Postoperative respiratory failure (PORF), defined as the inability to liberate from mechanical ventilation or unanticipated postoperative reintubation, is an adverse event with profound implications on morbidity and mortality for patients. Approximately 0.4–3.4% of patients undergoing surgical procedures develop PORF.(1,2)
Patients who develop PORF have an increased length of stay (LOS) of approximately 4 days, as well as an increased cost of care by $26,000 compared to patients who do not develop PORF.(2) Furthermore, the Centers for Medicare and Medicaid Services (CMS) monitor PORF as a marker for the quality of care provided to elective surgical patients.
Surgical procedures such as abdominal aortic aneurysm repair, thoracic surgery, neurosurgery, peripheral vascular surgery, and neck surgery carry increased risk for PORF.(1) Patient risk factors for PORF include chronic obstructive pulmonary disease (COPD), known or suspected obstructive sleep apnea (OSA), chronic kidney disease (CKD), and a higher American Society of Anesthesiologists (ASA) classification.(2)
Early identification of patients at risk for PORF allows for preoperative evaluation and optimization, as well as the customization of surgical and anesthetic plans to mitigate PORF.(3) Intraoperative lung protective ventilation, restrictive fluid strategies, minimization of opiates, and the use of regional anesthesia can improve postoperative respiratory outcomes.(3) Postoperative early mobilization, appropriate monitoring, and multimodal analgesia can reduce the incidence of PORF.(3)
While many studies have focused on individual interventions, few have examined outcomes associated with bundling interventions.(3) Bundling these interventions into enhanced recovery after surgery (ERAS) pathways can lead to superior patient outcomes. With so many interventions possible, a question arises: to what extent can postoperative respiratory complications be prevented and PORF averted through bundled implementation?
The intent of this study is to describe an approach to reducing the incidence of PSI 11 events in a large health system by implementing multiple bundled interventions. This effort was especially challenging because the interventions were implemented during the COVID-19 pandemic amid increased patient acuity, resource utilization, government-mandated operating room utilization restrictions, and staffing constraints.
Methods
The stated goal was to reduce PORF in the health system as part of our Zero Harm Commitment.(5) An enabling infrastructure was instituted by creating a multidisciplinary team with a broad system representation, which included administrative and clinical leaders. Opportunities were identified by reviewing evidence-based literature and analyzing the PSI 11 cases from the prior year. A key driver diagram was established (Figure 1). This serves as a conceptual tool to identify primary drivers related to an aim or outcome and map those primary drivers into intervention and change concepts.(6)
Figure 1. Postoperative Respiratory Failure Reduction Key Driver Diagram
Several interventions were put forth in the preoperative period to identify patients at risk for PORF and to engage risk-mitigation measures. The Arozullah respiratory failure index was implemented in pre-admission testing evaluation to identify patients at risk of PORF.(1) An elevated score was then used as a trigger to obtain preoperative pulmonary medicine optimization and clearance.
A deep venous thromboembolism assessment tool was implemented to guide ordering physicians on proper anticoagulation for patients. Education was provided during the preoperative period on the importance of pulmonary hygiene, early mobilization, and use of incentive spirometry.
During the intraoperative period, enhanced recovery after surgery pathways were developed for surgical procedures, including those at highest risk for PORF. Evidence-based medicine was used to developed anesthesia care plans and surgical approaches to reduce risk to patients. Examples of interventions included restrictive fluid strategy and minimization of intraoperative opiates.
In the postanesthesia care unit, the need for non-invasive ventilation early in the postoperative period was assessed. Care teams embraced multimodal analgesia to minimize respiratory depression and promoted the use of regional anesthesia if this had not been done preoperatively. Later in the postoperative period, patients could be managed by an acute pain service to control pain and minimize PORF.
A restrictive blood and fluid strategy was continued into the postoperative period to reduce adverse events. In critical care patients, the A-F bundle was deployed to help appropriately manage respiratory and cognitive issues early in the patient course.(7) Extensive education was provided for physicians on the definition of PORF and to provide appropriate supporting documentation of the diagnosis. Every PSI 11 case was reviewed by an interdisciplinary committee on a weekly basis.
Results
Observational data, including the number of surgical cases and the number of PORF events, was collected for the 4-year period from January 2018 through December 2021 across 11 regional medical centers and one quaternary referral medical center. There were 26,478 surgical cases included in the analysis and 103 individual instances of PORF were identified, representing an overall incidence of 3.89 cases of PORF per 1,000 surgeries over the study period.
