715
Contraceptive effects of extended lactational amenorrhoea: beyond the Bellagio Consensus
We have recorded the duration of lactational anovulation and amenorrhoea in a well-nourished group of Australian women who breastfed their babies throughout the study. The data enabled us to compare the theoretical cumulative probability of conception among breastfeeding women who had unprotected intercourse irrespective of their menstrual status with that of those who had
unprotected intercourse only during lactational amenorrhoea. Breastfeeding alone is not an effective form of contraception, since all the women in our study resumed normal ovulation while still breastfeeding. However, among women who have unprotected intercourse only during lactational amenorrhoea but adopt other contraceptive measures when they resume menstruation, only 1·7% would have become pregnant during the first 6 months of amenorrhoea, only 7% after 12 months, and only 13% after 24 months. Thus for our women it would be possible to extend the Bellagio Consensus Conference guidelines which stated that lactational amenorrhoea can only be relied on as a contraceptive for the first 6 months post-partum in women who are fully or almost fully breastfeeding. The lactational amenorrhoea method can be relied on for excellent contraceptive protection in the first 6 months of breastfeeding, irrespective of when supplements are introduced into the baby’s diet; for women who continue to breastfeed the method can also give good protection for up to 12 months post partum. Once menstruation has returned, other forms of contraception are essential to prevent pregnancy. Introduction
The Bellagio Consensus Conference on breastfeeding as a family planning method established that a mother who is fully or nearly fully breastfeeding her infant and who remains amenorrhoeic will have more than 98% protection from pregnancy in the first 6 months post partum.1,2 However, the contraceptive effect of longer periods of lactational amenorrhoea could not be quantified. In practice, many women may not be able to use the Bellagio guidelines to their full extent, since they will have started to introduce supplements into the baby’s diet before 6 months. We have completed a large prospective study of the duration of lactational anovulation and amenorrhoea in a well-nourished group of Australian women-members of
the Nursing Mothers’ Association of Australia-who breastfed for a long time.3
Subjects and methods Women were recruited into the study before their babies were 6 weeks old. Ovarian activity was monitored in all women by salivary progesterone determinations, and in a subgroup by means of urinary oestrogen and pregnanediol excretion rates. Saliva samples were collected 2-4 times per week. Twice a week the women also kept a 24 h record of the timing and duration of all breastfeeds and the timing, nature, and amount of all supplementary feeds. Ovulation was presumed to have occurred if there was a rise in salivary progesterone above the baseline established for that individual woman, and/or a urinary pregnanediol excretion rate of 1 mg/24 h or greater. An insufficient luteal phase was defmed as a rise in salivary progesterone above baseline but below 40 pg/ml, or a urinary pregnanediol excretion rate of more than 1 but less than 2 mg/24 h. A short luteal phase was defined as 11 days or less from ovulation to menstruation. Of the 142 women recruited, 41 dropped out because of the highly demanding nature of the experimental protocol. We have data on the time of first post-partum menstruation for 101 women. If anything, our data are therefore biased against women with longer durations of amenorrhoea. Some women did not collect saliva samples regularly, so we have complete hormonal data up to the time of resumption of menstrual cycles in only 89 women. The study was considered to be complete when a woman had had three menstrual cycles post partum. Each woman continued to breastfeed her baby throughout the time she was in the
study.
Results The
duration of lactational amenorrhoea was 9-5 months (in 101 women) and the mean duration of (SD 4-9) anovulation 10 6 (5-0) months (in 89 women). Although some of the women breastfed for as long as 2 years and were amenorrhoeic for over 18 months, the mean age of their infants at the introduction of supplementary feeding was 53 (11) months, and only 24 of the 101women would have been able to benefit by the full 6 months of the Bellagio mean
guidelines. We calculated the theoretical cumulative probability of pregnancy over 24 months for: a hypothetical population of
non-lactating women of normal fertility having unprotected intercourse, assuming a fecundability of 25 % per menstrual cycle;4 our breastfeeding women, if they had had unprotected intercourse for 24 months; and our ADDRESSES: Departments of Anatomy and Physiology, Monash University, Melbourne (Prof R. V. Short, ScD FRS, M. B. Renfree, DSc, G. Shaw, PhD), and Nursing Mothers Association of Australia, Nunawading, Australia (P.R. Lewis, PhD). Correspondence to Prof R. V. Short, Department of Physiology, Monash University, Melbourne, 3168, Australia.
