9.1 Predicted Value Equations, Reference Populations, and GLI 2012 / ATS/ERS 2022 Standards

9.1 Predicted Value Equations, Reference Populations, and GLI 2012 / ATS/ERS 2022 Standards

Interpretation of pulmonary function tests (PFTs) relies entirely on comparing a patient's measured parameters against predicted normative values derived from healthy reference populations. Without accurate reference equations, clinical technologists and pulmonologists cannot determine whether a patient's lung volumes, airflow rates, or gas transfer capabilities fall within normal physiological limits or reflect pathology.

This section reviews the anthropometric determinants of lung function, the evolution of reference equations from early single-population studies to the Global Lung Function Initiative (GLI) 2012 multi-ethnic equations, the ATS/ERS 2022 updated technical standards, the mathematical definition of the Lower Limit of Normal (LLN) versus the fixed 70% ratio, and Z-score severity grading essential for the NBRC CPFT examination.

---

Anthropometric Determinants of Predicted Values

Predicted values for spirometry ($FVC$, $FEV_1$, $FEV_1/FVC$, $PEFR$, $FEF_{25-75%}$), static lung volumes ($TLC$, $FRC$, $RV$), and diffusing capacity ($DLCO$) are calculated using statistical regression equations based on key anthropometric variables:

1. Height (Standing Height): Height is the primary structural determinant of lung volume. Taller individuals have larger thoracic cages and correspondingly larger lung capacities. Standing height must be measured accurately without shoes using a calibrated stadiometer.
   • Spinal Deformities and Amputations: In patients who cannot stand upright (e.g., severe kyphoscoliosis, spinal cord injury, or lower extremity amputation), arm span or sitting height must be used to estimate true standing height.
   • Arm Span Conversion: For adult Caucasian males, estimated height is calculated as $\text{Standing Height} = \frac{\text{Arm Span}}{1.06}$. For adult Caucasian females, $\text{Standing Height} = \frac{\text{Arm Span}}{1.03}$. Alternatively, a fixed ratio of $1:1$ is used in specific pediatric and adult demographic models.
2. Age: Lung volumes increase throughout childhood and peak between ages 20 and 25. After peak maturity, lung elasticity gradually declines, causing a progressive, physiological decrease in forced expiratory volumes ($FEV_1$) and vital capacity ($FVC$) at an average rate of 25 to 30 mL/year in non-smoking adults.
3. Biological Sex: On average, males have larger lung volumes and airway diameters than females of the same height and age, due to structural differences in chest wall geometry, diaphragm mass, and lung parenchyma.
4. Race and Ethnicity: Historically, reference equations incorporated self-reported race/ethnicity factors, reflecting observed population-level differences in thoracic proportions (e.g., sitting-to-standing height ratios). However, contemporary standards have re-evaluated these practices.

---

Evolution of Reference Equations and GLI 2012 Standards

Historically, PFT laboratories selected reference sets based on geographic proximity or local demographics. Early equations included Crapo (1981), Knudson (1983), and the NHANES III (Hankinson et al., 1999) dataset. While NHANES III established high-quality reference standards for Caucasian, African American, and Mexican American populations aged 8 to 80 years in the United States, it created artificial mathematical jump discontinuities when transitioning pediatric patients into adult equations.

Global Lung Function Initiative (GLI 2012)

To overcome these limitations, the European Respiratory Society established the Global Lung Function Initiative (GLI). Published in 2012, GLI 2012 provided the first continuous multi-ethnic reference equations for spirometry applicable from ages 3 to 95 years.

• Sample Size: Derived from 74,187 records from healthy non-smokers in 26 countries across five continents.
• Statistical Model: Uses the LMS method (Lambda-Mu-Sigma), accounting for age-related changes in median values (Mu), coefficient of variation (Sigma), and skewness (Lambda).
• Ethnic Modules: Included 5 distinct categories (Caucasian, African American, North East Asian, South East Asian, and Other/Mixed).

