Laboratory tests for Clostridioides difficile infection (CDI) rely on the detection of free toxin or molecular detection of toxin genes. The Singulex Clarity C. diff toxins A/B assay is a rapid, automated, and ultrasensitive assay that detects C. difficile toxins A and B in stool. We compared CDI assays across two prospective multicenter studies to set a cutoff for the Clarity assay and to independently validate the performance compared with that of a cell culture cytotoxicity neutralization assay (CCCNA).
Keywords: C. difficile, ultrasensitivity, single-molecule counting technology, toxins A and B, ultrasensitive
ABSTRACT
Laboratory tests for Clostridioides difficile infection (CDI) rely on the detection of free toxin or molecular detection of toxin genes. The Singulex Clarity C. diff toxins A/B assay is a rapid, automated, and ultrasensitive assay that detects C. difficile toxins A and B in stool. We compared CDI assays across two prospective multicenter studies to set a cutoff for the Clarity assay and to independently validate the performance compared with that of a cell culture cytotoxicity neutralization assay (CCCNA). The cutoff was set by two sites testing fresh samples from 897 subjects with suspected CDI and then validated at four sites testing fresh samples from 1,005 subjects with suspected CDI. CCCNA testing was performed at a centralized laboratory. Samples with discrepant results between the Clarity assay and CCCNA were retested with CCCNA when the Clarity result agreed with that of at least one comparator method; toxin enzyme immunoassays (EIA), glutamate dehydrogenase (GDH) detection, and PCR were performed on all samples. The cutoff for the Clarity assay was set at 12.0 pg/ml. Compared to results with CCCNA, the Clarity assay initially had 85.2% positive agreement and 92.4% negative agreement. However, when samples with discrepant results between the Clarity assay and CCCNA in the validation study were retested by CCCNA, 13/17 (76.5%) Clarity-negative but CCCNA-positive samples (Clarity+/CCCNA−) became CCCNA−, and 5/26 (19.2%) Clarity+/CCCNA− samples became CCCNA+, resulting in a 96.3% positive agreement and 93.0% negative agreement between Clarity and CCCNA results. The toxin EIA had 59.8% positive agreement with CCCNA. The Clarity assay was the most sensitive free-toxin immunoassay, capable of providing CDI diagnosis in a single-step solution. A different CCCNA result was reported for 42% of retested samples, increasing the positive agreement between Clarity and CCCNA from 85.2% to 96.3% and indicating the challenges of comparing free-toxin results to CCCNA results as a reference standard.
INTRODUCTION
Clostridioides (formerly Clostridium) difficile infection (CDI) is the main cause for nosocomial diarrhea in Europe and the United States, leading to substantial comorbidities, mortalities, and significantly increased health care costs (1, 2). CDI is a clinical diagnosis defined by the presence of symptoms, usually acute diarrhea, and a positive laboratory stool test for either free toxins A (TcdA) and B (TcdB) or toxigenic C. difficile (3–7). An accurate and prompt diagnosis of CDI aids decision making for appropriate treatment and infection control measures and may prevent the propagation and spreading of more virulent strains (3, 5, 7–9).
The CDI reference tests, cell culture cytotoxicity neutralization assay (CCCNA) and toxigenic culture (TC), are labor-intensive, technique variable, and have long turnaround times (4, 7–9). Currently available toxin enzyme immunoassays (EIAs) are specific, rapid, and low cost but have poor sensitivity, while nucleic acid amplification tests (NAATs) are sensitive and rapid but are costly and have poor specificity for disease (4, 6). Based on these diagnostic method challenges, testing with multistep algorithms is often recommended (7, 8).
The presence of toxins, not toxigenic C. difficile, in patients’ stools correlates with disease severity (10–12), and current guidelines state that the detection of toxins is most relevant for CDI diagnosis (7, 8). However, because of the poor sensitivity of currently available toxin EIAs, there is a need for a toxin test with high sensitivity that could be used as a stand-alone test (8). The Singulex Clarity C. diff toxins A/B assay (herein, Clarity), for use on the Singulex Clarity system (Singulex, Inc., Alameda, CA, USA), was developed to meet this need. The automated platform utilizes single-molecule counting technology for a rapid, ultrasensitive detection of analytes down to femtogram-per-milliliter concentrations (13, 14). Regulatory studies require comparison of free-toxin tests with CCCNA, a reference method that is known to have issues related to toxin stability and subjectivity (4). To address this issue, we compared different assays for C. difficile testing across two prospective multicenter studies. The aims of these studies were to set a diagnostic cutoff for the Clarity assay and to independently validate the performance compared with that of CCCNA, thus reflecting a study design based on the regulatory requirements.
