One lung anaesthesia in a patient with Fontan circulation for video-assisted thoracoscopic lung resection

INTRODUCTION

In normal heart physiology, the systemic and pulmonary circulations are supported by the two ventricles. A single ventricular physiology is observed in congenital abnormalities such as underdeveloped right ventricle (e.g., tricuspid atresia) and left ventricle (e.g., hypoplastic left heart syndrome), pulmonary atresia with intact ventricular septum, double inlet left or double outlet right ventricle, and complete atrioventricular septal defects.^[1] Surgical palliation in such patients was described in 1971 by Francis Fontan and Eugene Baudet,^[2] where a total cavopulmonary connection (TCPC) is created without a subpulmonary ventricle to pump blood into the pulmonary artery. In a Fontan physiology, the pulmonary blood flow is a passive, low-energy, sluggish circulation that is heavily dependent on favourable haemodynamics and elevated systemic venous pressure to overcome the pulmonary vascular resistance (PVR). Abnormalities in the systemic venous compartment (e.g., hypovolaemia) or high PVR (hypoxaemia, hypercarbia, acidosis, hypothermia, high intrathoracic pressure, and mechanical compression or looping of the pulmonary artery during thoracic surgery) adversely impact the cardiac output.

Moreover, pulmonary blood flow can be compromised by atrioventricular valve dysfunction, elevated systemic ventricular end-diastolic pressure, or arrhythmias. Till now, to the best of our knowledge, there are only four reports in the English language of a patient with Fontan circulation undergoing one lung ventilation (OLV).^[3,4,5,6] Our case should add valuable inputs in the perioperative preparation and implications of one lung anaesthesia in such patients.

CASE REPORT

A 15-year-old girl weighing 42 kg presented to our institute with haemoptysis for 6 months. She was a case of complex cyanotic congenital heart disease (univentricular heart, double outlet right ventricle, severe infundibular and valvular pulmonary stenosis, d-malposition of great arteries). She had undergone coil embolisation of major aortopulmonary collateral arteries, and subsequently a pulsatile bidirectional Glenn with atrial septectomy and division of the azygous vein performed at the age of 6 years. Thereafter, she was lost to follow-up for 7 years and finally presented with increasing cyanosis and dyspnoea on exertion for 1 year. She underwent extracardiac fenestrated Fontan with main pulmonary artery ligation at the age of 14 years. However, for the last 6 months, she started having occasional haemoptysis. On evaluation, her computed tomography (CT) chest showed cavitation with an internal solid component in the left upper lobe, suggestive of a tiny pseudoaneurysm or aspergilloma which after surgical resection, was confirmed to be a fungal ball [Figures 1a-b]. The patient was planned for video-assisted thoracoscopic lung resection, possibly left upper lobectomy. Her peripheral oxygen saturation (SpO₂) was 94% on room air, and there was no cyanosis or clubbing. Routine preoperative investigations were within normal limits. Her medications included tablet aspirin 75 mg, furosemide, and spironolactone 20 mg once daily. The electrocardiogram showed sinus rhythm with a heart rate of 75/min. 2D echocardiography was suggestive of situs solitus, well-functioning Fontan pathway, ejection fraction 45%, antegrade flow across pulmonary outflow tract, and a peak gradient of 80 mmHg. CT angiography was done, which confirmed the normal functionality of shunts. A polytetrafluoroethylene graft was seen from the inferior vena cava to the undersurface of the proximal right pulmonary artery (RPA) [Figure 2a-d]. No pulmonary venous stenosis was noted. RPA was of good calibre (29 × 24 mm at origin) with good arborisation. Left pulmonary artery (LPA) was, however, diffusely small in calibre (9x9 mm at origin) with suspicion of poor left upper lobe arborisation [Figure 2e]. Inhaled nitric oxide (iNO) and an extracorporeal membrane oxygenator were kept on standby in the operating room. Standard American Society of Anesthesiologists monitors were attached. Syringe pumps were loaded with noradrenaline, vasopressin, and pulmonary vasodilator (milrinone) infusions. Sequential pneumatic compression devices were started preinduction for their beneficial role in maintaining a forward blood flow and preventing venous stasis and thrombosis. An invasive radial artery cannula was secured on the right side under local anaesthesia and connected to the FloTrac sensor and Vigileo™ monitor system (Edwards Lifesciences, Irvine, CA, USA). Preoxygenation was done to achieve an end tidal oxygen (EtO₂) of >90%. Anaesthesia was induced with intravenous fentanyl 50 µg, propofol 1.2 mg/kg, and atracurium besylate 30mg was administered to achieve neuromuscular blockade. The trachea was intubated with a left-sided double-lumen endobronchial tube size 28F. A 7F triple-lumen central venous line was secured in the left femoral vein, and an arterial line was secured in the right radial artery. She was positioned in the right lateral decubitus, and OLV was commenced. Protective lung ventilation was titrated to maintain normal partial pressure of oxygen (PaO₂), partial pressure of carbon dioxide (PaCO₂), end-tidal carbon dioxide (EtCO₂), SpO₂, peak airway pressure < 32 cm H₂O, and plateau pressure <25 cm H₂O. The baseline central venous pressure recorded was 11 mmHg, which increased during OLV to a maximum reading of 14 mmHg. The target was to keep it within 20% of the baseline value throughout, as the Fontan circulation was preserved at that value. The baseline stroke volume recorded was 25-30 ml and remained within 10% of this value throughout. The baseline SVR was a little high at 1250 dynes•sec/cm^-5 and remained between 1100 and 1300 dynes•sec/cm^-5 throughout. Intraoperative analgesia was provided with intravenous fentanyl intermittently (total 150 µg), paracetamol 800 mg, and diclofenac sodium 50 mg. Surgical resection of a wedge of the apicoposterior segment of the left upper lobe containing the cavitary lesion was performed with a stapler. Local anaesthetic skin infiltration was done with 0.2% ropivacaine 20 ml. Following 120 minutes of surgery, including 70 minutes of one lung anaesthesia, the muscle relaxant was reversed with intravenous glycopyrrolate 0.008 mg/kg and neostigmine 0.05 mg/kg. The trachea was extubated on the table. She was shifted to the intensive care unit for monitoring in a haemodynamically stable condition without any inotropes. For postoperative analgesia, we used intravenous paracetamol 800 mg 8 hourly, diclofenac sodium 50 mg 12 hourly, and intravenous tramadol 50 mg for rescue analgesia.

