Authors (including presenting author) :
Tse KC(1), Yim CF(1)
Affiliation :
(1) Department of Anaesthesia & Operating Theatre Services, Tseung Kwan O Hospital
Introduction :
Transthoracic echocardiography (TTE) is increasingly adopted by anaesthetists as a "point-of-care" test (POCT). This is to distinguish from "comprehensive" (formal) echocardiography, which is usually performed by cardiologist, or by a qualified sonographer and subsequently reviewed by a cardiologist. It often serves to answer a specific clinical question e.g. the etiology of sudden-onset haemodynamic instability, or more in elective situations, the cause of a cardiac murmur. Preoperative TTE, where the patient referral source is the cases seen in a pre-anaesthetic assessment clinic, belongs to the latter category.
Objectives :
- To assess the degree to which the focused TTEs done by anaesthetists agree with formal TTEs performed by cardiologists (the "gold standard")
- To validate the findings or measurements
- To identify the clinical impact of an anaesthetist-led perioperative TTE service on the perioperative management of patients undergoing non-cardiac anaesthesia at a district general hospital.
Methodology :
The audit collects data from the past 5 years (November 2019-October 2024). A number of parameters are compared, including
- agreement on left ventricular ejection fraction (LVEF) between focused and formal TTE,
- agreement on presence of important valvular pathology in particular aortic and mitral stenosis
- agreement on tricuspid valve peak pressure gradient.
Cohen's kappa analysis is used to measure inter-observer agreeability for binary/categorical variables e.g. presence of valvular pathology, while Bland–Altman plot and intraclass correlation coefficients (ICC) are used for continuous data i.e. LVEF and tricuspid valve peak pressure gradient (TV PG).
Result & Outcome :
34 patients who have focused TTEs done in POME by anaesthetists with formal TTEs (performed or reported by cardiologists) within the 1-year timeframe were identified.
The Cohen's Kappa for diagnosing mitral stenosis (κ= 1, p<0.001) and aortic stenosis (κ= 0.87, p<0.001) showed that there was excellent agreement.
There was good agreement for LVEF and TV PG with little to no systematic bias, as indicated by the Bland–Altman plot and intraclass correlation coefficients. ICCs for LVEF and TVPG are 0.61 (95% CI 0.35-0.78, p < 0.001) and 0.7 (95% CI 0.41-0.86, p < 0.001) respectively.
In conclusion, the echocardiograms performed by anaesthetists have close agreement with those performed or reviewed by cardiologists. Our perioperative TTE service helps to avoid delay in surgery due to hospital-wide high demand for echocardiography involving cardiologists, while providing reliable diagnoses to answer clinical questions.
