Validity of Gram Staining of Stool Samples for Diagnosing Campylobacter Enteritis in Patients with Acute Diarrhea

抄録

BACKGROUND : Because of its high incidence, sensitivity to specific antibiotics, and rare but severe complications, campylobacter enteritis needs to be confirmed or excluded accurately and rapidly. We investigated the validity of Gram staining of a stool sample as a quick and useful method of diagnosing campylobacter enteritis in patients with acute diarrhea presenting at primary healthcare centers.
MATERIALS AND METHODS : Stool samples obtained from 64 patients with acute diarrhea were sent to a laboratory for Gram staining and culture. To estimate the usefulness of Gram staining, we calculated the sensitivity, specificity and likelihood ratio (LR) of Gram staining. Subject profiles, symptoms and peripheral white cell counts were also examined to see if they could raise the pre-test probability prior to the Gram staining test.
RESULTS : Of 64 subjects with acute diarrhea, 38 had C. jejuni (n=37) or C. coli (n=1) (campylobacter group), and 26 had other causes (control group). Gram staining revealed campylobacter-like bacteria (Cb-like bacteria) in 22 samples from the campylobacter group and 3 from the control group, yielding a sensitivity and specificity of 0.58 and 0.88, respectively. The positive LR was 5.02 (95%CI : 1.67-15.05), and the negative LR was 0.48 (0.32-0.71). Other factors such as patient age, disease duration, fever, abdominal pain and leucocytosis failed to raise the pre-test probability prior to Gram staining test. Taking a thorough history of food intake can raise the pre-test probability, although this may be difficult and was not evaluated in this study.
CONCLUSION : Gram staining can assist in making the diagnosis of campylobacter enteritis in patients with acute diarrhea, but it cannot be used alone to make or exclude the diagnosis.