716
By contrast, 50% of our breastfeeding women having unprotected intercourse irrespective of when their menses returned would become pregnant by 12 months post partum. Thus, although breastfeeding alone reduces fertility, it is not a particularly effective or reliable form of contraception. However, as intimated by the Bellagio guidelines, women in lactational amenorrhoea have a pronounced reduction in fertility. For the breastfeeding women who used other contraceptive measures when menses resumed, only 1-7% would have become pregnant by 6 months, 7% by 12 months, and 13% by the end of 2 months.
years. We found that it made little difference whether the
fully breastfeeding or had already introduced supplements into the baby’s diet. These findings are in general agreement with the demographic data from developing countries, which suggest that only 5-10% of breastfeeding women conceive before the resumption of
woman was
Fig 1-Cumulative breastfeeding.
of
pregnancy
during
of normal fertility
having unprotected having unprotected intercourse throughout 24 mo of lactation; -1-=our breastfeeding women having unprotected intercourse only during lactational amenorrhoea, and adopting effective contraceptive measures at resumption of menstruation. Percentage of women in lactational amenorrhoea by month post partum (V) is also shown. -
-
-
-
=
non-lactating
probability women
intercourse; -----= our breastfeeding
women
breastfeeding women, if they had had unprotected intercourse only during lactational amenorrhoea and used a 100% effective form of contraception from the resumption of menstruation. We assumed a fecundability of 25% for any ovulation that was followed by a normal luteal phase. Anovulatory cycles and those in which the luteal phase was short or insufficient were presumed to be infertile. When data on cycles were not available (for example, after the third post-partum cycle), all women were assumed to have fertile cycles of 30 days. Fig 1 illustrates the cumulative percentage probability of pregnancy by months post partum for these three groups of women. 50% of non-lactating women would be pregnant in less than 3 months, and 85% would be pregnant by 6
Fig 2-Monthly percentage probability of
pregnancy
lactational amenorrhoea with estimated standard
during
errors.
menstruation.5 In Fig 2, we have plotted the monthly percentage probability of our women becoming pregnant during lactational amenorrhoea. The results are calculated from the incidence of potentially fertile ovulations before the first menstruation in the women in our study over 2-month (1-12 months) or 3-month (13-24 months) intervals, by proportional hazards modelling.6 When continued data after the third post-partum cycle were lacking we assumed that each woman continued to have fertile cycles every 30 days. Risks of pregnancy were calculated as 0-25 times the total incidence of fertile ovulations divided by the total womenmonths exposure in that interval. The increasing risk of pregnancy from 12 months onwards is due to the increasing likelihood that ovulation followed by a normal luteal phase will precede the first post-partum menstruation.
Discussion We conclude that even in well-nourished Caucasian women, an extended period of breastfeeding can produce long periods of lactational amenorrhoea and thus exert a major contraceptive effect. Many such women who merely wish to space their children might therefore prefer to rely on the reduced fertility that occurs during lactational amenorrhoea, withholding the use of other contraceptives until the time of their first post-partum menstruation if it should occur sooner than expected. In developing countries, where modem contraceptives may not be freely available or affordable, and where double contraceptive cover is therefore particularly wasteful, prolonged lactational amenorrhoea deserves to be promoted for birth spacing, and other contraceptive use could be postponed until the time of the first post-partum menstruation. The ultimate goal must be to achieve a birth interval of at least 2 years, since this variable is critically important for infant survival .7 A study of the risk of ovulation during lactation in American and Filipino women8 suggested that no single measure (such as the time of resumption of menstruation) could give an adequate warning of incipient ovulation, so that only amenorrhoeic women practising exclusive breastfeeding during the first 6 months could achieve protection equivalent to that provided by intrauterine devices or the pill. Our Australian women could achieve substantial contraceptive protection from lactational amenorrhoea long after supplements were introduced, and for periods of 12 months or more.3 The fact that they breastfed their babies for much longer than the American or Filipino women highlights the influence of different
717
on the duration of lactational amenorrhoea. Much confusion has arisen in the past because of claims that breastfeeding by itself is an effective form of contraception. It is not, as our data show. All our breastfeeding women eventually resumed normal, fertile menstrual cycles before they had weaned their babies and were therefore at risk of becoming pregnant while still breastfeeding. However, our data also show that fertility is greatly reduced during lactational amenorrhoea. Breastfeeding women who take no other contraceptive precautions during amenorrhoea will have significant protection from pregnancy, although this protection declines with time. This lactational amenorrhoea method (LAM) is particularly important for women in many traditional societies who breastfeed for long periods of time. Reliance on the contraceptive effect of LAM, complemented if necessary by the use of other forms of contraception once menses have returned, has much to recommend it. It should help to achieve a 2 year birth interval, and it also has the benefits of cost-effectiveness, reduction of maternal breast cancer risk, and reduced incidence of gastrointestinal infections in the child.7 Once menses have returned, it is essential for breastfeeding mothers to use other forms of contraception if pregnancy is
breastfeeding patterns
to
be prevented.
