The research, led by Florian Wyler and colleagues at Bern University Hospital and the University of Bern in Switzerland, tackles a problem that has long frustrated clinicians and researchers alike. The main outputs of a washout test are the lung clearance index, or LCI, which measures how many lung turnovers are needed to clear the resident nitrogen, and the functional residual capacity, or FRC, the volume of air remaining in the lungs at the end of a normal breath out. Irregular breathing behaviors can bias both numbers, but the current gold standard for judging whether a measurement is trustworthy is expert quality control: a trained physician, nurse, or scientist visually inspects the signals and decides whether to accept or reject the test. That judgment is slow, requires rare expertise, and, crucially, is subjective.

The subjectivity is not a minor nuisance. Washout measurements typically last several minutes and are routinely performed in awake children from preschool age onward, so measurements full of minor, ambiguous irregularities are the norm rather than the exception. Two experienced raters can disagree about whether a given test should be kept, especially when the irregularities fall into a grey zone between clearly acceptable and clearly unusable. Existing guidelines also rely on fixed empirical thresholds, such as allowable variation in end-expiratory lung volume, without any direct way of estimating what those deviations actually do to the LCI or FRC. The Bern team set out to replace that guesswork with a quantitative, reproducible estimate of impact.

The core of the new approach is a mathematical model of how nitrogen is diluted during the washout. For any given breath, the model relates an input of breathing behavior to an expected output of nitrogen dilution, taking into account the current end-expiratory lung volume, physiological and machine dead spaces, and the fact that washout efficiency declines as the test proceeds, because well-ventilated lung regions empty first while poorly ventilated regions hold on to their nitrogen longer. The model’s parameters are estimated individually from each measurement, using the carbon dioxide and flow signals to locate dead space and fitting a linear efficiency curve over the course of the washout.

With this model in place, the algorithm runs paired simulations for each measurement. One simulation uses the observed breathing behavior, including whatever irregularities occurred; the other uses an idealized version in which the behavior in question, say a drifting end-expiratory lung volume, is replaced by a stable target. If the two simulations produce nearly identical LCI and FRC values, the irregularity was harmless. If they diverge, the size of the divergence, expressed as a percentage, becomes a behavior-specific quality control outcome. The team built seven such tests, covering changes in end-expiratory lung volume, tidal volume variability and size, leaks or trapped gas, the method used to determine end-tidal nitrogen concentration, signal synchronization, and breath detection. A measurement was deemed acceptable if no single test exceeded an estimated impact of 15 percent.

To validate the tool, the researchers re-analyzed an extraordinary dataset: 2,471 nitrogen washout measurements from 84 children with cystic fibrosis, collected between 2012 and 2019 across 754 visits, each visit including at least three completed measurements. The algorithm ran successfully on every single measurement without any manual intervention, taking roughly four seconds per test on an ordinary office computer. That speed matters, because expert quality control is one of the practical barriers to wider adoption of washout testing in clinical routine. The overall acceptance rate was 76 percent of visits, and the mean per-visit variance of LCI dropped from 0.72 turnovers when all measurements were included to 0.39 turnovers when only algorithm-accepted measurements were counted.

The most striking finding is how much of the natural variability in washout results can be attributed to quality control factors. The combined impact estimate from all tests explained around 45 percent of the within-visit variability of LCI in the children with cystic fibrosis, and the median Pearson correlation between estimated impact and observed within-visit LCI variation was 0.83. The largest single effect on LCI came from leaks or trapped gas, which exceeded the acceptability threshold in 31 percent of measurements, followed by changes in end-expiratory lung volume at 23 percent and variations in breath size around the end of the test at 19 percent. Leaks and trapped gas are particularly common in cystic fibrosis because the disease produces extended washout tails, in which the end-tidal nitrogen concentration hovers near the end-of-test threshold and occasionally ticks upward as poorly ventilated pockets of gas are slowly released.

