Modelling Thirty-day Mortality in the Acute Respiratory Distress Syndrome (ARDS) in an Adult ICU

Variables predicting thirty-day outcome from Acute Respiratory Distress Syndrome (ARDS) were analysed using Cox regression structured for time-varying covariates. Over a three-year period, 1996–1998, consecutive patients with ARDS (bilateral chest X-ray opacities, PaO2/FiO2 ratio of <200 and an acute precipitating event) were identified using a prospective computerized data base in a university teaching hospital ICU.

The cohort, 106 mechanically ventilated patients, was of mean (SD) age 63.5 (15.5) years and 37% were female. Primary lung injury occurred in 45% and 24% were postoperative. ICU-admission day APACHE II score was 25 (8); ARDS onset time from ICU admission was 1 day (median: range 0-16) and 30 day mortality was 41% (95% CI: 33%-51%). At ARDS onset, PaO2/FiO2 ratio was 92 (31), 81% had four-quadrant chest X-ray opacification and lung injury score was 2.75 (0.45). Average mechanical ventilator tidal volume was 10.3 ml/ predicted kg weight. Cox model mortality predictors (hazard ratio, 95% CI) were: APACHE II score, 1.15 (1.09-1.21); ARDS lag time (days), 0.72 (0.58-0.89); direct versus indirect injury, 2.89 (1.45-5.76); PaO2/FiO2 ratio, 0.98 (0.97-0.99); operative versus non-operative category, 0.24 (0.09-0.63). Time-varying effects were evident for PaO2/FiO2 ratio, operative versus non-operative category and ventilator tidal volume assessed as a categorical predictor with a cut-point of 8 ml/kg predicted weight (mean tidal volumes, 7.1 (1.9) vs 10.7 (1.6) ml/kg predicted weight). Thirty-day survival was improved for patients ventilated with lower tidal volumes.

Survival predictors in ARDS were multifactorial and related to patient-injury-time interaction and level of mechanical ventilator tidal volume.

Evaluation of the physiological properties of ventilatory ratio in a computational cardiopulmonary model and its clinical application in an acute respiratory distress syndrome population

BACKGROUND

Owing to complexities of measuring dead space, ventilatory failure is difficult to quantify in critical care. A simple, novel index called ventilatory ratio (VR) can quantify ventilatory efficiency at the bedside. The study objectives were to evaluate physiological properties of VR and examine its clinical applicability in acute respiratory distress syndrome (ARDS) patients.

METHODS

A validated computational model of cardiopulmonary physiology [Nottingham Physiology Simulator (NPS)] was used to evaluate VR ex vivo in three virtual patients with varying degrees of gas exchange defects. Arterial P(CO₂) and mixed expired P(CO₂) were obtained from the simulator while either dead space or CO₂ production was altered in isolation. VR and deadspace fraction was calculated using these values. A retrospective analysis of a previously presented prospective ARDS database was then used to evaluate the clinical utility of VR. Basic characteristics of VR and its association with mortality were examined.

RESULTS

The NPS showed that VR behaved in an intuitive manner as would be predicted by its physiological properties. When CO₂ production was constant, there was strong positive correlation between dead space and VR (modified Pearson's r 0.98, P<0.01). The ARDS database had a mean VR of 1.47 (standard deviation 0.58). Non-survivors had a significantly higher VR compared with survivors [1.70 vs 1.34, mean difference 0.35, 95% confidence interval (CI) 0.16-0.56, P<0.01]. VR was an independent predictor of mortality (odds ratio 3.05, CI 1.35-6.91, P<0.01).

CONCLUSIONS

VR is influenced by dead space and CO₂ production. In ARDS, high VR was associated with increased mortality.

Serum levels of CC16, SP-A and SP-B reflect tobacco-smoke exposure in asymptomatic subjects

Since the 16-kDa bronchiolar Clara cell protein (CC16) and the alveolar surfactant-associated proteins (SP)-A and -B leak into the circulation when parenchymal health is disturbed, the aim of this study was to determine whether their serum levels could serve as early peripheral markers of tobacco smoke-induced epithelial injury. Sixty-nine (51 yrs (32-54) median (25-75th percentile)) nonsmokers and 54 (42 yrs (31-53)) asymptomatic smokers were enrolled in the study. Serum levels of SP-A did not differ between subjects (270 (208-389) versus 259 (168-392) microg x L(-1)), however, CC16 levels decreased (10.6 (8.7-14.6) versus 7.6 (6.0-11.2) microg x L(-1)) and SP-B levels increased (2,529 (2,091-2,943) versus 3,053 (2,613-4,188) microg x L(-1)) in the smokers. When tobacco smoke exposure, serum creatinine (renal index), age and sex were used as independent variables, CC16 was negatively influenced by cumulative smoking and positively influenced by age. SP-A and -B were negatively influenced by creatinine and positively influenced by cumulative smoking. Serum SP-B was inversely correlated with forced expiratory volume in one second/vital capacity, suggesting an association between obstructive disease and parenchymal lung health. The authors suggest that serum surfactant-associated proteins-A and -B reflect increased alveolocapillary leakage whereas Clara cell secretory protein 16 reflects tobacco smoke-induced Clara cell toxicity. Their evaluation may allow the effects of tobacco smoke on different levels of the respiratory tract, cellular toxicity and epithelial leakage to be distinguished.

Effect of CPAP on intrinsic PEEP, inspiratory effort, and lung volume in severe stable COPD

BACKGROUND

Intrinsic positive end expiratory pressure (PEEPi) constitutes an inspiratory threshold load on the respiratory muscles, increasing work of breathing. The role of continuous positive airway pressure (CPAP) in alleviating PEEPi in patients with severe stable chronic obstructive pulmonary disease is uncertain. This study examined the effect of CPAP on the inspiratory threshold load, muscle effort, and lung volume in this patient group.

METHODS

Nine patients were studied at baseline and with CPAP increasing in increments of 1 cm H(2)O to a maximum of 10 cm H(2)O. Breathing pattern and minute ventilation (I), dynamic PEEPi, expiratory muscle activity, diaphragmatic (PTPdi/min) and oesophageal (PTPoes/min) pressure-time product per minute, integrated diaphragmatic (EMGdi) and intercostal EMG (EMGic) and end expiratory lung volume (EELV) were measured.

RESULTS

Expiratory muscle activity was present at baseline in one subject. In the remaining eight, PEEPi was reduced from a mean (SE) of 2.9 (0.6) cm H(2)O to 0.9 (0.1) cm H(2)O (p<0.05). In two subjects expiratory muscle activity contributed to PEEPi at higher pressures. There were no changes in respiratory pattern but I increased from 9.2 (0.6) l/min to 10.7 (1.1) l/min (p<0.05). EMGdi remained stable while EMGic increased significantly. PTPoes/min decreased, although this did not reach statistical significance. PTPdi/min decreased significantly from 242.1 (32.1) cm H(2)O.s/min to 112.9 (21.7) cm H(2)O.s/min). EELV increased by 1.1 (0.3) l (p<0.01).

CONCLUSION

High levels of CPAP reduce PEEPi and indices of muscle effort in patients with severe stable COPD, but only at the expense of substantial increases in lung volume.

