The Open Circuit Nitrogen Washout Technique for Thorax: First Published As 10.1136/Thx.54.9.790 on 1 September 1999

790 Thorax 1999;54:790–795

The open circuit nitrogen washout technique for Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from measuring the lung volume in infants: methodological aspects

Mohy G Morris

Abstract sophisticated equipment and a well trained Background—Lung volume measurement operator and cannot generally be performed at by nitrogen washout is widely used in the bedside or be used in sick infants. Gas dilu- infants, though a lack of accuracy and tion employs either helium or nitrogen as the changes of calibration over time have been tracer gas. The open circuit nitrogen washout reported. The potential sources of error technique is widely used because the dead were explored in order to increase the space and circuit resistance are lower than in accuracy and reliability of the technique. the helium closed circuit, making it suitable for Methods—A commercial system for nitro- small or sick infants. Moreover, data acquisi- gen washout and a 0.5 litre calibrating tion and calculation are easily programmed for syringe as a lung model were used to per- a personal computer.1–5 form over 2000 in vitro washouts, includ- While the measurement of lung volume has ing simulated rapid breathing, shallow been regarded as physiologically and clinically breathing, periodic breathing, sighs, and important,12the accuracy of the measurement brief apnoeas. A constant 10 l/min bias is undoubtedly equally important. A commer- flow of oxygen and extended equipment cial system has been used for the past 10 years warming times were employed. A collaps- or so in infant pulmonary function laboratories ible breathing bag was incorporated into to measure the functional residual capacity 4 the washout circuit. Following a single two (FRC) by nitrogen washout. Studies using this point calibration, known air volumes from system have occasionally explored the accuracy 42 ml to 492 ml were measured by nitro- of the nitrogen washout technique and possible 5 gen washout over a 14 hour period. The sources of errors. A change of calibration over flow waveform in the nitrogen mixing time was noted in a study on healthy infants chamber during a washout in vitro, with and, despite eVorts by the investigator to recheck calibration immediately prior to sub- and without the breathing bag in the http://thorax.bmj.com/ ject testing, a correction factor had to be used circuit, was also studied. 5 Results—The mean coeYcient of varia- on several test results. Indeed, a recent study tion of all volumes was 0.66%. The mean comparing lung volume measurements by diVerence between measured and known whole body plethysmography and nitrogen volumes was 0.30 ml (95% confidence washout reported a lack of accuracy and interval (CI) –0.18 to 0.79). This difference reliability of the two techniques in infants with was not statistically significant (p = 0.22). airway obstruction and concluded that no “gold standard” technique was available for use The mean percentage error was –0.1% 6 (range –0.47% to 0.46%). Nitrogen calibra- in this setting. on September 28, 2021 by guest. Protected copyright. tion remained stable for 14 hours. Without The primary aim of this investigation was the breathing bag flow transients were therefore to enhance the accuracy of measure- frequent in the mixing chamber during in ments and achieve a reliable calibration of the vitro washout. nitrogen washout technique by a systematic exploration of potential sources of errors in Conclusions—This technique increases the accuracy in vitro and the precision in vivo vitro using a calibrating syringe as a lung model, and ultimately to develop a method that of volume measurement by nitrogen wash- renders volume measurements by nitrogen out. Sources of potential errors including washout reliable and reproducible in vitro and baseline drifting and inadequate equip- in vivo. We also wanted to establish the linear- ment warming times were identified. The Department of ity of the system over a wide range of volumes breathing bag acted as a buVer reservoir, Pediatrics, Pulmonary so that a single two point calibration could be Medicine Section, preventing large swings in flows within the employed to measure the functional residual University of Arkansas nitrogen mixing chamber during wash- capacity (FRC) and, using a newly developed for Medical Sciences, outs, and should be an integral component Arkansas Children’s method, the residual lung volume (RV) in of the nitrogen washout circuit. Hospital, Little Rock, infants.7 Arkansas 72202-3591, (Thorax 1999;54:790–795) USA M G Morris Keywords: lung function tests; gas dilution; nitrogen Methods washout; infants A commercial system known as the Pediatric Correspondence to: Pulmonary Unit (PPU) 2600 (SensorMedics, Dr M G Morris. Anaheim, California, USA) was used. A Measurements of lung volumes in infants have 0.5 litre calibrating syringe (Hans Rudolph Received 6 May 1999 Accepted for publication been performed using whole body plethysmo- Inc, Kansas City, Missouri, USA) was em- 10 June 1999 graphy and gas dilution. The former requires ployed as a lung model. Open circuit nitrogen washout technique for measuring lung volume in infants 791

