Perioperative predictors of urinary tract infections after percutaneous nephrolithotomy

Table 5 Multivariate logistic regression analysis for the predictors (independent variables) of urinary tract infections after percutaneous nephrolithotomy (dependent variable) (n = 157; Assiut, Egypt, 2026)

Variables	OR1	95%CI	T2	VIF3	P value
Preoperative
Diabetes mellitus	0.15	(0.14-1.13)	0.85	1.18	0.019
History of pyuria	0.40	(0.14-1.13)	0.82	1.22	0.083
Stone size	2.15	(1.1-4.41)	0.55	1.82	0.036
Operative variables
Operation time	1.01	(0.99-1.02)	0.45	2.22	0.452
Multiple renal punctures	1.15	(0.44-3.02)	0.52	1.92	0.774
Blood transfusion	0.34	(0.09-1.3)	0.90	1.11	0.133
Intraoperative double-J placement	1.74	(0.38-7.86)	0.78	1.28	0.474
Nephrostomy tube insertion	0.14	(0.02-1.3)	0.70	1.43	0.084
Postoperative variables
Duration of urethral catheterization	1.1	(0.99-1.22)	0.81	1.23	0.062
Residual stones	0.16	(0.03-0.93)	0.65	1.54	0.041

¹Some odds ratios appear counterintuitive in direction (e.g., odds ratio < 1 for diabetes mellitus and residual stones despite being described as risk factors based on their P-values). This reflects coding direction in the regression model.

²Tolerance > 0.10 confirms the model is statistically sound.

³Variance inflation factor < 5 indicates that multicollinearity is not a concern.

CI: Confidence interval; OR: Odds ratio; T: Tolerance; VIF: Variance inflation factor.