D. Mascheroni

Relationships between lung computed tomographic density, gas exchange, and PEEP in acute respiratory failure.

Twenty-two patients with acute respiratory failure underwent lung computed tomography (CT) and physiological measurements at 5, 10, and 15 cm H2O positive end-expiratory pressure (PEEP) to investigate the relationship between morphology and function. Lung densities were primarily concentrated in the dependent regions. From the frequency distribution of CT numbers (difference in xray attenuation between water and lung) and lung gas volume measurements the authors obtained a quantitative estimate of normally inflated, poorly inflated, and non-inflated lung tissue weight. This estimated average lung weight was increased twofold nbove normal and excess lung weight correlated with the mean pulmonary artery pressure (P < 0.01). Venous admixture correlated with the non-inflated tissue mass (P < 0.01). Increasing PEEP caused progressive clearing of radiographic densities and increased the mass of normally inflated tissue (anatomic recruitment), while reducing venous admixture. The cardiac index decreased after increasing PEEP while oxygen delivery was unchanged. The authors conclude that CT scan lung density and oxygen exchange efficiency are correlated; the main effect of augmenting PEEP is to recruit perfused alveolar units that were previously collapsed.

Morphological response to positive end expiratory pressure in acute respiratory failure. Computerized tomography study

SummaryTen patients with acute respiratory failure (ARF), (4 pneumonia, 4 sepsis, 2 polytrauma), underwent computerized tomography (CT) of the lungs, (apex, hilum, base), at 5, 10, 15 cm H2O positive end expiratory pressure (PEEP). The ARF lungs, on CT scan, appeared as a patchwork of normal and dense areas with generally well defined boundaries. Most of the densities were found in the dependent regions. The areas of density were correlated with PaO2 (r=0.51). The PEEP increase resulted in a significant expansion of total cross-sectional lung surface area. The dense areas decreased significantly at the hilum and base when increasing PEEP while the changes at the apex were not significant. The changes of density with PEEP were highly correlated with the changes in oxygenation (r=0.91). In the individual patient, however, the modifications of gas exchange can not be entirely predicted from morphological changes, possibly due to a diversion of pulmonary blood flow.

The role of total static lung compliance in the management of severe ARDS unresponsive to conventional treatment

A group of 36 patients with severe adult respiratory distress syndrome (ARDS) meeting previously established blood gas criteria (mortality rate 90%) became candidates for possible extracorporeal respiratory support [low frequency positive pressure ventilation with extracorporeal CO2 removal (LFPPV-ECCO2R)]. Before connecting the patients to bypass we first switched the patients from conventional mechanical ventilation with positive end expiratory pressure (PEEP) to pressure controlled inverted ratio ventilation (PC-IRV), and then when feasible, to spontaneous breathing with continuous positive airways pressure (CPAP). Forty eight hours after the patients had entered the treatment protocol, only 19 out of the 36 patients in fact required LFPPV-ECCO2R, while 5 were still on PC-IRV, and 12 were on CPAP. The overall mortality rate of the entire population was 23%. The only predictive value of success or failure of a particular treatment mode was total static lung compliance (TSLC). No patients with a TSLC lower than 25 ml (cm H2O)-1 tolerated either PC-IRV or CPAP, while all patients with a TSLC higher than 30 ml (cm H2O)-1 were successfully treated with CPAP. Borderline patients (TSLC between 25 and 30 ml (cm H2O)-1) had to be treated with PC-IRV for more than 48 h, or were then placed on LFPPV-ECCO2R if Paco2 rose prohibitively. We conclude that TSLC is a most useful measurement in deciding on the best management of patients with severe ARDS, unresponsive to conventional treatment.

Volume/pressure curve of total respiratory system in paralysed patients: artefacts and correction factors

The volume/pressure (V/P) curve of the total respiratory system in paralysed patients is drawn assuming that volume changes of the respiratory system (ΔV resp) equals volume displacement of the measuring apparatus (ΔV syr), usually a supersyringe. However, in 93 VP curves we found that O2 removed from the lung-syringe system during the procedure (proportional to the time) largely exceedes the CO2 added to the lung-syringe system (ΔV gas). This results in a net loss of volume from the system (ΔV resp<ΔV-syr). Deflation compliance, hysteresis area and ratio are significantly affected by this phenomenon. Inflation compliance is less influenced by ΔV gas, partially compensated by the intrapulmonary gas expansion due to the temperature changes. We conclude that the parameters computed on the deflation limb of V/P curve are misleading if proper correction of the volume scale is not introduced.

Combined use of mask CPAP and minitracheotomy as an alternative to endotracheal intubation : preliminary observation

We describe the combined use of mask CPAP (continuous positive airway pressure) and minitracheotomy as an alternative to conventional endotracheal intubation in 3 patients requiring CPAP, secretion removal and diagnostic procedures such as bronchoalveolar lavage and bronchial cultures. These requirements were fulfilled with the combined technique approach, thus preserving glottic function and avoiding the disadvantages of endotracheal intubation or trachectomy. This approach seems particularly suitable in the treatment of immunocompromised patients because of its reduced invasiveness.

Mechanical pulmonary ventilation at high airway pressures: Is it safe?

