Current issues of ACP Journal Club are published in Annals of Internal Medicine


Etiology

Surgery for cancer, reduced renal function, history of congestive heart failure, and use of bronchodilator medication predicted 2-year mortality

ACP J Club. 1992 Nov-Dec;117:88. doi:10.7326/ACPJC-1992-117-3-088

Related Content in this Issue
• Companion Abstract and Commentary: Left ventricular hypertrophy, history of hypertension, diabetes mellitus, coronary artery disease, and digoxin use predicted postoperative myocardial ischemia


Source Citation

Browner WS, Li J, Mangano DT, for the Study of Perioperative Ischemia Research Group. In-hospital and long-term mortality in male veterans following noncardiac surgery. JAMA. 1992 Jul 8;268:228-32.


Abstract

Objective

To determine the risk factors for in-hospital and long-term all-cause mortality in men having major noncardiac surgery.

Design

Cohort study with a follow-up of 2 years.

Setting

Department of Veterans Affairs tertiary care hospital.

Patients

474 consecutive men (mean age, 68 y) with definite coronary artery disease or at high risk for coronary artery disease who were scheduled for major noncardiac surgery requiring general anesthesia. Patients with cardiac pacemakers were excluded. 6 patients were lost to follow-up.

Assessment of risk factors

Demographic and clinical data including cardiac history, cardiac risk factors, and medication use were collected preoperatively by chart review and clinical evaluation.

Main outcome measures

In-hospital and long-term mortality. Cause of death was determined by 2 investigators blinded to baseline data.

Main results

26 patients (5%) died during the in-hospital period, most commonly of sepsis (n = 6) and cardiac diseases (n = 6). A history of hypertension (odds ratio [OR] 3.7, 95% CI 1.1 to 13.0), severely limited activity level (OR 9.7, CI 2.5 to 37.0), and a creatinine clearance < 0.83 mL/s (< 50 mL/min) (OR 6.8, CI 2.8 to 16.0) were all independently associated with an increased risk for in-hospital mortality. Patients with ≥ 2 risk factors had a mortality of 20%, 8 times greater (CI 3.6 to 16.0) than the mortality in patients with ≤ 1 of the risk factors. 82 of the 442 patients (19%) followed over the next 2 years died. Independent preoperative predictors of long-term mortality were surgery for cancer (hazard ratio 2.7, CI 1.7 to 4.2), creatinine clearance < 0.83 mL/s (hazard ratio 2.4, CI 1.5 to 3.8), history of congestive heart failure (hazard ratio 2.5, CI 1.5 to 4.1), and use of bronchodilator medication (hazard ratio 2.0, CI 1.1 to 3.6). Mortality increased with the number of conditions present (P < 0.001).

Conclusions

A history of hypertension, severely limited activity, and a creatinine clearance < 0.83 mL/s were associated with an increased risk for in-hospital mortality after noncardiac surgery in men with, or at high risk for, coronary artery disease. Surgery for cancer, creatinine clearance < 0.83 mL/s, a history of congestive heart failure, and use of bronchodilator medication were associated with a 2-year mortality.

Source of funding: National Institutes of Health.

For article reprint: Dr. W. S. Browner, General Internal Medicine Section, Veterans Affairs Medical Center (111A1), San Francisco, CA 94121, USA.


Commentary

These 2 of a series of important Veterans Affairs studies examined predictive factors for postoperative ischemia and mortality after noncardiac surgery. The study by Hollenberg and colleagues found a high (41%) overall rate of postoperative myocardial ischemia. The strongest predictors were Holter-detected preoperative and intraoperative ischemia followed by 5 clinical predictors. 3 of the 5 (left ventricular hypertrophy on ECG, hypertension, and digoxin use) may be associated with ST-segment depression not necessarily caused by ischemia. Further, only one of these clinical markers, hypertension, predicted postoperative in-hospital mortality in the study by Browner and colleagues.

The possibility of confounding by various cardiac medications exists, although the authors tried to address this issue by stratifying patients into groups based on medication use. Other intraoperative factors that may be important in the development of postoperative ischemia and subsequent cardiac events are not mentioned, such as anesthetic agent, amount of blood loss or fluid shift, pain control, and other intraoperative events.

The study by Browner and colleagues showed an overall 5% (26 of 474 patients) in-hospital mortality, but only 6 deaths had a cardiac cause. They suggest that preoperative evaluation should pay particular attention to patients with hypertension, low creatinine clearance, or severe limitation in physical activity.

Previous studies have shown varying proportions of postoperative mortality due to cardiac causes. Goldman and colleagues (1) found that of the 59 of 1001 patients having noncardiac surgery who died, 19 died of cardiac causes. In contrast, Detsky and colleagues (2) found that 11 of 16 deaths were cardiac in nature in a sample of 524 patients. These mortality rates and causes of death probably differ because of differences in the sample characteristics.

Although many patients in the Browner sample had major surgery associated with a high rate of subsequent cardiac events (e.g., vascular and orthopedic surgery), their cardiac mortality rate was only 1%, perhaps reflecting improved patient selection and perioperative monitoring in recent years. This study also found that late mortality (up to 2 years after surgery) was predicted by cancer, decreased renal function, congestive heart failure, or bronchodilator use.

The results of both studies should be interpreted with caution. Patients were uniformly male, veterans, and had a definite or high likelihood of cardiac disease. All these factors affect the generalizability of the results.

Early postoperative ST changes appear to be quite important in predicting subsequent ischemic events. These patients may benefit from intensified medical management and consideration of further cardiac evaluation. Major questions remain, however, about the identification of these patients. Who should be monitored; how should they be monitored; and what is the cost-effectiveness of routine preoperative ST-segment monitoring? Further, the value of predicting noncardiac death will remain unclear until additional studies confirm these results and define the implications of attempting to modify risk factors.

Sumita D. Paul, MD, MPH
Nicholas I. Kondo, MD
Kim A. Eagle, MD
Massachusetts General HospitalBoston, Massachusetts, USA

Sumita D. Paul, MD, MPH
Massachusetts General Hospital
Boston, Massachusetts, USA

Nicholas I. Kondo, MD
Massachusetts General Hospital
Boston, Massachusetts, USA

Kim A. Eagle, MD
Massachusetts General Hospital
Boston, Massachusetts, USA


References

1. Goldman L, Caldera DL, Nussbaum SR, et al. N Engl J Med. 1977;297:845-50.

2. Detsky AS, Abrams HB, McLaughlan JR, et al. J Gen Intern Med. 1986;1:211-9.