- Email: [email protected]
Lung diffusing capacity for nitric oxide at lowered and raised ambient pressures
Respiratory Physiology & Neurobiology 189 (2013) 552–557
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Lung diffusing capacity for nitric oxide at lowered and raised ambient pressures Dag Linnarsson a,∗ , Tryggve E. Hemmingsson a,b , Claes Frostell a , Alain Van Muylem c , Yannick Kerckx d , Lars E. Gustafsson a a
Department of Physiology and Pharmacology, Karolinska Institutet, SE-17177 Stockholm, Sweden Department of Radiology, Karolinska University Hospital, Stockholm, Sweden c Chest Department, Erasme University Hospital, Brussels, Belgium d Biomedical Physics Laboratory, Université Libre de Bruxelles, Brussels, Belgium b
a r t i c l e
i n f o
Article history: Accepted 15 August 2013 Keywords: NO Gas density Diffusivity Hypobaria Hyperbaria
a b s t r a c t Lung diffusing capacity for NO (DLNO ) was determined in eight subjects at ambient pressures of 505, 1015, and 4053 hPa (379, 761 and 3040 mmHg) as they breathed normoxic gases. Mean values were 116.9 ± 11.1 (SEM), 113.4 ± 11.1 and 99.3 ± 10.1 ml min−1 hPa−1 at 505, 1015, and 4053 hPa, with a 13% difference between the two higher pressures (P = 0.017). The data were applied to a model with two serially coupled conductances; the gas phase (DgNO , variable with pressure), and the alveolo-capillary membrane (DmNO , constant). The data fitted the model well and we conclude that diffusive transport of NO in the peripheral lung is inversely related to gas density. At normal pressure DmNO was approximately 5% larger than DLNO , suggesting that the Dg factor then is not negligible. We also conclude that the density of the breathing gas is likely to impact the backdiffusion of naturally formed NO from conducting airways to the alveoli. © 2013 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Background Measurements of pulmonary nitric oxide (NO) are of scientific and clinical interest mainly in two areas; (1) to monitor the activity of inflammatory airway disease from exhaled NO of endogenous origin (Persson et al., 1994; Olin et al., 2007; ATS Guidelines, 2011), and (2) to measure lung diffusing capacity (DLNO ) from the rate of uptake of NO from an external source. In the former case the NO level is several orders of magnitude lower (ppb) than that inhaled during a DLNO maneuver (ppm) (Guénard et al., 1987; Zavorsky et al., 2008). 1.1.1. Exhaled NO Nitric oxide in the lungs has been implicated as a potentially protective factor against pulmonary manifestations of acute mountain sickness (Busch et al., 2001; Duplain et al., 2000; Erzurum et al., 2007). It has therefore been considered of interest to establish what the effects are of a reduced ambient air pressure on pulmonary NO in healthy individuals. A second consideration is to define normal values for altitude residents, when exhaled NO is
∗ Corresponding author. Tel.: +46 8 52486890; fax: + 46 8 304613. E-mail address: [email protected] (D. Linnarsson). 1569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resp.2013.08.008
used to monitor the activity of inflammatory airway diseases such as asthma (Olin et al., 2007; Taylor et al., 2006). Potential effects of altitude include the influence of hypoxia per se (Brown et al., 2006; Hemmingsson et al., 2009), of the reduced gas density (Shin et al., 2006; Van Muylem et al., 2003) and a combination of these two factors (Hemmingsson and Linnarsson, 2009). The latter authors showed that a short-lasting exposure to normobaric hypoxia down to 10% of an atmosphere had no significant effect on exhaled NO. We recently addressed the effects of gas density on exhaled NO by exposing healthy subjects to both reduced and increased ambient pressures while at the same time maintaining normoxia in the breathing gases. Based on previous experiments with heliumoxygen breathing (Shin et al., 2006; Van Muylem et al., 2003) we had expected that the lowered gas density at altitude and the associated increased diffusivity for NO in the lung gas would increase backdiffusion of NO to the alveoli. In turn this would increase the uptake of NO to the blood resulting in a reduced partial pressure of NO in the exhaled gas (PENO ). We also expected corresponding mechanisms to increase PENO at hyperbaric pressure. However, recent work from our laboratory has shown that PENO values were strikingly similar in an ambient pressure range from 0.5 to 4.0 atmospheres (Hemmingsson et al., 2012). We hypothesized that diffusive transport of NO in the gas phase in the lungs would indeed be influenced by changes of the gas density, but that these effects would be relatively small and possibly be concealed by simultaneous effects of density on convective gas transport acting
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
in opposite direction (Hemmingsson et al., 2012). We reasoned that direct determinations of the lung diffusing capacity for NO (DLNO ) would be a way to test this hypothesis, so we undertook an additional study, now focusing on DLNO while using an otherwise identical experimental design and studying the same subjects. 1.1.2. Lung diffusing capacity Pulmonary edema is a feared component of acute mountain sickness and determinations of DLNO have been used to detect interstitial pulmonary edema in otherwise healthy individuals during their stay at high altitude (de Bisschop et al., 2010, 2012). Conventionally, DLNO is used to estimate the capacity for diffusive transport through the alveolo-capillary membrane, under the assumption that any effects of diffusive transport through the gas phase in the respiratory zone of the lung can be neglected (Guénard et al., 1987; Zavorsky et al., 2008). We hypothesized that such an effect indeed is small but not necessarily negligible at all ambient pressures. We reasoned that by comparing DLNO data obtained with background gases of widely differing diffusivity for NO, we would be able to estimate the quantitative role of diffusive transport through the gas phase for DLNO .
553
possible to detect if conditions with sufficiently large differences in density and therefore also diffusivity were compared. Eq. (2) may be further developed by assuming that: (a) DgNO should vary in proportion to the diffusion coefficient for NO in the breathing gas. This parameter in turn varies inversely with the ambient pressure (Chang, 1985). (b) The term 1/DmNO is constant across conditions. For a range of pressures and breathing gas mixtures consistent with the above assumptions we then obtain for a given pressure: 1 = DLNO
1 ∗ P DgNO
P0
+
1 DmNO
(3)
where P/P0 is the ratio between the ambient pressure and the pressure for which DgNO is defined. The equation has the format of a linear relationship where the slope is 1/DgNO and the intercept with the Y axis is 1/DmNO . 2. Methods 2.1. Subjects
1.2. Theory The general way to characterize the passage of inhaled or endogenously formed gases from the airways to the pulmonarycapillary blood is to determine the diffusing capacity. Forster and Roughton (Forster et al., 1957) developed an algorithm to determine the diffusing capacity of the lung for a specific gas (DL), or transfer factor as it is usually called outside the United States. Below, the term DL will be used and, according to the classical model, expressed as serially connected resistances to diffusion, from the airway gas to the inside of the erythrocyte. Each subcomponent of the serial reaction chain is hence the inverse of a conductance involving both passive diffusion and chemical binding with hemoglobin (Hb). The complete transfer reaction of the lung DL for a gas x can be explained as three serially linked resistances (Cotton and Graham, 2005; Johnson et al., 1996); 1 1 1 1 + + = DL Dgx Dmx (x · Vc )
(1)
defined by the conductance for x in the background gas of the respiratory zone (Dgx ), the conductance for x through the alveolocapillary membrane (Dmx ), the conductance for uptake of gas x within the erythrocyte ( x ), and by the volume of the pulmonarycapillary blood (Vc ). The factor describing diffusion in the gas phase (Dgx ) is usually considered negligible for inhaled test gases such as CO, whereas both the amount and the oxygen saturation of the hemoglobin in the lung capillaries have measurable impacts on DLCO . The reaction of NO with hemoglobin in the erythrocyte is about 280 times faster than that of CO (Meyer and Piiper, 1989; Tamhane et al., 2001; Zavorsky et al., 2008). Therefore, the resistance to the NO transport within the erythrocyte has been considered negligible (Johnson et al., 1996), and the third term in Eq. (1) (1/ x × Vc ) could therefore be omitted in a corresponding equation for NO (Zavorsky, 2010). However, other authors have considered this term to have a finite value (Borland et al., 2010), see Martinot et al. (2013) for a detailed discussion. For the purpose of the present analysis, however, the results are based on the assumption that this third term could be neglected, thereby obtaining: 1 1 1 = + DLNO DgNO DmNO
