Usefulness of Nitric Oxide Treatment for Pulmonary Hypertensive Infants during Cardiac Anesthesia

The study was approved by an institutional review committee (Iwate Medical University Memorial Heart Center, Iwate, Japan), and informed consent was obtained for each patient. Fifty infants with congenital heart disease who underwent cardiac catheterization at Iwate Medical University Memorial Heart Center or Nagano Prefectural Children's Hospital from April 1995 to April 2000, were enrolled. The mean pulmonary arterial pressure was 25 mmHg or more.

Cardiac catheterization was scheduled within 2 weeks before surgery. Chloral hydrate (50 mg/kg administered orally), pethidine hydrochloride (1 mg/kg administered subcutaneously), and hydroxyzine hydrochloride (1 mg/kg administered subcutaneously) were given as premedication. Anesthesia was induced by intravenous injection of atropine sulfate (0.02 mg/kg), pentazocine (0.5 mg/kg), and lidocaine (1.0 mg/kg), followed by inhalation of sevoflurane and tracheal intubation. It was then maintained with inhalation of approximately 0.5% sevoflurane at an oxygen concentration of 21% during spontaneous respiration. After anesthesia was stabilized, cardiac catheters were introduced with additional local anesthesia. Peripheral arterial hemoglobin oxygen saturation was monitored continuously with a pulse oximeter (Durapalse PA2100; NEC, Tokyo, Japan), and end-tidal carbon dioxide and sevoflurane concentrations were monitored with capnography (PM8050; Drager, Lubeck, Germany). If the end-tidal carbon dioxide concentration exceeded 50 mmHg, ventilation was assisted according to the result of blood gas analysis. Blood pressure, hemoglobin concentration, and oxygen saturation were measured, followed by calculation of PVR, pulmonary arterial pressure/systemic blood pressure ratio (Pp/Ps), and pulmonary blood flow/systemic blood flow ratio (Q̇p/Q̇s) by the Fick formula:

where Q̇s is systemic blood flow (cardiac index; liters per minute per square meter), Cao²is systemic arterial oxygen content (milliliters per liter), CT̄vo²is mixed venous oxygen content (milliliters per liter), Q̇p is pulmonary blood flow, Cpvo²is pulmonary venous oxygen content, Cpao²is pulmonary arterial oxygen content, PVR is pulmonary vascular resistance (Wood units m²), mPAP is mean pulmonary arterial pressure (millimeters of mercury), and mPCWP is mean pulmonary capillary wedge pressure (millimeters of mercury). Oxygen consumption was estimated from nomograms based on surface area or age and heart rate. 20Oxygen content was calculated by multiplying the hemoglobin oxygen-carrying capacity (13.6 ml O²/l) and hemoglobin oxygen saturation.

When PVR was greater than or equal to 10 Wood units m², lung biopsy was performed during general anesthesia to determine whether surgical treatment was necessary.

In addition to the Heath-Edwards classification, 21an index of pulmonary vascular disease (IPVD), as proposed by Yamaki et al. , 14was evaluated. IPVD was defined by the formula:

where n¹, n², n³, and n⁴are the numbers of pulmonary arterial sections bearing the respective scores. A score from 1 to 4 was given to each arterial section according to the following histologic findings: (1) no intimal reaction, (2) cellular intimal proliferation, (3) fibroelastic proliferation of intima, and (4) partial and total destruction of media. This index varied in theory from 1.0 to 4.0.

The same premedication as described above were given. Anesthesia was induced by intravenous injection of atropine sulfate (0.02 mg/kg), fentanyl citrate (5 μg/kg), lidocaine (1.0 mg/kg), and vecuronium bromide (0.3 mg/kg) followed by tracheal intubation, and was maintained with intravenous injection of fentanyl citrate at a total dose of 100 μg/kg, midazolam at a total dose of 0.2 mg/kg, and vecuronium bromide as needed.

