Comparison of different culture conditions for human mesenchymal stromal cells for clinical stem cell therapy

OBJECTIVE

Mesenchymal stromal cells (MSCs) from adult bone marrow (BM) are considered potential candidates for therapeutic neovascularization in cardiovascular disease. When implementing results from animal trials in clinical treatment, it is essential to isolate and expand the MSCs under conditions following good manufacturing practice (GMP). The aims of the study were first to establish culture conditions following GMP quality demands for human MSC expansion and differentiation for use in clinical trials, and second to compare these MSCs with MSCs derived from culture in four media commonly used for MSC cultivation in animal studies simulating clinical stem cell therapy.

MATERIAL AND METHODS

Human mononuclear cells (MNCs) were isolated from BM aspirates by density gradient centrifugation and cultivated in a GMP-accepted medium (EMEA medium) or in one of four other media.

RESULTS

FACS analysis showed that the plastic-adherent MSCs cultured in EMEA medium or in the other four media were identically negative for the haematopoietic surface markers CD45 and CD34 and positive for CD105, CD73, CD90, CD166 and CD13, which in combined expression is characteristic of MSCs. MSC stimulation with vascular endothelial growth factor (VEGF) increased expression of the characteristic endothelial genes KDR and von Willebrand factor; the von Willebrand factor and CD31 at protein level as well as the capacity to develop capillary-like structures.

CONCLUSIONS

We established culture conditions with a GMP compliant medium for MSC cultivation, expansion and differentiation. The expanded and differentiated MSCs can be used in autologous mesenchymal stromal cell therapy in patients with ischaemic heart disease.

The influence of freezing and storage on the characteristics and functions of human mesenchymal stromal cells isolated for clinical use

BACKGROUND

Studies have shown that stem cell therapy could be a novel option for improving neovascularization and cardiac function in patients with ischemic heart disease. Human mesenchymal stromal cells (MSC) have generated wide interest in the clinical setting because of their ability to regenerate tissue. The aim of the study was to test whether freezing and storage of human BM mononuclear cells (BM-MNC) and ex vivo-expanded MSC influenced their phenotypic and functional characteristics as well as proliferation capacity.

METHODS

MNC were isolated from BM and divided into two portions: one part was immediately cultured (MSC P0) whereas the second part was frozen for a week before cultivation and analysis (F-MSC P1). Confluent MSC (P0) were harvested and divided: one was analyzed as MSC P1 and the other was frozen for a week before further cultivation and analysis as F-MSC P2.

RESULTS

MSC P1, F-MSC P1 and F-MSC P2 had similar proliferation capacities and demonstrated almost identical expression levels of markers characteristic for MSC. The capacity to form endothelial vascular structures was independent of freezing.

DISCUSSION

The proliferation and differentiation capacity as well as the cellular characteristics were identical in cultivated MSC derived from freshly isolated BM-MNC and MSC derived after freezing and storage of either freshly isolated BM-MNC or ex vivo-cultivated MSC. This highlights the potential clinical use of MSC in patients with cardiac and degenerative diseases, as it would be possible to inject MSC obtained from the same BM aspiration at different time points.

Labelling of human mesenchymal stem cells with indium-111 for SPECT imaging: effect on cell proliferation and differentiation

PURPOSE

Stem cell therapy seems to be a new treatment option within cardiac diseases to improve myocardial perfusion and function. However, the delivery and traceability of the cells represent a problem. Radioactive labelling with 111In could be a method for tracking mesenchymal stem cells (MSCs). However, 111In could influence the viability and differentiation capacity of MSCs, which would limit its use. Therefore, the aim of this study was to evaluate the influence of 111In labelling in doses relevant for SPECT imaging in humans on the viability and differentiation capacity of human MSCs.

METHODS AND RESULTS

Human MSCs isolated from bone marrow were incubated with 111In-tropolone (15-800 Bq/cell). The labelling efficiency was approximately 25% with 30 Bq/cell 111In. The MSC doubling time was 1.04+/-0.1 days and was not influenced by 111In within the range 15-260 Bq/cell. Using 30 Bq 111In/cell it was possible to label MSCs to a level relevant for clinical scintigraphic use. With this dose, 111In had no effect on characteristic surface and intracellular markers of cultured MSCs analysed both by flow cytometry and by real-time polymerase chain reaction. Further, the labelled MSCs differentiated towards endothelial cells and formed vascular structures.

CONCLUSION

It is possible to label human MSCs with 111In for scintigraphic tracking of stem cells delivered to the heart in clinical trials without affecting the viability and differentiation capacity of the MSCs. This creates an important tool for the control of stem cell delivery and dose response in clinical cardiovascular trials.

Changes in circulating mesenchymal stem cells, stem cell homing factor, and vascular growth factors in patients with acute ST elevation myocardial infarction treated with primary percutaneous coronary intervention

OBJECTIVE

To investigate the spontaneous occurrence of circulating mesenchymal stem cells (MSC) and angiogenic factors in patients with ST elevation acute myocardial infarction (STEMI) treated with primary percutaneous coronary intervention (PCI).

DESIGN

In 20 patients with STEMI, blood samples were obtained on days 1, 3, 7, 14, 21, and 28 after the acute PCI. Fifteen patients with a normal coronary angiography formed a control group. MSC (CD45-/CD34-), plasma stromal derived factor 1 (SDF-1), vascular endothelial growth factor A (VEGF-A), and fibroblast growth factor 2 (FGF-2) were measured by multiparametric flow cytometry and enzyme linked immunosorbent assay (ELISA).

