© 2004 American Thoracic Society
Weaning by ProtocolTo the Editor:The study by Krishnan and colleagues (1) recently published in AJRCCM suggests a weaning protocol (PW), which was proven to reduce duration of mechanical ventilation at one academic center (2), was no better than physician-directed weaning at a second center. The authors attribute this inconsistency to differences in intensive care unit (ICU) administration and manpower. It is possible that the authors are correct in this assertion, but for a different reason than they offer. It is clear from their results that nonphysicians complied highly with the PW. It is not clear, however, whether the physicians adhered to the demands of the protocol and may have impeded the prompt extubation of patients in the PW arm of the study.
There are several observations that suggest this may have been the case. First, the duration of spontaneous breathing trials (SBTs) was nearly twice as long in patients assigned to PW (3.0 vs. 1.6 hours, p < 0.01). Why was that so? The authors suggest that physicians, who were needed to order a trial of extubation following successful SBTs, were often "on rounds" during the times SBTs were performed by patients in the PW group: "Physicians on rounds, if interrupted, were also likely to postpone a decision if that patient had not yet been reviewed by the team" (1). The authors agree that physicians, who ultimately decided when to extubate, may have acted more rigorously in control patients: "In contrast, physicians were likely to review and promptly act on the results of a SBT that they had initiated in patients assigned to UC" (control group) (1). Accordingly, it is entirely possible that the reason no difference in weaning outcomes was found is because of such systematic bias. How could such bias have prolonged weaning in the PW group? If patients remained on excessively long SBTs some could have experienced fatigue (who would not necessarily have failed a trial of extubation) due to factors associated with the artificial airway, ventilator circuit, or both. Second, some patients may have passed the SBT at 30 minutes (when evidence would suggest that it is safe to perform a trial of extubation in successful patients [3]), but were never given a trial of extubation (without reason). Indeed, 65% of patients in the control group were extubated after their first SBT. This is consistent with other published series (4, 5). The PW group, on the other hand, was successful in only 54% of first attempts. It is possible that the PW group began trials earlier in the course of the patients' illnesses, due to more rigorous daily scrutiny, and therefore the first SBTs were unsuccessful more often due to insufficient resolution of respiratory failure. However, it is also possible that some patients were not successfully extubated because they did not receive a trial of extubation following a successful SBT. The authors can clarify this question if they offer how frequently patients had an SBT that was successful after 30 minutes but did not receive a trial of extubation that day. It is also worth noting that patients assigned to the control group were not constrained from performing SBTs as were patients in PW with Weaning protocols are created to obviate physicians' caprice in deciding when to perform SBTs, thereby detecting readiness for liberation on the earliest possible day. Accordingly, it is important to know whether the physicians in Dr. Krishnan's study contributed to lack of superiority of the PW. If, indeed, PW patients were not extubated because they were truly not responding to their SBTs, then perhaps more intensive physician participation in weaning in a closed ICU yields similar outcomes as PW. If not, this study may exemplify why PW is helpful.
Bridgeport Hospital and Yale University School of Medicine New Haven, Connecticut FOOTNOTES Conflict of Interest Statement: C.A.M. and Y.A.-A. do not have a financial relationship with a commercial entity that has an interest in the subject of this letter. REFERENCES
From the Authors: In our study (1), which showed that protocol-directed weaning did not hasten discontinuation of mechanical ventilation, we speculatively attributed this to physician assiduity in the usual care arm (1). Drs. Manthous and Amoateng-Adjepong offer several provocative alternative explanations that aggregate into two themes: that physicians systematically impeded weaning in the protocol arm, or that the protocol itself was flawed. In the protocol arm, spontaneous breathing trials that failed lasted 60 (45–79) minutes (median [interquartile range]). This does not support the speculation that their failure was due to excessive prolongation. Among patients in the protocol arm who passed their spontaneous breathing trial, one-third were not extubated the same day. This compares favorably with the study by Ely and colleagues (2) in which 45% of patients were not extubated the day they passed their spontaneous breathing trial. Of course, this includes patients in whom extubation was postponed because of rational clinical decisions. Although we lack comparable data for our Usual Care group, these data do not suggest that physicians subverted the protocol. Deficiency of the protocol is an argument without end, for "optimal" medical care will always be a moving target. We note that our protocol was very similar to others studied during the same time period (2–6), and it was more aggressive than that of Ely and colleagues (2). The institution of any protocol effectively freezes medical care in a form reflecting the state of knowledge at that time. The ability of caregivers to adapt their practice to evolving knowledge or individual patients is an agility that all protocols lack. Although some randomized trials of weaning protocols have found decreased duration of mechanical ventilation (2–4), others have not (5, 6), and some overall positive trials have found benefits limited to only some participating units (3, 4). We believe these protocols are just one of several tools that can speed recognition of patients' ability to breathe and may not be the right tool in all settings. We emphatically do not want our negative trial to be interpreted as condemnation of weaning protocols. When our study concluded, we reviewed its negative findings with our staff, expecting they would be relieved to jettison this extra effort. Instead, they voted unanimously to continue using the protocol (now modified to no longer require a f/VT measurement). They cited the satisfaction derived from active involvement in this aspect of care and how this facilitated dialog throughout the medical team. Thus, even a weaning protocol with no tangible benefits may yield intangible ones.
Johns Hopkins Medical Institutions Baltimore, Maryland FOOTNOTES Conflict of Interest Statement: H.E.F. and J.A.K. do not have a financial relationship with a commercial entity that has an interest in the subject of this letter. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
/VT greater than 105 breaths/minute/L and PaO2/FIO2 less than 200. Some patients with "unfavorable"