The effect of inhaled dry salt on pulmonary mucociliary clearance in obstructive lung disease: A randomised, placebo‐controlled, crossover study

Abstract

Background

Inhaled saline may improve mucus transport in respiratory diseases, for example, cystic fibrosis. This randomised, placebo‐controlled, crossover study examined the effect of inhaled dry NaCl on mucociliary clearance in patients with chronic obstructive pulmonary disease (COPD).

Methods

Twenty‐five patients with GOLD stage I‐III COPD (72% with excess mucus) were tested on two separate days after inhalation from a dry powder inhaler containing either 40 mg dry NaCl or placebo (empty). Pulmonary mucociliary clearance was assessed using inhalation of a radioactive ^99mTc‐labelled nanocolloid tracer and gamma camera imaging. Clearance was measured for 2 h post‐intervention. Co‐primary outcomes were clearance after 1 and 2 h.

Results

The study was terminated early due to COVID‐19 after enrolling 25 of 35 planned patients. No significant differences were observed in mucociliary clearance between NaCl and placebo after 1 h (11.3% ± 9.0% vs. 11.4% ± 7.0%, p = 0.97) or 2 h (15.3% ± 9.8% vs.16.1% ± 8.7%, p = 0.55). Normalised clearance based on initial radioaerosol deposition also showed no difference.

Conclusions

Inhalation of 40 mg dry NaCl was well tolerated without acute adverse effects in COPD patients. However, early termination left the study underpowered, increasing the risk of type II error. Non‐significant findings should not be interpreted as evidence of no effect. Larger, adequately powered trials are needed to clarify the impact of dry NaCl inhalation on mucociliary clearance in COPD.

1. INTRODUCTION

Salt inhalation therapy has been recognised for thousands of years as a treatment for respiratory diseases (Horowitz, 2010). The modern practice of salt therapy can be traced back to the use of salt mines and caves in Europe and other locations during the 19th century (Wasik & Tuuminen, 2021). Mechanistically, salt inhalation therapy is thought to enhance mucociliary function in the airways through a direct stimulatory effect of transient hyperosmolarity of the surface liquid on the airway cilia, thereby improving mucociliary clearance (MCC), i.e., refers to the ciliary transport of pathogens and inhaled particles within the mucus layer, facilitating their removal from the airways (Bustamante‐Marin & Ostrowski, 2017; Munkholm & Mortensen, 2014).

Previous studies have reported that salt inhalation therapy has beneficial effects on MCC in both healthy individuals and patients with asthma and cystic fibrosis (Daviskas et al., 1996; Robinson et al., 1996). Furthermore, while halotherapy—a treatment involving the inhalation of small salt particles in a controlled environment designed to replicate the natural microclimate of a salt cave—has been reported to provide symptom relief in patients with COPD, and its clinical effects often stated to be well‐documented in this context, the underlying evidence is limited (Rashleigh et al., 2014), and the acute effect of salt inhalation on MCC has not previously been investigated in this patient group.

The present study investigated the effect of dry powder sodium chloride inhalation on mucociliary function using a BREATHOX® device, a commercially available dry powder sodium chloride particle inhaler, in patients with COPD, chronic bronchitis, or both, in a randomised, placebo‐controlled cross‐over design. We hypothesised that dry powder sodium chloride inhalation would acutely improve scintigraphically measured MCC and reduce airway symptoms.

2. METHODS

2.1. Ethical approval

The study was approved by The Research Ethics Committee of the Capital Region of Denmark (H‐18017123). The study was conducted in accordance with the ethical standards described by the Declaration of Helsinki. All participants provided oral and written informed consent prior to participation.

2.2. Study design

The present study was conducted as a randomised, placebo‐controlled, crossover study. Participants acted as their own control on two different days with minimum 2 days and maximum 14 days between the investigation days, randomised to sequence with BREATHOX® dry powder sodium chloride on the first day/placebo on the second day or placebo on the first day/dry powder sodium chloride on the second day.

2.3. Setting

The study took place at Department of Physiology and Nuclear Medicine at Rigshospitalet in Copenhagen, Denmark, between 1st of September 2018 and 31st of December 2019.

