Comparison of Five Commercial Enzyme-Linked Immunosorbent Assays for Detection of Antibodies to Bordetella pertussis

Measuring antibodies to Bordetella pertussis antigens is mostly done by enzyme-linked immunosorbent assays (ELISAs). We compared the performance of five commercially available ELISA kits with the help of 65 serum specimens which were repetitively tested for evaluation of the kits. The specimens contained 20 paired serum samples from patients with clinical pertussis, 15 samples were from children vaccinated with a diphtheria-tetanus-acellular pertussis vaccine, seven specimens were taken from an interlaboratory comparison of ELISAs, and there were three reference preparations from the Food and Drug Administration's (FDA's) Laboratory of Pertussis and from our laboratory. Reference values were obtained from the FDA or from results obtained with an in-house ELISA. Commercial ELISAs were compared with respect to their reproducibility and variability, their ability to detect significant titer rises in paired serum samples, their ability to detect an immune response after vaccination, and the comparability of semiquantitative and quantitative results. Reproducibility was generally good (>89%), intra-assay variation ranged from 2.4 to 28.7%, and indeterminate results were recorded in up to 18.5% of all specimens. Most kits correctly identified the antibody response to an acellular pertussis vaccine. None of the commercial kits identified all cases of pertussis correctly, and the sensitivity ranged between 60 and 95%. All five commercial ELISAs showed great discrepancies when comparing semiquantitative results and contained obviously different antigen preparations. Our data suggest that the five commercial ELISAs tested here need further improvement and standardization.     

Evaluation of Three Commercially Available Influenza A Type-Specific Blocking Enzyme-Linked Immunosorbent Assays for Seroepidemiological Studies of Influenza A Virus Infection in Pigs

The reverse zoonotic transmission of the pandemic H1N1 2009 influenza virus to swine necessitates enhanced surveillance of swine for influenza virus infection. Using a well-characterized panel of naturally infected swine sera, we evaluated and optimized the performances of three commercially available competitive enzyme-linked immunosorbent assays (ELISAs), namely, the IDEXX Influenza A Ab test, IDEXX AI MultiS-Screen Ab test, and IDVet ID Screen influenza A antibody competition ELISA, for detecting influenza A virus-reactive antibodies in swine. Receiver operating characteristic (ROC) analysis suggests that adjustment of the manufacturer-recommended cutoff values optimizes the sensitivity and specificity of these assays, making them applicable for seroepidemiology studies of swine influenza. Using such optimized cutoff levels, the sensitivity and specificity of the IDEXX Influenza A Ab test were 86% and 89%, respectively; those for the IDEXX AI MultiS-Screen Ab test were 91% and 87%, respectively; and those for the IDVet ID Screen influenza A test were 95% and 79%, respectively.     

Comparison of Seven Commercial Antigen and Antibody Enzyme-Linked Immunosorbent Assays for Detection of Acute Dengue Infection

Seven commercial assays were evaluated to determine their suitability for the diagnosis of acute dengue infection: (i) the Panbio dengue virus Pan-E NS1 early enzyme-linked immunosorbent assay (ELISA), second generation (Alere, Australia); (ii) the Panbio dengue virus IgM capture ELISA (Alere, Australia); (iii) the Panbio dengue virus IgG capture ELISA (Alere, Australia); (iv) the Standard Diagnostics dengue virus NS1 antigen ELISA (Standard Diagnostics, South Korea); (v) the Standard Diagnostics dengue virus IgM ELISA (Standard Diagnostics, South Korea); (vi) the Standard Diagnostics dengue virus IgG ELISA (Standard Diagnostics, South Korea); and (vii) the Platelia NS1 antigen ELISA (Bio-Rad, France). Samples from 239 Thai patients confirmed to be dengue virus positive and 98 Sri Lankan patients negative for dengue virus infection were tested. The sensitivities and specificities of the NS1 antigen ELISAs ranged from 45 to 57% and 93 to 100% and those of the IgM antibody ELISAs ranged from 85 to 89% and 88 to 100%, respectively. Combining the NS1 antigen and IgM antibody results from the Standard Diagnostics ELISAs gave the best compromise between sensitivity and specificity (87 and 96%, respectively), as well as providing the best sensitivity for patients presenting at different times after fever onset. The Panbio IgG capture ELISA correctly classified 67% of secondary dengue infection cases. This study provides strong evidence of the value of combining dengue virus antigen- and antibody-based test results in the ELISA format for the diagnosis of acute dengue infection.     

