Outcomes of Endourologic Interventions in Patients with Preoperative Funguria

Introduction

Fungal colonization or infection of the urinary tract is increasingly common, especially among patients with significant comorbidity. Rates of funguria are widely disparate depending on the population studied ranging from <2% in healthy individuals to up to 40% of nosocomial Urinary Tract Infections (UTIs) in an intensive care unit (ICU) setting. ¹ Up to 95% of fungal urine cultures will be Candida species. ²

Despite the widespread prevalence of funguria, data regarding outcomes of patients with preoperative funguria undergoing endourologic intervention are extremely limited. Prior literature is limited to individual case reports that have demonstrated that postureteroscopy sepsis or fungemia can occur in patients with preoperative funguria. ^3,4 Treatment of asymptomatic funguria has not been associated with improved patient outcomes; however, it is unclear to what extent these data apply before urinary tract instrumentation. ⁵ In this study, we aim to assess the characteristics and outcomes of patients with preoperative funguria undergoing ureteroscopy or percutaneous nephrolithotomy (PCNL).

Methods

The study protocol was approved by the institutional review board at the University of Pittsburgh. Patients with preoperative funguria were retrospectively identified by cross-referencing lists of patients with funguria on urine cultures with operative reports of patients undergoing endourologic intervention from the years 2009 through 2016. Only adult patients undergoing ureteroscopy or PCNL were included. All patients underwent intervention by a single fellowship-trained endourologist at a tertiary referral center. Preoperative funguria was defined as a urine culture containing >10,000 colony forming units of fungus within 30 days of the operative intervention.

Collected preoperative data included patient demographics, comorbidities, including Charlson comorbidity index (CCI), presence of a neurogenic bladder diagnosis, urinary diversion, or chronic catheter usage, antibiotic exposure, and preoperative antibiotic and antifungal use. Antifungal use was recorded as either a single perioperative dose, or a full course of treatment, which ranged from 3 to 14 days of sensitivity-directed therapy. Operative data consisted of procedure performed, stone size, and operative duration. Assessed perioperative outcomes included length of stay, need for ICU admission, 30 day readmission and mortality rates, and postoperative systemic inflammatory response syndrome (SIRS), bacteremia, or fungemia rates. During ureteroscopy, normal saline compressed in a pressure bag was used as irrigation solution. Ureteral access sheaths were routinely used for treatment of renal stones. All PCNLs were performed using a standard 30F sheath with the patient in prone position. Univariable and multivariable regression was used to identify predictors of postoperative SIRS in patients with preoperative funguria. Postoperative SIRS was defined by meeting two or more of the following criteria: fever >38.0°C, heart rate >90 beats per minute, respiratory rate >20 breaths per minute, or white blood cell count >12,000/cc. A two-tailed p < 0.05 was used as the threshold for statistical significance.

Results

A total of 65 patients with preoperative funguria were identified, of whom 49 (75.4%) underwent ureteroscopy and 16 (24.6%) underwent PCNL (Table 1). Of them 66.2% of patients were women and average patient age was 55.1 ± 18.3 years. Average patient body mass index and CCI were 31.8 ± 11.0, and 2.52 ± 2.0, respectively. In total 35.4% of patients carried a known diagnosis of neurogenic bladder, 24.6% had an indwelling urethral or suprapubic catheter, 3.1% performed Clean intermittent catheterization (CIC), and 3.1% had a urinary diversion. Spinal cord injury and multiple sclerosis were the most common neurologic diagnosis present.

All cases of funguria were caused by Candida species. In total 49.2% of preoperative urine cultures grew Candida albicans, 26.2% grew Candida glabrata, and 24.6% grew Candida, species not specified. A total of 81.5% had been diagnosed with a UTI in the year before intervention and 87.7% of patients had been exposed to antibiotics in the 3 months before intervention. In total 64.6% of patients received antifungal treatment perioperatively, 12.3% received a single perioperative dose, whereas 52.3% received a full course of treatment.

Average patient stone burden, defined as the sum of the longest linear dimension of the identified stones was 3.4 ± 3.1 cm. In total 70.7% of patients had multiple stones, and 86.2% of patients had an indwelling stent or nephrostomy tube before the index procedure, with a median dwell time of 39 days (interquartile range 14–75 days). Eighty-seven percent of patients had stone analysis available for review, of which 50% were predominantly calcium oxalate, whereas 43% were predominantly carbonate apatite and 5% and 2% were uric acid and struvite, respectively (Table 2). Of 29 stone cultures performed, 25 (86.2%) grew Candida species, whereas 4 demonstrated no growth.

Eighteen patients (27.7%) met SIRS criteria postoperatively, and 16.9% required ICU admission. Three (4.6%) and two (3.1%) patients developed postoperative fungemia and bacteremia, respectively. All cases of fungemia were caused by C. glabrata. Of the 18 patients who developed postoperative SIRS, 10 had received a full course of antifungal therapy, 2 received a single perioperative dose of antifungal, and 6 received no antifungal therapy. In total 30.8% of patients were readmitted, half of whom were readmitted in the setting of postoperative SIRS, and no patients died within 30 days of intervention.

