Background
Perioperative cardiac complications depend on the patient's risk factors and comorbidities, the type and duration of the surgery. More specifically, postoperative atrial fibrillation (POAF) can manifest after surgical procedures and it affects more than 3% of all >45-year-old patients undergoing non cardiovascular surgery, around 30% of patients undergoing thoracic surgery and 40% of those experiencing cardiovascular surgery. POAF is associated to a higher risk of stroke and mortality.
Purpose
The aim of our study was to assess independent risk factors for POAF incidence in non-cardiac surgery. The secondary endpoint was to evaluate predictors of major cardiovascular events (MACE) which included non-fatal stroke, non-fatal myocardial infarction (MI), transient ischemic attack, decompensated acute heart failure and cardiac-related death.
Methods
Retrospective observational study including all patients attending our outpatient clinic for preoperative assessment from the 1st of January 2016 to the 31st of December 2018. For each enrolled patient we reviewed the clinical data referring to the period of their surgery and took note of all cases of POAF and MACE occurring before discharge.
Results
A total of 1383 patients were enrolled (910 men, age 71.6±12.9). 36 cases of POAF (2.6%), 4 cases of non-fatal MI (0.3%), 8 cases of acute heart failure (0.6%) and 1 cardiac death (0.1%) were registered. Systemic hypertension (HR 3.68, 95% CI 1.12–12.09, p=0.32) and thyroid dysfunction were independent predictor of POAF (HR 2.87, 95% CI 1.32–6.23, p=0.008) as well as age (HR 1.05 per each year, 95% CI 1.01–1.09, p=0.006). The Revised Cardiac Risk Index (RCRI) as defined by current guidelines was not able to predict POAF risk in our population, while low-risk surgery showed a highly protective effect towards POAF (HR 0.07, 95% CI 0.01–0.50, p=0.008). A four items point-based risk score called HALT (hypertension =1; age≥70 years = 1; low-risk surgery = −3; thyroid dysfunction = 1) >0 had an 81% sensitivity and a 61% specificity in detecting POAF in our population (AUC 0.753, figure 1). A rhythm control strategy was attempted in 25 patients (69.4%), mainly through pharmacological cardioversion (24 patients), while only one patient underwent emergency electrical cardioversion due to hemodynamic instability. Out of the 36 POAF events, oral anticoagulation was started in only one (3.1%) case while 33 patients were treated with low-molecular-weight heparin, either at anticoagulation doses (6, 16.7%) or more frequently at sub-therapeutic doses (27, 80.6%). Two patients (5.6%) were treated with aspirin.
Conclusions
POAF complicates 2.6% of all non-cardiac surgery, and is seldom treated as recommended by current guidelines, especially regarding thromboembolic risk reduction. A four items point-based risk score could be effective in implementing POAF screening. Low-risk surgery should not be screened for POAF, as the associated risk is very low.
Figure 1
Funding Acknowledgement
Type of funding source: None