2.4. Participants

The inclusion criteria were patients with COPD grade I‐III according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria (GOLD, 2024). The exclusion criteria were persons who had used salt inhalations within a month before the first examination, pregnant or breastfeeding women (fertile women must present a negative pregnancy test), persons who had had a radioactive examination within the last month, persons currently receiving antibiotic treatment, persons with symptoms of an acute airway infection within the last 3 weeks, persons with known airway hyperreactivity and/or known asthma and lastly persons under the age of 50 years old. The participants were permitted to use their regular medication and continued their treatment throughout the study period. They were instructed to take their inhaled medication at the same time on both test days. Twenty‐one of the 25 participants used inhaled β₂‐agonists (short‐acting and/or long‐acting), 17 of these received also a long‐acting muscarinic antagonist, and four participants did not use any respiratory medication.

Participants were recruited via advertising at outpatient lung clinics, pharmacies, using an online recruitment platform (forsoegsperson. dk) and on the sponsors’ website.

2.5. Randomisation and blinding

The study was double blinded, however in reality it was not possible to uphold the double‐blinding because it was not possible to disclose whether the BREATHOX® inhaler was with or without sodium chloride from the participants, because the taste of sodium chloride is immediately evident upon inhalation.

2.6. Intervention

The intervention was 10 times oral inhalation of 4 mg of dry powder sodium chloride, hence 40 mg in total, consisting of 95% NaCl and 5% lactose. BREATHOX® is a CE‐marked inhaler specifically designed to deliver such micronised, dry powder sodium chloride particles. The microparticle size varies from 1 to 10 μm, allowing general airway application. For the placebo‐group an inhaler without dry powder sodium chloride was used. If necessary, the participants could practice their inhalation technique with an empty inhaler supervised by experienced personnel.

2.7. Outcomes

The co‐primary outcome was between‐condition difference in lung retention of the radioactive tracer after 60 and 120 min. The key secondary outcome was airway symptoms.

2.8. Measurements

Lung function testing. At the beginning of each study day, lung function testing by dynamic spirometry was performed in accordance with the ERS/ATS guidelines (Culver et al., 2017; Graham et al., 2019) at a dedicated lung function facility using a MasterScreen (Vyaire Medical). Before measurements, standing height (to nearest 1 mm), weight (to the nearest 100 g) of the participants were obtained. The variables reported are forced expiratory volume in the first second (FEV₁) and forced expiratory volume (FVC). All manoeuvres were evaluated by established acceptability and repeatability criteria. These were reported as raw values, as % of predicted using ERS reference values for normalization for height, sex, and age (Quanjer et al., 1993).

Mucociliary clearance. MCC was measured scintigraphically by assessing the lung clearance of ^99mTc‐labelled albumin nanocolloid tracer (Venticoll®), as described in detail elsewhere (Marthin et al., 2023; Mortensen et al., 1994). After consuming 250 mL of water, participants underwent a 5‐min static baseline lung scintigraphy. Two radioactive markers were placed on their back at the C6 and Th12 vertebral levels to aid in the subtraction of background radiation. Participants inhaled approximately 10 MBq of Venticoll® (<0.1 mSv) from an ultrasound liquid nebuliser, followed by rinsing their mouth with water by gargling three times. The inhalation technique—consisting of slow inhalations and quick expirations to deposit particles in the central ciliated airways—was explained in person by hospital personnel experienced in the method. A second 5‐min static lung scintigraphy was then performed at time 0 min. For the intervention, participants performed 10 consecutive inhalations of either dry powder sodium chloride or a placebo, administered using the BREATHOX® inhaler, both provided by hospital personnel. This was followed by five lung scintigrams taken at the following time intervals after the intervention: 10–30 min (dynamic), 30–35 min (static), 35–55 min (dynamic), 60–65 min (static), and 120–125 min (static). At 65 min post‐intervention, participants underwent a 1‐min lung scintigraphy using Krypton‐81m inhalation (<0.1 mSv).

