Abstract
We analyzed 2462 episodes of suspected catheter-related bloodstream infection (CRBSI). The sensitivity, specificity, and positive and negative predictive values for detecting Candida CRBSI by gram staining of catheter tips were 66.1%, 99.4%, 84.4%, and 98.4%, respectively. Gram staining may be useful for the early detection of Candida CRBSI.
Keywords: CRBSI, Candida, catheter tips, gram stain, prediction
We analyzed 2462 episodes of suspected catheter-related bloodstream infection (CRBSI). The positive predictive value for detecting Candida CRBSI by Gram staining of catheter tips was 84.4%. Gram staining may be useful for the early detection of Candida CRBSI.
Candida species are important pathogens causing catheter-related bloodstream infections (CRBSIs), accounting for 10%–27% of all cases [1, 2]. Compared with culture tests, gram staining of catheter tips provides more rapid information. Furthermore, it has high specificity for predicting catheter colonization and CRBSI [3, 4]. However, previous studies analyzed microorganisms generally as causative organisms of CRBSIs and included only 7 fungal cases. Thus, the usefulness of gram staining of catheter tips for detecting Candida CRBSIs (C-CRBSIs) in clinical practice remains unclear. Thus, we aimed to investigate the predictive performance of gram staining of catheter tips for C-CRBSIs in patients with suspected CRBSIs.
METHODS
Patient Consent
This was a retrospective observational study with secondary use of routine medical records conducted following the Declaration of Helsinki and was approved by the ethics review committee of the National Center for Global Health and Medicine (approval number: NCGM-S-004402-00). Patients were allowed to withdraw consent for the use of their records for research. Information for opting out of this study is available on the hospital website.
Study Population, Design, and Data Collection
All patients admitted to the National Center for Global Health and Medicine (tertiary care hospital in Tokyo, Japan) between April 1, 2007, and September 21, 2021, with at least 1 central venous catheter (CVC) or peripherally inserted central catheter (PICC) submitted for culture were retrospectively reviewed. Demographic data, intravascular catheter type, results of gram staining and catheter tip culture, and results of blood culture in patients with positive gram staining for yeast-like fungi on the catheter tips were collected.
Laboratory Procedure
The catheter tips were cultured using the roll-plate method described by Maki et al. [5]. The external and internal catheter tip surfaces were rinsed with sterile purified water (0.5 mL). Some of the rinse water was processed with gram stain, and the rest and the catheter tips were together added to thioglycolate (TGC) medium and cultured for 7 days. Gram staining was performed using the following dyeing procedure: crystal violet, iodine solution, ethanol-acetone solution, and fuchsine for 30 seconds, 30 seconds, a few seconds, and 15 seconds, respectively. A catheter with at least 1 yeast-like fungus per 30 oil-immersion fields was considered positive on gram staining. Blood cultures were performed using BACTEC (Becton Dickinson, Sparks, MD, USA). The microbiological laboratory techniques used remained unchanged during the study. All laboratory procedures were performed by skilled microbiologists.
Definitions
-
Catheter colonization
Maki's roll-plate method with ≥15 colony-forming units (CFUs) or TGC medium fluid with ≥103 CFU/mL [6].
-
Catheter-related blood culture
Blood cultures obtained 3 days before to 3 days after catheter removal.
-
Suspected CRBSI
Patients for whom both catheter tip and catheter-related blood cultures were performed.
-
C-CRBSIs
Patients with at least 1 positive catheter-related blood culture and also a positive catheter tip culture for the same Candida species.
Outcome
The primary outcome was the diagnostic performance (ie, sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) of gram staining of catheter tips for C-CRBSI diagnosis.
