Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Theophylline versus terbutaline in treating critically ill children with status asthmaticus: A prospective, randomized, controlled trial.

Wheeler DS, Jacobs BR, Kenreigh CA, Bean JA, Hutson TK, Brilli RJ.

Pediatr Crit Care Med. 2005 Mar;6(2):142-147 [abstract]

Reviewed by Aditi Sharangpani MD and JoAnne Natale MD PhD, Critical Care Medicine, Children's National Medical Center, Washington, DC

Review posted November 17, 2005

---

I. What is being studied?:

The study objective:

To compare the effectiveness of theophylline alone, terbutaline alone or theophylline combined with terbutaline infusion in children with severe status asthmaticus who are receiving continuous nebulized albuterol and intravenous corticosteroids.

The study design:

Prospective, randomized, triple-blind clinical trial.

The patients included:

Criteria

• Age: 3 to 15 years old
• Medical condition: severe status asthmaticus defined as intractable wheezing requiring continuous nebulized albuterol and a modified Becker Clinical Asthma Score (CAS) ≥7. The CAS is a widely-used scoring system that provides a rapid, quantitative assessment of asthma clinical severity. Composed of four clinical characteristics (respiratory rate, wheezing, inspiration/expiration ratio, and accessory muscle use), the CAS ranges from 0 (minimal severity) to 12 (maximal severity).
• Admission location: intensive care unit
• Enrollment timing: within 2 hours of ICU admission
• Medical history: asthma defined as evaluation by physician for recurrent wheezing

Characteristics of those included

• 40 subjects enrolled, 36 subjects completed protocol
• Age 2.8-19 yrs
• 336 screened; 168 (50%) missed 2 hr enrollment window (night admissions); 168 remained eligible
• 168 eligible; 80 excluded with CAS <7; 88 remained eligible
• 88 eligible; 30 excluded due to other exclusion criteria; 58 remained eligible (paper states 59 screened)
• 59 "screened"; 19 parents refused consent; 40 subjects enrolled

The patients excluded:

Criteria

• Acute respiratory diagnoses other than asthma determined by history, physical exam, and chest x-ray
• Known allergy or hypersensitivity to methylxanthine, beta-agonist, or anticholinergic medication
• Pregnancy
• Use of either terbutaline or theophylline in the 24 hours before enrollment
• Underlying cardiovascular disease
• Failure to obtain consent
• A priori decision to withdraw patients who required mechanical ventilation
• A priori decision to withdraw patients with life threatening events (CPR or seizures)

The interventions compared:

• Group 1: 6.4 mg/kg iv theophylline followed by age-adjusted infusion dose of theophylline.
• Group 2: 20 mcg/kg iv terbutaline followed by 0.4 mcg/kg/min terbutaline infusion adjusted in response to change in CAS to a maximum of 3 mcg/kg/min
• Group 3: 6.4 mg/kg iv theophylline followed by age-adjusted infusion dose of theophylline and 20 mcg/kg iv terbutaline followed by 0.4 mcg/kg/min terbutaline infusion adjusted in response to change in CAS

The outcomes evaluated:

• Primary outcome: rate of change in CAS (yet in the Results, they state the primary outcome was simply change in CAS, with no mention of rate of change).
• Secondary outcomes: length of time to CAS≤3, length of stay in ICU, progression to mechanical ventilation, incidence of adverse effects, cost (drug and theophylline levels) comparison

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Unable to determine from the information provided. Although the authors state that randomization was done by inpatient hospital pharmacy using a sealed envelope technique, the actual technique of randomization was not specified (e.g. random number table, computer-generated, etc.).

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Yes, 40 subjects entered the trial; 2 patients withdrew at parental request due to agitation and behavioral disturbances (both in the theophylline-alone group), 1 subject received wrong study drug (theophylline-alone group), 1 subject withdrawn by treating physician (theophylline combined with terbutaline group). Of note, three participants withdrawn from theophylline-alone represents 23% of the 13 individuals enrolled in this group.

Was followup complete?

Not to the primary endpoint, the rate of change in CAS. This result was not presented. However, time to CAS≤3, a secondary endpoint, was reported but only after excluding the four patients who exited the study. Although not explicitly stated, it appears that an intention-to-treat analysis was carried out, i.e., even those removed from the study were included in the analysis. Therefore, the results of time to CAS≤3 are incomplete. In addition, other endpoints, such as the rate of readmissions to the PICU between groups, may have been another meaning outcome to report in this study.