(2)
Since DL for NO is about four times that for CO (Guénard et al., 1978; Zavorsky et al., 2004), we thought that the impact of diffusivity in the gas phase (as quantified by the 1/DgNO term) would be
Eight healthy non-smoking subjects without a history of inflammatory airway disease participated in the current study. The subjects came to the laboratory once for familiarization, physical examination, spirometry and baseline FENO measurement. They returned twice for experiments in increased or decreased ambient pressure in a combined hyperbaric and hypobaric pressure chamber. All eight subjects, four women, completed all tests at all pressure levels. Their age, height and weight ranged 21–37 years, 1.60–1.93 m and 58–87 kg, respectively. 2.2. Instrumentation and measurements Experiments were performed at 505 ± 0 (mean ± SEM), 1015 ± 3 and 4053 ± 0 hPa ambient pressure. The corresponding values in mmHg were 379, 761 and 3040. The pressure chamber (internal volume 8 m3 ) was pressurized with air but subjects breathed normoxic gas mixtures with oxygen fractions of 0.421, 0.2095 and 0.052 at 505, 1015 and 4053 hPa, respectively. Subjects were investigated at hypobaric and hyperbaric pressures on different days, in random order and always starting with control measurements at 1015 hPa. The lung diffusing capacity for NO (DLNO ) was measured in two subjects at a time, seated together with one of the test supervisors in the pressure chamber, using the standardized ATS/ERS techniques for diffusing capacity (ATS/ERS, 2005), which is based on the Jones and Meade methodology (Jones and Meade, 1961), with a modified 5 s breath-holding time (Zavorsky et al., 2008). After a change in ambient pressure, a 15 min waiting time was allowed to accommodate to the new environment while breathing the normoxic gas mixture. Decompression after the hyperbaric experiments was performed according to Swedish Navy standard tables and with correction for the increased nitrogen partial pressure compared to air breathing. There were no decompression symptoms in any of the subjects. Between DLNO determinations subjects were breathing through an oronasal mask and a non-rebreathing valve (Hans Rudolph Inc., Shawnee, KS, USA) from a 200 l Douglas bag via 40 mm inner diameter hoses. The supervisor kept the bag adequately filled by means of a needle valve connected to the reducing valve of tanks with compressed gas housed inside the chamber and quick-connect fittings allowed for simple change of gas source when required. Gas for argon (Ar) analysis was sampled to a mass spectrometer (lnnovision A/S, Odense, Denmark) located outside the pressure chamber. Gas for NO analysis was sampled to a chemiluminescence analyzer (Eco
554
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
Medics AG, Duernten, Switzerland) also located outside the pressure chamber. Sampling was performed first through a 1.5 m long Nafion catheter and then via a 2 mm inner diameter polytetrafluoroethylene (PTFE) tube connected via Luer fittings to the inlet of the analyzer (505 and 1015 hPa). During the hyperbaric experiments there was a three-way connector at the NO analyzer inlet. Sampling flow from the subject to the three-way connector was driven by the chamber pressure, whereas the pump of the analyzer sampled a part of this flow for analysis. The NO analyzer was calibrated with certified gas mixtures with 2000 ppb NO in nitrogen (AGA SpecialGas AB, Lidingö, Sweden). The mass spectrometer was calibrated with 10% Ar (AGA SpecialGas AB, Lidingö, Sweden). Respired flow was monitored with a Hans Rudolph 3700 series pneumotachometer coupled to a differential pressure transducer (Validyne Engineering, Northridge, CA, USA). The flow signal was fed through the chamber wall from the inside to the outside, where data including those from the gas analyzers, were collected at 100 Hz with a digital data acquisition system (Biopac Systems MP150CE (Biopac Systems Inc., Goleta, CA, USA)). 2.3. Diffusion capacity measurement procedure The subject shifted from the oronasal mask to the mouthpiece of the DLNO assembly while holding his/her breath and donning a nose clip. Then the diffusion capacity measurement was initiated by the subject taking 3–4 breaths at tidal breathing after which he/she performed a maximal expiration to residual volume (RV). The test supervisor then turned a rotational valve for inhalation of a normoxic test gas mixture with 10 ppm NO and 10% Ar, from a 10 l bag. After taking a deep, quick breath to total lung capacity, the subject made a 3–4 s breath-hold and then performed a rapid full expiration to RV. The supervisor then turned the rotational valve back to tidal breathing position. 2.4. Data analysis For each subject, a set of calibrated values for inhaled and exhaled NO and Ar, and flow were collected at each ambient pressure. DLNO (in ml min−1 hPa−1 ) was calculated according to current standards (ATS/ERS, 2005). A combined anatomical and instrument dead space of 400 ml was assumed. The timing of the maneuver was determined from the flow recording. Gas concentration signals were aligned with the flow recording by compensating for the sampling delays of the analyzers.
Fig. 1. Diffusing capacity for NO (DLNO ) in eight subjects inhaling 10 ppm NO in a stable normoxic environment at 505, 1015 and 4053 hPa ambient pressure, respectively. The horizontal line indicates the diffusing capacity through the alveolocapillary membrane (DmNO ), which represents the theoretical maximum value for DLNO , should there be an infinite rate of diffusion through the gas in the respiratory zone (1/DgNO = 0, see Eq. 2 for explanation).
In Fig. 2 the inverse of the same data as in Fig. 1 have been plotted as a function of the ratio between the ambient pressure and the sea level pressure, as described in Eq. (3) above. There was an excellent fit with a linear function, the intercept of which with the Y axis shows the value of 1/DLNO with an infinite diffusion coefficient, so that DLNO then must be equal to DmNO . The relationship in Fig. 2 is therefore compatible with the assumption that the value of DmNO is constant across conditions and varies with DgNO , which in turn varies with the diffusion coefficient and the inverse of the ambient pressure. The same relationship yields a DmNO value of 119.6 ml min−1 hPa−1 . This can be regarded as a theoretical maximum value for DLNO , should there be no resistance to NO transfer in the gas phase. From the slope of the function a
2.5. Ethics and statistics The study was approved by the Regional Ethical Review Board in Stockholm, Sweden and all subjects gave their written informed consent. Data are presented as group means and standard error of the mean (SEM) for each ambient condition. Data were analyzed by using a Wilcoxon matched pairs test (StatSoft, Tulsa, OK, USA) with a Bonferroni correction for multiple comparisons. A P-value of <0.05 was considered statistically significant. 3. Results Our group mean results are shown in Fig. 1. Values there are in ml min−1 hPa−1 . Expressed instead per unit mmHg, DLNO was 155.8, 151.1 and 132.4 ml min−1 at 505, 1015 and 4053 hPa, respectively. There was no significant difference between 505 hPa and control (P = 0.17), whereas DLNO was significantly reduced by 13% at 4053 hPa compared to sea level control (P = 0.017) and by 15% compared to 505 hPa (P = 0.012). Thus, there was a decreasing trend of DLNO with increasing ambient pressure (Fig. 1) and with an associated decreasing diffusivity for NO in the background gas.
Fig. 2. Same data as in Fig. 1, but with the inverse of DLNO plotted as a function of the ratio of ambient pressure over sea level pressure. A linear function has been fitted to the data; see also Eq. 3 for explanation.
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
value for DgNO of 2338 or about 2300 ml min−1 hPa−1 can be derived. As defined in Eq. (3) this is the value for 1015 hPa. Corresponding values for 505 and 4053 hPa would then be 4600 and 575 ml min−1 hPa−1 , respectively, where the diffusing coefficients are 2 and 0.25 times that at sea level. The alveolar volumes (VA ) averaged 7566 ± 859, 7420 ± 837, and 7567 ± 927 ml at 505, 1015, and 4053 hPa, respectively. Corresponding breath-hold times were 5.4 ± 0.1, 5.4 ± 0.2 and 5.9 ± 0.4 s. There were no significant differences between conditions for alveolar volume and breath-hold time. As a consequence of the constancy of VA across pressures, the DLNO /VA ratio (KNO ) differed in proportion to DLNO .
