A826 AGA ABSTRACTS GASTROENTEROLOGY, Vol. 108, No. 4 • FAMILIALITY IN CLINICAL CHARACTERISTICS OF CROHN'S DISEASE IN 69 FRENCH FAMILIES. C.Gower-Rousseau, S.Plegat, JP Evrard., B Grandbastien, JP Hugot, IL Dupas, JP Gendre, R Modigliani, J B61aiche, J Hostein, A Cortot, JF Colombel. Reglstre des MICI du Nord-Ouest de la France; INSERM CJF 9201, Institut Curie, Paris; GETAID. Familial aggregation argues for a genetic component in susceptibility to Crohn's disease (CD). The aims of this work were 1° to compare the age of onset of CD in familial and sporadic cases; 2 ° to study the concordance for location and transmural aggressiveness among first-degree relatives with CD. Patients and methods : 69 families with 2 (n=52), 3 (n=8), 4 (n=6) and 5 or more (n=3) affected first-degree relatives were studied. Twenty-six families were issued from an hospital and 43 from general population. All data were collected in a questionnaire by the same interviewer practitian. Results concerning the age of onset of CD were compared to those of sporadic cases referred by the Registre du Nord Ouest de la France (1), usmg Kruskal-Wallis test. Results : Data were obtained in 163 patients from the 69 families. The mean CD duration was 10 years. Familial occurence was the highest in siblings (82%). The median age of onset of CD was lower in familial cases than in sporadic cases : 22.5 years vs 26.5 years (p<0.01). Among the 104 patients from families with 2 affected first-degree relatives, 62 (60%) were concordant for location and 48 (46%) for transmural aggressiveness. Among the 59 patients from the 17 other families, 51 (86%) were concordant for location and 45 (76%) for transmural aggressiveness. Conclusions : Familiality was observed in clinical characteristics of first-degree relatives with CD and particularly in mutiplex families. These families may thus represent an homogeneous clinical group well fitted for genetic studies. (1). C.Gower-Rousseau et al. Gut 1994; 35: 1433-8. PROLONGED REMISSION IN CROHN'S DISEASE FOLLOWING THERAPY FOR MYCOBACTER/UM PARATUBERCULOSIS INFECTION. D.Y. Graham, M.T. AI-Assi, M. Robinson. Depts. of Medicine, VAMC and Baylor College of Medicine, Houston, TX and University of Oklahoma College of Medicine, Oklahoma City, OK. For more than a decade, there has been interest in M. paratubercu/osis (M. para) as a possible cause of Crohn's disease (CD). M. para is a very slow growing member of the MAI group of mycobacteria first isolated as the cause of Johne's disease of cattle and more recently associated with CD. tn vitro antimicrobial sensitivities of MAI correlate poorly with in vivo effectiveness. In contrast, antimicrobial activity done in mouse peritoneal macrophages correlates much better. Clarithromycin has been shown i~ be an effective anti-M, para antimicrobial (Eur J Gastroenterol Hepatol 1993;5:613). This study was designed to evaluate clarithromycin for the treatment of patients with active Crohn's disease. Patients with CD and a CDAI >200 were randomized to receive clarithromycin 500 mg b.i.d, or matching placebo for 3 months. Those whose CDAI remained >150 were then crossed over to the alternate medication. Those who achieved clinical remission continued on the same medication for 3 additional months. 15 patients completed the study. The mean CDAI pretreatment was 263. 5 0f the 7 patients who received clari first achieved remission by the end of the initial 3 months of therapy (pre-Rx CDAI = 246, Post-Rx CDAI = 98.2. Two other clad patients had marked falls in CDAI (from 445 to 292 and from 319 to 182). In contrast, only 1 placebo patient had a normalization of the CDAI. At the end of phase 2, another clari patient had normalization of the CDAI. Follow up approximately 1 year after completion of study drug shows that those who achieved remission ICDAI <150) have remained in remission. We conclude that it may be possible to achieve prolonged remission of Crohn's disease following a short course of clari, possibly due to its effect on M. para. • IL-I~ DOWNREGULATES COLLAGEN AND AUGMENTS GOLLAGENASE EXPRESSION IN HUMAN INTESTINALSMOOTH MUSCLE (HISM) CELLS M. Graham, A. Willey, J. Adams, D. Yager, H. Sugerman, R. Diegelmann. Depts. Pediatrics & Surgery, MCV/VCU, Richmond VA. Smooth muscle cells resident in the intestinal wall play a signiflcani role in the healing of the injured intestine and in the fibrosis that complicates Crohn's disease. In order to determinethe precise role of the inflammatory cytokine interleukin-t~ in these processes, the effect Of human, recombinant