The single-operator cholangiopancreatoscopy (SOCP) with the SpyGlass™ system is a endoscopy technique whose use has grown exponentially in recent years. The aims of this study were to evaluate the efficacy and safety of SOCP with SpyGlass™ and determine the factors related to the onset of adverse events (AEs).
Patients and methodsRetrospective study at a single tertiary institution with inclusion of all consecutive patients undergoing SOCP with SpyGlass™ from February-2009 to December-2021. No exclusion criteria were considered. A descriptive statistical analysis was performed. The factors associated with the existence of AE were analyzed using Chi-square and Student's t-test.
ResultsA total of 95 cases were included. The most common indications were biliary strictures (BS) evaluation (66.3%) or treatment of difficult common bile duct stones (27.4%). Technical and clinical success was attained in 98.9%. Single-session stone clearance was obtained in 84%. The AE rate was 7.4%. To detect malignancy in BS, optical diagnosis presents a sensitivity and specificity of 100% and 91.2%, respectively; while histology results were 36.4% and 100% respectively. A previous endoscopic sphincterotomy was associated with a lower rate of AEs (2.4% vs 41.7%; p<0.001).
ConclusionsSOCP with SpyGlass™ is a safe and effective technique to diagnose and treat pancreatobiliary pathology. The presence of sphincterotomy performed prior to the procedure could improve the technique's safety.
La colangiopancreatoscopia de un solo operador (SOCP) con el sistema SpyGlass® es una técnica endoscópica cuyo uso ha crecido exponencialmente durante los últimos años. Los objetivos de este estudio fueron evaluar la eficacia y seguridad de la SOPC con SpyGlass® y determinar los factores relacionados con la aparición de eventos adversos (EA).
Pacientes y métodosEstudio retrospectivo realizado en un único centro terciario, con inclusión consecutiva de todos los pacientes sometidos a SOCP con SpyGlass® desde febrero de 2009 hasta diciembre de 2021. No hubo criterios de exclusión. Se realizó un análisis estadístico descriptivo. Los factores asociados a la aparición de EA se analizaron mediante χ2 y la prueba t de Student.
ResultadosSe incluyeron un total de 95 casos. Las indicaciones más frecuentes fueron la evaluación de estenosis biliares (EB) (66,3%) o el tratamiento de coledocolitiasis difícil (27,4%). El éxito técnico y clínico se logró en 98,9%. La extracción de todas las litiasis en una sola sesión se obtuvo en 84%. La tasa de EA fue de 7,4%. Para la detección de malignidad en EB, el diagnóstico óptico presenta una sensibilidad y especificidad de 100% y 91,2%, respectivamente; mientras que los resultados de la histología fueron 36,4 y 100%, respectivamente. La esfinterotomía endoscópica previa se asocia con una menor tasa de EA (2,4 vs. 41,7%; p<0,001).
ConclusionesLa SOCP con SpyGlass® es una técnica segura y eficaz para diagnosticar y tratar la patología biliopancreática. La presencia de esfinterotomía previa al procedimiento podría mejorar la seguridad de la técnica.
Peroral cholangiopancreatoscopy is an emergent biliary endoscopy technique, performed using small-calibre endoscopic devices that allow direct visualization of the bile and pancreatic ducts. Since its inception, several systems have been introduced to perform these explorations, such as using an ultrathin gastroscope or the mother–baby scope system. However, its true expansion arose with the development of catheter-based devices such as the SpyGlass™ Direct Visualization System (Boston Scientific Corporation, Natick, MA, USA), approved by the FDA in 2005. This catheter includes an optical probe, a working channel and two irrigation channels that facilitate mucosal visualization. In addition, they are able to move in all four directions, facilitating the performance of therapeutic manoeuvres. It is introduced through the duodenoscope's working channel during endoscopic retrograde cholangiopancreatography (ERCP), which enables the entire procedure to be performed by a single endoscopist.1,2
Among its indications are the study of biliary strictures (BS).3–5 This enables improving the diagnosis of malignancy by means of endoscopic visualization of the lesions, in addition to allowing the taking of targeted biopsies.6,7 Another important application is the removal of stones from the bile or pancreatic duct by intraductal lithotripsy, after failure of the conventional ERCP techniques.8,9 Finally, it is useful to evaluate the extension of intraductal papillary mucinous neoplasms (IPMN).10,11
Among the adverse events (AEs) associated with this procedure are cholangitis, pancreatitis, haemobilia and biliary leakage, with higher occurrence rates compared to ERCP.11,12 However, the recent expansion of this technique and its low requirement mean that the factors related to the onset of AEs are not clearly established. The main aim of this study was to evaluate the efficacy and safety of the single-operator cholangiopancreatoscopy (SOCP) with the SpyGlass™ system. The secondary aim was to determine the factors related to the onset of AE.