The monthly number of surgical cases is noted in Figure 2, with the red line denoting implementation date for interventions. The monthly calculated incidence of PORF per 1,000 surgeries in each month of observation is in Figure 3, with the red line denoting implementation date for interventions.
Figure 2. Monthly Number of Surgical Cases; Blue dot represents April 2020 government-mandated suspension of elected cases due to COVID 19 pandemic.
Figure 3. Monthly Calculated Incidence of PORF per 1000 Surgeries; Blue dot represents April 2020 government-mandated suspension of elected cases due to COVID 19 pandemic. Trend line represents average incidence of PORF in each period.
Implementation of strategies to identify patients at increased risk of PORF and to mitigate their risk for PORF were initiated in January 2020. Incidence of PORF per 1,000 elective surgical cases before and after implementation were compared. An indicator variable for April 2020 was controlled for in the regression analysis because of the cancellation of all except emergency cases during the beginning of the COVID-19 pandemic resulting in a significant decrease in the number of surgical cases performed during that month (p < .01). The total number of cases per month was also controlled for to account for variation in the number of cases performed in each month. Monthly incidence of PORF per 1,000 surgeries was calculated and used for analysis.
Multivariate regression analysis was performed comparing pre-implementation incidence with post-implementation incidence of PORF controlling for the number of monthly surgical cases and for April 2020. Before implementation of mitigation strategies, overall baseline monthly incidence of PORF was 4.53 cases per 1,000. Following implementation, raw overall incidence of PORF fell to 3.5 cases per 1,000, with a multivariate regression-implied decrease of 1.59 cases/1000 (p = 0.028), representing a 35% reduction in incidence of PORF (Table 1).
Limitations
The COVID-19 pandemic and the varying incidence of these patients, as well as high patient acuity, increased resource utilization, staffing constraints, and change management, affected the study results. We do not know the impact of our intervention on pediatric patients. Patients included in this study were limited to adult elective surgical discharges with a diagnosis of documented respiratory failure that was not present on admission and excluded based on PSI 11 exclusion criteria found in PSI 11 technical specifications.(4)
We used the AHRQ PSI 11 definition of respiratory failure rather than a broader definition. Our study was observational, so we cannot establish a definitive relationship between the intervention and reduced PSI 11.
Additionally, although the study was conducted at a diverse group of hospitals, they were within a single health system, which limits our ability to generalize the results. All interventions were implemented as a bundle, making determination of the impact of individual interventions difficult to obtain.
Discussion
In our health system, the early identification of patients at risk for postoperative respiratory failure and implementation of perioperative prevention and mitigation strategies reduced the incidence of PSI 11 events by 35%. The pre-implementation incidence of postoperative respiratory failure was 4.53 events per 1,000 cases and the incidence of postoperative respiratory failure was 3.5 events per 1,000 cases in the post-implementation period, asserting a statistically significant reduction.
Many previous studies have focused on the correction of one particular variable or a restricted patient population.(8) Many studies have focused on pulmonary complications rather than PORF, which may lead to misinterpretation of results.
Smoking cessation in the perioperative population has been extensively studied, demonstrating a relative risk reduction of 41% when comparing former smokers to current smokers for pulmonary complications.(9) The importance of establishing obstructive sleep apnea protocols also has been well established as a method of reducing pulmonary sequelae after surgery.(10)
Many studies have focused on the identification of risk factors associated with PORF, by few have focused on the outcomes that can be achieved with a broad, diverse patient population.(11) Because of the ongoing nature of the COVID pandemic and the impact on the healthcare system, few studies have been completed to reflect sequelae on the current treatment of patients. In addition to reaching statistical significance, these results have important clinical implications.
While not measured in our study, the reduced incidence of postoperative respiratory failure directly benefits patients by reducing perioperative morbidity, mortality, and hospital length of stay.(12) Hospital systems benefit directly as the reduction in ventilator days and thus reduced intensive care and hospital length of stay lead to reduced hospital costs.
Additionally, postoperative respiratory failure is a metric that CMS uses to assess the quality of care for patients following elective surgery; therefore, reduced PORF correlates with improvements in quality of care and preservation of care reputation.(4,12)
One of the strengths of this study is the large sample size analyzed, leading to a robust data set. Another strength is that this study was conducted within a hospital system with multiple regional medical centers and one quaternary referral center. By including these multiple centers, this study shows that the intervention can be implemented throughout a hospital system at sites of various size and with differing patient acuity.
Ultimately, this study demonstrates a sustainable intervention that is able to be implemented at low cost while providing considerable benefit to the patients and the hospital system.