MATERIALS AND METHODS
Singulex Clarity C. diff toxins A/B assay.
The Clarity assay measures TcdA and TcdB in stool on the automated Singulex Clarity system, an in vitro diagnostic platform, and has been described previously (13). Briefly, the system is based upon a paramagnetic microparticle-based immunoassay powered by single-molecule counting technology that uses single-photon fluorescence detection for analyte quantitation. The quantitative limits of detection for TcdA and TcdB are 0.8 and 0.3 pg/ml in buffer and 2.0 and 0.7 pg/ml in stool, respectively (13). An unformed stool sample volume of 100 μl, or 0.1 g of semisolid stool sample, is diluted 1:20 with 1.9 ml of sample buffer and briefly vortexed. The sample is then centrifuged at 14,000 × g for 10 minutes, and 300 μl of the supernatant is transferred into a sample tube and loaded onto the Clarity instrument. The instrument automatically performs the immunoassay with a 1:1 mixture of paramagnetic microparticles precoated with anti-TcdA and anti-TcdB monoclonal antibodies (capture reagent) and toxin-specific monoclonal antibodies labeled with the fluorophore Alexa Fluor 647 (detection reagent). The Clarity software interpolates the data, using the fluorescent signal, into a combined TcdA/TcdB concentration reported in units of picograms/milliliter of stool. Time to the first result is 32 min, and the system can process 1 to 48 samples in an assay run.
Study design.
(i) Cutoff establishment. This multisite prospective study enrolled all consecutive stool samples from patients suspected of having CDI and submitted them to the clinical laboratories at two testing sites: Hennepin County Medical Center (Minneapolis Medical Research Foundation, Minneapolis MN, USA) and TriCore Reference Laboratories (Albuquerque, NM, USA). Inclusion criteria for study enrollment were stool samples submitted for routine C. difficile testing, with a minimum residual volume of 4 ml after completion of standard-of-care testing, and availability for study testing within 72 h of collection. Each residual stool sample was deidentified and stored at 2 to 8°C until testing with standard-of-care assays available at the laboratories: NAAT (Xpert C. difficile or Xpert C. difficile/Epi; Cepheid Inc., Sunnyvale, CA), a lateral flow EIA for detection of glutamate dehydrogenase (GDH) and TcdA and TcdB (C. Diff Quik Chek Complete [TechLab, Inc., Blacksburg, VA], with a limit of detection of ≥630 pg/ml for TcdA and ≥160 pg/ml for TcdB ), and Clarity. A separate aliquot was sent overnight at 2 to 8°C for CCCNA testing (C. difficile TOX-B Test [Techlab]; tested at Microbiology Specialists, Inc., Houston, TX). Samples were tested in singlicate by laboratory personnel on Clarity systems at both testing sites. Data on patient gender and age were collected, and Bristol scale assessments were made before testing; no additional clinical data were collected, and comparison with clinical diagnosis was not an objective of this study.
(ii) Validation study. This multisite prospective study enrolled all consecutive stool samples from patients suspected of having CDI and submitted them to the clinical laboratories at four testing sites: Hennepin County Medical Center, TriCore Reference Laboratories, the Johns Hopkins Hospital (Baltimore, MD, USA), and San Francisco General Hospital (San Francisco, CA, USA). Inclusion criteria for enrollment, sample storage, testing procedure, and CCCNA testing were similar to those in the cutoff establishment study. The samples were tested with NAAT (Xpert C. difficile or BD MAX Cdiff assay), a GDH and toxin EIA, and Clarity. For comparison with the Clarity toxin values, PCR cycle thresholds (CT) of ≤27 and >27 were used based upon the literature (8, 15, 16). Comparison of free-toxin tests with CCCNA is required by regulatory agencies, such as the U.S. Food and Drug Administration.
(iii) Discrepant analysis. CCCNA was retested on samples from the validation study with discrepant results between the Clarity assay and CCCNA when Clarity agreed with at least one other comparator method (NAAT, GDH EIA, and/or toxin EIA). Samples were frozen at –80°C for 1 to 3 months prior to repeat testing. A matched control consisting of a random selection of samples with concordant results between Clarity assay and CCCNA was retested to determine the repeatability of each CCCNA result.
Ethics.
The study was approved by the Hennepin County Medical Center Institutional Review Board (IRB; HSR no. 18-4498), Ethical and Independent Review Services on behalf of TriCore Reference Laboratories (no. 18009-01B), the Johns Hopkins Medicine Office of Human Subjects Research (no. IRB00173485), and the UCSF Human Research Protection Program (no. 18-25093).
