Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unknown cause (1,2), leading to inflammation and fibrosis of the biliary tree. More than 60% of patients with PSC have concomitant inflammatory bowel disease (IBD) (3–5), and 5% of patients with ulcerative colitis (UC) develop cholestatic liver disease with 40% of these developing PSC (6). Liver transplantation (LT) remains the only effective therapeutic option for patients with end-stage liver disease (2,7–11), with 5-year survival rates post-LT exceeding 70% (2,11). Some centers have reported an increase in the incidence of colorectal cancer (CRC) and dysplasia post-LT in patients with PSC and IBD compared with patients undergoing transplantation for other liver diseases or with isolated PSC (3,4,12,13). The course of IBD post-LT is variable, and in cases in which corticosteroids are withdrawn early, inadequate control of IBD has been observed in 35% to 50% (14,15). Also, there is no uniform colonic surveillance protocol for patients with IBD post-LT, ranging from regular colonoscopy to prophylactic colectomy at the time of the transplant (4,5). Colectomy for IBD in the setting of liver disease is a high-risk surgical procedure with mortality rates of 38%, even in those with Child-Pugh Class A or B cirrhosis (16).
True evidence of an increased risk of developing CRC post-LT in patients with PSC and UC is scarce, as are data of its impact on survival (5). This study reviews the experience of patients with PSC undergoing LT to study the incidence and risk factors related to the occurrence of CRC, its impact on survival, and the impact of colectomy on the transplant procedure.
PATIENTS AND METHODS
A total of 152 consecutive patients (108 men, 71%) with PSC between 1986 and May 2000 who underwent 173 transplants (including 19 second LTs and one fourth LT) were studied. The diagnosis of PSC was made on the basis of accepted radiologic and histologic criteria and confirmed by histologic examination of the explanted liver. Patients with an overlap diagnosis such as primary biliary cirrhosis or autoimmune hepatitis were excluded. IBD was diagnosed clinically and confirmed on colonic biopsy in 100 patients (65%): 92 with UC, and eight with Crohn’s disease. Colonoscopy results pre-LT were recorded in 57 of 90 patients (63%) with IBD and in 47 of 100 patients (47%) post-LT. The number of colonoscopic examinations pre-LT was 0.37 per patient per year, compared with 0.61 per patient per year in the post-LT group. LT was performed at a mean of 16.8 years (0.2–43.3 years) after the onset of IBD symptoms, and 10 patients developed IBD post-LT.
LT was performed in a standard fashion, and biliary reconstruction consisted of a Roux en Y choledochojejunostomy in all but four cases. Cyclosporine-based immunosuppression was used in 86.7% of patients and tacrolimus in 13.3%, along with low-dose steroid therapy (20 mg prednisolone) tapered with an aim to stop prednisolone 3 months post-LT. Most patients received azathioprine (90%). Rejection episodes were treated with bolus high-dose steroids (3×200 mg oral prednisolone).
The incidence of CRC in LT recipients with and without PSC was noted. To assess the risk of developing CRC in PSC, three groups were analyzed:
- PSC without IBD (n=52).
- PSC with colectomy (pre-LT and at LT) (n=17, 13 colectomies before assessment for LT and four simultaneous colectomies at LT).
- PSC with IBD and an intact colon (n=83). The following factors were studied: age, gender, liver and renal biochemistry, international normalized ratio, Child-Pugh stage, operative time, blood use, hospital stay, immunosuppression, risk of CRC, retransplantation rate, and mortality.
Survival and cumulative risk of developing CRC were estimated using the Kaplan-Meier method and compared using the log-rank test. Variables were tested for association with outcome using univariate analysis. The Wilcoxon rank-sum test was used to compare median values. A multivariate analysis (Cox proportional hazards regression model) was used to determine risk factors for CRC. A P value of less than 0.05 was considered significant. The cumulative risk of neoplasia was measured from the transplant date, and the patients were considered at risk until their death, last follow-up, or colectomy.
Of the patients with PSC in our study, 107 of 152 are alive to date, with an actuarial patient survival of 85%, 70%, and 60% at 1, 5, and 10 years, respectively. Death was caused by sepsis and multiple organ failure (n=18), hemorrhage (n=6), cancer (n=6), chronic rejection (n=5), recurrent disease (n=2), primary nonfunction (n=2), and others (n=6). The overall graft survival was 75%, 61%, and 45% after 1, 5, and 10 years, respectively. The following risk factors were similar in all three groups: age, gender, liver biochemistry, international normalized ratio, serum creatinine, operative time, Child-Pugh stage, intensive therapy unit (ITU), and inpatient stay (Table 1).
