EST has been generally accepted as a primary treatment for CBDS in current practice and is especially popular for elderly and high-risk patients. Multiple long-term follow-up studies have validated the superiority of EST in terms of safety and efficacy[8,9]. Over the past decade, OCT has been used less frequently each year but is still been performed at a reasonable scale in many institutions. However, short- and long-term follow-up of patient outcomes for CBDS remains unevaluated in the current literature for either classical OCT or minimally invasive EST, which is a cause for concern, especially for younger patients. This study is the first to report on this issue and may represent the best evidence comparing these two interventions.

In this study, anesthetic duration and procedure time were obviously and significantly shorter in the EST group than in the OCT group. These benefits of EST may reduce the risk of anesthesia and the incidence of procedure-related complications, especially in the elderly with poor cardiopulmonary function. Bacterial infection secondary to prolonged partial or complete CBD obstruction may result in AC. This effect varies from mild disease that generally responds well to medical management alone to life-threatening disease with a mortality rate of approximately 5%-10% that requires urgent biliary decompression[14]. In our study, the mean time from admission to biliary obstruction relief was longer for OCT (7.1 ± 3.7 d) than for EST (2.8 ± 1.6 d; P < 0.001). Theoretically, EST can relieve obstruction faster and prevent the transition of cholangitis into severe acute cholangitis, which ultimately reduces the mortality of CBDS[15]. However, a previous retrospective study and a recent Cochrane review of 16 randomized studies demonstrated that EST did not reduce the morbidity and mortality rates of CBDS compared with surgical intervention[16,17]. This finding may have been due to a selection bias because most of the critical and elderly patients choose EST, as it is minimally invasive and relieves obstruction rapidly, whereas those with mild disease and generally better condition choose OCT as their preference. For example, numerous studies suggest that patients presenting with toxic cholangitis, severe gallstone pancreatitis, multiple comorbid conditions, or medical risk factors should be managed by urgent EST for endoscopic drainage. These patients are at an increased risk of mortality due to the severity of their illness rather than from the use of EST.

Failure with the initial attempt to treat CBDS results in multiple procedures, incurring a significant increase in health and financial burden. Theoretically, EST is easier to perform than OCT and, thus, would be expected to lower hospitalization cost. However, in the current study, total hospitalization cost was similar between the two groups, which probably resulted from the requirement for multiple procedure sessions in the EST group (12.5% vs 2.5%, P = 0.009).

The results of this study indicated that EST was equivalent to OCT in terms of efficacy for CBDS treatment, with both having similarly high technical success rates. However, with regards to safety, we found that the primary procedure-related complications of EST were bleeding, cholangitis, hyperamylasemia, and PEP, with incidence rates of 1.8%, 1.2%, 4.8%, and 3.6%, respectively. These findings concurred with previous studies[15,18]. PEP, as the most frequent complication of ERCP, was probably due to sphincter of Oddi dysfunction. This likely resulted from postprocedural edema or sphincter spasm, which may worsen compression of the pancreatic orifice and thus increase the risk of PEP. Additionally, EST itself and the use of pancreatic duct indwelling guidewires were also two independent risk factors for PEP. The primary factors influencing the efficiency and safety of EST appeared to be the skill and experience of the endoscopists, regardless of patient age or general medical condition. OCT followed by T-tube drainage is a safe and effective method for postoperative biliary decompression to maximize healing of the CBD incision. However, it is not exempt from complications, which occur in up to 10% of patients[19]. The most frequent of these are biliary leakage and port site infection after removal, which occurred in 3.7% and 4.8% of cases in the current study, respectively. Keeping the T-tube in place for an appropriate period of time decreases the incidence of bile leakage into the peritoneal cavity. Therefore, dislodgement or fracture of the T-tube or its premature removal may increase the risk of biliary leakage and port site infection.

