Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/wolters-kluwer-health/predicting-urine-culture-results-in-candidates-for-lithotripsy-Vd0VWvLP7P
- Publisher
- Wolters Kluwer Health
- Copyright
- Copyright © 2022 The Authors. Published by Wolters Kluwer Health, Inc.
- ISSN
- 1661-7657
- eISSN
- 1661-7657
- DOI
- 10.1097/cu9.0000000000000117
- Publisher site
- See Article on Publisher Site
Abstract
Background: Urological guidelines assert that “urine culture should be obtained” before surgical management of ureteral or kidney stones. Thus, many surgeries are delayed by 1–3 days until the results of urine culture are available. During this time, the patient fre- quently experience pain and possible kidney damage. We investigated the hypothesis that it is possible to predict the results of urine culture in candidates for surgical intervention using parameters that are accessible immediately upon admission. Materials and methods: A database of 1000 patients who underwent either percutaneous nephrolithotomy (PCNL) or ureteroscopy/ retrograde intrarenal surgery was analyzed. Eleven parameters potentially related to urinary infections and accessible to the clinician at the emergency department were correlated with the preoperative urine culture results. Results: Of the patients, 234 (23.4 %) had positive cultures. On multivariate analysis, only sex, hydronephrosis grade, and history of previous nephrolithotomy were significantly associated with a positive preoperative urine culture. The risk of a positive culture can be easily determined from a simple table or an Excel-based calculator. This risk could be as low as 0.45% for a man without a history of PCNL and no hydronephrosis (4% in a woman with similar parameters) or as high as 79.5% in a man with a history of PCNL and hydronephrosis (85% in a woman with similar parameters). Conclusions: The risk of preoperative positive urine culture can be predicted using 3 parameters that are accessible upon admission. In low-risk cases, prompt surgical treatment can be provided, eliminating the anticipation time for urine culture results. Keywords: Ureteral calculi; Renal calculi; Urine culture; Ureteroscopy; Percutaneous nephrolithotomy 1. Introduction Several attempts were made to identify preoperative risk factors [9,10] of positive urine culture. A recent meta-analysis showed that sex, The American Urological Association/Endourological Society Guide- preoperative ureteric stent insertion, and diabetes mellitus were associ- [11] line states that before surgical management of ureteral or kidney ated with positive urine cultures in patients before URS. In another stones, urine culture should be obtained from patients with clinical study, a stone diameter greater than 2 cm was a risk factor in patients [1] [5] or laboratory signs of infection. The European Association of undergoing RIRS. The degree of hydronephrosis was reported to be [2] Urology guidelines are stricter: “A urine culture or urinary micros- associated with post-PCNL infection. Diameter of the largest stone [2] copy is mandatory before any treatment.” Because of these recom- and number of affected calyces were 2 preoperative predictors of post- [12] mendations, many patients are forced to postpone surgery until the operative complications in another study. The value of preopera- results of urine culture are available while experiencing pain and tive urinary microscopy and white blood cell count is controversial. [13] possibly from complications of upper urinary tract obstruction. Bacteriuria might not be a significant predictor of infection. The im- [14] Although emergency ureteroscopy (URS) is feasible in selected portance of leukocytosis and pyuria is debatable; Yilmaz et al. sug- cases, even in patients with urinary tract infection and symptoms gested that these are significant risk factors for positive urine culture, [3] [15] of mild sepsis, the guidelines are certainly justified. The risk of in- whereas Rohloff et al. showed that they only varied slightly in pos- fectious complications develops in 10.8% (range, 0%–32.1%) of itive and negative culture groups. Urinary nitrite and urinary leuko- [15] patients after percutaneous nephrolithotomy (PCNL), in 7.14% af- cyte esterase are potential indicators of a positive urinary culture. ter retrograde intrarenal surgery (RIRS), and in 2% (0%–7%) after In this study, using a large prospectively maintained database, [4–6] URS. Preoperative positive urine culture is a major risk factor we investigated the hypothesis that the parameters available upon for postoperative infection and should be treated according to urine admission to the emergency department can predict preoperative [7] [8] culture results, and upper tract drainage should be considered. urine culture results. We attempted to combine these parameters with a clinically useful tool. If such a tool can be developed, the waiting time for urine culture results can be eliminated in low-risk *Corresponding Author: Ofer N. Gofrit, Department of Urology, Hadassah University Hospital, patients and prompt surgical treatment can be provided, thus reduc- P.O. Box 12000, Jerusalem, 91120, Israel. E-mail address: ogofrit@gmail.com (O.N. Gofrit). ing patient suffering and possible kidney damage. In addition, there Current Urology, (2023) 17, 2, 113–117 is a potential economic benefit in reducing hospitalization time and Received March 14, 2021; Accepted April 6, 2021. easing the load on the urologic and microbiology wards. O.N.G. and R.A. shared first authorship. http://dx.doi.org/10.1097/CU9.0000000000000117 Copyright © 2022 The Authors. Published by Wolters Kluwer Health, Inc. This is 2. Materials and methods an open-access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it 2.1. Patient cohort is permissible to download and share the work provided it is properly cited. The The data were extracted from a prospectively maintained database work cannot be changed in any way or used commercially without permission from the journal. containing more than 3000 patients who were operated on 113 Gofrit et al. Volume 17 Issue 2 2023 www.currurol.org between September 2006 and April 2019. Data regarding the last Table 1 500 PCNL cases and the last 500 URS and RIRS cases were ana- Baseline characteristics of the patients included in the study.* lyzed. All patients had preoperative urine cultures that were avail- able for review. Patients with a high temperature (37.5°C) were not Characteristics Total included in the analysis. A retrospective analysis of the risk factors Age, mean (SD), yr 52.25 (15.17) of positive preoperative urine culture, defined as a culture with Sex, n (%) greater than 100,000 colony-forming units, was performed. This Male 685 (69) study was approved by the local institutional review board (ap- Female 311 (31) proval #0080-19-HMO). An informed consent was waved by the BMI, n (%), kg/m committee. Excluding body mass index (BMI), a complete data Underweight 16 (3) set was available for 970 patients, and the BMI was only 623. Normal 176 (28) The effects of the following variables on the risk of preoperative Overweight 244 (39) positive urine cultures were analyzed: Obese 187 (30) No. stones, n (%) 1. Age 1 558 (56) 2. Sex ≥2 380 (38) 3. BMI (underweight [<18.5], normal [18.5–25], overweight Staghorn 55 (6) [25.1–30], and obese [>30.1]) Hydronephrosis grade, n (%) 4. Number of stones to be treated (1 stone, ≥2 stones, staghorn 0 385 (39) stone) 1 240 (24) 5. Hydronephrosis grade (4 grades) 2 118 (12) 6. Largest stone mean density (in Hounsfield units) 3 166 (17) 7. Maximal diameter of the largest stone 480(8) 8. Stone location (calyceal stone [± ureter stone], pelvic stone Largest stone density, mean (SD), HU 1025.51 (358.79) [± calyceal or ureteral stones], or ureteral stone only) Maximal diameter of the largest stone, mean (SD), mm 18.62 (13.02) 9. Prior stent Prior stent, n (%) 10. Prior nephrostomy No 933 (93) 11. Prior surgery (no prior surgery, prior PCNL, prior URS—6pa- Yes 65 (7) tients had both PCNL and URS; therefore, these patients were Prior nephrostomy, n (%) excluded from the analysis) No 904 (91) 12. History of diabetes mellitus as a categorical variable Yes 90 (9) Prior surgery, n (%) None 769 (77) Because the intention of the study was to provide a useful tool for PCNL 208 (21) the admitting doctor at the emergency department level, only readily URS 15 (1.5) available parameters (from the patient’s history