Pancreatolithiasis is a pathological condition characterized by stones in the main pancreatic duct (MPD) or its branches during chronic pancreatitis. These lodged stones cause dilation and tissue pressure and result in pain.¹ Previous studies have demonstrated the efficacy and feasibility of endoscopic retrograde cholangiopancreatography (ERCP) and pancreatic extracorporeal shock wave lithotripsy (P-ESWL) in the treatment of pancreatic stones in patients with chronic pancreatitis.^2,3

Many Japanese institutions currently use electromagnetic ESWL systems for pancreatic stones. However, only a few institutions use the latest models of ESWL for treatment. Our center used three electromagnetic lithotripters: Lithostar (Siemens Medical Systems, Erlangen, Germany), Lithoskop (Siemens Medical Systems), and Modulith SLX-F2 (Storz Medical AG, Kreuzlingen, Switzerland). Although the efficacy of electromagnetic ESWL using different commercial electromagnetic lithotripters has been reported in the field of urology, there have been no comparative studies between the several types of commercial electromagnetic lithotripters available for pancreatolithiasis.

Recently, endoscopic treatment was recommended as an adjunctive treatment for pancreatic stones.^1,4 However, there are scarce data on the evaluation of additional endoscopic treatment and P-ESWL risk factors. These factors are crucial for justifying its use in treating pancreatolithiasis.

We retrospectively evaluated the feasibility of using different commercial electromagnetic ESWLs. Moreover, we evaluated the efficacy of additional endoscopic treatment and the predictive factors of P-ESWL in treating pancreatolithiasis.

1. Patients

We conducted a retrospective, observational, single-center, case-series study at Toho University Omori Medical Center, Tokyo, Japan. The study protocol was approved by the Ethics Committee of Toho University Omori Medical Center (approval number: M-20066). Informed consent was obtained from all patients, and the study in accordance with the principles of the Declaration of Helsinki. Details regarding this study were disclosed on our institutional website, and potential participantswere allowed to opt-out. A total of 258 consecutive patients (age, 18 to 83 years) with pancreatolithiasis and chronic pancreatitis who underwent ESWL between June 2003 and November 2019 were initially selected for the study. The exclusion criteria were: cases of endoscopic pancreatic stent placement and pseudocyst drainage without pancreatolithiasis, cases of pancreatic stones <5 mm that could be removed by endoscopic monotherapy treatment. After applying the exclusion criteria, 208 patients were selected (Fig. 1).

Figure 1. Strategy of pancreatic head and body stone treatments.
EHL, electrohydraulic lithotripter; ESWL, extracorporeal shock wave lithotripsy; MPD, main pancreatic duct. *Group A, Modulith SLX-F2 (Storz Medical AG Kreuzlingen); ^†Group B, Lithostar (Siemens Medical Systems), Lithoskop (Siemens Medical Systems) endoscopic pancreatic stenting.

2. Treatment indications

Treatment indications were as follows: (1) symptomatic stones and (2) presence of pancreatic duct stones (≥5 mm) in the Santorini duct or Wirsung duct and MPD dilatation (≥3 mm) detected by diagnostic imaging. We also included patients with asymptomatic pancreatic stones with rapid deterioration of glucose tolerance.

3. Methods

The strategy utilized in this study is illustrated in Fig. 1. All patients underwent radiography, contrast-enhanced computed tomography (CT), and magnetic resonance cholangiopancreatography as required before treatment to distinguish radiolucent stones from radiopaque ones. Before starting P-ESWL for pancreatic ductal stones, all patients underwent ERCP. All endoscopic procedures were performed using a TJF240 or TJF260V duodenoscope (Olympus Co., Tokyo, Japan). Endoscopic pancreatic sphincterotomy was routinely performed prior to ESWL to prevent pancreatitis from MPD obstruction by ESWL stone fragments. After confirming no ERCP-related complications, P-ESWL was commenced to fragment the stones. During this study, we used three current electromagnetic lithotripters for P-ESWL: Lithostar (Siemens Medical Systems) from June 2003 to September 2008; Lithoskop (Siemens Medical Systems) from October 2008 to January 2017; and a newer electromagnetic lithotripter, Modulith SLX-F2 (Storz Medical AG), from February 2017 to April 2019. We defined patients treated with the Modulith SLX-F2 as group A and those treated with the Lithostar and Lithoskop as group B. ESWL for pancreatic stones requires gentle focusing of the shockwaves on the stone followed by fine breathing adjustments (e.g., momentary breath holding) at certain times. Therefore, the use of sedatives during ESWL makes fine breathing adjustments difficult. For this reason, we do not use sedation and epidural anesthesia at our institution. We used diclofenac suppository (25 to 50 mg/body) and pethidine hydrochloride injection (35 to 50 mg/body) as a pretreatment analgesic. In each treatment session, ESWL was performed with Lithostar and Lithoskop on an intensity scale of 1 to 5.5, and Modulith SLX-F2 on an intensity scale of 1.9 to 8. The intensity of the ESWL was adjusted to the patient's individual pain tolerance. P-ESWL was conducted 2 to 3 times/wk with approximately 2,000 shocks/session at 45 to 60 pulses/min during one session. We concluded ESWL when the size of the stone was reduced to approximately 5 mm, which was sufficient for endoscopic treatment. After P-ESWL, combined endoscopic treatment such as endoscopic basket forceps removal or hydraulic lithotripsy were performed as additional treatments. Following stone removal by endoscopy, a final evaluation was made to determine complete stone clearance. These were defined as adjunctive endoscopic therapy (Fig. 2).

