Chapter 65: Controversies: Ventilator Weaning—Which Strategy is Better? RT-RN Versus Physician Driven

The advent of invasive mechanical ventilation (IMV) has been one of the greatest achievements in the field of respiratory and critical care medicine. In the United States alone, there were 790,257 hospitalizations involving IMV in 2005, and that number is expected to steadily increase, outpacing population growth by 2026.^1,2 Although advances in medical technology have facilitated the management of patients who require ventilatory support, mechanical ventilation remains associated with risks and complications.³ Thus, once clinical improvement occurs, significant emphasis is placed on rapidly weaning the patient from IMV. Over the past 25 years, numerous trials have evaluated the most effective approach to liberate patients from IMV, including the use of protocol-based weaning (PBW). Compared to physician-directed weaning, protocolized strategies driven by nurses and/or respiratory therapists (RTs) have gained general acceptance in many intensive care units (ICUs) because of reduction in morbidity, mortality, and health-care costs (Table 65–1).^4,5,6

Prior to the adoption of protocols, conventional weaning was at the sole discretion of the ICU physician(s) overseeing the patient’s care.⁷ Conventional weaning (ie, no protocol) was highly based on clinical experiences and institutional practices. Thus, the weaning of a particular patient varied considerably among physicians and across institutions. This broad range of traditions and practice patterns underscored an imperfect process of IMV termination. Data from the mid-to-late 1990s highlighted that clinical judgment alone was not always accurate in predicting successful extubation and that physicians were more likely to keep patients on IMV for longer periods than necessary.^6,8,9

The availability of registered nurses (RNs), RTs, and ancillary staff at the bedside in the ICU coupled with their intimate knowledge of patient’s characteristics offered an alternative pathway to liberation from IMV. Subsequent efforts focused on developing conventional, structured, reproducible, and safe methods of liberating patients from IMV, not only to improve patient outcomes but also to decrease existing variations in practice among healthcare providers. Protocol-driven weaning was thus born.⁴

Liberation from IMV is a process composed of several steps including readiness testing (or screening), weaning (itself), and extubation.

• Readiness testing (or daily screening)—The process uses several criteria to determine if a patient is ready to begin weaning or to undergo a so-called spontaneous breathing trial (SBT). The most frequent criteria include: (1) reversal of the condition(s) that prompted IMV; (2) acceptable PaO₂/FiO₂ ratio; (3) relatively low positive end-expiratory pressure (PEEP); (4) ability of the patient to initiate spontaneous breaths; and (5) stable hemodynamic parameters (Table 65–2).^7,9,10,11

• Weaning—It is the practice of either a gradual or total reduction in ventilator support during a SBT, thereby allowing patients to incrementally participate in the work of breathing.^6,8 Weaning a patient from IMV may take up to 40% of the total duration on the ventilator.⁶ A SBT is often performed concurrently with daily interruption of sedation.¹² The most popular weaning models are T-piece and continuous positive airway pressure (CPAP) trials, and pressure support ventilation (PSV) either alone or in combination with synchronized intermittent mandatory ventilation (SIMV). These strategies are largely institution-dependent and currently there is no general consensus as to which particular weaning mode leads to faster extubation rates.^6,8

• Extubation—It is the physical act of removing the artificial airway (ie, endotracheal tube) from the patient. Regardless of the weaning model chosen, extubation is considered once the patient demonstrates the ability to breathe with little or no support from the ventilator and both airway patency and protection have been established. Liberation and discontinuation of mechanical ventilation are terms used interchangeably to describe the process of weaning and extubation.¹³

A weaning protocol is an algorithm applied to the components of the traditional weaning process highlighted above, with the purpose of consistently identifying and determining the most appropriate time for extubation.¹⁴ A successful protocol usually incorporates the 3 basic components mentioned (ie, daily screening, weaning and criteria for extubation) but 2-step models have also been proposed. A protocol is usually driven by a RN and/or a RT.

