Low-frequency extracorporeal shock wave lithotripsy improves renal pelvic stone disintegration in a pig model


Rolf Gillitzer, Johannes-Gutenberg University, Department of Urology, Langenbeckstrasse 1 Mainz 55131, Germany.
e-mail: gillitze@mail.uni-mainz.de



To compare disintegration rates for renal stones treated by 60 vs 120 shock waves (SW)/min at the same energy settings, using standardized validated artificial stones in a pig model.


Gypsum artificial stones (13 × 6 mm) were inserted into the renal pelvis on either side of 12 anaesthetized pigs by open surgery. Extracorporeal SW lithotripsy (ESWL) was applied using a new electromagnetic lithotripter (Lithoskop®, Siemens AG Healthcare, Munich, Germany) at 60 and 120 SW/min; 3000 SW were applied to each kidney with the same energy settings. Stone fragments were collected after nephrectomy, passed through calibrated test sieves, and weighed. Fragment size categories were stratified according to the sieve hole size as set by the manufacturer. Fragments of ≤4.75 mm were defined as capable of spontaneous passage. For each pig the number of stone fragments of the respective size categories was counted and weighed. The results were analysed statistically using the Mann–Whitney U-test.


For fragments of >4.75 mm, the median (range) fragment counts were 0 (0–1) for 60 and 1 (0–3) for 120 SW/min (P = 0.006). For small fragments of 2.0–2.8 mm, the median fragment counts were 15 (4–24) for 60 and 10 (2–25) for 120 SW/min (P = 0.033); for fragments of 1.0–2.0 mm the respective values were 42.5 (9–81) and 21.5 (6–56) (P = 0.004). Of the total stone fragment mass in the 60 and 120 SW/min groups, 4.34% and 31.31% were >4.75 mm. There was complete disintegration yielding fragments capable of spontaneous passage in 10 of 12 renal units at 60 and in three of 12 renal units at 120 SW/min. The mean treatment time was 55.4 min for therapy at 60 and 34.3 min for therapy at 120 SW/min (P = 0.001). One parenchymal haematoma of 15 × 10 mm developed in the 60 SW/min group and another of 20 × 10 mm developed in the 120 SW/min group.


ESWL fragmentation with equal energy application yields significantly smaller fragments at 60 than at 120 SW/min. The theoretical stone passage rate could therefore be ≈80% for 60 vs 25% for 120 SW/min ESWL. Renal haematoma formation was comparable in both groups.


shock wave(s).


The higher efficacy of low frequency ESWL has been documented in a few in-vitro studies and in several prospective randomized clinical studies. However, fragmentation and stone-free rates were mostly judged by radiographic assessments. A previous animal study showed better disintegration rates at 30 than at 120 shock waves (SW)/min [1]. However, the use of very low frequencies is time-consuming and probably does not realistically reflect the clinical scenario. We compared disintegration rates of validated gypsum stones at 60 vs 120 SW/min, using identical energy settings and a new electromagnetic lithotripter in a pig model.


