Preoperative Urine Analysis is An Effective Tool to Predict Fever After Miniaturized Percutaneous Nephrolithotomy on Large Renal Stones Ze Hong Lu1, Tsung Yen Lin1, Ho Shiang Huang1,2, Chan Jung Liu1* Purpose: To investigate the preoperative and intraoperative potential risk factors associated with miniaturized percutaneous nephrolithotomy (mPCNL) fever in the treatment of patients with large renal stones. Materials and Methods: All patients with renal stones larger than 2.5 cm, who had undergone mPCNL, were included in the period between April 2018 and September 2019. Logistic regression analyses were performed to identify clinical variables associated with post-operative fever (>38°C). Results: A total of 53 patients were enrolled for whom the median maximal stone length was 3.08 cm. 24 (45%) patients had a fever after mPCNL. Significantly more patients with urine WBC ≥ 27(/HPF) had a fever after sur- gery (p = 0.004). No significant between-group differences in urine cultures were found for the fever and non-fever groups (p = 0.094). Stepwise and multivariable logistic regression analyses all revealed that urine WBC ≥ 27(/ HPF) is the only risk factor for developing post-mPCNL fever. Based on the highest body temperature, all of the patients were assigned into no fever, mild fever (37.5 ≤ Temp < 38.0), and fever groups, and an ordinal logistic regression analysis still supported the premise that the result of urine analysis is strongly associated with post-mP- CNL fever. Conclusion: Large renal stones are challenging to treat and associated with severe complications. Approximately 45% of large renal stone patients treated via mPCNL developed a fever. Urine WBC can easily and directly predict the risk of fever. Keywords: renal stone; urolithiasis; percutaneous nephrolithotomy; urine analysis; urinary tract infection; fever; sepsis INTRODUCTION Percutaneous nephrolithotomy (PCNL) is the stand-ard of care for the treatment of large renal stones, defined as larger than 2cm(1). Although PCNL is con- sidered to be the most effective therapy, it is definitely associated with high risks of complication. Some pub- lications have even reported complication rates up to 83% following PCNL(2). High complication rates con- tributed to less than 4% nonendourologists performing this surgery(3). PCNL carries two major concerns for complications. Bleeding accounts for most of the PCNL complications, and the incidence of blood transfusion has been reported from 5.5% to 18%(4,5). Given the ad- vancement in surgical techniques and equipment, min- iaturized PCNL (mPCNL) was developed in an effort to reduce bleeding related to standard PCNL. According to UAA (Urology Association of Asia) guidelines(6), mPCNL is recommended for renal stones size < 3.0- 3.5cm with good surgical outcome and less morbidi- ty. However, relatively small tract size restricted the efficacy of stone removal and therefore increased the risks of post operation fever(7). In consideration of in- fectious complications, few studies have used mPCNL to treat large renal stones, which was defined as “partial or complete renal stones filling the renal pelvis and one or more calices with diameter of at least 3 cm”(1,8). Even utilizing PCNL on large renal stones, experienced urol- ogists didn’t have universal consensus on preoperative antibiotics strategies to prevent infection(8). In this retro- spective study, we aimed to investigate the preoperative and intraoperative potential risk factors associated with post-mPCNL fever in the treatment of patients with large renal stones. MATERIAL AND METHODS Study Design and Population We retrospectively recruited reviewed patients from a single tertiary referral medical center between April 2018 and September 2019. The patients who fulfilled the definition of large renal stones and underwent mP- CNL were included(8). The reviewed data included pa- tient demographics, body weight, and height on the ad- mission day, and systemic diseases on medical records (e.g., diabetes mellitus (DM), hypertension (HTN), and cardiovascular disease). Preoperative laboratory in- vestigations included urine analysis, midstream urine culture, complete blood count, renal, liver function tests, and electrolytes. Differentiation of white blood cells was also done on preoperative survey. The plate- 1Department of Urology, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan 70403, Taiwan. 