Statistical Analysis
Summary statistics of patients’ background factors and Candida subspecies were calculated. The mean (SD), median (interquartile range), or percentage (%) was used as appropriate. In patients with suspected CRBSI, sensitivity, specificity, PPV, and NPV were calculated with 95% CIs to predict C-CRBSIs. Cases in which the catheter tip culture was positive for Candida species but no subspecies were identified were considered attributable to the same Candida species as that detected in the blood culture. An additional episode of catheter tip culture 24 hours after the previous episode was considered a separate episode. Moreover, based on previous studies, a period other than ±3 days of catheter removal was selected for catheter-related blood culture, and predictive performance was recalculated in the same way: (i) blood cultures obtained 7 days before to 2 days after catheter removal [7] and (ii) blood cultures obtained 2 days before to 1 day after catheter removal [8]. PPV is a particularly useful diagnostic test measure obtained from gram staining of catheter tips. To estimate the 95% CI of ±0.1, 61 cases with positive gram staining of the catheter tip were needed, estimated using the normal approximation to the binomial distribution [9]. Our center receives approximately 10–20 cases of C-CRBSI annually; thus, we determined that a 14-year study period would provide sufficient cases. Statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).
RESULTS
During the study, 2462 episodes of catheter tip and catheter-related blood cultures from 2381 patients were submitted, excluding 6 episodes for which data on bacterial quantity were missing. CVC and PICC accounted for 97.8% and 2.2% of the cases, respectively.
Ninety episodes of yeast-like fungus detection on gram staining from 88 patients were identified. Of the patients included, 64% were men; the median age was 76 years (Table 1). Of the 90 episodes of positive gram staining, Candida colonization was identified in 81 (90%); 5 episodes showed 2 Candida species on catheter tip culture; 76 episodes (84%) were accompanied by positive catheter-related blood culture and categorized as C-CRBSI. Overall, 2372 episodes were negative for yeast-like fungi by gram staining. Candida colonization was identified in 48 of these episodes (2%), and 39 (1.6%) were categorized as C-CRBSIs.
Table 1.
Characteristics of Intravenous Catheters With Positive Gram Staining for Yeast-Like Fungi
| Characteristic | |
|---|---|
| No. of catheters | 90 |
| Male sex, No. (%) | 58 (64.4) |
| Age, median (IQR), y | 76 (65–81) |
| Catheter type, No. (%) | … |
| CVC | 86 (95.6) |
| PICC | 4 (4.4) |
| No. of catheter-related blood cultures, No. (%) | … |
| 2 or more sets | 75 (83.3) |
| 1 set | 15 (16.7) |
| Time to obtain blood cultures, No. (%) | … |
| −3 d | 5 (5.6) |
| −2 d | 14 (15.6) |
| −1 d | 22 (24.4) |
| 0 d | 46 (51.1) |
| 1 d | 3 (3.3) |
| 2 d | 0 (0.0) |
| 3 d | 0 (0.0) |
| Timing of antifungal treatment initiation, No. (%) | … |
| Before blood culture | 10 (11.1) |
| After blood culture | 67 (74.4) |
| No antifungal treatment | 10 (11.1) |
| No data | 3 (3.3) |
| Candida species, No. (%) | … |
| Candida albicans | 34 (39.5) |
| Candida parapsilosis | 16 (18.6) |
| Candida glabrata | 19 (22.1) |
| Candida tropicalis | 8 (9.3) |
| Candida lusitaniae | 1 (1.2) |
| Candida guilliermondii | 1 (1.2) |
| Candida krusei | 2 (2.3) |
| Candida metapsilosis | 1 (1.2) |
| Candida paratropicalis | 1 (1.2) |
| Candida species | 3 (3.5) |
| Candida catheter colonization, No. (%) | 81 (90.0) |
| C-CRBSI, No. (%) | 76 (84.4) |
Abbreviations: C-CRBSI, Candida catheter-related bloodstream infection; CVC, central venous catheter; IQR, interquartile range; PICC, peripherally inserted central catheter.
The sensitivity, specificity, PPV, and NPV for C-CRBSI detection from gram staining of the catheter tip were 66.1% (95% CI, 56.7%–74.7%), 99.4% (95% CI, 99.0%–99.7%), 84.4% (95% CI, 75.3%–91.2%), and 98.4% (95% CI, 97.8%–98.8%), respectively (Table 2). Similar results were obtained after varying the period beyond ±3 days of catheter removal according to the definition of catheter-related blood culture (Supplementary Table 1).
Table 2.
