OBJECTIVES

To evaluate home noninvasive positive pressure ventilation (NIPPV) in adult patients with chronic respiratory failure in terms of initiation, continuation, effectiveness, adverse events, equipment parameters and required respiratory services. We evaluated respiratory failure primarily due to chronic obstructive pulmonary disease (COPD), thoracic restrictive disorders, neuromuscular disease, and obesity hypoventilation syndrome.

DATA SOURCES

National Guideline Clearinghouse, MEDLINE, EMBASE, SCOPUS, Cochrane Central Registrar of Controlled Trials, Cochrane Database of Systematic Reviews, and Scopus from January 1, 1995 to June 26, 2018.

REVIEW METHODS

We included randomized and comparative nonrandomized studies that enrolled adults with chronic respiratory failure who used NIPPV for ≥ 1 month at home (using a home mechanical ventilator [HMV], bi-level positive airway pressure [BPAP] device, or continuous positive airway pressure [CPAP] device).

RESULTS

We included 68 studies evaluating 53,733 patients. (36 studies), common criteria for NIPPV initiation were hypercapnia (PaCO2 ranging from >45 to >56mmHg), pH>7.35, FEV1 <50% of normal, and/or hypoxia (PaO2 ranging from <55 to <60mmHg or long term oxygen use). BPAP (compared with no device) was associated with reductions in mortality (moderate Strength of Evidence [SOE]), need for intubation (moderate SOE), and hospital admissions (low SOE). HMV (compared individually to BPAP, CPAP, or no device) was associated with fewer hospital admissions (low SOE). (8 studies), common criteria for NIPPV initiation were PaCO2>45mmHg, and FVC<40% normal or MIP<60cmH2O, or nocturnal SaO2<88% for ≥5 consecutive minutes. HMV (compared with no device) was associated with lower mortality (low SOE). (16 studies), common criteria for NIPPV initiation were PaCO2>45mmHg or FVC<50% or MIP <60cmH2O, or nocturnal SaO2 < 88% for ≥5 consecutive minutes. BPAP (compared with no device) was associated with reduced mortality (low SOE), and better quality of life (low SOE). (13 studies), common criteria for NIPPV initiation were hypercapnia (PaCO2 ranging from >45 to >53mmHg) and pH>7.35. HMV/BPAP mix (compared with no device) was associated with lower mortality. BPAP (compared with no device) was associated with improved sleep quality. In all conditions, evidence was insufficient to compare initiation criteria or determine the effect of specific home respiratory services on outcomes. Approximately one third of patients who use NIPPV via any device experienced non-serious adverse events such as facial rash, mucosal dryness, mask discomfort, etc. Based on direct comparisons, we found no significant differences in adverse events between devices or between devices and no device.

CONCLUSIONS

In patients with COPD, home BPAP (compared to no device) was associated with lower mortality, decreased need for intubations and hospital admissions, but no change in quality of life. In patients with COPD, HMV (compared individually with BPAP, CPAP, or no device) was associated with fewer hospital admissions. In patients with thoracic restrictive diseases, HMV (compared to no device) was associated with lower mortality. In patients with neuromuscular diseases, home BPAP (compared to no device) was associated with lower mortality and better quality of life. In patients with obesity hypoventilation syndrome, home HMV/BPAP (compared to no device) was associated with lower mortality. Current evidence is not available to assess the comparative effectiveness of many device capabilities on patient outcomes. Criteria to initiate home NIPPV and home respiratory services vary and are not validated in comparative studies.