Comparing outcomes of right verse left hepatic approach percutaneous biliary drainage catheters

Purpose Determine if there is a difference in adverse events (AE) between right or left hepatic percutaneous biliary drain placement (PTBD) in patients with biliary strictures. Materials & methods This retrospective study included patients with benign or malignant biliary stricture treated with PTBD at a single institution from 7/28/2004–3/30/2021. 357 patients met inclusion criteria, 77 (21.6 %) had PTBD on the left and 280 (78.4 %) on the right. AEs associated with the initial drain placement or during subsequent intervention were collected and categorized. AEs that were grouped as periprocedural included: surgery, infection, hemorrhage, and drain failure. AEs in the postprocedural group included: chills, catheter displacement, cholangitis, biliary stones, drain malfunction, fever resolving without treatment, and pericatheter leakage. Surgery was considered a major AE and the remaining AEs were categorized as minor. Statistical analyses were performed using Logistic Regression Analysis and p-values less than 0.05 were considered statistically significant. Results Overall, there was no statistically significant difference in AEs between right and left drains in the periprocedural and postprocedural period (p = 0.832, OR = 0.95 and p = 0.808, OR = 0.93 respectively). When analyzing minor AEs individually, only cholangitis occurred at a higher rate on the right side (p = 0.033, OR = 0.43). There was no statistical difference in the rate of major AEs in the periprocedural period between left and right drains (p = 0.311, OR = 1.37). Conclusion Current literature is equivocal when comparing right versus left percutaneous biliary drains. This analysis describes no statistically significant difference in AEs between right and left hepatobiliary drains aside from slightly higher incidence of cholangitis for right sided drains.


Introduction
Biliary stricture is an unfortunate clinical diagnosis and often treated by Interventional Radiologists.Benign biliary strictures may be secondary to several etiologies including trauma, cholelithiasis, pancreatitis, postprocedural and inflammatory disorders [1].Malignant biliary strictures are most frequently secondary to primary pancreaticobiliary malignancies, though lymphoma and metastatic disease are also potential causative tumors [2,3].Percutaneous transhepatic cholangiography (PTC) and percutaneous transhepatic biliary drainage (PTBD) are common procedures performed for diagnosis and treatment of both malignant and benign hepatobiliary disorders [1].When surgical or endoscopic therapies are suboptimal or not feasible, using fluoroscopic guided PTBD with possible drain or stent placement is a reasonable treatment option.Furthermore, as institutions continually adopt a benign biliary stricture treatment protocols (BBSP), PTBD will continue to be a mainstay therapy option for patients with biliary disease [10].
While benign and malignant etiologies may result in peripheral stricture specifically requiring a right or left drain, obstruction at or below the level of the biliary confluence may be treated with either approach.A specific example of laterality of drain being predetermined is a case of hilar cholangiocarcinoma that is resectable via ☆ Prior conference presentation: SIR 2023 Annual Scientific Meeting, submission number: 1325729 hemipatectomy; in this situation, there is evidence to support selectively draining the future remnant liver rather than the side being resected.Previous studies have shown unilobar hepatic drainage to be equally effective as bilobar drainage with obstruction at the confluence [4] and suggest relieving just 30 % of the hepatic parenchyma is sufficient to reduce symptoms of obstructive jaundice [5,6].These factors currently contribute to drain laterality often being selected based on the patients' personal preference (e.g.intercostal pain, ease of maintenance or concealment) rather than objective clinical parameters such as procedural complication rate or risk of subsequent drain failure.Long term maintenance for biliary drains are largely dependent on the institution.Typically, drains will be followed until either removal or permanency.The degree of which varies widely, depending on whether the placement is for benign, malignant, transient process, iatrogenic, or a different reason which determines the ultimate long-term course.Retrospective studies comparing right versus left approach PTBD are scarce with the current research comprised primarily of prospective studies that are limited in their long-term evaluation of drain adverse events (AE) and evaluate only malignant or benign strictures.Moreover, there is no current consensus on which approach is technically safer; one study suggested that the right approach carried an increased risk of bleeding [7], while others suggested the left approach [8,9] or have found no statistically significant difference whatsoever [2].This study aims to compare the periprocedural, postprocedural AE incidence and additional surgical intervention rates between right and left PTBD with additional long-term evaluations.

Materials and methods
Following institutional review board approval, patient medical records were reviewed in compliance with Health Care Portability and Accountability Act guidelines.Patients who received PTBD for treatment of benign and malignant biliary stricture were identified using Picture Archiving and Communications System (PACS) integrated data mining software (Illuminate Insight; Softek, Kansas City, Kansas).Study data was then managed using REDCap (Research Electronic Data Capture) electronic data capture tools.
Patients with benign stricture underwent repeat intervention based on the institution's BBSP or at the discretion of the interventional radiologists, depending on the patients change in status requiring reintervention.Patients with malignant stricture received palliative treatment with routine drain exchanges at set intervals; however, both groups underwent repeat intervention when the patient demonstrated clinical signs of drain failure.All procedures were performed by fellowship-trained interventional radiologists with an average of 10 years of experience (range 2-22 years).
Patients were followed from the time of initial biliary drain placement to the time of death, or last available note in the electronic medical record (median = 342 days, IQR = 67.5-1446.2days).The primary analysis grouped AEs occurring in intervention related periprocedural period versus the treatment related postprocedural period, which were further classified as minor or major.The periprocedural period consisted of intraprocedural, immediately after, and any required short-term hospitalization associated with the drain placement.The postprocedural period was quantified for any time thereafter.Among periprocedural associated AEs, only emergent surgical intervention was classified as a major AE.Minor periprocedural associated AEs included infection, hemorrhage, and drain failure.Among postprocedural associated AEs, no AEs were classified as a major.Minor postprocedural AEs included chills, catheter displacement, cholangitis, biliary stones, drain malfunction, fever resolving without treatment, and pericatheter leakage.Any clinical, laboratory, or radiologic evidence was used to evaluate for a malfunctioning drain.
The analysis was conducted using version 4.1.1 of the R statistical programming software.Univariate logistic regression models were employed to examine the association with biliary catheter laterality and the occurrence of adverse events.The outcome variables for each model were binary, with "No Complications" as the reference group compared to "Any Complications" (including major and minor complications) or "Major Complications" (major complications only).We utilized the 'glm ()' function in R with the default Wald test for hypothesis testing.The models were specified with the binomial family to account for the binary nature of the outcome variable.Model goodness-of-fit was assessed using standard techniques, including the Hosmer-Lemeshow test for logistic regression models.The significance level for all statistical tests was set at alpha = 0.05.

