Deranged CD95 system in a case of Churg–Strauss vasculitis

June 1998

compare these two techniques for evaluating more distal small bowel disease. Surely, Gasche et al. do not view it as unimportant that the SBFT added incidental information about the presence of duodenal disease. The comments by Drs. Kelvin and Maglinte regarding Figure 1 and the assessment of the duodenum with the SBE vis-a`-vis the SBFT and endoscopy underscore the fact that they have missed the main points of our study. The partially obstructing duodenal strictures in Figure 1A should not have ‘‘led to abandonment of the enteroclysis and substitution of a follow-through study.’’ This was a controlled study in which the patient underwent both procedures for the purpose of evaluating and comparing the results of both studies! It is of interest that Gasche et al. conclude that ‘‘regarding ileal disease, SBE is better or at least as good as SBFT.’’ If they are formulating that conclusion based on our data, they must have read a different paper, or if they are basing it on their own biases, they are falling into the trap of expounding the benefits of a procedure (the SBE) without any data (other than perhaps one false-positive SBFT in our data set) to support this. In fact, in our data the SBE did not ever reveal any data that were completely missed by the SBFT. We agree that the SBE may be as good as the SBFT (or vice versa) and, therefore, do not view patient preference or procedure safety as being frail reasons for choosing SBFT. Drs. Kelvin and Maglinte are so caught up in insisting that the SBE is the procedure of choice that they suggest the only way our study could have shown an equivalent efficacy of the SBFT is by producing inferior SBEs. We, too, have several beautiful double-contrast images of the small bowel by the SBE, but in our study these images in general did not add anything beyond what could be discerned with the SBFT and were not necessarily published in the figures. Dr. Maglinte and his group have been at the forefront of honing small bowel radiographic techniques. Unfortunately, we are unaware of any publications of theirs in which they have directly compared these techniques in the same patients with Crohn’s disease. We entered into this study without any bias for or against SBE. Our radiologists perform both studies equally well and with equal enthusiasm. We, in fact, were surprised that the SBFT performed as well as it did compared with the SBE. However, as with all other aspects of medicine, until a procedure is held out to rigorous study, as we have with small bowel radiography, one only can express biases. In North America, small bowel radiography is typically the procedure of choice for examining small bowel Crohn’s disease. If ultrasonography is proven to be as accurate, it may become more widely used on this continent as well. CHARLES N. BERNSTEIN, M.D. Inflammatory Bowel Disease Clinical and Research Centre University of Manitoba Winnipeg, Manitoba, Canada

Etiology of Fasting Hyperbilirubinemia Dear Sir: In the editorial entitled ‘‘Fasting Hyperbilirubinemia: Unraveling the Mechanism Involved,’’1 Dr. Fevery attempts to provide support for the concept that intestinal absorption of unconjugated bilirubin (UCB) accounts for the increase in serum UCB observed with fasting. To this end, he estimates that 50 µmol of UCB might be absorbed from the gut each day, and then calculates that this quantity of UCB would result in a 17-µmol increment in serum UCB (assuming distribution in 3 L of plasma). Although the latter calculation gives information on the increase in UCB concentration that would occur if 50 µmol were added to 3 L of plasma contained in a beaker, this calculation is irrelevant to the physiological situation. The steady-state plasma concentration of reabsorbed UCB is a

CORRESPONDENCE

1351

function of the amount absorbed and the efficiency of hepatic clearance of UCB and is independent of the pool size. Seemingly, the most logical approach to determining the influence of absorbed UCB on plasma level would be via a comparison of the absorbed load (50 µmol/day, according to Fevery) to the normal daily load of UCB resulting from heme turnover (about 500 µmol/day). Reabsorbed UCB represents only about 1/10 the nonreabsorbed load. Given linear (nonsaturating) hepatic removal of UCB, reabsorbed UCB would be expected to produce a negligible 10% increase in plasma UCB concentration. MICHAEL D. LEVITT, M.D. WILLIAM DUANE, M.D. Research Service Minneapolis Veterans Affairs Medical Center Minneapolis, Minnesota 1. Fevery J. Fasting hyperbilirubinemia: unraveling the mechanism involved (editorial). Gastroenterology 1997;113:1798–1800.

