Assessment of physiologic variables and subjective comfort under different levels of pressure support ventilation

Chest. 2004 Sep;126(3):851-9. doi: 10.1378/chest.126.3.851.


Study objectives: To evaluate the effects of 12 ventilator settings (pressure support ventilation [PSV] plus positive end-expiratory pressure [PEEP], 30 + 0 cm H(2)O; 25 + 5 cm H(2)O; 25 + 0 cm H(2)O; 20 + 5 cm H(2)O; 20 + 0 cm H(2)O; 15 + 5 cm H(2)O; 15 + 0 cm H(2)O; 10 + 5 cm H(2)O; 10 + 0 cm H(2)O; 5 + 5 cm H(2)O; 5 + 0 cm H(2)O; and 0 + 5 cm H(2)O) on physiologic variables; the percentage of ineffective efforts; patient comfort; and whether the diagnosis of COPD may influence results.

Design: Prospective, randomized, physiologic study.

Setting: Three weaning centers.

Patients: Thirty-six consecutive patients (20 patients with COPD).

Intervention: Patients were randomly submitted to the 12 settings.

Measurements and results: Breathing pattern, respiratory drive (p0.1), arterial oxygen saturation (Sato(2)), heart rate, percentage of ineffective efforts per minute, patient comfort measured by means of a visual analogue scale (VAS), and BORG scale were recorded under each setting. Under different levels of assistance, breathing pattern, Sato(2), and p0.1 significantly and linearly changed (p < 0.0001) while VAS and BORG scale presented a significant (p = 0.027) U-shaped trend; high or low assistance caused the most discomfort. Under high levels of assistance, a higher (analysis of variance, p = 0.023) frequency of ineffective effort percentage was observed in the subgroup of 26 patients who presented this phenomenon. Breathing pattern significantly (p = 0.013) changed when compared to PSV alone (PSV plus zero end-expiratory pressure [ZEEP]) at the same total inspiratory pressure assistance (PSV plus PEEP). A huge variability among patients in breathing pattern and comfort was found under the setting rated as the most comfortable by patients. The diagnosis of COPD did not influence the overall results.

Conclusions: The following conclusions are made: (1) physiologic variables followed a linear trend, while comfort followed a U-shaped trend under different levels of PSV (irrespective of COPD diagnosis); (2) high assistance caused an increase in ineffective efforts; (3) only the breathing pattern significantly changed when total assistance was given as PSV plus PEEP when compared to PSV alone (PSV plus ZEEP); and (4) the extreme levels of PSV are not associated with the best comfort.

Publication types

  • Clinical Trial
  • Comparative Study
  • Multicenter Study
  • Randomized Controlled Trial

MeSH terms

  • Aged
  • Aged, 80 and over
  • Dyspnea / physiopathology
  • Dyspnea / therapy
  • Female
  • Heart Rate / physiology
  • Humans
  • Hydrostatic Pressure
  • Italy
  • Linear Models
  • Male
  • Mathematical Computing
  • Middle Aged
  • Outcome and Process Assessment, Health Care / statistics & numerical data
  • Oxygen / blood
  • Positive-Pressure Respiration / methods*
  • Prospective Studies
  • Pulmonary Disease, Chronic Obstructive / physiopathology*
  • Pulmonary Disease, Chronic Obstructive / psychology
  • Pulmonary Disease, Chronic Obstructive / therapy
  • Quality of Life / psychology
  • Tidal Volume / physiology
  • Ventilator Weaning
  • Work of Breathing / physiology


  • Oxygen