Traditional predictors of in-hospital mortality after coronary artery bypass grafting: Current status

The most serious early adverse events after coronary artery bypass grafting (CABG) include in-hospital mortality, low cardiac output, perioperative myocardial infarction, renal dysfunction, stroke, atrial fibrillation and sternal wound infection. Prediction of outcome after CABG has a considerable importance which constitutes guidance for proper perioperative care and improved postoperative outcome. Advances in anesthesia and surgical techniques play a role in enrollment of patients with higher risk for CABG which indicates periodic updating of the traditional predictors of outcome. This review tries to highlight the current status of the important preoperative predictors of early mortality and morbidity after CABG. Correspondence to: Yasser Ali Kamal, Department of Cardiothoracic Surgery, Minia Cardiothoracic University Hospital, El-Minya El-Jadeda, 61768, Minia, Egypt, Tel: 002-01123234736; Fax: 20-086 2342501; E-mail: yaser_ ali_kamal@yahoo.com


Introduction
Awareness of perioperative risk factors associated with mortality and morbidity after cardiac surgery is an important issue that may provide valuable insights on areas to focus for better outcome and improved quality of care [1].
Research for perioperative predictors of outcome results in establishment of multiple risk stratification models, of which European system for cardiac operative risk evaluation (EuroSCORE) and Parsonnet score (Table 1) are the most frequently used models [2,3].
In the recent years more, complex preoperative patient profile had been referred for coronary artery bypass (CABG) including older patients, more advanced and diffuse coronary artery disease, impaired left ventricular function, failed previous percutaneous manipulations, more serious comorbidities, and multiple reoperations [4].
The increasing complexity and disability of patients is an important reason for periodic update of the perioperative predictors to account for the changes in patient characteristics and contemporary surgical techniques [5,6]. The aim of this review is to highlight the current status of the established preoperative predictors of adverse outcome after CABG.
The impact of obesity on early mortality after CABG remains uncertain [16]. Obesity has not been associated with increased inhospital or 3 months mortality [17], while morbid obesity was recognized as an independent predictor for late mortality after CABG [18].
Compared to non-obese patients, overweight and obese individuals have similar early mortality rate following CABG which can substantiate the presence of obesity paradox only in terms that elevated BMI patients have comparable outcome with non-obese [19].

Diabetes mellitus
Patients with diabetes and CAD often have more extensive and complicated atherosclerosis, thus they are at higher risk of developing major adverse events and death after CABG with prolonged hospital stay and greater hospital costs than are patients without diabetes [20].
Diabetes on insulin treatment is considered as a predictor of outcome after cardiac surgery [21]. Patients with insulin treated DM have a significantly higher rate of mortality and major adverse events compared to patients with non-insulin treated DM patients after CABG [

Chronic obstructive pulmonary disease
Chronic obstructive pulmonary disease (COPD) is traditionally established as a predictor of postoperative morbidity and mortality after CABG surgery, especially in patients older than 75 years with severe COPD and receiving steroids [28].
The known postoperative complications of COPD include respiratory failure, re-intubation, sternal dehiscence, prolonged Requiring opening of the pericardium 3 First 5

Second 10
Another heart procedure at the same time as CABG 2 CABG at time of valve surgery 2 Surgery on thoracic aorta for disorder of ascending, arch or descending aorta 3 Aortic surgery 5 Post-infarction VSD 4 LV aneurysm 5 Aortic surgery and aortic gradient ≥ 120 mmHg 7

Extra-cardiac arteriopathy
Evidence from recent studies suggests that peripheral vascular disease (PVD) still adversely affects the short-and long-term outcomes after CABG [31]. However, diversity of symptoms as well as the severity and location of occluded vessels may influence outcome [32].
Carotid artery disease is considered as an epiphenomenon that serves as a marker for diffuse systemic atherosclerotic disease [33]. Extracranial carotid artery stenosis (CAS) is a risk factor for perioperative stroke in patients undergoing CABG surgery [34]. In patients with 50% to 80% stenosis of the carotid arteries, the incidence of stroke after CABG ranged from 3% to 10%, and approached 22% in patients with >80% stenosis [35].

