Internal vs oral bile reinfusion for hepatic recovery after percutaneous transhepatic cholangiodrainage

Abstract

BACKGROUND

Percutaneous transhepatic cholangiodrainage (PTCD) is the gold standard for palliative management of malignant obstructive jaundice when endoscopic attempts fail or are contraindicated. However, long-term bile drainage externally produces a significant amount of biliary loss that can affect recovery of liver function and nutrition. Bile reinfusion is a promising approach for long-term maintenance of enterohepatic circulation; however, only a few studies have systematically compared different methods of bile reinfusion.

AIM

To explore whether hepatic function recovery after PTCD is better with internal than external bile reinfusion in patients with malignant obstructive jaundice.

METHODS

This retrospective study included a total of 100 patients with malignant obstructive jaundice treated with PTCD from January 2020 to December 2023. Patients were assigned to be treated with internal bile reinfusion through drainage tube (observation group, n = 55) or oral bile reinfusion (control group, n = 45). Between groups, baseline characteristics were similar. The primary outcomes were hepatic function parameters (serum total bilirubin, direct bilirubin, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, gamma-glutamyl transferase and albumin) measured at baseline and then 1 week, 2 weeks, and 4 weeks after the reinfusion. Secondary outcomes included the nutritional indices (prealbumin, hemoglobin and body weight), gastrointestinal symptoms, quality of life scores and safety data.

RESULTS

All outcomes were found to be significantly better in the observation group. At 4 weeks after reinfusion, total bilirubin decreased more significantly in the observation group (42.5 ± 12.3 μmol/L vs 68.7 ± 18.5 μmol/L; P < 0.001) and direct bilirubin (28.3 ± 8.5 μmol/L vs 45.2 ± 12.7 μmol/L; P < 0.001). At the end of intervention, serum albumin levels were more improved (38.5 ± 4.2 g/L vs 34.2 ± 5.1 g/L; P < 0.001), with normal serum albumin achieved by 78.2% compared with 46.7% in controls (P = 0.001). The liver transaminases (alanine aminotransferase: 38.5 ± 12.3 U/L vs 56.8 ± 18.5 U/L; P < 0.001; aspartate aminotransferase: 42.3 ± 14.5 U/L vs 62.5 ± 20.3 U/L; P < 0.001) of the patients in observation group were significantly recovered faster than those of control group respectively. Cholestatic enzymes normalized more quickly (alkaline phosphatase: 185.3 ± 42.5 U/L vs 248.7 ± 58.3 U/L; P < 0.001; gamma-glutamyl transferase: 125.8 ± 35.2 U/L vs 178.5 ± 48.7 U/L; P < 0.001). Prealbumin level significantly improved (285.5 ± 45.3 mg/L vs 215.8 ± 52.7 mg/L; P < 0.001), body weight loss ameliorated (2.3 ± 1.5 kg vs 5.8 ± 2.3 kg; P < 0.001) and quality of life scores increased (78.5 ± 9.2 vs 62.3 ± 11.5; P < 0∙001). The safety profile was similar between groups for catheter-related complications (12.7% vs 13.3%; P = 0.926). Multivariable analysis showed that internal bile reinfusion was independently associated with the recovery of good hepatic function after PTCD (adjusted odds ratio = 9.25, 95% confidence interval: 3.58-23.91; P < 0.001).

CONCLUSION

Internal bile reinfusion through the drainage tube markedly improves hepatic function recovery, shortens the time of total bilirubin normalization, enhances nutritional status, and significantly raises quality of life in patients with malignant obstructive jaundice patients after PTCD. It is a safe and effective option for postoperative bile therapy and may better preserve enterohepatic circulation than oral reinfusion.

Key Words: Percutaneous transhepatic cholangiodrainage; Malignant obstructive jaundice; Bile reinfusion; Internal reinfusion; Hepatic function

Core Tip: Malignant obstructive jaundice treated with percutaneous transhepatic cholangiodrainage often results in prolonged bile loss and delayed hepatic recovery. This retrospective study demonstrates that internal bile reinfusion via the drainage catheter restores enterohepatic circulation more effectively than oral bile reinfusion. Internal reinfusion significantly accelerates bilirubin normalization, improves hepatic synthetic and nutritional status, reduces gastrointestinal symptoms, and enhances quality of life without increasing complications. Internal bile reinfusion represents a safe, practical, and superior postoperative bile management strategy after percutaneous transhepatic cholangiodrainage.

INTRODUCTION

Malignant obstructive jaundice is a common clinical problem in oncology and hepatobiliary surgery, affecting 10%-40% of patients with pancreaticobiliary malignancies (e.g., pancreatic cancer, cholangiocarcinoma, ampullary carcinoma, and metastatic disease)[1]. The disease is caused by mechanical insufficiency of bile flow, with cholestasis, hepatocellular failure, coagulopathy and progressive malnutrition. Left untreated with effective biliary decompression, patients develop hepatic failure, sepsis and/or high-grade deterioration in quality of life[1].

