Article Text


Cardiovascular side effects of inhaled salbutamol in hypoxic asthmatic patients


BACKGROUND Beta-2 adrenoceptor agonists have been associated with sudden death in asthma patients but the cause and underlying mechanism are unclear. Animal experiments indicate that the combination of hypoxia and β2 agonists may result in detrimental cardiovascular effects. A study was undertaken to investigate the effect of hypoxia on the systemic vascular effects of salbutamol in patients with asthma who are hypoxic by assessing forearm blood flow (FBF) as a measure of peripheral vasodilatation.

METHODS Eight men with mild asthma underwent the following treatments: normoxia + placebo (NP), normoxia + salbutamol (NS), hypoxia + placebo (HP), and hypoxia + salbutamol (HS). The period of mask breathing started at t=0 minutes, lasted for 60 minutes, and at 30 minutes 800 μg salbutamol was inhaled. The experiment was completed 30 minutes after the inhalation (t=60 minutes). For the hypoxia treatment the Spo 2 level was 82%. Differences between treatments were sought using factorial ANOVA on percentage change from the pretreatment value.

RESULTS There were no significant differences in blood pressure and potassium levels between the treatments. After 60 minutes the increase in FBF was 13% (95% CI –12 to 39) more for HP treatment than for NP, 21% (95% CI –5 to 46) more for NS than for NP, and 32% (95% CI 7 to 58) more for HS than for HP (p=0.016). The inhalation of salbutamol during hypoxia resulted in a significant increase in FBF of 45% (95% CI 20 to 71) compared with NP (p=0.001).

CONCLUSION Patients with asthma who are hypoxic and inhale β2 agonists have serious systemic vascular side effects which may be an additional explanation for the association between asthma treatment and sudden death.

Statistics from

Beta-2 adrenoceptor agonists (β2 agonists) are the most effective and widely used bronchodilator drugs for treatment of acute exacerbations of asthma. However, ever since their introduction, β agonists have been associated with sudden death in asthma.1 Many explanations for the association of sudden death and asthma treatment have been proposed and refuted.2-7 Among the factors that may play an important role are the severity of asthma,8 treatment intensity,9 socioeconomic factors,10 cardiac arrhythmias (due to QT prolongation and/or changes in potassium levels), and the use of β agonists.5 11 It appears that most of the sudden deaths occur outside hospital.12 In contrast with the situation outside hospital, clinical treatment of exacerbations of asthma in hospital involves the concomitant administration of β agonists and oxygen. It can be hypothesised that administration of β agonists during hypoxia is potentially detrimental and provides an alternative explanation for the association with sudden death.13 Experimental studies have shown that β agonists are lethal in dogs when the animals are hypoxic.14 We have recently shown in volunteers exposed to hypoxia that a similar mechanism may also apply to humans.15 Furthermore, it has been shown that inhalation of β agonists can worsen hypoxia.16 The present study was designed to investigate the effect of hypoxia on the systemic vascular effects of salbutamol in patients with asthma who are hypoxic.


The study protocol was approved by the medical ethics committee of Leiden University Medical Center. After obtaining informed consent, eight men with mild asthma (aged 21–26 years, using β agonists on demand only) who complied with the international classification of intermittent mild asthma (NNHLBI/WHO Workshop 1995) participated in a double blind, placebo controlled, four way, crossover study. The following interventions were investigated: normoxia and placebo, normoxia and inhaled salbutamol, hypoxia and placebo, and hypoxia with salbutamol inhalation. Each intervention was separated by a one week washout period. During the experiments the subjects breathed ambient air or a variable N2/O2 mixture through a well fitting face mask. The N2/O2 mixture was continuously adjusted to obtain a peripheral oxygen saturation (Spo 2) of 80%. Salbutamol 800 μg and placebo were administered double blind as metered dose aerosols via a spacer device (Volumatic). After obtaining stable baseline haemodynamic values (forearm blood flow (FBF), mean arterial pressure (MAP), and heart rate) the subjects breathed either ambient air (normoxia) or an N2/O2 mixture (hypoxia) for 60 minutes. Salbutamol or placebo were administered after 30 minutes mask breathing. FBF was measured 30 and 60 minutes after baseline using computerised venous occlusion plethysmography.15Monitoring of one lead ECG and Spo 2 was done continuously. At regular time intervals blood pressure (oscillometric) was measured and 12-lead ECGs were recorded. Forearm vascular resistance (FVR) was calculated as MAP divided by FBF (in mmHg/%FBF). The interventions were compared using factorial ANOVA (factors subject and treatment) on the percentage change after 60 minutes relative to the baseline value. Contrasts between treatments within the ANOVA model are presented with 95% confidence intervals or by reporting significant contrasts.


