Background/aims: Although erythrocyte sedimentation rate (ESR) is included as a laboratory parameter in Truelove and Witts' classification, C-reactive protein (CRP) is also used for severity assessment in ulcerative colitis (UC). Frequently, the discordance between ESR and CRP is observed in clinical practice. The aim of this study was to determine which parameter is more related with clinical activity in UC patients.
Methods: A total of 155 patients with UC were identified from January 2004 to March 2005. Their medical records were reviewed within these patients, a total of 541 assessments of disease activity were made. Correlation of clinical activity and laboratory tests were evaluated by Pearson's correlation coefficient.
Results: Pearson's correlation coefficients of ESR and CRP with clinical symptoms were 0.376 and 0.258, respectively. The correlation coefficient between ESR and CRP was 0.403 (p=0.000). A total of 131 (24.2%) assessments revealed discordance between ESR and CRP. When discordance occurred, the correlation coefficients with clinical symptoms were 0.338 for ESR (p=0.000) and 0.034 for CRP (p>0.01). Dividing discordant patients into high ESR/low CRP group and low ESR/high CRP group, the coefficients were 0.420 for ESR and 0.226 for CRP in high ESR/low CRP group, and 0.333 for ESR and 0.068 for CRP in low ESR/high CRP group.
Conclusions: The correlation analysis indicates that ESR appears to be a more reliable laboratory parameter of disease activity than CRP in assessing the severity of UC. In particular, when the level of ESR and CRP is discordant, ESR is more useful in assessing the disease activity in UC patients.