Elevated plasma surfactant protein-B predicts development of acute respiratory distress syndrome in patients with acute respiratory failure

Surfactant protein-B is a lung specific protein secreted into the air spaces by pulmonary epithelial type II cells that leaks into the bloodstream in increased amounts in patients with ARDS. To test whether elevated plasma levels of surfactant protein-B would predict the development of ARDS in patients with acute hypoxemic respiratory failure, plasma and lung injury scores were collected at study entry and daily thereafter for 3 d from 54 patients admitted to our intensive care unit. ARDS was defined as a new bilateral infiltrate on chest radiograph and a lung injury score > or = 2.5. Twenty patients developed ARDS, of whom seven died. Although the initial lung injury score was not predictive of ARDS, the initial plasma surfactant protein-B was predictive (area under the curve = 0.77 [0.63 to 0.90], nonparametric receiver-operating characteristic analysis). In this cohort, plasma surfactant protein-B was particularly predictive of ARDS when applied to patients suffering a direct lung insult (area under the curve = 0.87 [0.72 to 1.02]), with a sensitivity of 85% (95% CI: 55 to 98%) and specificity of 78% (40 to 97%) at a cutoff of 4,994 ng/ml.

Lung function, permeability, and surfactant composition in oleic acid-induced acute lung injury in rats

Although acute lung injury (ALI) is associated with inflammation and surfactant dysfunction, the precise sequence of these changes remains poorly described. We used oleic acid to study the pathogenesis of ALI in spontaneously breathing anesthetized rats. We found that lung pathology can occur far more rapidly than previously appreciated. Lung neutrophils were increased approximately threefold within 5 min, and surfactant composition was dramatically altered within 15 min. Alveolar cholesterol increased by approximately 200%, and even though disaturated phospholipids increased by approximately 30% over 4 h, the disaturated phospholipid-to-total phospholipid ratio fell. Although the alveolocapillary barrier was profoundly disrupted after just 15 min, with marked elevations in lung fluid ((99m)Tc-labeled diethylenetriamine pentaacetic acid) and (125)I-labeled albumin flux, the lung rapidly began to regain its sieving properties. Despite the restoration in lung permeability, the animals remained hypoxic even though minute ventilation was increased approximately twofold and static compliance progressively deteriorated. This study highlights that ALI can set in motion a sequence of events continuing the respiratory failure irrespective of the alveolar surfactant pool size and the status of the alveolocapillary barrier.

Non.Invasive Ventilation for Adult Acute Respiratory Failure. Part II

OBJECTIVE

To discuss the clinical indications and complications of non-invasive ventilation.

DATA SOURCES

A review of articles published in peer-reviewed journals from 1966 to 1998 and identified through a MEDLINE search on non-invasive ventilation.

SUMMARY OF REVIEW

Non-invasive ventilation (NIV) has been used in patients with respiratory failure caused by cardiogenic pulmonary oedema, acute respiratory distress syndrome, acute asthma and chronic obstructive pulmonary disease. However, in patients with acute respiratory failure, it appears that acute cardiogenic pulmonary oedema and acute respiratory failure associated with Pneumocystis carinii pneumonia are the only disorders in which significant benefits have been associated with the use of the NIV mode of CPAP. The potential clinical benefit of CPAP in acute asthma and blunt chest trauma remains unclear. Pressure support ventilation is beneficial in patients with hypercapnic acute respiratory failure (ARF) secondary to respiratory muscle insufficiency, high inspiratory work loads, or reduced alveolar ventilation. It appears also to be associated with an improved outcome in COPD patients with hypercapnic ARF.

CONCLUSIONS

Non-invasive ventilation using the modes of CPAP, PSV, BiPAP and NIPPV should be considered in patients with respiratory disorders who remain in acute respiratory failure despite conventional therapy, before considering invasive mechanical ventilation.

Non.invasive ventilation for adult acute respiratory failure. Part I

OBJECTIVE

To detail the history, modes, physiological effects, and circuit geometry of non-invasive ventilation.

DATA SOURCES

A review of articles published in peer-reviewed journals from 1966 to 1998 and identified through a MEDLINE search on non-invasive ventilation.

SUMMARY OF REVIEW

Non-invasive ventilation (NIV) has been used for many years as an adjunct to standard therapy in patients with acute and chronic respiratory disorders. The newer modes of NIV which include continuous positive airway pressure (CPAP), pressure support ventilation (PSV), BiPAP (bi-level positive airway pressure) and controlled and assisted modes of intermittent non-invasive positive pressure ventilation (NIPPV) have additional advantages and are often used routinely in many respiratory diseases. These modes of ventilatory support have been found to improve arterial oxygenation, ventilation, work of breathing, and cardiac function, in patients with respiratory failure, although in normal subjects, respiration is often impaired..

CONCLUSIONS

Non-invasive ventilation using the modes of CPAP, PSV, BiPAP and NIPPV should be considered in patients with respiratory failure who are unresponsive to conventional therapy, before considering invasive mechanical ventilation.

Partitioning lung and plasma proteins: circulating surfactant proteins as biomarkers of alveolocapillary permeability

1. The alveolocapillary membrane faces an extraordinary task in partitioning the plasma and lung hypophase proteins, with a surface area approximately 50-fold that of the body and only 0.1-0.2 micron thick. 2. Lung permeability is compromised under a variety of circumstances and the delineation between physiological and pathological changes in permeability is not always clear. Although the tight junctions of the epithelium, rather than the endothelium, are regarded as the major barrier to fluid and protein flux, it is becoming apparent that the permeability of both are dynamically regulated. 3. Whereas increased permeability and the flux of plasma proteins into the alveolar compartment has dire consequences, fortuitously the flux of surfactant proteins from the airspaces into the circulation may provide a sensitive means of non-invasively monitoring the lung, with important implications for treatment modalities. 4. Surfactant proteins are unique in that they are present in the alveolar hypophase in high concentrations. They diffuse down their vast concentration gradients (approximately 1:1500-7000) into the circulation in a manner that reflects lung function and injury score. Surfactant proteins vary markedly in size (approximately 20-650 kDa) and changes in the relative amounts appear particularly diagnostic with regard to disease severity. Alveolar levels of surfactant proteins remain remarkably constant despite respiratory disease and, unlike the flux of plasma proteins into the alveolus, which may reach equilibrium in acute lung injury, the flux of surfactant proteins is unidirectional because of the concentration gradient and because they are rapidly cleared from the circulation. 5. Ultimately, the diagnostic usefulness of surfactant proteins as markers of alveolocapillary permeability will demand a sound understanding of their kinetics through the vascular compartment.

A simple bedside approach to measurement of respiratory mechanics in critically ill patients

OBJECTIVE

To describe and evaluate clinically applicable approaches to measurement of respiratory mechanics in critically ill patients.

DATA SOURCES

Methodological and evaluation studies of respiratory mechanics in critically ill patients from relevant MEDLINE searches.

SUMMARY OF REVIEW

In ventilated subjects clinically important respiratory system mechanics can be measured using airway pressure and flow data. However, since the respiratory system consists of the lung and chest wall, and chest wall mechanics can markedly alter respiratory system mechanics, it is preferable to compartmentalise these parameters with concurrent measurement of oesophageal, and preferably gastric pressure. Additional care must be taken with interpretation of these data since elastance and resistance may be influenced by frequency, volume, volume history and flow. Tissue viscoelasticity and non-homogeneity of regional time constants are responsible for stress adaptation, which can be measured simply, and accounts for some of these effects on elastance and resistance, and for a systematic difference between static and dynamic intrinsic PEEP. Elastance can be measured using the end-inspiratory occlusion technique, or from either static or dynamic volume-pressure curves. PEEP-mediated recruitment can be measured following referencing of these curves to FRC. Similarly, resistance can be measured from either end-inspiratory occlusion or dynamic pressure and flow data.

