Prognostic Impact of Stone-Scoring Systems After Percutaneous Nephrolithotomy for Staghorn Calculi: A Single Center's Experience Over 10 Years

Abstract

Introduction:

To investigate the prognostic factors associated with stone-free rate (SFR) and complications after percutaneous nephrolithotomy (PCNL) for the treatment of staghorn stone and to compare the predictive value and accuracy of three stone-scoring systems for the treatment success of staghorn stone.

Patients and Methods:

We retrospectively reviewed all patients undergoing PCNL at our center from June 2003 to June 2014. On the basis of noncontrast computed tomography (NCCT) scan images, we calculated Guy's score, S.T.O.N.E. nephrolithometry, and Clinical Research Office of the Endourological Society (CROES) nomogram to assess the association with stone-free status and complications. For statistical evaluation, univariate and multivariate logistic regression analyses were used.

Results:

During the study period, 886 cases had medical records available. Cases who underwent PCNL for the treatment of staghorn calculi accounted for 34.4% (305/886 cases). Preoperative NCCT was performed in 217 cases. The 217 procedures (205 patients, 12 simultaneous bilateral PCNLs) had a mean stone size of 1358.3 ± 760.7 mm², with 111 (51.2%) partial staghorn and 106 (48.8%) complete staghorn stones. The initial and overall SFRs of PCNL were 53.9% and 70.1%, respectively. The overall complication rate was 32.7% (71/217 cases). On a multivariate logistic regression analysis, independent predictors for SFR were number of involved calices, S.T.O.N.E. nephrolithometry, and pre-existent urinary tract infection (UTI) (odds ratios [ORs] = 1.311, 1.933, and 2.340, respectively). Stone burden was an independent risk factor for the development of complications on multivariate analysis (OR = 2.846 and p = 0.001).

Conclusion:

The results of this study show that multiple involved calices, high grades of S.T.O.N.E. nephrolithometry, and pre-existent UTIs were associated with lower SFR after PCNL for staghorn calculi. Stone burden was an independent risk factor for the development of complications.

Introduction

Since the first description of percutaneous renal stone extraction by Fernström and Johansson in 1976,¹ percutaneous nephrolithotomy (PCNL) has become the standard procedure for the treatment of renal stones and staghorn calculi.^2,3 Recent advances in surgical technique and technology have contributed to high stone-free rates (SFRs) (74%–83%) and low morbidity rates (7%–27%) after standard PCNL for the treatment of the staghorn calculi.² Nonetheless, PCNL is more challenging when used for the treatment of staghorn calculi, because the complete clearance of staghorn calculi may increase the need for multiple tracts and the incidence of secondary procedures. Although multiple-tract access did not lead to a more severe reduction in renal function than singe-tract access,⁴ multiple-tract access may lead to a higher complication rate.⁵ El-Nahas and colleagues⁴ and Akman and colleagues⁶ demonstrated that >60% of patients with residual stones after PCNL required a second intervention. To perform complete stone clearance with minimal morbidity, it is important to identify preoperative factors that affect the SFR and complication rate. In this article, we investigated the prognostic factors affecting the SFR and complication rate of PCNL for the treatment of staghorn calculi over 10 years at our single institute and analyzed the accuracy of the Guy's stone score,⁷ S.T.O.N.E. nephrolithometry,⁸ and the Clinical Research Office of the Endourological Society (CROES) nomogram⁹ for predicting the postoperative outcomes of PCNL for staghorn calculi.

Patients and Methods

Study design

We have been performing PCNL at our center since 1987. Imaging data were available from June 2003. We retrospectively reviewed the electronic medical records of all patients undergoing PCNL at our center from June 2003 to January 2015. In this period, 899 PCNLs were performed. Thirteen with incomplete data were excluded. Cases who underwent PCNL for the treatment of staghorn calculi accounted for 34.4% (305/886 cases). Preoperative noncontrast computed tomography (NCCT) was performed in 217 cases. The 217 procedures (205 patients, 12 simultaneous bilateral PCNLs) performed in patients had a mean stone size of 1358.3 ± 760.7 mm², with 111 (51.2%) partial staghorn and 106 (48.8%) complete staghorn stones. Complete staghorn calculi were defined as stones occupying ≥80% of the renal pelvis and caliceal or collecting system, and partial staghorn calculi were defined as stones occupying the renal pelvis and at least two calices.^2,10,11 We routinely recorded intraoperative events, nephroscopic findings, operative time according to surgical stage, number and location of tracts and stones, percutaneous puncture site and level, number of involved calices (NIC), initial and overall stone-free status (SFS), and stone burden. Stone burden was calculated using the longitudinal and transverse diameters (mm²).¹² To identify a cutoff value of stone burden, predictive SFR, and complication rates, the Youden index method of receiver operating characteristic (ROC) analysis was used. The criterion associated with stone burden for SFR and complication rate was a stone burden ≥1700 mm² and a stone burden of ≥990 mm², respectively (Table 1).