This work was supported by the Australian National Health and Medical Research Council, Monash University, the Lalor Foundation, the Jack Brockoff Foundation, the William Buckland Foundation, and Family Health International. We thank all the dedicated members of the Nursing Mothers Association of Australia, especially Lynne Hunt and Diane Lesley who coordinated the volunteers; Pascal Gelperowicz and Pauline Galvin for help with the salivary progesterone assays and data analysis; Peter Wagner, Jens Rasmussen, and David Caddy for computing assistance; and Dr Kathy Kennedy, Dr Nancy Williamson, and Dr Mark Belsey for helpful comments.
REFERENCES 1. Consensus Statement. Breastfeeding as a family planning method. Lancet 1988; ii: 1204-05. 2. Kennedy KI, Rivera R, McNeilly AS. Consensus statement on the use of as a family planning method. Contraception 1989; 39: 477-96. 3. Lewis PR, Brown JB, Renfree MB, Short RV. The resumption of ovulation and menstruation in a well-nourished population of women breastfeeding for an extended period of time. Fertil Steril (in press). 4. Wood JW. Fecundity and natural fertility in humans. Oxford Rev Reproduct Biol 1989; 11: 61-109. 5. Simpson-Hebert M, Huffman SL. The contraceptive effect of breastfeeding. Stud Family Plann 1981; 12: 125-33. 6. Rodriguez G, Diaz S. Breastfeeding and the length iof post-partum amenorrhea: a Hazards Model approach. IUSSP Committee on Comparative Analysis of Fertility and Family Planning, Dept of
breastfeeding
7.
Population Dynamics, Johns Hopkins University, 1988, pp. 14. Thapa S, Short RV, Potts M. Breast feeding, birth spacing and their
8.
Gray RH, Campbell OM, Apelo R,
effects on child survival. Nature 1988; 355: 679-82. et al. Risk of ovulation during lactation. Lancet 1990; 335: 25-29.
HYPOTHESIS One explanation of the asthma paradox: inhibition of natural anti-inflammatory mechanism by
&bgr;2-agonists
Our
understanding of the mechanisms contributing pathogenesis of bronchial asthma has increased substantially over the past decade. This has been accompanied by the introduction of a range of new drugs for the treatment of this disorder, and the usage of anti-asthma drugs is increasing. Despite these changes and an increased awareness of the disease, asthma remains the only "preventable" disease where the morbidity and mortality are still increasing in most parts of the world. This "asthma paradox" requires explanation, and this article is an attempt to provide a plausible scientific one. The hypothesis expresses concern that a long recognised, but little publicised, pharmacological property of the drug class most widely prescribed for the treatment of asthma, the &bgr;2-adrenoceptor agonists—namely, the inhibition of mast-cell degranulation—may be contributing to the world wide increase in morbidity and mortality from asthma via the inhibition of a natural anti-inflammatory
to the
mechanism. It is now five years since the Committee on Drugs of the American Academy of Allergy and Immunology expressed
that the adverse effects of (32 agonists might be contributing to the continuing trend towards increased morbidity from asthma. Their concerns have been supported by evidence suggesting that increased use of (32 agonists may be an important factor in the world wide increase in morbidity from asthma; regular use of (32 sympathomimetic agents seems to be associated with deterioration of asthma control. This clinical evidence had been preceded by four independent clinical studies showing that monotherapy with &bgr;2-agonists may exacerbate one of the underlying characteristics of asthma-that is, bronchial hyperresponsiveness to agents such as inhaled histamine when (32-agonists are used for between 15 days and a year. 3-6 These observations are worrying. I attempt to provide here a plausible hypothesis to explain these findings, based on what we now know of the pharmacology of 0,-agonists. concern
Mast cells and asthma cell has held a prominent place in allergy research, and undoubtedly mast-cell spasmogens contribute to acute bronchospasm after exposure to allergen.7 The discovery that sodium cromoglycate (SCG) was clinically The
mast
ADDRESS. Department of Pharmacology, King’s College, University of London, London SW3 6LX, UK (C. P. Page, PhD).