On univariable analysis (Table 3a), presence of an indwelling catheter (p = 0.009), presence of a known neurological diagnosis (p = 0.02), presence of C. glabrata on preoperative culture (p = 0.04), and longer operative time (p = 0.04) were predictive of development of postoperative SIRS. No significant predictors were identified on multivariable analysis. Notably, antifungal treatment or lack thereof was not associated with perioperative SIRS.

Discussion

This study demonstrates that patients with preoperative funguria are at high risk for postoperative infectious complications and readmission, with 27.7% of patients meeting SIRS criteria postoperatively and 30.8% of patients requiring readmission within 30 days. In contrast, a published series of 154 consecutive ureteroscopies performed by the same endourologist demonstrated a postoperative infection and readmission rate of 1.9%. ⁶ Preoperative treatment of funguria was not associated with a reduced rate of infectious complications. Patients with preoperative funguria were found to have high rates of obesity, catheter utilization, and frequently carry a diagnosis of neurogenic bladder. Prolonged operative time, presence of C. glabrata on preoperative culture, and presence of an indwelling catheter were predictive of postoperative SIRS.

The rate of postoperative infectious complications in this patient cohort was alarmingly high. Prior studies have identified that the presence of a preoperative ureteral stent, increased stent dwell time, and increased operative time, are independent risk factors for postureteroscopy sepsis. ^7,8 The patient cohort in this study had a very large average stone burden with subsequently long operative times, which may have contributed to the high infection rate; however, stent or nephrostomy dwell time was not a predictive factor for infection in this cohort. Further study is needed to characterize whether funguria is an independent risk factor for postureteroscopy infection or simply a marker for high-risk patients. Based on our results, urologists should be aware of the high risk for infectious complications in this patient cohort, limit surgical intervention whenever possible, and in said cases should counsel patients appropriately. In addition, this patient population could be targeted for specific interventions to minimize risk. Potential interventions that could be considered include the use of routine prestenting to allow for passage of larger ureteral access sheaths, lower irrigation pressures, limiting operative duration, using prolonged courses of preoperative antibiotic and antifungal therapy, documenting culture clearance before proceeding with intervention, and postoperative hospital admission for intensive monitoring. Consideration could even be given to preoperative placement of a nephrostomy tube before ureteroscopy to allow for renal irrigation while minimizing intrarenal pressure. These interventions lack sufficient evidentiary basis and would require further study; however, given the high infectious risk this patient cohort represents, they are reasonable options to consider empirically.

The finding that a significant portion of stones in this patient cohort were composed predominantly of carbonate apatite is significant in that it provides a potential target for metabolic therapy to limit the need for additional intervention in this high-risk patient population. Although calcium phosphate stones represent 12% to 20% of cases of nephrolithiasis in the general population, carbonate apatite was found in 43% of stone analysis within this cohort. ⁹ This mirrors the findings that patients with neurologically derived musculoskeletal deficiencies were found to have carbonate apatite stones in 53.8% of cases. ¹⁰ Carbonate apatite stones have been hypothesized to form more readily in the setting of bacterial colonization or infection. ¹¹ Whether this finding in this cohort of patients with preoperative funguria is caused by the funguria itself or high rates of associated bacterial colonization is unclear. Notably, carbonate apatite stones can effectively be prevented through urine acidification and solute dilution. ⁹ This finding underscores the need to metabolically evaluate patients with funguria and nephrolithiasis to identify potentially correctable abnormalities and thus avoid the significant risk of endourologic intervention.

Another significant finding in this study is that the concordance rate between preoperative funguria and fungal growth on stone culture was 86.2%. Prior studies have identified significant discordance between preoperative urine culture and stone culture results, with the positive predictive value of a preoperative urine culture for predicting stone culture results as low as 35%. ¹² Positive stone culture is a known risk factor for postoperative SIRS and sepsis. ¹³ Although limited data exist regarding whether antibiotic selection based on stone culture results affects patient outcomes, together these prior data points suggest that targeting antibiotic therapy of postureteroscopy infections based on preoperative urine cultures is typically insufficient. ¹⁴ However, based on the high concordance between fungal preoperative and stone cultures observed in this study, in patients with preoperative funguria who develop postoperative infectious complications, early empiric initiation of antifungal therapy in addition to antibiotics is necessary and appropriate.

The results of this study should be interpreted in light of its potential limitations. One limitation is that this study was retrospective and nonblinded in nature, introducing the potential for patient selection bias. This was minimized by including all patients who met prespecified criteria for preoperative funguria. This was a single-arm study, which limits the ability to identify predictive factors for preoperative fungemia or directly compare outcomes to nonfunguric patients. In addition, all procedures were performed by a single endourologist at a tertiary referral institution in a patient population with a larger aggregate stone burden than one would typically encounter in the community. This reduces bias from variation in surgical technique and proficiency but could reduce generalizability of results. Stone culture data were limited as it was not the practice of the operating surgeon to routinely send stone culture during ureteroscopy. Despite the inherent limitations present, this study is significant as it represents the first comprehensive examination of outcomes of endourologic procedures performed in patients with funguria.

Conclusion

Patients with preoperative funguria are at high risk for infectious complications after endourologic intervention.