Table 1.

Univariate Analysis of Patient Demographics, Stone Characteristics, and Perioperative Factors Affecting Stone-Free Rate and Complications After Percutaneous Nephrolithotomy

		Stone-free status, N (%)			Complications, N (%)
Demographics and factors	Total	Stone-free	Residual	p	Noncomplicated	Complicated	p
No. of procedures	217	117/217 (53.9)	100/217 (46.1)		146/217 (67.3)	71/217 (32.7)
Mean age (years)	55.2 ± 13.8	56.1 ± 13.9	54.2 ± 13.7	0.314	55.8 ± 13.4	53.9 ± 14.6	0.335
Sex—male (%)	127 (58.5)	75 (59.1)	52 (52.0)	0.071	84 (57.5)	43 (43.0)	0.671
Stone laterality—left (%)	143 (65.9)	86 (60.1)	57 (57.0)	0.011	99 (67.8)	44 (62.0)	0.395
ASA ≥3 (%)	12 (5.5)	8 (6.8)	4 (4.0)	0.623	8 (5.5)	4 (5.6)	0.998
CCI (%)	3.1 ± 1.8	3.1 ± 1.8	3.1 ± 1.8	0.874	3.1 ± 1.7	3.2 ± 2.1	0.784
≥3	85 (39.2)	49 (41.9)	36 (36.0)	0.376	58 (39.7)	27 (38.0)	0.810
Mean BMI (kg/m²)	24.7 ± 3.8	24.9 ± 3.3	24.5 ± 4.3	0.457	24.9 ± 3.9	24.2 ± 3.6	0.163
Genitourinary anomaly (%)	10 (4.6)	4 (3.4)	6 (6.0)	0.366	6 (4.1)	4 (5.6)	0.732
HN ≥moderate grade	49 (22.6)	20 (17.1)	29 (29.0)	0.037	33 (22.6)	16 (22.5)	0.991
Mean operative time (min)	88.2 ± 53.7	73.5 ± 35.7	105.4 ± 65.1	<0.001	84.2 ± 51.9	96.4 ± 56.6	0.006
Length of stay (days)	4.1 ± 2.9	3.7 ± 2.1	4.7 ± 3.5	0.011	3.3 ± 1.4	5.9 ± 4.1	<0.001
Change in Hb level (g/mL)	2.3 ± 1.52	2.2 ± 1.4	2.4 ± 1.6	0.350	2.2 ± 1.3	2.5 ± 1.9	0.139
Transfusion (%)	16 (7.4)	3 (2.6)	13 (13.0)	0.003	0 (0.0)	16 (22.5)	<0.001
Previous renal surgery (%)	25 (11.5)	18 (15.4)	7 (7.0)	0.054	20 (13.7)	5 (7.0)	0.150
Preoperative SWL (%)	38 (17.5)	21 (17.9)	17 (17.0)	0.855	29 (19.9)	9 (12.7)	0.191
Pre-existent UTI (%)	44 (20.3)	17 (14.5)	27 (27.0)	0.023	25 (17.1)	19 (26.8)	0.098
Stone opacity—lucent (%)	25 (11.5)	17 (14.5)	8 (8.0)	0.133	20 (13.7)	5 (7.0)	0.150
Complete staghorn (%)	106 (48.8)	38 (32.5)	68 (68.0)	<0.001	63 (43.2)	43 (60.6)	0.016
Percutaneous puncture site				0.001			0.483
Single subcostal	126 (58.1)	77 (65.8)	49 (49.0)		88 (60.3)	38 (53.5)
Single supracostal	50 (23.0)	28 (23.9)	22 (22.0)		31 (21.2)	19 (26.8)
Multiple	41 (18.9)	12 (10.3)	29 (29.0)		27 (18.5)	14 (19.7)
Mean stone density (HU)	917.5 ± 392.2	903.4 ± 422.1	933.9 ± 355.4	0.564	912.0 ± 399.1	928.8 ± 380.1	0.768
>950	114	57 (48.7)	57 (57.0)	0.223	77 (52.7)	37 (52.1)	0.931
Tract multiplicity	41 (18.9)	12 (10.3)	29 (29.0)	<0.001	27 (18.5)	14 (19.7)	0.829
No. of tract	1.2 ± 0.6	1.3 ± 0.6	1.2 ± 0.5	0.563	1.3 ± 0.6	1.2 ± 0.6	0.553
Mean tract length (mm)	73.2 ± 14.8	74.4 ± 12.6	71.9 ± 17.0	0.234	74.2 ± 14.5	71.4 ± 15.2	0.190
>100	9 (4.1)	3 (2.6)	6 (6.0)	0.206	7 (4.8)	2 (2.8)	0.721
No. of involved calices	3.9 ± 2.0	3.3 ± 1.5	4.7 ± 2.2	<0.001	3.8 ± 1.9	4.3 ± 2.1	0.015
≥4	119 (54.8)	51 (43.6)	68 (68.0)	<0.001	75 (51.4)	44 (62.0)	0.141
Mean stone burden (mm²)	1358.3 ± 760.7	1096.4 ± 495.1	1664.7 ± 894.0	<0.001	1258.1 ± 644.6	1564.4 ± 927.7	<0.001
≥990	131 (60.4)	NA	NA	NA	77 (52.7)	54 (76.1)	0.001
≥1700	54 (24.9)	13 (11.1)	41 (41.0)	<0.001	NA	NA	NA
Guy's score	3.5 ± 0.5	3.3 ± 0.5	3.7 ± 0.5	<0.001	3.4 ± 0.5	3.6 ± 0.5	0.016
S.T.O.N.E. score	9.5 ± 1.3	9.0 ± 1.1	10.1 ± 1.2	<0.001	9.4 ± 1.3	9.7 ± 1.2	0.110
CROES nomogram	73.5 ± 21.0	78.0 ± 22.2	68.3 ± 18.3	<0.001	74.2 ± 21.4	72.2 ± 20.3	0.503

