14.1 Spirometry Calculations & Reference Equations

Domain III.A: why calculations are scored separately from procedures

On the NBRC PFT Examination (high cut / RPFT), Domain III — Data Management includes III.A Calculate results and select reference values. Task groups III.A.3 a/b focus on spirometry: deriving indices from acceptable maneuvers and placing those indices against the correct reference framework. Domain II already decided which efforts were valid; Domain III asks whether you can compute, correct, and compare the numbers that clinicians read.

This section is the math and reference layer for forced spirometry—not coaching (Chapter 8) and not full clinical diagnosis essays. Exam stems often hand you raw or partially processed values and ask for %predicted, ratio, LLN interpretation of a flag, or which equation set is appropriate.

Primary spirometric indices (what you calculate and report)

Related values you may also select or verify: FEV6, FEV1/FEV6, FET (forced expiratory time), back-extrapolated volume flags, and pre/post bronchodilator absolute and percent changes (response math is often tied to II.C/III but uses the same %change logic: change = post − pre; %change = (post − pre)/pre × 100).

Worked mini-example — ratio and %predicted

Measured (BTPS): FEV1 = 2.40 L, FVC = 3.20 L. Predicted FEV1 = 3.00 L; predicted FVC = 3.75 L.

1. FEV1/FVC = 2.40 / 3.20 = 0.75 (75%).
2. FEV1 %predicted = (2.40 / 3.00) × 100 = 80%.
3. FVC %predicted = (3.20 / 3.75) × 100 = 85%.

Exam trap: mixing ratio as percent (75%) with percent predicted (80%). They answer different questions—obstruction pattern vs size relative to reference population.

FEF25–75 and PEF concepts

• FEF25–75 is flow averaged over the middle half of FVC. It is more effort-dependent at the edges of the maneuver quality and more variable than FEV1; technologists still report it when the system calculates it from an acceptable curve, but over-interpreting a lone low FEF25–75 without context is a clinical pitfall (your job is correct calculation and QC, not inventing a diagnosis).
• PEF is the maximum instantaneous flow. It is highly effort-dependent; a blunt peak on the flow–volume loop often means soft blast, not necessarily “true low PEF physiology.”

BTPS correction concept (III.A calculation literacy)

Spirometers measure gas near ambient temperature and pressure, saturated (ATPS) or under conditions the sensor firmware assumes. Lung volumes are conventionally reported at body temperature and pressure, saturated (BTPS) — approximately 37 °C, ambient barometric pressure, water vapor pressure ~47 mmHg at body temperature.

Why it matters: Gas volume expands when warmed from room temperature to body temperature. Without BTPS correction, reported FVC/FEV1 would be systematically low relative to physiologic volumes and to reference equations built on BTPS values.

Technologist responsibilities (concept level):

• Confirm the system is configured with correct ambient temperature, barometric pressure, and humidity inputs (or automatic sensors) so the BTPS factor is right.
• Do not manually “invent” a BTPS multiplier mid-report unless your IFU and medical director define a verified procedure—modern systems apply BTPS in software.
• Recognize that calibration syringe strokes are often evaluated at ATPS with temperature considerations per standards; patient results are reported BTPS. Confusing calibration display units with patient BTPS units is a classic Domain I/III interface error.

Conceptual BTPS idea (not a proprietary formula to memorize digit-for-digit): volume at BTPS is larger than the same quantity of gas measured cooler at ATPS; the correction depends on Tambient, Pbar, and water vapor assumptions. If ambient temperature is entered too high, the software may under-correct (or mis-correct) relative to true conditions—enter real lab conditions.

Percent predicted: the universal reporting math

$$\text{percent predicted} = \frac{\text{measured}}{\text{predicted}} \times 100$$

Predicted comes from a reference equation (or look-up) using demographics: typically age, sex, height (and ethnicity/ancestry variables depending on the equation set). Weight is not the primary predictor for spirometry the way height is—stature drives predicted lung size.

Worked mini-example — selecting the numerator

Always use the reported best values after Domain II validity rules (e.g., largest FVC and largest FEV1 from acceptable curves, even if not from the same maneuver when standards allow—know your lab’s ATS/ERS-aligned selection rules). Do not average a failed cough maneuver into the numerator.

If best FEV1 = 1.85 L and predicted = 2.50 L → %pred = 74%.

Reference equations: GLI-2012 and modern multi-ethnic sets

What “selecting reference values” means for the RPFT

III.A is not only arithmetic. You must select which predicted set the lab applies and ensure demographics entered match that set:

• Correct height (cm or in consistently), age, sex.
• Correct equation family configured in software (e.g., GLI-2012 global multi-ethnic equations widely adopted for spirometry).
• Consistent set for serial comparisons when possible—switching equations between visits changes %predicted even if physiology is unchanged.

