PFTs and Oxygen Delivery

In depth

The one-page summary above is what the PDF prints. These sections go further.

Lung volumes and how they are measured

Four volumes that do not overlap combine into four capacities. Functional residual capacity (FRC), the resting end-expiratory volume, is where lung recoil balances the outward spring of the chest wall. It is the oxygen store between breaths, and it shrinks when supine and with obesity, abdominal distension and anesthesia, which is why dependent lung collapses. Residual volume (RV) is the gas that cannot be exhaled.

Spirometry cannot measure RV, FRC or TLC. Helium dilution and nitrogen washout count only gas in contact with open airways, so they underestimate volume when gas is trapped, while a helium leak overestimates it. Body plethysmography applies Boyle's law as the patient pants against a closed shutter and counts all the gas in the chest, so it is preferred in obstruction.

TLC below the lower limit of normal (classically under 80% predicted) defines restriction; over 120% is hyperinflation. A high RV or RV/TLC ratio means air trapping, but the ratio (about 21% in the table) rises with age, so ERS/ATS 2022 uses the age-specific upper limit instead of the fixed 35–40% of older teaching.

Spirometry: technique, patterns and bronchodilator response

ATS/ERS 2019: the seated patient inhales fully, blasts out without hesitation and keeps blowing until volume changes by 0.025 L or less over 1 second, or for 15 seconds. The back-extrapolated volume must be under 5% of FVC or 0.100 L, whichever is greater, and a cough in the first second voids the FEV₁. Relative contraindications include MI within 1 week, unstable PE, pneumothorax and thoracic, abdominal or brain surgery within 4 weeks.

ERS/ATS 2022 reads each value against the lower limit of normal (z-score −1.645), and ATS 2023 recommends race-neutral GLI equations. GOLD's fixed 0.70 ratio overcalls obstruction in older adults and misses it in the young. Spirometry cannot confirm restriction: when obstruction comes with a low FVC, TLC is low in only about 10% (Barkous 2024).

Bronchodilator testing commonly gives four 100-µg puffs of albuterol by spacer and repeats spirometry about 15 minutes later. ERS/ATS 2022 counts a rise in FEV₁ or FVC of more than 10% of predicted; the 2005 rule, still on many reports, was at least 12% and 200 mL above baseline. Example: predicted FEV₁ 3.00 L, pre 1.80 L, post 2.15 L. The 0.35 L gain is 11.7% of predicted and 19% above baseline, positive by both rules. A negative test does not exclude asthma.

• Peak flow zones (NAEPP): green 80–100% of personal best, yellow 50–79%, red under 50%.
• MVV, measured over 12–15 seconds, is roughly FEV₁ × 35–40.
• Guillain–Barré (Lawn 2001): VC under 20 mL/kg, MIP weaker than −30 or MEP under 40 cm H₂O predicts intubation.

Flow-volume loops and upper-airway obstruction

Normally the expiratory limb peaks early and falls almost linearly, and the inspiratory limb is a smooth U. Obstruction scoops the expiratory limb, with a sharp drop after the peak in severe emphysema; restriction gives a small, tall, narrow loop.

A central-airway lesion flattens a limb into a plateau, and the pressure around the lesion decides which. Outside the chest a floppy segment narrows on forced inspiration, when airway pressure falls below atmospheric. Inside the chest it narrows on forced expiration, when pleural pressure exceeds airway pressure. A rigid lesion limits both.

Loops are insensitive: against endoscopy or CT, even a combined criterion found only 69% of obstructions (Modrykamien 2009, 475 loops), and COPD or asthma can hide the plateaus. Suspect a central lesion when dyspnea or stridor ignores bronchodilators, especially after prolonged intubation or tracheostomy, and confirm it with CT or bronchoscopy.

Diffusing capacity (DLCO)

In the single-breath test (ERS/ATS 2017) the patient exhales to RV, takes a full breath of gas with about 0.3% CO and a tracer, holds 10 seconds and exhales. Tracer dilution gives alveolar volume (VA) and CO uptake gives DLCO, about 25–30 mL/min/mm Hg in a healthy man. Repeat tests at least 4 minutes apart should agree within 2 mL/min/mm Hg.

Uptake depends on membrane area and thickness and on capillary blood volume and hemoglobin. Anemia and carboxyhemoglobin from smoking lower the result, so labs adjust for hemoglobin; polycythemia, alveolar hemorrhage, high pulmonary blood flow, lying supine and obesity raise it. KCO (DLCO ÷ VA) separates a small lung from a damaged one: weakness, chest wall disease, effusion or resection give a low DLCO with a normal or high KCO.