RESULTS

Circulating CD45-/CD34- cells were significantly decreased on day 7 compared with day 3. Cell counts normalised one month after the acute onset of STEMI. The changes were mainly seen in patients with a large infarction. Plasma SDF-1 increased significantly from day 3 to day 28, and VEGF-A and FGF-2 increased significantly from day 7 to day 28.

CONCLUSIONS

Spontaneous sequential fluctuations in MSC and the increase in vascular growth factor concentrations after STEMI suggest that the optimal time for additional stem cell therapy is three weeks after a myocardial infarction to obtain the maximum effects by stimulating endogenous growth factors on the delivered stem cells.

Complementation of the Magnaporthe grisea deltacpkA mutation by the Blumeria graminis PKA-c gene: functional genetic analysis of an obligate plant pathogen

Obligate plant-pathogenic fungi have proved extremely difficult to characterize with molecular genetics because they cannot be cultured away from host plants and only can be manipulated experimentally in limited circumstances. Previously, in order to characterize signal transduction processes during infection-related development of the powdery mildew fungus Blumeria graminis (syn. Erysiphe graminis) f. sp. hordei, we described a gene similar to the catalytic subunit of cyclic AMP-dependent protein kinase A (here renamed Bka1). Functional characterization of this gene has been achieved by expression in a deltacpkA mutant of the nonobligate pathogen Magnaporthe grisea. This nonpathogenic M. grisea deltacpkA mutant displays delayed and incomplete appressorium development, suggesting a role for PKA-c in the signal transduction processes that control the maturation of infection cells. Transformation of the deltacpkA mutant with the mildew Bka1 open reading frame, controlled by the M. grisea MPG1 promoter, restored pathogenicity and appressorium maturation kinetics. The results provide, to our knowledge, the first functional genetic analysis of pathogenicity in an obligate pathogen and highlight the remarkable conservation of signaling components regulating infection-related development in pathogenic fungi.

Risk factors for nosocomial intensive care infection: a long-term prospective analysis

BACKGROUND

To identify risk factors for nosocomial infection in intensive care and to provide a basis for allocation of resources.

METHODS

Long-term prospective incidence study of risk factors for nosocomial infection in the surgical-medical intensive care unit of a university hospital.

RESULTS

A total of 2671 patients were admitted during four years, and 562 of 574 patients staying >48 h were observed during 4921 patient days (median length of stay 5 days, range 2-114). Of these, 196 (34%) patients had 364 nosocomial infections after median 8-10 days, an infection rate of 14/100 admissions. Infection prolonged length of stay 8-9 days and doubled the risk of death. The infections were 17% blood stream, 26% pneumonias, 34% wound, 10% urinary tract and 13% other infections. The incidence of bloodstream infection declined significantly during the study years, from 12% to 5%. In multiple regression analysis, the important variables for infection were central venous catheter, mechanical ventilation, pleural drainage and trauma with open fractures. High age, immunosuppression and infection on admission did not influence the risk of acquiring infection. Trauma patients constituted 24% of the study population. Trauma with open fractures increased the risk of infection more than twice (P=0.003), mainly due to wound infections.

CONCLUSION

Trauma cases, with open fractures, were the patients most at risk of infection, despite low disease severity scores. Resources to prevent nosocomial infection should be allocated to these patients.

Evidence that the cAMP pathway controls emergence of both primary and appressorial germ tubes of barley powdery mildew

Development of conidia of barley powdery mildew involves the formation of a primary germ tube (PGT), an appressorial germ tube (AGT), and an appressorium. Previously, it was found that cyclic AMP (cAMP) was involved in these developmental processes. Comparison of development on the host surface with two types of cellulose membrane revealed that frequency of PGT emergence was surface independent. On one type of cellulose, where the frequencies of both AGT and appressorial differentiation were similar to that on the host surface, cAMP levels and protein kinase A (PKA) activities had a biphasic pattern with peaks at 15 min and 4 h after inoculation (prior to PGT and AGT emergence, respectively). The effect of manipulating cAMP levels was tested on another type of cellulose membrane, which stimulated a lower degree of AGT and appressorial formation than the host surface. Cholera toxin and forskolin, activators of adenylyl cyclase, significantly increased PGT emergence, but cAMP did not. Cholera toxin, forskolin, and cAMP increased the frequency of AGT and appressorial formation, but in a time-dependent manner.

Involvement of cAMP and protein kinase A in conidial differentiation by Erysiphe graminis f. sp. hordei

Erysiphe graminis f. sp. hordei, the causal agent of barley powdery mildew, is an obligate biotroph. On arrival on the host, a primary germ tube (PGT) emerges from the conidium. An appressorial germ tube (AGT) then appears, forms an appressorium, and effects host penetration. Such developmental precision may be due to multiple, plant-derived signals and to endogenous tactile and chemical signals. The transduction mechanism remains obscure. The isolation of an expressed sequence tag (EST) homologue of the catalytic subunit of cyclic AMP (cAMP)-dependent protein kinase A (PKA) enabled the corresponding gene to be characterized and the transcript to be identified in conidia and in PGT and AGT stage spores. cAMP-dependent PKA activity was detected in ungerminated conidia. These data suggest that PKA and cAMP are involved in conidial development. To substantiate this we exploited the responses of developing conidia to various surfaces, including exposure to the host leaf (fully inductive to AGT formation), cellulose membrane (semi-inductive), and glass (non-inductive). Assessment of fungal development, following application of exogenous cAMP or cAMP analogues, revealed that, at different concentrations and on different surfaces, cAMP either promoted or inhibited conidial differentiation. Various PKA inhibitors were tested for their effect on PKA activity and conidial development. A negative correlation was established between PKA inhibition in vitro and fungal development in vivo. Taken collectively, these data suggest that PKA and cAMP play a role in conidial differentiation in this obligate, plant-pathogenic fungus.