NITROGEN WASHOUT TECHNIQUE 15 mm OD × 10.5 mm ID. The oxygen ﬂow of Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from The open circuit nitrogen washout method for a high precision ﬂowmeter (Timemeter Instru-

assessment of FRC(N2) as described by ment Corporation, Lancaster, Pennsylvania, Gerhardt et al3 involves measuring the volume USA) was accurately set by adjusting the mid- of nitrogen expired after end tidal expiratory dle of the ﬂoat to the 10 l/min mark. This flow switching of the inspired gas from room air to was used for all tests. The oxygen tubing (King

100% oxygen. RV(N2) was also estimated after Systems Corporation, Noblesville, Indiana, end forced expiratory switching.7 At a constant USA) was connected to the 3.18 mm ID end bias flow that exceeds the infant’s inspiratory of an adapter (Hospitak Inc, Farmingdale, peak flow during tidal breathing, the integrated New York, USA), the other end of which expired nitrogen is multiplied by the constant (22 mm ID) fitted onto the 22 mm OD of a flow of oxygen to obtain the volume of expired “T” connection (“T” piece; Intersurgical Inc, nitrogen. A two point calibration is performed Cazenovia, New York, USA). A 0.5 litre with known air volumes. With the amount of collapsible breathing bag (Vital Signs Inc, nitrogen washed out measured and the initial Totowa, New Jersey, USA) was attached to the fractional alveolar nitrogen concentration centre port (22 mm ID) of the “T” connection × known (FAi(N2) room air = 0.79), the lung via an adapter (22 mm OD/19 mm ID volume at which the washout was initiated can 22 mm OD/17 mm ID; Baxter Healthcare be calculated578as follows: Corporation, Deerfield, Illinois, USA), and lung volume (FRC or RV) = volume N the distal (third) end (22 mm ID) of the “T” 2 connection was fitted onto an aerosol “T” washed out/FAi(N2) The PPU has an operator controlled adapter (22 mm OD; Hudson Respiratory pneumatic slide valve that switches the infant Care Inc, Temecula, California, USA). The to breathing 100% oxygen. The expired gas centre port (15 mm ID) of the aerosol “T” then enters a mixing chamber that is con- adapter was inserted onto the small port nected via a precision needle valve and a (15 mm OD) of the three way slide valve situ- vacuum pump to a nitrogen analyser and the ated at right angles to the mouth port. The nitrogen concentration is integrated electroni- opposite end (22 mm OD) of the aerosol “T” adapter was inserted into a distensible cou- cally by the PPU signal processing system. The × nitrogen washout curve is displayed in real pling connector (29 mm OD 17 mm ID; time on the computer monitor. When a 0% Marquest Medical, Aurora, Colorado, USA) nitrogen concentration is displayed on the to obtain a snug fit. A Concha Hose Adapter monitor, the slide valve is activated and the (Respiratory Care Inc, Arlington Heights, Illi- infant is switched back to breathing room air nois, USA) joined the other end of the and FRC(N )orRV(N) are automatically coupling connector and the proximal end of a 2 2 × calculated by the system. 