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 Prospective, Controlled Trial of a Protocol-based Strategy to Discontinue Mechanical Ventilation.

Krishnan JA, Moore D, Robeson C, Rand CS, Fessler HE.

Am J Respir Crit Care Med 2004;169 673-678 [abstract]

Reviewed by Anthony Lee and M. Nilufer Ozturk, Children's National Medical Center, Washington DC

Review posted September 27, 2004

---

I. What is being studied?:

The study objective:

The primary objective of this study was to evaluate whether a protocol for discontinuation of mechanical ventilation (based on previously-studied strategies) shortens the duration of mechanical ventilation (MV) compared to a standard approach.

The study design:

This was a prospective, controlled study in a closed, academic, intensivist-run medical ICU with high physician staffing levels and structured, system-based rounds. This study was powered to detect a 1-day difference in MV duration between groups.

The patients included:

All patients requiring invasive ventilation for 24 hours or more admitted between April 2000 and July 2001 were eligible, if they did not meet exclusion criteria. Out of 749 ICU admissions, 450 were excluded (60%) and 299 were included in the study. 154 were assigned to protocol wean (PW) and 145 were assigned to usual care (UC) groups.

The patients excluded:

• No MV (n=320).
• MV < 24 hours (n=73).
• Previous participants (n=4).
• Patient who were already enrolled in other studies that controlled weaning (ARDS network studies) (n=10).
• Patients who were transferred from other institutions already intubated (n=33).
• Although not stated as part of the exclusion criteria, patients could have been denied MD consent (n=6).
• Patients missed by study staff (n = 4).

The interventions compared:

In this trial, the protocol used was based on the weaning protocol used by Ely et al., previously shown to reduce duration of ventilation (1). The protocol consisted of daily screen for readiness. Readiness for spontaneous breathing trial (SBT) was determined by exclusion of the following criteria:

1. Known or suspected elevation of ICP
2. Unstable coronary artery disease
3. HR ≥ 140 bpm
4. Prohibition of the weaning screen by the physician
5. SpO2 < 92 %
6. PEEP > 5 cm H2O
7. FiO2 > 0.5
8. Receiving paralytics
9. Absent cough and gag reflex
10. Unresponsiveness to noxious stimuli

If the patient passed the screening, the ratio of respiratory frequency to mean Vt (f/Vt) was measured.

If f/Vt ≤ 105, a spontaneous breathing trial (SBT) was initiated with CPAP 5, PS 5 if ETT < 7mm. The patient was returned to the prior ventilator settings if any of the following occurred within 1 hr of SBT:

1. HR > 20 bpm above rate before SBT, persisting > 5min
2. SBP < 90 torr or > 30 change after SBT initiation, persisting > 5min
3. Chest pain or ECG changes (ischemia or new arrhythmia)
4. SpO2 < 88%, or PaO2 < 60 torr, persisting > 5 min
5. Marked distress, dyspnea, or agitation.

If SBT was tolerated for one hour, the physician was notified about tolerance. If not, the patient was rested for 24 hrs until next day. Physicians could also extubate PW patients based on their clinical judgment.

For patients randomized to UC, decision about the method and timing of discontinuation of MV was entirely left to MD. No scheduled screening was performed by the ancillary staff for UC, but f/Vt measurements could be requested or checked by the physician. Physicians specified the ventilator settings and the time of the SBT beginning and end with an individual order.

The outcomes evaluated:

Primary outcome:

1. Difference in the duration of MV between PW and UC groups (defined as time in hours, from the start of the mechanical ventilation to the beginning of SBT that ended with successful discontinuation of MV). Greater than 24 hrs was considered significant. Successful discontinuation was defined as being able to breathe unassisted for 48 hours.
2. Difference in the rates of failed extubations between PW and UC groups.

Secondary outcomes:

1. Duration of SBT
2. Reinstitution of MV less than 48 hr, > 48 hr
3. ICU length of stay
4. Location after ICU discharge
5. Hospital deaths

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

No. The assignment method was suboptimal because of lack of concealment. Assignment was based on the medical record number being even or odd, therefore patients were already "randomized" before they were screened for eligibility. Lack of concealment theoretically may destroy balance in prognostic factors that the investigators strive to achieve through randomization and could potentially induce selection bias (The authors admit this, calling the study "quasi-randomized" in the discussion section).