Past efforts designed to help diseased lungs to recover from a life threatening acute illness have traditionally focussed on the use of mechanical pulmonary ventilators. Yet when a patient with adult respiratory distress syndrome (ARDS) is maintained with a mechanical respirator, the outcome is by no means as favorable as when the breathing difficulties were due to non-parenchimal causes such as neuromuscular disease. In mild ARDS the mortality rate is between 10 and 20%, and in moderately severe ARDS the mortality rises to 50-60%. Severe ARDS, as defined by the National Heart, Lung, and Blood Institute ECMO (Extracorporeal Membrane Oxygenation) entry criteria, has a mortality rate over 90% with conventional mechanical pulmonary ventilation; while a randomly matched patient population treated with an extracorporeal membrane lung and still on mechanical ventilation had no better survival (1). Recently, Gattinoni et al. (University of Milan) have reported a substantial improvement in survival among severe ARDS patients treated with an extracorporeal membrane lung while the diseased lungs were ventilated at very low rates and at low airway pressure of 35 cm H20 (2). These results prompted us to question the role of mechanical ventilator in ARDS. In a previous editorial, we have explored some pulmonary and extrapulmonary effects of mechanical ventilation (3). Damage to the lung from barotrauma may consist of cystic dilation of airways, interstitial, subcutaneous and mediastinal emphysema, or pneumothorax, among others. The direct effects on pulmonary parenchyma are commonly ignored. The cardiovascular, renal , hepatic and CNS effects of mechanical ventilation are perhaps better known, but no better understood. While it is clear that mechanical ventilation could produce pulmonary and extrapulmonary changes, one has great difficulty looking for specific factors that block healing of the lungs. Treatment with a mechanical ventilator has been accepted as state of the art; unless patients met criteria for severe ARDS, such treatment with a mechanical ventilator was to be considered proper with no upper limits on peak airway pressures. We wish to question the safety on mechanical ventilation at elevated airway pressures in patients with healthy lungs, and particularly in patients with ARDS. Rarely is it remebered that the lungs are a metabolic organ, with some 80 or more cell lines, the homeostatis of which is likely to be of great importance. The lungs do much more than effect oxygen uptake, and C02 removal. And yet it is precisely to effect adequate pulmonary ventilation and to maintain sufficiently high P02 that high airway pressure, high tidal volume, high positive end expiratory pressure (PEEP), and the many modalitiesof administering the same are presently used. What effect might our current means of mechanical ventilation have on normal lungs, or the remaining normal areas of the lungs? A brief review of mechanical pulmonary ventilation in laboratory animals shows some surprises. Greenfield (4) has shown healthy dogs ventilated for but a few hours at 26-32 cm H20 airway pressure developing severe pulmoary atelectasis some 24 hours later. Unilaterally ventilating lungs to the same peak pressure of one hour showed atelectasis confined to that one lung, the other lung remaining normal. Such damaged lungs showed greatly elevated minimum surface tension values of saline lavage fluid. Other reports confirm the delicate nature of the lungs when ventilated at pressures in excess of 30 cm H20 (5).

Long term evaluation of gas exchange and hydrodynamic performance of a heparinized artificial lung: Comparison of two different hollow fiber pore sizes

We compared the performance of a heparinized hollow fiber artificial lung (Medtronic, Minimax) featuring standard hollow fibers (Group A) and experimental hollow fibers with a smaller pore size (Group B). Four sheep in each group underwent a veno-venous bypass for 72 hours. Every 6 hours, at 3 different blood flow rates (BFr) (400, 800, 1200 ml/min), at a constant gas flow rate (Gfr = 4 L/min), and at a constant blood inlet PCO2 (45-55 mmHg), we measured the oxygenation performance (O2 transfer = VO2 and blood outlet PO2 = PO2out), CO2 removal (CO2 transfer = VCO2 and PCO2 outlet = PCO2out) and pressure drop across the device (ΔP). A total of about 50 measurement sets were obtained for each group at different time points and blood flow rates. Both groups showed a good oxygenation performance (PO2out always higher than 200 mmHg) and no differences were observed between the two groups (at 1200 ml/min BFr, the average VO2 of all time points was 47 ± 15 ml/min in group A and 44 ± 11 ml/min in group B, mean ± SD, NS). During the first 24 hours, the VCO2 was higher in Group B than in Group A at each BFr (at 1200 ml/min BFr, 81 ± 18 vs 67 ± 20 ml/min, p<0.01), while no differences were observed during the subsequent 48 hours. Throughout the entire experiment, VCO2 increased with increasing BFr in both groups, (in group B, from 43 ± 14 ml/min at 400 ml/min BFr, to 73 ± 17 ml/min at 1200 ml/min BFr, average of all time points, p<0.01). In both groups the ΔP increased with the increasing BFr, but it was lower in Group B than in Group A at BFr 800 and 1200 ml/min (at 1200 ml/min BFr, 51 ± 15 mmHg vs 65 ± 17 mmHg, p<0.01), and remained stable for the entire experimental period.

Comment on the paper “The importance of the circuit capacity in the administration of CPAP”

The difference between the slope in the straight portion of the V/P curve and the TSLC was lower in group A 14.0 + 9.7 (data from the 7 pts presenting an inflection point) and in group B 15.0_+ 7 than in group C 21.96 _+ 8.4. Noting that 4 patients in group A did not present any inflection point and that the remaining 7 presented a lower slope TSLC it seems possible to suggest that patients in group A (undergoing LFPPV-ECCO2R ) presented not only a lower TSLC but also a straighter curve. Further studies are required in a larger number of patients to clarify this point. The number of patients appears too low to draw any conclusion about differences in terms of V/P curves, among survivors and non-survivors, in patients treated with LFPPV-ECCO2R. However, of 4 patients who did not present an inflection point in group A, 2 survived and 2 did not survive. The slope in the straight part of the V/P curve was slightly higher in survivors (38 _+ 13 ml cm H20-I) than in non-survivors (31 _+12 ml cm H2 Oi ) as was TSLC (28_+4; 25_+9), but in both cases the difference was not statistically significant.