Objectives: Subclassification of achalasia based on high-resolution manometry (HRM) may be clinically relevant because response to therapy may vary by subtype. However, the consistency and reliability of subtyping achalasia patients based on HRM remains undefined. The objectives of this study were to assess interrater and intrarater agreement (reliability) of achalasia subtyping using the Chicago classification, and to evaluate the diagnostic consistency between clinicians interpreting HRM. Methods: After receiving training on the classification criteria, five raters classified 20 achalasia and 10 non-achalasia cases in separate sessions 1 week apart. To further assess agreement, two raters classified all 101 available achalasia HRMs. Agreement for the classification of subtypes of achalasia was calculated using Cohen's and Krippendorff's α-reliability estimate. Results: Estimates of agreement among raters was good during both sessions (α=0.75; 95% confidence interval=0.69, 0.81 and α=0.75; 95% confidence interval=0.68, 0.81). Both interrater (κ0.86-1.0) and intrarater (κ0.86-1.0) agreement were very good for type III achalasia. Agreement between types I and II was more variable. Reliability was improved when type I and type II were combined (α=0.84; 95% confidence interval=0.78, 0.89). When all available cases were classified by two experienced raters, agreement was very good (κ0.81; 95% confidence interval=0.71, 0.91). Conclusions: Interobserver and intraobserver agreement for differentiating achalasia from non-achalasia patients using HRM and the Chicago classification was very good to excellent. More variability was seen in agreement when classifying achalasia subtypes. The most variation was observed in classification between type I and type II achalasia, which have similar characteristics. Clearly, differentiating between panesophageal pressurization and compartmentalization should improve discrimination between these subtypes.
ASJC Scopus subject areas