What is the clinical application and interpretation of Z-score spirometry in pediatric and young adult patients for assessing lung function?

Z-Score Spirometry in Pediatric and Young Adult Patients

Z-scores should be the preferred method for expressing spirometry results in pediatric and young adult patients because they account for age-related changes in lung function, eliminate biases related to height, sex, and ethnicity, and provide statistically robust interpretation across the entire age spectrum from early childhood through adulthood. 1, 2

What Z-Scores Represent

• Z-scores indicate how many standard deviations an observed value deviates from the predicted mean for a reference population, calculated as: (observed value - predicted mean) / residual standard deviation 1
• A z-score of 0 represents the population mean, with positive values above and negative values below the mean 1
• Z-scores can be interpreted in probability terms when data are normally distributed, indicating how likely a result is to occur within a normal population 1
• The lower limit of normal is defined as a z-score of -1.64 or below the 5th percentile of the reference population 3

Why Z-Scores Are Superior to Percent Predicted

• Percent predicted values are invalid in adults because the scatter around predicted values is constant regardless of height or age, leading to inappropriate comparisons between individuals 1
• Using percent predicted can suggest that two individuals with identical standard deviations from their predicted values have different lung function simply due to age or height differences 1
• Z-scores remove biases related to age, height, sex, and ethnic group and are clinically valid across all ages 2
• Percent predicted is only valid in children and adolescents where scatter is proportional to the mean, but z-scores work universally 1

Critical Advantages in Pediatric and Young Adult Populations

Addressing Growth Complexity

• In children and adolescents, lung growth lags behind height increase during the growth spurt, creating a shifting relationship between lung volume and height during adolescence 1
• Height growth in young males peaks approximately 1 year before FVC growth rate and 1.5 years before maximum flow growth 1
• A single equation or growth chart alone cannot completely describe the complex adolescent period, making z-scores essential for accurate interpretation 1

Seamless Transition Across Ages

• Modern reference equations using extended LMS (lambda, mu, sigma) methods provide smoothly changing curves from early childhood to old age, eliminating artificial discontinuities at age boundaries 4
• These models extend reference data down to 4 years of age and incorporate the height-age relationship in a biologically plausible manner 4
• Z-scores highlight that the range of normal values is highly dependent on age, which fixed cutoffs fail to capture 4

Clinical Application and Interpretation

Defining Abnormality

• Values with z-scores below -1.64 (5th percentile) are considered below the normal expected range 3
• For most pediatric measurements, z-scores between -2.0 and +2.0 are considered normal, with values outside this range being clinically significant 5
• The 95% confidence interval encompasses z-scores from -1.96 to +1.96 1

Practical Implementation

• Results should be expressed as z-scores rather than percentages of predicted values, as this provides far more information about interindividual variability 1
• Software for calculating z-scores should be provided with commercial spirometry equipment 1
• Reference data must match the patient's age, sex, height, and ethnicity, with the Global Lung Function Initiative (GLI) equations recommended for their wide age range and ethnic coverage 2, 6

Reference Equation Selection

Current Recommendations

• In the USA, ethnically appropriate NHANES III equations are recommended for ages 8-80 years, with Wang equations for children under 8 years 1
• The GLI equations provide the most comprehensive coverage, extending from 4 to 80+ years and including multiple ethnic groups 4, 6
• For children under 6 years, reference data from older subjects should never be extrapolated 1

Validation Requirements

• The validity of selected reference data should be checked by studying at least 30-50 healthy preschool children using identical techniques and comparing results with the reference population 1
• Laboratory directors should measure lung function in healthy subjects and compare with several reference equations to ensure appropriate selection 3

Common Pitfalls and How to Avoid Them

Age-Related Misdiagnosis

• Using fixed cutoff values (like FEV₁/FVC <0.70) without z-scores leads to 13.5% underdiagnosis in adults under 45 years and 33% overdiagnosis in elderly subjects 2
• This age-dependent bias is eliminated by using z-scores that account for normal age-related changes 2

Growth Assessment Errors

• In young infants, lung volumes should never be expressed per centimeter body length, as this creates false ratios and gross misinterpretation 1
• Height should be measured on the day of testing in infants and toddlers, as length can increase overnight by more than 1 cm due to saltatorial growth 1
• Measurements should be accurate to one decimal place in children and two decimals in infants to avoid excessive relative errors 1

Ethnic Considerations

• Most reference equations are derived from Caucasian populations, and differences due to ethnicity must be accounted for 1
• Reference values for lung volumes are approximately 12% lower in Black individuals compared to Whites 1, 3
• Failing to use ethnically appropriate reference equations leads to systematic misdiagnosis 7

Monitoring Disease and Treatment Response

• Z-scores allow tracking of changes in lung function with growth or treatment by providing standardized comparisons over time 1
• This is particularly valuable in conditions like cystic fibrosis, where early detection of lung function decline is critical 1
• Variability measurements should not be extrapolated from healthy children to those with disease; within-subject variability needs assessment in at least 30 subjects of similar age and diagnostic category 1

Integration with Clinical Context

• Spirometry results should never be interpreted in isolation but always in context with clinical presentation, symptoms, and other pulmonary function tests 3, 7
• For preschool children, at least two acceptable curves should be obtained where the second highest FVC and FEVt are within 0.1 L or 10% of the highest value 1
• If only a single satisfactory maneuver is recorded, results should not be excluded, but the number of technically satisfactory maneuvers and repeatability should always be reported 1