Individual lung blood flow during unilateral hypoxia: effects of inhaled nitric oxide

We hypothesized that the diversion of blood away from a hypoxic lung to the opposite oxygenated lung can be enhanced by inhaling nitric oxide (NO) into the oxygenated lung. We measured individual lung blood flow when 50 ppm NO was selectively inhaled to: a hyperoxic lung during contralateral hypoxia; a normoxic lung during bilateral normoxia; and a hyperoxic lung during bilateral hyperoxia. Twenty two patients with healthy lungs were studied during intravenous anaesthesia. The lungs were separately and synchronously ventilated. The relative perfusion of each lung was assessed by the inert gas elimination technique. Unilateral hypoxic (inspiratory oxygen fraction (FI,O2) 0.05) ventilation during contralateral hyperoxia reduced the perfusion of the hypoxic lung from a mean (SD) of 47 (9)% of cardiac output (Q'), to 30 (7)% (p<0.001) of Q'. NO inhalation to the hyperoxic lung increased its blood flow from 70 (7)% to 75 (6)% (p<0.05) of Q', and reduced the blood flow to the hypoxic lung to 25 (6)% (p<0.05). Unilateral NO inhalation during bilateral normoxia or hyperoxia had no effect on pulmonary blood flow distribution. Nitric oxide inhalation to a hyperoxic lung increases the perfusion to this lung by redistribution of blood flow if the opposite lung is hypoxic.

A retrospective analysis of nitric oxide inhalation in patients with severe acute lung injury in Sweden and Norway 1991-1994

BACKGROUND

Patients with severe acute lung injury (ALI) have been treated compassionately on doctors' initiative with inhaled nitric oxide (INO) in Sweden and Norway since 1991. In 1994 the previously used technical grade nitric oxide was replaced by medical grade nitric oxide.

METHODS

We have carried out a retrospective data collection on all identified adult patients treated with INO for >4 h during the period 1991-1994 focusing on safety aspects and patient outcome. We used the following exclusion criteria (1) Age <18 years, (2) Simultaneous treatment with extracorporeal removal of CO2 (3) NO inhalation period <4 h, (4) Incomplete or missing patient charts, (5) Use of INO in order to treat pulmonary hypertension following cardiac surgery, with little or no acute lung injury.

RESULTS

Inclusion criteria were met by 56 out of 73 identified patients. Mean age was 48+/-19 years and the median duration of INO treatment was 102 h. PaO2/FIO2 ratio at start of treatment was 85 +/- 33 mm Hg with a lung injury score (LIS) of 3.2+/-0.8. The aetiology of the lung injury was pneumonia (n= 27), sepsis (n=12) and trauma (n=8). Survival to hospital discharge was 41% and survival after 180 d was 38%. Three serious adverse events were identified, two from technical failures of the INO delivery device and one withdrawal reaction necessitating slow weaning from INO. No methaemoglobin values >5% were reported during treatment.

CONCLUSION

The overall mortality did not differ dramatically from historical controls with high mortality. Only a randomised study may determine whether INO as an adjunct to treatment alters the outcome in severe ALI. One cannot at present advocate the routine use of INO in patients with ALI outside such studies.

Hypoxic pulmonary vasoconstriction in human lungs. A stimulus-response study

BACKGROUND

A stimulus-response relation between alveolar oxygen tension and pulmonary vascular resistance has been observed in animals. This study investigated this relation in healthy human lungs. The distribution of pulmonary blood flow was measured during unilateral (1) graded hypoxia (fractional concentration of oxygen in inspired gas [FIO2] = 0.12, 0.08, and 0.05) and contralateral hyperoxia (FIO2 = 1.0; n = 6); (2) single-step hypoxia (FIO2 = 0.05) and contralateral hyperoxia (n = 5); and (3) normobaric hyperoxia and contralateral normoxia (FIO2 = 0.25; n = 6).

METHODS

Seventeen patients with healthy lungs were studied during intravenous anesthesia. The lungs were separately and synchronously ventilated. The relative perfusion of each lung was assessed by the inert gas (sulfurhexafluoride) elimination technique.

RESULTS

(1) Unilateral graded hypoxia reduced the perfusion of the hypoxic lung from a mean (+/-SD) of 52 (2)% of cardiac output (Q) during bilateral hyperoxia, to 47 (5)% (P > 0.05) 40 (3)% (P < 0.01), and 30 (8)% (P < 0.001) of Q, respectively. These progressive reductions in the perfusion of the hypoxic lung were all significantly different from each other. (2) Unilateral single-step hypoxia caused a blood flow diversion of the same magnitude as when the lung was previously ventilated with FiO2 of 0.12 and 0.08. The perfusion of the hypoxic lung was reduced from 46 (9)% of Q (bilateral hyperoxia) to 26 (4)% of Q (P < 0.01). (3) Unilateral hyperoxia did not significantly change the relative blood flow distribution between the two lungs or the pulmonary artery pressure.