Validation of a Commercially Available Monoclonal Antibody-Based Competitive-Inhibition Enzyme-Linked Immunosorbent Assay for Detection of Serum Antibodies to Neospora caninum in Cattle

A previously described monoclonal antibody (MAb)-based competitive-inhibition enzyme-linked immunosorbent assay (cELISA) was modified to optimize performance, and the assay was validated in various defined cattle populations for detection of serum antibody to Neospora caninum, a major cause of bovine abortion. Modifications to the cELISA included capturing native N. caninum antigen with a parasite-specific MAb (MAb 5B6-25) and directly conjugating the competitor MAb (MAb 4A4-2), with both MAbs binding different epitopes of a conserved, immunodominant 65-kDa tachyzoite surface antigen. The assay was validated using three serum sets, a "gold standard" set of 184 cow sera defined by fetal histopathology and N. caninum immunohistochemistry and by maternal N. caninum indirect fluorescence assay (IFA) at a 1:200 serum dilution, a relative standard set of 330 cow sera defined by IFA alone, and a set of 4,323 cow sera of unknown N. caninum status. A test cutoff of 30% inhibition was identified. The diagnostic sensitivity was 97.6%, and diagnostic specificity was 98.6% for the gold standard abortion-defined sera. The diagnostic sensitivity was 96.4%, and diagnostic specificity was 96.8% for the relative standard IFA-defined sera. Testing of the 4,323 bovine sera of unknown N. caninum status revealed a distinct bimodal distribution and steep sigmoid frequency curve with only 1.8% of samples within 5% of the test cutoff, indicating a sharp discrimination between test-positive and test-negative samples. In summary, the modified N. caninum cELISA provided a simple, rapid, and versatile method to accurately identify N. caninum infection status in cattle using a single cutoff value.     

Comparison of Five Commercial Anti-Tetanus Toxoid Immunoglobulin G Enzyme-Linked Immunosorbent Assays

