Sclerosing cholangitis in critically ill patients: an important and easily ignored problem based on a German experience

Abstract

Intensive care unit (ICU) is important in the rehabilitation of critically ill patients. In the past decades, many patients who received aggressive treatment in ICU developed sclerosing cholangitis in multiple centers. Sclerosing cholangitis in critically ill patients (SC-CIP) is a relatively new issue. To investigate the causes, clinical manifestation, treatment, and prognosis of SC-CIP, we searched for published cases in the databases of PubMed, Highwire, and Elsevier from 2001 to 2012. Data were extracted using a standard form and retrospectively analyzed. Twelve eligible studies covering 88 patients, with 64 men and 24 women, were enrolled in this analysis. The mean age was 49.8 years. All of the patients recovered from critical illnesses, such as trauma, infection, burn, and major surgeries. High pressure positive end-expiratory pressure (PEEP, peak level at 12.8 cm H₂O) was utilized for all patients, with the average duration of 36.3 d. In addition, vasopressor agents were administered in approximately 60% of SC-CIP. A rapid increase in cholestasis and irregular strictures in the intrahepatic bile ducts was observed in the following months. With an average follow-up period of 17.9 months, poor outcomes were observed in 54 patients, including 34 deaths. In conclusion, ischemic injury of the biliary tree, which may be affected by PEEP and/or vasopressor administration, affects cholangiopathic procedure. As a newly discovered type of secondary sclerosing cholangitis, SC-CIP is a severe progressive complication of patients in ICU and should be carefully monitored by clinicians.

Keywords

intensive care unit / sclerosing cholangitis / ischemic injury / prognosis / systemic review

Introduction

Sclerosing cholangitis represents a series of variably progressive cholestatic diseases of the intrahepatic and/or extrahepatic biliary system; this condition finally develops to end-stage liver disease. No apparent differences are observed between primary sclerosing cholangitis (PSC) and secondary sclerosing cholangitis (SSC) in terms of pathological and clinical manifestation. In particular, PSC and SSC exhibit similar clinical manifestation of irregular biliary strictures and biliary dilatations, which are usually complicated with progressive hepatic fibrosis. In contrast to PSC, SSC is commonly caused by biliary obstruction, surgical trauma, and ischemic injury of the bile duct during liver transplantation.

Sclerosing cholangitis, which occurs after a long-term treatment in ICU or sclerosing cholangitis in critically ill patients (SC-CIP), has been extensively studied. SC-CIP is characterized by fibrous stenosis and occlusion of the bile duct after treatment in ICU. These characteristics exclude biliary obstruction, biliary tract surgery, or other relevant history of hepatobiliary diseases before treatment in ICU. SC-CIP progresses rapidly with poor prognosis. This study aimed to analyze the clinical data and provide a comprehensive summary of the disease.

Materials and methods

Source of data

Data from available medical studies were systematically reviewed and pooled for analysis. Relevant studies written and published in the English language were retrieved using the following search terms: “sclerosing cholangitis,” “intensive care unit,” “sepsis,” and “trauma.” MEDLINE (US National Library of Medicine, Bethesda, MD), EMBASE (Elsevier Science, New York, NY), and HIGHWIRE (Highwire Press, Stanford, CA) bibliographic databases were included in the search. Relevant article references were also collected (from January 2001 to June 2012).

Data extraction and quality control

A medical information scientist (Dr. Kai Qu) performed literature retrieval and initial screening and a medical doctor (Dr. Ting Lin) reviewed and coded the studies. Irrelevant studies were removed by checking the title and abstract. Relevant studies were scrutinized to exclude any insufficient data or parameters. All of the selected documents were downloaded from the journal home page or obtained by direct contact and communication with the author. Discrepancies were resolved, and the review was iterated until a high reliability was obtained.

Database

The collected cases were evaluated individually. Details were extracted for further analysis. Coded potential prognostic determinants included patient demographics, clinical and laboratory findings, prognosis, diagnosis, and treatment methods. The obtained data were pooled at aggregate and patient levels to investigate the effect of primary disease, mechanical ventilation, vasopressor administration, and other pertinent variables.

