DETERMINATION OF INTRINSIC PEEP IN MECHAN ICALLY VENTILATED PATIENTS - VALIDATION OF A NEW OPTIONAL METHOD OF THE INTENSIVE-CARE VENTILATOREVITA

Intrinsic positive end-expiratory pressure (PEEP(i)) occurring during mechanical ventilation depends on expiratory time constants, expirator y volume and expiration time as well as on external flow resistance (t ubes, valves, etc.). It is not routinely determined in mechanically ve ntilated patients, but it is necessary to optimize respirator settings . The aim of the present study was the validation of an automated PEEP (i) determination method implemented in the respirator EVITA (Dragerwe rke, Lubeck) in mechanically ventilated patients with acute lung failu re. Patients. The method was validated in ten sedated, myorelaxed pati ents with respiratory insufficiency of different etiologies (five with restrictive, and five with obstructive pulmonary disease). PEEP(i) wa s determined using the volume constant ventilatory mode at ZEEP or at an external PEEP of 5 as well as 10 cm H2O. Method. PEEP(i) was first determined with the automated method implemented in the EVITA (five me asurements at each end-expiratory pressure level; PEEP(Evita). Steady- state was attained between each measurement. These values were compare d to the results obtained with end-expiratory occlusion (external, com puter-controlled valve in the inspiratory limb of the circuit) at the respective pressure levels (PEEP(EEO)). The average of five measuremen ts at each PEEP level with each method was defined as PEEP(i) for the particular ventilatory situation. Gas flow was measured at the proxima l end of the endotracheal tube with a heated pneumotachometer (Fleisch no. 2, Fleisch, Lausanne, Switzerland) and a differential pressure tr ansducer. Tracheal pressure was determined in the same position with a further differential pressure transducer (Dr. Fenyves & Gut, Basel, S witzerland). After A/D conversion, data were sampled with a frequency of 20 Hz and processed on an IBM compatible PC. Software for data coll ection and processing as well as for control of the occlusion valve wa s self-programmed. For the statistical analysis we used the Mann-Whitn ey U-test or Wilcoxon signed-ranks test; a P value less than 0.05 was considered significant. Results. At the given respiratory setting and without PEEP patients with obstructive lung disease had a higher PEEP( i) (median: 6.4 cm H2O; range: 5.0-9.6 cm H2O) than those with restric tive pulmonary disease (median: 2.3 cm H2O; range: 0. 8-3.0 cm H20) (P < 0.05). Increasing external PEEP to 5 or 10 cm H20 significantly dec reased the pressure difference between PEEP(i) and external PEEP (P < 0.05), but was unable to eliminate it completely. There was no statist ically significant difference between PEEP(EEO) and PEEP(Evita) (P = 0 .43; Wilcoxon signed-ranks tests). Regression analysis showed a highly significant correlation between PEEP(EEO) and PEEP(Evita) values (r = 0.985, P < 0.001; y = 1.03x-0.18). Discussion. PEEP(i) occurs during ventilation in patients with obstructive and restrictive lung disease. The difference between external end-expiratory pressure and PEEP(i) d ecreases with increasing external PEEP. However, PEEP(i) may increase with increasing external PEEP in some instances. The reason for this m ay be that the PEEP(i) determined at the proximal end of the endotrach eal tube represents only a mean value of different PEEP(i) values of v arious lung regions. Increasing extemal PEEP only partially alters thi s mean value due to an effect on PEEP(i) values lower than external PE EP. The PEEP(i) values measured by the EVITA respirator compared with classical end-expiratory occlusion with an external valve were nearly identical. Unfortunately, PEEP(i) measurement of the EVITA can only be performed during controlled and not during assisting (PSV, BIPAP etc. ) ventilation. Optimal respirator settings require a knowledge of PEEP (i) (i.e., adaption of external PEEP for lowering the work of breathin g in COPD patients or prolongation of the expiratory phase to avoid un wanted side effects of an occult PEEP(i) on the circulation). Since mo dem microprocessor-controlled respirators can easily be updated with t he necessary equipment, measurement of PEEP(i) should be a part of rou tine ventilatory monitoring today.