Pushing the Envelope: Ambulatory Percutaneous Nephrolithotomy for Patients with Complex Renal Calculi

Introduction

Percutaneous nephrolithotomy (PCNL) is well-established as the gold standard in the surgical management of large-volume kidney stones. ¹ Over the last several decades, a multitude of technological and procedural improvements have been introduced to make PCNL safer, more efficient, and less burdensome for patients. Modifications including tract miniaturization, omission of nephrostomy tubes and stents, and access techniques have turned a procedure that historically required inpatient admission into one that is conducive to same-day discharge in appropriately selected patients. ^{2 –4}

Although ambulatory PCNL has been shown to be safe in patients with minimal comorbidities and smaller stone burdens, the implementation of same-day discharge for patients with significant comorbidities or complex stones has not been widely adopted because of its higher risk of intraoperative blood loss and postoperative complications. ⁵ Herein, we report our experience performing same-day (ambulatory) PCNL on complex stones. We hypothesized that compared with inpatient PCNL, ambulatory PCNL in this subset of patients would yield similar safety outcomes and stone-free rates (SFRs).

Methods

We performed a retrospective review of all PCNLs performed at our institution from April 2019 through January 2021. Institutional review board (IRB) approval was obtained prior to study initiation (IRB #14-00879). Preoperative computed tomography (CT) images were used to determine stone complexity in all patients. Adult patients with complex stones by either Guy’s stone score (score 3 or 4) (Supplementary Table S1) or by S.T.O.N.E. nephrolithometry score (score ≥ 9) (Supplementary Table S2) who underwent unilateral PCNL were included in the study. ^6,7 Our practice of routinely admitting patients after PCNL was changed in large part out of necessity because of the overwhelming inpatient burden of the COVID-19 pandemic and coincided with hospital admission policies. We converted to a policy of same-day discharge after PCNL on October 16, 2020. Patients treated before this practice change defined the admission cohort—all of whom were admitted for at least 1 night postoperatively—whereas those treated after defined the ambulatory cohort. Patients with intended same-day discharge who were subsequently admitted for any reason were still included in the ambulatory cohort.

All procedures were performed by one of two high-volume, fellowship-trained endourologists and there were no significant changes in operative technique over the study duration. When retrograde access was possible, procedures were initiated with retrograde insertion of an occlusion balloon catheter. Percutaneous access was obtained under ultrasound guidance unless a pre-existing nephrostomy tube was present, in which case, the nephrostomy was used for access. Balloon dilation to 24 Fr was performed in all cases, and stone fragmentation was performed using the Swiss LithoClast Trilogy® Lithotripter (Boston Scientific). In general, we do not leave nephrostomy tubes at the end of PCNL. A stent is omitted in patients without infectious stones, absence of hydronephrosis preoperatively, and no evidence of inflammation or edema at the ureteropelvic junction or proximal ureter intraoperatively; all others receive a double pigtail stent at the end of the procedure. In the ambulatory cohort, urethral drainage (Foley) catheters were left in place for 1 hour postoperatively, and patients were required to void prior to discharge (except for those with incontinent urinary diversions). In the admission cohort, Foley catheters were left in place overnight and were removed on the morning of postoperative day 1 (POD#1) if the patient was hemodynamically stable.

Electronic medical records were reviewed for patient demographics, baseline characteristics, and stone information. Primary outcomes included 30-day rates of postoperative emergency department (ED) visits, readmissions, and Clavien–Dindo (CD) complications, as well as POD#1 complications. Secondary outcomes included the frequency of conversion to admission (for intended ambulatory patients) and SFR. In accordance with the Journal of Endourology criteria, SFR was calculated based on patients with postoperative noncontrast CT scans and classified as Grade A (no fragments), Grade B (fragments ≤ 2 mm), Grade C (fragments 2.1–4 mm), or not stone-free (fragments > 4 mm). Categorical variables were compared across groups via chi-square or Fisher’s exact tests, whereas continuous variables were compared using Mann–Whitney U tests. Multivariate logistic regression analysis was performed to identify whether ambulatory PCNL was a predictor of each primary outcome when compared with admitted patients. Multivariate covariates were identified either a priori or if the variable demonstrated a p-value of <0.1 on univariate analysis.