Were patients analyzed in the groups to which they were randomized?

Yes. There was no cross-over between groups.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

Yes, only the study pharmacist who adjusted theophylline dosing based on serum levels was aware of a subject's treatment group.

4. Were the groups similar at the start of the trial?

No. Theophylline/terbutaline group had more severe illness based on significantly higher PRISM scores (13.9±8.8 vs. 7.4±2.8 and 8.8±1.3 for theophylline and terbutaline groups respectively, p=0.006) and higher serum glucose (322±2.7 vs. 200±21 and 242±21 mg/dL for theophylline and terbutaline groups respectively, p=0.005).

5. Aside from the experimental intervention, were the groups treated equally?

Probably. However, although no "statistically significant" differences in adjunct therapies were identified, there are notable differences in the proportion of participants receiving ipratropium bromide (31% vs. 3% vs. 27%) and ketamine (23% vs. 6% vs. 0%) in the theophylline alone, terbutaline alone, and theophylline + terbutaline groups respectively. Although the "significance" of such treatment differences between groups remain speculative, the fact that the adjunctive use was not "statistically significant" between the groups may have simply been because the study was underpowered to detect difference these variables.

III. What were the results?

1. How large was the treatment effect?

Interim analysis demonstrated no significant differences in the effectiveness of theophylline-alone, terbutaline-alone, or theophylline combined with terbutaline for the treatment of severe status asthmaticus in children. Because it would require "several hundred" additional participants to detect a difference between these therapies, this study was halted. Details of these study results are presented below

Three different kinds of outcomes were analyzed: Illness / PICU related outcomes, adverse events and cost. In each analysis, the groups are compared to one another.

Illness / PICU related

• CAS improved significantly from enrollment to discharge from the study and similarly in all three groups (reported as mean ± standard error of mean): theophylline alone 8.8 ± 0.3 to 3.9 ± 1.0, terbutaline alone 8.3 ± 0.4 to 4.3 ± 1.1, and theophylline + terbutaline 8.7 ± 0.4 to 4.3 ± 1.1, p<0.05 in each of the three groups. However, since to exit the study all patients had to have improvement in the CAS, this is not surprising. It is surprising that they do not present in tabular or graphical form the rate of change of CAS over time between the three groups.
• Length of PICU stay and time needed to achieve CAS ≤ 3 did not significantly differ between the three groups when analyzed as intention-to-treat. When data from only subjects who completed the study were analyzed, those children treated in the theophylline group achieved a CAS ≤ 3 in a significantly shorter time period (24.2 ± 12.1 h) than did children in the terbutaline only group (51.6 ± 33.3 h) or the terbutaline + theophylline group (47.1 ± 38.3) p<0.05. However, such an analysis is weaker than an intention to treat analysis.
• Progression to mechanical ventilation did not occur in any of the study patients.

Adverse Events

• Supratherapeutic theophylline levels (20-23 mcg/ml) occurred in 2 of 13 patients who received theophylline alone and in 2 of 11 patients who received theophylline in conjunction with terbutaline.
• More children in the theophylline + terbutaline group experienced nausea (10 / 11 patients) than in the terbutaline only group (5 / 16 patients). No difference was detected when compared to the theophylline only group (9 / 13 patients).
• The theophylline + terbutaline group had a higher median number of complaints of nausea (3 reports / patient) and tremors 2 reports / patient) than the terbutaline group alone (0 reports/patient for each). No difference was detected when compared to the theophylline only group (nausea: 1 report / patient, tremors: 3 report / patient).

Cost

• Median hospital cost of medication and theophylline concentrations was lower in the theophylline only group ($280) than in either the terbutaline only group ($3908) or terbutaline + theophylline group ($4045)

2. How precise was the estimate of the treatment effect?

Illness / PICU

• 95% confidence intervals (CI) around the reported primary outcome (CAS at baseline; CAS at study completion) by treatment group: theophylline alone (8.21, 9.39; 1.64, 5.86), terbutaline alone (7.52, 9.08; 2.14, 6.46), and terbutaline + theophylline (7.92, 9.48; 2.14, 6.46).
• 95% CI around length of PICU stay in hours by treatment group: theophylline alone (0.7, 48.1), terbutaline alone (-13.7, 116.9), and terbutaline + theophylline (-28.0, 122.2). Such wide CIs indicates the study has low precision. When reporting a result that is not normally distributed, such as length of stay, the median is a better measure of central tendency.