Impact of Colectomy on Liver Transplantation
Of the 100 patients with IBD in our study, 27 underwent colectomy (13 pre-, 4 during, and 10 post-LT) for the following indications: refractory colitis (n=16, 59%), cancer or dysplasia (n=10, 37%), and colonic ischemia (n=1, 4%) (Fig. 1). Ten of these patients underwent an ileal pouch reconstruction. Colectomy pre- or during LT did not have a major impact on early (<30 days) post-LT outcome. There were no significant differences between the colectomy and non-colectomy groups for operation time, blood loss, ITU, and inpatient stay. Graft loss in the pre-LT colectomy group (3/13) was caused by hepatic artery thrombosis (n=2) and chronic rejection (n=1). Of the patients who underwent colectomy at the time of the LT, one required retransplantation for hepatic artery thrombosis, and one died from hemorrhage. The deaths in the post-LT colectomy group were cancer related in 5 of 10 cases, leading to a decreased survival compared with the remainder of patients with PSC and UC who underwent transplantations with an intact colon (25% vs. 55% at 10 years, not significant). Conversely, patients with the colon removed pre- or during LT showed a better survival than those with an intact colon at the time of the transplant (10-year survival, 87% vs. 55%, not significant), reflecting the fact that the CRC- and colitis-related morbidity had been avoided (Fig. 2).
The incidence of CRC in patients without PSC was 7 of 1,184 (0.6%) (P <0.001). CRC occurred in 8 of 152 patients (5.3%) who underwent transplantation for PSC and 8 of 83 patients (9.6%) with UC and an intact colon at the time of transplant (Fig. 3). Of these eight patients, six were diagnosed by colonoscopy and one by barium enema at 5 months, and another was diagnosed with metastatic disease at 23 months. The location and differentiation of the cancers are shown in Table 2. The mean interval between LT and cancer presentation was 46 months (range 21–68 months) in the PSC group and 52.4 months (range 6–108 months) in the non-PSC group. The cumulative risk of developing CRC in 83 patients with an intact colon and IBD was 14% and 17% after 5 and 10 years, respectively; in contrast, the PSC non-IBD group demonstrated a 0% risk after 10 years (P <0.05) (Fig. 4). All the CRCs presenting post-LT occurred in patients older than 45 years (P <0.02), with a 22% risk of developing malignancy after 10 years (P <0.002). The mean interval between the diagnosis of UC and LT was longer in patients who subsequently developed CRC (22.5 years vs. 14.8 years). Patients with a history of UC more than 10 years pre-LT demonstrated a higher risk of developing CRC 6 years post-LT (30%) (P <0.03).
Patients undergoing more than three colonoscopic examinations had an earlier diagnosis of their cancers (mean 58 months vs. 84 months) and less advanced disease at diagnosis (50% with stage I vs. 100% with stage III) compared with patients with less than two colonoscopies. The presence of polyps on colonoscopy (n=11) was strongly related to the development of CRC with a risk of 55% after 5 years versus 5% in patients with no polyps (P <0.001). However, more than half of these polyps on microscopy were simply regenerative (n=7), one was adenomatous, and three were dysplastic. Cancer developed in three of these patients, one in each group (adenomatous, regenerative, and dysplastic). Dysplasia seen in colonic biopsies post-LT led to an increased risk of 11% of CRC compared with 1% in the non-dysplasia group (P <0.001). The extent of colitis post-LT was also associated with the development of CRC. Five years post-LT, 35% of patients with pancolitis developed CRC compared with 7% in patients with less than total involvement (P <0.03).
The multivariate analysis identified three variables contributing significantly to the risk of developing CRC: dysplasia post-LT (P <0.0003), duration of colitis more than 10 years (P <0.002), and pancolitis (P <0.004). When dysplasia was not taken into account, the significant variables were the following: time between LT and UC (<10 or >10 years) (P =0.001), age (<45 or >45 years) (P =0.004), and the presence of polyps or no polyps at colonoscopy (P =0.03). Patient gender, immunosuppressive treatment, extent of IBD, and pre- and posttransplant colitis activity were not associated with an increased risk of CRC. Patients with two of the previously mentioned significant variables (time between LT and UC, age, and polyps) have a risk of 0%, 20%, and 38% of developing CRC 1, 5, and 10 years post-LT, respectively, whereas patients with all three variables will develop CRC 5 years post-LT (risk: 0%, 100%, 100% 1, 5, and 10 years post-LT, respectively).
The cause of death in patients with CRC was cancer related in three of four of the cases and chronic rejection after a reduction in immunosuppression because of the cancer in one case. The survival rate in this group was 53% after 5 years compared with 73% in the non-CRC group (P <0.7) (Fig. 5).