Results

Of the 569 patients screened for eligibility, 190 (79 ambulatory, 111 admitted) met inclusion criteria for complexity and were included in the study. Although age, gender, body mass index (BMI), and American Society of Anesthesiologists (ASA) score were not significantly different between groups, patients in the overnight admission group were more likely to have neurological disease, be on antiplatelet or anticoagulation therapy, and to have a history of recurrent urinary tract infections (Table 1). Moreover, the admitted group had a higher proportion of patients with Guy’s score 4 or STONE nephrolithometry score ≥ 9, multiaccess PCNL, and generally had a greater stone burden and a longer total operative time. In addition, although intraoperative stents were placed in the vast majority of patients in both groups, same-day discharge did not significantly increase the likelihood of receiving a stent.

The frequency of 30-day postoperative complications (ambulatory = 20.3%, admitted = 25.2%, p = 0.423) and POD#1 complications (ambulatory = 1.3%, admitted = 7.2%, p = 0.083) was lower among ambulatory patients, but these differences were not significant (Table 2). Furthermore, the rates of complications across each CD category also did not significantly differ across groups. The proportion of patients undergoing either an ED visit or readmission was similar across ambulatory (13.9%) and admitted (11.7%) groups, p = 0.651. Furthermore, for both cohorts, the median time to presentation after surgery for an ED visit or readmission was 4 days. However, compared with the overnight group, the frequency of having an ED visit only was significantly higher among ambulatory patients (8.9% vs 1.8%, p = 0.024), whereas the likelihood of undergoing readmission was lower (5.1% vs 9.9%, p = 0.222)—although this difference was not significant. No patients in the admitted cohort required overnight ED observation, whereas only one of the seven ambulatory cohort patients with an ED visit only required overnight ED observation.

Reasons for 30-day ED visit or hospital readmission were similar across groups and included pain-related (ambulatory = 3.8%, admitted = 0.9%, p = 0.309), bleeding-related (ambulatory = 3.8%, admitted = 1.8%, p = 0.651), infection-related (ambulatory = 3.8%, admitted = 4.5%, p = 1.000), and other (ambulatory = 2.5%, admitted = 4.5%, p = 0.701). Overall, only one patient in the ambulatory group was converted to admission, which was because of a fever requiring overnight intravenous antibiotics. The CT-determined SFR was similar in both groups (no fragments: ambulatory = 80.6%, admitted = 70.6%), but admitted patients were more likely to have larger residual fragments (fragments ≤ 2 mm: ambulatory = 16.1%, admitted = 17.6%; fragments 2.1–4 mm: ambulatory = 0.0%, admitted = 5.9%; fragments > 4 mm: ambulatory = 3.2%, admitted = 5.9%), although none of the differences in SFR were statistically significant.

On multivariate analysis, compared with admitted patients, same-day discharge was an independent risk factor for undergoing an ED visit only (odds ratio = 16.095 [1.544–167.794], p = 0.020) (Supplementary Table S3. However, same-day discharge was not a significant predictor of any other primary outcome including 30-day complications, POD#1 complications, and readmissions.

Discussion

Historically, PCNL has required hospital admission. However, in recent years, several studies have evaluated and confirmed the feasibility of outpatient PCNL in select patients. ^{8 –13} Still, the majority of these studies utilize very specific inclusion criteria, which limits the studied population to a highly selected set of patients who often do not reflect a typical PCNL patient. For example, common exclusion criteria included BMI > 30 kg/m², ASA score > 2, preoperative ureteral stent or nephrostomy tube, anatomical abnormalities of the urinary tract, and certain comorbidities such as active cardiac disease or poor renal function. ¹⁴ Of note, these studies also often excluded patients with multiple stones or staghorn calculi. ¹⁵