2.0 m long hose (Tygon, 9.53 mm ID 15.88 mm OD; Baxter Healthcare Corpora- A systematic exploration of possible sources http://thorax.bmj.com/ of error when using the nitrogen washout tech- tion) with a snug fit. The distal end of the hose nique was undertaken. Potential sources of was inserted into the 11 mm OD end of an error encountered, suggested, or reported adapter (Marquest Medical, Aurora, Colo- included (1) changes in background oxygen,5 rado, USA) whose other end (22 mm ID) fit- (2) the infant’s peak flow exceeding the ted snugly into the inlet port of the nitrogen background flow, (3) a change in calibration mixing chamber. A 1.8 m long (22 mm ID) over time,5 (4) switching at end tidal expiration corrugated tube (Baxter Healthcare Corpora- when FRC57was measured or leaks in the cir- tion) was attached to the outlet port of this cuit, including the face mask, and in infants chamber and was loosely coiled in an open box on September 28, 2021 by guest. Protected copyright. with airway obstruction, (5) the unanswered on the side of the PPU. This tube prevented question of the final nitrogen concentration, ambient air from diVusing back into the nitro- and (6) length of the washout time and the gen mixing chamber. minimal time interval between consecutive measurements.1578 Over 2000 in vitro wash- Calibration of the PPU outs were performed using the calibrating As recommended by the manufacturer, using syringe as a lung model. Furthermore, simula- separate electrical outlets the vacuum pump tions were performed using unequal stroke vol- with its oxygen flow was turned on 30 minutes before calibration and the PPU measurement umes and diVerent rates with the calibrating syringe to simulate periodic breathing, sighs, module and computer at least 20 minutes rapid shallow breathing, pauses to simulate before calibration. The breathing bag was apnoeas, or temporary upper airway obstruc- squeezed manually several times to wash out tion. Of several connections, equipment warm- any nitrogen. Since the computer software ing times, and flow rates of background oxygen automatically converts measured washout vol- tested, the methodology which yielded the best umes to BTPS, we chose to make all measure- results is given below. ments under ambient (ATPS) conditions by entering, before testing, a temperature of 37°C and a barometric pressure of 760 mm Hg Nitrogen washout circuit when prompted by the software program. This A three way pneumatic slide valve (8540 facilitated the comparison between integrated series, 9.5 mm flow bore size; Hans Rudolph nitrogen signals obtained when injecting a Inc) was used with a mouth port of 22 mm known volume during calibration and using outer diameter (OD) × 15 mm inner diameter this same volume to confirm calibration. The (ID) and two smaller inlet/outlet ports of calculated volume of the latter would otherwise 792 Morris

