Helium/Oxygen Breathing Improved Hypoxemia After Cardiac Surgery

Helium possesses a low specific gravity, has a coefficient of diffusion twice that of nitrogen and is half as soluble [1]. The use of a helium-oxygen mixture (He/O2) has been reported to result in the temporary relief of major airway obstruction or chronic obstructive lung disease, including asthma [2,3], by reducing turbulent flow with a concomitant reduction in flow resistance. Patients undergoing open heart surgery can develop increases in extravascular lung water and microatelectasis, resulting in a compromised pulmonary mechanism and gas exchange. These effects are usually reversible by increasing the level of positive end-expiratory pressure (PEEP) to 15-20 cm H2 O (high PEEP). However, high PEEP has undesirable cardiovascular effects in patients with severely deteriorated circulation. We hypothesized that the substitution of helium for nitrogen would effectively resolve the abnormalities in pulmonary gas exchange. After institutional approval and informed consent, three hypoxemic postcardiac surgical patients were ventilated with He/O2 after failing to improve sufficiently with the usual therapeutic techniques. Case Reports Case 1 A 60-yr-old, 154-cm, 57-kg woman had a 20-yr history of rheumatic mitral stenosis combined with aortic stenosis and insufficiency. She was New York Heart Association functional class III on admission, and preoperative respiratory function tests showed mild peripheral obstruction (FEV1 = 77%). She underwent mitral valve commissurotomy, aortic valve plasty, and a Maze operation. Cardiopulmonary bypass (CPB) time was 4 h 12 min. During weaning from the CPB, her arterial blood oxygenation was poor (PaO2 70-80 mm Hg with a FIO2 of 1.0 with inspiration-to-expiration ratio (1:E ratio) 0.75, tidal volume 10 mL/kg, and respiratory rate 14 breaths/min under volume-controlled ventilation without muscle relaxants, which was sufficient to maintain normocapnea) and PaO2 did not improve. After admission to the intensive care unit (ICU), her arterial blood oxygenation failed to improve despite vigorous therapy, including nebulization of a bronchodilator, an I:E ratio of 0.75, and PEEP of up to 10 cm H2 O. Use of high PEEP (15-20 cm H2 O) was avoided in order to prevent hemodynamic deterioration. (Her cardiac index [CI] and central venous pressure were 2.0-2.3 L centered dot min-1 centered dot m-2 and 9-13 mm Hg, respectively, with support by intravenous dopamine and dobutamine at a rate of 5 micro gram centered dot kg-1 centered dot min-1 each.) Bronchial secretion was not pathologic, and her chest radiograph showed no significant change Figure 1. On the second postoperative day, when PaO2 had decreased to 56 mm Hg with a FIO2 of 1.0 and the intrapulmonary shunt fraction (Qs/Qt) was calculated to be 35%, He/O2 was initiated. Helium (purity, 99.9%) was connected to the compressed-air port of a Servo-ventilator 900-C. The required FIO2 decreased and the fraction of helium thus increased: PaO2 was 50 mm Hg with a FIO2 of 40.5 and Qs/Qt of 31% some 15 min after the initiation of He/O2Figure 2. PaCO2 was slightly decreased (42 right arrow 37 mm Hg). Peak inspiratory pressure (PIP, 24 right arrow 24 cm H2 O), CI (2.4 right arrow 2.5 L centered dot min-1 centered dot m-2), and a systolic systemic blood pressure (127 right arrow 129 mm Hg) showed no change. After 2 h of inhaling He/O2, oxygenation remained good and reached the normal range after 12 h.Figure 1: Chest radiographs at the initiation of helium inhalation in Case 1 (left), Case 2 (center), and Case 3 (right).Figure 2: The time course of the effects of He/O2 treatment in Patient 1. Throughout the period shown, the patient received positive end-expiratory pressure (PEEP) of 10 cm H2 O. Note that inhalation of He/O2 improved PaO2/FIO 2 and Qs/Qt.Case 2 A 58-yr-old, 169-cm, 68-kg man had a 20-yr history of hypertension with a 3-yr history of ischemic heart disease. He underwent elective coronary artery bypass grafting which was complicated by intraoperative myocardial infarction; intraaortic balloon pumping was used for severe low output