ASA = American Society of Anesthesiology; BMI = body mass index; CCI = Charlson comorbidity index; CROES = Clinical Research Office of the Endourological Society; Hb = hemoglobin; HN = hydronephrosis; HU = Hounsfield unit; NA = not applicable; SWL = extracorporeal shock wave lithotripsy; UTI = urinary tract infection.

During the study period, plain radiography of the kidneys, ureters, and bladder (KUB) and an NCCT scan were performed in all patients preoperatively. We calculated the Guy's stone score, S.T.O.N.E. nephrolithometry, and CROES nomogram scoring systems. All scores of each stone-scoring system were scored by one fellow (S.W. Choi) according to the descriptive methods of each original study.^7
–9 On postoperative day (POD) 1, a plain KUB radiography was routinely performed in all patients. A postoperative NCCT scan was performed to evaluate patients with preoperative radiolucent stones, intraoperatively suspected residual fragments, or complications. All patients were generally followed up at postoperative 1 week and 3 months. An SFS was defined as completely clear of stones or clinically insignificant residual fragments. Patients with clinically insignificant residual fragments were those with asymptomatic, nonobstructing, and noninfectious fragments that were ≤4 mm in diameter.¹³ Plain chest radiography was performed to check a hydro/pneumothorax if a supracostal approach was used intraoperatively. We classified all complications according to the modified Clavien classification of the CROES PCNL study group.¹⁴ Our institutional review board approved this study.

Surgical technique

Our surgical techniques were previously described.^15,16 Prophylactic antibiotics were administrated to all patients. Patients with positive urine cultures were treated preoperatively. All PCNLs were performed under general anesthesia with a standard or tubeless PCNL procedure. The patient was placed in the prone position with all pressure points padded. Percutaneous renal access was accomplished under C-armed fluoroscopy. The access tract was dilated using Amplatz or balloon dilators of as much as 24F or 30F. Staghorn stones were disintegrated using a pneumatic or ultrasonic lithotripter. The residual fragments were removed using retrieval graspers or various forceps. After nephroscopic and fluoroscopic assessment of residual fragments and collecting system status at the completion of the procedure, nephrostomy tubes and/or Double-J stents were left in place if needed. In the standard PCNL procedure, the 20F-sized nephrostomy tubes were placed after the end of the procedure. All procedures were performed by three experienced endourologists. Ancillary treatments such as extracorporeal shock wave lithotripsy (SWL), re-PCNL, and ureterorenoscopic lithotripsy were considered to remove residual fragments when indicated. On POD 2–4, the nephrostomy tube was removed if there was no evidence of pain or leakage after clamping the tube. Double-J stents were removed as an outpatient department procedure on POD 10–14.

