Effects of HI-antihistamine drug regimen on histamine release by nonlesionai skin mast cells of patients with chronic urticaria

Transcription

1 VOLUME Richman PG, Khan HA, Turkeltaub PC, Malveaux FJ, Baer H. The important sources of German cockroach allergens as determined by RAST analysis. J ALLERGY CLIN IMMUNOL 1984;73: Bernton HS, McMahon TF, Brown H. Cockroach asthma. Br J Dis Ches 1972;66: Steinberg DR, Bemstein DI, Gallagher JS, Arlian L, Bemstein IL. Cockroach sensitization in laboratory workers. J ALLERGY CLIN IMMUNOL 1987;80: Kang B, Chang JL. Allergenic impact of inhaled arthropod material. Clin Rev Allergy 1985;3: Wu CH, Lan JL. Cockroach hypersensitivity: isolation and partial characterization of major allergens. J ALLERGY CLrN IMMUNOL 1988;82: Matthews KP. Inhalant insect-derived allergens. Immunol Allergy Clin N Am 1989;9: Characterization of cockroach allergens 10. Stankus RP, O'Neil CE. Antigenic/allergenic characterization of American and German cockroach extracts. J ALLERGY CLIN IMMUNOL 1988;81: Bemton HS, Brown H. Insect allergy: the allergenicity of the excrement of the cockroach. Ann Allergy 1970;28: Menon P, Hillman B, Menon V, Stankus R, Lehrer S. Skin test reactivity to whole body and fecal extracts from American and German cockroach [in press]. Ann Allergy. 13. Homer WE, Kailas J, Stankus RP, Lehrer SB. Common German cockroach whole body and fecal allergens: immunoprint inhibition studies [submitted for publication]. 14. Lehrer SB, Homer WE, Menon PK, Stankus RP. Comparison of cockroach allergenic activity in whole body and fecal extracts [submitted for publication]. Effects of HI-antihistamine drug regimen on histamine release by nonlesionai skin mast cells of patients with chronic urticaria Chantal Brunet, BSc, Pierre-Michel Bddard, MD, and Jacques Hdbert, MD Ste-Foy, Quebec, Canada Profiles of compound 48~80-induced histamine release (HR ) from mast cells of uninvolved skin from patients with chronic urticaria (CU) and from a normal control (NC) group were compared, and the effects of anti-h1 medications were assessed versus placebo. Then, patients with CU (15) and NC subjects (10) were randomly assigned to take either hydroxyzine (100 mg/day), terfenadine (120 mg/day), or placebo for 28 days. The effects of such treatment on the clinical response and on the profile of compound 48~80-induced HR during a 4-hour period were analyzed. Treatment with hydroxyzine in patients with CU improved the clinical symptoms and modified the profile of HR; more histamine was recovered at 1 hour (p < 0.05) and 2 hours (p < 0.05), as compared with baseline. Terfenadine and placebo had no effect on the clinical response or on the profiles of HR. In the NC group, the amounts of histamine recovered at 1 hour after challenge with compound 48/80 were lower than amounts of the pretherapy values (p < 0.01). It could be concluded that (1) the profile of HR in patients with CU is reproducible during a period of 28 days, (2) only hydroxyzine modifies both the clinical response and the profile of HR, and (3) anti-ht compounds decrease the HR in the NC group. (J ALLERGY CLIN IMMUNOL 1990;86: ) From the Unit6 de Recherche en Inflammation et Immunologie- Rhumatologie, Centre Hospitalier de l'universit6 Laval, Ste-Foy, Quebec, Canada. Received for publication March 20, Revised May 1, 1990, Accepted for publication May 15, Reprint requests: Jacques H6bert, MD, Unit6 de Recherche en Inflammation et Immunologie-Rhumatologie, Centre Hospitalier de l'universit6 Laval, No. 9800, 2705 Blv. Lanrier, Ste-Foy, Qu6bec, Canada. G 1V 4G2. 1/1/23577 Histamine is released by skin mast cells on immunologic and nonimmunologic stimuli. A large body of evidence supporting its role in the pathophysiology of idiopathic CU has been demonstrated."5 Indeed, clinical manifestations of CU are comparable to manifestations produced by intradermal injection of histamine, and both are suppressed by anti,h: antihistamine drugs. Furthermore, the amounts of histamine found in lesional skin or in skin blister fluids are more elevated in patients with active CU than in 787

