Skip to main content
NCBI home page
BMC Pulmonary Medicine logoLink to BMC Pulmonary Medicine
. 2026 May 16;26:302. doi: 10.1186/s12890-026-04348-2

Plasma short-chain fatty acids are associated with COPD and COPD phenotypes

Anders Ørskov Rotevatn 1,2,✉, Tomas Mikal Eagan 1,2, Adrian McCann 3, Johnny Laupsa-Borge 3, Solveig Tangedal 2, Gunnar Reksten Husebø 1,2, Rune Nielsen 1,2
PMCID: PMC13343913  PMID: 42141402

Abstract

Background and aim

Short-chain fatty acids (SCFAs) are metabolites largely produced by the gut microbiota, except for formate, which is also produced by the host. SCFAs may influence COPD development by modulating the immune system. This study examined the relationships between circulating SCFA concentrations and COPD and COPD phenotypes.

Methods

We report findings from a single-centre observational study with 130 patients with COPD and 103 controls. Eight SCFAs (formate, acetate, propionate, butyrate, isobutyrate, valerate, isovalerate and α-methylbutyrate) were quantified in plasma using isotope-labelled gas chromatography-tandem mass spectrometry. Associations between SCFAs, COPD status, and COPD phenotypes (severely obstructive COPD, emphysema, eosinophilic COPD, frequent exacerbations, and symptom burden) were evaluated using multivariable regression analyses adjusted for age, sex, smoking, and body composition.

Results

Plasma formate concentrations were inversely associated with COPD status (adjusted OR 0.33; 95% CI 0.15 to 0.66; p = 0.004). Propionate and isobutyrate concentrations were inversely associated with severely obstructive COPD (adjusted ORs 0.46; 95% CI 0.24 to 0.86; p = 0.018, and 0.33; 95% CI 0.14 to 0.75; p = 0.012, respectively). Propionate concentrations were also inversely associated with frequent exacerbations of COPD (adjusted OR 0.23; 95% CI 0.07 to 0.62; p = 0.007). Isovalerate was positively associated with symptom burden (CAT score) (estimate 1.76; 95% CI 0.20 to 3.32; p = 0.027). No consistent associations were observed for emphysema or eosinophilic COPD.

Conclusions

SCFA formate was inversely related to COPD. Several SCFAs were associated with specific COPD phenotypes. Future studies should investigate the underlying mechanisms linking SCFAs to COPD and include the often-overlooked formate, while acknowledging formate’s dual origin.

Keywords: COPD Chronic obstructive pulmonary disease, Metabolome, Formate, SCFA Short-chain fatty acids, Propionate, Isobutyrate

Introduction

Chronic obstructive pulmonary disease (COPD) is a progressive lung disease characterised by emphysema and chronic inflammation in the airways. The clinical presentation is heterogeneous, with patients differing in degree of airflow impairment, symptom burden, markers of type 2 inflammation (e.g., blood eosinophil count), exacerbation frequency, and extent of emphysema. COPD develops in response to repeated injury from inhaled toxic gases, such as cigarette smoke. However, not all smokers develop COPD. What drives inflammation, and to varying degrees in different patients, is not fully understood. The immune response resembles a chronic infection, and microbial dysbiosis is a possible factor. Studies have shown a dysbiosis in the airway microbiota in patients with COPD [1, 2], but whether it is a cause or an effect remains unknown.

The gut microbiota, the body’s largest microbial community, may also be relevant in chronic lung disease [3]. Differences in the gut microbial composition have been observed between individuals with COPD or emphysema and healthy controls [4–8]. Although the clinical relevance of these findings is still unclear, faecal transplants from patients with COPD have induced lung inflammation and COPD-like changes in mice [6]. The underlying mechanisms are not well understood, but short-chain fatty acids (SCFAs) have been proposed as possible mediators.

SCFAs are gut-derived fatty acids with 1–6 carbon atoms, primarily produced by bacterial fermentation of undigested dietary fibre. However, branched SCFAs (such as isobutyrate, isovalerate and α-methylbutyrate) mainly arise from bacterial fermentation of branched amino acids. Although some SCFAs (acetate, propionate and butyrate) can be produced endogenously in negligible amounts, only formate is produced in substantial amounts by the host. Therefore, except for formate, circulating SCFA concentrations predominantly reflect gut microbial activity. Gut microbiota-derived SCFAs are detectable in the lungs [9], and both alveolar macrophages and alveolar type 2 cells express SCFA receptors [9]. Microbial metabolites, such as propionate and lipopolysaccharides, can influence alveolar immune tone and macrophage metabolism [9].

Only two small studies have previously compared SCFA concentrations in plasma between patients with COPD and controls, with conflicting results. One reported reduced acetate concentrations in COPD [10]. In the other, COPD smokers were compared with non-COPD smokers without finding differences, but the number of participants (6 per group) was likely too low to detect biological differences [11]. Others have reported lower acetate in feces of patients with COPD [6] and emphysema [12]. Additionally, faecal isobutyrate and isovalerate were reported to be lower in patients with severely obstructive COPD [6].

