Comparison of endoscopic versus CT assessment of stone-free status after percutaneous nephrolithotomy (PCNL)

Urolithiasis is one of the most common diseases worldwide with increasing prevalence and incidence in the recent decades [1‐3]. Consequently, diagnosis, treatment and prevention of stone recurrence are associated with significant costs [4]. Since the late 1980s, the percutaneous nephrolithotomy (PCNL) technique has been regarded as the standard therapy for large kidney stones with very good stone-free rates (SFR) [5, 6]. According to the European Association of Urology (EAU) guidelines, PCNL is the first-line therapy for kidney stones > 20 mm, whereas kidney stones between 10 and 20 mm can be treated by extracorporeal shock wave lithotripsy (ESWL), ureterorenoscopy (URS) and PCNL [7]. The objective of PCNL is to achieve stone-free-status (SFS) to minimize the risk of future stone-related events and accompanying surgeries and interventions [8]. However, residual fragments (RF) after PCNL are common [9]. To detect RF, imaging methods such as kidney-ureter-bladder radiography (KUB), sonography or low dose computed tomography (LDCT) can be used after the first intervention [9, 10]. Compared to KUB and sonography, LDCT might expose patients to higher doses of radiation but can be performed quickly, does not require contrast agent administration, and detects RF of all sizes with high sensitivity [11, 12]. Only few studies have investigated the correlation between intraoperative SFS defined by the surgeon compared with postoperative LDCT [13, 14].

The aim of this study is to investigate how reliable the surgeon's intraoperative assessment of SFS is and to clarify whether an additional postoperative LDCT is necessary.

The descriptive analysis included a total of 610 PCNL interventions performed between February 2009 and September 2020 (Fig. 1). Patient’s characteristics, as well as intraoperative and postoperative patient´s cohort parameters, are presented in Table 1. In 126 cases, no postoperative imaging was performed because additional surgical interventions were planned or SFS was diagnosed by the surgeon and a CT was not necessary (e.g. in recurrent stone formers) or an ordered CT was not available or a CT was performed 3–4 weeks postoperatively in an outpatient setting.

A study by Harraz et al. from 2017 analyzed data from 306 patients undergoing PCNL which showed that the sensitivity of the intraoperative assessment of SFS by the surgeon was 49.6% and the PPV was 92.8% (specificity: 97.1% NPV: 72%) [16]. In 2018 Nevo et al. analysed data from 312 patients undergoing PCNL. In this study RF < 4 mm were additionally evaluated as stone-free by the surgeon as well as by postoperative imaging. The NPV of the surgeon’s assessment for SFS was 100% (sensitivity: 100%, specificity: 12.5%, PPV: 75%) [14]. In another study, conducted by Jannello et al. in 2022, the match rate between intraoperative SFS and CT-based SFS was determined after vacuum-assisted PCNL [13]. In this study the surgeon was able to accurately predict SFS in 77% of cases. Our results support the high false negative rate and low NPV, which in this study was 50%. Patients with kidney abnormalities and large stones were included in our study, and stones were defined as RF regardless of their size. In addition, it must be noted that in the context of early postoperative imaging, just as with our study, false-positive CT findings e.g. stone dust are possible, which can influence the NPV. Then again if CT imaging is performed 3 months postoperatively, as in the study by Jannello et al. RF can spontaneously go off or even increase in size in the time between surgery and imaging. The EAU recently updated its guidelines and recommends that regular imaging (plain film X-ray and/or sonography) should take place up to 60 months after treatment in case of RF, independent of their size. CT should be considered if patient is symptomatic or if intervention is planned [7]. Regarding RF and aftercare, it can be mentioned at this point that there is only limited data regarding spontaneous stone passage (SSP) according to stone size in case of a RF outlet into the ureter. A study by Yallappa et al. from 2018 analysed 6600 patients with ureteral stones and showed that SPP was reported for 68% of distal ureteral stones, 58% of mid ureteral stones and 49% of upper ureteral stones. Considering stone size almost 75% of stones < 5 mm and 62% of stones > 5 mm passed spontaneously in this study [17]. A more recent multi-center study by Shah et al. from 2019 reported SPP for ureteral stones ≤ 5 mm to be 89% if localized in the distal ureter, 80% if localized in the middle ureter and 71% if localized in the upper ureter [18].

In their studies, Harraz et al. Nevo et al. and Jannello et al. also addressed predictors for an accurate surgeon´s assessment regarding RF. Harraz et al. were able to show that a low GSS had the highest predictive power for a correct surgeon´s assessment for the absence of RF in a multivariate analysis. Nevo et al. showed that the surgeon had missed RF more frequently in patients with increasing stone size and quantity. Jannello et al. were able to identify a larger stone volume, a higher rate of multiple stones, a higher rate of several calyx-groups affected by stones, and a higher rate of GSS II degrees as independent predictors of a discordant stone-free status. Our results confirm the results of Jannello et al. regarding the number of stones and the affected groups of calyxes as well as the results of Nevo et al. regarding the number of stones.

Our multivariable logistic regression analysis shows that an operation time of ≥ 110 min, a GSS degree ≥ II and an initial maximum stone size ≥ 15 mm was associated with a higher rate of postoperative RF. ASA-Classification and bore size was not associated with a higher rate of postoperative RF in our multivariable logistic regression analysis.

The results of this study show that despite the surgeon's ability to detect RF as such after PCNL, postoperative imaging is still necessary to avoid missing RF. Instead of postoperative CT, intraoperative imaging with low radiation exposure, which is sensitive for the assessment of RF, could be performed. The surgeon’s intraoperative assessment of stone-free status would thus be supported by imaging and an improvement in the SFR could be achieved.

The Uro Dyna-CT (ArtisZee Siemens Healthcare Sector, Erlangen, Germany) already exists for the intraoperative detection of RF. It causes lower radiation exposure at CT-like image quality when identifying kidney stones compared to standard CT [19]. Meister et al. showed that RF can be detected using the Uro Dyna-CT and their size can be measured intraoperatively with high accuracy. In 2017 Vincentini et al. reported a case in which Uro Dyna-CT was performed intraoperatively with simultaneous removal of fragments residual stones that were not found by digital fluoroscopy and flexible nephroscopy at the end of surgery [20]. However, installing this system is costly and not widely available [10].

At this point, the possibility of endoscopic combined intrarenal surgery (ECIRS) should also be considered, which, according to a meta-analysis by Widyokirono et al. [21], is an effective and safe treatment particularly for large and complex nephrolithiasis, with significantly higher one-step SFR, a lower necessity for auxiliary procedures, and a lower complication rate compared with PCNL.

This study has limitations due the retrospective design and the unicentric location. The preoperatively collected data on the affected side and the stone location did only provide information about the stones that were planned to be removed as part of the PCNL procedure. Furthermore, a patient could have been treated several times with PCNL in the context of different stone episodes, which leads to limitations in the descriptive statistics of patient history data. Furthermore, nephroscopy was performed anterograde through the tract which may have less adequate access to all calyces as opposed to a supine approach with a flexible ureteroscope.

The surgeon’s intraoperative evaluation on its own is not reliable in evaluating the absence RF. We recommend performing LDCT after PCNL to ensure a stone free status. LDCT may be omitted, if the surgeon reports intraoperative RF. Measures to improve SFR after the first intervention should be taken such as antegrade and retrograde nephroscopy and intraoperative CT imaging if available.