Many scholars believe that endoscopic therapy shortens the length of hospital stay, but this advantage is offset by a substantially higher rate of recurrent biliary symptoms, some of which demand readmission. Conversely, our results indicated that the OCT group had a higher rate of CBDS recurrence than the EST group, although this difference was not statistically significant (20.5% vs 17.0%, P = 0.418). Presently, there are no relevant studies confirming our findings. Additionally, patients with recurrent CBDS complicated by AC were observed significantly more frequently after OCT, but there was no significant difference when compared with EST (P = 0.054). It is possible that hyperplasia of fibrous scar tissue at the CBD incision site after OCT could cause CBD angulation or malformation. This effect may lead to chronic inflammation and thus cholangitic stenosis of the CBD, contributing to biliary tract bacterial infection by disturbing bile excretion and ultimately resulting in AC. With EST, there has been concern expressed that duodenobiliary reflux due to permanent loss of sphincter of Oddi function may lead to recurrence of stones or ascending cholangitis[20,21].

In the current study, we found a significant association between the two groups in terms of the length of time to initial recurrence of CBDS. Accumulating studies have reported that most recurrences occurred within the first 2 years after EST[22]. Similarly, 39.3% of recurrences in our study occurred within 2 years of EST. These studies found that sphincter of Oddi basal pressure and amplitude decreased markedly after EST but could recover to some extent 2 years following this procedure[23]. However, except for inflammatory stenosis, no studies have fully explained the underlying pathogenesis of stone recurrence after OCT. Thus, further study on this topic is needed.

Christos et al[24] proposed that patients with recurrent CBDS were at an increased risk of a subsequent recurrence. In our study, 21.2% of patients with one recurrence had at least one further episode, while the first-time recurrence rate was 18.8%. However, there was no significant difference in the rates of a second recurrence between the EST and OCT groups in our study (P = 0.700). Furthermore, the rates of reintervention for an initial recurrence were similar in the EST and OCT groups (17.0% vs 15.9%, P = 0.839). However, EST was the approach chosen more frequently for reintervention for recurrent patients, with over 50.0% (7/14) of patients transferring from the OCT group to EST therapy. This could be due to EST being more favorable for postoperative recovery, with decreased pain and an increase in quality of life in the short term. Moreover, free external drainage from T-tubes may lead to slow wound healing, anorexia, and constipation, as well as increasing the risk of dehydration and electrolyte imbalance[25]. Mortality was very low, and there was no significant difference in the postprocedure death rate between the two groups in our study. Our results concur with that of a retrospective study in which 1514 patients with CBDS were treated by OCT and the operative mortality was 2%. During our follow-up, we found that the primary causes of death were cardiopulmonary complications and unknown etiology rather than procedure-related adverse events.

The recurrence rate for CBDS quoted in the literature ranges from 4%-25%[11]. However, the risk factors for stone recurrence remain unknown. Numerous studies have found that bile stagnation, duodenal-biliary reflux, bacterial colonization of the CBD, ML, and CBD angulation or deformity all contribute to recurrence[26]. However, our study demonstrated that ML and CBD bacterial infection did not increase recurrence incidence. Although univariate analysis revealed that the stone recurrence rate was higher in patients with stones ≥ 20 mm, this was not statistically significant in multivariate analysis. Multivariate analysis revealed that CBD diameter, CBD angle ≤ 145°, and multiple CBDS were independent risk factors for CBDS recurrence. The underlying mechanism for CBD dilation, which predisposes to stone reformation, may be a decrease or loss of normal CBD functional movement or a decrease in the hydrostatic force of bile. Both of these events could slow down bile excretion, thus leading to CBDS formation. Additionally, cholesterol in the bile of patients with multiple stones might be prone to crystallizing and nucleating, causing a tendency toward stone formation.

The comprehensive and long-term follow-up in our study provides novel insights into the different outcomes with EST and OCT, and the risk factors for subsequent stone recurrence. However, there are several limitations to our study. First, we could not rule out the possibility that recurrent CBDS might be asymptomatic and evade detection during follow-up. Second, due to its retrospective nature, a selection bias may have occurred with the two treatment groups. Last, we did not ascertain the composition of the stones at the initial intervention. Therefore, we could not examine whether differences in stone composition affected the risk of recurrent CBDS.

In conclusion, EST currently remains the most commonly performed procedure for CBDS management. This may be because it is minimally invasive and has satisfactory short-term outcomes including shorter time to biliary obstruction relief, anesthetic duration, procedure time, and hospital stay. Furthermore, EST can be performed safely without posing a higher risk of subsequent recurrent CBDS or overall mortality than OCT. The recurrence rate appears to be associated with CBD diameter ≥ 15 mm, multiple CBDS, and distal CBD angle ≤ 145°.