Quantifying Lives Lost Due to Variability in Emergency General Surgery Outcomes: Why We Need a National Emergency General Surgery Quality Improvement Program. Hashmi ZG, Jarman MP, Havens JM, Scott JW, Goralnick E, Cooper Z, Salim A, Haider AH. J Trauma Acute Care Surg. 2021 Apr 1;90(4):685-693.
The purpose of this study was to estimate the number of emergency general surgery (EGS) deaths that occur as a result of hospital-level variability, as an argument for the implementation of a national EGS-specific quality improvement program.
The authors begin by highlighting the total annual costs of EGS care in the United States and commenting on the absence of a dedicated EGS quality improvement program, as well as discussing the existence of dedicated quality improvement programs for specialties with far less cost and healthcare system burden. They go on to describe the morbidity and mortality improvements that have been observed in both elective and trauma surgery following implementation of the American College of Surgeons’ National Surgical Quality Improvement Program (NSQIP) and Trauma Quality Improvement Program (TQIP), as well as the belief that no small part of these improvements is the effectiveness of NSQIP and TQIP in addressing hospital-level variation. This hospital-level variability should be assumed to exist in the setting of EGS as well, and the authors set out to use the variability between high-performing and low-performing hospitals to quantify the potential cost in lives of EGS hospital variability.
The study utilized the National Emergency Department Sample dataset for years 2006-2014. Adult patients admitted with a primary diagnosis of an EGS condition were identified using American Association for the Surgery of Trauma (AAST) defined ICD-9 codes. Using in-hospital mortality as the outcome, hierarchical logistic regression was performed in order to stratify hospitals into quintiles of reliability-adjusted mortality. Using the best performing quintile of hospitals as the standard, lower-performing quintiles were then evaluated in comparison, allowing for the calculation of relative-risk of mortality in the lower performing quintiles, as well as the subsequent calculation of “excess deaths” seen at lower performing quintiles as compared with the best performing quintile.
Just over 26 million EGS patients were identified over the course of the study period, with 6 million of those patients undergoing surgical intervention. Mortality in the best performing quintile (Q1) was 1.04%, and rose incrementally in each subsequent quintile until reaching a maximum of 2.32% in the worst performing quintile (Q5), a trend repeating when looking at operative EGS alone. The relative risk of mortality in Q5 was 2.37 (95% CI 2.32-2.42) as compared with Q1, resulting in a calculated 74,462 excess deaths for this quintile. When evaluating the cumulative relative risks of Q2-5, 158,177 excess deaths were calculated as compared with the best performing quintile. When divided by the number of years included in this study, this results in 17,575 deaths annually.
The idea that hospital variability is major influencer of outcomes within the healthcare delivery macro-environment is not a controversial one. The acknowledgement of this variability (and the potential for improved outcomes through its mitigation) was a central component in the genesis of numerous quality improvement programs. However, a dedicated EGS quality improvement program does not exist at this time. The authors of this paper make a strong argument for development of such a program, and beyond simply appealing to rational thought, make a tangible and quantifiable estimate of the potential such a program could yield.
Association of Early vs Late Tracheostomy Placement With Pneumonia and Ventilator Days in Critically Ill Patients: A Meta-analysis. Chorath K, Hoang A, Rajasekaran K, Moreira A. JAMA Otolaryngol Head Neck Surg. 2021 May 1;147(5):450-459.
The timing of tracheostomy placement in critically ill patients and its effect on complications and outcomes is a topic that has been studied a multitude of times and ways, without the establishment of any universal consensus. This study used the most recent prospective literature on the subject to better investigate how the timing of tracheostomy placement influences rates of ventilator-associated pneumonia (VAP) and the duration of mechanical ventilation. This has implications for trauma PI/QI in that timing to tracheostomy and rate of tracheostomies performed within 7 days are reported in TQIP, with noted variably among centers.
In this review, the authors included 17 unique randomized controlled trials (totaling 3,145 cumulative patients) to compare the outcomes of early vs late tracheostomy placement. All 17 studies were included for review, with only 14 being included in meta-analysis. Early tracheostomy was defined as tracheostomy placement ≤7 days following initiation of mechanical ventilation. Tracheostomy placement >7 days after initiation of mechanical ventilation or patients without tracheostomy placement at any point were categorized into the “late tracheostomy” group. The primary outcome measures were the incidence of VAP and the duration of mechanical ventilation, with secondary outcomes of all-cause mortality and intensive care unit (ICU) days. All patients were adults (≥18 years of age) and admitted to an ICU level of care, and all surgical approaches to tracheostomy were included (percutaneous and open). Of note, all studies included were prospective, randomized, and controlled in nature, with retrospective studies excluded.
This meta-analysis found a lower incidence of VAP in patients who underwent early vs late tracheostomy (OR 0.59, 95% CI 0.35-0.99). When evaluating the duration of mechanical ventilation, 10 trials compared “ventilator days” and the remaining 4 compared “ventilator-free days”. There was no association found between early tracheostomy and a change in ventilator days (Mean difference -2.4 days, 95% CI -5.09 to +0.29 days), however early tracheostomy was associated with a greater number of ventilator-free days (Mean difference 1.74 days, 95% CI 0.48 to 3.00 days). In regard to secondary endpoints, early tracheostomy was not associated with reduced short term all-cause mortality (OR 0.66, 95% CI 0.38-1.15), but was associated with reduced duration of ICU stay (Mean difference -6.25 days, 95% CI -11.22 to -1.28 days).
While acknowledging that every patient is unique, and that the particular timing of tracheostomy placement is a decision that requires a full understanding of a patient’s clinical status, this study lends support for the general idea that tracheostomy placement within 7 days of the onset of mechanical ventilation is associated with a decreased incidence of VAP and reduced ICU duration, while not necessarily translating to improved mortality. Clinicians should apply these findings to the unique clinical circumstances of their patients to assist in decision making regarding the optimal timing of tracheostomy placement, especially in resource limited environments.