The MicroCOPD study included 130 patients with COPD and 103 controls with clinical data on exacerbation frequency, blood eosinophil level, lung function, symptom score, and emphysema. The current study aimed to investigate whether plasma SCFA concentrations differed between patients with COPD and controls, and according to COPD phenotypes.

Methods

Study design and population

The MicroCOPD study is a single-centre observational study (2012–2015, Haukeland University Hospital, Norway). The main objective was to investigate the role of microbiota in COPD, using bronchoscopy to sample the lower airways. Eligible participants were ≥ 40 years of age and included never, former, and current smokers. Inclusion was postponed if subjects had received antibiotics or oral corticosteroids within the preceding two weeks, or if they had symptoms of an ongoing COPD exacerbation. COPD diagnosis and control status were confirmed by experienced pulmonologists based on medical history and post-bronchodilator spirometry. Exclusion criteria were high bleeding risk, cardiac valve prosthesis, acute coronary syndrome within 6 weeks, cancer within the past five years, or significant immunosuppression. The controls had no history of lung or airway disease. A detailed protocol for the MicroCOPD study has previously been published [13].

Measurements

At inclusion, participants underwent blood sampling and structured interviews, including smoking history, symptom burden (COPD Assessment Test; CAT), and time since last antibiotic treatment. Spirometry was performed using a Viasys Vmax ENCORE 30 min after inhalation of 400 µg of salbutamol administered through a large-volume spacer. Body composition was measured by bioelectrical impedance. Cachexia was defined as having a fat-free mass index (kg/m2) of < 14 for women and < 17 for men, while obesity was defined as a fat mass index (kg/m2) of > 13.5 in women and > 9.3 in men [14]. Thoracic computed tomography (CT thorax) scans were performed using a Siemens Somatom Definition Flash scanner. Emphysema was defined as > 10% low attenuating area (LAA) at density threshold −950 Hounsfield units (HU), using the free 3D Slicer software [15]. Blood counts were measured using standard laboratory methods. The plasma samples were collected according to standard phlebotomy techniques into EDTA-vacutainer tubes. Plasma separation was performed by centrifugation at 2,150 g for 15 min at 4 °C, after which the plasma was aliquoted and stored at −80 °C until analysis.

Targeted analyses of SCFAs

Eight SCFAs – formate, acetate, propionate, butyrate, isobutyrate, valerate, isovalerate and α-methylbutyrate – were targeted for quantification in plasma using isotope-labelled internal standards for each metabolite and gas chromatography–tandem mass spectrometry (GC–MS/MS) at Bevital AS, Bergen, Norway (https://bevital.no/). The limits of detection for the metabolites were as follows: formate, 7.0 µmol/L; acetate, 5.0 µmol/L; propionate, 0.06 µmol/L; butyrate, 0.2 µmol/L; isobutyrate, 0.05 µmol/L; valerate, 0.05 µmol/L; isovalerate, 0.02 µmol/L; and α-methylbutyrate, 0.05 µmol/L. Within- and between-day coefficient of variation ranged from 3 to 7% and 3–8%, respectively.

In addition, we also calculated the total sum of SCFAs and the ratio of total to branched SCFAs (isobutyrate, isovalerate and α-methylbutyrate), the latter serving as an inverse marker of the degree of protein fermentation.

Statistical analysis

Statistical analyses were conducted in R (version 4.4.1). Continuous data are presented as mean ± SD, and categorical variables as n (%) unless otherwise specified. Left-censored missing values of SCFA concentrations due to lower than the limit of detection or quantification were considered missing not at random and imputed using the GSimp procedure [16].

Associations between COPD and SCFA concentrations were assessed using unconditional logistic regression with COPD (yes/no) as the binary outcome: Three models were fitted: (i) unadjusted, (ii) adjusted for age and sex, and (iii) adjusted for age, sex, smoking, and body composition. Due to their multiplicative nature, SCFA concentrations were log2-transformed. With this transformation, the odds ratio reflects the change in odds associated with a doubling in SCFA concentrations.

Among participants with COPD, associations between COPD phenotypes and log2-transformed SCFA concentrations were investigated similarly using logistic regression with the following binary outcomes: emphysema vs. no emphysema, severely obstructive COPD as defined as forced expiratory volume in 1 s (FEV1) < 50% vs. FEV1 ≥50% of predicted values, high vs. low eosinophils (≥/<0.3 × 109/L), and frequent vs. infrequent exacerbations (≥ 2/0–1 exacerbations of COPD the preceding year). For frequent exacerbations, adjustments were limited to age, sex, and smoking. Associations between symptom burden and SCFA concentrations were analysed using linear regression. Models were fitted (i) unadjusted, (ii) adjusted for age and sex, and (iii) additionally adjusted for smoking and body composition. With log2-transformation, the regression estimates the expected change in CAT score associated with a doubling of SCFA concentrations.