Oxygen devices and FiO₂ delivery

Low and high flow compare the device's flow with the patient's demand; they say nothing about FiO₂. A resting adult inhales at about 30 L/min (500 mL in 1 second), and peak flow in distress often exceeds 60 L/min. Below demand, room air makes up the difference and FiO₂ falls as breathing gets faster and deeper. A device that meets demand fixes FiO₂, so a 24% Venturi mask is high flow and a nonrebreather is low flow with a reservoir.

Cannula FiO₂ rises about 4% per L/min up to 6 L/min, even with mouth breathing. A simple mask needs at least 5 L/min to flush CO₂. Published FiO₂ values for the valveless partial rebreather range from about 35% to 85%. Keep a nonrebreather's bag from collapsing on inspiration; leaks lower its FiO₂.

Venturi masks fix FiO₂ by jet and port size. More oxygen raises total flow, not FiO₂; BTS 2017 advises more than the printed minimum when the rate is over 30. Covered ports or kinked or water-filled tubing add back-pressure, so FiO₂ rises and total flow falls.

Mixed gases: 30 L/min of air plus 10 L/min of O₂ gives (30 × 0.21 + 10) ÷ 40 = 0.41. Cylinder minutes = psi × factor ÷ L/min, with factors of 0.16 (D), 0.28 (E) and 3.14 (H or K); a full E cylinder at 2,200 psi lasts about 60 minutes at 10 L/min.

High-flow nasal cannula and the ROX index

HFNC delivers gas at about 37 °C and full humidity at up to 60 L/min, with FiO₂ 0.21–1.0. Matching inspiratory flow keeps the delivered FiO₂ near the set value, and flushing nasopharyngeal dead space lowers the minute ventilation needed. The flow adds a small pressure: about 2–3 cm H₂O at 30–50 L/min with the mouth closed, and less with it open (Parke 2011). Patients can talk and eat.

FLORALI 2015 (n = 310, nonhypercapnic, P/F ≤300) found intubation in 38% with HFNC, 47% with standard oxygen and 50% with NIV (P = 0.18), but lower 90-day mortality with HFNC. Rochwerg 2020 strongly recommends HFNC over conventional oxygen in hypoxemic failure and suggests it after extubation, where Hernández 2016 cut low-risk reintubation from 12.2% to 4.9%. ERS 2022 prefers NIV for high-risk extubations and a trial of NIV first in hypercapnic COPD.

Start at 40–60 L/min (FLORALI used 50) with FiO₂ titrated to the SpO₂ target and prongs filling no more than half of each naris. Weaning protocols vary; a common one lowers FiO₂ to about 0.40 first, then flow in 5–10 L/min steps to 20–30 L/min.

Example: SpO₂ 92% on FiO₂ 0.60 at a rate of 32 gives 153 ÷ 32 = 4.8. A ROX that does not climb over 12 hours is a warning. Late intubation is linked to worse outcomes: after 48 hours of failing HFNC, ICU mortality was 66.7%, against 39.2% with earlier intubation (Kang 2015, retrospective).

Oxygen toxicity, targets and hyperbaric oxygen

Excess oxygen forms reactive oxygen species that injure the lung: tracheobronchitis and a falling vital capacity first, diffuse alveolar damage later. High FiO₂ also causes absorption atelectasis as nitrogen washes out of poorly ventilated alveoli, constricts coronary and cerebral vessels, worsens hypercapnia in patients at risk and causes retinopathy in preterm infants. After bleomycin or in paraquat poisoning, TSANZ 2022 targets an SpO₂ of about 85%.

BTS 2017 and AARC 2022 aim for 94–98% in most acutely ill adults and 88–92% with hypercapnic risk; TSANZ 2022 uses 92–96% and 88–92%. BTS allows 94–98% for an at-risk patient whose PaCO₂ is normal, unless they have needed NIV or intubation for hypercapnic failure before. DETO2X-AMI 2017 (n = 6,629) found no benefit from oxygen in suspected MI without hypoxemia.

The trials show harm from liberal oxygen and from a PaO₂ of 55–70 in ARDS, and no difference between a PaO₂ of about 60 and 90 mm Hg. Turn FiO₂ down whenever SpO₂ sits at the top of its range, and check a low-normal SpO₂ with a gas in patients with darker skin (Sjoding 2020).

Hyperbaric oxygen gives 100% O₂ at 2.0–3.0 atmospheres absolute (ATA). At 3 ATA PaO₂ nears 2,000 mm Hg and dissolved oxygen (0.003 × PaO₂) reaches about 6 mL/dL, enough for resting needs, and Boyle's law shrinks bubbles to a third. Carboxyhemoglobin's half-life falls from about 5 hours on air to 74 minutes on 100% O₂ (Weaver 2000) and about 20 minutes in the chamber. UHMS indications include gas embolism, decompression sickness, CO poisoning, gas gangrene, necrotizing infection, crush injury, selected diabetic wounds, radiation injury, compromised flaps and sudden hearing loss.