2Department of Urology, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan. *Correspondence: Department of Urology, National Cheng Kung University Hospital, College of Medicine, National Cheng Kung University, Tainan 70403, Taiwan.dragon2043@hotmail.com. Received September 2020 & Accepted September 2021 ENDOUROLOGY AND STONE DISEASE Urology Journal/Vol 18 No. 6/ November-December 2021/ pp. 600-607. [DOI: 10.22037/uj.v18i.6463] let-to-lymphocyte ratio (PLR) and neutrophil-to-lym- phocyte ratio (NLR) were defined as the ratios of the absolute platelet, lymphocyte, and neutrophil counts, respectively. At the last outpatient clinics visit before surgery, midstream urine culture was collected from all the patients. All the patients were admitted one day be- fore surgery. If the urine culture was negative, prophy- lactic intravenous (IV) broad-spectrum antibiotic was given after admission based on the recommendations from American Urology Association guideline(9). In our hospital, cefuroxime is given as prophylactic antibiot- ics prior to operation in a negative urine culture patient. The patients who had positive urine culture were given with appropriate oral form or IV form antibiotics for 7 days according to sensitivity tests. All the patients had at least one abdominal computed tomography (CT) before surgery. The maximal stone length was calculated based on CT images by the opera- tor, and in cases of multiple stones, the stone length was calculated by adding the length of the longest axis of each stone. The mean attenuation levels in Hounsfield units (HUs) were measured by CT. We used the biggest circular diameters to cover the stone and calculated the average HU values. Stone clearance was assessed in- traoperatively by direct renoscopy and postoperatively by radiography images. All patients underwent a plain abdominal film one month after mPCNL to see any re- sidual stones. Stone free was defined as either complete clearance or clearance with insignificant residual frag- ments less than 4 mm in size on the follow-up imag- ing(10). Operation details All the operations were performed by the same expe- rienced surgeon at our hospital using a 1- stage pro- cedure. After induction of general anesthesia, a ureter occlusive catheter was retrogradely placed to the target kidney by cystoscopy. Percutaneous access was per- formed using an 18-gauge needle under combined echo and fluoroscopic assistance. After successful access, a guidewire was inserted into the collecting system and Table 1. Demographics and baseline characteristics of the patients. Non fever (N=29) n (%) Fever (N=24) n (%) p value a Age (years) median (IQR) 62.00 (55.00, 65.00) 63.00 (55.00, 66.50) 0.642 Gender Male 19 (65.52) 11 (45.83) 0.246 Female 10 (34.48) 13 (54.17) BMI (kg/m2) < 25.0 13 (44.83) 11 (45.83) 1.000 ≥ 25.0 16 (55.17) 13 (54.17) median (IQR) 25.60 (22.70, 29.99) 25.05 (22.95, 26.98) 0.655 Stone size < 30 12 (41.38) 12 (50.00) 0.726 ≥ 30 17 (58.62) 12 (50.00) median (IQR) 30.84 (25.58, 38.09) 30.73 (23.60, 42.52) 0.964 Operation time < 120 9 (31.03) 3 (12.50) 0.202 ≥ 120 20 (68.97) 21 (87.50) median (IQR) 130.00 (110.00, 180.00) 120.00 (175.00, 180.00) 0.123 Urine WBC(/HPF) < 27 20 (68.97) 6 (25.00) 0.004 ≥ 27 9 (31.03) 18 (75.00) median (IQR) 19.00 (6.00, 33.00) 87.00 (22.50, 278.00) 0.007 WBC < 10000 28 (96.55) 20 (83.33) 0.164 ≥ 10000 1 (3.45) 4 (16.67) median (IQR) 6500.00 (5100.00, 7600.00) 7250.00 (6250.00, 7950.00) 0.133 GFR < 90 13 (44.83) 12 (50.00) 0.921 ≥ 90 16 (55.17) 12 (50.00) median (IQR) 96.14 (72.35, 107.66) 87.43 (63.45, 115.39) 0.480 PLR <110 11 (37.93) 7 (29.17) 0.705 ≥ 110 18 (62.07) 17 (70.83) median (IQR) 125.83 (92.05, 173.27) 157.37 (106.57, 226.79) 0.085 NLR <5 28 (96.55) 18 (75.00) 0.038 ≥5 1 (3.45) 6 (25.00) median (IQR) 1.91 (1.43, 2.68) 2.01 (1.34, 4.62) 0.416 Hydronephrosis No 13 (44.83) 7 (29.17) 0.376 Yes 16 (55.17) 17 (70.83) HU 900 No 7 (24.14) 6 (25.00) 1.000 Yes 22 (75.86) 18 (75.00) Diabetes mellitus No 25 (86.21) 20 (83.33) 1.000 Yes 4 (13.79) 4 (16.67) UC No 25 (86.21) 15 (62.50) 0.094 Yes 4 (13.79) 9 (37.50) achi-square test or Fisher’s exact test for categorical variables / Mann-whitney U test for continuous variables. Preoperative urine WBC predicts post-PCNL fever - Lu et al. Vol 18 No 6 November-December 2021 601 Endourology and Stones diseases 602 the tract was dilated using balloon dilators until an 18 Amplatz sheath can be placed. Mini-nephroscopy (12 Fr Richard Wolf) was inserted into the Amplatz sheath and stones were disintegrated using Holmium laser. Holmium:yttrium-aluminum-garnet (Ho:YAG) laser 60 W is generated by Sphinx 60(LISA Laser, Pleas- anton, CA, USA) with setting of energy from 0.5-1.5J and frequency from 6-20 Hz for fragmentation(11). The stone fragments were removed with forceps. After the completion of stone extraction, a 6 Fr double J catheter was inserted. A 14F nephrostomy tube was placed at the end of each surgery. The operative time was calculated from the insertion of the cystoscopy to the completion of nephrostomy tube placement. Fever definition and management Ear temperatures were recorded every 2 hours after sur- gery on all patients. Fever was defined as body tem- perature > 38 °C. For every patient, the highest body temperatures were recorded. The normal range of ear temperature is between 35.7 to 37.5 degree(12). Based on the highest body temperature, we further separat- ed non-fever group into no fever and mild fever group (37.5 ≤ Temp < 38.0). All fever patients are treated with IV form antibiotics which are adjusted by urine culture results. The choices of definite antibiotics in fever group are listed in sup- plementary data. The treatment duration is 7 to 14 days with oral or IV form antibiotics according to European Association of Urology (EAU) infections guidelines(13). If there are no available culture results, antibiotics is given with second or third generation of cephalosporin by EAU guidelines recommendation(13). Crude OR p-value Adjusted OR a p-value Adjusted OR b p-value Adjusted OR c p-value (95 % CI) (95 % CI) (95 % CI) (95 % CI) Age (years) 1.02 (0.97-1.08) 0.412 Age (per 10 years) 1.24 (0.74-2.06) 0.412 Gender Male Ref. Female 2.25 (0.74-6.81) 0.153 BMI (kg/m2) 0.93 (0.80-1.08) 0.334 BMI <25.0 Ref. ≥25.0 0.96 (0.32-2.85) 0.942 Stone size 1.01 (0.96-1.06) 0.723 Stone size <30 Ref. ≥30 0.71 (0.24-2.10) 0.531 Operation time 1.01 (0.99-1.02) 0.217 Operation time < 120 Ref. Ref. ≥ 120 3.15 (0.74-13.34) 0.119 5.30 (1.02-27.55) 0.047 Urine WBC 1.00 (0.99-1.00) 0.860 Urine WBC (/HPF) < 27 Ref. Ref. Ref. Ref. ≥ 27 6.67 (1.98-22.44) 0.002 5.48 (1.57-19.10) 0.008 5.08 (1.39-18.60) 0.014 8.86 (2.35-33.42) 0.001 WBC 1.00 (1.00-1.00) 0.094 WBC < 10000 Ref. ≥ 10000 5.60 (0.58-53.94) 0.136 GFR 0.99 (0.98-1.01) 0.406 GFR < 90 Ref. ≥ 90 0.81 (0.28-2.40) 0.707 PLR 1.01 (0.99-1.02) 0.121 PLR < 110 Ref. ≥ 110 1.48 (0.47-4.72) 0.503 NLR 1.20 (0.87-1.66) 0.261 NLR < 5 Ref. Ref. Ref. ≥ 5 9.33 (1.04-84.02) 0.046 5.82 (0.58-58.46) 0.135 5.22 (0.48-56.94) 0.175 Hydronephrosis No Ref. Yes 1.97 (0.63-6.20) 0.245 HU 900 No Ref. Yes 0.95 (0.27-3.35) 0.942 Diabetes mellitus No Ref. Yes 1.25 (0.28-5.63) 0.771 UC No Ref. Ref. Yes 3.75 (0.98-14.33) 0.053 1.52 (0.27-8.58) 0.633 Table 2. Results of operations. a Multivariable logistic regression analysis of variables (p-value < 0.05 in univariate logistic regression analysis). AIC: 65.61 b Multivariable logistic regression analysis of variables (p-value < 0.1 in univariate logistic regression analysis). AIC: 67.45 c Stepwise logistic regression for variables entry in model p < 0.1 p < 0.05 & stay in model p < 0.1 p < 0.05. AIC: 63.97 Preoperative urine WBC predicts post-PCNL fever - Lu et al. Statistical analysis All categorical variables were analyzed by chi-square test or Fisher’s exact test. The Mann-Whitney U test was used to compare continuous variables. For the comparison of three groups, Kruskal-Wallis tests were used to analyze continuous variables. Multiple logistic regression analysis was used to determine any risk fac- tors associated with fever. The variables were selected if their p values were less than 0.10 in univariate logistic regression analysis. For the comparison of three groups, ordinal logistic regression analysis was conducted. All analyses were conducted using SPSS statistical software (versions 16; SPSS Inc., Chicago, CA, USA). Two-tailed p < 0.05 was considered statistically signif- icant. RESULTS A total of 53 patients were enrolled and 56.6% of them were male. The median maximal stone length was 3.08 cm (95% CI=2.98 to 3.57). Most of the patients were above 60 years old (58.5%). The mean age was 59.91 years old (SD=10.99). The overall stone-free rate was 67.9 % (36 of 53 patients). 