The liver function test of the SC-CIP cases was analyzed, but incomplete results were obtained from every patient. We selected the patients with complete results and showed the results as cohort A. We excluded studies or individuals with missing data from specific analyses. As a result, the number of patients in cohort A varied. For instance, Gelbmann et al. [1] (29 cases) and Kulaksiz et al. [2] (26 cases) reported only aggregate results. We considered the 29 cases reported by Gelbmann et al. [1] as cohort B and the 26 cases reported by Kulaksiz et al. [2] as cohort C. We compared the three cohorts (from cohort A to cohort C) to observe the differences among populations.

Statistical analysis

The collected data were transcribed in a Microsoft Excel spreadsheet and analyzed using SPSS version 14.0 (SPSS Inc. Chicago, IL). Continuous variables were expressed as \overline{x}±s and categorical variables were expressed as percentage. Continuous variables were compared using student’s t-test analysis and categorical variables were compared using χ² test. P<0.05 was considered significantly different.

Results

Baseline characteristics of SC-CIP

Excluding irrelevant journal or literature, we collected a total of 12 articles covering 97 cases from 217 publications in international journals [1–12]. Another article was included as supplement by searching among the cited references, adding the consequent number of selected articles to seven. Nine cases [5] were included in another article reported by the same center [2], but these articles were not gathered repeatedly for analysis. Hence, 88 patients were selected. Among these patients, 64 were males and 24 were females with an average age of 49.8 years (Table 1). The global distribution of the reported cases was of high importance. Although the cases of SC-CIP were reported by researchers from different regions worldwide, the greatest portion (96.6%) was reported by centers located in Germany (9 articles, including 85 cases; Fig. 1).

Primary diseases of SC-CIP

The statistics of the primary diseases of SC-CIP showed emergency diseases in 55 patients (62.5%), post major surgery in 24 patients (27.3%), and other diseases in 9 patients (10.2%). Among these emergency diseases, trauma in 24 cases (27.3%), infection in 23 cases (26.1%), and burn in 8 cases (9.14%) were found. Among the patients who suffered from major surgeries, 22 patients (25.0%) were subjected to cardiothoracic surgery and 2 patients (2.3%) were subjected to abdominal surgery. Other diseases were observed in 9 cases (10.2%), including cerebral hemorrhage in 3 cases, pancreatitis in 2 cases, HELLP syndrome in 1 case, acute myocardial infarction in 1 case, status asthmaticus in 1 case, and rhabdomyolysis in 1 case. The detailed data from eachβreport were presented in Table 2.

Evaluation of ICU-related factors eliciting a possible effect on SC-CIP

Complete intensive care records were available only for 66 patients (Table 3). All of these patients (66/66) in ICU accepted positive end-expiratory pressure (PEEP) treatment, with average treatment duration of 36.3 d. The mean pressure of PEEP treatment was 12.8 cm H₂O. In addition, vasopressor administration, as a common treatment to achieve hemodynamic stabilization, was applied in 39 cases (59.1% of the SC-CIP) to control life-threatening illnesses (catecholamines single or combined with other drugs).

Liver function parameters of SC-CIP

We observed a rapid increase in cholestasis parameters in patients, particularly total bilirubin (TB), alkaline phosphatase (ALP) and γ-glutamyltransferase (GGT). TB reached the peak level of 12.5±9.4, 12.0±10.6, and 10.4±9.1 mg/dl in cohorts A, B, and C, respectively (normal range= 0 mg/dl to 1.0 mg/dl). SC-CIP exhibited>10 times higher TB level than normal people. No significant difference was determined among the three cohorts (Fig. 2A). Similar results were observed in ALP and GGT levels. ALP reached the peak level of 1398.7±871.4, 687.0±368.0, and 1182.2±1058.3 U/L in cohorts A, B, and C, respectively (normal range= 0 U/L to 115.0 U/L; Fig. 2B). Although the average ALP value of cohort B was lower than that of the two other cohorts, this parameter was six times higher than that of normal people. GGT reached 885.4±576.1 and 641.0±383.0 U/L in cohorts A and B, respectively. These results were approximately 10 times higher than normal (normal range= 0 U/L to 48.0 U/L; Fig. 2C). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) results were available only in 20 SC-CIP cases, whose obtained ALT and AST levels mildly increased to 140.0±151.9 and 138.6±101.3 U/L, respectively. These results were three times higher than normal (normal range= 0 U/L to 40.0 U/L; Fig. 2D and 2E).