IL-I~ on human intestinal smooth muscle cell proliferation and collagen metabOlism was analyzed in vitro. Procollagen and collagen synthesis and secretion were determined by the quantitation of newly synthesised, radiolabelled procollagen accumulating in cells and culture medium by slab gel electrophoresis. Steady state levels of procollagen I and III mRNA were determined by Northern blot analysis, and Of collagenase mRNA levels by RNAse protection assay. IL-I~ caused a 3-fold increase in [3H] thymidine:uptake by HISM ceils at 100 pM. This mitogenic effect was equipotent with that of: platelet derived growth factor when cells were exposed to IL-[~, for 48 ve 24 hours. IL- z~ (100 pM) inhibited the secretion of procollagen into culture medium by 70 % and the accumulation of newly-synthesised procollagen in cells by 55 % suggesting that the predominant effect of IL-I~ was on the synthesis of procollagen and not on secretion: Pepsin digestion of procollagen demonstrated marked inhibition of both type I and III collagen synthesis. IL-11~ caused a concentration-dependent inhibition of steady state levels of procollagen I and III mRNA (85 % inhibition at 100 pM) and a 3-5 fold, concentration-dependent, augmentation of collagenase mRNA levels. The effects on collagen and collagenase expression were all abrogated by dexamethasone (10-6 M) and were specific to IL-1 I~ because they were not seen in reponse to PDGF. These data demonstrate that interleukin-[~ is mitogenic for human intestinal smooth muscle cells, but that the mitogenic action of this cytokine is associated with a concomitant, transcriptionally-mediated downregulation of collagen synthesis and secretion, and augmentation of collagenase expression. We propose that the mitogenic and collagenolytic effects of 11.-1!3 on human intestinal smooth muscle ceils are involved in an initial phase of intestinal repair - the proliferation and movement of the resident smooth muscle cells into the area of injury~ and that the elaboration of the healing collagenous matrix follows, in a second phase, in response to TGF-~. Failure of healing, seen in fistula formation and toxic megacolon may result from the unopposed expression of IL-I~. NIH grant DK34151 HOME TOTAL PARENTERAL NUTRITION (HTPN) IN ADVANCED AIDe. L:M. Gramlich. E. A. Mascioli. Deaconess Hospital, Harvard University, Boston, MA. Objective: TO describe characteristics of patienls with AIDS referred for HTPN and Io determine predictors of change in ft~hcti0nal and nu|ritional'status as well as the complications of HTPN. Patients: 31 consecutive AIDS patierlts who were referred for HTPN on the basis of intractable weight 10ss and/or gastrointestinal iG.!,) symptoms were studied. Parameters studied included weight, albumin, lymphocyte count, Karnofsky performance score(KPS), durationof HIV positivily, CO4 count, AIDS defining illnesses and AIDS related G:I. disease. Metabolic and line- related complications were also noted. Patients were studied at the initiation of HTPN, and after 4 and 8 weeks. Results: 31 patients were studied; of whom 28 were men and 25 had homosexual acquisition. Average known HIV posiiivity was 42+/:20 months and CD4 count was 23+1-19/mm'3;. Patients had and average of 1.6+/-1 AIDS defining illnesses, The HTPN was given for 22+/-21 weeks. At the stad of the study the patients weighed 84+/-11% of their ideal body weight (IBW). Initial serum albumin was 3.4+l-0.7g/dl which correlated positively with %IBW(p=O,O07). Nutritional parameters did not change with HTPN. KPS was 48+/-15 at the start of HTPN. indicating a significant degree of functional impairment, which significantly improved from initiation of HTPN to 4 weeks (KPS=S4+/-20, p=0.013) and this effect was maintained at 8 weeks(KPS=64+/-24, p<O.O0t). There was a trend tewards greatest improvement in KPS in those whose starting KPS was greater than 30. The incidence of line-related complications was 0.591100 catheter days and proven infection rate was 0.481100 catheter days. There were no metabolic complicalions associated with the HTPN. Conclusions: Patients are significantly malnourished and functionally impaired when referred for HTPN HTPN does not result in change in nutritional markers but there is weight maintenance. HTPN had a Positive impact on nutritional status but this is seen to a lesser degree in patients with low KPS tested. HTPN in this group of patients does not appear to be associated with an increased rate of line related infections compared to other AIDe patients with indwelling lines.