In this study of 152 patients undergoing LT for PSC over a 14-year period, the overall outcome was not as good as for other chronic hepatic diseases, in which the 1- and 5-year survival rates were 87% and 80%, respectively. Several factors may be responsible for this: (1) the presence of PSC in conjunction with IBD (two diseases) in up to 66% of cases, (2) the higher risk of developing cancer (colorectal or biliary tree) compared with other chronic liver diseases, (3) the increased technical difficulty of performing LT in patients with previous biliary or colorectal intervention, and (4) the recurrence of PSC post-LT. In addition, it seems that PSC is different when associated with IBD. The combination of PSC and IBD is more common in men than PSC alone and can be more severe, leading to a reduced survival. (17). In this study, the male to female ratio varied from 1.6 to 1 in PSC to 3.2 to 1 in combined PSC and IBD.
The management and follow-up of patients with PSC post-LT requires attention to the control of IBD (medical and surgical), the surveillance for colonic cancer, and the possibility of further colonic surgery. However, there is no current consensus regarding the postoperative course for these patients. Some studies have shown little or no change in colitis activity post-LT (prednisolone and cyclosporine based) (18), others have found improvement in most patients (19), and some have demonstrated an increased disease activity post-LT (14,15). The reports of increased IBD activity may be related to the improvement in T-cell function after grafting (15). In our series, colectomy for symptom control was necessary in 14.4% of patients pre-LT and 6% post-LT. The impact of colectomy pre-, during, and post-LT has not been previously described. Some reports show that colectomy in patients with PSC has failed to show any advantage in graft survival (20), and complications such as stomal variceal bleeding and ileal pouch-related problems have been described (21,22).
The perioperative technical factors at transplantation (operation time, blood loss, ITU, and hospital stay) in both the pre-LT colectomy and simultaneous LT and colectomy groups were no different compared with the non-colectomy group, indicating that the colectomy procedure in these selected patients did not represent a significant additional mortality risk. One fourth of the grafts were lost when the colectomy was performed at the time of the transplant, similar to 3 of 13 when it was performed pre-LT.
In this series, most colectomies pre-LT were performed for poor symptomatic control of IBD (10/13, with three for colonic dysplasia), compared with post-LT (5/10 for cancer and 5/10 for progressive IBD). In view of the high percentage of patients with cancer in the latter group, overall survival was better in patients in whom the colon was removed pre- and at LT. Thus, colectomy pre- or during LT is a safe procedure in selected patients with PSC and progressive severe IBD and is associated with an improved survival compared with patients requiring a colectomy post-LT, indicating a role for colectomy before the development of CRC in longstanding severe colitis.
The risk for cancer associated with chronic IBD, particularly UC, is up to fourfold higher than the standard population risk for CRC. Factors that influence the incidence of cancer include the duration and extent of colonic involvement and severity of inflammation (23). Most series show a cumulative risk of between 9% and 14% at 25 years from the onset of symptoms. An association between PSC and the presence of dysplasia or colonic carcinoma with UC has been described (13). Brentnall et al. showed a significantly higher incidence of dysplasia (45%) in patients with PSC and UC compared with control UC groups (16%) (12).
Liver transplantation has been shown to be an additional risk factor in the development of CRC in patients with PSC and IBD (5). In this series, the univariate analysis identified an increased risk of CRC in patients with the following: age more than 45 years, long-term IBD (>10 years pre-LT), and more severe colitis (dysplasia, pancolitis). Patients undergoing regular colonoscopies (more than three post-LT) were more likely to be diagnosed in the early stage of cancer. Colonoscopic surveillance should be performed at least on a yearly basis pre- and post-LT and continue long term if the diagnosis of cancer was made at a median of approximately 4 years post-LT. Early colonic resection in patients with PSC and UC post-LT, similar to colonic resection in high-risk patients without PSC and UC with acute, chronic, or recurrent inflammation not well controlled with medical therapy, will help to lower cancer risk in the long term (24).
CRC results in significant post-LT mortality in patients with PSC and UC. Colectomy is a relatively safe procedure when associated with LT for PSC, and this strengthens the argument for considering colectomy pre-, during, or shortly post-LT in selected patients with two or more of the risk factors for CRC. Frequent colonoscopic surveillance and early colectomy in selected patients with longstanding severe bowel disease will reduce the risks of cancer.
1. Lee YM, Kaplan MM. Primary sclerosing cholangitis. N Engl J Med 1995; 332: 924–933.
2. Graziadei IW, Wiesner RH, Marotta PJ, et al. Long-term results of patients undergoing liver transplantation for primary sclerosing cholangitis. Hepatology 1999; 30: 1121–1127.
