Spirometry: read the curve, get the pattern right

The report arrives with three numbers highlighted and a ready-made sentence: mild restrictive ventilatory disorder. The patient is in their early sixties, has smoked for a long time and complains of shortness of breath when climbing stairs. The doctor accepts the report, thinks of interstitial disease and orders a CT scan. Nobody looked at the curve, nobody checked whether the maneuvers were good and nobody measured lung volumes. The word restrictive stayed in the chart for years.

Spirometry is an easy test to order and deceptively easy to read. The numbers look objective, but interpretation depends on the quality of the maneuvers, the reference equation, the limit used to define normal and the clinical context. Errors at any of these steps change the pattern and, with it, the diagnosis.

This text reviews how to check test quality, how to use the lower limit of normal, how to separate obstruction from restriction, what to do with the bronchodilator test and which pitfalls appear most often when reading the curve.

Key points

• Before interpreting, check whether there were acceptable and repeatable maneuvers.
• The lower limit of normal and the z-score are more correct than the 80 percent of predicted cutoff.
• The fixed 0.7 ratio can overdiagnose obstruction in older adults and underdiagnose it in young people (Agustí et al., 2023).
• Spirometry suggests restriction, but only total lung capacity confirms it.
• The pattern with a preserved ratio and reduced FEV1 (PRISm) is common and has prognostic meaning.
• The chosen reference equation can change the classification of many tests.

Before the curve: indication and quality

Spirometry is indicated to investigate lung disease in people with respiratory symptoms such as shortness of breath, wheezing, cough or chest tightness (Aggarwal et al., 2019). The Indian guideline does not recommend screening spirometry to diagnose COPD in people without symptoms, because many of the changes found this way do not call for intervention (Aggarwal et al., 2019). The clinical question, therefore, comes before the test.

What makes a maneuver acceptable

The 2019 standardization by the American Thoracic Society and the European Respiratory Society updated the acceptability and repeatability criteria and created a new quality grading system (Graham et al., 2019). The maneuver starts with maximal inspiration, followed by a blast and complete expiration, which may last up to 15 seconds (Graham et al., 2019). The system should signal when expiration reaches a plateau, defined as a volume change of up to 0.025 L for at least 1 second, or when the time reaches 15 seconds (Graham et al., 2019).

The start also matters. The back-extrapolated volume must be below 5 percent of the forced vital capacity or 0.100 L, whichever is greater, and the hesitation between maximal inspiration and the start of expiration should be 2 seconds or less (Graham et al., 2019). A large back-extrapolated volume usually produces a falsely high FEV1 (Graham et al., 2019). A maneuver in which the forced inspiratory vital capacity exceeds the FVC by more than 0.100 L or 5 percent of the FVC, whichever is greater, is also not acceptable, a sign that expiration did not start from maximal inspiration (Graham et al., 2019). Cough or glottic closure in the first second makes the FEV1 unacceptable (Graham et al., 2019).

Repeatability

The goal is to obtain at least three acceptable FEV1 and FVC measurements before and after the bronchodilator (Graham et al., 2019). Above 6 years of age, the difference between the two highest FVC values should be up to 0.150 L (Graham et al., 2019). The overlap of the flow-volume curves helps assess repeatability and detect submaximal efforts (Graham et al., 2019).

Most people can do the test well. Studies show that even very young children and older adults produce acceptable and repeatable tests when given good instructions (Haynes, 2018). A poor maneuver, on the other hand, can mimic disease: ending expiration early makes the FVC unacceptable (Graham and colleagues, 2019) and, if unnoticed, can suggest a restriction that does not exist.

Expiration time

In a study cited by the standardization, more than 95 out of 100 obstructed patients had a forced expiration time under 15 seconds, and more than 95 out of 100 normal people stayed under 11 seconds (Graham and colleagues, 2019). Some patients with restrictive disease and young people with strong elastic recoil empty their lungs quickly and may not sustain the plateau for 1 second (Graham and colleagues, 2019).

What normal means: lower limit and z-score

After confirming quality, the next step is to compare the result with what is expected for the person. Here is one of the most frequent errors in practice.