CONCLUSIONS

A stimulus-response relation between graded hypoxia and blood flow diversion defines hypoxic pulmonary vasoconstriction in the normal human lung. Hyperoxia has no significant effect on vascular resistance in the normal human lung.

Surface heparinization of central venous catheters reduces microbial colonization in vitro and in vivo: results from a prospective, randomized trial

OBJECTIVE

To evaluate in vitro and in vivo the efficacy of covalent end point-attached heparin to single-lumen polyurethane central venous catheters in reducing microbial adherence and colonization.

DESIGN

In vitro study: A controlled bench study. In vivo study: A prospective, randomized, double-blind, clinical trial.

SETTING

Intensive care unit in a 1200-bed teaching hospital.

INTERVENTIONS

In vitro study: Adhesion of 17 radiolabeled clinical isolates of Staphylococci to catheters was examined in vitro. In vivo study: The outcome of heparinized and control catheters was compared in vivo in patients receiving long-term parenteral nutrition. Fifty-five adult patients were prospectively, blindly randomized to heparinized or control central venous catheters. The catheters, removed on clinical grounds, were analyzed with semiquantitative and quantitative cultures. Blood cultures were done at catheter removal.

MEASUREMENTS AND MAIN RESULTS

In vitro study: Coagulase-negative Staphylococci adhered less in vitro to heparinized catheters than to control catheters (p < .05). In vivo study: Among 32 central venous catheters, or patients who completed the study, catheter-associated bacteremia or fungemia was observed in five patients in the control group (n = 19) and in no patient with a heparinized catheter (n = 13) (p = .047). Four of 13 catheters in the heparin group were colonized compared with 14 of 19 in the control group (p = .03). Coagulase-negative Staphylococci were the most frequent microorganisms in both groups. The numbers of organisms found on colonized catheters were larger in the control group than in the heparin group.

CONCLUSIONS

Covalent end point surface heparinization appears to have a great impact on both in vitro and in vivo bacterial colonization of central venous catheters. Such heparinization can be a practical and economical approach to the prevention of catheter-associated bacteremia or fungemia.

Intestinal amino acid content in critically ill patients

BACKGROUND

The purpose of the study was to determine the concentrations of free amino acids and the total protein content of the human intestinal mucosa during critical illness.

METHODS

The free amino acid and protein concentrations in endoscopically obtained biopsy specimens from the duodenum and the distal colonic segments were determined on 19 critically ill patients. The free amino acids were separated by ion exchange chromatography and detected by fluorescence, and the protein content was quantified by the method of Lowry.

RESULTS

In general, the typical amino acid pattern of the intestinal mucosa was seen, with very high levels of taurine, aspartate and glutamic acid. The main difference, as compared to a reference series of healthy subjects, was the elevated glutamine concentration of the duodenal mucosa. This amino acid was unaltered in the descending colon and depressed in the rectum. At the same time, the glutamatic acid concentrations were unaltered, suggesting that the degradation of glutamine was not increased in the septic state of the majority of the patients studied. Phenylalanine and the two branched-chain amino acids, valine and leucine, were elevated in the duodenal mucosa, and in the colonic mucosa, methionine and phenylalanine were elevated; otherwise, all the other individual amino acids were unaltered or depressed.

CONCLUSIONS

The alterations seen in mucosal free amino acid and protein concentrations in connection with critical illness are different in many respects and contrast with the findings seen after starvation or moderate surgical trauma.

Extracorporeal carbon dioxide removal performed with surface-heparinized equipment in patients with ARDS

To avoid the drawbacks of systemic anticoagulation during prolonged extracorporeal circulation in patients with adult respiratory distress syndrome (ARDS) a heparinization technique has been developed by which partially degraded heparin can be covalently end-point attached to the surface of the equipment constituting the extracorporeal circuit (Carmeda Bio-Active Surface, CBAS) thereby localizing the anticoagulatory effect. Since 1986 we have used extracorporeal circuits and membrane lungs coated with the CBAS for extracorporeal lung assistance (ECLA) in 14 patients suffering from ARDS. The patients were on ECLA for 3 to 55 days with a survival rate of 43%. Our experience so far is that by using equipment coated with CBAS it is possible to perform long-term extracorporeal circulation with a minimum of intravenously administered heparin, thus avoiding the risk of major coagulation defects.