3. Knechtle SJ, D’Alessandro AM, Harms BA, et al. Relationships between sclerosing cholangitis, inflammatory bowel disease, and cancer in patients undergoing liver transplantation. Surgery 1995; 118: 615–619; discussion 619–620.
4. Higashi H, Yanaga K, Marsh JW, et al. Development of colon cancer after liver transplantation for primary sclerosing cholangitis associated with ulcerative colitis. Hepatology 1990; 11: 477–480.
5. Loftus EV, Jr., Aguilar HI, Sandborn WJ, et al. Risk of colorectal neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis following orthotopic liver transplantation. Hepatology 1998; 27: 685–690.
6. Olsson R, Danielsson A, Jarnerot G, et al. Prevalence of primary sclerosing cholangitis in patients with ulcerative colitis. Gastroenterology 1991; 100: 1319–1323.
7. Hay JE. Liver transplantation for primary biliary cirrhosis and primary sclerosing cholangitis: does medical treatment alter timing and selection? Liver Transpl Surg 1998; 4: S9–17.
8. Abu-Elmagd KM, Malinchoc M, Dickson ER, et al. Efficacy of hepatic transplantation in patients with primary sclerosing cholangitis. Surg Gynecol Obstet 1993; 177: 335–344.
9. Marsh JW Jr, Iwatsuki S, Makowka L, et al. Orthotopic liver transplantation for primary sclerosing cholangitis. Ann Surg 1988; 207: 21–25.
10. Farges O, Malassagne B, Sebagh M, et al. Primary sclerosing cholangitis: liver transplantation or biliary surgery. Surgery 1995; 117: 146–155.
11. Neuberger J, Gunson B, Komolmit P, et al. Pretransplant prediction of prognosis after liver transplantation in primary sclerosing cholangitis using a Cox regression model. Hepatology 1999; 29: 1375–1379.
12. Brentnall TA, Haggitt RC, Rabinovitch PS, et al. Risk and natural history of colonic neoplasia in patients with primary sclerosing cholangitis and ulcerative colitis [see comments]. Gastroenterology 1996; 110: 331–338.
13. Broome U, Lindberg G, Lofberg R. Primary sclerosing cholangitis in ulcerative colitis—a risk factor for the development of dysplasia and DNA aneuploidy? [see comments]. Gastroenterology 1992; 102: 1877–1880.
14. Miki C, Harrison JD, Gunson BK, et al. Inflammatory bowel disease in primary sclerosing cholangitis: an analysis of patients undergoing liver transplantation. Br J Surg 1995; 82: 1114–1117.
15. Papatheodoridis GV, Hamilton M, Rolles K, et al. Liver transplantation and inflammatory bowel disease. J Hepatol 1998; 28: 1070–1076.
16. Post AB, Bozdech JM, Lavery I, et al. Colectomy in patients with inflammatory bowel disease and primary sclerosing cholangitis. Dis Colon Rectum 1994; 37: 175–178.
17. Rabinovitz M, Gavaler JS, Schade RR, et al. Does primary sclerosing cholangitis occurring in association with inflammatory bowel disease differ from that occurring in the absence of inflammatory bowel disease? A study of sixty-six subjects. Hepatology 1990; 11: 7–11.
18. Gavaler JS, Delemos B, Belle SH, et al. Ulcerative colitis disease activity as subjectively assessed by patient- completed questionnaires following orthotopic liver transplantation for sclerosing cholangitis. Dig Dis Sci 1991; 36: 321–328.
19. Stephens J, Goldstein R, Crippin J, et al. Effects of orthotopic liver transplantation and immunosuppression on inflammatory bowel disease in primary sclerosing cholangitis patients. Transplant Proc 1993; 25: 1122–1123.
20. Cangemi JR, Wiesner RH, Beaver SJ, et al. Effect of proctocolectomy for chronic ulcerative colitis on the natural history of primary sclerosing cholangitis. Gastroenterology 1989; 96: 790–794.
21. Wiesner RH, LaRusso NF, Dozois RR, et al. Peristomal varices after proctocolectomy in patients with primary sclerosing cholangitis. Gastroenterology 1986; 90: 316–322.
22. Fucini C, Wolff BG, Dozois RR. Bleeding from peristomal varices: perspectives on prevention and treatment. Dis Colon Rectum 1991; 34: 1073–1078.
23. Sugita A, Sachar DB, Bodian C, et al. Colorectal cancer in ulcerative colitis. Influence of anatomical extent and age at onset on colitis-cancer interval. Gut 1991; 32: 167–169.
24. Levin B. Risk of cancer in ulcerative colitis. Gastrointest Endosc 1999; 49: S60–2.