Correlation Between Catheter Tip Gram Stains and C-CRBSIs
| C-CRBSI Present | C-CRBSI NOT Present | Total | |
|---|---|---|---|
| Gram stain | … | … | … |
| Positive | 76 | 14 | 90 |
| Negative | 39 | 2333 | 2372 |
| Total | 115 | 2347 | 2462 |
Sensitivity, 76/115 (66.1%); specificity, 2333/2347 (99.4%), positive predictive value, 76/90 (84.4%); negative predictive value, 2333/2372 (98.4%).
Abbreviation: C-CRBSI, Candida catheter bloodstream infection.
Supplementary Table 2 shows the characteristics of 14 false-positive episodes. Two or more sets of blood cultures were obtained in 12 episodes and 1 in 2 episodes. Thirteen episodes had a negative blood culture. One episode had a positive blood culture for Candida species and was treated for possible C-CRBSI. Detailed information on antifungal treatment was available for 11 episodes. None of them had antifungal treatment preceding the blood culture examination.
DISCUSSION
Our study demonstrated that gram staining of the catheter tip afforded a good PPV (84.4%) for C-CRBSI detection in patients with suspected CRBSIs. Gram staining of a catheter tip is a simple and rapid test. Previously, the predictive performance of gram staining for CRBSI detection was investigated for all microorganisms, including Candida species, and sensitivity, specificity, PPV, and NPV, respectively, of 73%, 98%, 90%, and 94% were obtained [10]. These results are generally similar to those of the present study. However, the study included only 7 catheters colonized by Candida species. To our knowledge, our study is the first large study to investigate the feasibility of using gram staining to detect C-CRBSIs.
Routine empirical antifungals in the initial CRBSI treatment are associated with side effects, the emergence of antifungal resistance, and high costs [11]. However, as the incubation time until Candida species detection is several days [12], considering the high mortality in candidemia [13], a delay in initiating appropriate antifungal therapy is associated with adverse outcomes [14, 15]. The high PPV of gram staining of catheter tips in this study supports the validity of initiating antifungal therapy for suspected CRBSIs as soon as yeast-like fungi are identified by gram staining of the removed catheter tip.
This study had several limitations. First, other apparent sources of bloodstream infection were not ruled out, which may have led to the inclusion of infections other than C-CRBSI. Second, because of the retrospective nature of the study, the submission of catheter tips for culture was at the attending physician's discretion. Therefore, not all CRBSI cases were included in the present study. Third, in 87 of the 90 episodes with positive gram staining, we could mention the presence or absence of prior antifungal therapy but not for all the 2462 episodes included in this study; thus, some C-CRBSI cases may have been missed because of false-negative results. Finally, whether the results of this study will improve the prognosis of C-CRBSI remains unclear, and further studies are needed.
In conclusion, we found that gram staining of catheter tips had good specificity, PPV, and NPV for detecting C-CRBSIs and may be useful for the early detection of C-CRBSIs.
Supplementary Material
Acknowledgments
Financial support. This research was supported by the National Center for Global Health and Medicine Intramural Research Fund (grant number: 20A05).
Author contributions. All authors have read and approved the final manuscript and contributed significantly to the work. Conceptualization: G.Y., N.I., M.I., A.M., M.K., K.M., and N.O. Data curation: G.Y., A.M., M.K., and K.M. Formal analysis: G.Y. Methodology: G.Y., N.I., M.I., and M.I. Project administration: N.I., M.I., and N.O. Resources: A.M., M.K., and K.M. Supervision: N.I., M.I., K.M., and N.O. Writing—original draft: G.Y., N.I., M.I., A.M. Writing—review and editing: K.M., K.M., and N.O.
Data availability. Data are not publicly available.
Contributor Information
Gen Yamada, Disease Control and Prevention Center, National Center for Global Health and Medicine, Tokyo, Japan.
Noriko Iwamoto, Disease Control and Prevention Center, National Center for Global Health and Medicine, Tokyo, Japan.
Masahiro Ishikane, Disease Control and Prevention Center, National Center for Global Health and Medicine, Tokyo, Japan.
Ataru Moriya, Clinical Laboratory Department, National Center for Global Health and Medicine, Tokyo, Japan.
Masami Kurokawa, Clinical Laboratory Department, National Center for Global Health and Medicine, Tokyo, Japan.