2 788 Brunet et al. J. ALLERGY CLIN. IMMUNOL. NOVEMBER 1990 Abbreviations used HR: Histamine release NC: Normal control CU: Chronic Urticaria PBS: Phosphate-buffered saline NC subjects. Indeed, Kaplan et al. 3 reported that 93% of patients with CU have a sixfold increase in histamine levels into the blister fluids overlying lesions, and more than half of these patients also exhibited elevated histamine release in areas of normal skin. Phanuphak et al. 4 found an increase in skin histamine content in both urticarial lesions and normal skin sites from patients with CU. Recent developments have shed some light on the reasons for the augmented HR by skin mast cells in these patients. Indeed, using the skin chamber model, we have previously reported that, on challenge with a nonspecific degranulator, compound 48/80, the nonlesional skin mast cells of patients with CU could release more histamine than skin mast ceils of an NC group in a 4-hour time-course experiment. 6 Indeed, the histamine increase was, respectively, two times and two and one-half times higher at first and second hour after compound 48/80 stimulation in patients with CU as compared to that in NC subjects. This brisk increase of histamine secretion was not attributable to a difference in the number of dermal mast cells or their histamine content. These findings were also supported by a previous study of Cohen and Rosenstreich 7 who measured responses to the intracutaneous injection of another nonspecific mast cell degranulator, codeine sulfate. They found that patients with active CU and those who had a prior history of urticaria exhibited a mean reactivity to codeine that is almost 100 times that of the subjects with asthma, the allergic subjects, and normal subjects, whose increased sensitivity to codeine could not be explained by an increased sensitivity to histamine. We further enlarged on these findings by demonstrating a difference in the magnitude of HR in the skin milieu between a group of patients with CU and an NC population, 8 which could be explained by a difference in the threshold of maximum skin mast cell activation. Altogether, these results, suggesting that the increase in HR on nonspecific stimulation is due to an enhanced skin mast cell releasability, support the concept of an intrinsic defect at the mast cell level itself. However, the persistence of this abnormality over time in symptomatic patients remains to be demonstrated. In contrast, antihistamines are the mainstay of CU management. Antihistamines provide symptomatic re- lief, and their prolonged use may bring about a complete remission. Antihistamines are also known to block histamine at the Hi-receptor site of the target organ, and most antihistamines have minimal effect on the release of mediators by mast cells or basophils.9 Their effect on the secretory response of skin mast cells and their histamine releasability in patients with CU is unknown. The current study was therefore designed, first, to verify the profile of compound 48/80-induced skin mast cell HR in patients with CU during a 28-day period. Second, we wanted to analyze the effects of two anti-h1 blockers, hydroxyzine (100 mg/day) and terfenadine (120 mg/day), on the profile of compound 48/80-induced skin HR in patients with CU and NC subjects and to correlate the results with clinical response to treatment. MATERIAL AND METHODS Study groups ~ Twenty-five subjects were studied: (1) 10 healthy normal volunteers, aged 21 to 24 years, and (2) 15 patients with CU, aged 28 to 52 years, who had persistent or recurrent urticaria lasting for at least 6 weeks and displaying minimal dermographism. All patients were symptomatic on admission to the protocol. Patients had not received systemic corticosteroid for at least 1 month, and classic H1 and H2 blockers were stopped for at least 72 hours before the study. Informed consent was obtained from all subjects. Design of the study Patients with CU were assigned to receive either (1) hydroxyzine (100 mg/day), (2) terfenadine (120 mg/day), or (3) placebo for 28 days, according to a randomized, doubleblind, placebo-controlled trial. Patients were blindly assessed by an investigator before starting treatment, after 14 days of treatment, and on termination of the study. The group of patients receiving antihistamine was later matched with a group of normal volunteers, taking blindly, either (1) hydroxyzine (100 mg/day) or (2) terfenadine (120 mg/day). Therapeutic responses were assessed by the patients' scores for itching, hive number, hive size, and erythema. Each patient completed a diary card, which was checked for accuracy by the same clinical observer throughout the trial period. On arising and at bedtime of each treatment day, the patient scored the average severity of itch, number of hives, size of hives, and the intensity of erythema, using a defined scoring system: 0, no itching, no hives, no erythema; 1, mild itching, 1 to 6 hives <1.5 cm, and mild erythema~ 2, moderate itching, 7 to 12 hives > 1.5 cm, and moderate erythema; 3, severe itching, 12 hives plus >2.5 cm, and severe erythema. The mean symptom scores were reported for each patient. Before initiation of therapy and after a period of 28 days of treatment, the profile of HR by skin mast cells on challenge with compound 48/80 was done by the skin chamber