and noncontrast PCNL and URS 6 (0.5) computed tomography) were analyzed. Therefore, the maximum di- Stone location ameter of the largest stone was included in the analysis and not more Calyx (± ureter) 243 complex parameters, such as total stone surface area or volume. Pelvis (± calyceal ± ureter) 512 However, these parameters were available for revision and showed Ureter 215 a good correlation with the maximal diameter of the largest stone Diabetes mellitus, n (%) (total stone surface area: R = 0.78, p < 0.0001; total stone volume: No 814 (82) R = 0.59, p < 0.0001). Yes 175 (18) Urine culture, n (%) Negative 766 (77) 2.2. Statistical analysis Positive 234 (23) Statistical analysis was performed using IBM Statistical Package for the Social Sciences (version 25.0) and Stata/SE version 15.0 *Complete data set was available for only 970 patients. BMI = body mass index; HU = Hounsfield unit; PCNL = percutaneous nephrolithotomy; SD = standard (StataCorp). Descriptive statistics are presented as mean with standard deviation; URS = ureteroscopy. deviation or as frequency with percentage, according to the scale of the variable. The associations between all variables and preoperative positive urine culture were assessed using logistic regression models, culture: sex, prior stent placement (but not prior nephrostomy), in which all variables with p < 0.2 in the bivariate analysis were in- hydronephrosis grade, and prior PCNL (but not prior URS). Age, cluded in the multivariable logistic regression model. Statistical sig- BMI, number of stones, stone density, location, size, and history nificance was set at p < 0.05, and all reported p values were 2-tailed. of diabetes mellitus did not affect the risk of positive cultures. Parameters with a p value less than 0.2 were included in the mul- 3. Results tivariate analysis. These included age, sex, hydronephrosis, history of prior stenting or nephrostomy, and prior PCNL or URS (Table 3). The baseline characteristics of the study population are presented Sex, hydronephrosis grade, and prior PCNL were the only significant in Table 1. A positive preoperative urine culture result was found factors associated with positive preoperative urine culture results. The in 234 patients (23.4%). The most frequently cultured bacteria calculated risks of positive urine cultures using these parameters are were Escherichia coli (86 cases), Enterococcus faecalis (45 cases), presented in Table 4. The table shows that the risk of positive urine Klebsiella pneumoniae (29 cases), Proteus mirabilis (24 cases), culture can be as low as 0.45% for a man without history of PCNL and Pseudomonas aeruginosa (22 cases). and no hydronephrosis and 4% for a woman with the same parameters. Bivariate analysis (Table 2) showed that the following character- In contrast, when there is a history of PCNL and grade 1 hydronephrosis, istics were significantly associated with preoperative positive urine the risk of positive urine culture is as high as 79.5% in men and 114 Gofrit et al. Volume 17 Issue 2 2023 www.currurol.org Table 2 Table 4 Bivariate analysis of baseline characteristics potentially associated with positive Risk of positive urine culture results.* preoperative urine culture results. Sex Hydronephrosis Prior PCNL Positive urine culture (%) 95% CI for OR Male No N 1/222 (0.45) Characteristics p Adjusted OR Lower Upper Grade 1 N 5/131 (3.8) Grade 2 N 9/61 (14.8) Age 0.09 1.01 0.99 1.09 Grade 3 N 13/96 (13.5) Sex, female <0.001 3.10 2.29 4.21 Grade 4 N 13/45 (28.9) BMI* 0.21 Male No Y 31/53 (58.4) Underweight 0.30 0.56 0.19 1.67 Grade 1 Y 35/44 (79.5) Overweight 0.09 0.39 0.13 1.16 Grade 2 Y 2/10 (20) Obese 0.15 0.45 0.15 1.33 Grade 3 Y 2/8 (25) Stone location 0.86 Grade 4 Y 3/7 (42.8) Pelvis (± calyceal) 0.72 1.07 0.74 1.54 Female No N 2/49 (4.1) Ureter 0.58 1.12 0.74 1.71 Grade 1 N 7/45 (15.6) No. stones 0.57 Grade 2 N 11/40 (27.5) ≥2 0.81 1.04 0.76 1.42 Grade 3 N 21/57 (36.8) Staghorn 0.29 1.40 0.75 2.63 Grade 4 N 12/26 (46.2) Hydronephrosis grade 0.04 Female No Y 43/58 (74.1) 1 0.04 1.43 0.98 2.09 Grade 1 Y 17/20 (85) 2 0.06 0.95 0.57 1.60 Grade 2 Y 1/6 (16.7) 3 0.31 1.25 0.81 1.93 Grade 3 Y 4/5 (80) 4 0.005 2.12 1.26 3.57 Grade 4 Y 0/2 (0) Largest stone mean