syndrome. The CI was improved from 1.5 to 1.8 L centered dot min-1 centered dot m-2 using balloon pumping and catecholamine support (i.e., dopamine 8 micro gram centered dot kg-1 centered dot min-1 and norepinephrine 0.15 micro gram centered dot kg-1 centered dot min-1). After CPB, arterial oxygenation was unstable: PaO2 varied between 59 and 212 mm Hg with a FIO2 of 1.0 and PEEP of up to 6 cm H2 O. Intraoperative bronchofiberscopy showed no abnormal finding. After admission to the ICU, his arterial blood oxygenation did not improve, despite PEEP of up to 10 cm H2 O with an I:E ratio of 0.75, tidal volume of 10 mL/kg, and respiratory rate of 12 breaths/min under volume-controlled ventilation without muscle relaxant. Use of higher PEEP was avoided in order to prevent further hemodynamic deterioration (CI, central venous pressure, and pulmonary capillary wedge pressure were 2.2 L centered dot min-1 centered dot m-2, 9 mm Hg, and 12 mm Hg, respectively, using dopamine, dobutamine, and norepinephrine of 8, 5, and 0.15 micro gram centered dot kg-1 centered dot min-1, respectively). A pulmonary monitor (CP-100; Bicore Monitoring Systems, Irvine, CA) with a VarFlex flow transducer and SmartCath esophageal balloon catheter were used to monitor PIP and dynamic lung compliance (Cdyn; see Appendix). Six hours later, when PaO2 had decreased to 62 mm Hg with a FIO2 of 1.0 and Qs/Qt was calculated to be 37%, He/O2 was started. Some 20 min after the start of He/O2, PaO2 had increased to 90 mm Hg with a FIO2 of 0.75 and Qs/Qt of 25% Figure 3. PaCO2 was slightly decreased (45 right arrow 41 mm Hg). PIP (18 right arrow 18 cm H2 O), CI (2.1 right arrow 2.2 L centered dot min-1 centered dot m-2) and systemic blood pressure (115 right arrow 117 mm Hg) showed no change. After the cessation of the 6-h inhalation of He/O2, oxygenation remained good. Cdyn also showed an increase, from 52 mL/cm H2 O before the start of He/O2 to 58 mL/cm H2 O at 150 min into the He/O2, and 54 mL/cm H2 O after the cessation of He/O2. PIP (21 cm H2 O) remained unchanged throughout.Figure 3: The time course of the effects of He/O2 treatment in Patient 2. Throughout the period shown, the patient received positive end-expiratory pressure (PEEP) of 10 cm H2 O.Case 3 A 64-yr-old, 148-cm 55-kg woman had a 20-yr history of hypertension with a 10-yr history of ischemic heart disease. Respiratory function testing was not done preoperatively because of her low tolerance to exercise. She underwent elective coronary artery bypass grafting. Just after CPB, her arterial oxygenation and CI (2.8 L centered dot min-1 centered dot m-2) were both normal. After admission to ICU, the required FIO2 gradually decreased to 0.5 and PaO2 was then 110 mm Hg at 4 cm H2 O PEEP. On the first postoperative day her arterial blood oxygenation deteriorated gradually without any obvious cause. It did not improve, despite PEEP of up to 10 cm H2 O with an I:E ratio of 0.75 under controlled ventilation. Immediately before the start of He/O2 inhalation (21.5 h after admission to the ICU), blood gas analysis revealed a PaO2 of 134 mm Hg with a FIO2 of 1.0 and Qs/Qt was calculated to be 28%; PEEP was 8 cm H2 O at a tidal volume of 10 mL/kg, respiratory rate 12 breaths/min, and an I:E ratio 0.6. Twenty minutes after the start of He/O2 inhalation, PaO2 had increased to 152 mm Hg with a FIO2 of 0.8 Figure 4. PaCO2 (41 right arrow 41 mm Hg), PIP (18 right arrow 18 cm H2 O), CI (2.1 right arrow 2.2 L centered dot min-1 centered dot m-2) and systolic systemic blood pressure (115 right arrow 117 mm Hg) showed little or no change. Cdyn also showed an increase from 40 mL/H2 O before He/O (2) to 46 mL/H2 O at 20 min and 50 mL/H2 O at 120 min after the start of He/O2 although PIP (21 cm H2 O) remained unchanged. Even after FIO2 was decreased to 0.5 and PEEP to 4 cm H2 O with light sedation, oxygenation remained good. Synchronous intermittent mandatory ventilation was initiated and 3 h later her trachea was extubated.Figure 4: The time course of the effects of He/O2 treatment in Patient 3. PEEP = positive end-expiratory pressure.Discussion Whether to institute high