Neurological dysfunction
Preoperative neurological events are determined as risk factors for postoperative neurologic events and in-hospital mortality particularly with cardiopulmonary bypass [36].
Preoperative stroke was reported to be associated with mortality, increased risk of early and late postoperative stroke, and prolonged length of stay [37]. However, optimal time interval between stroke and surgery has a debate, as there is no contemporary evidence that more recent preoperative stroke predisposed patients undergoing CABG to have adverse perioperative outcomes [38].

Chronic renal impairment
Preoperative renal impairment is a well-established predictor of adverse events in after CABG [39]. However, when compared with expected survival, only advanced renal impairment led to worse outcome, and patients with moderate renal impairment had an outcome similar to that expected [40].
Serum creatinine is a main predicting factor in the scoring systems for risk estimation with a significant impact on surgical outcome after cardiac surgery [41]. For a better estimate of kidney failure degree, current risk scores, such as EuroSCORE II, have included creatinine clearance (CrCl) calculation rather than serum creatinine to predict mortality after cardiac surgery [42,43].
Cystatin C which is less affected by dietary protein intake has been shown to have a greater predictive power than serum creatinine for overall mortality after elective CABG [44].

Left main stem disease
A significant Left main stem (LMS) disease, defined as reduction of ≥ 50% of the vessel diameter, is an important predictor of cardiopulmonary morbidity and mortality after CABG [45]. Multiple clinical variables have been shown as prognostic factors in LMS lesion, including: cardiogenic shock, ST-elevation MI, reduced LVEF, older age, multi-vessel disease, and diabetes mellitus [46].

Severity of angina
High Canadian Cardiovascular Society (CCS) class of angina before CABG is still determined as an independent predictor of adverse outcome and quality of life after CABG [47].
Acute coronary syndrome (unstable angina and non-ST-segment elevation myocardial infarction) is determined as an independent predictor of early mortality, however the long-term outcomes after CABG are similar between patients with acute coronary syndrome and stable angina [48].

Perioperative myocardial infarction
Perioperative myocardial infarction (PMI) is a well-known cause of morbidity and mortality after CABG surgery [49]. Ninety days threshold is traditionally accepted as minimal time interval between MI and CABG [20]. However, a recent study showed that patients operated on 1 to 2 days and 3 to 7 days after MI had a similar mortality rate [50].
It has been suggested that CABG may be safely performed in non-ST elevation myocardial infarction (NSTEMI) patients at any time after the first 6 hours of the event in patients with cardiac troponin I (cTnI) <0.15 ng/ml [51]. The contemporary findings show that it is safe to wait until cTnT levels decrease to the 1 ng/ml threshold value in cTnT positive patients having a stable course [52].

Urgency of CABG
Urgency of CABG (urgent, emergent or salvage operation) was determined as a significant predictor of outcome after CABG in Euroscore II risk model [53]. Urgency is associated with significantly higher adverse outcomes, including death, when compared with elective CABG [54].
Early mortality in patients undergoing emergent and salvage CABG is substantial, especially in salvage patients. Long-term survival is acceptable in both emergent and salvage patients [55].

Low left ventricular ejection fraction (LVEF)
Preoperative low LVEF is still an important predictor of mortality and morbidity after CABG [56,57]. However, surgery remains superior to medical therapy alone in patients with low LVEF [58,59]. Accurate selections of patients and risk/benefit evaluation, as well as planning of surgical and anesthesiological management are mandatory to improve outcome [60].

Conclusions
Predicting outcome after CABG is multifactorial. In the current literature, traditional predictors of adverse outcome are subjected to isolated or combined revision and validation. The cut-off value of age as a predictor of worse outcome does not be stand at 60 years. The predictive value of female gender is further investigated in respecting of gender difference in preoperative comorbidities. A debate exists regarding the impact of obesity on early outcome after CABG. Not only suffering from DM, but also elevation of preoperative HbA1c has a useful predictive value. The role of COPD in predicting outcome is related to presence of other risk factors especially older age and steroid therapy. Carotid artery stenosis > 50% remains a significant predictor of postoperative stroke. The optimal time interval between preoperative stroke and CABG is still under debate. The predictive role of serum creatinine is questionable in comparison to creatinine clearance and Cystatin C. Left main stem disease ≥ 50%, severity of angina, perioperative MI, urgency of surgery and low preoperative LVEF are still recognized as important predictors of adverse outcome after CABG. Timing between NSTEMI and CABG is related to reduction of the level of cardiac troponin. In patients with low LVEF, proper selection of patients for surgery and myocardial viability still has a