Pulmonary function tests are essential to detecting impairments in gas exchange. The diffusing capacity of the lungs for carbon monoxide (DLCO) is a key marker of alveolar-capillary integrity, as it reflects both structural aspects of the lung and pulmonary blood flow [1].

Scientific evidence indicates that reduced DLCO also reflects systemic inflammatory and metabolic alterations. A drop in DLCO may represent one of the earliest indicators of pulmonary impairment [2–4]. These early alterations may be influenced by perinatal conditions [5], smoking [2,6], and excess weight [7]. In individuals with chronic obstructive pulmonary disease (COPD), for example, reduced DLCO is frequently associated with emphysema, even when spirometry remains within normal limits [8].

Most available data come from clinical samples or disease-specific cohorts, which limit our understanding of these parameters in the general population [9–11]. This gap hampers the identification of early functional changes that may precede overt disease, hindering the development of targeted prevention strategies.

Only a few population studies [2,12] have investigated DLCO and its components—AV and the transfer coefficient for carbon monoxide (KCO)—while accounting for demographic and health factors. The 1993 Pelotas Birth Cohort [13] helps address assessments of DLCO, AV, and KCO at age 22.

We aimed to describe DLCO, AV, and KCO at 22 years of age in the 1993 Pelotas Birth Cohort and investigate the associations with smoking, sex, and early-life and current health factors in a population-based setting.

Table 1 illustrates the sample characteristics by sex. About two-thirds of participants self-reported as white, with similar distributions between sexes. Women were more likely to be in the lowest socioeconomic quintile, not reach physical activity recommendations, and be obese. A higher proportion of men reported smoking vs women (15.3%).

DLCO analyses included only tests with quality grades A–D, totaling 2743 participants (1283 men and 1460 women), ensuring high reproducibility and a sufficient number of valid maneuvers (Supplementary Table S2).

Tables 2 and 3 present the averages of DLCO, AV, and KCO (and their respective 95% CI) by early-life, sociodemographic, behavioral, and health-related variables. Regarding early-life and demographic variables, among men, lower DLCO and AV were observed in those born prematurely, with low birth weight, or exposed to maternal smoking. Women born with low birth weight showed lower DLCO. Black participants had lower DLCO and AV, and higher KCO, in both sexes.

When current variables were assessed, overweight and obesity were linked to higher DLCO and KCO, especially in women, though AV tended to be lower in obese individuals of both sexes. Current individuals’ smoking status showed distinct sex-specific patterns in DLCO, AV, and KCO. Among men, DLCO, AV, and KCO were lower across all levels of exposure, with lower values observed in those who smoked ≥11cigarettes per day. Among women, DLCO and AV showed higher values with smoking exposure, while KCO was lower in daily smokers, particularly among those consuming ≥11cigarettes per day (Tables 2 and 3).

Tables 4 and 5 show the adjusted models. Early life exposures showed that, among men, maternal smoking during pregnancy was linked to lower DLCO (β=−0.96; 95%CI, −1.67, −0.25) and AV (β=−0.19; 95%CI, −0.33, −0.05) vs non-exposed. In women, low birth weight remained associated with lower DLCO (β=−1.21; 95%CI, −2.13, −0.29) and AV (β=−0.29; 95%CI, −0.47, −0.11) vs individuals with normal birth weight. Black individuals had lower AV in both sexes vs those related being whit. Among men, black skin color was also associated with higher KCO (β=0.29; 95%CI, 0.14, 0.44), while no significant association with KCO was found in women after adjustment.

Regarding late-life exposures, obesity was associated with lower AV and higher KCO in both sexes. Additionally, obese women showed higher DLCO vs those with normal BMI (β=0.99; 95%CI, 0.32, 1.66). Among men, current smokers had significantly lower DLCO, AV, and KCO vs never smokers. These associations were more pronounced among those smoking ≥11cigarettes per day: DLCO (β=−4.80, 95%CI, −6.01, −3.59), AV (β=−0.40; 95%CI, −0.64, −0.17), and KCO (β=−0.51, 95%CI, −0.70, −0.32). In female smokers, DLCO and AV were higher vs never smokers. However, KCO was significantly lower among those smoking ≥11cigarettes per day (β=−0.25; 95%CI, −0.49, −0.01) (Tables 4 and 5).

Sensitive analyses using pack-years to classify smoking exposure confirmed distinct sex-specific patterns. High smoking exposure (≥4.01 pack-years) was more common in men (Supplementary Table S3). Among men, pack-years were consistently associated with lower DLCO, AV, and mainly KCO across all exposure levels (Supplementary Tables S4 and S5). Among women, adjusted models showed that only those with high exposure (≥4.01 pack-years) had higher DLCO and AV, but lower KCO (β=−0.27; 95%CI, −0.47, −0.06) (Supplementary Tables S4 and S5).