Figure 2. Adjunctive treatment for pancreatic stone. (A) A 14.5-mm-sized stone (arrowhead) found in the pancreatic head. (B) Impacted main pancreatic duct stone was identified by endoscopic retrograde pancreatography. (C) X-ray image at the beginning of extracorporeal shock wave lithotripsy (ESWL). (D) X-ray image after 7 sessions of ESWL. (E) Successful removal by adjunctive treatment using basket forceps.

Drainage of the pancreatic duct may be necessary due to significant abdominal pain. In such cases, we tried to perform endoscopic pancreatic stenting at the same time as ESWL. A 10-F S-type plastic stent (Olympus Co.) was placed and stents remained in place for one uninterrupted year and were then exchanged every three months. A dilatation device, such as a Soehendra^TM biliary dilatation catheter (Cook Medical, Winston-Salem, NC, USA), Soehendra^TM stent retriever catheter (Cook Medical), Maxpass^TM (Olympus Co.), or REN^TM (Kaneka Medix, Osaka, Japan), was used before endoscopic pancreatic stenting. In refractory cases of pancreatic stone lithotripsy, surgical treatment was considered.

4. CT protocol

Non-contrast CT was used for preoperative radiological evaluation of the CT scan. The slice thickness was 5 mm. Images showing pancreatic stones with the largest dimensions were analyzed. To record the average density of the stone, we measured attenuation in Hounsfield units (HU) by drawing a region of interest on the stone in the axial CT image.^5,6

5. Outcomes and definitions

The primary aim of this study was to determine ESWL disintegration and stone clearance of adjunctive endoscopic treatments between groups A and B and the stone location. We defined stone location in the MPD as thehead or the body and tail. Successful ESWL disintegration was defined as the pancreatic stone being crushed to 5 mm in size. The additional endoscopic lithotripsy rate was defined as the rate of stone clearance after adjunctive endoscopic treatment. MPD stenosis was defined as: (1) severe stenosis of the MPD duct itself (Fig. 3A), and (2) obstruction of the duct due to filled/impacted stones (Fig. 3B). We also discussed the rates of surgical conversion and details of ESWL complications. We evaluated patient sex, age, etiology of pancreatitis (alcoholic or non-alcoholic), number of stones (single or multiple), maximum stone diameter, stone density (HU), number of ESWL sessions, number of total ESWL shots, number of ERCP sessions, and the presence of MPD stricture as potential predictors. Patient medical records were collected to obtain clinicopathological data. We used the classification proposed by Li et al.⁷ to define major adverse events in ESWL.

Figure 3. Definition of main pancreatic duct (MPD) strictures. (A) Severe MPD stricture shown by endoscopic retrograde pancreatography (arrowhead). (B) Obstruction of MPD due to stone impaction (arrowhead).

6. Statistical analysis

Continuous values of the results are presented as mean±standard deviation or as frequency (percentage). We used the Student t-test and Mann-Whitney U test to compare continuous variables. Categorical variables were analyzed using the chi-square test and Fisher exact test. Statistical significance was set at p<0.05. Factors significantly affecting complete stone removal in univariate analysis were further analyzed using the logistic regression. Odds ratios (OR) and 95% confidence intervals (CI) were calculated. Statistical analysis was performed with JMP 13 (SAS Institute Inc., Cary, NC, USA).