Regional ventilation distribution: The regional ventilation distribution was visualised by a ^81mKr‐gas scintigram. The initial ^99mTc aerosol distribution on the initial O min image was compared to the ^81mKr distribution in the peripheral and central lung regions by calculation of the penetration index (PI) to indicate how far the radioaerosol had penetrated the airways and lungs, and hence had to move before being cleared from the airways and lung regions (Mortensen et al., 1994):

$$PI=(Ominperipheral/centralregion)/({}^{81m}Krperipheral/centralregion)$$

Lung retention after 60 min (LR60) was calculated as whole lung retention on the 60 min image in percentage of the initial aerosol deposition (0 min image), after correction for background and physical decay. It can also be expressed as the 60 min clearance ( = 100%‐LR60%). Similarly, LR30 and LR120 retention was calculated. The predicted LR60 and LR120 was calculated based on a regression model based on age, sex and PI using established reference equations (Mortensen et al., 1994). The z‐score for LR60 (and LR120) was calculated by subtracting the predicted LR60 (LR120) from the measured LR60 (LR120), and this difference was then divided by the SD for LR60 (and LR120 SD). In a similar way, calculations were performed on the right lung only (excluding the left lung) to diminish the possible influence by radioactivity in the stomach which lays near the left lung.

The dynamic images were used to calculate the tracheobronchial bolus velocity (TBV) of single mucus boluses being transported up the trachea and main bronchi (normal rate: 2–6 mm/min) (Marthin et al., 2023).

Airway symptoms: On each study day, at 35 and 125 min after receiving the inhalant (BREATHOX® or placebo), each participant responded to the following questions: 1) Do you feel any changes in your lung condition? (options: Yes/No); 2) Do you breathe more easily or heavily after receiving the inhalant? (options: Easier/Heavier); 3) On a scale of −5 to +5, how much overall change in your airway condition have you experienced? (−5 = major change for the worse; 0 = no change; +5 = major change for the better); 4) On a scale of −5 to +5, was coughing today more difficult or easier than normal? (−5 = much more difficult; 0 = no change; +5 = much easier). In addition to these subjective assessments, the number of coughs occurring within the 125 min after receiving the inhalant was recorded by the operator conducting the MCC.

Questionnaires. On each study day, prior to any measurement, daily symptom burden and health‐related quality of life was assessed with the COPD Assessment Test (CAT) (Jones et al., 2009) and The Clinical COPD Questionnaire (CCQ) (van der Molen et al., 2003).

2.9. Safety and adverse events

Assessments for adverse events (AEs) and serious adverse events (SAEs) were conducted during and between study days, as well as for 2 days following the last study day. AEs were defined as any condition leading to hospitalisation for 24 h or less. SAEs were defined as any condition necessitating hospital admission for more than 24 h, as well as events that were life‐threatening, resulted in death, or caused persistent or significant disability/incapacity. Any SAE was reported to the Regional Ethics Committee of the Capital Region of Denmark within 7 days.

AEs were specifically defined as persistent coughing, bronchial constriction, or hoarseness. Coughing was considered a potential and acceptable effect of particle inhalation unless it persisted for more than 3 min or failed to subside after the participant drank or gargled three mouthfuls of water. Coughing lasting less than 3 min was not recorded as an AE, and drinking or gargling water was not considered an intervention. However, persistent coughing was reported as an SAE.

Bronchial constriction has been documented in individuals with hyperreactive airways following the inhalation of hypertonic saline. The quantities of sodium chloride inhaled in this study were lower than those reported to induce bronchial constriction. Should bronchial hyperresponsiveness occur, it was classified as an AE, and if it required intervention (e.g., beta₂ agonist inhalation), it was classified as an SAE.

Hoarseness was considered an SAE if it was persistent and/or required intervention. A salty taste was anticipated in all participants following dry salt and could occasionally increase salivary secretion; these were not considered AEs and were not registered as such.

2.10. Sample size considerations

To detect a 25% between‐condition difference in lung retention at 60 min, favouring lower lung retention after dry powder sodium chloride inhalation, a sample size of 35 participants would have been required to achieve 80% statistical power [(1–β) × 100%] with a two‐sided significance level of 0.05.