The team also compared the automated judgments with human experts using a second dataset of 100 measurements from six-year-old children, half with cystic fibrosis and half healthy, drawn from the Bern and Swiss infant lung development cohorts. Agreement between the algorithm and expert raters was moderate, with Cohen’s kappa ranging from 0.32 against the rating made on the day of the test to 0.46 against a retrospective rating. Notably, that level of agreement was comparable to the agreement between two experienced human raters themselves, which stood at 0.65. Some disagreements were instructive: current guidelines instruct raters to reject any measurement with a leak, no matter how small, whereas the automated system tolerates minor leaks as long as their estimated impact on outcomes remains negligible. Conversely, experts sometimes struggled to judge visually whether a transient shift in end-expiratory lung volume mattered, while the model could quantify it directly.

The implications extend beyond cystic fibrosis clinics. Because the algorithm judges every irregularity on the same scale, its estimated impact on the primary outcomes, it replaces a patchwork of device-specific and behavior-specific thresholds with a single interpretable metric. The threshold itself can be tuned to the purpose of the test: a strict 15 percent cutoff for research repeatability, or a looser criterion when a clinical LCI is already far above the upper limit of normal and a modest bias would not change the diagnosis. The tool can also be applied retrospectively when measurements are re-analyzed with updated algorithms, and it inherently distinguishes between signal anomalies caused by genuine leaks and those fully explained by the patient’s underlying breathing pattern, a distinction that is especially valuable in patients who cannot achieve high-quality tidal breathing.

Limitations remain. The algorithm was developed for nitrogen washout data from the Ecomedics Exhalyzer D and relies on carbon dioxide, tracer gas, and flow signals; some tests are specific to that device setup, though the authors note that only minor adaptations would be needed for sulfur hexafluoride measurements or other devices. The current implementation runs after the measurement is complete, so a live version that warns operators mid-test to restart a compromised measurement would require further development. Still, the study demonstrates a feasible, fast, and reproducible method for assigning impact estimates to breathing irregularities, and it explains a substantial share of the measurement noise that has long plagued washout testing. For a technique whose value depends on detecting subtle changes in ventilation inhomogeneity over years of follow-up, an objective quality control standard could make the difference between a noisy research tool and a dependable clinical instrument.

Subject of Research: Automated model-based quality control for multiple-breath washout lung function testing in children

Article Title: Quality control in multiple-breath washout measurements: an automated, model-based approach quantifies effects on primary outcomes

Article References: Wyler, F., Borer, S., Curdy, M., Bovermann, X., Frauchiger, B. S., & Latzin, P. (2026). Quality control in multiple-breath washout measurements: an automated, model-based approach quantifies effects on primary outcomes. Frontiers in Medical Engineering, 4, Article 1774073. https://doi.org/10.3389/fmede.2026.1774073

Image Credits: AI Generated

DOI: 10.3389/fmede.2026.1774073

Keywords: multiple-breath washout, lung clearance index, quality control, cystic fibrosis, functional residual capacity, pediatric lung function, nitrogen washout, ventilation inhomogeneity, automated algorithm, respiratory medicine, clinical engineering, Bern University Hospital

The research team, led by Walid Al-Qerem and colleagues, framed their work around a persistent problem in respiratory medicine. Reference equations developed in one population or one age range often fail to transport cleanly to another. Growth patterns, body proportions, environmental exposures, and genetic backgrounds all shape how lung function scales with body size, and distributional models that capture nonlinear growth in one group may be systematically biased in another. At the same time, an open methodological question has hovered over the field: do modern machine-learning algorithms, with their capacity to capture intricate interactions, genuinely improve prediction beyond simpler regression models built on well-chosen transformations of the data?

To answer that question, the investigators designed a two-phase cross-sectional study comparing sex-specific predictive models for the three core spirometric outcomes: FEV1, forced vital capacity (FVC), and the ratio between them, FEV1/FVC, which is the standard marker of obstructive lung physiology. The first phase used the same derivation dataset of 1,576 children on which the original Jordanian reference equation, known as the Al-Qerem equation, had been built. That choice was deliberate. By training candidate models on identical data, the researchers could compare predictive modeling strategies on equal footing, isolating the effect of model class from the effect of the underlying sample.