have been automatically converted to BTPS. 15 minutes for room air to be sampled. The Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from This allowed the operator to pinpoint potential displayed nitrogen concentration was 79.0 sources of errors during measurements. The (0.2)% after all tests. computer software nitrogen calibration menu was accessed. The displayed nitrogen concen- ACCURACY STUDY WITH A LUNG MODEL tration was 0.0% and, if not, it was zeroed using When the 0.5 litre calibrating syringe with its the “autozero” mode of the PPU software. A piston pushed in to the 500 ml mark was con- check for the presence of a baseline drift was nected to the mouth port of the slide valve then performed. The menu of the “low” (or (22 mm OD × 15 mm ID) their combined vol- “high”) volume calibration was accessed. ume was found to be 42 ml. This was When the displayed nitrogen concentration estimated by volume replacement and was fur- was 0.0%, the slide valve was activated and the ther confirmed by Hans Rudolph Inc. This mouth port was left open to room air for 10 volume was not subtracted during calibration. seconds in order to wash out the air within the The gas space within the slide valve piston port with the pure oxygen. The port was then body was ignored because it was flushed with occluded to prevent room air from diVusing oxygen during testing for baseline drifts. back into the port. The nitrogen concentration Following a two point calibration the syringe and the integrated percentage nitrogen signal was used as a lung model to wash out known air were observed for 90 seconds for a stable 0.0% volumes from 42 ml to 492 ml. The back- reading. If the integrated nitrogen signal rose, a ground flow of oxygen was set at 10 l/min. baseline drift was presumed to be present, the Each measurement was repeated five times. “escape” key on the computer keyboard was Using stroke volumes of 50–75 ml, diVerent pressed, and the calibration menu was re- known air volumes were randomly used over a accessed. Although the displayed nitrogen con- 14 hour period to determine whether changes centration was still 0.0%, a further decrease of in calibration occurred over time. the baseline towards zero was performed using Introduction of the collapsible breathing bag 1–3 keyboard strokes in the manual mode of into the oxygen circuit enhanced the reproduc- the program, the space bar of the keyboard was ibility and precision of volume measurements. then pressed to enter the baseline change, Therefore, using stroke volumes of approxi- followed by a repeat check for a baseline drift. mately 70 ml to simulate tidal breathing, as was Overcorrection resulting in a negative percent- routinely done for calibration, the influence of age nitrogen concentration reading was not the breathing bag was studied by comparing allowed. Additional checks for baseline drifts the waveform of the flow and volume signals (and correction if needed) were undertaken before and after introducing the breathing bag during the waiting periods between measure- into the circuit. This in vitro experiment was ments on an infant with the slide valve being performed by inserting the PNT into the distal activated via the “patient test” menu rather end of the corrugated tube, the proximal end of http://thorax.bmj.com/ than the “calibration” menu used initially. which was connected to the outlet port of the Calibration of the nitrogen analyser needle nitrogen mixing chamber. valve was performed before each test. The needle was removed from the outlet port of the STATISTICAL ANALYSIS mixing chamber and allowed to hang for at Data were expressed as arithmetic mean and least 45 minutes so that room air was sampled. standard deviation (SD). The Student’s t test The procedure of “peaking the needle”, as for paired data was used to analyse data from described by the manufacturer, was performed the accuracy study to compare known volumes with volumes measured by the nitrogen to obtain the optimum negative pressure in the on September 28, 2021 by guest. Protected copyright. analyser to achieve maximum nitrogen ionisa- washout technique. p values of less than 0.05 tion. When the needle was replaced into the were considered significant. outlet port the displayed nitrogen concentration was 0.0%. The low and high volume cali- Results bration was performed with 42 ml and 342 ml ACCURACY STUDY WITH THE LUNG MODEL air, respectively, for all infant tests. When the Results from this in vitro study are presented in integrated nitrogen signal stopped rising it was table 1. The mean coeYcient of variation (CV) recorded, the “escape” key was pressed, and of all volumes was 0.66%. The mean difference each volume calibration was repeated until two Table 1 Accuracy study: measurement of known air successive integrated nitrogen signals within volumes by nitrogen washout 1% of each other were obtained before the Known volume Measured volume calibration was entered. The accuracy of the (ml) (ml) Error (%) calibration was rechecked immediately after calibrating by measuring known volumes of air 42 41.8 (0.4)† −0.47 92 92.2 (0.6) 0.22 (see below). It was also rechecked after a test 142 141.9 (1.1) −0.07 had been completed using a sterilised nitrogen 192 191.1 (1.2) 0.46 242 242.4 (1.9) −0.16 circuit (slide valve, connectors, and a new col- 292 290.9 (1.4) −0.37 lapsible breathing bag) to prevent contamina- 342 341.6 (1.8) −0.11 tion of the calibrating syringe. A known volume 392 391.9 (2.6) −0.02 442 442.8 (2.9) −0.18 equivalent to the infant’s measured lung 492 490.4 (1.4) −0.32 volume was included in rechecking the calibra- −0.10 (0.28)* tion. Finally, the nitrogen analyser needle was *Mean (SD). removed from the outlet port of the nitrogen †Values are mean (SD) of five measurements performed mixing chamber and allowed to hang for about randomly over a 14 hour period. Open circuit nitrogen washout technique for measuring lung volume in infants 793