Conclusions
Early identification of patients at risk for postoperative respiratory failure, the creation of an enabling infrastructure, and implementation of preventative strategies were effective steps in reducing the observed incidence of PSI 11 events in the study hospital system.
Using this approach is beneficial to patients and hospital systems as a reduced PSI 11 incidence is correlated with reduced morbidity, mortality, hospital length-of-stay, and visit cost.(2) Our hope is that other healthcare systems will employ similar methodology, as outlined in the key driver diagram, to attempt to eliminate patient harm.
References
Arozullah AM, Daley J, Henderson WG, Khuri SF. Multifactorial Risk Index for Predicting Postoperative Respiratory Failure in Men After Major Noncardiac Surgery. The National Veterans Administration Surgical Quality Improvement Program. Ann Surg. 2000 Aug;232(2):242–253. doi: 10.1097/00000658-200008000-00015. PMID: 10903604; PMCID: PMC1421137.
Melamed R, Boland LL, Normington JP, Prenevost RM, Hur LY, Maynard LF, et al. Postoperative Respiratory Failure Necessitating Transfer to the Intensive Care Unit in Orthopedic Surgery Patients: Risk Factors, Costs, and Outcomes. Perioper Med (Lond). 2016 Aug 2;5:19. doi: 10.1186/s13741-016-0044-1. PMID: 27486512; PMCID: PMC4969722.
Eikermann M, Santer P, Ramachandran SK, Pandit J. Recent Advances in Understanding and Managing Postoperative Respiratory Problems. F1000Res. 2019 Feb 18;8: F1000 Faculty Rev-197. doi: 10.12688/f1000research.16687.1. PMID: 30828433; PMCID: PMC6381803.
Patient Safety Indicator 11 (PSI 11) Postoperative Respiratory Failure Rate. Agency for Healthcare Research and Quality; https://qualityindicators.ahrq.gov/Downloads/Modules/PSI/V2020/TechSpecs/PSI_11_Postoperative_Respiratory_Failure_Rate.pdf . Accessed September 17, 2022.
Gandhi TK, Freeley D, Schummers D. Zero Harm in Health Care. New England Journal of Medicine Catalyst. 2020 Feb, 19. Doi: 10.1056/CAT.19.1137
The EvidenceNOW Key Driver Diagram. Agency for Healthcare Research and Quality; https://www.ahrq.gov/evidencenow/tools/keydrivers/index.html
Marra A, Ely EW, Pandharipande PP, Patel MB. The ABCDEF Bundle in Critical Care. Crit Care Clin. 2017 Apr;33(2):225–243. doi: 10.1016/j.ccc.2016.12.005. PMID: 28284292; PMCID: PMC5351776.
Ruscic KJ, Grabitz SD, Rudolph MI, Eikermann M. Prevention of Respiratory Complications of the Surgical Patient: Actionable Plan for Continued Process Improvement. Curr Opin Anaesthesiol. 2017 Jun;30(3):399–408. doi: 10.1097/ACO.0000000000000465. PMID: 28323670; PMCID: PMC5434965.
Mills E, Eyawo O, Lockhart I, Kelly S, Wu P, Ebbert JO. Smoking Cessation Reduces Postoperative Complications: A Systematic Review and Meta-Analysis. Am J Med. 2011 Feb;124(2):144–154.e8. doi: 10.1016/j.amjmed.2010.09.013. PMID: 21295194.
Bolden N, Smith CE, Auckley D, Makarski J, Avula R. Perioperative Complications During Use of an Obstructive Sleep Apnea Protocol Following Surgery and Anesthesia. Anesthesia & Analgesia. 2007;105(6):1869–1870. doi: 10.1213/01.ane.0000295223.31946.b5
Gupta H, Gupta PK, Fang X, Miller WJ, Cemag S, Forse A, Morrow LE. Development and Validation of a Risk Calculator Predicting Postoperative Respiratory Failure. Chest. 2011 Nov; 140(5):1207–1215. doi 10.1378/chest.11-0466.
Stocking JC, Drake C, Aldrich JM, Ong MK, Amin A, Marmor RA, et al. Outcomes and Risk Factors for Delayed-Onset Postoperative Respiratory Failure: A Multi-Center Case-Control Study by the University of California Critical Care Research Collaborative (UC3RC). BMC Anesthesiol. 2022 May 14;22(1):146. doi: 10.1186/s12871-022-01681-x. PMID: 35568812; PMCID: PMC9107656.