2.11. Statistical analysis

Statistical analyses were performed using Microsoft Excel [Data Analysis Toolpak] (Microsoft Corporation, 2021). Data are reported as mean (SD). Assumptions of normality were evaluated by visual inspection of histograms, and pairwise comparisons between time points were conducted using two‐tailed paired t‐tests, and adjustments for multiplicity were made where appropriate. A significance level of 0.05 was applied.

3. RESULTS

3.1. Participant characteristics

Because of the COVID‐19 pandemic, the study had to be terminated after the inclusion of n = 25. Participant characteristics are presented in Table 1. The participants had a diagnosis of chronic obstructive pulmonary disease (COPD, n = 9), chronic bronchitis (n = 5), or both conditions (n = 11). Among the 20 participants diagnosed with COPD, 4 (20%) were classified as GOLD III, 14 (70%) as GOLD II, and 2 (10%) as GOLD I. Two participants reported never having smoked, and two were current smokers. The remaining 21 participants had abstained from smoking for an average of 12.2 years (SD = 8.8 years). Seven participants (28%) reported no daily expectoration, while the remaining 18 (72%) reported daily expectoration, ranging from 0.5 to 10 teaspoons per day (mean 2.69, (2.52)).

Table 1.

Participant characteristics.

	n (%)
	25 (100)
Age (years)	70 (6)
Sex, female/male	14/11 (56/44)
Diagnosis
Chronic bronchitis	5 (20)
COPD	9 (36)
Both	11 (44)
COPD classification
GOLD‐stage I	2 (10)
GOLD‐stage II	14 (70)
GOLD‐stage III	4 (20)
Smoking status
Never smoked	2 (8)
Previous smoker	21 (84)
Current smoker	2 (8)
Pack years (for previous and current smokers)	31.3 (10.3)
Expectoration
No	7 (28)
Yes	18 (72)
Amount (teaspoonful)	2.7 (2.5)
CAT‐score	16.4 (7.0)

Note: Data are represented as n (%) and mean (SD).

Abbreviations: CAT‐score, chronic obstructive pulmonary disease assessment test; COPD, chronic obstructive pulmonary disease; GOLD, Global Initiative for Chronic Obstructive Lung Disease.

3.2. Mucociliary clearance

Prior to the study intervention, there were no significant differences between study days in FEV₁ (1.63 (0.72) vs. 1.82 (0.78) litres; p = 0.76), FVC (2.68 (0.75) vs. 2.77 (0.77) litres; p = 0.65), PI (0.46 (0.18) vs. 0.50 (0.21); p = 0.07), or CCQ scores (17.9 (9.3) vs. 18.2 (7.8); p = 0.76). Results from the MCC analysis following inhaled dry powder sodium chloride versus placebo are provided for the right lung in Table 2, and for both lungs in Table 3. Apart from a marginally lower PI on the days immediately before dry powder sodium chloride was inhaled, no differences in lung retention between conditions were observed at any time point.

Table 2.

Mucociliary clearance in the right lung.

	Dry NaCL inhalation		Placebo		P‐value
Lung retention (%)
30 min	91.38 (9.82)	[68.00–107.05]	90.94 (6.42)	[77.70–101.00]	0.82
60 min	89.15 (9.62)	[69.75–104.75]	88.64 (7.02)	[76.00–99.30]	0.75
120 min	84.81 (9.97)	[63.75–102.20]	83.95 (8.73)	[70.30–97.15]	0.50
Clearance (%)
30 min	9.27 (9.26)	[0–32.00]	9.11 (6.48)	[0–22.30]	0.94
60 min	11.30 (8.98)	[0–30.25]	11.36 (7.02)	[0.70–24.00]	0.97
120 min	15.28 (9.8)	[0–36.25]	16.05 (8.73)	[2.85–29.70]	0.55
Coughing (number of coughs)	16.10 (36.5)	[0–185.00]	8.6 (14.50)	[0–68.00]	0.13

Note: Data are represented as mean (SD) and [range]. P‐values indicates the difference between the two groups.

Table 3.

Mucociliary clearance of both lungs.