Radio-opaque cylindrical gypsum artificial stones (BegoStone®, Bego USA, Smithfield, RI, USA) of 13 × 6 mm each were inserted through a pyelotomy incision into the renal pelvis on either side of 12 juvenile farm pigs (mean body weight 37.7 kg, range 33–45) by open surgery. BegoStone artificial stones are comparable to hard kidney stones such as calcium oxalate monohydrate; they are abrasion-resistant and do not soften in urine [2]. One stone was placed in each kidney. JJ ureteric stents (16 cm; 4.7 F) were inserted through the same incision into each reno-ureteric unit to avoid stone particle migration into the ureter during ESWL, and to prevent hydronephrosis. Anaesthesia was maintained with thiopental 10 mg/kg i.v. plus fentanyl 1 µg/kg i.v. The pigs were transported from the operating room to the lithotripsy unit, placed supine and secured onto the ESWL table. The skin was completely shaved at the coupling area. The stone was targeted with biplanar fluoroscopy. Lithotripsy was applied using a third-generation electromagnetic lithotripter (Lithoskop®, Siemens AG, Munich, Germany) at a rate of 60 SW/min for one kidney and 120 SW/min for the contralateral kidney. The side to be treated first (right/left) and the frequency rate (60 vs 120) were alternated every other case. Energy increase was continuous until an energy level of 3.0 (0.02 J/pulse) was reached; 3000 shocks were applied to each kidney, mimicking ESWL settings for kidney stone treatment in clinical practice. Stone target monitoring was achieved with repeated fluoroscopy every 20–50 shocks. The blood pressure was kept stable throughout the procedure. In none of the cases was repositioning of the pig on the ESWL table necessary. Immediately after ESWL, the pigs were transferred to the operating room and through a laparotomy both reno-ureteric units were removed, with primary ligature and transection of the ureter at the vesico-ureteric junction, and the pigs were then killed humanely. Each ureter was opened retrogradely up to the renal pelvis and the kidney was opened longitudinally. The kidneys were washed with saline and stone fragments were collected. The collecting system was thoroughly inspected to ascertain complete stone collection. Tissue debris was removed from the fragments with forceps. The saline fluid collection containing the stone particles was passed through a 0.5 × 0.5 mm mesh and decanted. The renal collecting system was measured and the kidneys inspected for haematoma formation. The collected stone fragments were left to dry overnight, then passed through calibrated sieves and weighed. Fragment size categories were stratified according to the sieve hole size as set by the manufacturer, and were >4.75, 4.0, 2.8, 2.0, and 1.0 mm. Fragments of ≤4.75 mm were defined as capable of spontaneous passage.

Using the McNemar test for paired data and a two-sided significance level of 5%, case numbers were calculated assuming a 60% fraction of successful treatments (stone size ≤4.75 mm) and a fraction of 70% of discordant observations, given a statistical power of 80%. The calculated sample number for the study was 12. The results from the two groups were analysed statistically using the Mann–Whitney U-test, with statistically significant differences considered at a one-sided P < 0.05.


The ESWL energy settings and kidney measurements are shown in Table 1. Generally, fluoroscopy during ESWL indicated a better disintegration rate at 60 than at 120 SW/min; this difference was visually evident at the time of fragment collection (Figs 1,2).

Table 1.  ESWL settings and kidney measurements
Pig no. weight, kg/sexKidneySW/minTreatment time, minTotal energy, JDimension, mmRemarks
Kidney lengthKidney widthCalyx diameterPelvis lengthPelvis widthUreter diameter
  1. ph, parenchymal haematoma.

1/37/fleft606062 45811055 819124 
2/40/mleft605562 47710555 815104 
3/37/fleft605462 49010555 61510415 × 10 mm ph
4/40/mright605462 381110501020154 
5/40/mright605262 405110501020153 
6/38/mright605362 010120551020154 
7/45/mleft605362 489110501025155 
8/35/mleft605562 30810852 920114 
9/35/mright605262 489112551017115 
10/33/wleft605576 72511455 921155 
11/35/mright606462 15411053 925134 
12/37/wright605862 448117561021155 
Mean37.7 55.463 569110.953.4 9.119.813.14.3 
1/37/fright1202862 45011155 920124 
2/40/mright1202662 45710558 515103 
3/37/fright1205162 45810050 514104 
4/40/mleft1203162 45211055 830154 
5/40/mleft1204462 357115501119145 
6/38/mleft1203062 48112050102515420 × 10 mm ph
7/45/mright1203162 453115451019124 
8/35/mright1202962 14210952 918125 
9/35/mleft1202862 472112541018104 
10/33/fright1204162 45811356 819154 
11/35/mleft1204062 337108531023143 
12/37/fleft1203362 465116551122145 
Mean37.7 34.362 415111.252.8 8.820.212.84.1 
Figure 1.