Kazuhisa Mezaki, Clinical Laboratory Department, National Center for Global Health and Medicine, Tokyo, Japan.
Norio Ohmagari, Disease Control and Prevention Center, National Center for Global Health and Medicine, Tokyo, Japan.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
References
- 1. Chitnis AS, Edwards JR, Ricks PM, Sievert DM, Fridkin SK, Gould CV. Device-associated infection rates, device utilization, and antimicrobial resistance in long-term acute care hospitals reporting to the National Healthcare Safety Network, 2010. Infect Control Hosp Epidemiol 2012; 33:993–1000. [DOI] [PubMed] [Google Scholar]
- 2. Novosad SA, Fike L, Dudeck MA, et al. Pathogens causing central-line-associated bloodstream infections in acute-care hospitals—United States, 2011–2017. Infect Control Hosp Epidemiol 2020; 41:313–9. [DOI] [PubMed] [Google Scholar]
- 3. Guembe M, Pérez-Granda MJ, Rivera ML, Martín-Rabadán P, Bouza E. Performing gram stain directly on catheter tips: assessment of the quality of the observation process. Eur J Clin Microbiol Infect Dis 2015; 34:1091–5. [DOI] [PubMed] [Google Scholar]
- 4. Coutlée F, Lemieux C, Paradis JF. Value of direct catheter staining in the diagnosis of intravascular-catheter-related infection. J Clin Microbiol 1988; 26:1088–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Maki DG, Weise CE, Sarafin HW. A semiquantitative culture method for identifying intravenous-catheter-related infection. N Engl J Med 1977; 296:1305–9. [DOI] [PubMed] [Google Scholar]
- 6. Mermel LA, Allon M, Bouza E, et al. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 update by the Infectious Diseases Society of America. Clin Infect Dis 2009; 49:1–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Buetti N, Lo Priore E, Atkinson A, et al. Low incidence of subsequent bacteraemia or fungaemia after removal of a colonized intravascular catheter tip. Clin Microbiol Infect 2018; 24:548.e1–3. [DOI] [PubMed] [Google Scholar]
- 8. Peacock SJ, Eddleston M, Emptage A, King A, Crook DW. Positive intravenous line tip cultures as predictors of bacteraemia. J Hosp Infect 1998; 40:35–8. [DOI] [PubMed] [Google Scholar]
- 9. Van Belle G, Fisher LD, Heagerty PJ, Lumley T. Biostatistics: A Methodology for the Health Sciences. Wiley; 1993. [Google Scholar]
- 10. Bouza E, Alvarado N, Alcalá L, Muñoz P, Rabadán PM, Rodríguez-Créixems M. An instant procedure to demonstrate catheter-tip colonization may help clinicians. Diagn Microbiol Infect Dis 2006; 56:255–60. [DOI] [PubMed] [Google Scholar]
- 11. Martin-Loeches I, Antonelli M, Cuenca-Estrella M, et al. ESICM/ESCMID Task Force on Practical Management of Invasive Candidiasis in Critically Ill Patients. Intensive Care Med 2019; 45:789–805. [DOI] [PubMed] [Google Scholar]
- 12. Fernandez J, Erstad BL, Petty W, Nix DE. Time to positive culture and identification for Candida blood stream infections. Diagn Microbiol Infect Dis 2009; 64:402–7. [DOI] [PubMed] [Google Scholar]
- 13. Ishikane M, Hayakawa K, Kutsuna S, Takeshita N, Ohmagari N. The impact of infectious disease consultation in candidemia in a tertiary care hospital in Japan over 12 years. PLoS One 2019; 14:e0215996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Parkins MD, Sabuda DM, Elsayed S, Laupland KB. Adequacy of empirical antifungal therapy and effect on outcome among patients with invasive Candida species infections. J Antimicrob Chemother 2007; 60:613–8. [DOI] [PubMed] [Google Scholar]
- 15. Garey KW, Rege M, Pai MP, et al. Time to initiation of fluconazole therapy impacts mortality in patients with candidemia: a multi-institutional study. Clin Infect Dis 2006; 43:25–31. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.