555
intraacinar gas transport on one hand, tending to reduce DLNO , and facilitated diffusive intraacinar gas transport and altitude acclimatization on the other, tending to increase DLNO . Interestingly, they found that one week of medication with an endothelin receptor A blocker improved DLNO significantly, suggesting that an increased pulmonary vascular resistance and interstitial edema impaired the DmNO component at altitude in the control subjects who received no medication (de Bisschop et al., 2012). In summary, the experimental design of these two very ambitious field studies were not such that the specific influences of the physical characteristics of the breathing gas (density and diffusivity in the gas phase) could be singled out. 4.1. DLNO and gas density
4. Discussion Our principal finding was that DLNO indeed was influenced by ambient pressure, as shown by the decreasing difference between the estimated DmNO and the measured DLNO values as ambient pressure is reduced from hyperbaric to hypobaric (Figs. 1 and 2). The present data also suggest that the DgNO factor is not really negligible at normal atmospheric pressure, rather they suggest that the resistance 1/DgNO then amounts to some 5% of the total resistance 1/DLNO . However, within the range of ambient pressures found at the altitudes where humans normally dwell the impact of pressure on DLNO is smaller and in most cases not of practical importance when comparing data obtained at different altitudes. In contrast, our results show that estimates of DmNO from DLNO values obtained during hyperbaric studies of lung diffusing capacity should be corrected for a 1/DgNO specific for the composition and pressure of the breathing gas. Thus, at 4053 hPa (4 atmospheres absolute, equivalent depth 30 m sea water) breathing a normoxic nitrogen-oxygen mixture, DmNO is underestimated by 17% if equated with DLNO. The above conclusions rest on the assumptions that DmNO and x × Vc are the same across ambient pressures. We consider these as highly reasonable assumptions, since there was normoxia in all three conditions, thereby avoiding the risk for changes in DmNO and Vc from hypoxia at altitude (de Bisschop et al., 2010) or pulmonary oxygen toxicity in hyperbaria (Lowry, 2002). To our knowledge DLNO has not been studied before during hypobaric, normobaric, and hyperbaric conditions while maintaining normoxia at all ambient pressures. The present group mean values found at 1015 hPa agree well with previous data in the literature. Thus, Zavorsky et al. (2008) found a group mean value of 108 ml min−1 hPa−1 (144 ml min−1 mmHg−1 ) in 130 healthy subjects of both genders as compared to the present group mean value of 113.4 ml min−1 hPa−1 (151.1 ml min−1 mmHg−1 ). Their subjects were on the average older and not as tall as the present group, which may account for the difference in mean values between the groups. Two earlier studies have reported data from measurements of single-breath DLNO obtained at altitudes similar to that used in the present study. Experimental conditions in these studies differed from that of the present study since subjects were exposed to the normal hypoxia of altitude and there were several days of acclimatization to the new environment (de Bisschop et al., 2010, 2012). In one study, lowlanders were investigated four days after arrival to about 4000 m altitude, and there was a 16% reduction of DLNO compared to sea level (de Bisschop et al., 2010). In another study, lowlanders were investigated 48 h after arrival to 5050 m altitude and after having hiked for one week, starting at 2800 m. Here, there was no difference between DLNO values obtained during rest at sea level and at extreme altitude. These authors discuss their findings as the result of counteracting influences between subclinical interstitial lung tissue edema and attenuated convective
The effects of the density of the pulmonary gas on DL has been addressed in earlier work using helium-oxygen as the breathing gas and determinations of DLCO with a steady-state technique which included tidal breathing. Both Guénard et al. (1978) and Kvale et al. (1975) found reductions of DLCO with reduced gas density and ascribed that to less efficient convective penetration of the inspired gas into the lung periphery, thereby prolonging the pathway for diffusive transport to the alveoli. It was suggested that the dominating effect of the increased diffusion coefficient for CO in the helium-oxygen gas, as compared to air, would not be faster axial diffusive transport but rather increased Taylor diffusion in small airways with a laminar flow pattern. Briefly, Taylor diffusion means a blunting of the central peak of the front of fresh inspired gas in an airway. Both Paiva and Engel (1979) and Nixon and Pack (1980), however, showed on theoretical grounds that there was no need to invoke Taylor dispersion as the principal mechanism for an inverse relationship between the diffusion coefficient for gases such as CO, O2 and NO and the efficiency of acinar gas transport. Rather, these authors argued that a denser background gas promoted a more optimal distribution of the inspired gas. Regardless of mechanism, there is much evidence that during tidal breathing and with a less dense background gas, the net effect is a slowing of acinar transport of inhaled gas species such as CO, O2 and NO, which are taken up by the blood, and a corresponding faster acinar transport with a denser gas (Christopherson and Hlastala, 1982; Johnson and Van Liew, 1974; Martin et al., 1972; Wood et al., 1976). There are, however, experimental evidence, that under certain other experimental conditions there can be a proportional rather than an inverse relationship between diffusion coefficient for NO in the lung gas and DLNO . Hsia et al. (2003) studied dogs which had been allowed to mature after a unilateral pulmectomy. They determined DLNO with a steady-state method and two different tidal volumes, and with helium-oxygen, air, and sulphur hexafluoride-oxygen as breathing gases. It was found that DLNO was reduced to half when going from helium to sulphur hexafluoride and using a relatively low tidal volume. Sham-operated control animals showed no differences between breathing gases, and the authors concluded that the elongated acinar airways in the pneumonectomized animals had created a longer diffusion pathway which made them sensitive to the lower diffusing coefficient for NO in the densest gas mixture. The steady-state techniques with tidal breathing in the above human and animal experiments are likely to have a different relationship between convective and diffusive acinar gas transport than the present single-breath technique. In the present study, subjects performed a rapid vital capacity inspiration of the NO-containing gas, presumably leading to as deep a penetration of the NO as possible into the periphery of the lungs, with little or no difference between gas densities. Thus, the remaining pathway for diffusive transport during the 5 s breath-holding period is likely to be relatively short compared to a situation with tidal breathing, but nevertheless giving rise to a small but significant
556
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
effect of gas density in the hyperbaric experiments, in which diffusivity for NO was reduced by a factor of four and eight respectively, i.e. compared to the normobaric and hypobaric cases. Gas density also increases the flow resistance, and therefore may potentially impair the ability to perform rapid inhalations and exhalations at the start and end of the DLNO maneuvers at the highest ambient pressure. The constancy of alveolar volumes between conditions, however, speak against such an influence of gas density on the ability to rapidly fill the lungs in a consistent manner at the start of the DLNO maneuvers in the present study. 4.2. Relevance for determination of exhaled NO and related parameters Exhaled NO is the net result of several parallel processes; NO formation, diffusive backtransport to the alveolar compartment with uptake to the pulmonary blood (Van Muylem et al., 2003), convective backtransport also contributing to alveolar uptake (Hemmingsson et al., 2012; Paiva and Engel, 1979) and dilution with the exhaled carrier gas (Pietropaoli et al., 1999). An implicit assumption so far in this paper has been that the diffusive and convective backtransport in the gas phase of the lung periphery, which can be estimated with the DLNO maneuver, reasonably well mimics that during diffusive and convective backtransport of naturally generated NO from conducting airways to the alveoli. A closer look at the two maneuvers, however, shows that they differ somewhat with respect to both the inhalation and exahalation phases: (a) Inhalation. Both the DLNO maneuver and the standard FENO maneuver (Fraction of exhaled NO, ATS Guidelines, 2011) start with a deep inspiration, which, however, is faster in the case of DLNO . This is so since in order to determine a breath-hold time of the order 3–4 s reasonably accurately it must start and end with fast inhalations and exhalations. A faster inhalation may tend to push the diffusion front more deeply and thereby shorten the diffusion pathway (Paiva, 1973). In contrast, there is no requirement for a very fast inhalation in the standard FENO maneuver (ATS Guidelines, 2011). (b) Breath-hold/Exhalation. Thereafter, there is a 3–4 s breath-hold in the case of DLNO , and this is the period over which DLNO and its subcomponents are determined. The FENO maneuver includes no explicit breath-hold but a slow exhalation at a rate of 50 ml s−1 (ATS Guidelines, 2011), which from a gas exchange point of view also is a breath-hold in the sense that no fresh gas then enters the lung periphery. This pseudo-breath-hold is usually of the order of 10 s. Due to the longer duration of the phase dominated by diffusion in the FENO maneuver, it is likely that the impact of diffusion here is larger than in the DLNO maneuver. On this basis we propose that the presently found impacts on DLNO of diffusive transport in the peripheral lung gas represent underestimates of the corresponding impacts on the backdiffusion of NO during a typical FENO maneuver. 5. Conclusions We have here established that the ambient pressure has an impact on the lung diffusing capacity for NO, most likely since changes in gas density cause inverse changes of diffusive NO transport in the peripheral lung gas. Quantitatively, this effect appears to be without practical consequence for the interpretation of DLNO values obtained at different altitudes. The same effect is still small but not really negligible at normal atmospheric pressure. Furthermore, the impact of gas density seen here on DLNO likely underestimates the impact on backdiffusion when analyzing the
naturally occurring NO turnover in the lungs in terms of alveolar estimates and peripheral conductive-airway contributions to the exhaled NO. In summary therefore, the present data are compatible with the model we presented in a recent paper (Hemmingsson et al., 2012), namely that with increasing ambient pressure, diffusive backtransport of NO is attenuated, but this effect appears to be offset by enhanced convective backtransport and/or feed-back control of the NO formation.
Disclosures No conflicts of interest, financial or otherwise, are declared by the authors.
Acknowledgements This study was supported by the Swedish National Space Board, Fraenckel’s Fund for Medical Research and the European Space Agency.