A, Complete disintegration of left renal stone (red arrow) with 3000 SWs at 60 SW/min. Notice the untreated right renal stone (blue arrow); B, Partial disintegration of left renal stone with 3000 SW at 120 SW/min (red arrow).

Figure 2.

A, A kidney treated with 3000 SW at 60 SW/min (same pig as in Fig. 1A). Notice the parenchymal haematoma formation (arrow): B, Partial disintegration of a pelvic stone treated with 3000 SW at 120 SW/min.

The median (range) fragment counts for each size category are shown in Table 2; 4.34% of the stone mass treated with 60 and 31.31% of that treated with 120SW/min were >4.75 mm (Fig. 3). There was complete disintegration yielding fragments capable of spontaneous passage in 10 of 12 renal units at 60 and in three of 12 renal units at 120 SW/min. The mean treatment time was 55.4 min for therapy at 60 and 34.3 min for therapy at 120 SW/min (P < 0.001).

Table 2.  The median (range) fragment counts for the different size categories at 60 and 120 SW/min
Sizecategory, mmMedian (range) countP
≥4.75 0 (0–1) 1.0 (0–3) 0.006
4.0–4.75 0.0 (0–3) 0.5 (0–2) 0.05
2.8–4.0 6.0 (2–12) 4.0 (3–11)>0.05
2.0–2.815.0 (4–24)10.0 (2–25) 0.033
1.0–2.042.5 (9–81)21.5 (6–56) 0.004
Figure 3.

The mean relative percentage of stone mass for fragments of different size categories.

A parenchymal haematoma of 15 × 10 mm developed in the 60 SW/min group and another of 20 × 10 mm in the 120 SW/min group.


The introduction of ESWL in the 1980s revolutionized stone treatment. Currently, ESWL remains the first-line therapy for renal stones of <20 mm [3]. To improve stone comminution, several technical modifications have been introduced, such as the use of various energy sources, focusing devices, and shock-delivery modes [4].

In previous in-vitro studies better fragmentation of stones was recognized by using lower SW frequencies [5–7]. Greenstein and Matzkin [6] compared SW delivery at rates of 30, 60, 90, 120, and 150 SW/min for treating 118 standardized stones with a mean diameter of 9.5 mm. They used an electrohydraulic lithotripter with energy intensities of 15, 20 and 22.5 kV, and found the most effective fragmentation rate was 60 SW/min. There was no statistically different fragmentation rate between 30 and 60 SW/min at all energy settings. As expected, stones had overall better fragmentation at higher energy levels [6]. The frequency effect had also been reported for spark-gap lithotripsy at comparable energy output, thus explaining the difference in stone fragmentation by mechanisms of stone disintegration and not as a function of the electrode [7]. Hard stones such as weddellite, carbapatite, whitlockite, and uric acid were shown in in-vitro studies to disintegrate better at lower frequencies of 1.25 and 2.50 SW/s than at 5 and 10 SW/s. By comparison, friable stones such as struvite fragmented well at either frequency setting [5].

In a live animal model, Paterson et al.[1] first reported better disintegration rates for low-frequency ESWL, comparing 30 and 120 SW/min. They selected these frequency settings because their previous in-vitro studies showed no difference in disintegration rates for 60 vs 120 SW/min [1].

Different from most previous in vitro and animal studies, which have used predominantly electrohydraulic and piezoelectric SW sources [8], the present study allowed us to compare two different ESWL frequencies at comparable energy settings in a new-generation electromagnetic lithotripter. Using validated standardized stones in the present pig model allowed a more precise evaluation of fragmentation results than obtained from clinical studies, and controls for confounding variables such as stone composition, location and size. Furthermore, no patient- and operator-selection bias needs to be considered, nor the influence of varying methods of anaesthesia. Maloney et al.[9] successfully used the same pig model and stones previously, to investigate the influence of an increasing lithotripter output voltage on stone comminution.

