Prolonged intubation after cardiac surgery increases the risk of morbidity and mortality and lengthens hospital stays. Factors that influence the ability to extubate patients with speed and efficiency include the operation, the patient’s baseline physiological condition, workflow processes, and provider practice patterns.
Progression to extubation lacked consistency and coordination across the team. The purpose of the project was to engage interprofessional stakeholders to reduce intubation times after cardiac surgery by implementing fast-track extubation and redesigned care processes.
This staged implementation study used the Define, Measure, Analyze, Improve, and Control approach to quality improvement. Barriers to extubation were identified and reduced through care redesign. A protocol-driven approach to extubation was also developed for the cardiothoracic intensive care unit. The team was engaged with clear goals and given progress updates.
In the preimplementation cohort, early extubation was achieved in 48 of 101 patients (47.5%) who were designated for early extubation on admission to the cardiothoracic intensive care unit. Following implementation of a fast-track extubation protocol and improved care processes, 153 of 211 patients (72.5%) were extubated within 6 hours after cardiac surgery. Reintubation rate, length of stay, and 30-day mortality did not differ between cohorts.
The number of early extubations following cardiac surgery was successfully increased. Faster progression to extubation did not increase risk of reintubation or other adverse events. Using a framework that integrated personal, social, and environmental influences helped increase the impact of this project.
This article has been designated for CE contact hour(s). The evaluation tests your knowledge of the following objectives:
Identify the importance of reducing intubation times in cardiac surgery patients.
Distinguish process, people, and patient barriers to extubation in postoperative cardiac surgery patients.
Explain the advantage of using a behavioral change framework to improve staff engagement in the improvement initiative.
To complete evaluation for CE contact hour(s) for activity C2131, visit www.ccnonline.org and click the “CE Articles” button. No CE fee for AACN members. This activity expires on June 1, 2023.
The American Association of Critical-Care Nurses is accredited as a provider of nursing continuing professional development by the American Nurses Credentialing Center’s Commission on Accreditation, ANCC Provider Number 0012. AACN has been approved as a provider of continuing education in nursing by the California Board of Registered Nursing (CA BRN), CA Provider Number CEP1036, for 1 contact hour.
For patients undergoing cardiac surgery, early extubation after surgery is associated with fewer adverse outcomes, shorter intensive care unit (ICU) stays, and lower costs.1–4 The Society of Thoracic Surgeons (STS) defines early extubation as extubation within 6 hours after ICU admission following cardiac surgery.3 Prolonged intubation of more than 24 hours is a key quality indicator following cardiac operations and is reported to STS-participating institutions as a morbidity metric for star quality ratings.5 Prolonged intubation is associated with adverse outcomes and increased costs.4,6 Examples of adverse outcomes reported in the literature are infectious complications, renal failure, stroke, ICU readmission, reintubation, and operative mortality.4,7
In our 32-bed dedicated cardiothoracic ICU embedded within a large academic medical center in the southeastern United States, patient extubation times were inconsistent. Specifically, our group lacked a uniform approach to advance patients toward extubation. Other challenges were a lack of staff awareness of the time-dependent extubation goal of the STS quality metric for early extubation and poor care coordination across disciplines (eg, among nurses, respiratory therapists, ICU clinicians, and physicians). For example, during handoff to the ICU team a surgeon would identify a patient as appropriate for early extubation, but no set criteria, workflow algorithm, or timeline for progression were in place. Therefore, progression to extubation lacked consistency and coordination across the team. Early extubation was dependent not only on patient readiness but also on communication and staff workflow (eg, delays during handoffs).
We sought to implement a fast-track extubation (FTE) program to improve patient outcomes and reduce costs associated with prolonged intubation times in adult patients after cardiac surgery. The purpose of this quality improvement initiative was to achieve early extubation in patients after cardiac surgery by using an extubation protocol and redesigned care processes. This study was approved by the Duke University Hospital institutional review board and met criteria for exemption as a quality improvement study. The framework used for this report of our quality improvement study was the Standards for Quality Improvement Reporting Excellence 2.0 publication guideline.8,9
Progression to extubation lacked consistency and coordination across the team.