Diagnosis and treatment of SC-CIP

SC-CIP is a rare disease of unknown pathophysiology but with distinct characteristics in endoscopic retrograde cholangiopancreatography (ERCP). The 42 patients with records received ERCP and exhibited different manifestations at various periods of the disease. In the first few weeks of the bile duct injury, ERCP showed that the intrahepatic bile ducts were filled with bile cast. Although a small amount of mold residue remained visible in the larger bile duct after several weeks, the peripheral bile duct manifested stenosis and inflammation, aggravating to complete occlusion [1] (Fig. 3). A total of 33 patients (78.6%) who underwent liver biopsy confirmed this pathological process (Table 4).

Endotherapy combined with ursodeoxycholic acid (UDCA) administration was recommended for sclerosing cholangitis. Endotherapies, including endoscopic dilatations, endoscopic sphincterotomy of the papilla, and sludge extraction, which aimed to improve bile duct drainage, were performed on 56.2% of the patients (18/32). Antibiotic drugs were also used to control infection in 64.3% of the patients (27/42). UDCA treatment was performed in 19 patients (Table 4). Many cases [3,5,9,10] demonstrated liver function improvement after treatment at doses of 10 mg·kg^-1·d^-1 to 15 mg·kg^-1·d^-1.

Prognosis

During the average 17.9 months of follow-up of the remaining 87 cases (one patient was lost), death occurred in 34 cases (39.1%); among these cases, 31 died of hepatic failure (35.6%). A total of 53 patients survived (60.9%); among the survivors, 31 patients with cholestatic cirrhosis were in a stable condition and another 20 cases were considered for liver transplantation. To date, nine patients continuously wait for liver transplantation. Poor outcomes were found in 54 patients (including death and liver transplantation; 62.1%). The detailed data were presented in Table 5.

Discussion

SSC exhibits a definite etiology, including immune injury, obstruction, and chemotherapy [13,14]. SC-CIP, as a newly developed SSC, has been extensively studied. Many relevant cases have been reported since the first description of Scheppach et al. [3]. Individuals have attributed the etiology to various primary diseases, and treatment in ICU as an independent cause has been gradually accepted. However, no relevant studies and description have been conducted yet.

The symptoms of SC-CIP emerge early and progress rapidly with a poor prognosis. The levels of TB, ALP, and GGT reach 10 times higher than normal levels. End-stage liver diseases and even liver failure likely occur if these patients fail to undergo appropriate treatment. During the follow-up period of the 87 patients, 34 cases died of liver failure. The average survival time was only 1.1 year, particularly when no liver transplantation was performed [2].

Biliary ischemia is considered as a direct cause of sclerosing cholangitis [1]. In contrast to the liver, the biliary tract accepts blood supply only from hepatic artery. The formation of a capillary network of hepatic artery branches around the hepatic ducts and the intrahepatic bile ducts constitutes the anatomical basis of biliary ischemic injury [15]. Our results showed that 66 patients underwent high-level PEEP (average peak level of 12.8 cm H₂O), with an average treatment time of 36.3 d. As one of the commonly used modes of assisted ventilation, PEEP aims to increase the functional residual gas volume and oxygen levels to improve lung function. However, studies have shown that high PEEP levels (>10 cm H₂O) can reduce cardiac output and venous return, thereby increasing the risk of visceral ischemia [16]. Previous reports also demonstrated gastrointestinal ischemic complications caused by PEEP [17]. The blood supply of liver in critically ill patients is affected by primary diseases, particularly when such diseases are complicated with sepsis [18]. Therefore, the appropriate level of PEEP should be further investigated. In our retrospective analysis, approximately 60% of SC-CIP received vasopressor administration. Relevant studies have shown that the combined use of vasoconstrictor and PEEP may aggravate visceral ischemia [19].

Considering that no specific pathological manifestation has been described, we should apply ERCP as the primary means of diagnosis once the disease is suspected. Among the currently limited treatments, endoscopic therapy, including endoscopic balloon dilation, stenting, and nasobiliary drainage, can significantly alleviate the clinical symptoms and biochemical parameters in patients with sclerosing cholangitis [20]. Studies have reported that maintaining the smooth drainage of the biliary tract in the early stage of the disease prevents secondary infection and alleviates or even reverses the disease [2]. Moreover, adequate dose of antibiotics is important. For patients with end-stage liver disease, in which the obstruction caused by hardened biliary tract is difficult to avoid, liver transplantation should also be considered [21].