45.3% (24 of 53 patients) patients had fever after the operation. We compared the baseline characteristics between the fever and the non-fever groups (Table 1). The demographic char- acteristics were generally similar in each group. Only urine WBC was significantly different between the two groups. Among all, only 9 patients didn’t have pyuria before surgery. Significantly more patients with urine WBC ≥ 27 had fever after surgery (p = 0.004). No sig- nificant difference in urine culture was found between the two groups (p = 0.094). The logistic regression anal- ysis (Table 2) indicated that urine WBC ≥ 27(/HPF) is the risk factor for developing post-mPCNL fever. The association between urine culture and post-mPCNL fe- Non fever (N=21) n (%) Mild fever (N=8) n (%) Fever (N=24) n (%) p valuea Age (years) median (IQR) 58.00 (52.00, 65.00) 62.50 (57.50, 66.50) 63.00 (55.00, 66.50) 0.661 Gender Male 15 (71.43) 4 (50.00) 11 (45.83) 0.232 Female 6 (28.57) 4 (50.00) 13 (54.17) BMI (kg/m2) < 25.0 8 (38.10) 5 (62.50) 11 (45.83) 0.570 ≥ 25.0 13 (61.90) 3 (37.50) 13 (54.17) median (IQR) 25.90 (23.92, 30.08) 23.65 (22.62, 26.40) 25.05 (22.95, 26.98) 0.512 Stone size < 30 11 (52.38) 1 (12.50) 12 (50.00) 0.156 ≥ 30 10 (47.62) 7 (87.50) 12 (50.00) median (IQR) 27.77 (23.50, 36.93) 36.14 (31.65, 42.28) 30.73 (23.60, 42.52) 0.160 Operation time < 120 7 (33.33) 2 (25.00) 3 (12.50) 0.246 ≥ 120 14 (66.67) 6 (75.00) 21 (87.50) median (IQR) 120.00 (110.00, 155.00) 160.00 (115.00, 240.00) 175.00 (120.00, 180.00) 0.130 Urine WBC (/HPF) < 27 15 (71.43) 5 (62.50) 6 (25.00) 0.006 ≥ 27 6 (28.57) 3 (37.50) 18 (75.00) median (IQR) 15.00 (5.00, 33.00) 24.50 (9.00, 101.50) 87.00 (22.50, 278.00) 0.016 WBC < 10000 20 (95.24) 8 (100.00) 20 (83.33) 0.355 ≥ 10000 1 (4.76) 0 (0.00) 4 (16.67) median (IQR) 6600.00 (4800.00, 8500.00) 6000.00 (5450.00, 6650.00) 7250.00 (6250.00, 7950.00) 0.188 GFR < 90 11 (52.38) 2 (25.00) 12 (50.00) 0.430 ≥ 90 10 (47.62) 6 (75.00) 12 (50.00) median (IQR) 88.56 (72.35, 110.80) 100.32 (71.58, 105.49) 87.43 (63.45, 115.39) 0.671 PLR < 110 8 (38.10) 3 (37.50) 7 (29.17) 0.798 ≥ 110 13 (61.90) 5 (62.50) 17 (70.83) median (IQR) 121.22 (89.56, 165.01) 145.98 (94.22, 180.08) 157.37 (106.57, 226.79) 0.188 NLR < 5 20 (95.24) 8 (100.00) 18 (75.00) 0.091 ≥ 5 1 (4.76) 0 (0.00) 6 (25.00) median (IQR) 1.78 (1.36, 2.50) 2.31 (1.69, 3.19) 2.01 (1.34, 4.62) 0.519 Hydronephrosis No 9 (42.86) 4 (50.00) 7 (29.17) 0.440 Yes 12 (57.14) 4 (50.00) 17 (70.83) HU 900 No 3 (14.29) 4 (50.00) 6 (25.00) 0.136 Yes 8 (85.71) 4 (50.00) 18 (75.00) Diabetes mellitus No 18 (85.71) 7 (87.50) 20 (83.33) 1.000 Yes 3 (14.29) 1 (12.50) 4 (16.67) UC No 19 (90.48) 6 (75.00) 15 (62.50) 0.094 Yes 2 (9.52) 2 (25.00) 9 (37.50) Table 3. Difference in clinical features and laboratory findings of mPCNL patients subsequently happening fever (Temp ≥ 38.0), mild fever (37.5 ≤ Temp < 38.0) or not (Temp < 37.5). a chi-square test or Fisher’s exact test for categorical variables / Kruskal-Wallis Test for continuous variables. Preoperative urine WBC predicts post-PCNL fever - Lu et al. Vol 18 No 6 November-December 2021 603 Endourology and Stones diseases 604 ver revealed a marginal trend toward significance be- fore adjustment (p = 0.053). However, the significance blunted after adjustment (p = 0.369). In ordinal logistic regression analysis (Table 4), urine culture and NLR revealed significance in univariate analysis, but there was no statistical significance in multivariable logistic regression analysis. Stepwise and multivariable logistic regression analysis also suggested that urine WBC ≥ 27(/HPF) is still the risk factor for developing post-mP- CNL fever. According to Akaike information criterion (AIC) which is listed in Table 2 and Table 4, stepwise logistic regression is the best-fit model. Based on the highest body temperature, all the patients were as- signed to no fever, mild fever, and fever groups (Table 3). Only 8 patients were in the mild fever group. Most clinical characteristics were not significantly different. Only urine WBC was significantly different between the three groups. We used ordinal logistic regression analysis to find any risk factors for developing fever (Table 4). Only urine WBC ≥ 27(/HPF) could predict whether the patients had fever after mPCNL. The area under the curve for WBC ≥ 27 (/HPF) was 0.72 (Figure 1). Using the cutoff of WBC ≥ 27 (/HPF), the sensitiv- ity was 75% and specificity 69%, with an odds ratio of 6.67 (1.98-22.44; p value = 0.002). The bacteria type of urine culture and stone composi- tion of those patients are listed in supplementary ta- ble. Gram-negative bacteria, such as Proteus mirabilis, Klebsiella pneumoniae and Escherichia coli were the most common pathogens from urine culture. In both fe- ver and non-fever groups, calcium oxalate stone was the leading composition from stone analysis. In fever group, average fever lasting days was 1.67 days, only 4 patients had fever lasting more than two days including the operation day, and the onset day of fever was on post-operative day (POD) 0 and 1 in most Crude OR (95 % CI) p-value Adjusted OR a (95 % CI) p-value Adjusted OR b (95 % CI) p-value Adjusted OR c (95 % CI) p-value Age (years) 1.03 (0.98-1.08) 0.236 Age (per 10 years) 1.34 (0.83-2.17) 0.236 Gender Male Ref. Ref. Female 2.48 (0.86-7.11) 0.091 1.83 (0.54-6.13) 0.331 BMI (kg/m2) 0.91 (0.79-1.05) 0.210 BMI < 25.0 Ref. ≥ 25.0 1.28 (0.46-3.55) 0.639 Stone size 1.02 (0.98-1.08) 0.347 Stone size < 30 Ref. ≥ 30 1.04 (0.37-2.88) 0.944 Operation time 1.01 (0.99-1.02) 0.116 Operation time < 120 Ref. Ref. ≥ 120 2.86 (0.81-10.13) 0.104 5.29 (1.24-22.57) 0.025 Urine WBC 1.00 (0.99-1.00) 0.611 Urine WBC (/HPF) < 27 Ref. Ref. Ref. Ref. ≥ 27 5.70 (1.88-17.25) 0.002 4.16 (1.29-13.36) 0.017 3.83 (1.18-12.48) 0.026 8.26 (2.43-28.02) 0.001 WBC 1.00 (1.00-1.00) 0.150 WBC < 10000 Ref. ≥ 10000 4.93 (0.56-43.54) 0.152 GFR 1.00 (0.98-1.01) 0.542 GFR < 90 Ref. ≥ 90 1.05 (0.38-2.91) 0.921 PLR 1.01 (0.99-1.02) 0.104 PLR <110 Ref. ≥110 1.42 (0.48-4.16) 0.525 NLR 1.22 (0.89-1.69) 0.217 NLR <5 Ref. Ref. Ref. ≥5 8.43 (1.00-71.13) 0.050 3.78 (0.41-34.89) 0.241 4.59 (0.47-44.82) 0.190 Hydronephrosis No Ref. Yes 1.68 (0.59-4.82) 0.336 HU 900 No Ref. Yes 0.66 (0.20-2.17) 0.492 Diabetes mellitus No Ref. Yes 1.19 (0.29-4.95) 0.813 UC No Ref. Ref. Ref. Yes 4.03 (1.08-15.06) 0.038 1.81 (0.40-8.20) 0.439 1.40 (0.28-6.87) 0.681 Table 4. Ordinal logistic regression analysis of risk factors for fever among mPCNL patients. a Multivariable logistic regression analysis of variables (p-value < 0.05 in univariate logistic regression analysis). AIC: 103.91 b Multivariable logistic regression analysis of variables (p-value < 0.10 in univariate logistic regression analysis). AIC: 104.98 c Stepwise logistic regression for variables entry in model p < 0.10 p < 0.05 & stay in model p < 0.10 p < 0.05. AIC: 99.18 Preoperative urine WBC predicts post-PCNL fever - Lu et al. cases. The details fever pattern were listed in supple- mentary table. DISCUSSION In the current study, we analyzed the risk factors for developing fever after mPCNL treatment on large re- nal stones. Numerous studies have studied the contrib- uting factors for infectious complications after PCNL, but few aimed at mPCNL. Lai et al. had conducted a meta-analysis on 2018 and a total of 24 studies were recruited, of which 12 were prospective and 12 were retrospective(14). In all the prospective studies, preop- erative urine culture, intraoperative renal pelvic urine culture, and stone culture have been associated with fever after PCNL. Only preoperative urine culture and stone culture were found to be significantly associated with infection of all the retrospective studies. Howev- er, stone culture is not a common preoperative exam in all medical facilities. Besides, the exam should rely on urinary tract stone specimens, which are usually taken from surgery. Therefore, it is unlikely to have results soon after surgery. Taken together, although the stone culture appears to be the strongest risk factor on liter- ature, stone culture is only available after PCNL and, therefore, cannot be used to prevent infectious compli- cations. Urine cultures, including preoperative midstream urine and intraoperative renal pelvis urine, are also associat- ed with post PCNL infectious complications(15,16). Even some studies found that intraoperative renal pelvis urine was more predictable than preoperative urine culture (17,18). The finding was offset by the meta-analysis re- sults(14). Besides, intraoperative