CONCLUSIONS

Some of this information is available on modern ventilators, but greater insight requires measurement and manipulation of flow and pressure data using a pneumotachograph and pressure transducers. Given the importance of respiratory mechanics in the management of many critically ill patients, and given how poorly the respiratory system is monitored compared with the cardiovascular system, it is worth considering making this simple but additional effort.

Measurement of tidal volume by using transthoracic impedance variations in rats

The application of impedance pneumography for monitoring respiration in small animals has been limited by problems with calibration. With improved instrumentation, we describe the calibration of tidal volume in anesthetized rats. The detection of changes in voltage, reflecting the electrical impedance variations associated with respiration, was optimized by using disposable adhesive silver-silver chloride electrodes, advanced circuitry, and analog-to-digital recording instrumentation. We found a linear relationship between change in impedance and tidal volume in individual rats (R2 >/= 98%), which was strongly influenced by rat weight. Consequently, a calibration equation incorporating change in impedance and rat weight was derived to predict tidal volume. Comparison of the predicted and true tidal volumes revealed a mean R2 >/= 98%, slopes of approximately 1, intercepts of approximately 0, and bias of approximately 0.07 ml. The predicted volumes were not significantly affected by either frequency of respiration or pulmonary edema. We conclude that impedance pneumography provides a valuable tool for the noninvasive measurement of tidal volume in anesthetized rats.

The use of antimicrobials in ten Australian and New Zealand intensive care units. The Australian and New Zealand Intensive Care Multicentre Studies Group Investigators

A prospective standardized collection of clinical, microbiological and pharmaceutical information on antibiotic use was conducted in Australia and New Zealand intensive care units (ICUs) involving 481 consecutive critically ill patients who were receiving antibiotics for any reason while in ICU. Patients had a mean SAPS II score of 34.1 +/- 17.8 with an expected mortality of 15.6% (actual mortality 12%). Of these, 292 (60.8%) were admitted to the ICU within 72 hours of surgery. Among such surgical patients, 233 (79.9%) received antibiotics for "surgical prophylaxis" while in ICU (48% of sample population). The second largest group of patients treated with antibiotics in ICU included those with systemic inflammatory response syndrome and clinical suspicion of infection (38%). Antibiotics were prescribed for the treatment of clinically diagnosed infection in 268 patients. Clinical response was apparent in 62.6% and in most (71%) was achieved in the first 72 hours of treatment. The incidence of antimicrobial-related side-effects was 4%, mostly in the form of diarrhoea or rash (75% of all side-effects). The most commonly prescribed antimicrobials were gentamicin (n = 146), ceftriaxone (n = 98), vancomycin (n = 94) and metronidazole (n = 111). Three times daily prescription of aminoglycosides was uncommon (< 1%). Forty-one patients had a documented infection (positive culture) with a gram-negative organism. Of these, 17 received therapy with a single antibiotic and 24 received therapy with two antibiotics. Despite similar illness severity, there were six deaths in the former group and only two in the latter.

Respiratory mechanics and surfactant in the acute respiratory distress syndrome

1. Although abnormalities in pulmonary surfactant were initially implicated in the pathogenesis of the acute respiratory distress syndrome (ARDS) 30 years ago, most subsequent research has focused on mediators of the parenchymal acute lung injury (ALI) and the associated increase in alveolocapillary permeability. 2. Surfactant is essential for normal breathing and the severity of ALI correlates with surfactant dysfunction and abnormalities in surfactant composition; however, no relationship has been shown with respiratory system compliance. In neonates and most animal models, respiratory system compliance will directly reflect the elastic properties of the lung. However, the greater vertical height of the chest wall in adults, in combination with the increase in lung density due to ALI, results in dependent collapse of alveoli. Because simple, global measurement of compliance is strongly influenced by the volume of aerated lung, alternative measures of respiratory mechanics may reflect surfactant dysfunction. 3. Using a dynamic, volume-dependent model of respiratory mechanics to indirectly reflect this heterogeneous inflation, we have found direct relationships with surfactant composition in patients with ARDS. A failure of surfactant to increase surface tension in large alveoli may also explain why lung overdistension occurs at relatively low pressures. Furthermore, surfactant dysfunction will exaggerate heterogeneous lung inflation, augmenting regional overinflation, and is essential for ALI secondary to repetitive opening and closing of alveoli during tidal ventilation. 4. Ventilation-induced ALI has also been shown to result in massive increases in pro-inflammatory cytokines within the lung. Because ALI itself fails to compartmentalize cytokines, with spillover into the systemic circulation resulting in distant organ dysfunction, surfactant dysfunction may have widespread implications.

Clearance of Clara cell secretory protein 16 (CC16) and surfactant proteins A and B from blood in acute respiratory failure

Surfactant proteins A and B (SP-A and SP-B) enter the circulation in a manner that acutely reflects changes in pulmonary function in patients with acute respiratory failure (ARF). There is a small but significant gradient in SP-A and SP-B from arterial to mixed venous (A-V) blood, and since we have detected both proteins in urine, the kidney may be a major site of their systemic clearance. Clara cell secretory protein 16 (CC16), which leaks from the respiratory tract, is known to be freely eliminated by the kidney. Lung plasma protein levels will depend on the rates of both protein entry into and clearance from plasma. In order to study the limiting variable determining these levels, we compared plasma CC16, SP-A, and SP-B in matching A-V blood samples from 37 ARF patients with indices of lung dysfunction and glomerular filtration rate (GFR) (of plasma cystatin C and creatinine). Cystatin C, CC16, SP-A, and SP-B were reduced in mixed venous plasma (all p < 0.001) and their A-V gradients were directly related to their arterial levels (all p < 0.03). Whereas CC16, SP-A, and SP-B reflected blood oxygenation (all p < 0.05), only SP-A and SP-B were related to lung injury score (LIS) (both p < 0.05). In contrast, whereas the clearances of both CC16 and cystatin C were related to that of creatinine (p < 0.02 for both), the clearances of SP-A and SP-B were not. Our study confirms that all three lung proteins are acutely cleared from the circulation of patients with ARF (half-lives < 18 min), and we conclude that whereas the plasma concentration of CC16 depends on GFR, plasma concentrations of SP-A and SP-B reflect lung function independently of this variable.

Measurement of overinflation by multiple linear regression analysis in patients with acute lung injury

Strategies to optimize alveolar recruitment and prevent lung overinflation are central to ventilatory management of patients with acute lung injury (ALI). The recent description of overinflation using multilinear regression analysis of airway pressure (Paw) and flow (V') data allows a functional assessment of lung mechanics. However, this technique has not been studied in ALI patients. During 15 positive end-expiratory pressure (PEEP) trials in 10 ALI patients, respiratory elastance was partitioned into volume-independent (E1) and volume-dependent (E2VT) components, where Paw=(E1+E2VT)V+RrsV'+Po; where V is volume, VT is tidal volume, Rrs is respiratory resistance and Po is static recoil pressure at end-expiration (equivalent to total PEEP). Then, %E2 was calculated as (100E2VT)/(E1+E2VT); a measure of lung overinflation when %E2>30%. Alveolar recruitment, assessed as a PEEP-induced increase in V>50 mL at a constant Paw occurred in 14 of 15 trials (299+/-34 mL, mean+/-SEM), but was independent of the degree of lung inflation. Lung overinflation was common (six of 15 clinically set PEEP levels) and occurred despite a dynamic elastic distending pressure (Pel,dyn) <30 cmH2O during 18 of 36 PEEP titrations. During a PEEP titration the resultant %E2 was directly related to delta(peak airway pressure-Po) (rs=0.86, p<0.001) and delta(Pel,dyn-Po) (rs=0.89, p<0.001). The 95% predictive intervals for a 2 cmH2O change in either driving pressure were %E2 values of 30.4-68.1% and 32.8-69.2%, respectively. Single or continuous measurement of %E2 (a measure of lung inflation) is a readily available method for titrating ventilatory parameters. Further, during a positive end-expiratory pressure titration a change in ventilatory driving pressure > or =2 cmH2O is indicative of overinflation.

