Introduction

Since 1980, extracorporeal shock wave lithotripsy (SWL) has become the first-line treatment option for <2 cm kidney stones in adults [1]. In general, factors predicting the success rate of SWL can be divided into stone-related and patient-related factors. Patient-related factors are age, sex, stone laterality (right or left), body surface area and body mass index (BMI). Stone-related factors are stone size, intrarenal stone location, skin-to-stone distance (SSD) and stone fragility [2].

In the last decade, noncontrast computed tomography (NCCT) has become the first choice for diagnosis of kidney stones with high sensitivity and specifity [3]. NCCT enables determination of urinary stone density with the probability of 0.5 % difference ranges [4]. In addition, several studies established that the Hounsfield units (HU) and Hounsfield density (HD) of stones determined by NCCT were highly correlated in terms of fragility with SWL [5–8].

Significance of the aforementioned parameters have not been evaluated in our country, so we aimed to evaluate the predictivity of various HU values in addition to HUave and also SSD for SWL success in a Turkish patient group since racial differences may have an effect on outcomes.

Materials and methods

After an approval obtained from the Local Ethics Committee, we retrospectively evaluated patient records who have undergone SWL for upper urinary tract stone of 5–20 mm diameter from January 2008 to June 2012. Only patients with documented radiographic evaluation of the urinary tract by NCCT before SWL were included. The exclusion criteria included stones of <5 or >20 mm in diameter, staghorn stones, obstructive and multiple stones, stones requiring drainage, patients with solitary kidney and patients with congenital urinary tract anomalies. All patients with SWL failures underwent ureterorenoscopic treatment for ureteral stones and mini-percutaneous nephrolithotomy or retrograde intrarenal surgery for kidney stones.

NCCT images using 2-mm sections with the liver’s dome as cranial border and pubis joint as caudal border at 100 mA 120 kV (Brilliance 64, Philips®, Best, the Netherlands) were taken. HU values were measured in the largest diameter of the stone (longitudinal or transverse) with bone window and large magnification. After serial measurements of the highest HU value (HUmax) and lowest HU value (HUmin), HU values were calculated as the average (HUave) of these two values. The HD was calculated as the HUmax divided by stone size [9]. The average SSD for kidney stones was determined as previously described by Pareek et al. [10] and briefly calculated by using measurements in the coronal plane, sagittal plane and at 45° between these two planes from the center of the stone to the skin on NCCT. All measurements were calculated by radiologists. All treatments were performed with Elmed Lithotripsy® (Elmed, Ankara, Turkey) Systems. Stone fragmentation was monitored by fluoroscopy throughout the procedure. Processing began with 13 kV and frequency of 60/min, and sessions were completed with maximum 20 kV and 2000 shocks. After each session, location and fragmentation of stones were followed with kidney-ureter-bladder (KUB) plain abdominal radiograph. Time interval between the sessions was 1 week, and the SWL success was evaluated by radiographic imaging (KUB, intravenous urography or NCCT) 6 weeks after the last session and defined as stone-free. Efficiency quotient (EQ) was calculated as originally described [11]. Then, aforementioned parameters were compared between stone-free patients (SF group) and patients who have residual fragments after last SWL session (RF group). Thereafter, stones with >750 and ≤750 HUave values were compared regarding SWL success [6].

Statistical analysis

Data were analyzed using the Statistical Package for Social Sciences, version 15.0 (SPSS, Chicago, Ill) software program. Independent Student’s t test analysis was used for comparing demographic and radiologic parameters between SF and RF groups. Data were given as mean ± SD. Bivariate correlations between possible predictive parameters for SWL success were compared by using Pearson’s correlation test. Statistical significance was defined as p < 0.05.

Results

A total of 254 patients (157 men and 97 women) were included. Mean age was 51.0 ± 14.6 (18–87) years, and mean stone size was 10.9 ± 3.7 mm for the whole group. Stone localization and stone side and average SWL sessions are given in Table 1. Considering the stone localization, there were 113 patients with kidney stones and 141 patients with ureteral stones. Stone localizations were 21 lower pole (8.3 %), 10 middle calyx (3.9 %), 8 upper pole (3.1 %), 74 renal pelvis (29.2 %), 58 proximal ureter (22.8 %), 35 mid-ureter (13.8 %) and 48 distal ureter (18.9 %).