Patient demographics and perioperative factors

Demographics and perioperative factors included mean age, sex, stone laterality, the American Society of Anesthesiology (ASA) classification, Charlson comorbidity index,¹⁷ mean body mass index, genitourinary anomaly, grade of hydronephrosis (categorized as the Fetal Urology Society classification system),¹⁸ mean operative time, length of hospital stays (LOS), changes in hemoglobin, transfusion rate, prior renal surgery on the ipsilateral side, presence of SWL, pre-existent urinary tract infection (UTI), stone opacity, staghorn type, percutaneous puncture site, tract multiplicity, NIC, mean stone burden, Guy's stone score, S.T.O.N.E. nephrolithiometry, and CROES nomogram score (Table 1).

Statistical analysis

For statistical evaluation, univariate and multivariate logistic regression analyses were used. Preoperative parameters were subjected to univariate analysis. Continuous variables were expressed as means ± standard deviation and compared using the independent samples (t) test. Categorical variables were presented as numbers with percentages and compared using chi-squared and Fisher's exact tests. A multivariate logistic regression analysis was performed to determine prognostic factors affecting SFRs and complication rates if there were significant preoperative parameters on univariate analysis. The area under the curve (AUC) of the ROC analysis was applied to evaluate the accuracies of prognostic factors and scoring systems for prediction of SFR. Statistical significance was considered at a p-value <0.05. All statistical analyses were performed using the Statistical Package of Social Science for Windows (SPSS, Inc., Chicago, IL) version 20.0 software. Pairwise comparison and Youden index of ROC curves were analyzed using the MedCalc software, version 14.8.1 (website: www.medcalc.org/).

Results

Univariate analysis of perioperative data

The patient demographics and stone characteristics are shown in Table 1. The study included a total of 217 PCNL procedures (205 patients, 12 simultaneous bilateral PCNLs). Staghorn stones were classified as 111 (51.2%) partial staghorn and 106 (48.8%) complete staghorn stones. Multi-tract PCNLs were needed in 41 out of 217 cases (18.9%). The mean stone burden was 1358.3 ± 760.7 mm². The initial SFR was 53.9% (117/217). The overall SFR after auxiliary procedures was 70.1% (152/217). The stone laterality, grade of hydronephrosis, mean operative time, pre-existent UTI, LOS, transfusion rate, staghorn type, percutaneous puncture site, tract multiplicity, NIC, and mean stone burden were significantly different between SFS and nonstone-free status (NSFS). When compared with patients with SFS, patients with residual fragments had more additional tracts (SFS vs NSFS: 10.3% vs 29.0%). The stone characteristics in patients with NSFS had larger burdens, more complete staghorn calculi and involved calices than those with SFS (1096.4 ± 495.1 mm² vs 1664.7 ± 894.0 mm², 32.5% vs 68.0%, and 3.3 vs 4.7, respectively). In addition, patients with residual fragments had a more positive association with pre-existent UTI (14.5% vs 27.0%).

All complications are summarized in Table 2. A total of 86 (40.6%) complications occurred in 71 (32.7%) patients. Most complications were minor (63.4%, 45/71). There were 26 out of 71 patients (36.6%) with major complications (grade IIIa–V). A univariate analysis according to complication status showed statistically significant differences in the staghorn stone type, the NIC, and stone burden. (p = 0.016, p = 0.015, and p < 0.001, respectively). Univariate analysis of prognostic factors affecting the SFR and complication rate of PCNL for staghorn stones is shown in Table 1.

Table 2.