Data analysis.
Performance of all assay methods, including multistep algorithms, was compared to that of the CCCNA as the reference method. The area under the receiver operating characteristics (ROC) curve (AuROC) was calculated to quantify the overall diagnostic accuracy of a preliminary quantitative assay relative to that of CCCNA and to set an ideal cutoff. Using a range of concentrations across the ROC curve, the minimum difference between the sensitivity-specificity and the Youden methods were used to set the cutoff (17). Subsequently, standard diagnostic performance metrics were determined with associated exact 95% confidence intervals (CI) (18). Spearman’s rank correlation coefficient was used to measure the strength and direction of the linear association between Clarity and tcdB PCR cycle threshold (CT) values, and analyses were also performed using established CT values (19, 20). Statistical differences in C. difficile toxin concentrations were assessed by a Mann Whitney U test. Statistical analysis was done with SAS (version 9.4), Analyze-It for MS Excel (version 4.51), and GraphPad Prism (version 5.0) software.
RESULTS
Cutoff establishment study.
A total of 914 subject samples were collected between April and May 2018. Five subject samples had missing CCCNA results, and 12 subject samples had missing or invalid Clarity results. Final analyses were performed on the remaining 897 patient samples: 118 (13.2%) CCCNA-positive (CCCNA+) and 779 (86.9%) CCCNA-negative (CCCNA−). The demographics of the study subjects stratified by CCCNA results are presented in Table S1 in the supplemental material. Four of the included patients had samples with missing PCR results. The AuROC curve demonstrated a concordance (C) statistic of 0.971 (95% CI, 0.955 to 0.988). An optimized cutoff for the Clarity assay was set at 12.0 pg/ml using the ROC analysis (Fig. S1).
There was a significant inverse linear correlation between tcdB PCR cycle threshold values and Clarity TcdA/TcdB concentrations in the 207 PCR-positive (PCR+) samples (Spearman’s correlation coefficient, −0.62; P value, <0.001) (Fig. 1). However, there were 18 samples toxin negative (toxin−) by Clarity, with CT values of <27.0 (22.8% of all PCR+/toxin− samples) and 36 toxin-positive (toxin+) samples with CT values of >27.0 (14.3% of all PCR+/toxin+ samples) (Table S2).
FIG 1.
Correlation between tcdB PCR CT values and toxin concentration in 207 PCR-positive samples tested with Clarity. Spearman’s correlation coefficient was −0.62 (95% CI, −0.70, −0.53). The dashed lines show Clarity assay cutoff (12.0 pg/ml) and the chosen tcdB CT value cutoff (27.0).
Validation study and discrepant analysis.
A total of 1,010 subject samples were collected between June and September 2018. Five samples were excluded due to the absence of either CCCNA or Clarity results. Final analyses were performed on the remaining 1,005 patient samples. Table 1 presents the baseline demographics for study patients stratified by CCCNA results after the discrepant analysis. Compared to CCCNA, the Clarity assay initially had 91.5% overall agreement, with 85.2% positive agreement (sensitivity) and 92.4% negative agreement (specificity) (Table 2).
TABLE 1.
Baseline demographics for study subjects in the validation study after discrepant analysis
| Parameter | Value for the group |
||
|---|---|---|---|
| CCCNA positivea | CCCNA negative | Total | |
| No. of study subjects (%) | 107 (10.6) | 898 (89.4) | 1,005 (100) |
| TriCore Laboratories, NM | 58 (54.2) | 363 (40.4) | 421 (41.9) |
| Johns Hopkins Hospital, MD | 32 (29.9) | 315 (35.1) | 347 (34.5) |
| Hennepin County Medical Center, MN | 11 (10.3) | 116 (12.9) | 127 (12.6) |
| San Francisco General Hospital, CA | 6 (5.6) | 104 (11.6) | 110 (10.9) |
| Sex (no. [%]) | |||
| Female | 56 (52.3) | 487 (54.2) | 543 (54.0) |
| Male | 51 (47.7) | 411 (45.8) | 462 (46.0) |
| Mean (SD) age (yr) | 59.7 (19.2) | 56.5 (18.9) | 56.9 (19.0) |
| No. of patients (%) by age (yr) | |||
| <5 | 0 (0.0) | 2 (0.2) | 2 (0.2) |
| 5–21 | 3 (2.8) | 31 (3.5) | 34 (3.4) |
| 22–59 | 45 (42.1) | 428 (47.7) | 473 (47.1) |
| >60 | 59 (55.1) | 436 (48.6) | 495 (49.3) |
| Missing | 0 (0.0) | 1 (0.1) | 1 (0.1) |
| Bristol scale (no. of patients [%]) | |||
| 3 | 1 (0.9) | 8 (0.9) | 9 (0.9) |
| 4 | 1 (0.9) | 17 (1.9) | 18 (1.8) |
| 5 | 14 (13.1) | 157 (17.5) | 171 (17.0) |
| 6 | 34 (31.8) | 268 (29.8) | 302 (30.0) |
| 7 | 57 (53.3) | 448 (49.9) | 505 (50.2) |
CCCNA, cell culture cytotoxicity neutralization assay.
