Glutamate Dehydrogenase for Laboratory Diagnosis of Clostridium difficile Infection

We read with interest the paper “Evaluation of the C.Diff Quik Chek Complete Assay, a New Glutamate Dehydrogenase and A/B Toxin Combination Lateral Flow Assay for Use in Rapid, Simple Diagnosis of Clostridium difficile Disease” by Sharp and colleagues (10).

Interest in the laboratory diagnosis of Clostridium difficile has been encouraged by research that has shown the commonly used toxin enzyme immunoassays (EIAs) to be inadequate when used alone (1, 9).

We agree with Sharp et al. (10) that glutamate dehydrogenase (GDH) is a very sensitive (but poorly specific) assay and can accurately rule out the presence of C. difficile in stool samples. Sharp and colleagues (10) used a commercially available combined GDH and toxin A/B assay (C.Diff Quik Chek Complete) in the first step of their testing algorithm, as others have done (2, 8, 11). However, we note that there were no true-positive samples that were GDH negative but EIA toxin positive in the 284 specimens they tested.

We made the same observation in our own study of 500 specimens (3), and others have reported similar results (6, 7, 11). Thus, it is very rare to have a GDH-negative, EIA toxin-positive result for a true-positive sample. Therefore, we believe the toxin component of the C.Diff Quik Chek Complete assay is redundant. We prefer to use the GDH-only assay (C.Diff Chek-60), which is less expensive and allows for automated processing using the Dynex DS2 platform. This saves hands-on time and avoids possible misinterpretation of ambiguous results, since it produces a numerical result.

We note that Sharp and colleagues had one false-positive result with their testing algorithm, giving a sensitivity and specificity of 100% and 99.6%, respectively (10). Using those authors' figures, we calculated performance characteristics using our preferred algorithm of screening with the C.Diff Chek-60 GDH assay, followed by confirmation with GeneXpert PCR, and found a sensitivity and specificity of 100% and 100%, respectively. Furthermore, the GeneXpert system can simultaneously indicate the presence of the potentially “hypervirulent” ribotype 027 strain (4), giving important epidemiological information not provided by the C.Diff Quik Chek Complete test.

Using the Sharp and colleagues' figures for material cost per test (10), we also calculated the cost of each algorithm for testing 284 specimens. Those authors suggested the algorithm would cost $4,200.64, and ours cost slightly less at $3,923.30 (Table 1). Turnaround times for these algorithms would not differ significantly (depending upon degree of batching), but significant reductions may be possible by using GeneXpert as a point-of-care test. Further study is warranted in this area.

TABLE 1.

Material cost comparison for two suggested diagnostic algorithms

Diagnostic algorithm	Component assay	Material cost per test ($)	No. of tests required	Total material cost ($)
Chek Complete (GDH plus toxin A/B), followed by PCR	C. Diff Chek Complete	11.50	284	3,266.00
	GeneXpert PCR	33.38	28	934.64
Total cost				4,200.64
Chek-60 (GDH alone), followed by PCR	C. Diff Chek 60	7.35	284	2,087.40
	GeneXpert PCR	33.38	55	1,835.90
Total cost				3,923.30

It is clear that the GDH test is sufficiently sensitive to work well as an initial screen, but we do not believe any benefit is derived by coupling it with a toxin EIA. Furthermore, a dedicated, automated GDH test is easier to interpret.

In England, we are now seeing a decrease in the prevalence of Clostridium difficile infection (which is a mandatory reportable disease) (5), and this will inevitably have an impact upon the positive and negative predictive values of any diagnostic tests. It is critical, therefore, that we have robust, accurate methods of identifying patients with this organism.