3 VOLUME 86 H 1 antihistamines in chronic urticaria 789 TABLE I. Characteristics and distribution of study groups No. of studied subjects* Mean age Sex Total Group (yr) (M/F) Placebo Hydroxyzine Terfenadine (N) CU 42 6/9 5 (2) 5 (0) 5 (0) 15 (2) NC 22 4/ (0) 5 (0) 10 (0) *The number in parentheses represents the number of patients who withdrew from the study. technique. The last dose of study medication was taken 90 minutes before the skin window challenge. Skin chamber technique Skin chamber studies were performed as previously reported. 6 Briefly, skin blisters were induced on each forearm by combination of heat and suction. A collection chamber was appended directly over the denuded blister base. After irrigation of the site, each chamber was filled with either PBS (control chamber) or with 2.4 mg/ml of compound 48/80 in PBS for 30 minutes. The skin chamber fluids were replaced by PBS at 1.2.3, and 4 hours after the beginning of the challenge. The fluids then harvested were stored at 70 ~ C until they were used. Histamine assay The quantitation of the amounts of histamine in skin chamber fluids was determined by a single radioenzymatic assay, as previously reported by us. 6 This assay is not affected by compound 48/80. Statistical analysis of the data The HR was calculated in each subject by subtracting the histamine concentration at the PB S-challenged site from that measured at the compound 48/80-challenged site. Results are expressed in mean SEM, and comparative findings on each subject were analyzed by paired Student's t test. RESULTS The characteristics and distribution of subjects under study are displayed in Table I. Fifteen patients with CU were assigned to receive one of the three drug regimens (hydroxyzine, terfenadine, or placebo). From this number, two patients, who were both receiving placebo, withdrew from the study because of treatment failure. The demographic variables (age, race, sex, and duration of disease condition) were comparable among these three groups (not presented). Ten normal volunteers were also assigned to receive, blindly, either hydroxyzine or terfenadine. Symptom scores of patients with CU obtained before and after 28 days of treatment are presented in Table II. They represent the average of the 14 days before the two clinic visits. Over- TABLE II. Total symptom scores of patients with CU before (day 0) and after (day 28) treatment with hydroxyzine, terfenadine, or placebo Day Placebo Hydroxyzine Terfenadine i all, the average of daily symptom scores recorded by each patient matched the average obtained during the medical visit by the investigator. At baseline, no significant difference of symptom scores was observed between the three treatment groups (placebo, , hydroxyzine, ; and terfenadine, ). Significant subsequent change of symptom scores was observed only in patients receiving hydroxyzine ( ), as opposed to patients receiving terfenadine ( ) and placebo ( ). Profiles of HR by nonlesional skin mast cells either spontaneously or on challenge with compound 48 / 80 (2.4 mg/ml) were obtained in patients with CU and NC subjects before and after 4 weeks of treatment. First, there was no statistical difference in the spontaneous release of histamine (PBS sites) by skin mast cells of patients with CU and NC subjects (Table III). Treatment with hydroxyzine or terfenadine did not alter the spontaneous release of histamine in the two study groups. This study confirms also a previous observation demonstrating higher compound 48/80- driven release of histamine in untreated patients with CU than in NC subjects (Fig. 1). Indeed, the amounts of histamine released during the first hour were elevated and then decreased during the last 3 hours. For patients receiving placebo, the profile of HR at 28 days was comparable at each point of time to the profile obtained at the beginning of the study. This observation is in keeping with the absence of clinical response and suggests that the increased mast cell