density 0.57 1 0.99 1.00 Maximal diameter of the largest stone 0.46 1.004 0.99 1.02 *Based on data available from 985 patients. Prior stent 0.008 2.65 1.28 5.47 PCNL = percutaneous nephrolithotomy. Prior nephrostomy 0.13 1.44 0.90 2.32 Prior surgery <0.0001 Prior PCNL <0.0001 15.20 10.57 21.85 4. Discussion Prior URS 0.09 2.74 0.86 8.79 Diabetes mellitus 0.34 0.82 0.55 1.23 The cultivation of microorganisms on culture media was credited *Compared with normal BMI. to Louis Pasteur (1822–1895). It was Koch (1843–1910) that pop- Compared with calyceal stone. ularized the word “colony” to describe the discrete growth of bac- Compared with a single stone. teria on solid media. Julius Richard Petri (1852–1921) manufac- Compared with “no hydronephrosis.” tured a shallow circular dish with a loose-fitting cover, and Arthur BMI = body mass index; CI = confidence interval; OR = odds ratio; PCNL = percutaneous nephrolithotomy; Parker Hitchens (1877–1947) replaced gelatin with agar in solid URS = ureteroscopy. [16] media. Not much has been changed since then. We hope that culture-independent diagnostic techniques will become available [17] in the future. They might be used for immunoassays and nucleic 85% in women. A calculator of the risk of a positive culture that acid amplification testing. When available, they are expected to combines all parameters with a p value less than 0.2 is provided provide fast and reliable microbial diagnosis. Currently, however, in the supplementary material. we are still dependent on bacterial growth in Petri dishes, a time-consuming process that takes 1 to 3 days, which is long, from the point of view of a patient in pain and the surrounding medical Table 3 system that is often ready and eager to provide treatment. Multivariate analysis of baseline characteristics associated with positive urine The clinical guidelines state that before surgical management of culture results. ureteral or kidney stones, urine culture should be obtained from [1] patients with clinical or laboratory signs of infection. However, 95% CI for OR the clinical and laboratory signs of infection have not been defined Characteristics p OR Lower Upper in the guidelines. Using a database of 1000 recently operated cases, Age 0.46 1.01 0.99 1.02 we sought simple and accessible parameters that could predict pos- Sex, female <0.0001 3.10 2.07 4.38 itive urinary culture in candidates for stone surgery. We found that Hydronephrosis grade* 23% of the candidates for stone surgery had positive preoperative 1 0.001 2.34 1.40 3.92 urine cultures. This percentage is higher than that reported by 2 0.02 2.15 1.01 4.20 [15] Rohloffetal. (18.4%). This difference may stem from the differ- 3 <0.0001 4.01 2.21 7.29 ent populations that were studied. In our study, candidates for 4 <0.0001 7.75 3.94 15.26 stone surgery were studied, whereas only patients diagnosed with Prior stent 0.08 1.82 0.93 3.55 [15] ureteral calculi were studied by Rohloff et al. Prior nephrostomy 0.25 1.41 0.79 2.51 We then looked for parameters available to the admitting doctor Prior surgery in the emergency department that could predict the risk of positive Prior PCNL <0.0001 25.2 15.96 39.81 urine culture. Only 3 parameters were significant in the multivari- Prior URS 0.11 2.76 0.79 9.62 ate analysis: sex, hydronephrosis, and history of previous PCNL. *Compared with “no hydronephrosis.” Using these parameters, it was possible to predict situations in BMI = body mass index; CI = confidence interval; OR = odds ratio; PCNL = percutaneous nephrolithotomy; URS = ureteroscopy. which the risk of a positive urine culture was as low as 0.45% or 115 Gofrit et al. Volume 17 Issue 2 2023 www.currurol.org as high as 85% (Table 4). A more accurate calculator that com- with “clinical or laboratory signs of infection.” When the risk of a bines the 6 parameters is provided in the supplementary material. positive culture is low, surgical treatment can be administered imme- This information can be used to eliminate the waiting time for cul- diately, eliminating the anticipation time for urine culture results. ture results in low-risk cases and enable the treating team to pro- ceed immediately to definitive care. Acknowledgments The finding that