Tubeless procedure reduces hospitalization and pain after percutaneous nephrolithotomy: results of a multivariable analysis.

The objective of the study was to assess the efficacy and safety of tubeless percutaneous nephrolithotomy (t-PCNL) in comparison with standard PCNL (s-PCNL). We retrospectively evaluated 317 consecutive PCNL and compared perioperative results, time of hospitalization and analgesic requirement of t-PCNL (114; 36.0 %) to s-PCNL (203; 64.0 %). The decision to perform a tubeless PCNL was made at the end of the procedures depending on the surgeon's preference and according to the following inclusion criteria: (a) no serious bleeding or perforation in the collecting system during the procedure; (b) patients with no more than one access; and (c) residual stone burden needing a second-stage nephroscopy. Staghorn stones and anatomic anomalies were not considered as exclusion criteria for t-PCNL. Univariate analyses were conducted with one-way ANOVA, Fisher's exact test, Pearson's Chi-square and linear-by-linear association test as appropriate. Stepwise multivariable regression analyses were used to assess the independent correlation between demographics and clinical variables and the clinical outcomes. There were no significant differences between the two groups in terms of stone-free rate, hemoglobin decrease, blood transfusion and complication rate. Mean hospital stay was significantly shorter in the t-PCNL group (3.3 vs. 4.6 days; P < 0.001). Tubeless PCNL was associated with less analgesia requirement (68.4 vs. 86.7 %; P < 0.001) and with lower analgesic dose requirement (1.6 vs. 2.1 mean doses; P = 0.010). Multivariable analyses showed that t-PCNL (P < 0.001), postoperative fever (P < 0.001), transfusions (P < 0.001), operative time (P = 0.002), postoperative hydronephrosis (P = 0.005) and residual fragment dimension (P = 0.024) were independently correlated with duration of hospitalization, while analgesic dose requirement was independently influenced by hemoglobin decrease (P < 0.001), t-PCNL (P = 0.005) and stone number (P = 0.044). Our study confirmed that t-PCNL has similar outcomes to s-PCNL in terms of stone-free rate without increasing complications in selected cases. t-PCNL is a factor independently associated with shorter hospitalization and lower analgesic requirement.