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

Effects of a multifaceted, multidisciplinary, hospital-wide quality improvement program on weaning from mechanical ventilation

Smyrnios NA, Connolly A, Wilson MM, et al.

Crit Care Med. 2002;30(6):1224-30. [abstract; full-text for subscribers]

Reviewed by Satid Thammasitboon, MD, MHPE, Pediatric Critical Care Medicine, Baylor College of Medicine, Houston, TX, and Supat Thammasitboon, MD, Pulmonary and Critical Care Medicine, Tulane University, New Orleans, LA

Review posted September 6, 2002

---

I. What is being studied?:

The study objective:

The objective of this study is to examine the effects of mechanical ventilation (MV) weaning management protocol that was implemented as a hospital-wide, quality improvement program on clinical and economic outcomes.

The study design:

This is a prospective, before-and-after intervention study. Data from a preimplementation year are compared with those of the two-year period after protocol implementation.

The patients included:

1. Patients who had received 2 weeks of MV
2. Patients who received MV for 7 days after acute surgical issues were resolved
3. Patients who had failed to wean after 72 hours of attempts

For data analysis, they evaluated hospital administrative data on patients from two Diagnosis Related Groups (DRGs); respiratory failure with mechanical ventilation and tracheotomy (not related to airway disease).

The patients excluded:

There were no exclusion criteria.

The interventions compared:

1. MV weaning management in a preimplementation year: Weaning was performed in the manner prescribed by individual attending physician. There was no standardization of care throughout one year.
2. MV weaning management protocol: It was a hospital-wide, quality improvement program using a single mode of weaning, i.e., CPAP.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

No. Since this is a before-and-after intervention study, randomization was not applicable.

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

Was followup complete?

Complete data from 99.5% of patients included in the study were available. However, we do not know how long they follow the patients after discontinuation of MV. Thirty-three percent of patients in the baseline year were reintubated. Would the patients be considered having failure of MV discontinuation if the reintubation occurred 2 days afterward? It was not mentioned in the article.

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

There was no randomization. Patients were analyzed in the groups of baseline and post-intervention. No cross-over possible in this type of study design.

Secondary questions:

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

No. Blinding is not applicable in this study design. As a multidisciplinary, hospital-wide quality improvement program, all personnel must be informed to assure the success of the program.

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

No. The total number of patients in nontrached group increased from 86 to 156 accompanied by a decrease in the trached group from 134 to 115 (p < 0.0005). Although it could be favorable outcome of the weaning protocol, this significant difference might also affect the ICU length of stay. It was probably easier to discontinue MV on patients with tracheostomy than without for several reasons (e.g., effectiveness of bronchopulmonary toilet).

There were more patients with respiratory failure attributable to pneumonia in year 2 of intervention (32% vs. 75%, p < 0.0005). This was consistent with the increase in severity of the illness measured by APACHE II (22.2 ± 7.4 vs. 26.2 ± 7.8, p = 0.006). These characteristics might put patients in higher risk to have prolonged mechanical ventilation and failure of weaning from MV. These trends, however, could be expected to reduce the impact of a weaning protocol, so the effect might have been greater if the patient population had not changed.

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

It is unlikely that patients would receive equal treatment aside from the experimental intervention. During the course of quality improvement program, there were several educational sessions conducted to facilitate the program. Theses might enhance overall quality of patient care. The multidisciplinary approach was also emphasized as an essential principle of the program. Patients would have received better care just because the enthusiasm and motivation among participants. This is known as the "Hawthorne effect."

In addition, there have been multiple studies on new strategies involving MV management published during this 3-year study period. Theses well-received evidences could have significant impact on MV management and the way we treated patients with respiratory failure overall. Indeed, this is one reason why this study design is inherently weaker, as many other aspects of care that impacted outcome could have changed as well during this time period.

III. What were the results?

1. How large was the treatment effect?

(All the analyses below include all patients, combining both DRGs - of trached and non-trached patients. The data on each DRG cohort alone did not demonstrate statistically significant changes over time.)

a. Clinical outcomes

Although only tracheostomy and reintubation rate were proved statistically significant reduction, the hospital mortality and destination of discharge deserve attention with the p-values close to significant levels.

