- Email: [email protected]
Consequences of Inspired Oxygen Fraction Manipulation on Myocardial Oxygen Pressure, Adenosine and Lactate Concentrations: A Combined Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
J Mol Cell Cardiol 32, 493–504 (2000) doi:10.1006/jmcc.1999.1094, available online at http://www.idealibrary.com on
Consequences of Inspired Oxygen Fraction Manipulation on Myocardial Oxygen Pressure, Adenosine and Lactate Concentrations: A Combined Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs E. M. Siaghy1, Y. Devaux1∗, N. Sfaksi1∗, J. P. Carteaux1, D. UngureanuLongrois1,3, F. Zannad1, J. P. Villemot1, C. Burlet2 and P. M. Mertes1,2 1
Laboratoire de Chirurgie Expe´rimentale, UPRES 971068, Faculte´ de Me´decine de Nancy, Universite´ Henri Poincare´, Nancy I, France, 2Laboratoire de Biologie Cellulaire, and 3De´partement d’Anesthe´sieRe´animation Chirurgicale, CHU Brabois, Vandoeuvre-les-Nancy, France (Received 29 June 1999, accepted in revised form 31 December 1999) E. M. S, Y. D, N. S, J. P. C, D. U-L, F. Z, J. P. V, C. B P. M. M. Consequences of Inspired Oxygen Fraction Manipulation on Myocardial Oxygen Pressure, Adenosine and Lactate Concentrations: A Combined Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs. Journal of Molecular and Cellular Cardiology (2000) 32, 493–504. Adenosine is a potent vasodilator whose concentration has been shown to increase in cardiac tissue in response to hypoxia. However, the time-dependent relationship between the levels of myocardial interstitial adenosine and tissue oxygenation has not yet been completely established. Therefore, the purpose of this study was to investigate the complex relationship between tissue myocardial oxygen tension (PtiO2) and interstitial myocardial adenosine and lactate concentrations by developing a new technique which combines a cardiac microdialysis probe and a Clark-type P2 electrode. The combined and the single microdialysis probes were implanted in the left ventricular myocardium of anesthetized pigs. The consequences of the combined use of microdialysis and P2 probes on myocardial PtiO2 and microdialysis performances against glucose were evaluated. A moderate but significant reduction in the relative recovery against glucose of the combined probe was observed when compared to that of the single microdialysis probe (42±2 v 32±1%, mean±...; n= 5; P<0.05), at 2 ll/min microdialysis probe perfusion flow. Similarly, myocardial oxygen enrichment, measured by the P2 electrode, was negligible when microdialysis probe perfusion flow was 2 ll/min. Systemic hypoxia (FiO2= 0.08) resulted in a significant decrease in PtiO2 from 30±4 to 11±2 mmHg, limited increase in coronary blood flow (CBF), and a significant increase in myocardial adenosine and lactate concentrations from 0.34±0.05 to 0.98±0.06 lmol/l and from 0.45±0.05 to 0.97±0.06 mmol/l respectively (P<0.05). Increasing the FiO2 to 0.3 restored the PtiO2 and hemodynamic parameters to baseline values with no changes in interstitial adenosine and lactate concentrations. Nevertheless, myocardial interstitial adenosine remained significantly higher than baseline values. In conclusion, this study demonstrates the ability of a combined probe to measure simultaneously regional myocardial PtiO2 and metabolite concentration during hypoxia. The hypoxia-induced increase in myocardial adenosine persists after correction of hypoxia. The physiological significance of this observation requires further studies. 2000 Academic Press
K W: Adenosine; Hypoxia; Myocardial oxygenation; PtiO2; Cardiac microdialysis.
∗ Research Fellow. Please address all correspondence to: Dan Ungureanu-Longrois, De´partement d’Anesthe´sie-Re´animation Chirurgicale, CHU de Brabois, Alle´e du Morvan, 54511 Vandoeuvre-les-Nancy, France. Tel: 33 (0) 3-83-15-41-66; Fax: 33 (0) 3-83-15-36-88; E-mail: d.longrois@ chu.nancy.fr