4 DAY 790 Brunet et ah J. ALLERGY CLIN. IMMUNOL. =] NOVEMBER (2.2 _ 0.8 versus ng/ml). However, the amounts of HR in NC subjects significantly decreased for the first 2 hours after compound 48 / 80 challenge. The change in histamine concentrations was about four to five times lower at 30 minutes ( versus 3.8 _ 1.1 ng/ml), 1 hour (1.2 _ 0.3 versus 5.0 _ 1.3 ng/ml), and 2 hours (0.4 _ 0.1 versus ng/ml), compared with baseline TIME (hr) FIG. 1. Comparison of HR after stimulation with compound 48/80 before (day O) and after (day 28) placebo administration in patients with CU. releasability is reproducible and stable during that period of time. The effects of hydroxyzine treatment on the profile of HR induced by compound 48/80 was assessed in both study groups (Fig. 2). In patients with CU (Fig. 2, A), more histamine was recovered in compound 48/80-challenged sites after 28 days of treatment with hydroxyzine compared with baseline values (day 0). The amounts of HR were about three times greater on day 28 than on day 0 at 30 minutes ( versus ng/ml, p < 0.05), 1 hour ( versus ng/ml, p < 0.05), 2 hours ( versus 2.9 _ 0.9 ng/ml, p < 0.05), and 3 hours ( versus ng/ml, p < 0.01). These significant increases of HR occurred while patients were all clinically improved. In contrast, in NC subjects treated with hydroxyzine (Fig. 2, B), less histamine was measured in compound 48/80-challenged sites at 30 minutes (2.1 _ 0.7 versus ng/ml, p < 0.05) and 1 hour ( versus ng/ml, p < 0.05) compared with baseline. In contrast to what was observed with hydroxyzine, terfenadine treatment did not modify the profile of HR induced by compound 48/80 in patients with CU but did modify in NC subjects. (Fig. 3). Indeed, in patients with CU, no significant difference was observed that concerns the release of histamine before and after terfenadine treatment in compound 48/80-challenge sites at 30 minutes ( versus 3.1 -,- 1.8 ng/ml), 1 hour ( versus 6.2 -,- 1.7 ng/ml), 2 hours ( versus 3.3 _ 1.0 ng/ml), 3 hours ( versus ng/ml), and 4 hours DISCUSSION A whole body of evidence supports the role of histamine in the pathophysiology of CU. ls Our group has also confirmed this finding 6 and demonstrated that there is an increased histamine releasability by skin mast cells of patients with CU. Indeed, we found, in a 4-hour time-course experiment, an increase in both spontaneous skin release of histamine and release induced by a nonspecific mast cell degranulator, compound 48/80. We have observed that this increased releasability in patients with CU was not due to mast cell accumulation in normal-appearing skin. Indeed, we have demonstrated that the greater capacity of the skin of patients with CU to release histamine does not reflect a large number of mast cells or higher skin histamine content in areas free of urticarial lesions. In contrast, Natbony et al.lo found a tenfold increase in mast cell numbers in urticarial lesional skin as compared with the skin of normal volunteers. Consequently, it could be hypothesized that this enhanced releasability observed in individuals with CU was attributable to a failure of the feedback mechanism by which histamine released in the skin milieu normally blocks further HR through its action on putative H2 receptor on cell surface, as reported for basophils. 9 Indeed, Ting et a1.11 have reported that the addition of exogenous histamine to the skin blister fluid inhibited the subsequent antigen-mediated HR by skin mast cells of ragweed-sensitive subjects. It could be postulated further that the enhanced mast cell releasability in CU could also be due to the production of larger amounts of one or more newly described cytokines synthesized by activated T-lymphocytes at the skin lesion level, called histamine-releasing factor, and known to enhance the histamine secretion of skin mast cells. This point remains to be demonstrated in CU disease. More recently, we enlarged on our results by demonstrating that the difference in the magnitude of HR in the skin milieu between an NC group and a group with CU was observed only when skin mast cells were activated with a critical amount of some mast cell activator. 8 Even though these findings were reproducible from patient to patient, the question of reproducibility over time in the same patients remained unanswered.

