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

Effect of Exogenous Surfactant (Calfactant) in Pediatric Acute Lung Injury: A Randomized Controlled Trial.

Willson DF, Thomas NJ, Markovitz BP et al.

JAMA 2005;293 470-476 [abstract]

Reviewed by Angela Shieh Czaja MD, Children's Hospital & Regional Medical Center, University of Washington, Seattle, WA

Review posted July 30, 2005

---

I. What is being studied?:

The study objective:

To determine if endotracheal instillation of calfactant in infants, children and adolescents with Acute Lung Injury (ALI) would shorten the course of respiratory failure.

The study design:

A multi-center, randomized, double-blind, placebo-controlled trial

The patients included:

A total of 152 patients from 21 pediatric ICUs were enrolled and randomized. Entry criteria included: (1) ages 1 week to 21 years, (2) radiologic evidence of bilateral lung disease, and (3) oxygenation index (OI) of greater than 7. Enrollment was initially required to be within 24 hours of endotracheal intubation. However, after the first 50 patients, this was subsequently extended to 48 hours, presumably due to slow recruitment.

The patients excluded:

The exclusion criteria were prematurity (corrected gestational age < 37 wks), status asthmaticus, head injury with Glasgow Coma Scale < 8, chronic lung disease (defined as home oxygen or chronic diuretic use), brain death, do not resuscitate orders, ongoing CPR, life support limitations, significant airway disease that might delay extubation, uncorrected congenital heart disease, pre-existing myocardial dysfunction, or cardiogenic pulmonary edema.

The interventions compared:

The experimental group received 80 ml/m2 calfactant via intratracheal instillation. The control group received an equal volume of air placebo. For infants weighing less than 10 kg, 3 ml/kg doses were administered. A second and final intervention (in both groups) was performed 12 ± 2 hr later if the OI remained higher than 7.

The outcomes evaluated:

The primary outcome evaluated was the duration of respiratory failure measured by ventilator-free days at 28 days. Secondary outcomes were PICU and hospital lengths of stay, hospital charges, duration of supplemental oxygen therapy, adverse events, and failure of conventional mechanical ventilation (the use of high-frequency oscillatory ventilation, inhaled nitric oxide, or extracorporeal membrane oxygenation).

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. Patients were centrally randomized and stratified to balance the severity of lung injury between study groups at entry using blocks of 2 and 4. The randomization scheme used was effective in generating two similar groups. Of the 152 patients enrolled, 77 were randomized to receive calfactant and 75 were randomized to receive placebo.

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

Was followup complete?

Yes. All 152 patients entered were included in the final analysis. In evaluation of the primary outcome, all entered patients were followed for 28 days post-study entry or until death. For secondary outcomes, patients were followed until hospital discharge.

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

Yes. There was no cross-over between treatment groups. Except for one patient that withdrew consent prior to treatment, all patients entered in the study were included in an intention-to-treat analysis. Eight patients with protocol violations (equal occurrences in each group) were analyzed in the groups they were randomized.

Secondary questions:

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

Yes. A pharmacist drew an opaque envelope from the next appropriate fast or slow entry file which had been previously randomized centrally in blocks of 2 and 4. An opaque container with syringes of either calfactant or placebo were sent to the PICU. A respiratory therapist not involved with care of the patient then placed opaque tape on the endotracheal tube and performed the intervention. The investigators as well as the physicians and nurses caring for the patients remained blinded throughout the study.

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

Yes, there were no significant differences between groups in the following characteristics: a) demographic profile (age, sex, race), b) severity of illness at randomization (PRISM score, mean initial oxygenation index, mean initial PaO2/FiO2, and the number of patients in fast entry [OI of 13 or higher] vs slow entry [OI higher than 7 but less than 13]), c) co-existing diagnoses or comorbidities.

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

Yes and no. The ventilator management guidelines used limited tidal volumes to less than 8 ml/kg, FiO2 of less than 0.6, peak inspiratory pressure (PIPs) of less than 40 mm Hg, and PaCO2 in the range of 40-60mm Hg. Adherence to these guidelines was similar between both groups, with FiO2 and PIPs being within guidelines greater than 90% of the time and PaCO2 within guidelines greater than 80% of the time. The clinical care team, however, determined all other aspects of the patient care. Escalation of therapy was left to bedside clinicians because the authors did not define conventional mechanical ventilation failure a priori.

