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

A comparison of two methods to perform a breathing trial before extubation in pediatric intensive care patients

Farias JA, Retta A, I. Alía, et al.

Intensive Care Med 2001; 27:1649-1654. [abstract; full-text for subscribers].

Commentary: Brochard L. Weaning from mechanical ventilation. When paediatric intensive care medicine profits from adult experience and vice-versa. Intensive Care Med 2001; 27:1564-1566L. [full-text for subscribers]

Reviewed by Ayse Akcan Arikan, MD Mona McPherson, MD, Texas Children's Hospital, Baylor Medical School, Houston TX

Review posted July 22, 2002

---

I. What is being studied?:

The study objective:

To compare two types of spontaneous breathing trials on patients who are on mechanical ventilation (T-piece vs pressure support ventilation of 10 cm H20) as a means to determine extubation failure (defined as reintubation within 48 hours).

The study design:

This is a multi-institutional, unblinded, randomized prospective clinical trial of two interventions: T-piece vs pressure support of 10 cm H2O

The patients included:

257 children who were admitted to three pediatric intensive care units over a 19 month period. Patients must have received mechanical ventilation for at least 48 hours to be included. Their primary physician had to judge them ready for a breathing trial and they had to satisfy the following criteria:

1. Age between 1 month and 15 years
2. Improvement or resolution of the underlying respiratory problem
3. Evidence of adequate gas exchange as demonstrated by PaO2 > 60 mm Hg on FiO2 of .4 or less and PEEP of 5 cm H2O or less
4. Afebrile (Core temperature < 38.5 C)
5. Alert after discontinuation of sedatives
6. Hemoglobin > 10 gm/dl
7. No further requirement for vasoactive agents

The patients excluded:

Patients who were tracheostomized or who had audible leaks around endotracheal tubes were excluded from the study

The interventions compared:

Two hour trials of T-piece breathing versus two hour trials of pressure support of 10 cm H2O. When patients were judged ready for a breathing trial, mechanical ventilation was stopped and respiratory function indices were measured while the patients breathed spontaneously for 5 minutes on a T-piece. Tidal volume and frequency to tidal volume ratio were calculated and recorded. The patients were then randomly assigned to one of the two types of spontaneous breathing trial. One group underwent a breathing trial on a T-piece and the other group on pressure support of 10 cm H2O. The breathing trials in both groups lasted up to 2 hours. Patients in the pressure support group could also be placed on PEEP of up to 5 cm H2O.

The trial was stopped by the primary physicians if any of the following occurred: a) the respiratory rate or the heart rate were higher than the 90th percentile for age, b) the blood pressure was lower than the third percentile for age, c) oxygen saturation on pulse oximetry was lower than 90%, d) pCO2 was above 50 mm Hg or the pH was lower than 7.30 on arterial blood gas measurement, e) increased work of breathing, f) diaphoresis and anxiety, or g) change in mental status.

The outcomes evaluated:

The main outcomes measure used to compare the success of the breathing trials was extubation failure within 48 hours. Secondary endpoints included length of stay in intensive care, length of stay in hospital and in-unit and in-hospital mortality rates. The authors also found reintubation was strongly associated with increased mortality but this seems to be a post hoc finding.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The patients were assigned to different breathing test groups using a random number table.

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

Was followup complete?

Yes. Although, there is a discrepancy (of 3 patients) in numbers of patients mentioned in the text and the numbers of patients given in Table 3. We assume that the authors have left out a subgroup comprised of patients requiring mechanical ventilation secondary to neuromuscular disease when constructing the table. Therefore, we have based our calculations on the numbers given in the text.

Yes. All patients who were included in the study were followed until they were discharged from the intensive care unit and until they were discharged from the hospital. No patient was lost to follow-up.

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

Yes, there was no cross-over between the two test groups.

Secondary questions:

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

No. The patients were not blinded; however, this likely did not impact the study results. The respiratory personnel who were doing the measurements for respiratory functional indices were not blinded. Although the attending physicians responsible for the primary care of the patients were blind to the results of these data, they were not blinded to different breathing trials since the T-piece is a visibly different apparatus. This creates a possibility for bias since these physicians were responsible for stopping the breathing trial based on both objective and subjective criteria. However, blinding physicians to the trial would be quite impractical if not impossible.

