2009. Fungal colonization and/or infection in non-neutropenic
478 critically ill patients: results of the EPCAN observational study. Eur. J. Clin. Microbiol.
479 Infect. Dis. 28:233-242.
480 23. Magill SS, Swoboda SM, Johnson EA, Merz WG, Pelz RK, Lipsett PA, and
481 Hendrix CW. 2006. The association between anatomic site of Candida colonization,
482 invasive candidiasis, and mortality in critically ill surgical patients. Diagn. Microbiol.
483 Infect. Dis. 55:293-301.
484 24. Magill SS, Swoboda SM, Shields CE, Colantuoni EA, Fothergill AW, Merz WG,
485 Lipsett PA, and Hendrix CW. 2009. The epidemiology of Candida colonization and
486 invasive candidiasis in a surgical intensive care unit where fluconazole prophylaxis is
487 utilized: follow-up to a randomized clinical trial. Ann. Surg. 249:657-665.
488 25. Piarroux R, Grenouillet F, Balvay P, Tran V, Blasco G, Millon L, and Boillot A.
489 2004. Assessment of preemptive treatment to prevent severe candidiasis in critically ill
490 surgical patients. Crit. Care Med. 32:2443-2449.
491 26. Van Horn KG, Audette CD, Tucker KA, and Sebeck D. 2008. Comparison of 3
492 swab transport systems for direct release and recovery of aerobic and anaerobic
493 bacteria. Diagn Microbiol Infect Dis 62:471-473.
494 27. Van Horn KG, Audette CD, Sebeck D, and Tucker KA. 2008. Comparison of the
495 Copan ESwab system with two Amies agar swab transport systems for maintenance of
496 microorganism viability. J Clin Microbiol 46:1655-1658.
497 28. Lau A, Sorrell TC, Chen S, Stanley K, Iredell J, and Halliday C. 2008. Multiplex
498 Tandem PCR: a Novel Platform for Rapid Detection and Identification of Fungal
499 Pathogens from Blood Culture Specimens. J. Clin. Microbiol. 46:3021-3027.
500 29. Lau A, Sorrell TC, Lee O, Stanley K, and Halliday C. 2008. Colony multiplex-
501 tandem PCR for rapid, accurate identification of fungal cultures. J. Clin. Microbiol.
502 46:4058-4060.

Page 20 of 21
503 30. Lau A, Halliday C, Chen SCA, Playford EG, Stanley K, and Sorrell TC. 2010.
504 Comparison of Whole Blood, Serum and Plasma for Early Detection of Candidemia by
505 Multiplex-Tandem PCR. J. Clin. Microbiol. 48:811-816.
506 31. Karabinis A, Hill C, Leclercq B, Tancrede C, Baume D, and Andremont A. 1988.
507 Risk factors for candidemia in cancer patients: a case-control study. J. Clin. Microbiol.
508 26:429-432.
509 32. Wey SB, Mori M, Pfaller MA, Woolson RF, and Wenzel RP. 1989. Risk factors for
510 hospital-acquired candidemia. A matched case-control study. Arch. Intern. Med.
511 149:2349-2353.
512 33. Garbino J, Lew DP, Romand JA, Hugonnet S, Auckenthaler R, and Pittet D. 2002.
513 Prevention of severe Candida infections in nonneutropenic, high-risk, critically ill
514 patients: a randomized, double-blind, placebo-controlled trial in patients treated by
515 selective digestive decontamination. Intensive Care Med. 28:1708-1717.
516 34. Agvald-Ohman C, Klingspor L, Hjelmqvist H, and Edlund C. 2008. Invasive
517 candidiasis in long-term patients at a multidisciplinary intensive care unit: Candida
518 colonization index, risk factors, treatment and outcome. Scand. J. Infect. Dis. 40:145-
519 153.
520 35. Pfaller MA, and Diekema DJ. 2007. Epidemiology of invasive candidiasis: a
521 persistent public health problem. Clin. Microbiol. Rev. 20:133-163.
522 36. Bassetti M, Righi E, Costa A, Fasce R, Molinari MP, Rosso R, Pallavicini FB, and
523 Viscoli C. 2006. Epidemiological trends in nosocomial candidemia in intensive care.
524 BMC Infect. Dis. 6:21.
525 37. Comert F, Kulah C, Aktas E, Eroglu O, and Ozlu N. 2007. Identification of Candida
526 species isolated from patients in intensive care unit and in vitro susceptibility to
527 fluconazole for a 3-year period. Mycoses 50:52-57.
528 38. Horn DL, Neofytos D, Anaissie EJ, Fishman JA, Steinbach WJ, Olyaei AJ, Marr
529 KA, Pfaller MA, Chang CH, and Webster KM. 2009. Epidemiology and outcomes
530 of candidemia in 2019 patients: data from the prospective antifungal therapy alliance
531 registry. Clin. Infect. Dis. 48:1695-1703.
532

533

534 Figure 1. Cumulative incidence of IC and percentage of patients colonized at any site or

535 specifically in the throat, perineum or urine over the first three time points.