Table 1 Patient and stone characteristics of the study population. Data regarding SWL sessions were also provided
Full size table

When we compare patients in SF and RF groups, we determined that factors affecting SWL success for both kidney and ureter stones were stone diameter, HUmax, HUmin and HUave (Tables 2, 3). For patients with kidney stones, increasing SSD and BMI was also another factor decreasing SWL success rates. In correlation analysis, we see that all these factors were correlated with SWL success (Table 4). Further analysis regarding the cutoff HUave value 750 HU, we found that SWL success for kidney stones was 50 and 20.2 % for ≤750 HU and >750 HU groups, respectively (p < 0.05). Stone diameters did not differ between these two groups (12.2 ± 4.6 mm vs. 12.3 ± 4.2 mm, p = 0.544). For ureteral stones, SWL success were 75.6 and 42.1 % for stones ≤750 HU and >750 HU HUave values, respectively (p < 0.05). However, there was also significant difference in terms of stone diameter between these two groups (8.1 ± 3.5 mm for ≤750 HU vs. 9.8 ± 3.4 mm for >750 HU, p < 0.05).

Table 2 Comparison of possible success predictors between SF and RF group in kidney stones
Full size table
Table 3 Comparison of possible success predictors between SF and RF group in ureteral stones
Full size table
Table 4 Correlation analysis of possible success predictive parameters for SWL
Full size table

Discussion

SWL has gained much popularity for the last three decades as an alternative to surgical management with high success rates. However, there are some factors affecting the success of SWL which are the stone size, stone density and configuration previously evaluated with KUB abdominal plain radiograms. Several reports indicated that stone density lower than bone on KUB is an augmenting factor for SWL success [12–16]. However, effective measurement of the size of stone and evaluation of density may not be done truly sometimes because of bowel gases [5]. Today, most patients with nephrolithiasis are now evaluated with NCCT instead of KUB and IVU [6]. NCCT is a noninvasive radiologic imaging modality, which determines the stone density and anatomical relationships better than KUB. Predictors of SWL success provided by NCCT are mostly stone density, stone location and SSD [6, 10]. As mentioned previously, NCCT is more sensitive than KUB abdominal plain graphies for determining the stone density [4, 17].

Joseph et al. reported that higher HU values needed increasing number of pulses for stone clearance with SWL and success rates are lower for high HU values. They proposed surgical treatment with a cutoff stone density value denser than 950 HU according to their results [5]. In another study considering the SWL session number for stone clearance, Gupta et al. [6] demonstrated that 80 % of stones ≤750 HU need <3 SWL sessions with a stone-free rate of 88 %, whereas 72 % of stones >750 HU need more than three sessions with a stone-free rate of 60 %. When we compare stone clearance rates for stones ≤750 HUave and >750 HUave in our study, it is clearly seen that denser stones are more resistant to SWL. Relatively, low success rates in our study when compared with the aforementioned study may be attributed to definition of different success criteria. Ouzaid et al. [18] defined 970 HU as cutoff HU value and gave that clearance rates for stones <970 HU was 96 %, whereas 38 % stone-free rates was documented for stones ≥970 HU in the same study.

Factors predicting SWL success rates in our study are presented as stone size, HUmin, HUmax and HUave for kidney and ureteral stones; additionally, SSD and BMI for kidney stones. A cutoff value of 10 cm was proposed for SSD in terms of predicting SWL success by Pareek et al. [10]. Another study presented SSD as an important parameter for predicting SWL success rates for calyxeal stones [2]. In our study, there was a significant difference between SSD values; RF group have significantly higher SSD values compared to SF group.

Major drawbacks of the present study include retrospective analysis of nonrandomized patients and limited number of patients; lack of metabolic workup, stone analysis of most patients and auxiliary re-treatment rates for renal and ureteral stones are also missing. Evaluation of patients with KUB after SWL may also have an adverse effect on outcomes since sensitivity and specificity of these methods vary. Further prospective-randomized trials may enlighten these findings.

In conclusion, as a gold standard diagnostic method for the urinary tract stone disease, NCCT also gives valuable information in terms of prediction of success before SWL. Definitive stone size measurement and stone location, enabling calculation of HUave value as stone density parameter and SSD measurements, are well-documented success predictors of NCCT. With this study, we provided that all these parameters are also significant for Turkish patient population. We recommend NCCT evaluation and measurement of these parameters for all patients with urinary tract stone disease in order to predict SWL success.