Fortunately, several recent studies have begun to examine the feasibility of ambulatory PCNL in more complex patients. For instance, Hosier et al. studied the safety of utilizing extended selection criteria for ambulatory PCNL, which included patients with characteristics such as BMI > 30 kg/m², age > 75 years, ASA > 2, solitary or transplant kidney, bilateral stones, staghorn calculi, and multiple tracts. ¹⁶ When comparing the 92 extended criteria patients with 26 patients meeting “standard criteria,” the authors found no significant difference in terms of complication rates, ED visits, or readmissions. While this study did provide initial evidence supporting the expansion of ambulatory PCNL inclusion criteria, the majority of the patients in the extended criteria group only had a single stone, and only 6% had complete staghorn calculi. In a different study, Zhu et al. evaluated ambulatory discharge among patients undergoing multiple-tract PCNL. ¹⁷ Among the 46 patients, all of whom had multiple stones or a staghorn, the authors demonstrated a complication rate of 17.4% and an SFR of 84.8%. Limitations of this study include a small sample size, the absence of a comparison group, no assessment of stone complexity via metrics such as Guy’s or STONE score, and the definition of ambulatory as discharge within 24 hours—rather than same-day discharge as was used in our study. Although these studies provide further guidance on how to approach ambulatory PCNL in more difficult patients, their limitations still warrant a formal analysis of the safety of same-day discharge PCNL for clearly defined complex renal stones when compared with admitted patients.

Our work builds on these prior analyses by providing the first study to explicitly assess the outcomes of same-day discharge after PCNL for complex renal stones. Our group’s protocol change amid the COVID-19 pandemic was largely successful, resulting in all but one of the intended ambulatory patients being safely discharged on the same day. When comparing ambulatory and admitted patients, there was no significant difference in rates of complications. Notably, our 30-day complication rate of 20.3% in outpatients was largely in line with the complication rates identified in prior studies on ambulatory PCNL, which ranged from 0% to 20%, along with the overall PCNL complication rate of 20.5% identified by the Clinical Research Office of the Endourological Society Percutaneous Nephrolithotomy Global Study. ^{8 –10,12,13,16,18 –24} Moreover, only one ambulatory patient experienced a complication on POD#1 (anemia), and it did not require a subsequent ED visit or readmission. An important consideration when opting to implement ambulatory PCNL is the potential to delay detection or miss high-grade complications on POD#1 that would have otherwise been identified via an overnight admission. Given that we identified only one low-grade (CD II) complication, our work demonstrates that such serious POD#1 complications may be rare for ambulatory PCNL for complex stones. Furthermore, it is worth noting that the most serious complications from PCNL tend to be either delayed in presentation—such as urinoma, fistula, or pseudoaneurysm—and their course would thus not be significantly altered by an overnight admission, or present immediately—such as sepsis or pneumothorax. ¹¹ These immediate postoperative complications may require unintended admission, as they did for one of our intended ambulatory patients, but should not preclude the adoption of a predominantly outpatient PCNL program assuming postoperative screening for such complications is implemented.

While our study did identify significantly higher rates of ED visits without admission among same-day discharge patients, they also experienced a lower rate of readmissions compared with overnight PCNL patients. This discrepancy may be because of ambulatory patients presenting to the ED with generally lower-grade complications. Furthermore, for ambulatory patients, the any-ED visit rate (including those subsequently readmitted) of 13.9% was comparable with rates identified in prior studies of ambulatory PCNL, ranging from 0% to 18.6%. ^{8,9,12,16,18 –21,23} Similarly, our readmission rate of 5.1% also fell within the range (0–11.6%) identified in prior ambulatory PCNL studies. ^{8 –10,13,16,18 –23} Lastly, compared with admitted patients, ambulatory patients experienced a comparable CT-determined SFR that is similar to rates identified in other studies on ambulatory PCNL. ^8,12,16 Altogether, our study suggests that same-day discharge PCNL may be safe and efficacious even in patients with complex renal stones.