All subjects completed the study without clinically significant adverse events. Subjects could not distinguish whether they had been exposed to hypoxia or normoxia. Hypoxia was easily achieved and maintained at an Spo 2 level of 82 (3)%. In three subjects who were administered salbutamol during hypoxia Spo 2 levels rapidly decreased below 80% (to 60–70%), requiring cessation of N2 for several minutes. Salbutamol during normoxia did not influence Spo 2. Hypoxia did not greatly influence MAP, QTc interval, or potassium levels compared with normoxia (table 1). Salbutamol had no effect on MAP and only a minor effect on QTc (<7% prolongation) and potassium levels (<5% decrease) during both hypoxia and normoxia. After 60 minutes the increase in heart rate was 16% (95% CI 6 to 26) more for hypoxia/placebo than for normoxia/placebo. The increase in heart rate was 9% (95% CI –2 to 19) more for normoxia/salbutamol than for normoxia/placebo, and 13% (95% CI 3 to 24) more for hypoxia/salbutamol than for hypoxia/placebo. After 60 minutes the increase in FBF was 13% (95% CI –12 to 39) more for hypoxia/placebo than for normoxia/placebo (fig 1). The increase in FBF was 21% (95% CI –5 to 46) more for normoxia/salbutamol than for normoxia/placebo, and 32% (95% CI 7 to 58) more for hypoxia/salbutamol than for hypoxia/placebo (p=0.016). Inhalation of salbutamol during hypoxia resulted in a significant increase in FBF of 45% (95% CI 20 to 71) compared with normoxia/placebo (p=0.001).

Table 1

Mean (SD) values (n=8) for mean arterial pressure (MAP), heart rate (HR), QTc interval, forearm vascular resistance (FVR), and serum potassium levels (K+) at baseline (t=0) and at the end of the experiment (t=60)

Figure 1

Mean (SD) percentage change in forearm blood flow (FBF) induced by placebo (normoxia + placebo), hypoxia (hypoxia + placebo), salbutamol alone (normoxia + salbutamol), and hypoxia with salbutamol (hypoxia + salbutamol). The arrow indicates inhalation of placebo/salbutamol.


This study has shown that patients with asthma who were hypoxic and inhaled salbutamol at a relatively low dose experienced significant and potentially detrimental cardiovascular effects because of substantial vasodilatation and possibly pulmonary shunting. The 45% increase in FBF (or 30% decrease in FVR) is of the same order of magnitude as that observed after vasodilators such as 0.6–0.9 mg sublingual nitrogylcerin (35% decrease in FVR)17 or 10 mg oral felodipine (30% increase in FBF).18

The conclusion that salbutamol during hypoxia may have caused pulmonary shunting was based upon the observation that, in three of the eight subjects, Spo 2 levels declined rapidly after inhalation. Obviously, this prompted immediate cessation of breathing of N2 and made it impossible to test formally whether or not this observation was related by chance to the combination of hypoxia and salbutamol. Nevertheless, the observation seems to confirm previous data showing that β2 agonists can induce pulmonary shunting.16

This study therefore provides an additional explanation for the association between the use of β agonists and sudden death in asthma. Salbutamol did not greatly influence the QTc interval or potassium level. This is in agreement with previous studies showing that, with the salbutamol dose used in this study, only minor or no changes in these parameters occurred.19 It was found that both hypoxia and inhalation of salbutamol induced peripheral vasodilatation, but these effects hardly reached statistical significance. In contrast, the combination of salbutamol and hypoxia resulted in profound cardiovascular effects evidenced by a significant increase in FBF, reflecting a reduction in peripheral vascular resistance. Such a mechanism may result in a diminished venous return, especially when the patient is in the upright position, eliciting the Bezold-Jarisch reflex and subsequent cardiac arrest.20

In conclusion, inhaled salbutamol during hypoxia causes significant cardiovascular effects that can be detrimental in compromised patients. Asthmatic patients in respiratory distress should be given β2 agonists and oxygen concomitantly whenever possible.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.