Multiple anesthetics in a child with severe pulmonary hypertension
By Chad Cripe MD, and Aruna Nathan, MD
The Children’s Hospital of Philadelphia
At the time of his initial admission to our hospital the patient was a 4-year-old male with severe pulmonary hypertension (PH), supra systemic right ventricular (RV) pressure, significantly elevated pulmonary vascular resistance (PVR) of 9.6 Woods units, and left coronary ostial stenosis. He required multimodal pharmacologic therapy and general anesthesia for multiple cardiac catheterizations and cardiac surgery.
Medical History
The patient was a former 37 week gestation infant, hospitalized at an outside institution for one week after birth for transitional tachypnea of newborn. He underwent evaluation for a murmur diagnosed on day 3 of life. A transthoracic echocardiogram (TTE) revealed near systemic RV pressures, possible bicuspid aortic valve, a moderate patent ductus arteriosus (PDA) with bidirectional shunting, and good biventricular function. The patient was apparently discharged after one week and further follow up was not documented.
The patient presented at four years of age to his primary care physician complaining of abdominal pain, emesis and lethargy. A chest x-ray at that time was significant for an enlarged cardiac silhouette. Subsequently, a TTE revealed supra systemic RV pressure, moderate RV dilation and diminished function, severe main pulmonary artery (PA) dilation and leftward atrial septal bowing. At that time the patient was transferred to our facility for further management.
Cardiac catheterization and anesthetic #1
One day after admission to the cardiac intensive care unit (CICU), the patient underwent general anesthesia for hemodynamic cardiac catheterization with possible device closure of PDA. The patient presented to the cardiac catheterization lab on dopamine (3 mcg/kg/min), oxygen two liters per minute [lpm] and inhaled nitric oxide 40 parts per million [ppm] via nasal cannula. The patient’s weight was 13.1 kg. Midazolam (1 mg) was administered intravenously (IV) for anxiolysis. The standard monitors were placed and the patient was pre-oxygenated with 100% oxygen and 40 ppm iNO. General anesthesia was induced with a second dose of midazolam 1mg IV and ketamine 5mg. Neuromuscular blockade was achieved with pancuronium 0.1mg/kg IV. The trachea was intubated without incident and the patient was continued on iNO 40 ppm and a fractional inspired oxygen concentration [FiO2] of 1. Anesthesia was maintained with isoflurane. Right and left heart catheterizations were performed without complications. The PDA was occluded with an Amplatzer ductal occluder device, without further increase in PA pressure or desaturation. Hemodynamic data obtained before PDA occlusion confirmed the presence of elevated PA pressures (>2/3 systemic), PVR (9.3 WU) and Qp: Qs of 1.2 on dopamine and an FiO2 1.0 /iNO 40 ppm. Isoflurane was discontinued, neuromuscular blockade reversed with glycopyrrolate(0.01mg/kg) and neostigmine (0.07mg/kg) IV, and upon return to consciousness the patient’s trachea was extubated. No adverse events were observed throughout the procedure. The patient was transported uneventfully to the CICU while continuing to receive dopamine, oxygen, and iNO.
Cardiac catheterization and anesthetic #2
The patient remained hospitalized for the next month, in which time dopamine and iNO were weaned off; additional therapies were added including treprostinil (48 ng/kg/min), sildenafil (20 mg every 8 hours) and oxygen (2 lpm) via nasal cannula. The patient presented again for a hemodynamic cardiac catheterization and drug study to assess response to therapy. The patient was preoxygenated with 100% oxygen, and given midazolam 1mg IV, and ketamine 10mg IV which produced loss of consciousness. Anesthesia was maintained with isoflurane and pancuronium 0.1mg/kg IV was followed by tracheal intubation. Oxygen concentration was not reduced prior to diagnostic measurements. Catheterization results revealed continued near-systemic RV pressures and elevated PVR (13 WU on FiO2 0.21 and 10.8 WU on FiO2 0.21 with iNO 40 ppm despite IV treprostinil and sildenafil. A long term central venous line was placed for continuous treprostinil infusion. The patient was extubated without incident following return of consciousness, and reversal of neuromuscular blockade; and returned to the CICU uneventfully.