UDCA is a hydrophilic, tertiary bile acid used to treat various chronic cholestatic diseases. In several clinical studies, UDCA exhibits potency for the treatment of PSC [22,23]. Lindor et al. [24] also conducted a randomized controlled trial of high-dose UDCA (28 mg·kg^-1·d^-1 to 30 mg·kg^-1·d^-1) for five years in 150 patients with PSC. The results demonstrated that high-dose UDCA therapy is associated with improved liver function, but survival remains low. Other studies have also shown that the high-dose UDCA treatment is associated with increased risk of colonic neoplasia [25]. Considering that no randomized controlled trial has evaluated the therapeutic value of UDCA in patients with SSC, we administered low-dose UDCA (10 mg·kg^-1·d^-1 to 15 mg·kg^-1·d^-1) to these patients. In addition, the combined endoscopic and medical treatment with UDCA is beneficial for patients with SSC [26].

Therefore, SC-CIP should be prevented considering the lack of effective treatment. Although SC-CIP is a rare kind of serious cholestatic disease, approximately 40% of patients in ICU present varied degrees of jaundice [27,28]; however, whether or not this condition likely develops to SC-CIP remains unknown. Hence, any stimulus, particularly high levels of PEEP and vasoconstrictor, that potentially induce biliary ischemia, should be carefully monitored. At present, acid-suppressing agents are routinely administered to severely affected patients to prevent bleeding caused by stress ulcer [29].

In conclusion, SC-CIP, which mostly occurs in middle-aged patients at rapid progression and high mortality, seriously affects the prognosis of patients in ICU. Ischemic injury of the biliary tree induced by PEEP is important in cholangiopathic procedure. Therefore, specific manifestation on ERCP should be considered to facilitate diagnosis. Among the currently limited treatments, endoscopic treatment combined with drug therapy is the first choice. Clinicians, particularly ICU physicians, should enhance and develop procedures that prevent SC-CIP. Hence, further studies should be conducted to develop such procedures.

References

[1] Gelbmann CM, Rümmele P, Wimmer M, Hofstädter F, Göhlmann B, Endlicher E, Kullmann F, Langgartner J, Schölmerich J. Ischemic-like cholangiopathy with secondary sclerosing cholangitis in critically ill patients. Am J Gastroenterol2007; 102(6): 1221–1229

[2] Kulaksiz H, Heuberger D, Engler S, Stiehl A. Poor outcome in progressive sclerosing cholangitis after septic shock. Endoscopy2008; 40(3): 214–218

[3] Scheppach W, Druge G, Wittenberg G, Mueller JG, Gassel AM, Gassel HJ, Richter F. Sclerosing cholangitis and liver cirrhosis after extrabiliary infections: report on three cases. Crit Care Med2001; 29(2): 438–441

[4] ter Borg PC, van Buuren HR, Depla AC. Bacterial cholangitis causing secondary sclerosing cholangitis: a case report. BMC Gastroenterol2002; 2(1): 14

[5] Engler S, Elsing C, Flechtenmacher C, Theilmann L, Stremmel W, Stiehl A. Progressive sclerosing cholangitis after septic shock: a new variant of vanishing bile duct disorders. Gut 2003; 52(5): 688–693

[6] Benninger J, Grobholz R, Oeztuerk Y, Antoni CH, Hahn EG, Singer MV, Strauss R. Sclerosing cholangitis following severe trauma: description of a remarkable disease entity with emphasis on possible pathophysiologic mechanisms. World J Gastroenterol2005; 11(27): 4199–4205

[7] Jaeger C, Mayer G, Henrich R, Gossner L, Rabenstein T, May A, Guenter E, Ell C. Secondary sclerosing cholangitis after long-term treatment in an intensive care unit: clinical presentation, endoscopic findings, treatment, and follow-up. Endoscopy2006; 38(7): 730–734

[8] Esposito I, Kubisova A, Stiehl A, Kulaksiz H, Schirmacher P. Secondary sclerosing cholangitis after intensive care unit treatment: clues to the histopathological differential diagnosis. Virchows Arch2008; 453(4): 339–345

[9] Schnitzbauer AA, Tsui TY, Kirchner G, Scherer MN, Bein T, Schlitt HJ, Obed A. Liver transplantation for sclerosing cholangitis in a polytraumatized patient. Nat Clin Pract Gastroenterol Hepatol2009; 6(2): 121–126

