Inflammatory bowel disease: a raised platelet count made it more
likely
|
|
|
Clinical bottom line (level 4)
- The following tests made inflammatory bowel disease more
likely
- elevated platelet count (LR + )
- diarrhoea for > 14 days (LR + )
- low serum albumin (LR + 4.2)
- No test could safely rule out inflammatory bowel
disease.
| |
Harries et al: Journal of Infection 1991; 22 : 247-250
|
Expires June 2004 |
The study Setting: infectious disease unit, acute hospital, UK
239 patients (aged mean 42, 53% male) referred with a diagnosis of
gastroenteritis
Excluded if
Independent unblinded
reference standard, applied in all patients from a non-consecutive
appropriate spectrum. Reference standard:
- Faecal pathogen detected by culture, or inflammatory bowel disease
detected by characteristic histological and/or radiographic
abnormalities
Diagnostic test: clinical features, blood count,
serum albumin
The evidence pre-test probability of inflammatory bowel disease:
11%, (95% CI: 7.3% to 15%)
| diagnostic test |
inflammatory bowel disease |
no inflammatory bowel disease |
LR+ (95% CI) |
post-test probability |
LR- (95% CI) |
post-test probability |
| platelet count > 450 x 109/l |
16 |
3 |
42 (13 to 130) |
84% |
0.41 (0.26 to 0.65) |
5% |
| diarrhoea for > 14 days |
11 |
10 |
8.6 (4.1 to 18) |
52% |
0.62 (0.45 to 0.85) |
7% |
| serum albumin < 35 g/dl |
16 |
30 |
4.2 (2.7 to 6.6) |
35% |
0.47 (0.30 to 0.75) |
6% |
| white cell count > 10 x 109/l |
15 |
53 |
2.2 (1.5 to 3.3) |
22% |
0.59 (0.39 to 0.91) |
7% |
| anaemia (Hb < 12 g/dl in women, < 14 g/dl in men) |
16 |
47 |
2.7 (1.8 to 4.0) |
25% |
0.52 (0.33 to 0.83) |
6% |
| ESR > 30 mm/h |
17 |
72 |
1.9 (1.3 to 2.6) |
19% |
0.56 (0.34 to 0.93) |
7% |
| blood in faeces |
12 |
45 |
2.1 (1.3 to 3.4) |
21% |
0.71 (0.50 to 0.995) |
8% |
| total |
212 |
27 |
Comments
- Patients were selected using a retrospective chart review and only
included if had a definitive diagnosis made. Consequently pre and
post-test probabilites should be treated with caution.
Citation
- Harries AD, Beeching NJ, Rogerson SJ, et al: the platelet count as a
simple measure to distinguish inflammatory bowel disease from infective
diarrhoea. Journal of Infection 1991; 22 : 247-250
Search Terms:
? Contributor: Chris Ball, June 2002 Reviewer:
Clinical Question.
| Patient |
suspected gastroenteritis |
| Intervention or Exposure |
clinical features, blood tests |
| Outcome |
inflammatory bowel disease | |
|