ATS/ERS 2022 Update: Race-Neutral Reference Equations

In 2022, a joint ATS/ERS taskforce recommended transitioning to race-neutral reference equations (such as GLI Global). The taskforce noted that race is a social construct rather than a biological determinant of lung function. Self-reported race-based equations can perpetuate health disparities by underestimating disease severity or delaying disability qualification in minority populations. The ATS/ERS 2022 standards recommend applying GLI Global multi-ethnic average equations across all patient groups paired with Z-score severity interpretation.

---

Lower Limit of Normal (LLN) vs. Fixed 70% Ratio

A central focus of the CPFT examination is the distinction between defining airflow obstruction using the Lower Limit of Normal (LLN) versus a fixed 70% (0.70) $FEV_1/FVC$ ratio.

Defining the Lower Limit of Normal (LLN)

The Lower Limit of Normal (LLN) represents the 5th percentile of a healthy, non-smoking reference population. Statistically, in a normally distributed population, 95% of healthy individuals fall above the 5th percentile, while 5% fall below. The LLN is calculated as:

$\text{LLN} = \text{Predicted Value} - (1.645 \times \text{Standard Error of the Estimate})$ $\text{Or using Z-scores: } \text{LLN corresponds to } Z = -1.645$

The Physiological Flaw of the Fixed 70% Ratio

The GOLD (Global Initiative for Chronic Obstructive Lung Disease) guidelines historically defined airflow obstruction as a post-bronchodilator $FEV_1/FVC < 0.70$. While simple to apply, using a fixed 70% ratio introduces severe clinical misclassifications:

1. Overdiagnosis in the Elderly: As humans age, loss of alveolar elastic recoil naturally decreases $FEV_1$ faster than $FVC$. Consequently, healthy elderly non-smokers (aged > 65) frequently exhibit a normal, age-related $FEV_1/FVC$ ratio between 63% and 68%. Using a fixed 0.70 threshold incorrectly diagnoses up to 20% of healthy elderly adults with obstructive lung disease (false positives).
2. Underdiagnosis in Young Adults: Healthy young adults (aged 18 to 30) normally have an $FEV_1/FVC$ ratio of 82% to 88%. The LLN for a 20-year-old may be 78%. A 20-year-old asthmatic patient with an $FEV_1/FVC$ ratio of 72% has experienced a significant physiological reduction in airflow, yet the fixed 0.70 cutoff misses the diagnosis (false negative).

\hline \text{Patient Population} & \text{Fixed 0.70 Ratio Metric} & \text{Physiological LLN Metric} \\ \hline \text{Elderly (Age > 65)} & \text{False Positives (Overdiagnosis)} & \text{Correct Normal Classification} \\ \text{Young Adults (Age 18-30)} & \text{False Negatives (Underdiagnosis)} & \text{Correct Obstructive Identification} \\ \hline \end{array}$$ --- ## Z-Scores and ATS/ERS 2021/2022 Severity Grading Traditional PFT interpretation classified disease severity using **percentage of predicted** cutoffs (e.g., 80% predicted = normal, 65-79% = mild, 50-64% = moderate, <50% = severe). However, percent predicted fails to account for variance; a 75% predicted value may be normal for an 80-year-old but severely abnormal for a 20-year-old. ### Z-Score Calculation A **Z-score** (standard deviation score) expresses how many standard deviations a measured value lies above or below the predicted mean: $$Z\text{-score} = \frac{\text{Measured Value} - \text{Predicted Mean}}{\text{Residual Standard Deviation (RSD)}}$$ * A Z-score of **$0$** indicates the measured value exactly equals predicted. * A Z-score of **$-1.645$** marks the 5th percentile (**LLN**). * Any measured parameter with a Z-score **$< -1.645$** is classified as abnormally low. ### Contemporary Severity Grading (ATS/ERS Standards) $$\begin{array}{cc} \hline \text{Z-Score Range} & \text{Severity Impairment Classification} \\ \hline \ge -1.645 & \text{Normal (Within Expected Range)} \\ -1.65 \text{ to } -2.50 & \text{Mild Impairment} \\ -2.51 \text{ to } -4.00 & \text{Moderate Impairment} \\ < -4.00 & \text{Severe Impairment} \\ \hline \end{array}$$