Perioperative Complications According to the Modified Clavien Classification System

Complication grading	No. of complications (%)
Grade I	28 cases (12.9)
Urine leakage managed by watchful waiting	10
Postoperative fever (>38.0°C) managed by observation without antibiotics	6
Bleeding managed using IV fluid without need for blood transfusion	5
Bleeding that requires skin compression/pressure dressing	2
Ureteric clot managed by watchful waiting	2
Wound infection managed by dressing without antibiotics	1
Postoperative pain managed by opioid with or without adjunct analgesic regimen	1
Renal pelvic perforation managed by watchful waiting	1
Grade II	27 cases (12.4)
Bleeding requiring blood transfusion	16
Symptomatic UTI managed using antibiotics	3
Postoperative ileus managed by nasogastric decompression and TPN	2
Pulmonary edema managed by diuretics	1
Postoperative fever (>38.0°C) managed with antibiotics in the ward	1
Colon perforation managed conservatively using IV fluid and antibiotics without controlled colocutaneous fistula	1
Hyposaturation managed by oxygen in the ward	1
Minor atelectasis requiring medical management	1
Delirium managed by medication	1
Grade IIIa	12 cases (5.5)
Ureteric clot obstruction managed by ureteral stenting without general anesthesia	6
Urine leakage managed by ureteral stenting without general anesthesia	5
Febrile UTI or suspected sepsis without organ failure requiring supportive therapy and enhanced monitoring	4
Renal pelvic perforation managed by ureteral stenting without general anesthesia	1
Bleeding requiring multiple bladder washouts/irrigations	1
Grade IIIb	6 cases (2.8)
Bleeding managed by angioembolization	2
Colon perforation managed by colostomy	2
Avulsion of the ureteropelvic junction managed by surgical repair	1
Bleeding managed by nephrectomy	1
Grade IVa	1 case (0.5)
Pulmonary edema requiring ICU management	1
Grade IVb	4 cases (1.8)
Urosepsis with multiple organ failure requiring ICU management	4
Grade V	4 cases (1.8)
Any complication leading to death	4
Total complications	86 cases in 71 patients, 40.6% in 32.7%

ICU = intensive care unit; IV = intravenous; TPN = total parenteral nutrition.

Postoperative data

Mean operative time (73.5 ± 35.7 minutes vs 105.4 ± 65.1 minutes, p < 0.001) and LOS (3.7 ± 2.1 vs 4.7 ± 3.5, p = 0.011) in patients with residual fragments were significantly longer than those of stone-free patients. When compared with noncomplicated patients, complicated patients had a significantly longer operation time (84.2 minutes vs 96.4 minutes, p = 0.006) and LOS (3.3 ± 1.4 days vs 5.9 ± 4.1 days, p < 0.001), and a greater transfusion rate (0.0% vs 22.5%, p < 0.001).

Multivariate analysis of prognostic factors and scoring systems

All three scoring systems had a significant association with initial SFR (each p < 0.001, Table 1). On univariate analysis according to complication status, Guy's stone score had the only significant association (3.4 ± 0.5 vs 3.6 ± 0.5, p = 0.016, Table 1). On a multivariate logistic regression analysis (Table 3), independent predictors for SFR were NIC, S.T.O.N.E. nephrolithometry, and pre-existent UTI (odds ratios [ORs] = 1.311, 1.933, and 2.340, respectively). Stone burden ≥990 mm² was an independent risk factor for the development of complications on multivariate analysis (OR = 2.846 and p = 0.001, Table 3). In terms of prediction of SFR, the S.T.O.N.E. nephrolithometry revealed the highest AUC (0.746) than the other predictors (NIC: 0.687, pre-existent UTI; 0.562) and scoring systems (Guy's score: 0.678 and CROES nomogram: 0.627, Table 4).

Table 3.

Independent Predictors of Stone-Free Rates and Risk Factor of Complications in a Multivariate Logistic Regression Analysis

Variables	B	Exp(B)	95% CI	p
1. Independent predictors of stone-free rates
NIC	0.271	1.311	1.092, 1.573	<0.001
S.T.O.N.E. score	0.659	1.933	1.441, 2.593	0.004
Pre-existent UTI	0.850	2.340	1.082, 5.062	0.031
2. Risk factors of complications
Stone burden ≥990 mm²	1.046	2.846	1.509, 5.39	0.001

CI = confidence interval; NIC = number of involved calices.

Table 4.

Receiver Operating Characteristic Curve Values by Prognostic Factors and Scoring Systems

Prognostic factors scoring systems	AUC	p	95% CI
NIC	0.687	<0.001	0.615, 0.758
S.T.O.N.E. score	0.746	<0.001	0.680, 0.811
Pre-existent UTI	0.562	0.114	0.485, 0.639
Guy's score	0.678	<0.001	0.605, 0.750
CROES nomogram	0.627	0.001	0.554, 0.701

AUC = area under the curve.