Therapeutic options include numerous approaches, with percutaneous transhepatic cholangiodrainage (PTCD) being an indispensable palliative measure when endoscopic retrograde cholangiopancreatography with stent placement has failed, is technically unfeasible because of altered anatomy or has been contraindicated by duodenal obstruction. PTCD is a feasible alternative with technical success rates greater than 95% and prompt reduction of serum bilirubin levels correlating positively with symptom relief. The procedure involves percutaneous introduction of a drainage catheter through liver parenchyma into dilated intrahepatic bile ducts under fluoroscopic or ultrasound guidance, creating external or internal-external drainages[2,3].

Although PTCD is effective in relieving the obstructive pathways, long-term external bile drainage leads to considerable physiological challenges. This results in continued loss of bile on a daily basis (500-1000 mL/day), which disrupts the enterohepatic circulation, an important physiologic mechanism leading to intestinal reabsorption and hepatic recirculation of bile acids. The interruption results in several negative effects: (1) Impaired absorption of liposoluble vitamins (A, D, E, K); (2) Maldigestion of dietary fats; (3) Imbalance of electrolytes (especially losses of sodium and chloride); (4) Progressive protein-energy malnutrition; and (5) Delay in the recovery of the hepatic function. Furthermore, loss of bile-related immunoglobulins causes intestinal immune deficiency and diminished bile acid pool leads to reduced hepatocyte synthetic ability and coagulopathy. These complications significantly affect the patient’s prognosis, tolerance to treatment and quality of life[4,5].

The physiological and rational use of bile reinfusion is emerging as a potential solution to overcome these complications by restoring enterohepatic circulation. There are mainly two strategies followed: (1) Oral reinfusion, that is collecting the external drainage and pouring it to the mouth of patients; and (2) Internal reinfusion, which characterized by directly returning bile into gastrointestinal (GI) tract through drain catheter after converting the external drainage to internal-external configuration. Despite the identical objective of restoring bile acid pool and preserving enterohepatic physiology, both techniques are fundamentally different in relation to delivery modality, bioavailability of bile acids, need for patient compliance and potential to preserve physiological pattern of bile flow[6,7].

Benefits associated with bile reinfusion have been described in the literature including improved nutritional parameters, enhanced hepatic function recovery and lower complication rates. However, few studies have systematically compared internal and oral reinfusion with the majority focusing on feasibility instead of comparative effectiveness. There is no definitive superiority between internal and oral reinfusion with respect to liver function recovery or normalization of cholestatic enzymes, nutritional status improvement, and quality of life enhancement in patients with malignant obstructive jaundice. In addition, questions remain about the best time to initiate, how long to continue, as well as the safety profile and technical aspects of bile reinfusion protocols[8,9].

We assessed the clinical effectiveness and safety of internal bile reinfusion vs oral bile reinfusion in patients with malignant obstructive jaundice after percutaneous transhepatic cholangial drainage (PTCD) operation. We theorized that internal reinfusion, which more closely approximates physiological bile delivery and eliminates external collection with oral administration, would result in: (1) Better recovery of hepatic function; (2) Greater improvements in cholestatic parameters and nutritional markers measuring absorption; (3) Enhanced quality of life measures; (4) Reduced GI symptoms; and (5) A reasonable safety profile compared to oral reinfusion.

MATERIALS AND METHODS

Study design and patient selection

This retrospective study was performed at our tertiary hepatobiliary institution to determine the clinical effectiveness and safety of internal bile reinfusion compared with oral bile reinfusion following PTCD in patients with malignant obstructive jaundice. Institutional Review Board approval was obtained for the study (Approval No. 2024-IRB-095) and the need for informed consent waived in light of its retrospective nature. One hundred patients were included from January 2020 to December 2023 with strict inclusion criteria: (1) Age between 18 and 80 years; (2) Histopathological confirmation of malignancy causing biliary obstruction; (3) Total bilirubin ≥ 85.5 μmol/L before PTCD; (4) Successful PTCD with adequate drainage; (5) Eastern Cooperative Oncology Group (ECOG) performance status ranging from 0 to 2; and (6) Complete follow-up data. In total, 100 consecutive patients [55 in observation group (internal reinfusion), 45 in control group (oral reinfusion)] were allocated to treatment according to evolution of institutional protocol; groups matched well for baseline characteristics including age, sex, tumor type, performance status, and hepatic function parameters.

Intervention protocols and outcome measures

Bile reinfusion began after spontaneous biliary drainage using PTCD within 5-7 days following clinical stabilization based on objective signs as follows: (1) Body temperature < 37.5 °C for two consecutive days; (2) Hemodynamic stability without the need for vasopressor support; (3) Bile drainage output > 300 mL/24 hours with clear appearance of bile drainage; and (4) Decreasing trend in serum total bilirubin over two measurements and no signs of acute abdominal pain or peritoneal signs. The observation group received the catheter for internal-external exchange, and progressive clamping was performed in 4 weeks to promote internal bile flow. The control group was provided with collected external drainage orally (150-450 mL/day) for the duration of 4 weeks. Volume ranges for oral reinfusion were determined by individual patient tolerance and GI symptoms. The initial volume of daily reinfusion was 150 mL, and it was titrated based on bile output and gut tolerance to a maximum volume of 450 mL per day. Once a thresholding volume was defined per patient (this commonly took 3-5 days), the volume remained stable for the duration of treatment. The mean volume per day was 285 ± 78 mL, and patients showed stable volumes after a period of adjustment. Primary endpoints were full hepatic function parameters (bilirubin, transaminases, cholestatic enzymes, albumin and coagulation) measured at baseline and at 1 week, 2 weeks, and 4 weeks after reinfusion. The secondary outcomes included nutritional status (prealbumin, hemoglobin, body weight), GI symptoms, quality of life [European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30)] and safety parameters.