We used the Define, Measure, Analyze, Improve, and Control quality improvement approach to design and conduct this implementation study.10 In the definition phase, we recognized high rates of variability in extubation time among patients who had undergone cardiac surgery and used the STS criteria for early extubation in patients in stable condition following surgery.
During the measurement phase, we collected data prospectively in a preintervention convenience sample of 101 cardiac surgery patients during 1 quarter in 2016 to determine the current state of practice and to define the scope of the problem in our local patient population. We evaluated intubation times and barriers to extubation in patients identified as eligible for progression to extubation. In this preintervention (baseline) cohort, 53 patients (52%) did not meet standard STS guidelines for early extubation. Of these, 21 (40%) had no identifiable barriers to extubation but still missed the goal.
We used root cause analysis to identify extubation processes and patient and system barriers contributing to prolonged intubation times. Barriers to extubation were categorized into 3 groups: process-specific, people-specific, and patient-specific barriers.11 Process issues included a lack of clarity about eligibility for early extubation by the surgical team, inappropriate use of sedation for hemodynamic control (eg, increasing the propofol infusion dose to lower blood pressure), inadequate pain management (possibly related to oversedation), and lack of a clear plan for initiating the weaning and extubation process. In the category of people-specific barriers, we found problems with interdisciplinary communication, poor patient progression during shift change, and an absence of cross-coverage when respiratory therapists were off the unit transporting patients. The most common patient-specific barrier was metabolic acidosis. Others included respiratory acidosis, hemodynamic instability, bleeding, and altered mental status.
Clinicians will change practice if they believe it is important and they are able to do what is needed.
In the analysis phase, we used an interdisciplinary team to analyze preintervention data, consider the most common root causes, and posit solutions to minimize variability in extubation times. Stakeholders from all disciplines (nursing, anesthesiology, cardiac surgery, intensive care, respiratory therapy, pharmacy, and advanced practice) were invited and involved in eliminating barriers and improving time to extubation. Each discipline took ownership and collaborated to eliminate or minimize the impact of the barriers to extubation. These barriers and solutions are summarized in Table 1.
As part of the root cause analysis, we reviewed extubation protocols that successfully reduced duration of mechanical ventilation, weaning time, and ICU length of stay for adoption in our local setting.11–13 Previous studies have demonstrated shortened intubation times without increased risk by using standard extubation protocols.14 Using this information, our team developed an FTE algorithm and protocol rooted within our enhanced patient care processes (Figures 1 and 2).
During the improvement phase we implemented a protocol-based approach focused on improving extubation variability. This quality improvement study was conducted in our high-volume, high-acuity, 32-bed cardiothoracic ICU. Postoperative ICU care is delivered for all types of cardiothoracic surgery, including coronary artery bypass graft and valvular surgeries; mechanical circulatory support (eg, ventricular assist devices and extracorporeal membrane oxygenation); and cardiac and pulmonary transplantation. The unit is staffed on each shift with dedicated intensivists, fellows, advanced practice providers, respiratory therapists, and registered nurses. The total number of staff members who required orientation to our newly developed processes included 170 nurses, 12 respiratory therapists, 14 intensivists, 40 advanced practice providers, 9 cardiothoracic surgery trainees, and 15 surgeons. Because of the large number of clinicians needed to fully staff the unit, approximately 70 nurses and more than 20 others typically receive orientation annually.
