Left-sided hepatic resection for hepatolithiasis: a longitudinal study of 110 patients

Abstract

Background

Hepatolithiasis affecting the left hepatobiliary system is common in the Asia Pacific region. This aim of this study was to describe an experience with the diagnosis and treatment of patients with isolated left-sided hepatolithiasis.

Methods

One hundred and ten patients with isolated left-sided hepatolithiasis who underwent a left-sided hepatic resection between January 1999 and February 2010 were included for further analysis. The clinical profile, cholangiograms, operative procedures and early and late results were examined.

Results

Analysis of magnetic resonance cholangiopancreatography (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP) findings revealed left hepatic ductal anatomical details in 101 patients. Three types of left hepatic ductal variations were observed (type 1 in 90.1%, type 2 in 5.9% and type 3 in 4.0% patients). Eighty-four (76.4%) patients had the presence of strictures in the left hepatic ductal system and in 26 (23.6%) patients no strictures could be seen. Out of 84 patients with strictures, 78 could be classified (type I in 89.7% and type II in 10.3%). Of the 110 patients, 75 (68.2%) underwent a left lateral sectionectomy, 33 (30%) a left hepatectomy and 2 (1.8%) patients were treated with a left hepatectomy combined with a caudate lobe resection. Of the patients who underwent a left hepatectomy 11.4% developed a bile leak. Eight per cent of patients who underwent a left lateral sectionectomy had infective complications. During a median follow-up period of 63 (range 3–134) months, 2 (2.1%) patients were discovered to have residual stones and five (5.2%) others possessed recurrent stones.

Conclusions

This study highlights the impact of the presence or absence of bile duct stricture on the clinical and histological profile of patients as well as their operative and the post-operative behaviour. It is concluded that hepatic resection is an appropriate treatment modality in localized left-sided hepatolithiasis.

Introduction

Hepatolithiasis is characterized by the appearance of intraductal calculi proximal to the confluence of the right and left hepatic ducts. Hepatolithiasis is rare in the West but common in East Asia where it can be detected in 31–50% of patients undergoing surgery for cholelithiasis.^1,2 Aggressive treatment is necessary to prevent complications such as suppurative cholangitis, septicemia, secondary biliary cirrhosis, portal hypertension, bleeding varices and hepatic failure.^3,4 A hepatic resection seems to be an ideal procedure in selected patients because it, not only removes all the stones with associated strictures, but also reduces the risk of recurrent stones and eradicates the risk of a malignant change.^5,6 Hepatolithiasis is more frequent in the left lobe possibly because the left hepatic duct coalesces with the common hepatic duct at an acute angle which tends to induce bile stasis.⁷ Thus in localized left-sided hepatolithiasis, surgical resection has been considered as a potentially curative treatment.^8,9

The present study pertains to a series of patients who had isolated left-sided hepatolithiasis and were subjected to hepatic resection. It reviews the clinical observations, surgical procedures undertaken, histological findings and the post-operative outcome. It particularly focuses on the impact of biliary stricture on the clinical profile and patient management.

Methods

This retrospective study is derived from 110 patients suffering from left-sided hepatolithiasis who reported to the Department of Surgical Gastroenterology Sher-i-Kashmir Institute of Medical Sciences, between January 1999 and February 2010. All these patients were subjected to left-sided liver resection. Information was collected on the personal particulars of these patients, their presenting symptoms and the nature of treatment that they had received in the past. The extent and severity of the disease were evaluated in each patient using appropriate biochemical tests, imaging modalities including hepatic ultrasonography (USG), an abdominal spiral computed tomography (CT) scan, magnetic resonance cholangiopancreatography (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP) supported by histological examination.