Deranged CD95 System in a Case of Churg–Strauss Vasculitis Dear Sir: In the October issue of GASTROENTEROLOGY, Pensati et al.1 report a case of autoimmune lymphoproliferative syndrome (ALPS) with autoimmune hepatitis that was caused by mutations of the CD95 (Apo-1/Fas) gene. Providing an in-depth analysis of heredity, they suggest that ALPS might be explained as a digenic autosomal recessive disorder. Because augmentation of CD42 CD82 T-cell subsets, autoantibody production, and defective CD95-mediated apoptosis are immunologic hallmarks of the syndrome, we would like to discuss a case of rare Churg–Strauss vasculitis (CSV) with similar immunologic features. A 47-year-old woman with CSV presented with peripheral blood eosinophilia, multiple gastrointestinal and cutaneous ulcerations with underlying lesions corresponding to granulomatous vasculitis, asthma with recurrent pulmonary infiltrations, and ischemia. Unlike ALPS, CSV is a nonhereditary disease whose underlying pathomechanisms are not yet clear. However, our patient with CSV shared several criteria for ALPS, among them a severe derangement of the CD95 system and a 2.6-fold increase in CD42 CD82 T cells. Resistance of peripheral blood lymphocytes (PBLs) to CD95-mediated apoptosis was found in this patient. Only 20% of the PBLs underwent apoptosis on exposure to an anti-CD95 antibody (100 ng/mL), whereas more than 90% of the PBLs from 5 healthy donors became apoptotic under these conditions. CD95 was functional in this patient except that, for a yield of .90% apoptotic PBLs, fivefold higher agonistic anti-CD95 antibody concentrations were required. PBLs from healthy donors acquired secondary resistance to CD95-mediated apoptosis when incubated with supernatants derived from PBLs of the CSV patient. These supernatants were less protective when they were pretreated with a neutralizing anti-CD95 antibody. The findings were suggestive of an intervening factor competing with transmembrane CD95 for binding to its natural ligand (CD95L) or the anti-CD95 antibody in our experiment rather than an inherited mutation of the CD95 gene as described by Pensati et al. The effect of the neutralizing anti-CD95 antibody indicated that the soluble CD95 isoform was present in the supernatant and was responsible for PBL protection against CD95mediated apoptosis. The presence of CD95 isoforms was confirmed by quantitative reverse-transcription polymerase chain reaction; PBLs from healthy donors predominantly expressed the messenger RNA (mRNA) of the membrane-bound CD95 isoform, whereas this expression was diminished about 50% in PBLs from the CSV patient.

1352

CORRESPONDENCE

Here, the CD95 expression switched to sevenfold increased soluble CD95 isoform mRNA levels. In addition, CD95L mRNA levels were increased fourfold in PBLs from the patient with CSV. Regarding intraepithelial lymphocytes (IELs) infiltrating the gastric mucosa, CD95L mRNA levels were about 20-fold higher in ulcerative mucosa compared with normal gastric mucosa, but distribution of CD95 isoforms was not altered. Because these data suggested the presence of cytotoxic autoreactive T-cell clones, PBLs and IELs from gastric mucosa were analyzed by a reverse-transcription polymerase chain reaction technique that determines the b-chain CDR3 size.2 We found deviations from the normal polyclonal diversity of Vb chain repertoires in six Vb families in PBLs from the CSV patient. When we studied IELs from ulcerative gastric mucosa, oligoclonal expansions were found in two Vb families; one of them was also detected in PBLs. Analysis of these two Vb families by Jb specific primers revealed two oligoclonal expansions, which were present in five independent ulcerative gastric tissue samples from the stomach. These clones are likely to be identical with cytotoxic T cells overexpressing CD95L. The persistence of these clones may be favored by overexpression of the soluble CD95 isoform in the peripheral blood. Blocking CD95L by the soluble CD95 isoform in peripheral blood has been implicated in failure of termination of the immune response and autoimmunity.3 Because of recurrent gastric bleeding, the patient with CSV underwent gastrectomy and was treated with corticosteroids for 14 days. Interestingly, the above-mentioned immunologic features of this disease normalized in parallel with the clinical improvement. Sensitivity of PBLs toward CD95-mediated apoptosis, as well as the predominant expression of the membrane-bound CD95 isoform, were restored. Moreover, after 2 weeks of treatment, the oligoclonal expansions disappeared in PBLs and the altered Vb families regained almost normal diversity. MARKUS MU¨SCHEN ULRICH WARSKULAT, Ph.D. DIETER HA¨USSINGER, M.D. Department of Gastroenterology and Infectiology CORDULA MOERS DIETMAR SIMON, M.D. Department of General Surgery Heinrich-Heine-University Du¨sseldorf, Germany JOS EVEN, M.D. Department of Immunology Institut Pasteur Paris, France 1. Pensati I, Costanzo A, Ianni A, et al. Fas/APO1 mutations and autoimmune lymphoproliferative syndrome in a patient with type 2 autoimmune hepatitis. Gastroenterology 1997;113:1384–1389. 2. Pannetier C, Even J, Kourilsky P. T-cell repertoire diversity and clonal expansions in normal and clinical samples. Immunol Today 1995;16:176–181. 3. Cheng J, Zhou T, Liu C, et al. Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 1994;263: 1759–1762.

