Higher incidence of discontinuation of angiotensin converting enzyme inhibitors due to cough in black subjects

PHARMACOEPIDEMIOLOGY AND DRUG UTILIZATION

Higher incidence of discontinuation of angiotensin converting enzyme inhibitors due to cough in black subjects

Objective: To compare the rates of discontinuation of angiotensin converting enzyme (ACE) inhibitors in patients with different racial and ethnic backgrounds. Methods: A registry from a tertiary hypertension clinic consisting of 892 patients who received their first-ever dose ofACE inhibitor therapy was examined. Surveillance for cough was prospective, systematic, and constant beginning in 1986 and routinely included a trial of sinusitis therapy, followed by withdrawal and rechallenge before discontinuation of drug. Results: The prevalence (per 100 patients) of cough requiring discontinuation of ACE inhibitor therapy was 62 of 644 (9.6 per 100) patients among black subjects compared with six of 248 (2.4 per 100) patients among others (odds ratio, 4.0; 95% confidence interval, 1.7 to 9.1; p < 0.001). There were no significant differences in discontinuation rates across the three most commonly used ACE inhibitors: captopril(6.6%; all black subjects), enalapril(6.1%; 94% black subjects), and lisinopril(7.3%; 90% black subjects). Cough was more common among women (70% of subjects). After adjustment (by backward stepwise multiple logistic regression analysis) for baseline differences, black subjects had a relative risk of 2.58 (95% confidence interval, 1.21 to 4.65; p = 0.01) of discontinuation of ACE inhibitor due to cough. Conclusions: These data suggest that there may be a race- or ethnicity-related difference in the prevalence of cough attributed to ACE inhibitor therapy. Although a race-related difference in ACE gene polymorphism has been suggested, further work is necessary to define the biological reason and pathophysiology for such a difference. (Clin Pharmacol Ther 1996;60:582-8.)

William J. Elliott, MD, PhD Chicago, Ill.

There has been increasing interest and concern in recent years about race- and ethnic&related differences in adverse event and efficacy rates for different classes of medications. One of the most widely touted differences in efficacy of antihypertensive medications is the alleged disparity of blood

From the Department of Preventive Medicine, Rush- Presbyterian-St. Luke’s Medical Center.

Received for publication April 10, 1996; accepted July 10, 1996. Reprint requests: William J. Elliott, MD, PhD, Department of

Preventive Medicine, Rush-Presbyterian-St. Luke’s Medical Center, 1725 West Harrison St., Suite 117, Chicago, IL 60612- 3824.

Copyright 0 1996 by Mosby-Year Book, Inc. 0009-9236/96/$5.00 + 0 13/l/76394

pressure-lowering effects of angiotensin converting enzyme (ACE) inhibitors, which some have de- scribed as less effective in black subjects and black Americans than in other ethnic groups (e.g., white subjects).r Critics of this concept point to differences in sodium intake and balance, intravascular volume differences, and other baseline differences between races to explain the differential hypotensive efficacy of these drugs.

Recently, Brown and Nadeau’ have suggested that race or ethnicity may play a role in differential rates of angioedema with ACE inhibitors, with the risk of this potentially life-threatening complication being higher in black subjects. One of the reasons a recently approved ACE inhibitor is allegedly not

582

Elliott 583

being marketed is because of a high risk of angioedema seen particularly more commonly in black Americans in clinical efficacy and safety studies (A.A. Taylor, Jr., personal communication, 1995); these data have apparently not yet been published in the medical literature. However, few data exist about potential race- or ethnicity-related differences3-6 in a much more common (although perhaps less serious) adverse effect of ACE inhibitors: cough.7-’ ’ Previous work has established two subgroups at higher risk for ACE inhibitor- associated cough: women 12-20 or nonsmok- ers. ‘23143’6321 We therefore gathered data to see if there is a difference across racial and ethnic groups with regard to the prevalence of ACE inhibitor- associated cough.

METHODS The records of 2829 consecutive patients seen at a

tertiary hypertension clinic were reviewed to identify patients who had received their first exposure to ACE inhibitors as part of their care for hypertension. There was (since the first locally recognized case in 1985) a special interest in the prevalence and cause of ACE inhibitor-associated cough among the physicians in this clinic.