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

Was followup complete?

Yes. All patients enrolled were followed to hospital discharge.

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

Yes. The data were analyzed with an intention to treat. Although 10 of the 154 patients in PW group were withdrawn by treating physicians, all 154 were included in the analysis All 145 patients in the UC group were included in the analysis.

Secondary questions:

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

The authors admit that the trial was unblinded and, therefore, subject to a change in practice because physicians, nurses, and respiratory therapists knew the trial was underway, i.e., the Hawthorne effect. The authors try to counter this limitation by stating that they did not find evidence of aggressive MV weaning during the 50 charts reviewed in the UC group.

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

There were no statistically significant differences between treatment groups in regards to in baseline characteristics, related to age, gender, ethnicity, severity of illness scores (SAPS II), oxygenation indices (PaO2/ FiO2) first 24 hrs of ICU stay, or presence of chronic respiratory disease.

Patients assigned to PW were less likely to be admitted from another ICU (7.1 vs. 12.4%), hospital (2 vs. 5.5%), or nursing home (1.3 vs. 2.1%) and were more likely to have pneumonia/ALI on initiation of MV (33.2 vs. 21.4%). Overall differences between groups in source of admission and reason for initiating MV were not significantly different between the two groups (p=0.27 and p=0.29, respectively).

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

Yes and no. There is no mention of other, non-MV-related aspects of patient care in the article, but the on-line data supplement states that other aspects of medical care were left to physician discretion in both groups. As a part of the routine care, practice standards in use included protocols controlling electrolyte repletion, vasopressors and sedation, the latter titrating midazolam and/or fentanyl to a goal based upon modified Ramsey scale. However, there are a couple points which are important to mention:

1. There was a longer interval between the initiation of the final SBT and extubation in the patients weaned by protocol. Physicians were likely to review the results and promptly act on the results of SBT that they had initiated in their patients assigned to UC. The duration of SBT preceding successful discontinuation of MV was longer among the patients assigned to PW compare to UC. (median duration [IQR] 3 hrs [1.3-5.6] vs 1.6 hrs [0-3.9], p < 0.01) The authors accounted for this difference by describing the time differences in the usual completion of the SBT. They mention there likely was a reluctance to interrupt physician rounds for an order to discontinue MV and also an increased likelihood that physicians would want to evaluate a patient on rounds before deciding to discontinue MV.

2. The physicians could extubate the PW patients based on their clinical judgment, presumed to be after completion of the SBT (it's not quite clear from the authors). The therapy group has to rely on how well the criteria were applied, which is the protocol adherence. The criteria for screening and performing a SBT were clear. However, if a physician's "clinical judgment" was used for determination of extubation, these additional criteria were not specified.

3. It is not clear from the article whether UC was very different from protocol based care. Although no scheduled screening tests were performed, the providers seem to have free access to this information (like f/Vt measurements). We do not know how often it was used in the UC group, and what the results were. The most important point we learned from previous adult studies which showed that a shorter weaning duration is probably achieved by the evaluation of spontaneous breathing, based on some objective criteria, such as a SBT trial. In this study, the UC group also used this evaluation tool, which suggests that this was already implemented in the UC group. The difference was primarily whether the challenging occurred based on a protocol performed by respiratory therapists, rather than upon a physician order.

III. What were the results?

1. How large was the treatment effect?

This study showed no difference in duration of MV weaning in PW and UC (p=0.61). The median values [interquartile ranges] for PW and UC groups were 60.4 hours [28.6-167.0] vs. 68.0 hours [27.1-169.3]. The sample size provided 82% power to detect a one-day difference in MV duration (assuming two sided type I error = 0.05).

There was also no difference in successful extubation rate (PW 74.7%, UC 75.2%, p=0.92), ICU mortality (PW 25.35, UC 28.3%), hospital mortality (36.6% vs 33.1%), ICU length of stay (median 115hrs [IQR 67-259] vs 146 [81-291], p=0.10, or rate of reinstituting MV (10.35 vs 9%).