TABLE 2.
The performance of the Clarity assay compared to that of CCCNA before and after discrepant analysis
| Singulex Clarity C. diff toxins A/B assay result | No. of CCCNA resultsa |
Positive % agreement (95% CI)b | Negative % agreement (95% CI) | Overall % agreement (95% CI) | ||
|---|---|---|---|---|---|---|
| Positive | Negative | Total | ||||
| Before discrepant analysis | ||||||
| Positive | 98 | 68 | 166 | 85.2 (77.4–91.2) | 92.4 (90.4–94.0) | 91.5 (89.7–93.2) |
| Negative | 17 | 822 | 839 | |||
| Total | 115 | 890 | 1,005 | |||
| After discrepant analysis | ||||||
| Positive | 103 | 63 | 166 | 96.3 (90.7–99.0) | 93.0 (91.1–94.6) | 93.3 (91.6–94.8) |
| Negative | 4 | 835 | 839 | |||
| Total | 107 | 898 | 1,005 | |||
CCCNA, cell culture cytotoxicity neutralization assay.
CI, confidence interval.
When samples with discrepant results between Clarity and CCCNA were retested with CCCNA, 13/17 (76.5%) Clarity−/CCCNA+ samples became CCCNA−, and 5/26 (19.2%) of Clarity+/CCCNA− samples became CCCNA+. No differences in original test results were observed for any of the controls. All retested control samples with concordant results between Clarity and CCCNA (19 Clarity+/CCCNA+ and 27 Clarity−/CCCNA−) remained concordant (Table S3). A total of 839 (83.5%) samples had a toxin concentration below the Clarity cutoff, of which 835 (99.5%) samples were CCCNA negative, with a 93.3% overall agreement, 96.3% positive agreement, and 93.0% negative agreement between Clarity and CCCNA results (Table 2). Performance for other testing methods and algorithms is outlined in Table 3. The positive agreement between Clarity and PCR was 64.7% (Table 4). There were 66 PCR+/Clarity− samples, of which 63 (95.5%) were CCCNA negative.
TABLE 3.
The performance of toxin EIA and multistep algorithms utilizing toxin assays as last tests compared to that of CCCNA after discrepant analysisa
| Test method(s) and result | No. of CCCNA results |
Positive % agreement (95% CI)b | Negative % agreement (95% CI) | ||
|---|---|---|---|---|---|
| Positive | Negative | Total | |||
| Toxin EIA | |||||
| Positive | 64 | 8 | 72 | 59.8 (49.9–69.2) | 99.1 (98.3–99.6) |
| Negative | 43 | 890 | 933 | ||
| Total | 107 | 898 | 1,005 | ||
| PCR → toxin EIA | |||||
| PCR positive → toxin EIA positive | 64 | 2 | 66 | 59.8 (49.9–69.2) | 99.8 (99.2–100) |
| PCR positive → toxin EIA negative | 43 | 896 | 939 | ||
| Total | 107 | 898 | 1,005 | ||
| PCR → Clarity toxinc | |||||
| PCR positive → Clarity toxin positive | 102 | 19 | 121 | 95.3 (89.4–98.5) | 97.9 (96.7–98.7) |
| PCR positive → Clarity toxin negative | 5 | 879 | 884 | ||
| Total | 107 | 898 | 1,005 | ||
PCR-positive samples were reflexed to toxin EIA or Clarity. CCCNA, cell culture cytotoxicity neutralization assay.
CI, confidence interval.
A total of 45 specimens were PCR negative and toxin positive (Table 4).
TABLE 4.