Finally, weaning is a fastidious process with sometimes unpredictable outcomes that are dependent on several variables. Patient-specific factors include underlying illness severity, functional status before ICU admission, and comorbid conditions. The application of evidence-based practice by healthcare providers (eg, low tidal volume ventilation strategy for acute respiratory distress syndrome, promotion of sleep hygiene, avoidance of delirium-generating medications and interventions, and early ICU mobilization) has a definite impact on weaning. In reality, some patients are easily and promptly extubated while others linger on the ventilator for long periods of time. To improve standardization, patients who undergo weaning can be classified as simple, difficult and prolonged.¹⁵ The “simple weaning” group incorporates patients that are successfully liberated from IMV after 1 SBT. Patients in the “difficult weaning” group require up to 3 SBTs or up to 7 days on IMV (after failing the first SBT) prior to successful liberation. Those in the “prolonged weaning” group fail at least 3 SBTs and require more than 7 days of weaning before they can be successfully disconnected from the ventilator.^15,16

The evidence for PBW dates back to the early 1990s. One of the earliest trials was a computer-based protocol that was compared to conventional or physician-controlled weaning. The study reported fewer arterial blood gas samples, shorter weaning times, and less time spent outside acceptable respiratory rate and tidal volume parameters in the group assigned to the computer-based protocol.¹⁷ In 1996, Ely et al demonstrated that, compared to a physician’s judgment alone, a protocol executed by nurses and RTs was superior in decreasing the total duration of IMV, associated complications, and costs.⁴ Subsequent studies have shown substantial evidence in favor of nursing and/or RT-directed PBW to expedite liberation from IMV.

A 2011 Cochrane systematic review and meta-analysis of 11 randomized and quasi-randomized controlled trials (primarily RN/RT and computerized protocols) concluded that PBW resulted in a decrease in the duration of mechanical ventilation, weaning and ICU length of stay (LOS), without adverse outcomes.¹⁸ A retrospective study of patients mechanically ventilated for more than 24 hours (at an institution with PBW) corroborated a decrease in duration of IMV and a 12.7% reintubation rate, well within the standard of 5% to 15% proposed by the multisociety task force guidelines in 2001.^9,19 Similarly, a multicenter prospective cohort study by Teixeira et al concluded that incorporating a protocol in difficult-to-wean patients (ie, patients failing their first SBT) led to significantly higher extubation rates and improved mortality.²⁰ Extubation failure in the protocol group was 13.3% compared to 30.4% in the nonprotocol group.²⁰ Another prospective randomized controlled trial in a coronary care unit demonstrated significantly faster times to extubation and lower reintubation rates (16.7%), once protocol weaning was initiated.²¹

The majority of the early PBW trials were primarily RT/RN- or RT-driven. Recent evidence from solely RN-driven weaning has emerged (Table 65–3). Danckers et al showed that a protocol for liberation from IMV driven by ICU nurses decreased the duration of IMV and ICU LOS in patients mechanically ventilated for more than 24 hours without adverse effects and was well accepted by ICU physicians.²²

In 2001, a study looking at a neurosurgical population found no benefit to an RT-driven protocol (compared to physician judgment alone) in terms of duration of IMV, mortality, reintubation rates, and development of ventilator-associated pneumonia (VAP).²³ Similarly, another study looking at an RT-driven protocol (vs physician judgment) in a trauma ICU also concluded no difference in outcomes (ie, self-extubation rates, ventilator days, ICU LOS, and costs).²⁴ Krishnan et al studied the outcomes of 2 groups of patients who required mechanical ventilation for more than 24 hours: the experimental group underwent a weaning protocol driven by RTs and/or nursing staff; weaning in the control group was managed by various physicians at different levels of training who staffed a closed ICU.²⁵ Reported outcomes were successful extubation, duration of mechanical ventilation, hospital mortality, ICU-LOS, and rates of reintubation. The authors reported no difference in outcomes between the 2 groups. Finally, Rose et al found no reduction in duration of IMV in surgical and trauma ICUs that compared Smartcare automated weaning versus usual care.²⁶

The study by Krishnan et al highlighted the importance of the “organizational context,” which relates to factors such as staffing, “usual care,” type of ICU (surgical, medical, neurosurgical, etc), bed availability, and even an open versus a closed ICU model. It points to redundancies in hospitals and medical centers where a high staffing level is likely to promote early recognition and transition to weaning and extubation without the need for a choreographed structure.^25,26 Usual care refers to the customary practices performed at a particular ICU. At Johns Hopkins Hospital where the Krishnan study was conducted, high-level, evidence-based practice with 24-hour ICU physician coverage (ie, usual care) may explain the lack of difference in outcomes and the apparent ineffectiveness of PBW.²⁷ Protocolized weaning may be more suitable to hospitals that lack 24-hour coverage by intensivists or other physicians versed in the weaning cultures.