	Dry NaCl inhalation		Placebo		P‐value
Lung retention (%)
30 min	91.41 (9.64)	[67.2–104.53]	92.10 (5.94)	[82.73–102.10]	0.67
60 min	88.66 (9.67)	[67.03–102.38]	89.32 (6.96)	[78.13–100.80]	0.65
120 min	84.46 (9.82)	[66.93–100.93]	84.60 (8.76)	[70.20–97.43]	0.90
Clearance (%)
30 min	8.95 (9.82)	[0–32.80]	7.90 (5.94)	[−2.1–17.28]	0.51
60 min	11.52 (9.42)	[0–32.98]	10.68 (6.96)	[−0.8–21.88]	0.56
120 min	15.58 (9.75)	[0–33.08]	15.40 (8.76)	[2.57–29.80]	0.88
PI (ratio)	0.59 (0.24)	[0.27–1.21]	0.64 (0.27)	[0.30–1.48]	0.036*

Note: Data are represented as mean (SD) and [range]. P‐values indicates the difference between the two groups.

Abbreviation: PI, penetration index.

3.3. Airway symptoms

Self‐reported airway symptoms and coughing are presented in Table 4. No differences were observed between conditions.

Table 4.

Airway symptoms after inhalation from the BREATHOX® inhaler.

	Dry NaCl inhalation		Placebo		P‐value
35 min after inhalation:
Breathing more easily	10/25		6/25		0.21
Experiences a change in airway condition*	1.14 (1.55)	[0–4]	0.54 (1.02)	[0–3]	0.099
Experiences change in cough**	0.28 (0.96)	[0–4]	0.24 (0.65)	[0–2]	0.85
125 min after inhalation:
Breathing more easily	11***/25		10/25		0.75
Experiences a change in airway condition*	0.96 (1.44)	[−2–4]	0.94 (1.22)	[0–3]	0.96
Experiences change in cough**	0.32 (1.05)	[−1–4]	0.24 (0.81)	[0–3]	0.79

Note: Data are represented as mean (SD) and [range]. P‐values indicates differences between the two groups.

^*Score from −5 to 5 where −5 is a change for the worse and +5 is a change to the better.

^**Score from −5 to 5 where −5 is more difficult breathing and +5 is easier breathing.

^***One participant with more difficult breathing.

3.4. Adverse events

None of the participants experienced any AE or SAE during the study.

4. DISCUSSION

The main finding of this randomised, placebo‐controlled, crossover study is that inhalation of 40 mg dry powder sodium chloride did not affect pulmonary MCC or airway symptoms during the first 2 h after administration in patients with COPD, chronic bronchitis, or both. However, our findings also demonstrate that dry powder sodium chloride inhalation is a safe intervention in these patients, given that no SAE or AEs occurred.

The scintigraphic assessment of MCC is the reference method for evaluating mucociliary function in vivo in humans (Mortensen et al., 1994; Munkholm & Mortensen, 2014). As such, the present findings provide compelling evidence that MCC is not acutely affected to any clinically relevant extent by dry powder sodium chloride inhalation. MCC is known to be impaired in COPD, which is associated with increased vulnerability to respiratory tract infections, an impairment that is attributed to the shortening of cilia and airway epithelial dysfunction (Petit et al., 2023; Tilley et al., 2015). Given that a hyperosmolar microenvironment is thought to stimulate the release of mediators that increase ciliary beat frequency and enhance the interaction between cilia and mucus, thereby improving MCC (Daviskas & Anderson, 2006), the mechanistic basis of dry powder sodium chloride inhalation in these patients seems sound. Nevertheless, the present findings may be interpreted to contrast somewhat with this proposed mechanistic basis, at least in the context of COPD.

While it may be suggested that inflamed airways in COPD render airway cilia resistant to the hyperosmolar stimulus evoked by dry powder sodium chloride inhalation, the finding that MCC is unaffected by dry powder sodium chloride inhalation is somewhat surprising, given that marked effects have been demonstrated in previous studies on cystic fibrosis, asthma, and healthy individuals using the same imaging modality as in the present study (Daviskas & Anderson, 2006; Daviskas et al., 1996; Robinson et al., 1996). However, these differences may to some extent relate to variations in the mode of administration and dosage. In contrast to the dry powder sodium chloride administered in the present study, previous studies used nebulised hyperosmolar saline delivered directly into the airways at doses of NaCl up to 10 times higher than those employed here. Indeed, the effect of a hyperosmolar saline is known to be dose‐dependent (Robinson et al., 1997). In any event, the results provided here stand in contrast to widespread claims outside the scientific literature, including those promoted in mass media and in the context of various alternative medicine practices, that both halotherapy and speleotherapy, the latter involving the use of a natural subterranean salt cave environment, provide immediate respiratory symptom relief in COPD, at least partly through effects on mucociliary function (Rashleigh et al., 2014).