1. Patient characteristics

Table 1 summarizes the baseline characteristics and imaging findings of all patients enrolled in the study. A total of 208 patients fulfilled the inclusion criteria for the study, including 46 patients in group A and 162 patients in group B. Patients in group A were significantly older than those in group B (61.6±12.6 vs 55.1±12.8, p=0.003), and the rate of symptomatic patients was significantly higher in group B (84.8% vs 96.8%, p=0.005). There was no significant difference in imaging findings between the groups. In addition, we examined stone diameter, CT values, single/diffuse stones, and MPD stricture between head and body stones. Pancreatic stones tended to be larger in the pancreatic head stone group (0.7±4.1 mm vs 9.0±3.6 mm, p=0.007). However, there was no significant difference in CT values (1,203±429 HU vs body and tail 1,063±503 HU, p=0.054), single and diffuse stones (42/113 vs 21/32, p=0.118), and MPD stricture (presence/absence) between head and body/tail stones (69/86 vs 25/28, p=0.752).

Table 1 . Baseline Patient Characteristics and Imaging Findings (n=208).

Variable	Group A (n=46)	Group B (n=162)	p-value
Patient characteristics
Male sex†	40 (86.9)	135 (83.3)	0.55
Age, yr*	61.6±12.6	55.1±12.8	0.003‡
Etiology (alcohol)†	38 (82.6)	129 (79.6)	0.65
Symptoms (symptomatic)†	39 (84.8)	157 (96.9)	0.005‡
Imaging findings
No. of stones (single)†	18 (39.1)	45 (27.7)	0.13
Stone location (head)†	32 (69.5)	126 (77.8)	0.25
Stone diameter, mm*	9.7±3.2	10.4±4.2	0.24
Stone density, HU*	1,216±400	1,155±465	0.23
Main pancreatic duct stricture†	20 (43.5)	74 (45.7)	0.79

Data are presented as number (%) or mean±SD. Group A was defined as patients treated with Modulith SLXF2 and group B as patients treated with Lithostar and Lithoskop..

HU, Hounsfield units..

*Mann-Whitney U test; ^†Chi-square test; ^‡p<0.05 was considered statistically significant..

2. Treatment outcomes

Table 2 presents the treatment outcomes for pancreatic head stones and body and tail stones. There were 158 cases of pancreatic head stones and 50 cases of body and tail stones. An average of >10 ESWL sessions were performed in both groups, with an average of 13.3±15.7 sessions for head stones and 11.0±15.7 sessions for body and tail stones. In group A, for pancreatic head stones, the ESWL disintegration rate was 93.7%, significantly higher than that of group B (93.7% vs 69.0%, p=0.004). Additional endoscopic treatment improved the total success rate to 96.8% in group A, which was significantly higher than that in group B (73.0%, p=0.003). The total rate of surgical conversion was 10.1% for pancreatic head stone. Of these, group A had a significantly lower rate of surgical conversion than group B (0% vs 12.7%, p=0.04). For body and tail stones, group A had a significantly higher ESWL disintegration rate than group B (85.7% vs 52.7%, p=0.03). For pancreatic body and tail stones, the surgical conversion rate was 18.0%, with no significant difference between group A and group B (14.2% vs 19.4%, p=0.66).

Table 2 . Outcomes of Treatments.

Variable	Group A	Group B	Total	p-value
Head stones	32	126	158
Power of ESWL*	5.8±1.2	2.2±0.9‖	-	<0.001§
No. of ESWL sessions*	10.4±7.1	14.1±17.1	13.3±15.7	0.06
No. of ESWL total shots*	20,991±14,343	28,193±30,264	26,734±27,897	0.03§
ESWL disintegration (success)†	30 (93.7)	87 (69.0)#	117 (74.1)	0.004§
Additional endoscopic lithotripsy (success)†	31 (96.8)	92 (73.0)	123 (77.8)	0.003§
MPD stricture†	14 (43.8)	57 (45.2)	71 (44.9)	0.87
Transition rates to surgery	0	16 (12.7)	16 (10.1)	0.04
ESWL complications	1 (3.1)	6 (4.8)	7 (4.4)	0.16
Body and tail stones	14	36	50
Power of ESWL*	5.3±0.7	2.3±1.2¶	-	<0.001§
No. of ESWL sessions*	8.0±7.4	12.2±17.8	11.0±15.7	0.7
No. of ESWL total shots*	16,776±15,098	29,185±38,803	25,710±34,170	0.11
ESWL disintegration (success)	12 (85.7)	19 (52.7)**	31 (62.0)	0.03§
Additional endoscopic lithotripsy (success)‡	12 (85.7)	23 (63.9)	35 (70.0)	0.13
MPD stricture‡	6 (42.9)	17 (47.2)	23 (46.0)	0.78
Transition to surgery‡	2 (14.2)	7 (19.4)	9 (18.0)	0.66
ESWL complications	0	0	0	-

