Sensitivity, Specificity Higher With PCR in C Difficile-Associated Diarrhea

November 30, 2009 (Orlando, Florida) — Testing for Clostridium difficile toxin with real-time polymerase chain reaction (PCR) can improve the laboratory diagnoses of C difficile–associated diarrhea, compared with A/B enzyme immunoassay (EIA), according to results from a new study presented here at the Association for Molecular Pathology (AMP) 2009 Annual Meeting.

Dr. Cynthia Essmyer

In fact, real-time PCR demonstrated a sensitivity of 100% and a specificity of 96.9%, whereas the A/B EIA was found to generate both false-positive and false-negative results.

"Although C difficile–associated diarrhea has increased in prevalence and severity, the inaccuracy of conventional C difficile toxin EIA for diagnosis is well documented in the literature," said lead investigator Cynthia Essmyer, MD, medical pathologist at Saint Luke's Hospital in Kansas City, Missouri, during a poster presentation. "The toxin A/B [EIA] is just not cutting it anymore."

Because of this, the investigators sought to evaluate a rapid real-time PCR assay that targets C difficile toxin genes.

"We wanted to determine the feasibility of replacing toxin A/B EIA with real-time PCR at our labs without resorting to a cumbersome test algorithm that would be detrimental to turn-around time," explained Dr. Essmyer.

A Collaborative Effort

In the first part of the study, 204 fresh stool samples were tested with EIA (121 were negative, 83 were positive), and then with real-time PCR on a LightCycler 2.0 (Roche) using MagNA Pure Compact extraction, a method developed and provided by Texas Children's Hospital in Houston. This PCR targeted both C difficile tcdA and tcdB genes.

"Our colleagues down at Texas Children's Hospital had already developed an assay that they were kind enough to share with us," said Dr. Essmyer. "They were testing a different population, looking at pediatrics, and hadn't used this in an adult population. This is actually the first time this has been tested in this way in this population."

In the second part of the study, 66 stool samples from 18 patients with multiple specimens that had inconsistent EIA results were evaluated with PCR.

Samples with discordant results were sent to Mayo Medical Laboratories as the reference laboratory for testing with real-time PCR using alternate primers.

Superior Performance Characteristics

At the end of the study, 179 of the 204 samples (88%) showed concordant results between PCR and EIA on initial testing.

A total of 21 of the 25 discordant samples (14 negative, 7 positive) showed agreement between the reference laboratory and the initial PCR result. The remaining 4 samples, positive on initial PCR, were negative on EIA and reference laboratory PCR.

"We used Mayo's PCR as the gold standard," explained Dr. Essmyer. "And compared with Mayo's PCR, our PCR had 100% sensitivity and 97% specificity. This was a little surprising because we expected it to be more like the low 90s. So we're very pleased with these results."

The second part of the study showed consistent PCR results (13 negative and 5 positive cases) for the patients who had had multiple specimens with inconsistent EIA results. In addition, real-time PCR detected the toxin earlier than EIA in 3 of the 5 positive cases, "precluding the need to test multiple specimens prior to therapy," noted Dr. Essmyer.

"The performance characteristics of C difficile real-time PCR are clearly superior to standard toxin A/B EIA," said Dr. Essmyer. "Plus, these results suggest that real-time PCR may provide the diagnosis earlier in the course of the disease."

She added that "some commercial assays only look for the A gene or for the B, which makes me a little nervous because, although sometimes we find patients who are A only or B only, the majority are both. I think the PCR, which looks for both genes, is invaluable."

Dr. Essmyer reported that her lab has now switched from the A/B EIA to PCR, saving them considerable time. "Unlike before, when we were sent a sequence of 3 or 4 specimens, we're now getting 1 specimen per patient and the PCR clearly shows whether it's positive or negative. So it's just easier on everyone and less time-consuming."

Real-Time PCR — A Useful Replacement

"This is a disease I know well,” said Timothy J. O'Leary, MD, PhD, program chair for the AMP and soon-to-be editor (starting in January) of The Journal of Molecular Diagnostics, which is published by the AMP and the American Society for Investigative Pathology.

“In this study, they're comparing 2 different techniques for diagnosing a particularly common and increasingly aggressive hospital-acquired infection, one that can be life-threatening to individuals," said Dr. O'Leary, who was not involved with this study.

"The toxin assays that have been used for some time to confirm the diagnosis of [C difficile], while fairly specific, are not as sensitive as you would like. Real-time PCR, however, detects DNA associated with the organism and is much more sensitive, detecting a vast majority — if not 100%, very close to it."

He noted that "it's not necessarily the perfect assay because, while it is extremely sensitive, it's possible, for example, that an individual may harbor this organism and not have the disease. You will detect that they are a carrier, which is important knowledge, but it's not actually diagnosing the cause of their diarrhea. So, as with all laboratory tests, this has to be put into context."

Overall, however, "it looks like a very good assay," concluded Dr. O'Leary. "It represents a real advance and I think it's going to be very common that PCR-based tests are used rather than toxin-based tests in the future."

The study was supported by a grant from the Saint Luke's Hospital Foundation. Dr. Essmyer and Dr. O'Leary have disclosed no relevant financial relationships.

Association for Molecular Pathology (AMP) 2009 Annual Meeting: Abstract ID19. Presented November 20, 2009.