Hypoxia-induced pulmonary vasoconstriction in the human lung. The effect of isoflurane anesthesia

The influence of isoflurane on hypoxic pulmonary vasoconstriction (HPV) was studied in eight subjects prior to elective surgery. The lungs were ventilated separately with a double-lumen endobronchial catheter. After oxygen ventilation of both lungs for 30 min during intravenous barbiturate anesthesia, the test lung was rendered hypoxic by ventilation with 8% O2 in nitrogen. The control lung was ventilated continuously with 100% O2. Isoflurane was added to the inspired gas, so that end-tidal concentrations of 1% and 1.5% were obtained. Cardiac output (QT) was determined by thermodilution, and the distribution of blood flow between the lungs was assessed from the excretion of a continuously infused, poorly soluble gas (SF6). The hypoxic challenge during intravenous anesthesia resulted in a reduction in the fractional perfusion of the test lung from 54% to 41% of QT. Mean pulmonary arterial pressure increased by 46%, and pulmonary vascular resistance (PVR) of the test lung more than doubled. Arterial oxygen tension fell from 375 mmHg (50 kPa) to 101 mmHg (13.5 kPa). Adding isoflurane to the inhalation gas, first at a concentration of 1%, then 1.5%, caused no further significant change in the distribution of pulmonary blood flow, although six of the eight subjects showed a small increase in test lung blood flow at isoflurane 1.5%. There was no change in PVR or in any other circulatory variable. Arterial blood gases remained essentially unaltered. When the hypoxic challenge was discontinued, all variables returned to control values.(ABSTRACT TRUNCATED AT 250 WORDS)

Hypoxia-induced vasoconstriction in human lung exposed to enflurane anaesthesia

The degree of hypoxic pulmonary vasoconstriction was studied in eight subjects during enflurane anaesthesia and was compared with that during intravenous pentobarbital anaesthesia in the same subjects. The lungs were ventilated separately with the aid of a double-lumen endobronchial catheter. After preoxygenation of both lungs for 30 min, during intravenous anaesthesia, the right lung (test lung) was rendered hypoxic by ventilation with 6% O2 in nitrogen. The left lung (control lung) was ventilated continuously with 100% oxygen. Cardiac output (QT) was determined by thermodilution, and the distribution of blood flow between the lungs was assessed from the elimination of a continuously infused, poorly soluble inert gas (SF6). The hypoxic challenge resulted in a reduction of the distribution of perfusion to the test lung from 57% to 36% of QT. Mean pulmonary arterial pressure increased by 37% and pulmonary vascular resistance of the test lung doubled. Arterial oxygen tension decreased from 45.9 to 9.5 kPa. Administration of enflurane to an end-tidal concentration of 2% to both lungs caused no significant change in the distribution of the pulmonary blood flow, PVR, or any other circulatory variable. The arterial blood gases remained unaltered. When the hypoxic challenge was discontinued, all variables returned towards control values. The findings suggest that the inhalational anaesthetic enflurane does not reduce the hypoxic vasoconstrictor response in the human lung.

Reversal of nitrous oxide-induced depression of hypoxic pulmonary vasoconstriction by lignocaine hydrochloride during collapse and ventilation hypoxia of the left lower lobe

The blood flow to the left lower lobe (QL), and total (QT) pulmonary blood flow, were measured in 10 open-chest dogs using electromagnetic flowmeters. Ventilation of the left lower lobe with 7% oxygen in nitrogen produced a greater reduction in QL/QT (41%) than lobar ventilation with 7% oxygen in nitrous oxide (33%). Lobar collapse reduced QL/QT by 65%, but there was no change in QL/QT when 50% nitrous oxide was administered to the right lung. The i.v. infusion of lignocaine hydrochloride during ventilation of the lobe with 7% oxygen in nitrogen produced no change in QL/QT. However, lignocaine infusion during lobar ventilation with 7% oxygen in nitrous oxide produced a further reduction in QL/QT to a value which was not significantly different from that observed during ventilation with 7% oxygen in nitrogen. Lignocaine had no effect on QL/QT during lobar collapse whether the right lung was ventilated with 50% oxygen in nitrogen or 50% oxygen in nitrous oxide. It is concluded that lignocaine reverses the depression of hypoxic pulmonary vasoconstriction produced by lobar ventilation with nitrous oxide.

Effect of lignocaine and nitrous oxide on hypoxic pulmonary vasoconstriction in the dog constant-flow perfused left lower lobe preparation

The effects of lignocaine hydrochloride on hypoxic pulmonary vasoconstriction in the presence and absence of nitrous oxide were examined during constant-flow perfusion of the left lower lobe. Lignocaine had no effect on lobar pulmonary vascular resistance during lobar ventilation with oxygen or 7% oxygen in nitrogen. The substitution of nitrous oxide for nitrogen diminished the hypoxic vasoconstrictor response, whilst the subsequent infusion of lignocaine during 7% oxygen in nitrous oxide increased the response to a value which was not significantly different from that produced by 7% oxygen in nitrogen alone.

Lung and chest wall mechanics during differential ventilation with selective PEEP

Eight patients free from cardio-pulmonary disease and with a mean age of 46 years were studied during general anaesthesia in the lateral position. Measurements of hemithoracic mechanics were made during four different modes of ventilation: 1. Conventional ventilation (free distribution of ventilation) with no positive end-expiratory pressure (PEEP) (CV), 2. differential ventilation (50% of ventilation to each lung) with no PEEP (DV:0), and 3 and 4. DV with selective PEEP of 0.8 and 1.6 kPa, respectively, to the dependent lung only (DV:8, DV:16). During CV, 60% of ventilation was distributed to the non-dependent lung. Non-dependent hemithoracic compliance was 64% greater and inspiratory resistance 39% lower than those of the dependent hemithorax. No significant differences between the two hemithoraces were noted during DV:0, but on application of selective PEEP the compliance of the dependent hemithorax increased and its resistance decreased. With DV:16, the compliances of the two hemithoraces were essentially equal, as were their resistances. Selective PEEP caused a larger volume increase in the dependent lung than general PEEP. Selective PEEP reduced the volume of the non-dependent lung but only by 1/3 of the simultaneous increase in that of the dependent lung. Oesophageal pressure increased only slightly on selective inflation of the dependent lung, and remained negative within the 21 volume range studied. It is suggested that the altered mechanics of the dependent lung during selective PEEP result in a more even distribution of the inspired gas within that lung.