Using radiological investigations, the system suggested by Cho et al.¹⁰ was adopted to describe the nature of the left biliary duct anatomy. In accordance with this system, type 1 anatomy is defined as the union of segment 2 and 3 ducts to form a single left lateral segmental duct, and thence with one or two segment 4 ducts to form the left hepatic duct. Type 2 anatomy was defined as a common trunk of segment 3 and 4 ducts joining the segment 2 duct to form the left hepatic duct. Type 3 was defined as a union of segment 2, 3 and 4 ducts directly forming the left hepatic duct.

A stricture was defined as a stenosis usually less than 2 mm in diameter within an intrahepatic duct distal to the stones, or when upon imaging during choledochoscopy, acute bile duct narrowing was observed leading to a dilated area containing stones. A gradual narrowing of the duct towards the dilated portion was considered to be a physiological tapering. Left hepatic duct stenosis was graded as a type 1 stricture when confined to one or two segmental ducts of the left liver located on the peripheral side to the bifurcation point of bile duct of segment 4, and usually associated with dilatation of segments 2 and 3 of liver. It was graded as a type II stricture when located close to the hilar confluence with dilatation of segments 2, 3 and 4 of liver, rarely involving caudate lobe ducts with dilatation or stone impaction. The presence or absence of a stricture was finally confirmed on gross and histopathological examination of the resected part of the liver.

Patients with left-sided hepatolithiasis with a pre-operative identification of a right posterior or anterior sectorial duct draining into left hepatic duct or imaging suggesting a pre-operative diagnosis of cholangiocarcinoma associated with intrahepatic stones were excluded from the study. The indications for hepatic resection included the following: the presence of stones leading to fibrosis or atrophy of the liver, the existence of fibrotic strictures, or the presence of cholangitic abscesses of the left hepatic duct or its secondary/tertiary branches. The International Hepato-Pancreato-Biliary Association (IHPBA) terminology was used for left-side liver segmentation and resection.¹¹

Post-operative outcome morbidity and mortality, macroscopic and histopathological findings regarding degree of fibrosis (central and interlobular); hyperplasia of peribiliary glands (intramural and extramural); and inflammation and the presence of dysplastic or malignant cells at the level of hepatic bile ducts, were duly recorded. In the presence of malignant transformation, staging was done as per TNM classification.¹²

Patients were followed by regular hospital visits and telephone contact (3 monthly for first 2 years, 6 monthly for next 3 years and yearly thereafter). Follow-up details were obtained by direct clinical evaluation, laboratory results and using MRCP and ERCP if indicated. Calculi appearing within the intrahepatic ducts within 3 months of a hepatic resection were treated as residual and those detected in the intrahepatic ducts, more than 3 months after surgery, were classified as recurrent stones as suggested by Lee et al.¹³ Data analysis was performed using the Statistical Package for Social Sciences (SPSS) version 11.5 software (SPSS Inc., Chicago, IL, United States). Yates corrected χ² test was applied for statistical analysis and a P-value of <0.05 was considered statistically significant.

Results

This series of 110 patients treated with left hepatic resection comprised 65 (59.1%) females. The median age was 33 years (range 19–52) The frequency of clinical presentation is shown in Table 1. Sixteen (14.5%) patients in the series had undergone surgical treatment in the past which included a cholecystectomy in 9, a cholecystectomy with common bile duct exploration in 3, a cholecystectomy with choledochoduodenostomy in 3 and a left hepatotomy in one patient. MRCP and CT scans (Fig. 1) were conducted in 86 (78.2%) and 18 (16.4%) patients, respectively. Diagnostic ERCP was performed in 62 (56.4%) patients. This included its therapeutic application in 16 (25.8%) patients entailing papillotomy in 5 (8.1%); clearance of common bile duct stones in 10 (16.1%); biliary stenting in 3(4.8%); and endoscopic nasobiliary drainage in 3 (4.8%) patients. Thirty-eight (34.5%) patients had both ERCP and MRCP.

Table 1.