As a complementary analysis aimed at estimating group differences in SCFA concentrations, we fitted generalised least squares (GLS) models allowing for unequal residual variances between groups. Group was the main predictor, comparing COPD vs. controls, and each binary COPD phenotype (emphysema vs. no emphysema, severe airflow obstruction vs. non-severe, high vs. low blood eosinophils, and frequent vs. infrequent exacerbations). Because most biochemical variables fit a log-normal (multiplicative) distribution equally well or better than a normal (additive) distribution [17], SCFA values were transformed using the natural logarithm (ln). Relative between-group differences are expressed as sympercents (s%), which are additive and symmetric percentage differences on the 100 ln scale [18]. This relative measure is calculated as the difference between the natural logs of two numbers multiplied by 100. GLS was also used to evaluate the associations between symptom burden (CAT score) and natural log-transformed SCFA concentrations, with CAT score included as a continuous predictor. In this context, the relative measure represents the symmetric percentage change (s%) in SCFA concentration per 1-point increase in CAT score. All GLS models were fitted (i) unadjusted, (ii) adjusted for age and sex, and (iii) additionally adjusted for smoking and body composition.

Because analyses of SCFAs in relation to COPD status and phenotypes were explicitly exploratory and hypothesis-generating, we did not apply formal adjustments for multiple testing. A nominal two-sided alpha of 0.05 was used, and we report exact p-values with effect estimates and 95% confidence intervals to allow readers to judge the strength and precision of the evidence. All findings should be interpreted as provisional and in need of confirmation in independent cohorts.

Results

Participant demographics and clinical characteristics are summarised in Table 1. Compared with controls, patients with COPD had greater smoking exposure, reduced lung function, and a higher prevalence of cachexia. In the COPD group, 24% had emphysema, 39% had FEV1 <50%, 35% had eosinophilic COPD, and 13% reported frequent COPD exacerbations.

Table 1.

The demographic and clinical characteristics of the study participants

Characteristic Control COPD Overall
N = 103 N = 130 N = 233
Age 65.25 (± 8.64) 67.22 (± 7.35) 66.35 (± 7.99)
Sex (Male) 60 (58%) 76 (58%) 136 (58%)
Smoking status
 Never 19 (18%) 0 (0%) 19 (8.2%)
 Former 59 (57%) 100 (77%) 159 (68%)
 Current 25 (24%) 30 (23%) 55 (24%)
Body composition
 Normal 78 (76%) 81 (62%) 159 (69%)
 Cachexia 10 (9.8%) 27 (21%) 37 (16%)
 Obese 14 (14%) 22 (17%) 36 (16%)
 Missing 1 0 1
FVC % of predicted 112 (± 13) 94 (± 18) 102 (± 18)
FEV1% of predicted 104 (± 12) 57 (± 19) 78 (± 29)
FEV1/FVC-ratio 0.73 (± 0.05) 0.47 (± 0.12) 0.59 (± 0.17)
Eosinophilic COPD (Blood eosinophils ≥ 0.3 × 109) 46 (35%)
Emphysema (≥ 10% LAA at -950 HU threshold) 28 (24%)
 Missing 14
FEV1 <50% 51 (39%)
Frequent COPD exacerbations (≥ 2 COPD exacerbations previous 12 months) 17 (13%)
 Missing 1
Symptom burden (CAT score) 16 (± 8)
 Missing 2
Open in a new tab

Data are presented as n (%) or mean ± SD

BMI body mass index, LAA low attenuating area, HU Hounsfield units, FEV1 forced expiratory volume in 1 s, FVC forced vital capacity, CAT COPD Assessment Test

COPD vs. controls

The geometric mean (SD range) of the plasma SCFA concentrations in controls and patients with COPD is presented in Table 2. Plasma acetate concentrations were highest, followed by formate. Valerate had the lowest overall plasma concentrations and the highest number of left-censored data (missing not at random) at 111 (47.6%), with butyrate second highest at 8 (3.4%) left-censored data.

Table 2.

Concentrations of plasma SCFAs in COPD and controls

SCFAs, µmol/L Control COPD p-value2
N = 103 N = 130
Formate 13.0 (10.5–16.1) 11.3 (7.35–17.5) 1a < 0.001
Acetate 35.78 (20.56–62.25) 35.49 (16.06–78.45) 0.644
Propionate 0.95 (0.65–1.39) 0.99 (0.64–1.53) 0.282
Butyrate 0.44 (0.20–0.98)1b 0.45 (0.23–0.87)1c 0.732
Isobutyrate 0.25 (0.19–0.34) 0.26 (0.19–0.37) 0.334
Valerate 0.05 (0.03–0.10)1d 0.06 (0.03–0.11)1e 0.908
Isovalerate 0.32 (0.17–0.62) 0.30 (0.16–0.57) 0.418
α-methylbutyrate 0.12 (0.09–0.17) 0.12 (0.08–0.18)1f 0.792
Total SCFAs 51.66 (33.07–80.72) 51.02 (25.93–100.39) 0.752
Ratio of total to branched SCFAs 70.59 (44.37–112.30) 71.80 (35.10–146.89) 0.560
Open in a new tab

Geometric mean (SD range). 1n (%) missing not at random, n (%) values imputed groupwise by the GSimp method; 1a 6 (4,6%), 1b 3 (2,9%), 1c 5 (3,8%), 1d 46 (44,7%), 1e 65 (50,0%), 1f 1 (0,8%). 2P-values from linear regression modeling (generalised least squares), including a variance function structure allowing for different variances across groups and corrected for age, sex, smoking status and body composition

A doubling in plasma formate concentrations was associated with 67% lower odds of COPD (adjusted OR 0.33; 95% CI 0.15 to 0.66; p = 0.004; Fig. 1). None of the other SCFAs were significantly associated with COPD status (all p > 0.05). Consistent with these findings, formate concentrations were lower in patients with COPD than in controls, with an adjusted difference of −15.5 s% (95% CI −24.1 to −6.8; p < 0.001) after adjustment for age, sex, smoking status and body composition (Fig. 2).

