Search

P320 Thrombocytosis in acute inflammatory bowel disease: a useful biomarker?

K. Gazelakis1, I. Chu1, C. Martin2, M. Ward1,3, M. Sparrow1,3

1Department of Gastroenterology, The Alfred Hospital, Melbourne, Australia, 2Department of Epidemiology and Preventative Medicine, Monash University, Melbourne, Australia, 3Department of Gastroenterology, Monash University, Melbourne, Australia

Background

Differentiating between infectious gastroenteritis and a flare of inflammatory bowel disease (IBD) can be difficult. Small studies have shown that thrombocytosis may not occur in infectious gastroenteritis. We aimed to determine whether thrombocytosis is a reliable biomarker in distinguishing between these two diagnoses in patients presenting with diarrhoea.

Methods

A retrospective cohort study was conducted at a tertiary referral IBD centre. From January 2000 and December 2018, patients admitted with acute diarrhoea were included. Inclusion criteria were infective gastroenteritis, IBD flare or both. IBD diagnosis was confirmed by standard clinical, radiological and histopathological criteria. Clinical and biochemical parameters were collected.

Results

There were 351 infectious and 506 IBD flare cases. Among these 216 (42.8%) had Crohn’s disease, 276 (54.7%) ulcerative colitis, and 13(2.6%) had IBD-unclassified. Table 1 summarises the main results. Those with acute IBD flare had a longer duration of diarrhoea, bloody diarrhoea, lower albumin and anaemia (p < 0.05 for all comparisons). Patients with infectious diarrhoea were more likely to be older, female, have vomiting and fever and leucocytosis (p < 0.05 for all comparisons). Median platelet count was higher in patients with IBD flares, 334 vs. 220 (p < 0.001) and persisted on multivariate analysis (p < 0.001, OR1.45). On multivariate analysis, other significant associations for IBD flare were age (OR.85, p < 0.001) female sex (OR.23, p < 0.110), blood in faeces (OR 5.98, p < 0.001) vomiting (OR .17, p < 0.001) and albumin (OR.83, p = 0.02). A sub-analysis compared patients with known IBD and infectious gastroenteritis with an identified pathogen (n = 47), with those with an IBD flare alone showed no significant difference in platelet count between groups (419 vs. 465, respectively, p = 0.17).

Table 1.
IBDInfectiousp-value
N506351
Age, mean(SD)37.8(15.6)47.4(22.4)<0.001
Female239(47.2%)193(55.0%)0.026
Diarrhoea, days, median(IQR)14.0(5.0, 28.0)2.0(1.0, 4.0)<0.001
Bloody diarrhoea376(80.2%)104(32.3%)<0.001
Vomiting42(9.5%)146(52.1%)<0.001
Fever50(11.3%)70(24.9%)<0.001
Haemoglobin, mean(SD)123.3(22.1%)141.3(19.3)<0.001
White cells, median (IQR)9.5(7.0,12.5)10.1(7.7, 13.2)0.015
Platelet, median (IQR)344.5(267.0, 465.0)220.0(179.0, 267.0)<0.001
Albumin, median (IQR)31.0(26.0, 35.0)37.0(34.0, 41.0)<0.001

Conclusion

Our study shows significant differences between clinical and biological markers in patients with acute IBD flares compared with those with infectious gastroenteritis. In particular, thrombocytosis occurs in IBD flares but not in infectious gastroenteritis. This biomarker can be used to differentiate between these diagnoses and guide management.