Clinical profile of patients

Clinical presentation	No. of patients (n = 110)	%
Acute cholangitis	14	12.7
Abdominal pain	78	70.9
Jaundice	4	3.6
Acute pancreatitis	3	2.7
Liver abscess	4	3.6
Subphrenic abscess	1	0.9
Cholecystitis	3	2.7
Incidental	3	2.7
Associated conditions
Gall stones	22	20.0
Common bile duct stones	20	18.1
Ascariasis (live or dead)	8	7.3
Oesophageal varices	3	2.7
Splenomegaly	4	3.6

Figure 1.

Non-contrast computed tomography (CT) abdomen showing large hyperdense calculi in the left hepatic duct causing upstream dilatation of left biliary ductal system

Analysis of MRCP (Figs 2–4) and ERCP (Fig. 5) findings revealed anatomical details of a left hepatic duct in 101 patients, whereupon 3 types of variations were observed. Type I anatomy of the left hepatic duct was observed in 91 (90.1%) patients; type 2 anatomy in 6 (5.9%) and type 3 anatomy in 4 (4.0%). In 9 (8.2%) patients ductal details could not be defined. On screening the left hepatic ducts, 84 (76.4%) had the presence of strictures; however, in only 78 patients they were classified: type I strictures in 70 (89.7%) and type II in 8 (10.3%) (Fig. 6). In 75(68.2%) patients the stones were located in segments 2/3; in 33(30.0%) in segments 2–4; in 2(1.8%) patients segments 1–4 of the affected liver. Besides cholecystectomy, the other operative procedures performed on these patients are shown in Table 2. Ninety-four (85.5%) patients had atrophy of the affected liver parenchyma. Additional drainage procedures were performed in 23 (20.9%) patients, as a result of a main duct stricture or grossly dilated common bile duct. This included 13 (11.8%) choledochoduodenostomies, 2 (1.8%) hepaticojejunostomies and 8 (7.3%) t-tube drainage procedures. A patulous papilla at laparotomy was detected in 16 out of 23 patients who were subjected to additional drainage procedures.

Figure 2.

Magnetic resonance cholangiopancreatogram showing a stricture (type I) at the union of segment 2 and 3 intrahepatic ducts with upstream dilatation and presence of multiple filling defects within these liver segments. Note the presence of type 1 ductal anatomy

Figure 4.

Magnetic resonance cholangiopancreatogram showing isolated left intahepatic and extrahepatic ductal dilitation with multiple stones within them. Note the absence of left intrahepatic ductal stricture

Figure 5.

Balloon inflated endoscopic retrograde cholangiogram showing isolated dilitation of the left intrahepatic ductal system with multiple stones and strictures(type II). Note the presence of type 3 ductal anatomy

Figure 6.

Diagram showing various types of anatomical and stricture details of the left intrahepatic biliary ductal system

Table 2.

Surgical procedures undertaken

Surgical procedure	No. of patients (n = 110)	%
Left lateral sectionectomy (S2,3)	75	68.1
Left hepatectomy (S2,3,4)	33	30.0
Left hepatectomy (S2,3,4) + caudate lobe resection	2	1.8
Bile duct drainage	23	20.9
T-tube drainage	8	34.8
Choledochoduodenostomy	13	56.5
Hepaticojejunostomy	2	8.7

Figure 3.

Magnetic resonance cholangiopancreatogram demonstrating dilated left intrahepatic and extrahepatic ductal system containing multiple stones. Note the presence type 3 ductal anatomy with a stricture (type II) at the conflucence of left intrahepatic ducts

Post-operative complications in patients who underwent a left lateral sectionnectomy or a left hepatectomy are presented in Table 3. The frequency of complications was relatively more after a left hepatectomy; a bile leak being the main complication observed among four patients.

Table 3.