0022–2828/00/030493+12 $35.00/0
2000 Academic Press
494
E. M. Siaghy et al.
Introduction Previous studies have suggested that adenosine is one of the metabolic mediators that account for the changes in coronary blood flow during conditions of hypoxemia-triggered decreased oxygen supply and increased oxygen demand.1–3 Most of the evidence in support of the adenosine hypothesis for regulation of coronary blood flow (CBF) during hypoxemia comes from studies in which changes in cardiac tissue or coronary venous adenosine levels are correlated with changes in myocardial oxygen consumption and/or CBF.1,4–6 Nevertheless, controversies concerning the regulatory role of adenosine on CBF during conditions of hypoxemia still persist. For example, both an increase7 or no change8 in myocardial interstitial adenosine levels during hypoxia have been observed. One of the possible explanations for the difficulties in clearly establishing the regulatory role of adenosine during hypoxemia is that the release of adenosine mainly depends on myocardial oxygen availability which is more accurately reflected by regional tissue myocardial oxygen tension (PtiO2) than by CBF. Moreover, it is well recognized that cardiac tissue or coronary venous adenosine levels may not accurately reflect the myocardial interstitial adenosine concentration to which the coronary resistance vessels are exposed.2,3,7 Therefore, a technological approach that permits simultaneous measurements of both regional PtiO2 and myocardial interstitial adenosine concentration is still required. The purpose of the present study was to investigate the complex relationship between PtiO2 and interstitial myocardial adenosine and lactate concentrations by developing a new technique which combines a cardiac microdialysis probe and a Clark-type P2 electrode.
Materials and Methods Combined microdialysis-PO2 probe A flexible polarographic microcatheter Clark-type P2 probe (diameter 500 lm, length 20 mm, computer-supported LICOX system; GMS, Kiel-Mielkendorf, Germany) was combined with a concentric flexible microdialysis probe (membrane outer diameter 340 lm, length 20 mm, molecular weight cut-off (MWCO): 50 kD; Medicorp, Vandoeuvre, France). The sensitive areas of both the microdialysis probe and the P2 electrode were linked
using a metallic collar allowing their implantation within the same site through a fine custom-made guiding needle specially designed for this purpose. For correction of the myocardial PtiO2 measurements, the temperature within the myocardium was monitored and myocardial PtiO2 values were adjusted to myocardial temperature by means of a computer software (LICOX system; GMS, KielMielkendorf, Germany). Microdialysis probes were perfused with a Ringer solution (Fandre, Ludres, France) using a microinjection pump (Harvard Pump, Cambridge, MA, USA). An expanded view of the assembled combined P2/microdialysis probe is illustrated in Figure 1. Probes recovery performances, as determined in vitro, were 76±2% for adenosine and 56±2% for lactate.
Animal preparation Domestic 4–6-month-old pigs weighing 30–40 kg were chosen as experimental animals. All animals were housed and treated in accordance with accepted practices for humane care of laboratory animals. Animals were fasted overnight and premedicated with 10 mg/kg ketamine (Park-Davis, Courbevoie, France), and 0.2 mg/kg diazepam (Roche, Neuilly-Sur-Seine, France), given intramuscularly. Anesthesia was induced with 5 mg/ kg of thiopentone sodium (Specia, Rhone-PoulencRhorer, Paris, France) and maintained with a continuous infusion at a rate of 100–150 mg/h. This anesthetic regimen was continued during the entire study period in all animals. Neuromuscular blockade was obtained with pancuronium bromide (0.6 mg/kg, i.v.), with additional doses of 0.3 mg/ kg administered at 30 min intervals. The trachea was intubated with a cuffed endotracheal tube, and the lungs were mechanically ventilated with a mixture of oxygen and N2O at a minute volume of 150 ml/kg. Adjustments were made to maintain the PaCO2 in the range of 35–45 mmHg. A median sternotomy was performed and an aortic catheter was inserted via the right common carotid artery to measure mean arterial pressure (MAP). The electrocardiogram was monitored continuously. An electromagnetic flowmeter (Skalar, Stockholm, Sweden) was placed on the left anterior descending coronary artery (LAD) to measure coronary blood flow. A combined microdialysis-P2 probe and a single microdialysis probe as control were successively implanted into the free wall of the left ventricle as close as possible to the LAD as previously described.9 Briefly, the probes were inserted into a fine guiding needle which was then
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
495
Figure 1 Diagram showing the assembly and implantation of the combined microdialysis and Clark-type oxygen P2 electrode within the pig myocardium. The insert represents the active area of both microdialysis probe and Clark-type oxygen P2 electrode.
passed through the myocardium. As soon as the needle traversed the myocardium, it was removed and the probe was gently pulled back until the microdialysis membrane and/or active area of the P2 electrode were completely embedded within the myocardium, and then sutured into place. The microdialysis effluent was collected in plastic Eppendorf microtubes kept on ice during the entire collection period.
Experimental design At the end of the surgical preparation period the animals were allowed to stabilize for 2 h at FiO2 of
0.3. Baseline recordings of hemodynamic, myocardial PtiO2 and blood gas measurements were performed and microdialysate collection was started with 20 min sampling periods.