Adverse Events

• The risk ratio (RR) for a child to experience nausea while receiving terbutaline + theophylline, as opposed to terbutaline alone is 2.94, (95% CI 1.5, 3.8).

IV. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

The patient sample studied is similar to most populations in pediatric intensive care units. All are established treatments for asthma, particularly for asthma severe enough to require ICU admission. The inclusion criteria are broad and inclusive of most PICU patients with asthma. The exclusion criteria are not unnecessarily restrictive, and with the exception of "use of theophylline or terbutaline in the preceding 24 hours", they are appropriate for the standard use of these medications. However, because the treatment effect was small and study enrollment was low, the lack of significant differences among the three treatment groups in the primary outcome variable does not provide evidence that these treatments either benefit or harm patients with severe asthma.

2. Were all clinically important outcomes considered?

Length of PICU stay, the adverse events and cost of the medication are all clinically significant outcomes. In addition to those, we would have considered the following:

Illness / PICU related

• Although "change in CAS [clinical asthma score] over time" was the primary outcome, a description of rate of improvement in the CAS was not presented (e.g. change CAS / hour).
• The additional outcome of length of hospital stay would have also been useful, particularly if adverse events (e.g. inability to maintain hydration orally due to nausea) or a decrease in improvement following discontinuation of the study drug resulted in additional floor days.
• Subjective patient variables, e.g. resolution of dyspnea or agitation, patient satisfaction.

Adverse events

• Neither agitation nor behavioral disturbances (cited as reasons parents withdrew their children following randomization) were included in the list of adverse outcomes.
• The type of cardiac dysrhythmia (e.g. sinus tachycardia versus ventricular ectopy) was not stated.
• The magnitude of each adverse event was not described. In the case of hypokalemia for instance, for instance, did 3 of 11 children have potassium concentrations of 3.0-3.5 mEq/L or 1.5-2.0 mEq /L?

Costs

• An analysis of total PICU and hospital costs - which would include cost of additional medications needed to treat the asthma exacerbation as well as treatment of adverse events - would be useful

3. Are the likely treatment benefits worth the potential harms and costs?

We would not change our practice based on this study. This study addresses the question of "If a child requires escalation of asthma therapy beyond supplemental oxygen, parenteral steroids and inhaled beta agonists, which of the following is the most efficacious: the addition of theophylline, the addition of terbutaline or the addition of both theophylline and terbutaline?" In the study, the stated primary outcome--change in CAS over time--was not reported and could not be calculated because the duration of the study infusions for each group was not reported. Combination therapy seemed to be associated with a higher incidence of certain side effects when compared to terbutaline alone. Additionally, two children were withdrawn from the theophylline group by their parents due to side effects. These suggest that terbutaline is better tolerated than theophylline. On the other hand, the costs of the medications and associated blood levels were more than 10-fold lower in children receiving theophylline than in children in either of the other two groups.

Here, a more detailed description of the adverse effects would be useful. Hypokalemia severe enough to require multiple parenteral doses of potassium along with severe nausea and tremors as well as occasional complex ventricular ectopy is not the same as mild hypokalemia not requiring treatment, mild nausea and tremors and sinus tachycardia. The clinical significance exists despite the lack of statistical significance. Of note, the study design does not allow for this determination (most adverse events are "yes/no"), even had sufficient enrollment numbers been reached for statistical significance. Also, costs related to adverse effects may outstrip the cost of the medication itself. The number of patients in each arm was likely too small to be able to detect these differences; the authors reported difficulty in achieving a sufficient number of enrollees which may also be an indication of how well standard therapies work, and how infrequently a patient with asthma needs more aggressive treatment. Both theophylline and terbutaline given as intravenous infusions require some skill and experience, as do therapies such as heliox, positive airway pressure and ketamine. Perhaps selecting a few of these as "first line" escalation therapies will promote physician, bedside nurse and respiratory therapist proficiency in their use - whatever the choices are. Given the lack of differences in improvement in the CAS as well as the above concerns, this study does not support the use of either agent over the other or over the combination of the two.

Comments

Back to the EB Journal Club Index