Discussion

The first report by Clayman and colleagues¹⁹ of the treatment of staghorn stones using PCNL was a considerable advancement in surgical techniques and instruments. In 2005, the American Urological Association guidelines recommended PCNL monotherapy as the first-line treatment for patients with staghorn calculi.² Although PCNL has high efficacy and low morbidity, its use for the treatment of staghorn calculi is more challenging. This is attributable to the complexity, large size, and anomalous morphology of staghorn calculi. The treatment of staghorn calculi aims at achieving complete stone clearance with minimal morbidity.

To achieve an SFS with minimal morbidity, it is important to determine the preoperative factors that affect the SFR and complication rate of PCNL for the treatment of staghorn calculi. We considered that recently proposed stone-scoring systems^7
–9 could be predictive of post-PCNL outcomes. The Guy's stone score, S.T.O.N.E. nephrolithometry, and CROES nomogram have been externally validated.^20

–24 Labadie and colleagues evaluated these three stone-scoring systems in 246 patients who underwent PCNL at three academic institutions.²² The three stone-scoring systems were found to be equally predictive of SFS. No association between the CROES nomogram and complications was found. However, the multivariate logistic regression analysis of Bozkurt and colleagues showed a significant relationship between the CROES nomogram and SFR (p < 0.001) and complication rate (p < 0.001).²⁴ In our study, the CROES nomogram was not significantly associated with the SFR and complication rate by multivariate logistic regression analysis. Sfoungaristos and colleagues compared the three stone-scoring systems in patients who underwent PCNL for the treatment of staghorn stones. They reported that S.T.O.N.E. nephrolithometry was the only predictor of the SFR after PCNL for staghorn stones.²⁵

The three stone-scoring systems had some similar and dissimilar variables. They, respectively, incorporated individual variables in a different fashion. The stone count and presence of staghorn stone are important common factors of the three stone-scoring systems. The Guy's score included renal abnormality and did not directly reflect stone burden and density. Although the Guy's score is easy to implement in daily practice, it is limited in describing technical difficulties and risk stratification of the procedure and clinical variability because of its simplicity. The CROES nomogram reflected prior treatment and surgeon case volume but not hydronephrosis, renal abnormality, and stone density. This nomogram is complex to implement in the clinical setting. Unlike the other scoring systems, S.T.O.N.E. nephrolithometry included tract length, presence of obstruction, NIC, and stone density. It had greater accuracy than any of the individual variables alone. NIC is a well-defined concept reflecting the structural complexity of renal stones. This concept is consistent with our results. Although its variables are exclusively obtained from NCCT and its validation showed a high concordance rate, its variables are limited to calculate objective measurement. In addition, S.T.O.N.E. nephrolithometry did not include surgeon experience, renal abnormality, and previous history of renal surgery. A more objective method to measure each variable is necessary to improve the reproducibility and predictive ability of this system. However, this system is clinically practical for surgical planning, because its variables directly reflect technical difficulties and risk stratification of PCNL, and clinical variability. In our study, this system has more predictive ability than the Guy's score and CROES nomogram.

In our study, the initial SFR and overall SFR were 53.9% and 70.1%, respectively. This is comparable to the 56.9% of the Desai study and the 56.6% in the review by El-Nahas.^26,27 However, our SFR was lower than the 78% reported by Soucy and colleagues and than the 65.8% in the review by Sfoungaristos.^11,25 This could be associated with the large mean stone burden (1365.4 ± 759.1 mm²) and the high proportion of complete staghorn stones (106/217, 48.8%) in our study. Additionally, our patients with partial staghorn stone were included using a stricter definition of staghorn stone.

We noted that the NIC and S.T.O.N.E. nephrolithometry had higher accuracies than the other factors or scoring systems in the AUROC (area under the receiver operating characteristic) analysis (Table 4). A recent study of the three stone-scoring systems for predicting outcomes of PCNL for staghorn stones demonstrated that S.T.O.N.E. nephrolithometry was a significant independent predictor of post-PCNL SFS in a multivariate analysis.²⁵ In this study, S.T.O.N.E. nephrolithometry exhibited a higher AUC (0.743) than Guy's stone score (0.635) and CROES nomogram (0.687). These are consistent with the AUROCs in our study (AUC: 0.746, 0.678, and 0.627, respectively). However, the study included a small sample size (a total of 73 staghorn calculi), was performed by a single surgeon, and had a short follow-up duration. In addition, their study did not demonstrate that preoperative factors other than S.T.O.N.E. nephrolithometry were significantly associated with SFS and complications.