The type of ICU can play an important role in the weaning process. For example, a postanesthesia care unit (PACU) or a surgical ICU may not embrace the time-consuming and restrictive PBW. The daily screen is often disregarded in favor of an expeditious SBT in the majority of patients in the PACU or surgical ICU, thereby avoiding the burden of fulfilling stringent weaning criteria. No benefits were demonstrated when PBW was tried in neurosurgical and trauma ICUs, reinforcing the limitations associated with certain types of ICU.^23,24 Although safe and effective, PBW may miss many patients that could be extubated sooner than what is predicted by established protocol criteria. As previously reported, half of the patients who self-extubate prematurely, do not require reintubation within a 24-hour period.^28,29

Another major concern regarding PBW lies in the heterogeneity of the published studies and their conflicting and inconsistent results. Hospital and ICU cultures, resources and settings heavily influence the methodology and protocol designs of these trials, resulting in a vast array of treatment algorithms. In the meta-analysis of 11 randomized controlled trials (RCTs) by Blackwood et al, readiness to wean criteria (daily screen) differed in every single study.¹⁸ Likewise, there was variability in the staff members (RTs, RNs, or physicians) involved in the application of the study protocols. Of the 11 studies, only 2 employed the same weaning protocol while the remainder used different combinations of weaning methods for liberation from IMV. The weaning modality (eg, PSV, T-piece trials, SIMV, or CPAP) also differed between the studies but did not impact the process of adequately identifying prospective candidates for liberation from IMV.^10,20 The results of these studies are hardly reproducible and centers should be mindful and perform careful assessment of a protocol before its implementation.

Despite these limitations, a large body of evidence argues in favor of an RN/RT-led approach to weaning when compared to conventional (ie, no protocol) weaning by physicians (Table 65–3). Indeed, RN- and RT-directed protocols appear to be safe, beneficial and cost-effective. Certainly, healthcare providers who spend a lot of time with patients are well equipped to recognize the best time for extubation. PBW is relatively easy to incorporate into the culture of an ICU because RNs and RTs are widely available in many ICUs and they are consistently present at a patient’s bedside.

RN/RT-led PBW appears ideal for patients in the “simple weaning” category and for most patients in the “difficult weaning” group. Patients in the “prolonged weaning” group and a few patients in the “difficult weaning” category should undergo further workup to identify barriers to successful liberation from IMV. Imaging studies, electromyography and nerve conduction testing, serologic markers for vasculitides, myopathies and endocrinopathies, esophageal manometry, etc may be necessary to diagnose medical conditions that impede weaning and to develop appropriate plans of treatment.^30,31

In the past few years, a new paradigm has emerged with the concepts of limited/no sedation and early ICU mobilization (early rehabilitation therapy) as adjuncts in liberating patients faster from IMV.^{32,33,34,35,36} A small number of RCTs and prospective studies have demonstrated that early mobilization of patients requiring mechanical ventilation was not only safe and feasible but also associated with increased ventilator-free days, shorter return time to baseline functional capacity, and reduced ICU and hospital LOS, mortality and costs.^33,34,35 In the majority of cases, early mobilization was possible at a median of 1.5 days (range of 1-2.1 days) after initiation of IMV.³⁵

Ventilator weaning protocols can aid ICU staff (physicians, RNs, and RTs) in identifying patients ready to be liberated from IMV. However, protocols have limitations. Notwithstanding, protocols are safe and, for the majority of patients, they yield real benefits that reduce morbidity, mortality and costs. RNs and RTs play an instrumental role in successful PBW strategies but the appropriate weaning strategy is likely dependent on the available resources at a specific institution. We advocate RN/RT PBW for patients in the “simple” and “difficult” to wean groups in hospitals that have limited human and financial resources. At our institution, we use a wide-ranging integrative approach of limited sedation and early mobilization that involves physicians at different levels of experience (ie, residents, fellows, and full-time intensivists), critical care RNs, nurse practitioners, physician assistants, RTs, and physical and occupational therapists to determine readiness to wean and achieve liberation from IMV. We have achieved great success without the need for a specific weaning protocol.