Hypoxic pulmonary vasoconstriction in man: effects of hyperventilation

The pulmonary vasoconstriction response to hypoxia was studied in eight anaesthetized supine subjects. One lung was made hypoxic while the other was ventilated with 100% oxygen. This was achieved by separating the tidal gas-distribution to the lungs by means of a double-lumen tracheal catheter. The hypoxic pulmonary vasoconstriction (HPV) response was estimated from the blood flow diversion away from the hypoxic lung. Blood flow distribution between the lungs was calculated from the regional expired carbon dioxide production, assuming regional carbon dioxide production to be proportional to blood flow. The subjects were studied during six different conditions. Firstly, when ventilated with 100% oxygen to both lungs at a PaCO2 of about 6 kPa. Secondly, with 100% oxygen to the left lung and 5% oxygen in nitrogen to the right (test) lung. The ratio between carbon dioxide output from right and left lung was calculated. These measurements were repeated during two states of hyperventilation (PaCO2 of about 4.5 kPa and 3.5 kPa, respectively) with and without hypoxia (conditions 3-6). During normoventilation, blood flow distribution between the lungs was equal. During hypoxia, blood flow distribution to the hypoxic lung decreased by 35% of the pre-hypoxic value. Furthermore, a decrease in arterial oxygen tension from 51.5 +/- 4.5 to 11.5 +/- 2.1 kPa was observed. During excessive hyperventilation (PaCO2 3.2 +/- 0.2 kPa), blood flow distribution to the hypoxic right lung decreased by only 10% of its pre-hypoxic value. A further decrease in arterial oxygen tension to 8.5 +/- 1.8 kPa was observed. This decrease in PaO2 was possibly due to an increased venous admixture caused by an abolished HPV response. It is concluded that hyperventilation counteracts hypoxic pulmonary vasoconstriction in man.

Hypoxic pulmonary vasoconstriction in the human lung: effect of repeated hypoxic challenges during anesthesia

Six patients, ages 29-58 yr, were investigated during barbiturate and fentanyl anesthesia. After intubation with a double-lumen bronchial catheter, one lung was ventilated continuously with 100% O2, and the other was rendered hypoxic during three 15-min periods by ventilation with 95% N2 + 5% O2, with intervening 15-min periods of oxygen ventilation. Cardiac output was determined by thermodilution, and the distribution of blood flow between the lungs was assessed from the excretion of a continuously infused poorly soluble gas (SF6). The first hypoxic challenge resulted in a 10% increase in cardiac output (QT) and a reduction in the fractional perfusion of the test lung from 57% to 31% of QT. The pulmonary artery mean pressure increased by 54%, and the vascular resistance of the test lung increased threefold. The venous admixture increased from 19% to 40% of QT, whereas the inert gas shunt remained unaltered at 15% (inert gases also being eliminated by nitrogen-ventilated areas). The arterial oxygen tension decreased from 353 mmHg to 79 mmHg. On resumption of the control state, central hemodynamics and gas exchange returned to the initial values. The second and third hypoxic challenges resulted in reductions in the fractional perfusion of the test lung to 35% and 37% of QT. All other variables were altered to the same degree as during the first challenge. The authors conclude that hypoxic challenge of one lung in an intravenously anesthetized human subject elicits a maximum vasoconstrictor response within the first 15 min, and this response cannot be potentiated by repeated challenges.

Hypoxic pulmonary vasoconstriction in the human lung: the effect of prolonged unilateral hypoxic challenge during anaesthesia

The influence of time on the pulmonary vasoconstrictor response to hypoxia was studied in six subjects during general anaesthesia and artificial ventilation prior to elective surgery. The lungs were intubated separately with a double-lumen bronchial catheter. After preoxygenation of both lungs for 30 min, the test lung was rendered hypoxic for 60 min by ventilation with 5% O2 in N2, with the control lung still being ventilated with 100% O2. Cardiac output was determined by thermodilution, and the distribution of blood flow between the lungs was assessed from the excretion of a continuously infused poorly soluble gas (SF6). The fractional perfusion of the test lung decreased from 53% to 25% of cardiac output within the first 15 min of unilateral hypoxia. The pulmonary artery mean pressure increased by 14% and the pulmonary vascular resistance (PVR) of the test lung increased by 54%. Venous admixture increased from 21% to 39% of cardiac output, while the "true" shunt was maintained at about 15%. Arterial oxygen tension (Pao2) fell from 45 kPa to 12 kPa. Prolonging the unilateral hypoxic challenge caused no further change in the redistribution of the pulmonary blood flow, but cardiac output and pulmonary artery mean pressure continued to increase to 40%-50% above control values after 1 h of hypoxia. The PVR of the test lung remained unchanged. The findings suggest that there is an immediate vasoconstrictor response to hypoxia in the human lung and that there is no further potentiation or diminution, of the response during a 60-min period of hypoxia.

