Does delay in diagnosing colorectal cancer in symptomatic patients affect tumor stage and survival? A population-based observational study

Abstract

Background: Diagnosing colorectal cancer (CRC) at an early stage improves survival. To what extent any delay affects outcome once patients are symptomatic is still unclear.Our objectives were to evaluate the association between diagnostic delay and survival in symptomatic patients with early stage CRC and late stage CRC.

Methods: Prospective population-based observational study evaluating daily clinical practice in Northern Holland. Diagnostic delay was determined through questionnaire-interviews. Dukes' stage was classified into two groups: early stage (Dukes A or B) and late stage (Dukes C or D) cancer. Patients were followed up for 3.5 years after diagnosis.

Results: In total, 272 patients were available for analysis. Early stage CRC was present in 136 patients while 136 patients had late stage CRC. The mean total diagnostic delay (SE) was 31 (1.5) weeks in all CRC patients. No significant difference was observed in the mean total diagnostic delay in early versus late stage CRC (p = 0.27).In early stage CRC, no difference in survival was observed between patients with total diagnostic delay shorter and longer than the median (Kaplan-Meier, log-rank p = 0.93).In late stage CRC, patients with a diagnostic delay shorter than the median had a shorter survival than patients with a diagnostic delay longer than the median (log-rank p = 0.01). In the multivariate Cox regression model with survival as dependent variable and median delay, age, open access endoscopy, number and type of symptoms as independent variables, the odd's ratio for survival in patients with long delay (>median) versus short delay (</=median) was 1.8 (95% confidence interval (CI) 1.1 to 3.0; p = 0.01). Tumor-site was not associated with patient survival. When separating late stage CRC in Dukes C and Dukes D tumors, a shorter delay was associated with a shorter survival in Dukes D tumors only and not in Dukes C tumors.

Conclusion: In symptomatic CRC patients, a longer diagnostic and therapeutic delay in routine clinical practice was not associated with an adverse effect on survival. The time to CRC diagnosis and initiation of treatment did not differ between early stage and late stage colorectal cancer.

Figure 1

Mean total diagnostic delay in early and late stage colorectal cancer.

Figure 2

Mean patient's delay in early and late stage colorectal cancer.

Figure 3

Mean healthcare delay in early and late stage colorectal cancer.

Figure 4

Mean total diagnostic delay in rectal and colon cancer.

Figure 5

Survival analysis in early stage colorectal cancer. Survival in early stage CRC for diagnostic delay longer and shorter than the median delay of 23.5 weeks. On the Y-axis the proportion of patients surviving is plotted. On the X-axis the time is plotted in weeks. Log-rank p = 0.93 using Kaplan-Meier analysis. In the Cox regression model with survival as dependent variable and median delay, age, open access endoscopy, number and type of symptoms as independent variables, the odd's ratio for survival in patients with long delay (>median) versus short delay (≤median) was 1.1 (95% confidence interval 0.5 to 2.6; p = 0.76). Blue line: delay < median delay. Green line: delay > median delay

Figure 6

Survival analysis in late stage colorectal cancer. Survival in late stage CRC for diagnostic delay longer and shorter than the median delay of 23.5 weeks. On the Y-axis the proportion of patients surviving is plotted. On the X-axis the time is plotted in weeks. Log-rank p = 0.01 using Kaplan-Meier analysis. In the multivariate Cox regression model with survival as dependent variable and median delay, age, open access endoscopy, number and type of symptoms as independent variables, the odd's ratio for survival in patients with long delay (>median) versus short delay (≤median) was 1.8 (95% confidence interval (CI) 1.1 to 3.0; p = 0.01). Blue line: delay < median delay. Green line: delay > median delay