Statistical analysis

Statistical analyses were accomplished with SPSS version 26.0 (IBM Corp., Armonk, NY, United States). Continuous variables were tested for normality using the Shapiro-Wilk test and are expressed as mean ± SD for normally distributed data or median (interquartile range) if non-normally distributed. The n (%) were used to describe categorical variables. Baseline characteristics were compared between the observation and control groups using independent samples t-tests for normally distributed continuous variables, Mann-Whitney U tests for non-normally distributed continuous variables, and χ² tests or Fisher’s exact test (when expected cell counts < 5) for categorical variables. Repeated measures analysis of variance with time as the within-subject factor and treatment group as the between-subject factor was used to analyze changes in hepatic function parameters, nutritional indicators, and quality of life scores over time. For cases where the sphericity assumption was violated (tested with Mauchly’s test), Greenhouse-Geisser correction was employed. Independent t-test with Bonferroni correction for multiple comparisons was performed for post-hoc pairwise comparisons at each time point. Multivariate logistic regression analysis was carried out to determine independent predictors of favorable hepatic function recovery, defined as total bilirubin < 51 μmol/L and albumin ≥ 35 g/L both at 4 weeks. Variables with P < 0.10 on the univariate analysis entered into the multivariate model using enter method. Results are presented as adjusted odds ratio (OR) with 95% confidence interval (CI). Statistical analyses were performed using a two-tailed approach and P < 0.05 was considered statistically significant. Handling of missing data points (< 3% for all variables) was approached with complete case analysis.

RESULTS

Baseline characteristics

In this study, there were a total of 100 patients enrolled; 55 into the observation group and 45 into the control group. The baseline characteristics were comparable between the two groups: (1) Mean age was 64.3 ± 10.5 years vs 63.8 ± 11.2 years (P = 0.812); and (2) Male ratio was 58.2% vs 55.6% (P = 0.786). The distribution of primary tumor types was similar (P = 0.972), with pancreatic cancer being the most common (52.7% vs 53.3%), followed by cholangiocarcinoma (29.1% vs 26.7%), ampullary carcinoma (10.9% vs 11.1%), and metastatic disease (7.3% vs 8.9%). Distribution of ECOG performance status was well balanced (P = 0.845), ECOG 0-1 proportion 69.1% vs 66.7%. There were no statistical differences in baseline hepatic function parameters including total bilirubin (152.3 ± 35.8 μmol/L vs 148.7 ± 38.2 μmol/L; P = 0.616), direct bilirubin (105.8 ± 28.5 μmol/L vs 102.3 ± 30.1 μmol/L; P = 0.542), alanine aminotransferase (ALT) (95.3 ± 28.5 U/L vs 92.7 ± 31.2 U/L; P = 0.653), aspartate aminotransferase (AST) (108.5 ± 32.3 U/L vs 105.2 ± 35.8 U/L; P = 0.625), alkaline phosphatase (385.5 ± 95.3 U/L vs 378.2 ± 102.5 U/L; P = 0.718) and albumin (32.5 ± 4.8 based on the indicator of the effectiveness of subsequent stability comparison between groups, we provide a solid foundation for Table 1).

Table 1 Baseline characteristics of study participants, n (%)/mean ± SD.

Characteristic	Observation group (n = 55)	Control group (n = 45)	P value
Age (years)	64.3 ± 10.5	63.8 ± 11.2	0.812
Male	32 (58.2)	25 (55.6)	0.786
BMI (kg/m2)	22.5 ± 3.2	22.3 ± 3.5	0.758
Primary tumor			0.972
Pancreatic cancer	29 (52.7)	24 (53.3)
Cholangiocarcinoma	16 (29.1)	12 (26.7)
Ampullary carcinoma	6 (10.9)	5 (11.1)
Metastatic disease	4 (7.3)	4 (8.9)
ECOG performance status			0.845
0-1	38 (69.1)	30 (66.7)
2	17 (30.9)	15 (33.3)
Baseline laboratory values
Total bilirubin (μmol/L)	152.3 ± 35.8	148.7 ± 38.2	0.616
Direct bilirubin (μmol/L)	105.8 ± 28.5	102.3 ± 30.1	0.542
ALT (U/L)	95.3 ± 28.5	92.7 ± 31.2	0.653
AST (U/L)	108.5 ± 32.3	105.2 ± 35.8	0.625
ALP (U/L)	385.5 ± 95.3	378.2 ± 102.5	0.718
Albumin (g/L)	32.5 ± 4.8	32.1 ± 5.2	0.690

ALT: Alanine aminotransferase; ALP: Alkaline phosphatase; AST: Aspartate aminotransferase; BMI: Body mass index; ECOG: Eastern Cooperative Oncology Group.