III. What were the results?

1. How large was the treatment effect and how precise were the results?

1. Mortality: significantly greater in the placebo group (27/75) vs. the treated group (15/77) with a relative risk of 1.9 and a 95% confidence interval (CI) of 1.1 to 3.2. This gives an absolute risk reduction (ARR) of 17% (CI 2, 29) and the number needed to treat (NNT) as 6 (CI 3, 43). Subgroup analysis of infants younger than 12 months (26% of the population studied) showed that mortality was also greater in the placebo group with an relative risk of 5.4 and a 95% confidence interval 1.2 to 22.8. Thus, the ARR for infants was 33% (CI 4, 52) and the NNT was 3 (CI 1.9, 21.5). However, the use of a subgroup analysis needs to be interpreted with caution given that the study is already underpowered (for the primary outcome) when including all patients.
2. Ventilator-free days at 28 day evaluation: Calfactant group had a mean of 13.2 (standard deviation of 10) and placebo group had a mean of 11.5 (standard deviation of 10.5) with p=0.21. Note that death or need for extracorporeal membrane oxygenation were considered equivalent to no ventilator-free days. Subgroup analysis of infants younger than 12 months showed that the treatment group had statistically significant more ventilator-free days (15.2 days [SD 10.3] vs 7.0 days in the placebo group [SD 9.9], p=0.01).
3. Oxygenation as measured by OI: After the first intervention, mean OI after 12 hours was 15 with SD 14-16 in the placebo group and 14 with SD 13-15 in the calfactant group, p=0.01. After the second intervention (given to 70% of the calfactant group and 79% of the placebo group), mean OI after 12 hours was 16 with SD 15-17 in the placebo group and 13.5 with SD 12.5-15, p=0.02 (exact numbers not reported, estimated from graphic appearance) After the first intervention, the change in OI ranged from 9.8-15.5 in the calfactant group vs 9.0-16.2 in the placebo group. After the second intervention, the change in OI ranged from 6.7-10.0 in the calfactant group vs. 7.6-8.6 in the placebo group.
4. Conventional mechanical ventilation failure: 21% in the calfactant group and 42% in the placebo group p=0.02. This means an absolute risk reduction of 21% (CI 7.2, 34.8) and the number needed to treat is 4.7 (CI 3, 14). Of concern, in interpreting these results, is the lack of protocol definition for "conventional mechanical ventilation failure." This leaves much interpretation on the individual clinician's side as to when to switch modes of ventilation or add other therapies that could effect oxygenation such as nitric oxide.
5. Length of stay in the PICU and hospital, length of oxygen requirement and hospital changes: no statistical differences seen.

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

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

Yes, the range of ages and types of patients are commonly seen in the pediatric ICU.

2. Were all clinically important outcomes considered?

The trial examined many of the clinically important outcomes. There were other secondary outcomes that could have been included as well such as numbers of organ failure, incidence of chronic lung disease, tracheostomy with or without home ventilation, functional status, disposition to home or nursing facility, etc.

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

The complications seen were more common in the calfactant patients. Hypotension was seen in 9% of calfactant patients and 1% of placebo patients (p=0.005), all of whom responded to volume. (Absolute risk increase = 8%, number needed to harm = 12.5) Transient hypoxia was seen in 12% of calfactant patients vs 3% of placebo patients (p=0.008) but resolved with slowing of the instillation and/or positive pressure ventilation was transiently increased. (Absolute risk increase = 9%, number needed to harm = 11) There was no statistically significant difference in incidence of air leaks or nosocomial pneumonias. No patients were withdrawn from the study due to complications. The study did not look specifically at a cost analysis.

The number needed to treat to effect mortality was 6 with a fairly wide 95% confidence interval (3-43). However, as the authors discuss, there was a larger number of immunocompromised patients in the placebo group compared to the treatment group (30 vs 22). Given the higher mortality rate in immunocompromised patients, this discrepancy may have influenced the difference in mortality seen. When adjusted for immunocompromised patients, the effect on mortality may become statistically insignificant with an OR was 2.11 with 95% CI of 0.93-4.95, p-value 0.07.

There was no statistically significant difference in ventilator-free days between treatment and placebo groups. It is important to note, however, that they did not meet their goal enrollment of 300 patients which was originally calculated to detect a 25% decrease in the 13 day average ventilator course with an alpha level of 0.05 and beta level of 0.10. Therefore, the study was underpowered so that we don't know if we might have seen a difference in ventilator-free days if there had been 300 patients enrolled rather than the 153 actually enrolled. Although there was a statistically significant difference seen with infants, the meaning of a subgroup analysis with an already underpowered study can become difficult to interpret.

In addition, when immunocompromised status is controlled for, the treatment effect on mortality just loses significance (lower CI crosses one), further clouding the generalizability of this trial.

Given the low incidence of side effects without major adverse outcomes, the treatment benefits likely outweigh the potential harms. This treatment may not improve outcomes (mortality and ventilator-free days) in every patient with ALI but there may be certain patients that could benefit from calfactant (number needed to treat = 6, although it could be higher). Unfortunately, until a larger study can be performed, we don't know which patient would benefit and which would not. Given the evidence from this article, the use of calfactant, if readily available, may be considered early in the treatment of pediatric ALI.

Comments

Back to the EB Journal Club Index