Consequences of microdialysis-PO2 probes coupling on microdialysis and PO2 electrode performances The consequences of spatial hindrance due to the presence of the P2 probe on microdialysis performance was assessed by comparing the in vivo relative recovery (Rrinvivo) of glucose of the single microdialysis probe and the combined probe. The in vivo relative recovery rate as a function of perfusion flow was determined. The probes were perfused at increasing flow from 0.5 to 12 ll/min. An
496
E. M. Siaghy et al.
equilibration period of at least 30 min was observed after each change in perfusion flow followed by three consecutive 20 min collection periods. Simultaneous plasma glucose concentration measurements were performed at the end of each sampling period. In vivo relative recovery rate was calculated according to the following formula: [Glucose]microdialysate , [Glucose]plasma
Rrinvivo=
based on the assumption that the concentration of glucose in the extracellular fluid of myocardial tissue is the same as that in plasma under steady state conditions.10,11 The consequences of oxygen enrichment of myocardial interstitial space due to the microdialysis probe perfusion with the Ringer solution containing 115 mmHg of O2 on P2 electrode measurements was assessed by measuring PtiO2 at increasing microdialysis probe perfusion flows (from 0 to 12 ll/ min). Repeated LAD occlusions were performed at each microdialysis probe perfusion flow to confirm the ability of the combined probe to rapidly detect an acute drop in PtiO2.
Consequences of changes in FiO2 on myocardial PtiO2, hemodynamic parameters and myocardial interstitial adenosine and lactate concentrations For these experiments, microdialysis probes were perfused at 2 ll/min flow rate. A 10 min equilibration period was observed following every FiO2 modification before microdialysis collection was started. During the hypoxemia period, FiO2 was lowered from 0.3 to 0.08 by replacing O2 with N2O. Following a 30 min hypoxemia period, the FiO2 was switched to 0.3 again. In order to achieve complete hemodynamic and metabolic recovery, a 1 h period with FiO2 at 0.3 was observed between the hypoxemia and hyperoxemia protocols. During the hyperoxemia period, FiO2 was raised from 0.3 to 0.8 and readjusted after 30 min to 0.3 again. Simultaneous measurements of myocardial PtiO2, hemodynamic recordings, microdialysis samples of adenosine and lactate and blood gas measurements were performed at each period.
Analytic procedures Arterial blood gas and pH determinations were performed on an ABL 4 automated blood gas analyzer (Radiometer A/S, Copenhagen, Denmark). In
order to determine the relative recovery of probes against glucose, serial arterial blood samples were drawn into chilled tubes at the end of each microdialysis sample collection period and centrifuged for 10 min at 2500 rpm. After centrifugation, the plasma was stored at −80°C until assay. Glucose in microdialysis and plasma samples and lactate in the myocardial microdialysis samples were analyzed using enzymatic kinetic methods (Glucose: Biotrol, Paris, France; Lactate: Kit 256773, Boehringer Mannheim GmbH Diagnostica, Mannheim, Germany). Adenosine in myocardial microdialysis samples was analyzed by high-performance liquid chromatography as previously described.12 Briefly, separation of compounds was achieved using a Symmetry C18-column (Symmetry, Waters, St Quentin-Yvelines, France) and a 1% (pH 5.3) to 25% (pH 5.58) methanol in 1 m KH2PO4 gradient. All peaks were detected at 254 nm and unknown samples were identified and quantified by comparing retention time and peak areas to known standards.
Statistical analysis Results are expressed as the mean±... Linear regression was performed to study the correlation between PaO2 or PtiO2 values and myocardial interstitial adenosine concentrations. Mann–Whitney U-test was performed to compare myocardial interstitial concentrations of adenosine and lactate during hypoxemia or hyperoxemia to normoxemia period. One-way analysis of variance with repeated measures was performed to compare the changes in PtiO2 as a function of perfusion flow, followed by Fisher test when statistical significance was detected. The criterion of significance was P<0.05.
Results Consequences of the combined use of microdialysis and PO2 probes on glucose relative recovery measurements and myocardial tissue PO2
Consequences on microdialysis probe glucose recovery Typical in vivo relative recovery results obtained with a 50 kD MWCO microdialysis membrane against glucose are displayed in Figure 2. As expected, the relative recovery rate was found to decrease in an exponential manner with increasing perfusion flow (n=5). A moderate but significant reduction in the recovery rate against glucose of
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
497
Figure 2 Relation between relative recovery of the combined and single microdialysis probes. Values are mean±... ∗ P<0.05 v single microdialysis probe for each microdialysis probe perfusion flow. Ρ, Single microdialysis probe; Ε, combined probe.
the combined probe was observed when compared to that of the single microdialysis probe ranging from 42±2% v 32±1% for the single and combined probe respectively (P=0.0017) at a 2 ll/min perfusion flow.