Patients and methodsThis was a retrospective study at a single tertiary-care public institution (Hospital Universitario de Ourense, Galicia, Spain), in an endoscopy unit with a high volume of ERCPs (more than 500 per year). All patients undergoing SOCP with the SpyGlass™ system were consecutively recruited in a dedicated database and enrolled in the study. The inclusion period was from the implementation of this technique in our centre in February 2009 to December 2021. No exclusion criteria were considered.
The study was approved by the health area's ethics committee. All patients provided written informed consent for the procedure. None of the authors has received funding or aid to perform the study.
ProceduresAll procedures were performed by a single senior endoscopist. Prophylactic intravenous antibiotic therapy was administered before all cases and continued post procedure for at least 5 days. The procedures were carried out using deep sedation controlled by the endoscopist or under anaesthesia monitoring. The choice was uncontrolled and depended on the endoscopist's criteria. When sedation was controlled by the endoscopist, the use of propofol in addition to an opioid was usually employed, in some cases adding a low dose of midazolam.
In all cases, the SOCP was performed with the SpyGlass™ system (Boston Scientific Corporation, Natick, MA, USA), initially using the legacy system and later the digital system when it became available (SpyGlass™ DS and SpyGlass™ DS II). This device was inserted through the duodenoscope's therapeutic working channel over a previously placed guidewire in the duct of interest (common bile duct or main pancreatic duct). During the ERCP procedure a duodenoscope was used (TJF-145, TJF-Q180V, Olympus, Europe), using the wire-guided cannulation technique with a standard papillotome plus long-guidewire (Autotome™; 0.035 inch Jagwire, Boston Scientific). The use of needle knife or other advanced cannulation technique was not required in any patient. Biliary sphincterotomy was performed in all cases when it had not been performed during a previous ERCP. Pancreatic sphincterotomy was performed in the cases of pancreatoscopy.
During the SOCP, detritus was aspirated and saline solution irrigated to facilitate an adequate visualization or perform the lithotripsy. Tissue sampling was performed using the SpyBite™ mini-forceps. To the extraction of difficult CBDS, a SOCP-guided electrohydraulic lithotripsy (EHL) was performed (Autolith Touch EHL System; Boston Scientific Corporation, Northgate Technologies Inc, MA, USA) with a 1.9-Fr probe.
Post-procedure care and follow-upIn accordance with our centre's protocol, there was a minimum hospital stay of 24h after the procedures. On discharge, patients were followed up in outpatient clinics, between 2 and 4 weeks after the SOCP to evaluate the procedure's success and the absence of late AE.
DefinitionsTechnical success was defined as the ability to reach the lesion using the SpyGlass™ system. Following the study of Pons-Beltrán et al.,13 clinical success was defined differently for specific biliary lesions. In cases of BS or IPMN, a clinical success of the procedure was defined as the ability to reach the lesion, adequately visualize it and implement the appropriate diagnostic or therapeutic technique (biopsy, passage of the guidewire, or dilation). For biliary stones, clinical success was defined as the ability to access the stone and to perform lithotripsy if needed.
In the cases of lithotripsy, single-session stone clearance was defined as successful stone fragmentation and full removal in the first SOCP-guided EHL. Complete stone clearance was defined as successful stone removal after all endoscopic procedures.
AEs were defined and graded according to the ASGE lexicon's severity grading system Endoscopy.14 Procedure safety was considered within the first two weeks.
Procedure time was defined from insertion of the duodenoscope to its removal.