The lower limit of normal

The lower limit of normal corresponds to the 5th percentile of the healthy population, meaning a value at this limit appears in only 1 out of every 20 normal people (Haynes, 2018). In the British statement on pulmonary function tests, the z-score, which says how many standard deviations the value is from predicted, is the preferred method, and the lower limit of normal corresponds to a z-score of minus 1.645 (Sylvester and colleagues, 2020). The Indian guideline recommends using the statistical limit instead of fixed cutoffs (grade 1A) (Aggarwal and colleagues, 2019).

Why 80 of predicted misleads

A common approach is to consider any value below 80 out of 100 of predicted abnormal. Haynes calls this practice statistically invalid, because the true limit usually sits above 80 of predicted in young people and well below that in older adults (Haynes, 2018). The British statement shows that, in older people, the lower limit of FEV1 is below 80 of predicted, while for total lung capacity and for the FEV1/FVC ratio it is above (Sylvester and colleagues, 2020).

Using 80 of predicted and the 0.7 ratio as limits has already been shown to misclassify more than 20 out of 100 patients (Haynes, 2018). In older adults, the error tends to create disease; in young people, to hide it.

Pretest probability also counts

A value at the lower limit of normal should be read with the patient in front of you. In someone with symptoms and risk factors, a value at or below the limit suggests more chance of disease than of health; in someone with low probability, it calls for caution, because 1 in 20 healthy people will be at that point (Haynes, 2018). Likewise, normal values do not exclude disease when pretest probability is high and the numbers are near the limit (Haynes, 2018). The text on pretest probability goes deeper into this logic.

Obstruction: the ratio that decides

The obstructive pattern is defined by the FEV1/FVC ratio below the lower limit of normal, usually with FEV1 also low (Haynes, 2018). FVC can be normal in mild disease and fall in advanced disease (Haynes, 2018). On the flow-volume curve, obstruction appears with a lower peak flow and a concave expiratory limb, like a spoon (Haynes, 2018).

Fixed ratio or lower limit?

The GOLD 2023 report requires an FEV1/FVC ratio below 0.7 after the bronchodilator to confirm COPD (Agustí and colleagues, 2023). The report itself acknowledges that this fixed cutoff can underdiagnose young adults and overdiagnose older adults, especially in mild disease (Agustí and colleagues, 2023). Haynes describes the same problem (Haynes, 2018).

A Swedish study followed 984 people aged 65 to 100 for 6 years (Luoto et al., 2016). The incidence of flow limitation was 28.2 per 1000 person-years with the fixed ratio and 11.7 with the lower limit of normal (Luoto et al., 2016). Flow limitation by the lower limit was associated with more mortality, but limitation flagged only by the fixed ratio, in people normal by the lower limit, was not associated (Luoto et al., 2016).

One exam is not enough

GOLD recommends repeating spirometry on another occasion when the post-bronchodilator ratio falls between 0.60 and 0.80, because it can change with biological variation (Agustí et al., 2023). When the initial ratio is below 0.6, it is very unlikely to rise above 0.7 (Agustí et al., 2023).

A Russian study shows how much this matters. Of the 102 people with post-bronchodilator obstruction at the first assessment, only 60.8 out of every 100 kept the obstruction at the second exam (Andreeva et al., 2017). Among 65 people with obstruction only before the bronchodilator, 23.1 out of every 100 had post-bronchodilator obstruction at the second assessment (Andreeva et al., 2017). A single abnormal exam does not close the diagnosis, and the text on overdiagnosis discusses the cost of labeling too early.

Asthma and spirometry

In asthma, spirometry has a double role. The Indian guideline cites a study in which asthma went undiagnosed in 21 out of every 100 people who sought care when spirometry was not used (Aggarwal et al., 2019). At the same time, normal spirometry does not rule out asthma (Aggarwal et al., 2019). Some people with asthma, such as older adults and those with long-standing or severe disease, notice few symptoms despite a significant drop in FEV1 (Aggarwal et al., 2019).

Restriction that is not restriction

Spirometry suggests restriction when FEV1 and FVC are below the lower limit and the FEV1/FVC ratio is preserved (Haynes, 2018). On the flow-volume curve, the loop becomes narrow, because the volumes are small (Haynes, 2018). But spirometry does not measure total lung capacity.

