Background. The safety of epidural anaesthesia in patients at risk for right ventricular
pressure overload remains controversial. We compared the haemodynamic effects of vascular
and cardiac autonomic nerve block, induced by selective lumbar (LEA) and high thoracic epidural
anaesthesia (TEA), respectively, in an animal model subjected to controlled acute right
ventricular pressure overload.
Methods. Eighteen pigs were instrumented with epidural catheters at the thoracic (T) and
lumbar (L) level and received separate injections at T2 (1 ml) and L3 (4 ml) with saline (s) or
bupivacaine 0.5% (b). Three groups of six animals were studied: (i) a control group (LsþTs), (ii)
LEA group (LbþTs), and (iii) TEA group (LsþTb). Haemodynamic measurements including
biventricular pressure-volumetry were performed. Right ventricular afterload was then
increased by inflating a pulmonary artery (PA) balloon. Measurements were repeated after
30 min of sustained right ventricular afterload increase.
Results. LEA decreased systemic vascular resistance (SVR) and did not affect ventricular function.
TEA had minor effects on SVR but decreased left ventricular contractility while baseline
right ventricular function was not affected. Control and LEA-treated animals responded similarly
to a PA balloon occlusion with an increase in right ventricular contractility and heart rate.
Animals pretreated with a TEA did not show this positive inotropic response and developed
low cardiac output in the presence of right ventricular pressure overload.
Conclusions. In contrast to LEA, TEA reduced the haemodynamic tolerance to PA balloon
occlusion by inhibiting the right ventricular positive inotropic response to acute pressure overload

Placement of the PLMA was achieved by threading a gum elastic bougie (Portex-SIMS; Hythe, Kent, UK) into the drain tube, and, under gentle direct laryngoscopy, its distal end was introduced into the esophagus. Positive-pressure ventilation was instituted and, via the multiport adaptor, the BB was passed into the left mainstem bronchus under bronchoscopic guidance, carefully avoiding the tracheal mass. A gastric drainage tube was gently inserted via the drainage port to drain the stomach. Adequate one-lung ventilation (OLV) was achieved after inflation of the BB cuff in the lateral decubitus position. Throughout the period of OLV, peak inspiratory pressures ranged from 30 to 35 cmH2O on pressure-controlled ventilation with tidal volumes of 400 to 450 mL, a respiratory rate of 12, and end-tidal CO2 of 45 to 48 mmHg. Surgical exposure was good, and surgery proceeded uneventfully. Postprocedure bronchoscopy did not reveal any evidence of aspiration in the larynx or tracheobronchial tree.

The patient had a friable tracheal mass close to the vocal cords that precluded the placement of an endotracheal tube and had the potential for tumor trauma. We used a supraglottic airway to mitigate this and chose the PLMA in view of the patient’s aspiration risk, taking advantage of the probable added protection of a gastric drainage tube. Although a reduction in aspiration risk with the PLMA (compared with the CLMA) has not been shown, the drainage tube successfully vented regurgitated fluid and protected against aspiration. 2 In more than 2,000 PLMA placements, there were 12 cases of clinically apparent regurgitation with no clinical or bronchoscopic evidence of aspiration. 2

The measured peak airway pressures of 30 to 35 cmH2O were high probably because of reduced lung compliance. These pressures are within the appropriate “lung-protective ventilation” strategy of limiting the peak inspiratory pressure to <35 cmH2O during OLV. 3 The reported improved seal of the PLMA as compared with the CLMA was to our advantage. Median seal pressures with PLMA are reported to be approximately 30 cmH2O, 4 with pressures exceeding 40 cmH2O in 20% of cases. 5 The airway seal with CLMA is reported to exceed 30 cmH2O in only 4% of cases. 5 Our approach was an effective strategy in the management of this uncommon yet challenging scenario. We suggest that it be considered as part of the armamentarium for such cases requiring lung separation.