Positive, negative, and overall percent agreement between the Clarity assay and PCR
| Singulex Clarity C. diff toxins A/B assay result | No. of PCR results |
Positive % agreement (95% CI)a | Negative % agreement (95% CI) | Overall % agreement (95% CI) | ||
|---|---|---|---|---|---|---|
| Positive | Negative | Total | ||||
| Positive | 121 | 45 | 166 | 64.7 (54.7–71.5) | 94.5 (92.7–95.9) | 88.9 (86.8–90.8) |
| Negative | 66 | 770 | 836 | |||
| Total | 187 | 815 | 1,002 | |||
CI, confidence interval.
DISCUSSION
Currently available testing options for the diagnosis of CDI lack either sensitivity (toxin EIAs), clinical specificity (NAATs), or rapid time to result (CCCNA and TC). To overcome these limitations, guidelines recommend multistep testing algorithms and/or implementation of collection gatekeepers to CDI diagnostics (7, 8), both of which add complexity and turnaround time but not necessarily diagnostic certainty. In these two prospective, multicenter studies, using fresh samples from almost 2,000 patients with suspected CDI, samples were tested with PCR, toxin and GDH EIAs, CCCNA, and the Clarity toxin assay. This allowed for a large, real-world comparison of multiple diagnostic methods, using consecutive samples submitted to clinical laboratories, both commercial and academic, in geographically diverse populations.
CDI is a toxin-mediated disease, and the presence of toxins better correlates with disease and outcome than the presence of toxigenic organisms alone (10, 11). For example, in a comparison of TC and CCCNA methods, CCCNA correlated with clinical outcome while the presence of toxigenic organism alone (TC+/CCCNA−) was not associated with an increase in mortality (11). The Clarity assay detects free TcdA and TcdB in stool, allowing for the high clinical specificity intrinsic to toxin tests at a clinical sensitivity level that approaches that of NAATs and reference tests. Compared to results with the CCCNA, the Clarity assay yielded 96.3% positive agreement and 93.0% negative agreement after discrepant analysis, while the toxin EIA had 59.8% positive agreement after discrepant analysis, an expected finding given its higher detection limit. Algorithms did not improve accuracy over single-assay testing; the last test used in an algorithm drove the performance of the full algorithm. In the cutoff establishment and validation studies, 38% and 35%, respectively, of PCR-positive samples were toxin negative by Clarity, and there was a 64.7% positive agreement between PCR and Clarity results in the validation study. In a study comparing PCR with toxin assays, over 50% of PCR-positive samples were toxin EIA negative, and patients with PCR+/toxin− samples had outcomes similar to those of patients with PCR-negative samples (10), indicating that the difference in clinical specificities between NAATs and ultrasensitivity toxin detection may be significant.
A limitation of the study is the lack of clinical correlation, and ongoing studies are addressing that. An analytical cutoff for the Clarity assay was set at 12.0 pg/ml by comparing it statistically to CCCNA, the reference free-toxin test. Asymptomatic carriers can also have positive toxin assays (12), and CCCNA testing can lack sensitivity in clinically confirmed cases of CDI (21). Thus, future studies validating the clinical performance of the Clarity assay may provide additional insight into the role of ultrasensitive toxin detection.
In a previous study, the Clarity assay detected toxins in 23% of PCR+/EIA−/CCCNA− samples (13), indicating that Clarity may be more sensitive than CCCNA. Compared to the specificity of CCCNA, the specificity of the Clarity assay in this study may have been underestimated. CCCNA is a functional, user-dependent assay, and given a variable sensitivity, the use of CCCNA as reference method has been questioned (22). Testing in this study was performed on fresh samples without delay, in conjunction with standard-of-care testing, and testing with CCCNA was performed after overnight shipment at 2 to 8°C. Studies using CCCNA and EIA have shown that toxins may deteriorate after extended periods at room temperature but are stable in longer-term storage under refrigerated and frozen conditions (23–25). Using the Clarity assay, samples for C. difficile toxin testing were stable for up to 8 h at room temperature, 1 week at 2 to 8°C, and 3 months at –70°C and up to three freeze-thaw cycles (13). In this study, samples with discrepant results between the Clarity assay and CCCNA were retested with CCCNA when the Clarity result agreed with that of at least one other comparator method. A different CCCNA result was reported for 42% of retested samples, indicating the challenges of comparing free-toxin results to CCCNA results. Currently, new free-toxin tests are required to be compared with CCCNA for regulatory purposes while NAATs should be compared with TC, a method that is known to yield more positive results (26). The contemporary diagnostic standard for CDI remains an area of debate (6) and requires knowledge of clinical variables, which are infrequently reported in large-scale diagnostic studies. CCCNA and broth-enriched TC are standards that have been used to assess diagnostic performance. Differences in gold standard reporting often raise questions pertaining to the actual sensitivity of the reported toxin assay, which may vary by up to 10% to 30%, depending on the comparator test used (27). With the advent of ultrasensitive toxin assays, the use of CCCNA as a reference method for free-toxin detection may be reevaluated.
