Does High Alveolar Fluid Reabsorption Prevent HAPE in Individuals with Exaggerated Pulmonary Hypertension in Hypoxia?

Abstract

Betz, Theresa, Christoph Dehnert, Peter Bärtsch, Kai Schommer, and Heimo Mairbäurl. Does high alveolar fluid reabsorption prevent HAPE in individuals with exaggerated pulmonary hypertension in hypoxia? High Alt Med Biol 16:283–289, 2015.—An exaggerated increase in pulmonary arterial systolic pressure (PAsP) is a highlight of high altitude pulmonary edema (HAPE). However, the incidence of HAPE at 4559 m was much lower in altitude-naïve individuals with exaggerated pulmonary vasoconstriction (HPV) in normobaric hypoxia than in known HAPE-susceptibles, indicating that elevated PAsP alone is insufficient to induce HAPE. A decreased nasal potential difference (NPD) has been found in HAPE-susceptibles, where, based on animal models, NPD serves as surrogate of alveolar epithelial ion transport. We hypothesize that those HAPE-resistant individuals with high HPV may be protected by elevated alveolar Na and fluid reabsorption, which might be detected as increased NPD. To test this hypothesis, we measured NPD in normoxia of subjects who were phenotyped in previous studies as high altitude tolerant (controls), known HAPE-susceptibles with high HPV (HP+HAPE), as well as individuals with high HPV but without HAPE (HP-no-HAPE) at 4559 m. NPD and amiloride-sensitive NPD were lower in HP+HAPE than in controls, whereas HP-no-HAPE were not different from either group. There were no differences in Cl-transport between groups. Our results show low nasal ion transport in HAPE but higher transport in those individuals with the highest HPV but without HAPE. This indicates that in some individuals with high PAsP at high altitude high alveolar fluid reabsorption might protect them from HAPE.

Introduction

Exposure to hypoxia causes pulmonary vasoconstriction (HPV) and an increase in pulmonary arterial systolic pressure (PAsP), where the pressure increase correlates directly with the degree of hypoxia (Sylvester et al., 2012) and altitude (Hultgren et al., 1964). It also causes an increase in pulmonary capillary pressure (Maggiorini et al., 2001) leading to filtration of protein into the alveolar space (Swenson et al., 2002).

Some individuals develop high altitude pulmonary edema (HAPE), a noncardiogenic edema characterized by patchy infiltrations into the lung (Bärtsch and Gibbs, 2007) and high levels of protein and red blood cells in broncho-alveolar lavage fluid (Swenson et al., 2002). These individuals respond to hypoxia with an exaggerated increase in PAsP (Bärtsch and Gibbs, 2007) and increased pulmonary capillary pressure (Maggiorini et al., 2001). Decreasing PAsP at high altitude by prophylactic intake of the Ca-antagonist nifedipine (Bärtsch et al., 1991) or the PDE5 inhibitor tadalafil (Maggiorini et al., 2006) decreases the incidence of HAPE, which indicates that exaggerated PAsP might be the major pathophysiologic mechanisms (e.g., (Dehnert et al., 2007)). Both, the high recurrence rate of 60%–70% of HAPE in susceptible individuals and their exaggerated pressure response to brief normobaric hypoxia (Bärtsch et al., 2005) point to a genetic background of HAPE susceptibility.

Interestingly, there are individuals presenting with an exaggerated increase in PAsP in hypoxia but who do not develop HAPE: Sartori and colleagues (1999) studied individuals who had experienced perinatal hypoxia and found that those show an exaggerated increase in PAsP at high altitude, which was similar to that in HAPE-susceptible individuals, but that those individuals did not develop HAPE upon ascent to an altitude of 4559 m (Sartori et al., 2000). Grünig and colleagues found that approximately 10% of a study population older than 11 years shows an abnormally high increase in PAsP when exposed to normobaric hypoxia and exercise in normoxia (Grünig et al. 2009).

Based on these findings, Dehnert and colleagues (2015) tested whether the incidence of HAPE in this subgroup of high altitude-naïve individuals without perinatal hypoxic events but with this exaggerated increase in PAsP in normobaric hypoxia was identical to that observed when exposing HAPE-susceptible individuals in an identical setting. Interestingly, they found that the incidence for HAPE is not significantly higher than that of a high altitude-naïve population with normal HPV. This indicates that besides exaggerated HPV, additional factors may be necessary to cause HAPE, and that some individuals with high PAsP in hypoxia express mechanisms protecting them from HAPE.

An efficient system of reabsorption protects the lung from accumulation of fluid in the alveolar space, which is a requirement for efficient alveolar gas exchange. Fluid reabsorption is driven by an osmotic gradient across the alveolar epithelium that is generated by vectorial Na-transport involving apical epithelial Na-channels (ENaC) and basolateral Na/K-ATPase [for reviews, see Matthay et al., (1996); Matalon and O'Brodovich, (1999)]. Total knock out of αENaC subunits in mice is lethal because it prevents the removal of alveolar fluid upon birth (Hummler et al., 1996), and partial rescue of the lethal phenotype that had a lower than normal αENaC expression resulted in decreased alveolar fluid reabsorptive capacity and an increased wet-to-dry lung weight ratio indicating edema (Egli et al., 2004).

