Comment on 'The comparison of two exhaled nitric oxide analyzers: NIOX VERO and SUNVOU-CA2122'

We read with interest the article by Lei et al [1]. The authors compared the performance of the Sunvou-CA2122 and NIOX VERO instruments for measuring fractional exhaled nitric oxide (FeNO). FeNO levels were obtained using both devices in patients with asthma, chronic obstructive pulmonary disease and chronic cough. It was concluded that the devices can be replaced by each other and are essentially interchangeable. This conflicts with the work of previous investigators that examined the NIOX VERO, Sunvou-CA2122 and other devices.

In a prior comparison of the Sunvou-CA2122 and NIOX VERO, FeNO values were 'strongly correlated but significantly different' [2]. Significantly higher FeNO values were observed with the Sunvou-CA2122 compared to the NIOX VERO (median 87.0 ppb, range 16–276 ppb and median 58.0 ppb, range 9–228 ppb, respectively; p < 0.001), and these differences increased at higher levels. Despite a high correlation between devices (r_s = 0.878; p < 0.001) the research demonstrated a 'low degree of agreement' and 'low degree of consistency between the devices', leading to the conclusion that 'the FeNO values obtained with (Sunvou-CA2122) cannot be interpreted according to the American Thoracic Society/European Respiratory Society (ATS/ERS) criteria, but need to be converted by a regression equation before being used for clinical purposes'. The authors speculated that the different technologies utilized by the two devices (reported as a Nano Coulomb current sensor for Sunvou-CA2122 and electrochemical sensor for NIOX VERO) might be a factor contributing to these results [2].

Similarly, other investigators have compared the NIOX VERO to other FeNO-measuring instruments. In one study of NOX VERO, Vivatmo-PRO (Bosch) and HypAir-FeNO (Medisoft) analyzers, mean differences in paired FeNO values ranged from −0.7 to 4.3 ppb and measures correlated (r² always >0.9; all p < 0.001) [3]. However, even though the Bland–Altman assessments of FeNO results strongly suggested that the devices agreed with each other, it was concluded that 'absolute FeNO measurements may differ to a clinically relevant extent using (different analyzers), and that they cannot be used interchangeably' [3].

Comparing the FeNO measurements between the Sunvou CA-2122 and NIOX VERO devices in the current study [1], the authors state that the Bland–Altman plot only showed a moderate degree of agreement, with the Sunvou-CA2122 analyzer reading significantly higher than the NIOX VERO to a modest degree (mean difference 2.6 ppb; p < 0.001). Additionally, 7% of all values were outside the 95% Confidence Interval of −13.8–19.0 ppb for the differences between the two devices. The range of the differences in FeNO values was wide, from −41 to 31 ppb, and appears to be more pronounced at higher FeNO levels [1]. Given that many asthma management guidelines, including the widely regarded ATS clinical practice guideline for interpreting FeNO [4] utilize cut points for FeNO when diagnosing and evaluating patients (e.g. <25 ppb, 25–50 ppb and >50 ppb), these large differences could result in differing or altered treatment decisions.

The advantages described for the Sunvou-CA2122 were in part based on 0.3 fewer measurement attempts and a total time of 7 s less required for a successful measurement (mean values) [1]. We agree that fewer attempts and shortening measurement time can improve efficiency, but while these differences were reported to be statistically significant (p = 0.009 for both), they cannot be considered meaningful. Also, the authors noted that the NIOX VERO did not require calibration with standard gases prior to conducting testing as this is done during the manufacturing process, making it unnecessary for the user to calibrate the device [1].

Based on these observations, we would suggest that the Sunvou-CA2122 and NIOX VERO devices for measuring FeNO are not interchangeable.

No new data were created or analyzed in this study.