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

Confidence intervals for the differences between the groups are not provided. The authors mention interquartile ranges, which is a measure of dispersion describing the difference between the 25th and 75th percentiles. Broad overlap of the interquartile range suggests a lack of meaningful difference between the two groups.

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

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

Yes and no. No benefits to the protocolized approach were demonstrated in this study compared to the usual physician-directed approach. The authors mention high staffing ratios as a possible explanation for their findings. In the earlier studies by Esteban et al (2), and Brochard et al (3) using a protocol based approach, despite the presence of apparent disease stability/reversal prior to performing a SBT, the managing physicians did not recognize that discontinuation was feasible in 2/3 of the patients. It was a learning experience for medical personnel. The task force guidelines for weaning and discontinuing ventilatory support by ACCP, AARC and ACCCM recommends performing SBT based on level A evidence (4). Since this study used SBT's in both the PW and UC group, we are not sure whether checking f/Vt would have been important, even if there was a difference shown in this study. The ability of traditional weaning indices like f/Vt to discriminate between children successfully extubated and children re-intubated has been shown to be poor (5).

The largest question regarding applicability of these results to our patients, of course, relates to the patient population - this was an adult study and its applicability to children is uncertain. The lack of difference between UC and PW in regards to successful extubation rates and duration was shown in the randomized controlled study of children by Randolph et al., although the protocol adherence in this study was 66%. This study also had extensive exclusion criteria (6).

Last but not least, the reasons for the failed extubations would be helpful to know. The largest study in pediatrics done by Kuracheck et al. (7) revealed that extubation failure was due to upper airway obstruction in 37% of the cases. Unfortunately, there are no good predictors to foresee that especially in younger children during weaning.

This study does not tell us that we do not need protocols, but rather that we may not need them "if they merely codify a set of behaviors already in use". Most ICU's do not have physicians circling through the unit challenging their patients based upon formal algorithms as reliably as RT's might be able to do.

2. Were all clinically important outcomes considered?

Yes.

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

Not applicable, since there were no treatment benefits demonstrated in this study. It could be inferred that the study supports the notion that adequate staffing - medical, nursing or respiratory therapy - are necessary to wean patients from mechanical ventilation efficiently, with or without guidelines.

References

1. Ely EW, Baker AM, Dunagan DP, et al. Effect on the duration of mechanical ventilation of identifying patients who are capable of breathing spontaneously. N Eng J Med 1996;335:1864-1869 [abstract]
2. MacIntyre NR, Cook DJ, Ely EW Jr, Ebstein SK, Fink JB, Heffner JE, et al. American College of Chest Physicians; American Association for Respiratory Care; American College of Critical Care Medicine. Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians; American Association for Respiratory Care; Chest 2001; 120(6 Suppl) p375S-95S. [citation]
3. Brochard L, Rauss A, Benito S, et al. Comparison of three methods of gradual withdrawal from ventilatory support during weaning from mechanical ventilation. Am J Respir Crit Care Med 1994; 150: 896-903. [abstract]
4. Esteban A, Frutos F, Tobin MJ, et al. Comparison of four methods of weaning patients from mechanical ventilation. N Eng J Med 1995; 332: 345-350. [abstract]
5. Farias JA, Olazarri F RA, Fernandez A et al. An evaluation of extubation failure predictors in mechanically ventilated infants and children. Intensive Care Med 2002; 28(6): 752-7. [abstract]
6. Randolph AG, Wypij D, Venkataraman ST, Hanson JH, Gedeit RG, Meert KL, et al. Effect of mechanical ventilator weaning protocols on respiratory outcomes in infants and children: A randomized controlled trial. JAMA 2002; 288:2561-2568. [abstract] [PedsCCM EBJC Review]
7. Kurachek SC, Newth CJ, Quasney MW, Rice T, Sachdeva RC, Patel NR, et al. Extubation failure in pediatric intensive care: A multiple- center study of risk factors and outcomes. Crit Care Med 2003; 31 (11):2657-2664. [abstract]

---

Related Reviews

Osmond MH, Klassen TP. Efficacy of ipratropium bromide in acute childhood asthma: a meta-analysis. Acad Emerg Med 1995; 2:651-656. [abstract]

• Review, by A. Torres

Comments

Back to the EB Journal Club Index