5 VOLUME 86 H1 antihistamines in chronic urticaria 791 i+ 15: 10 5 A. DAY o---- DAY 28 I s 2) 15' Ii 84 I0' = DAY o---- DAY 28 TIME (hr) i TIME 0at) FIG. 2. Profile of HR before (day 0) and after (day 28) hydroxyzine treatment. A, In patients with CU. B, In NC subjects. 251 A o---. DAY0 28 ~" B 9 DAY o----- DAY ~ ~ 15 I0 i ~ I0" = 5 = s (b.r) i i l TIME Oar) FIG. 3. Profile of HR before (day 0) and after (day 28) terfenadine treatment. A, In patients with CU. B, In NC subjects. TABLE III. Spontaneous release of histamine (nanograms per milliliter SEM) before and after treatment with hydroxyzine or terfenadine in patients with CU and in NC subjects (p = NS) CU NC Hydroxyzine Terfenadine Hydroxyzine Terfenadine Time (hr) Day 0 Day 28 Day 0 Day 28 Day 0 Day 28 Day 0 Day _ _ _ _ _ 0, Therefore, the current study included subjects with CU receiving placebo to allow the assessment of this point by use of the same skin window model with a 4-hour time-course experiment. Indeed, the levels of HR in the skin chamber fluids after stimulation with compound 48/80 were comparable at each point of time when levels were measured at days 0 and 28. Consequently, these data support more strongly the

6 792 Brunet et al. J. ALLERGY CLIN, IMMUNOL, NOVEMBER 1990 finding that the increased histamine releasability by skin mast cell is a persistent pathophysiologic mechanism in CU. Since antihistamines are known to provide symptomatic relief in patients with CU and sometimes full remission after sustained treatment, it would be interesting to verify whether or not antihistamines influence the skin mast cell abnormality that we have described. In our group with CU, terfenadine had little or no influence on the skin lesions or on the profile of HR, suggesting that this drug, at the dose regimen used, did not affect skin mast cells in vivo and is a poor H~ blocker at the skin level in this group of patients. In contrast, more histamine was recovered in the skin blister fluids in patients with CU while they were receiving hydroxyzine than before this drug treatment. Other investigators reported a significant reduction of histamine release after hydroxyzine therapy. Indeed, Ting et al. la have demonstrated a significant reduction of skin blister fluid histamine levels after a 30-minute challenge with ragweed in a group of seven ragweed-sensitive patients after 3 days of hydroxyzine therapy. This inhibitory effect of hydroxyzine was not observed in our model, but a nonspecific degranulator, not an antigen, was used, since CU is not an IgE-mediated process, as opposed to ragweed-mediated allergy. Finally, our patients were treated for a period of 28 days: and not for 3 days. In addition, cetirizine, which is an active metabolic product of hydroxyzine, has been demonstrated to inhibit compound 48/80 or histaminednduced wheal-andflare reaction in both CU and dermographism. ~2 These increased histamine levels in CU after hydroxyzine treatment could be explained by an effective blockade of peripheral H] receptors, making more histamine available for measuring. This is in keeping with the good clinical response to hydroxyzine as compared to terfenadine when the amounts of histamine were similar at days 0 and 28. Moreover, it can be stated that effective anti-h~ blockers elicit symptomatic relief in CU but do not appear to correct the basic mast cell