This study investigated DLCO at the 22-year follow-up of the 1993 Pelotas Birth Cohort, to our knowledge, the first population-based study in Brazil to examine DLCO in early adulthood. Even in this young and healthy population, lower DLCO, AV, and KCO were associated with perinatal exposures, skin color, BMI, and smoking. Adequate diffusing capacity is essential for tissue oxygenation and carbon dioxide elimination, key processes for cardiovascular and body tissues [3]. Therefore, the assessment of alveolar-capillary diffusion extends beyond diagnosing lung disease, serving as an early indicator of systemic health [4].

Perinatal factors such as prematurity, low birth weight, and maternal smoking during pregnancy were associated with lower pulmonary diffusing capacity, in line with previous studies [5,18,19]. These associations likely reflect disruptions in lung development during the critical period between the 22nd and 32nd gestational weeks, when alveolar and capillary structures essential for gas exchange begin to form [20]. Lower birthweight has been associated with reduced adult diffusing capacity, with no consistent sex interactions reported previously. In our study, the association was observed only in women. Among men, although the direction of the association was similar, it was not statistically significant, and the magnitude of the association may be underestimated, possibly due to differential losses associated with birthweight (Supplementary Table S1).

Our results reinforce evidence of ethnic disparities in pulmonary function, as demonstrated in previous population-based studies [21,22]. Black individuals tend to present lower pulmonary function values for a given height, a pattern partly attributed to anthropometric differences, such as a lower trunk-to-limb length ratio. This characteristic may result in reduced thoracic dimensions and lung volumes [23], without necessarily implying impaired gas exchange efficiency [24,25]. Although the literature is inconsistent on whether this reflects actual enhanced efficiency, the combination of elevated KCO with reduced DLCO and AV may represent a compensatory physiological adaptation to lower average hemoglobin concentrations and greater dead space ventilation, optimizing gas-exchange efficiency per unit of VA [24]. On the other hand, this increase may reflect a mathematical artifact of reduced AV rather than a genuine physiological difference [22]. Despite socioeconomic-related losses being more pronounced among men, the pattern and strength of the associations were essentially preserved. This aligns with prior evidence indicating that individuals with higher socioeconomic status tend to have more favorable pulmonary outcomes [26].

BMI was associated with lower AV but higher DLCO and KCO values, consistent with patterns reported in previous studies [7,27]. These findings do not necessarily indicate enhanced gas exchange efficiency; instead, they likely reflect a compensatory mechanism involving increased pulmonary blood volume and vascular recruitment [31]. The elevated circulating blood volume commonly observed in obesity may further contribute to higher KCO [27].

Overall, performing at least 150min of leisure-time physical activity per week improved almost all parameters in both sexes, especially the DLCO in men and KCO in women. Exercise enhances oxygen uptake by increasing pulmonary capillary recruitment and distensibility, improving diffusing capacity, and supporting a higher oxygen consumption (VO2) peak [29]. As this relationship persists even after adjusting for body and lung size, regular physical activity likely contributes to better gas-exchange efficiency, helping explain our findings [29]. There is biological plausibility for a link between sleep duration and diffusing capacity, as adequate sleep contributes to autonomic regulation, inflammatory balance, and cellular repair [30,31], all of which contribute to maintaining alveolar–capillary function [22]. However, we did not observe this association in our analysis.

Cigarette smoking was associated with alterations in lung diffusing capacity, and our findings align with previous evidence showing smoking-related reductions in pulmonary diffusion capacity [2,6,12,32]. The lower diffusion capacity observed among smokers likely reflects early impairment of the alveolar–capillary membrane and reduced gas-exchange efficiency [33]. These changes are consistent with the known biological effects of cigarette smoke—chronic inflammation, oxidative stress, and profibrotic mechanisms that promote alveolar destruction, loss of lung elasticity, and fibrotic remodeling [34,35]. In addition, smoking may further reduce measured DLCO because increased carboxyhemoglobin (COHb) competes with the tracer gas for hemoglobin binding, reducing the adequate capacity for gas transfer [33,36]. In our study, DLCO values were not corrected for hemoglobin or COHb levels because these measurements were unavailable at the 22-year follow-up. This limitation could result in a slight underestimation of diffusion capacity, particularly among individuals with anemia or recent carbon monoxide exposure (e.g., smokers) [22]. However, as all analyses were adjusted for the time since the last cigarette, any bias related to transient COHb elevation is likely to be minimal [37].