The candidate models spanned a spectrum of complexity. On the simpler end sat generalized linear models, or GLMs, which predict outcomes through straightforward mathematical relationships after appropriate transformations. On the more complex end sat gradient boosting machines, or GBMs, an ensemble machine-learning technique that builds predictions by sequentially combining many weak learners, each correcting the errors of its predecessors. GBMs can, in principle, capture nonlinearities and interactions that a GLM would miss, but they also carry a risk of overfitting, learning quirks of the training sample that do not generalize. The study assigned the best-performing model class separately for each outcome and sex: GBMs for FEV1 in both sexes, GLMs for FVC in both sexes, a GLM for FEV1/FVC in girls, and a GBM for FEV1/FVC in boys.

A crucial technical detail shaped the evaluation. Because several models were trained on log-transformed outcomes, the researchers assessed performance on the original measurement scale, after back-transformation of predictions. This matters more than it might appear. Error metrics computed on a log scale can obscure clinically meaningful biases in absolute volume units, and a model that looks sharp in transformed space may perform unevenly where it counts, at the extremes of the distribution where clinical decisions are made. By evaluating on the natural scale, the team kept the comparison anchored to the quantities clinicians actually use.

The second phase provided the real test. The models, alongside established reference equations including the Global Lung Function Initiative 2012 equations, the updated GLI 2022 equations, and the original Al-Qerem equation, were evaluated in an external validation sample of 1,007 healthy Jordanian children aged 6 to 18 years. External validation, testing on data not used in model development, is the gold standard for judging whether a predictive tool is ready for real-world use, and it is precisely where many promising models stumble.

The results revealed a clear hierarchy of predictability. FEV1 and FVC were predicted substantially more accurately than FEV1/FVC across every model class and every established equation. The explained variance of the ratio, based on age and height alone, remained low regardless of the modeling approach. This finding has physiological logic: the ratio is designed to be relatively independent of body size, so the very variables that powerfully predict absolute lung volumes carry less information about the proportion between them. For clinicians, it underscores that FEV1/FVC reference limits rest on a narrower statistical foundation than the volume measures, a point often underappreciated in routine reporting.

In external validation, the picture grew more nuanced. The study’s own model achieved the lowest mean squared error for female FEV1 and FVC, a genuine win for the comparative modeling framework in girls. But it did not consistently outperform the GLI 2012, GLI 2022, or Al-Qerem equations in boys, nor for FEV1/FVC in either sex. Complexity, in other words, bought no blanket advantage. The best-performing model differed by outcome and by sex, and carefully chosen transformations and calibration steps mattered more than the sophistication of the algorithm itself. A gradient boosting machine trained carelessly could lose to a well-calibrated linear model, and often did.

One of the most clinically consequential findings emerged from age-stratified analyses. When the equations were applied to boys younger than 10 years, all four reference standards, the study model, the Al-Qerem equation, GLI 2012, and GLI 2022, produced elevated rates of FEV1 and FVC values falling below the lower limit of normal, a threshold that typically flags potentially abnormal lung function. In a validation sample of healthy children, such elevations signal systematic miscalibration: the equations are, in effect, setting the bar too high for young boys in this population, risking false-positive classifications. By contrast, below-limit proportions for FEV1/FVC were generally closer to the nominal values expected from the statistical definition of the lower limit of normal, suggesting the ratio equations were better calibrated even where their predictive precision was weaker.

The authors are careful about what their findings do and do not license. Comparative predictive modeling, they conclude, is a useful development framework, a disciplined way to test candidate equations against one another before deployment. But the study does not demonstrate superiority of complex machine-learning methods, and it does not establish readiness for clinical use without further calibration and additional external validation. For pediatric respiratory practice in Jordan and comparable populations, the message is both sobering and constructive: the path to better lung function reference equations runs not through ever-more-elaborate algorithms, but through meticulous attention to transformation, calibration, population fit, and honest out-of-sample testing, with special vigilance for the youngest children in whom existing equations appear least reliable.

