STORZ MEDICAL – Literature Databases
STORZ MEDICAL – Literature Databases
Literature Databases
Literature Databases

Erkoc M. et al., 2021: Success of Extracorporeal Shock Wave Lithotripsy based on CT texture analysis.

Erkoc M, Bozkurt M, Besiroglu H, Canat L, Atalay HA.
Department of Urology, Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey.
Department of Urology, Faculty of Medical School, Istanbul-Cerrahpasa University, Istanbul, Turkey.
Department of Urology, Beylikduzu State Hospital, Istanbul, Turkey.

Abstract

Objective: The aims of the study were to evaluate whether computerised tomography texture analysis (CTTA) based on non-contrast computed tomography (NCCT) has predictive value for the success of extracorporeal-shockwave lithotripsy (ESWL) in upper urinary tract stones (UUTS).

Methods: This study included 156 of 356 patients undergoing ESWL for UUTS sized 0.5-2 cm from 2015 to 2019. Patients with congenital kidney anomalies, radiolucent stones, multiple stones, treated for upper urinary tract stones previously and lower pole stones were excluded from study. The number of ESWL sessions of the patients was as follows: 78 (50%) patients had 1 session, 30 (19.2%) patients had 2 sessions and 48 (30.8%) patients had >2 sessions. First- and second-order CTTA properties of patients' UUTS were evaluated using texture analysis software (LIFEx Software). Other clinical features such as Hounsfield Unit (HU), initial stone size, body-mass index (BMI) and skin to stone distance (SSD) was recorded. The patients were divided into two groups according to ESWL success. Cases with residual stones larger than 4 mm were considered failed cases.

Results: BMI, the standard deviation of HU, SSD, skewness, kurtosis, entropy and all second-order statistics were found to be statistically different between the two groups except for correlation (P < .05). Multivariate analysis showed longer SSD and four new parameters of CTTA (kurtosis, entropy, dissimilarity and energy by the distribution of pixel grey levels in the UUTS) to be significant predictors for unsuccessful ESWL outcomes. SSD and second-order CTTA properties (dissimilarity and energy) had an area under ROC curve of 0.802, 0.850 and 0.824 at a 95% confidence interval. ESWL success rate in all patients was 76.9%.

Conclusion: CTTA can help select patients who will undergo ESWL for upper urinary tract stones. Thus, we can reduce treatment costs and ESWL complications by preventing unnecessary ESWL applications.
Int J Clin Pract. 2021 Sep 7:e14823. doi: 10.1111/ijcp.14823. Online ahead of print. PMID: 34491588

0
 

Comments 1

Hans-Göran Tiselius on Tuesday, 14 December 2021 09:30

There is in the literature a continuously increasing number of variables suggested for prediction of SWL success. Great attention often is paid to even small differences in measurements recorded in such studies. For a non-radiologist it is difficult to fully understand what the different measures listed below really stand for. The current publication refers to an entirely retrospective study but how powerful the set of data is for prediction of SWL outcome remains to be proven.
It is of note that some observations in commonly used variables are different from what we commonly consider as discriminators. In a comparison between successful (S) and unsuccessful (US) SWL the following notations were recorded (unexpected findings indicated in bold face):
Stone height: S > US
Stone volume: no difference!
Max HU: S > US
MIN HU S > US
SD HU: S >US
SSD: S US
Histoenergy: S US
GLCM dissimilarity S > US
Generally, the recorded differences are small and to me it is not clear to which extent this advanced technology can be used clinically. These measurements are probably of more theoretical than clinical interest. If of clinical value, it might be necessary to summarize the measurements in one index. It seems to be of limited value to use several variables for predicting the outcome of SWL when the method is applied so differently in different units.

Hans-Göran Tiselius

There is in the literature a continuously increasing number of variables suggested for prediction of SWL success. Great attention often is paid to even small differences in measurements recorded in such studies. For a non-radiologist it is difficult to fully understand what the different measures listed below really stand for. The current publication refers to an entirely retrospective study but how powerful the set of data is for prediction of SWL outcome remains to be proven. It is of note that some observations in commonly used variables are different from what we commonly consider as discriminators. In a comparison between successful (S) and unsuccessful (US) SWL the following notations were recorded (unexpected findings indicated in bold face): Stone height: S > US Stone volume: no difference! Max HU: S > US MIN HU S > US SD HU: S >US SSD: S US Histoenergy: S US GLCM dissimilarity S > US Generally, the recorded differences are small and to me it is not clear to which extent this advanced technology can be used clinically. These measurements are probably of more theoretical than clinical interest. If of clinical value, it might be necessary to summarize the measurements in one index. It seems to be of limited value to use several variables for predicting the outcome of SWL when the method is applied so differently in different units. Hans-Göran Tiselius
Friday, 24 May 2024