Evaluation of the biliary strictures with the SOCPLesions were endoscopically diagnosed as malignant if they had characteristic features, such as tumour-like vessels (irregular, dilated and tortuous blood vessels), intraductal masses, ulcerated areas, irregular surfaces with papillary areas or spontaneous bleeding.6,7,15,16 Histological diagnosis with the SOCP was determined based on the biopsies obtained using the SpyBite™ mini-forceps.
Following the study protocol by Pons-Beltrán et al.,13 the final diagnosis of the BS was determined based on the histological findings of the lesion and/or follow-up imaging. The biopsies to the final diagnosis could be obtained using SpyBite™, endoscopic ultrasonography or surgery. In the absence of a definitive histological diagnosis, the final diagnosis was considered as malignant when there was clinical deterioration and progression of the lesions after follow-up imaging. In contrast, lesions were defined as benign when biopsies were negative for malignancy and there was no clinical and radiological progression after 6 months follow-up.
Statistical analysisDescriptive statistics was performed using median with the interquartile range (IQR) for continuous variables and frequency with the corresponding percentage for categorical variables. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were calculated for optical and histological diagnosis. Factors associated with the presence of AE were analyzed using Chi-square and Student's t-test for the categorical and continuous variables, respectively. p values below 0.05 were considered statistically significant. Data were analyzed using R version 4.0.5 (R Core Team, 2020).
ResultsPatient characteristicsDuring the study period, a total of 5727 ERCP were performed. SOCP with SpyGlass™ was performed in 95 cases (1.66%). Most procedures (93 out of 95) were cholangioscopy and 2 were pancreatoscopy. The patients’ demographic data are detailed in Table 1. As shown in Fig. 1, SOCP use has gradually increased over the years, going from performing between 1 and 6 examinations per year after implementation of the technique to 23 explorations during the last year reported in our series.
Demographic data and baseline information of the procedures.
| n=95 procedures | |
|---|---|
| Age, median (IQR), years | 72 (63–79) |
| Sex male, n (%) | 53 (55.8) |
| ASA, n (%) | |
| I | 8 (8.4) |
| II | 42 (44.4) |
| III | 43 (45.3) |
| IV | 2 (2.1) |
| Indication, n (%) | |
| Biliary stricture | 63 (66.3) |
| Difficult CBDS | 26 (27.4) |
| Intrahepatic stones | 4 (4.2) |
| IPMN extension study | 2 (2.2) |
| Biliary stricture location, n (%) | |
| Proximal | 20 (31.8) |
| Medial | 17 (27.0) |
| Distal | 21 (33.4) |
| Multifocal | 5 (7.9) |
ASA: American Society of Anesthesiologists physical status classification; CBDS: common bile duct stones; IPMN: intraductal papillary mucinous neoplasm.
Table 2 summarizes observations of the procedures, outcomes and AE. Technical and clinical success was attained in 98.9% of procedures (94/95). The only failed case was a difficult CBDS, in which the EHL probe could not be inserted due to the patient's poor tolerance to the examination.
Observations of the procedures, outcomes and adverse events.
| n=95 procedures | |
|---|---|
| Procedure year, median (IQR) | 2019 (2016–2020) |
| Staff responsible for sedation, n (%) | |
| Anaesthetist | 65 (68.4) |
| Endoscopist | 30 (31.6) |
| Drugs in sedation by endoscopist, median (IQR) | |
| Propofol, mg | 265 (203–378) |
| Fentanyl, mcg | 50 (50–56) |
| Midazolam, mg | 1 (1–2) |
| Previous sphincterotomy, n (%) | |
| Yes | 83 (87.4) |
| No | 12 (12.6) |
| Procedure time, median (IQR), min | 60 (49–95) |
| Technical success, n (%) | 94 (98.9) |
| Clinical success, n (%) | 94 (98.9) |
| Single-session stone clearance*, n (%) | 21 (84) |
| Complete stone clearance*, n (%) | 24 (96) |
| Adverse events†, n (%) | 7 (7.4) |
| Perforation rate, n (%) | 3 (3.2) |
| Pancreatitis rate, n (%) | 2 (2.1) |
| Cholangitis rate, n (%) | 2 (2.1) |
| Severity of adverse events, n (%) | |
| Mild | 2 (2.1) |
| Moderate | 1 (1.1) |
| Severe | 2 (2.1) |
| Fatal | 2 (2.1) |
Of the 26 cases in our series whose indication of SOCP was the treatment of difficult CBDS, intraductal EHL was performed in 25 cases (in the remaining case the lithiasis could be removed using a Fogarty balloon, after the temporary placement of a plastic stent). Single-session stone clearance was obtained in 21/25 (84%). As previously indicated, one of the failed cases was due to the patient's poor tolerance to the examination, while in the remaining 3 cases there were residual CBDS after the initial EHL procedure. In these cases, it was necessary to temporarily stent placement and finalize the extraction in all of them during a second endoscopy session (complete stone clearance rate of 96%).
