Long-term oxygen therapy remains one of only three interventions — alongside smoking cessation and, in selected patients, pulmonary rehabilitation or lung volume reduction — that extends survival in COPD. The evidence anchoring this is four decades old, the thresholds have not moved materially, and yet in Indian practice, LTOT prescription is persistently miscalibrated. Patients who would benefit never get it. Patients who do get it are told to use it “whenever breathless” rather than ≥15 hours daily. Patients who do not meet criteria end up on 4 L/min continuously because a well-meaning prescriber did not want to refuse a distressed family. This article lays out what GOLD 2024 actually says, how to titrate, and where Indian prescribing habits drift off course.
The audience is the prescribing pulmonologist, the respiratory therapist, and the home-care dealer interpreting prescriptions that may or may not be fully specified.
The evidence — still NOTT and MRC
Two trials, both from 1980–1981, define modern LTOT and have never been displaced.
The Nocturnal Oxygen Therapy Trial (NOTT) enrolled 203 COPD patients with documented arterial hypoxaemia (PaO₂ ≤55 mmHg, or 55–59 mmHg with cor pulmonale or polycythaemia) and randomised them to ~12 hours of nocturnal oxygen or ~18+ hours of continuous oxygen. The continuous-oxygen group had approximately half the two-year mortality of the nocturnal-only group. The dose-response was the critical finding — more hours of oxygen, lower mortality.
The Medical Research Council trial enrolled 87 similarly severe COPD patients with chronic hypoxaemia and randomised them to 15 hours/day of oxygen or no oxygen. At 5 years, mortality was ~45% in the oxygen group vs ~67% in controls.
Combined, the findings have defined LTOT for four decades: in COPD patients with documented resting arterial hypoxaemia, supplemental oxygen reduces mortality; the effect is dose-dependent; the minimum duration producing benefit is approximately 15 hours/day.
The LOTT trial (2016) later tested oxygen in patients with moderate hypoxaemia — SpO₂ 89–93% or exertional desaturation — and found no mortality or hospitalisation benefit. The LOTT result did not dilute NOTT/MRC; it reinforced the threshold boundary. Oxygen helps severe resting hypoxaemia. It does not help moderate hypoxaemia.
GOLD 2024 LTOT criteria
GOLD 2024 restates the LTOT indication (GOLD Report):
Prescribe LTOT for stable COPD patients with:
- PaO₂ ≤55 mmHg (SaO₂ ≤88%) at rest breathing room air, measured at least twice, three weeks or more apart, during a period of clinical stability;
or
- PaO₂ 55–60 mmHg (SaO₂ ~89%) in the presence of at least one of:
- Echocardiographic, ECG, or clinical evidence of cor pulmonale
- Polycythaemia (haematocrit > 55%, haemoglobin > 17 g/dL)
- Clinically significant pulmonary hypertension
- Peripheral oedema attributable to right heart failure
Stability conditions: not within six weeks of an exacerbation, on optimised medical therapy (long-acting bronchodilators, ICS where indicated per ABCD or ABE grouping), and not actively smoking.
The target of therapy is SaO₂ ≥ 90% (roughly SpO₂ 92%) at rest, during sleep, and during exertion. The minimum duration for mortality benefit is ≥15 hours/day, and most guidelines and textbooks favour 15–24 hours. Practically, the prescription translates as: wear oxygen during sleep, during quiet time at home, during any activity that brings on breathlessness, and during meals.
Indian Chest Society consensus
The Indian Chest Society’s LTOT guidance is broadly aligned with GOLD with some additional context specific to the Indian population (Indian Chest Society):
- Tuberculosis must be ruled out or adequately treated before chronic hypoxaemia is attributed to COPD. TB sequelae (bronchiectasis, destroyed-lung syndromes) remain a major respiratory-failure aetiology across India and affect LTOT selection.
- Smoking cessation is a hard precondition — the fire-and-burn risk with nasal oxygen plus active smoking is unacceptable, and many centres withhold LTOT until confirmed cessation.
- In the absence of ABG, two stable SpO₂ readings ≤88% three weeks apart are accepted as a working surrogate for PaO₂ ≤55 mmHg, with referral for ABG confirmation within 90 days where feasible.