Cardiac catheterization and anesthetic #3
The patient was discharged from the hospital one day after the last anesthetic while receiving IV treprostininil 54ng/kg/min, sildenafil 20 mg every 8 hours, furosemide 20mg every 8 hours, enoxaparin 0.13ml subcutaneously twice a day, and oxygen 1 lpm at via nasal cannula at night. The patient was then followed at the outpatient pulmonary hypertension clinic for 2 ½ months. He presented with worsening dyspnea, fatigue and abdominal pain. Laboratory and clinical investigation revealed the presence of worsening LV dilation/function with an LV shortening fraction (SF) of 16% and rising brain type natriuretic peptide (BNP) levels of 4206 pg/ml. The patient was found to be PCR + for parvovirus B19. He was admitted to the CICU, receiving 25 grams intravenous immunoglobulin (IVIG) therapy and inotropic support with milrinone (0.5 mcg/kg/min), in addition to the continuation of IV treprostinil infusion (61 ng/kg/min) and sildenafil (20 mg every 8 hours). Approximately 1 week after admission the patient presented for cardiac catheterization. At this time, his BNP levels had dropped to 2850 pg/ml. All medications were continued during the anesthetic. Anesthesia was induced with 2 doses of ketamine (10mg IV), glycopyrrolate (0.01mg/kg) IV and pancuronium (0.1 mg/kg IV), and the trachea intubated after preoxygenation with FiO2 1.0. At the completion of the procedure, neuromuscular blocakade was reversed and the the trachea extubated. No adverse events were observed throughout the procedure. Dexmedetomidine (1mcg/kg IV) was administered for post catheterization agitation over 10 minutes. The patient recovered in the CICU, as the patient was agitated. Hemodynamic variables demonstrated significant left ventricular diastolic dysfunction (LVEDP 10 mmHg, LAP 13 mmHg) in addition to similar elevations in PVR (9 WU) which was unresponsive to vasoreactivity testing. Coronary angiography revealed right and left coronary ostial stenosis with the left ostium and proximal left main being severely narrowed. At the multi-disciplinary case conference, it was believed that ischemia resulting from left coronary stenosis was contributing to worsening LV function, and coronary revascularization might improve chances of survival to transplantation. The patient was listed for a heart-lung transplant. Carvedilol was added to the therapeutic regimen.
Cardiac revascularization and anesthetic #4
Two weeks after the last cardiac catheterization and medical optimization, the patient presented to the cardiac operating room (OR) for repair of left coronary ostial stenosis. Intravenous treprostinil (61 ng/kg/min) was continued throughout the procedure. All cardiac medications including sildenafil, bumetanide and enoxaparin, were continued peri-operatively. Ketamine (1mg/kg) was used for parental separation just outside the OR. Once in the OR, standard ASA monitors were applied. After preoxygenation the anesthetic induction was accomplished with incremental ketamine (20mg) to unresponsiveness, and pancuronium (0.1mg/kg). After the trachea was intubated, additional IV access and arterial line placement was secured. Although the heart rate increased to 170, no evidence of myocardial ischemia occurred. Fentanyl 5 mcg/kg was titrated prior to sternotomy and the patient was maintained on 0.4 MAC isoflurane. The patient tolerated the pre-bypass period without hypotension, desaturation, evidence of myocardial ischemia or arrhythmias. Hemodynamic support was not escalated. Ventilation was controlled to maintain a respiratory alkalosis, and systemic oxygenation with FiO2 of 1.0. Once on cardiopulmonary bypass (CPB) and after cardioplegia administration with diastolic arrest of the heart, an incision was made down the left main coronary ostium. A patch of pulmonary homograft was used to create a large osteoplasty. The patient was weaned from CPB on dopamine 3 mcg/kg/min, a milrinone loading dose of 300 mcg/kg, and the treprostinil infusion. In addition, ventilation was resumed with a FiO2 of 1.0 and iNO at 40 ppm for separation from CPB. The decision was made to keep the patient sedated, paralyzed, and intubated with positive pressure ventilation in the CICU overnight to avoid immediate postoperative lability in the PVR, given the severely elevated pulmonary vascular resistance. The patient was extubated uneventfully the next day. His cardiac function has not improved significantly with persistently severely diminished biventricular function, and ¾ systemic RV pressures. The patient discussed in this case is at home, on treprostinil, sildenafil, carvedilol, bumetanide and oxygen therapy. He is currently listed as status 7 for a heart-lung transplant.