Fig. 1.

Fig. 1

Open in a new tab

Associations between log2-transformed metabolite concentrations and COPD status (COPD vs. controls) from logistic regression analyses. Odds ratios represent the multiplicative change in odds associated with a doubling of metabolite concentrations. ** p < 0.01

Fig. 2.

Fig. 2

Open in a new tab

Group differences in SCFA concentrations expressed as sympercents (s%) with 95% confidence intervals according to A COPD status (COPD vs. controls), and among participants with COPD with: B FEV1 < 50%, C frequent exacerbations and D symptom score, all adjusted for age, sex, smoking status and body composition. * p < 0.05. ** p < 0.01. *** p < 0.001

Severely obstructive COPD

FEV1 <50% of predicted was associated with lower plasma propionate (adjusted OR 0.46; 95% CI 0.24 to 0.86; p = 0.018) and isobutyrate (adjusted OR 0.33; 95% CI 0.14 to 0.75; p = 0.012). The ratio of total to branched SCFAs was positively associated with severely obstructive COPD (adjusted OR 1.59; 95% CI 1.10 to 2.45; p = 0.021) (Fig. 3).

Fig. 3.

Fig. 3

Open in a new tab

Associations between log2-transformed metabolite concentrations and COPD phenotypes. Panel A shows associations with dichotomic phenotypes, and Panel B shows associations with symptom score, numeric). Model 1 is unadjusted. Model 2 is adjusted for age and sex. Model 3a is adjusted for age, sex, smoking status and body composition. Model 3b is adjusted for age, sex and smoking. Because metabolite concentrations are log2-transformed, odds ratios from logistic regression represent the multiplicative change in odds associated with a doubling of metabolite concentration, whereas regression coefficients from linear regression represent the change in symptom score per doubling of metabolite concentration. * p < 0.05. ** p < 0.01

Consistently, propionate and isobutyrate concentrations were lower in those with FEV1 <50% (−18.28 s%; 95% CI −33.23 to −3.33 s%; p = 0.018 and −16.04 s%; 95% CI −26.85 to −5.23 s%; p = 0.004), while the ratio of total to branched SCFAs was greater (30.28 s%; 95% CI 3.35 to 57.22 s%; p = 0.029). All estimates were adjusted for age, sex, smoking status and body composition (Fig. 2).

Frequent exacerbators

Plasma propionate concentration was inversely associated with frequent exacerbations of COPD (adjusted OR 0.23; 95% CI 0.07 to 0.62; p = 0.007). Valerate and α-methylbutyrate did not reach the predefined significance threshold, but they showed a trend towards inverse associations with frequent exacerbations (adjusted p = 0.093 and 0.068, respectively).

Consistently, propionate concentrations were lower in patients with COPD with frequent exacerbations (−29.00 s%; 95% CI −46.72 to −11.27; p = 0.002; adjusted for age, sex, smoking and body composition). Valerate and α-methylbutyrate concentrations were also lower in frequent exacerbators in unadjusted analyses (−33.11 s%; 95% −58.01 to −8.21; p = 0.010 and −18.21 s%; 95% CI −35.67 to −0.76; p = 0.043). When adjusting for age and sex, the reduction remained significant (−27.59 s%; 95% CI −52.54 to −2.64; p = 0.032 and −18.87; 95% CI −38.90 to −2.68; p = 0.037, respectively). However, these associations were no longer significant after additional adjustment for smoking status and body composition (p = 0.104 and 0.169, respectively).

Eosinophilic COPD

No SCFAs were significantly associated with eosinophilic COPD (p > 0.05), and there were no significant differences in SCFA concentrations between patients with high and low blood eosinophils (p > 0.05).

Emphysema

Valerate was barely significantly associated with emphysema according to unadjusted analysis (OR 1.56; 95% CI 1.01 to 2.48; p = 0.049) but was not significant in the adjusted models (Fig. 3). None of the other SCFAs were significantly associated with emphysema among patients with COPD (p > 0.05).

Symptom burden

A doubling of isovalerate concentrations was associated with a multiplicative increase in CAT score of 1.76 (95% CI 0.20 to 3.32; p = 0.027), adjusted for age, sex, smoking status and body composition. Consistently, isovalerate concentrations increased by 1.56 s% per one-point increase in CAT score (95% CI 0.19 to 2.93; p = 0.027; adjusted for age, sex, smoking and body composition).

Discussion

In this study, we investigated the associations between plasma SCFA concentrations in patients with COPD compared with controls, as well as in COPD patients with different disease manifestations. Overall, formate concentrations were lower in patients with COPD. Specific SCFAs were related to distinct COPD phenotypes. Propionate was lower in those with FEV1 <50% and in patients with frequent exacerbations, isobutyrate was lower in patients with severely obstructive COPD, and isovalerate was positively associated with symptom burden.