in vitro testing (table 1) as well as after infant Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from –200 testing had been completed.7 The nitrogen washout circuit can be quickly assembled into a –50 virtually leak-free unit that can be disinfected between tests. Inclusion of the collapsible breathing bag in 100

Flow (ml/s) the nitrogen washout circuit had a significant stabilising eVect that led to exceptional repro- 250 ducibility in vivo7 and accuracy in vitro (table 1) in measured washout volumes. The in vitro 10 15 20 study indicated that flow transients were more frequent in the absence of the breathing bag 100 (fig 1). To ensure that the observed waveform diVerence was not the result of a technical variation in injecting air from the calibrating 60 syringe, tidal breathing indices (tidal volume, the inspiratory, expiratory and total time and Volume (ml) Volume frequency) were computed from the in vitro 20 study. Using the unpaired t test, no significant diVerence was found between the mean values of the indices before and after the breathing 10 15 20 bag was placed in the circuit (data not shown). Given the low resistance in the open circuit of –200 nitrogen washout, the breathing bag probably acted as a buVer reservoir that prevented large –50 swings in the bias flow of oxygen in the nitrogen mixing chamber. These swings inevitably occur 100 during the breathing cycle of an older child Flow (ml/s) with a larger tidal volume. More importantly, 250 the breathing bag minimised the retrograde movement of mixed oxygen and nitrogen after it had passed beyond the sampling needle port. 10 15 20 This explanation was supported by the observation that, when a tube shorter than the 1.8 m 100 tube used in our study was attached to the outlet port of the nitrogen mixing chamber, a sharp rise in the nitrogen washout curve above

60 http://thorax.bmj.com/ baseline occurred when an older child took a

Volume (ml) Volume deep breath. This was presumably due to the 20 retrograde movement of gas enriched with room air nitrogen into the outlet port and through the sampling needle which is located at 10 15 20 a depth of 30 mm from the port opening (15 mm ID)—that is, separated from the 3 Time (s) opening by only 5.3 cm . An in vitro simulation

using a sharp pull on the piston of the calibrat- on September 28, 2021 by guest. Protected copyright. Figure 1 Influence of the collapsible breathing bag on gas flow in the nitrogen mixing chamber during in vitro ing syringe lends support to this hypothesis. washout. Using stroke volumes from the calibrating syringe The breathing bag served also as a constant of about 70 ml to simulate tidal breathing, the flow and monitor for tidal breathing. In the author’s volume signals pattern were compared using a pneumotachometer connected to the outlet of the nitrogen experience, an occasional transient upper mixing chamber before (lower two panels) and after (upper airway obstruction in the sedated infant was panels) the breathing bag was introduced into the washout easily detected and promptly corrected during circuit. Note the presence of flow transients in the absence of the breathing bag. washouts. Pauses during periodic breathing were also noted. If either of these two phenom- between measured and known volumes was ena were to occur close to the end of a washout, 0.30 ml (95% CI –0.18 to 0.79). This they could conceivably lead to a premature diVerence was not statistically significant (p = return of the nitrogen washout curve to 0.22). The mean percentage error was –0.1% baseline and termination of the test by the (range –0.47% to 0.46%). operator resulting in an underestimation of the The breathing bag had a clear influence on measured washout volume. Interestingly, the the waveform of the flow and volume signals in detection of a transient airway obstruction vitro. In the absence of the breathing bag, flow during FRC measurement of an infant was transients were frequent (fig 1). slightly delayed, resulting in the return of the nitrogen washout curve to baseline. The Discussion infant’s head was repositioned which relieved Of several connections, equipment warming the obstruction, which in turn was confirmed times, and magnitudes of background oxygen by the breathing bag. The nitrogen curve rose flow, the methodology described here gave the again above baseline when the washout re- most reproducible results in vitro and in vivo.7 sumed. This produced the second peak of what Calibration remained stable during 14 hours of became a biphasic curve whose calculated 794 Morris