Consequences on PO2 electrode measurements Basal myocardial PtiO2 level measured in the absence of microdialysis probe perfusion (perfusion flow 0 ll/min) was 23±2 mmHg (n=5). As could be expected, perfusion of the microdialysis probe with Ringer solution containing a partial P2 pressure of 115 mmHg resulted in a progressive increase in myocardial PtiO2 values in relation to microdialysis perfusion flow. These were of 23±2 mmHg
at 0 ll/min and 64±7 mmHg at 12 ll/min. However, oxygen enrichment was negligible when microdialysis probe perfusion flow was 2 ll/min (Fig. 3). Regression analysis showed a significant linear correlation of PtiO2 with microdialysis perfusion flow (r=0.97). Moreover, when coronary occlusion was performed, myocardial PtiO2 values in the LAD region rapidly decreased from 23±2 mmHg to 2±4 mmHg at 0 ll/min perfusion flow, and from 30±2 to 6±2 mmHg at 2 ll/min perfusion flow (Fig. 4). Based on these results, a perfusion flow of 2 ll/min was used for all the subsequent in vivo experiments. Using this perfusion flow, we reached a compromise between the ability to detect a rapid change in interstitial metabolite concentration and a minimum oxygen enrichment
498
E. M. Siaghy et al.
Figure 3 Relation between myocardial tissue P2 (PtiO2) and perfusion speed of microdialysis probe. Values are mean±... ∗ P<0.05 v single microdialysis probe for each microdialysis probe perfusion flow.
of myocardial interstitial space due to the microdialysis probe perfusion.
Consequences of FiO2 changes on myocardial PtiO2 and myocardial interstitial adenosine and lactate concentrations
Myocardial PtiO2, blood gas and hemodynamic parameters Decreasing the FiO2 from 0.3 to 0.08 reduced arterial PaO2 from a baseline of 128±7 to 63±5 mmHg (0.16 FiO2) and 36±5 mmHg (0.08 FiO2), respectively. Arterial hypoxemia resulted in a significant increase in heart rate, a slight but not significant decrease of mean arterial pressure, and
an increase in CBF through the LAD, as an attempt to maintain oxygen delivery to the left ventricular myocardium close to its baseline value [Table 1(a)]. Under basal conditions (0.3 FiO2 and 2 ll/min perfusion flow of microdialysis probe), PtiO2 values were 30±4 mmHg within the left ventricular area during the control recording. No changes were noted during the subsequent control periods until the end of the experiments. During hypoxia, a significant decrease of myocardial PtiO2 in relation with FiO2 and arterial PaO2 was observed (Fig. 5). A significant linear correlation between myocardial PtiO2 and both arterial PaO2 (r=0.98) and FiO2 (r=0.97) was observed. During the hyperoxemia period, increasing FiO2 from 0.3 to 0.8 raised arterial PaO2 from a baseline of 128±7 mmHg to 170±9 mmHg (0.5 FiO2) and
499
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
Figure 4 Diagram showing the evolution of myocardial PtiO2 measured by combined probe at different microdialysis probe perfusion flow and the effect of LAD occlusion on myocardial PtiO2.
Table 1 Hemodynamic parameters at hypoxemia (a) and hyperoxemia (b) challenges. (a)
FiO2 0.3
HR (beats/min) MAP (mmHg) CBF (ml/min) PaO2 (mmHg)
94±2 98±2 49±8 128±7
0.16
0.08
116±6 95±7 62±7∗ 63±5
162±6∗ 75±3 85±11∗ 36±5
(b)
110±5 102±3 53±6 132±7
FiO2 0.3
HR (beats/min) MAP (mmHg) CBF (ml/min) PaO2 (mmHg)
0.3
110±5 102±3 53±6 132±7
0.5
0.8
92±2.74 107±8 44±6 170±9
88±2.42 98±5 42±8 271±12
0.3 105±6 96±4 44±7 130±5
Results are presented as mean±... ∗ P<0.05 as compared to values measures at FiO2 0.3. HR: heart rate, MAP: mean arterial pressure, CBF: coronary blood flow, PaO2: arterial P2.
271±12 mmHg (0.8 FiO2). No significant difference was observed in hemodynamic parameters (MAP, heart rate, CBF) when compared with control values. A significant increase in myocardial PtiO2 in relation with FiO2 and PaO2 was observed [Table 1(b)].