The ability to implement same-day discharge after PCNL for patients with complex stones carries numerous benefits. When given the choice, patients typically favor ambulatory surgery. This preference is driven by several factors including shorter waiting times, improved continuity of care, home recovery, and generally fewer days to return to normal activity. ^22,25 Furthermore, the ability to recover at home may confer improved rest given that hospitalized patients frequently experience poor-quality sleep. ²⁶ This may have a subtle but important impact on the recovery process, given that decreased sleep quality may increase the risk of onset of pain and exacerbate existing pain in the perioperative period. ²⁷ Outside of patient preference, ambulatory surgery lowers the risk of nosocomial infections, which can otherwise lead to prolonged hospital stays, additional complications, and increased health care costs for patients. ^28,29 Lastly, another significant potential benefit of ambulatory PCNL is reduced cost. In fact, a study in Indiana found that ambulatory PCNL confers an absolute cost savings of approximately $5327 ± 442 per case when compared with inpatient PCNL. ²³

After the success of this ambulatory PCNL experience, our group has continued to pursue same-day discharge in nearly all patients. Currently, all surgeries are performed in the ambulatory wing of a hospital, and all patients are preoperatively intended for same-day discharge unless they are currently admitted patients. The main contraindications to same-day discharge in our practice are postoperative complications such as significant bleeding, clinical signs of systemic inflammatory response syndrome, or other complications that necessitate further medical and/or surgical management. In general, there are important factors urologists must consider before choosing to adopt a same-day discharge protocol. The cornerstone of a successful outpatient PCNL program relies on minimizing the incidence of high-grade complications that cannot be managed remotely. While our study, along with others, has begun to demonstrate the feasibility of ambulatory PCNL in more complex patients, surgeon discretion must be used to carefully assess the suitability of certain higher-risk patients for outpatient PCNL. Furthermore, for patients undergoing ambulatory PCNL, it may be beneficial to schedule an outpatient appointment shortly after surgery or to implement a postoperative follow-up phone call program in order to proactively address complications that may arise and to reduce ED visits. In addition, social reasons represent salient factors toward the success of ambulatory PCNL. Situations such as patients not having transportation home or nobody to stay with the patient overnight can often lead to unplanned conversion to inpatient PCNL. ¹⁶ Accordingly, preoperative planning should include clear communication between patients, family members, and members of the urology team regarding the intended discharge plan.

Our study has several limitations. First, in addition to limitations inherent to retrospective analyses, our study was not randomized, and instead our two patient groups were separated temporally by a practice change because of COVID-19. This carries a potential cohort and/or historical bias because of differences in both the patient population and health care environment at different stages during the pandemic. Of note, patients in the admitted group tended to have more complex stones and were preoperatively more likely to have a positive urine culture along with a stent or nephrostomy tube. While we attempted to control for these confounding factors in multivariate analysis, these significant differences in preoperative characteristics represent possible selection biases in the study. Second, POD#1 complications may be undercounted in the ambulatory group as they may have experienced minor, unreported complications that resolved on their own compared with admitted patients who may have been more likely to have minor complications reported because of close hospital observation. Third, our study did not evaluate pain or other patient-reported outcomes, which may represent important considerations when deciding whether to pursue ambulatory PCNL for complex stones. Lastly, all patients in our study were treated at a high-volume PCNL center by endourology fellowship-trained surgeons, meaning that our findings may not be generalizable to all practice settings.

Conclusions

Same-day discharge after PCNL for complex renal stones did not confer an increased risk for complications, readmissions, or poorer stone-free outcomes when compared with patients admitted overnight. These data indicate a similar safety profile between outpatient and inpatient PCNL, suggesting that exclusion criteria, which preclude patients with complex stones from ambulatory PCNL, may not be strictly necessary. Future prospective trials should further evaluate the safety and efficacy of same-day discharge after PCNL in patients with complex renal stones.