Pulmonary Function Tests & Hypoxemia Workup

Spirometry: is there obstruction?

The ratio comes first. Spirometry measures FEV1, the volume forcefully exhaled in the first second, and FVC, the total volume forcefully exhaled. A post-bronchodilator FEV1/FVC <0.70 defines airflow obstruction, as in asthma, COPD, bronchiectasis, CF, and bronchiolitis obliterans. Severity is graded by FEV1 % predicted: ≥80% mild, 50–79% moderate, 30–49% severe, and <30% very severe (GOLD 1–4). Because the ratio normally declines with age, a fixed 0.70 cutoff overdiagnoses obstruction in older adults and misses it in the young; the lower limit of normal (LLN) corrects for this.

Bronchodilator response. A rise in FEV1 or FVC of ≥12% and ≥200 mL after an inhaled SABA is a significant response that supports asthma. COPD can improve partially, but the post-bronchodilator ratio remains <0.70, whereas complete normalization favors asthma. Short-acting bronchodilators are withheld for several hours before testing, and long-acting ones longer, so that a true response is not masked.

A low FVC is not proof of restriction. A reduced FVC with a normal or high ratio suggests restriction, but air trapping in obstructive disease and poor effort produce the same picture. Restriction is confirmed only by lung volumes showing a TLC <80% predicted (or below the LLN).

Lung volumes and DLCO

Lung volumes. Body plethysmography or gas dilution measures TLC and RV; plethysmography is more accurate when air is trapped behind obstructed airways. In obstruction RV rises (air trapping), TLC is normal or high (hyperinflation), and RV/TLC increases. In restriction TLC falls, usually with RV. A ratio <0.70 together with a low TLC indicates a mixed defect.

Parenchymal vs extraparenchymal restriction. Parenchymal disease (ILD, pneumonitis, pulmonary edema) lowers both TLC and DLCO. Extraparenchymal restriction from obesity, kyphoscoliosis, ankylosing spondylitis, pleural thickening, or neuromuscular disease (myasthenia gravis, ALS, Guillain-Barré, diaphragm paralysis) lowers TLC while DLCO stays normal, because the lung itself is healthy. In neuromuscular weakness RV may stay normal or rise because the expiratory muscles cannot empty the lung, maximal inspiratory and expiratory pressures fall, and FVC falls further when the patient lies supine if the diaphragm is weak.

In Guillain-Barré syndrome or myasthenic crisis, SpO2 falls late. Prepare to intubate when FVC <20 mL/kg, maximal inspiratory pressure is weaker than −30 cm H2O, or maximal expiratory pressure is <40 cm H2O (the 20/30/40 rule).

DLCO. The single-breath carbon monoxide diffusing capacity reflects alveolar surface area, membrane thickness, pulmonary capillary blood volume, and hemoglobin, and it is corrected for hemoglobin. It is low in emphysema (lost surface), ILD (thickened, scarred membrane), pulmonary vascular disease (PE, pulmonary hypertension), anemia, and after lung resection. It is normal in chronic bronchitis, asthma, and chest-wall or neuromuscular restriction, which is how DLCO separates emphysema from chronic bronchitis.

Isolated low DLCO. Unexplained dyspnea with normal spirometry and lung volumes but a low hemoglobin-corrected DLCO points to pulmonary vascular disease, either pulmonary arterial hypertension or chronic thromboembolic disease, or to early ILD. The next steps are echocardiography, a V/Q scan, and HRCT. Combined pulmonary fibrosis and emphysema in a heavy smoker can also produce near-normal spirometry and volumes, because the two processes offset each other, with a strikingly low DLCO.

High DLCO. DLCO rises when more hemoglobin sits near the alveoli: alveolar hemorrhage (red cells in the airspaces bind CO, so a rising DLCO tracks active bleeding), polycythemia, obesity and the supine position (larger pulmonary blood volume), left-to-right shunts, and asthma.