Close

Figure 1:

(a): Computed tomography image showing left upper lobe cavitary lesion with aspergilloma/pseudoaneurysm (black arrow), (b) Resected specimen with the cavity and fungal ball

Export to PPT

Close

Figure 2:

CT image showing the Glenn and Fontan shunts. LPA is hypoplastic. (a): CT images showing Glenn and Fontan shunts, (b): CT image of the single ventricle physiology, (c) CT cross sectional image showing the atrial septal defect (ASD) and the common ventricle, (d) The Glenn and Fontan shunts, (e) CT reconstruction image showing the Glenn and Fontan shunts and hypoplastic left pulmonary artery. CT: Computed tomography, RA: Right atrium, LA: Left atrium, LPA: Left pulmonary artery, RPA: Right pulmonary artery, MPA: Main pulmonary artery

Export to PPT

DISCUSSION

Fontan patients have a much higher perioperative complication rate (31%).^[7] Restrictive lung physiology, pulmonary hypoplasia, pleural effusion, and plastic bronchitis (airway casts from leakage of proteinaceous material) are common in Fontan patients. A hypercoagulable state develops owing to deficiencies of protein C, protein S, and antithrombin III^[8], augmented by sluggish systemic and pulmonary circulations and atrial arrhythmias. Protein-losing enteropathy^[9], seen as a late complication, can cause hypoproteinaemia and hypoalbuminaemia, which impacts the protein binding of anaesthetic drugs. High systemic venous pressure-induced congestive hepatopathy with decreased perfusion can impair liver function and drug metabolism.^[10] By the third decade of life, 50% of patients develop a reduction in glomerular filtration rate and albuminuria. Pulmonary blood flow distribution can be helpful in estimating the impact of OLV on haemodynamics and pulmonary vascular pressures. Our patient had hypoplastic LPA, and that was the likely reason she did not have much of an effect of OLV on her Fontan circulation and haemodynamics. OLV and lateral decubitus position, as in our patient, can increase ventilation-perfusion mismatch and PVR due to hypoxic pulmonary vasoconstriction.^[6,11] The challenge was to maintain an optimal transpulmonary gradient (mean pulmonary artery pressure – common atrial pressure) to achieve reasonable pulmonary blood flow, systemic ventricular preload, and cardiac output. In the absence of guidelines, the authors recommend the use of vasopressin as the vasopressor of choice, as it has minimal effect on the PVR as compared to norepinephrine. The inodilator agent milrinone is also useful for its lusitropic and pulmonary vasodilatory properties. Hypothermia should be avoided as it increases the PVR.^[12] On-table tracheal extubation and restoration of spontaneous negative pressure ventilation are desirable for maintaining preload and cardiac output. As a part of a multimodal postoperative pain management strategy, acetaminophen, nonsteroidal anti-inflammatory drugs, opioids, peripheral nerve, and fascial plane blocks are used, as was done in our patient.

CONCLUSION

Successful one-lung anaesthesia in a patient with Fontan circulation requires a thorough preoperative evaluation, a detailed understanding of the physiology, functional status of the shunts, pulmonary blood flow distribution, and calibre of the pulmonary arteries. Our case report highlights the feasibility of one lung anaesthesia in patients with Fontan physiology and also the importance of targeting normal PVR and transpulmonary gradient.