An evaluation of different laboratory methods in the diagnosis of Clostridium difficile associated diarrhoea (CDAD) in a District General Hospital setting.

Background: CDAD is the most important cause of hospital-acquired diarrhoea, with high morbidity and mortality. Although HPA has reported 35% reduction in incidence during the year 2008/09, rapid and accurate diagnosis of CDAD is essential for improving patient outcome and reducing the nosocomial spread. Recent studies1,2 have shown that current laboratory testing for CDAD is quite variable across various centres and majority of the laboratories rely on the enzyme immunoassay (EIA) which may miss up to 40% of cases. The “Gold standard tests” (cytotoxin assay and culture) are expensive and need technical expertise, and take up to 48 hours to give a definitive result. This study, evaluates the laboratory methods of CDAD diagnosis in a DGH setting.
Methodology:
Setting: 612 bedded busy multispecialty district general hospital, England. All the stool specimens from in-patients submitted to our laboratory during November 08 to January 09 for C. difficile toxin testing, who met the criteria according to laboratory SOP, were included in the study. When multiple samples were sent from same patient, random first sample was selected. Stool specimens from outpatients, requests from GPs, other known infectious cause of diarrhoea, and patients tested positive for toxin within previous 4-8 weeks were excluded from the study. Methods used were rapid immunoassay using TOX A/B QUIK CHEK kit (TechLab, USA)®, Cell cytotoxicity neutralisation assay, detection of Lactoferrin by Leuko test®, isolation of organism by culture on CCEY plate followed by latex agglutination for confirmation. Retrospective case notes review was done to record clinical variables including demographic and epidemiologic features, clinical presentation including stool characters, and risk factors for CDAD including a detailed antibiotic history.
Results:
Demographic features:
A total number of enrolled patients = 100, Average age: 72.58 years, Male:Female = 43:57. Patients were admitted from their own home (87%), nursing home (6%) or residential home (7%).
The average time interval from admission to onset of symptoms = 14days. Proportion of community onset =26%.
 Test results: Stool samples from all 100 patients were tested for 3 tests for the diagnosis of CDAD (EIA, Cytotoxin assay, Culture) and 1 test for the detection of intestinal inflammation (Lactoferrin).
Total number of stool sample processed = 100
Individual test result:
EIA positive = 36, Culture positive = 31, Cytotoxin assay positive = 23, Lactoferrin positive = 32
All three tests for CDAD diagnosis negative = 60
All three tests for CDAD positive = 23
At least one test for CDAD positive = 40
[Please note: All 23 cytotoxin positive tests were Culture and EIA positive as well]
Discrepant results amongst three tests = 17
(EIA + Culture - Cytotoxin - ) = 9, EIA + Culture + Cytotoxin -) =4 (EIA – Culture + Cytotoxin - ) =4
Based on this finding the patients were classified into 3 groups,
Non-CDAD (all 3 tests for CDAD negative) = 60
Definite CDAD (all 3 tests for CDAD positive) = 23
Probable CDAD (at least one test for CDAD positive) = 17
Conclusion: Toxigenic culture seems like an ideal test, with both excellent sensitivity and specificity, but with a prolonged turnaround time.
Newer tests (such as GDH and PCR) and approaches using two or three steps testing may overcome some of the hurdles of CDAD diagnosis.