Formate and COPD status

Formate concentrations were the second highest among the SCFAs in our study participants, with concentrations more than tenfold greater than the third highest. While formate has not been included in previous SCFA-focused studies on patients with COPD [6, 10–12], untargeted metabolomic approaches using nuclear magnetic resonance (NMR) for metabolome analysis have reported altered formate levels in COPD. One study analysing exhaled breath condensate (EBC) observed lower formate concentrations in patients with COPD compared with controls [19]. Another study reported that serum formate concentrations and lung function increased in patients with COPD after three months of doxycycline treatment, but not in controls [20], and a third study demonstrated a positive association between urinary formate and lung function, but not with plasma formate, possibly due to the limited sensitivity of their analytical methods [21]. Additionally, serum formate was found to be lower in patients with acute respiratory failure than those with stable COPD [22].

In contrast, Zheng et al. found higher serum formate in 54 patients with COPD compared with 74 controls [23], but they used untargeted NMR for quantification, which is less sensitive and specific than the targeted GC–MS/MS methods we applied. Airoldi et al. reported higher formate in EBC from alpha1-antitrypsin-deficient patients with emphysema compared with healthy controls [24], but the study included only 22 participants. Circulating formate has also been found to be reduced in other maladies such as HER2 + breast cancer, non-small cell lung cancer and obesity [25].

Formate is an intermediate metabolite in the one-carbon metabolism [26], produced by both host mitochondrial pathways (primarily serine and glycine catabolism via SHMT2 and MTHFD2) and by anaerobic bacterial fermentation in the gut. Circulating formate therefore reflects the host’s metabolic state, mitochondrial one-carbon flux, and microbiome activity. As a key one-carbon donor, formate supports de novo purine and pyrimidine synthesis, histone and protein methylation, and immune cell proliferation. Its dual origin complicates interpretation, but the inverse association observed here suggests that reduced circulating formate in COPD reflects broader metabolic stress —specifically impaired mitochondrial one-carbon metabolism and systemic energy deficits — rather than gut-derived immunomodulation alone. These findings highlight the need for further investigation of formate as a potential marker of COPD-related metabolic dysfunction.

Propionate

Propionate concentrations were lower in those with FEV1 <50% and in frequent exacerbations, somewhat consistent with Otake’s study, which reported a positive correlation between propionate and FEV1/FVC [11], although Li et al. found no significant difference in propionate according to severity of COPD [6]. Propionate has previously been reported to be increased in EBC from patients with COPD [19] and alpha1-antitrypsin-deficient patients with emphysema [24] compared with controls. In mice, propionate enhances haematopoiesis of dendritic cell precursors and leads to impaired activation of TH2 effector cells in the lung and resolution of allergic airway inflammation. These effects depended on G protein–coupled receptor 41 (Free fatty acid receptor 3) [27]. Propionate also represses IL-17 by human γδ T cells [28]. However, we found no significant association between propionate (or any other SCFAs) and eosinophilic COPD.

Isobutyrate and isovalerate

Isobutyrate was also lower in patients with severely obstructive COPD, consistent with an earlier study focusing on protein degradation in COPD [29]. Isovalerate was positively associated with symptom burden, a previously unreported finding. Branched SCFAs such as isobutyrate and isovalerate are understudied, although high colonic and systemic concentrations have been linked to reduced gut and metabolic health [30].

Acetate

Acetate had the highest concentrations of the measured SCFAs in our material. We found no differences in acetate between COPD and controls, consistent with Otake et al. [11]. Other studies have reported lower serum [10] and faecal [6, 12] acetate in patients with COPD and emphysema. Otake et al. also reported reduced acetate and propionate in current smokers compared with non-smokers [11]. We adjusted for smoking in our analysis, but this did not significantly affect associations between COPD, COPD phenotypes and SCFAs.

Emphysema

Contrary to expectations, we found no strong associations between emphysema and SCFAs. A weak positive correlation with valerate was no longer significant after adjusting for age and sex. Valerate had the lowest concentrations among the SCFAs, with many missing values imputed, which limits the reliability of our observations. Our results differ from those of Lee et al., who reported lower faecal acetate in patients with emphysema [12]. There is a considerable concentration gradient between colonic and systemic SCFA levels [30], which could allow for different patterns in serum and faecal samples. Additionally, while emphysema usually slowly evolves, reflecting chronic pathological processes, we measured SCFAs only at a single time point, and cannot exclude temporal variation in SCFA concentrations. Lee et al. reported therapeutic effects of acetate and propionate in a smoking-induced murine model of emphysema [12]. This finding aligns with Kim et al., who found that antibiotics aggravated emphysema development and changed the microbiota, whereas subsequent SCFA administration attenuated this effect [31]. Other murine studies have shown that high-fibre diets increase SCFA and attenuate emphysema [32, 33]. Large human studies similarly associate higher dietary fibre with lower COPD risk [34–36] and higher lung function [37]. Long-term dietary intervention trials should further explore whether systemic SCFA changes mediate protective effects. Additionally, future studies on COPD and SCFAs should include formate.