volume was still reliable. An in vitro simulation was attached to the mixing chamber outlet Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from with the calibrating syringe performed by port, allowing room air to diVuse back into the pausing during washout resulted in a biphasic port and sampling needle. Another potential curve that reliably reflected the known volume source was a discontinuity in any of the (M G Morris, personal observation). Indeed, it connections, and this happened only once would be reasonable not to terminate an infant through a visible break in the aerosol “T” con- test immediately when a zero nitrogen concen- nection port. tration is displayed on the monitor unless the Using the calibrating syringe as a lung breathing bag is still synchronously inflating model, the author has used a wide range of vol- and deflating with the infant chest wall umes for several two point calibration experi- excursions. A substantial or nearly complete ments. These in vitro experiments suggested deflation of the breathing bag during a washout that the system remained linear for at least by the inspiring child could presumably alert 50 ml beyond the two point calibration. We the operator to the possibility of a peak inspira- chose 42 ml and 342 ml for all infant tests7 so tory flow exceeding the bias flow of oxygen. that we would become familiar with their The insertion of a pneumotachometer between respective integrated nitrogen signal and easily the face mask and the mouth port of the slide detect aberrant numbers. This added a quality valve could provide alternative continuous assurance element to our operating conditions. monitoring for tidal breathing during The use of a small volume (42 ml) has the washouts.8 However, this has the disadvantage advantage of shortening the calibration time. of increasing the dead space and the resistance After an infant test had been completed, a of the washout circuit without providing the known volume equivalent to the patient’s buVering capability of the breathing bag for measured lung volume was routinely included large tidal volumes. Taken together, we con- in re-checking the calibration. clude that the breathing bag should be an inte- Inadequate equilibration of the calibrating gral component of the open nitrogen washout syringe with room air nitrogen after it had been circuit. The optimal location of the breathing used for calibration or to confirm calibration bag in the circuit was between the patient and seemed a real and frequent, but avoidable, the oxygen source, and closer to the former. source of error by underestimating “pre- Placement distal to the patient tended to retain sumed” known volumes. This was likely to some of the washed out nitrogen and, unless occur when a large volume such as 492 ml had the bag was slowly and completely squeezed been used by the hurried operator. We used a just before the washout was completed, 15 litre flow of air via a cannula inserted for a measured volumes would be underestimated brief period into the calibration syringe assisted (M G Morris, unpublished data). by a few piston strokes to speed up the equili- Adequate equipment warming time was an bration process. After the cannula had been important factor in performing reliable nitro- pulled out, a few more piston strokes were used http://thorax.bmj.com/ gen washouts. Hanging the needle valve for at for a final equilibration with room air. This was least 45 minutes to sample room air was found, preferable to loosening the volume adjustment by using longer periods up to two hours, to be metal ring to achieve larger excursions of the essential to achieve a stable nitrogen concen- piston in order to preserve the exact known tration reading in our PPU. Premature termi- volume for a repeat injection. Similarly, the nation of this step contributed to the change in mouth port of the slide valve should be flushed nitrogen calibration over time and with the with room air after a washout. After a washout finding, after infant testing had been com- had been completed and the slide valve had