Data are presented as mean±SD or number (%). Group A was defined as patients treated with Modulith SLXF2 and group B as patients treated with Lithostar and Lithoskop..

ESWL, extracorporeal shock wave lithotripsy; MPD, main pancreatic duct..

*Mann-Whitney U test; ^†Chi-square test; ^‡Fisher exact test; ^§p<0.05 was considered statistically significant; ^‖Lithostar vs Lithoskop (2.2±0.9 vs 2.2±1.0, p=0.9); ^¶Lithostar vs Lithoskop (2.2±0.8 vs 2.3±1.3, p=0.8); ^#Lithostar vs Lithoskop (83.3% vs 64.5%, p=0.07); **Lithostar vs Lithoskop (58.3% vs 50%, p=0.72)..

Finally, though not shown in Table 2, a comparison of surgical conversion rates between pancreatic head versus body and tail stones showed that the pancreatic head showed no significant difference compared with other stones, although the pancreatic head group tended have a lower rate (head stones 10.1% vs body and tail stones 18.0%, p=0.14; OR,1.94; 95% CI, 0.70 to 5.08).

Table 3 demonstrates the factors that predict ESWL disintegration for pancreatic head stones. Compared with those in group B, pancreatic head stones for which lithotripsy was performed in group A had a significantly higher fragmentation rate (OR, 6.99; 95% CI, 1.56 to 31.33; p<0.01). MPD stricture was a significant factor in multivariate analysis (OR, 2.87; 95% CI, 1.27 to 6.45; p<0.01).

Table 3 . Univariate and Multivariate Analysis of Factors Predicting ESWL for Pancreatic Head Stones.

Variable	Univariate analysis				Multivariate analysis
	ESWL success	ESWL failure	p-value		OR (95% CI)	p-value
Sex			0.71
Male	97 (73.5)	35 (26.5)
Female	20 (76.9)	6 (23.1)
Age, yr	56.9±1.2	53.1±1.9	0.86
Etiology			0.98
Alcohol	94 (74.1)	33 (25.9)
Not alcohol	23 (74.2)	8 (25.8)
Symptoms			0.78
Yes	102 (73.4)	37 (26.6)
No	15 (81.8)	4 (18.2)
P-ESWL devices			0.003*		6.99 (1.56–31.33)	0.01*
Group A	30 (93.7)	2 (6.3)
Group B	87 (69.1)	39 (30.1)
No. of stones (single)			0.43
Single	33 (78.6)	9 (21.4)
Multiple	84 (72.4)	32 (27.6)
Stone diameter			0.36
≥9.55 mm	60 (70.6)	25 (29.4)
<9.55 mm	57 (78.1)	16 (21.9)
Stone density			0.11
≥1,144 HU	60 (68.9)	27 (31.1)
<1,144 HU	57 (80.3)	14 (19.7)
MPD stricture			0.005*		2.87 (1.27–6.45)	0.01*
Presence	60 (84.5)	11 (15.5)
Absence	57 (65.5)	30 (34.5)

Data are presented as number (%) or mean±SD. Group A was defined as patients treated with Modulith SLXF2 and group B as patients treated with Lithostar and Lithoskop..

ESWL, extracorporeal shock wave lithotripsy; OR, odds ratio; CI, confidence interval; P-ESWL, pancreatic ESWL; HU, Hounsfield units; MPD, main pancreatic duct..

*p<0.05 was considered statistically significant..

Table 4 shows factors that predict ESWL disintegration for pancreatic body and tail stones. Both univariate and multivariate analyses showed that ESWL success rates in group A were higher than those in group B (OR, 12.34; 95% CI, 1.43 to 106.29; p<0.02).

Table 4 . Univariate and Multivariate Analysis of Factors Predicting ESWL Disintegration of Pancreatic Body and Tail Stones.