Thus, it has been postulated that impaired alveolar fluid reabsorption might be one of the factors that can cause HAPE (Höschele and Mairbäurl, 2003), but experimental evidence is sparse. Positive results on the use of potential stimulators of alveolar Na-transport in HAPE prophylaxis [salbutamol (Sartori et al., 2002), dexamethasone (Maggiorini et al., 2006)] allow no direct conclusion that stimulated fluid clearance was in fact the protective mechanism because of possible other action of these drugs (Bärtsch and Mairbäurl, 2002).

Two independent groups showed that the potential difference across the nasal mucosa was decreased in HAPE-susceptible individuals (Sartori et al., 2002; Mairbäurl et al., 2003), even when measured at low altitude. Nasal potential difference (NPD) might serve as a surrogate of lung alveolar ion transport activity. This conclusion is based on similarities in the expression profile of ENaC subunits in airway and alveolar epithelium (Rochelle et al., 2000) and on results by Hardiman et al. (2001), who demonstrated a direct relation between alveolar fluid reabsorption and the NPD in mice genetically modified to have altered Na-transport.

If, in fact, NPD is a surrogate of alveolar epithelial reabsorptive ion transport, then the low NPD in HAPE-susceptible individuals might reflect a decreased activity of alveolar Na and fluid reabsorption even in normoxia, and those individuals with low NPD may develop alveolar edema at high altitude, because increased fluid filtration may not be compensated by adequate clearance. Similarly to the experimental animals, also in HAPE the decreased ion transport might have a genetic background, which can be detected even in normoxia.

Based on this work, we hypothesize that a high capacity of alveolar fluid reabsorption protects from HAPE even when a PAsP above physiological values in hypoxia enhances fluid filtration into the alveolar space. Thus, individuals with an exaggerated increase in PAsP but without HAPE at high altitude should not show the decreased NPD that was observed in HAPE-susceptible individuals. To test this hypothesis, we measured the NPD in normoxia in individuals with normal HPV who had never developed HAPE, as well as in individuals with exaggerated HPV and diagnosed HAPE in earlier studies, and compared them with those individuals identified by Dehnert et al. (2015) presenting with exaggerated HPV but without HAPE when exposed to high altitude.

Methods

Subjects

The nasal potential difference was measured at low altitude as a surrogate marker of the activity of ion transport in alveolar epithelium and of alveolar fluid reabsorption on three groups of individuals: The control group consisted of individuals with a “physiological” increase in PAsP at high altitude but who had never developed HAPE in at least one of our previous high altitude studies (n = 14) (Swenson et al., 2002; Maggiorini et al., 2006; Dehnert et al. 2010; 2015).

Individuals in the second group were previously altitude naïve and were found to develop high PAsP in acute, normobaric hypoxia, but who did not develop HAPE after fast ascent to 4559 m despite exaggerated PAsP (Dehnert et al., 2015) (HP-no-HAPE; only 14 of the individuals participating in the study by Dehnert et al. (2015) could be recruited).

The third group were individuals with well documented higher-than normal PAsP and HAPE at 4559 m in previous studies on the pathophysiology of HAPE (HP+HAPE; n = 14; (Swenson et al., 2002; Maggiorini et al., 2006; Dehnert et al. 2010; 2015).

Anthropometric data are shown in Table 1. Individuals were not exposed to high altitudes > 2000 m in the 2 weeks prior to nasal potential measurements, and viral infections within 2 weeks prior to NPD measurements was an exclusion criterion. NPD was measured in Heidelberg (altitude ∼110 m) after written informed consent. The study was approved by the ethics committee of the University of Heidelberg and was performed according to the Declaration of Helsinki and its current amendments.

Table 1.

Anthropometric Data and Pulmonary Arterial Systolic Pressure at Low and High Altitude

	Controls	HP-no-HAPE	HP+HAPE	P value
number, gender	14 m	3 f/11 m	1 f/13 m
Age (years)	38.5 ± 8.3	45.1 ± 9.2	46.9 ± 7.8^*	0.040
PAsP-LA (mmHg)	22.9 ± 2.6	27.9 ± 3.7^*	24.1 ± 4.8	0.006
PAsP-HA (mmHg)	34.5 ± 5.7	55.5 ± 11.0^*	53.6 ± 7.7^*	<0.001
ΔPAsP (mmHg)	11.5 ± 5.7	27.5 ± 9.7^*	29.5 ± 8.0^*	<0.001

Mean values ± SD from the indicated number of female (f) and male (m) individuals. HP-no-HAPE are 14 of the 26 individuals from the group studied by Dehnert and colleagues (2015). HP-no-HAPE have high pulmonary arterial systolic pressure (PAsP) in hypoxia but no high altitude pulmonary edema (HAPE); HP-HAPE are those with high pressure and HAPE after rapid ascent to 4559 m. Pulmonary artery pressures from the HP-no-HAPE group are the same as those reported by Dehnert (2015). PAsP from the other groups were from the respective studies for which individuals were recruited (Maggiorini et al. 2006; Dehnert et al. 2010; 2015). ΔPAsP is the increase in PAsP between low (LA) and high altitude (HA). P values are from one way ANOVA. ^* indicates significant difference to controls (p < 0.05 in post-hoc tests).