Surfactant composition reflects lung overinflation and arterial oxygenation in patients with acute lung injury

Pulmonary surfactant abnormalities have consistently been documented in patients with acute lung injury (ALI), however, there is little evidence directly correlating them to altered respiratory mechanics. To explore this further, surfactant composition was measured in lung aspirate fluid collected on 15 occasions from 10 patients with ALI. The composition was compared with lung aspirate fluid from 11 intubated patients prior to elective cardiac surgery (CS), and bronchoalveolar lavage fluid from 16 normal subjects. In both the ALI and cardiac groups the proportion of disaturated phospholipids (DSP) and phosphatidylcholine was reduced. Plasma levels of surfactant proteins-A and -B (SP-A and -B) were elevated, but were unrelated to alveolar surfactant levels. In the ALI group, and the ALI + CS group, DSP, normalized to the total phospholipid content, sphingomyelin (SPH), and urea, showed strong direct correlations with arterial oxygen tension/inspiratory oxygen fraction (all p < or = 0.01). In the ALI group, normalized DSP was also directly related to the elastance of the positive end-expiratory pressure-induced increase in the end-expiratory lung volume (all p < or = 0.02), and indirect correlations were found with a measure of lung overinflation (%E2; all p < or = 0.01). We conclude that surfactant composition correlates with lung function abnormalities in acute lung injury and cardiac patients, and that both groups had elevated plasma surfactant proteins-A and -B levels, consistent with a concurrent increase in alveolocapillary permeability.

Haemolysis associated with continuous venovenous renal replacement circuits

Extracorporeal circuits can cause haemolysis resulting in an increase in plasma-free haemoglobin (PFHb). High pressures and clots within the circuit have been identified as factors increasing the likelihood of haemolysis. Continuous venovenous haemodiafiltration (CVVHD) is associated with high circuit pressures as the pump-driven circuit clots over a period of time. PFHb was measured during CVVHD to determine if circuit life, maximum circuit pressure or the clotting of the haemofilter was associated with evidence of haemolysis. Circuit life up to 50 hours, circuit pressures or haemofilter clotting had no significant effect on PFHb. There was a small rise in PFHb in the circuits lasting beyond 50 hours. CVVHD circuits can be run up to 50 hours without concern for haemolysis.

Surfactant proteins-A and -B are elevated in plasma of patients with acute respiratory failure

Surfactant protein-A (SP-A) leaks into the circulation of patients with acute respiratory distress syndrome (ARDS) or acute cardiogenic pulmonary edema (APE) in a manner inversely related to lung function. Since surfactant protein-B (SP-B) is synthesized as a precursor considerably smaller than alveolar SP-A, we investigated whether it enters the circulation more readily. Reactivities consistent with SP-B proprotein (approximately 42 to approximately 45 kD) and the approximately 25 kD processing intermediate were detected in plasma. Plasma immunoreactive SP-B levels were significantly higher in ARDS (8,007+/-1,654 ng/ml [mean+/-SEM], n = 22) and APE (3,646+/-635 ng/ml, n = 10) patients compared with normal subjects (1,685+/-58 ng/ml, n = 33) and ventilated patients with no cardiorespiratory disease (1,829+/-184 ng/ml, n = 7). All groups had plasma SP-B/SP-A ratios approximately 6- to approximately 8-fold higher than in normal lavage or ARDS tracheal aspirate fluid, consistent with protein sieving. During admission, both plasma SP-B and the SP-B/SP-A ratio were inversely related to blood oxygenation (PaO2/FIO2) (p < 0.0001 and p < 0.025, n = 260 from 39 patients; Spearman) and static respiratory system compliance (deltaV/deltaP) (p < 0.0001 and p < 0.01, n = 168 from 25 patients). We describe in detail three patients and conclude that immunoreactive SP-B enters more readily than SP-A, is cleared acutely, and provides a better indicator of lung trauma.

Endotoxin alters the systemic disposition of nitric oxide synthase inhibitors in the awake sheep

1. We evaluated the haemodynamic effects and systemic disposition of the nitric oxide synthase (NOS) inhibitor NL-nitro-L-arginine (NOLA) after intravenous (i.v.) administration of two different doses (5 and 20 mg/kg) in awake healthy sheep and awake sheep given a continuous i.v. infusion of endotoxin (lipopolysaccharide, 12 ng/kg per h, i.v., for 18 h). In addition, we determined the systemic disposition of another NOS inhibitor, NL-nitro-L-arginine methylester (L-NAME; 20 mg/kg, i.v.) in awake healthy sheep only. 2. NL-Nitro-L-arginine produced a dose-dependent decrease in heart rate (HR) and cardiac output (CO) together with a dose-dependent increase in mean arterial pressure (MAP) and peripheral vascular resistance (PVR) when compared to baseline. In endotoxic sheep NOLA produced a greater increase in MAP and mean pulmonary arterial pressure (MPAP). 3. In healthy sheep there was a dose-related increase in total body clearance (Cl) of NOLA. The Cl increased from 0.028 L/min after the lower dose to 0.032 L/min after the higher dose. The infusion of endotoxin caused an increase in Cl of NOLA to 0.040 and 0.047 L/min, respectively, and a decrease in plasma slow half-life (t1/2) from 825 to 546 min and from 780 to 453 min, respectively. 4. NL-Nitro-L-arginine methylester was rapidly cleared from the plasma with a slow half-life of approximately 7.5 min and there was a simultaneous appearance of NOLA in the plasma. 5. These results support the view that nitric oxide has a significant role in regulating vascular tone in healthy and endotoxic sheep and indicate that the increases in Cl of NOLA with an increase in its dose and the presence of endotoxin will be important in influencing appropriate dosage regimens in clinical studies.

Surfactant replacement therapy in ARDS: white knight or noise in the system?

Although one would predict that surfactant replacement therapy would be effective in acute respiratory distress syndrome (ARDS), a recent large trial proved unsuccessful, possibly reflecting the nature of the surfactant used. Given the importance of the unique proteins in the action of surfactant, these would seem vital components of any exogenous surfactant. The ability to identify patients at risk of ARDS and to characterise their surfactant might allow prophylactic treatment with a nebulised, complementary, tailor-made preparation of surfactant. Advanced cases might undergo bronchoscopic focal lavage to remove plasma proteins and inflammatory mediators prior to focal instillation of surfactant to areas of greatest need. Ventilation regimens might be adjusted both to minimise trauma and to conserve endogenous surfactant.

Aerosol kinetics and bronchodilator efficacy during continuous positive airway pressure delivered by face mask

BACKGROUND

Rates of fresh gas flow (FGF) commonly used when continuous positive airway pressure (CPAP) is delivered by face mask theoretically reduce the delivery and availability of therapeutic aerosols. As it may be hazardous for patients with acute respiratory failure to interrupt mask CPAP, the effects of CPAP on aerosol kinetics and bronchodilator efficacy were investigated.