References ATS Guidelines, Dweik, R.A., Boggs, P.B., Erzurum, S.C., Irvin, C.G., Leigh, M.W., Lundberg, J.O., Olin, A.C., Plummer, A.L., Taylor, D.R., 2011. American Thoracic Society Committee on Interpretation of Exhaled Nitric Oxide Levels (FENO) for Clinical Applications. An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical application (with on-line supplement, pp 1–33). Am. J. Respir. Crit. Care Med. 184, 602–615. ATS/ERS task force, 2005. Standardisation of lung function testing. Edited by V. Brusasco R. Crapo and G. Viegi. Number 4 in this Series Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur. Respir. J. 26, 720–735. Borland, C.D., Dunningham, H., Bottrill, F., Vuylsteke, A., Yilmaz, C., Dane, D.M., Hsia, C.C., 2010. Significant blood resistance to nitric oxide transfer in the lung. J. Appl. Physiol. 108 (5), 1052–1060. Brown, D.E., Beall, C.M., Strohl, K.P., Mills, P.S., 2006. Exhaled nitric oxide decreases upon acute exposure to high-altitude hypoxia. Am. J. Hum. Biol. 18, 196–202. Busch, T., Bärtsch, P., Pappert, D., Grünig, E., Hildebrandt, W., Elser, H., Falke, K.J., Swenson, E.R., 2001. Hypoxia decreases exhaled nitric oxide in mountaineers susceptible to high-altitude pulmonary edema. Am. J. Respir. Crit. Care Med. 163, 368–373. Chang, H.K., 1985. General concepts of molecular diffusion. In: Engel, L.A., Paiva, M. (Eds.), Gas Mixing and Distribution in the Lung. Marcel Dekker, Inc., New York, USA., pp. 1–22. Christopherson, S.K., Hlastala, M.P., 1982. Pulmonary gas exchange during altered density gas breathing. J. Appl. Physiol. 52 (1), 221–225. Cotton, D.J., Graham, B.L., 2005. Single-breath carbon monoxide diffusing capacity or transfer factor. In: Hamid, Q., Shannon, J., Martin, J.G. (Eds.), Physiologic Basis of Respiratory Disease. B.C. Decker Inc, Canada, Chapter 57. de Bisschop, C., Kiger, L., Marden, M.C., Ajata, A., Huez, S., Faoro, V., Martinot, J.B., Naeije, R., Guénard, H., 2010. Pulmonary capillary blood volume and membrane conductance in Andeans and lowlanders at high altitude: a cross-sectional study. Nitric Oxide 23 (3), 187–193. de Bisschop, C., Martinot, J.B., Leurquin-Sterk, G., Faoro, V., Guénard, H., Naeije, R., 2012. Improvement in lung diffusion by endothelin A receptor blockade at high altitude. J. Appl. Physiol. 112 (1), 20–25. Duplain, H., Sartori, C., Lepori, M., Egli, M., Allemann, Y., Nicod, P., Scherrer, U., 2000. Exhaled nitric oxide in high-altitude pulmonary edema: role in the regulation of pulmonary vascular tone and evidence for a role against inflammation. Am. J. Respir. Crit. Care Med. 162, 221–224. Erzurum, S.C., Ghosh, S., Janocha, A.J., Xu, W., Bauer, S., Bryan, N.S., Tejero, J., Hemann, C., Hille, R., Stuehr, D.J., Feelisch, M., Beall, C.M., 2007. Higher blood flow and circulating NO products offset high-altitude hypoxia among Tibetans. Proc. Natl. Acad. Sci. U.S.A. 104, 17593–17598. Forster, R.E., Roughton, F.J., Cander, L., Briscoe, W.A., Kreuzer, F., 1957. Apparent pulmonary diffusing capacity for CO at varying alveolar O2 tensions. J. Appl. Physiol. 11, 277–289. Guénard, H., Chaussain, M., Lebeau, C., 1978. ratory gas exchange under normobaric helium-oxygen breathing at rest and during muscular exercise. Bull. Eur. Physiopathol. Respir. 14 (4), 417–429. Guénard, H., Varene, N., Vaida, P., 1987. Determination of lung capillary blood volume and membrane diffusing capacity in man by the measurement of NO and CO transfer. Respir. Physiol. 70 (1), 113–120. Hemmingsson, T., Horn, A., Linnarsson, D., 2009. Measuring exhaled nitric oxide at high altitude. Respir. Physiol. Neurobiol. 167, 292–298. Hemmingsson, T., Linnarsson, D., 2009. Lower exhaled nitric oxide in hypobaric than in normobaric acute hypoxia. Respir. Physiol. Neurobiol. 169, 74–77.
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557 Hemmingsson, T.E., Linnarsson, D., Frostell, C., Van Muylem, A., Kerckx, Y., Gustafsson, L.E., 2012. Effects of ambient pressure on pulmonary nitric oxide. J. Appl. Physiol. 112 (4), 580–586. Hsia, C.C., Yan, X., Dane, D.M., Johnson Jr., R.L., 2003. Density-dependent reduction of nitric oxide diffusing capacity after pneumonectomy. J. Appl. Physiol. 94 (5), 1926–1932. Jones, R.S., Meade, F., 1961. A theoretical and experimental analysis of anomalies in the estimation of pulmonary diffusing capacity by the single breath method. Q. J. Exp. Physiol. Cogn. Med. Sci. 46, 131–143. Johnson Jr., R.L., Heigenhauser, G.J.F., Hsia, C.C.W., Jones, N.L., Wagner, P.D., 1996. Determinants of gas exchange and acid-balance during exercise. In: Rowell, L.B., Shepard JT (Eds.), Handbook of Physiology. Section 12: Exercise: Regulation and Integration of Multiple Systems. Oxford University Press, New York, pp. 522–545. Johnson, L.R., Van Liew, H.D., 1974. Use of arterial PO2 to study convective and diffusive gas mixing in the lungs. J. Appl. Physiol. 36 (1), 91–97. Kvale, P.A., Davis, J., Schroter, R.C., 1975. Effect of gas density and ventilatory pattern of steady-state co uptake by the lung. Respir. Physiol. 24 (3), 385–398. Lowry, C., 2002. Oxygen toxicity. In: Edmonds, C., Lowry, C., Pennefather, J., Walker, R. (Eds.), Diving and Subaquatic Medicine. Edward Arnold Ltd, London, pp. 207–222. Martin, R.R., Zutter, M., Anthonisen, N.R., 1972. Pulmonary gas exchange in dogs breathing SF 6 at 4 ata. J. Appl. Physiol. 33 (1), 86–92. Martinot, J.B., Mulè, M., de Bisschop, C., Overbecek, M.J., Le-Dong, N.N., Naeije, R., Guénard, H., 2013. Lung membrane conductance and capillary volume derived from the NO and CO transfer in high altitude newcombers. J. Appl. Physiol. 115, 157–166. Meyer, M., Piiper, J., 1989. Nitric oxide (NO), a new test gas for study of alveolarcapillary diffusion. Eur. Respir. J. 2, 494–496. Nixon, W., Pack, A., 1980. Effect of altered gas diffusivity on alveolar gas exchange a theoretical study. J. Appl. Physiol. 48 (1), 147–153.
557
Paiva, M., 1973. Gas transport in the human lung. J. Appl. Physiol. 35, 401–410. Paiva, M., Engel, L.A., 1979. Pulmonary interdependence of gas transport. J. Appl. Physiol. 47 (2), 296–305. Persson, M.G., Zetterstrom, O., Agrenius, V., Ihre, E., Gustafsson, L.E., 1994. Singlebreath nitric oxide measurements in asthmatic patients and smokers. Lancet 15 (343), 146–147. Pietropaoli, A.P., Perillo, I.B., Torres, A., Perkins, P.T., Frasier, L.M., Utell, M.J., Frampton, M.W., Hyde, R.W., 1999. Simultaneous measurement of nitric oxide production by conducting and alveolar airways of humans. J. Appl. Physiol. 87, 1532–1542. Olin, A-C., Bake, B., Torén, K., 2007. Fraction of exhaled nitric oxide at 50 mL/s: reference values for adult lifelong never-smokers. Chest 131, 1852–1856. Shin, H-W., Condorelli, P., George, S.C., 2006. Examining axial diffusion of nitric oxide in the lungs using heliox and breath hold. J. Appl. Physiol. 100, 623–630. Tamhane, R.M., Johnson, R.L., Hsia, C.C.W., 2001. Pulmonary membrane diffusing capacity and capillary blood volume measured during exercise from nitric oxide uptake. Chest 120, 1850–1856. Taylor, D.R., Pijnenburg, M.W., Smith, A.D., De Jongste, J.C., 2006. Exhaled nitric oxide measurements: clinical application and interpretation. Thorax 61, 817–827. Van Muylem, A., Noël, C., Paiva, M., 2003. Modeling of impact of gas molecular diffusion on nitric oxide expired profile. J. Appl. Physiol. 94, 119–127. Wood, L.D., Bryan, A.C., Bau, S.K., Weng, T.R., Levison, H., 1976. Effect of increased gas density on pulmonary gas exchange in man. J. Appl. Physiol. 41 (2), 206–210. Zavorsky, G.S., Quiron, K.B., Massarelli, P.S., Lands, L.C., 2004. The relationship between single-breath diffusion capacity of the lung for nitric oxide and carbon monoxide during various exercise intensities. Chest 125 (3), 1019–1027. Zavorsky, G.S., Cao, J., Murias, J.M., 2008. Reference values of pulmonary diffusing capacity for nitric oxide in an adult population. Nitric Oxide 18, 70–79. Zavorsky, G.S., 2010. No red cell resistance to NO? I think not! J. Appl. Physiol. 108 (5), 1027–1029.