During CPB, mean arterial pressure was maintained at approximately 30 mmHg by manipulation of pump flow between 100 and 180 ml/kg, bolus administration of chlorpromazine hydrochloride (total dose of 1 mg/kg), and continuous infusion with nitroglycerine (3–6 μg · kg⁻¹· min⁻¹) and alprostadil (prostaglandin E1; 0.05–0.1 μg · kg⁻¹· min⁻¹). During weaning from CPB, pulmonary arterial pressure (Pp), ascending aortic pressure (Ps), and left atrial pressure were directly monitored. Weaning was attempted by continuous infusion with dopamine hydrochloride (5–15 μg · kg⁻¹· min⁻¹), dobutamine hydrochloride (5–15 μg · kg⁻¹· min⁻¹), milrinone (0.2–0.5 μg · kg⁻¹· min⁻¹), along with continuous use of vasodilators, 100% oxygen inhalation, and hyperventilation. Drugs were administered to keep left atrial pressure less than 12 mmHg and systolic ascending aortic pressure greater than 60 mmHg. If pulmonary arterial pressure exceeded ascending aortic pressure (Pp/Ps > 1) in the presence of maximal doses of vasodilators, 40 ppm NO was added to the inspired gas mixture. NO was supplied from NO cylinders at 500 ppm through a flow meter (Taiyotoyo Co., Ltd., Osaka, Japan). Exhaled gas was collected by a TMS-100 (Taiyotoyo Co., Ltd.) placed distally to the endotracheal tube to monitor NO and NO²concentrations.

The Pp/Ps ratios at the time of weaning from CPB were compared with preoperative PVR values, Pp/Ps ratios, and Q̇p/Q̇s ratios measured during cardiac catheterization to find any possible correlation between these parameters. In the four patients who developed postbypass pulmonary hypertension that exceeded systemic pressure, Pp/Ps ratios were calculated before and 15 min after inhalation of 40 ppm NO. During NO inhalation, breathing conditions and the use of pulmonary vasodilators were kept constant.

Data are presented as mean ± SD. The regression curve of PVR was depicted by Stat View 4.0 software (Abacus Concepts, Berkeley, CA). Significant differences between groups were tested by the Scheffé F test, and P < 0.05 was considered a significant difference.

Data were obtained from 46 patients (25 male, 21 female). The mean age was 5.1 ± 2.8 months (range, 1–15 months) at the time of surgery, and mean weight was 5.0 ± 1.8 kg (range, 2.7–11.5 kg). The mean PVR measured during cardiac catheterization was 4.1 ± 2.7 Wood unit m²(range, 0.7–15 Wood unit m²). Mean Pp/Ps ratio was 0.7 ± 0.2 (range, 0.25–1), and mean Q̇p/Q̇s ratio was 2.8 ± 1.2 (range, 1–5.3). The mean Pp/Ps ratio at the time of weaning from CPB was 0.5 ± 0.2 (range, 0.25–1.2). Data are presented in table 1. Preoperative PVR correlated well with Pp/Ps at the time of weaning from CPB (r²= 0.86;P < 0.05; n = 46;fig. 1), but preoperative Pp/Ps did not correlate with Pp/Ps at the time of weaning from CPB (fig. 2). Preoperative Q̇p/Q̇s did not correlate with Pp/Ps at the time of weaning from CPB (fig. 3).

Characteristics of patients who developed postbypass pulmonary hypertension that exceeded systemic pressure are presented in table 2. In no case was Pp/Ps greater than or equal to 1 seen in a patient with preoperative PVR values less than 7 Wood units m². In three patients (cases no. 11, 25, 31), Pp/Ps ratios decreased significantly after NO inhalation (1.07 ± 0.12 vs.  0.67 ± 0.06;P < 0.05; n = 3). Case no. 7, who showed a PVR value of 15 Wood units m², did not respond to inhaled NO. He failed to wean from CPB and died. After obtaining informed consent, lung autopsy was performed in this patient.

The histopathologic evaluation of pulmonary arterial walls in two cases whose preoperative PVR values were 10 Wood units m²or more were as follows: case no. 7: type of intimal reaction was fibrous, state of media of arteries and arterioles were hypertrophied (Heath-Edwards classification grade 3, IPVD score 2.1); case 25: type of intimal reaction was none, state of media of arteries and arterioles were hypertrophied (Heath-Edwards classification grade 1, IPVD score 1.0).

The histopathologic evaluation of pulmonary arterial walls in case no. 7 after death was as follows: type of intimal reaction was fibrous, state of media of arteries and arterioles were hypertrophied (Heath-Edwards classification grade 3, IPVD score 2.2), the same as preoperative findings. However, pathologic obstruction of the proximal lumen and secondary atrophy of the media of the peripheral small pulmonary arteries were observed.