Primary outcomes

Compared with those in the control group, both groups exhibited a marked decrease in bilirubin levels after reinfusion of bile (all time points) (Table 2 and Figure 1), but those in the observation group showed markedly more favorable results. The temporal course leading to total bilirubin normalization was significantly different between cohorts, as internal reinfusion resulted in a more rapid and complete recovery. At 1 week post-reinfusion, total bilirubin was reduced to 98.5 ± 22.3 μmol/L in the observation group compared with 115.8 ± 26.5 μmol/L in controls (P = 0.001). This differential increased over time: (1) With 2 weeks, 68.3 ± 15.8 μmol/L vs 89.5 ± 21.2 μmol/L (P < 0.001); and (2) At 4 weeks, 42.5 ± 12.3 μmol/L vs 68.7 ± 18.5 μmol/L (P < 0.001). The reduction from baseline achieved by the observation group (72.1%) was significantly greater than that of controls (53.8%), resulting in a clinically meaningful 18.3% point difference. Direct bilirubin also improved in parallel, 28.3 ± 8.5 μmol/L vs 45.2 ± 12.7 μmol/L at 4 weeks (P < 0.001), suggesting that internal reinfusion led to increased resolution of cholestasis.

Figure 1 Temporal changes in total bilirubin levels. Changes in serum total bilirubin levels over time in patients receiving internal bile reinfusion (observation group, blue line with circles) vs oral bile reinfusion (control group, red dashed line with squares). The observation group demonstrated significantly greater reduction in bilirubin at all time points following initiation of reinfusion (P = 0.001 at 1 week; P < 0.001 at 2 weeks and 4 weeks). Error bars represent standard deviation. ^cP < 0.001 between groups.

Table 2 Temporal changes in bilirubin and hepatocellular enzymes, mean ± SD.

Parameter	Time	Observation group (n = 55)	Control group (n = 45)	Difference	P value
TBIL (μmol/L)	Baseline	152.3 ± 35.8	148.7 ± 38.2	3.6	0.616
	1 week	98.5 ± 22.3	115.8 ± 26.5	-17.3	0.001
	2 weeks	68.3 ± 15.8	89.5 ± 21.2	-21.2	< 0.001
	4 weeks	42.5 ± 12.3	68.7 ± 18.5	-26.2	< 0.001
ALT (U/L)	Baseline	95.3 ± 28.5	92.7 ± 31.2	2.6	0.653
	4 weeks	38.5 ± 12.3	56.8 ± 18.5	-18.3	< 0.001

ALT: Alanine aminotransferase; TBIL: Total bilirubin.

Recovery of hepatocellular enzymes followed the same trend favoring internal reinfusion. ALT and AST levels normalized significantly faster compared with the observation group. At 4 weeks, ALT level (38.5 ± 12.3 U/L vs 56.8 ± 18.5 U/L [P < 0.001]) and AST (42.3 ± 14.5 U/L vs 62.5 ± 20.3 U/L [P < 0.001]) were significantly lower in the observation group than in the control group. Normal ALT (< 40 U/L) was seen in 58.2% of the observation group compared with 33.3% of controls (P = 0.009). These data demonstrate more effective restoration of hepatocellular integrity in addition to attenuated ongoing hepatic injury with internal reinfusion (Table 2 and Figure 1).

Hepatic synthetic function and recovery of cholestatic enzyme

Internal reinfusion produced a greater improvement of cholestatic markers in line with better biliary obstruction resolution and bile flow (Table 3). Alkaline phosphatase levels declined in both groups but were significantly lower in the observation group after 4 weeks: 185.3 ± 42.5 U/L vs 248.7 ± 58.3 U/L (P < 0.001). The same pattern was also observed for gamma-glutamyl transferase: 125.8 ± 35.2 U/L vs 178.5 ± 48.7 U/L (P < 0.001). These differential improvements suggest that internal reinfusion more effectively restores physiological bile flow dynamics and mitigates cholestatic stress on hepatocytes.

Table 3 Cholestatic enzymes and synthetic function at 4 weeks, n (%)/mean ± SD.

Parameter	Observation group (n = 55)	Control group (n = 45)	P value
ALP (U/L)	185.3 ± 42.5	248.7 ± 58.3	< 0.001
GGT (U/L)	125.8 ± 35.2	178.5 ± 48.7	< 0.001
Albumin (g/L)	38.5 ± 4.2	34.2 ± 5.1	< 0.001
Normal albumin (≥ 35 g/L)	43 (78.2)	21 (46.7)	0.001
INR	1.15 ± 0.18	1.32 ± 0.25	< 0.001

ALP: Alkaline phosphatase; GGT: Gamma-glutamyl transferase; INR: International normalized ratio.