Subject of Research: Comparative predictive modeling of pediatric spirometry reference equations in Jordanian children

Article Title: Comparative predictive modeling of pediatric spirometry reference equations in Jordanian children: Complex versus simple models

Article References: Al-Qerem, W., Jarab, A., Eberhardt, J., Smairan, K., Mimi, Y., & Khdour, M. (2026). Comparative predictive modeling of pediatric spirometry reference equations in Jordanian children: Complex versus simple models. PLOS Digital Health, 5(9), e0001746. https://doi.org/10.1371/journal.pdig.0001746

Image Credits: AI Generated

DOI: 10.1371/journal.pdig.0001746

Keywords: spirometry, pediatric lung function, reference equations, machine learning, gradient boosting, FEV1, FVC, Jordan, GLI 2012, external validation, GAMLSS, predictive modeling

Spirometry, the familiar test in which a patient blows forcefully into a mouthpiece, measures how much air can be exhaled and how quickly. It is cheap, widely available and reproducible, which explains its dominance in both clinics and research studies. Yet it fundamentally measures airflow through large and medium airways. The injury pattern left by extreme preterm birth, often described as chronic lung disease of prematurity or bronchopulmonary dysplasia in its severe forms, is primarily one of disrupted alveolar and vascular development. The distal lung, where gas exchange actually happens, and the pulmonary circulation, which must carry the entire cardiac output through a finely branched vascular tree, are the structures most affected. Spirometry can remain deceptively normal in people whose gas transfer capacity and pulmonary vasculature are substantially impaired.

The evidence for this blind spot is accumulating rapidly. A study by Harris and colleagues, published in Pediatric Research, examined school-aged children born preterm and found that reduced diffusion of carbon monoxide across the lung, a marker of impaired gas exchange surface, was associated with elevated pulmonary artery pressures. This finding is significant because it links two silent abnormalities: a lung that cannot transfer oxygen efficiently and a heart that must pump against higher resistance in the pulmonary circulation. Neither abnormality would necessarily be flagged by a routine spirometry test, and both may progress quietly for decades before producing symptoms.

Longitudinal data reinforce the concern. Research published in the European Respiratory Journal followed individuals born extremely preterm into adulthood and documented persistent deficits in lung diffusion capacity that did not normalize with growth. A systematic review and meta-analysis by Gibbons and colleagues in the European Respiratory Review synthesized findings across lung volumes, gas transfer and oscillometry, a technique that measures airway resistance during quiet tidal breathing rather than forced expiration. Together, these studies sketch a consistent picture: preterm birth leaves a durable imprint on the distal lung and its mechanics that forced expiratory volumes capture only partially, if at all.

The clinical implications extend beyond the lungs. Survivors of extreme preterm birth appear to face elevated long-term risks of pulmonary vascular disease and cardiovascular complications, a concern articulated in a lifespan-oriented framework by Simpson and colleagues in The Lancet Respiratory Medicine. The developing lung and its vasculature are built in tandem during the canalicular and saccular stages of gestation, precisely the window that extreme preterm birth interrupts. When alveolarization is curtailed, the vascular tree that should accompany each new alveolus is curtailed with it. The result is a lung with fewer, larger airspaces and a reduced, remodeled vascular bed, a combination that increases the workload on the right ventricle and may set the stage for pulmonary hypertension and early cardiovascular decline in adult life.

What would a comprehensive assessment look like in practice? The commentary by Beaven and Finlay points toward a battery of complementary tests, each interrogating a different component of the cardiopulmonary system. Lung clearance index, derived from multiple-breath washout, quantifies ventilation inhomogeneity in the small airways and peripheral lung units, and has proven sensitive to early lung disease in other pediatric populations. Body plethysmography measures total lung volumes, revealing the restricted or hyperinflated patterns that spirometry alone can miss. The single-breath or rebreathing diffusing capacity for carbon monoxide directly probes the gas exchange surface. Oscillometry adds information about peripheral airway resistance without requiring maximal effort, an important advantage in young children who cannot reliably perform forced maneuvers.