Nasal potential difference

The electrical potential difference across the nasal mucosa (nasal potential difference [NPD]) was recorded between the surface of the nasal mucosa in the inferior turbinate of the nose (measuring electrode) and a subcutaneous reference electrode placed in the right forearm. Different buffers containing inhibitors and stimulators of transport, warmed to have a temperature of approximately 32°C at the nasal mucosa, were applied to the nasal mucosa close to the measuring electrode at a rate of 3 mL per minute as described (De Boeck et al., 2011). This protocol allows discriminating different ion transport systems as shown in Figure 1.

FIG. 1.

Typical recording of the nasal potential difference (NPD; subject #29) showing the measuring conditions and how individual components of ion transport are defined. The nasal mucosa was perfused with Ringers buffer. Lines indicate where buffers were switched from Ringers buffer to Ringers buffer with amiloride (100 μM), Ringers buffer with amiloride where Cl has been replaced with gluconate (0 Cl), 0 Cl with amiloride and isoproterenol (iso; 10 μM), and 0 Cl with amiloride, isoproterenol, and ATP (100 μM). Basal NPD (NPD_bas); the portion inhibited by amiloride (NPD_Δamil); the portion of NPD stimulated by the Cl-gradient (NPD_0Cl); Cl-transport stimulated by isoproterenol; Cl-transport stimulated by ATP detecting Ca-activated Cl-channels (NPD_0Cl-ΔATP); NPD_Cl-tot is the sum of all components measured during perfusion with zero-Cl-Ringer (De Boeck et al., 2011). NPD and its components are shown as positive values.

Briefly, basal NPD (NPD_bas) was recorded during perfusion with Ringer-buffer composed of (in mM) 148 NaCl, 4.05 KCl, 2.25 CaCl₂, 2.4 K₂HPO₄, 0.4 KH₂PO₄, 1.2 MgCl₂. The amiloride-sensitive component of NPD (NPD_Δamil) representing transport by epithelial Na-channels (ENaC) was determined as the difference in NPD between NPD_bas and the NPD during perfusion with Ringer-buffer containing 100 μM amiloride.

For measuring Cl-transport, all chloride in the Ringer-buffer was replaced with gluconate except for MgCl₂, which was replaced with MgSO₄ (0 Cl Ringer) to generate a blood-to-nasal surface gradient. Basal chloride-sensitive NPD (NPD_ΔCl) was the difference in NPD measured at the end of amiloride application and the NPD during perfusion with 0 Cl Ringer in presence of 100 μM amiloride. The stimulation of NPD_ΔCl by cAMP (NPD_ΔCl,iso) was measured as the decrease in NPD during perfusion with zero-Cl Ringer in presence of amiloride and isoproterenol (10 μM), where the sum of NPD_ΔCl + NPD_ΔCl,iso is an indicator of the capacity of the cystic fibrosis transmembrane regulator [CFTR; NPD_CFTR; (De Boeck et al., 2011)]. The change in NPD upon administration of ATP resembling Ca-dependent Cl-channels was measured by perfusion with 0 Cl Ringer in presence of amiloride, isoproterenol, and ATP (100 μM; NPD_ΔCl-ATP). The sum of all Cl-driven components of NPD resembles total Cl-transport capacity, NPD_Cl-tot.

NPD was measured with Ag/AgCl electrodes (WPI, Germany) connected to the measuring site with agar/3 M KCl-filled polyethylene tubing (measuring electrode) and a hypodermic needle (reference), respectively, using a BMA-200 Bioamplifier and an analog-to-digital converter. NPD was recorded on a PC with the PowerLab software (ADinstruments, Germany). In practical terms, NPD_bas was determined during perfusion with Ringer-buffer after locating the area with the most negative potential. Perfusates were switched upon achieving a stable recording in the order Ringer + amiloride, 0Cl-Ringer + amiloride, 0Cl-Ringer + amiloride + isoproterenol, 0Cl-Ringer + amiloride + isoproterenol + ATP, and Ringer-buffer.

Pulmonary arterial systolic pressures presented have been measured at low and high altitude in the respective studies in which the phenotyping of the individuals was performed (see above, study population). Thus, the group means presented here have not been published before but individual data were part of previous publications. PAsP was determined with conventional echocardiographic equipment by measuring peak-flow velocities of tricuspid valve regurgitation jets using continuous-wave Doppler, guided by color-flow Doppler. Right atrial to right ventricular pressure gradient was calculated from a modified Bernoulli equation adding 5 mmHg for estimated right atrial pressure to obtain PAsP as described previously [e.g., (Bärtsch et al. 1991)]. Values of PAsP shown here were measured in the pre-altitude tests and on the first morning after arrival at high altitude at the Capanna Regina Margherita (4559 m); ΔPAsP denotes to the increase in PAsP at high altitude.