Five commercially available enzyme-linked immunosorbent assays for the measurement of anti-tetanus toxoid immunoglobulin G (IgG) antibodies were evaluated for performance. The data suggest that there are manufacturer-dependent differences in sensitivity and accuracy for the determination of tetanus toxoid IgG antibodies that could result in different diagnostic interpretations.The measurement of anti-tetanus toxoid immunoglobulin G (IgG) levels is of value in (i) determining the rates of immunity within broad populations or the immune status of individuals who may be at risk of infection (15), (ii) assessing responses to vaccination and immunization schedule efficacy (2, 5, 15), and (iii) evaluating individuals for potential immunodeficiency disorders (1, 3).The World Health Organization guidelines suggest that IgG antibody titers above 0.01 IU/ml afford minimal protection against infection and titers above 0.1 IU/ml provide substantial protection (2).Significant differences in performance against international reference preparations and clinical interpretation of a number of patient samples have been found previously in commercial anti-tetanus toxoid enzyme-linked immunosorbent assays (ELISAs) (14). The present study confirms these findings, increases the number of commercial assays evaluated, and extends the comparison of their performance in terms of diagnostic interpretations.Normal human serum samples were obtained from Research Sample Bank, Inc., Pompano Beach, FL, and Golden West Biologicals Inc., Temecula, CA, and stored at −20°C prior to testing. The reference sample NIBSC 76/589 (NIBSC, Hertfordshire, United Kingdom) was used to evaluate the calibration of the ELISA kits. It was chosen because it is correlated against the mouse in vivo neutralization test, the concentration was more appropriate for the measuring ranges of clinically relevant ELISA kits, and it has been used previously (14). For use, NIBSC 76/589 was reconstituted according to the manufacturer''s instructions (working concentration of 1 IU/ml), serially diluted with distilled water to a final concentration of 0.03 IU/ml, and further diluted immediately into the appropriate sample diluents.Anti-tetanus toxoid IgG antibodies were measured according to the manufacturers'' instructions by using the following ELISA kits, with the measuring ranges in parentheses: Euroimmun, Lübeck, Germany (0.01 to 10 IU/ml); Scimedx Corp., Denville, NJ (0.1 to 5 IU/ml); Serion-Verion, Würzburg, Germany (0.1 to 5 IU/ml); The Binding Site, Birmingham, United Kingdom (the TBS assay; 0.01 to 7 IU/ml); and Genzyme Virotech, Rüsselsheim, Germany (0.1 to 5 IU/ml). Results were generated in accordance with the manufacturer''s instructions. Assays were considered valid when quality control parameters were in the range specified in the manufacturer''s product insert. Intraassay precision was measured by using three serum samples (low, medium, and high levels) and assayed in seven-well repeats at the same time. For interassay precision, the same measurements were performed in triplicate over 3 consecutive days. Intra- and interassay precision was assessed by calculating the coefficient of variation. Calibration was assessed by calculating the “recovery” of NIBSC 76/589. To calculate results, values (IU/ml) for serially diluted NIBSC 76/589 were obtained from a calibration curve and compared to the expected values according to the equation (obtained NIBSC value/expected NIBSC value) × 100. The results are expressed as a percentage.All P values were two tailed, and results were considered significant at a P of <0.05.The intra- and interassay precision of each kit was calculated (Table ​(Table1).1). The measuring ranges of the Virotech, Serion, and Scimedx assays were only designed to measure antibody levels as low as 0.1 IU/ml. The precision data for Scimedx were limited because the high-level sample values were significantly lower (P < 0.0001) than those obtained in the other four assays. Precision ranged from 3% to 23%, with the lowest values achieved with the TBS (3.2% to 9.8%) and Virotech (3.6% to 10.8%) assays. Poor interassay precision (>20%) was evident with the high-level samples in the Scimedx and Serion ELISAs.

TABLE 1.

Calculated intra- and interassay precision of five commercial ELISAs

Comparison of Two Commercial Tick-Borne Encephalitis Virus IgG Enzyme-Linked Immunosorbent Assays

Despite the availability of protective vaccines, tick-borne encephalitis virus (TBEV) infections have been increasingly reported to the European Centre for Disease Prevention and Control in the past 2 decades. Since the diagnosis of TBEV exposure relies on serological testing, we compared two commercial enzyme-linked immunosorbent assays (ELISAs), i.e., Immunozym FSME IgG assay (ELISA-1) and Euroimmun FSME Vienna IgG assay (ELISA-2). Both assays use whole TBEV antigens, but they differ in viral strains (Neudoerfl for ELISA-1 and K23 for ELISA-2) and cutoff values. In testing of samples from 398 healthy blood donors, ELISA-1 showed higher reactivity levels than ELISA-2 (P < 0.001), suggesting different assay properties. This finding was supported by Bland-Altman analysis of the optical density at 450 nm (OD₄₅₀) (mean bias, +0.32 [95% limits of agreement, −0.31 to +0.95]) and persisted after transformation into Vienna units. Concordant results were observed for 276 sera (69%) (44 positive and 232 negative results). Discordant results were observed for 122 sera (31%); 15 were fully discordant, all being ELISA-1 positive and ELISA-2 negative, and 107 were partially discordant (101 being ELISA-1 indeterminate and ELISA-2 negative and 6 having positive or indeterminate reactivity in both ELISAs). Neutralization testing at a 1:10 dilution yielded positive results for 33 of 44 concordant positive sera, 1 of 15 fully discordant sera, and 1 of 33 partially discordant sera. Indirect immunofluorescence testing revealed high antibody titers of ≥100 for yellow fever virus in 18 cases and for dengue virus in one case, suggesting that cross-reactivity contributed to the ELISA-1 results. We conclude that (i) cross-reactivity among flaviviruses remains a limitation of TBEV serological testing, (ii) ELISA-2 revealed reasonable sensitivity and specificity for anti-TBEV IgG population screening of human sera, and (iii) neutralization testing is most specific and should be reserved for selective questions.     