A standard protocol for assessment of cough potentially due to ACE inhibitors was commonly followed, which included the following: inquiring of both the patient and significant other about the presence of cough at the first visit (typically 2 to 4 weeks) after starting an ACE inhibitor (“Have you noticed any problem with cough since your last visit?“); routine treatment of presumed sinusitis (with an antihistamine and antibiotic for at least 2 weeks) whenever coughing was reported as present; discontinuation (for 2 to 4 weeks) of ACE inhibitor therapy if the cough did not improve after anti-sinusitis treatment; and rechallenge if the cough disappeared after the ACE inhibitor was discontinued. In only a seven subjects (in whom it was thought to be particularly difficult to ascribe causality) was an “n of 1” trial done with placebo and blinded medications. This typically involved a double-blind dispensing of six identical bottles (each containing a week’s supply of blue opaque No. 00 capsules): three bottles that contained sucrose and three bottles that contained tablets of ACE inhibitor and sucrose. The patient was instructed not to open the capsules. On return to the clinic, the patient was asked to indicate which bottles (lettered “A through F”) of

tablets caused cough; ACE inhibitor-associated cough was diagnosed if five of six bottles were correctly identified.

Patient data were excluded from the analysis if (1) subjects had received ACE inhibitor therapy previously (e.g., from another physician); or (2) angioedema developed that was attributed to the ACE inhibitor (eight cases: five in black subjects and three in others, which was recognized in all subjects before the 2- to 4-week interval between first administration of ACE inhibitor and first return visit, when standard query about cough was performed).

We report here discontinuation rates for ACE inhibitor therapy that were attributed to cough. There were four black patients in the clinic who had congestive heart failure or renal impairment who elected to continue with their ACE inhibitor therapy, despite the presence of an ACE inhibitor- associated cough; their exclusion makes the following analysis more conservative (i.e., biases toward the null).

The assignment of patients was done according to the patient’s own identification of his or her racial or ethnic group. “Black’ is used as a descriptor in this context (rather than “black American”) because some of the patients were not American. Patients in the “other” racial or ethnic category included: 218 white subjects, three Asians, two Native Americans, and one each of the following: Arabic, Cuban, His- panic, Mexican, and Puerto Rican.

Prevalence rates were calculated per 100 patients exposed. Statistical comparisons were performed with use of Statistics/Mac 4.1 (StatSoft, Inc., Tulsa, Okla.), with x2 tests (for categorical variables) or t tests (for continuous variables). For multivariate comparisons, backward stepwise multiple logistic regression analysis was performed. For this calcula- tion, the date of initial administration of ACE inhibitor was stratified by year, which resulted in only seven possible covariate entries, compared with 2557 (if individual dates had been used). Statistical significance was accepted for p 5 0.05.

RESULTS The baseline characteristics of the 892 patients in

the clinic who had never previously received ACE inhibitor therapy are shown in Table I, grouped according to whether or not their ACE inhibitor therapy was discontinued due to the appearance of cough. Univariate characteristics of those who discontinued treatment included the following: black, women, older, or given a prescription for an ACE

584 Elliott CLINICAL P HABMACOLOGY & THERAPEUTICS

NOVEMBER 1996

Black

Other 6 of 248 I I

P < 0.001 I I

Disc~ntinuatiof Rate due:o Cough Fer 100 Paients Trelt%

Fig. 1. Comparison of prevalence of discontinuation rates for angiotensin converting enzyme inhibitor therapy for black and other patients.

Table I. Baseline characteristics of patients studied

Characteristic Cough No cough p Value

No. of patients Age (~4 Gender (% male) Race (% black) Cigarette use (%) Initial SBP (mm Hg) Initial DBP (mm Hg) Diuretic treated (%) Date of first ACE

inhibitor prescription

68 824 58.1 ? 1.4 54.2 t 0.5

28 46 92 71 25 20

181.2 2 3.0 179.4 t 1.0 106.3 + 1.5 104.4 2 0.5

65 64 8124189 L 167 days 6113188 ? 39 days

- 0.03 0.006 0.001 0.37 0.62 0.28 0.98 0.04

SBP, Systolic blood pressure; DBP, diastolic blood pressure; ACE, angiotensin converting enzyme.

inhibitor at a later date. There were no significant differences between those who had ACE inhibitor therapy discontinued due to cough in cigarette use, initial systolic or diastolic blood pressure, or diuretic treatment.