Data evaluation

From the continuous recordings, NPD was averaged over the 30 sec before changing to the next perfusate. Regardless of the original polarization, NPD is always shown positive. Results are mean values ± standard deviation (SD) of 14 individuals in each group. Comparisons between all three groups were performed by analysis of variance and Least Squares Difference post-hoc testing for group comparisons. A p ≤ 0.05 indicates statistically significant differences. P values shown in the text are from post-hoc testing, p values in the tables are from ANOVA. Unfortunately we were unable to study additional individuals to increase the power, because this would have required the screening of additional altitude-naïve individuals for exaggerated pulmonary hypertension in acute hypoxia [incidence approximately 10%; (Grünig et al., 2009)] and evaluation of these individual's HAPE-susceptibility by exposure to high altitude, which was beyond the financial resources of this NPD-study.

Results

Table 1 shows anthropometric data and PAsP from all three groups. Values for PAsP from controls and HP+HAPE shown are those measured in pre-altitude tests and at high altitude in the respective studies where the subjects participated and where the phenotyping was performed. Therefore the mean values in Table 1 represent unique data not previously published. PAsP at low altitude (normoxia) was slightly but significantly higher in the HP-no-HAPE group (p = 0.002) than in controls, whereas there was no statistically significant difference between HP+HAPE and controls (p = 0.191), and between HP+HAPE and HP-no-HAPE (p = 0.084).

Table 1 also shows that exposure to high altitude significantly increased PAsP in controls by ∼11.6 mmHg (p = 0.001), but by ∼27.6 and 29.5 mmHg in HP-no-HAPE and HP+HAPE, respectively (p < 0.001) and that in the HP-groups PAsP was higher at high altitude than in the controls (p < 0.001). There was no difference in PAsP at high altitude between HP-no-HAPE and HP+HAPE (p = 0.343).

Figure 2 shows that NPD_bas was ∼20 mV in controls and that it was approximately 4 mV lower in individuals with HP+HAPE (p = 0.022) in normoxia confirming earlier reports (Sartori et al., 2002; Mairbäurl et al., 2003). In contrast to the earlier studies, we found here a lower NPD_Δamil in HP+HAPE than in controls (p = 0.046), indicating that HAPE-susceptible individuals might have decreased ENaC activity. Figure 2 also shows that there was no difference in NPD_bas and NPD_Δamil between controls and HP-no-HAPE (p = 0.228 and p = 0.274, respectively) and between HP-no-HAPE and HP+HAPE (p = 0.104 and p = 0.151, respectively). Table 2 shows that there was no significant difference among groups in NPD_amil-IS and in Cl-transport and its components.

FIG. 2.

Basal nasal potential difference (NPD_bas) and portion inhibited by amiloride (NPD_Δamil) in HAPE-resistant individuals (co), in individuals found to have increased pulmonary arterial systolic pressure (PAsP) in hypoxia but who did not develop HAPE after ascent to 4559 m (HP-no-HAPE; (Dehnert et al., 2015)), and individuals with high PAsP in hypoxia and known HAPE susceptibility (HP+HAPE). NPD_bas is the NPD measured during perfusion with Ringer buffer, NPD_Δamil is the portion of NPD inhibited by amiloride. Mean values ± SD from 14 individuals in each group. * indicates p < 0.05 between controls and HP+HAPE.

Table 2.

Components of NDP from Controls, HP-no-HAPE, and HP+HAPE

	Controls	HP-no-HAPE	HP+HAPE	P value
NPD_amil-IS	13.6 ± 3.9	12.5 ± 2.9	10.9 ± 2.2	0.161
NPD_Δ0Cl	13.3 ± 7.5	17.3 ± 8.7	13.0 ± 8.4	0.314
NPD_CFTR	23.3 ± 11.9	29.4 ± 10.3	26.1 ± 12.0	0.387
NPD_0Cl-ΔATP	7.9 ± 4.6	6.8 ± 4.9	5.2 ± 4.4	0.484
NPD_Cl-tot	31.2 ± 12.6	37.3 ± 11.0	32.8 ± 13.5	0.416

NPD was measured according to the suggestions for the diagnosis of cystic fibrosis (De Boeck et al., 2011). NPD_amil-IS is the NPD during perfusion with amiloride, NPD_Δ0Cl is the change in NPD during perfusion with Cl-free Ringer-buffer, NPD_CFTR is the change in NPD by zero Cl plus isoproterenol, which is a measure of the activity of the cystic fibrosis transmembrane regulator (CFTR), NPD_0Cl-ΔATP is the change in NPD by adding ATP to perfusion with zero-Cl Ringer indicating Ca-activated Cl-channels, and NPD_Cl-tot is the sum of all Cl-sensitive components of NPD (De Boeck et al., 2011). Mean values ± SD (in mV), n = 14 in each group. P values are from one-way ANOVA.