METHOD

The effect of CPAP at 10 cm H2O at a FGF rate of 50 l/min on the delivery of technetium labelled aerosol generated from a readily available jet nebuliser was measured using a bench model of spontaneous respiration. In a separate clinical study the bronchodilator responses to incremental doses of nebulised salbutamol were measured in nine stable asthmatic subjects in a random sequence of conventional nebulisation (control) or nebulisation whilst receiving CPAP via a tight fitting face mask. Each patient acted as his or her own control.

RESULTS

CPAP significantly reduced total aerosol delivery to the face mask from 6.85 (1.52)% to 1.3 (0.37)% of the initial nebuliser charge. In the clinical study a significant bronchodilator response to nebulised salbutamol was seen during both conventional nebulisation and nebulisation whilst receiving CPAP by face mask. The shape of the dose-response curves and the magnitude of the total increase in the forced expiratory volume in one second (FEV1) was identical for CPAP and control conditions.

CONCLUSIONS

Despite a reduction in aerosol presented to the proximal airway, the bronchodilator response to inhaled beta 2 agonists in stable asthmatic subjects was not affected when CPAP was delivered by face mask. Despite a high rate of FGF, nebulised beta 2 agonists are effective when administered in conjunction with CPAP delivered by face mask.

Continuous renal replacement therapy in critically ill patients: monitoring circuit function

There is currently no universally accepted method to monitor circuit function or guidelines for circuit replacement during continuous renal replacement therapies (CRRT). The objectives of this study were to diagnose the causes of circuit failure, identify factors responsible for circuit clotting and determine a predictive monitor of circuit function. The CRRT technique used in this study was continuous venovenous haemodialysis (CVVHD). Continuous monitoring of circuit pressures (pre- and post-haemofilter and their difference: the transfilter pressure gradient) was used to diagnose the causes of circuit failure. In circuits ceasing due to clotting, the factors thought to contribute, anticoagulation, haematocrit and platelet count, were measured at the commencement of CVVHD and every eight hours thereafter until circuit failure. Monitors of circuit function, creatinine clearance and plasma to diafiltrate urea ratio were measured every eight hours and compared to the transfilter pressure gradient. During a three-month period data was collected on five consecutive patients (41 consecutive haemofilters). Clotting of the haemofilter (63%) and air detection chamber (7.5%) were the most common identifiable causes of circuit failure. The duration of their circuit life was described using multiple regression analysis, i.e. hours of filter life = -82.8 + (delta platelet count x 0.25) + (delta haematocrit x 3.6) + (circuit flow [ml/min] x 4) R2 = 0.77. A rise in transfilter pressure gradient and a fall in haemofilter function discriminated clotted filters with falling function (decrease in creatinine clearance and urea ratio) from unclotted filters. In any circuit an increase of 26 mmHg or more in the transfilter pressure gradient accurately predicted circuit failure due to clotting and imminent cessation of function. Increases in platelet count, haematocrit, and low circuit flows are important determinants of haemofilter life. The measurement of transfilter pressure gradient across the haemofilter is an accurate bedside monitor of circuit function.

Vasoactive drugs and the importance of renal perfusion pressure

Despite the often multifactorial nature of renal insults in critically ill patients, inadequate renal blood flow (RBF) is common and frequently causes a reduction in the glomerular filtration rate (GFR). Renal autoregulation acts to maintain both the RBF and GFR constant across a broad range of renal perfusion pressure (RPP) levels; however, the lower limit of this range (approximately 80 mm Hg for RBF, and 10-15 mm Hg higher for GFR) is often above the RPP achieved in critically ill patients. Furthermore, renal autoregulation is often lost, resulting in a linear pressure-flow relationship in the "at-risk" kidney. Consequently, maintenance of an adequate RPP level is needed to optimize RBF. While this may require the use of vasopressor catecholamines with their attendant risk of renal vasoconstriction and a reduction in RBF, both laboratory studies and clinical data suggest that such reactions rarely occur with intravenous infusions of these drugs, and that RBF and renal function usually improve when RPP is augmented during shock. Preliminary data, using nitric oxide (NO.) synthase inhibitors to augment blood pressure, showed a detrimental effect on renal perfusion, perhaps due to the central role of NO. in the normal vasoregulation of the kidney. Dopaminergic agonists have been commonly used as renal vasodilators; however, their actions are complex and include a proximal tubular diuretic effect, renal vasodilation, and systemic hemodynamic effects. Their specific action to increase RBF and GFR has not been demonstrated in clinically relevant studies and no prospective randomized study has shown a reduction in the incidence of renal impairment or acute renal failure.

Serum surfactant protein-A levels in patients with acute cardiogenic pulmonary edema and adult respiratory distress syndrome

Detection of alveolo-capillary damage has important implications for treatment modalities in ventilated patients. Although surfactant protein-A (SP-A) is normally only found in appreciable amounts in the lung, we describe significantly elevated concentrations in the sera of patients with acute cardiogenic pulmonary edema (median, 250 ng/ml; range, 180 to 364; n = 10) and in those with the adult respiratory distress syndrome (ARDS) (median, 378 ng/ml; range, 215 to 1,378; n = 15) relative to healthy control subjects (median, 175 ng/ml; range, 123 to 248; n = 15) and ventilated patients with no cardiorespiratory disease (median, 169 ng/ml; range, 126 to 253; n = 6) (p < 0.01, in all cases). Serum SP-A was inversely related to blood oxygenation and to static respiratory system compliance both at the time of the patient's entry into the study (p < 0.005, rs = -0.51, n = 31; p < 0.001, rs = 0.82, n = 17; respectively) and during the course of admission (p < 0.001, rs = -0.34, n = 168; p < 0.001, rs = -0.50, n = 111; respectively). In addition, we describe in detail three cases of ARDS where lung function either improved, remained static, or deteriorated. We conclude that serum SP-A is an acute indicator of lung function and alveolo-capillary membrane injury.

Renovascular interaction of epinephrine, dopamine, and intraperitoneal sepsis

OBJECTIVE

To determine the effect of intraperitoneal sepsis on the systemic and renal actions of the continous infusion of epinephrine or dopamine, and during the concurrent administration of both drugs.

DESIGN

Prospective, randomized study.

SETTING

Laboratory at a university hospital.

SUBJECTS

Seven conscious, chronically catheterized, adult merino sheep.

INTERVENTIONS

Epinephrine at 40 micrograms/min or dopamine at 2 micrograms/kg/min, or both drugs concurrently were infused for 4 hrs on separate study days in healthy sheep. This protocol was then repeated following the induction of sepsis after the intraperitoneal injection of 10(11) Escherichia coli, 10(12) Bacteroides fragilis, and bran.

MEASUREMENTS AND MAIN RESULTS

Systemic oxygen delivery (DO2) and consumption were measured using thermodilution cardiac output and measured oxygen content. Renal blood flow was measured using an electromagnetic flow transducer, and creatinine clearance was calculated as the quotient of renal blood flow and the renal extraction ratio of creatinine. Infusion of epinephrine augmented systemic DO2 and mean arterial pressure (MAP) during both healthy and septic studies. Systemic oxygen consumption was only increased during epinephrine infusion in the septic study. During the healthy animal study, renal blood flow was initially decreased during epinephrine infusion, but increased to 36% above baseline (p = .003). However, creatinine clearance remained unchanged. During the experimental sepsis study, the infusion of epinephrine had less marked effects on renal blood flow (unchanged from baseline), while an initial reduction (15 mins) in creatinine clearance (p = .04) was not sustained and had returned to baseline by 3 hrs. Dopamine alone produced no change in systemic oxygen variables or MAP during the studies on healthy or septic animals. Although dopamine produced renal vasodilation and an increase in renal blood flow in the healthy state, these results were not found during the septic state. In addition, concurrent infusion of dopamine with epinephrine did not alter the systemic or renal effects of epinephrine during the healthy or septic states.