Evaluation of Enzyme-Linked Immunosorbent Assays for Detection of Mycoplasma bovis-Specific Antibody in Bison Sera

Mycoplasma bovis has recently emerged as a significant and costly infectious disease problem in bison. A method for the detection of M. bovis-specific serum antibodies is needed in order to establish prevalence and transmission patterns. Enzyme-linked immunosorbent assays (ELISAs) validated for the detection of M. bovis-specific serum IgG in cattle are commercially available, but their suitability for bison sera has not been determined. A collection of bison sera, most from animals with a known history of infection or vaccination with M. bovis, was tested for M. bovis-specific IgG using commercially available kits as well as an in-house ELISA in which either cattle or bison M. bovis isolates were used as a source of antigen. Comparison of the results demonstrates that ELISAs optimized for cattle sera may not be optimal for the identification of bison seropositive for M. bovis, particularly those with low to moderate antibody levels. The reagent used for the detection of bison IgG and the source of the antigen affect the sensitivity of the assay. Optimal performance was obtained when the capture antigen was derived from bison isolates rather than cattle isolates and when a protein G conjugate rather than an anti-bovine IgG conjugate was used for the detection of bison IgG.     

Glycoprotein-Based Enzyme-Linked Immunosorbent Assays for Serodiagnosis of Infectious Laryngotracheitis

For detection of infectious laryngotracheitis virus (ILTV) antibody, glycoprotein B-, C-, and D-based enzyme-linked immunosorbent assays (B-, C-, and D-ELISAs, respectively) were developed. The B- and D-ELISAs showed enhanced detection of anti-ILTV antibodies in infected chickens compared to that of the commercial ELISA. Furthermore, the D-ELISA was efficient in detecting seroconversion with vectored vaccine, using recombinant Newcastle disease virus (rNDV) expressing glycoprotein D (gD) as the vaccine vector.     

Evaluation of a New Commercially Available Immunoglobulin M Capture Enzyme-Linked Immunosorbent Assay for Diagnosis of Japanese Encephalitis Infections

A new commercial enzyme-linked immunosorbent assay (ELISA) for the diagnosis of Japanese encephalitis virus infections showed a sensitivity of 88% with sera and 81% with cerebrospinal fluid and a specificity of 97% with sera from patients with primary and secondary dengue virus infections. Specificity was 100% when samples from nonflavivirus infections were tested.     

Development of Two Antibody Detection Enzyme-Linked Immunosorbent Assays for Serodiagnosis of Human Chronic Fascioliasis

Coprological examination based on egg detection in stool samples is currently used as the gold standard for the diagnosis of human fascioliasis. However, this method is not effective during the acute phase of the disease and has poor sensitivity during the chronic phase. Serodiagnosis has become an excellent alternative to coprological examination in efforts to combat the effects of fascioliasis on human and animal health. Two novel recombinant Fasciola hepatica proteins, i.e., a ferritin (FhFtn-1) and a tegument-associated protein (FhTP16.5), were used as antigens to develop in-house enzyme-linked immunosorbent assay (ELISA) methods. The assays were optimized and validated using 152 serum samples from humans with a known infection status, including healthy subjects, patients with chronic fascioliasis, and patients with other parasitic diseases. The FhFtn-1 ELISA was shown to be 96.6% sensitive and 95.7% specific; the respective parameters for the FhTP16.5 ELISA were 91.4% and 92.4%. The performances of the FhFtn-1 and FhTP16.5 ELISAs were compared with that of an available commercial test (the DRG test) using a subset of serum samples. Our in-house tests were slightly more sensitive than the DRG test in detecting antibodies against F. hepatica, but the differences were not statistically significant. In conclusion, the present study provides evidence for the potential of the FhFtn-1 and FhTP16.5 ELISAs as diagnostic tools for human fascioliasis, as might be implemented in conjunction with standard assays for large-scale screenings in areas where the disease is endemic and for the detection of occasional cases in clinical laboratories.     