It is also possible that there is a relation between ΔPAsP at HA and ion transport activity (i.e., NPD), which is not necessarily associated with the occurrence of HAPE, in which case there should be a correlation between both parameters. Figure 3 shows that, when combining data from all groups, there was no correlation between ΔPAsP and NPD and its components (p values in figure legends) and that in the groups with exaggerated HPV ΔPAsP is always higher than in controls, independent of NPD, and independent of the occurrence of HAPE, although it appears that correlations in controls are different from those individuals with exaggerated HPV. Correlations using absolute values of PAsP at high altitude instead of ΔPAsP provide a similar picture (not shown).

FIG. 3.

Relation between the increase in PAsP in hypoxia and the NPD and its amiloride- and Cl-sensitive components measured in normoxia. NPD_bas is the NPD measured during perfusion with Ringer buffer, NPD_Δamil is the portion of NPD inhibited by amiloride, and NPD_Cl-tot is the Cl-dependent potential change measured during perfusion with zero-Cl and in presence of amiloride, isoproterenol, and ATP. PAsP values were not available for all individuals. Correlation coefficients and p values are indicated in the inserts. Solid lines are drawn where a statistically significant correlation was found as indicated by the regression coefficients and p values in the figure inserts. P values for regressions when combining data from all groups are: panel (A) 0.206, (B) 0.956, (C) 0.315.

Discussion

In this study we confirm that HAPE-susceptible individuals have a decreased NPD_bas in normoxia (Sartori et al., 2002; Mairbäurl et al., 2003), indicating insufficient removal of fluid filtered into the alveolar space, assuming that NPD is a valid surrogate of alveolar transport processes. However, we found no statistically significant elevation of NPD in individuals with HP-no-HAPE but also no difference to controls.

Exaggerated HPV was long considered the main cause of HAPE [for review, see (Dehnert et al., 2007)]. However, two independent publications report that altitude naïve individuals with exaggerated HPV in normobaric hypoxia and at high altitude have no increased risk to suffer from HAPE compared to a non-selected population (Sartori et al., 1999; Dehnert et al., 2015), indicating that additional factors are required to explain HAPE-susceptibility, or that there are mechanisms protecting those with high HPV from developing HAPE. Decreased lung diffusion capacity (Steinacker et al., 1998), low hypoxic ventilatory response (Hohenhaus et al., 1995), inflammatory events before ascent to high altitude and subsequent leakiness in hypoxia in animal models (Carpenter et al., 1998), ineffective lymphatic drainage, and insufficient fluid reabsorption (Sartori et al., 2002; Höschele and Mairbäurl, 2003) of fluid filtered into the alveolar space by hydrostatic forces (Maggiorini et al., 2001) may all cause pulmonary edema in hypoxia. Here we speculated that a high activity of alveolar fluid reabsorption might protect those individuals with abnormally high PAsP in hypoxia from HAPE (Höschele and Mairbäurl, 2003).

The clearance of fluid from the alveolar space is driven by an osmotic gradient generated by Na-entry via epithelial Na-channels (ENaC) in the apical membrane of alveolar epithelial cells and basolateral Na/K-ATPase, and Cl-transport mediated by various pathways in order to maintain electro-neutrality (for review, see Matalon and O'Brodovich, 1999; Matthay et al., 2002; Mutlu and Sznajder, 2005). Whereas alveolar Na and fluid reabsorption can be determined quite easily in animal models and in cultured cells, there is no noninvasive and simple technique to measure these parameters in humans in vivo. Thus, one has to rely on indirect measures at sites that are more easily accessible.

One such measure is the electrical potential difference across the nasal mucosa. This method has been established to evaluate defects of Na- and Cl-dependent ion transport in cystic fibrosis (Knowles et al., 1981; Middleton et al., 1994). Several results indicate that NPD qualifies as a surrogate of alveolar Na reabsorption, all of which are derived from animal experiments but not from measurements on humans. There are similarities in the expression of Na-transporters in airway and alveolar epithelium (Rochelle et al., 2000). A direct correlation between NPD and alveolar fluid clearance has been found in iNOS knockout mice, because NO regulates the activity of Na-transport (Hardiman et al., 2001). A correlation between NPD and alveolar clearance was also found in partial rescues of αENaC knockout mice (Egli et al., 2004). Together, these studies imply that modulation of Na-transport by directly modifying ENaC abundance or one of the regulators of ENaC activity ubiquitously in the entire organism can also be detected at the nasal mucosa.