Variable	Univariate analysis				Multivariate analysis
	ESWL success	ESWL failure	p-value		OR (95% CI)	p-value
Sex			0.65
Male	27 (62.7)	16 (37.3)
Female	5 (71.4)	2 (28.5)
Age, yr	59.4±14.6	57.2±12.8	0.58
Etiology			0.76
Alcohol	26 (65.0)	14 (35.0)
Not alcohol	6 (60.0)	4 (40.0)
Symptoms			1.00
Yes	32 (65.3)	17 (34.7)
No	0	1 (100)
P-ESWL devices			0.008*		12.34 (1.43–106.29)	0.02*
Group A	13 (92.8)	1 (7.1)
Group B	19 (52.7)	17 (47.2)
No. of stones (single)			0.79
Single	13 (61.9)	8 (38.1)
Multiple	19 (65.5)	10 (34.5)
Stone diameter			0.22
≥9.55 mm	14 (73.7)	5 (26.3)
<9.55 mm	18 (58.1)	13 (41.9)
Stone density			1.00
≥1,144 HU	13 (65)	7 (35)
<1,144 HU	19 (63.3)	11 (36.7)
MPD stricture			0.45
Presence	16 (69.4)	7 (30.4)
Absence	16 (59.3)	11 (40.7)

Data are presented as number (%) or mean±SD. Group A was defined as patients treated with Modulith SLXF2 and group B as patients treated with Lithostar and Lithoskop..

ESWL, extracorporeal shock wave lithotripsy; OR, odds ratio; CI, confidence interval; P-ESWL, pancreatic ESWL; HU, Hounsfield units; MPD, main pancreatic duct..

*p<0.05 was considered statistically significant..

3. Details of complications

Among 158 patients with pancreatic head stones, seven patients (4.4%) developed pancreatitis after P-ESWL (Table 5), of which four were mild cases, two were moderate, and one was severe. Among the seven patients, pancreatitis occurred in one patient in the first session, two patients in the second session, one patient in the third session, one patient in the ninth session, one patient in the 16th session, and one patient in the 21st session. There was no case of post-ESWL pancreatitis in the pancreatic body or tail. Table 5 shows predictors of pancreatitis based on P-ESWL. We examined sex, alcohol use, lithotripter, number of stones, stone location, stone diameter, stone density, MPD stenosis, common bile duct stenosis, and endoscopic pancreatic stenting or endoscopic nasopancreatic drainage placement. However, none of these factors were significantly associated with P-ESWL pancreatitis.

Table 5 . Univariate Analyses (Fisher Exact Test) of a Prognostic Factor for P-ESWL Pancreatitis.

Variable	P-ESWL pancreatitis		OR (95% CI)	p-value
	Yes	None
Sex			0.88 (0.02–7.65)	1
Male	6	169
Female	1	32
Etiology			0.31 (0.05–2.23)	0.140
Alcohol	3	38
Not alcohol	4	163
Symptoms			0.35 (0.04-17.41)	0.344
Yes	6	190
No	1	11
P-ESWL devices			0.58 (0.01–4.97)	1
Group A	6	156
Group B	1	45
No. of stones			1.09 (0.17–11.73)	1
Single	2	61
Multiple	5	140
Presence of stones			0.00 (0.00–2.18)	0.200
Head	7	151
Body and tail	0	50
Stone diameter			2.56 (0.41–27.53)	0.445
≥9.55 mm	5	99
<9.55 mm	2	102
Stone density
≥1,144 HU	5	101	2.54 (0.40–27.28)	0.445
<1,144 HU	2	100
MPD stricture			1.64 (0.27–11.49)	0.704
Presence	4	90
Absence	3	111
P-ESWL start with EPS or ENPD			0.71 (0.26–11.26)	0.705
Yes	3	110
No	4	91

Group A was defined as patients treated with Modulith SLXF2 and group B as patients treated with Lithostar and Lithoskop..

P-ESWL, pancreatic extracorporeal shock wave lithotripsy; OR, odds ratio; CI, confidence interval; HU, Hounsfield units; MPD, main pancreatic duct; EPS, endoscopic pancreatic stenting; ENPD, endoscopic nasopancreatic drainage..

In addition, two cases of pancreatic head stones and four cases of pancreatic body stones caused basket forceps impaction during basket stone removal. The average diameter of stones with basket forceps impaction was 6.8 mm. In all cases, rescue was performed using a salvage lithotripter (BML 110A-1; Olympus Co.).