We chose the Influencer behavioral change framework created by Patterson and colleagues15 to guide the improvement intervention. This framework is based on evidence suggesting that people will change behavior if they believe it is important and they are able to do what is needed. Historically, practice change in our unit has been difficult to sustain and new information has been disseminated primarily through written or verbal communication. To improve engagement with this initiative, the nursing research team used the model of change described by Patterson et al,15 focusing on the conceptualization of behavior as motivated by 6 sources of influence. Using this evidence, we chose personal, social, and structural sources of influence to guide the interventions for FTE protocol implementation (Table 2). The stakeholder groups were educated by using an online module. This module included information about our internal performance data, the STS quality metrics, and the new FTE extubation protocol and redesigned care processes. We also posted flyers around the unit and provided information at staff meetings. We drew attention to the new protocol with a kickoff party that included snacks and unit decorations with a fast-track race car theme. We implemented the FTE protocol on September 1, 2017.
We created a monitoring and measurement plan to evaluate the success of the intervention. The primary end point for the quality improvement study was early extubation, as defined by the STS (< 6 hours following admission to the ICU from the operating room). In addition, patient demographics, type of surgery, operating room and ICU medications, arterial blood gas values, and barriers to extubation were recorded. We selected standard definitions for all data points, including categorical definitions for barriers to extubation. Environmental reminders such as copies of the FTE protocol and signs with the extubation time goal (Figure 3) placed on the door of each patient room drew staff attention and defined expected practice.
Members of the unit nursing research team collected data for all FTE patients. We created a log of all FTE patients for daily tracking and follow-up with data collection. A standard data sheet was completed for every patient designated for FTE. Data were collected prospectively from the electronic health records. Barriers were identified through review of electronic health records for items meeting barrier definition criteria or with additional input from the care team. Team performance was reported as the percentage of eligible patients achieving early extubation each week. We used a racetrack that was updated weekly to include the names of the “pit crew” (nurse, respiratory therapist, and advanced practice provider) who successfully extubated patients within the recommended period (Figure 4). Social influence can be a key factor in individuals’ adoption of new innovations.16 The racetrack generated enthusiasm and captured the power of peer influence. Having names on the racetrack created healthy competition between peers, making best practices socially desirable.
Members of the interdisciplinary team met monthly to review data and modify interventions if needed. Data were collected for 3 months following implementation. As the need for clarification or modifications arose, ongoing education was disseminated to the care team after discussion among stakeholders and protocol implementation.
In the final phase of the study (control), the research team agreed to continue to monitor every FTE patient for a full year with a modified data form to evaluate the 12-month sustained effect of the intervention. Staff member enthusiasm for seeing their car awarded on the racetrack was high, so the racetrack theme also continued for a year (September 1, 2017, to August 31, 2018).
Our results demonstrated that an early extuba tion protocol can safely decrease ventilation time with out increasing morbidity or mortality. All patients designated for progression to extubation were included in the analysis except 4 patients who received FTE but underwent unplanned reoperation following admission to the ICU. These 4 patients were excluded from the analysis according to STS database criteria for unplanned reoperations. The final analysis included 312 patients (101 in the preintervention cohort and 211 in the FTE cohort). Patients’ procedures included coronary artery bypass graft surgery, valve repair or replacement, ascending aortic aneurysm repair, maze procedure, septal myectomy, and combinations of these procedures. Eighty-three patients (28 in the preintervention cohort, 55 in the FTE cohort) had combined procedures. Baseline demographic characteristics of the 2 cohorts were not significantly different (Table 3). The mean (SD) patient age for the combined cohorts was 62.0 (13.2) years. Most patients were male (n = 209 [67%]) and White (n = 253 [81%]). Patients’ mean (SD) body mass index was 29.2 (6.3). The preoperative glomerular filtration rate was greater than 60 mL/min for most patients (n = 229 [73.4%]). Most patients had preoperative essential hypertension (n = 217 [69.6%]) and a diagnosis of heart failure (n = 183 [58.7%]). Current smokers made up a small proportion of the patient population (n = 32 [10.3%]), as did patients with chronic lung disease. Most patients underwent elective surgery (n = 270 [86.5%]).