Attention and Distraction in Pain Experiments Dear Sir: I read with interest the article by Accarino et al.1 exploring the effect of different psychological conditions in 8 healthy college

GASTROENTEROLOGY Vol. 114, No. 6

students subjected to intestinal balloon distentions. They showed that subjects reported a higher grade of perceived pain in the attention condition vs. the distraction condition. The attention condition consisted of anticipatory knowledge of intestinal distention signaled by a light source, whereas distraction consisted of a mental concentration task. The study of experimental pain is riddled with methodological problems, one of which is the instruction given to the subject to create the desired experimental condition.2 Despite the precautions taken, the condition that was labeled attention may well have been a condition of anxiety, with its focus (the light source) related to pain. Anxiety has been shown to have an influence on the experience of pain with the focus of the anxiety, related or unrelated to pain, to be important.3 Anxiety with focus related to the painful stimulus would be expected to increase pain perception.3 However, the fact remains that Accarino et al. have shown the ability to alter pain perception under different psychological conditions, whatever their label. This supports a cognitive component to acute visceral pain in healthy subjects. The authors’ extrapolation of these experimental data to the chronic pain of functional gut disorders has limitations. Accarino et al. speculate in their discussion ‘‘that patients with functional disorders pay more attention to their gut and that mental interventions could, theoretically, provide some therapeutic benefit.’’1 The reader should be aware that in the therapeutic literature, the term attention refers to a cognitive process that is used in attention-based coping strategies to help patients.4 This use of the term needs to be differentiated from an experimental attention condition. Although Accarino et al. showed a lower grade of perceived pain in the distraction condition, the reader should be cautioned against concluding that patients with functional pain need to be distracted from their guts. Pain coping may also be achieved by directing attention toward the pain and reinterpreting its meaning.4 It remains to be shown whether the chronic pain experienced by our patients with functional disorders will be better managed with attention-based cognitive strategies vs. distraction-based ones.5 ALEXANDRA ILNYCKYJ, M.D. Section of Gastroenterology Department of Internal Medicine University of Manitoba Manitoba, Canada 1. Accarino AM, Azpiroz F, Malagelada JR. Attention and distraction: effects on gut perception. Gastroenterology 1997;113:415–422. 2. Eccleston C. The attentional control of pain: methodological and theoretical concerns. Pain 1995;63:3–10. 3. Janssen SA, Arntz A. Anxiety and pain: attentional and endorphinergic influences. Pain 1996;66:145–150. 4. Devine DP, Spanos NP. Effectiveness of maximally different cogitive strategies and expectancy in attenuation of reported pain. J Pers Soc Psychol 1990;58:672–678. 5. Eccleston C. Chronic pain and distraction: an experimental investigation into the role of sustained and shifting attention in the processing of chronic persistant pain. Behav Res Ther 1995;33: 391–405.

Reply. We appreciate the comments by Dr. Ilnyckyj to our recent paper. We agree there are possible limitations inherent to studies of mental activity on perception, a message we purposely made very clear in our manuscript. Dr. Ilnyckyj reflects on the undissociability of attention and anxiety, adapting arguments from the somatic pain literature. To this