This study investigated outcomes after the clearance of pancreatic stones via electromagnetic lithotripters, with adjustments for patients’ clinical characteristics and treatment site, using a single-center database. The results revealed significant differences in stone fragmentation effects between the ESWL models. Therefore, multiple logistic regression analysis of the measured variables was conducted. The results showed that P-ESWL was effective for pancreatic head stones and body and tail stones in group A. Furthermore, additional endoscopic treatment was effective after P-ESWL for pancreatic head stones. MPD stenosis was correlated with pancreatic head stone lithotripsy.

P-ESWL is an excellent option for patients with large stones that are difficult to treat with adjunctive endoscopic treatments. ESWL lithotripters can be classified into electromagnetic generators, spark-gap systems, and piezoelectric generators. According to a Japan-wide survey, electromagnetic generators, spark gap systems, and piezoelectric generators have similar stone fragmentation rates.³ In recent years, electromagnetic systems have been widely used because of their simplicity. Previous studies have reported mixed results on the factors of P-ESWL, such as the location of the pancreatic stone, background of the stone, principle, and timing of ESWL.^6,8,9 P-ESWL disintegration success rate using electromagnetic P-ESWL lithotripters for pancreatic head stones ranges from 10.17% to 81.9%.^6,10,11 A prospective Chinese study reported more fragmented stones among 214 pancreatic head stones.¹² Several studies have indicated that the presence of a single stone and the absence of MPD stricture are prognostic factors for complete stone removal.^13-15 In this study, it was notable that there was a difference in the disintegration rate between the same third-generation electromagnetic devices. We focused on this point and considered the following differences in the mechanisms of the devices: SLX-F2 and Lithoskop/Lithostar are the same electromagnetic models but have different characteristics in terms of the shock wave source, focal pressure, and focal size. First, SLX-F2 uses a cylindrical electromagnetic system to source the shock wave, whereas Lithoskop uses a flat system. A cylindrical system can output approximately 1.5 times more energy than a flat-type electromagnetic system. Second, SLX-F2 has a shock wave focal pressure range of 5 to 160 MPa, which is broader than that of Lithoskop (10 to 75 MPa), allowing it to handle various types of stones, from fragmentable ones to resistant ones. Third, SLXF2 has a compact focus size of 28×6 mm, which minimizes its impact on peripheral organs. These three characteristics may contribute to efficient lithotripsy of pancreatic stones. These considerations may explain the differences in fragmentation rates.

In a Japanese retrospective multicenter study including 555 patients, the complete stone clearance rate was 72.6%. In addition, in a single-center study of more than 1,000 patients, the complete stone clearance rate was 76%.^1,15,16 In this study, patients in group B showed an increase in stone removal rate from 69% to 73% with adjunctive endoscopic treatment. This rate is similar to that reported in previous studies. In contrast, the success rate using SLX-F2 improved from 93.7% to 96.8% with adjunctive endoscopic therapy in cases of diffuse stones requiring frequent ESWL. The present study showed that endoscopic adjuvant therapy was more effective using a newer third-generation electromagnetic lithotripter.

We examined the relationship between the patient characteristics and the outcome data. In this study, there was a significant difference between group A (61.6±12.6 years) and group B (55.1±12.8 years) in terms of age group. One reason may be that the population of group A was small because of the single center retrospective design; however, we could not find a clearer reason than this. The large number of symptomatic patients in group B was thought to be due to the recent increase in referrals of patients with asymptomatic pancreatic stone to our institution. We found that there was no significant difference in stone removal in asymptomatic patients. In addition, we also examined Lithostar and Lithoskop from the same Siemens company set up in group B. In pancreatic head stones, the ESWL disintegration rates for Lithostar and Lithoskop were 83.3% and 64.2%, respectively, showing no difference between the two groups (p=0.07). ESWL disintegration rates in body/tail stones were 58.3% for Lithostar and 50% for Lithoskop with no difference between the two groups (p=0.72). Based on the above, no issue was identified with setting the two lithotripters of the same Siemens company as group B, as these were considered to have the same crushing performance.