sex affects the risk of positive urine culture (odds ratio [OR], 3.1) is not surprising. Asymptomatic bacteriuria None. is far more common in women than in men (0.1% vs. 10% in 30- to [18] 65-year-olds and 5% vs. 15% in 65- to 85-year-olds). In addi- Statement of ethics tion, female sex was found to be a risk factor for infectious compli- [11] cations after URS in a meta-analysis by Ma et al., although with This study was approved by the local institutional review board a lower OR (1.82). (approval #0080-19-HMO). An informed consent was waived The finding that hydronephrosis is a risk factor for positive urine by the committee. All procedures performed in this study involving culture is also not surprising (OR of 2.344 for grade 1 hydronephrosis human participants were in accordance with the ethical standards and 7.75 in grade 4). Hydronephrosis may stem from urinary ob- of the institutional and national research committee and with the struction, previous obstruction, or the direct effect of bacterial li- 1964 Helsinki Declaration and its later amendments or compara- popolysaccharide on smooth muscle mediated by nonendothelial ble ethical standards. [19] overproduction of nitric oxide. Hydronephrosis was also one of the preoperative factors associated with postoperative infection Conflict of interest statement [20] after PCNL in a prospective data analysis by Lai and Assimos. History of PCNL was the third most common risk factor. This is No conflict of interest has been declared by the author. also not a surprise; approximately 30% of stones treated with PCNL are infected, and when a candidate for stone surgery has a Funding source history of PCNL, it would be fair to assume that residual and po- [21] tentially infected stone fragments are present. This study was not supported by any external funding. Other factors, such as prior nephrostomy, parameters of stone burden, and history of diabetes mellitus, were not significantly as- Author contributions sociated with positive preoperative culture. Prior stent placement was a significant parameter in the bivariate analysis ( p =0.008) All authors were deeply involved in the design of the study, data but not in the multivariate analysis ( p = 0.08). The presence of analysis, and composition of the manuscript. prior stent was a significant risk factor for infectious complications [11] after URS in a meta-analysis by Ma et al. However, these au- References thors did not address the issue of hydronephrosis, which would [1] Assimos D, Krambeck A, Miller NL, et al. Surgical management of stones: have absorbed the effect of prior stenting. American Urological Association/Endourological Society Guideline, part I. The absence of an effect of stone burden on the risk of positive cul- JUrol 2016;196(4):1153–1160. ture contrasts with several previous reports. Diameter of the largest [2] Türk C, Petřík A, Sarica K, et al. EAU guidelines on interventional stone and number of affected calyces were associated with the risk of treatment for urolithiasis. Eur Urol 2016;69(3):475–482. [12] post-PCNL complications in the study by El-Nahas et al. This dif- [3] Bakr M, Abdelhalim KM. Safety and efficacy of emergency ureteroscopy ference could be attributed to the different outcomes that were ex- with intracorporeal lithotripsy in patients presented with urinary tract infection with mild sepsis. JEndourol 2020;34(3):262–266. amined. While the stone burden is expected to increase operative [4] Seitz C, Desai M, Häcker A, et al. Incidence, prevention, and management time, technical complexity, and complication rate, it may not be di- of complications following percutaneous nephrolitholapaxy. Eur Urol rectly related to the risk of preoperative positive culture. The same 2012;61(1):146–158. consideration applies to a comparison with the study of infectious [5] Zhang H, Jiang T, Gao R, et al. Risk factors of infectious complications [5] complications after RIRS by Zhang et al. These authors showed after retrograde intrarenal surgery: A retrospective clinical analysis. JInt Med Res 2020;48(9):300060520956833. that a stone size greater than 2 cm is a significant risk factor for pos- [6] Preminger GM, Tiselius HG, Assimos DG, et al. 2007 Guideline