Hepatic synthetic function, assessed via albumin and coagulation parameters, improved significantly in both groups but was more apparent with internal reinfusion. Serum albumin rose from 32.5 ± 4.8 g/L to 38.5 ± 4.2 g/L at 4 weeks in the observation group, from 32.1 ± 5.2 g/L to 34.2 ± 5.1 g/L in controls (P < 0.001). The rates of normal albumin (≥ 35 g/L) were 78.2% vs 46.7% respectively (P = 0.001). International normalized ratio normalized more rapidly: 1.15 ± 0.18 vs 1.32 ± 0.25 at 4 weeks (P < 0.001), demonstrating enhanced vitamin K absorption and coagulation factor synthesis after supplementation. These results show that internal reinfusion restores the enterohepatic circulation more effectively, resulting in better absorption of fat-soluble vitamins and stimulation hepatitis protein synthesis (Table 3).

Secondary outcomes: Nutritional status and quality of life

The nutritional recovery in the observation group was better than that in the control group in all aspects, indicating a more effective restoration of enterohepatic circulation and nutrient absorption (Table 4). Prealbumin is a sensitive marker of nutritional status and hepatic synthetic capacity, with a short half-life of 2-3 days; in the observation group, prealbumin increased to 4 weeks from 185.3 ± 38.5 mg/L at baseline to 285.5 ± 45.3 mg/L (P < 0.001) compared with controls which saw an increase from 180.8 ± 42.3 mg/L to 215.8 ± 52.7 mg/L (P < 0.001). This large differential improvement (100.2 mg/L vs 35.0 mg/L increase) suggests considerable modulation of protein synthesis and nutritional recovery using internal reinfusion.

Table 4 Nutritional status and quality of life outcomes at 4 weeks, mean ± SD.

Outcome	Observation group (n = 55)	Control group (n = 45)	P value
Prealbumin (mg/L)	285.5 ± 45.3	215.8 ± 52.7	< 0.001
Hemoglobin (g/L)	118.5 ± 15.3	105.8 ± 18.5	< 0.001
Body weight change (kg)	-2.3 ± 1.5	-5.8 ± 2.3	< 0.001
BMI change (kg/m2)	-0.9 ± 0.6	-2.3 ± 0.9	< 0.001
EORTC QLQ-C30 global health	78.5 ± 9.2	62.3 ± 11.5	< 0.001
Physical functioning	82.3 ± 10.5	68.5 ± 12.8	< 0.001

BMI: Body mass index; EORTC QLQ-C30: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Core 30.

Body weight, a clinically relevant marker of overall nutritional status, was maintained better in the observation group. Mean weight loss at follow-up was significantly less: 2.3 ± 1.5 kg vs 5.8 ± 2.3 kg (P < 0.001) representing 3.5 kg differential preservation. Concurrently, body mass index decreased 0.9 ± 0.6 kg/m² vs 2.3 ± 0.9 kg/m² (P < 0.001). Hemoglobin levels, which reflect iron stores and nutritional status overall, also improved more significantly: 118.5 ± 15.3 g/L vs 105.8 ± 18.5 g/L at 4 weeks (P < 0.001). These all-encompassing nutritional advances led to improved quality of life and tolerance of treatment.

Quality of life, measured by EORTC QLQ-C30, was significantly improved in all functional and symptom domains within the observation group. At 4 weeks, global health scores were higher in the observation group (36.91% improvement from baseline) than in controls (22.00% improvement from baseline), with an average difference of 16.2 points (P < 0.001). Significantly higher physical functioning (82.3 ± 10.5 vs 68.5 ± 12.8; P < 0.001), role functioning (76.8 ± 11.2 vs 61.5 ± 13.5; P < 0.001) and emotional functioning recovery scores were also observed (75.5 ± 10 vs 63.2 ± 12.5; P < 0.001). Symptom scales showed corresponding benefits with reduced fatigue (32.5 ± 8.5 vs 48.7 ± 11.2; P < 0.001) and pain scores (28.3 ± 7.8 vs 42.5 ± 10.5; P < 0.001) in the observation group (Table 4).

GI symptoms and safety profile

Oral reinfusion was associated with significantly higher rates of GI adverse effects, which negatively influenced patient compliance and treatment intolerance (Table 5). Nausea was reported in 8 (14.5%) of the observation group vs 18 (40.0%) controls (P = 0.003), mostly within the first week after reinfusion. Vomiting was observed in 3 patients (5.5%) vs 12 patients (26.7%) in the control group (P = 0.003), and 2 control group patients needed antiemetic therapy. Diarrhea, a particularly worrying symptom, was present in 5 patients in the observation group (9.1%) vs 15 controls (33.3%) (P = 0.002), frequently leading to reduced oral intake and further worsening nutritional deficiencies. Abdominal discomfort occurred in 6 patients (10.9%) compared with 16 patients (35.6%; P = 0.002). The mean GI symptom score (0-10 Visual Analogue Scale) was significantly lower in the observation group: 1.8 ± 1.2 vs 4.5 ± 2.3 (P < 0.001), suggesting markedly better control of symptoms with internal reinfusion.