- Altitude is flagged but not given a numeric modifier — Indian prescribers at altitude apply sea-level thresholds with clinical judgement.
Titration — why the target is SpO₂ 88–92%
A common Indian prescribing habit is to set the patient at 4 L/min continuous with an instruction to “keep SpO₂ above 95%”. This is usually wrong on three counts.
First, the target in chronic hypoxaemic COPD is not SpO₂ 95+%. It is SpO₂ 88–92%, corresponding roughly to SaO₂ 88–92% and PaO₂ 55–65 mmHg — enough to move the patient off the steep portion of the dissociation curve onto the plateau, without saturating to higher than physiologically necessary.
Second, most stable COPD LTOT prescriptions titrate to this target at 1–3 L/min via nasal cannula. A patient who needs 4+ L/min continuously at rest either (a) has end-stage disease and deserves a rehab and palliative-care conversation, (b) has a comorbidity like ILD or PH that is the actual driver, or (c) is being over-prescribed. Raising flow to 4+ L/min as a blanket setting without a titration record is poor practice.
Third — and this is the clinically dangerous part — in the subgroup of COPD patients who are chronic CO₂ retainers, excess oxygen suppresses hypoxic respiratory drive and raises PaCO₂. The SpO₂ creeps up, the patient gets drowsy, the family attributes it to “deep sleep”, and the patient develops hypercapnic respiratory failure. The BTS guidance in acute COPD exacerbation explicitly targets SpO₂ 88–92% in CO₂-retainer-suspected patients (British Thoracic Society). In chronic LTOT, the same logic applies — titrate to the lowest flow that achieves SpO₂ 88–92%, not to the highest flow the concentrator can deliver.
Typical LTOT titration flow:
- Start at 1 L/min nasal cannula with patient at rest.
- Measure SpO₂ after 20 minutes. If <88%, increase to 2 L/min. Reassess.
- Continue in 1 L/min increments until SpO₂ is 88–92%.
- Separately titrate for exertion (6-minute walk) and nocturnal use; exertional and nocturnal flow may exceed resting flow.
- For any patient requiring > 3 L/min at rest, perform an ABG to confirm PaO₂ and check PaCO₂ before finalising the prescription.
When LTOT is not the right prescription
Exertional desaturation in a patient with normal resting saturation. If resting PaO₂ is above 60 mmHg (or SpO₂ above 92%) and the patient desaturates only on walking — say SpO₂ 85% on a 6MWT — the LOTT-era evidence does not support continuous LTOT. The appropriate prescription is ambulatory oxygen during exertion, typically with a portable concentrator or ambulatory cylinder sized to the walking dose. Adherence is low in this group and the mortality benefit is unproven. Many patients find that the effort of carrying the device exceeds the breathlessness benefit. This should be honestly discussed before prescription.
Nocturnal desaturation only. A patient with resting daytime SpO₂ above 92% who desaturates during sleep (SpO₂ < 88% for > 30% of the night) may benefit from nocturnal-only oxygen if sleep-disordered breathing has been ruled out or addressed. Obstructive sleep apnoea should be treated with CPAP first; obesity-hypoventilation deserves BiPAP. Nocturnal oxygen alone in an isolated-desaturation patient without OSA or OHS is a defensible prescription but the mortality evidence is weaker than for resting-hypoxaemic LTOT.
Dyspnoea without hypoxaemia. This is the commonest misprescription in Indian practice. A patient presents breathless, SpO₂ reads 94%, the family expects oxygen, and a prescription is written. There is no evidence that supplemental oxygen relieves dyspnoea in non-hypoxaemic patients. Pulmonary rehabilitation, inhaler optimisation, anxiety management, and fan therapy have more evidence than oxygen in this group.
Acute exacerbation settings. In acute exacerbation, oxygen is titrated to SpO₂ 88–92% (BTS/ICS guidance) and reassessed after the exacerbation resolves. A discharge oxygen prescription should not be issued on exacerbation-era gas values; LTOT qualification requires stable-state measurements 6 weeks out.
Indian practice gaps
The common drift patterns we see in prescription review:
“4 L/min continuous” with no titration record. Often written by non-pulmonology prescribers under family pressure. Almost always over-prescription. The correct answer is 1–3 L/min titrated to SpO₂ 88–92%.