Total lung capacity confirms

Haynes is clear: restriction must be confirmed by measuring lung volumes, with total lung capacity below the lower limit (Haynes, 2018). A restrictive pattern on spirometry is also not specific to interstitial disease (Aggarwal et al., 2019). On the other hand, normal spirometry has good negative predictive value: only 2.4 out of every 100 of 1361 patients with normal vital capacity had restriction on measurement of total lung capacity (Aggarwal et al., 2019).

When vital capacity and the FEV1/FVC ratio are both reduced, the pattern may be obstructive or mixed, and total lung capacity is needed to separate the two (Aggarwal et al., 2019). In the opening scene, a low FVC in a smoker may reflect advanced obstruction, in which FVC also falls (Haynes, 2018), or a short maneuver, and not interstitial disease.

PRISm: preserved ratio, reduced FEV1

The pattern called PRISm describes people with an FEV1/FVC ratio of 0.7 or more and FEV1 below 80 percent of predicted after the bronchodilator (Agustí et al., 2023). The GOLD report cites prevalence between 7.1 and 20.3 out of every 100, with more cases in smokers and former smokers, and notes that PRISm can progress to normal, obstructive or restrictive spirometry (Agustí et al., 2023).

In the COPDGene study, with 10,192 smokers and former smokers, PRISm appeared in 12.3 out of every 100 (Wan et al., 2014). Compared with people with normal spirometry, people with PRISm had more shortness of breath, more emphysema and lower total lung capacity (Wan et al., 2014). In a Chinese study with 1439 participants, total lung capacity measured on CT was lower in PRISm, 4.15 versus 4.78 L, and there was more small airway dysfunction (Zhao et al., 2022).

Why PRISm matters

PRISm is not a neutral finding. In 3032 Japanese people followed for a median of 5.3 years, it was associated with a higher risk of death from any cause, with a hazard ratio of 2.20, of cardiovascular death, with 4.07, and of developing flow limitation, with 2.48 (Washio and colleagues, 2022). In the Japanese follow-up, 131 participants died and 218 developed flow limitation (Washio and colleagues, 2022). In the United States, NHANES data showed a prevalence of COPD between 13.1 and 14.3 per 100 and of PRISm between 9.6 and 10.2 per 100, with a hazard ratio of 2.3 for death from any cause in PRISm (Cadham and colleagues, 2024). In that survey, PRISm was more frequent in women, higher in middle age and notably higher among non-Hispanic Black people, unlike COPD, which increased with age and was more common in men (Cadham and colleagues, 2024). The profile of someone with PRISm, therefore, does not always resemble the typical COPD patient.

Bronchodilator and the flow-volume curve

The bronchodilator test is done by repeating spirometry after a short-acting bronchodilator (Haynes, 2018). Haynes recommends waiting at least 10 minutes between administration and the new test (Haynes, 2018).

How to define response

Haynes and the Indian guideline use as a significant response an increase in FEV1 or FVC of 12 per 100 and 200 mL relative to baseline (Haynes, 2018) (Aggarwal and colleagues, 2019). The decision to withhold bronchodilators before the test is clinical: when the goal is to diagnose a disease, withholding helps; when it is to assess the response to treatment in use, they are generally not withheld (Graham and colleagues, 2019).

GOLD 2023 notes that post-bronchodilator spirometry is needed to diagnose and assess COPD, but that assessing the degree of reversibility to decide treatment is no longer recommended (Agustí and colleagues, 2023). A large response to a bronchodilator, therefore, does not by itself separate asthma from COPD.

The shape of the curve

The flow-volume curve shows what the numbers hide. In obstruction, the expiratory loop becomes concave; in restriction, narrow (Haynes, 2018). A loop with a typical flattened shape may suggest central airway obstruction, according to the Indian guideline (Aggarwal and colleagues, 2019). A report without the curve, therefore, loses information.

The same guideline does not recommend using intermediate flows, such as FEF 25 to 75, for interpretation, because they add no advantage to the main parameters (Aggarwal and colleagues, 2019). Many reports still highlight small airways based on this isolated number.