Thus, if a generalized defect in ion transport that also impairs alveolar fluid clearance contributes to the high recurrence rate of HAPE in susceptible individuals (Bärtsch et al., 2003), it should be detectible also at the nasal mucosa, even in normoxia. In fact, a decreased nasal ion transport activity has been found in HAPE-susceptible individuals prior to ascent to high altitude (Sartori et al., 2002; Mairbäurl et al., 2003). Here we confirm these findings on a different study population. It is not known yet, whether this difference in reabsorptive activity in fact reflects a genetic variation in one or several ion transporters or their regulators such as in pseudohypoaldosteronism-1, where a defective epithelial Na-channel causes lung fluid accumulation (Schaedel et al., 1999). However, no symptoms of this disease such as hyponatremia, hyperkalemia, a high Cl concentration in sweat, and elevated aldosterone (Schaedel et al., 1999) have been described in HAPE. It is unclear, whether a more subtle expression of this disease without a renal phenotype exists in HAPE-susceptible individuals that might decrease alveolar Na and fluid reabsorption.

It is not clear which ion transport pathway might be impaired in HAPE-susceptible individuals. Here we show a decreased NPD_Δamil in HAPE-susceptible individuals, which confirms a result described earlier (Sartori et al., 2002), but which we had not seen in a previous study (Mairbäurl et al., 2003). However, the difference in NPD_Δamil between controls and HAPE-susceptible individuals is much smaller than the one in NPD_bas, indicating that other pathways may be affected too. Cl transport was not different. We had previously shown that NPD_Δamil tended to be decreased at high altitude (Mairbäurl et al., 2003), which might indicate inhibition of alveolar Na and fluid reabsorption by hypoxia, consistent with results from animal experiments (Vivona et al., 2001; Guney et al., 2007). Hypoxic inhibition in addition to an already impaired reabsorption in normoxia might further impair alveolar clearance and therefore enhance the risk to develop HAPE (Höschele and Mairbäurl, 2003).

Taken together, on the basis that NPD and its components actually reflects ion transport activity at the alveolar epithelium, these findings support the notion that HAPE-susceptible individuals have impaired alveolar ion transport that might contribute to formation of alveolar edema upon exposure to hypoxia and thus to HAPE susceptibility.

We had speculated that individuals with an exaggerated increase in PAsP in acute hypoxia and at high altitude, but who did not develop HAPE at 4559 m, might be protected by having a higher alveolar fluid reabsorption than HAPE-susceptible individuals. However, our results seem not to support this hypothesis, as we found no statistically significant difference in NPD_bas, NPD_Δamil, and any of the other components that determine NPD between HP-no-HAPE and the HAPE susceptible individuals.

However, NPD was also not significantly different from controls, indicating nearly normal ion transport activity. Thus, our results allow no direct conclusion on an actually improved alveolar fluid reabsorption in HP-no-HAPE. Unfortunately, the number of individuals in the study may have been too small to obtain statistically significant differences, which is certainly a limitation of this study. Increasing the number would have required additional screening for exaggerated HPV in normobaric hypoxia and tests for HAPE-susceptibility in a controlled high altitude setting, which was beyond the scope of this study.

Although, as outlined above, there appears to be a direct relation between alveolar fluid reabsorption and NPD in animal models, it is not clear from these experiments, how high ion transport activity actually has to be to prevent fluid accumulation in the alveolar space, and how this would translate into NPD. It is therefore possible that some individuals with a high NPD may be protected from HAPE when filtration into the alveolar space can be assumed to be high with increased PAsP.

Another possible constraint of this study is that ion transport has only been measured in normoxia, which was based on the assumption of an ubiquitous defect in ion transport related to HAPE susceptibility. Another possibility is, however, that the activity of regulators of Na reabsorption changes differently among our study groups upon exposure to high altitude. If this were the case, then measurements of NPD at high altitude might have had a higher discriminative strength. Due to the retrospective design of our study these values are not available.

Alveolar fluid reabsorption determines the volume of alveolar lining fluid (Bove et al., 2010) and thus the thickness of the diffusion barrier for respiratory gasses. Therefore, impaired fluid clearance increases the volume of alveolar lining fluid, impairs oxygen diffusion across the alveolar barrier, and causes hypoxemia, which might subsequently aggravate hypoxic pulmonary vasoconstriction. In that case, ion transport activity and PAsP might correlate in a way that a low increase in PAsP in hypoxia might be associated with high NPD and vice versa. Thus, controls and HAPE-susceptible individuals would follow the same correlation line, where data from HP+HAPE would be located on the low-NPD/high-ΔPAsP side and controls at the other end of the line, whereas HP-no-HAPE would not follow this correlation because of the same increase in PAsP but slightly higher NPD.

In contrast to this hypothesis, the results shown in Figure 3 indicate no such relation. Over the wide range of NPDs, ΔPAsP of most individuals with an exaggerated HPV, regardless of whether HAPE-susceptible or not, is higher than that of controls. We had found earlier that there was no correlation between NPD and arterial PO₂ and arterial SO₂ (Mairbäurl et al., 2003), which supports this notion. It appears therefore that the capacity of ion transport in normoxia does not affect PAsP in hypoxia by modulating alveolar fluid volume.

In summary, our results indicate group differences in NPD between controls and HAPE-susceptible individuals that might have a genetic background. However, data on genetic analyses of ion transporters are lacking. In contrast to our hypothesis, significantly higher nasal epithelial Na- and Cl-transport was not found in those individuals with a high HPV but who are HAPE-resistant (HP-no-HAPE). The fact that NPDs in this group were also not different from controls might be interpreted as a weak indication that some individuals with high PAsP at high altitude but without HAPE may have a sufficiently high alveolar fluid reabsorptive capacity that might protect them from HAPE.