Evaluation of a New Commercially Available Immunoglobulin M Capture Enzyme-Linked Immunosorbent Assay for Diagnosis of Dengue Virus Infection

We evaluated the FDA-cleared InBios dengue virus (DENV) IgM capture enzyme-linked immunosorbent assay (ELISA) for qualitative detection of anti-DENV IgM antibodies from 79 serum samples obtained from dengue virus-infected patients or suspected dengue cases. The agreement, sensitivity, and specificity of the InBios assay compared to the gold standard in-house DENV IgM capture ELISA were 94, 92, and 94%, respectively. We conclude that the InBios DENV IgM capture ELISA can be effectively used for rapid diagnosis of acute or recent DENV infection.     

Enzyme-Linked Immunosorbent Assays for Detection of Equine Antibodies Specific to Sarcocystis neurona Surface Antigens

Sarcocystis neurona is the primary causative agent of equine protozoal myeloencephalitis (EPM), a common neurologic disease of horses in the Americas. We have developed a set of enzyme-linked immunosorbent assays (ELISAs) based on the four major surface antigens of S. neurona (SnSAGs) to analyze the equine antibody response to S. neurona. The SnSAG ELISAs were optimized and standardized with a sample set of 36 equine sera that had been characterized by Western blotting against total S. neurona parasite antigen, the current gold standard for S. neurona serology. The recombinant SnSAG2 (rSnSAG2) ELISA showed the highest sensitivity and specificity at 95.5% and 92.9%, respectively. In contrast, only 68.2% sensitivity and 71.4% specificity were achieved with the rSnSAG1 ELISA, indicating that this antigen may not be a reliable serological marker for analyzing antibodies against S. neurona in horses. Importantly, the ELISA antigens did not show cross-reactivity with antisera to Sarcocystis fayeri or Neospora hughesi, two other equine parasites. The accuracy and reliability exhibited by the SnSAG ELISAs suggest that these assays will be valuable tools for examining the equine immune response against S. neurona infection, which may help in understanding the pathobiology of this accidental parasite-host interaction. Moreover, with modification and further investigation, the SnSAG ELISAs have potential for use as immunodiagnostic tests to aid in the identification of horses affected by EPM.     

Specific Enzyme-Linked Immunosorbent Assays for Quantitation of Antibody-Cytokine Fusion Proteins

Preliminary testing has shown in vitro and in vivo that antitumor activity can be obtained with fusion proteins linking tumor-reactive monoclonal antibodies to cytokines, such as granulocyte-macrophage colony-stimulating factor or interleukin 2 (IL-2). Preclinical and clinical testing of these reagents requires their in vitro and in vivo quantitation and pharmacokinetic evaluation. We have focused on the detection of a fusion protein which links one human IL-2 molecule to the carboxy terminus of each heavy chain of the tumor-reactive human-mouse chimeric anti-GD2 antibody, ch14.18. We have developed enzyme-linked immunosorbent assays (ELISAs) to evaluate intact tumor-reactive fusion proteins. By these ELISAs we can reliably measure nanogram quantities of intact ch14.18-IL-2 fusion protein and distinguish the intact protein from its components (ch14.18 and IL-2) in buffer, mouse serum, and human serum with specificity and reproducibility. The measurement of intact ch14.18-IL-2 fusion protein is not confounded by free IL-2 or free ch14.18 when 100 ng or less of total immunoglobulin per ml is used during the assay procedure. Our results indicate that these ELISAs are suitable for preclinical and clinical testing and with slight modifications are applicable to the analysis of a variety of other fusion proteins.     