536

Page 21 of 21
 Table 1. Risk of invasive candidiasis based on screening time point and anatomical site of colonization (derived from culture results

only; entire patient cohort included).

CIb 95% CI 95%

Variable N RRa P-value Sensitivity (%) Specificity (%) PPVc (%) NPVd (%)
(low) (high)
TIME POINT 1* (N studied = 6,015)
Any site colonized (CIe ≥ 0.3) 3511 2.35 0.002 1.35 4.08 78 40 2 99
At least throat colonized 2927 2.04 0.004 1.25 3.34 68 49 2 99
At least perineum colonized 1967 1.76 0.0155 1.1 2.77 48 66 2 99
At least urine colonized 630 2.25 0.003 1.32 3.84 23 88 3 99
TIME POINT 2 (N studied = 3,244)
Any site colonized (CI ≥ 0.3) 2025 3.69 0.001 1.66 8.17 86 38 2 99
At least throat colonized 1555 2.97 0.0007 1.58 5.56 74 51 2 99
At least perineum colonized 1313 1.54 NSf 0.89 2.67 52 59 2 99
At least urine colonized 450 1.81 NS 0.95 3.44 24 85 3 99

TIME POINT 3 (N studied = 1,820)

Any site colonized (CI ≥ 0.3) 1162 1.41 <0.0001 1.23 1.61 89 37 2 100
At least throat colonized 837 5.04 0.001 1.93 13.21 82 53 3 99
At least perineum colonized 762 2.04 NS 0.96 4.34 61 57 2 99
At least urine colonized 284 - NS - - - - - -

*Time point 1 (days 3-4 post-ICU admission), 2 (days 6-8 post-ICU admission), and 3 (days 9-11 post-ICU admission).
a
Relative risk (RR); bConfidence Interval (CI); cPositive predictive value (PPV); dNegative predictive value (NPV); eColonization
index (CI); fNot significant (NS).
Table 2. Performance characteristics using CI > 0.5 and CCI ≥ 0.3 (derived from culture results only; entire patient cohort included).

CIb 95% CI 95% Sensitivity Specificity PPVc NPVd

Variable N RRa P-value
(low) (high) (%) (%) (%) (%)
TIME POINT 1 (N studied = 6,015)
At least 2 sites colonized (CIe ≥ 0.5) 1671 2.25 0.0005 1.4 3.5 48 71 2 99
All 3 sites colonized 342 2.25 0.016 1.16 4.34 14 94 3 99
At least one site heavy density (CCIf ≥ 0.3) 1549 3.7 <0.0001 2.36 5.93 58 74 3 99
At least 2 sites heavy density 448 3.1 0.001 1.77 5.4 21 92 3 99
At least throat heavy density 1025 3.77 <0.0001 2.39 5.94 45 82 3 99
At least perineum heavy density 703 2 0.01 1.15 3.46 22 88 2 99
At least urine heavy density 327 - NSg - - - - - -
TIME POINT 2 (N studied = 3,244)
At least 2 sites colonized (CI ≥ 0.5) 1056 2.4 0.002 1.38 4.16 54 67 3 99
All 3 sites colonized 238 - NS - - - - - -
At least one site heavy density (CCI ≥ 0.3) 915 3.24 <0.0001 1.86 5.63 56 72 3 99
At least 2 sites heavy density 287 2.9 0.001 1.5 5.6 22 91 4 99
At least throat heavy density 530 3.93 <0.0001 2.26 6.81 44 84 4 99
At least perineum heavy density 481 - NS - - - - - -
At least urine heavy density 233 - NS - - - - - -
TIME POINT 3 (N studied = 1,820)
At least 2 sites colonized (CI ≥ 0.5) 596 3.12 0.003 1.47 6.62 61 67 3 99
All 3 sites colonized 122 - NS - - - - - -
At least one site heavy density (CCI ≥ 0.3) 480 2.1 0.05 0.99 4.39 43 74 3 99
At least 2 sites heavy density 143 - NS - - - - - -
At least throat heavy density 253 2.83 0.009 1.3 6.2 32 86 4 99
At least perineum heavy density 263 - NS - - - - - -
At least urine heavy density 126 - NS - - - - - -

*Time point 1 (days 3-4 post-ICU admission), 2 (days 6-8 post-ICU admission), and 3 (days 9-11 post-ICU admission).
a
Relative risk (RR)
b
Confidence Interval (CI)
c
Positive predictive value (PPV)
d
Negative predictive value (NPV)
e
Colonization index (CI)
f
Corrected colonization index (CCI)
g
Not significant (NS)