Interstitial myocardial adenosine and lactate concentration As expected, following a rapid increase in microdialysate adenosine concentration due to probe implantation (data not shown), adenosine and lact-
500
E. M. Siaghy et al.
Figure 5 Relation between myocardial PtiO2 and FiO2. Values are mean±SEM. ∗ P<0.05 as compared to values measures at FiO2 0.3. There was a significant correlation between PtiO2 and FiO2 (r=0.97).
ate concentrations remained stable during the entire equilibration period. Decreasing the FiO2 from 0.3 to 0.08 resulted in a significant increase in myocardial interstitial adenosine and lactate concentrations [Table 2(a)]. Increasing the FiO2 from 0.08 to 0.3 restored the PtiO2 and hemodynamic parameters to baseline values. Nevertheless, myocardial interstitial adenosine concentration remained significantly higher then baseline values. Increasing FiO2 from 0.3 to 0.8 had no effect on myocardial interstitial adenosine and lactate concentrations in comparison with baseline values [Table 2(b)]. Finally, both myocardial interstitial adenosine and lactate concentrations were similar when meas-
ured with the single and the combined probe for the entire study period [Table 2(a,b)].
Discussion The main results of this study demonstrate that a technique which combined cardiac microdialysis with polarographic oxygen electrode permits simultaneous measurements of myocardial interstitial adenosine concentration and myocardial PtiO2. This study demonstrates that hypoxemia results in decreased myocardial oxygen availability as assessed by decreased PtiO2, leading to an increase in myocardial interstitial adenosine concentration and the
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
501
Table 2 Myocardial interstitial adenosine and lactate concentrations measured by the combined and single microdialysis probes at hypoxemia (a) and hyperoxemia (b) challenges. (a)
Adenosine (lmol/l) Single probe Combined probe Lactate (mmol/l) Single probe Combined probe
FiO2 0.3 (baseline)
0.16
0.08
0.3
0.34±0.05 0.29±0.05
0.69±0.11∗ 0.53±0.13∗
0.98±0.06∗ 0.80±0.10∗
0.85±0.15∗ 0.71±0.2∗
0.4±0.05 0.31±0.06
0.53±0.11 0.39±0.13
0.97±0.06∗ 0.7±0.1∗
0.73±0.2∗ 0.53±0.22
(b)
Adenosine (lmol/l) Single probe Combined probe Lactate (mmol/l) Single probe Combined probe
FiO2 0.3 (baseline)
0.5
0.8
0.3
0.22±0.05 0.19±0.07
0.15±0.05 0.18±0.06
0.18±0.09 0.18±0.07
0.15±0.08 0.13±0.05
0.34±0.04 0.30±0.03
0.36±0.09 0.32±0.06
0.34±0.05 0.22±0.05
0.33±0.05 0.29±0.08
Results are presented as mean±... ∗ P<0.05 as compared to baseline values measured at FiO2 0.3. HR: heart rate, MAP: mean arterial pressure, CBF: coronary blood flow, PaO2: arterial P2.
onset of myocardial anaerobic metabolism assessed by the increased myocardial lactate concentration. Adenosine is a potent vasodilatator whose concentration has been shown to increase in cardiac tissue in response to hypoxia.2,7,8 Much of the previous work used various experimental models,13–15 and most of the evidence for a role of adenosine came from studies showing that hypoxia-induced hyperemia was attenuated by adenosine deaminase,16 adenosine receptor blockade,15 or inhibition of adenosine production.7 In the present study, acute systemic hypoxia resulted in a three-fold increase of the myocardial interstitial adenosine and lactate concentrations. Lactate concentration has been employed as an indicator of onset of anaerobic metabolism.17–19 Our results are similar to those previously observed7 and consistent with the concept that during acute systemic hypoxia, myocardial interstitial adenosine plays a key role in stimulating coronary vasodilatation.7,12 It should be noted, however, that the observed sustained elevation in myocardial interstitial adenosine concentration during the recovery period is in conflict with the hypothesis that a significant correlation exists between hemodynamic parameters and interstitial adenosine release.7,20,21 Similar findings have been reported in our study of brain-dead pigs,12 and taken together these data
are consistent with more recent reports concerning the mechanism of local control of CBF. Adenosine is considered to exhibit a “retaliatory” effect when energy delivery does not match energy demand.22 The lack of correlation between hemodynamic events or coronary flow and adenosine release is usually regarded as a consequence of the multiplicity of the factors involved in CBF regulation.12 It has been suggested that other compounds interact with adenosine to mediate the coronary vasodilatation associated with hypoxia. For example, it has been demonstrated that adenosine stimulates ATP-sensives K+ channels in skeletal muscle to open with resulting release of K+ which acts as a vasodilatator.23 More recently, it has been shown that adenosine release elicited by hypoxia was dependent on nitric oxide production.24 Therefore, it will be important in the future to measure several putative regulators when investigating the mechanisms associated with hypoxia-induced vasodilatation. Nevertheless, the delayed decrease to baseline values of myocardial interstitial concentrations demonstrated in the present study after the return of FiO2 to 0.3 could have other explanations. This delay could be due to the length of the microdialysis sampling period used in the present study, which was 20 min, thus leading to an underestimation of