Contraceptive effects of extended lactational amenorrhoea: beyond the Bellagio Consensus
We have recorded the duration of lactational anovulation and amenorrhoea in a well-nourished group of Australian women who breastfed their babies throughout the study. The data enabled us to compare the theoretical cumulative probability of conception among breastfeeding women who had unprotected intercourse irrespective of their menstrual status with that of those who had
unprotected intercourse only during lactational amenorrhoea. Breastfeeding alone is not an effective form of contraception, since all the women in our study resumed normal ovulation while still breastfeeding. However, among women who have unprotected intercourse only during lactational amenorrhoea but adopt other contraceptive measures when they resume menstruation, only 1·7% would have become pregnant during the first 6 months of amenorrhoea, only 7% after 12 months, and only 13% after 24 months. Thus for our women it would be possible to extend the Bellagio Consensus Conference guidelines which stated that lactational amenorrhoea can only be relied on as a contraceptive for the first 6 months post-partum in women who are fully or almost fully breastfeeding. The lactational amenorrhoea method can be relied on for excellent contraceptive protection in the first 6 months of breastfeeding, irrespective of when supplements are introduced into the baby’s diet; for women who continue to breastfeed the method can also give good protection for up to 12 months post partum. Once menstruation has returned, other forms of contraception are essential to prevent pregnancy. Introduction
The Bellagio Consensus Conference on breastfeeding as a family planning method established that a mother who is fully or nearly fully breastfeeding her infant and who remains amenorrhoeic will have more than 98% protection from pregnancy in the first 6 months post partum.1,2 However, the contraceptive effect of longer periods of lactational amenorrhoea could not be quantified. In practice, many women may not be able to use the Bellagio guidelines to their full extent, since they will have started to introduce supplements into the baby’s diet before 6 months. We have completed a large prospective study of the duration of lactational anovulation and amenorrhoea in a well-nourished group of Australian women-members of
the Nursing Mothers’ Association of Australia-who breastfed for a long time.3
Subjects and methods Women were recruited into the study before their babies were 6 weeks old. Ovarian activity was monitored in all women by salivary progesterone determinations, and in a subgroup by means of urinary oestrogen and pregnanediol excretion rates. Saliva samples were collected 2-4 times per week. Twice a week the women also kept a 24 h record of the timing and duration of all breastfeeds and the timing, nature, and amount of all supplementary feeds. Ovulation was presumed to have occurred if there was a rise in salivary progesterone above the baseline established for that individual woman, and/or a urinary pregnanediol excretion rate of 1 mg/24 h or greater. An insufficient luteal phase was defmed as a rise in salivary progesterone above baseline but below 40 pg/ml, or a urinary pregnanediol excretion rate of more than 1 but less than 2 mg/24 h. A short luteal phase was defined as 11 days or less from ovulation to menstruation. Of the 142 women recruited, 41 dropped out because of the highly demanding nature of the experimental protocol. We have data on the time of first post-partum menstruation for 101 women. If anything, our data are therefore biased against women with longer durations of amenorrhoea. Some women did not collect saliva samples regularly, so we have complete hormonal data up to the time of resumption of menstrual cycles in only 89 women. The study was considered to be complete when a woman had had three menstrual cycles post partum. Each woman continued to breastfeed her baby throughout the time she was in the
study.
Results The
duration of lactational amenorrhoea was 9-5 months (in 101 women) and the mean duration of (SD 4-9) anovulation 10 6 (5-0) months (in 89 women). Although some of the women breastfed for as long as 2 years and were amenorrhoeic for over 18 months, the mean age of their infants at the introduction of supplementary feeding was 53 (11) months, and only 24 of the 101women would have been able to benefit by the full 6 months of the Bellagio mean
guidelines. We calculated the theoretical cumulative probability of pregnancy over 24 months for: a hypothetical population of
non-lactating women of normal fertility having unprotected intercourse, assuming a fecundability of 25 % per menstrual cycle;4 our breastfeeding women, if they had had unprotected intercourse for 24 months; and our ADDRESSES: Departments of Anatomy and Physiology, Monash University, Melbourne (Prof R. V. Short, ScD FRS, M. B. Renfree, DSc, G. Shaw, PhD), and Nursing Mothers Association of Australia, Nunawading, Australia (P.R. Lewis, PhD). Correspondence to Prof R. V. Short, Department of Physiology, Monash University, Melbourne, 3168, Australia.