CONCLUSIONS

These results do not support the routine use of low-dose dopamine, and demonstrate a change in renovascular responses to catecholamines during intraperitoneal sepsis. The infusion of epinephrine at 40 micrograms/min had few deleterious effects on the kidney, and augmented both MAP and systemic DO2. Its role as a catecholamine in the management of sepsis may need to be reconsidered.

Determination of NG-nitro-L-arginine and NG-nitro-L-arginine methyl ester in plasma by high-performance liquid chromatography

An HPLC method has been developed for the measurement of the nitric oxide synthase inhibitors, NG-nitro-L-arginine (L-NOLA) and NG-nitro-L-arginine methyl ester (L-NAME), in sheep plasma. Using an ion-exchange HPLC column (JWAS 150, 100 x 3.9 mm I.D., Millipore-Waters, Australia) and post-column ninhydrin detection, L-NOLA was separated from valine and other plasma amino acids. When added to sheep plasma, good recovery (mean 102%) and precision (mean coefficient of variation 2.7%) in the measurement of L-NOLA was obtained over the range 2-50 mg/l. L-NAME was unstable in sheep plasma at 37 degrees C, and was converted to L-NOLA with a half-life of 250 min. This method will permit pharmacokinetic parameters to be determined for these potential drugs, and will allow plasma drug concentrations to be correlated with the pharmacodynamic effects of these compounds.

Epinephrine infusion in sheep: systemic and renal hemodynamic effects

OBJECTIVE

To evaluate the dose-response effects of graded epinephrine infusions on systemic and renal hemodynamics.

DESIGN

Prospective, dose-response study.

SETTING

Laboratory at a university hospital.

SUBJECTS

Thirteen conscious, chronically catheterized, adult merino sheep.

INTERVENTIONS

Ten sheep received five infusions of epinephrine (5, 10, 20, and 40 micrograms/min; the 40-micrograms/min dose was repeated) and a placebo (saline) on separate days (at least 1 day apart). Each drug infusion was administered for 4 hrs on separate days after a 90-min baseline was established. Plasma catecholamine values and renin activity were measured in an additional three sheep infused with 40 micrograms/min epinephrine.

MEASUREMENTS AND MAIN RESULTS

Renal blood flow was measured, using an electromagnetic flow transducer; these data, along with aortic and pulmonary arterial pressure, were continuously recorded after analog-to-digital conversion. Cardiac output was intermittently measured by thermodilution. Epinephrine resulted in a dose-dependent increase in mean arterial pressure (p < .001), and in cardiac output at 30, 60, and 120 mins after the start of the infusion. Concurrently, systemic vascular resistance was initially depressed below baseline, but then gradually increased during the 4-hr infusion period. Although a dose-dependent increase in renal vascular resistance was found, 5- and 10-micrograms/min of epinephrine failed to alter renal vascular resistance. However, 20- and 40-micrograms/min of epinephrine increased renal vascular resistance by 77% and 94% respectively, at 10 mins, but these values decreased to 17% and 16% of baseline by 120 mins. Consequently, an early dose-dependent decrease in renal blood flow was also time dependent, with renal blood flow increasing back to or above baseline at all studied infusion rates of epinephrine.

CONCLUSIONS

Using a clinically relevant dose regimen, epinephrine increased mean arterial pressure and cardiac output. Renal blood flow decreased transiently, but returned to baseline within 30 to 60 mins.

Intensive care costing methodology: cost benefit analysis of mask continuous positive airway pressure for severe cardiogenic pulmonary oedema

Costing data for intensive care admissions is important, not only for unit funding, but also for cost outcome analysis of new therapies. This paper presents an intensive care episode costing methodology using the example of a cost-benefit analysis of mask CPAP for severe cardiogenic pulmonary oedema (CPO). This analysis examines the intervention of admitting all patients with severe CPO to the intensive care unit for mask CPAP, compared with the previous practice of admitting only patients failing conventional non-CPAP treatment and requiring mechanical ventilation. The episode costs were determined from a prospective study which showed mask CPAP reduced the need for mechanical ventilation from 35% to 0%. The mean cost of a mask CPAP episode was $1,156, with a mean stay of 1.2 days, compared with ventilated patients, $5,055 and 4.2 days. The major contributors to cost in both groups were nursing and medical salaries, and hospital overheads. The cost of previous estimated yearly caseload of 35 ventilated patients ($176,925) was greater than the cost associated with an increased caseload of 100 mask CPAP patients ($115,600). We conclude that, despite an increase in admissions, mask CPAP for severe CPO is cost-effective.

Efficacy of pressure support in compensating for apparatus work

Breathing through an endotracheal tube, connector, and ventilator demand valve imposes an added load on the respiratory muscles. As respiratory muscle fatigue is thought to be a frequent cause of ventilator dependence, we sought to examine the efficacy of five different ventilators in reducing this imposed work through the application of pressure support ventilation. Using a model of spontaneous breathing, we examined the apparatus work imposed by the Servo 900-C, Puritan Bennett 7200a, Engstrom Erica, Drager EV-A or Hamilton Veolar ventilators, a size 7.0 and 8.0 mm endotracheal tube, and inspiratory flow rates of 40 and 60 l/min. Pressure support of 0, 5, 10, 15, 20 and 30 cm H2O was tested at each experimental condition. Apparatus work was greater with increased inspiratory flow rate and decreased endotracheal tube size, and was lowest for the Servo 900-C and Puritan Bennett 7200a ventilators. Apparatus work fell in a curvilinear fashion when pressure support was applied, with no major difference noted between the five ventilators tested. At an inspiratory flow rate of 40 l/min, a pressure support of 5 and 8 cm H2O compensated for apparatus work through size 8.0 and 7.0 endotracheal tubes and the Servo 900-C and Puritan Bennett 7200a ventilators. However, the maximum negative pressure was greater for the Servo 900-C. The added work of breathing through endotracheal tubes and ventilator demand valves may be compensated for by the application of pressure support. The level of pressure support required depends on inspiratory flow rate, endotracheal tube size, and type of ventilator.

Prospective evaluation of residents and nurses as severity score data collectors

OBJECTIVES

To determine the interobserver reliability of residents and nurses collecting Acute Physiology and Chronic Health Evaluation (APACHE II) data and the subsequent effect of these data collections on individual patient mortality prediction.

DESIGN

In a prospective study, residents and nurses independently collected data to derive APACHE II scores. When their scores differed, a standard score was determined by one of the investigators.

SETTING

A general medical and surgical ICU.

PATIENTS

A total of 120 consecutive patients were included; of these patients, 79 had standard scores determined because resident and nurse scores differed.

MAIN RESULTS

There was overall agreement between the residents and nurses with no significant difference between mean APACHE II scores or mean predicted mortality rates. Intraclass correlation coefficients confirmed good overall agreement between observer groups for predicted mortality rate: resident vs. nurse r2 = .94, resident vs. standard r2 = .94, and nurse vs. standard r2 = .90. However, clinically significant lack of agreement was demonstrated in 5% of the patients by the 95% confidence limits of agreement: resident vs. nurse -14 to +14%, resident vs. standard -10 to +14%, and nurse vs. standard -14 to +20%.

