- Email: [email protected]
Ursodeoxycholic acid for symptomatic primary biliary cirrhosis
332
Journal of Hepatology, 1993; 17:332-338 © 1993 ElsevierScientificPublishers Ireland Ltd. All rights reserved. 0168-8278/93/$06.00
HEPAT 01255
Ursodeoxycholic acid for symptomatic primary biliary cirrhosis Preliminary analysis of a double-blind multicenter trial P.M. Battezzati, M. P o d d a , F.B. Bianchi, R. N a c c a r a t o , F. Orlandi, C. Surrenti, L. Pagliaro, F. M a n e n t i a n d the Italian Multicenter G r o u p for the Study of U D C A in PBC* (Received 3 December 1991)
The administration of ursodeoxycholic acid, a hydrophilic bile acid not hepatotoxic to humans, has been suggested for treatment of primary biliary cirrhosis to improve cholestasis and reduce hepatocellular damage. Efficacy of treatment has been studied mainly in patients with asymptomatic or early-stage disease. In January 1988, to establish the efficacy and safety of ursodeoxycholic acid in a population with more severe disease, we started a multicenter, double-blind, placebo-controlled trial in patients with symptomatic disease, that is, with pruritus or serum bilirubin exceeding 2 mg/dl. Forty-four patients were assigned to ursodeoxycholic acid, 500 mg daily (corresponding to about 8.7 mg/kg body weight in these patients), and 44 to a placebo. As planned at the beginning of the study, a preliminary analysis was performed when all patients had been followed for at least 6 months (33 patients up to 12 months). Pruritus, self-evaluated by the patients, and cholestyramine consumption, as recorded in a diary, decreased significantly (p < 0.01) in both groups. In patients who initially had abnormal levels, serum bilirubin decreased significantly (p < 0.05) in the ursodeoxycholic acid group compared to placebo. After 6 months the following were also significantly better in the ursodeoxycholic acid than in the placebo group: a composite weighted biochemical index taking into account the changes in serum bilirubin, alkaline phosphatase, ),-GT and AST (p<0.001); serum prealbumin (p<0.05); IgG (p < 0.01) and IgM (p < 0.01) levels. The positive effects of ursodeoxycholic acid administration on serum bilirubin, the most important prognostic factor in primary biliary cirrhosis, and on liver protein synthesis suggest that ursodeoxycholic acid may be useful for patients with more advanced disease than those so far included in therapeutic trials. K e y words: Ursodeoxycholic acid; Bile acids; Chronic cholestasis; Primary biliary cirrhosis; Pruritus; Icteric
A beneficial effect of ursodeoxycholic acid (UDCA) in patients with primary biliary cirrhosis (PBC) was first reported in uncontrolled pilot studies (1) and was confirmed more recently in placebo-controlled, double-blind studies (2,3). In these studies, mainly subjects with mild symptoms and early stages of the disease were included, and the effects of treatment could not be analyzed separately for the subset of patients with advanced PBC. This is important, since progression of the disease in patients
with asymptomatic PBC is slow (4,5) and there are other preliminary reports indicating that UDCA may fail in advanced stages of PBC (6-8). With this in mind, we performed a randomized controlled trial including only patients with symptomatic disease, that is, with pruritus or serum bilirubin exceeding 2mg/dl, to establish whether or not UDCA is effective against the symptoms and indices of the severity of PBC.
Correspondence to: Professor Mauro Podda, Cattedra di Medicina Interna, Istituto di Scienze BiomedicheS. Paolo, UniversitY.di Milano, via di
Rudini, 8, 20142 Milano, Italy. *The Italian MulticenterGroup for the Study of UDCA in PBC consisted of the following:Investigators and Clinical Centers: M. Podda, P.M. Battezzati,E. Bertolini,M. Camisasca,A. Crosignani,M. Zuin (Universityof Milan);F.B. Bianchi,M. Fusconi,G. Ballardini,F. Cassani(University of Bologna);F. Orlandi, G.P. Macarri, A. Benedetti(Universityof Ancona); R. Naccarato, M. Chiaramonte,A.R. FIoreani (Universityof Padova); C. Surrenti, M. Pozzi, M.R. Biagini(Universityof Firenze); L. Pagliaro, P. Almasio,G. Provenzano,I. Scalisi(Universityof Palermo); F. Manenti, S. Bellentani, G. Tabarroni (University of Modena); Bile acid analysis: A. Roda (University of Bologna); Determination of AMA: F.B. Bianchi (University of Bologna);Measurement ofPIllP: G. Annoni (Universityof Milan); Measurement ofprealbumin: G. Bisiani(Universityof Milan).
UDCA FOR SYMPTOMATICPBC Patients and Methods
Patient selection Patients with clinical histories and biochemical profiles consistent with PBC, who presented with pruritus or serum bilirubin exceeding 2 mg/dl, were considered for admission into the study. Diagnosis of the disease was re-evaluated according to the following criteria, which were modified from Taal et al. (9): a positive test (titer ~>1:40) for antimitochondrial antibodies (AMA), a pre-entry liver specimen considered to be consistent with or diagnosis of PBC (10). If one of these major criteria was not met, three of the following were required for the diagnostic of PBC to be confirmed: serum levels of alkaline phosphatase higher than 2-times the upper reference value, IgM levels >t280 mg/dl, pruritus, serum bilirubin >2 mg/dl, a positive result in Schirmer's test. Patients also had to have radiological or ultrasonographic evidence of the absence of extrahepatic biliary obstruction. Inclusion criterion was presence of symptomatic disease, as defined by presence of pruritus and/or serum bilirubin >2 mg/dl. Pruritus had to have been present in the 6 months prior to evaluation, serum levels of bilirubin had to have been >2 mg/dl at two separate check-ups in the previous 6 months. Percutaneous liver biopsy had to be repeated if it had been performed more than 12 months prior to admission for patients who had been classified as stages I-III. Serum bilirubin levels higher than 10 mg/dl, ascites, previous episodes of variceal bleeding or encephalopathy, evidence of malignant conditions, and alcohol abuse were criteria for exclusion. No patient was taking any medication known to be hepatotoxic nor had been treated with corticosteroids, immunosuppressant agents, coichicine, penicillamine or UDCA in the previous 6 months.