The primary results of the study demonstrated significant, sustained improvement in the rate of extubation less than 6 hours after surgery (Figure 5). As shown in the control chart, the proportion of patients extubated within the recommended 6-hour window in the preintervention cohort was 47.5% (n = 48), demonstrating a large opportunity for improvement. Following the intervention, the proportion of patients extubated in less than 6 hours increased to 72.5% (n = 153; P < .001). In the FTE cohort, people- and process-related barriers decreased from 48% to 17%.
Factors associated with successful extubation in less than 6 hours were absence of heart failure within the previous 2 weeks (Wald χ2 = 4.09; P = .04) and being in the FTE cohort (Wald χ2 = 15.48; P < .001). We also found a tendency toward shorter ICU and hospital stays, but these differences were not significant (P = .41 and P = .31, respectively). The mean number of arterial blood gas samples obtained was 5.3 in patients extubated after more than 6 hours and 3.3 in patients extubated in less than 6 hours. This finding suggests a potential cost reduction, although cost data were not collected in this study. Reintubation rates during the hospital stay did not appreciably differ between the 2 cohorts (preintervention cohort, 2 of 101 patients [2%]; FTE cohort, 3 of 211 patients [1%]). At the end of 1 year, the monthly early extubation rate ranged from 62% to 85%, with an overall rate of 73.6% (559 of 760 patients).
The team selected surgical patients with similar medical conditions for the preintervention cohort but included some surgical patients with more complicated conditions in the early extubation quality initiative. We analyzed data for this group of patients separately because of lack of similarity with the preintervention cohort. This discovery cohort was defined as patients who were considered eligible for early extubation but met 1 or more of the prespecified considerations for exclusion (Figure 2). The study team chose to list exclusions as “considerations” rather than specifying limits for FTE protocol use. The apparent precision and safety of our FTE protocol ushered in the inclusion of high-risk patients in protocol-based care. Although these patients would not previously have qualified for inclusion in the quality improvement project according to the initial criteria, they were ultimately included and safely progressed to extubation. Of these 23 patients, 9 (39%) were successfully extubated within 6 hours. We likely achieved this result because safety is designed into the protocol: continuation or cessation of progression toward extubation depends on prespecified safety parameters (eg, patient free from acidosis, patient hemodynamically stable, chest tube output minimal) delineated in the FTE protocol.
The results of this quality improvement study demonstrate that a multidisciplinary approach can successfully improve the proportion of patients extubated within a 6-hour window according to the STS quality recommendations. Similar to other published studies, our results demonstrated that an early extubation protocol can safely decrease ventilation time without increasing morbidity or mortality.17
Previous studies have successfully used quality improvement approaches to reduce variability in intubation times.11–13 Extubating patients soon after cardiac surgery is based on a number of factors, including the patient’s baseline physiological condition and comorbidities, workflow processes/culture, and provider practice patterns. Examination of all factors that influence the ability to decrease unnecessary intubation time is necessary to improve practice. In addition, the use of standardized extubation protocols has been shown to reduce intubation time without increasing patient risk.14,17,18 Gutsche et al11 identified workflow barriers, physical inconveniences, and interdisciplinary miscommunication as causes of increased intubation times. They used a coordinated approach to mitigate these factors, improving early extubation from 27% (53 of 195 patients) to 50% (85 of 171 patients).11 Similarly, Fitch and colleagues12 used a sequential approach to integrate a new process aimed at improving early extubation. Their approach consisted of monitoring current practice (phase 1); implementing an extubation protocol (phase 2); and continuing to use a protocol combined with reversal of paralytic agents, extubation at lower body temperature, and reminders in the patient room (phase 3). Early extubation improved from 11.8% in phase 1 (139 of 1174 patients) to 23.9% in phase 2 (151 of 631 patients) and 38.3% in phase 3 (98 of 256 patients).12 In addition to implementing an extubation protocol and improving communication, Hefner et al13 amended the postoperative order set to reduce intubation times and reported sustaining the reduction in intubation times for 4 years. Our study demonstrated an improvement in extubation time at 3 months, as was found in studies by Gutsche et al11 and others,12,13 but we were also able to show a sustained effect at 1 year.