renal pelvis urine cul- ture was performed during operation and the culture may take 5 days to have the results, which indicates that intraoperative renal pelvis urine culture was not a practical tool for predicting post-PCNL fever. Preoper- ative midstream urine culture is a common practice to detect latent bacteria in the urinary tract in most facili- ties. However, the accuracy of midstream urine culture for predicting infectious complications after PCNL is always questioned. A prospective study revealed near half positive stone culture patients had negative pre- operative midstream urine culture. Consequently, the author concluded that although preoperative midstream urine culture should be collected, neither a positive nor a negative midstream urine culture influences the risk of postoperative systemic inflammatory response syndrome (SIRS). In the current study, we only had the results of midstream urine culture rather than stone culture or renal pelvis urine culture. No matter in univariable or multivariable analysis, the result of midstream urine culture can’t be the predictor to dis- tinguish whether the patient will have fever after mPC- NL or not. In contrast, the result of urine analysis can strongly predict post mPCNL fever even by the use of stepwise logistic regression. In analysis of mild fever and fever groups, urine WBC ≥ 27 was still strongly associated with post-surgery fever using ordinal logistic regression. All the solid evidence above highly suggest- ed that urine WBC alone can reliably predict the risk of post-mPCNL fever. The first introduction of the technique of mPCNL was in 1997, which was using an 11~15Fr sheath on pediat- ric stone patients by Jackman et al. and Helal et al(19,20). Afterward, mPCNL is generally accepted as tract sizes between 14 Fr and 22 Fr, although a clear definition re- mains controversial(21). Echo, fluoroscopy or combined guided tract creation are applied in mPCNL currently(22). In our hospital protocol, we combine ultrasound and fluoroscopy guidance to create tract. The first step is ultrasound guided needle placement, and then position confirmation by fluoroscopy. Ultrasonography guided calyx access has been proved feasibility, but some pit- falls have been found such as minimal hydronephrosis, superior pole approach or high lying kidneys(23) with Figure 1. ROC curve of urine WBC on post-mPCNL fever. AUC, area under curve. Preoperative urine WBC predicts post-PCNL fever - Lu et al. Vol 18 No 6 November-December 2021 605 Endourology and Stones diseases 606 bare ultrasound guidance. A prospective and rand- omized trial(22) showed combined ultrasonographic and fluoroscopic guidance for percutaneous renal access in mini-percutaneous nephrolithotomy is safe and effec- tive especially in complex renal stone. Ultrasonic and pneumatic lithotripsy devices have showed efficacy and safety in PCNL(24) Due to narrow working channel of nephroscopes in mP- CNL, Holmium laser (Sphinx 60, LISA Laser, Pleas- anton, CA, USA) is applied in our institute. However, small diameter dual energy lithotripsy has showed com- parable stone clearing in mPCNL(25). It needs further in- vestigations and large size studies in the future. Most available evidence support the role of mPCNL is more suitable for smaller rather than larger renal stones >20 mm(21). The main reasons for the limitations are increased operation time(26) and concerning infectious complications. In the literature, fever occurred in 21%- 39.8% of patients who underwent PCNL, but small number of patients progressed to sepsis or even mor- tality(27). During any endoscopic surgeries, irrigation is always requested to maintain a clear visual field. High intrapelvic pressure (IPP) caused by irrigation can lead to pyelovenous and pyelolymphatic backflow, which will transmit bacteria and endotoxin into the systemic circulation and infectious complications develop(28). Comparing with standard PCNL, miniaturized percuta- neous sheath restricted efficient circulation of irrigation fluid and then IPP increased during mPCNL (29). The- oretically, mPCNL could be prone to have post-surgery fever, and a study confirmed this hypothesis with the re- sult of nearly two times higher incidence of fever after mPCNL compared with the standard(7). In our results, near half of the patients had