CONCLUSIONS

While interobserver variability between resident and nurse data collection has minimal effect on derived predicted mortality rate with large patient groups, significant variability may occur in individual patients. Residents were more accurate data collectors than nurses.

The effect of various sympathomimetics on the regional circulations in hyperdynamic sepsis

Because sepsis is characterized by a depression in vascular reactivity, we hypothesized that changes in organ blood flows (Q) would differ between the nonseptic and septic state during the infusion of sympathomimetics. Therefore we examined the (sepsis x organ Q) interaction during the infusion of five sympathomimetics in 36 mature, awake sheep before and after cecal ligation and perforation produced hyperdynamic sepsis. A 3-hour infusion of dobutamine, norepinephrine, dopamine, dopexamine, or salbutamol was compared with that of placebo during both nonseptic and septic studies; drug infusion was titrated to an increase in cardiac index of greater than 20%. Increased plateau infusion doses of norepinephrine (+305%), salbutamol (+275%), dopamine (+70%), and dobutamine (+49%) were required to achieve predefined treatment guidelines during the septic versus nonseptic study. Few differences in the regional effects of individual sympathomimetics were found in the nonseptic study, although infusion of sympathomimetics was accompanied by a redistribution of systemic Q toward the heart and away from the brain, kidney, small intestine, liver, and pancreas. In the septic study, however, the sympathomimetic infusions were not accompanied by the redistribution of Q away from small intestine and liver that was demonstrated in the nonseptic study. Therefore (1) the depressed vascular reactivity in hyperdynamic sepsis altered the dose profile of exogenous sympathomimetics required to augment systemic Q, and (2) the (sepsis x sympathomimetic) interaction was characterized by a depression in the anticipated redistribution of organ Q from "nonvital" to "vital" circulations.

Dopamine and renal salvage in the critically ill patient

Dopamine is a catecholamine used widely in critically ill patients and those undergoing major surgery, often as a 'renal protective' agent. Direct renal vasodilatation with 'low-dose' dopamine is the widely accepted basis for its use--hence the term 'renal dose' dopamine. However, recent evidence has revealed that the renal effects of this agent are far more complex. Moreover, some of these effects may be undesirable in the 'at-risk' kidney. The increased renal blood flow (RBF) of dopamine may be largely attributable to its inotropic (myocardial) action, even with low doses (i.e. less than 5 micrograms/kg/min). Similar increases in RBF can also be demonstrated with other (non-dopaminergic) inotropes. The early evidence for direct renal vasodilatation in response to dopamine has been brought into question by more recent research. The diuresis and natriuresis commonly seen following dopamine administration is now known to be due to a direct renal tubular (or 'diuretic') action. Furthermore, increasing knowledge regarding the pathophysiology of acute (ischaemic) renal failure, including RBF and the concept of 'oxygen supply and demand' in relation to tubular function, suggests that dopamine may mask important signs of renal ischaemia. Whether or not dopamine is truly beneficial to renal function currently remains unanswered. As it stands however, there is sufficient evidence to question its routine use in the setting of renal dysfunction in the critically ill patient.

Interaction of sepsis and sepsis plus sympathomimetics on myocardial oxygen availability

The ability to regulate myocardial blood flows (Q) in accord with changing myocardial O2 needs may be depressed in sepsis. This could be an important concern when sympathomimetics are used to augment systemic oxygen delivery (QO2) in this syndrome as increased myocardial O2 needs may accompany an infusion of this class of drugs. Therefore after measuring the effect of sepsis on myocardial O2 metabolism, we then infused various sympathomimetics to evaluate the sepsis+sympathomimetic interaction on myocardial QO2. We measured Q to the left (LV) and right (RV) ventricles by the radioactive microsphere technique in 36 unanesthetized mature sheep, before and during the infusion of dopamine, dobutamine, dopexamine, norepinephrine, salbutamol, or placebo. Randomly selected for infusion, these drugs were titrated to augment the thermodilution-derived cardiac index (CI) by greater than 20%. This study was repeated 24-48 h after cecal ligation and perforation had resulted in a hyperdynamic septic state [change (delta) in CI = sepsis - baseline = +54%; P less than 0.01]. During the septic study, both Q-LV (delta = +80%; P less than 0.01) and Q-RV (delta = +84%; P less than 0.01) were increased above baseline values; the augmented Q to both LV and RV was directly correlated with the arterial perfusion pressure (PA) x CI product and the mean pulmonary artery pressure (PPA) x CI product, respectively. Only 23% of study animals demonstrated net transmyocardial lactate production during the septic study. When the infusion of sympathomimetics was accompanied by an increase in the PPA x CI and PA x CI products, a further increase in both Q-RV and Q-LV, respectively, occurred. Also, neither the ventricular endocardial-to-epicardial flow ratios nor transmyocardial lactate extraction were modified by the sympathomimetics infusion. We conclude that the septic response to infection in this animal model was not accompanied by significant abnormalities in the metabolic regulation of myocardial QO2 (R. E. Cunnion, G. L. Scher, and M. M. Parker, Circulation 73: 637-644, 1986).

Treatment of severe cardiogenic pulmonary edema with continuous positive airway pressure delivered by face mask

BACKGROUND

Severe cardiogenic pulmonary edema is a frequent cause of respiratory failure, and many patients with this condition require endotracheal intubation and mechanical ventilation. We investigated whether continuous positive airway pressure delivered by means of a face mask had physiologic benefit and would reduce the need for intubation and mechanical ventilation.

METHODS

We randomly assigned 39 consecutive patients with respiratory failure due to severe cardiogenic pulmonary edema to receive either oxygen alone or oxygen plus continuous positive airway pressure delivered through a face mask. It was not possible to blind the investigators to the assigned treatment. Physiologic measurements were made over the subsequent 24 hours, and the patients were followed to hospital discharge.

RESULTS

After 30 minutes, both respiratory rate and arterial carbon dioxide tension had decreased more in the patients who received oxygen plus continuous positive airway pressure. The mean (+/- SD) respiratory rate at 30 minutes decreased from 32 +/- 6 to 33 +/- 9 breaths per minute in the patients receiving oxygen alone and from 35 +/- 8 to 27 +/- 6 breaths per minute in those receiving oxygen plus continuous positive airway pressure (P = 0.008); the arterial carbon dioxide tension decreased from 64 +/- 17 to 62 +/- 14 mm Hg in those receiving oxygen alone and from 58 +/- 8 to 46 +/- 4 mm Hg in those receiving oxygen plus continuous positive airway pressure (P less than 0.001). The patients receiving continuous positive airway pressure also had a greater increase in the arterial pH (oxygen alone, from 7.15 +/- 0.11 to 7.18 +/- 0.18; oxygen plus continuous positive airway pressure, from 7.18 +/- 0.08 to 7.28 +/- 0.06; P less than 0.001) and in the ratio of arterial oxygen tension to the fraction of inspired oxygen (oxygen alone, from 136 +/- 44 to 126 +/- 47; oxygen plus continuous positive airway pressure, from 138 +/- 32 to 206 +/- 126; P = 0.01). After 24 hours, however, there were no significant differences between the two treatment groups in any of these respiratory indexes. Seven (35 percent) of the patients who received oxygen alone but none who received oxygen plus continuous positive airway pressure required intubation and mechanical ventilation (P = 0.005). However, no significant difference was found in in-hospital mortality (oxygen alone, 4 of 20 patients; oxygen plus continuous positive airway pressure, 2 of 19; P = 0.36) or the length of the hospital stay.