502
E. M. Siaghy et al.
peak values and overestimation of low values during dynamic processes such as the decrease in adenosine interstitial concentrations. This is one of the limitations of the microdialysis technique which gives information on a period of time. Another explanation could be related to the interaction between diazepam used to anesthetize the animals and the nucleoside transporter which might have decreased adenosine re-uptake.25 Cardiac microdialysis is an established technique for the study of metabolites in a clearly defined compartment, the interstitial space,26 the site at which many of the metabolites are either produced or exert their effects.27,28 Thus, the microdialysis technique provides information not readily obtained from other conventional methods, such as arteriovenous differences and tissues biopsies.2,3 The microdialysis technique uses the dialysis principle, therefore in vivo recovery rate is directly proportional to the size of the dialysis membrane area which defines the dialysis fraction volume.29 In the present study, glucose relative recovery rate obtained with the combined probe was significantly lower as compared to the single microdialysis probe at lower perfusion rate (32±2 v 42±1% at a 2 ll/min perfusion rate). This was related to the proximity of the P2 electrode which interferes with free diffusion of water and solutes through the microdialysis membrane, thus reducing microdialysis probe recovery. However, despite hindrance to free diffusion, the combined probe still provides a satisfactory probe recovery rate. Myocardial tissue oxygen tension is assumed to be the central parameter in local control of blood flow.30,31 There are numerous reports on oxygen measurement in biological systems and oxygen sensor fabrication, and new techniques for oxygen monitoring are spread throughout the scientific and engineering literature. Extensive reviews and books on this subject have been published. In the present study we used flexible polarographic catheters for continuous measurement of myocardial oxygen tension. It has been demonstrated under experimental and clinical conditions that this technique allows the determination of P2 profiles within human tumours or P2 distribution within the brain and the skeletal muscle of patients.32–34 The resulting P2 values are considered to reflect local tissue oxygenation, giving the net effect of local blood flow, microcirculatory flow distribution, and oxygen consumption.35 Some authors have advocated the use of surface rather than needle electrodes on the grounds that the latter can cause tissue damage.35,36 However, surface electrodes are susceptible to significant errors arising from in-
terfacial gas exchanges and their use is clearly limited in tissues such as muscle where the structure of the surface layer is known to differ significantly from the tissue beneath.37–39 The polarographic needle electrode provides clear advantages because it is representative of the transmural distribution of myocardial oxygen pressure.37 Furthermore, previous studies demonstrate that no histological evidence of significant tissue damage can be attributed to electrode placement.37 When an oxygen sensor is implanted directly into the myocardium, oxygen diffuses through the coated membrane toward the electrode. Therefore, the observed current is a direct reflection of tissue oxygen pressure. With the combined probe, the oxygen dissolved in the buffer used to perfuse the microdialysis probe diffuses through the implanted microdialysis membrane and subsequently enriches the microenvironment within the myocardial interstitial space. Consequently, as should be expected, the myocardial PtiO2 values obtained with perfused combined probe were higher but not significantly different at 2 ll/min perfusion rate as compared to those obtained at zero perfusion rate of microdialysis probe. Furthermore, tissue oxygen enrichment is linearly correlated with the perfusion rate of the microdialysis probe. The in vivo response time of the oxygen probe was tested by performing a brief LAD occlusion, the response of the P2 electrode reached 90% of the steady state value in less than 1 min. Moreover, a significant correlation was obtained for myocardial tissue P2 with arterial P2 and inspired oxygen fraction. All these results indicate a rapid response of the P2 electrode probe despite the close proximity of the microdialysis probe. Therefore, this study has demonstrated the ability of a combined probe to measure simultaneously regional myocardial interstitial tissue oxygenation and metabolite concentration. We present evidence that the close proximity of the dual probes causes no significant perturbation on P2 and microdialysis measurements under the experimental conditions of the study. Validation of this technique has demonstrated that myocardial interstitial adenosine concentration is increased as a result of decreased PtiO2 induced by hypoxemia. The technique described here will permit the study of the role of myocardial adenosine in other pathophysiological circumstances such as myocardial ischemia.