Post-operative complications by the type of surgery

Complications	Left lateral sectionectomy (n = 75)		Left hepatectomy (n = 35)
	No	%	No
Bile leaks	1	1.3	4
Wound infection	5	6.7	2
Intra-abdominal pus collection	1	1.3	2
Pulmonary atelectasis	2	2.7	2
Pleural effusion	1	1.3	2

The profile of stricture positive patients in comparison with that of the stricture negative patients is presented in Table 4. Histopathological findings in stricture positive patients revealed predominant fibrosis with mild hyperplasia of the peribiliary ductal glands. In the stricture negative patients, hyperplasia of the peribiliary ductal gland with mild fibrosis was the dominant finding (Figs 7,8).

Table 4.

Comparative profile of stricture positive and stricture negative patients

Features	Group I	Group II	P-value
	Stricture +	Stricture −
	No (%) (n = 84)	No (%) (n = 26)
Presentation
Acute cholangitis	4 (4.8)	10	0.000
Abdominal pain	70 (83.3)	8	0.000
Acute pancreatitis	0 (0.0)	3	0.014
Jaundice	0 (0.0)	4	0.002
Liver abscess	4 (4.8)	0	0.593
Subphrenic abscess	1 (1.2)	0
Associated condition
Gall stones	10 (11.9)	12	0.000
Common bile duct stones	4 (4.8)	16	0.000
Ascariasis (live or dead)	2 (2.4)	6	0.002
Oesophageal varices	0 (0.0)	3	0.014
Splenomegaly	0 (0.0)	4	0.002
Patulous papilla
Intrahepatic stone location
Segments: 2,3	70 (83.3)	5	0.000
Segments: 2,3,4	12 (14.2)	21	0.000
Segments: 1,2,3,4	2 (2.3)	0	1.000
Operative procedures indicated
Left lateral sectionectomy (S2,3)	70 (83.3)	5	0.000
Left hepatectomy (S2,3,4)	12 (14.2)*	21**	0.000
Left hepatectomy (S2,3,4) + caudate lobe resection	2 (2.3)	0	1.000
Bile duct drainage	7 (8.3)	16	0.000
Follow-up development
Residual stones	0 (0.0)	2	0.084
Recurrent stones	1 (1.2)	2	0.276

Includes 6 (*) and 3(**) patients where pre-operative ductal anatomy was not defined.

Figure 7.

Photomicrograph of a resected liver specimen showing extensive periductal fibrosis (grade 3) of a medium-sized bile duct with mild inflammation

Figure 8.

Photomicrograph of a resected liver tissue showing a pigmented stone within a large bile duct with surrounding tissue showing mild fibrosis, extensive peribiliary ductular glandular hyperplasia (grade 3) with formation of lymphoid follicles. The surrounding liver parenchyma shows mild portal triaditis

Of the 110 patients, 14 (12.7%) were lost to follow-up, 5 died during follow-up (cholangiocarcinoma 2, acute septic cholangitis 1,variceal bleeding 1 and myocardial infarction 1) and the remaining 96 were followed up for a median time period of 63 months (range 3 to 134 months). Sixty-eight (95.7%) out of 71 patients in the stricture positive group and 16 of the 20 patients within the stricture negative group remained asymptomatic. However, 3 patients (2 recurrent stones, 1 residual stones) in stricture positive group and 4 patients (3 recurrent stones, 1 residual stones) in stricture negative group reported with symptoms on follow-up. In stricture positive group, segment 1 was involved in two patients (1 each with recurrent and residual stones) and in third patient recurrent stones appeared in the common bile duct. In the stricture negative group, recurrent stones were detected in the main right hepatic duct (2 patients) and in the common bile duct(1 patient). Furthermore, in 1 patient residual stones were seen in the common bile duct. Persistent attacks of pain with recurrent stones in segment 1 necessitated partial excision of the segment (Spiegel's lobe of liver); the patient with a residual stone in segment 1 was followed up without intervention. As to the remaining 5 patients with stones, 2 was subjected to endoscopic extraction and the other 3 to re-operative surgery involving bilio-digestive anastomosis (2 hepaticojejunostomies and 1 choledochoduodenostomy).