Strengths and limitations

Strengths include detailed clinical characteristics of participants, use of highly sensitive isotope-labelled GC–MS/MS, and measurement of circulating SCFAs. Limitations include (1) a lack of dietary data, as fibre and protein intake influence SCFAs, (2) the inability to assess other COPD phenotypes, such as chronic bronchitis and rapidly progressing COPD, (3) the cross-sectional design, which limits causal inference. Consequently, our data cannot infer the directionality of the associations, which may result from SFCAs’ effect on COPD disease, or secondary to COPD-related changes in activity, nutritional intake, or frailty-related microbiome changes, rather than being primary contributors to disease development. (4) multiple statistical testing across multiple SCFA measures and several COPD-related outcomes without formal correction for multiple testing, increasing the risk of chance findings — results should therefore be interpreted as exploratory and hypothesis-generating, and (5) SCFAs are microbiota-derived, and although we required a 14-day antibiotic washout, residual effects cannot be ruled out. However, most participants had a substantially interval since their last antibiotic exposure. Time since last antibiotic treatment was recorded in 194 of 233 participants and exceeded 12 weeks for most participants (89 participants with COPD (92%), 90 controls (93%)). (6) Lack of specific data on ongoing inflammation in the small airways, the primary site of inflammation in COPD, which is where any SCFA-related immunomodulatory effects in COPD are most likely to occur. Specific markers from immunohistology or CT (e.g., mucus plug scores, airway wall thickness) could enable further patient stratification and reveal associations not apparent in the current analysis.

Conclusions

Plasma formate concentrations were consistently lower in patients with COPD compared with controls. Several SCFAs aligned with key COPD phenotypes, including FEV1 <50%, frequent exacerbations and symptom burden. Together, our findings outline a potential metabolic signature, while keeping the dual origin of formate in mind. Future studies should elucidate the mechanisms linking SCFAs to COPD, explicitly including the often-overlooked formate, and employ longitudinal and interventional designs to test therapeutic potential.

Acknowledgements

In addition to all the participants, the authors are indebted to Eli Nordeide, Lise Monsen, Hildegunn Fleten, Ingvild Haaland, Harald Wiker, Christine Drengenes, Tharmini Kalananthan, Sverre Lehmann, Einar Marius Helgeland Martinsen, Elise Orvedal Leiten, Per Bakke, Kristel Knudsen, Øistein Svanes and Kristoffer Ostridge for their participation in the planning, data collection and analyses of data from the MicroCOPD study.

Abbreviations

CAT

COPD Assessment Test

CI

Confidence interval

COPD

Chronic obstructive pulmonary disease

CT thorax

Thoracic computed tomography

EBC

Exhaled breath condensate

FEV1

Forced expiratory volume in 1 s

FVC

Forced vital capacity

GC-MS/MS

Gas chromatography–tandem mass spectrometry

GLS

Generalised least squares

HU

Hounsfield units

LAA

Low attenuating area

NMR

Nuclear magnetic resonance

OR

Odds ratio

SCFA

Short–chain fatty acid

SD

Standard deviation

s%

Sympercent

Authors' contributions

AØR – Statistical analysis, Writing. TME – Concept and design. Data collection. Supervision. Writing – review and editing. AMC – Writing – review and editing. JLB – Statistical analysis. Writing – review and editing. ST – Supervision. Writing – review and editing. GRH – Data collection. Writing – review and editing. RN – Concept and design. Supervision. Data collection. Writing – review and editing.

Funding

Open access funding provided by University of Bergen. The MicroCOPD study was funded by unrestricted grants and fellowships from Helse Vest, Bergen Medical Research Foundation, the Endowment of Timber Merchant A. Delphin and Wife through the Norwegian Medical Association, AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline and Novartis. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The datasets generated and/or analysed during the current study are available in the Dryad repository, [10.5061/dryad.mw6m9069h].

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki. The Regional Committee for Medical and Health Research Ethics (REK Nord) approved the study, case number 2011/1307. REK Nord is affiliated with the University of Tromsø and the Ministry of Education and Research, Norway. All participants provided both oral and written informed consent.

Consent for publication

Not applicable.

