Do Multiple Ureteroscopies Alter Long-Term Renal Function? A Study Using Estimated Glomerular Filtration Rate

Introduction

Over the last 15 years, the use of ureteroscopy (URS) and the holmium laser for fragmentation of upper tract calculi has greatly expanded. ¹ Under direct visualization, calculi can be fragmented using targeted laser energy while minimizing direct tissue damage. URS is considered a safe and efficacious procedure with a complication rate of 3.5%, the most common complication being postoperative fever. ² Only one study, however, has evaluated long-term renal function in patients undergoing URS for stone management. ³ Potential mechanisms of injury include direct damage from inadvertent contact between the laser and tissue, and pyelovenous backflow resulting from forceful irrigation.

In contrast to URS, multiple studies have evaluated the long-term effects of shockwave lithotripsy (SWL) and percutaneous nephrolithotomy (PCNL) on renal function. The mechanisms of injury of concentrated shockwaves and renal parenchymal invasion intuitively pose a risk to renal tissue. Nonetheless, the majority of literature available suggests neither SWL or PCNL has deleterious long-term effects on renal function. ^{4 –8}

The existence of a functional contralateral kidney confounds assessment of overall renal function, because this renal unit is presumably unaffected during URS and may compensate for any potential renal injury secondary to URS. We therefore assessed patients with baseline mild to moderate renal insufficiency undergoing multiple URS procedures to enhance the sensitivity of our study to detect a change in estimated glomerular filtration rate (eGFR). To our knowledge, this is the first study assessing patients undergoing multiple URS and assessing long-term change in renal function.

Methods

With Institutional Review Board approval, chart review was performed for patients with a baseline eGFR below 90 mL/min/1.73 m² who underwent at least two URS with laser lithotripsy for urolithiasis at our institution from 2004 to 2012. Hand controlled bulb irrigation was used in all cases. Patients undergoing SWL or PCNL at any point in their history and those without follow-up creatinine assays at least 5 months postoperatively were excluded. From nearly 2000 URS procedures performed over the 9-year period, 26 patients (mean of 2.3±0.6 URS procedures each) met all criteria and were included in the analysis. The eGFR at baseline (defined as 3–12 months before acute stone presentation) and at the last recorded nonemergent follow-up visit were calculated by the four variable Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation based on creatinine, age, sex, and race. ⁹

Follow-up time was defined as elapsed time from the patient's most recent URS to last recorded creatinine assay. Age, sex, body mass index (BMI), comorbidities including hypertension and diabetes mellitus, stone location, total stone burden, number of URS, and elapsed time between first and last URS were recorded.

The percent change in eGFR was calculated as follows, with a positive number representing an increase in eGFR, and a negative number indicating a decrease in eGFR: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6} \begin{document} \begin{align*}\hbox{[(follow-up eGFR)} - \hbox{(baseline eGFR)] / (baseline eGFR)}\end{align*} \end{document}

Annualized change in eGFR was calculated as the percent change divided by years of follow-up. Continuous variables were compared across groups using the Mann-Whitney U test.

Results

The study population included 14 females and 12 males, with a mean age of 54.3±14.5 years, and BMI of 33.9±9.4 kg/m². Ten patients had DM, 15 had HTN, and 8 had both. Mean stone burden for the sum of each patient's procedures was 25.4±11.8 mm. Mean time between first and last URS was 19.3±26.1 months (range 0–108), with 13/26 (50%) patients undergoing all URS within 12 months. Characteristics and procedural data are summarized in Table 1.

The mean eGFR changed from 68.0±13.3 to 75.4±23.0 mL/min/1.73 m² (mean increase of 10.1±25.0%; mean annual increase of 3.8±15.3%) over a mean follow-up period of 28.1 months (range 5–75 months). There was no significant difference in eGFR change between patients with stones treated in the kidney alone vs the ureter and kidney combined (12.1% vs 8.3% mean increase; P=0.74). Age, baseline creatinine, total stone burden, number of URS, presence of DM and HTN, elapsed time between first and last URS, and single sided vs bilateral treatment were not significantly associated with change in eGFR (Table 2). No patients experienced major complications or postoperative obstruction.

Two patients had renal abnormalities. One with a horseshoe kidney experienced a 9.8% decrease in eGFR measured at 5 months follow-up (59.2–53.4 mL/min/1.73 m²) after a two-stage URS. The second had a solitary kidney with extensive history of fistulizing Crohn's disease and experienced a 35.7% decrease in eGFR at 81 months follow-up (40.3–25.9 mL/min/1.73 m²).