Based on a preoperative NCCT scan, El-Nahas and colleagues investigated the factors affecting the SFR and complication rate of PCNL for the treatment of staghorn stones.²⁷ They demonstrated that complete staghorn stone (OR = 2.2 and p = 0.005) and the presence of secondary caliceal stones (OR = 3.1 and p < 0.001) were independent predictors of SFR and unskilled endourologists (OR = 2.2 and p = 0.017) and positive preoperative urine cultures (OR = 2.1 and p = 0.029) were independent risk factors for the development of complications. These findings for SFR are somewhat analogous to our results. The concept of the NIC is similar to the presence of secondary caliceal stones. In our study, positive preoperative UTIs were added to independent predictors of SFR, and stone burden was an important prognostic factor for the development of complications. Unlike our study, El-Nahas and colleagues excluded renal anomalies, tubeless PCNL, and bilateral simultaneous PCNL for staghorn stones and did not analyze three stone-scoring systems. However, we did not include surgeon experience in the preoperative factors. These discrepancies may have led to the difference in the results of the studies.

Our study was limited by the following factors. First, it was of a retrospective design at a single center. However, we evaluated three stone-scoring systems using a relatively large sample size, an appropriate statistical method, and a strict definition of staghorn stone status. Second, to identify prognostic factors, we acknowledge that other factors such as individual surgeon experience, newer surgical techniques, and instruments may need to be investigated. These factors may need to be incorporated in a future, larger-scale multi-center study. Lastly, all patients were not assessed postoperatively by NCCT but KUB was performed, which may have underestimated the presence of residual fragments. However, we performed a postoperative NCCT scan on selected patients with the suspected residual stones, radiolucent stones, or postoperative complications because of cost problem and radiation exposure. The various external validations of these scoring systems reflect that the predictive tools are important to patient counseling and surgical planning before the PCNL. Although these scoring systems externally validated, in general, PCNL cohorts, we believe that various applications such as respective PCNL types, patient characteristics, and stone type are essential to corroborate the general applicability of these scoring systems. In addition, we need to develop more elaborate scoring systems to evaluate the association between these scoring systems and complications. To our knowledge, this is the first study to evaluate the prognostic factors affecting SFR and complication rates after PCNL for the treatment of staghorn stones and the application of three stone-scoring systems to the prediction of post-PCNL outcomes.

Conclusion

Our results show that multiple involved calices and pre-existent UTI were associated with lower SFR after PCNL for staghorn calculi. A large stone burden was an independent risk factor for the development of complications. S.T.O.N.E. nephrolithotomy was associated with SFR. Larger, prospective studies to assess the SFR and complication rate after PCNL for staghorn stones are warranted.

Footnotes

Acknowledgments

This study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. This study has been approved by our institutional review board (KC15RISI0462). The authors wish to acknowledge the financial support of the Catholic Medical Center Research Foundation made in the program year of 2015. They gratefully thank Sangwon Seo and Yu Jin Baek for their help with data collection. The abstract with preliminary data has been published at the Scientific Program of the 33rd World Congress of Endourology.

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Fernström

, Johansson

. Percutaneous pyelolithotomy. A new extraction technique. Scand J Urol Nephrol, 1976; 10:257–259.

Preminger

, Assimos

, Lingeman

, Nakada

, Pearle

, Wolf

Jr.

Chapter 1: AUA guideline on management of staghorn calculi: Diagnosis and treatment recommendations. J Urol, 2005; 173:1991–2000.

Turk

, Petrik

, Sarica

, et al. EAU guidelines on interventional treatment for urolithiasis. Eur Urol, 2016; 69:475–482.

El-Nahas

, Eraky

, Shokeir

, et al. Long-term results of percutaneous nephrolithotomy for treatment of staghorn stones. BJU Int, 2011; 108:750–754.

El-Nahas

, Shokeir

, El-Assmy

, et al. Post-percutaneous nephrolithotomy extensive hemorrhage: A study of risk factors. J Urol, 2007; 177:576–579.

Akman

, Binbay

, Kezer

, et al. Factors affecting kidney function and stone recurrence rate after percutaneous nephrolithotomy for staghorn calculi: Outcomes of a long-term followup. J Urol, 2012; 187:1656–1661.

Thomas

, Smith

, Hegarty

, Glass

. The Guy's stone score—Grading the complexity of percutaneous nephrolithotomy procedures. Urology, 2011; 78:277–281.