The onset of AE was observed in 7.4% of cases (7/95). The 2 cases of pancreatitis in our series were mild according to the ASGE severity classification.14 The 2 cases of cholangitis occurred during the study of BS, one of them moderate and the other severe. Regarding the perforations, one was severe (labelled as type IV according to the Stapfer classification) while in the other 2 the result was fatal (type I and III of the Stapfer classification). The 3 perforations occurred during explorations whose indication was the study of BS. No AE were observed in any of the SOCP-guided EHL.
Diagnostic of indeterminate biliary stricturesDuring the study period, 63 SOCP were performed to determine the aetiology of BS. In all cases, the required area with the SpyGlass™ system was accessed for a correct visualization of the lesions and optical categorization of the nature of the stenosis as benign or malignant (Fig. 2). In 2 cases, the final diagnosis according to our study protocol could not be established because they had a survival of less than 6 months, so they were excluded from this part of the analysis.
Among the 61 BS evaluated, optical diagnosis by SOCP was considered malignant in 30 cases, and final diagnosis confirmed malignancy in 27 of them. None of the 31 cases considered benign after SOCP visualization was finally a malignant stenosis. Endoscopic visualization sensitivity and specificity for the detection of malignancy were 100% and 91.2%, respectively.
In 28 (44.4%) of the BS under study, an attempt was made to obtain a histological diagnosis guided by SOCP using the SpyBite™ mini-forceps. In most in which biopsies were taken (75%), the suspected diagnosis according to the optical findings was malignant stenosis. The biopsy samples obtained were considered adequate to make the diagnosis in 20/28 cases (71.4%). In 4/4 cases labelled histologically as malignant, the final result was malignant. In contrast, among the 16 cases classified histologically as benign, in 7 the final diagnosis was malignant. The histological sensitivity and specificity for detection of malignancy were 36.4% and 100%, respectively.
Table 3 compares the diagnostic accuracy of the endoscopic visualization and the histological result of SOCP.
Diagnostic capacity of SOCP through endoscopic visualization and histological study.
| Optical diagnosis | Histological diagnosis | |
|---|---|---|
| Sensitivity, (%) | 100 | 36.4 |
| Specificity, (%) | 91.2 | 100 |
| PPV, (%) | 90.0 | 100 |
| NPV, (%) | 100 | 56.3 |
| Accuracy, (%) | 95.1 | 65.0 |
PPV: positive predictive value; NPV: negative predictive value.
In order to analyze the SOCP's safety, the different variables potentially associated with the onset of AE was analyzed. As shown in Table 4, among all the variables analyzed, only the presence of a previous sphincterotomy revealed a statistically significant association with the occurrence of AE, acting as a protective factor. Thus, only 2/83 patients with sphincterotomy prior to SOCP presented AE, compared to 5/12 who underwent sphincterotomy during the same procedure (2.4% vs 41.7%; p<0.001). To rule out the influence of the learning curve effect and endoscopist's grooving experience with SpyGlass™, we analyzed the proportion of previous sphincterotomies in the first and second half of the study and the rate of AEs over these periods. In the first 48 cases, the percentage of previous sphincterotomies was 81.25%, while in the remaining 47 it was 93.62% (p=0.132). The AE rate during these periods was 6.25% and 8.51%, respectively (p=0.977).