Severity is not diagnosis

After defining the pattern, the report usually grades severity. The Indian guideline proposes classifying the disorder, obstructive or restrictive, as mild, moderate or severe when FEV1 is at 70 or more, between 50 and 69 or below 50 per 100 of predicted (Aggarwal and colleagues, 2019). This grading describes the size of the change, but does not say which disease it is. A very low FEV1 may come from COPD, from poorly controlled asthma or from a poor maneuver, and only the combination of quality, pattern, curve and clinical picture answers the question.

Reference equations, race and common errors

The predicted value depends on the equation chosen. For years, laboratories used equations specific to each racial or ethnic group. The American Thoracic Society began recommending average equations, neutral as to race, and names GLI Global as an option (Bhakta and colleagues, 2023).

What changes in practice

Changing the equation changes the classification of many tests. At an American academic center, with 2722 Black people and 5709 white people, the neutral equation increased restriction among Black people from 26.8 to 37.5 per 100 and reduced it among white people from 22.6 to 18.0 per 100, with no significant change in obstruction (Moffett and colleagues, 2023). At the Mayo Clinic, applying GLI Global to 109,447 tests changed the interpretation category in 10.2 per 100 of the tests in a population with equal representation of the groups (Kanj and colleagues, 2024). Obstruction increased 9.7 per 100 in white people, and possible restriction increased 51.1 and 37.1 per 100 in Black and Southeast Asian people, respectively (Kanj and colleagues, 2024).

The rationale from the American Thoracic Society is that race and ethnicity do not capture individual variation in lung function well and that specific equations can mask the effects of social and environmental factors (Bhakta and colleagues, 2023). GLI Global is a weighted average of the data from the original GLI ethnic equations (Bhakta and colleagues, 2023). The same statement acknowledges concerns about possible harms of the change and calls for education and policies to protect patients (Bhakta and colleagues, 2023).

For those who read the report, the message is to know which equation was used and to interpret values near the limit with the clinical context. The table below summarizes the main points for reading.

Step	What to check	Common error
Quality	Three acceptable maneuvers, repeatability of 0.150 L, plateau or 15 seconds (Graham and colleagues, 2019)	Interpreting a short maneuver as restriction
Limit of normal	Lower limit of normal or z-score of minus 1.645 (Sylvester and colleagues, 2020)	Using 80 of predicted as the limit
Obstruction	FEV1/FVC ratio below the limit; repeat if between 0.60 and 0.80 (Agustí and colleagues, 2023)	Labeling an older adult as COPD based only on the fixed ratio
Restriction	Total lung capacity below the limit	Calling it restriction without measuring volumes
PRISm	Preserved ratio with reduced FEV1	Ignoring the finding as normal
Bronchodilator	Increase of 12 per 100 and 200 mL (Haynes, 2018)	Using reversibility to separate asthma from COPD
Equation	Which reference equation was used	Comparing tests done with different equations

The numbers in the table come from the 2019 standardization, the Haynes review, the British statement and GOLD (Graham and colleagues, 2019) (Haynes, 2018) (Sylvester and colleagues, 2020) (Agustí and colleagues, 2023).

Errors that repeat

• Interpreting without checking acceptability and repeatability (Graham and colleagues, 2019).
• Using 80 of predicted as the limit of normal (Haynes, 2018).
• Diagnosing COPD in an older adult from a single test based on the fixed ratio (Luoto and colleagues, 2016).
• Calling a pattern restriction without a measured total lung capacity (Haynes, 2018).
• Treating PRISm as normal (Washio and colleagues, 2022).
• Ignoring the reference equation used in the report (Kanj and colleagues, 2024).

The text on clinical decision-making discusses how each test should answer a clear clinical question, and the text on dupilumab in COPD with type 2 inflammation shows how spirometry fits into the follow-up of people who already have a diagnosis.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of use of the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page. Training differential diagnosis with clues and searching for the right ICD-10 code complete this practice.

Frequently asked questions

Can I use 80 of predicted as the limit of normal?

It is not ideal. The lower limit of normal, at the 5th percentile, is more correct, because the cutoff of 80 of predicted is above the real limit in older adults and below it in young people, which can misclassify more than 20 per 100 of patients (Haynes, 2018).

Does an FEV1/FVC ratio below the fixed cutoff always indicate COPD?