Table 5 Gastrointestinal symptoms during reinfusion period, n (%)/mean ± SD.

Symptom	Observation group (n = 55)	Control group (n = 45)	P value
Nausea	8 (14.5)	18 (40.0)	0.003
Vomiting	3 (5.5)	12 (26.7)	0.003
Diarrhea	5 (9.1)	15 (33.3)	0.002
Abdominal discomfort	6 (10.9)	16 (35.6)	0.002
GI symptom score (0-10)	1.8 ± 1.2	4.5 ± 2.3	< 0.001

GI: Gastrointestinal.

Safety profiles were similar between the groups, with comparable tolerability of both reinfusion approaches (Table 6). Catheter-related complications in 7 (12.7%) observation vs 6 patients (13.3%) controls (P = 0.926). Notable complications included: (1) Minor catheter site infections (4 patients vs 4 patients), successfully managed with local care and oral antibiotics; (2) Catheter obstruction (2 patients vs 1 patient), resolved by flushing; and (3) Catheter dislodgement (1 vs 1 patients), requiring replacement. No major catheter-related adverse events occurred in either group. Cholangitis episodes were rare: 2 patients (3.6%) in the observation group and 3 patients (6.7%) in controls (P = 0.661), all responded rapidly to intravenous antibiotic therapy, with no need for drainage revision.

Table 6 Safety profile and complications, n (%).

Adverse event	Observation group (n = 55)	Control group (n = 45)	P value
Catheter-related complications	7 (12.7)	6 (13.3)	0.926
Catheter site infection	4 (7.3)	4 (8.9)	0.760
Catheter obstruction	2 (3.6)	1 (2.2)	0.665
Catheter dislodgement	1 (1.8)	1 (2.2)	0.873
Cholangitis	2 (3.6)	3 (6.7)	0.661
Hyponatremia	3 (5.5)	8 (17.8)	0.048
Hypokalemia	2 (3.6)	7 (15.6)	0.038
Protocol completion	52 (94.5)	38 (84.4)	0.081

Electrolyte abnormalities were less frequent among the observation group, likely reflecting reduced GI losses. Hyponatremia occurred in 3 patients (5.5%) vs 8 patients (17.8%) (P = 0.048), all mild (sodium 130-135 mmol/L) and easily corrected with oral supplementation. Hypokalemia was seen in 2 patients (3.6%) compared with 7 patients (15.6%) (P = 0.038), equally mild and easily corrected. No patients had severe electrolyte disturbances that required hospitalization or IV correction. Adherence to treatment was higher in the observation group: 52 patients (94.5%) completed the total protocol of 4 weeks compared with 38 patients (84.4%) in controls (P = 0.081), with discontinuations mostly due to intolerable GI symptoms in controls (Tables 5 and 6).

Multivariate analysis of predictors of good hepatic recovery

Multivariate logistic regression analysis was carried out to evaluate independent predictors of good hepatic function recovery, defined as total bilirubin ≤ 51 μmol/L and albumin ≥ 35 g/L at 4 weeks after reinfusion (Figure 2). This composite endpoint reflects clinically significant hepatic recovery that is pertinent to chemotherapy eligibility and survival. Internal bile reinfusion was the most powerful independent predictor (adjusted OR = 9.25, 95%CI: 3.58-23.91; P < 0.001) that remained strongly significant also adjusting for all other potential confounders. These are the baselines and confounders that are eliminated by what we observe with internal reinfusion treatment effects.

Figure 2 Multivariate predictors of favorable hepatic function recovery. Forest plot illustrating independent predictors of favorable hepatic function recovery, defined as achieving both total bilirubin < 51 μmol/L and albumin ≥ 35 g/L at 4 weeks post-reinfusion. Internal bile reinfusion emerged as the strongest independent predictor (adjusted odds ratio [OR] = 9.25, 95% confidence interval [CI]: 3.58-23.91; P < 0.001), maintaining robust significance even after controlling for all potential confounders. Other significant independent predictors included baseline bilirubin < 170 μmol/L (adjusted OR = 3.82, 95%CI: 1.45-10.05; P = 0.007), Eastern Cooperative Oncology Group (ECOG) performance status 0-1 (adjusted OR = 3.15, 95%CI: 1.28-7.75; P = 0.012), and age < 65 years (adjusted OR = 2.68, 95%CI: 1.12-6.41; P = 0.027). Primary tumor type did not significantly predict outcomes (P = 0.412). The dashed vertical line represents OR = 1 (no effect). Data point size reflects the relative importance of each predictor.

Other independent predictors of interest were lower baseline bilirubin (< 170 μmol/L: Adjusted OR = 3.82, 95%CI: 1.45-10.05; P = 0.007), reflecting less severe cholestasis and preserved hepatic reserve; younger age (< 65 years: Adjusted OR = 2.68, 95%CI: 1.12-6.41; P = 0.027), likely related to a better regenerative capacity; and better ECOG performance status (0-1: Adjusted OR = 3.15, 95%CI: 1.28-7.75; P = 0.012), indicative of overall better physiological reserve. Importantly, primary tumor type was not significantly predictive of outcome (P = 0.412), suggesting the benefits of bile reinfusion were similar across various malignancy etiologies for obstructive jaundice.

