High-sensitivity CRP at diagnosis and during follow-up in patients with Crohn's disease: Is it a marker for patient classification?

Aim: C-reactive protein (CRP) is a traditional non-specific marker of inflammation and Crohn's disease (CD) is associated with a strong CRP response. No clear cut-off values exist. Our aim was to investigate whether a classification based on the high-sensitivity (hs)-CRP value at diagnosis is useful for the prediction of disease phenotype and relapse rate in patients with CD.

Methods: Two-hundred-sixty well-characterized, unrelated, consecutive CD patients (m/f: 120/140, duration: 7.0±6.1years) with a complete clinical follow-up were included. Medical records including disease phenotype according to Montreal classification, extraintestinal manifestations, smoking habits, medical therapy and surgical events were analyzed retrospectively. Hs-CRP and clinical activity according Harvey-Bradshaw Index was followed-up consecutively between 1 January, 2008 and 1 June, 2010. Results: 32.3% of CD patients had normal hs-CRP at diagnosis. Elevated hs-CRP at diagnosis was associated with disease location (p=0.002), non-inflammatory disease behavior (p=0.058) and need for later azathioprine/biological therapy (p<0.001 and p=0.024). The accuracy of hs-CRP for identifying patients with active disease later during the course of the disease was good (AUC: 0.82, cut-off: 10.7mg/L). AUC was better in patients with an elevated hs-CRP at diagnosis (AUC: 0.92, sensitivity: 96%, specificity: 79%, PPV: 83%, NPV: 95%). In Kaplan-Meier and Cox-regression analysis, hs-CRP was independent predictor of 3- (p=0.007) or 12-month (p=0.001) clinical relapses in patients with an elevated hs-CRP at diagnosis. In addition, perianal involvement (p=0.01) was associated with the 12-month relapse frequency.

Conclusion: Our data suggest that classification based on hs-CRP value at diagnosis is useful for identifying complicated disease phenotype, active disease and risk of relapses during follow-up.