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

Not quite. Both groups were similar in terms of gender, duration of ventilator support before weaning, reason for initiation of mechanical ventilation, and endotracheal tube sizes. In patients requiring mechanical ventilation secondary to acute respiratory failure, causes for acute respiratory failure were also similar in both groups. However, patients who were randomized to T-piece were both younger and physically smaller than the patients in the pressure support group. (Both of these factors could easily be associated with extubation failure since younger kids may be more prone to atelectasis secondary to having more compliant chest walls.) Moreover, patients in the pressure support group had a tendency toward higher PRISM scores (p=0.07) making them arguably sicker.

In terms of respiratory function indices measured before the breathing trial was started, patients in both groups were similar in PaO2/FiO2 ratio, respiratory rate, and maximal inspiratory pressure. However, patients in the pressure support group had a tendency toward higher tidal volumes and lower frequency to tidal volume ratio. Again, these differences were not statistically significant.

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

Unknown. The authors did not provide other treatment or intervention information.. The authors also didnÕt mention if any type of non-invasive ventilatory support was provided after extubation.

III. What were the results?

1. How large was the treatment effect?

Of the 132 patients in the T-piece group, 102 patients were extubated at the completion of the trial (22.7% "failed" the trial and were not extubated) but 13 patients required reintubation in 48 hours. Of the 125 patients randomized to the pressure support group, 99 were extubated (20.8% "failed "the trial and were not extubated) after completing the trial but 15 patients were reintubated in the next 48 hours. The ratios were 67.4 vs 67.2% for successful extubation (of the total groups) and 12.7 vs 15.1% for reintubation (of those extubated) in the T-piece and pressure support groups respectively. None of the differences reached statistical significance.

If we treat these two tests as predictors of successful extubation , then the positive predictive value for successful extubation after a successful pressure support trial would be 84.8% and 87.3% for a T-piece trial; there is no significant difference between the two groups.

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

Not applicable.

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

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

Yes and no. This particular study was done on a group of patients with a relative short course of mechanical ventilation and ICU stay. The mean duration of mechanical ventilation was 6 days and ICU stay was around 10 days. These findings would likely not be applicable to patients requiring long courses of mechanical ventilation with resultant muscle weakness. The majority of patients required institution of mechanical ventilation secondary to acute respiratory failure (73% in both groups) and only 13% of the patients in either group had acute exacerbation of chronic respiratory failure. In our experience, many patients have chronic diseases and tend to require longer courses on the ventilator. It can be expected that these patients might require increased support for weaning, i.e., increased pressure support, and may not do so well with a breathing trial on the T-piece as many require some type of non-invasive pressure after extubation. However, in a subgroup of patients with acute respiratory failure who require mechanical ventilation for a considerably short period of time the results of this study could be useful.

2. Were all clinically important outcomes considered?

Yes. Clinically important outcomes such as success in breathing trial, reintubation within 48 hours, length of ICU stay, length of hospital stay, in-unit mortality, and in-hospital mortality have all been compared between two groups and found not to be significantly different. However, performing the breathing trial on the T-piece is marginally more costly secondary to the requirement for extra apparatus whereas pressure support can be performed easily on the same ventilator without any extra intervention. For units with mechanical ventilators not capable of pressure support ventilation, these data may be reassuring that the T-piece trial is as good (or as bad) a predictor of extubation success as PSV trials.

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

If we assume that the treatment is the pressure support group and T-piece breathing trial is serving as the control group, then the answer is yes. Since the pressure support breathing trial seemed to be as effective as the T-piece trial in predicting who would remain successfully extubated and there was no apparent harm in using pressure support, it could be more cost-effective to use pressure support to predict readiness for extubation if the patient is already on a ventilator capable of generating pressure support so a separate apparatus or expertise would not be required.

Comments

Back to the EB Journal Club Index