Fig. 1 shows the prevalence of discontinuation of ACE inhibitor therapy due to cough for black subjects and other subjects. The rates were 9.6 per 100 treated black patients, compared with 2.4 per 100 treated patients of other racial and ethnic backgrounds. The odds ratio for this comparison was 3.98, with a 95% confidence interval of 1.74 to 9.08 (p < 0.001).

The backward stepwise multiple logistic regression analysis resulted in a model that included only two statistically significant covariates: race and gen-

der. In this set of calculations, race was always the most powerful risk factor for discontinuation of ACE inhibitor therapy due to cough, with a relative risk varying between 2.11 and 3.31 for black subjects compared with other subjects. The results of the final model (including the last nonsignificant covariate) are shown in Table II.

Fig. 2 shows the discontinuation rates for the three most commonly used ACE inhibitors. The shading and the text within each bar describes the racial or ethnic grouping of those affected. Not shown are the rates for the one black (of nine total) patient in whom cough developed during treatment with benazepril. Also, data are not shown for other ACE inhibitors that did not cause development of a cough that required a patient to quit taking the drug.

(:LIiXlC:AL PHARMACOLOGP icr ‘1’HERAPE~‘TICS ~OI.lIME 60, NL’MRER 5 Elliott 585

There were no significant differences across the three drugs in the prevalence of discontinuing the ACE inhibitor due to cough.

In an attempt to examine whether there was a race- or ethnicity-related difference in overall rates of discontinuation of ACE inhibitors (e.g., for reasons other than cough), the proportion of black subjects in the group who discontinued ACE inhibitor therapy (for any reason, but not including changes within class; e.g., for reasons of compliance or economy) was compared with the group who successfully continued ACE inhibitor therapy. No significant differences were found (p = 0.16): 76.4% of those who discontinued therapy were black; 71.1% of those who remained with ACE inhibitor therapy were black. When the discontinuations due to documented lack of hypotensive efficacy were excluded from this analysis (i.e., retaining those patients who discontinued ACE inhibitor therapy for signs or symptoms attributed to ACE inhibitors), the race- and ethnic&y-related difference was even smaller and less significant (p = 0.41): 74.7% of the patients discontinuing therapy due to signs or symptoms were black and 71.1% of patients who successfully remained with ACE inhibitor therapy were black.

DISCUSSION Cough due to ACE inhibitor therapy is now

widely recognized as a side effect of this important class of drugs, although the prevalence varies widely (0% to 44%)9 and is dependent on the method used to detect cough. l3 Published rates of discontinuation of ACE inhibitors due to this side effect also vary widely (0% to 22%)9,” and are inversely propor- tional to the complexity of protocols used to ascribe causality to the ACE inhibitor.22 This report is apparently the first to suggest a race- or ethnicity- related difference in the prevalence of a clinically relevant consequence; that is, discontinuation of ACE inhibitors due to cough. The observed difference retained statistical significance in multivariate analysis even after adjustment for baseline differences, including gender, age, and year of initial ACE inhibitor treatment.

The reason black subjects may have a higher rate of ACE inhibitor-associated cough (and discontinuation of therapy for this reason) is not yet apparent. Some researchers have pointed to race- or ethnic@related differences in ACE gene polymorphism as a potential explanation for the differences in prevalence of cough due to ACE inhibitors,23 but

Table II. Results of multivariate stepwise backward multiple logistic regression analysis

Risk factor

9.5% Relative Confidence p

risk interval Value

Black race or ethnic@ 2.58 1.21-4.65 0.01 Female gender 1.09 1.003-1.16 0.049 Year of first ACE inhibitor 1.03 0.87-1.15 0.11

prescription (per year)

a significant difference has not been observed in two genetic screening programs designed to demon- strate it.24,25 The race- or ethnic@related difference in cough due to ACE inhibitor may have a similar mechanism as the apparent race- or ethnic&y-related difference in angioedema observed with some of these drugs2 but the underlying pathophysiology of this adverse effect has not been com- pletely elucidated.2h Bradykinin has been implicated in both the cough27,28 and angioedema29330 observed with ACE inhibitor therapy.8,26 Others have suggested that there is a race- or ethnic@-related difference in both circulating levels of, and susceptibility to, bradykinin. African American patients have lower urinary kallikrein levels than whites31 as well as increased sensitivity to intradermally adminis- tered bradykinin, 32 which is at least consistent with (although clearly not sufficient to explain) both types of bradykinin-mediated adverse effects of ACE inhibitors occurring more commonly in black Ameri- cans.