For CO, three hyperbaric sessions in 24 hours cut 6-week cognitive sequelae from 46.1% to 25.0% (Weaver 2002); Hampson 2012 advises hyperbaric oxygen for loss of consciousness, neurologic deficits, cardiac ischemia, severe acidosis or COHb over 25%. Trials conflict, so ACEP 2017 accepts either hyperbaric or high-flow normobaric oxygen.

Humidification and aerosol delivery

Inspired gas reaches 37 °C and 44 mg H₂O/L at the isothermic saturation boundary, about 5 cm below the carina; an artificial airway moves it deeper, drying mucosa and thickening secretions until they plug the tube. AARC 2012: humidify every invasively ventilated patient, with 33–44 mg/L at 34–41 °C at the Y-piece from a heated humidifier or at least 30 mg/L from an HME, and avoid HMEs with low tidal volumes and in NIV. Classic HME contraindications (AARC 1992): thick, copious or bloody secretions, exhaled volume under 70% of delivered, temperature under 32 °C, minute ventilation over 10 L/min and in-line aerosol. Drain condensate away from the patient, never back into the humidifier.

Particles of 1–5 µm reach the lower airways; larger ones impact in the throat, more so with fast inhalation. AARC 2012 (Ari and Restrepo): a nebulizer or pMDI with valved holding chamber for children 4 or under and adults who cannot coordinate, DPIs from age 4 with enough flow, and no blow-by. In acute asthma a pMDI with chamber matches a nebulizer (Cochrane 2013).

On the ventilator, fire a pMDI into a chamber spacer in the inspiratory limb at the Y-piece as a breath starts. Nebulizer position is debated, and nebulizers put under 20% of the dose into the lungs (Dugernier 2017). A jet nebulizer on a separate flowmeter adds flow, raising tidal volume and FiO₂ and hindering triggering, so prefer the ventilator's nebulizer function or a mesh nebulizer. Remove the HME, and judge response by a smaller peak-to-plateau difference and less auto-PEEP.

Airway clearance and lung expansion

Lung expansion raises transpulmonary pressure, either by lowering pleural pressure (deep breathing, incentive spirometry) or by raising airway pressure (CPAP, PEP, IPPB). AARC 2011 does not recommend incentive spirometry alone for routine postoperative use; pair it with deep breathing, directed coughing, early mobilization and analgesia. Teach a slow inhalation, a 3–5 second hold and about 10 breaths an hour while awake; patients who cannot reach a vital capacity of about 10 mL/kg need positive pressure.

AARC 2013 (patients without cystic fibrosis): no chest physiotherapy for uncomplicated pneumonia, no routine clearance in COPD, early mobilization after surgery, and cough assistance in neuromuscular disease when peak cough flow is under 270 L/min. In cystic fibrosis the order is bronchodilator, mucoactive agent, airway clearance, then inhaled antibiotic or steroid.

NBRC emphasis and exam pearls

The 2027 NBRC RT outline lists spirometry, lung volumes, DLCO, peak flow, MIP, MEP, MVV, SpO₂/FiO₂ and oxygen titration with exercise. Device troubleshooting fills 19 of 25 Devices items, from flowmeters and cylinders to heated high-flow systems, humidifiers, nebulizers and inhalers. Hyperbaric oxygen is not a named task.

• Restriction needs a low TLC; a low FVC with a normal ratio is only a clue.
• Plethysmography finds trapped gas that helium dilution misses.
• Bronchodilator response: more than 10% of predicted (ERS/ATS 2022); older rule at least 12% and 200 mL.
• Flat inspiratory limb: variable extrathoracic; flat expiratory limb: variable intrathoracic; both flat: fixed stenosis.
• Emphysema lowers DLCO, asthma does not; a lung that cannot expand keeps a normal or high KCO.
• Venturi 40% is 3:1 (4 parts): 8 L/min gives 32 L/min; blocked ports raise FiO₂ and cut total flow.
• ROX = (SpO₂ ÷ FiO₂) ÷ RR; 4.88 or more reassures, under 3.85 at 12 hours predicts HFNC failure.
• Targets 94–98%, or 88–92% with hypercapnic risk; HOT-ICU found equal 90-day mortality at PaO₂ 60 and 90.
• No HME with thick or bloody secretions, a large leak, temperature under 32 °C or in-line aerosol.
• Untreated pneumothorax rules out hyperbaric oxygen and IPPB.