Study design Patients were informed of the nature, purpose and requirements of the study before giving oral consent. The study was carried out according to the principles of the Declaration of Helsinki for research in human subjects. At the time of enrollment patients underwent clinical and laboratory evaluations. An ultrasound examination of the upper abdomen was performed to confirm the absence of extrahepatic biliary obstruction or of liver tumors. UDCA, 500 mg daily, or placebo were administered double-blind in two divided doses at mealtime. Randomization of treatment assignments was performed separately for each center: patients were consecutively given indistinguishable medications, which had been assigned
333 by the central pharmacy according to a computergenerated list. UDCA and an identical-appearing placebo were obtained through the courtesy of ABC Farmaceutici, Torino, Italy. Patients were allowed to continue taking supportive medication, such as calcium and vitamins. When required for control of pruritus, cholestyramine could be taken by the patients, who were instructed to take it at least 4 h before or after the study drug. Clinical examination, routine hematological tests and blood chemistries were performed every 3 months. Ultrasonography was repeated every 6 months. Assessment of intensity of pruritus at each visit was expressed by a numerical score, taking into account the symptom's intensity, cholestyramine consumption, and presence of skin marks. During the week preceding each visit, patients were asked to report intensity of pruritus and cholestyramine consumption in a diary. Intensity of pruritus was rated on a scale from 0 (absent) to 2 (severe); the need for cholestyramine for control of pruritus was taken into account as follows: 1 point was added to the intensity score of patients needing no more than 8 g daily and without skin marks, 2 points were added otherwise. The intensity score used for statistical purposes was the mean of the assessment at the visit and the daily reports during the week preceding the visit. Clinical observations and blood test results at each visit were recorded on standard data-collection forms and forwarded to the study monitor with the patient's diary. Compliance with the treatment was assessed by counting the capsules returned to the clinic by the patient and by the determination of serum UDCA levels. The study started in January 1988 and patient enrollment was terminated in June 1988. According to the study protocol, a preliminary analysis was conducted on data available on January 1989, when all patients had been followed for at least 6 months. Patients agreed that if this analysis showed sufficient evidence of a beneficial effect of UDCA, they could be placed on this treatment and continue the study according to a different protocol. The pre-defined major parameters for evaluation of treatment efficacy were decrease in serum bilirubin, improvement of pruritus and improvement of serum liver enzymes, which was assessed by a composite biochemical index to avoid the need for repetitive significance testing. The index was calculated for each time point as the weighted mean of changes (expressed as ratio to baseline values) of serum bilirubin, aspartate aminotransferase (AST), alkaline phosphatase (ALP),
334 and ~,-glutamyl transpeptidase (GGT), according to the following formula: [2 x (bilirubin ratio) + (AST ratio) + (ALP ratio) + ( G G T ratio)I/5 Values higher than ! indicate an overall worsening of serum liver enzymes, lower than 1 indicate improvement. The sample size determination was based on the results of a pilot study of the effects of U D C A administration to symptomatic patients with PBC (11). The composite biochemical index for that group of patients indicated that a reduction of about 0.60 standard deviations from baseline values had been achieved during administration of 500 mg/day UDCA. We calculated that a sample size of 45 patients for each treatment group would yield a probability of a type II error of 0.80, at the 5% level of significance, in detecting such a difference between U D C A and placebo-treated patients, assuming that no improvement in the composite biochemical index occurred in the latter group.
P.M, BATTEZZATIet al. to the reference values for each laboratory and are expressed as multiples of the upper limit of reference values. To compare changes over time between the two treatment groups in the continuous variables, the ratio of values at a particular time to baseline values was computed for each patient. Changes in the pruritus score were expressed as the difference between values at a particular time and baseline values. Quantitative data were analyzed by the Wilcoxon rank signs test and the Mann-Whitney test. Relative frequencies were compared by the Chi-square test and Fisher's exact test. Correlations were calculated by linear regression analysis; data were transformed logarithmically when departures from normality had been detected. Two-tailed p values were used for statistical comparisons. The total period of observation was used in the analysis of data for each patient, according to the 'intention to treat' principle.
Results
Clinical and laboratory investigations Fasting blood samples were collected from patients at each control visit. Laboratory tests included full blood counts, standard liver function tests, AMA determination, serum lipids, albumin, globulins, quantitative immunoglobulins, prothrombin time, serum creatinine, urea, serum prealbumin, serum bile acids and serum procollagen III peptide (PIIIP). Serum bile acids and serum P I I I P were determined on patients enrolled at four of the participating centers (53 patients from the centers of Milan, Bologna, Palermo and Modena). Hematological tests and blood chemistries were done in the clinical laboratory of the participating hospitals by routine automated techniques. Some of the tests were performed centrally, namely serum AMA, serum prealbumin (radial immunodiffusion commercial kit, Behringwerke AG, Marburg, Germany) (12), P I I I P (commercial radioimmunoassay, Boehringer RIA), individual bile acids (solid phase immunoenzymatic assays) (13). For this purpose, aliquots of serum were collected from each patient at 6-month intervals and kept frozen at - 20 °C until the time of testing.
Analysis of data As stated in the protocol, the interest of this analysis was focused on changes in pre-defined parameters after at least 6 months of treatment. To allow a more detailed analysis over the entire study period, results were also reported for data obtained at different times, up to 1 year. Liver enzyme values were standardized according
The study began on January 1, 1988 and patient enrollment was terminated in June 1988. During that period, 88 patients with PBC were admitted to the trial in the seven centers. Forty-four patients were randomly assigned to U D C A treatment, and the same number to placebo. The two treatment groups were similar at entry into the study for the most relevant clinical and laboratory characteristics (Tables !-3). Mean weight was 54 kg for both groups. In the U D C A group, the treatment dose of 500 mg per day corresponded to an average daily dose of 8.7 mg/kg body weight (range 5.4-1 i.6). During the study period, six of the patients (five in TABLE 1 Clintcal and laboratory characteristics of patients with primary bdiary cirrhosis, at entry into the study Characterisuc UDCA Placebo Subjects 44 44 Mean age (years) 54+2 55+2 Women ('~-) 84 93 Mean duration of disease (years) 7+ I 7+ I AMA positive (%) 84 91 Histologic stage IV (%) 48 50 Varices (%)a 26 45 Pruritus present (%/ 91 89 Serum bilirubin > 2 mg/dl (%) 34 32 Mean ALT l× N)u 2.76+0.25 2.38+0.15 Mean AST Ix N) 2.51 +0.17 2.49+0.17 Mean alkaline phosphatase {× N) 4.99-+0.43 4.55+2.44 Mean -;-glutamyltransferase(× N) 14.90_+2.09 12 87 + 1.44 Plus/minus values are means+ SE. aData for 35 subjects in the UDCA and for 30 patients in the placebo group. bData for serum liver enzymes are reported as multiples of the upper limit of normal reference values (x N).
UDCA FOR SYMPTOMATIC PBC
335
the U D C A group) were classified as d r o p - o u t s . T w o of
TABLE 3
them, both in the U D C A group, were lost to follow-up
Serum bile acids at entry and after 6 months of treatment
before the first m o n t h of t r e a t m e n t was completed: o n e
At entry
left the study due to personal choice, and the o t h e r due
After 6 months
Serum UDCA (.umol/I)' UDCA <5 23.5 + 3.9 Placebo <5 <5 Serum cholic acid (,umol/l}" UDCA 12.7±2.5 7.44-1.6 Placebo 9.6 ± 2. I 9.2 ± 1.5 Serum chenodeoxycholic acid (,umol/l)° UDCA 15.6+4.2 10.3±2.7 Placebo 8.0 4-1.4 10.4 ± 1.9
to the d e v e l o p m e n t of a skin rash. T w o patients on
UDCA an d one on p la c e b o were unable to tolerate the medication because of w o r s e n i n g of pruritus. O n e patient was w i t h d r a w n from t r e a t m e n t with U D C A as a result of a recurrent episode of h e m o l y t i c anemia. She was taking U D C A and was s u b s e q u e n t l y treated with ste-
p<0.01 vs. placebo p<0.01 vs. placebo
Plus/minus values are means ± SE.
roids. All the r e m a i n i n g subjects took m o r e than 80%
aData for 26 subjects in the UDCA group and for 27 subjects in the
of the prescribed dose, as assessed from the counts of
placebo group.