Association of different variables to the appearance of AE.
| Adverse events | No adverse events | p value | |
|---|---|---|---|
| Age, mean (SD), years | 63.7 (8.12) | 70.1 (13.5) | 0.091 |
| Sex, n (%) | 0.235 | ||
| Female | 2 (2.11) | 51 (53.68) | – |
| Male | 5 (5.26) | 37 (38.95) | – |
| ASA, n (%) | 0.492 | ||
| I | 1 (1.05) | 7 (7.37) | – |
| II | 4 (4.21) | 38 (40.0) | – |
| III | 2 (2.11) | 41 (43.16) | – |
| IV | 0 (0) | 2 (2.11) | – |
| Indication, n (%) | 0.104 | ||
| Biliary stricture | 6 (6.32) | 57 (60.0) | – |
| Difficult CBDS | 0 (0) | 26 (27.37) | – |
| Intrahepatic stones | 0 (0) | 4 (4.21) | – |
| IPMN extension study | 1 (1.05) | 1 (1.05) | – |
| Biliary stricture location, n (%) | 0.449 | ||
| Proximal | 3 (4.76) | 17 (26.98) | |
| Medial | 1 (1.59) | 16 (23.40) | |
| Distal | 1 (1.59) | 20 (31.75) | |
| Multifocal | 1 (1.59) | 4 (6.35) | |
| Staff responsible for sedation, n (%) | 0.094 | ||
| Anaesthetist | 7 (7.37) | 58 (61.05) | – |
| Endoscopist | 0 (0) | 30 (31.58) | – |
| Previous sphincterotomy, n (%) | <0.001 | ||
| Yes | 2 (2.11) | 81 (85.26) | – |
| No | 5 (5.26) | 7 (7.37) | – |
| Procedure year, mean (SD) | 2017 (2.5) | 2018 (3.59) | 0.676 |
ASA: American Society of Anesthesiologists physical status classification; CBDS: common bile duct stones; IPMN: intraductal papillary mucinous neoplasm.
Since its inception, SOCP has improved the diagnosis and endoscopic treatment of multiple biliopancreatic diseases, mainly thanks to the advantage provided by direct visualization of the ducts. This improvement has meant that its use has grown exponentially in recent years and has become a fundamental exploration for the management of some biliopancreatic pathologies.5
In part, the major expansion of its use could be due to the fact that it is a relatively affordable technique for operators accustomed to ERCP, as shown by our study's technical success rate (98.9%). This result is slightly higher than reported in older series, which was approximately 90%,2,17 but similar to the rates reported more recently, ranging from 95% to 100%.13,18 The technical improvements in the devices in recent years have probably contributed to further improve results that were already initially excellent.
Regarding the success in the management of difficult CBDS, our single-session stones clearance and complete stone clearance was 84% and 96%, respectively. This result is similar or slightly higher than some other studies: Pons-Beltrán et al.13 reported respective rates of 65% and 91% while Brewer et al.19 reported rates of 77.4% and 97.3%. Finally, in a meta-analysis involving 2786 patients, stone removal in a single session was attained in 71.1% (95% CI: 62.1%–79.5%) and complete stone clearance was 94.3% (95% CI: 90.2%–97.5%).20 Despite the variability of devices, the distinct modalities of lithotripsy (laser or EHL) used in the different series, and the variability of what is deemed “a difficult CBDS”, these results show the major utility of SOCP to extract the stones that cannot be managed by conventional ERCP procedures. However, there is a lack of randomized clinical trials that support these efficacy data.
To the strengths of clinical success of the SOCP we must add its good safety profile. The overall AE rate in our study was 7.4%, similar results to those reported in other series, with rates between 6% and 7.5%.2,12,13,15,20 The most commonly noted are cholangitis (related to the need for intraductal fluid irrigation to perform the examination), pancreatitis (mainly associated with pancreatoscopy and papillary manipulation), and perforations (mainly related to traumatic damage caused by the cholangioscope). Among the AEs in our study, it should be noted that of the 2 cases of pancreatitis, one appeared in one of only two explorations of pancreatoscopy that we performed. Despite this limited sample size these results would be consistent with the data provided in a recent study of the European Cholangioscopy study group, in which the greatest risk of this technique is collected reporting an AE rate for pancreatoscopy of 12%,21 which would double the rate with respect to the series with the global results of SOCP.
