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

Extubation failure due to post-extubation stridor is better correlated with neurologic impairment than with upper airway lesions in critically ill pediatric patients

Harel Y, Vardi A, Quigley R, Brink LW, Manning SC, Carmody TJ, Levin DL.

Int J Pediatr Otorhynolaryngal 1997; 39: 147-158. [abstract]

Reviewed by Barry P. Markovitz

Review posted August 25, 1997

---

I. What is being studied?:

The study objective (s):

1. To describe the epidemiology of extubation failure from postextubation stridor in PICU patients,
2. To assess the correlation between a stridor score and extubation failure, and
3. To assess the efficacy of dexamethasone prophylaxis to prevent secondary extubation failure (following initial failure from postextubation stridor and reintubation).

EB Journal Club editorial note: the analysis below focuses on this paper's third objective only.

The study design:

Prospective, randomized, controlled, double-blind trial

The patients included:

All patients in two PICU's who failed initial extubation for stridor

The patients excluded:

1. Patients who received steroids in the previous week, 2. medical contraindication to steroid use

The interventions compared:

Dexamethasone (0.5 mg/kg/dose IV, max. dose 15 mg) at 6 hours prior to planned extubation, just prior to and at 6 and 12 hours following extubation, vs. placebo (0.9% saline).

The outcomes evaluated:

Stridor score between treated and control groups. This score assigns 0 to 3 points for severity of stridor, and 0 to 2 points for cyanosis, sternal retractions, and age-adjusted respiratory and heart rates. The range of possible scores is then 0 to 11. (The study was designed to detect a 4 point difference in stridor scores between the groups.)

Extubation "failure" is assessed in the results, but no criteria for failure are given.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes.

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

Was followup complete?

Yes. Although the initial analysis excluded 3 patients who inadvertently received dexamethasone (on the order of another physician), an "intention to treat" analysis was also performed, including all patients.

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

Yes.

Secondary questions:

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

Yes. Identical appearing solutions of dexamethasone and saline were prepared by a pharmacist uninvolved with patient treatment. Nurses and physicians caring for the patients were unaware of the assignment.

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

Unknown. No comparison of treated and control patient data (age, sex, diagnosis, etc.) is presented. Furthermore, although patients selected were those "who failed extubation due to PES (postextubation stridor)," only 17% of patients enrolled had airway obstruction documented as the reason for reintubation, and it is unclear whether the groups were similar with regard to this important feature. Additionally, since specific criteria were not applied to determine "failure" and thus reintubation (following the first or subsequent extubations), the severity of airway obstruction may have been dissimilar in the two groups.

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

Apparently. All study patients had endotracheal tubes with air leak pressures between 15-25 cm H2O. If an air leak was not present, patients were electively reintubated with a smaller tube before the study extubation was attempted. The authors do not report how many patients underwent this manipulation, what the actual air leak values were, or under what conditions the air leaks were checked (sedated, neuromuscular blockade, or awake).

In addition, no mention of other aspects of patient care is reported, e.g., the use of sedative and/or muscle relaxants, total number of airway manipulations, etc. These factors may play a role in the development of postextubation stridor. There is also no mention of the use of aerosolized racemic epinephrine for the treatment of postextubation stridor.

III. What were the results?

1. How large was the treatment effect?

There was no significant difference in stridor scores following the study extubation between dexamethasone and placebo treated patients (5.083 vs. 6.182, p=0.60). There was, not surprisingly, a significantly higher stridor score in those patients that failed extubation (8.875 vs. 3.867, p=0.0027). These analyses are subject to question, however, for stridor scores are treated as continuous data; in fact they represent a ranked score, and arithmetic manipulation (reporting of means) should not have been performed (although the proper non-parametric Mann-Whitney rank sum test was used) .

Three of 12 dexamethasone treated patients failed extubation, while 5 of 11 controls failed (25% vs. 46%). This represents an absolute risk reduction (ARR) of 21%, i.e., 21% fewer dexamethasone treated patients failed extubation compared to controls. However, the 95% confidence intervals for this ARR are -18% to 60%.

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

Because of the small number of patients enrolled, the confidence intervals for the ARR cross zero, i.e., the true ARR might be anywhere between -18% (treated patients did worse) and 60%. The effect of the small sample size can be appreciated by observing the effect on the confidence intervals of a ten-fold higher number of patients. If the risk of extubation failure proved to be similar (25% of treated vs. 46% of controls) with 120 treated and 110 control patients, instead of 12 and 11 respectively, the 95% confidence intervals for the ARR narrow to 12%, or between 9% and 33%.

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

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

No. (see below)

2. Were all clinically important outcomes considered?

No. Dexamethasone is used to prevent extubation failure secondary to laryngeal or subglottic edema. Although extubation failure was defined as an objective of this study, there were insufficient numbers of patients to assess the impact of this therapy on extubation failure. Approximately 80 patients in each group would have been necessary to demonstrate a 20% difference in failure rates (from 40% in controls to 20% in treated patients) with alpha = 0.05 and beta = 0.20.

The potential side effects of dexamethasone, e.g., hyperglycemia, hypertension, gastrointestinal bleeding, were also not reported.

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

Unclear. This study does not adequately address either benefits or harms.

---

Related Reviews:

Tellez DW, Galvis AG, Storgion SA, Amer HN, Hoseyni M, Deakers TW. Dexamethasone in the prevention of postextubation stridor in children. J Pediatr 1991;118:289-294.

• Review, by A. Randolph

Anene O, Meert KL, Uy H, Simpson P, Sarniak AP. Dexamethasone for the prevention of postextubation airway obstruction: a prospective, randomized, double-blind, placebo-controlled trial. Crit Care Med 1996; 24: 1666-1669.

• Review, by A. Randolph

Comments

Back to the EB Journal Club Index