Discussion

In hepatolithiasis, the goal of treatment is to provide complete stone clearance, decompression of the biliary tree and control of infection; thus the management has to be individualized according to different presentation of the disease. Treatment options include surgical treatment with biliary decompression and/or hepatic resection as the first approach.^14,15 Non-surgical procedures such as percutaneous transhepatic lithotriphy or endoscopic therapy are usually reserved to treat stone recurrences.^16,17 Morbidity and mortality rates for hepatic resection have decreased significantly in the last decade.^5,18 Thus hepatic resection (especially for unilateral disease) has been considered as a potentially curative treatment.^5,9,18

Many patients with hepatolithiasis have ductal dilatation without any strictures. Koga et al.¹⁹ observed that 37% of patients with hepatolithiasis showed dilatation of bile ducts without biliary stricture. Azuma et al.²⁰ described two groups of patients: 25 with strictures and 4 without strictures. The present study covers 84 patients with strictures and 26 without strictures. Such observations make it difficult to explain the cause of hepatolithiasis on the basis of biliary stricture alone. Although the two groups (stricture positive and stricture negative) in the present study share similarities in several aspects, major difference were observed in modes of presentation, histological features and final outcome (Table 4).

The frequent association of stones in the gall bladder and common bile duct and the recurrent involvement of segments 2, 3 and 4 of liver in the stricture negative patients demand a left hepatectomy with common bile duct drainage. As segments 2 and 3 of the liver are mostly involved in the stricture positive patients, a left lateral sectionectomy is an ideal procedure. Nevertheless, concerns have been raised about injury to the left medial section of the segment 4 hepatic duct during a left lateral segmentectomy and because of this some authors have recommended a left hepatectomy instead. In a comparison study between the two groups, one which underwent a left lateral sectionectomy and the other a left hepatectomy, Sun et al.²¹ reported a higher incidence of residual and recurrent stones in the first group and accordingly recommended a left hepatectomy in patients with left-sided hepatolithiasis. In a similar study, Kim et al.²² reported that the incidence of complications (bile leakage, abscess and appearance of inflammatory pseudotumors) after a left hepatectomy were significantly lower than after a left lateral sectionectomy and suggested that where the anatomy of a medial segmental bile duct was not identifiable pre-operatively, a left hepatectomy should be considered. Lee et al.,²³ in an extensive study of left-sided hepatic resection in 181 patients, also documented higher rate of bile leak in the left hepatectomy group (7.6%) as compared with the left lateral sectionectomy group (3.8%). In this study out of the five patients who had post-operative bile leakage, four were subjected to a left hepatectomy and one to a left lateral sectionectomy.

A higher incidence of bile leak in a left hepatectomy is ascribed to: (i) small bile duct openings at the transected surface that remain open for a long time owing to inflammation; and (ii) injury caused to the caudate lobe and the segment 4 bile ducts which drain into the left hepatic duct. A thorough pre-operative study of the anatomy of the left ductal system, especially of segment 4, is therefore mandatory. A careful search for bile ducts and their meticulous closure is also advised to prevent bile leak. Pertinently hepatectomies that include segment 4, usually exposed the major glissonian sheath and hepatic hilum on the cut surface, thereby increasing the risk of damage to the bile duct wall. Thus whenever a left hepatectomy is contemplated, a careful dissection of the glissonian sheath should be performed by lowering the hilar plate, which can help prevent bile duct injury.