Discussion
Pulmonary hypertension (PH) remains a significant predictor of peri-operative risk of adverse events including cardiopulmonary arrest and cardiovascular morbidity in pediatric cardiac anesthesia1,2. Anesthetic management of a patient with PH can be extremely challenging even with isolated PH. This patient had other significant risk factors namely left coronary ostial stenosis, with ventricular dysfunction (either due to myocardial ischemia, or myocarditis). Cormosino et al reviewed the data from 256 anesthetic exposures in 156 patients with PH undergoing noncardiac surgery or cardiac catheterization. Children with suprasystemic PH were at significant risk of major perioperative complications including cardiac arrest and pulmonary hypertensive crisis irrespective of age, etiology of PH, type of anesthetic or airway management. Taylor et al reviewed perioperative adverse events during cardiac catheterization in 75 patients undergoing cardiac catheterization. Risk was highest in patients with severe idiopathic PH, who were symptomatic with chest pain, syncope or dizziness. These studies quote a risk of 4.5-6% for intra-operative resuscitation or death. Presence of PH has been linked to increased postoperative morbidity and mortality, although the level of pulmonary vascular disease that precludes operability might be both patient and procedure specific3,4.
In the course of his hospitalization, this patient required general anesthesia 4 times in a 5 month period of which 3 were for cardiac catheterization and vaso-reactivity testing. The cardiac anesthesia service was consulted to assist in managing the procedures. We believed that general anesthesia and endotracheal intubation provided the ability to optimally control physiologic variables and overall was the safest approach. We are extremely conservative in our management of patients with PH. Most often, these patients are not pre-medicated, and if they are, it is usually intravenous, under the direct supervision of the attending anesthesiologist with immediate access to oxygen therapy and respiratory support. The rationale is to avoid respiratory depression with either hypoxemia or hypercarbia and the subsequent effects on PVR. Patients are admitted routinely to the cardiac or pediatric ICU for postoperative recovery for 24 hours for ongoing vigilance and monitored for adverse events. Our general approach consists of the cautious use of sedatives in the peri-operative period. The challenges and techniques that can be used in patients with PH have been reviewed by Friesen5. We find that ketamine has some desirable properties as an induction agent in this population i.e effective amnesia, preservation of hemodynamics, and avoidance of adverse events. Although there is an increase in the incidence of tachycardia with our technique i.e induction with ketamine and neuromuscular blockade with pancuronium, the maintenance of higher than normal perfusion pressures, and preservation of systemic vascular resistance, seems to mitigate against myocardial ischemia while preserving RV function. Our preference is an anesthetic technique that allows for early return of spontaneous ventilation and maintenance of a natural airway following the anesthetic. We are aggressive with ventilation managment and prevention of hypercarbia, acidosis or hypoxia. Factors that may elevate PVR acutely are monitored and aggressively corrected. Ventilatory strategies are aimed at preserving RV preload, and reducing RV afterload. If necessary, inotropic support of the RV will be initiated early upon transition to positive pressure ventilation. The availability of iNO and rapid acting vasopressor agents is immediate in the entire peri-operative period. We consider oxygen to be a therapeutic agent in PH, and wean inspired concentrations gradually prior to catheterization if tolerated. Extra-corporeal membrane oxygenation (ECMO) may be required as a rescue mechanism for severely ill children, and is discussed in the pre-procedure planning phase. However, outcome is ill defined for patients with PH receiving ECMO support. Post-operatively, sedation strategies in the ICU have involved lower doses of ketamine, midazolam, or pentobarbital that achieve effective sedation without cardio-respiratory depression. Non pharmacologic measures such as the presence of child-life specialists, parents, distraction techniques, imagery, and music are used during the post operative phase to provide anxiolysis. The key issue has been the maintenance of a patent airway, spontaneous ventilation, with normocapnia, normoxia, and avoidance of acidosis. A drug that offers some promise in the future is dexmedetomidine. In a small set of patients following congenital cardiac surgery, dexmedetomidine has been shown to decrease PA pressure, while maintaining LV function and respiration6. We have used dexmedetomidine occasionally for sedation and management of emergence delirium in children with PH following cardiac catheterization without adverse events.
