Do we need to place a biliary stent in patients waiting for cholecystectomy?

MEMBER ARTICLE

Role of Biliary Stent in Recurrence of Biliary Stones and Complications After Stone Clearance in Patients Awaiting Cholecystectomy: A Randomized Trial

[membership level="0"] [/membership] [membership level="7,8,9"]

Role of Biliary Stent in Recurrence of Biliary Stones and Complications After Stone Clearance in Patients Awaiting Cholecystectomy: A Randomized Trial

 Sasani A, Mandavdhare HS, Sharma V, et al. Am J Gastroenterol. 2023;118:1864-1870.

Question

To study the effect of biliary stent vs. no stent on recurrence of choledocholithiasis and biliary complications in those awaiting cholecystectomy after biliary clearance.

 Design 

A single-center randomized controlled trial.

Patients

Patients with choledocholithiasis with concomitant cholelithiasis in whom biliary clearance was achieved. Biliary clearance was diagnosed based on an occlusion cholangiogram showing no filling defect on scout film.

Interventions

Randomization was conducted in a 1:1 allocation using a computer-generated sequence of random numbers with permuted block randomization with block sizes of 2 and 4. Masking was not possible. Patients were randomized into 2 groups. Group A—biliary clearance with stent placement (either a 7-Fr and 7cm or a 10-Fr and 7-cm double-pigtail stent was placed as per the endoscopist's discretion) and group B—biliary clearance without stent placement.

ERCP was performed by experts with experience in performing an average of 300 endoscopic retrograde cholangiopancreatography per year for 5 years or more, along with experience in performing at least 50 precut rescue cannulation techniques. All the patients underwent complete biliary sphincterotomy before clearance. Endoscopic papillary balloon dilation was also used at the discretion of the endoscopist. After documenting total clearance, patients were randomized into 2 groups.

During the waiting period for cholecystectomy, patients were followed up after a week of ERC and then monthly. The patients were informed to report immediately if they developed abdominal pain, jaundice, and fever and were admitted for further management. Diagnostic endoscopic ultrasound (EUS) was performed at 3 months of ERC to detect recurrence of choledocholithiasis. Those unwilling for EUS were advised MRCP. Those who were unwilling for either EUS/MRCP were advised ultrasonography of the abdomen and liver function test at follow-up.

Those randomized to the stent group had their stent removed 3 months after the follow-up evaluation on the same day. In patients with evidence of recurrence of choledocholithiasis, repeat ERC was performed.

Outcome

The primary outcome was the choledocholithiasis recurrence rate. The secondary objectives were complications (cholecystitis, cholangitis, and pancreatitis) rate and the need for repeat ERC for 3 months. The definition, severity, and management of cholecystitis and cholangitis were per the Tokyo Guidelines 2018. The definition and severity of biliary pancreatitis were per the revised Atlanta classification, and the management was done per the institute’s protocol.

Data analysis

Sample size calculation was done with a power of the study at 80% and alpha error of 5%, expecting 30% in the non-stented group vs 5% in the stented group.

 (70 patients, 35 in each group). Analyses were performed according to per protocol (PP) principles. A comparison of categorical outcomes was performed between study groups using the x² test/Fisher exact test. For quantitative parameters, the independent sample t-test (2 groups) or the Mann-Whitney U test was used depending on the normality of distribution.

Funding

None

 Results 

Between September 2021 and July 2022, 359 patients were screened. Seventy patients were randomized, 35 in each group. After the last patient was recruited in July 2022, all patients were followed up until October 2022. One patient in group A and 3 in group B were lost to follow-up. Hence, 34 patients in group A and 32 in group B were available for final analysis as PP.

Group A had a higher rate of cholecystectomy (79.4% [27/34] vs 53.1% [17/32], P=0.024). For the primary outcome, 10 (15.2%) had a recurrence of choledocholithiasis (group A: 7/34 [20.6%] and group B: 3/34 [9.4%], P = 0.306). None of the patients with recurrence were symptomatic or had abnormal LFT. Nine underwent ERC and had evidence of stones, while 1 patient was awaiting ERC. In the secondary outcome, five patients in the stent group while none from the no stent group developed complications (5 vs 0, P = 0.024). Urgent/emergency ERC for cholangitis was not needed in any of the patients.

COMMENTARY

Why Is This Important?

In clinical practice, a frequent scenario is deciding if you place or not one biliary stent (and the patient will need another procedure to remove the stent). For most of us, the decision is based on the date of the future cholecystectomy. If cholecystectomy is soon enough (I have heard different times, from 2 weeks until 3 months), you do not place the stent, but if the waiting is more prolonged or you do not know when it will be, you place the stent.

According to the authors, their trial “suggests that in patients with choledocholithiasis with concomitant cholelithiasis, prophylactic biliary stent placement after biliary clearance may increase the risk of complications while providing no benefit in terms of reducing choledocholithiasis recurrence.”

Key Study Findings

Urgent/emergency ERC was not needed in either group of patients.

None of the patients with “recurrence” were symptomatic or had abnormal LFT.

 Caution

I want to consider three points: a) the author´s analysis: they did it PP analysis. Much has been written about the importance of the intention to treat analysis (ITT) vs. PP. Sometimes, it is valid (and necessary) to do PP analysis (1). However, I think that this paper is not the case. In Figure 1, the authors show us that in group B (no stent group), 3 patients were lost in the follow-up and 1 patient in group A (stent group). The appropriate analysis must be done with “imputation.” That means the patients lost in the follow-up need to be analyzed as “failures” (in this case, as if they were presented with the outcome). We do not know why the patients were lost in the follow-up (they could die because of a severe complication such as cholangitis or pancreatitis). If you consider those patients in ITT analysis, the numbers change dramatically: recurrence of choledocholithiasis (group A: 8/35 [22.8%] and group B: 6/35 [17.1%]), and in secondary outcome, 6/35 (17.1%) vs 3/35 (8.5%). The second point is the gold standard for cholelithiasis or choledocholithiasis. The authors never clarify the gold standard. In clinical trials, cholecystectomy is the gold standard for cholelithiasis, and the ERCP is the gold standard for choledocholithiasis. The authors used the EUS/MRI to decide which patients underwent ERCP (n = 10). However, what about if they missed patients with choledocholithiasis with these methods? ALL patients were asymptomatic or had abnormal LFT. In that case, we will never know these patients' true recurrence of choledocholithiasis. The same applied to cholecystectomy; not all patients had it.

And the third one, and (in my opinion) the most significant, is none of the patients with recurrence were symptomatic or had abnormal LFT. When we placed one stent while waiting for the cholecystectomy, we intended to keep the patients asymptomatic. Based on this, the study is practical, and the main message (from my point of view) is that patients without biliary stents did not have a higher rate of symptomatic recurrences of choledocholithiasis. Would the p-value remain significant if the ITT analysis were performed for complications?

 My Practice

Regularly, if I do not know when the cholecystectomy will be done or if it will be in more than 4-6 weeks, I prefer to place a biliary stent. I could be comfortable extending the period to 3 months with these results.

For Future Research

One study specifically designed to evaluate complication rates could be attractive.

 REFERENCES 

1.- Ford I, Norrie J. Pragmatic Trials. N Engl J Med. 2016;375:454-63

[/membership]