Despite its hypothetical increased risk, it should be noted that none of the EHL cases in our series presented AEs. In the previously mentioned meta-analysis,20 which evaluated the role of SOCP in the management of difficult CBDS, cholangitis was the predominant AE. They theorize that this may be related to a showering effect of fragmented stones throughout the biliary tree from irrigation, so prophylactic antibiotics may need to be administered. In our centre, prophylactic intravenous antibiotic was routinely administered before and after the procedures, which could justify the absence of cholangitis events associated with the practice of SOCP-guided EHL and would support our practice.
In our study, optical diagnosis using SOCP presents very high sensitivity and specificity (100% and 91.2%, respectively). However, despite the high specificity of the histological diagnosis (100%), its sensitivity is much lower (36.4%), which incorrectly classifies several malignant diagnoses. These results would be consistent with those previously published. In a systematic review and meta-analysis that evaluate the diagnostic performance of SOCP on indeterminate biliary lesions, optical diagnosis also led to better results detecting malignant lesions. A combined sensitivity and specificity of 90% (95% CI, 73%–97%) and 87% (95% CI, 76%–94%) was reported for the optical diagnosis, while histological results were 69% (95% CI, 57%–79%) and 98% (95% CI, 92%–99%) respectively.22 According to these results, SOCP is a good technique to differentiate malignant and benign biliary lesions, but its poor histological diagnostic capacity is a major limitation. The development during the last few years of personalized cancer treatments, makes it necessary to improve or develop new devices that enable a greater capacity for SOCP-guided histological diagnosis.
It is important to highlight the lack of studies with major statistical power that provide clear evidence for the identification of factors associated with the onset of AE during SOCP. In our report, among all the parameters evaluated, the only associated variable was the existence of a previous sphincterotomy, with a clearly statistically significant p value <0.001. Thus, of the total patients who had already undergone a sphincterotomy in a previous procedure, only 2.4% presented AE; compared to 41.7% of patients without previously performed sphincterotomy. It is difficult to compare our results with the existing literature, since most series provide global SOCP safety results, including cases in which sphincterotomy was performed before or during the same procedure, but without providing the data broken down between the different strategies. However, it is important to highlight that the series by Sethi et al.12 and Tieu et al.23 have already reported that the combined practice of ERCP and SOCP revealed an AE rate twofold higher than that of ERCP alone; and endoscopic sphincterotomy (required to be performed before the cholangioscope enters the bile duct) could be one of the possible causes accounting for these differences.20 It should also be taken into account that the SpyGlass™ is a device with a more relevant calibre and rigidity compared to other devices commonly used during ERCP. We hypothesize that its introduction by means of a recently performed endoscopic sphincterotomy could be much more traumatic than its introduction by means of a sphincterotomy that has already healed, which would justify the difference in AE rates. According to our findings, the performance of SOCP should be deferred whenever possible from the practice of endoscopic sphincterotomy.
This current study has some strengths and limitations that require further comment. Among the main limitations are that it is a single centre experience, a retrospective analysis (possibility of loss of cases and underdiagnosed AEs) and the strategy choice to perform the procedures with or without prior sphincterotomy was not controlled and randomized. In cases without a malignant histological diagnosis, another limitation could be define the lesions as benign when there was no clinical and radiological progression after 6 months follow-up, since this interval could be insufficient in some cases.24 However, it is also important to highlight the homogeneity of our cohort, the consecutive inclusion of patients during a period of more than 12 years and the fact that all procedures were performed by the same endoscopy team with a standardized surveillance.
Finally, this work adds an important new message to the growing SOCP-related literature: the presence of sphincterotomy performed prior to the procedure could significantly decrease the AE rates. To date, there are no studies clearly reporting these findings, and the increasing use of this technique pushes us to investigate all those measures that could improve its safety. Although these clinical results are attractive, larger prospective and randomized studies are needed to confirm these findings.
Ethical approvalThe work described has been carried out in accordance with the Ethical Code of the World Medical Association (Declaration of Helsinki). All patients provided written informed consent for the procedures. The study was approved by the health area's ethics committee.
FundingThis research has not received specific support from public sector agencies, the commercial sector or non-profit entities.
Conflict of interestsE. Sánchez has received consulting fees from Boston Scientific.
C. Tejido, M. Puga, C. Regueiro, M. Francisco and L. Rivas declare that they have no conflict of interest.
The following are the supplementary data to this article:
