Successful Transplantation of Lungs From an Uncontrolled Donor After Circulatory Death Preserved In Situ by Alveolar Recruitment Maneuvers and Assessed by Ex Vivo Lung Perfusion.

We developed a protocol to procure lungs from uncontrolled donors after circulatory determination of death (NCT02061462). Subjects with cardiovascular collapse, treated on scene by a resuscitation team and transferred to the emergency room, are considered potential donors once declared dead. Exclusion criteria include unwitnessed collapse, no-flow period of >15 min and low flow >60 min. After death, lung preservation with recruitment maneuvers, continuous positive airway pressure, and protective mechanical ventilation is applied to the donor. After procurement, ex vivo lung perfusion (EVLP) is performed. From November 2014, 10 subjects were considered potential donors; one of these underwent the full process of procurement, EVLP, and transplantation. The donor was a 46-year-old male who died because of thoracic aortic dissection. Lungs were procured 4 h and 48 min after death, and deemed suitable for transplantation after EVLP. Lungs were then offered to a rapidly deteriorating recipient with cystic fibrosis (lung allocation score [LAS] 46) who consented to the transplant in this experimental setting. Six months after transplantation, the recipient is in good condition (forced expiratory volume in 1 s 85%) with no signs of rejection. This protocol allowed procurement of lungs from an uncontrolled donor after circulatory determination of death following an extended period of warm ischemia.

Effect of body mass index in acute respiratory distress syndrome.

BACKGROUND: Obesity is associated in healthy subjects with a great reduction in functional residual capacity and with a stiffening of lung and chest wall elastance, which promote alveolar collapse and hypoxaemia. Likewise, obese patients with acute respiratory distress syndrome (ARDS) could present greater derangements of respiratory mechanics than patients of normal weight. METHODS: One hundred and one ARDS patients were enrolled. Partitioned respiratory mechanics and gas exchange were measured at 5 and 15 cm H2O of PEEP with a tidal volume of 6-8 ml kg(-1) of predicted body weight. At 5 and 45 cm H2O of PEEP, two lung computed tomography scans were performed. RESULTS: Patients were divided as follows according to BMI: normal weight (BMI≤25 kg m(-2)), overweight (BMI between 25 and 30 kg m(-2)), and obese (BMI>30 kg m(-2)). Obese, overweight, and normal-weight groups presented a similar lung elastance (median [interquartile range], respectively: 17.7 [14.2-24.8], 20.9 [16.1-30.2], and 20.5 [15.2-23.6] cm H2O litre(-1) at 5 cm H2O of PEEP and 19.3 [15.5-26.3], 21.1 [17.4-29.2], and 17.1 [13.4-20.4] cm H2O litre(-1) at 15 cm H2O of PEEP) and chest elastance (respectively: 4.9 [3.1-8.8], 5.9 [3.8-8.7], and 7.8 [3.9-9.8] cm H2O litre(-1) at 5 cm H2O of PEEP and 6.5 [4.5-9.6], 6.6 [4.2-9.2], and 4.9 [2.4-7.6] cm H2O litre(-1) at 15 cm H2O of PEEP). Lung recruitability was not affected by the body weight (15.6 [6.3-23.4], 15.7 [9.8-22.2], and 11.3 [6.2-15.6]% for normal-weight, overweight, and obese groups, respectively). Lung gas volume was significantly lower whereas total superimposed pressure was significantly higher in the obese compared with the normal-weight group (1148 [680-1815] vs 827 [686-1213] ml and 17.4 [15.8-19.3] vs 19.3 [18.6-21.7] cm H2O, respectively). CONCLUSIONS: Obese ARDS patients do not present higher chest wall elastance and lung recruitability.

Monitoring lung impedance changes during long-term ventilator-induced lung injury ventilation using electrical impedance tomography.

OBJECTIVE: The target of this methodological evaluation was the feasibility of long-term monitoring of changes in lung conditions by time-difference electrical impedance tomography (tdEIT). In contrast to ventilation monitoring by tdEIT, the monitoring of end-expiratory (EELIC) or end-inspiratory (EILIC) lung impedance change always requires a reference measurement. APPROACH: To determine the stability of the used Pulmovista 500® EIT system, as a prerequisite it was initially secured on a resistive phantom for 50 h. By comparing the slopes of EELIC for the whole lung area up to 48 h from 36 pigs ventilated at six positive end-expiratory pressure (PEEP) levels from 0 to 18 cmH2O we found a good agreement (range of r 2 = 0.93-1.0) between absolute EIT (aEIT) and tdEIT values. This justified the usage of tdEIT with its superior local resolution compared to aEIT for long-term determination of EELIC. MAIN RESULTS: The EELIC was between -0.07 Ωm day-1 at PEEP 4 and -1.04 Ωm day-1 at PEEP 18 cmH2O. The complex local time pattern for EELIC was roughly quantified by the new parameter, centre of end-expiratory change (CoEEC), in equivalence to the established centre of ventilation (CoV). The ventrally located mean of the CoV was fairly constant in the range of 42%-46% of thorax diameter; however, on the contrary, the CoEEC shifted from about 40% to about 75% in the dorsal direction for PEEP levels of 14 and 18 cmH2O. SIGNIFICANCE: The observed shifts started earlier for higher PEEP levels. Changes of EELI could be precisely monitored over a period of 48 h by tdEIT on pigs.

Understanding international differences in terminology for delirium and other types of acute brain dysfunction in critically ill patients.

BACKGROUND: Delirium (acute brain dysfunction) is a potentially life threatening disturbance in brain function that frequently occurs in critically ill patients. While this area of brain dysfunction in critical care is rapidly advancing, striking limitations in use of terminology related to delirium internationally are hindering cross-talk and collaborative research. In the English literature, synonyms of delirium such as the Intensive Care Unit syndrome, acute brain dysfunction, acute brain failure, psychosis, confusion, and encephalopathy are widely used. This often leads to scientific "confusion" regarding published data and methodology within studies, which is further exacerbated by organizational, cultural and language barriers. OBJECTIVE: We undertook this multinational effort to identify conflicts in terminology and phenomenology of delirium to facilitate communication across medical disciplines and languages. METHODS: The evaluation of the terminology used for acute brain dysfunction was determined conducting communications with 24 authors from academic communities throughout countries/regions that speak the 13 variants of the Romanic languages included into this manuscript. RESULTS: In the 13 languages utilizing Romanic characters, included in this report, we identified the following terms used to define major types of acute brain dysfunction: coma, delirium, delirio, delirium tremens, délire, confusion mentale, delir, delier, Durchgangs-Syndrom, acute verwardheid, intensiv-psykose, IVA-psykos, IVA-syndrom, akutt konfusion/forvirring. Interestingly two terms are very consistent: 100 % of the selected languages use the term coma or koma to describe patients unresponsive to verbal and/or physical stimuli, and 100% use delirium tremens to define delirium due to alcohol withdrawal. Conversely, only 54% use the term delirium to indicate the disorder as defined by the DSM-IV as an acute change in mental status, inattention, disorganized thinking and altered level of consciousness. CONCLUSIONS: Attempts towards standardization in terminology, or at least awareness of differences across languages and specialties, will help cross-talk among clinicians and researchers.

Does high PEEP prevent alveolar cycling?

Acute respiratory distress syndrome (ARDS) patients need mechanical ventilation to sustain gas exchange. Animal experiments showed that mechanical ventilation with high volume/plateau pressure and no positive end-expiratory pressure (PEEP) damages healthy lungs, while low tidal volumes and the application of higher PEEP levels are protective. PEEP makes the lung homogeneous, reducing the pressure multiplication at the interface between lung units with different inflation statuses and keeps the lung open through the whole respiratory cycle, avoiding intratidal opening and closing. Four randomized clinical trials tested a higher PEEP strategy compared to a lower PEEP strategy but failed to show any survival benefit. These results, which apparently contradict preclinical data, may be explained by CT scanning, which investigates the behaviour of ARDS lung upon inflation and deflation demonstrating that: (1) 15 cmH2O PEEP is insufficient to overcome the closing pressures of the lung and keep it open through the whole respiratory cycle; (2) lung recruitment is continuous along the volume-pressure curve. The application of a PEEP level around 15 cmH2O does not abolish opening and closing, but the lung region undergoing opening and closing is simply shifted downward, i. e. becomes more vertebral in the supine patient. (3) Recruited lung tissue becomes poorly inflated and not well inflated; poorly inflated tissue is inhomogeneous: while increasing PEEP the reduction in lung inhomogeneity is small or non-existent.

The DAM gene family encodes a new group of tumor-specific antigens recognized by human leukocyte antigen A2-restricted cytotoxic T lymphocytes.

The DAM family of genes has a high degree of homology with MAGE, both in nucleotide sequence and in neoplastic tissue-specific expression. This study describes, for the first time, the identification of CTLs specific for a peptide epitope encoded by DAM genes. A human leukocyte antigen (HLA)-A2-restricted CTL clone was raised against a peptide, D10/6-271, encoded by codons 271-279 in the DAM cDNA. The corresponding peptide in the MAGE-3 sequence, M3-271, has been shown previously to be a natural T-cell epitope for HLA-A2-restricted CTLs recognizing the MAGE-3 protein. The D10/6-271-specific CTL clone required approximately 3 nM exogenous peptide for half-maximal lysis of target cells and was able to specifically recognize endogenous DAM antigen on HLA-A2+ melanoma cells infected with a vaccinia vector recombinant for gene DAM-6. These data suggest that DAM genes might encode a new group of tumor-specific antigens useful for the design of specific antitumor vaccines.

Deterioration of Lung Function in a Pig Model of Uncontrolled Cardiac Death.

INTRODUCTION: Uncontrolled donors after circulatory determination of death (uDCDD) represent a yet unexplored pool of organs potentially available for transplantation. The aims of this study were to validate a protocol of cardiac death in the pig and to investigate lung function during the process. MATERIALS AND METHODS: Cardiac death was induced in preanesthetized animals with an injection of 600 mg propofol; once systolic blood pressure was <50 mm Hg (Agonal Phase), a 20 mEq bolus of KCl was given and, after asystolia was documented, cardiopulmonary resuscitation (CPR) started, followed by 5 minutes no touch (end-CPR). Invasive blood pressure (BP) and heart rate (HR) were recorded; blood samples taken at baseline, 15 minutes after CPR, and after the no touch period (end-CPR). Computed tomography (CT) scans were taken at baseline and at end-CPR. RESULTS: Agonal phase was reached in 6 ± 1 minutes and lasted 3 ± 1 minutes; average HR was 49 ± 16 beats/min, and BP was 41 ± 12 mm Hg. CPR lasted 35 ± 3 minutes; average HR and BP were 113 ± 32 beats/min and 86 ± 63 mm Hg, respectively. PaO2/FiO2 decreased from 442 ± 31 mm Hg at baseline to 63 ± 36 at end-CPR (P < .001). pH decreased from 7.378 ± 0.045 to 6.931 ± 0.042 (P < .001), with a corresponding increase of lactate from 0.9 ± 0.2 to mmol/L to 12.8 ± 2.1 (P < .001). As assessed using CT scan, total lung volume decreased (baseline vs end-CPR 1107 ± 106 mL vs 617 ± 95; P < .001), whereas noninflated tissue (ie, atelectasis) significantly increased (46 ± 10 g vs 131 ± 89; P = .008). CONCLUSIONS: Lung function greatly deteriorated after cardiac death. The model we set may constitute a reproducible platform for future investigations on lung uDCDD.

Ventilator-related causes of lung injury: the mechanical power.

PURPOSE: We hypothesized that the ventilator-related causes of lung injury may be unified in a single variable: the mechanical power. We assessed whether the mechanical power measured by the pressure-volume loops can be computed from its components: tidal volume (TV)/driving pressure (∆P aw), flow, positive end-expiratory pressure (PEEP), and respiratory rate (RR). If so, the relative contributions of each variable to the mechanical power can be estimated. METHODS: We computed the mechanical power by multiplying each component of the equation of motion by the variation of volume and RR: [Formula: see text]where ∆V is the tidal volume, ELrs is the elastance of the respiratory system, I:E is the inspiratory-to-expiratory time ratio, and R aw is the airway resistance. In 30 patients with normal lungs and in 50 ARDS patients, mechanical power was computed via the power equation and measured from the dynamic pressure-volume curve at 5 and 15 cmH2O PEEP and 6, 8, 10, and 12 ml/kg TV. We then computed the effects of the individual component variables on the mechanical power. RESULTS: Computed and measured mechanical powers were similar at 5 and 15 cmH2O PEEP both in normal subjects and in ARDS patients (slopes = 0.96, 1.06, 1.01, 1.12 respectively, R (2) > 0.96 and p < 0.0001 for all). The mechanical power increases exponentially with TV, ∆P aw, and flow (exponent = 2) as well as with RR (exponent = 1.4) and linearly with PEEP. CONCLUSIONS: The mechanical power equation may help estimate the contribution of the different ventilator-related causes of lung injury and of their variations. The equation can be easily implemented in every ventilator's software.

Early outcome of liver transplantation performed with organs procured from brain death donors with transient or sustained cardio-circulatory collapse.

BACKGROUND: Aim of this study was to compare early graft function after transplantation of recipients transplanted with livers procured from donors after brain death who experienced transient or sustained cardio-circulatory collapse. METHODS: We retrospectively analysed patients who underwent liver transplantation (LTx) at our Institution from January 2010 to May 2012. Recipients were divided into 3 groups: those who received livers from brain death donors who experienced reversible cardio-circulatory arrest before organ procurement (RCA); those who experienced sustained cardio-circulatory collapse, treated with extra-corporeal membrane oxygenation support as rescue therapy of refractory cardiogenic shock (ECMO). Standard donors were considered as reference group (REF). Postoperative graft function, Primary Non-Function (PNF), and complications during the first 30 days were analysed. RESULTS: 102 LTx were analysed (76 REF, 22 RCA and 4 ECMO). The main cause of donor's death was post-anoxic coma in RCA and ECMO, cerebrovascular accident in REF. SGOT in REF, RCA, and ECMO donors were 27 [17-43], 54 [34-92], 716 [190-962] respectively, SGPT 17 [12-34], 46 [27-73], 84 [51-175] UI/L respectively, both P<0.01. All recipients had similar SGOT (P=0.48), SGPT (P=0.75) and Model for End-Stage Liver Disease scores (P=0.98) before LTx; similar graft cold and warm ischemia time and serum lactate levels at the end of surgery. After LTx, Intensive Care Unit stay and the incidence of PNF were similar. CONCLUSION: The use of livers procured from donors after brain death that experienced transient or sustained cardio-circulatory collapse was associated with early graft function comparable to that of standard donors.

The prone position in ARDS patients. A clinical study.

The gas exchange and hemodynamics were evaluated before, during, and after a two-hour period of prone position in 13 moderate-severe ARDS patients. Lung computerized tomography was obtained in both the supine and prone positions in two of these patients. Average arterial oxygenation improved after prone positioning (p less than 0.01). A PaO2 improvement of at least 10 mm Hg after 30 minutes of prone position was used as a criterion to discriminate between responders and nonresponders to the postural change. Eight patients met the "responders" group criterion, and in the five nonresponder patients, the PaO2 did not change significantly throughout the study. Computerized tomograms in the prone position showed disappearance of posterobasal densities and appearance of new densities in the anterior regions, in both patients studied. One of these was a responder, the other a nonresponder. A brief test period in prone position is indicated in ARDS patients to identify those who may benefit from this postural treatment. The definite mechanism of the arterial oxygenation improvement observed remains to be clarified.

An alternative to breathing.

If carbon dioxide is removed by an extracorporeal membrane lung ventilated with room air, the natural lung can be used for oxygen transport alone; we have demonstrated this in lambs by maintaining lungs "inflated" with 100 percent oxygen at constant pressure and removing all carbon dioxide through the membrane lung. This process is a variant of "apneic oxygenation" without its disadvantages, because the arterial pH, PCO2, and PO2 all remain normal. No nitrogen washout is needed. These studies were carried out in five lambs anesthetized and paralyzed for 24 hours. For carbon dioxide removal, blood from the subclavin artery was pumped through an extracorporeal membrane lung and was returned into the external jugular vein. For oxygen delivery, the lungs were inflated through a tracheostomy tube with 100 percent oxygen to a pressure of 5 cm. H2O. There was no significant change in arterial blood PO2 after perfusion had begun or at the end of the perfusion 24 hours later. The arterial PCO2 remained steady, and there was no change in acid-base balance. The functional residual capacity (FRC) and static lung compliance remained unchanged. The total dead space was 10 to 15 ml. All animals recovered and survived in good health. At equilibrium, alveolar nitrogen partial pressure was always equal to the partial pressure of nitrogen in the ventilating gas of the membrane lung and was the sole determining factor in controlling alveolar oxygen concentration. Direct measurement of pulmonary gas showed alveolar gas at the level of the carina.

Total respiratory system, lung, and chest wall mechanics in sedated-paralyzed postoperative morbidly obese patients.

OBJECTIVE: To study the relative contribution of the lung and the chest wall on the total respiratory system mechanics, gas exchange, and work of breathing in sedated-paralyzed normal subjects and morbidly obese patients, in the postoperative period. SETTING: Policlinico Hospital, University of Milan, Italy. METHODS: In ten normal subjects (normal) and ten morbidly obese patients (obese), we partitioned the total respiratory mechanics (rs) into its lung (L) and chest wall (w) components using the esophageal balloon technique together with airway occlusion technique, during constant flow inflation. We measured, after abdominal surgery, static respiratory system compliance (Cst,rs), lung compliance (Cst,L), chest wall compliance (Cst,w), total lung (Rmax,L) and chest wall (Rmax,w) resistance. Rmax,L includes airway (Rmin,L) and "additional" lung resistance (DR,L). DR,L represents the component due to viscoelastic phenomena of the lung tissue and time constant inequalities (pendelluft). Functional residual capacity (FRC) was measured by helium dilution technique. RESULTS: We found that morbidly obese patients compared with normal subjects are characterized by the following: (1) reduced Cst,rs (p < 0.01), due to lower Cst,L (55.3 +/- 15.3 mL x cm H2O-1 vs 106.6 +/- 31.7 mL x cm H2O-1; p < 0.01) and Cst,w (112.4 +/- 47.4 mL x cm H2O-1 vs 190.7 +/- 45.1 mL x cm H2O-1; p < 0.01); (2) increased Rmin,L (4.7 +/- 3.1 mL x cm H2O x L-1 x s; vs 1.0 +/- 0.8 mL x cm H2O x L-1 x s; p < 0.01) and DR,L (4.9 +/- 2.6 mL x cm H2O x L-1 x s; vs 1.5 +/- 0.8 mL x cm H2O x L-1 x s; p < 0.01); (3) reduced FRC (0.665 +/- 0.191 L vs 1.691 +/- 0.325 L; p < 0.01); (4) increased work performed to inflate both the lung (0.91 +/- 0.25 J/L vs 0.34 +/- 0.08 J/L; p < 0.01) and the chest wall (0.39 +/- 0.13 J/L vs 0.18 +/- 0.04 J/L; p < 0.01); and (5) a reduced pulmonary oxygenation index (PaO2/PAO2 ratio). CONCLUSION: Sedated-paralyzed morbidly obese patients, compared with normal subjects, are characterized by marked derangements in lung and chest wall mechanics and reduced lung volume after abdominal surgery. These alterations may account for impaired arterial oxygenation in the postoperative period.

The value of portable chest roentgenography in adult respiratory distress syndrome. Comparison with computed tomography.

In 17 patients with adult respiratory distress syndrome, we used data derived from computed tomographic (CT) scan densitometric analysis to validate the value of portable chest roentgenograms in objectively estimating the amount of pulmonary edema. Chest roentgenograms and CT scans were taken in the same ventilatory conditions (apnea at 10 cm H2O of positive end-expiratory pressure [PEEP]); blood gas samples and hemodynamic parameters were collected at the same time. Roentgenographic analysis was undertaken by independent observers using two standardized scoring systems proposed in the literature. CT scan analysis was performed using the CT number frequency distribution and the gas lung volume (measured by helium dilution technique) to estimate quantitatively the lung density, the lung weight, and the percentage of normally aerated and nonaerated tissue. Knowing the mean CT number of the pulmonary parenchyma in a group of normal subjects, we also inferred the ideal lung weight expected in the study population and computed the excess tissue mass as the difference between actual and ideal lung weight. Both the roentgenographic scoring systems showed direct correlation with the pulmonary impairment as detected by CT scan densitometric analysis (CT number, percentage of nonaerated tissue, lung weight, and excess tissue mass; p less than 0.01) and inverse relation with the percentage of normally aerated tissue (p less than 0.01). We also found a relationship between roentgenographic scores and the impairment in gas exchange as detected by shunt fraction (p less than 0.05). We conclude that standardized reading of portable chest roentgenograms by means of scoring tables is a valuable tool in estimating the amount of pulmonary edema in a patient with adult respiratory distress syndrome.

Unrelated mismatched cord blood transplantation in an adult with secondary AML.

Umbilical cord blood (CB) has been widely used for related and unrelated transplants in pediatric patients. We present the case of an adult with secondary AML who received an unrelated, one-antigen mismatched CB transplant due to the lack of a matched donor. The patient was a 26-year-old female (35 kg/bw) who had received an autologous bone marrow transplant for Hodgkin's disease in April 1994 and, 6 months later, developed secondary MDS (RAEB, 46, XX, -7, +mar), which slowly evolved into acute myelogenous leukemia. In May 1995, she was transplanted with a 165 ml CB unit containing a total of 1.6 x 10(9) nucleated cells, 11 x 10(6) CD34+ cells and 7.2 x 10(5) CFU-GM. GVHD prophylaxis consisted of standard CsA and methotrexate. Myeloid engraftment occurred on day +28 (PMN > 500) and full donor chimerism was confirmed twice (on days +33 and +56) by means of cytogenetics and DNA microsatellite analysis. Erythroid and megakaryocytic engraftment was documented by immunohistochemical analysis of a bone marrow biopsy on day +40, showing the presence of erythroblastic islands and isolated CD61+ immature cells. The patient did not develop GVHD but died on day +56 from idiopathic interstitial pneumonia and multiorgan failure. To our knowledge, this is one of the first case reports of unrelated mismatched CB transplantation in an adult.

Short-term effects of anastrozole treatment on insulin-like growth factor system in postmenopausal advanced breast cancer patients.

Insulin-like growth factors (IGFs) play a fundamental role in cancer development by acting in both an endocrinal and paracrinal manner, and hormone breast cancer treatments affect the IGF system by modifying circulating growth factor levels. We evaluated total IGF-1, IGF-2, IGF binding protein (IGFBP)-1 and IGFBP-3 in the blood of 34 postmenopausal advanced breast cancer patients (median age 63 years, range 41-85) treated with anastrozole, a non-steroidal structure aromatase inhibitor (NSS-AI). The plasma samples were obtained at baseline, and after 2, 4, 8 and 12 weeks of treatment. The IGFs were quantitated by means of sensitive radioimmunoassays (RIAs). IGF-1 significantly increased during anastrozole treatment (baseline versus 12 weeks, P=0.031), IGF-2 showed a trend towards an increase, and IGFBP-1 constantly but not significantly decreased; IGFBP-3 did not seem to be affected at all. The anastrozole-induced changes in IGFs and IGFBP-1 appeared to be different in the patients receiving a clinical benefit from those observed in non-responders. We have previously shown that letrozole (a different type of NSS-AI) modifies blood IGF-1 levels, and the results of this study of the biological effects of anastrozole on the components of the IGF system confirm our previous observations.

Simple estimate of the true right to left shunt (Qs/Qt) at maintenance F1O2 by sulphur hexafluoride retention.

Differentiating Qs/Qt over Qva/Q (as measured by the standard O2 content formula) appears to be of great interest in applying and evaluating different therapeutic approaches. The estimation of Qs/Qt by 100% O2 breathing may alter "per se" the lung condition and is unsatisfactory. We used Sulphur Hexafluoride (SF6; lambda = 6.10(-3) ml ml-1 760 mmHg-1) to identify the true shunt (VA/Q less than 0.05) at maintenance F1O2. A simple and rapid determination of SF6 retention is performed by ECD gas chromatography from contemporaneous arterial and mixed venous blood samples, taken during i.v. infusion of an SF6 containing solution. QS/Qt estimate is then given by the ratio: PaSF6/PvSF6. It is not necessary to know the absolute gas partial pressures or concentration, hence absolute gas calibrations are not required. This method is suggested as feasible and satisfactory for clinical use, allowing the determination of QS/Qt at the maintenance F1O2.

Venous admixture (Qva/Q) and true shunt (Qs/Qt) in ARF patients: effects of PEEP at constant FIO2.

Venous admixture (Qva/Q) in ARF patients is due to both true right to left shunt (Qs/Qt: perfusion of truly unventilated areas) and to maldistribution [Qva-Qs)/Qt: effects of unevenness of ventilation/perfusion ratio). Using the retention rate of sulphur hexafluoride we determined the effects of PEEP on Qs/Qt and (Qva-Qs)/Qt at a constant FIO2 for each patient (0.57 +/- 0.19 SD, range 0.4-0.95). Eleven patients with ARF (treated either by CPPV or CPAP) were studied on 16 occasions. Each measurement was repeated at two levels of PEEP, 5 cm H2O below and 5 cm H2O above the patient's clinically determined PEEP level. The increase in PEEP resulted in: - a decrease in Qva/Q (from 0.37 +/- 0.13 to 0.27 +/- 0.12, p less than 0.01); - a parallel decrease in Qs/Qt (from 0.29 +/- 0.16 to 0.22 +/- 0.14, p less than 0.01); there was a positive correlation between Qva/Q and Qs/Qt changes (r = 0.53, p less than 0.05). No significant variation was demonstrated in (Qva-Qs)/Qt (from 0.074 +/- 0.045 to 0.054 +/- 0.048). On the other hand there was a negative correlation between the fraction of Qva/Q due to the maldistribution and FIO2: (Qva-Qs)/Qva = 0.75-0.86 FIO2 (r = 0.74, p less than 0.01). We conclude that: PEEP decreased Qva/Q mainly through changes in Qs/Qt but did not have a definite effect on (Qva-Qs)/Qt. Maldistribution was responsible for a significant portion of Qva/Q in those ARF patients tolerating a relatively low FIO2 (0.4-0.6).