During cardiac displacement for off-pump coronary artery bypass(OPCAB) surgery, the presence of air underneath the displaced heart compromises the transgastric (TG) window for transoesophageal echocardiography(TOE). The investigators hypothesised that placing a saline-filled glove would enhance TG transmission of ultrasound and facilitate TOE imaging for monitoring left-ventricular regional wall motion (LV-RWM). For left circumflex coronary artery (LCX) grafting in OPCAB surgery (n=13), mid-oesophageal (ME) and TG TOE images are recorded before cardiac displacement (T control), after displacing and stabilising the heart (T-displaced), and after placing a saline bag (saline-filled surgical glove) underneath the displaced heart (T-saline bag). Following data are determined by integrated TG and ME TOE views (ME+TG) at T-control, T-displaced and T-saline bag: number of readable segments (NRS) in a 17-segment model; NRSs in basal and mid-TG short axis views; NRS in 5-LV segments of the LCX territory; the incidence of inadequate monitoring of LV-RWM (NRS < 14/16 except for the apex in 17-segment model).

A multi-plane TOE probe (6T, GE Healthcare, Milwaukee, WI, USA) is inserted immediately after anaesthesia induction. One of authors (anaesthesiologist) performs the routine intraoperative TOE study, including analysis of RWM, using the 17-segment LV model 11-14 and a TOE imaging system (Vivid 7, GE Healthcare, Milwaukee, WI, USA) and the following six TOE images: ME 4-chamber (4-CH), ME 2-chamber (2-CH), ME long-axis (LAX), mid and basal TG short-axis (SAX), and TG LAX.

For the present study, the three-beat clips of two-dimensional (2D) mid-oesophageal (ME) views, including ME 4-chamber (ME 4-CH), ME 2-chamber (ME 2-CH), and ME long-axis (ME-LAX) views, and those of transgastric (TG) views, including the basal TG short-axis (basal TG SAX), mid short-axis (mid TG SAX), and TG-LAX views, are stored on the hard disc of the TOE imaging system as digital recordings at each of the following time points:

(i) T-control: during harvesting the vascular grafts after sternotomy. (ii) T-displaced: after completion of cardiac displacement and stabilisation for the distal graft construction to the LCX territory by stabilisers to maintain both maximal surgical access and minimal haemodynamic instability under the guidance of haemodynamic monitoring.

(iii) T-saline bag: after placing the saline bag underneath the displaced heart and further fine positioning.

Using a program (EchoPAC, GE Healthcare, Milwaukee, WI,), a cardiologist, who is unaware of the aim of the present study, undertook a postoperative off-line analysis of the stored images to determine the following data under T-control, T-displaced, and T-saline bag conditions: 1) NRS out of a LV 17-segment model using ME views only and integrated ME+TG views; 2) NRS out of five LV segments in the LCX territory including basal inferolateral, mid-inferolateral, basal anterolateral, mid anterolateral, and apical lateral segments using ME views only and integrated ME + TG views; 3) NRS out of six LV segments in the TG basal SAX view, including basal anterior, basal anterolateral, basal anteroseptal, basal inferior, basal inferolateral, and basal inferoseptal segment; 4) NRS out of six LV segments in the TG mid-SAX view, including mid-anterior, mid-anterolateral, mid-anteroseptal, mid-inferior, mid-inferolateral, and mid-inferoseptal segments; and 5) the incidence (%) of inadequate RWM monitoring, defined as fewer than 14 out of the 17 segment model (except for the apex) were readable, using the integrated ME + TG views.

Each LV segment is defined as readable/unreadable when more than 50% of the endocardial and epicardial borders were visible/invisible or more than 90% of endocardial border was visible or 10% was invisible.

The primary objectives of the present study are to evaluate the efficacy of placing a saline filled glove in enhancing TOE's ability for global and regional LV RWM monitoring and to determine NRS out of the LV 17-segment model using ME views alone and integrated ME+TG views.

As secondary objectives, NRS out of five LV segments in the LCX territory by using ME views alone vs. integrated ME+TG views as well as NRS out of six LV segments in the TG SAX view are determined, to confirm whether TG TOE imaging can be applied for regional LV RWM monitoring during cardiac displacement. The incidence (%) of inadequate RWM monitoring in 17-segment model using ME views alone vs. integrated ME+TG views is also compared to determine whether the possible increase in NRS facilitates the adequate TOE imaging.

Inclusion criteria:

• three-vessel disease during a week at a tertiary university hospital participated in this prospective observational study after we obtained institutional review board approval (KUH1160037) and written consent.

Exclusion criteria:

• patients with oesophageal pathology (such as spasm, stricture, laceration, perforation, and diverticula),
• diaphragmatic hernia,
• history of extensive radiation to the mediastinum and upper gastrointestinal bleeding are excluded.

Intraoperative exclusion criteria for the study are patients with inadequate RWM monitoring, defined as fewer than 14 out of the 17 segment LV model, except for the apex, are readable using the integrated ME and TG TOE images immediately after TOE probe insertion and patients who required cardiopulmonary bypass due to severe myocardial ischaemia and haemodynamic instability during cardiac displacement.

elective OPCAB surgery patients