About sex differences, male smokers exhibit lower DLCO, AV, and KCO values, which aligns with the expected effects of tobacco exposure on diffusion capacity [2,6,12,32]. Among female smokers, DLCO and AV are higher vs never smokers, a pattern that is not expected. Even so, smoking still appeared to impair diffusion efficiency in women: daily cigarette consumption was associated with lower KCO, even after adjustment for lung volume. Sensitivity analyses confirmed lower gas-exchange efficiency, showing that women with ≥4.01 pack-years had lower KCO values (Supplementary Table S5). Former studies have reported sex-specific differences in the respiratory effects of smoking [6,12,38]. Impairments tend to be more pronounced in men. At the same time, in women, the changes are generally subtler and more closely related to cumulative exposure [38], likely reflecting differences in smoking patterns and lifetime exposure between the sexes. As secondary hypotheses, we considered alternative explanations, including hormonal influences, such as menstrual-cycle fluctuations, which are known to affect DLCO among women of reproductive age [39]. Of note, sex-related differences were most pronounced in women for KCO. This reinforces the need to interpret DLCO along with AV and KCO, since increases in AV may mask underlying impairments in gas-exchange efficiency [4,22]. These findings indicate a deleterious impact of smoking on diffusion capacity [2,6,12,32], even when DLCO alone does not appear reduced.

Our study has several strengths. We used data from a large, population-based birth cohort with standardized protocols and robust follow-up, supporting a broader applicability of the conclusions. On the other hand, this study has limitations that should be considered when interpreting the results: (1) Participant differences: despite differential losses related to socioeconomic status and birthweight among men, the direction and magnitude of the associations remained unchanged; however, these associations can be underestimated. (2) DLCO measurements: Values were not corrected for hemoglobin or COHb, which were not measured at 22 years. This fact slightly underestimates diffusion capacity, particularly in anemic individuals or recent smokers [33]. Given the young, generally healthy profile of the cohort and the adjustment for time since the last cigarette, the risk of bias is likely to be low. (3) Lung volume measurement: Because TLC reflects both alveolar volume and residual volume (dead space), relying solely on AV omits important information [40]. Increases in RV—common in smokers (air trapping) and in individuals with obesity (restricted inspiratory expansion)—can reduce AV and consequently affect DLCO and KCO. As a result, our estimates may underestimate the true magnitude of the associations, and this limitation should be taken into consideration when interpreting the findings.

In this population of young adults, our findings demonstrate that pulmonary diffusing capacity is influenced by early-life exposures, skin color, body mass index, and tobacco smoking. These results underscore the importance of assessing DLCO, KCO, and AV together, even in asymptomatic individuals, as early alterations may reflect subclinical impairments and signal increased vulnerability to long-term respiratory and systemic diseases. As an indirect measure of gas exchange, DLCO reflects not only pulmonary function but also the integration of ventilation, perfusion, and systemic health. Therefore, it may serve as a sensitive early marker of susceptibility to pulmonary and systemic damage, helping to identify individuals who could benefit from targeted preventive strategies.

However, its use as a population-based screening tool may not be cost-effective, as feasibility and resource availability vary across settings. Instead, early DLCO assessment may be more appropriate in selected vulnerable groups, such as smokers or individuals with obesity, who are more likely to present subclinical alterations in alveolar-capillary function. From a public health perspective, focusing on interventions that modify exposures at this stage, particularly smoking cessation, weight control, and improvements in perinatal health, may represent a more effective strategy to reduce the future burden of respiratory diseases, including COPD.

L. Picanço: Contributed to the study design, data analysis, drafting of the manuscript, and critical revision for important intellectual content. A.M.B. Menezes: Contributed to data management, study coordination, and critical revision of the manuscript. P.D. de Oliveira: Responsible for data handling, project supervision, and manuscript editing. P. Weber: Contributed to the critical review and editing of the manuscript. R. Perez-Padilla: Provided intellectual input during manuscript review and editing. F.C. Wehrmeister: Involved in study conception, data supervision, project coordination, statistical analysis, manuscript drafting, and critical revision.

The study protocol for the 22-year follow-up was approved by the Ethics and Research Committee of the School of Medicine, Federal University of Pelotas (protocol No. 1.250.366). All procedures involving human participants were conducted in accordance with the ethical standards of the institutional and national research committees and with the principles of the Declaration of Helsinki and its subsequent amendments. Written informed consent was obtained from all participants before each assessment.

Artificial intelligence tools (ChatGPT, OpenAI) were used solely to assist with English grammar, spelling, and language clarity. The authors reviewed and approved all content.

This study used data from the 1993 Pelotas Birth Cohort, conducted by the Postgraduate Program in Epidemiology at the Federal University of Pelotas. The 22-year follow-up was funded by the Department of Science and Technology (DECIT), Brazilian Ministry of Health. Throughout the years, the cohort has received support from multiple centers, including the Coordination for the Improvement of Higher Education Personnel (CAPES), Wellcome Trust (supporting capacity building in low- and middle-income countries), National Program for Centers of Excellence (PRONEX), National Council for Scientific and Technological Development (CNPq), Research Support Foundation of the State of Rio Grande do Sul (FAPERGS), Brazilian Ministry of Health, and the Brazilian Association of Collective Health (ABRASCO).

None declared.

Data from the 1993 Pelotas Birth Cohort are not publicly available due to ethical restrictions and data protection regulations. Requests for access may be considered on a case-by-case basis upon reasonable request to the corresponding author.

[14,28].