Evaluation of Four Enzyme Immunoassays for the Detection of Giardia lamblia Antigen in Stool Specimens

 Four enzyme immunoassays for the detection of Giardia lamblia antigen in stool specimens were evaluated: the ProSpecT Giardia Microplate Assay (Alexon, USA), the Giardia CELISA (Cellabs, Australia), the DSl-Giardia-ELISA (DSL, Germany), and the Melotest Giardiasis Ag (Melotec, Spain). Microscopic examination and enzyme immunoassays were performed on 168 stool specimens collected from 168 patients suspected to have giardiasis. All assays were easy to perform. The ProSpecT Giardia assay had the highest sensitivity of the assays evaluated (91%), and its interpretation was the easiest. The sensitivity of the three other assays ranged from 63 to 81%. The ProSpecT Giardia assay can be useful to detect Giardia lamblia and may replace microscopic examination in areas of high endemicity.     

Detection of Bovine Herpesvirus 1 Glycoprotein E Antibodies in Individual Milk Samples by Enzyme-Linked Immunosorbent Assays

The purpose of this study was to determine whether individual milk samples can replace serum samples for the detection of bovine herpesvirus 1 (BHV1) glycoprotein E (gE)-specific antibodies. Serum and milk samples were collected at the same time from cattle in BHV1-free herds, cattle in unvaccinated herds, and cattle in herds that were vaccinated twice with a BHV1 marker vaccine. The samples were tested in two gE enzyme-linked immunosorbent assay (ELISA) systems. In comparison to serum, the results showed that the gE-blocking ELISA was highly sensitive for testing milk samples (0.96). In contrast, the gE ELISA was less sensitive (0.79). The specificities of the gE-blocking ELISA and the gE ELISA for testing milk samples were very high (1.00 and 0.99, respectively). The presented results indicate that individual milk samples, which can be collected relatively easily and inexpensively, can be used instead of individual serum samples in the gE-blocking ELISA for the screening of cattle for BHV1 gE antibodies.     

Comparison of Two Enzyme-Linked Immunosorbent Assays and One Rapid Immunoblot Assay for Detection of Herpes Simplex Virus Type 2-Specific Antibodies in Serum

The sensitivities and specificities of three immunoassays for the detection of herpes simplex virus type 2 (HSV-2)-specific immunoglobulin G antibodies in serum, including the one-strip rapid immunoblot assay (RIBA; Chiron Corporation) and two indirect enzyme immunosorbent assays (EIA; Gull Laboratories and Centocor), were compared by testing a panel of 1,250 serum samples from individuals attending an outpatient clinic for sexually transmitted diseases. A qualitative agreement among the three assays was observed with 1,080 serum samples (86.4%); 291 of the serum samples (23.3%) were positive, 789 samples (63.1%) were negative, and 170 serum samples (13.6%) gave a discordant result. Results were considered conclusive when a concordant result was obtained with two of three assays. The sensitivities and specificities of the RIBA, the Gull EIA, and the Centocor EIA proved to be 99.2, 99.7, and 89.9% and 97.1, 96.7, and 99.3%, respectively. These results indicate that the Chiron RIBA and the Gull EIA are especially useful and reliable for the detection of HSV-2-specific antibodies in serum.     

