Beta-Agonists- Friends Or Foes?

Eur Respir J LETTER TO THE EDITOR 1992,5,894-895 Beta-agonists- friends or foes?

We would like to respond to some comments about a higher dose of a ~-agonist with higher intrinsic fenoterol in the editorial by C.G. LOfdabl and N. activity. In this situation, epidemiological studies can Svedmyr, "Beta-agonists-friends or foes?" [1]. produce artificial, non-causal associations. None of the The increase in asthma deaths in New Zealand New Zealand Case Control Studies solved this prob- began in 1975. It peaked in 1979 and, for unknown !em. While, as LofdabJ and Svedmyr point out, the reasons, declined to near baseline by 1987. The "epi- most recent New Zealand analysis [8] removed demic" of asthma deaths was temporarily related to a control selection bias, the unbiased control group re- large per capita increase in the consumption of both vealed odds ratios not only for inhaled fenoterol, but inhaled salbutamol and fenoterol, and to the wide- also for oral salbutamol, nebulized salbutamol, theo- spread introduction of regular ~-agonist therapy. As phylline and oral steroids. The odds ratio for the editorial points out, asthma mortality fell, while the salbutamol MDI is analytically linked to fenoterol, ap- consumption of fenoterol and salbutamol further proximately the inverse. The design of the Saskatch- increased. Simultaneously, parallel increases in ewan Asthma Epidemiology Project permitted the odds beclomethasone use occurred (fig. 1). The argument ratios for fenoterol and salbutamol MDis to be uncou- that fenoterol caused the most recent New Zealand pled. This showed use related increase in death rate asthma mortality epidemic was based on selective pres- not only for fenoterol but also for salbutamol MDI, as entation of data [2]. Fenoterol was never sold over well as elevated odds ratios for nebulized and oral the counter in New Zealand. salbutamol. In the United States, where fenoterol has never been Fenoterol has a higher intrinsic activity than marketed, asthma mqrtality is rising. The asthma salbutamol at the ~2 - as well as the ~ 1 -adrenoceptor. death rate in Australia, a country with a low fenoterol Salbutamol is a partial agonist at both receptor consumption, has also been increasing. Over the coun- subtypes. According to binding studies (Birke F, ter sale of P-agonists is common in Australia. Biochemical characterization of the new ~-mimetic In Australia, analysis of the Victorian Mortality SOM 1122 MS in comparison with fenoterol and Study [3] data showed that at most 16 of the 163 salbutamol: ~/~2 -adrenergic receptor affinities and asthma fatalities had ever used fenoterol during the selectivity, Boehringer Ingelheim KG 1989), there is study period. Many of these patients had also received no relevant difference between fenoterol and other ~-agonists. This study did not demonstrate salbutamol in terms of P-selectivity. The pharmaco- an overrepresentation of fenoterol in fatal asthma logical relevance in humans has been studied and re- cases. In Sweden, of the 110 asthma deaths between suits are expected soon. 1986 and 1990, only 3% used fenoterol [4], not more Drugs with higher intrinsic activity, like than expected with a market share between 1.7 and fenoterol, have steeper dose response curves and have 3.4%. the potential for greater bronchodilatory and There is now considerable evidence from The extrapulmonary effects (Kazim F, Statistical R_eport Netherlands, Germany and New Zealand that fenoterol on: Cumulative Dose Response Curves to Nebulized was selectively prescribed to higher risk asth- Solutions of Fenoterol and Salbutamol in Stable matics [5-7]. This is not surprising, since the Asthmatics; Investigators Newhouse M and Dolo- fenoterol metered dose inhaler (MDI) 200 J.t.g delivered vich M, Data on File, Boehringer Ingelheim). 18or--_.:..._Aerosol puffs______in millions Death/100,000__;_---,4 .5

1~ 4~ 140 3.5

1~ 3~

100 2~ 00 M

60 1~ 40 . . 1.0

~ M 0 ~---=-~~!-=--1~~~~~~~~~~~~--"!':~:t--"';~ 0.0 80 81 82 83 84 85 86 87 Year Fig. 1. - J3-agonist inhalers in New Zealand (1970-1987): aerosol puffs vs mortality rate (age 5-34 yrs).liZZi!l: salbutamol; t::=l : fenoterol; -X- : monality rate. BETA-AGONIST$- FRIENDS OR FOES? 895