Competing interests

Dr. Rotevatn reports grants from Boehringer Ingelheim, AstraZeneca and the Endowment of Timber Merchant A. Delphin and Wife during the conduction of the study and honoraria for a presentation from AstraZeneca. Dr. Eagan reports grants from Helse Vest Regional Health Trust during the conduction of the study, honoraria and support for attending meetings from Boehringer Ingelheim and AstraZeneca outside the submitted work, serves as DSMB leader for the RART study on malnutrition in Norway, attended advisory board meetings for Boehringer Ingelheim, and is a member of the Norwegian Committee for Training of Pulmonologists (unpaid). Dr. Nielsen reports grants from the Endowment of Timber Merchant A. Delphin, AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, and Novartis during the conduction of the study. Dr. McCann, Mr. Laupsa-Borge, Dr. Tangedal, and Dr. Husebø report no conflicts of interest.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  • 1.Ramsheh MY, Haldar K, Esteve-Codina A, Purser LF, Richardson M, Müller-Quernheim J, et al. Lung microbiome composition and bronchial epithelial gene expression in patients with COPD versus healthy individuals: a bacterial 16S rRNA gene sequencing and host transcriptomic analysis. Lancet Microbe. 2021;2(7):e300–10. [DOI] [PubMed] [Google Scholar]
  • 2.Tangedal S, Nielsen R, Aanerud M, Drengenes C, Husebø GR, Lehmann S, et al. Lower airway microbiota in COPD and healthy controls. Thorax. 2024;79(3):219–26. [DOI] [PubMed] [Google Scholar]
  • 3.Wang L, Cai Y, Garssen J, Henricks PAJ, Folkerts G, Braber S. The Bidirectional Gut-Lung Axis in Chronic Obstructive Pulmonary Disease. Am J Respir Crit Care Med. 2023;207(9):1145–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Bowerman KL, Rehman SF, Vaughan A, Lachner N, Budden KF, Kim RY, et al. Disease-associated gut microbiome and metabolome changes in patients with chronic obstructive pulmonary disease. Nat Commun. 2020;11(1):5886. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Lai T, Luo C, Yuan Y, Fang J, Wang Y, Tang X, et al. Promising Intestinal Microbiota Associated with Clinical Characteristics of COPD Through Integrated Bioinformatics Analysis. Int J Chron Obstruct Pulmon Dis. 2024;19:873–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Li N, Dai Z, Wang Z, Deng Z, Zhang J, Pu J, et al. Gut microbiota dysbiosis contributes to the development of chronic obstructive pulmonary disease. Respir Res. 2021;22(1):274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Rotevatn AØ, Mikal ET, Solveig T, Reksten HG, Kristoffer O, Nielsen R. Gut microbiota in chronic obstructive pulmonary disease varies by CT-verified emphysema status. Eur Clin Respiratory J. 2025;12(1):2470499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wei Y, Lu X, Liu C. Gut microbiota and chronic obstructive pulmonary disease: a Mendelian randomization study. Front Microbiol. 2023;14:1196751. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Liu Q, Tian X, Maruyama D, Arjomandi M, Prakash A. Lung immune tone via gut-lung axis: gut-derived LPS and short-chain fatty acids’ immunometabolic regulation of lung IL-1β, FFAR2, and FFAR3 expression. Am J Physiology-Lung Cell Mol Physiol. 2021;321(1):L65–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Kirschner SK, Deutz NEP, Jonker R, Olde Damink SWM, Harrykissoon RI, Zachria AJ, et al. Intestinal function is impaired in patients with Chronic Obstructive Pulmonary Disease. Clin Nutr. 2021;40(4):2270–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Otake S, Chubachi S, Miyamoto J, Haneishi Y, Arai T, Iizuka H, et al. Impact of smoking on gut microbiota and short-chain fatty acids in human and mice: Implications for COPD. Mucosal Immunol. 2025;18(2):353–65. [DOI] [PubMed] [Google Scholar]
  • 12.Lee SH, Kim J, Kim NH, Kim O-H, Shon C-H, Kim SJ, et al. Gut microbiota composition and metabolite profiling in smokers: a comparative study between emphysema and asymptomatic individuals with therapeutic implications. Thorax. 2023;78(11):1080–9. [DOI] [PubMed] [Google Scholar]
  • 13.Grønseth R, Haaland I, Wiker HG, Martinsen EMH, Leiten EO, Husebø G, et al. The Bergen COPD microbiome study (MicroCOPD): rationale, design, and initial experiences. Eur Clin Respir J. 2014;1(1):1. 10.3402/ecrj.v1.26196. [DOI] [PMC free article] [PubMed]
  • 14.Eagan TML, Gabazza EC, D’Alessandro-Gabazza C, Gil-Bernabe P, Aoki S, Hardie JA, et al. TNF-α is associated with loss of lean body mass only in already cachectic COPD patients. Respir Res. 2012;13(1):48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fedorov A, Beichel R, Kalpathy-Cramer J, Finet J, Fillion-Robin J-C, Pujol S, et al. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magn Reson Imaging. 2012;30(9):1323–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wei R, Wang J, Jia E, Chen T, Ni Y, Jia W. GSimp: A Gibbs sampler based left-censored missing value imputation approach for metabolomics studies. PLoS Comput Biol. 2018;14(1):e1005973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Limpert E, Stahel WA. Problems with Using the Normal Distribution – and Ways to Improve Quality and Efficiency of Data Analysis. PLoS ONE. 2011;6(7):e21403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Cole TJ. Sympercents: symmetric percentage differences on the 100 log(e) scale simplify the presentation of log transformed data. Stat Med. 2000;19(22):3109–25. [DOI] [PubMed] [Google Scholar]