unused capsules at each visit. In five patients serum U D C A levels were lower than 5/amol/l, and they were considered to be n o n - c o m p l i a n t . N o n e of the patients
following analyses and, with the ex cep t i o n of the patient
in the p l aceb o g r o u p had serum levels of U D C A higher
w h o was given steroid t r e a t m e n t because of the recur-
than 5 ~umol/i. All the patients were included in the
rence of h e m o l y t i c anemia, data from their total periods
TABLE 2 Climcal and laboratory values at entry and after 6 months of treatment At entry Pruritus score 2.3±0.2 UDCA 1.9±0.2 Placebo Bilirubin, all patients (mg/dl) 1.85 ± 0.24 UDCA Placebo 1.91 ± 0.22 Bdlrubin, patients with abnormal values at entry (mg/dl)~ 2.67 _+0.32 UDCA 2.59 ± 0.30 Placebo Composite biochemical index I UDCA 1 Placebo Albumin (g/dl) 4.07±0.10 UDCA 4.09 + 0.08 Placebo ;,-Globulins (g/dll 1.91+0.11 UDCA 1.87+0.09 Placebo IgG (mg/dl) 2018±132 UDCA I891 ± 110 Placebo IgM (mg/dl} 501 ± 47 UDCA 570 ± 56 Placebo IgA [mg/dl) 333±30 UDCA 329 ± 29 Placebo Fotal cholesterol (mg/dl) 263±12 UDCA 266± 13 Placebo HDL cholesterol (mg/dl) 78±6 UDCA 61±5 Placebo Prealbumin (mg/dl, n.y. > 20) 16.0±1.1 UDCA 16.9±1.1 Placebo PIIIP (ng/ml, n.v. < 14.5)b 19.6±1.4 UDCA 22.9±1.9 Placebo
After 6 months 1.4±0.2 1.3±0.2 1.60 ± 0.20 1.93 ± 0.28 2.24 ± 0.28 2.67 4-0.39
p < 0.05 vs. placebo
0.82 ± 0.04 0.96 __+0.03
p<0.001 vs. placebo
4.19+0.09 4.04 ± 0.08 1.88±0.12 1.98±0.10 2077±141 2038±95
p=0.01 vs. placebo
504 ± 54 607 ± 69
p = 0.05 vs. placebo
351±34 362 ± 31 263+14 261 ±11 72±6 61+4 17.7±1.2 16.4±I.0 18.8±1.4 23.3±1.6
Plus/minus values are means+ SE. 'There were 24 subjects in the UDCA group and 27 subjects in the placebo group. ~'Data for 26 subjects in the UDCA group and for 27 subjects in the placebo group.
p<0.05 vs. placebo at entry p<0.05 vs. placebo
336
P.M. BATTEZZATIet al.
of observation were used. At the time of this analysis the median observation period was 9 months, 86 patients (42 in the UDCA group) had been followed for at least 6 months, and in 33 patients (17 in the UDCA group) the observation period lasted one year. The changes in the pruritus scores, serum bilirubin and the liver function tests of the composite biochemical index, and other laboratory variables of interest after 6 months of treatment are reported in Table 2. Highly significant differences were observed between the UDCA and the placebo groups in the composite biochemical index, and they were even more marked when this index was re-calculated without the contribution of serum bilirubin. Changes with time up to 1 year of treatment for pruritus and serum bilirubin are reported in Fig. 1. The pruritus score was significantly lower than at entry in both groups. Eleven of the 25 patients in the UDCA
Pruritus score 3,
1.5
o
~
~
~
1'2 mos afterentry
Bilirubin (mg/dL), all patients
T .....
I. . . . . . . . . I...
group who were initially taking cholestyramine could discontinue it after 6 months of treatment, compared to 5 of 19 patients in the placebo group. There was a significant decrease in serum bilirubin values in the UDCA group after 6 months of treatment, considering only the 51 patients with abnormal bilirubin values at entry (Fig. 1). Changes in serum UDCA and primary serum bile acids are reported in Table 3. There were no correlations found between the dose of UDCA, as mg/kg body weight, and changes in any of the laboratory values or changes in the pruritus scores in the UDCA group. Worsening of pruritus was the major complaint of five patients in the UDCA group and of four placebo patients. Of these, two patients in the UDCA and one in the placebo group were withdrawn from the study; one other patient, who was taking UDCA, had to temporarily reduce the dosage to 1 capsule daily and tolerated well the reinstitution of the protocol dose. One patient given UDCA was withdrawn from the study because of the development of a skin rash after 2 weeks of treatment. This side effect disappeared after discontinuation of therapy, and the patient reported that a similar episode had occurred after she took the drug 1 year before entering the study. Minor side effects reported were development of diarrhea during the first days of treatment, which occurred in four patients in the UDCA group, and minor and transient abdominal complaints. They required no treatment nor any modifications of the treatment schedule.
I
1.5
Discussion
UDCA Placebo
42 (§) 44
42 44
41 44
26 30
17 16
~. ~, o"o~e
Bilirubin (mg/dL), patients with abnormal values at entry T
3
0
UDCA Placebo
24 (§) 27
24 27
24 27
14 16
9 11
,t & ~- J o ~ ~o
(§) no of patient~ F i g . I . C h a n g e s w i t h t i m e o f the pruritus score a n d o f s e r u m b i l i r u b i n .
Among patients with abnormal values at entry, serum bilirubin was significantly(p < 0.05) improved in the UDCA compared to the placebo group after 6 months.