Our findings support the use of multidisciplinary protocol-driven care in improving safety, timeliness, efficiency, and consistency of care delivery associated with patient extubation.
By developing and implementing an extubation protocol with members of the interdisciplinary team as key stakeholders in the care redesign process, we were able to reduce barriers related to people (eg, individual-level variations in order writing or propofol weaning) and processes (eg, ventilator weaning, handoffs, and work-flow issues). Our findings, consistent with those of Chan and colleagues,18 support the use of multidisciplinary protocol-driven care in improving safety, timeliness, efficiency, and consistency of care delivery associated with patient extubation.19
We observed a decline in the number of patients achieving early extubation 6 to 7 months into the project. We attributed this observation to a high influx of new nurses, respiratory therapists, and residents with the new academic year. To improve compliance and reengage the team, we used the same strategies to raise awareness of the FTE protocol. These strategies included education for new team members and inclusion of the FTE content in the orientation curriculum.
Unique to our study as compared with others in the literature was the use of a change model to guide personal, social, and structural sources of influence to motivate change.15 Implementation of the model resulted in universal and overwhelming enthusiasm for change, not only from members of the investigative team but also across all disciplines and staff, from the unit secretary to the chief of cardiothoracic surgery. The interventions and activities associated with improving quality of care for patients were adopted by all.
We encountered several limitations, including initial barriers to buy-in by the full multidisciplinary team, resulting in implementation delays at the outset. Staff nurses in patient care roles also supported data entry needs, resulting in an occasional backlog of data and delays in internal feedback reporting. Unequal numbers of patients were included in the pre- and postintervention cohorts because of the use of convenience sampling. These limitations are considered part of the natural evolution of adoption of a change in practice and were not thought to detract from the overall improvements that resulted from the study.
Collaborative implementation of a standardized extubation protocol and redesigned patient care processes helped us successfully achieve the goal of reducing intubation time in patients in stable condition after cardiac surgery. We markedly improved the proportion of patients extubated in less than 6 hours following cardiac surgery. Use of the FTE protocol reduced intubation time without increasing reintubation rate or mortality. The use of a change model with multiple sources of influence was vital to the success of adoption of the FTE protocol. The fidelity of long-term implementation is driven by a team process that includes ongoing feedback and education to ensure success.
The authors would like to thank and acknowledge the interprofessional leadership team for support of this project: Jill Engel, DNP, RN, Kelly Kester, MSN, RN, Lynn McGugan, DNP, ACNP, Jacob Schroder, MD, John Haney, MD, Raquel Bartz, MD, Shawn Kram, PharmD, Peter Henkel, PharmD, Brian Costa, RCP, Robert Slonac, RCP, and Yasmin Maisonave, MD.
To purchase electronic or print reprints, contact the American Association of Critical-Care Nurses, 27071 Aliso Creek Rd, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; email, email@example.com.
Dr Ghadimi is a remunerated consultant for UpToDate in the section for coagulation and blood management, receives grant support from the National Institutes of Health (T32GM008600) and Duke Health, and is a nonremunerated consultant and subject matter expert for ERAS Cardiac (Enhanced Recovery After Cardiac Surgery Society). He has received previous grant support from Octapharma. Dr Granger has received research support from Novartis, Janssen Pharmaceuticals, AstraZeneca, the American Heart Association, and the National Institutes of Health. She reports no relevant conflicts of interest related to this manuscript. Other authors have no conflicts of interest to report. Duke Heart Center departmental funding supported this project.
To learn more about intubation of patients in the critical care setting, read “Intubation Setting, Aspiration, and Ventilator-Associated Conditions” by Talbert et al in the American Journal of Critical Care, 2020; 29(5):371-378. Available at www.ajcconline.org.