Ventilation and perfusion of each lung during differential ventilation with selective PEEP

Lung perfusion was studied in 10 patients (mean age 58 yr) in the lateral position during enflurane anesthesia. They were ventilated through a double-lumen endotracheal catheter: 1) by one ventilator with free distribution of ventilation between the lungs, with no (zero) end-respiratory pressure (ZEEP); 2) as above but with a general positive end-expiratory pressure (PEEP) of 9 cmH2O; or 3) by two ventilators with equal distribution of ventilation between the lungs and with a selective PEEP of 8 cmH2O to the dependent lung only. Total ventilation was on average 8 l/min (BTPS) throughout the study. During the first method, 34% of ventilation was distributed to the dependent and 66% to the nondependent lung. Cardiac output (thermodilution) was 4.5 l/min, 57% being distributed to the dependent lung as assessed by iv boli of Xenon 133. During the second method, ventilation was assumed to be distributed equally between the lungs. Cardiac output was decreased to 3.8 l/min, and the dependent lung received 81% of lung blood flow. During the third method, cardiac output was significantly greater than during the second method (4.1 l/min), 51% passing to the dependent lung. Peak and end-inspiratory airway pressures were 5-18 cm H2O lower during selective than during general PEEP. Arterial oxygen tension was significantly greater during the third method than during either of the other ventilator settings and the alveolar-arterial oxygen tension difference was almost halved compared with the first method. It is concluded that differential ventilation with selective PEEP improves ventilation-perfusion matching and thus oxygenation.

Selective PEEP in acute bilateral lung disease. Effect on patients in the lateral posture

Seven patients with acute respiratory failure due to diffuse and fairly uniform lung disease were studied during mechanical ventilation in the lateral decubital position with: (a) zero end-expiratory pressure (ZEEP) through a double-lumen oro-bronchial tube to permit a recording of the ventilation to each lung; (b) bilateral positive end-expiratory pressure (PEEP) of 1.2 kPa, with maintenance of ventilation distribution between lungs as observed during ZEEP; (c) selective PEEP of 1.2 kPa, applied to the dependent lung only, with ventilation as during ZEEP; and (d) conventional PEEP of 1.2 kPa applied to both lungs through a single-lumen tube, with free distribution of ventilation between the lungs. During ZEEP, 69% of ventilation was distributed to the non-dependent and 31% to the dependent lung; cardiac output was 6.51 X min-1, venous admixture (QS/QT) 40% and arterial oxygen tension (PaO2) 8.3 kPa. With bilateral PEEP, functional residual capacity (FRC) increased by 0.331, cardiac output was reduced to 5.11 X min-1 and venous admixture to 32%. PaO2 increased to 10.1 kPa. With selective PEEP the dependent lung FRC increased by 0.211 and the FRC of the non-dependent lung decreased by 0.081. Cardiac output increased to 6.11 X min-1, which was no longer significantly different from that during ZEEP. Venous admixture remained at the same level as with bilateral PEEP.(ABSTRACT TRUNCATED AT 250 WORDS)

Differential ventilation in acute bilateral lung disease. Influence on gas exchange and central haemodynamics

Eight patients with acute respiratory failure (ARF) due to diffuse and rather uniform lung disease were intubated with a double-lumen bronchial tube and ventilated in the lateral decubital position by two synchronized ventilators. Ventilation of each lung was individually adjusted to match the expected regional blood flow (differential ventilation). When ventilation with equal volumes (i.e. 50% of tidal volume to each lung) was performed, a 19% reduction of venous admixture (P less than 0.001) and a 22% increment in arterial oxygen tension (P less than 0.001) were seen. Comcomitantly, the cardiac output increased by 17% (P less than 0.001), to which a reduced pulmonary vascular resistance may have contributed. The net result was a 14% increment of the oxygen availability (P less than 0.001). An attempt to go further, giving 2/3 of the tidal ventilation to the dependent lung, was made on six of the patients. However, this ventilatory pattern did not further improve the gas exchange and also had detrimental effects on the haemodynamics. It is concluded that differential ventilation with equal tidal volumes in the lateral position can substantially improve gas exchange and central haemodynamics in patients with ARF due to diffuse lung disease.

Regional differences in lung function during anaesthesia and intensive care: clinical implications

Anaesthesia and most frequently acute respiratory failure are accompanied by a lowered functional residual capacity (FRC). This lowering promotes airway closure in dependent lung units and forces ventilation to non-dependent regions. Perfusion, on the other hand, is forced towards dependent lung units. A ventilation-perfusion mismatch is created and hypoxaemia may develop. General PEEP counters airway closure, but impedes cardiac output and forces perfusion further to dependent regions. In addition, barotrauma may occur. Improved matching of ventilation and perfusion can be achieved by: (1) positioning the subject in the lateral posture; (2) ventilating each lung separately in proportion to its perfusion (differential ventilation); and (3) applying PEEP only to the dependent lung (selective PEEP). Because of less overall intrathoracic pressure and lung expansion, interference with the total lung blood flow and the danger of barotrauma should be less than with general PEEP. Improved gas exchange with a 50-100% increase in PaO2 has been observed in a limited number of patients with acute bilateral lung disease studied so far during differential ventilation and selective PEEP.

Pressure-volume and airway closure relationships in each lung in anaesthetized man

Airway closure, functional residual capacity (FRC) and transpulmonary pressure-volume curves were assessed for each lung separately in the anaesthetized subject by means of a double lumen tracheal catheter. In the supine position airway closure occurred synchronously in the two lungs and 0.2-0.31 above FRC. The pressure- volume curves in both lungs were rather similar and critical closing pressure (CP) was approximately 3 cmH2O in each lung. In the left lateral posture, FRC was increased in the non-dependent and reduced in the dependent lung, while closing capacity (CC) remained unaltered. Airway closure was asynchronous and discontinuous between the two lungs. This was caused by the non-linear transpulmonary pressure-volume curve in the lungs, in conjunction with the vertical pleural pressure gradient. An interpulmonary "pendelluft" phenomenon was observed in the left lateral posture, increasing inhomogeneity of ventilation. It may depend on regional differences in compliance.

Ventilation-perfusion distribution during inhalation anaesthesia. Effects of spontaneous breathing, mechanical ventilation and positive end-expiratory pressure

Ventilation-perfusion (VA/Q) ratios were studied by means of an inert gas elimination technique in healthy subjects with an average age of 51 years in the supine posture (a) when awake, (b) during inhalational anaesthesia, spontaneously breathing, (c) during mechanical ventilation, and (d) when a positive end-expiratory pressure (PEEP) was applied. In the awake subject a bimodal distribution of VA/Q was recovered in most patients, one mode centered around the ratio of 1 and another, smaller mode, within low VA/Q-regions. Any shunt was less than 3% of cardiac output. With anaesthesia and spontaneous breathing, the low VA/Q mode was reduced and the shunt increased to an average of 6.2%. With mechanical ventilation, the major VA/Q mode was widened while the shunt was further increased in 4 of 10 subjects (mean 8.6%). With PEEP, the shunt was reduced and a new mode within high VA/Q-regions appeared. The shunt and low VA/Q-regions appeared. The shunt and low VA/Q-regions may be explained in terms of airway closure while the high VA/Q mode with PEEP may be attributed to the development of a zone I.

Distribution of inspired gas to each lung in anesthetized human subjects

The distribution of ventilation in man during halothane anesthesia was studied in a two-compartment lung model in which each lung was ventilated separately by means of a double-lumen tracheal tube. Eight subjects were studied prior to scheduled surgery. Tidal volume distribution was even between the lungs in the supine position (horizontal distribution) as was distribution of dynamic lung compliance, resistance and dead space. The vertical distribution was assessed when the patient was in the left lateral position. Dependent dynamic lung compliance and dead space were lower and lung resistance was higher than in the non-dependent lung. These factors favoured a non-dependent lung ventilation and, moreover, caused a re-distribution from dependent to non-dependent lung during an end-inspiratory pause (EIP), thus increasing the inhomogeneity of ventilation. The application of a positive end-expiratory pressure (PEEP) of 10 cmH2O improved dependent ventilation and abolished redistribution between the lungs. In conclusion, uneven distribution of dynamic lung compliance and lung resistance causes inhomogeneous ventilation distribution, favouring the non-dependent lung. An EIP enhances and a PEEP reduces the inhomogeneity of ventilation.

Airway closure in each lung of anesthetized human subjects

Airway closure and functional residual capacity (FRC) were assessed for each lung separately in the anesthetized subject by means of a double-lumen tracheal catheter. Airway closure was studied by argon-bolus and nitrogen-washout techniques, and FRC was calculated from single-breath nitrogen washout. Recordings were done with subjects in the supine and lateral postures. In the supine position, closing capacity (CC) exceeded FRC in each lung. Airway closure occurred synchronously in the two lungs. Argon CC was 0.05-0.1 liter larger than nitrogen CC of either lung. Minor gas trapping occurred during the vital capacity (VC) maneuver, so that inspired VC exceeded expired VC by 3%. In the left lateral posture, CC remained unaltered in either lung, whereas FRC was markedly increased in the nondependent and reduced in the dependent lung. Airway closure occurred asynchronously in the two lungs, and its distribution was discontinuous between them. Onset of airway closure in the dependent lung caused an early (60% VC) upstroke on the overall tracer gas recording (sampling of mixed expirate at the mouth), whereas onset of airway closure in the nondependent lung caused an additional upstroke at 10% VC. Gas trapping was more marked in the dependent lung than in the supine position, but some gas was released (expired VC greater than inspired VC) n the nondependent lung.

Airway closure during anaesthesia, and its prevention by positive end expiratory pressure

Airway closure, functional residual capacity (FRC) and the transpulmonary pressure volume relationship of each lung were studied in the anaesthetized subject in the supine and the left lateral positions. In the supine posture, FRC was of approximately the same size in each lung as was closing capacity (CC). CC exceeded FRC in either lung. In the left lateral position, FRC was increased by 0.91 in the non-dependent lung and was reduced by 0.21 in the dependent lung, while CC was unaltered in either lung. Consequently, FRC exceeded CC in the non-dependent lung and was further lowered beneath CC in the dependent lung. Airway closure did not occur in the non-dependent lung until an average of 0.51 of gas had been expelled after the dependent lung had ceased to empty. The addition of positive end-expiratory pressure (PEEP) in the range 0.5-2 kPa, increased FRC more in the non-dependent than the dependent lung. The findings suggest that airway closure is evenly distributed in the horizontal level, while it has a discontinuous distribution between the dependent and non-dependent lung. Moreover, the increase in lung volume caused by PEEP has a distribution that is by no means ideal for the purpose of countering airway closure.