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Lung diffusing capacity for nitric oxide at lowered and raised ambient pressures Dag Linnarsson a,∗ , Tryggve E. Hemmingsson a,b , Claes Frostell a , Alain Van Muylem c , Yannick Kerckx d , Lars E. Gustafsson a a
Department of Physiology and Pharmacology, Karolinska Institutet, SE-17177 Stockholm, Sweden Department of Radiology, Karolinska University Hospital, Stockholm, Sweden c Chest Department, Erasme University Hospital, Brussels, Belgium d Biomedical Physics Laboratory, Université Libre de Bruxelles, Brussels, Belgium b
a r t i c l e
i n f o
Article history: Accepted 15 August 2013 Keywords: NO Gas density Diffusivity Hypobaria Hyperbaria
a b s t r a c t Lung diffusing capacity for NO (DLNO ) was determined in eight subjects at ambient pressures of 505, 1015, and 4053 hPa (379, 761 and 3040 mmHg) as they breathed normoxic gases. Mean values were 116.9 ± 11.1 (SEM), 113.4 ± 11.1 and 99.3 ± 10.1 ml min−1 hPa−1 at 505, 1015, and 4053 hPa, with a 13% difference between the two higher pressures (P = 0.017). The data were applied to a model with two serially coupled conductances; the gas phase (DgNO , variable with pressure), and the alveolo-capillary membrane (DmNO , constant). The data fitted the model well and we conclude that diffusive transport of NO in the peripheral lung is inversely related to gas density. At normal pressure DmNO was approximately 5% larger than DLNO , suggesting that the Dg factor then is not negligible. We also conclude that the density of the breathing gas is likely to impact the backdiffusion of naturally formed NO from conducting airways to the alveoli. © 2013 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Background Measurements of pulmonary nitric oxide (NO) are of scientific and clinical interest mainly in two areas; (1) to monitor the activity of inflammatory airway disease from exhaled NO of endogenous origin (Persson et al., 1994; Olin et al., 2007; ATS Guidelines, 2011), and (2) to measure lung diffusing capacity (DLNO ) from the rate of uptake of NO from an external source. In the former case the NO level is several orders of magnitude lower (ppb) than that inhaled during a DLNO maneuver (ppm) (Guénard et al., 1987; Zavorsky et al., 2008). 1.1.1. Exhaled NO Nitric oxide in the lungs has been implicated as a potentially protective factor against pulmonary manifestations of acute mountain sickness (Busch et al., 2001; Duplain et al., 2000; Erzurum et al., 2007). It has therefore been considered of interest to establish what the effects are of a reduced ambient air pressure on pulmonary NO in healthy individuals. A second consideration is to define normal values for altitude residents, when exhaled NO is
∗ Corresponding author. Tel.: +46 8 52486890; fax: + 46 8 304613. E-mail address: [email protected] (D. Linnarsson). 1569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.resp.2013.08.008
used to monitor the activity of inflammatory airway diseases such as asthma (Olin et al., 2007; Taylor et al., 2006). Potential effects of altitude include the influence of hypoxia per se (Brown et al., 2006; Hemmingsson et al., 2009), of the reduced gas density (Shin et al., 2006; Van Muylem et al., 2003) and a combination of these two factors (Hemmingsson and Linnarsson, 2009). The latter authors showed that a short-lasting exposure to normobaric hypoxia down to 10% of an atmosphere had no significant effect on exhaled NO. We recently addressed the effects of gas density on exhaled NO by exposing healthy subjects to both reduced and increased ambient pressures while at the same time maintaining normoxia in the breathing gases. Based on previous experiments with heliumoxygen breathing (Shin et al., 2006; Van Muylem et al., 2003) we had expected that the lowered gas density at altitude and the associated increased diffusivity for NO in the lung gas would increase backdiffusion of NO to the alveoli. In turn this would increase the uptake of NO to the blood resulting in a reduced partial pressure of NO in the exhaled gas (PENO ). We also expected corresponding mechanisms to increase PENO at hyperbaric pressure. However, recent work from our laboratory has shown that PENO values were strikingly similar in an ambient pressure range from 0.5 to 4.0 atmospheres (Hemmingsson et al., 2012). We hypothesized that diffusive transport of NO in the gas phase in the lungs would indeed be influenced by changes of the gas density, but that these effects would be relatively small and possibly be concealed by simultaneous effects of density on convective gas transport acting
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
in opposite direction (Hemmingsson et al., 2012). We reasoned that direct determinations of the lung diffusing capacity for NO (DLNO ) would be a way to test this hypothesis, so we undertook an additional study, now focusing on DLNO while using an otherwise identical experimental design and studying the same subjects. 1.1.2. Lung diffusing capacity Pulmonary edema is a feared component of acute mountain sickness and determinations of DLNO have been used to detect interstitial pulmonary edema in otherwise healthy individuals during their stay at high altitude (de Bisschop et al., 2010, 2012). Conventionally, DLNO is used to estimate the capacity for diffusive transport through the alveolo-capillary membrane, under the assumption that any effects of diffusive transport through the gas phase in the respiratory zone of the lung can be neglected (Guénard et al., 1987; Zavorsky et al., 2008). We hypothesized that such an effect indeed is small but not necessarily negligible at all ambient pressures. We reasoned that by comparing DLNO data obtained with background gases of widely differing diffusivity for NO, we would be able to estimate the quantitative role of diffusive transport through the gas phase for DLNO .
553
possible to detect if conditions with sufficiently large differences in density and therefore also diffusivity were compared. Eq. (2) may be further developed by assuming that: (a) DgNO should vary in proportion to the diffusion coefficient for NO in the breathing gas. This parameter in turn varies inversely with the ambient pressure (Chang, 1985). (b) The term 1/DmNO is constant across conditions. For a range of pressures and breathing gas mixtures consistent with the above assumptions we then obtain for a given pressure: 1 = DLNO
1 ∗ P DgNO
P0
+
1 DmNO
(3)
where P/P0 is the ratio between the ambient pressure and the pressure for which DgNO is defined. The equation has the format of a linear relationship where the slope is 1/DgNO and the intercept with the Y axis is 1/DmNO . 2. Methods 2.1. Subjects
1.2. Theory The general way to characterize the passage of inhaled or endogenously formed gases from the airways to the pulmonarycapillary blood is to determine the diffusing capacity. Forster and Roughton (Forster et al., 1957) developed an algorithm to determine the diffusing capacity of the lung for a specific gas (DL), or transfer factor as it is usually called outside the United States. Below, the term DL will be used and, according to the classical model, expressed as serially connected resistances to diffusion, from the airway gas to the inside of the erythrocyte. Each subcomponent of the serial reaction chain is hence the inverse of a conductance involving both passive diffusion and chemical binding with hemoglobin (Hb). The complete transfer reaction of the lung DL for a gas x can be explained as three serially linked resistances (Cotton and Graham, 2005; Johnson et al., 1996); 1 1 1 1 + + = DL Dgx Dmx (x · Vc )
(1)
defined by the conductance for x in the background gas of the respiratory zone (Dgx ), the conductance for x through the alveolocapillary membrane (Dmx ), the conductance for uptake of gas x within the erythrocyte ( x ), and by the volume of the pulmonarycapillary blood (Vc ). The factor describing diffusion in the gas phase (Dgx ) is usually considered negligible for inhaled test gases such as CO, whereas both the amount and the oxygen saturation of the hemoglobin in the lung capillaries have measurable impacts on DLCO . The reaction of NO with hemoglobin in the erythrocyte is about 280 times faster than that of CO (Meyer and Piiper, 1989; Tamhane et al., 2001; Zavorsky et al., 2008). Therefore, the resistance to the NO transport within the erythrocyte has been considered negligible (Johnson et al., 1996), and the third term in Eq. (1) (1/ x × Vc ) could therefore be omitted in a corresponding equation for NO (Zavorsky, 2010). However, other authors have considered this term to have a finite value (Borland et al., 2010), see Martinot et al. (2013) for a detailed discussion. For the purpose of the present analysis, however, the results are based on the assumption that this third term could be neglected, thereby obtaining: 1 1 1 = + DLNO DgNO DmNO