abnormality, at least in the early phase of the treatment. However, in NC subjects, treatment with these two anti-h, antihistamine drugs inhibits the release of histamine in the skin chamber fluids, suggesting that, when skin mast cell releasability is not increased, as it is in the case of CU, both anti-h~ antihistamines have an antirelease activity on normal skin mast cells. The difference of compound 48 / 80-induced histamine release in patients with CU and NC subjects on various regimen (placebo, hydroxyzine, and terfenadine) is not explained by difference of the spontaneous release of histamine, since the amounts of histamine released at PBS sites are comparable for all groups under study. CU evolves in a wave of active symptoms and a period of remission, and it is generally accepted that treatmerit with anti-h~ medications brings most patients into remission for a period of time. Therefore, it is difficult to use a crossover-design study for assessing the efficacity of drug regimen in the treatment of CU because there is no guaranty that the disease will return to the initial level of activity after a washout period. This finding is the reason we chose to include a placebo-treated group as control. Our human in vivo observations parallel the previous results of Mota and da Silva 13 who reported some inhibition by antihistamine drugs of the compound 48 / 80 and the antigeninduced HR from guinea pig and rat mast cells. Identical findings were reported by Lichtenstein and Gillespie 9 who demonstrated that some antilh] antihistamine drugs, in appropriate concentrations, inhibit the in vitro antigen-induced HR from sensitized human leukocytes. The mechanism of this inhibition was then unknown but was not due to partial agonist activity because anti-h1 drugs caused a significant drop in the intracellular cyclic adenosine-3'-5'- monophosphate levels In summaur these results suggest that anti-h~antihistamine drugs do not term correct, in the short, the enhanced skin mast cell releasability of patients with CU in spite of a distinct clinical improvement. Consequently, it could be interesting to verify whether patients with CU in complete remission and not receiving treatment continue to display this skin mast cell hyperreactivity, making CU an intrinsic skin defect expressed only under a variety of circumstances. Two other points are worth mentioning. First, it is evident that anti-hi compounds have different effects on the skin of normal subjects than on the skin of patients with urticaria. Therefore, caution should be used in applying data regarding anti-h, drugs tested on control populations alone. Second, our study demonstrated that the greater the amount of histamine recovered under therapy, the more significant the clinical improvement. This finding might constitute a criterion for evaluating the effectiveness of a drug as an H~ blocker at the skin level. REFERENCES 1. Lewis T. Blood vessels of the human skin and their response. London: Shaw and Sons, Juhlin L. Localization and content of histamine in normal and diseased skin. Acta Derm Venereol (Stockh) 1967: Kaplan AP, Horakova Z, Katz SI. Assessment of tissue fluid histamine levels in patients with urticaria. J ALLERGY CLIN tmmonol 1978;61: Phanuphak P, Schocket AL, Arroyave CM, Kohler PF. 5kin histamine in chronic urticaria. J ALLERGY CLIN IMMUNOL 1980; 65:371-5.