The aim of this study was to assess the therapeutic value of UDCA administration for patients with symptomatic or advanced-stage PBC, as reflected by pruritus or high serum bilirubin levels. Assessment of treatment efficacy in this group of patients was considered important in view of some preliminary reports indicating that UDCA may fail in advanced stages of PBC (6-8). Compared to other controlled trials (2,3) patients included in our study had longer duration of disease, stage IV was found more frequently at histology, jaundice was more frequently present, thus indicating that as a whole they had more advanced disease. Serum bilirubin has been repeatedly identified as an independent prognostic indicator in PBC (4,5,14,15), and elevated values indicate a poor prognosis and advancing disease (16). Presence of symptoms has also been considered to indicate a poor prognosis in PBC (4). Pruritus is an often distressing symptom, with great individual variability, which can severely compromise quality of
UDCA FOR SYMPTOMATICPBC life (14). Fatigue, another symptom which is considered to be commonly reported by patients with PBC, is even more difficult to assess objectively and may be largely influenced by patient beliefs. For this reason we did not include it in the definition of symptomatic patients for this trial. In planning our study we chose not to include histological assessment in the evaluation of treatment efficacy. The patchy nature of hepatic damage in PBC makes serial biopsies difficult to interpret, and performing control liver biopsies in this group of PBC patients with advanced disease was considered unethical. In a recently published mathematical model, liver biopsy was also not required for predicting survival in patients with PBC (5). For simplicity of execution UDCA was administered at a fixed dose (500 mg daily) to each patient rather than on an individual basis. This dosage was chosen to minimize the risk of untoward effects in our patients, in view of a few case reports of patients with late-stage PBC who deteriorated on UDCA (6,8). The average daily dose was 8.7 mg/kg body weight, which was previously shown to be effective against the biochemical expression of liver damage in icteric patients with PBC (11). UDCA was well tolerated at this dosage, and no major side effects were noted during the study. Pruritus scores, as recorded by the patients in a diary and at each visit, decreased in both groups. Decreased pruritus following the administration of placebo has also been observed by others in a double-blind study (3), thus suggesting that there is an important placebo effect on this symptom. To obtain an objective estimate of pruritus, several devices have been proposed, including sensitive limb movement meters (17), but their accuracy was similar to the patient's subjective assessment. There was a significant improvement in serum bilirubin levels during UDCA administration in those patients with abnormal values at entry. Serum prealbumin significantly improved during UDCA treatment. Prealbumin is a protein synthesized
337 in the liver and is easily and reliably measured in the plasma without requiring the active cooperation of the patient. Its short half-life, 1.9 days, renders serum prealbumin a sensitive indicator of liver function capacity (18,19). Serum PIIIP levels, which are correlated with fibrogenesis in many liver diseases (20), did not change during UDCA administration suggesting that this drug does not influence collagen metabolism, at least over 6 months. Our results c o n f r m the striking improvements of serum liver enzymes following UDCA administration, even in the short term (2,3,11). Serum immunoglobulins, both IgG and IgM, did not deteriorate or even improved, thus confirming data by others on the beneficial effects of UDCA on some immunological features of PBC (2,3,21). There were significant decreases in serum cholic and chenodeoxycholic acids during UDCA administration. This is in agreement with the hypothesis that UDCA inhibits the intestinal reabsorption of endogenous bile acids (22,23). In conclusion, the data from our study confirm that UDCA administration for at least 6 months significantly improves biochemical abnormalities in patients with PBC. Although the design of our study does not allow us to assess the clinical importance of such changes, the positive effects we observed on serum bilirubin levels and on liver protein synthesis suggest th~.t UDCA may also be effective for patients with more advanced disease than those so far included in therapeutic trials. Available data indicate that the same group of patients can be safely given colchicine as an anti-fibrotic agent (24-26). Accordingly, after the completion of this analysis all patients were assigned to the UDCA treatment and colchicine or placebo was added in a double-blind fashion.
Acknowledgements This work was presented, in part, at the 40th Annual Meeting of the American Association for the Study of Liver Diseases in Chicago, IL, on October 28-31, 1989.
References I Poupon R, Chr&ien Y, Poupon RE, Ballet F, Calmus Y, Darnis F. Is ursodeoxycholicacid an effectivetreatment for primary biliary cirrhosis? Lancet 1987; i: 834-6. 2 Leuschner U, Fischer H, Kurz W, et al. Ursodeoxycholicacid in primary biliary cirrhosis: results of a controlled double-blind trial. Gastroenterology 1989;97: 1268-74. 3 Poupon RE, Balkau B, Eschwtge E, et al. A multicenter,controlled trial of ursodiol for the treatment of primary biliary cirrhosis. N Engl J Med 1991; 324:1548-54. 4 Roll J, Boyer JL, Barry D, et al. The prognostic importance of
5 6 7 8
clinical and histologicfeatures in asymptomatic and symptomatic primary biliary cirrhosis. N Engl J Med 1983;308: I-7. Dickson ER, Grambsch PM, Fleming TR, Fisher LD, Langworthy A. Prognosis in primary biliary cirrhosis: Model for decision making. Hepatology 1989; 10: I-7. VogelW, Kathrein H, Judmaier G, et al. Deterioration of primary biliary cirrhosis during treatment with ursodeoxycholicacid. Lancet 1988; i: 1163. Hadziyannis S, Hadziyannis E. A randomised controlled trial of ursodeoxychilic acid (UDCA) in primary biliary cirrhosis (PBC) (Abstr.). Hepatology 1988; 5: 1421. Raedsch R, Stiehl A, Thielmann L, et al. Influence of ursodeox-
338 ycholic acid on primary biliary cirrhosis depending on stage of the disease (Abstr.). Gastroenterology 1989; 96: A647. 9 Taal BG, Schalm SW, ten Kate FWJ, Hermans J, Geertzen RGM, Feltkamp BEW. Clinical diagnosis of primary biliary cirrhosis: A classification based on major and minor criteria. Hepatogastroenterology 1983; 30: 178-82. 10 Scheuer PM. Liver Biopsy Interpretation. 2nd Edn. Baltimore: Williams & Wilkins, 1973: 33-8. 11 Podda M, Ghezzi C, Battezzati PM, et al. Ursodeoxycholic acid for chronic liver diseases. J Clin Gastroenterol 1988; 10: $25-$32. 12 Mancini G, Carbonara A, Heremans JF. Immunochemical quantitation of antigens by immunodiffusion, lmmunochemistry 1965; 2: 235--40. 13 Podda M, Ghezzi C, Battezzati PM, Crosignani A, Zuin M, Roda A. Effects of ursodeoxycholic acid and taurine on serum liver enzymes and bile acids in chronic hepatitis. Gastroenterology 1990; 98:1044-50. 14 Kaplan MM. Primary biliary cirrhosis. N Engl J Med 1987; 316: 521-8. 15 Christensen E, Neuberger J, Crowe J, et al. Beneficial effects of azathioprine and prediction of prognosis in primary biliary cirrhosis. Gastroenterology 1985; 89: 1084-91. 16 Shapiro JM, Smith H, Schaffner F. Serum bilirubin: A prognostic factor in primary biliary cirrhosis. Gut 1979; 20: 137-40. 17 Summerfield JA, Welch ME. The measurement of itch with sensitive limb movement meters. Br J Dermatol 1980; 102: 275-81. 18 Hutchinson DR, Smith MG, Parke DW. Prealbumin as an index
P.M. BATTEZZAT1 et al.
19
20
21
22
23
24
25
26
of liver function after acute paracetamol poisoning. Lancet 1980; ii: 121-3. Teppo A-M, Maury CPJ. Serum prealbumin, transferrin and immunoglobulins in fatty liver, alcoholic cirrhosis and primary biliary cirrhosis. Clin Chim Acta 1983; 129: 279-86. Frei A, Zimmermann A, Wiegand K. The n-terminal peptide of collagen type III in serum reflects activity and degree of fibrosis in patients with liver disease. Hepatology 1984; 4: 830-4. Calmus Y, Gane P, Rouger P, Poupon R. Hepatic expression of class I and class II major histocompatibility complex molecules in primary biliary cirrhosis: effect of ursodeoxycholic acid. Hepatology 1990; I1: 12-5. Stiehl A, Raedsch R, Rudolph G. Acute effects of ursodeoxycholic and chenodeoxycholic acid on the small intestinal absorption of bile acids. Gastroenterology 1990; 98: 424-8. Marteau P, Chazouillrres O, Myara A, Jian R, Rambaud J-C, Poupon R. Effect of chronic administration of ursodeoxycholic acid on the ileal absorption of endogenous bile acids in man. Hepatology 1991; 12: 1206-8. Kaplan MM, Alling DW, Zimmerman H J, et al. A prospective trial of colchicine for primary biliary cirrhosis. N Engl J Med 1986; 315: 1448-54. Warnes TW, Smith A, Lee FI, Haboubi NY, Johnson P J, Hunt L. A controlled trial of colchicine in primary biliary cirrhosis: trial design and preliminary report. J Hepatol 1987; 5: I-7. Bodenheimer H Jr, Schaffner F, Pezzullo J. Evaluation ofcolchicine therapy in primary biliary cirrhosis. Gastroenterology 1988; 95: 124-9.