No. GOLD uses this cutoff after the bronchodilator, but it acknowledges that it can overdiagnose older adults (Agustí et al., 2023). In older Swedish adults, the limitation identified only by the fixed ratio was not associated with higher mortality (Luoto et al., 2016).

Can spirometry confirm restriction?

No. It can suggest restriction, but confirmation requires total lung capacity below the lower limit of normal (Haynes, 2018).

What is PRISm?

It is the pattern with a preserved FEV1/FVC ratio and a reduced FEV1 (Agustí et al., 2023). It has been associated with higher mortality and a greater risk of progressing to airflow limitation (Washio et al., 2022).

Does the bronchodilator test separate asthma from COPD?

Not on its own. A significant response is defined as an increase of 12 in every 100 and 200 mL (Haynes, 2018), but GOLD does not recommend using the degree of reversibility to decide treatment (Agustí et al., 2023).

Take-home messages

• Check the quality of the maneuvers before reading any number.
• Use the lower limit of normal or the z-score, not 80 of predicted.
• The fixed ratio cutoff in older adults can create disease; confirm with repetition and context.
• Restriction is only confirmed with total lung capacity.
• PRISm is common and has prognostic meaning.
• Know which reference equation the report used.

References

1. Aggarwal AN, Agarwal R, Dhooria S, et al. Joint Indian Chest Society-National College of Chest Physicians (India) guidelines for spirometry. Lung India. 2019. PMID 31006703.
2. Agustí A, Celli BR, Criner GJ, et al. Global Initiative for Chronic Obstructive Lung Disease 2023 Report: GOLD Executive Summary. Eur Respir J. 2023. PMID 36858443.
3. Andreeva E, Pokhaznikova M, Lebedev A, et al. Spirometry is not enough to diagnose COPD in epidemiological studies: a follow-up study. NPJ Prim Care Respir Med. 2017. PMID 29138407.
4. Bhakta NR, Bime C, Kaminsky DA, et al. Race and Ethnicity in Pulmonary Function Test Interpretation: An Official American Thoracic Society Statement. Am J Respir Crit Care Med. 2023. PMID 36973004.
5. Cadham CJ, Oh H, Han MK, et al. The prevalence and mortality risks of PRISm and COPD in the United States from NHANES 2007-2012. Respir Res. 2024. PMID 38750492.
6. Graham BL, Steenbruggen I, Miller MR, et al. Standardization of Spirometry 2019 Update. An Official American Thoracic Society and European Respiratory Society Technical Statement. Am J Respir Crit Care Med. 2019. PMID 31613151.
7. Haynes JM. Basic spirometry testing and interpretation for the primary care provider. Can J Respir Ther. 2018. PMID 31164790.
8. Kanj AN, Scanlon PD, Yadav H, et al. Application of Global Lung Function Initiative Global Spirometry Reference Equations across a Large, Multicenter Pulmonary Function Lab Population. Am J Respir Crit Care Med. 2024. PMID 37523681.
9. Luoto JA, Elmståhl S, Wollmer P, et al. Incidence of airflow limitation in subjects 65-100 years of age. Eur Respir J. 2016. PMID 26677939.
10. Moffett AT, Bowerman C, Stanojevic S, et al. Global, Race-Neutral Reference Equations and Pulmonary Function Test Interpretation. JAMA Netw Open. 2023. PMID 37261830.
11. Sylvester KP, Clayton N, Cliff I, et al. ARTP statement on pulmonary function testing 2020. BMJ Open Respir Res. 2020. PMID 32631927.
12. Wan ES, Castaldi PJ, Cho MH, et al. Epidemiology, genetics, and subtyping of preserved ratio impaired spirometry (PRISm) in COPDGene. Respir Res. 2014. PMID 25096860.
13. Washio Y, Sakata S, Fukuyama S, et al. Risks of Mortality and Airflow Limitation in Japanese Individuals with Preserved Ratio Impaired Spirometry. Am J Respir Crit Care Med. 2022. PMID 35549659.
14. Zhao N, Wu F, Peng J, et al. Preserved ratio impaired spirometry is associated with small airway dysfunction and reduced total lung capacity. Respir Res. 2022. PMID 36316732.