Preoperative cardiac risk assessment in vascular surgery: risk stratification, novel cardiac biomarkers, and their importance in abdominal aortic aneurysm surgery

Bryce, Gavin John (2011) Preoperative cardiac risk assessment in vascular surgery: risk stratification, novel cardiac biomarkers, and their importance in abdominal aortic aneurysm surgery. MD thesis, University of Glasgow.

Abstract

Major vascular surgery is associated with a substantial risk of cardiovascular events and death. This risk is of increased importance in prophylactic elective open Abdominal Aortic Aneurysm (AAA) repair, where a balance of risk of rupture and postoperative outcome is assessed prior to management decisions. Further, the UK Small Aneurysm Trial has shown that prophylactic repair of an AAA has no survival benefit for the first three years due to the major adverse cardiac event (MACE) rate of 5-15%. There is however no ideal method of predicting this risk.

Cardiac Troponin I (cTnI) is a contractile protein that is a highly sensitive and specific marker of myocardial necrosis. A few case reports have commented on the finding of preoperative asymptomatic elevated cTnI levels and poor outcome in a small number of patients undergoing major vascular surgery. There are however no studies looking at its incidence in the vascular surgical population or its utility as a preoperative marker.

Several studies have noted that B-type natriuretic peptide (BNP), a diagnostic and prognostic marker of heart failure, may have a role in predicting MACE in settings including major vascular surgery. There are no studies that have investigated this role in AAA repair alone.

The aim of this thesis is to investigate the incidence of, and to determine a possible role for, preoperative elevated cTnI in major vascular surgery. The further aim is to determine if a single preoperative BNP level correlated with MACE and all-cause mortality in elective open AAA repair in both the short and long-term. Comparisons to current accepted risk indices in AAA, and a possible role for BNP in EndoVascular Aneurysm Repair (EVAR) will also be investigated.

Patients were recruited in two cohorts:
Firstly, a prospective, 2 year observational single centre cohort study of all patients undergoing a vascular procedure, with an expected cardiac event rate >5%, recruited patients who had no clinical or ECG evidence of myocardial ischaemia. Preoperative cTnI was performed in all and postoperative screening (clinical assessment, ECG and cTnI) for cardiac events was performed at days 2, 5 and 30.
213 patient were recruited, of whom 11 (5.2%) had an asymptomatic elevated preoperative cTnI (>0.02 ng/ml). Eight of these patients proceeded directly to theatre, and 2 were delayed but later underwent surgery with a persistently elevated cTnI. Of these 10 patients, 5 (50%) died and 4 (40%) suffered MACE. The remaining patient was delayed due to the poor outcome of the preceding patients, and later underwent an uncomplicated aortic bifurcation graft with a normal cTnI level which had been preceded by coronary intervention.

Secondly, a prospective, 2 year observational multi-centre cohort study in the 3 largest vascular units in Glasgow (Gartnavel General Hospital, Glasgow Royal Infirmary and Southern General Hospital) was performed between August 2005 and August 2007, recruiting all patients who were admitted for both elective open AAA repair and EVAR. Preoperative BNP levels were performed and batch analysed at the end of the study. Postoperative screening for cardiac events was performed as described above. Data was collected to allow calculation of risk indices associated with outcome in AAA repair (Glasgow Aneurysm Score [GAS], Vascular physiology only Physiological and Operative Severity Score for enUmeration of Mortality [V{p}-POSSUM], Vascular Biochemical and Haematological Outcome Model [VBHOM], Revised Cardiac Risk Index [RCRI] and Preoperative Risk Score of the Estimation of Physiological Ability and Surgical Stress Score [PRS of E-PASS]). Follow-up was continued to a minimum of 3 years, where possible, with cause of death recorded.
106 of 111 patients were recruited. The median [interquartile range] BNP concentrations in the 16 patients (15%) who suffered immediate postoperative MACE was 206 [118-454] vs 35 [17-61] pg/ml in the remainder (p=0.001). ROC analysis indicated a BNP concentration of 99.5 pg/ml best predicted MACE (area under the curve 0.927), with sensitivity of 88% and specificity of 89%. The BNP in patients who suffered cardiac death was significantly higher than in those that did not (median BNP 496 [280-881] vs 38 [18-84] pg/ml, p=0.043). ROC analysis revealed a cut-off of 448 pg/ml (AUC 0.963), with sensitivity 80%, specificity 100%, positive predictive value 100% and negative predictive value 99%.
Not only did higher values of BNP predict MACE, but it was also found to predict all-cause mortality in the immediate (median BNP 100 [84-521] vs 35 [17-81], p=0.028), intermediate (median BNP 201 [97-496] vs 35 [17-73], p<0.001) and long-term (median BNP 98.5 [58-285] vs 32 [17-71.5], p<0.001) postoperative periods. ROC analysis revealed decreasing BNP levels to predict outcome over time, with a BNP of >60.5 pg/ml (AUC 0.761) found to best predict death at 3 years.
Whilst BNP was found to predict outcome, most risk indices performed poorly. The GAS, VBHOM and RCRI performed poorly and did not predict any outcome measure. V(p)-POSSUM was, however, found to predict all outcome measures (p=0.028, p=0.030, p=0.038 for MACE, cardiac death and all-cause mortality respectively). The PRS component of E-PASS was found to predict MACE (p=0.019) and cardiac death (p=0.017). BNP performed better than any risk index.

During the study period only 40 of 42 patients admitted for elective EVAR were recruited. Of these 40, only 3 suffered a non-fatal MI and 1 died of respiratory failure. BNP was not found to predict MACE or death in this cohort, and due to the small number of patients, and events, no strong conclusions could be drawn.

Whilst preoperative elevated cTnI was found to identify patients that were at an increased risk of both postoperative MACE and death following their major vascular surgical procedure, its use in elective open AAA repair is limited due to infrequent occurrence. Preoperative serum BNP concentration, however, predicted postoperative MACE, cardiac death and all-cause mortality in patients undergoing elective open AAA repair on immediate, intermediate and long term follow-up. Further, BNP performed better than any current risk index for elective open AAA surgery. This simple blood test, therefore, offers valuable information regarding risk stratification of prospective surgical patients and should be considered a part of routine preoperative assessment in this prophylactic procedure.

• Item Type: Thesis (MD)
• Qualification Level: Doctoral
• Keywords: Preoperative cardiac risk, B-type natriuretic peptide, cardiac troponin I, vascular surgery, abdominal aortic aneurysm, preoperative risk scores, risk stratification, cardiac biomarkers, postoperative cardiac events, Glasgow aneurysm score, Vascular physiological and operative severity score for enumeration of mortality, Vascular biochemical and haematological outcome model, Revised cardiac risk index, Estimation of physiological ability and surgical stress score
• Subjects: R Medicine > R Medicine (General); R Medicine > RD Surgery
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cardiovascular & Metabolic Health
• Supervisor's Name: Delles, Dr. Christian
• Date of Award: 2011
• Depositing User: Dr Gavin J Bryce
• Unique ID: glathesis:2011-2628
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 31 May 2011
• Last Modified: 10 Dec 2012 13:58
• URI: https://theses.gla.ac.uk/id/eprint/2628