Journal of Hepatology, 1993; 17:332-338 © 1993 ElsevierScientificPublishers Ireland Ltd. All rights reserved. 0168-8278/93/$06.00
HEPAT 01255
Ursodeoxycholic acid for symptomatic primary biliary cirrhosis Preliminary analysis of a double-blind multicenter trial P.M. Battezzati, M. P o d d a , F.B. Bianchi, R. N a c c a r a t o , F. Orlandi, C. Surrenti, L. Pagliaro, F. M a n e n t i a n d the Italian Multicenter G r o u p for the Study of U D C A in PBC* (Received 3 December 1991)
The administration of ursodeoxycholic acid, a hydrophilic bile acid not hepatotoxic to humans, has been suggested for treatment of primary biliary cirrhosis to improve cholestasis and reduce hepatocellular damage. Efficacy of treatment has been studied mainly in patients with asymptomatic or early-stage disease. In January 1988, to establish the efficacy and safety of ursodeoxycholic acid in a population with more severe disease, we started a multicenter, double-blind, placebo-controlled trial in patients with symptomatic disease, that is, with pruritus or serum bilirubin exceeding 2 mg/dl. Forty-four patients were assigned to ursodeoxycholic acid, 500 mg daily (corresponding to about 8.7 mg/kg body weight in these patients), and 44 to a placebo. As planned at the beginning of the study, a preliminary analysis was performed when all patients had been followed for at least 6 months (33 patients up to 12 months). Pruritus, self-evaluated by the patients, and cholestyramine consumption, as recorded in a diary, decreased significantly (p < 0.01) in both groups. In patients who initially had abnormal levels, serum bilirubin decreased significantly (p < 0.05) in the ursodeoxycholic acid group compared to placebo. After 6 months the following were also significantly better in the ursodeoxycholic acid than in the placebo group: a composite weighted biochemical index taking into account the changes in serum bilirubin, alkaline phosphatase, ),-GT and AST (p<0.001); serum prealbumin (p<0.05); IgG (p < 0.01) and IgM (p < 0.01) levels. The positive effects of ursodeoxycholic acid administration on serum bilirubin, the most important prognostic factor in primary biliary cirrhosis, and on liver protein synthesis suggest that ursodeoxycholic acid may be useful for patients with more advanced disease than those so far included in therapeutic trials. K e y words: Ursodeoxycholic acid; Bile acids; Chronic cholestasis; Primary biliary cirrhosis; Pruritus; Icteric
A beneficial effect of ursodeoxycholic acid (UDCA) in patients with primary biliary cirrhosis (PBC) was first reported in uncontrolled pilot studies (1) and was confirmed more recently in placebo-controlled, double-blind studies (2,3). In these studies, mainly subjects with mild symptoms and early stages of the disease were included, and the effects of treatment could not be analyzed separately for the subset of patients with advanced PBC. This is important, since progression of the disease in patients
with asymptomatic PBC is slow (4,5) and there are other preliminary reports indicating that UDCA may fail in advanced stages of PBC (6-8). With this in mind, we performed a randomized controlled trial including only patients with symptomatic disease, that is, with pruritus or serum bilirubin exceeding 2mg/dl, to establish whether or not UDCA is effective against the symptoms and indices of the severity of PBC.
Correspondence to: Professor Mauro Podda, Cattedra di Medicina Interna, Istituto di Scienze BiomedicheS. Paolo, UniversitY.di Milano, via di
Rudini, 8, 20142 Milano, Italy. *The Italian MulticenterGroup for the Study of UDCA in PBC consisted of the following:Investigators and Clinical Centers: M. Podda, P.M. Battezzati,E. Bertolini,M. Camisasca,A. Crosignani,M. Zuin (Universityof Milan);F.B. Bianchi,M. Fusconi,G. Ballardini,F. Cassani(University of Bologna);F. Orlandi, G.P. Macarri, A. Benedetti(Universityof Ancona); R. Naccarato, M. Chiaramonte,A.R. FIoreani (Universityof Padova); C. Surrenti, M. Pozzi, M.R. Biagini(Universityof Firenze); L. Pagliaro, P. Almasio,G. Provenzano,I. Scalisi(Universityof Palermo); F. Manenti, S. Bellentani, G. Tabarroni (University of Modena); Bile acid analysis: A. Roda (University of Bologna); Determination of AMA: F.B. Bianchi (University of Bologna);Measurement ofPIllP: G. Annoni (Universityof Milan); Measurement ofprealbumin: G. Bisiani(Universityof Milan).
UDCA FOR SYMPTOMATICPBC Patients and Methods
Patient selection Patients with clinical histories and biochemical profiles consistent with PBC, who presented with pruritus or serum bilirubin exceeding 2 mg/dl, were considered for admission into the study. Diagnosis of the disease was re-evaluated according to the following criteria, which were modified from Taal et al. (9): a positive test (titer ~>1:40) for antimitochondrial antibodies (AMA), a pre-entry liver specimen considered to be consistent with or diagnosis of PBC (10). If one of these major criteria was not met, three of the following were required for the diagnostic of PBC to be confirmed: serum levels of alkaline phosphatase higher than 2-times the upper reference value, IgM levels >t280 mg/dl, pruritus, serum bilirubin >2 mg/dl, a positive result in Schirmer's test. Patients also had to have radiological or ultrasonographic evidence of the absence of extrahepatic biliary obstruction. Inclusion criterion was presence of symptomatic disease, as defined by presence of pruritus and/or serum bilirubin >2 mg/dl. Pruritus had to have been present in the 6 months prior to evaluation, serum levels of bilirubin had to have been >2 mg/dl at two separate check-ups in the previous 6 months. Percutaneous liver biopsy had to be repeated if it had been performed more than 12 months prior to admission for patients who had been classified as stages I-III. Serum bilirubin levels higher than 10 mg/dl, ascites, previous episodes of variceal bleeding or encephalopathy, evidence of malignant conditions, and alcohol abuse were criteria for exclusion. No patient was taking any medication known to be hepatotoxic nor had been treated with corticosteroids, immunosuppressant agents, coichicine, penicillamine or UDCA in the previous 6 months.