  • 19.de Laurentiis G, Paris D, Melck D, Montuschi P, Maniscalco M, Bianco A, et al. Separating Smoking-Related Diseases Using NMR-Based Metabolomics of Exhaled Breath Condensate. J Proteome Res. 2013;12(3):1502–11. [DOI] [PubMed] [Google Scholar]
  • 20.Singh B, Jana SK, Ghosh N, Das SK, Joshi M, Bhattacharyya P, et al. Metabolomic profiling of doxycycline treatment in chronic obstructive pulmonary disease. J Pharm Biomed Anal. 2017;132:103–8. [DOI] [PubMed] [Google Scholar]
  • 21.McClay JL, Adkins DE, Isern NG, O’Connell TM, Wooten JB, Zedler BK, et al. 1H Nuclear Magnetic Resonance Metabolomics Analysis Identifies Novel Urinary Biomarkers for Lung Function. J Proteome Res. 2010;9(6):3083–90. [DOI] [PubMed] [Google Scholar]
  • 22.Fortis S, Lusczek ER, Weinert CR, Beilman GJ. Metabolomics in COPD Acute Respiratory Failure Requiring Noninvasive Positive Pressure Ventilation. Can Respir J. 2017;2017(1):9480346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Zheng H, Hu Y, Dong L, Shu Q, Zhu M, Li Y, et al. Predictive diagnosis of chronic obstructive pulmonary disease using serum metabolic biomarkers and least-squares support vector machine. J Clin Lab Anal. 2021;35(2):e23641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Airoldi C, Ciaramelli C, Fumagalli M, Bussei R, Mazzoni V, Viglio S, et al. 1H NMR To Explore the Metabolome of Exhaled Breath Condensate in α1-Antitrypsin Deficient Patients: A Pilot Study. J Proteome Res. 2016;15(12):4569–78. [DOI] [PubMed] [Google Scholar]
  • 25.Pietzke M, Arroyo SF, Sumpton D, Mackay GM, Martin-Castillo B, Camps J, et al. Stratification of cancer and diabetes based on circulating levels of formate and glucose. Cancer Metabolism. 2019;7(1):3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Pietzke M, Meiser J, Vazquez A. Formate metabolism in health and disease. Mol Metabolism. 2020;33:23–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Trompette A, Gollwitzer ES, Yadava K, Sichelstiel AK, Sprenger N, Ngom-Bru C, et al. Gut microbiota metabolism of dietary fiber influences allergic airway disease and hematopoiesis. Nat Med. 2014;20(2):159–66. [DOI] [PubMed] [Google Scholar]
  • 28.Dupraz L, Magniez A, Rolhion N, Richard ML, Da Costa G, Touch S, et al. Gut microbiota-derived short-chain fatty acids regulate IL-17 production by mouse and human intestinal γδ T cells. Cell Rep. 2021;36(1):109332. [DOI] [PubMed] [Google Scholar]
  • 29.Ubhi BK, Riley JH, Shaw PA, Lomas DA, Tal-Singer R, MacNee W, et al. Metabolic profiling detects biomarkers of protein degradation in COPD patients. Eur Respir J. 2011;40(2):345–55. [DOI] [PubMed] [Google Scholar]
  • 30.Blaak EE, Canfora EE, Theis S, Frost G, Groen AK, Mithieux G, et al. Short chain fatty acids in human gut and metabolic health. Benef Microbes. 2020;11(5):411–55. [DOI] [PubMed] [Google Scholar]
  • 31.Kim NH, Choi B-Y, Kim ES, Kim SJ, Hong JY, Heo S-H, et al. Systemic antibiotics cause deterioration of emphysema associated with exaggerated inflammation and autophagy. Exp Mol Med. 2023;55(10):2260–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Jang YO, Kim O-H, Kim SJ, Lee SH, Yun S, Lim SE, et al. High-fiber diets attenuate emphysema development via modulation of gut microbiota and metabolism. Sci Rep. 2021;11(1):7008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Tomoda K, Kubo K, Dairiki K, Yamaji T, Yamamoto Y, Nishii Y, et al. Whey peptide-based enteral diet attenuated elastase-induced emphysema with increase in short chain fatty acids in mice. BMC Pulm Med. 2015;15(1):64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kaluza J, Harris H, Wallin A, Linden A, Wolk A. Dietary Fiber Intake and Risk of Chronic Obstructive Pulmonary Disease: A Prospective Cohort Study of Men. Epidemiology. 2018;29(2):254–60. [DOI] [PubMed] [Google Scholar]
  • 35.Szmidt MK, Kaluza J, Harris HR, Linden A, Wolk A. Long-term dietary fiber intake and risk of chronic obstructive pulmonary disease: a prospective cohort study of women. Eur J Nutr. 2020;59(5):1869–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kan H, Stevens J, Heiss G, Rose KM, London SJ. Dietary Fiber, Lung Function, and Chronic Obstructive Pulmonary Disease in the Atherosclerosis Risk in Communities Study. Am J Epidemiol. 2007;167(5):570–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Hanson C, Lyden E, Rennard S, Mannino DM, Rutten EPA, Hopkins R, et al. The Relationship between Dietary Fiber Intake and Lung Function in the National Health and Nutrition Examination Surveys. Annals Am Thorac Soc. 2016;13(5):643–50. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets generated and/or analysed during the current study are available in the Dryad repository, [10.5061/dryad.mw6m9069h].

Articles from BMC Pulmonary Medicine are provided here courtesy of BMC

RESOURCES