There has been recent interest in utilizing real-time PCR CT values for prediction of the presence or absence of C. difficile toxin, which is supported by multiple studies (16, 19, 28, 29). CT values for samples from patients with CDI were significantly lower than those from colonized individuals (30), and there was an inverse correlation between CT values and C. difficile fecal load (31), suggesting that CT could be used as a surrogate marker for bacterial load and disease activity. Clinical laboratories may now consider implementing tcdB CT values in CDI diagnostics for prediction of free toxin and estimation of disease severity for treatment guidance. In our study, there was indeed a correlation between toxin concentration and tcdB CT values, but there were a significant number of discordant samples. In the current study, 22.8% of all PCR+/toxin− samples had tcdB CT values of <27.0, and 14.3% of all PCR+/toxin+ samples had CT values of >27.0. Similar results have also recently been reported (13, 32). Although there is a correlation between tcdB CT values and toxin concentration, the accuracy appears to be suboptimal for use in clinical practice. There is a significant risk of misclassifying patients and either treating incorrectly or inappropriately refraining from treatment and infection control measures.
In summary, the Singulex Clarity C. diff toxins A/B assay is a novel assay that uses single-molecule counting technology, leading to improved detection of C. difficile toxins A and B. In this study, there was high overall percent agreement with CCCNA results and significantly higher positive percent agreement than with a toxin EIA. The assay allows for a rapid, direct detection of more reliable clinical CDI markers, free toxins, at a level at which the clinical sensitivity is greatly improved over other free-toxin testing methods. With clinical correlation, this assay could offer an alternative to toxin EIAs and CCCNA as a single-test solution for C. difficile toxin testing in stool. Unfortunately, at the time of publication, Singulex has chosen not to pursue commercial application of the Clarity assay. However, our results shed further light on the potential of single-molecule counting technology for free-toxin detection, which should continue to inform C. difficile diagnostics.
Supplementary Material
ACKNOWLEDGMENTS
We thank the microbiology staff who participated in testing and Kristi Whitfield who helped with illustrations.
J.E., J.S., and E.F. are employees of Singulex, Inc.
This study was supported by Singulex, Inc.
Footnotes
Supplemental material for this article may be found at https://doi.org/10.1128/JCM.00719-19.
REFERENCES
- 1.Lessa FC, Mu Y, Bamberg WM, Beldavs ZG, Dumyati GK, Dunn JR, Farley MM, Holzbauer SM, Meek JI, Phipps EC, Wilson LE, Winston LG, Cohen JA, Limbago BM, Fridkin SK, Gerding DN, McDonald LC. 2015. Burden of Clostridium difficile infection in the United States. N Engl J Med 372:825–834. doi: 10.1056/NEJMoa1408913. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cassini A, Plachouras D, Eckmanns T, Abu Sin M, Blank H-P, Ducomble T, Haller S, Harder T, Klingeberg A, Sixtensson M, Velasco E, Weiß B, Kramarz P, Monnet DL, Kretzschmar ME, Suetens C. 2016. Burden of six healthcare-associated infections on European population health: estimating incidence-based disability-adjusted life years through a population prevalence-based modelling study. PLoS Med 13:e1002150. doi: 10.1371/journal.pmed.1002150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bartlett JG, Gerding DN. 2008. Clinical recognition and diagnosis of Clostridium difficile infection. Clin Infect Dis 46(Suppl 1):S12–S18. doi: 10.1086/521863. [DOI] [PubMed] [Google Scholar]
- 4.Burnham C-A, Carroll KC. 2013. Diagnosis of Clostridium difficile infection: an ongoing conundrum for clinicians and for clinical laboratories. Clin Microbiol Rev 26:604–630. doi: 10.1128/CMR.00016-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Dubberke ER, Burnham C-A. 2015. Diagnosis of Clostridium difficile infection: treat the patient, not the test. JAMA Intern Med 175:1801–1802. doi: 10.1001/jamainternmed.2015.4607. [DOI] [PubMed] [Google Scholar]
