Off-line exhaled nitric oxide measurements in children

The concentration of exhaled nitric oxide (eNO) is a useful marker of asthm atic bronchial inflammation. eNO can now be measured away from the laborato ry (off-line), even in children. Short exhalation maneuvers (8 sec) and sma ll samples (1 L) of exhaled gas are probably sufficient in children, but mo re information is needed about the effect of different measurement conditio ns. As a preliminary step before conducting epidemiological studies in scho olchildren, we investigated the effects of expiratory flow, dead space, and expiratory time on eNO concentrations collected in 1-L mylar collection ba gs. We studied 101 cooperative subjects (62 males) aged 5-18years (30 healt hy volunteers, 51 asthmatics, and 20 children with various other respirator y diseases) in our pulmonary function laboratory. On-line and off-line eNO were compared in a single session, and analyzed with a Sievers NOA (TM) 280 nitric oxide analyzer. For both methods of collecting expired gas, subject s did a single exhalation without breath-holding against an expiratory pres sure 10 cm H2O, We investigated the effects of expiratory flow, dead space, and exhalation time on eNO; we also compared on-line and off-line eNO meas urements, and the repeatability of both techniques at a given flow rate. Expiratory flows of 58 mL/sec provided more reproducible data than lower fl ows (coefficient of repeatability 1.1 ppb for 58 mL/sec vs. 2.8 for 27 mL/s ec vs. 5.7 for 18 mL/sec). eNO concentrations were about 25% higher in off- line than in on-line recordings if the initial 250 mL of exhaled gas were n ot eliminated, and 37% higher if exhalation lasted longer (16 sec vs, 8 sec ). Eliminating 250 mL of dead space and shortening the filling time to 8 se c yielded off-line eNO values close to those on-line (geometric mean off-li ne eNO 14.4 ppb, 95% confidence interval: 12.2-17.0) vs. on-line eNO 13.8 p pb (95% confidence interval: 11.6-16.5). On-line and off-line results were highly correlated (r = 0.996, P = 0.000) and had similar coefficients of va riation (online eNO 2.6%, off-line 2.8%). Neither agreement nor repeatabili ty of eNO measurements were affected by disease status or baseline FEV1 (% predicted values). Once standardized, the off-line eNO technique using 1 -L gas collection bags will provide results similar to those recorded on-line . (C) 2001 Wiley-Liss. Inc.