Perioperative myocardial infarction in peripheral vascular surgeryBMJ 1996; 312 doi: http://dx.doi.org/10.1136/bmj.312.7043.1396 (Published 01 June 1996) Cite this as: BMJ 1996;312:1396
- N Mamode, registrara,
- R N Scott, senior registrara,
- S C McLaughlin, research assistantb,
- A McLelland, top grade biochemistc,
- J G Pollock, consultant vascular surgeona
- a Peripheral Vascular Unit, Glasgow Royal Infirmary, Glasgow G31 2ER
- b Department of Medical Cardiology, Glasgow Royal Infirmary
- c Department of Biochemistry, Glasgow Royal Infirmary
- Correspondence to: Mr Mamode.
- Accepted 13 February 1996
The commonest major complication in patients undergoing peripheral vascular surgery is perioperative myocardial infarction.1 No study in Britain has prospectively assessed this risk, but a recent retrospective study found an incidence of 6.3% in patients undergoing aortic surgery.2 We report the incidence of perioperative myocardial infarction in our unit, which is a regional centre for peripheral vascular surgery.
Patients, methods, and results
We studied consecutive patients undergoing peripheral vascular surgery after excluding those who were undergoing surgery for trauma, venous surgery, and minor procedures. Patients thought to be at high risk of perioperative myocardial infarction were referred to a cardiologist for further preoperative assessment. Concentrations of creatinine kinase MB isoenzymes were measured (by Imx STAT) for the first three days after surgery, and electrocardiograms were recorded on admission, on discharge from hospital, and at doctors' discretion. Because the concentration of creatinine kinase MB isoenzyme may be raised by skeletal muscle ischaemia, myocardial infarction was diagnosed only when a raised total creatinine kinase concentration was associated with an MB subunit concentration of > 10 ng/ml and a ratio of MB subunit to total creatinine kinase of >/=5%. Cardiac death was defined as death unequivocally related to myocardial infarction, heart failure, or arrhythmia. The primary end points of our study were myocardial infarction or cardiac death within 30 days of surgery. We performed statistical analysis with the χ2 test or, when appropriate, Fisher's exact test.
The 191 patients included in our study (128 men, median age 65, and 63 women, median age 70) underwent 204 operations, of which 100 were elective procedures, 70 were urgent (requiring surgery within the same hospital admission), and 34 were emergency (requiring surgery within 24 hours). Ninety one of the operations were carried out for critical ischaemia, 32 were aortic procedures, and 120 patients (of the 182 in whom data were available) had preoperative evidence of ischaemic heart disease.
The overall incidence of perioperative myocardial infarction and cardiac death was 7.3%—6% for those undergoing elective or urgent procedures and 12% for those undergoing emergency surgery. Table 1 gives details of the six myocardial infarctions and eight cardiac deaths that occurred. Four of the non-fatal infarctions were clinically silent. Two patients had surgery performed under regional anaesthesia (amputation and brachial embolectomy), while the rest had a general anaesthetic. Three patients were submitted to necropsy, which showed subendocardial infarction in two patients and transmural infarction (probably preceding surgery) in the other. The factors associated with perioperative myocardial infarction were age over 70 χ2=4.642, P=0.03), sex (χ2=3.989, P=.05), perioperative shock (Fisher's exact test χ2=6.949, P=0.05), and angina (Fisher's exact test χ2=7.317, P=0.008).
This is the first prospective study in Britain to assess the incidence of perioperative myocardial infarction in patients undergoing peripheral vascular surgery. Infarction is clinically silent in about 30% of patients,3 as we found, and is fatal in 50%.4 Perioperative cardiac events were not restricted to patients obviously at high risk: myocardial infarctions occurred in patients undergoing elective non-aortic surgery as well as in those undergoing repair of ruptured aortic aneurysm. We found angina to be a strong predictor of risk, but previous myocardial infarction was not. Reviews have confirmed age, diabetes, and heart failure at operation as the only consistent predictors of risk,5 but larger studies with better design might shed more light on this.
We found an overall rate of perioperative myocardial infarction of 7.3%, which remained substantial at 6% if we excluded patients undergoing emergency surgery. This implies that further attempts at risk stratification are justified. We urgently require studies to ascertain the best methods of preoperative stratification in order to minimise the risks of vascular surgery.
Conflict of interest None.