CONCLUSIONS

Continuous positive airway pressure delivered by face mask in patients with severe cardiogenic pulmonary edema can result in early physiologic improvement and reduce the need for intubation and mechanical ventilation. This short-term study could not establish whether continuous positive airway pressure has any long-term benefit or whether a larger study would have shown a difference in mortality between the treatment groups.

Hyperdynamic sepsis modifies a PEEP-mediated redistribution in organ blood flows

Changes in organ blood flow (Q) produced by 20 cm H2O positive end-expiratory pressure (PEEP) were measured before and after the induction of hyperdynamic sepsis in nine unanesthetized sheep. During the baseline nonseptic study, PEEP was associated with a 9% fall in thermodilution-measured systemic Q, although arterial perfusing pressures were unaffected. Concurrently, microsphere-derived Q was maintained to the brain and heart, but fell to liver, spleen, pancreas, kidney, large intestine, and gastrocnemius. Twenty-four to 36 h after cecal ligation and perforation, a pre-PEEP septic study demonstrated an increase in all of the cardiac index (CI) (+43%) and systemic O2 delivery (+54%) when compared with the nonseptic study, whereas whole-body O2 extraction (-30%) was depressed. Although PEEP depressed systemic Q (-17%) during the septic study to a greater extent than during the nonseptic study (p less than 0.02), absolute organ Q fell only to pancreas, liver, and spleen. Relative to the simultaneous fall in the CI, Q to some splanchnic organs was not depressed by PEEP to the same magnitude in the septic as in the nonseptic study. When an infusion of Ringer's lactate (993 +/- 295 ml) subsequently restored systemic Q to pre-PEEP septic levels, individual flows that had been depressed by PEEP were not restored. Furthermore, Q-kidney continued to fall, such that the postfluid Q-kidney (-19%) was significantly less than was demonstrated in the pre-PEEP septic study. We postulate that differences noted in the distribution of organ Q between the nonseptic and hyperdynamic septic studies after the application of PEEP were secondary to the vasculopathy of sepsis and/or an alteration in the function of specific organ microcirculations. However, these data do not address whether the changes in organ Q distribution after a PEEP-mediated depression in systemic Q during sepsis significantly restricted tissue DO2. The inability to acutely reverse the PEEP-mediated changes in organ Q after restoring systemic Q by a fluid infusion also suggests the need to evaluate alternative methods of support to organ Q in acute respiratory failure secondary to sepsis when the addition of PEEP acutely depresses systemic DO2.

Histologic and ultrastructural changes in nonpulmonary organs during early hyperdynamic sepsis

Previous studies describing the histologic elements of multi-system organ failure caused by bacterial sepsis may have been complicated by a significant interaction on tissue injury from either a preterminal low-flow state or the effects of therapy immediately before death. Therefore we evaluated the nonpulmonary histologic findings of sepsis during a 3-day period that followed cecal ligation and perforation. In this septic model, mean arterial perfusion pressures remained unchanged from baseline, systemic flows rose by 54%, and laboratory evidence of organ dysfunction including an elevation of the serum bilirubin levels and a depression of the serum total protein values was considered mild. Concurrently, development of the hyperdynamic central circulatory septic state was associated with widespread histologic changes in myocardium, striated muscle, liver, gut, and pancreas. Lesions common to these organs included high-protein interstitial and intracellular edema, mitochondrial destruction, and patchy cell necrosis. Lesions within the pancreas were exaggerated over those noted in other organs. Of all organs examined, only the liver demonstrated microvascular neutrophil accumulation. Unlike models of shock caused by sepsis, fibrin thrombi were not seen in the microvasculature of any organ. We conclude that tissue injury characterized by the accumulation of protein-rich extravascular fluid and the development of reversible and irreversible cell injury antedated significant multiple-system organ failure in this animal model of normotensive sepsis.

Additional work of breathing imposed by endotracheal tubes, breathing circuits, and intensive care ventilators

A disadvantage of spontaneous breathing through an endotracheal tube (ETT) and connector attached to a breathing circuit and/or ventilator (breathing device) is an increase in the work of breathing. The work of breathing associated with ETT of 6 to 9-mm diameter and eight breathing devices was determined, using a lung simulator to mimic spontaneous inspiration at flow rates of 20 to 100 L/min and a tidal volume of 500 ml, at both zero end-expiratory pressure (ZEEP) and 10 cm H2O continuous positive airway pressure (CPAP). Work associated with the breathing devices alone (WCIR) ranged from -0.002 kg.m/L (Servo 900-C ventilator, 7-mm ETT, 20 L/min, ZEEP) to 0.1 kg.m/L (continuous flow circuit, 7-mm ETT, 100 L/min, CPAP), the latter representing 196% of the work of normal breathing. When the devices were attached to ETT, total apparatus work (WAPP) ranged from 0.009 kg.m/L (Mapleson-D circuit, 9-mm ETT, 20 L/min, ZEEP) to 0.25 kg.m/L (Drager EV-A, 6-mm ETT, 100 L/min, ZEEP), the latter representing 490% of the work of normal breathing. This additional work imposed by the ETT varied considerably among devices. Spontaneous breathing through modern ventilators, circuits and ETT imposes a burden of increased work, most of which is associated with the presence of the ETT and connector. Whether this burden represents an impediment to the weaning patient, or has training value for the ultimate resumption of unassisted spontaneous ventilation, remains to be determined.

Central venous catheter stiffness and its relation to vascular perforation

Delayed central venous perforation is an uncommon but serious complication of central venous catheter insertion. An increase in catheter stiffness may have been responsible for our association of venous perforation with use of a guidewire insertion technique. A bench model was used to investigate the stiffness characteristics of thirty-four different types of catheters. The initial stiffness is poorly described by material or catheter gauge. A large range of values is seen between apparently similar catheters--the 16 gauge polyethylene catheter associated with two perforations at our institution had an initial stiffness value 7.5 Nm2 X 10(-5) at 37 degrees C in comparison with our previous standard--the 16 gauge Deseret Intracath with an initial stiffness of 2 Nm2 X 10(-5). Multilumen catheters had a similar range of stiffness to single lumen catheters, while paediatric catheters in general were less stiff. Dialysis catheters were up to five times as stiff as the stiffest central venous catheter. Stiffness decayed at a rate and to an extent which differed from catheter to catheter. Absorption of water by the catheter appears to be one factor involved in stress relaxation.

New generation ventilators

Desirable features of new generation intensive care ventilators include the ability to ventilate a wide range of patient sizes, an uncomplicated control panel, an appropriate but not excessive variety of ventilatory patterns, adequate patient monitoring and alarm functions, and simplicity of cleaning and routine maintenance. Examples of currently available ventilators include the Servo 900-C, CPU-1, Engstrom Erica, Bear 5, Drager EV-A and Hamilton Veolar. The incorporation of microcomputer control into some of these ventilators has resulted in improved flexibility and a limited number of automatic responses to detected patient changes. However, the function of components provided to allow spontaneous ventilation, such as demand valves, requires considerable improvement. Current trends in ventilator design include further refinement of computer control and the provision of graphic displays showing the results of continuous sophisticated analysis of respiratory function. The extent to which these developments will prove clinically useful will require careful evaluation.