Next, we discussed the number of ESWL sessions/shots and the success rate of stone removal in this study. It is noteworthy that the number of ESWL sessions/shots was more frequent and the success rate of stone removal was higher than in other reports. In previous reports, the number of electromagnetic P-ESWL lithotripter sessions required for clearance was mainly within five sessions or 10,000 shots.^{11,13,14,16-20} In contrast, our results for P-ESWL sessions for pancreatic head stones were 10.4±71 sessions and 20,990±114,343 shots, which were much higher than those previously reported. Only one report mentioned the number and frequency of P-ESWLs. The highest number of total P-ESWL shots and sessions was 13 (2 to 74) and 22,011 (1,700 to 150,900), respectively.¹³ According to the same report, the total number of sessions and shots of P-ESWL increased due to the lower energy level of shockwaves, suggesting the importance of pretreatments such as analgesics and sedatives. There are several reasons for the high number of P-ESWL sessions in this study. First, the rate of MPD stenosis was higher than that in other previously reported cases.^13-21 According to previous reports, MPD stenosis reduces pancreatic stone removal rates.¹³ Multivariate analysis showed that the number of P-ESWL sessions was higher in patients with concomitant MPD stenosis in the pancreatic head stone group. The results suggest that frequent ESWL should be considered, particularly in cases of pancreatic head stones with MPD strictures. Second, the number of shots administered in one P-ESWL session was a factor to consider. While most other reports used a maximum of 5,000 shots per session, we limited the number of shots to 2,000 per session. This was based on the number of shots recommended by each company based on the burden on the pancreatic duct and parenchyma. According to reports of patients treated with a maximum of 5,000 shots per session, stones were safely treated with an average total of 10,000 shots or with an average of five sessions or less.^{6,11,12,22-24} This suggests that the maximum number of shots per session should be considered in the future. Third, stone density on non-contrast CT was an important factor. A recent retrospective study reported a baseline mean stone density of 822.6±352.0 HU, and that P-ESWL resulted in significant complete stone removal with a density <820.5 HU.⁶ In our study, the mean CT values were higher than those previously reported, which may have affected the number of fragments. Fourth, we discussed the power of ESWL. In Table 2, there was no difference in ESWL power values between Lithostar and Lithoskop in group B. However, there was a significant difference between groups A and B. The values (level) of the intensity of each device in the Siemens and Storz crushers are not constant (although they should be). It would be desirable to compare the total energy for each device and treatment rather than the maximum power of ESWL. However, there are no standardized data available. The present data were based on a unified treatment strategy implemented at a single center institution. The analgesic dosage was also standardized for each treatment. The intensity of the ESWL was adjusted to the patient’s individual pain tolerance, suggesting that influence of these factors may be minimal. Therefore, it is considered that there was no difference in intensity between the two groups. Fifth, we discussed whether the expertise in ESWL of individual physicians may affect treatment outcomes. Regarding the learning curve in urologic ESWL, it has been noted that ongoing supervision and mentoring may be effective during the first year.²⁵ However, there is no medical literature available on learning curves in the field of P-ESWL. The most important point in P-ESWL is the stone focusing, which is determined by several physicians while checking the X-rays. After setting the stone focusing, the position and intensity of the fragmentation can be easily adjusted by checking the X-rays. Therefore, there is almost no learning curve as in endoscopic or surgical procedures, and we consider that the outcomes of P-ESWL at our institution are not affected by differences in physicians’ experience. For these reasons, SLX-F2 is considered superior to the other two lithotripters.

ESGE recommends ESWL for MPD stones larger than 5 mm located in the head/body of the pancreas. However, there are no detailed reports on the results of endoscopic therapy and/or ESWL in pancreatic tail stones.⁴ Body/tail stones showed lower ESWL disintegration rates than head stones, despite the smaller diameter of the stones. Additional endoscopic lithotripsy did not improve the outcomes, and the result was less favorable than that of the pancreatic head stones group. One of the reasons is the distance between the stone and the main papilla. Longer MPD distances affect the fragmentation of ESWL in stone removal. Furthermore, cases of complicating MPD stricture make it difficult to manipulate the basket forceps, which leads to treatment failure. However, no specific factor was found to explain the difference in ESWL disintegration rates of body tail stones compared to head stones in the ESWL group. Further investigation is needed to clarify ESWL disintegration and endoscopic lithotomy of body/tail stones.

We also examined treatment-related complications. Although there are widespread reports of complications after P-ESWL, such as renal hematoma, intestinal perforation, splenic rupture, and severe acute pancreatitis, these incidents are rare and limited.²⁶ Fortunately, we did not observe adverse events, such as renal hematoma, intestinal perforation, or splenic rupture, although we encountered basket impaction and ESWL.