fever episodes after mP- CNL, but only 4 patients (7.5%) persisted fever more than two days after the operation (including the opera- tion day). None of them had septic shock or sepsis. The findings can be explained by the hypothesis mentioned above that transient peak IPP leads to fever but is soon ameliorated under an adequate control of outflow with a JJ catheter insertion or nephrostomy tube placement. There are limited studies assessing mPCNL on large renal stones until very recently. Kandemir et al. and Güler et al. all introduced the outcomes of mPNL in the treatment of renal stones ≥ 3cm(30,31). The stone free rate (SFR) reported in two studies were 75.0% and 76.5%, respectively. In discordance with the literature, we have found an obvious lower SFR (67.9%). The reasons for the different results obtained in the present study might be that 8 of them (15.0%) were cases with complete staghorn stones. Accumulative evidence suggests that staghorn stones are the most difficult to achieve stone clearance. Besides, the number of cases enrolled in the present cohort is relatively small. The difference in a few cases could easily affect the proportion of the out- come. There are some limitations to this study. First, the study was based on retrospective patient data from a single center. Large-scale and prospective design studies will be needed for further analysis. Second, we did not dis- cuss the relationship between fever and residual stone. Besides, the stone sizes in the current study ranged too wide, which would limit the specificity of the analy- sis. In contrast, it is worthy to mention that this is the first study to analyze the possible factors contributing to fever after mPCNL in the treatment of kidney stones larger than 3cm. Our investigation is also the first one to use the peak body temperature to ordinally evaluate post-mPCNL infectious complications rather than fever or not. The ordinal logistic regression analysis definite- ly strengthens our findings. CONCLUSIONS In patients with large renal stones, mPCNL is associated with adequate stone clearance rate but high incidence of post-surgery fever. Urine WBC alone rather than urine culture can reliably predict the risk of post-mPCNL fe- ver. Using the cutoff of WBC ≥ 27, the predictive sen- sitivity was 75% and specificity was 69%. ACKNOWLEDGEMENTS We are grateful to Associate Professor Sheng-Hsiang Lin for providing the statistical consulting services from the Biostatistics Consulting Center, Clinical Med- icine Research Center, National Cheng Kung Univer- sity Hospital. This study was funded by the National Cheng Kung University Hospital (NCKUH-11002035). CONFLICTS OF INTEREST The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. APPENDIX https://journals.sbmu.ac.ir/urolj/index.php/uj/libraryFiles/downloadPublic/28 REFERENCES 1. Turk C, Petrik A, Sarica K, et al. EAU Guidelines on Interventional Treatment for Urolithiasis. Eur Urol. 2016;69:475-82. 2. Michel MS, Trojan L, Rassweiler JJ. Complications in percutaneous nephrolithotomy. Eur Urol. 2007;51:899-906; discussion 3. Oberlin DT, Flum AS, Bachrach L, Matulewicz RS, Flury SC. Contemporary surgical trends in the management of upper tract calculi. J Urol. 2015;193:880-4. 4. Liatsikos EN, Kallidonis P, Stolzenburg JU, et al. Percutaneous management of staghorn calculi in horseshoe kidneys: a multi-institutional experience. J Endourol. 2010;24:531-6. 5. De S, Autorino R, Kim FJ, et al. Percutaneous nephrolithotomy versus retrograde intrarenal surgery: a systematic review and meta- analysis. Eur Urol. 2015;67:125-37. 6. Taguchi K, Cho SY, Ng AC, et al. The Urological Association of Asia clinical guideline for urinary stone disease. Int J of Urol. 2019;26:688-709. 7. Wu C, Hua LX, Zhang JZ, Zhou XR, Zhong W, Ni HD. Comparison of renal pelvic pressure and postoperative fever incidence between standard- and mini-tract percutaneous nephrolithotomy. Kaohsiung J Med Sci. 2017;33:36-43. 8. Schoenthaler M, Hein S, Seitz C, et al. The stone surgeon in the mirror: how are German- speaking urologists treating large renal stones Preoperative urine WBC predicts post-PCNL fever - Lu et al. Vol 18 No 6 November-December 2021 607 today? World J Urol. 2018;36:467-73. 9. Lightner DJ, Wymer K, Sanchez J, Kavoussi L. Best Practice Statement on Urologic Procedures and Antimicrobial Prophylaxis. J Urol. 2019101097JU0000000000000509. 