PEEP in an attempt to rectify against respiratory dysfunction in critically ill patients after open heart surgery is a difficult problem because high PEEP has definite disadvantages, such as it 1) reduces right ventricular function [4], 2) decreases urine output [5], 3) does not always improve ventilation/perfusion (VA/Q) abnormality [6], 4) increases the risk of pulmonary barotrauma, and 5) promotes right-to-left interatrial shunting through a patent foramen ovale [7]. Microatelectasis from 100% oxygen breathing is markedly accentuated by the elimination of nitrogen from the alveolar gas [8]. Theoretically, the substitution of helium for nitrogen (via inhalation of He/O2) should enhance both convective and diffusive forces, effectively recruiting areas of low VA/Q, and thus increasing the alveolar oxygen exchange while reducing the need for breathing 100% oxygen [3]. In the present cases, the improvements in the PaO2/FIO2 index and in Qs/Qt were striking, despite the absence of any potentially confounding additional treatments during the period of helium administration. Just before the administration of helium, it was difficult to improve oxygenation by using nitrogen/oxygen to the extent achieved by 100% oxygen. We believe that, in the situations in which we initiated the administration of helium, substitution of nitrogen for helium would not have allowed us to prevent further deterioration of oxygenation. Furthermore, in our experience, no patients undergoing conventional intervention with high PEEP and 100% oxygen for the treatment of deteriorated oxygenation showed such an immediate improvement in oxygenation as did our present cases, who showed rather rapid improvement after the initiation of helium administration. Concerning the change of respiratory variables, Cdyn increased during inhalation of He/O2 and, even after its cessation, Cdyn remained at the new high level although PIP showed no change throughout. During mechanical ventilation, PIP is determined by the bronchioles larger than 3 mm in diameter (the seventh generation), while Cdyn is a sensitive indicator of changes in the smaller airways which are difficult to detect by measurements of airway resistance. These results suggest that He/O2 may have recruited obstructed small airways and alveoli in our patients who were obese, or who had mild peripheral airway obstruction and a tendency to suffer microatelectasis, and that it would improve VA/Q ratio variance and PaO2[9]. Furthermore, the improvement in oxygenation persisted beyond the period of helium administration. We think that the recruited small airways and alveoli might have remained patent even after the cessation of helium, because the decrease in the oxygen fraction reduced their tendency to be reobstructed. We did not try a reverse switch back to oxygen-nitrogen during the period in which oxygenation was improved because the patients were in a critical condition both in terms of oxygenation and hemodynamics. Helium has some clinical benefits: 1) it is a biologically inert gas, and long-term inhalation of He/O2 at 1 atm produces no deleterious effects; 2) the method of inhalation of He/O2 is simple and safe; 3) helium is not prohibitively expensive (three times the price of oxygen); 4) helium should be effective for the prevention or reversal of absorption microatelectasis which can be easily caused by the inhalation of a high FIO2; and 5) substitution of He/O2 for high PEEP reduces the need for heavy sedation during mechanical ventilation. We conclude that He/O2 is a useful alternative therapeutic option to improve hypoxemia in postoperative patients with low cardiac function who, although they show no obvious physical findings, might suffer from possible microatelectasis which would resist conventional respiratory therapeutics, and who might otherwise need high PEEP with all its attendant risks. Appendix Cdyn is measured at points of no flow at peak inspiration and peak expiration of continuous respiratory cycles. Matching points in the respiratory cycle for tidal volume (VT) and transpulmonary pressure (Ptp) provide the Cdyn equation: Equation 1