7 VOLUME Atkins PC, Zweiman B. Mediator release in local heat urticaria. J ALLERGY CLIN IMMUNOL 1981;68: BGdard PM, Brunet C, Pelletier G, HGbert J. Increased compound 48/80 induced local histamine release from nonlesional skin of patients with chronic urticaria. J ALLERGY CLIN IM- MUNOL 1986;78: Cohen RW, Rosenstreich DL. Discrimination between urticariaprone and other allergic patients by intradermal skin testing with codeine. J ALLERGY CLIN IMMUNOL 1986;77: Brunet C, BGdard PM, HGbert J. Analysis of compound 48/80-induced skin histamine release and leukotriene production in chronic urticaria. J ALLERGY CLIN IMMUNOL 1988;82: Lichtenstein LM, Gillespie E. The effect of the Hi and H~ antihistamines on "allergic" histamine release and its inhibition by histamine. J Pharmacol Exp Ther 1975;192: H~ antihistamines in chronic urticaria 10. Natbony SF, Phillips ME, Elias JM, Godfrey HP, Kaplan AP. Histologic studies of chronic idiopathic urticaria. J ALLERGY CLIN IMMUNOL 1983;71: Ting S, Zweiman B, Lavker R, Dunsky EH. Histamine suppression of in vivo eosinophil accumulation and histamine release in human allergic reactions. J ALLERGY CLIN IMMUNOL 1981;68: Juhlin L, de Vos C, Rihoux JP. Inhibiting effect of cetirizine on histamine-induced and 48/80-induced wheals and flares, experimental dermographism, and cold-induced urticaria. J AL- LERGY CLIN IMMUNOL 1987;80: Mota I, da Silva WD. The antianaphylactic and histaminereleasing properties of the antihistamines: their effect on the mast cells. Br J Pharmacol 1960;15: Increased bronchial hyperresponsiveness after inhaling salbutamol during 1 year is not caused by subsensitization to salbutamol C. P. van Schayck, MSc,* S. J. Graafsma, PhD,** M.B. Visch, MSc,* E. Dompeling, MSc,* C. van Weel,* and C. L. A. van Herwaarden*** Nijmegen. The Netherlands Recently, it was suggested that long-term administration of an inhaled fl2-agonist might increase bronchial hyperresponsiveness (BHR) to histamine, possibly as a consequence of subsensitization to the inhaled fl2-agonist. To test this hypothesis, we studied two groups of patients with asthma or with chronic obstructive pulmonary disease. An experimental group of 15 patients, inhaling 400 lzg of salbutamol four times daily during 1 year and subsequently, 40 Ixg of ipratropium bromide four times daily for 6 months, and a control group, consisting of 22 patients with the opposite treatment regimen. The BHR, the response in FEVI to cumulative doses of salbutamol, and the number of [3e-adrenoceptors and antagonist affinity of these receptors on circulating lymphocytes were assessed at the start of the study and at 6-month intervals for 11/2 years. The BHR increased significantly (p = 0.001) during the year salbutamol was inhaled and returned to about the value at the start of the study after inhaling ipratropium bromide for 6 months. No change occurred in the bronchodilating responses to cumulative doses of salbutamol, nor was any change observed in the number and the affinity of fl2-adrenoceptors on lymphocytes. It was concluded that long-term use of salbutamol caused a small but significant increase in BHR. The increase in BHR was not caused by subsensitization of fl2-adrenoceptors to salbutamol. ( J ALLERGY CLIN IMMUNOL 1990;86: ) From the *Department of Family Medicine, **Department of Internal Medicine, and ***Department of Pulmonary Diseases, Nijmegen University, Nijmegen, The Netherlands. Supported by Dutch Asthma Foundation and Boehringer Ingelheim, Alkmaar, The Netherlands. Received for publication Feb. 21, Revised June 14, Accepted for publication June 28, Reprint requests: C. P. van Schayck, MSc, Department of Family Medicine, Nijmegen University, PO Box 9101, 6500 HB Nijmegen, The Netherlands. 1 / 1/23571 Selective 13z-adrenergic stimulants, used by inhalation, play an important role in the treatment of patients with asthma or COPD. The swift bronchodilating response and the generally mild side effects make these inhalants useful in the day-by-day control of symptoms of bronchial obstruction. Since [3zadrenergic drugs do not have an effect on late-phase asthmatic reactions and on allergen-provoked increase in BHR, it is generally assumed that thes~ drugs have no long-term effect on nonspecific BHR. Recently, 793