INTRODUCTION: Inflammatory markers such as white cell count (WCC) and C-reactive protein (CRP) and, more recently, bilirubin have been used as adjuncts in the diagnosis of appendicitis. The aim of this study was to determine the diagnostic accuracy of the above markers in acute and perforated appendicitis as well as their value in excluding the condition. METHODS: A retrospective analysis of 1,169 appendicectomies was performed. Patients were grouped according to histological examination of appendicectomy specimens (normal appendix = NA, acute appendicitis = AA, perforated appendicitis = PA) and preoperative laboratory test results were correlated. Receiver operating characteristic (ROC) curve area analysis (area under the curve [AUC]) was performed to examine diagnostic accuracy. RESULTS: ROC analysis of all laboratory variables showed that no independent variable was diagnostic for AA. Good diagnostic accuracy was seen for AA when all variables were combined (WCC/CRP/bilirubin combined AUC: 0.8173). In PA, the median CRP level was significantly higher than that of AA (158mg/l vs 30mg, p<0.0001). CRP also showed the highest sensitivity (100%) and negative predictive value (100%) for PA. CRP had the highest diagnostic accuracy in PA (AUC: 0.9322) and this was increased when it was combined with WCC (AUC: 0.9388). Bilirubin added no diagnostic value in PA. Normal levels of WCC, CRP and bilirubin could not rule out appendicitis. CONCLUSIONS: CRP provides the highest diagnostic accuracy for PA. Bilirubin did not provide any discriminatory value for AA and its complications. Normal inflammatory markers cannot exclude appendicitis, which remains a clinical diagnosis.
INTRODUCTION: Inflammatory markers such as white cell count (WCC) and C-reactive protein (CRP) and, more recently, bilirubin have been used as adjuncts in the diagnosis of appendicitis. The aim of this study was to determine the diagnostic accuracy of the above markers in acute and perforated appendicitis as well as their value in excluding the condition. METHODS: A retrospective analysis of 1,169 appendicectomies was performed. Patients were grouped according to histological examination of appendicectomy specimens (normal appendix = NA, acute appendicitis = AA, perforated appendicitis = PA) and preoperative laboratory test results were correlated. Receiver operating characteristic (ROC) curve area analysis (area under the curve [AUC]) was performed to examine diagnostic accuracy. RESULTS: ROC analysis of all laboratory variables showed that no independent variable was diagnostic for AA. Good diagnostic accuracy was seen for AA when all variables were combined (WCC/CRP/bilirubin combined AUC: 0.8173). In PA, the median CRP level was significantly higher than that of AA (158mg/l vs 30mg, p<0.0001). CRP also showed the highest sensitivity (100%) and negative predictive value (100%) for PA. CRP had the highest diagnostic accuracy in PA (AUC: 0.9322) and this was increased when it was combined with WCC (AUC: 0.9388). Bilirubin added no diagnostic value in PA. Normal levels of WCC, CRP and bilirubin could not rule out appendicitis. CONCLUSIONS:CRP provides the highest diagnostic accuracy for PA. Bilirubin did not provide any discriminatory value for AA and its complications. Normal inflammatory markers cannot exclude appendicitis, which remains a clinical diagnosis.