Endoscopic lithotripsy is the first choice for the treatment of choledocholithiasis. Minimally invasive endoscopic treatment, which is safe even among frail elderly patients with underlying diseases, has been increasingly used. However, in actual clinical practice, patients may not endure long procedures. Thus, the procedure can be discontinued, thereby leaving a biliary stent alone. Bergman et al. assessed the outcomes of long-term treatment with 10-F polyethylene stents among elderly patients (n = 58).[16] This method was initially successful. However, over time, 38% of patients developed recurrent cholangitis, and in 12% of cases, it was fatal. Chopra et al. found that ductal clearance was consistently associated with a higher rate of procedural AEs in a randomized comparison between ductal clearance and long-term biliary stenting (16% vs. 7%). However, the incidence of long-term biliary AEs was lower (14% vs. 36%).[17] By contrast, there are some reports showing that long-term stenting is beneficial for high-risk elderly patients with choledocholithiasis.[9–11] However, the validity of long-term EBS remains controversial.
In retrospective studies comparing complete stone removal and long-term EBS, selection bias is more likely to occur because a high number of elderly patients and those with poor general health are included in the EBS group. In this study, this bias was eliminated by including only patients aged 75 years and older in the target population and by matching patient factors between the two groups via PSM.
Interestingly, when we compared the background characteristics of patients before PSM in our study, we found significant differences in terms of patient factors (such as age and underlying diseases) and anatomical factors (including periampullary diverticulum and surgically altered anatomy). However, there was no significant difference in terms of the number of stones. Furthermore, the patients in the EBS group were more likely to have larger stones. Nevertheless, the median diameter was only 10 mm. In recent years, the technique used to remove the so-called difficult stones has significantly improved. New ampullary interventions including EPLBD were found to be useful in the treatment of large and multiple stones.[21–23] Moreover, lithotripsy techniques including electronic hydraulic lithotripsy using a digital-single-operator cholangioscopy have also been developed.[24, 25] This suggests that the general condition of the patients rather than stone factors may affect the choice of palliative EBS in our study. In addition, before PSM, the EBS group was likely to have a surgically altered anatomy. In recent years, ERCP with a balloon-assisted endoscope was found to be effective in patients with surgically altered anatomy.[26–28] However, the technique is still challenging, and the procedure was longer in such cases. This might increase the number of cases in which the stones were not completely removed even after a successful ERCP and the procedure was discontinued with stenting alone.
In the current study, the incidence of AE and mortality rate did not significantly differ. However, the median time to cholangitis was significantly shorter in the EBS group than in the complete stone removal group. There was no significant difference between the two groups in terms of treatment safety or length of hospital stay. Nevertheless, about half of the patients in the EBS group eventually required reintervention and re-hospitalization. More importantly, one patient in the EBS group died due to severe cholangitis. In the EBS group, the main cause of stent dysfunction was stent migration, which might have been caused by the fact that the stent was placed in the bile duct without stenosis. Additionally, in our study, there was a case of stent–stone complex in the EBS group. Kaneko et al. showed that long-term EBS increases the risk of stent–stone complex.[29] Stent–stone complex formation can lead to difficulties in removing old stents via conventional endoscopic procedures.
By contrast, the median duration of cholangitis-free periods in patients with palliative EBS was 596 days. Therefore, palliative EBS may be acceptable in patients with malignancies who have a poor prognosis. However, the indications should be limited. Even in cases in which biliary stenting was unavoidable at the time of initial treatment, it may be necessary to perform the procedure again after the patient’s general condition improves to achieve complete stone removal or to consider planned stent replacement.[30]
In addition, in some elderly patients with choledocholithiasis, even if the bile duct stones are removed via ERCP, the gallbladder stones may not be surgically resected thereafter. In our study, even in the matched cohort adjusted for the proportion of patients with residual gallbladder with gallstones after ERCP, the incidence of cholangitis was significantly lower in the stone removal group. Yasui et al. showed that the recurrence of choledocholithiasis did not increase even if the gallbladder with gallstones is preserved after endoscopic treatment of choledocholithiasis among elderly patients.[31] Therefore, regardless of whether cholecystectomy is feasible in the future, a reasonable therapeutic strategy should be used to completely remove bile duct stones.
The current study had several limitations. First, patients with inadequate follow-up were included. This study included a high number of elderly patients with underlying medical conditions. In some cases, regular outpatient visits were challenging. Therefore, long-term prognosis could not be assessed, and some patients were censored using the Kaplan–Meier curve. Second, this is a retrospective study, which may not be as statistically reliable as randomized control trials. However, in this study, the background characteristics of the stone removal group and the EBS group were adjusted using PSM to ensure homogeneity between the two groups. We believe that the statistical reliability of this study is sufficient. Third, because the outcome of the study was the occurrence of cholangitis, the exact recurrence rate of choledocholithiasis in the stone removal group remains unknown. However, we believe that the incidence of cholangitis is more important than the recurrence rate of stones in clinical practice. Therefore, this study is considered more relevant to actual clinical practice.
In conclusion, palliative EBS was effective in controlling cholangitis for a certain period of time among frail elderly patients with choledocholithiasis. However, a significantly higher number of patients required reintervention and re-hospitalization for cholangitis in the EBS group than in the complete stone removal group. The median duration of cholangitis-free periods in the palliative EBS group was significantly shorter than that in the complete stone removal group even after adjusting for background characteristics using PSM. Furthermore, one patient in the EBS group died due to severe cholangitis. Thus, palliative EBS should be indicated only in patients with choledocholithiasis who have a poor prognosis.