References 1. E SW, K RM, B AN, R R, B RL. Measurements of coronary plasma and peripheral
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
2.
3.
4. 5.
6. 7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
infusate adenosine concentrations during exercise in conscious dog: Relationship to myocardial oxygen consumption and coronary blood flow. J Mol Cell Cardiol 1983; 15: 673–683. ML DA, S LI, L U. Systemic hypoxia elevates skeletal muscle interstitial adenosine levels in humans. Circulation 1998; 98: 1990–1992. V W DG, W J, S J, W RJ, L RD, M RM J et al. Cardiac microdialysis to estimate interstitial adenosine and coronary blood flow. Am J Physiol 1990; 258: H1642–H1649. B RM, K RM, E SW, R R. Adenosine in the local regulation of blood flow: A brief overview. Fed Proc 1983; 42: 3136–3142. E SW, S CW, K RM, G JM, R R, B RM. Adenosine and metabolic regulation of coronary blood flow in dogs with renal hypertension. Hypertension 1983b; 5: 943–950. O RA, B R. Metabolic control of coronary blood flow. Prog Cardiovasc Dis 1987; 29: 369–387. W T, S J, M RM, V W DG. Changes in interstitial adenosine during hypoxia: Relationship to oxygen supply:demand imbalance, and effects of adenosine deaminase. Cardiovasc Res 1994; 28: 1320–1325. D UK, J E, K H. Interstitial transudate concentration of adenosine and inosine in rat and guinea pig hearts. Am J Physiol 1988; 254: H1125–H1132. M PM, B B, J Y, P G, C JP, A K et al. In vivo monitoring of myocardial interstitial neuropeptide tyrosine release following brain death. Transplant Proc 1993; 25: 3167–3168. H Y, M T, K C, H Y, Y N, T M. In vivo calibration of microdialysis probe by use of endogenous glucose as an internal recovery marker: Measurement of skin distribution of tranilast in rats. J Pharma Pharmacol 1998; 50: 621–626. P LJ, K JK, B J. Microdialysis of the interstitial water space in human skin in vivo: Quantitative measurement of cutaneous glucose concentrations. J Invest Dermatol 1992; 99: 357– 360. H-D P, S EM, D Y, U-L D, R JP, B B et al. Increase in myocardial interstitial adenosine and net lactate production in brain-dead pigs: An in vivo microdialysis study. Transplantation 1998; 66: 1278– 1284. H JP, M GP, B RM. Myocardial adenosine formation during hypoxia: Effects of ecto5′-nucleotidase inhibition. J Mol Cell Cardiol 1992; 24: 295–303. H SC, F EO. Adrenergic blockade blunts adenosine concentration and coronary vasodilation during hypoxia. Circ Res 1992; 70: 1203–1216. S J, B H. Assessment of vasoactive metabolites released from the isolated guinea pig during heart hypoxia and beta-adrenergic stimulation. Basic Res Cardiol 1981; 76: 365–368. M GF, D HF, J CE. Adenosine deaminase attenuates canine coronary vasodilation during systemic hypoxia. Am J Physiol 1986; 250: H579–H583.