- 6.Fang FC, Polage CR, Wilcox MH. 2017. Point-counterpoint: what is the optimal approach for detection of Clostridium difficile infection? J Clin Microbiol 55:670–680. doi: 10.1128/JCM.02463-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.McDonald LC, Gerding DN, Johnson S, Bakken JS, Carroll KC, Coffin SE, Dubberke ER, Garey KW, Gould CV, Kelly C, Loo V, Shaklee Sammons J, Sandora TJ, Wilcox MH. 2018. Clinical practice guidelines for Clostridium difficile infection in adults and children: 2017 update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis 66:e1–e48. doi: 10.1093/cid/cix1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Crobach MJT, Planche T, Eckert C, Barbut F, Terveer EM, Dekkers OM, Wilcox MH, Kuijper EJ. 2016. European Society of Clinical Microbiology and Infectious Diseases: update of the diagnostic guidance document for Clostridium difficile infection. Clin Microbiol Infect 22(Suppl 4):S63–S81. doi: 10.1016/j.cmi.2016.03.010. [DOI] [PubMed] [Google Scholar]
- 9.Surawicz CM, Brandt LJ, Binion DG, Ananthakrishnan AN, Curry SR, Gilligan PH, McFarland LV, Mellow M, Zuckerbraun BS. 2013. Guidelines for diagnosis, treatment, and prevention of Clostridium difficile infections. Am J Gastroenterol 108:478–498. doi: 10.1038/ajg.2013.4. [DOI] [PubMed] [Google Scholar]
- 10.Polage CR, Gyorke CE, Kennedy MA, Leslie JL, Chin DL, Wang S, Nguyen HH, Huang B, Tang Y-W, Lee LW, Kim K, Taylor S, Romano PS, Panacek EA, Goodell PB, Solnick JV, Cohen SH. 2015. Overdiagnosis of Clostridium difficile infection in the molecular test era. JAMA Intern Med 175:1792–1801. doi: 10.1001/jamainternmed.2015.4114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Planche TD, Davies KA, Coen PG, Finney JM, Monahan IM, Morris KA, O'Connor L, Oakley SJ, Pope CF, Wren MW, Shetty NP, Crook DW, Wilcox MH. 2013. Differences in outcome according to Clostridium difficile testing method: a prospective multicentre diagnostic validation study of C difficile infection. Lancet Infect Dis 13:936–945. doi: 10.1016/S1473-3099(13)70200-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pollock NR, Banz A, Chen X, Williams D, Xu H, Cuddemi CA, Cui AX, Perrotta M, Alhassan E, Riou B, Lantz A, Miller MA, Kelly CP. 2019. Comparison of C. difficile stool toxin concentrations in adults with symptomatic infection and asymptomatic carriage using an ultrasensitive quantitative immunoassay. Clin Infect Dis 68:78–86. doi: 10.1093/cid/ciy415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sandlund J, Bartolome A, Almazan A, Tam S, Biscocho S, Abusali S, Bishop JJ, Nolan N, Estis J, Todd J, Young S, Senchyna F, Banaei N. 2018. Ultrasensitive detection of Clostridioides difficile toxins A and B by use of automated single-molecule counting technology. J Clin Microbiol 56:e00908-18. doi: 10.1128/JCM.00908-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sandlund J, Mills R, Griego-Fullbright C, Wagner A, Estis J, Bartolome A, Almazan A, Tam S, Biscocho S, Abusali S, Nolan N, Bishop JJ, Todd J, Young S. 2019. Laboratory comparison between cell cytotoxicity neutralization assay and ultrasensitive single molecule counting technology for detection of Clostridioides difficile toxins A and B, PCR, enzyme immunoassays, and multistep algorithms. Diagn Microbiol Infect Dis 95:20–24. doi: 10.1016/j.diagmicrobio.2019.04.002. [DOI] [PubMed] [Google Scholar]
- 15.Schwenk HT, Bio LL, Kruger JF, Banaei N. 22 November 2018. Clinical impact of Clostridium difficile PCR cycle threshold-predicted toxin reporting in pediatric patients. J Pediatr Infect Dis Soc doi: 10.1093/jpids/piy117. [DOI] [PubMed] [Google Scholar]
- 16.Senchyna F, Gaur RL, Gombar S, Truong CY, Schroeder LF, Banaei N. 2017. Clostridium difficile PCR cycle threshold predicts free toxin. J Clin Microbiol 55:2651–2660. doi: 10.1128/JCM.00563-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Habibzadeh F, Habibzadeh P, Yadollahie M. 2016. On determining the most appropriate test cut-off value: the case of tests with continuous results. Biochem Med (Zagreb) 26:297–307. doi: 10.11613/BM.2016.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhou X, McClish DK, Obuchowski NA. 2011. Statistical methods in diagnostic medicine, 2nd ed Wiley, Hoboken, NJ. [Google Scholar]