Evaluation of Three Commercial Varicella-Zoster Virus IgG Enzyme-Linked Immunosorbent Assays in Comparison to the Fluorescent-Antibody-to-Membrane-Antigen Test

Commercial serologic assays for varicella-zoster virus (VZV), which enable reliable determination of VZV immune status and are amenable to automation, are needed. The present study compares the automated performance of the VZV whole-cell enzyme-linked immunosorbent assay (ELISA) Enzygnost anti-VZV/IgG, the Euroimmun anti-VZV ELISA (IgG) based on highly purified viral proteins, and the VZV glycoprotein (gp)-based Serion ELISA Classic VZV IgG. The fluorescent-antibody-to-membrane-antibody (FAMA) test was used as a reference. A total of 638 serum samples from VZV-negative children, blood donors, varicella vaccinees, and bone marrow transplant recipients were included. The Enzygnost anti-VZV/IgG and the Serion ELISA Classic VZV IgG showed sensitivities of 99.6% and 99.2%, respectively, and the Euroimmun anti-VZV ELISA (IgG) had a significantly lower sensitivity of 90.5%. Specificity was calculated as 100% for both the Euroimmun anti-VZV ELISA (IgG) and for the Enzygnost anti-VZV/IgG, and the Serion ELISA Classic VZV IgG had a significantly lower specificity of 89.4%. Quantitative results of all ELISAs correlated well, but there was a poor quantitative correlation between the ELISAs and FAMA. In conclusion, this study does not show any superiority of a gp- and a protein-based ELISA compared to a whole-cell ELISA for the automated detection of VZV-specific IgG. The automated performance of the Enzygnost anti-VZV/IgG assay correlated best with the FAMA reference assay.     

Evaluation of ColorPAC Giardia/Cryptosporidium Rapid Assay and ProSpecT Giardia/Cryptosporidium Microplate Assay for Detection of Giardia and Cryptosporidium in Fecal Specimens

Detection of Giardia and Cryptosporidium in clinical stool specimens using the ColorPAC and ProSpecT enzyme immunoassays revealed 98.7% agreement for Giardia detection and 98.1% agreement for Cryptosporidium detection. Sensitivities were uniformly 100%. The specificities of the ColorPAC immunoassay for Giardia and Cryptosporidium detection were 100 and 99.5%, respectively, and those for the ProSpecT assay were 98.4 and 98.6%, respectively. The false-positive reactions with the ProSpecT assay occurred with specimens that were grossly bloody.     

Development and Validation of Digital Enzyme-Linked Immunosorbent Assays for Ultrasensitive Detection and Quantification of Clostridium difficile Toxins in Stool

The currently available diagnostics for Clostridium difficile infection (CDI) have major limitations. Despite mounting evidence that toxin detection is paramount for diagnosis, conventional toxin immunoassays are insufficiently sensitive and cytotoxicity assays too complex; assays that detect toxigenic organisms (toxigenic culture [TC] and nucleic acid amplification testing [NAAT]) are confounded by asymptomatic colonization by toxigenic C. difficile. We developed ultrasensitive digital enzyme-linked immunosorbent assays (ELISAs) for toxins A and B using single-molecule array technology and validated the assays using (i) culture filtrates from a panel of clinical C. difficile isolates and (ii) 149 adult stool specimens already tested routinely by NAAT. The digital ELISAs detected toxins A and B in stool with limits of detection of 0.45 and 1.5 pg/ml, respectively, quantified toxins across a 4-log range, and detected toxins from all clinical strains studied. Using specimens that were negative by cytotoxicity assay/TC/NAAT, clinical cutoffs were set at 29.4 pg/ml (toxin A) and 23.3 pg/ml (toxin B); the resulting clinical specificities were 96% and 98%, respectively. The toxin B digital ELISA was 100% sensitive versus cytotoxicity assay. Twenty-five percent and 22% of the samples positive by NAAT and TC, respectively, were negative by the toxin B digital ELISA, consistent with the presence of organism but minimal or no toxin. The mean toxin levels by digital ELISA were 1.5- to 1.7-fold higher in five patients with CDI-attributable severe outcomes, versus 68 patients without, but this difference was not statistically significant. Ultrasensitive digital ELISAs for the detection and quantification of toxins A and B in stool can provide a rapid and simple tool for the diagnosis of CDI with both high analytical sensitivity and high clinical specificity.