716
By contrast, 50% of our breastfeeding women having unprotected intercourse irrespective of when their menses returned would become pregnant by 12 months post partum. Thus, although breastfeeding alone reduces fertility, it is not a particularly effective or reliable form of contraception. However, as intimated by the Bellagio guidelines, women in lactational amenorrhoea have a pronounced reduction in fertility. For the breastfeeding women who used other contraceptive measures when menses resumed, only 1-7% would have become pregnant by 6 months, 7% by 12 months, and 13% by the end of 2 months.
years. We found that it made little difference whether the
fully breastfeeding or had already introduced supplements into the baby’s diet. These findings are in general agreement with the demographic data from developing countries, which suggest that only 5-10% of breastfeeding women conceive before the resumption of
woman was
Fig 1-Cumulative breastfeeding.
of
pregnancy
during
of normal fertility
having unprotected having unprotected intercourse throughout 24 mo of lactation; -1-=our breastfeeding women having unprotected intercourse only during lactational amenorrhoea, and adopting effective contraceptive measures at resumption of menstruation. Percentage of women in lactational amenorrhoea by month post partum (V) is also shown. -
-
-
-
=
non-lactating
probability women
intercourse; -----= our breastfeeding
women
breastfeeding women, if they had had unprotected intercourse only during lactational amenorrhoea and used a 100% effective form of contraception from the resumption of menstruation. We assumed a fecundability of 25% for any ovulation that was followed by a normal luteal phase. Anovulatory cycles and those in which the luteal phase was short or insufficient were presumed to be infertile. When data on cycles were not available (for example, after the third post-partum cycle), all women were assumed to have fertile cycles of 30 days. Fig 1 illustrates the cumulative percentage probability of pregnancy by months post partum for these three groups of women. 50% of non-lactating women would be pregnant in less than 3 months, and 85% would be pregnant by 6
Fig 2-Monthly percentage probability of
pregnancy
lactational amenorrhoea with estimated standard
during
errors.
menstruation.5 In Fig 2, we have plotted the monthly percentage probability of our women becoming pregnant during lactational amenorrhoea. The results are calculated from the incidence of potentially fertile ovulations before the first menstruation in the women in our study over 2-month (1-12 months) or 3-month (13-24 months) intervals, by proportional hazards modelling.6 When continued data after the third post-partum cycle were lacking we assumed that each woman continued to have fertile cycles every 30 days. Risks of pregnancy were calculated as 0-25 times the total incidence of fertile ovulations divided by the total womenmonths exposure in that interval. The increasing risk of pregnancy from 12 months onwards is due to the increasing likelihood that ovulation followed by a normal luteal phase will precede the first post-partum menstruation.
Discussion We conclude that even in well-nourished Caucasian women, an extended period of breastfeeding can produce long periods of lactational amenorrhoea and thus exert a major contraceptive effect. Many such women who merely wish to space their children might therefore prefer to rely on the reduced fertility that occurs during lactational amenorrhoea, withholding the use of other contraceptives until the time of their first post-partum menstruation if it should occur sooner than expected. In developing countries, where modem contraceptives may not be freely available or affordable, and where double contraceptive cover is therefore particularly wasteful, prolonged lactational amenorrhoea deserves to be promoted for birth spacing, and other contraceptive use could be postponed until the time of the first post-partum menstruation. The ultimate goal must be to achieve a birth interval of at least 2 years, since this variable is critically important for infant survival .7 A study of the risk of ovulation during lactation in American and Filipino women8 suggested that no single measure (such as the time of resumption of menstruation) could give an adequate warning of incipient ovulation, so that only amenorrhoeic women practising exclusive breastfeeding during the first 6 months could achieve protection equivalent to that provided by intrauterine devices or the pill. Our Australian women could achieve substantial contraceptive protection from lactational amenorrhoea long after supplements were introduced, and for periods of 12 months or more.3 The fact that they breastfed their babies for much longer than the American or Filipino women highlights the influence of different
717
on the duration of lactational amenorrhoea. Much confusion has arisen in the past because of claims that breastfeeding by itself is an effective form of contraception. It is not, as our data show. All our breastfeeding women eventually resumed normal, fertile menstrual cycles before they had weaned their babies and were therefore at risk of becoming pregnant while still breastfeeding. However, our data also show that fertility is greatly reduced during lactational amenorrhoea. Breastfeeding women who take no other contraceptive precautions during amenorrhoea will have significant protection from pregnancy, although this protection declines with time. This lactational amenorrhoea method (LAM) is particularly important for women in many traditional societies who breastfeed for long periods of time. Reliance on the contraceptive effect of LAM, complemented if necessary by the use of other forms of contraception once menses have returned, has much to recommend it. It should help to achieve a 2 year birth interval, and it also has the benefits of cost-effectiveness, reduction of maternal breast cancer risk, and reduced incidence of gastrointestinal infections in the child.7 Once menses have returned, it is essential for breastfeeding mothers to use other forms of contraception if pregnancy is
breastfeeding patterns
to
be prevented.