- 19.Crobach MJT, Duszenko N, Terveer EM, Verduin CM, Kuijper EJ. 2018. Nucleic acid amplification test quantitation as predictor of toxin presence in Clostridium difficile infection. J Clin Microbiol 56:e01316-17. doi: 10.1128/JCM.01316-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wilmore S, Goldenberg SD. 2018. Potential of real-time PCR threshold cycle (CT) to predict presence of free toxin and clinically relevant C. difficile infection (CDI) in patients with cancer: a reply. J Infect 76:424–426. doi: 10.1016/j.jinf.2018.01.001. [DOI] [PubMed] [Google Scholar]
- 21.Lozniewski A, Rabaud C, Dotto E, Weber M, Mory F. 2001. Laboratory diagnosis of Clostridium difficile-associated diarrhea and colitis: usefulness of Premier Cytoclone A+B enzyme immunoassay for combined detection of stool toxins and toxigenic C. difficile strains. J Clin Microbiol 39:1996–1998. doi: 10.1128/JCM.39.5.1996-1998.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Planche T, Wilcox M. 2011. Reference assays for Clostridium difficile infection: one or two gold standards? J Clin Pathol 64:1–5. doi: 10.1136/jcp.2010.080135. [DOI] [PubMed] [Google Scholar]
- 23.Nancy OM. 2014. Fecal specimens for Clostridium difficile diagnostic testing are stable for up to 72 hours at 4°C. J Med Microbiol Diagn 3:140. [Google Scholar]
- 24.Schora DM, Peterson LR, Usacheva EA. 2018. Immunological stability of Clostridium difficile toxins in clinical specimens. Infect Control Hosp Epidemiol 39:434–438. doi: 10.1017/ice.2018.20. [DOI] [PubMed] [Google Scholar]
- 25.Modi C, DePasquale JR, Nguyen NQ, Malinowski JE, Perez G. 2010. Does the handling time of unrefrigerated human fecal specimens impact the detection of Clostridium difficile toxins in a hospital setting? Indian J Gastroenterol 29:157–161. doi: 10.1007/s12664-010-0040-1. [DOI] [PubMed] [Google Scholar]
- 26.Bouza E, Peláez T, Alonso R, Catalán P, Muñoz P, Créixems MR. 2001. “Second-look” cytotoxicity: an evaluation of culture plus cytotoxin assay of Clostridium difficile isolates in the laboratory diagnosis of CDAD. J Hosp Infect 48:233–237. doi: 10.1053/jhin.2001.1000. [DOI] [PubMed] [Google Scholar]
- 27.Reigadas E, Alcalá L, Marín M, Muñoz-Pacheco P, Catalán P, Martin A, Bouza E. 2016. Clinical significance of direct cytotoxicity and toxigenic culture in Clostridium difficile infection. Anaerobe 37:38–42. doi: 10.1016/j.anaerobe.2015.10.003. [DOI] [PubMed] [Google Scholar]
- 28.Kim HN, Kim H, Moon H-W, Hur M, Yun Y-M. 2018. Toxin positivity and tcdB gene load in broad-spectrum Clostridium difficile infection. Infection 46:113–117. doi: 10.1007/s15010-017-1108-y. [DOI] [PubMed] [Google Scholar]
- 29.Kamboj M, Brite J, McMillen T, Robilotti E, Herrera A, Sepkowitz K, Babady NE. 2018. Potential of real-time PCR threshold cycle (CT) to predict presence of free toxin and clinically relevant C. difficile infection (CDI) in patients with cancer. J Infect 76:369–375. doi: 10.1016/j.jinf.2017.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Biswas JS, Patel A, Otter JA, van Kleef E, Goldenberg SD. 2015. Contamination of the hospital environment from potential Clostridium difficile excretors without active infection. Infect Control Hosp Epidemiol 36:975–977. doi: 10.1017/ice.2015.79. [DOI] [PubMed] [Google Scholar]
- 31.Dionne L-L, Raymond F, Corbeil J, Longtin J, Gervais P, Longtin Y. 2013. Correlation between Clostridium difficile bacterial load, commercial real-time PCR cycle thresholds, and results of diagnostic tests based on enzyme immunoassay and cell culture cytotoxicity assay. J Clin Microbiol 51:3624–3630. doi: 10.1128/JCM.01444-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sandlund J, Wilcox MH. 2019. Ultrasensitive detection of Clostridium difficile toxins reveals suboptimal accuracy of toxin gene cycle thresholds for toxin predictions. J Clin Microbiol 57:e01885-18. doi: 10.1128/JCM.01885-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.