Sleep-hour-only oxygen for a resting-hypoxaemic patient. Often driven by concerns about mains electricity bills (oxygen concentrators at 5 LPM draw ~350 W, ~₹1,500–3,500/month depending on state tariff). The NOTT data specifically showed that 12 hours is not enough — mortality benefit requires ≥15 hours. Limiting a qualifying patient to sleep-hours only nullifies most of the benefit.
LTOT started on a single SpO₂ reading. GOLD requires two stable readings ≥3 weeks apart. Single-reading initiation happens in Indian OPD practice under time pressure; the prescription should still be reconfirmed within 90 days.
Concentrator size mismatch. A patient titrated to 2 L/min does not need a 10 LPM concentrator. Over-sizing is a common sales-driven error; the 10 LPM machines are louder, pull more mains current, and cost more to buy and run. Most COPD LTOT patients are appropriately served by a 5 LPM concentrator with its flowmeter in the 1–3 LPM range. 10 LPM units are for ILD, pulmonary hypertension, or CPAP-blend applications.
Ignoring CO₂ retention risk. A COPD patient with morning headache, daytime somnolence, or plethora should have ABG before LTOT. The concentrator titrated to SpO₂ 88–92% is a safe target; the concentrator cranked up to 4+ L/min without an ABG is not.
Contraindications and risks
Active smoking. Oxygen vigorously supports combustion. Facial burns, home fires, and deaths have been reported in patients smoking while on nasal cannula. Smoking cessation is a hard precondition for LTOT. In practice, the honest approach is to verify cessation (CO monitor, cotinine) before initiation, and to withdraw the prescription if smoking resumes.
Unstable coronary or cerebrovascular disease with CO₂ retention. Oxygen-induced hypercapnia in this subgroup is particularly risky. ABG-documented prescription and careful titration are mandatory.
Home fire risk. Indian household kitchens, incense, diya/agarbatti, and gas hobs are all ignition sources. Family education on keeping the cannula and tubing away from any flame is part of the prescription, not optional.
Electrical infrastructure. A stationary concentrator at 5 LPM uses ~350–450 W. Over 15+ hours/day, monthly electricity is ₹1,500–3,500 depending on state tariff and tier. The family’s ability to afford this should be assessed and discussed before the patient goes home with the device. Power-cut areas need a UPS or inverter sized for concentrator startup surge.
The ambulatory and exercise question
For the subgroup of COPD patients with resting PaO₂ above 60 mmHg but significant exertional desaturation, ambulatory oxygen is an option with honest caveats. The LOTT trial found no mortality or hospitalisation benefit from ambulatory/supplemental oxygen in moderate-hypoxaemia patients. Some patients report symptomatic benefit — ability to walk further, shop, attend family events. Some patients find the equipment burden exceeds the benefit and stop carrying the device. The prescription should be issued with a realistic conversation about trial periods and expected outcomes.
Portable equipment choice matters here. Continuous-flow portables (5–6 kg, 2–3 LPM continuous, 4–5 hours battery) carry oxygen delivery that matches home concentrator flow. Pulse-dose portable concentrators (2–3 kg, 3–5 hour battery) deliver a bolus on inhalation trigger; the effective minute ventilation of oxygen is less than the numbered setting suggests, and patients with high respiratory rates or mouth-breathing patterns under-dose on pulse settings. For exertional desaturators who walk fast enough to trigger 30+ breaths/min, pulse-dose often fails to maintain saturation.
Clinical takeaway
Prescribe LTOT for COPD patients with PaO₂ ≤55 mmHg or SpO₂ ≤88% at rest, or PaO₂ 55–60 mmHg with cor pulmonale, polycythaemia, or pulmonary hypertension — measured in two stable readings at least three weeks apart. Target SpO₂ 88–92% at the lowest flow that achieves it, for ≥15 hours daily. Most stable COPD LTOT patients are correctly prescribed at 1–3 L/min; 4+ L/min continuous is almost always over-prescription or a signal that the diagnosis is not uncomplicated COPD. Ambulatory oxygen for exertional desaturators is an option with honest trial-period caveats, not a default.
Consult your pulmonologist before initiating or changing oxygen therapy; titration and CO₂-retention screening are not optional components of a safe prescription.