(2)
Since DL for NO is about four times that for CO (Guénard et al., 1978; Zavorsky et al., 2004), we thought that the impact of diffusivity in the gas phase (as quantified by the 1/DgNO term) would be
Eight healthy non-smoking subjects without a history of inflammatory airway disease participated in the current study. The subjects came to the laboratory once for familiarization, physical examination, spirometry and baseline FENO measurement. They returned twice for experiments in increased or decreased ambient pressure in a combined hyperbaric and hypobaric pressure chamber. All eight subjects, four women, completed all tests at all pressure levels. Their age, height and weight ranged 21–37 years, 1.60–1.93 m and 58–87 kg, respectively. 2.2. Instrumentation and measurements Experiments were performed at 505 ± 0 (mean ± SEM), 1015 ± 3 and 4053 ± 0 hPa ambient pressure. The corresponding values in mmHg were 379, 761 and 3040. The pressure chamber (internal volume 8 m3 ) was pressurized with air but subjects breathed normoxic gas mixtures with oxygen fractions of 0.421, 0.2095 and 0.052 at 505, 1015 and 4053 hPa, respectively. Subjects were investigated at hypobaric and hyperbaric pressures on different days, in random order and always starting with control measurements at 1015 hPa. The lung diffusing capacity for NO (DLNO ) was measured in two subjects at a time, seated together with one of the test supervisors in the pressure chamber, using the standardized ATS/ERS techniques for diffusing capacity (ATS/ERS, 2005), which is based on the Jones and Meade methodology (Jones and Meade, 1961), with a modified 5 s breath-holding time (Zavorsky et al., 2008). After a change in ambient pressure, a 15 min waiting time was allowed to accommodate to the new environment while breathing the normoxic gas mixture. Decompression after the hyperbaric experiments was performed according to Swedish Navy standard tables and with correction for the increased nitrogen partial pressure compared to air breathing. There were no decompression symptoms in any of the subjects. Between DLNO determinations subjects were breathing through an oronasal mask and a non-rebreathing valve (Hans Rudolph Inc., Shawnee, KS, USA) from a 200 l Douglas bag via 40 mm inner diameter hoses. The supervisor kept the bag adequately filled by means of a needle valve connected to the reducing valve of tanks with compressed gas housed inside the chamber and quick-connect fittings allowed for simple change of gas source when required. Gas for argon (Ar) analysis was sampled to a mass spectrometer (lnnovision A/S, Odense, Denmark) located outside the pressure chamber. Gas for NO analysis was sampled to a chemiluminescence analyzer (Eco
554
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
Medics AG, Duernten, Switzerland) also located outside the pressure chamber. Sampling was performed first through a 1.5 m long Nafion catheter and then via a 2 mm inner diameter polytetrafluoroethylene (PTFE) tube connected via Luer fittings to the inlet of the analyzer (505 and 1015 hPa). During the hyperbaric experiments there was a three-way connector at the NO analyzer inlet. Sampling flow from the subject to the three-way connector was driven by the chamber pressure, whereas the pump of the analyzer sampled a part of this flow for analysis. The NO analyzer was calibrated with certified gas mixtures with 2000 ppb NO in nitrogen (AGA SpecialGas AB, Lidingö, Sweden). The mass spectrometer was calibrated with 10% Ar (AGA SpecialGas AB, Lidingö, Sweden). Respired flow was monitored with a Hans Rudolph 3700 series pneumotachometer coupled to a differential pressure transducer (Validyne Engineering, Northridge, CA, USA). The flow signal was fed through the chamber wall from the inside to the outside, where data including those from the gas analyzers, were collected at 100 Hz with a digital data acquisition system (Biopac Systems MP150CE (Biopac Systems Inc., Goleta, CA, USA)). 2.3. Diffusion capacity measurement procedure The subject shifted from the oronasal mask to the mouthpiece of the DLNO assembly while holding his/her breath and donning a nose clip. Then the diffusion capacity measurement was initiated by the subject taking 3–4 breaths at tidal breathing after which he/she performed a maximal expiration to residual volume (RV). The test supervisor then turned a rotational valve for inhalation of a normoxic test gas mixture with 10 ppm NO and 10% Ar, from a 10 l bag. After taking a deep, quick breath to total lung capacity, the subject made a 3–4 s breath-hold and then performed a rapid full expiration to RV. The supervisor then turned the rotational valve back to tidal breathing position. 2.4. Data analysis For each subject, a set of calibrated values for inhaled and exhaled NO and Ar, and flow were collected at each ambient pressure. DLNO (in ml min−1 hPa−1 ) was calculated according to current standards (ATS/ERS, 2005). A combined anatomical and instrument dead space of 400 ml was assumed. The timing of the maneuver was determined from the flow recording. Gas concentration signals were aligned with the flow recording by compensating for the sampling delays of the analyzers.
Fig. 1. Diffusing capacity for NO (DLNO ) in eight subjects inhaling 10 ppm NO in a stable normoxic environment at 505, 1015 and 4053 hPa ambient pressure, respectively. The horizontal line indicates the diffusing capacity through the alveolocapillary membrane (DmNO ), which represents the theoretical maximum value for DLNO , should there be an infinite rate of diffusion through the gas in the respiratory zone (1/DgNO = 0, see Eq. 2 for explanation).
In Fig. 2 the inverse of the same data as in Fig. 1 have been plotted as a function of the ratio between the ambient pressure and the sea level pressure, as described in Eq. (3) above. There was an excellent fit with a linear function, the intercept of which with the Y axis shows the value of 1/DLNO with an infinite diffusion coefficient, so that DLNO then must be equal to DmNO . The relationship in Fig. 2 is therefore compatible with the assumption that the value of DmNO is constant across conditions and varies with DgNO , which in turn varies with the diffusion coefficient and the inverse of the ambient pressure. The same relationship yields a DmNO value of 119.6 ml min−1 hPa−1 . This can be regarded as a theoretical maximum value for DLNO , should there be no resistance to NO transfer in the gas phase. From the slope of the function a
2.5. Ethics and statistics The study was approved by the Regional Ethical Review Board in Stockholm, Sweden and all subjects gave their written informed consent. Data are presented as group means and standard error of the mean (SEM) for each ambient condition. Data were analyzed by using a Wilcoxon matched pairs test (StatSoft, Tulsa, OK, USA) with a Bonferroni correction for multiple comparisons. A P-value of <0.05 was considered statistically significant. 3. Results Our group mean results are shown in Fig. 1. Values there are in ml min−1 hPa−1 . Expressed instead per unit mmHg, DLNO was 155.8, 151.1 and 132.4 ml min−1 at 505, 1015 and 4053 hPa, respectively. There was no significant difference between 505 hPa and control (P = 0.17), whereas DLNO was significantly reduced by 13% at 4053 hPa compared to sea level control (P = 0.017) and by 15% compared to 505 hPa (P = 0.012). Thus, there was a decreasing trend of DLNO with increasing ambient pressure (Fig. 1) and with an associated decreasing diffusivity for NO in the background gas.
Fig. 2. Same data as in Fig. 1, but with the inverse of DLNO plotted as a function of the ratio of ambient pressure over sea level pressure. A linear function has been fitted to the data; see also Eq. 3 for explanation.
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
value for DgNO of 2338 or about 2300 ml min−1 hPa−1 can be derived. As defined in Eq. (3) this is the value for 1015 hPa. Corresponding values for 505 and 4053 hPa would then be 4600 and 575 ml min−1 hPa−1 , respectively, where the diffusing coefficients are 2 and 0.25 times that at sea level. The alveolar volumes (VA ) averaged 7566 ± 859, 7420 ± 837, and 7567 ± 927 ml at 505, 1015, and 4053 hPa, respectively. Corresponding breath-hold times were 5.4 ± 0.1, 5.4 ± 0.2 and 5.9 ± 0.4 s. There were no significant differences between conditions for alveolar volume and breath-hold time. As a consequence of the constancy of VA across pressures, the DLNO /VA ratio (KNO ) differed in proportion to DLNO .
