The HOT-HMV study
The HOT-HMV study

The HOT-HMV study

51% decrease in risk of hospital readmission or death in hypercapnic COPD patients treated with home non-invasive ventilation and oxygen therapy

Results of the HOT-HMV trial published in JAMA1

Chronic obstructive pulmonary disease (COPD) causes the progressive deterioration of respiratory function and often leads to intermittent severe exacerbations as the disease advances. Patient outcomes following hypercapnic exacerbations requiring hospitalization and acute non-invasive ventilation (NIV) are poor, and few treatment options exist for these patients. Exacerbations and hospitalizations also place a significant burden on healthcare systems. The multi-center, open label, parallel-group, randomized controlled HOT-HMV1 trial investigated the effect of home NIV and oxygen on time to readmission or death in patients with persistent hypercapnia after an acute COPD exacerbation (AECOPD) requiring acute NIV. It aimed to identify whether this treatment strategy could improve outcomes for this patient group. 

Study design and study population

The study recruited severe hypoxic and hypercapnic COPD patients who had been hospitalized for acute decompensated hypercapnic exacerbation of COPD requiring NIV.

116 patients were randomized into two groups: the HOT group treated with home oxygen therapy (HOT) and the HOT-HMV group treated with both HOT and home non-invasive ventilation (HMV).

The primary outcome was admission-free survival, a combined endpoint of time to readmission to hospital for any cause or death within 12 months after randomization. Patients met the primary outcome if they experienced either endpoint. 

The recruitment process was designed to ensure that the effect of the therapy was assessed in patients who did not have any significant cause of sleep-disordered breathing and/or respiratory failure other than COPD (e.g. obesity, obstructive sleep apnea, neuromuscular disease, chest wall disease).

A computer-assisted stratified randomization was performed to guarantee the balance of the two groups of the study with regard to the following factors: age (<65years, ≥65 years); body mass index (BMI) (≤20, >20), current long-term oxygen therapy (yes, no); frequency of COPD-related readmissions during previous 12 months (<3, ≥3); recruitment center. 

Hypoxia and hypercapnia were assessed by blood gas analysis (ABG) two to four weeks after the resolution of the hypercapnic acidosis to ensure that enrolled patients presented with a chronic condition.

Inclusion criteria

  • FEV1 <50% of predicted - FEV1/FVC <60%.

  • Inpatient admission with acute hypercapnic exacerbation of COPD.

  • Persistent hypercapnia (pH >7.30, PaCO2 ≥53 mmHg) evaluated two to four weeks after the resolution of the hypercapnic acidosis.

  • Chronic hypoxia PaO2 <55 mmHg or <60 mmHg with secondary polycythaemia, pulmonary hypertension, peripheral edema or significant nocturnal hypoxia (SpO2 <90% for >30% sleep time).

  • Smoking history of greater than 20 pack-years.

Exclusion criteria*

  • Declined n=296 (16%)

  • Inability to consent n=237 (12%)

  • Admission not due to an acute exacerbation of COPD n=157 (8%)

  • Died prior to screening n=128 (7%)

  • Unable to screen within trial protocol n=46 (2%)

  • Unable to wean from NIV (pH <7.30) n=252 (13%)

  • Post-decannulation or extubation on index admission n=51 (3%)

  • Unable to tolerate NIV n=131 (7%)

  • Decompensated with oxygen therapy n=8 (<1%)

  • Obstructive sleep apnea n=76 (4%)

  • BMI >35kg/m2 n=96 (5%)

  • Arterial blood gases not meeting inclusion criteria n=419 (22%)

  • Other reasons n=8 (<1%)


Baseline characteristics

Reproduced with permission from Murphy 2017, Table 1.


  • Baseline characteristics were well matched between the intervention and control groups.

  • Baseline characteristics of the enrolled patients show a population with severely compromised pulmonary function, high levels of PaCO2, and a high rate of hospitalizations per year.

  • Over 50% of patients had ≥3 COPD-related hospital admissions in the previous year.

  • On room air, mean PaO2 was 48 mmHg and PaCO2 was 59 mmHg, indicating hypoxemia with hypercapnia in both patient groups.

  • Health-related quality of life (HRQOL) was significantly impaired, as measured by the St. George’s Respiratory Questionnaire (SGRQ), Severe Respiratory Insufficiency (SRI) questionnaire, and the Medical Research Council (MRC) breathlessness scale, which indicates degree of dyspnea.

Study results

64 patients completed the 12-month study period (28 in the HOT group, 36 in the HOT-HMV group).

All primary and secondary analyses were undertaken on the intention-to-treat principle.

51% decrease in risk of hospital readmission or death within 12 months

There was a 51% reduction in the risk of hospital readmission or death in the HOT-HMV group compared to the HOT group. Median admission-free survival time was 4.3 months in the HOT-HMV group compared to 1.4 months for those in the control group. This translates to an increase of over 90 days in the median time to first event for the HOT-HMV group. 

Reproduced with permission from Murphy 2017, Figure 2


17% absolute risk reduction

The risk of hospital readmission or death at the end of the 12 months was 63.4% in the HOT-HMV group and 80.4% in the HOT group, meaning an absolute risk reduction of 17% (95% CI, 0.1%-34.0%). This translates to a need to treat 6 patients to avoid one hospital readmission or death in 12 months.

Given the significant cost of hospital admissions for severe COPD, this implies that HOT-HMV could help to reduce the economic burden of this disease.2

Positive results driven by hospital readmission

These results were driven by a reduction in hospital readmission. There were no statistically significant differences in mortality at 12 months or for the event triggering the primary outcome. 