for the itive urine culture, but they ignored the factor of hydronephrosis. management of ureteral calculi. Eur Urol 2007;52(6):1610–1631. In addition, diabetes mellitus, a well-known risk factor for asymp- [7] Gonen M, Turan H, Ozturk B, Ozkardes H. Factors affecting fever tomatic bacteriuria, was not a risk factor for a positive urine culture. following percutaneous nephrolithotomy: A prospective clinical study. J This can be explained by the much higher prevalence of positive cul- Endourol 2008;22(9):2135–2138. tures in the current study (23.4%), which overwhelmed the preva- [8] Haleblian G, Kijvikai K, de la Rosette J, Preminger G. Ureteral stenting and [22] lence of bacteriuria in diabetic patients (7.9% in diabetic women). urinary stone management: A systematic review. J Urol 2008;179: 424–430. This study is limited by its single-center design. External valida- [9] Gehringer C, Regeniter A, Rentsch K, Tschudin-Sutter S, Bassetti S, Egli A. tion of the findings is certainly needed, and it is hoped that an even Accuracy of urine flow cytometry and urine test strip in predicting better combination of parameters can be achieved. Another limita- relevant bacteriuria in different patient populations. BMC Infect Dis tion was the inclusion of patients treated with antibiotics before 2021;21(1):209. surgery. However, this limitation represents a real-life situation. [10] Nadeem S, Badawy M, Oke OK, Filkins LM, Park JY, Hennes HM. Pyuria and urine concentration for identifying urinary tract infection in young children. Pediatrics 2021;147(2):e2020014068. [11] Ma YC, Jian ZY, Yuan C, Li H, Wang KJ. Risk factors of infectious 5. Conclusions complications after ureteroscopy: A systematic review and meta-analysis based on adjusted effect estimate. Surg Infect (Larchmt) 2020;21(10):811–822. It is possible to determine the risk of a positive urine culture at the time [12] El-Nahas AR, Nabeeh MA, Laymon M, Sheir KZ, El-Kappany HA, of admission to the emergency department using 3 accessible param- Osman Y. Preoperative risk factors for complications of percutaneous eters. This information fills a gap in the guidelines by defining patients nephrolithotomy. Urolithiasis 2021;49(2):153–160. 116 Gofrit et al. Volume 17 Issue 2 2023 www.currurol.org [13] Cheung F, Loeb CA, Croglio MP, Waltzer WC, Weissbart SJ. Bacteria on [19] Kim SC, Seo KK, Kim IK, Kal WJ, Lee MY. Effects of bacterial endotoxin urine microscopy is not associated with systemic infection in patients on the contraction and relaxation responses of the rabbit cavernous with obstructing urolithiasis. JEndourol 2017;31(9):942–945. smooth muscles. JUrol 1999;161(3):964–969. [20] Lai WS, Assimos D. Factors associated with postoperative infection after [14] Yilmaz S, Pekdemir M, Aksu NM, Koyuncu N, Cinar O, Akpinar E. A percutaneous nephrolithotomy. Rev Urol 2018;20(1):7–11. multicenter case-control study of diagnostic tests for urinary tract [21] Bag S, Kumar S, Taneja N, Sharma V, Mandal AK, Singh SK. One week of infection in the presence of urolithiasis. Urol Res 2012;40(1):61–65. nitrofurantoin before percutaneous nephrolithotomy significantly reduces [15] Rohloff M, Shakuri-Rad J, McElrath C, et al. Which objective parameters upper tract infection and urosepsis: A prospective controlled study. are associated with a positive urine culture in the setting of ureteral calculi: The Urology 2011;77(1):45–49. ureteral calculi urinary culture calculator. J Endourol 2018;32(12):1168–1172. [22] Zhanel GG, Harding GK, Nicolle LE. Asymptomatic bacteriuria in [16] Hitchens AP, Leikind MC. The introduction of agar-agar into bacteriology. J patients with diabetes mellitus. Rev Infect Dis 1991;13(1):150–154. Bacteriol 1939;37(5):485–493. [17] Bursle E, Robson J. Non-culture methods for detecting infection. Aust Prescr 2016;39(5):171–175. How to cite this article: Gofrit ON, Abudi R, Lorber A, Duvdevani M. [18] Lipsky BA. Urinary tract infections in men. Epidemiology, pathophysiology, Predicting urine culture results in candidates for lithotripsy. Curr Urol diagnosis, and treatment. AnnInternMed 1989;110(2):138–150. 2023;17(2):113–117. doi: 10.1097/CU9.0000000000000117
Journal
Current Urology – Wolters Kluwer Health
Published: Jun 2, 2023