In this randomized, multicentre, placebo-controlled trial no benefit was found of a loading dose IV salbutamol versus normal saline in children admitted to a PICU with SAA (majority already on IV salbutamol infusion), with regard to clinical asthma score, co-medication, respiratory support, and PICU length of stay (LOS). Neither were there side effects. We found an association between heart rate and asthma score, whilst no association was found between heart rate and salbutamol plasma levels, suggesting that tachycardia in SAA children is mainly caused by respiratory distress, not dose of salbutamol IV.
Our findings are only partly consistent with previous studies. Single center randomized studies done in Australia (1994-1995, 1997-1998) showed shorter recovery time (e.g. cessation of inhaled medication), which resulted in earlier hospital discharge, in SAA children who received a bolus of IV salbutamol (15 mcg/kg in 10 minutes), whereas no side effects were reported. However the results of these 2 studies cannot be compared to our present study. The setting was the emergency department (ED) and a loading dose salbutamol was not followed by or added to continuous salbutamol infusion [10, 11].
In 2007 Bogie performed a randomized, double blind, placebo-controlled trial on children with SAA presenting to an ED. Patients were randomized to receive either IV terbutaline (a loading dose followed by continuous infusion) or IV normal saline while on continuous high-dose nebulized albuterol. Outcome measures revealed a trend toward improvement in the terbutaline group [12]. In the present study we looked at the efficacy of a loading dose salbutamol while already on continuous infusion with salbutamol, in a PICU setting. We found no clinical benefit of an additional loading dose salbutamol in this population.
The efficacy of IV salbutamol was suggested in 1984 by Bohn et al, showing a decrease of PaCO2 in 11 of 16 SAA patients after a loading dose of 10 mcg/kg followed by continuous infusion of salbutamol [24]. This study is not comparable with the present study (e.g. different loading dose and outcome variable). None of these 4 studies described PK data [10-12, 24].
How can our findings, e.g. lack of efficacy of a loading dose IV salbutamol, be explained?
The pharmacologic activity of salbutamol resides predominantly in the (R)-isomer, with little or no activity, and concerns about adverse reactions, attributed to the (S)-isomer [6, 25]. Based on a previous population PK model of IV R- and S-salbutamol in SAA children, a loading dose seemed valid to reach higher initial R-salbutamol concentrations with a possible therapeutic advantage [6].
All of our patients received nebulized salbutamol before IV. Furthermore, the majority of our children were already on IV salbutamol infusion (median dose at baseline of 0.5 mcg/kg/min) before inclusion. Although a salbutamol loading dose results in significant higher plasma levels 10 minutes after administration of the loading dose, this effect did not remain significant after 1 hour. Treatment with IV salbutamol resulted in high inter-individual differences in plasma salbutamol levels, with no clear correlation with pharmacodynamic parameters (e.g. asthma score and heart rate). Better knowledge of the pharmacokinetics of both R- and S-salbutamol is important to find the optimal dose for bronchodilation without adverse reactions.
Our study does not exclude a possible benefit of a loading dose of IV salbutamol in children with severe or near-fatal SAA in a pre-hospital setting or ED with very low baseline concentration salbutamol, as previously illustrated in a case report [26].
In SAA treatment, the goal of salbutamol as a selective β2-adrenoreceptor agonist with potent smooth muscle relaxant properties, is bronchodilation. Could it be that inflammation or mucus plugging are more predominant in the cause of severe airflow obstruction in SAA? The pathophysiology or “clinical phenotype” of SAA children should be better understood.
IV salbutamol is more likely than inhaled salbutamol to produce systemic adverse reactions related to interaction with non-airway β1 receptors, such as tachycardia and dysrhythmias. However, data on toxicity related to dose or serum concentrations are scarce. In a recent review on SAA in children, the limiting factor in increasing the dose of salbutamol is tachycardia. Also an NHS clinical guideline recommends to reduce the infusion rate of salbutamol in case of tachycardia. In the present study, there was no association between heart rate and serum salbutamol concentration, implicating that tachycardia was mainly caused by severity of respiratory distress, not by the dose of salbutamol. The PK-PD data could implicate that clinicians at the ED or PICU should titrate the dose of continuous salbutamol infusion on the severity of respiratory distress (including heart rate), taking into account that tachycardia after initiation of salbutamol inhalation or infusion was not found to be a side effect of SAA treatment in our study.
Limitations/strengths
Strengths of the present study include the randomized placebo-controlled design, the PK-PD analysis for (side)effects and the participation of the majority of Dutch PICU’s.
There are some limitations as well. First as to the design of the study, most children received continuous infusion with salbutamol for more than 1 hour before start of study medication. Unfortunately, it was unfeasible to administer a blinded loading dose of IV salbutamol in regional hospitals (before the continuous infusion of IV salbutamol) as part of a research project. Second, the asthma score by Qureshi is a subjective description for determining asthma severity and change in clinical condition. The asthma score is widely used in studies regarding SAA. To our knowledge there have never been subsequent studies validating this score for different populations, and the precise characteristics of this score have not been determined; the original article mentions an interrater reliability of 80%. It may therefore be that this score could be improved to detect clinical improvement in studies like the present one. However, since also co-medication, respiratory support and PICU LOS did not differ between both groups, it is unlikely that the asthma score was inappropriate.