Okhunov

, Friedlander

, George

, et al. S.T.O.N.E. nephrolithometry: Novel surgical classification system for kidney calculi. Urology, 2013; 81:1154–1159.

Smith

, Averch

, Shahrour

, et al. A nephrolithometric nomogram to predict treatment success of percutaneous nephrolithotomy. J Urol, 2013; 190:149–156.

10.

Akman

, Binbay

, Akcay

, et al. Variables that influence operative time during percutaneous nephrolithotomy: An analysis of 1897 cases. J Endourol, 2011; 25:1269–1273.

11.

Soucy

, Ko

, Duvdevani

, Nott

, Denstedt

, Razvi

. Percutaneous nephrolithotomy for staghorn calculi: A single center's experience over 15 years. J Endourol, 2009; 23:1669–1673.

12.

Tiselius

, Andersson

. Stone burden in an average Swedish population of stone formers requiring active stone removal: How can the stone size be estimated in the clinical routine?. Eur Urol, 2003; 43:275–281.

13.

Lingeman

, Coury

, Newman

, et al. Comparison of results and morbidity of percutaneous nephrostolithotomy and extracorporeal shock wave lithotripsy. J Urol, 1987; 138:485–490.

14.

de la Rosette

, Opondo

, Daels

, et al. Categorisation of complications and validation of the Clavien score for percutaneous nephrolithotomy. Eur Urol, 2012; 62:246–255.

15.

Choi

, Kim

, et al. Totally tubeless versus standard percutaneous nephrolithotomy for renal stones: Analysis of clinical outcomes and cost. J Endourol, 2014; 28:1487–1494.

16.

Cho

, Lee

, Kim

, Hwang

, Hong

. Percutaneous nephrolithotomy for complex renal calculi: Is multi-tract approach ok?. Can J Urol, 2012; 19:6360–6365.

17.

Charlson

, Pompei

, Ales

, MacKenzie

. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis, 1987; 40:373–383.

18.

Fernbach

, Maizels

, Conway

. Ultrasound grading of hydronephrosis: Introduction to the system used by the Society for Fetal Urology. Pediatr Radiol, 1993; 23:478–480.

19.

Clayman

, Surya

, Miller

, Castaneda-Zuniga

, Amplatz

, Lange

. Percutaneous nephrolithotomy. An approach to branched and staghorn renal calculi. JAMA, 1983; 250:73–75.

20.

Ingimarsson

, Dagrosa

, Hyams

, Pais

Jr.

External validation of a preoperative renal stone grading system: Reproducibility and inter-rater concordance of the Guy's stone score using preoperative computed tomography and rigorous postoperative stone-free criteria. Urology, 2014; 83:45–49.

21.

Akhavein

, Henriksen

, Syed

, Bird

. Prediction of single procedure success rate using S.T.O.N.E. nephrolithometry surgical classification system with strict criteria for surgical outcome. Urology, 2015; 85:69–73.

22.

Labadie

, Okhunov

, Akhavein

, et al. Evaluation and comparison of urolithiasis scoring systems used in percutaneous kidney stone surgery. J Urol, 2015; 193:154–159.

23.

Noureldin

, Elkoushy

, Andonian

. Which is better? Guy's versus S.T.O.N.E. nephrolithometry scoring systems in predicting stone-free status post-percutaneous nephrolithotomy. World J Urol, 2015; 33:1821–1825.

24.

Bozkurt

, Aydogdu

, Yonguc

, et al. Comparison of guy and clinical research office of the endourological society nephrolithometry scoring systems for predicting stone-free status and complication rates after percutaneous nephrolithotomy: A single center study with 437 cases. J Endourol, 2015; 29:1006–1010.

25.

Sfoungaristos

, Gofrit

, Pode

, Landau

, Duvdevani

. Percutaneous nephrolithotomy for staghorn stones: Which nomogram can better predict postoperative outcomes?. World J Urol, 2015. doi:10.1007/s00345-015-1743-9

26.

Desai

, De Lisa

, Turna

, et al. The clinical research office of the endourological society percutaneous nephrolithotomy global study: Staghorn versus nonstaghorn stones. J Endourol, 2011; 25:1263–1268.

27.

el-Nahas

, Eraky

, Shokeir

, et al. Factors affecting stone-free rate and complications of percutaneous nephrolithotomy for treatment of staghorn stone. Urology, 2012; 79:1236–1241.