NOTT

Published

NOTT Group. "Continuous or nocturnal oxygen therapy in hypoxemic chronic obstructive lung disease: A clinical trial. Nocturnal Oxygen Therapy Trial Group". Annals of Internal Medicine. 1980. 93(3):91-398.
PubMed • Full text

Clinical Question

Does continuous oxygen therapy reduce mortality in patients with COPD and hypoxemia when compared to nocturnal oxygen therapy?

Bottom Line

In patients with COPD and hypoxemia, continuous oxygen therapy significantly reduces mortality when compared to nocturnal oxygen therapy.

Major Points

Therapies for COPD that were few in the 1960s and included antibiotics, theophylline, ephedrine, the mucolytic potassium iodide, and phlebotomy. A trial by Levine and colleagues in 1967 suggested benefit of oxygen therapy for patients with COPD and hypoxemia though a larger multicenter trial to evaluate for optimal oxygen therapy frequency and a potential mortality benefit had not been performed.

The 1980 Nocturnal Oxygen Therapy Trial (NOTT) randomized 203 patients with COPD and hypoxemia to either nocturnal or continuous oxygen therapy. Participants had extensive evaluation of physiologic parameters at baseline and on regular intervals throughout the study. With an average follow-up of 19.3 months, continuous oxygen therapy was associated with a significant reduction in mortality as well as reductions in HCT and pulmonary vascular resistance. These findings were augmented by the 1981 British Medical Research Council Hypoxemia Trial that demonstrated a mortality benefit of oxygen therapy for >15h/day over room air in a similar population. Subsequent studies did not demonstrate a similar mortality benefit for milder hypoxemia.

Guidelines

ACP/ACCP/ATS/ERS COPD (2011, adapted)

• Continuous oxygen therapy for COPD and resting PaO₂ ≤55 mmHg or SpO₂ ≤88% (strong recommendation, moderate quality evidence)

GOLD COPD (2014, adapted)

• Long-term therapy with >15 hours/day of supplemental oxygen improves survival in patients with severe hypoxemia at rest (evidence B) and is indicated if either of the following are present:
  • PaO₂ ≤ 55 mmHg or SaO₂ ≤88% with or without hypercapnia confirmed twice in three weeks (evidence B)
  • PaO₂ 55-60 mmHg or SaO₂ of 88% if pulmonary hypertension, HF suggested by peripheral edema, or HCT >55% (evidence D)

Design

• Multicenter, parallel-group, open-label, randomized controlled trial
• N=203
  • Nocturnal Oxygen Therapy (n=102)
  • Continuous Oxygen Therapy (n=101)
• Setting: 6 treatment centers (unspecified locations)
• Enrollment: Not specifically stated, for a 27 months duration before 1980
• Average follow-up: 19.3 months
• Analysis: Intention-to-treat
• Primary outcome: Mortality

Population

Inclusion Criteria

• Age >35 with clinically diagnosed COPD
• Hypoxemia, measured on ≥2 occasions >1 week apart in a 3 week exacerbation-free observation period while on bronchodilators but off oxygen:
  • PaO₂ ≤55 mmHg
  • PaO₂ ≤59 mmHg plus one of the following:
    • Edema
    • HCT ≥55%
    • P pulmonale on ECG defined by P wave of 3mm in II, III, and aVF
• Lung function:
  • FEV1/FVC <70% after bronchodilator
  • TLC ≥80%

Exclusion Criteria

• Previous O₂ therapy for 12 hours/day for 30 days in prior 2 months
• Other disease influencing mortality, morbidity, compliance, or ability to provide informed consent

Baseline Characteristics

From the nocturnal group.

• Demographics: Age 65.7 years, male 80.4%, white 78.4%
• Laboratory analysis: PaO₂ 51.5 mmHg, PaCO₂ 43.9 mmHg, pH 7.41, HCT 47.3%
• Pulmonary function: FEV1 29.9%, FVC 53.6%, FRC 177.6%
• SpO₂: Sleeping with room air 83.5%, sleeping with O₂ 94%
• Other measurements: Maximum workload on air 37.3 W, HR 92.6 BPM, PA pressure 29.0 mmHg, CI 2.91 L/min*m², PVR 330 dyne/s*cm⁵
• QOL and psychiatric evaluation:
  • Overall rating: 4.5 (normal is 3.5)
  • Halstead impairment index: 0.78 (normal is 0.63)
  • Russell-Neuringer average impairment index: 2.3 (normal is 1.8)
  • Brain age quotient: 75.4 (normal is 89)
  • Minnesota Multiphasic Personality Inventory: 60.9 (normal is 54.5)
  • Sickness Impact Profile:
    • Physical scale: 20.5 (normal is 0.6)
    • Psychosocial scale: 23.9 (normal is 1.6)
  • Profile of Mood States: 48.4 (normal is 26.4)

Interventions

• Randomized to a group:
  • Nocturnal oxygen therapy
  • Continuous oxygen therapy
• All patients were treated with theophylline orally and inhaled beta-2 agonists. Use of steroids, cardiac glycosides, sedatives, tranquilizers, antidepressants, and oral beta agonists was discouraged. Therapeutic phlebotomies were not performed.
• Extensive baseline evaluation including laboratory analysis, EKG, chest radiography, PFTs, sleep studies, and right heart catheterization
• Repeat evaluation (except sleep study) every 6 months

Outcomes

Comparisons are nocturnal oxygen therapy vs. continuous oxygen therapy. Data and statistics presented only when given by authors.

Primary Outcomes

Mortality

Overall: Higher for nocturnal therapy group (RR 1.94; 95% CI 1.17-3.24)

At 12 months: 20.6% vs. 11.9% (P=0.01)

At 24 months: 40.8% vs 22.4%

Secondary Outcomes

Change in HCT from baseline

At 6 months: No difference

At 18 months: -2.0% vs. -9.2% (P=0.008)

Change in pulmonary vascular resistance

At 6 months: +6.5% vs. -11.1% (P=0.04)

Hospitalization rate

No difference

Additional Analyses

Average oxygen use

12.0 vs. 17.7 h/day

Subgroup Analysis

Significance for mortality benefit with continuous oxygen therapy was preserved for PaCO₂ ≥43, pH <7.40, HCG <47.4%, FVC <1.89 L, FRC ≥ 6.06 L, mean SaO₂ while asleep <85%, maximum work ≥35 W, mean PA pressure <27 mmHg, PVR <279 dyne/S*cm⁵, neuropsychological rating ≥4.5, Russell-Neuringer average impairment index ≥2.17, Halstead Impairment Index ≥0.75, and mood disturbance ≥43

Criticisms

• Unclear if similar benefit for those with less severe hypoxemia
• Lack of data for multiple outcomes in this article

Funding

Lung Diseases of the National Heart, Lung, and Blood Institute (NHLBI)