Interobserver variability of C lavien– D indo scoring in urology

Abstract

OBJECTIVES: To assess the interobserver variability of Clavien-Dindo scoring in urology. METHODS: A thematic survey was carried out simultaneously in nine Polish urological centers among 171 responders - 81 urologists and 90 urologists-in-training, made up of 16 women and 155 men. After Clavien-Dindo classification was presented, respondents were asked to grade nine typical urological complications according to Clavien-Dindo classification. Interobserver agreement, reproducibility of answers, as well as the influencing factors were analyzed. Questions in the survey related to complications of variable severity after basic urological operations, including four open, one laparoscopic and four endourological. RESULTS: Agreement on the grade of complication was moderate (κ = 0.45) and it was noticed in 69% of respondents (range 38-87%). No effect of surgical approach (classic vs endoscopic, 70% vs 67%, P = 0.64), or physician professional experience (resident vs urologist, 71% vs 66%, P = 0.77) was observed. The most significant discrepancies were found in cases of local complications after transurethral surgery (accordance in 39%) and laparoscopic radical prostatectomy (accordance in 55%), and in cases of severe general complications after Bricker operation (accordance in 58%). CONCLUSIONS: The variability of Clavien-Dindo classification scoring among urologists is significant. Thus, the Clavien-Dindo classification might require detailing before its eventual implementation in urology, together with appropriate training of specialists. However, despite some disadvantages, the simplicity, reproducibility and logical scheme of the Clavien-Dindo classification make it a promising tool for quality assessment in different fields of urology.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 36 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.