	Year 0	Year 1	Year 2	p-value
Hospital mortality (%)	32	29	28	0.06
Discharge to home (%)	16	17	19	0.06
Tracheostomy (%)	61	50	42	< 0.0005
Reintubation (%)	33	36	26	0.039

	Absolute Risk Reduction (RRR)	95% CI	Number needed to treat (NNT)	95% CI
Hospital mortality rate reduction	32-28 = 4%	-4.16,12.16	25	8.2, inf.
Tracheostomy rate reduction	61-42 = 19%	10.28, 27.72)	5.3	5.3, 7.7
Reintubation rate reduction	33-26 = 7%	-1.12, 15.12	14.3	6.6, inf.

b. Resource utilization and financial outcomes

	Year 0	Year 1	Year 2	p-value*
Ventilator days (mean ± SD)	23.9 ± 22	21.9 ± 25.8	17.5 ± 17.2	0.004
LOS-hospital (mean ± SD)	37.5 ± 31.6	31.6 ± 30.7	24.7 ± 20.6	< 0.0005
LOS-ICU (mean ± SD)	30.5 ± 28.7	25.9 ± 28.9	20.3 ± 18.5	< 0.0005
Total cost/case	92,933	78,624	63,687	< 0.0005

*p-value for trend year 0 through 2

The magnitude of improvement in the study was greater than any previously reported and maintained for a longer period than previously reported (days vs. hours reduction) (1, 2). The explanation for this finding could be the difference in patient characteristics. This study focused on a group of patients who had very prolonged course of MV at the beginning of the study. The degree of reduction would be more consequently. It is curious that these significant improvements were not seen when the two groups (trached and non-trached patients) were evaluated separately.

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

Not very precise. The standard deviation of each variable was very wide. The 95% CI of significant findings (tracheotomy rate reduction) were also wide.

3. Did the treatment effect differ in subgroups of patients?

As noted above, the effect of the intervention did not demonstrate any significant improvement in two groups when the data were analyzed separately. Only hospital length of stay was found to be significantly lower after intervention in trached group when the data was analyzed separately.

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

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

The concept of this MV management protocol could be applied to pediatric intensive care unit. The principle of using a single mode of weaning, involving fewer decision points should be beneficial for the patients and for trainees (residents, fellows, and respiratory therapists) since they could develop expertise in MV weaning. The identification of objective reason for weaning failure is crucial and need to be done early and systematically. I believe this was the key element of this successful protocol. Caution is warranted, of course, because of the different acute and chronic causes of respiratory failure between adults and children. For example, it is not clear that most children requiring prolonged ventilation have underlying respiratory muscle inadequacy as the primary reason.

The actual protocol requires significant modification. Further study on protocol development for pediatric patients could use data from multiple studies conducted previously on evaluating predictors of weaning from MV in pediatric patients (3, 4).

2. Were all clinically important outcomes considered?

The study was carefully designed to obtain all important clinical outcomes. Resource utilization and financial outcomes were also included providing interesting findings and interpretation.

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

The initial phase of program implementation might require some extra cost. This should be cost effective program from the data provided in the study. There should not be any potential harm from this protocol. Some physicians might not want to loose their autonomy in MV management. High level of physician acceptance is a prerequisite for a successful implementation.

References

1. Kollef MH, Shapiro SD, Silver P, et al. A randomized, control trial of protocol-directed versus physician-directed weaning from mechanical ventilation. Crit care med 1997; 25: 567-574 [abstract]
2. Marelich GP, Murin S, Battistella F. Protocol weaning of mechanical ventilation in medical and surgical patients by respiratory care practitioners and nurses. Chest 2000: 118: 459-467 [abstract]
3. Baumeister BL, El-Khatib M, Smith PG, et al. Evaluation of predictors of weaning from mechanical ventilation in pediatric patients. Pediatric Pulmonology 1997; 24: 344-52. [abstract] [PedsCCM EB Journal Club Review, by S. Venkataraman]
4. Farias JA, Alia I, Esteban A, et al. Weaning from mechanical ventilation in pediatric intensive care patients. Intensive care med 1998; 24: 1070-1075. [abstract] [PedsCCM EB Journal Club Review by P. Jines, A. Madikians]

Comments

Back to the EB Journal Club Index