Our overall rate of discontinuation of ACE inhibitor therapy due to cough is actually quite close to previous estimates from the literature using similar methods. The overall rate for the patients reported here is 7.6%; others have reported 4% in 164 patients with hypertension rechallenged with a different ACE inhibitor;33 6.0% (in 36 patients, with a 95% confidence interval of 4.5% to 7.5%),‘” 6.25% (in 80 patients),34 or 7.1% (in 199 patients, although only 1% were withdrawn from therapy for “severe cough”).35 These rates (obtained after withdrawal and rechallenge) are considerably lower than spon- taneously reported rates of cough due to ACE inhibitor, consistent with recent prospective studies, in which patients with a history of ACE-associated cough are challenged, withdrawn, and rechallenged, when the prevalence of cough is only 65% to 75%.36,37

The lack of a significant difference across racial

586 Elliott CLINICAL PHARMA COLOGY &THERAPEUTICS NOVEMBER 1996

Captopril

Enalapril

Lisinopril

% Oaf Patknts &scoSinuir?g &-I Du6e to c7ough”

Fig. 2. Comparison of prevalence of discontinuation rates for the three most commonly prescribed angiotensin converting enzyme inhibitors (ACE-I). The dark shading of each bar corresponds to the percentage of black patients prescribed each ACE inhibitor.

and ethnic groups in discontinuation rates of ACE inhibitors for reasons other than cough is important because it suggests that the predominance of discontinuations due to cough in black patients was a specific finding and not a feature of some broader predilection of black patients to withdraw from therapy with these drugs. There was also no significant difference in rates of angioedema (0.8 discontinuations per 100 treated patients in both black and other subjects), but only eight cases were discovered in all.

The results of the multiple logistic regression analysis should be interpreted cautiously because the number of cases of discontinuation due to cough was only 68. As the number of covariates are increased, or the number of possibilities for each covariate is increased, the reliability and absolute mag- nitude of the relative risk estimate decreases. Thus, if the analysis was performed with use of all 821 actual dates of beginning ACE inhibitor therapy (rather than only the 7 years, as reported in Table II), the relative risk for black subjects decreases to 1.09 (although it continues to be statistically significant at p < 0.01 and the most powerful of all of the univariate predictors).

Unlike several recent reports suggesting that specific ACE inhibitors may be associated with lower risk of cough, 38-41 Fig. 2 shows no significant differ-

ence across the three most commonly used ACE inhibitors in the prevalence of discontinuation due to cough. During the early days of ACE inhibitor prescribing, there were four patients whose initial ACE inhibitor (associated with cough) was changed to a different agent of the same class; no improve- ment in the cough was seen in any case. Our data are consistent with others’ hypotheses that the cough due to ACE inhibitors may be a class effect.42

The conclusions drawn from these data have several limitations. The number of affected patients is small (only 68) so multivariate analyses and confidence intervals are, by necessity, prone to error and mis-estimation. Patients who elected to continue ACE inhibitor therapy despite the presence of a reasonably well-documented cough due to the drug therapy were not included in the analysis, although the estimates given are more conservative because of this exclusion. The attribution of the cough to the ACE inhibitor was not done blindly, nor was there ‘placebo control, except in the seven cases when an “n of 1” trial was done. It was thought that this higher degree of endeavor was unlikely to be accepted by many patients and physicians in the clinic, so the more rigorous method was pursued only in situations in which there was doubt about whether or not the ACE inhibitor was associated with cough (despite anti-sinusitis treatment, washout, and re-

challenge, and the passage of at least 2 months’ time). Lastly, it is possible (and even likely) that there was a “learning curve” in the clinic, such that cough was more likely to be discovered by query (thus launching the sequence to attribute it to the ACE inhibitor) in the later years than early on; this is seen in the univariate analyses (in which patients who later had discontinuations of their ACE inhibitor were seen, on average, about a year later than those who had no discontinuation due to cough) but did not achieve statistical significance in the multivariate analysis. On the other hand, prescribing of ACE inhibitors was more commonly done in the 1990-1992 time period than 19851987, when fewer agents of this class were available.

In summary, these data suggest that there may be a race- or ethnic@-related difference in both the ACE inhibitor associated cough and discontinuations due to this adverse effect. Although the mechanism for this disparity is not known, it may well be related to the differences in bradykinin levels or sensitivity that have been shown in different racial and ethnic groups. Physicians may wish especially to inquire of their black patients about cough related to ACE inhibitor therapy and make appropriate adjustments to their pharmacologic regimen if sub- sequent studies indicate that this problem occurs more often in specific racial or ethnic groups.