Por high and very high doses, a clinically relevant especially the Saskatchewan Asthma Epidemiology assessment of beneficial pulmonary versus unsought Project, [1 3] it appears advisable to avoid chronic extrapulmonary effects can only be made under emer­ overuse of ~ -a gooists. In addition to product infor­ gency room conditions. Two major emergency room mation changes reflecting modern understanding of studies comparing fenoterol and salbutamol are in asthma therapy and recommending appropriate use of progress. Since by far the majority of asthmatics die anti-inflammatory therapy, Boehringer lngelheim has - with evidence of suffocation, not with signs of cardiac in addition to the already lower dose fenoterol/ toxicity such as rhythm disturbances, higher intrinsic ipratropium combinations - extended the availability of activity could translate into additional therapeutic ben­ a lower dose fenoterol MDI (Berotec®lOO). efit in the severe acute attack. The majority of the studies on extrapulmonary H.W. Staudinger and J.F. Haas effects of fenoterol cited in the editorial have been Boehringer lngelheim GmbH performed by one group in New Zealand and suffer Abteilung Medizin from a variety of methodological difficulties. For ex­ D-6507 Ingelheim am Rhein ample, based on cumulative dose response data, BuR­ Germany GESs et al. [9] claim that the tachycardia produced by fenoterol is greater than that of isoprenaline. The References study design ignored the much longer duration of action of fenoterol compared to isoprena1ine. Such a l. Lofdahl CG, Svedmyr N. - Beta-agonists - friends design necessarily creates a bias unfavourable for any or foes? Eur Respir J, 1991; 4: 1161-1165. longer acting drug than isoprenaline, in d1is case 2. Crane J, F1att A, Jackson R, Ball M, Pearce N, fenoterol. Combined with the use of endpoints reflect­ Burgess C, Kwong T, Beasley R. - Prescribed fcnotero1 and death from asthma in New Zealand, 1981- 1983: ing both ~ .- and ~ 2 -effects, like tachycardia, the study design does not allow valid conclusions about case-control study. Lancet, 1989; i: 917-922, DN: P89-0833. ~-selectivity. 3. Robertson CF, Rubinfeld AR, Bowes G. - Deaths Findings consistent with some deterioration of rrom asthma in Victoria: a 12-month survey. Med J Aust, asthma control on regular ~-agonists have been 1990; 152 (10): 511-517. reported for fenoterol, salbutamol and terbutaline. The 4. Jackeviciute I, Horte I, Saldeen T. - Selective observations are not consistent and the issue has to be ~2 -slimulators and deaths in asthma. Annual Meeting of the regarded as still open. Swedish Medical Association. 1991; p. 321. The crossover study performed by SEARS et al. [10] 5. Garrett JE, Turner P. - The severity of asthma in was the most comprehensive study suggesting inferior relation to ~-agonist prescribing. NZ Med J, 1991; 104: asthma control with regular ~-agonist treatment. 3~0. 6. Rea HH, Mulder J, Fenwick J, Garrett J, Kolbe J. - Virtually all previous studies with other ~-agonists , in­ Prescribing for asthma and severity markers. NZ Med J, cluding long-acting drugs, have used parallel group 1990; !03: 491. designs. The Sear's study administered quantities of 7. Wilson JD. - Selective prescribing of fenoterol and fenoterol in excess of that recommended as initial salbutamol in New Zealand in general practice. Post treatment for mild to moderate asthmatics in the com­ Marketing Surveillance, 1991; 5: 105-118. pany's basic product information. The question of 8. Grainger J, Woodman K, Pearcc N, Crane J, Burgess whether similar findings would have occurred with C, Keane A, Bcasley R. - Prescribed fenoterol and death customary doses remains open. from asthma in New Zealand, 1981-1987, a further case­ In pointing out that TR.EMBA TH et al. [ 11] showed a control study. Thorax, 1991; 46: I 05-11 I. reduction in airway calibre for fenoterol but not for 9. Burgess CD, Windom H, Pearcc N et al. - Lack of terbutaline (used at about half the equivalent dose), the ~ 2 -receptor selectivity of fenoterol compared to orci­ prenaline and isoprenaline. Aust NZ J Med, 1990; 20 editorial seems to suggest that a better result with (Suppl.): 525. p.r.n. compared to regular therapy is specific to 10. Sears MR, Taylor DR, Print CG, Lake DC, Quingqing fenoterol. In fact, BESWICK et al. [12] showed a LI, Flanncry EM, Yates DM, Lucas MK, Herrison GP. - greater increase in morning peak expiratory flow (from Regular inhaled beta-agonist treatment in bronchial asthma. a baseline of 320 to 352 /·min·1) after 10-12 months Lancet, 1990; 336: 139!- 1396. of on demand treatment with salbutamol compared to 11. Trembath PW, Greenacre JK, Anderson M, Dimmock a minimal increase (from 293 to 301 /·min·•) with reg­ S, Mansfield L, Wadsworth J, Green M. - Comparison ular treatment. Increases in bronchial hyperreactivity of 4 weeks treatment with fenoterol and tcrbutaline aero­ have occasionally been observed with salbutamo1, sols in adult asthmatics. A double-blind cross-over study. terbutaline and fenoterol. J Allergy Clin lmmunol. 1979; 63: 395-400. 12. Bcswick KB, Pover GM. Sampson S. - Long-term In conclusion, the epidemiological evidence does not regularly inhaled salbutarnol. Curr Med Res Opin, 1986; support a unique association between fenoterol and 10: 228-234. asthma mortality. Neither does the pharmacological or 13. Spitzer WO, Suissa S, Emst P, Horwitz Rl, Habbick clinical evidence suggest that fenoterol is uniquely B, Cockroft D, Boivim JF, McNutt M, Buist S, Rcbuck AS. different from other ~-agonists. - The Use of ~-Agonists and the Risk of Death and Near­ Nonetheless, based on the findings available, Death from Asthma. N Engl J Med, 1992; 326: 501-506.