10. Resorlu B, Oguz U, Resorlu EB, Oztuna D, Unsal A. The impact of pelvicaliceal anatomy on the success of retrograde intrarenal surgery in patients with lower pole renal stones. Urology. 2012;79:61-6. 11. Ganesamoni R, Sabnis RB, Mishra S, et al. Prospective randomized controlled trial comparing laser lithotripsy with pneumatic lithotripsy in miniperc for renal calculi. J Endourol. 2013;27:1444-9. 12. Geneva, II, Cuzzo B, Fazili T, Javaid W. Normal Body Temperature: A Systematic Review. Open Forum Infect Dis. 2019;6:ofz032. 13. Bonkat G, Pickard R, Bartoletti R, et al. EAU guidelines on urological infections. Eur Urol. 201722-6. 14. Lai WS, Assimos D. Factors Associated With Postoperative Infection After Percutaneous Nephrolithotomy. Rev Urol. 2018;20:7-11. 15. Mariappan P, Smith G, Bariol SV, Moussa SA, Tolley DA. Stone and pelvic urine culture and sensitivity are better than bladder urine as predictors of urosepsis following percutaneous nephrolithotomy: a prospective clinical study. J Urol. 2005;173:1610-4. 16. Korets R, Graversen JA, Kates M, Mues AC, Gupta M. Post-percutaneous nephrolithotomy systemic inflammatory response: a prospective analysis of preoperative urine, renal pelvic urine and stone cultures. J Urol. 2011;186:1899-903. 17. Dogan HS, Sahin A, Cetinkaya Y, Akdogan B, Ozden E, Kendi S. Antibiotic prophylaxis in percutaneous nephrolithotomy: prospective study in 81 patients. J Endourol. 2002;16:649- 53. 18. Margel D, Ehrlich Y, Brown N, Lask D, Livne PM, Lifshitz DA. Clinical implication of routine stone culture in percutaneous nephrolithotomy--a prospective study. Urology. 2006;67:26-9. 19. Jackman SV, Docimo SG, Cadeddu JA, Bishoff JT, Kavoussi LR, Jarrett TW. The "mini-perc" technique: a less invasive alternative to percutaneous nephrolithotomy. World J Urol. 1998;16:371-4. 20. Helal M, Black T, Lockhart J, Figueroa TE. The Hickman peel-away sheath: alternative for pediatric percutaneous nephrolithotomy. J Endourol. 1997;11:171-2. 21. Ruhayel Y, Tepeler A, Dabestani S, et al. Tract Sizes in Miniaturized Percutaneous Nephrolithotomy: A Systematic Review from the European Association of Urology Urolithiasis Guidelines Panel. Eur Urol. 2017;72:220-35. 22. Zhu W, Li J, Yuan J, et al. A prospective and randomised trial comparing fluoroscopic, total ultrasonographic, and combined guidance for renal access in mini‐percutaneous nephrolithotomy. BJU int. 2017;119:612-8. 23. Basiri A, Kashi AH, Zeinali M, Nasiri M, Sarhangnejad R, Valipour R. Ultrasound- guided access during percutaneous nephrolithotomy: entering desired calyx with appropriate entry site and angle. Int Braz J Urol. 2016;42:1160-7. 24. Radfar MH, Basiri A, Nouralizadeh A, et al. Comparing the efficacy and safety of ultrasonic versus pneumatic lithotripsy in percutaneous nephrolithotomy: a randomized clinical trial. Eur Urol focus. 2017;3:82-8. 25. Timm B, Farag M, Davis NF, et al. Stone clearance times with mini-percutaneous nephrolithotomy: Comparison of a 1.5 mm ballistic/ultrasonic mini-probe vs. laser. Can Urol Assoc J. 2021;15:E17. 26. Lahme S. Miniaturisation of PCNL. Urolithiasis. 2018;46:99-106. 27. Wollin DA, Joyce AD, Gupta M, et al. Antibiotic use and the prevention and management of infectious complications in stone disease. World J Urol. 2017;35:1369-79. 28. Jung H, Norby B, Frimodt-Moller PC, Osther PJ. Endoluminal isoproterenol irrigation decreases renal pelvic pressure during flexible ureterorenoscopy: a clinical randomized, controlled study. Eur Urol. 2008;54:1404-13. 29. Tepeler A, Akman T, Silay MS, et al. Comparison of intrarenal pelvic pressure during micro-percutaneous nephrolithotomy and conventional percutaneous nephrolithotomy. Urolithiasis. 2014;42:275-9. 30. Kandemir E, Savun M, Sezer A, Erbin A, Akbulut MF, Sarilar O. Comparison of Miniaturized Percutaneous Nephrolithotomy and Standard Percutaneous Nephrolithotomy in Secondary Patients: A Randomized Prospective Study. J Endourol. 2019. 31. Guler A, Erbin A, Ucpinar B, Savun M, Sarilar O, Akbulut MF. Comparison of miniaturized percutaneous nephrolithotomy and standard percutaneous nephrolithotomy for the treatment of large kidney stones: a randomized prospective study. Urolithiasis. 2019;47:289- 95. Preoperative urine WBC predicts post-PCNL fever - Lu et al.