555
intraacinar gas transport on one hand, tending to reduce DLNO , and facilitated diffusive intraacinar gas transport and altitude acclimatization on the other, tending to increase DLNO . Interestingly, they found that one week of medication with an endothelin receptor A blocker improved DLNO significantly, suggesting that an increased pulmonary vascular resistance and interstitial edema impaired the DmNO component at altitude in the control subjects who received no medication (de Bisschop et al., 2012). In summary, the experimental design of these two very ambitious field studies were not such that the specific influences of the physical characteristics of the breathing gas (density and diffusivity in the gas phase) could be singled out. 4.1. DLNO and gas density
4. Discussion Our principal finding was that DLNO indeed was influenced by ambient pressure, as shown by the decreasing difference between the estimated DmNO and the measured DLNO values as ambient pressure is reduced from hyperbaric to hypobaric (Figs. 1 and 2). The present data also suggest that the DgNO factor is not really negligible at normal atmospheric pressure, rather they suggest that the resistance 1/DgNO then amounts to some 5% of the total resistance 1/DLNO . However, within the range of ambient pressures found at the altitudes where humans normally dwell the impact of pressure on DLNO is smaller and in most cases not of practical importance when comparing data obtained at different altitudes. In contrast, our results show that estimates of DmNO from DLNO values obtained during hyperbaric studies of lung diffusing capacity should be corrected for a 1/DgNO specific for the composition and pressure of the breathing gas. Thus, at 4053 hPa (4 atmospheres absolute, equivalent depth 30 m sea water) breathing a normoxic nitrogen-oxygen mixture, DmNO is underestimated by 17% if equated with DLNO. The above conclusions rest on the assumptions that DmNO and x × Vc are the same across ambient pressures. We consider these as highly reasonable assumptions, since there was normoxia in all three conditions, thereby avoiding the risk for changes in DmNO and Vc from hypoxia at altitude (de Bisschop et al., 2010) or pulmonary oxygen toxicity in hyperbaria (Lowry, 2002). To our knowledge DLNO has not been studied before during hypobaric, normobaric, and hyperbaric conditions while maintaining normoxia at all ambient pressures. The present group mean values found at 1015 hPa agree well with previous data in the literature. Thus, Zavorsky et al. (2008) found a group mean value of 108 ml min−1 hPa−1 (144 ml min−1 mmHg−1 ) in 130 healthy subjects of both genders as compared to the present group mean value of 113.4 ml min−1 hPa−1 (151.1 ml min−1 mmHg−1 ). Their subjects were on the average older and not as tall as the present group, which may account for the difference in mean values between the groups. Two earlier studies have reported data from measurements of single-breath DLNO obtained at altitudes similar to that used in the present study. Experimental conditions in these studies differed from that of the present study since subjects were exposed to the normal hypoxia of altitude and there were several days of acclimatization to the new environment (de Bisschop et al., 2010, 2012). In one study, lowlanders were investigated four days after arrival to about 4000 m altitude, and there was a 16% reduction of DLNO compared to sea level (de Bisschop et al., 2010). In another study, lowlanders were investigated 48 h after arrival to 5050 m altitude and after having hiked for one week, starting at 2800 m. Here, there was no difference between DLNO values obtained during rest at sea level and at extreme altitude. These authors discuss their findings as the result of counteracting influences between subclinical interstitial lung tissue edema and attenuated convective
The effects of the density of the pulmonary gas on DL has been addressed in earlier work using helium-oxygen as the breathing gas and determinations of DLCO with a steady-state technique which included tidal breathing. Both Guénard et al. (1978) and Kvale et al. (1975) found reductions of DLCO with reduced gas density and ascribed that to less efficient convective penetration of the inspired gas into the lung periphery, thereby prolonging the pathway for diffusive transport to the alveoli. It was suggested that the dominating effect of the increased diffusion coefficient for CO in the helium-oxygen gas, as compared to air, would not be faster axial diffusive transport but rather increased Taylor diffusion in small airways with a laminar flow pattern. Briefly, Taylor diffusion means a blunting of the central peak of the front of fresh inspired gas in an airway. Both Paiva and Engel (1979) and Nixon and Pack (1980), however, showed on theoretical grounds that there was no need to invoke Taylor dispersion as the principal mechanism for an inverse relationship between the diffusion coefficient for gases such as CO, O2 and NO and the efficiency of acinar gas transport. Rather, these authors argued that a denser background gas promoted a more optimal distribution of the inspired gas. Regardless of mechanism, there is much evidence that during tidal breathing and with a less dense background gas, the net effect is a slowing of acinar transport of inhaled gas species such as CO, O2 and NO, which are taken up by the blood, and a corresponding faster acinar transport with a denser gas (Christopherson and Hlastala, 1982; Johnson and Van Liew, 1974; Martin et al., 1972; Wood et al., 1976). There are, however, experimental evidence, that under certain other experimental conditions there can be a proportional rather than an inverse relationship between diffusion coefficient for NO in the lung gas and DLNO . Hsia et al. (2003) studied dogs which had been allowed to mature after a unilateral pulmectomy. They determined DLNO with a steady-state method and two different tidal volumes, and with helium-oxygen, air, and sulphur hexafluoride-oxygen as breathing gases. It was found that DLNO was reduced to half when going from helium to sulphur hexafluoride and using a relatively low tidal volume. Sham-operated control animals showed no differences between breathing gases, and the authors concluded that the elongated acinar airways in the pneumonectomized animals had created a longer diffusion pathway which made them sensitive to the lower diffusing coefficient for NO in the densest gas mixture. The steady-state techniques with tidal breathing in the above human and animal experiments are likely to have a different relationship between convective and diffusive acinar gas transport than the present single-breath technique. In the present study, subjects performed a rapid vital capacity inspiration of the NO-containing gas, presumably leading to as deep a penetration of the NO as possible into the periphery of the lungs, with little or no difference between gas densities. Thus, the remaining pathway for diffusive transport during the 5 s breath-holding period is likely to be relatively short compared to a situation with tidal breathing, but nevertheless giving rise to a small but significant
556
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557
effect of gas density in the hyperbaric experiments, in which diffusivity for NO was reduced by a factor of four and eight respectively, i.e. compared to the normobaric and hypobaric cases. Gas density also increases the flow resistance, and therefore may potentially impair the ability to perform rapid inhalations and exhalations at the start and end of the DLNO maneuvers at the highest ambient pressure. The constancy of alveolar volumes between conditions, however, speak against such an influence of gas density on the ability to rapidly fill the lungs in a consistent manner at the start of the DLNO maneuvers in the present study. 4.2. Relevance for determination of exhaled NO and related parameters Exhaled NO is the net result of several parallel processes; NO formation, diffusive backtransport to the alveolar compartment with uptake to the pulmonary blood (Van Muylem et al., 2003), convective backtransport also contributing to alveolar uptake (Hemmingsson et al., 2012; Paiva and Engel, 1979) and dilution with the exhaled carrier gas (Pietropaoli et al., 1999). An implicit assumption so far in this paper has been that the diffusive and convective backtransport in the gas phase of the lung periphery, which can be estimated with the DLNO maneuver, reasonably well mimics that during diffusive and convective backtransport of naturally generated NO from conducting airways to the alveoli. A closer look at the two maneuvers, however, shows that they differ somewhat with respect to both the inhalation and exahalation phases: (a) Inhalation. Both the DLNO maneuver and the standard FENO maneuver (Fraction of exhaled NO, ATS Guidelines, 2011) start with a deep inspiration, which, however, is faster in the case of DLNO . This is so since in order to determine a breath-hold time of the order 3–4 s reasonably accurately it must start and end with fast inhalations and exhalations. A faster inhalation may tend to push the diffusion front more deeply and thereby shorten the diffusion pathway (Paiva, 1973). In contrast, there is no requirement for a very fast inhalation in the standard FENO maneuver (ATS Guidelines, 2011). (b) Breath-hold/Exhalation. Thereafter, there is a 3–4 s breath-hold in the case of DLNO , and this is the period over which DLNO and its subcomponents are determined. The FENO maneuver includes no explicit breath-hold but a slow exhalation at a rate of 50 ml s−1 (ATS Guidelines, 2011), which from a gas exchange point of view also is a breath-hold in the sense that no fresh gas then enters the lung periphery. This pseudo-breath-hold is usually of the order of 10 s. Due to the longer duration of the phase dominated by diffusion in the FENO maneuver, it is likely that the impact of diffusion here is larger than in the DLNO maneuver. On this basis we propose that the presently found impacts on DLNO of diffusive transport in the peripheral lung gas represent underestimates of the corresponding impacts on the backdiffusion of NO during a typical FENO maneuver. 5. Conclusions We have here established that the ambient pressure has an impact on the lung diffusing capacity for NO, most likely since changes in gas density cause inverse changes of diffusive NO transport in the peripheral lung gas. Quantitatively, this effect appears to be without practical consequence for the interpretation of DLNO values obtained at different altitudes. The same effect is still small but not really negligible at normal atmospheric pressure. Furthermore, the impact of gas density seen here on DLNO likely underestimates the impact on backdiffusion when analyzing the
naturally occurring NO turnover in the lungs in terms of alveolar estimates and peripheral conductive-airway contributions to the exhaled NO. In summary therefore, the present data are compatible with the model we presented in a recent paper (Hemmingsson et al., 2012), namely that with increasing ambient pressure, diffusive backtransport of NO is attenuated, but this effect appears to be offset by enhanced convective backtransport and/or feed-back control of the NO formation.
Disclosures No conflicts of interest, financial or otherwise, are declared by the authors.
Acknowledgements This study was supported by the Swedish National Space Board, Fraenckel’s Fund for Medical Research and the European Space Agency.