pleted, of a nitrogen concentration below 79% switched the calibrating syringe to room air, the on September 28, 2021 by guest. Protected copyright. when the needle was used to resample room piston was routinely pulled back before discon- air. necting the syringe from the mouth port to Drifting of the baseline has been a major flush the latter with room air. challenge during testing and a potential source Since clinical settings are usually associated for a systematic error by overestimating with time constraints, the calibration of the measured washout volumes. Frequent checks PPU described here seemed adaptable for use for baseline drifts and correction were there- in these settings.7 Whereas the manufacturer fore essential to obtain reproducible data. recommended warming periods of 30 minutes Interestingly, drifting might not occur for 60 and 20 minutes for the nitrogen pump and seconds but then the integrated nitrogen signal computer module, respectively, there was no would start rising rapidly. The author has mention of a time period being required for the observed a stability of the baseline lasting up to nitrogen analyser needle to equilibrate with 15 minutes. Hence, in testing patients with air- ambient nitrogen. Similarly, there was no way obstruction it is important to test for drifts information regarding baseline drifting. These for at least 90 seconds because of the expected were discovered during our in vitro studies. A prolonged washout times.7 An immediate rise 90–100 minute calibration period was usually above baseline of the nitrogen washout curve suYcient for the PPU to be ready for clinical after the slide valve was switched indicated the testing. This included 30 minutes for pump presence of a drift if air had not yet entered the and computer module, 45 minutes for the bias flow of oxygen from either the calibrating sampling needle equilibration with room air syringe or the apnoeic infant during measure- nitrogen, “peaking the needle”, testing for ment of RV.7 A true rise in the integrated nitro- baseline drifts, the two point calibration (see gen signal was found when a tube shorter than above), and confirmation of calibration imme- the 1.8 m corrugated tube used in our study diately after calibration with at least two known Open circuit nitrogen washout technique for measuring lung volume in infants 795

volumes. The ultimate result, however, is a one 1 American Thoracic Society/European Respiratory Society. Respiratory mechanics in infants: physiological evaluation Thorax: first published as 10.1136/thx.54.9.790 on 1 September 1999. Downloaded from day stable calibration that allows testing of in health and disease. Am Rev Respir Dis 1993;147:474–96. more than one infant per day. Moreover, the 2 Gaultier C. Lung volume in neonates and infants. Eur Respir described methodology rewards the operator J 1989;2(Suppl 4):130–4s. 3 Gerhardt T, Hehre D, Bancalari E, et al. A simple method with excellent repeatability of measurements for measuring functional residual capacity by nitrogen such that fewer reliable tests need to be washout in animals and newborn infants. Pediatr Res 1986; 7 20:668–71. performed on a sedated infant. 4 Sivan Y, Deakers TW, Newth CJL. An automated bedside In conclusion, we developed a method that method for measuring functional residual capacity by nitrogen washout in mechanically ventilated children. Pedi- increased the accurancy in vitro and the preci- atr Res 1990;28:446–50. sion in vivo7 of volume measurement using the 5 Gappa M, Fletcher ME, Dezateux CA, et al. Comparison of nitrogen washout and plethysmographic measurements of open circuit nitrogen washout technique. lung volume in healthy infants. Am Rev Respir Dis Sources of potential errors that included base- 1993;148:1496–501. 6 Eber E, Steinbrugger B, Modl M, et al. Lung volume meas- line drifting and inadequate equipment warm- urements in wheezy infants: comparison of plethysmogra- ing times were evaluated. We conclude that a phy and gas dilution. Eur Respir J 1994;7:1988–94. 7 Morris MG. A novel noninvasive technique for measuring collapsible breathing bag should be incorpo- the residual lung volume by nitrogen washout with rapid rated in the nitrogen washout circuit. thoracoabdominal compression in infants. Thorax 1999 (in press). 8 Tepper RS, Merth IT, Newth CJL, et al. Measurement of Supported by a Clinical Research Grant (CG-008-N) co- functional residual capacity in infants by helium dilution funded by the American Lung Association (ALA) and the and nitrogen washout techniques. In: Stocks J, Sly PD, Arkansas Chapter of ALA. The author was supported in part by Tepper RS, Morgan WJ, eds. Infant respiratory function test- the Department of Pediatrics, UAMS. ing. New York: John Wiley & Sons Inc, 1996: 165–89. http://thorax.bmj.com/ on September 28, 2021 by guest. Protected copyright.