Bronchoprovocation testing

Methacholine challenge. When symptoms suggest asthma but spirometry and the bronchodilator response are normal, inhaled methacholine, a muscarinic agonist, is given in increasing doses; a fall in FEV1 of ≥20% at a low dose indicates airway hyperresponsiveness. The test is highly sensitive, so a negative result essentially excludes current asthma. A positive result is less specific and also occurs in COPD, allergic rhinitis, CF, heart failure, and some smokers.

Safety. The challenge is avoided when baseline obstruction is significant (FEV1 <60% predicted) and after a recent MI or stroke, with uncontrolled hypertension, or with an aortic aneurysm. Exercise or eucapnic hyperventilation challenges are used when exercise-induced bronchoconstriction is the question.

Flow-volume loops

Basic shapes. A normal loop has a sharp expiratory peak, a linear decline in flow, and a rounded inspiratory limb. Obstruction scoops out the expiratory limb (a concave shape) and shifts the loop toward higher lung volumes; severe emphysema gives a "church and steeple" loop, a brief peak followed by an abrupt collapse in flow as floppy airways close. Restriction produces a small, narrow loop with preserved or even high flows at low volumes.

Why the limbs differ. During forced inspiration, pressure inside the airway falls below atmospheric pressure and pulls a floppy extrathoracic segment closed, while negative pleural pressure holds the intrathoracic airway open. During forced expiration, positive pleural pressure compresses a floppy intrathoracic segment, while the extrathoracic airway is pushed open. A fixed lesion cannot change caliber, so both limbs plateau.

Hypoxemia and the A-a gradient

Five mechanisms. Low inspired oxygen, hypoventilation, V/Q mismatch (the most common), shunt, and diffusion limitation. The A-a gradient separates the first two, which have a normal gradient, from the last three, which widen it; the response to supplemental oxygen then separates V/Q mismatch from shunt.

Alveolar gas equation. PAO2 = FiO2 × (Patm − 47) − PaCO2/0.8. On room air at sea level this simplifies to about 150 − 1.25 × PaCO2, and the A-a gradient is PAO2 − PaO2. The normal gradient is about 5–10 mm Hg in young adults and widens with age. For example, a PaO2 of 60 mm Hg with a PaCO2 of 40 mm Hg gives a PAO2 of about 100 and a gradient of about 40 mm Hg, clearly widened.

Matching treatment to mechanism. Hypoventilation is corrected by restoring ventilation with naloxone or noninvasive ventilation, not by oxygen alone. In V/Q mismatch, partly ventilated units still receive the added oxygen, so PaO2 rises easily; hypoxic pulmonary vasoconstriction normally limits the mismatch. In shunt, blood crosses alveoli that are not ventilated at all (V/Q = 0), so extra oxygen never reaches it.

Anatomic shunts. Refractory hypoxemia with clear lungs suggests a right-to-left shunt: an ASD or PFO with high right-sided pressures (Eisenmenger physiology), pulmonary AVMs in hereditary hemorrhagic telangiectasia, or hepatopulmonary syndrome in cirrhosis. Agitated-saline (bubble) echocardiography detects them: bubbles reaching the left heart within about three beats indicate an intracardiac shunt, and later arrival indicates an intrapulmonary one; CT angiography then maps pulmonary AVMs for embolization. Platypnea-orthodeoxia, dyspnea and desaturation when upright that ease lying down, points to basilar AVMs, hepatopulmonary syndrome, or a PFO.

Diffusion limitation. A thickened or shrunken alveolar-capillary membrane may allow a normal resting PaO2, but faster capillary transit during exercise leaves too little time for equilibration, so saturation falls with exertion. A 6-minute walk test documents exertional desaturation and guides oxygen prescription.

Oximetry pitfalls: carbon monoxide produces a falsely normal SpO2, and methemoglobinemia reads about 85% regardless of the true oxygenation. In both the PaO2 is normal, and co-oximetry is required.