GLI-2012 (Global Lung Function Initiative) and related modern equations:

• Were developed from large international datasets spanning ages (including transition from childhood to older adults in continuous models).
• Provide predicted values, lower limit of normal (LLN), and z-scores rather than relying only on a fixed percent-predicted cutoff.
• Incorporate ethnic/ancestry groupings as defined by the equation authors (software may label populations differently—use the lab’s validated mapping, not guesswork).

Other historical sets (NHANES III, ECSC, Knudson, etc.) still appear in older reports and some legacy software. RPFT literacy: know that equation choice changes predicted values, document which set was used, and prefer lab SOP / standards-aligned modern multi-ethnic equations when configuring systems.

z-scores and LLN vs fixed cutoffs

Debate at technologist literacy (exam-ready):

• Fixed 0.70: widely used in COPD framework discussions; easy threshold; age bias is the criticism.
• LLN / z-score (GLI and similar): statistically aligned with the reference population; preferred in many PFT interpretation standards discussions for defining “below normal.”
• Your role: ensure the report shows ratio, %predicted, LLN/z-score when the system provides them, and that the configured equation matches lab policy. Do not silently swap fixed-ratio flags for LLN flags without understanding what the software is set to display.

Worked mini-example — LLN concept

Measured FEV1/FVC = 0.68. Patient is 75 years old. Predicted mean ratio might be near 0.72 with LLN near 0.64 (illustrative numbers—not a universal table).

• Against fixed 0.70, 0.68 is “low.”
• Against LLN 0.64, 0.68 may still be within the age-adjusted reference range.

Exam items test whether you understand why both numbers appear and that equation/LLN context matters, not that you memorize every GLI coefficient.

Upright vs supine spirometry comparison math

Some orders request supine (or semi-recumbent) spirometry—e.g., neuromuscular disease screening, diaphragmatic weakness evaluation, or perioperative risk pathways. Domain III expects you to calculate and present the comparison cleanly.

Procedure math, not new indices

1. Obtain acceptable upright FVC (and often FEV1) per lab protocol.
2. Obtain acceptable supine FVC (same quality rules; coach carefully—orthopnea and positioning).
3. Compute absolute change: ΔFVC = FVC_supine − FVC_upright (often negative when supine falls).
4. Compute percent change relative to upright (common framing):

$$\%\Delta FVC = \frac{FVC_{supine} - FVC_{upright}}{FVC_{upright}} \times 100$$

Worked mini-example

Upright FVC = 3.50 L; supine FVC = 2.80 L.

• ΔFVC = 2.80 − 3.50 = −0.70 L
• %ΔFVC = (−0.70 / 3.50) × 100 = −20%

Report both postures, quality grades, and the calculated change. Clinical thresholds for “significant” supine fall are interpretation-layer; technologist layer is accurate paired measurements and transparent math.

Exam trap: Computing percent change from the smaller denominator incorrectly, or mixing FEV1 change when the order asked for FVC positional comparison—match the ordered index.

Selecting which spirometric values go on the report (III.A.3)

After validity (II.C):

1. Select best FVC, best FEV1, compute ratio from the selected pair rules your lab uses (ATS/ERS-aligned selection).
2. Apply BTPS (system) and confirm demographics.
3. Attach predicted, %predicted, LLN/z-score from the configured equation (e.g., GLI-2012).
4. Include FEF25–75, PEF, and curve QC metrics as required by SOP.
5. For BD response or upright/supine, add absolute and percent change tables.
6. Never “fix” a predicted value by hand-picking an equation that makes the patient look normal—select references per policy, not convenience.

RPFT calculation vignettes

Vignette A: Software shows FEV1 2.1 L ATPS-looking raw and FVC 2.8 L; temperature sensor failed and defaulted to 37 °C ambient. Action: recognize BTPS/ambient configuration error risk; correct environmental inputs and recalculate per IFU—do not release volumes from known bad ambient data.

Vignette B: Adult report uses pediatric equation set after a height entry in inches was treated as centimeters. Action: demographic/equation mismatch; correct height units and regenerate predicted/%pred/z-scores.

Vignette C: FEV1/FVC = 0.69 in a 28-year-old with LLN 0.75. Action: ratio is below both fixed 0.70 and likely LLN—ensure numbers are correctly calculated; interpretation severity is clinical, but flags should reflect configured LLN logic.

Link across Domain III

Spirometry calculations feed reliability checks (III.B) and clinical selection of results (III.C). Lung volumes, DLCO, blood gas derived indices, and exercise math follow the same spirit: correct arithmetic, correct reference, transparent corrections—developed in sections 14.2–14.4.