In recent years there has been an increase in surgery of infants with pulmonary hypertension, and the indication for surgery is often decided according to the results of lung biopsy, even if hemodynamics suggest the presence of Eizenmenger syndrome. 14–17 

Morphologic diagnosis of the operative indication in cases of congenital heart diseases with pulmonary hypertension is generally based on the Heath-Edwards classification. However, this classification is not adequate for the evaluation of pulmonary vascular disease in the entire pulmonary arterial system. To determine the severity of pulmonary vascular disease in the entire pulmonary arterial system, Yamaki et al. , 14introduced an IPVD, which is a quantitative assessment of the plexogenic pulmonary arteriopathy. They defined the histopathologic criteria for operative indication using IPVD and reported significant positive correlations between IPVD and preoperative PVR. 14 

In this context, PVR seems to be an important measure in the preoperative assessment of patients. The results of our study suggest that preoperative PVR is a sensitive factor to the prediction of Pp/Ps at the time of weaning from CPB and is therefore very useful in predicting the development of postbypass pulmonary hypertension and in decision-making for treatment options. Preoperative high Pp/Ps has been reported to be a risk factor for postoperative pulmonary hypertension. 22However, if Pp/Ps is high and Q̇p/Q̇s is also high, pulmonary hypertension is secondary to excessive pulmonary perfusion, and pulmonary vascular degeneration is not yet progressive. Patients with these conditions respond to oxygen therapy and pulmonary vasodilators if pulmonary perfusion is corrected by surgery, and as a result, their Pp/Ps should decrease after surgery. In contrast, if Pp/Ps is high and Q̇p/Q̇s is low before surgery, pulmonary vascular degeneration is already progressive, and patients may be resistant to pharmacologic treatment. Therefore, for anesthetic management, PVR seems to be more useful than preoperative Pp/Ps. Furthermore, Yamaki et al.  15also reported that the Q̇p/Q̇s and Pp/Ps were not useful in determining operative indications, whereas PVR proved to be the most reliable index of hemodynamics. PVR is also useful for predicting effectiveness of NO inhalation therapy in the preoperative period. In 42 patients, the postbypass pulmonary hypertension that exceeds systemic pressure was not observed in any cases in which the PVR values were less than 7 Wood units m², even if the preoperative Pp/Ps ratio was 1.0.

This suggests that patients with PVR values less than 7 Wood units m²may be treated with high-dose fentanyl, 12,13oxygen, 6hyperventilation, 7and pulmonary vasodilators, 8–11which have been reported to be effective without additional NO inhalation.

According to the literature about PVR, pulmonary hypertension persisted after surgery if PVR values were more than 6 Wood units m², and the prognosis was poor. 19,23,24These articles support the results of our study, which suggest that additional NO inhalation may be useful in the anesthetic management of patients whose PVR values are greater than 7 Wood units m².

Of patients who had already received oxygen, hyperventilation, and pulmonary vasodilators during anesthesia, three responded to additional inhaled NO. However, one patient with a PVR value of 15 Wood units m²did not respond to NO inhalation, suggesting that this therapy may have limitations. In this patient (case no. 7), cardiac repair was performed 1 month after lung biopsy. It was possible that pulmonary vascular disease had progressed, because pathologic obstruction of the proximal lumen and secondary atrophy of the media of the peripheral small pulmonary arteries were observed, which were not observed in the preoperative lung biopsy. These pathologic degenerations might be one reason why this patients did not respond to NO inhalation.

In summary, preoperative PVR is a sensitive factor to the prediction of postbypass pulmonary hypertension and seems to be a measure that would predict effectiveness of inhaled NO treatment.

In the current study, only four patients showed PVR values of more than 7 Wood units m²and received NO inhalation. Therefore, further studies are needed to determine the effect and usefulness of NO inhalation in patients with high PVR values and to define the criteria for inhaled NO indication using preoperative PVR values in a larger group of patients through a multiinstitutional clinical trial.

The authors thank Jun Ohata, M.D. (Director, Department of Anesthesiology, Nagano Prefectural Children's Hospital, Nagano, Japan), and Katuhiro Kawakami, M.D., and Ken Iwasawa, M.D. (Staff Anesthesiologists, Department of Anesthesiology, Shinsyuu University School of Medicine, Nagano, Japan), for cooperation and advice.