I thank my former colleagues, especially Delores Ze- brauskas, RN, CRNA, for assistance in managing the patients reported.

References 1. Materson BJ, Reda DJ, Cushman WC, Massie BM,

Freis ED, Kochar MS, et al. Single-drug therapy for hypertension in men: a comparison of six antihypertensive agents with placebo. N Engl J Med 1993;328: 914-21.

2. Brown NJ, Nadeau JH. Does race predispose to angiotensin-associated angioneurotic edema? [letter]. Ann Intern Med 1993;119:1224.

3. Woo J, Chan TY. A high incidence of cough associated with combination therapy of hypertension with isradipine and lisinopril in Chinese subjects. Br J Clin Pratt 1991;45:178-80.

4. Chan WK, Chan TY, Luk WK, Leung VK, Li TH, Critchley JA. A high incidence of cough in Chinese subjects treated with angiotensin converting enzyme inhibitors [letter]. Eur J Clin Pharmacol 1993;44:299- 300.

5. Woo KS, Nicholls MG. High prevalence of persistent

6.

7.

8.

9.

10.

11.

12.

13.

14.

1.5

16.

17.

18.

19.

20.

Elliott 587

cough with angiotensin converting enzyme inhibitors in Chinese. Br J Clin Pharmacol 1995;40:141-4. Woo KS, Norris RM, Nicholls G. Racial difference in incidence of cough with angiotensin-converting enzyme inhibitors (a tale of two cities). Am J Cardiol 1995;75:967-8. Sesoko S, Kaneko Y. Cough associated with the use of captopril. Arch Intern Med 198.5;145:1524. Karlberg BE. Cough and inhibition of the renin- angiotensin system. J Hypertens 1993;11(suppl 3): S49-52. Fletcher AE, Palmer AJ, Bulpitt CJ. Cough with angiotensin converting enzyme inhibitors: how much of a problem? J Hypertens 1994;12(suppl 2):S43-7. Cough caused by ACE inhibitors. Drug Ther Bull 1994;32:28. Lacourciere Y, Lefebvre J. Modulation of the renin- angiotensin-aldosterone system and cough. Can J Cardiol 1995;11(suppl F):33F-9F. Fuller RW. Cough associated with angiotensin- converting enzyme inhibitors. J Hum Hypertens 1989; 1:159-61. Yeo WW, Ramsay LE. Persistent dry cough with enalapril: incidence depends on method used. J Hum Hypertens 1990;4:517-20. Yeo WW, Foster G, Ramsay LE. Prevalence of persistent cough during long-term enalapril treatment: controlled study versus nifedipine. Q J Med 1991;SO: 763-70. Capella D, Torres F, Avila P, Moreno V, Laporte JR. [Cough caused by angiotensin-converting enzyme inhibitors: a series of cases collected by spontaneous notification of adverse reactions]. Med Clin (Bare) 1991;96:126-8. Fujimori K. [Angiotensin converting enzyme (ACE) inhibitor-induced cough in non-smoking hypertensive patients]. Arerugi 1991;40:1327-33. Kaku T, Yamasaki H, Harada N, Tsujino M, Inoue G. [Dry cough in the elderly patients treated with angiotensin converting enzyme inhibitor]. Nippon Ronen Igakkai Zasshi 1991;28:365-70. Efstratopoulos AD, Meikopoulos M, Voyaki S. Fre- quency of cough during therapy with ACE inhibitors in Greek hypertensives. J Hum Hypertens 1993;7: 607-9. OS I, Bratland B, Dahlof B, Gisholt K, Syvertsen JO, Tretli S. Female preponderance for lisinopril-induced cough in hypertension. Am J Hypertens 1994;7: 1012-5. Visser LE, Stricker BH, van der Velden J, Paes AH, Bakker A. Angiotensin converting enzyme inhibitor associated cough: a population-based case-control study. J Clin Epidemiol 1995;48:851-7.

21. Chalmers D, Whitehead A, Lawson DH. Postmarket- ing surveillance of captopril for hypertension. Br J Clin Pharmacol 1992;34:215-23.

