Percutaneous nephrolithotomy versus retrograde intrarenal surgery for treatment of renal stones in adults

Background Kidney stones (also called renal stones) can be a source of pain, obstruction, and infection. Depending on size, location, composition, and other patient factors, the treatment of kidney stones can involve observation, shock wave lithotripsy, retrograde intrarenal surgery (RIRS; i.e. ureteroscopic approaches), percutaneous nephrolithotomy (PCNL), or a combination of these approaches. Objectives To assess the effects of percutaneous nephrolithotomy (PCNL) versus retrograde intrarenal surgery (RIRS) for the treatment of renal stones in adults. Search methods We performed a comprehensive search of the Cochrane Library, MEDLINE, Embase, Scopus, and two trials registries up to 23 March 2023. We applied no restrictions on publication language or status. Selection criteria We included randomized controlled trials that evaluated PCNL (grouped by access size in French gauge [Fr] into three groups: ≥ 24 Fr [standard PCNL], 15–23 Fr [mini‐PCNL and minimally invasive PCNL], and < 15 Fr [ultra‐mini‐, mini‐micro‐, super‐mini‐, and micro‐PCNL]) versus RIRS. Data collection and analysis Two review authors independently selected studies and extracted data from the included studies. Our primary outcomes were stone‐free rate, major complications, and need for secondary interventions. Our main secondary outcomes were unplanned medical visits to emergency/urgent care or outpatient clinic, length of hospital stay, ureteral stricture or injury, and quality of life. We performed statistical analyses using a random‐effects model. We rated the certainty of evidence using GRADE criteria. We adopted a minimally contextualized approach with predefined thresholds for minimal clinically important differences (MCIDs). Main results We included 42 trials assessing the effects of PCNL versus RIRS in 4571 randomized participants. Twenty‐two studies were published as full‐text articles, and 20 were published as abstract proceedings. The average size of stones ranged from 10.1 mm to 39.1 mm. Most studies did not report sources of funding or conflicts of interest. The main results for the most important outcomes are summarized below. Stone‐free rate PCNL compared with RIRS may improve stone‐free rates (risk ratio [RR] 1.13, 95% confidence interval [CI] 1.08 to 1.18; I2 = 71%; 39 studies, 4088 participants; low‐certainty evidence). Based on 770 participants per 1000 being stone‐free with RIRS, this corresponds to 100 more (62 more to 139 more) stone‐free participants per 1000 with PCNL (an absolute difference of 10%, where the predefined MCID was 5%). Major complications PCNL compared with RIRS probably has little or no effect on major complications (RR 0.86, 95% CI 0.59 to 1.25; I2 = 15%; 34 studies, 3649 participants; moderate‐certainty evidence). Based on 31 complications in the RIRS group, this corresponds to six fewer (13 fewer to six more) major complications per 1000 with PCNL (an absolute difference of 0.6%, where the predefined MCID was 2%). Need for secondary interventions PCNL compared with RIRS may reduce the need for secondary interventions (RR 0.31, 95% CI 0.17 to 0.55; I2 = 61%; 21 studies, 2005 participants; low‐certainty evidence). Based on 222 secondary interventions in the RIRS group, this corresponds to 153 fewer (185 fewer to 100 fewer) secondary interventions per 1000 with PCNL (an absolute difference of 15.3%, where the predefined MCID was 5%). Unplanned medical visits No studies reported unplanned medical visits. Length of hospital stay PCNL compared with RIRS may extend length of hospital stay (mean difference 1.04 days more, 95% CI 0.27 more to 1.81 more; I2 = 100%; 26 studies, 2804 participants; low‐certainty evidence). This effect size is greater than the predefined MCID of one day. Ureteral stricture or injury PCNL compared with RIRS may have little or no effect on the occurrence of ureteral strictures (RR 0.93, 95% CI 0.39 to 2.21; I2 = 0%; 13 studies, 1574 participants; low‐certainty evidence). Based on 14 ureteral strictures in the RIRS group, this corresponds to one fewer (nine fewer to 17 more) ureteral strictures per 1000 with PCNL (an absolute difference of 0.1%, where the predefined MCID was 2%). Quality of life No studies reported quality of life. Authors' conclusions Based on a large body of evidence from 42 trials, we found that PCNL compared with RIRS may improve stone‐free rates and may reduce the need for secondary interventions, but probably has little or no effect on major complications. PCNL compared with RIRS may have little or no effect on ureteral stricture rates and may increase length of hospital stay. We found no evidence on unplanned medical visits or participant quality of life. Because of the considerable shortcomings of the included trials, the evidence for most outcomes was of low certainty. Access size for PCNL was less than 24 Fr in most studies that provided this information. We expect the findings of this review to be helpful for shared decision‐making about management choices for individuals with renal stones.