DISCUSSION

We present that internal bile reinfusion through the drainage catheter significantly improves hepatic function recovery, nutritional status, and quality of life compared with oral bile reinfusion in patients suffering from malignant obstructive jaundice post-PTCD: Results from a retrospective study. These results show that internal reinfusion provides superior outcomes on all measured metrics, without a significant increase in adverse events, and lends further support for its use as the desired bile management strategy for this population.

More rapid return of hepatic functions was observed with internal reinfusion, which is consistent with basic physiological principles underlying enterohepatic circulation. Bile acids are overwhelmingly absorbed (> 95%) in the intestine, predominantly in the distal ileum through active transport and returned to the liver via the portal venous system where they activate hepatocyte activity and response as well as bile recycling. Through internal reinfusion, bile is introduced in a physiological concentration and flow rates into the proximal duodenum preserving this cycle preserved. By contrast, oral reinfusion presents multiple hurdles with gastric acid degradation of bile acids, altered pH regulating micelle formation, delayed gastric emptying and variable intestinal absorption. Studies have demonstrated that only 50%-70% of the orally administered bile acids are bioavailable compared to physiological delivery, thus accounting for the reduced efficacy we observed[10,11]. In addition to delivering bile acids directly, internal reinfusion most effectively restores the ileal-hepatic fibroblast growth factor 19 (FGF19) signaling axis, an important feedback control system for bile acid production. Bile acids signal enterocytes in the ileum to secrete FGF19, via farnesoid X receptor activation, allowing feedback control of hepatic cytochrome p450 family 7 subfamily A member 1 expression and bile acid synthesis. Internal reinfusion preserves a more physiological bile acid challenge to the terminal ileum, sustaining this FGF19-mediated feedback loop and further enhancing bile acid homeostasis. Oral reinfusion, with its variable bioavailability and real-time intermittent dosing, disrupts this finely tuned homeostatic system which has implications for the suboptimal hepatic function recovery seen in our study. Moreover, internal reinfusion stays closer to physiological bile acid exposure of gut bacteria and may thus preserve more microbial diversity and metabolic functions. In the gut, bile acids undergo extensive modification by the microbiome involving deconjugation and dehydroxylation among other transformations which leads to increased production of secondary bile acids with distinct signaling capabilities. With regards to microbiome metabolism, the depletion of both primary and secondary bile acids due to external drainage interferes with microbiome composition. Internal reinfusion provides continuous delivery of bile, preserving microbial communities compared to intermittent delivery by oral administration. Moreover, continuous delivery of bile acids through internal reinfusion provides better maintenance of intestinal barrier function. Activation of farnesoid X receptor by bile acids modulates tight junction proteins and strengthens epithelial integrity. Particularly, the diverse spectra of bile acid concentrations due to intermittent oral dosing disrupts the homeostasis that is a prerequisite for maintaining a healthy barrier and thus increases bacterial translocation into circulation and inflammation throughout the body. Internal reinfusion’s steady-state delivery is more effective at maintaining tight junction integrity, with a resultant decrease in infectious complications and improved nutritional absorption.

The 72.1% decrease in total bilirubin with internal reinfusion compared with a 53.8% decrease with oral reinfusion reflects clinically meaningful improvement. Normalization of bilirubin is important not just for symptom control but also in allowing the administration of chemotherapy, since many regimens require a bilirubin < 51 μmol/L for safe dosing. The 78.2% threshold achievement rate in the observation group compared with 46.7% in controls has important implications for treatment eligibility and potential survival. Moreover, the 73.3% drop of direct bilirubin after going for internal reinfusion suggests a more profound regression in cholestasis thus minimizing the possible hepatocellular injury due to bile acid back up[12,13].

Such augmented nutritional recuperation with internal reinfusion is highly clinically significant. Malignant obstructive jaundice leads to severe nutritional depletion via several mechanisms: Impaired fat absorption due to absence of bile acids, maldigestion, anorexia and cancer-associated cachexia[3]. Our observation group had significantly better prealbumin (285.5 mg/L vs 215.8 mg/L), less body weight loss (2.3 kg vs 5.8 kg) and better albumin recovery. Such improvements correspond to a more effective restoration of the enterohepatic circulation, an improved absorption of fat-soluble vitamins (mainly vitamin K that influences coagulation) and general metabolic homeostasis. Improved nutritional status indicates better treatment tolerance, fewer infectious complications and possibly greater survival[14,15].

Importantly, the significantly decreased GI symptoms with internal reinfusion is a major benefit for patient compliance and quality of life. The oral reinfusion subjects patients to the unpleasant experience of swallowing bile, which has a bitter taste to an extreme degree even with attempts at masking. Our control group had significantly higher nausea (40% vs 14.5%), vomiting (26.7% vs 5.5%) and diarrhea (33.3% vs 9.1%). These not only decrease quality of life but also affect compliance, with 15.6% of patients in the control group unable to finish the protocol. Internal reinfusion circumvents these issues completely, and secretes bile into the intestine without patient knowledge, thus increasing compliance and treatment efficacy[16,17].