[10] Al-Benna S, Willert J, Steinau HU, Steinstraesser L. Secondary sclerosing cholangitis, following major burn injury. Burns2010; 36(6): e106–e110

[11] Tian M, Lv Y, Li J. Hepatobiliary and pancreatic: sclerosing cholangitis associated with critical illness. J Gastroenterol Hepatol2010; 25(4): 842

[12] Kwon ON, Cho SH, Park CK, Mun SH. Biliary cast formation with sclerosing cholangitis in critically ill patient: case report and literature review. Korean J Radiol2012; 13(3): 358–362

[13] Abdalian R, Heathcote EJ. Sclerosing cholangitis: a focus on secondary causes. Hepatology2006; 44(5): 1063–1074

[14] Ruemmele P, Hofstaedter F, Gelbmann CM. Secondary sclerosing cholangitis. Nat Rev Gastroenterol Hepatol2009; 6(5): 287–295

[15] Deltenre P, Valla DC. Ischemic cholangiopathy. J Hepatol2006; 44(4): 806–817

[16] Paramythiotis D, Kazamias1 P, Grosomanidis1 V, Kotzampassi K. Splanchnic ischemia during mechanical ventilation. Annals Gastroenterol2008; 21(1): 45–52

[17] Putensen C, Wrigge H, Hering R. The effects of mechanical ventilation on the gut and abdomen. Curr Opin Crit Care2006; 12(2): 160–165

[18] Träger K, Radermacher P, Georgieff M. PEEP and hepatic metabolic performance in septic shock. Intensive Care Med1996; 22(11): 1274–1275

[19] Krejci V, Hiltebrand LB, Sigurdsson GH. Effects of epinephrine, norepinephrine, and phenylephrine on microcirculatory blood flow in the gastrointestinal tract in sepsis. Crit Care Med2006; 34(5): 1456–1463

[20] Lee JG, Schutz SM, England RE, Leung JW, Cotton PB. Endoscopic therapy of sclerosing cholangitis. Hepatology1995; 21 (3): 661–667

[21] Narumi S, Roberts JP, Emond JC, Lake J, Ascher NL. Liver transplantation for sclerosing cholangitis. Hepatology1995; 22(2): 451–457

[22] Harnois DM, Angulo P, Jorgensen RA, Larusso NF, Lindor KD. High-dose ursodeoxycholic acid as a therapy for patients with primary sclerosing cholangitis. Am J Gastroenterol2001; 96(5): 1558–1562

[23] Triantos CK, Koukias NM, Nikolopoulou VN, Burroughs AK. Meta-analysis: ursodeoxycholic acid for primary sclerosing cholangitis. Aliment Pharmacol Ther2011; 34(8): 901–910

[24] Lindor KD, Kowdley KV, Luketic VA, Harrison ME, McCashland T, Befeler AS, Harnois D, Jorgensen R, Petz J, Keach J, Mooney J, Sargeant C, Braaten J, Bernard T, King D, Miceli E, Schmoll J, Hoskin T, Thapa P, Enders F. High-dose ursodeoxycholic acid for the treatment of primary sclerosing cholangitis. Hepatology2009; 50(3): 808–814

[25] Eaton JE, Silveira MG, Pardi DS, Sinakos E, Kowdley KV, Luketic VA, Harrison ME, McCashland T, Befeler AS, Harnois D, Jorgensen R, Petz J, Lindor KD. High-dose ursodeoxycholic acid is associated with the development of colorectal neoplasia in patients with ulcerative colitis and primary sclerosing cholangitis. Am J Gastroenterol2011; 106(9): 1638–1645

[26] Katsinelos P, Kountouras J, Chatzimavroudis G, Zavos C, Pilpilidis I, Paroutoglou G. Combined endoscopic and ursodeoxycholic acid treatment of biliary cast syndrome in a non-transplant patient. World J Gastroenterol2008; 14(33): 5223–5225

[27] Bansal V, Schuchert VD. Jaundice in the intensive care unit. Surg Clin North Am2006; 86(6): 1495–1502

[28] Brienza N, Dalfino L, Cinnella G, Diele C, Bruno F, Fiore T. Jaundice in critical illness: promoting factors of a concealed reality. Intensive Care Med2006; 32(2): 267–274

[29] Spirt MJ, Stanley S. Update on stress ulcer prophylaxis in critically ill patients. Crit Care Nurse2006; 26(1): 18–20, 22–28, quiz 29

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg