Is the Goligher classification a valid tool in clinical practice and research for hemorrhoidal disease?

Background: The most widely used classification for hemorrhoidal disease (HD) is the Goligher classification, which ranks presence and severity of prolapse in four grades. Since physicians base this gradation on medical history and physical examination, it might be prone to interobserver variability. Furthermore, the gradation impacts the treatment of choice which makes reproducibility of utmost importance. The aim of this study was to determine the interobserver variability of Goligher classification among surgeons in the Netherlands. Methods: A single-choice survey was used. The first part consisted of questions concerning baseline characteristics and the use of the Goligher classification in routine clinical practice. In the second part, to assess interobserver variability, we asked gastrointestinal surgeons and residents who routinely treat HD to review 25 photographs (with given timing as during rest or push) of patients with HD and classify the gradation using the Goligher classification. The survey was sent by email on April 19, 2021 and was available online until July 5, 2021. Interobserver variability was assessed using Fleiss’ Kappa test. Results: A total of 329 gastrointestinal surgeons, fellows and residents were sent an invitation email, of whom 95 (29%) completed the survey. Among the respondents, 87% indicated that they use the Goligher classification in clinical practice. Eighty-one percent found the classification helpful and 63% classified HD according to Goligher and followed the guidelines for treatment of HD accordingly. The interobserver variability showed an overall fair strength of agreement, with a Fleiss’ Kappa (κ) of 0.376 (95% CI 0.373–0.380). There was a moderate agreement for grade I and IV HD with a κ statistic of 0.466 and 0.522, respectively. For grades II and III, there was a lower (fair) strength of agreement with 0.206 and 0.378, respectively. Conclusions: The fair interobserver variability is disappointing and demonstrates the need for a more reliable, and internationally accepted, classification for HD. A new classification should enable more uniformity in treating HD and in comparing outcomes of future trials and prospective registries. The protocol for a Delphi study for a new classification system is currently being prepared and led by an international research group.