Footnotes

Acknowledgments

We thank Mrs. Sonja Engelhardt and Mrs. Christiane Herth for excellent technical assistance. We also thank Dr. Erik Swenson, Seattle, for fruitful data discussions. This work was supported by a grant from the BMBF to the German Center for Lung Research (DZL, TLRC-H) and the Mukoviszidose e.V.

Author Disclosure Statement

No conflicting financial interests exist.

References

Bärtsch

, and Gibbs

JSR

. (2007). Effect of altitude on the heart and the lungs. Circulation, 116:2191–2202.

Bärtsch

, Maggiorini

, Ritter

, Noti

, Vock

, and Oelz

. (1991). Prevention of high-altitude pulmonary edema by nifedipine. N Engl J Med, 325:1284–1289.

Bärtsch

, and Mairbäurl

. (2002). Salmeterol for the prevention of high altitude pulmonary edema. N Engl J Med, 347:1283.

Bärtsch

, Mairbäurl

, Maggiorini

, and Swenson

. (2005). Physiological aspects of high-altitude pulmonary edema. J Appl Physiol, 98:1101–1110.

Bärtsch

, Mairbäurl

, Swenson

, and Maggiorini

. (2003). High altitude pulmonary oedema. Swiss Med Weekly, 133:377–384.

Bove

, Grubb

, Okada

, Ribeiro

, Rogers

, Randell

, O'Neal

, and Boucher

. (2010). Human alveolar type II cells secrete and absorb liquid in response to local nucleotide signaling. J Biol Chem, 285:34939–34949.

Carpenter

, Reeves

, and Durmowicz

. (1998). Viral respiratory infection increases susceptibility of young rats to hypoxia-induced pulmonary edema. J Appl Physiol, 84:1048–1054.

De Boeck

, Derichs

, Fajac

, de Jonge

, Bronsveld

, Sermet

, Vermeulen

, Sheppard

, Cuppens

, Hug

, Melotti

, Middleton

, and Wilschanski

. (2011). New clinical diagnostic procedures for cystic fibrosis in Europe. J Cyst Fibros, 10:S53–S66.

Dehnert

, Berger

, Mairbäurl

, and Bärtsch

. (2007). High altitude pulmonary edema: A pressure-induced leak. Respir Physiol Neurobiol, 158:266–273.

10.

Dehnert

, Luks

, Schendler

, Menold

, Berger

, Mairbäurl

, Faoro

, Bailey

, Castell

, Hahn

, Vock

, Swenson

, and Bartsch

. (2010). No evidence for interstitial lung oedema by extensive pulmonary function testing at 4559 m. Eur Respir J, 35:812–820.

11.

Dehnert

, Mereles

, Greiner

, Albers

, Scheurlen

, Zügel

, Bohm

, Vock

, Maggiorini

, Grünig

, and Bartsch

. (2015). Exaggerated hypoxic pulmonary vasoconstriction without susceptibility to high altitude pulmonary edema. High Alt Med Biol, 16:11–17.

12.

Egli

, Duplain

, Lepori

, Cook

, Nicod

, Hummler

, Sartori

, and Scherrer

. (2004). Defective respiratory amiloride sensitive sodium transport predisposes to pulmonary oedema and delays its resolution in mice. J Physiol, 560:857–865.

13.

Grünig

, Weissmann

, Ehlken

, Fijalkowska

, Fischer

, Fourme

, Galie

, Ghofrani

, Harrison

, Huez

, Humbert

, Janssen

, Kober

, Koehler

, Machado

, Mereles

, Naeije

, Olschewski

, Provencher

, Reichenberger

, Retailleau

, Rocchi

, Simonneau

, Torbicki

, Trembath

, and Seeger

. (2009). Stress Doppler echocardiography in relatives of patients with idiopathic and familial pulmonary arterial hypertension: Results of a multicenter European analysis of pulmonary artery pressure response to exercise and hypoxia. Circulation, 119:1747–1757.

14.

Guney

, Schuler

, Ott

, Höschele

, Baloglu

, Bärtsch

, and Mairbäurl

. (2007). Dexamethasone prevents transport inhibition by hypoxia in rat lung and alveolar epithelial cells by stimulating activity and expression of Na/K-ATPase and ENaC. Am J Physiol, 293:L1332–L1338.

15.

Hardiman

, Lindsey

, and Matalon

. (2001). Lack of amiloride-sensitive transport across alveolar and respiratory epithelium of iNOS(-/-) mice in vivo. Am J Physiol, 281:L722–L731.

16.

Hohenhaus

, Paul

, McCullough

, Kucherer

, and Bärtsch

. (1995). Ventilatory and pulmonary vascular response to hypoxia and susceptibility to high altitude pulmonary oedema. Eur Respir J, 8:1825–1833.