We encountered basket impaction in six patient cases. Fortunately, we were able rescue patients using a salvage lithotripter. We also found that the average diameter of the stones was 6.8 mm (i.e., more than 5 mm), and we removed the stones using basket forceps. When removing stones that are approximately 5 mm in size using basket forceps, we considered the possibility of basket impaction and continuing ESWL.

The incidence of post-ESWL pancreatitis in our study was 4.4% (only in pancreatic head stones), which was similar to previous reports.^7,27 However, in one case in group A, pancreatic head stones complicating pancreatic pseudocysts led to severe pancreatitis and infection after the second ESWL session. In the present study, we did not detect significant factors in the incidence of pancreatitis. Although not shown in Table 5, pancreatitis tended to develop in the first three sessions of P-ESWL. This suggests that the incidence of pancreatitis in early sessions should be considered in choosing P-ESWL for pancreatic head stones.

Finally, we evaluated the predictors of stone removal treatment outcomes. This study showed that P-ESWL and MPD strictures are the two most important factors behind stone fragmentation. Previous studies have also shown that stones located in the pancreatic head respond better than those in the body or tail.⁴ However, the present study using SLX-F2 showed a significantly higher fragmentation rate for combined endoscopic therapy than other reports. As noted above, SLX-F2 includes technical refinements, such as a shock wave source, focal pressure, and focal size. This is considered to have many advantages for patients undergoing pancreatic lithotripsy treatment.

Second, we discussed a case of concurrent MPD stenosis. Previous studies have shown that complicating MPD strictures with pancreatic stones have a suboptimal outcome.⁴ Spontaneous stone clearance cannot be expected in cases of MPD stenosis. It is important to perform pancreatic duct dilatation procedures to facilitate stone drainage and endoscopic treatment. In this study, multivariate analysis showed an unexpected OR of 2.87 for patients with MPD stricture. When defining MPD stricture, three situations need to be considered: pure stenosis, obstruction of the MPD due to stones (Fig. 3), and the coexistence of both factors. In addition, it is important to consider background factors such as stone diameter and CT values. It is clinically difficult to clearly distinguish true MPD stricture until stone removal is completed. Therefore, we broadly defined “MPD stricture” as cases in which a pancreatic stent was placed for pure stricture and MPD obstruction due to stones before P-ESWL. The following procedural factors are also considered to affect the stricture of MPD: (1) whether the MPD on the peripheral side of the stenosis can be contrast-enhanced by endoscopic retrograde pancreatography; (2) whether the guidewire can pass through the peripheral side of the stenosis; (3) whether the stenosis dilation procedure was successful; and (4) whether the pancreatic duct stent placement was successful or not. However, we could not obtain detailed data on these four points in this study. Because the data presented in this study are limited to a retrospective single-center study, further multicenter analysis will be necessary in the future to take into account these background and procedural factors.

This study focused on an electromagnetic model disintegration device for P-ESWL from Japan, but certain limitations were experienced. First, as mentioned above, this was a single-center, retrospective study. Second, the number of patients for whom SLX-F2 was used was small, and more cases are needed to study the disintegration-predictive factors in detail. In addition, there are no reports on the use of this lithotripter for P-ESWL; therefore, it was not possible to make comparisons with the same device. Furthermore, we focused only on the disintegration effect and did not include the pain relief effect of the models.

In conclusion, our study suggests that SLX-F2 is more effective than other lithotripters in treating pancreatolithiasis. Improved ESWL lithotripters are considered to have many advantages for patients undergoing pancreatic lithotripsy treatment. There are still many facilities in Japan that do not utilize ESWL for patients with pancreatic stones. For this reason, it is necessary to increase the number of institutions that can perform ESWL in Japan.

The authors thank the paramedical, medical, and endoscopy staff at the Division of Gastroenterology and Hepatology, Department of Internal Medicine, Toho University Omori Medical Center, for making this study possible.

No potential conflict of interest relevant to this article was reported.

Study concept and design: K.I., Y.I. Data acquisition: K.I., K.W., Y. Kimura, Y.Y., K.Y. Data analysis and interpretation: K.I., S.I., K.T., S.H., Y. Kishimoto. Drafting of the manuscript: K.I. Critical revision of the manuscript for important intellectual content: K.I., N.O., T.M. Statistical analysis: K.I. Administrative, technical, or material support; study supervision: K.T., S.I., K.W., Y. Kimura, Y.Y., K.Y., S.H., Y. Kishimoto. Study supervision: Y.I., T.M. Approval of final manuscript: all authors.