Several clinical studies have shown better stone disintegration with low frequency ESWL [8,10–13]. Madbouly et al.[10] examined different SW rates for treating renal and ureteric stones in a prospective randomized study. The success rate was higher in the 60 than in the 120 SW/min group, and the former group required significantly fewer total SW [10]. Pace et al.[11] examined fragmentation differences for renal stones in a prospective randomized double-blind study comprising 220 patients, comparing 60 vs 120 SW/min. Success was defined as being stone-free or left with asymptomatic residual fragments of <5 mm at 3 months after treatment. They reported a statistically significant overall better fragmentation rate of 75% for 60, vs 61% for 120 SW/min. The difference was even greater for larger stones with a cross-sectional area of ≥100 mm2. However, stone sizes in that study, as measured on X-ray films, varied considerably, with a mean (sd) initial cross-sectional stone area of 82.4 (62.5) mm2. Stones with a cross-sectional area of <100 mm2 had similar success rates at low- and high-frequency ESWL. Another prospective randomized study showed no differences in disintegration rates for patients treated with 60 or 120 SW/min, using an electromagnetic lithotripter to treat solitary renal calculi [8]. However, stones were smaller in that study than in the previous ones that had shown improved disintegration with lower frequencies. In that particular study, which was performed as an outpatient treatment with no anaesthesia, ESWL was stopped early if pain or nausea were intolerable under pain medication, or if the stone could not be visualized. Others have also been unable to show different disintegration outcomes for smaller stones (30–90 mm2 stone surface area) using different frequencies [13]. However, in the present experiment treating standardized stones with a cross-sectional area of 78 mm2 in a few pigs, there was a statistically significant difference of stone disintegration between the low- and high-frequency SW groups. Thus, the beneficial effect of lower frequency rates for stone comminution might not be limited to larger stones, as has been suggested by clinical studies [14,15].

The physical explanation for improved comminution by lower SW frequencies has still not been completely elucidated. It has been postulated that with an increase in SW frequency, cavitation bubble formation from the previous SW would increasingly dissipate the incoming new SW and reduce its shearing and tearing forces. However, cavitation bubbles dissolve within milliseconds, which would leave any negative effect on the next wave unlikely [16]. Another possible explanation is that with increasing frequency, interference between the SW propagated in the stone and the following one are multiplied, and might reduce pressure effects [5]. On the other hand, cavitation bubbles might implode at the stone surface and promote enlargement of cracks at the stone surface [17,18].

In the present experimental model, open pyelotomy must be considered as allowing air to enter into the collecting system and thus reduce SW penetration. Air which has been retrogradely injected into the collecting system through a ureteric catheter remains visible on fluoroscopy for up to 2 h after injection [1]. However, even if there was a negative effect of air on stone comminution, we altered the sequence of which side was treated first and of the frequency settings from pig to pig. Thus, any possible presence of air within the collecting system would have affected disintegration rates in both frequency groups similarly.

We acknowledge that our threshold of 4.75 mm for spontaneous stone passage was set arbitrarily. Spontaneous stone passage can be expected in up to 80% of patients with stones of ≤4 mm, and there is still a chance of passage of stones with diameters up to 7 mm [19]. For a fragment size of ≤4.0 mm, in the present study 58% of renal units in the 60 SW/min group, and only 17% in the 120 SW/min group, would be exclusively left with fragments capable of spontaneous passage.

Our results add new evidence that low-frequency ESWL improves stone disintegration in a pig model with a cross-sectional stone surface size of <100 mm2, using a new electromagnetic lithotripter. Comparison of frequencies of 60 and 120 SW/min reflect realistic clinical alternatives with the result of better fragmentation of stones at 60 SW/min.


We thank Prof. Achim Tresch (Department of Biomedicine and Biostatistics, Johannes Gutenberg University, Mainz) for performing the statistical analysis.


None declared.