Study design Patients were informed of the nature, purpose and requirements of the study before giving oral consent. The study was carried out according to the principles of the Declaration of Helsinki for research in human subjects. At the time of enrollment patients underwent clinical and laboratory evaluations. An ultrasound examination of the upper abdomen was performed to confirm the absence of extrahepatic biliary obstruction or of liver tumors. UDCA, 500 mg daily, or placebo were administered double-blind in two divided doses at mealtime. Randomization of treatment assignments was performed separately for each center: patients were consecutively given indistinguishable medications, which had been assigned
333 by the central pharmacy according to a computergenerated list. UDCA and an identical-appearing placebo were obtained through the courtesy of ABC Farmaceutici, Torino, Italy. Patients were allowed to continue taking supportive medication, such as calcium and vitamins. When required for control of pruritus, cholestyramine could be taken by the patients, who were instructed to take it at least 4 h before or after the study drug. Clinical examination, routine hematological tests and blood chemistries were performed every 3 months. Ultrasonography was repeated every 6 months. Assessment of intensity of pruritus at each visit was expressed by a numerical score, taking into account the symptom's intensity, cholestyramine consumption, and presence of skin marks. During the week preceding each visit, patients were asked to report intensity of pruritus and cholestyramine consumption in a diary. Intensity of pruritus was rated on a scale from 0 (absent) to 2 (severe); the need for cholestyramine for control of pruritus was taken into account as follows: 1 point was added to the intensity score of patients needing no more than 8 g daily and without skin marks, 2 points were added otherwise. The intensity score used for statistical purposes was the mean of the assessment at the visit and the daily reports during the week preceding the visit. Clinical observations and blood test results at each visit were recorded on standard data-collection forms and forwarded to the study monitor with the patient's diary. Compliance with the treatment was assessed by counting the capsules returned to the clinic by the patient and by the determination of serum UDCA levels. The study started in January 1988 and patient enrollment was terminated in June 1988. According to the study protocol, a preliminary analysis was conducted on data available on January 1989, when all patients had been followed for at least 6 months. Patients agreed that if this analysis showed sufficient evidence of a beneficial effect of UDCA, they could be placed on this treatment and continue the study according to a different protocol. The pre-defined major parameters for evaluation of treatment efficacy were decrease in serum bilirubin, improvement of pruritus and improvement of serum liver enzymes, which was assessed by a composite biochemical index to avoid the need for repetitive significance testing. The index was calculated for each time point as the weighted mean of changes (expressed as ratio to baseline values) of serum bilirubin, aspartate aminotransferase (AST), alkaline phosphatase (ALP),
334 and ~,-glutamyl transpeptidase (GGT), according to the following formula: [2 x (bilirubin ratio) + (AST ratio) + (ALP ratio) + ( G G T ratio)I/5 Values higher than ! indicate an overall worsening of serum liver enzymes, lower than 1 indicate improvement. The sample size determination was based on the results of a pilot study of the effects of U D C A administration to symptomatic patients with PBC (11). The composite biochemical index for that group of patients indicated that a reduction of about 0.60 standard deviations from baseline values had been achieved during administration of 500 mg/day UDCA. We calculated that a sample size of 45 patients for each treatment group would yield a probability of a type II error of 0.80, at the 5% level of significance, in detecting such a difference between U D C A and placebo-treated patients, assuming that no improvement in the composite biochemical index occurred in the latter group.
P.M, BATTEZZATIet al. to the reference values for each laboratory and are expressed as multiples of the upper limit of reference values. To compare changes over time between the two treatment groups in the continuous variables, the ratio of values at a particular time to baseline values was computed for each patient. Changes in the pruritus score were expressed as the difference between values at a particular time and baseline values. Quantitative data were analyzed by the Wilcoxon rank signs test and the Mann-Whitney test. Relative frequencies were compared by the Chi-square test and Fisher's exact test. Correlations were calculated by linear regression analysis; data were transformed logarithmically when departures from normality had been detected. Two-tailed p values were used for statistical comparisons. The total period of observation was used in the analysis of data for each patient, according to the 'intention to treat' principle.
Results
Clinical and laboratory investigations Fasting blood samples were collected from patients at each control visit. Laboratory tests included full blood counts, standard liver function tests, AMA determination, serum lipids, albumin, globulins, quantitative immunoglobulins, prothrombin time, serum creatinine, urea, serum prealbumin, serum bile acids and serum procollagen III peptide (PIIIP). Serum bile acids and serum P I I I P were determined on patients enrolled at four of the participating centers (53 patients from the centers of Milan, Bologna, Palermo and Modena). Hematological tests and blood chemistries were done in the clinical laboratory of the participating hospitals by routine automated techniques. Some of the tests were performed centrally, namely serum AMA, serum prealbumin (radial immunodiffusion commercial kit, Behringwerke AG, Marburg, Germany) (12), P I I I P (commercial radioimmunoassay, Boehringer RIA), individual bile acids (solid phase immunoenzymatic assays) (13). For this purpose, aliquots of serum were collected from each patient at 6-month intervals and kept frozen at - 20 °C until the time of testing.
Analysis of data As stated in the protocol, the interest of this analysis was focused on changes in pre-defined parameters after at least 6 months of treatment. To allow a more detailed analysis over the entire study period, results were also reported for data obtained at different times, up to 1 year. Liver enzyme values were standardized according
The study began on January 1, 1988 and patient enrollment was terminated in June 1988. During that period, 88 patients with PBC were admitted to the trial in the seven centers. Forty-four patients were randomly assigned to U D C A treatment, and the same number to placebo. The two treatment groups were similar at entry into the study for the most relevant clinical and laboratory characteristics (Tables !-3). Mean weight was 54 kg for both groups. In the U D C A group, the treatment dose of 500 mg per day corresponded to an average daily dose of 8.7 mg/kg body weight (range 5.4-1 i.6). During the study period, six of the patients (five in TABLE 1 Clintcal and laboratory characteristics of patients with primary bdiary cirrhosis, at entry into the study Characterisuc UDCA Placebo Subjects 44 44 Mean age (years) 54+2 55+2 Women ('~-) 84 93 Mean duration of disease (years) 7+ I 7+ I AMA positive (%) 84 91 Histologic stage IV (%) 48 50 Varices (%)a 26 45 Pruritus present (%/ 91 89 Serum bilirubin > 2 mg/dl (%) 34 32 Mean ALT l× N)u 2.76+0.25 2.38+0.15 Mean AST Ix N) 2.51 +0.17 2.49+0.17 Mean alkaline phosphatase {× N) 4.99-+0.43 4.55+2.44 Mean -;-glutamyltransferase(× N) 14.90_+2.09 12 87 + 1.44 Plus/minus values are means+ SE. aData for 35 subjects in the UDCA and for 30 patients in the placebo group. bData for serum liver enzymes are reported as multiples of the upper limit of normal reference values (x N).
UDCA FOR SYMPTOMATIC PBC
335
the U D C A group) were classified as d r o p - o u t s . T w o of
TABLE 3
them, both in the U D C A group, were lost to follow-up
Serum bile acids at entry and after 6 months of treatment
before the first m o n t h of t r e a t m e n t was completed: o n e
At entry
left the study due to personal choice, and the o t h e r due
After 6 months
Serum UDCA (.umol/I)' UDCA <5 23.5 + 3.9 Placebo <5 <5 Serum cholic acid (,umol/l}" UDCA 12.7±2.5 7.44-1.6 Placebo 9.6 ± 2. I 9.2 ± 1.5 Serum chenodeoxycholic acid (,umol/l)° UDCA 15.6+4.2 10.3±2.7 Placebo 8.0 4-1.4 10.4 ± 1.9
to the d e v e l o p m e n t of a skin rash. T w o patients on
UDCA an d one on p la c e b o were unable to tolerate the medication because of w o r s e n i n g of pruritus. O n e patient was w i t h d r a w n from t r e a t m e n t with U D C A as a result of a recurrent episode of h e m o l y t i c anemia. She was taking U D C A and was s u b s e q u e n t l y treated with ste-
p<0.01 vs. placebo p<0.01 vs. placebo
Plus/minus values are means ± SE.
roids. All the r e m a i n i n g subjects took m o r e than 80%
aData for 26 subjects in the UDCA group and for 27 subjects in the
of the prescribed dose, as assessed from the counts of
placebo group.