The anesthetic provided for the coronary revascularization surgery was associated with the highest anesthetic risk. The role the patient’s coronary stenosis played in the development of LV dysfunction or PH remains unclear. Revascularization of the coronary arteries in children is rare carrying significant risk7. Isolated left main coronary ostial narrowing is extremely uncommon. The patient had a bicuspid aortic valve, although not noted on most echocardiograms or catheterizations,may have lead to turbulent flow near the coronary ostial opening. Severe ostial narrowing was noted during an angiogram of the coronary arteries in the face of new onset LV dysfunction. The RV function was also significantly reduced, and the BNP >2000 pg/ml despite maximal therapy. The picture was also complicated by a recent viral infection and concern for myocarditis. Diastolic dysfunction of the LV could have further worsened the lability of the pulmonary vascular bed in this patient. And, the presence of PH with the risk of RV failure increased the risk of inability to separate from cardiopulmonary bypass (CPB). Therefore, we felt that ketamine would be a safer induction agent in this patient, especially to help with parental separation. It is possible that trying to minimize CPB circuit, and time, use of modified ultrafiltration techniques, minimizing exposure to blood and component therapy and optimizing PH therapy could have helped prevent peri-operative surges in PVR. We did not have to escalate PH therapy in the pre CPB phase, and due to a judicious combination of ketamine, low dose fentanyl, and low concentrations of inhalational agent we were able to maintain a balanced general anesthetic without adversely affecting the pulmonary vascular bed or myocardial oxygenation.
References
- Cormosino MJ, Friesen RH, Doran A, Ivy DD. Perioperative complications in children with pulmonary hypertension undergoing noncardiac surgery or cardiac catheterization. Anesth Analg. 2007;104(3):521-527
- Taylor CJ, Derrick G, McEwan A, Haworth SG, Sury MRJ. Risk of cardiac catheterization under anesthesia in children with pulmonary hypertension. BJA 2007; 98:657-661.
- Reich D, Bodian DA, Krol M, Kuroda M, Osinski T, Thys DM. Intraoperative hemodynamic predictors of mortality, stroke, and myocardial infarction after coronary artery bypass surgery. Anesth Analg 1999;89:814-22
- Giglia TM, Humpl T. Preoperative pulmonary hemodynamics and assessment of operability: Is there a pulmonary vascular resistance that precludes cardiac operation? Pediatr Crit Care Med 2010; 11:S57-S69
- Friesen RH, Williams GD. Anesthetic management of children with pulmonary arterial hypertension. Pediatr Anesth 2008; 18:208-216
- Lazol JP, Lichtenstein SE, Jooste EH et al. Effect of dexmedetomidine on pulmonary artery pressure after congenital cardiac surgery: a pilot study. Pediatr Crit Care Med 2010; 11: 589-592
- Borgoend E, Raisky O, Degandt A, Tamisier D, Sidi D, Vouhe P. Myocardial revascularization in infants and children by means of coronary artery proximal patch arterioplasty or bypass grafting: A single institution experience. J Thorac Cardiovasc Surg 2008; 136:298-306