17.

Höschele

, and Mairbäurl

. (2003). Alveolar flooding at high altitude: Failure of reabsorption?. News Physiol Sci, 18:55–59.

18.

Hultgren

, Lopez

, Lundberg

, and Miller

. (1964). Physiologic studies of pulmonary edema at high altitude. Circulation, 29:393–408.

19.

Hummler

, Barker

, Gatzy

, Beermann

, Verdumo

, Boucher

, and Rossier

. (1996). Early death due to defective neonatal lung liquid clearance in à-ENaC-deficient mice. Nature Genetics, 13:325–328.

20.

Knowles

, Carson

, Collier

, Gatzy

, and Boucher

. (1981). Measurements of nasal transepithelial electric potential differences in normal human subjects in vivo. Am Rev Respir Dis, 124:484–490.

21.

Maggiorini

, Brunner-La Rocca

, Peth

, Fischler

, Böhm

, Bernheim

, Kiencke

, Bloch

, Dehnert

, Naeije

, Lehmann

, Bärtsch

, and Mairbäurl

. (2006). Both tadalafil and dexamethasone may reduce the incidence of high-altitude pulmonary edema. Ann Intern Med, 145:497–506.

22.

Maggiorini

, Melot

, Pierre

, Pfeiffer

, Greve

, Sartori

, Lepori

, Hauser

, Scherrer

, and Naeije

. (2001). High-altitude pulmonary edema is initially caused by an increase in capillary pressure. Circulation, 103:2078–2083.

23.

Mairbäurl

, Weymann

, Möhrlein

, Swenson

, Maggiorini

, Gibbs

JSR

, and Bärtsch

. (2003). Nasal epithelium potential difference at high altitude (4559 m): Evidence for secretion. Am J Respir Crit Care Med, 167:862–867.

24.

Matalon

, and O'Brodovich

. (1999). Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance. Ann Rev Physiol, 61:627–661.

25.

Matthay

, Folkesson

, and Clerici

. (2002). Lung epithelial fluid transport and the resolution of pulmonary edema. Physiol Rev, 82:569–600.

26.

Matthay

, Folkesson

, and Verkman

. (1996). Salt and water transport across alveolar and distal airway epithelia in the adult lung. Am J Physiol, 270:L487–L503.

27.

Middleton

, Geddes

, and Alton

EWFW

. (1994). Protocols for in vivo measurements of the ion transport defects in cystic fibrosis nasal epithelium. Eur Respir J, 7:2050–2056.

28.

Mutlu

, and Sznajder

. (2005). Mechanisms of pulmonary edema clearance. Am J Physiol, 289:L685–L695.

29.

Rochelle

, Li

, Ye

, Lee

, Talbot

, and Boucher

. (2000). Distribution of ion transport mRNAs throughout murine nose and lung. Am J Physiol, 279:L14–L24.

30.

Sartori

, Allemann

, Duplain

, Lepori

, Egli

, Lipp

, Hutter

, Turini

, Hugli

, Cook

, Nicod

, and Scherrer

. (2002). Salmeterol for the prevention of high-altitude pulmonary edema. N Engl J Med, 346:1631–1636.

31.

Sartori

, Allemann

, Trueb

, Delabays

, Nicod

, and Scherrer

. (1999). Augmented vasoreactivity in adult life associated with perinatal vascular insult. Lancet, 353:2205–2207.

32.

Sartori

, Allemann

, Trueb

, Lepori

, Maggiorini

, Nicod

, and Scherrer

. (2000). Exaggerated pulmonary hypertension is not sufficient to trigger high-altitude pulmonary oedema in humans. Schweiz Med Wochenschr, 130:385–389.

33.

Schaedel

, Marthinsen

, Kristoffersson

, Kornfalt

, Nilsson

, Orlenius

, and Holmberg

. (1999). Lung symptoms in pseudohypoaldosteronism type 1 are associated with deficiency of the alpha-subunit of the epithelial sodium channel. J Pediat, 135:739–745.

34.

Steinacker

, Tobias

, Menold

, Reissnecker

, Hohenhaus

, Liu

, Lehmann

, Bärtsch

, and Swenson

. (1998). Lung diffusing capacity and exercise in subjects with previous high altitude pulmonary oedema. Eur Respir J, 11:643–650.

35.

Swenson

, Maggiorini

, Mongovin

, Gibbs

JSR

, Greve

, Mairbäurl

, and Bärtsch

. (2002). Pathogenesis of high-altitude pulmonary edema: Inflammation is not an etiologic factor. J Am Med A, 287:2228–2235.

36.

Sylvester

, Shimoda

, Aaronson

, and Ward

. (2012). Hypoxic pulmonary vasoconstriction. Physiol Rev. 92:367–520.

37.

Vivona

, Matthay

, Chabaud

, Friedlander

, and Clerici

. (2001). Hypoxia reduces alveolar epithelial sodium and fluid transport in rats: Reversal by beta-adrenergic agonist treatment. Am J Respir Cell Mol Biol, 25:554–561.