unused capsules at each visit. In five patients serum U D C A levels were lower than 5/amol/l, and they were considered to be n o n - c o m p l i a n t . N o n e of the patients
following analyses and, with the ex cep t i o n of the patient
in the p l aceb o g r o u p had serum levels of U D C A higher
w h o was given steroid t r e a t m e n t because of the recur-
than 5 ~umol/i. All the patients were included in the
rence of h e m o l y t i c anemia, data from their total periods
TABLE 2 Climcal and laboratory values at entry and after 6 months of treatment At entry Pruritus score 2.3±0.2 UDCA 1.9±0.2 Placebo Bilirubin, all patients (mg/dl) 1.85 ± 0.24 UDCA Placebo 1.91 ± 0.22 Bdlrubin, patients with abnormal values at entry (mg/dl)~ 2.67 _+0.32 UDCA 2.59 ± 0.30 Placebo Composite biochemical index I UDCA 1 Placebo Albumin (g/dl) 4.07±0.10 UDCA 4.09 + 0.08 Placebo ;,-Globulins (g/dll 1.91+0.11 UDCA 1.87+0.09 Placebo IgG (mg/dl) 2018±132 UDCA I891 ± 110 Placebo IgM (mg/dl} 501 ± 47 UDCA 570 ± 56 Placebo IgA [mg/dl) 333±30 UDCA 329 ± 29 Placebo Fotal cholesterol (mg/dl) 263±12 UDCA 266± 13 Placebo HDL cholesterol (mg/dl) 78±6 UDCA 61±5 Placebo Prealbumin (mg/dl, n.y. > 20) 16.0±1.1 UDCA 16.9±1.1 Placebo PIIIP (ng/ml, n.v. < 14.5)b 19.6±1.4 UDCA 22.9±1.9 Placebo
After 6 months 1.4±0.2 1.3±0.2 1.60 ± 0.20 1.93 ± 0.28 2.24 ± 0.28 2.67 4-0.39
p < 0.05 vs. placebo
0.82 ± 0.04 0.96 __+0.03
p<0.001 vs. placebo
4.19+0.09 4.04 ± 0.08 1.88±0.12 1.98±0.10 2077±141 2038±95
p=0.01 vs. placebo
504 ± 54 607 ± 69
p = 0.05 vs. placebo
351±34 362 ± 31 263+14 261 ±11 72±6 61+4 17.7±1.2 16.4±I.0 18.8±1.4 23.3±1.6
Plus/minus values are means+ SE. 'There were 24 subjects in the UDCA group and 27 subjects in the placebo group. ~'Data for 26 subjects in the UDCA group and for 27 subjects in the placebo group.
p<0.05 vs. placebo at entry p<0.05 vs. placebo
336
P.M. BATTEZZATIet al.
of observation were used. At the time of this analysis the median observation period was 9 months, 86 patients (42 in the UDCA group) had been followed for at least 6 months, and in 33 patients (17 in the UDCA group) the observation period lasted one year. The changes in the pruritus scores, serum bilirubin and the liver function tests of the composite biochemical index, and other laboratory variables of interest after 6 months of treatment are reported in Table 2. Highly significant differences were observed between the UDCA and the placebo groups in the composite biochemical index, and they were even more marked when this index was re-calculated without the contribution of serum bilirubin. Changes with time up to 1 year of treatment for pruritus and serum bilirubin are reported in Fig. 1. The pruritus score was significantly lower than at entry in both groups. Eleven of the 25 patients in the UDCA
Pruritus score 3,
1.5
o
~
~
~
1'2 mos afterentry
Bilirubin (mg/dL), all patients
T .....
I. . . . . . . . . I...
group who were initially taking cholestyramine could discontinue it after 6 months of treatment, compared to 5 of 19 patients in the placebo group. There was a significant decrease in serum bilirubin values in the UDCA group after 6 months of treatment, considering only the 51 patients with abnormal bilirubin values at entry (Fig. 1). Changes in serum UDCA and primary serum bile acids are reported in Table 3. There were no correlations found between the dose of UDCA, as mg/kg body weight, and changes in any of the laboratory values or changes in the pruritus scores in the UDCA group. Worsening of pruritus was the major complaint of five patients in the UDCA group and of four placebo patients. Of these, two patients in the UDCA and one in the placebo group were withdrawn from the study; one other patient, who was taking UDCA, had to temporarily reduce the dosage to 1 capsule daily and tolerated well the reinstitution of the protocol dose. One patient given UDCA was withdrawn from the study because of the development of a skin rash after 2 weeks of treatment. This side effect disappeared after discontinuation of therapy, and the patient reported that a similar episode had occurred after she took the drug 1 year before entering the study. Minor side effects reported were development of diarrhea during the first days of treatment, which occurred in four patients in the UDCA group, and minor and transient abdominal complaints. They required no treatment nor any modifications of the treatment schedule.
I
1.5
Discussion
UDCA Placebo
42 (§) 44
42 44
41 44
26 30
17 16
~. ~, o"o~e
Bilirubin (mg/dL), patients with abnormal values at entry T
3
0
UDCA Placebo
24 (§) 27
24 27
24 27
14 16
9 11
,t & ~- J o ~ ~o
(§) no of patient~ F i g . I . C h a n g e s w i t h t i m e o f the pruritus score a n d o f s e r u m b i l i r u b i n .
Among patients with abnormal values at entry, serum bilirubin was significantly(p < 0.05) improved in the UDCA compared to the placebo group after 6 months.
The aim of this study was to assess the therapeutic value of UDCA administration for patients with symptomatic or advanced-stage PBC, as reflected by pruritus or high serum bilirubin levels. Assessment of treatment efficacy in this group of patients was considered important in view of some preliminary reports indicating that UDCA may fail in advanced stages of PBC (6-8). Compared to other controlled trials (2,3) patients included in our study had longer duration of disease, stage IV was found more frequently at histology, jaundice was more frequently present, thus indicating that as a whole they had more advanced disease. Serum bilirubin has been repeatedly identified as an independent prognostic indicator in PBC (4,5,14,15), and elevated values indicate a poor prognosis and advancing disease (16). Presence of symptoms has also been considered to indicate a poor prognosis in PBC (4). Pruritus is an often distressing symptom, with great individual variability, which can severely compromise quality of
UDCA FOR SYMPTOMATICPBC life (14). Fatigue, another symptom which is considered to be commonly reported by patients with PBC, is even more difficult to assess objectively and may be largely influenced by patient beliefs. For this reason we did not include it in the definition of symptomatic patients for this trial. In planning our study we chose not to include histological assessment in the evaluation of treatment efficacy. The patchy nature of hepatic damage in PBC makes serial biopsies difficult to interpret, and performing control liver biopsies in this group of PBC patients with advanced disease was considered unethical. In a recently published mathematical model, liver biopsy was also not required for predicting survival in patients with PBC (5). For simplicity of execution UDCA was administered at a fixed dose (500 mg daily) to each patient rather than on an individual basis. This dosage was chosen to minimize the risk of untoward effects in our patients, in view of a few case reports of patients with late-stage PBC who deteriorated on UDCA (6,8). The average daily dose was 8.7 mg/kg body weight, which was previously shown to be effective against the biochemical expression of liver damage in icteric patients with PBC (11). UDCA was well tolerated at this dosage, and no major side effects were noted during the study. Pruritus scores, as recorded by the patients in a diary and at each visit, decreased in both groups. Decreased pruritus following the administration of placebo has also been observed by others in a double-blind study (3), thus suggesting that there is an important placebo effect on this symptom. To obtain an objective estimate of pruritus, several devices have been proposed, including sensitive limb movement meters (17), but their accuracy was similar to the patient's subjective assessment. There was a significant improvement in serum bilirubin levels during UDCA administration in those patients with abnormal values at entry. Serum prealbumin significantly improved during UDCA treatment. Prealbumin is a protein synthesized
337 in the liver and is easily and reliably measured in the plasma without requiring the active cooperation of the patient. Its short half-life, 1.9 days, renders serum prealbumin a sensitive indicator of liver function capacity (18,19). Serum PIIIP levels, which are correlated with fibrogenesis in many liver diseases (20), did not change during UDCA administration suggesting that this drug does not influence collagen metabolism, at least over 6 months. Our results c o n f r m the striking improvements of serum liver enzymes following UDCA administration, even in the short term (2,3,11). Serum immunoglobulins, both IgG and IgM, did not deteriorate or even improved, thus confirming data by others on the beneficial effects of UDCA on some immunological features of PBC (2,3,21). There were significant decreases in serum cholic and chenodeoxycholic acids during UDCA administration. This is in agreement with the hypothesis that UDCA inhibits the intestinal reabsorption of endogenous bile acids (22,23). In conclusion, the data from our study confirm that UDCA administration for at least 6 months significantly improves biochemical abnormalities in patients with PBC. Although the design of our study does not allow us to assess the clinical importance of such changes, the positive effects we observed on serum bilirubin levels and on liver protein synthesis suggest th~.t UDCA may also be effective for patients with more advanced disease than those so far included in therapeutic trials. Available data indicate that the same group of patients can be safely given colchicine as an anti-fibrotic agent (24-26). Accordingly, after the completion of this analysis all patients were assigned to the UDCA treatment and colchicine or placebo was added in a double-blind fashion.
