Intraoperative Management of Severe Pulmonary Hypertension during Cardiac Surgery with Inhaled Iloprost

A 78-yr-old female patient (height, 1.75 m; weight, 74 kg) presented with a history of syncope and congestive heart failure. Cardiac catheterization revealed severe aortic valve stenosis (aortic valve area, 0.49 cm²; mean pressure gradient 58 mmHg), mitral valve insufficiency (degree II), critical stenosis of the left main coronary artery, impaired left ventricular function, and hypokinesia of the anterior and apical left inferior wall. Furthermore, severe pulmonary hypertension was diagnosed (pulmonary artery pressure, 80/30 mmHg; mean pulmonary artery pressure, 65 mmHg; pulmonary artery occlusion pressure, 45 mmHg).

After the induction of anesthesia with sufentanil and midazolam, anesthesia was maintained with isoflurane and sufentanil. Hemodynamic monitoring consisted of arterial, central venous, and pulmonary artery catheterization. Hemodynamic parameters are presented in table 1. In addition, transesophageal echocardiography (Omniplane II T6210 probe; Sonos 5500, Philips Medical Systems, Best, The Netherlands) was performed intraoperatively. Before CPB, transesophageal echocardiography confirmed the diagnoses obtained by cardiac catheterization and revealed severe RV dysfunction. Detailed echocardiographic data are listed in table 2.

After the induction of anesthesia, nitroglycerin was administered intravenously to decrease PVR; however, the nitroglycerin was not effective (table 1). After sternotomy, PVR increased, probably because of increased RV preload caused by the reduction in intrathoracic pressure. Therefore, we administered 12.5 μg aerosolized iloprost (Ilomedin®; Schering Deutschland GmbH, Berlin, Germany) over 15 min via  a commercially available nebulizer (Aeroneb® Pro; Aerogen Inc., Mountain View, CA) connected to the inspiratory limb of the ventilator circuit. The administration of iloprost significantly decreased pulmonary artery pressure and PVR and was accompanied by an increase in cardiac output. CPB was performed using moderate hypothermia (30°C), and cardioplegic arrest was instituted with 2 l of crystalloid cardioplegia. The patient underwent aortic valve replacement, mitral valve repair, and aorto-coronary bypass grafting to the left anterior descending and circumflex arteries. The duration of ischemia was 140 min. After 80 min of reperfusion, 12.5 μg inhaled iloprost were again administered over 15 min. Weaning from CPB was completed after a reperfusion time of 97 min. Moderate doses of vasoactive agents were administered to achieve adequate hemodynamic parameters. Transesophageal echocardiography showed an improvement in RV-function parameters after CPB: the RV-fractional area change increased from 18% (pre-CPB) to 38% (post-CPB). The patient was transferred to the intensive care unit, and endotracheal extubation was performed 13 h postoperatively.

Impaired RV function is associated with a poor outcome in the surgical and nonsurgical settings. 1,4The mortality of patients with combined arterial hypotension and severe RV dysfunction after CPB (defined as RV-fractional area change < 35%) can reach 86%. 3 

Adequate treatment of RV failure consists of different strategies. The main goal is to decrease RV afterload by using vasodilating agents. The use of intravenously applied vasodilators is limited, as they are not selective to the pulmonary circulation and often cause arterial hypotension. Therefore, the administration of selective pulmonary vasodilators such as inhaled nitric oxide and prostacyclin may be beneficial. 5,6Inhaled prostacyclin seems to be the more favorable agent because of its lack of toxicity, ease of application, and reduced costs. 5Iloprost is the stable carbacyclin derivative of prostacyclin and can be administered intermittently, as the hemodynamic effects of a single dose are sustained for approximately 60–120 min. 7Although the plasma half-life time of intravenously administered iloprost is known (20–30 min), no pharmacokinetic data are available concerning the plasma half-life time and the bioavailability after administration of inhaled iloprost. 8 

Similar to inhaled prostacyclin, inhaled iloprost causes a more pronounced increase in cardiac output and a greater degree of PVR-reduction when compared with inhaled nitric oxide. 7Inhaled iloprost has been successfully used in the long-term therapy of pulmonary hypertension and in the testing of pulmonary vascular responsiveness. 9,10To our knowledge, only three reports are available concerning the use of inhaled iloprost during cardiac surgery, two of them in patients awaiting or having undergone heart transplantation. 11–13 

In the present case, we used inhaled iloprost as part of a stepwise approach to prevent RV failure in a patient with severe pulmonary hypertension undergoing combined valve surgery and coronary artery bypass grafting. Administration of inhaled iloprost before CPB showed that the substance acted as an effective pulmonary vasodilator in our patient. Despite a concomitant decrease in mean arterial pressure and systemic vascular resistance (SVR), iloprost led to a more pronounced reduction of pulmonary artery pressure and PVR, so that the PVR/SVR ratio was remarkably decreased before CPB. During reperfusion, iloprost was again administered. PVR and pulmonary artery pressure were significantly decreased when compared with the preoperative values. However, the PVR/SVR ratio was increased after CPB, which can be attributed to an increase of PVR due to CPB-induced pulmonary vascular injury and to a decrease in SVR. Reduction of SVR after CPB is a well-known phenomenon mainly caused by hemodilution and activation of inflammatory mechanisms by extracorporal circulation. The additional use of milrinone contributed to the decrease in SVR.

We used inhaled iloprost during weaning from CPB as an integral part of the therapy and not as a rescue medication. This is in contrast to other case reports, in which inhaled nitric oxide, prostacyclin, or iloprost were used after RV failure had already occurred. 14,15The most effective dose and the best time for the administration of iloprost are still unknown. We used a dose of iloprost that is within the range described in the literature, 7,11and we administered the second dose before starting the weaning from CPB. Thus, an effective RV unloading could be expected in the immediate post-CPB period. RV failure with the need for an excessive dosage of catecholamines or even for reinstitution of CPB could be avoided. Despite the use of positive inotropic substances and surgical correction of valvular disease, echocardiographic parameters indicated a significant impairment of left ventricular function after CPB, most probably caused by severe myocardial stunning. Thus, it seems unlikely that improvement of RV function was caused solely by the surgical procedure.