Clinical utility of C-reactive protein-to-albumin ratio in the management of patients with inflammatory bowel disease

Abstract

Background and Aim

C-reactive protein (CRP)-to-albumin ratio (CAR) is a novel score with prognostic value in inflammatory conditions. This study assessed the performance of CAR as an objective marker of disease activity and prediction of subtherapeutic infliximab trough levels in patients with inflammatory bowel disease (IBD).

Methods

A retrospective study was conducted on three different patient cohorts with IBD: patients who had (i) fecal calprotectin (FC) measurements; (ii) Mayo Endoscopic Scores; and (iii) infliximab trough levels available. The relative performances of CAR, albumin, and CRP were compared in predicting disease activity (based on FC or Mayo Endoscopic Score) and infliximab trough levels.

Results

In both the FC (n = 289) and endoscopy (n = 65) cohorts, albumin and CAR correlated with objective disease activity. CAR (area under the curve [AUC] 0.70) was only marginally better at detecting active disease, measured by FC, compared to CRP (AUC 0.68). A CAR >0.15 was able to detect Mayo 3 disease (AUC 0.83, sensitivity 81%, specificity 89%). Albumin (r = 0.38) and CAR (r = −0.42) correlated with infliximab trough levels (n = 204). The optimal CAR for detecting subtherapeutic infliximab trough levels was >0.08 (AUC 0.70, sensitivity 66%, specificity 64%). Both albumin and CAR were independent predictors of subtherapeutic infliximab trough levels but correlated poorly with infliximab trough levels longitudinally in the same patient.

Conclusion

CAR was only a modest discriminator of subtherapeutic infliximab levels and offers little more than CRP in detecting active disease. CAR has potential to detect severe Mayo 3 disease and could be calculated in patients admitted with suspected acute severe ulcerative colitis.