This work was supported by the Australian National Health and Medical Research Council, Monash University, the Lalor Foundation, the Jack Brockoff Foundation, the William Buckland Foundation, and Family Health International. We thank all the dedicated members of the Nursing Mothers Association of Australia, especially Lynne Hunt and Diane Lesley who coordinated the volunteers; Pascal Gelperowicz and Pauline Galvin for help with the salivary progesterone assays and data analysis; Peter Wagner, Jens Rasmussen, and David Caddy for computing assistance; and Dr Kathy Kennedy, Dr Nancy Williamson, and Dr Mark Belsey for helpful comments.
REFERENCES 1. Consensus Statement. Breastfeeding as a family planning method. Lancet 1988; ii: 1204-05. 2. Kennedy KI, Rivera R, McNeilly AS. Consensus statement on the use of as a family planning method. Contraception 1989; 39: 477-96. 3. Lewis PR, Brown JB, Renfree MB, Short RV. The resumption of ovulation and menstruation in a well-nourished population of women breastfeeding for an extended period of time. Fertil Steril (in press). 4. Wood JW. Fecundity and natural fertility in humans. Oxford Rev Reproduct Biol 1989; 11: 61-109. 5. Simpson-Hebert M, Huffman SL. The contraceptive effect of breastfeeding. Stud Family Plann 1981; 12: 125-33. 6. Rodriguez G, Diaz S. Breastfeeding and the length iof post-partum amenorrhea: a Hazards Model approach. IUSSP Committee on Comparative Analysis of Fertility and Family Planning, Dept of
breastfeeding
7.
Population Dynamics, Johns Hopkins University, 1988, pp. 14. Thapa S, Short RV, Potts M. Breast feeding, birth spacing and their
8.
Gray RH, Campbell OM, Apelo R,
effects on child survival. Nature 1988; 355: 679-82. et al. Risk of ovulation during lactation. Lancet 1990; 335: 25-29.
HYPOTHESIS One explanation of the asthma paradox: inhibition of natural anti-inflammatory mechanism by
&bgr;2-agonists
Our
understanding of the mechanisms contributing pathogenesis of bronchial asthma has increased substantially over the past decade. This has been accompanied by the introduction of a range of new drugs for the treatment of this disorder, and the usage of anti-asthma drugs is increasing. Despite these changes and an increased awareness of the disease, asthma remains the only "preventable" disease where the morbidity and mortality are still increasing in most parts of the world. This "asthma paradox" requires explanation, and this article is an attempt to provide a plausible scientific one. The hypothesis expresses concern that a long recognised, but little publicised, pharmacological property of the drug class most widely prescribed for the treatment of asthma, the &bgr;2-adrenoceptor agonists—namely, the inhibition of mast-cell degranulation—may be contributing to the world wide increase in morbidity and mortality from asthma via the inhibition of a natural anti-inflammatory
to the
mechanism. It is now five years since the Committee on Drugs of the American Academy of Allergy and Immunology expressed
that the adverse effects of (32 agonists might be contributing to the continuing trend towards increased morbidity from asthma. Their concerns have been supported by evidence suggesting that increased use of (32 agonists may be an important factor in the world wide increase in morbidity from asthma; regular use of (32 sympathomimetic agents seems to be associated with deterioration of asthma control. This clinical evidence had been preceded by four independent clinical studies showing that monotherapy with &bgr;2-agonists may exacerbate one of the underlying characteristics of asthma-that is, bronchial hyperresponsiveness to agents such as inhaled histamine when (32-agonists are used for between 15 days and a year. 3-6 These observations are worrying. I attempt to provide here a plausible hypothesis to explain these findings, based on what we now know of the pharmacology of 0,-agonists. concern
Mast cells and asthma cell has held a prominent place in allergy research, and undoubtedly mast-cell spasmogens contribute to acute bronchospasm after exposure to allergen.7 The discovery that sodium cromoglycate (SCG) was clinically The
mast
ADDRESS. Department of Pharmacology, King’s College, University of London, London SW3 6LX, UK (C. P. Page, PhD).
