Abstract
Background/Aim: Acute diverticulitis represents a common surgical condition and one of the leading gastrointestinal causes of surgical admissions in Western societies. Complicated diverticulitis increases the length of the hospital stay and the risk of requiring surgical intervention. In areas of limited availability or long waiting times for CT scanning, biochemical predictors of complicated diverticulitis might be valuable. In the available literature, there is no consensus on cut-off values of C-reactive protein or the value of a white cell count in the diagnosis of complicated diverticulitis. Additional studies among different populations are required to add to the existing literature to reach a consensus on diagnostic cut-off levels of inflammatory markers to diagnose complicated diverticulitis. The aim of the present study is to evaluate the predictive value of a white cell count and C-reactive protein, and their sensitivity and specificity in differentiating complicated from uncomplicated diverticulitis.
Methods: This case-control study was performed for patients with acute diverticulitis in Lyell McEwin Hospital in Adelaide, South Australia. Data were collected for consecutive patients admitted from January 2015 to December 2017. Patients with acute diverticulitis confirmed by computed tomography were included in the study. Data of patients with complicated diverticulitis were compared to those of patients with uncomplicated diverticulitis as a control group. Patient characteristics, symptoms, number of attacks of diverticulitis, presence of immunosuppression, past history of complicated diverticulitis, inflammatory markers (white cell count and C-reactive protein), and computed tomography findings were collected and compared.
Results: A total of 106 consecutive cases were recruited for the period from 2015 to 2017. There were 44 cases of complicated diverticulitis and 62 cases with uncomplicated diverticulitis (control group). A white cell count (WCC) and C-reactive protein (CRP) were collected at the time of presentation from the clinical records and pathology reports. A receiver operating characteristic (ROC) analysis was performed and multiple cut-off values for both WCC and CRP were reported. For WCC, the area under curve (AUC) was 0.69 (0.582-0.797) with a P-value of 0.001. At a cut-off of 14, sensitivity was found to be 56.8% and specificity of 80.7%. The sensitivity gradually decreased and specificity gradually increased as the cut-off value increased. At 18 the sensitivity was 25% and specificity was 79%. The positive predictive value for the study sample at WCC of 18 × 109/L or above is 79.5%. For CRP, the AUC was 0.828 (0.729-0.927) with a P-value of <0.001. At a cut-off value of 100 mg/L, the sensitivity was 72.7% and specificity was 80.6%. Sensitivity gradually decreased and specificity increased as the cut-off increased in value. At 160 mg/L, sensitivity was 36.36% and specificity was 97.22% with a positive predictive value of 76%.
Conclusion: Contrary to what has been previously reported in the literature, we found that WCC remains a significant test in diagnosing complicated diverticulitis. A high cut-off value of 18 × 109/L is useful in predicting complicated diverticulitis with high positive predictive value. When compared to WCC, CRP is a more sensitive test in detecting complicated diverticulitis. We recognized a cut-off value of 160 mg/L to be a significant value to rule in complicated diverticulitis with a significant positive predictive value. WCC and CRP are very specific predictors of complicated diverticulitis with high positive predictive value at high cut-off values of 18 × 109/L and 160 mg/L, respectively.