Statistical analysis
Categorical data were presented with number and percentage of rows. Only
ESWL, preoperative stent requirement, and complication rates were
presented as percentages of column for the convenience of comparison for
the reader. Continuous data were evaluated using the Kolmogorov-Smirnov
test to verify the normality of distribution of variables. Normally
distributed data were expressed with mean + standard deviation (SD), and
non-normally distributed data with median and percentile
(25-75th) values. The independent-samples t-test was
used to compare two independent normally distributed data, while the
Mann-Whitney U test was conducted for the comparison of data without
normal distribution. In the comparison of categorical variables,
Pearson’s or Yate’s chi-square test was used as appropriate. The
relationship of stone size and stone surface area with SFS was evaluated
with the multivariate logistic regression analysis, and stone surface
area was determined as an independent predictive factor [odds ratio
(OR) = 1.004, p = 0.025] (Table 1). Therefore, the measurement of
surface area, which is used in both computed tomography and kidney,
ureter, bladder radiography in clinical practice, was undertaken to
predict stone volume. Possible predictive variables associated with SFS
were evaluated with the multivariate logistic regression analysis, and
the Backward elimination (Wald) method was used to construct a model.
The exclusion criterion for the model was set at p < 0.1. A
new nomogram including stone surface area was created using the
regression coefficients of independent predictive variables. The
predictive ability of the nomogram was evaluated with the receiver
operating characteristic (ROC) analysis. Then, the T.O.HO., STONE and
modified T.O.HO. scores were calculated for each patient. The ability of
the scores to predict SFS was analyzed using the ROC analysis, and
sensitivity and specificity values were calculated by determining the
cut-off value for each scoring. A p value of <0.05 was
considered statistically significant. SPSS software (version 23.0; IBM
Corporation, Armonk, NY, USA) was used for statistical analyses and the
R-project statistical software and “rms” package for the construction
of the nomogram.