Langenauer J et al, 2018: Advanced non-contrasted computed tomography post-processing by CT-Calculometry (CT-CM) outperforms established predictors for the outcome of shock wave lithotripsy.
Langenauer J, Betschart P, Hechelhammer L, Güsewell S, Schmid HP, Engeler DS, Abt D, Zumstein V.
Department of Urology, Kantonsspital St. Gallen, Rorschacherstrasse 95, 9007, St. Gallen, Switzerland.
Department of Radiology and Nuclear Medicine, Kantonsspital St. Gallen, Rorschacherstrasse 95, 9007, St. Gallen, Switzerland.
Biostatistics, Clinical Trials Unit, Rorschacherstrasse 95, 9007, St. Gallen, Switzerland.
Department of Urology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Abstract
OBJECTIVES: To evaluate the predictive value of advanced non-contrasted computed tomography (NCCT) post-processing using novel CT-calculometry (CT-CM) parameters compared to established predictors of success of shock wave lithotripsy (SWL) for urinary calculi.
MATERIALS AND METHODS: NCCT post-processing was retrospectively performed in 312 patients suffering from upper tract urinary calculi who were treated by SWL. Established predictors such as skin to stone distance, body mass index, stone diameter or mean stone attenuation values were assessed. Precise stone size and shape metrics, 3-D greyscale measurements and homogeneity parameters such as skewness and kurtosis, were analysed using CT-CM. Predictive values for SWL outcome were analysed using logistic regression and receiver operating characteristics (ROC) statistics.
RESULTS: Overall success rate (stone disintegration and no re-intervention needed) of SWL was 59% (184 patients). CT-CM metrics mainly outperformed established predictors. According to ROC analyses, stone volume and surface area performed better than established stone diameter, mean 3D attenuation value was a stronger predictor than established mean attenuation value, and parameters skewness and kurtosis performed better than recently emerged variation coefficient of stone density. Moreover, prediction of SWL outcome with 80% probability to be correct would be possible in a clearly higher number of patients (up to fivefold) using CT-CM-derived parameters.
CONCLUSIONS: Advanced NCCT post-processing by CT-CM provides novel parameters that seem to outperform established predictors of SWL response. Implementation of these parameters into clinical routine might reduce SWL failure rates.
World J Urol. 2018 May 29. doi: 10.1007/s00345-018-2348-x. [Epub ahead of print]
Comments 1
This is work in progress and the authors’ enthusiasm is damped a little bit by the fact that “no re-intervention needed” includes up to 2 SWL retreatments:” Treatment was performed until complete stone fragmentation occurred, including a maximum of three subsequent applications during one inpatient stay. If after three SWL sessions, stones showed no disintegration or disintegration was insufficient (i.e. fragments ≥ 0.4 cm present) and/or there was a need for a reintervention (URS, PCNL or SWL), the patient was considered as treatment failure.”
The number of patients requiring > 1 session is not given. Also, surprisingly, stone analysis is not mentioned, despite the fact that some stones composed of cystine, brushite or whewellit do not respond well to SWL.
312 patients were treated within 5 ½ years. Although it says “312 consecutive patients” a selection bias cannot be excluded as no data about patient selection are given.
It is not mentioned if the CT settings have been changed during those 5 ½ years and radiation doses may be another issue.
Finally the authors state: “…, the number in whom SWL outcome could be predicted with high accuracy based on a single parameter was still rather low (maximum 26% of the patients met the thresholds allowing for prediction with ≥ 80% accuracy). However, … the combination of CT-CM-derived parameters to prognostic models might further improve NCCT-based prediction of SWL outcome, which was beyond the scope of the present study.”
More studies from different authors can be expected as the software needed for CT-CM is a free open-source application (http://www.slicer.org) and semi-automatic determination of all parameters used in this study only took 4 min per patient.