Multivariable Predictors of Postoperative Cardiac Adverse Events after General and Vascular Surgery: Results from the Patient Safety in Surgery Study

Department of Surgery, University of Kentucky, Lexington, KY 40536-0298, USA.
Journal of the American College of Surgeons (Impact Factor: 5.12). 07/2007; 204(6):1199-210. DOI: 10.1016/j.jamcollsurg.2007.02.065
Source: PubMed


Cardiac adverse events (CAEs) are relatively infrequent, but highly lethal, after noncardiac operations. The value of available risk scoring systems is uncertain and these systems can be outdated. We used the Patient Safety in Surgery Study database to develop and test a model to predict patient risk for CAEs after general and vascular surgical operations.
As part of the Patient Safety in Surgery Study, following the National Surgical Quality Improvement Program's protocol, multiple demographic, preoperative, perioperative, and outcomes variables were measured during a 3-year period. Data from 128 Veterans Affairs medical center hospitals and from 14 academic medical centers on 183,069 patients were used in a logistic regression analysis to model multivariable predictors of serious CAEs (cardiac arrest or acute myocardial infarction within 30 days of operation).
CAEs occurred in 2,362 patients (1.29%) and of these, 59.44% expired. Multivariable stepwise logistic regression identified 20 independent predictors of CAEs, which excluded most cardiac-specific risk factors. The most important multivariable predictors of CAE were American Society of Anesthesiologists physical status classification, work relative value units of the most complex procedure, age, and type of operation. A risk prediction scoring system using the logistic regression odds ratios proved to be a useful prediction tool when tested using a random sample from the database.
CAEs after noncardiac operations are relatively infrequent but highly lethal. Operation type and urgency and American Society of Anesthesiologists physical status assessment are important independent predictors of cardiac morbidity, but angina, recent MI, and earlier cardiac operation are not. A prediction scoring system based on the Patient Safety in Surgery Study multivariable odds ratios is likely to be predictive of future events in a similar population requiring noncardiac procedures. This risk model can also serve as a tool to measure quality and effectiveness of care by providers who perform noncardiac operations.

15 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this article is to describe the background, design, and patient populations of the Patient Safety in Surgery Study, as a preliminary to the articles in this journal that will report the results of the Study. The Patient Safety in Surgery Study was a prospective cohort study. Trained nurses collected preoperative risk factors, operative variables, and 30-day postoperative mortality and morbidity outcomes in patients undergoing major general and vascular operations at 128 Veterans Affairs (VA) medical centers and 14 selected university medical centers between October 1, 2001 and September 30, 2004. An Internet-based data collection system was used to input data from the different private medical centers. Semiannual feedback of observed to expected mortality and morbidity ratios was provided to the participating medical centers. During the 3-year study, total accrual in general surgery was 145,618 patients, including 68.5% from the VA and 31.5% from the private sector. Accrual in vascular surgery totaled 39,225 patients, including 77.8% from the VA and 22.2% from the private sector. VA patients were older and included a larger proportion of male patients and African Americans and Hispanics. The VA population included more inguinal, umbilical, and ventral hernia repairs, although the private-sector population included more thyroid and parathyroid, appendectomy, and operations for breast cancer. Preoperative comorbidities were similar in the two populations, but the rates of comorbidities were higher in the VA. American Society of Anesthesiologists classification tended to be higher in the VA. The National Surgical Quality Improvement Program methodology was successfully implemented in the 14 university medical centers. The data from the study provided the basis for the articles in this issue of the Journal of the American College of Surgeons.
    Full-text · Article · Jul 2007 · Journal of the American College of Surgeons
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: OBJECTIVES: Due to the dramatic medical breakthroughs and an increasingly ageing population, the proportion of patients who are at risk of dying following surgery is increasing over time. The aim of this study was to evaluate the outcomes and the epidemiology of non-cardiac surgical patients admitted to the intensive care unit. METHODS: A multicenter, prospective, observational, cohort study was carried out in 21 intensive care units. A total of 885 adult surgical patients admitted to a participating intensive care unit from April to June 2006 were evaluated and 587 patients were enrolled. Exclusion criteria were trauma, cardiac, neurological, gynecologic, obstetric and palliative surgeries. The main outcome measures were postoperative complications and intensive care unit and 90-day mortality rates. RESULTS: Major and urgent surgeries were performed in 66.4 percent and 31.7 percent of the patients, respectively. The intensive care unit mortality rate was 15 percent, and 38 percent of the patients had postoperative complications. The most common complication was infection or sepsis (24.7 percent). Myocardial ischemia was diagnosed in only 1.9 percent of the patients. A total of 94 percent of the patients who died after surgery had co-morbidities at the time of surgery (3.4 ± 2.2). Multiple organ failure was the main cause of death (53 percent). CONCLUSION: Sepsis is the predominant cause of morbidity in patients undergoing non-cardiac surgery. In this patient population, multiple organ failure prevailed as the most frequent cause of death in the hospital
    Full-text · Article · Jan 2008 · Revista Brasileira de Terapia Intensiva
  • [Show abstract] [Hide abstract]
    ABSTRACT: Valvular heart disease can be an important finding in patients presenting for noncardiac surgery. Valvular heart disease and resulting comorbidity, such as heart failure or atrial fibrillation, significantly increase the risk for perioperative adverse events. Appropriate preoperative assessment, adequate perioperative monitoring, and early intervention, should hemodynamic disturbances occur, may help prevent adverse events and improve patient outcome. This review article aims to guide the practitioner in the various aspects of anesthetic management in the perioperative care of patients with valvular heart disease. The pharmacological approach to optimization of patient outcome with drugs, such as betablockers and lipid-lowering medications (statins), is an evolving field, and recent developments are discussed in this article.
    No preview · Article · Apr 2008 · Seminars in Cardiothoracic and Vascular Anesthesia
Show more