Abstract
Purpose: To create a pre-operatively usable tool to identify patients at high risk of early death (within 90 days post-operatively) after radical cystectomy and to assess potential risk factors for post-operative and surgery related mortality.
Materials and methods: Material consists of 1099 consecutive radical cystectomy (RC) patients operated at 16 different hospitals in Finland 2005–2014. Machine learning methodology was utilized. For model building and testing, the data was randomly divided into training data (n = 733, 66.7%) and independent testing data (n = 366, 33.3%). To predict the risk of early death after RC from baseline variables, a binary classifier was constructed using logistic regression with lasso regularization. Finally, a user-friendly risk table was constructed for practical use.
Results: The model resulted in an area under the receiver operating characteristic curve (AUROC) of 0.73 (95% CI = 0.59–0.87). The strongest risk factors were: American Society of Anesthesiologists physical status classification (ASA), congestive heart failure (CHF), age adjusted Charlson comorbidity index (ACCI) and chronic pulmonary disease.
Conclusion: This study with a novel methodological approach adds CHF and chronic pulmonary disease to previously known independent prognostic risk factors for early death after RC. Importantly, the risk prediction tool uses purely pre-operative data and can be used before surgery.
Author contribution
Dr Riku Klén created the risk tool, designed the study and wrote the manuscript. MD Antti Salminen gathered the data, designed the study and wrote the manuscript. MSc Mehrad Mahmoudian participated in the analyses and edited the manuscript. MD, PhD Kari T. Syvänen edited the manuscript. Dr Laura Elo designed the study and participated in writing the manuscript. MD, PhD Peter J. Boström designed the study and participated in writing the manuscript.
Collaborators
Ilmari Koskinen and Jukka Sairanen, Department of Urology, University of Helsinki and Helsinki University Hospital; Ileana Montoya Perez, Department of Information Technology, University of Turku; Teemu J. Murtola and Petri Virtanen, Department of Urology, University of Tampere and Tampere University Hospital; Markku H. Vaarala and Venla Syri, Department of Urology, University of Oulu and Oulu University Hospital; Timo K. Nykopp, Department of Urology, University of Eastern Finland and Kuopio University Hospital; Marjo Seppänen, Department of Surgery, Division of Urology, Central Hospital of Pori; Taina Isotalo, Department of Surgery, Division of Urology, Central Hospital of Lahti; Timo Marttila and Samuli Virtanen, Department of Surgery, Division of Urology, Central Hospital of Seinäjoki; Lasse Levomäki, Department of Surgery, Division of Urology, Central Hospital of Jyväskylä; Sebastian Becker, Department of Surgery, Division of Urology, Central Hospital of Lappeenranta; Mikael Anttinen and Tapani Liukkonen, Department of Surgery, Division of Urology, Central Hospital of Mikkeli; Matti Säily, Department of Surgery, Division of Urology, Central Hospital of Rovaniemi; Dimitri Pogodin-Hannolainen, Department of Surgery, Division of Urology, Central Hospital of Hämeenlinna; Jouko Viitanen, Department of Surgery, Division of Urology, Central Hospital of Joensuu; Christian Palmberg, Department of Surgery, Division of Urology, Central Hospital of Vaasa; Juhani Ottelin, Department of Surgery, Division of Urology, Central Hospital of Kemi
Disclosure statement
None of the authors have anything to report.