It should be noted that the study was not powered to detect differences in mortality outcomes.2

Reproduced with permission from Murphy 2017, Supplement 2, eFigure


In addition, a post-hoc analysis showed that in the first 28 days after randomization there was a 74% reduction in the risk of hospital readmission for those in the HOT-HMV group. Two-thirds fewer readmissions were observed in this period in this patient group compared to the HOT group. 

Exacerbation rate reduced by 34%

In addition to the positive effect on time to first readmission or death, further analysis showed that the exacerbation rate over 12 months was reduced by 34% in the HOT-HMV group. 

This suggests that patients receiving HOT-HMV may experience better outcomes.2

QOL maintained and therapy well tolerated

Health-related quality of life for patients in the HOT-HMV arm was significantly higher at 6 weeks according the results of the Severe Respiratory Insufficiency (SRI) questionnaire and at 3 months according the results of the St George’s Respiratory Questionnnaire (SGRQ). 

These benefits became less marked over time, with no statistically significant difference thereafter.

HMV usage increased from a median of 4.7 hours per night at 6 weeks to 7.6 hours at 12-month follow up.

These results indicate that the high-pressure ventilation strategy was well tolerated.






The therapy was well tolerated and QOL was maintained despite the use of high pressures. Hours of use increased over the course of the study.

The modest effect on QOL is unsurprising: the patient population had severe disease and high levels of physical impairment at baseline. After the first 3 months, there was a dilution of treatment effect as 18 patients from the HOT group were allowed to receive the ventilation therapy, in line with study protocol.2

Therapy initiation and settings

Oxygen therapy (HOT)

  • Both groups received home oxygen therapy.

  • Oxygen was started in both groups, at the lowest flow rate required to increase PaO2 above 60mmHg without producing a decompensated respiratory failure.

  • Both groups received a median of 1 liter/minute of oxygen.

Home NIV therapy

  • The HOT-HMV group received HMV in addition to HOT.

  • The study used a high-pressure ventilation strategy.

  • In-patient NIV titration was performed during the night after a daytime acclimatization, and with O2 therapy set at daytime flow rate.

  • Inspiratory pressure was initially set at 18 cmH2O and was titrated to the highest level tolerated by the patient under SpO2 and tcCO2 monitoring, reaching a median IPAP of 24 cmH2O.

  • The back-up rate was moderate (median 14 bpm), as high rates have not been found to be beneficial in previous trials.3

Home NIV effectively corrected hypoventilation and reduced CO2 level

Mean / Max tcCO2

Significant improvements were observed in nocturnal mean tcCO2 and maximum tcCO2 in the HOT-HMV arm showing that ventilation therapy was effective in correcting hypoventilation.

Mean PaCO2

HMV therapy was effective in reducing daytime levels of CO2, as measured by ABG. Patients receiving HOT-HMV obtained a statistically significant benefit at 6 weeks and 3 months.

The dilution in the therapy effect on TcCO2 and PaCO2 may be explained by the 18 patients from the HOT group who required and were permitted to receive ventilation therapy.2

What are the implications for clinical practice?

The HOT-HMV trial has important implications for the clinical treatment of patients with severe COPD and chronic respiratory failure who experience a life-threatening exacerbation.

Wider adoption of home NIV for hypercapnic COPD patients

The HOT-HMV study should prompt changes in the clinical management of severe COPD patients with chronic respiratory failure following a life-threatening exacerbation and also supports the argument that home NIV should be adopted more widely as part of the therapy strategy for severely hypercapnic COPD patients after hospitalization for AECOPD.

Systematic screening

The positive results of the HOT-HMV trial suggest that patients with severe COPD should be systematically screened following a hospitalization for AECOPD requiring acute NIV to assess their suitability for home NIV therapy.

GOLD guidelines4

The HOT-HMV findings confirm the value of the new recommendations provided by the GOLD guidelines. The guidelines now include home NIV as a therapy to consider for the treatment of hypercapnic COPD patients.

High-pressure strategy

HOT-HMV confirms the efficacy and feasibility of using high pressures to treat COPD patients. High pressures were effective in correcting hypoventilation and reducing hypercapnia while QOL assessments and usage statistics indicate that patients tolerated the therapy well.

What are the implications for health economics?

Home NIV has the potential to reduce the healthcare costs associated with the management of patients with severe COPD. HOT-HMV significantly increased time to hospital readmissions, which place a significant burden on healthcare systems. Reductions in the exacerbation rate also imply that HOT-HMV has the potential to reduce the costs and resources required to treat this group outside the hospital.

Furthermore, HOT-HMV may have amplified effects on cost reduction in some systems which apply penalties for recurrent hospital readmissions due to AECOPD.5


* Percentage relative to the total number of excluded patients

  1. P. Murphy et al., Effect of Home Noninvasive Ventilation With Oxygen Therapy vs Oxygen Therapy Alone on Hospital Readmission or Death After an Acute COPD Exacerbation. A Randomized Clinical Trial, JAMA. Published online May 21, 2017. doi:10.1001/jama.2017.4451.
  2. ERS 2017, Milan, Italy - Accessed July 18th 2017
  3. Murphy PB, Brignall K, Moxham J, Polkey MI, Davidson AC, Hart N. High pressure versus high intensity noninvasive ventilation in stable hypercapnic chronic obstructive pulmonary disease: a randomized crossover trial. Int J Chron Obstruct Pulmon Dis 2012;7:811-8
  4. Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease.2017 Report.
  5. Fingar K, Washington R. Trends in hospital readmission for four high-volume conditions, 2009-2013. Agency for Healthcare Research and Quality (AHRQ). Healthcare Cost and Utilization Project. Statistical brief #196. November 2015.