1995 OPEN FORUM Abstracts
EVALUATION OF ELECTROCHEMCIAL NITRIC OXIDE ANALYZERS.
Edward P. Purtz, BS, RRT; Dean Hess PhD, RRT; Robert M. Kacmarek, PhD, RRT. Respiratory Care and Anesthesia, Massachusetts General Hospital and Harvard Medical School, Boston, MA.
Although it remains investigational, inhaled nitric oxide (NO) is being used increasingly in the treatment of PPHN, ARDS, and pulmonary hypertension. It is used clinically in dosesៀ ppm and often at dosesង ppm. NO is usually mixed with O_2 and delivered through a ventilator to the patient. The purpose of this study was to evaluate the accuracy of electrochemical NO analyzers that have recently become available.
Methods: We evaluated the following NO analyzers: Pulmonox II (Pulmonox, Alberta, Canada), NOxBOX (Bedfont, Kent, England), and the Saan (Taiyo Sanso, Osaka, Japan). They were provided by their manufacturers, calibrated as recommended and used per manufacturer's specifications. A Puritan-Bennett 7200 ventilator was used to produce serial dilutions of NO with O_2 to deliver [NO] of 0 - 80 ppm (Respir Care 1994;39:1113). FIO_2 settings of 0.90, 0.70, 0.50, 0.30, and 0.21 were used to produce serial dilutions of NO. The high pressure O_2 inlet of the ventilator was attached to 50 psi O_2 and the high pressure air inlet was attached to NO (80, 40, or 20 ppm in N_2). The following ventilator settings were used: flow-by 10 L/min, CPAP 5 cm H_2O; PEEP 5 cm H_2O, PC 10 cm H_2O; PEEP 7.5 cm H_2O, PC 15 cm H_2O; PEEP 10 cm H_2O, PC 20 cm H_2O; PEEP 12.5 cm H_2O, PC 25 cm H_2O; PEEP 15 cm H_2O, PC 30 cm H_2O. Other than the CPAP setting, a rate of 15 and Ti of 2 s were used. The ventilator was attached to a Michigan Instruments test lung (resistance = 20 cm H_2O/L/s, compliance = 20 mL/cm H_2O). The gas from the ventilator was not humidified. The analyzer was inserted into the inspiratory limb of the circuit using either a sidestream or mainstream technique. Measurements of NO and NO_2 were also performed using a calibrated Ecophysics Chemiluminesent analyzer. Bias ± precision were used to compare the [NO] from the chemiluminescence and electrochemical analyzers.
Results: Bias ± precision were:
[NO] (ppm) Peak Pressure (cm H_2O)FIO_2
analyzeroverall< = 20 ppm >20 ppm 0 - 15 22.5 - 30 37.5 - 45 < = 0.50 > 0.50
0.11± -0.03 ±0.27 ± 0.09 ± 0.11 ± 0.12 ±0.12 ± 0.09 ±
Saan 0.670.37 0.870.94 0.570.380.72 0.58
1.83± 0.95 ±2.88 ± 1.17 ± 1.53 ± 2.79 ±2.31 ± 1.11 ±
Pulmonox1.870.73 2.241.30 1.302.402.16 0.93
-0.77± -0.25 ± -1.38 ±1.34 ±-0.68 ±-0.28 ± -1.02 ±-0.39 ±
NOxBOX 1.04 0.67 1.07 1.430.64 0.541.13 0.75
There were significant differences in bias between the analyzers (Pɘ.001). For the Saan, there was no significant difference in bias for pressure (P = 0.99)or FIO_2 (P = 0.8), and a small but significant difference for [NO] (P = 0.03). For the Pulmonox, there were significant differences in bias for pressure (P = 0.001), FIO_2 (P = 0.002), and [NO] (Pɘ.001). For the NOxBOX, there were also significant differences in bias for pressure (P = 0.0002), FIO_2 (P=0.004), and [NO] (Pɘ.001).
Conclusions: Despite differences between devices, the bias and precision of these analyzers may be acceptable for clinical use. The devices tended to be most accurate at [NO]< = 20 ppm, higher airway pressures and higher FIO_2 levels - the clinical conditions at which NO is most commonly used. Considering that electrochemical NO analyzers have only been available for several years, the accuracy and precision of these devices is exceptional. (Supported in part by Puritan-Bennett)