Preoperative nomogram to predict the likelihood of complications after radical nephroureterectomy

Jay D. Raman, Yu Kuan Lin, Shahrokh F. Shariat, Laura Maria Krabbe, Vitaly Margulis, Alex Arnouk, Costas D. Lallas, Edouard J. Trabulsi, Sarah J. Drouin, Morgan Rouprêt, Gregory Bozzini, Pierre Colin, Benoit Peyronnet, Karim Bensalah, Kari Bailey, David Canes, Tobias Klatte

Research output: Contribution to journalArticlepeer-review

29 Scopus citations

Abstract

Objectives: To construct a nomogram based on preoperative variables to better predict the likelihood of complications occurring within 30 days of radical nephroureterectomy (RNU). Patients and Methods: The charts of 731 patients undergoing RNU at eight academic medical centres between 2002 and 2014 were reviewed. Preoperative clinical, demographic and comorbidity indices were collected. Complications occurring within 30 days of surgery were graded using the modified Clavien–Dindo scale. Multivariate logistic regression determined the association between preoperative variables and post-RNU complications. A nomogram was created from the reduced multivariate model with internal validation using the bootstrapping technique with 200 repetitions. Results: A total of 408 men and 323 women with a median age of 70 years and a body mass index of 27 kg/m2 were included. A total of 75% of the cohort was white, 18% had an Eastern Cooperative Oncology Group (ECOG) performance status ≥2, 20% had a Charlson comorbidity index (CCI) score >5 and 50% had baseline chronic kidney disease (CKD) ≥ stage III. Overall, 279 patients (38%) experienced a complication, including 61 events (22%) with Clavien grade ≥ III. A multivariate model identified five variables associated with complications, including patient age, race, ECOG performance status, CKD stage and CCI score. A preoperative nomogram incorporating these risk factors was constructed with an area under curve of 72.2%. Conclusions: Using standard preoperative variables from this multi-institutional RNU experience, we constructed and validated a nomogram for predicting peri-operative complications after RNU. Such information may permit more accurate risk stratification on an individual cases basis before major surgery.

Original languageEnglish (US)
Pages (from-to)268-275
Number of pages8
JournalBJU International
Volume119
Issue number2
DOIs
StatePublished - Feb 1 2017

All Science Journal Classification (ASJC) codes

  • Urology

Fingerprint

Dive into the research topics of 'Preoperative nomogram to predict the likelihood of complications after radical nephroureterectomy'. Together they form a unique fingerprint.

Cite this