Both recurrent and residual stones that appear after treatment for hepatolithiasis are a source of serious concern. In such patients Lee et al.²⁴ and Jan et al.²⁵ reported a recurrent stone rate of 4.2% and 9.5%, respectively. In the current series residual and recurrent stone rates were observed in1.8% and 2.7% patients, respectively; in three patients, the stones were located in the common bile duct and were effectively treated by endoscopic or re-operative surgical methods. In one patient, the recurrent stones were located in Spiegel's part of the caudate lobe of liver, which needed surgical resection. The majority of Spiegel's lobe ducts drain into the left hepatic ducts. To prevent recurrent or residual stones or even bile leaks affecting the Spiegel's lobe, it may be safer to resect this part of the caudate lobe when a left-sided hepatic resection is intended. In this study, caudate lobe involvement was observed in four patients: pre-operatively in 2 and on follow-up in the other 2. Hwang et al.²⁶ noticed 12 patients with stricture close to hilar confluence; to prevent complications such as bile leak, they resected caudate lobe in 7 patients. It is the authors' opinion that in patients with type II strictures, caudate lobe resection along with removal of segment 2, 3 and 4 may be advisable as was done in two patients in the present study. This not only reduces the risk of post-operative bile leak but also prevents any recurrence in that part of liver. It may be noted in this study the frequency of post-operative bile leaks and infective complications, also, long-term after effects such as residual or recurrent stone formation was lower than observed in various published reports.

It is not uncommon to find a lax sphincter of Oddi (SO) in patients with hepatolithiasis;²⁷ a choledochoscope can pass easily through the SO into the duodenum without resistance. However, whether these patients have SO dysfunction or not, and if so, what is the cause and effect relationship between SO dysfunction, and the presence of hepatolithiasis remains unclear. It is imaginable that recurrent passage of stones may injure the SO, leading to SO dysfunction in the long run. In the present study, a patulous papilla at a laparotomy was detected in 14 of 16 stricture negative patients.

Kashmir is an endemic zone for ascariasis; the role of biliary ascariasis in hepatolithiasis has been well documented.²⁸ In the present study, 23.1% patients in the stricture negative group had biliary ascariasis as opposed to 2.4% patients in the stricture positive group. This higher rate of ascariasis in the stricture negative group may be attributed to the existence of patulous papilla in this group that allows easier passage of worms into the bile duct.

In the stricture negative group, hepatolithiasis was frequently (61.5%) found to be associated with common bile duct stones. Repeated acute episodes of cholangitis are often seen in such patients. Without treatment, such an infection can lead to liver abscess, secondary biliary cirrhosis, portal hypertension and death from sepsis or hepatic failure. Scheuer et al.²⁹ reported that secondary biliary cirrhosis develops 7 years after the onset of obstruction from a stricture; 4.5 years after a calculous obstruction and 10 months after the onset of malignant strictures. The presence of portal hypertension in three patients within the stricture negative group may be attributed to the frequent involvement of the extra hepatic duct by stones.

The predominant presence of fibrosis in the stricture positive group and hyperplasia of peribiliary ductal glands in the stricture negative group may be crucial in understanding the pathogenesis of this disease. However, the striking differences observed in this study between the stricture positive and stricture negative groups needs further study. A comprehensive long-term study is indicated to verify these facts, also to unfold the pathogenesis of stricture development to explain the contrasting profile of the two groups of patients.

Conclusion

On the basis of this study it may concluded that a hepatic resection is an appropriate treatment modality in localized left-sided hepatolithiasis as (i) it is a safe and effective procedure that ensures the removal of the involved liver segments with strictures, dilatations or stones; (ii) it contributes towards the prevention of a malignant change; and (iii) it secures an acceptable long-term treatment outcome. This study highlights the impact of the presence or absence of bile duct stricture on the clinical and histological profile of patients as well as their operative and the post-operative behaviour. Furthermore, for excellent short- and long-term results, it is advisable to subject the non-stricture group of patients with left-sided hepatolithiasis involving 2, 3 and 4 segments of liver to a left hepatectomy and the stricture positive group of patients mostly having 2 and 3 segment involvement, to a left lateral sectionectomy. However, in the type II stricture positive group, caudate lobe involvement may necessitate resection of segments 1, 2, 3 and 4.

Conflict of interest

None declared.