Most participants were receiving regular inhaled β₂‐agonist therapy. β₂‐agonists may enhance mucociliary clearance, and they may also induce bronchodilation, which could theoretically alter the deposition pattern of the radioaerosol used in the mucociliary clearance assessments. However, it is unlikely that the inhaled medication had a substantial impact on the results for two reasons. First, participants took their inhaled medication at the same time on both study days, minimizing any differential effect. Second, markedly higher doses of β₂‐agonists than those typically used for bronchodilation are required to meaningfully enhance mucociliary clearance, as demonstrated in prior studies (Bennett, 2002; Mortensen, 1998).

This study has several limitations. First, the double‐blind design could not be maintained, as participants were able to identify the intervention days due to the perceptible saline taste of the particles. Second, the sample size was relatively small, and the targeted number of participants could not be reached due to external factors, notably the COVID‐19 pandemic. Consequently, the negative findings are associated with a high risk of a Type II error. Although the smaller‐than‐planned sample size increases the risk of a type II error, we did not perform a post‐hoc power calculation, as statistical power is a pre‐trial design concept and post‐hoc power estimates are determined by the observed P value and do not add information beyond the reported effect sizes and confidence intervals (Berg et al., 2024). It is a limitation that the study group was relatively heterogeneous, comprising participants with either COPD, chronic bronchitis, or both conditions, resulting in variability among the cohort. Additionally, the study only evaluated the acute effects of a single treatment on one domain of respiratory function, that is, MCC. In clinical practice, such treatments would typically involve extended administration over months or years rather than a single dose. This limits the generalisability of the findings, which should therefore be interpreted with caution. The results thus cannot be extrapolated to conclude on clinically relevant effects on MCC, respiratory symptoms, or other domains of respiratory function in COPD. To draw robust conclusions, larger randomised studies are thus warranted, in which dry sodium chloride inhalation is compared with placebo over an extended period.

5. CONCLUSIONS

Administration of 40 mg dry powder sodium chloride via a BREATHOX® inhaler appears safe in individuals with COPD and/or chronic bronchitis. However, our findings do not indicate any immediate clinically relevant effects on pulmonary MCC or airway symptoms. These conclusions are limited by the premature termination of the study due to the COVID‐19 pandemic. With only 25 participants enrolled, the study was underpowered, increasing the risk of type II error; therefore, non‐significant results should not be interpreted as evidence of no effect. Larger, adequately powered trials are needed to determine the true effect size.

AUTHOR CONTRIBUTIONS

Clara Mikkelsen: First draft; data interpretation; revisions. Rie Skovly Thomsen: First draft; data interpretation; preparation of tables; revisions. Naja Enevold Olsen: Data collection; data interpretation; revisions. Hrefna Sæunn Einarsdóttir: Data collection; data interpretation; revisions. Mohamed Attauabi: Design; revisions. Ronan M. G. Berg: First draft; data collection; data interpretation; revisions; supervision. Henrik Harboe: Conception; design; revisions. Jann Mortensen: Conception; design; data collection; data interpretation; revisions; supervision.

CONFLICT OF INTEREST STATEMENT

Henrik Harboe is employed by Liita Care ApS. The remaining authors declare no conflicts of interest.

Thomsen, R.S. , Mikkelsen, C. , Olsen, N.E. , Einarsdóttir, H.S. , Attauabi, M. , Berg, R.M.G. et al. (2026) The effect of inhaled dry salt on pulmonary mucociliary clearance in obstructive lung disease: a randomised, placebo‐controlled, crossover study. Clinical Physiology and Functional Imaging, 46, e70060. 10.1111/cpf.70060

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.