Diverticular Inflammation and Complication Assessment classification, CODA score and fecal calprotectin in clinical assessment of patients with diverticular disease: A decision curve analysis

Abstract

Background and aims: The Diverticular Inflammation and Complication Assessment (DICA) classification and the Combined Overview on Diverticular Assessment (CODA) were found to be effective in predicting the outcomes of Diverticular Disease (DD). We ascertain whether fecal calprotectin (FC) can further aid in improving risk stratification.

Methods: A three-year international, multicentre, prospective cohort study was conducted involving 43 Gastroenterology and Endoscopy centres. Survival methods for censored observations were used to estimate the risk of acute diverticulitis (AD) in newly diagnosed DD patients according to basal FC, DICA, and CODA. The net benefit of management strategies based on DICA, CODA and FC in addition to CODA was assessed with decision curve analysis, which incorporates the harms and benefits of using a prognostic model for clinical decisions.

Results: At the first diagnosis of diverticulosis/DD, 871 participants underwent FC measurement. FC was associated with the risk of AD at 3 years (HR per each base 10 logarithm increase: 3.29; 95% confidence interval, 2.13-5.10) and showed moderate discrimination (c-statistic: 0.685; 0.614-0.756). DICA and CODA were more accurate predictors of AD than FC. However, FC showed high discrimination capacity to predict AD at 3 months, which was not maintained at longer follow-up times. The decision curve analysis comparing the combination of FC and CODA with CODA alone did not clearly indicate a larger net benefit of one strategy over the other.

Conclusions: FC measurement could be used as a complementary tool to assess the immediate risk of AD. In all other cases, treatment strategies based on the CODA score alone should be recommended.

Keywords: CODA score; DICA score; acute diverticulitis; diverticular disease; diverticulosis; fecal calprotectin.

FIGURE 1

Kaplan–Meier curves of cumulative incidence of diverticulitis from patients categorized into high and low fecal calprotectin (FC) levels at baseline. The threshold (90 μg/g) was selected by maximally selected rank statistics. This threshold is used purely for illustrative reasons and should not be considered as the best threshold to adopt for long‐term decision making.

FIGURE 2

Box plots displaying (a) baseline fecal calprotectin (FC) in patients who developed diverticulitis versus those who did not; (b) by Diverticular Inflammation and Complication Assessment (DICA) endoscopic classification levels; (c) and by Combined Overview on Diverticular Assessment (CODA) score. The p‐values reported correspond to a two‐sample Wilcoxon rank‐sum test (a) and the Kruskal–Wallis test (b and c).

FIGURE 3

Decision curve analysis plotting the net benefit of management strategies adopted on the basis of three prognostic tools predicting the 3‐year risk of diverticulitis in patients with Diverticular Disease (DD). The net benefit corresponding to using the Diverticular Inflammation and Complication Assessment (DICA) classification (in orange), the Combined Overview on Diverticular Assessment (CODA) score (in green) and the CODA score plus fecal calprotectin (FC) are compared to strategies to “treat all” (diagonal dashed line) and “treat none” (horizontal dashed line). The net benefit, plotted on the y axis, is a metric representing the benefit of a certain intervention minus its harms multiplied by an exchange rate (x axis). The unit of net benefit is true positive. A net benefit of 0.05, for instance, means to find “5 true positives for every 100 patients in the target population” with no harms (i.e., benefit is “net”). The net benefit is plotted over a range of possible decision thresholds/exchange rates (i.e., individual predicted probabilities derived by applying the prognostic tool). The net benefit also incorporates any consequence (i.e., clinical actions taken) of knowing the individual risk of a subsequent diverticulitis. The net benefit of five different strategies is compared. The two extreme—default—strategies are “treat all” (diagonal dashed line) and “treat none” (horizontal dashed line), meaning enacting clinical actions as if all patients with DD will develop diverticulitis (i.e., “treat all”), or as if nobody of them will develop diverticulitis (i.e., “treat none”). The x axis can also be renamed preference: clinicians more worried about the harms of a missed diverticulitis (i.e., true positive) will adopt thresholds closer to a predicted probability of zero (i.e., left side of the graph), while clinicians more worried about the harms/costs of unnecessary interventions/visits (i.e., on false positives) will adopt higher thresholds (i.e., right side of the graph). The x axis is also called “exchange rate”, which is an odds ratio and represents how many false positives are worth one true positive (i.e., adopting a threshold probability of 20% means that a patient with a predicted probability over 20% will be considered likely to develop diverticulitis—and treated accordingly—and by adopting this classification rule/threshold, one accepts that one true positive is worth four false positives). Interventions associated with different harms/costs may need the adoption of different thresholds/exchange rates. The prognostic tool corresponding to the highest net benefit over the largest range of threshold probabilities should be adopted. When this is unclear, the simpler prognostic tool or the one corresponding to lower costs/harms/inconvenience should be used. For a comprehensive guidance regarding interpreting the decision curve analysis, please see eAppendix 2.

FIGURE 4

Flow‐chart suggesting the possible short‐term (3‐month) risk stratification of patients with newly diagnosed colonic diverticulosis detected on endoscopy. CODA, Combined Overview on Diverticular Assessment; DICA, Inflammation and Complication Assessment; FC, fecal calprotectin.