References ATS Guidelines, Dweik, R.A., Boggs, P.B., Erzurum, S.C., Irvin, C.G., Leigh, M.W., Lundberg, J.O., Olin, A.C., Plummer, A.L., Taylor, D.R., 2011. American Thoracic Society Committee on Interpretation of Exhaled Nitric Oxide Levels (FENO) for Clinical Applications. An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical application (with on-line supplement, pp 1–33). Am. J. Respir. Crit. Care Med. 184, 602–615. ATS/ERS task force, 2005. Standardisation of lung function testing. Edited by V. Brusasco R. Crapo and G. Viegi. Number 4 in this Series Standardisation of the single-breath determination of carbon monoxide uptake in the lung. Eur. Respir. J. 26, 720–735. Borland, C.D., Dunningham, H., Bottrill, F., Vuylsteke, A., Yilmaz, C., Dane, D.M., Hsia, C.C., 2010. Significant blood resistance to nitric oxide transfer in the lung. J. Appl. Physiol. 108 (5), 1052–1060. Brown, D.E., Beall, C.M., Strohl, K.P., Mills, P.S., 2006. Exhaled nitric oxide decreases upon acute exposure to high-altitude hypoxia. Am. J. Hum. Biol. 18, 196–202. Busch, T., Bärtsch, P., Pappert, D., Grünig, E., Hildebrandt, W., Elser, H., Falke, K.J., Swenson, E.R., 2001. Hypoxia decreases exhaled nitric oxide in mountaineers susceptible to high-altitude pulmonary edema. Am. J. Respir. Crit. Care Med. 163, 368–373. Chang, H.K., 1985. General concepts of molecular diffusion. In: Engel, L.A., Paiva, M. (Eds.), Gas Mixing and Distribution in the Lung. Marcel Dekker, Inc., New York, USA., pp. 1–22. Christopherson, S.K., Hlastala, M.P., 1982. Pulmonary gas exchange during altered density gas breathing. J. Appl. Physiol. 52 (1), 221–225. Cotton, D.J., Graham, B.L., 2005. Single-breath carbon monoxide diffusing capacity or transfer factor. In: Hamid, Q., Shannon, J., Martin, J.G. (Eds.), Physiologic Basis of Respiratory Disease. B.C. Decker Inc, Canada, Chapter 57. de Bisschop, C., Kiger, L., Marden, M.C., Ajata, A., Huez, S., Faoro, V., Martinot, J.B., Naeije, R., Guénard, H., 2010. Pulmonary capillary blood volume and membrane conductance in Andeans and lowlanders at high altitude: a cross-sectional study. Nitric Oxide 23 (3), 187–193. de Bisschop, C., Martinot, J.B., Leurquin-Sterk, G., Faoro, V., Guénard, H., Naeije, R., 2012. Improvement in lung diffusion by endothelin A receptor blockade at high altitude. J. Appl. Physiol. 112 (1), 20–25. Duplain, H., Sartori, C., Lepori, M., Egli, M., Allemann, Y., Nicod, P., Scherrer, U., 2000. Exhaled nitric oxide in high-altitude pulmonary edema: role in the regulation of pulmonary vascular tone and evidence for a role against inflammation. Am. J. Respir. Crit. Care Med. 162, 221–224. Erzurum, S.C., Ghosh, S., Janocha, A.J., Xu, W., Bauer, S., Bryan, N.S., Tejero, J., Hemann, C., Hille, R., Stuehr, D.J., Feelisch, M., Beall, C.M., 2007. Higher blood flow and circulating NO products offset high-altitude hypoxia among Tibetans. Proc. Natl. Acad. Sci. U.S.A. 104, 17593–17598. Forster, R.E., Roughton, F.J., Cander, L., Briscoe, W.A., Kreuzer, F., 1957. Apparent pulmonary diffusing capacity for CO at varying alveolar O2 tensions. J. Appl. Physiol. 11, 277–289. Guénard, H., Chaussain, M., Lebeau, C., 1978. ratory gas exchange under normobaric helium-oxygen breathing at rest and during muscular exercise. Bull. Eur. Physiopathol. Respir. 14 (4), 417–429. Guénard, H., Varene, N., Vaida, P., 1987. Determination of lung capillary blood volume and membrane diffusing capacity in man by the measurement of NO and CO transfer. Respir. Physiol. 70 (1), 113–120. Hemmingsson, T., Horn, A., Linnarsson, D., 2009. Measuring exhaled nitric oxide at high altitude. Respir. Physiol. Neurobiol. 167, 292–298. Hemmingsson, T., Linnarsson, D., 2009. Lower exhaled nitric oxide in hypobaric than in normobaric acute hypoxia. Respir. Physiol. Neurobiol. 169, 74–77.
D. Linnarsson et al. / Respiratory Physiology & Neurobiology 189 (2013) 552–557 Hemmingsson, T.E., Linnarsson, D., Frostell, C., Van Muylem, A., Kerckx, Y., Gustafsson, L.E., 2012. Effects of ambient pressure on pulmonary nitric oxide. J. Appl. Physiol. 112 (4), 580–586. Hsia, C.C., Yan, X., Dane, D.M., Johnson Jr., R.L., 2003. Density-dependent reduction of nitric oxide diffusing capacity after pneumonectomy. J. Appl. Physiol. 94 (5), 1926–1932. Jones, R.S., Meade, F., 1961. A theoretical and experimental analysis of anomalies in the estimation of pulmonary diffusing capacity by the single breath method. Q. J. Exp. Physiol. Cogn. Med. Sci. 46, 131–143. Johnson Jr., R.L., Heigenhauser, G.J.F., Hsia, C.C.W., Jones, N.L., Wagner, P.D., 1996. Determinants of gas exchange and acid-balance during exercise. In: Rowell, L.B., Shepard JT (Eds.), Handbook of Physiology. Section 12: Exercise: Regulation and Integration of Multiple Systems. Oxford University Press, New York, pp. 522–545. Johnson, L.R., Van Liew, H.D., 1974. Use of arterial PO2 to study convective and diffusive gas mixing in the lungs. J. Appl. Physiol. 36 (1), 91–97. Kvale, P.A., Davis, J., Schroter, R.C., 1975. Effect of gas density and ventilatory pattern of steady-state co uptake by the lung. Respir. Physiol. 24 (3), 385–398. Lowry, C., 2002. Oxygen toxicity. In: Edmonds, C., Lowry, C., Pennefather, J., Walker, R. (Eds.), Diving and Subaquatic Medicine. Edward Arnold Ltd, London, pp. 207–222. Martin, R.R., Zutter, M., Anthonisen, N.R., 1972. Pulmonary gas exchange in dogs breathing SF 6 at 4 ata. J. Appl. Physiol. 33 (1), 86–92. Martinot, J.B., Mulè, M., de Bisschop, C., Overbecek, M.J., Le-Dong, N.N., Naeije, R., Guénard, H., 2013. Lung membrane conductance and capillary volume derived from the NO and CO transfer in high altitude newcombers. J. Appl. Physiol. 115, 157–166. Meyer, M., Piiper, J., 1989. Nitric oxide (NO), a new test gas for study of alveolarcapillary diffusion. Eur. Respir. J. 2, 494–496. Nixon, W., Pack, A., 1980. Effect of altered gas diffusivity on alveolar gas exchange a theoretical study. J. Appl. Physiol. 48 (1), 147–153.
557
Paiva, M., 1973. Gas transport in the human lung. J. Appl. Physiol. 35, 401–410. Paiva, M., Engel, L.A., 1979. Pulmonary interdependence of gas transport. J. Appl. Physiol. 47 (2), 296–305. Persson, M.G., Zetterstrom, O., Agrenius, V., Ihre, E., Gustafsson, L.E., 1994. Singlebreath nitric oxide measurements in asthmatic patients and smokers. Lancet 15 (343), 146–147. Pietropaoli, A.P., Perillo, I.B., Torres, A., Perkins, P.T., Frasier, L.M., Utell, M.J., Frampton, M.W., Hyde, R.W., 1999. Simultaneous measurement of nitric oxide production by conducting and alveolar airways of humans. J. Appl. Physiol. 87, 1532–1542. Olin, A-C., Bake, B., Torén, K., 2007. Fraction of exhaled nitric oxide at 50 mL/s: reference values for adult lifelong never-smokers. Chest 131, 1852–1856. Shin, H-W., Condorelli, P., George, S.C., 2006. Examining axial diffusion of nitric oxide in the lungs using heliox and breath hold. J. Appl. Physiol. 100, 623–630. Tamhane, R.M., Johnson, R.L., Hsia, C.C.W., 2001. Pulmonary membrane diffusing capacity and capillary blood volume measured during exercise from nitric oxide uptake. Chest 120, 1850–1856. Taylor, D.R., Pijnenburg, M.W., Smith, A.D., De Jongste, J.C., 2006. Exhaled nitric oxide measurements: clinical application and interpretation. Thorax 61, 817–827. Van Muylem, A., Noël, C., Paiva, M., 2003. Modeling of impact of gas molecular diffusion on nitric oxide expired profile. J. Appl. Physiol. 94, 119–127. Wood, L.D., Bryan, A.C., Bau, S.K., Weng, T.R., Levison, H., 1976. Effect of increased gas density on pulmonary gas exchange in man. J. Appl. Physiol. 41 (2), 206–210. Zavorsky, G.S., Quiron, K.B., Massarelli, P.S., Lands, L.C., 2004. The relationship between single-breath diffusion capacity of the lung for nitric oxide and carbon monoxide during various exercise intensities. Chest 125 (3), 1019–1027. Zavorsky, G.S., Cao, J., Murias, J.M., 2008. Reference values of pulmonary diffusing capacity for nitric oxide in an adult population. Nitric Oxide 18, 70–79. Zavorsky, G.S., 2010. No red cell resistance to NO? I think not! J. Appl. Physiol. 108 (5), 1027–1029.