Acknowledgements This work was presented, in part, at the 40th Annual Meeting of the American Association for the Study of Liver Diseases in Chicago, IL, on October 28-31, 1989.
References I Poupon R, Chr&ien Y, Poupon RE, Ballet F, Calmus Y, Darnis F. Is ursodeoxycholicacid an effectivetreatment for primary biliary cirrhosis? Lancet 1987; i: 834-6. 2 Leuschner U, Fischer H, Kurz W, et al. Ursodeoxycholicacid in primary biliary cirrhosis: results of a controlled double-blind trial. Gastroenterology 1989;97: 1268-74. 3 Poupon RE, Balkau B, Eschwtge E, et al. A multicenter,controlled trial of ursodiol for the treatment of primary biliary cirrhosis. N Engl J Med 1991; 324:1548-54. 4 Roll J, Boyer JL, Barry D, et al. The prognostic importance of
5 6 7 8
clinical and histologicfeatures in asymptomatic and symptomatic primary biliary cirrhosis. N Engl J Med 1983;308: I-7. Dickson ER, Grambsch PM, Fleming TR, Fisher LD, Langworthy A. Prognosis in primary biliary cirrhosis: Model for decision making. Hepatology 1989; 10: I-7. VogelW, Kathrein H, Judmaier G, et al. Deterioration of primary biliary cirrhosis during treatment with ursodeoxycholicacid. Lancet 1988; i: 1163. Hadziyannis S, Hadziyannis E. A randomised controlled trial of ursodeoxychilic acid (UDCA) in primary biliary cirrhosis (PBC) (Abstr.). Hepatology 1988; 5: 1421. Raedsch R, Stiehl A, Thielmann L, et al. Influence of ursodeox-
338 ycholic acid on primary biliary cirrhosis depending on stage of the disease (Abstr.). Gastroenterology 1989; 96: A647. 9 Taal BG, Schalm SW, ten Kate FWJ, Hermans J, Geertzen RGM, Feltkamp BEW. Clinical diagnosis of primary biliary cirrhosis: A classification based on major and minor criteria. Hepatogastroenterology 1983; 30: 178-82. 10 Scheuer PM. Liver Biopsy Interpretation. 2nd Edn. Baltimore: Williams & Wilkins, 1973: 33-8. 11 Podda M, Ghezzi C, Battezzati PM, et al. Ursodeoxycholic acid for chronic liver diseases. J Clin Gastroenterol 1988; 10: $25-$32. 12 Mancini G, Carbonara A, Heremans JF. Immunochemical quantitation of antigens by immunodiffusion, lmmunochemistry 1965; 2: 235--40. 13 Podda M, Ghezzi C, Battezzati PM, Crosignani A, Zuin M, Roda A. Effects of ursodeoxycholic acid and taurine on serum liver enzymes and bile acids in chronic hepatitis. Gastroenterology 1990; 98:1044-50. 14 Kaplan MM. Primary biliary cirrhosis. N Engl J Med 1987; 316: 521-8. 15 Christensen E, Neuberger J, Crowe J, et al. Beneficial effects of azathioprine and prediction of prognosis in primary biliary cirrhosis. Gastroenterology 1985; 89: 1084-91. 16 Shapiro JM, Smith H, Schaffner F. Serum bilirubin: A prognostic factor in primary biliary cirrhosis. Gut 1979; 20: 137-40. 17 Summerfield JA, Welch ME. The measurement of itch with sensitive limb movement meters. Br J Dermatol 1980; 102: 275-81. 18 Hutchinson DR, Smith MG, Parke DW. Prealbumin as an index
P.M. BATTEZZAT1 et al.
19
20
21
22
23
24
25
26
of liver function after acute paracetamol poisoning. Lancet 1980; ii: 121-3. Teppo A-M, Maury CPJ. Serum prealbumin, transferrin and immunoglobulins in fatty liver, alcoholic cirrhosis and primary biliary cirrhosis. Clin Chim Acta 1983; 129: 279-86. Frei A, Zimmermann A, Wiegand K. The n-terminal peptide of collagen type III in serum reflects activity and degree of fibrosis in patients with liver disease. Hepatology 1984; 4: 830-4. Calmus Y, Gane P, Rouger P, Poupon R. Hepatic expression of class I and class II major histocompatibility complex molecules in primary biliary cirrhosis: effect of ursodeoxycholic acid. Hepatology 1990; I1: 12-5. Stiehl A, Raedsch R, Rudolph G. Acute effects of ursodeoxycholic and chenodeoxycholic acid on the small intestinal absorption of bile acids. Gastroenterology 1990; 98: 424-8. Marteau P, Chazouillrres O, Myara A, Jian R, Rambaud J-C, Poupon R. Effect of chronic administration of ursodeoxycholic acid on the ileal absorption of endogenous bile acids in man. Hepatology 1991; 12: 1206-8. Kaplan MM, Alling DW, Zimmerman H J, et al. A prospective trial of colchicine for primary biliary cirrhosis. N Engl J Med 1986; 315: 1448-54. Warnes TW, Smith A, Lee FI, Haboubi NY, Johnson P J, Hunt L. A controlled trial of colchicine in primary biliary cirrhosis: trial design and preliminary report. J Hepatol 1987; 5: I-7. Bodenheimer H Jr, Schaffner F, Pezzullo J. Evaluation ofcolchicine therapy in primary biliary cirrhosis. Gastroenterology 1988; 95: 124-9.