588 Elliott CLINICAL PHARMACOLOGY&THERAPEUTICS NOVEMBER 1996

22. Reisin L, Schneeweiss A. Complete spontaneous re- mission of cough induced by ACE inhibitors during chronic therapy in hypertensive patients. J Hum Hy- pertens 1992;6:333-5.

23. Furuya K, Yamaguchi E, Hirabayashi T, Itoh A, Hizawa N, Ohnuma N, et al. Angiotensin I converting enzyme gene polymorphism and susceptibility to cough [letter]. Lancet 1994;343:354.

24. Chadwick IG, Yeo WW, Higgins KS, Jackson PR, Ramsay LE, Morice AH. ACE inhibitors, cough, and genetics [letter; comment]. Lancet 1994;343:740-1.

25. Kreft-Jais C, Laforest L, Bonnardeaux A, Dumont C, Plouin PF, Jeunemaitre X. ACE inhibitors, cough, and genetics [letter; comment]. Lancet 1994;343:740.

26. Israili ZH, Hall WD. Cough and angioneurotic edema associated with angiotensin-converting enzyme inhibitor therapy: a review of the literature and pathophysiology. Ann Intern Med 1992;117:234-42.

27. Fuller RW, Dixon CM, Cuss FM, Barnes PJ. Bradykinin-induced bronchoconstriction in humans: mode of action. Am Rev Respir Dis 1987;135:176-80.

28. Katsumata U, Sekizawa K, Ujiie Y, Sasaki H, Tak- ishima T. Bradykinin-induced cough reflex markedly increases in patients with cough associated with captopril and enalapril. Tohoku J Exp Med 1991;164:103-9.

29. Schapira M, Silver LD, Scott CF, Schmaier AH, Pro- grais LJ Jr, Curd JG, et al. Prekallikrein activation and high-molecular weight kininogen consumption in hereditary angioedema. N Engl J Med 1983;308: 1050-4.

30. Slater EE, Merrill DD, Guess HA, Roylance PJ, Coo- per WD. Clinical profile of angioedema associated with angiotensin converting-enzyme inhibition. JAMA 1988;260:967-70.

31. Levy SB, Lilley JJ, Frigon RP, Stone RA. Urinary kallikrein and plasma renin activity as determinants of renal blood flow. J Clin Invest 1977;60:129-38.

32. Gainer J, Nadeau J, Ryder D, Brown NJ. Increased sensitivity to bradykinin in African Americans and hypertensives [abstract]. J Invest Med 1995;43:295A.

33. Ravid D, Lishner M, Lang R, Ravid M. Angiotensin- converting enzyme inhibitors and cough: a prospective evaluation in hypertension and in congestive heart failure. J Clin Pharmacol 1994;34:1116-20.

34. Town GI, Hallwright GP, Maling TJ, O’Donnell TV. Angiotensin converting enzyme inhibitors and cough. N Z Med J 1987;100:161-3.

35. Lefebvre J, Poirier L, Lacourciere Y. Prospective trial on captopril-related cough. Ann Pharmacother 1992; 26:161-4.

36. Charlon V, Dollow S, Fidel J, Hoglund C, Honkanen T, Kobrin I, et al. Reproducibility of angiotensin converting enzyme inhibitor induced cough: a double- blind randomised study. Br J Clin Pharmacol1995;39: 125-9.

37. Lacourciere Y, Brunner H, Irwin R, Karlberg BE, Ramsay LE, Snavely DB, et al. Effects of modulators of the renin-angiotensin-aldosterone system on cough: Losartan Cough Study Group. J Hypertens 1994;12:1387-93.

38. Ogihara T, Mikami H, Katahira K, Otsuka A. Com- parative study of the effects of three angiotensin converting enzyme inhibitors on the cough reflex. Am J Hypertens 1991;4(suppl 2):46S51S.

39. Saruta T, Nishikawa K. Characteristics of a new angiotensin converting enzyme inhibitor: delapril. Am J Hypertens 1991;4(suppl 2):23S-8s.

40. Punzi HA. Safety update: focus on cough. Am J Car- diol 1993;72:45H-8H.

41. Sharif MN, Evans BL, Pylypchuk GB. Cough induced by quinapril with resolution after changing to fosino- pril. Ann Pharmacother 1994;28:720-2.

42. Roth A. Chronic cough caused by angiotensin converting enzyme inhibitors. Ann Allergy 1990;64:47-8.

Documents

Higher incidence of discontinuation of angiotensin converting enzyme inhibitors due to cough in black subjects