Discussion

Alhough the potential causes of renal injury during URS are not as evident as in SWL and PCNL, two proposed mechanisms exist: Direct tissue injury as a result of contact with the holmium laser and pyelovenous backflow from saline irrigation. The combination of advanced optics allowing precise application of the laser, presence of saline irrigant dissipating thermal energy, and direct contact injury zones of less than 1 mm make the laser an unlikely cause of renal damage. ¹⁰ A more likely scenario, however, is pyelovenous backflow from manual saline irrigation used to maintain a clear field of vision endoscopically.

Routine flexible URS can produce renal pelvic pressures above 100 mm Hg, with vigorous hand irrigation generating pressures up to 410 mm Hg. ¹¹ Although no human studies investigating the effect of elevated renal pelvic pressures on renal injury exist, animal studies have shown high pressure (>150 mm Hg) irrigation can cause irreversible damage to both the urothelium and parenchyma. At 4 to 6 weeks, kidneys subjected to high-pressure irrigation showed significant renal tubule vacuolization and degeneration, as well as focal scarring within the parenchyma. ¹² The repetitive renal insults from pyelovenous backflow over the course of multiple URS potentially translates to a decrease in overall renal function.

Until now, only Lee and Bagley ³ have investigated the effects of URS on long-term renal function. These authors studied a cohort of 18 patients with preexisting mild to moderate renal insufficiency undergoing a single URS using pre- and postoperative eGFR (calculated by the Cockcroft and Gault formula) and reported a 5.9% improvement in eGFR over a mean follow-up of 18 months. In their study, renal insufficiency was defined as baseline creatinine level of ≥1.5 mg/dL. Baseline creatinine values were determined preoperatively, potentially overestimating the degree of chronic renal insufficiency secondary to obstruction and/or dehydration during an acute stone event. Also, of the 18 patients in the study, 6 had undergone previous SWL and 3 had prievious PCNL.

To control for these variables and maximize the accuracy of the eGFR, we used stable baseline creatinine values measured 3 to 12 months before a patient's acute stone event. This, along with the use of the CKD-EPI formula that has been shown to perform better across a mild to moderate range of eGFR values, ⁹ served to identify patients with genuine renal insufficiency. Only including patients undergoing multiple URS and excluding those with a history of SWL or PCNL further strengthened our study in attempting to identify the true impact of URS on long-term renal function.

Multiple limitations exist, however, primarily the retrospective design and small patient cohort (N=26). As discussed previously, the presence of an unaffected, “normal” contralateral kidney also confounds the measurement of long-term renal function. This study spanned the transition from paper to electronic medical records, and operative time (a potential reflection of exposure time to pressure irrigation) was unavailable for the majority of patients. Finally, drawbacks still exist in using creatinine level as a sole measure of renal function, because body composition, diet, medications, endogenous substances, and comorbidities can all alter the creatinine value and calculated eGFR. For example, our only study patient with a solitary kidney experienced a 35.7% decrease in eGFR at 81-month follow-up, likely secondary to severe fistulizing Crohn disease with multiple bouts of sepsis, recurrent exploratory laparotomies, bowel resection, and extended hospitalizations. In patients like this with ongoing systemic insults, determining the true impact of URS on long-term renal function is extremely challenging.

The development of novel urinary markers of renal injury may provide further insight into the effects of URS on renal function by bypassing the use of creatinine altogether. A recent study ¹³ measured pre- and postoperative secretion of kidney injury molecule 1 (KIM-1), and N-acetyl-β-D-glucosaminidase (NAG) in kidney stone patients undergoing either SWL (n=50) or URS (n=10). They reported a significant increase in both biomarkers post-SWL (P<0.001 for each), but no change in KIM-1 or NAG post-URS (P=0.893, P=0.674, respectively). ¹³ Increases in both biomarkers have been shown to correlate with decreased eGFR, ¹⁴ supporting our findings that URS is unlikely to negatively affect renal function.

We measured a 10.1% mean increase in eGFR over an average of follow-up of 28.1 months, with no statistically significant difference noted for age, baseline creatinine level, total stone burden, number of URS, presence of DM and HTN, or elapsed time between URS procedures. Nevertheless, all surgical procedures carry inherent risk, and careful use of the holmium laser with minimum necessary irrigation should be employed.

Conclusions

Renal function is a complex entity with a multitude of acute and chronic effectors. Our study comparing baseline and long-term eGFR suggests URS did not impact long-term renal function, specifically in patients with preexisting renal insufficiency undergoing multiple URS procedures.