Results
This retrospective study included all patients diagnosed with benign or malignant biliary stricture who underwent PTBD in the interventional radiology division at a single institution from July 2004 to March 2021.Patients were excluded if they initiated or completed percutaneous therapy at an outside institution or had bilateral biliary drains placed.1.
The rate of no AE in the intervention related periprocedural period was similar between left (50.6 %) and right (49.6 %) hepatic approach drains with logistical regression analysis demonstrating no statistical significance (p = 0.832, OR = 0.95).Furthermore, the rate of no AE in the treatment related postprocedural period was also similar for left (36.4 %) and right (37.5 %) hepatic approach drains (p = 0.808, OR = 0.93).When individual minor AEs were evaluated, there was no statistically significant difference between left and right drains for any minor AEs in the periprocedural period.However, when individual AEs in the postprocedural treatment period were analyzed individually there was a statistically significant difference in the rate of cholangitis between left sided drains (10.4 %) and right sided drains (21.4 %) (p = 0.033, OR = 0.43).In the periprocedural period, there was no statistically significant difference in the rates of surgery between left (23.4 %) and right drains (18.2 %) (p = 0.311, OR = 1.37).See Table 2.
Lastly, there was no difference in odds or having major vs none nor any complications vs none between left and right drains.See Table 3.

Discussion
This retrospective study included patients with benign or malignant biliary stricture treated with PTBD and evaluated adverse events regarding the periprocedural and postprocedural periods.There were no statistically significant differences in major AEs between right and left drains in the periprocedural and postprocedural period.When considering minor AEs individually, cholangitis in the postprocedural period was the sole AE that occurred at a statistically significant different rate between left and right sided drains.There was no difference in odds or having major vs none nor any complications vs none between left and right drains.
This study suggests there may be greater chance of developing cholangitis during the postprocedural period with right hepatic approach biliary drains but found no difference in any other AEs between left and right approaches.There is no current, available literature that corroborates this correlation between right sided interventions and cholangitis, but the anatomy of the biliary tree may provide further insight.The right hepatic duct is generally much shorter and bifurcates earlier than the left hepatic duct, measuring 0.9 cm on the right versus 3.0 cm on the left [5].The left hepatic duct also generally branches at a more acute angle from the common hepatic duct than the much more obtuse or streamlined right hepatic duct, see Fig. 1.This difference in branch angle and length of the duct prior to bifurcation may cause the progression of infection or inflammation along the right hepatic more frequently than the left hepatic duct.
Current literature regarding efficacy and safety of right versus left percutaneous biliary drains is equivocal and each side may have its advantages and disadvantages.Prior studies have been limited due to prospective small group analysis or limiting patient populations strictly to malignant or benign biliary strictures.Nonetheless, the consensus favors right hepatic approach drains even though there is research that favors left biliary drains due to decreased bleeding [7].Frequently, radiologists may choose whichever ductal system is the most dilated allowing for a less complicated placement, but often the laterality is purely at the discretion of the provider.The decision to place a right versus left percutaneous biliary drain is universally operator dependent.Many technical and physical challenges are rationalized when making this decision.They vary from patient tolerability (i.e.intercostal versus peritoneal), isolated biliary systems on imaging, degree of dilation, future planned surgeries, etc.However, right-sided percutaneous biliary drains are generally easier for wound management and maintenance flushes.
There are several limitations to this study.Patients were not randomized to left or right drains and provider preference and variance in technique may have introduced bias.The study group also revealed asymmetry in terms of number of left versus right biliary drains included.This is likely due to a subjective preference of providers who placed the drains in this single institution study.AE data was captured retrospectively, limiting the ability to properly standardize or catch all  instances of AEs.Additionally, some patients may have had follow-up or received treatment for AEs at outside institutions, and these instances would not have been captured in this data set.Finally, there is a potential for bias due to provider preference for right versus left biliary drains in patients where either side was suitable.This retrospective analysis describes no statistically significant difference in AEs between right and left hepatobiliary drains aside from slightly higher incidence of cholangitis for right sided drains.Future research including a randomized controlled trial may further explore this finding.Current research suggests providers may continue to choose either a right or left sided drain based on their own or the patients' preference.

Fig. 1 .
Fig. 1.Typical intrahepatic biliary branching pattern demonstrates the right and left hepatic lobe.Created with Biorender.com

Table 2
Perioperative and postoperative period AE rates.

Table 3
Odds ratios between left and right drains.