Improvement in quality of life observed in our study should be highlighted. EORTC QLQ-C30 global health status improved to 78.5 in the observation group vs 62.3 in controls, consistent with substantial clinical significance (minimal important difference: 5-10 points). There were significant improvements in physical, role and emotional functioning. Lessening symptom burden – especially fatigue, pain and loss of appetite – contributed directly to improved well-being. Quality of life optimization is a crucial therapeutic objective in patients with advanced malignancy and limited life expectancy. The ability for internal reinfusion to enhance both objective and subjective outcomes makes it especially beneficial in this palliation population[18,19].

Internal vs oral reinfusion no longer poses a procedural risk, given that they have similar safety profiles. Although internal reinfusion does necessitate the catheter to switch between its internal-external configuration, this exchange can safely take place under fluoroscopy with only a small risk of complication. In our data, catheter-related complications (12.7% vs 13.3%) and cholangitis (3.6% vs 6.7%) occurred at similar rates. The lower rates of electrolyte abnormalities with internal reinfusion (5.5% vs 17.8% for hyponatremia) would represent reflect reduced GI losses. These results corroborate that higher efficacy of internal reinfusion is not counter-balanced by higher risk[20,21].

In multivariate analysis, internal bile reinfusion was the most significant independent predictor of favorable recovery of hepatic function (OR = 9.25) with strong effectiveness remaining after controlling for confounding factors. The consistency of benefits across several outcome domains – hepatic function, nutrition, symptoms, quality-of-life strengthens confidence in the treatment effect. Other factors such as baseline bilirubin, age and performance status were also associated with outcome but the magnitude of effect that could be attained by internal reinfusion far outweighed any associations with these other markers thus underscoring the more vigorous role of internal reinfusion in shaping a patients trajectory[22,23].

There are several empirical reasons for adopting internal reinfusion. The techniques described only require standard interventional radiology equipment with applicable expertise, which is available in most centers performing PTCD. Delivers simple protocols for clamping after initial catheter exchange setup. The convenience to the patient is significantly increased, as they do not need to collect and store bile or consume it orally. Healthcare resource utilization could be impacted by a decrease in the need for symptom management and improved treatment compliance. Factors such as these, together with better clinical outcomes, have provided a strong rationale for standard practice of internal reinfusion[24,25]. Internal reinfusion should only be attempted provided progressive clamping has been performed without adverse event (a definition of clinical stabilization is described in Methods). For clinicians interested in adopting internal reinfusion we propose the following protocol. Perform PTCD with standard external drainage initially with the aim of biliary decompression, once clinical stabilization is achieved (as defined by the criteria outlined), typically within 5-7 days from PTCD post-procedure exchange to internal-external drainage catheter under fluoroscopic guidance and continue an average initial progressive clamp schedule starting 2 hours twice daily week one followed by 4 hour twice daily week two, 8 hours twice daily week three and continuous clamped monitoring for biliary drainage by week four. Patients remain monitored for symptoms suggestive of infection (fever), abdominal pain or complications at the bile duct vicinity; tolerance to clamping assessed through control on bile output, vital signs and development of symptoms during this period. In cases where clamping intolerances develop (approximately 5% based on our experience) return continuously drain before resuming progressive clamp approach after resolution of presenting symptoms. Selection of patients should be based on ways that include ECOG performance status 0-2, good drainage connotation as well as no serious disease. This technique is simple and does not require any additional resources other than standard PTCD equipment; it could be safely applied in most centers performing percutaneous biliary drainage.

This study had several limitations. The retrospective design allows for selection bias; however, groups were well matched at baseline and treatment allocation was determined by the evolution of institutional protocol rather than physician preference. The uni-center nature, however may restrict generalizability even though our patient population and procedures are widely practiced at hepatobiliary centers. Short follow-up (4 weeks), while adequate to assess acute hepatic recovery, prevents long-term outcomes assessment. Prospective randomized controlled trials of longer duration are needed to confirm these results and determine clinical impact on survival rates and cancer treatment completion rates.

CONCLUSION

Systemic bile reinfusion through the drainage catheter markedly ameliorated hepatic function recovery, shortened bilirubin clearance time, enhanced nutritional therapy, diminished symptomatology and greatly improved quality of life compared with oral administration in malignant obstructive jaundice patients after PTCD. The intervention shows superior effectiveness in several clinically important domain outcomes with equivalent safety. These findings also provide support for the integration of internal bile reinfusion in standard perioperative care protocols for this population, significantly improve functional outcomes and quality of life while potentially improving treatment eligibility after biliary decompression. This method is safe, practical, can be done using routine interventional gadgets and shows high effectiveness that holds even after controlling for confounding factors. Internal bile reinfusion should be established as the optimal bile management strategy in patients with malignant obstructive jaundice after PTCD, as it is a significant advancement in palliative care for this difficult-to-manage population.