Cardiovascular assessment deserves equal billing. Echocardiography can estimate pulmonary artery pressures and right ventricular function, while newer techniques such as cardiac magnetic resonance imaging offer precise quantification of right ventricular mass and ejection fraction. Exercise testing integrates the entire system, revealing whether oxygen delivery keeps pace with metabolic demand, and cardiopulmonary exercise testing can unmask limitations that are invisible at rest. Reduced exercise capacity has been documented in preterm-born cohorts, and the physiological basis, whether ventilatory limitation, gas exchange impairment, vascular limitation or deconditioning, can only be disentangled by measuring multiple variables simultaneously during exertion.

The concept of treatable traits offers a compelling rationale for this expanded toolkit. As Zanetto and colleagues argued recently in the American Journal of Respiratory and Critical Care Medicine, clinicians caring for preterm-born individuals with chronic lung disease of prematurity should identify and target specific, modifiable characteristics across the lifespan rather than relying on a single diagnostic label. A patient with preserved spirometry but reduced diffusion capacity and elevated pulmonary artery pressures needs a very different management plan from one with airflow obstruction and air trapping. Without comprehensive phenotyping, the first patient may be falsely reassured and the second may receive therapies that do not address their dominant abnormality. Treatable traits might include airway reactivity, ventilation inhomogeneity, impaired gas transfer, pulmonary vascular dysfunction, muscle weakness and physical deconditioning, each with its own therapeutic levers.

There are practical hurdles to overcome before comprehensive assessment becomes standard care. Many of the recommended tests require specialized equipment, trained personnel and cooperative patients, which limits their availability in ordinary clinical settings and in the low-resource environments where much of the world’s preterm births occur. Reference ranges for children and young adults born preterm are incomplete for several modalities, complicating interpretation. Radiation exposure argues against routine computed tomography, and the research community continues to seek safe, non-invasive imaging surrogates of alveolar and vascular structure. Standardizing protocols across centers and across the age spectrum, from infancy through adulthood, remains an unfinished project, and longitudinal cohorts that follow preterm-born individuals with repeated multi-modal assessments are still relatively rare.

Nevertheless, the direction of travel is clear. The generation born after the widespread adoption of surfactant therapy and gentle ventilation is now reaching adulthood in large numbers, and their long-term cardiopulmonary trajectory is becoming a matter of public health, not just neonatology. Detecting silent gas exchange impairment and early pulmonary vascular change during the school years, when interventions such as exercise programs, vaccination, smoking prevention and targeted follow-up can still alter the course, is far preferable to discovering these problems when symptomatic disease emerges in mid-life. The commentary by Beaven and Finlay makes the case that spirometry, whatever its virtues, cannot carry this burden alone. A comprehensive cardiopulmonary assessment, combining measures of lung volume, ventilation distribution, gas transfer, airway mechanics, pulmonary hemodynamics and exercise capacity, offers the best chance of understanding, monitoring and ultimately improving the lifelong health of the fastest-growing population of adult survivors of neonatal intensive care.

Subject of Research: Cardiopulmonary assessment of lung and cardiovascular outcomes in survivors of extreme preterm birth

Article Title: Beyond spirometry: towards comprehensive cardiopulmonary assessment after extreme preterm birth

Article References: Beaven, M. L., & Finlay, P. (2026). Beyond spirometry: towards comprehensive cardiopulmonary assessment after extreme preterm birth. Pediatric Research. https://doi.org/10.1038/s41390-026-05399-1

Image Credits: AI Generated

DOI: 10.1038/s41390-026-05399-1

Keywords: extreme preterm birth, bronchopulmonary dysplasia, spirometry, lung diffusion capacity, pulmonary hypertension, chronic lung disease of prematurity, lung clearance index, oscillometry, cardiopulmonary exercise testing, treatable traits, pediatric lung function, right ventricular function