503
17. A H, T J, H M, N I, S I. Implication of myocardial lactate metabolism during coronary artery bypass grafting (see comments). Cardiovasc Surg 1997; 5: 210–215. 18. N M, F Y, K T, Y T, T K, U M et al. Changes in myocardial lactate metabolism during ramp exercise in patients with effort angina and microvascular angina. JCJ 1996; 60: 876–888. 19. R GM, K JC, A S, P WI, H P, C P et al. Failure to demonstrate myocardial ischaemia in patients with angina and normal coronary arteries. Evaluation by continuous coronary sinus pH monitoring and lactate metabolism (see comments). Eur Heart J 1996; 17: 1175–1180. 20. E SW, K RM, R R, B RM. Comments on “Pericardial superfusion to measure cardiac interstitial adenosine concentration” (letter). Am J Physiol 1983; 245: H1085–H1086. 21. M WL, B L, B A, F DH, R R, B RM. Canine myocardial adenosine and lactate production, oxygen consumption, and coronary blood flow during stellate ganglia stimulation. Circ Res 1979; 45: 708–718. 22. M P, P JD, S LL. Regulation of extracellular adenosine production by ectonucleotidases of adult rat ventricular myocytes. Am J Physiol 1992; 263: H40–H47. 23. M JM, T T, T L. A link between adenosine, ATP-sensitive K+ channels, potassium and muscle vasodilatation in the rat in systemic hypoxia. J Physiol 1993; 472: 1–9. 24. S MR, M JM. Studies on the roles of ATP, adenosine and nitric oxide in mediating muscle vasodilatation induced in the rat by acute systemic hypoxia. J Physiol 1996; 495: 553–560. 25. S TW. Actions of benzodiazepines and the benzodiazepine antagonist flumazenil may involve adenosine [letter]. J Neurol Sci 1999; 163: 199–201. 26. M PM, B B, J Y, S A, C JP, P G et al. Microdialysis in the estimation of interstitial myocardial neuropeptide Y release. Regul Pept 1993; 49: 81–90. 27. M T, L H, Z W, R Z, G O. Penetration of rifampicin into the brain tissue and cerebral extracellular space of rats. J Antimicrob Chemother 1993; 31: 731–737. 28. S L. Drug distribution studies with microdialysis: I. Tissue dependent difference in recovery between caffeine and theophylline. Life Sci 1991; 49: 1835–1842. 29. B H, D J, S A, D NH. Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 1984; 43: 1369–1374. 30. M W, S S, F H. Intramyocardial oxygen pressure and coronary blood flow during experimental coronary stenosis. Adv Exp Med Biol 1984; 169: 341–350. 31. S JA, D J. Theoretical analysis of coronary blood flow and tissue oxygen pressurecontrol. Adv Exp Med Biol 1993; 346: 189–195. 32. B P, W¨ S, K T, W
504
E. M. Siaghy et al.
K. Skeletal muscle partial pressure of oxygen in patients with sepsis. Crit Care Med 1994; 22: 640– 650. 33. B P, W P, V D G, N CA, A M, R TC et al. Preservation of regional myocardial function and myocardial oxygen tension during acute ischemia in pigs: Comparison of selective synchronized suction and retroinfusion of coronary veins to synchronized coronary venous retroperfusion. J Am Coll Cardiol 1994; 23: 459–469. 34. D EM, Z A, B R, W JD, F PP, Y HF. Correlations between brain tissue oxygen tension, carbon dioxide tension, pH, and cerebral blood flow—A better way of monitoring the severely injured brain? Surg Neurol 1998; 49: 650–654. 35. C PF, H H, C M, B H, H-
36. 37.
38. 39.
J, V B et al. Left ventricular surface tissue oxygen pressures determined by oxygen sensitive multiwire electrodes in pigs. Cardiovasc Res 1991; 25: 207–216. L N. Skeletal and cardiac muscle oxygenation. Adv Exp Med Biol 1985; 191: 37–43. G AR, E PJ, M S, O’H D, P KH, G CJ et al. Measurements of oxygenation and perfusion in skeletal muscle using multiple microelectrodes. J Muscle Res Cell Motil 1997; 18: 149–159. G G, L N, A AL, M C. Relationship of venous P2 to muscle O2 during hypoxemia. J Appl Physiol. 1989; 67: 1093–1099. O S, L N. Data acquisition and information processing in MDO oxygen electrode measurement of tissue oxygen pressure. Acta Anaesthesiol Scand 1980; 24: 161–165.
Consequences of Inspired Oxygen Fraction Manipulation on Myocardial Oxygen Pressure, Adenosine and Lactate Concentrations: A Combined Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs E. M. Siaghy1, Y. Devaux1∗, N. Sfaksi1∗, J. P. Carteaux1, D. UngureanuLongrois1,3, F. Zannad1, J. P. Villemot1, C. Burlet2 and P. M. Mertes1,2 1
Laboratoire de Chirurgie Expe´rimentale, UPRES 971068, Faculte´ de Me´decine de Nancy, Universite´ Henri Poincare´, Nancy I, France, 2Laboratoire de Biologie Cellulaire, and 3De´partement d’Anesthe´sieRe´animation Chirurgicale, CHU Brabois, Vandoeuvre-les-Nancy, France (Received 29 June 1999, accepted in revised form 31 December 1999) E. M. S, Y. D, N. S, J. P. C, D. U-L, F. Z, J. P. V, C. B P. M. M. Consequences of Inspired Oxygen Fraction Manipulation on Myocardial Oxygen Pressure, Adenosine and Lactate Concentrations: A Combined Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs. Journal of Molecular and Cellular Cardiology (2000) 32, 493–504. Adenosine is a potent vasodilator whose concentration has been shown to increase in cardiac tissue in response to hypoxia. However, the time-dependent relationship between the levels of myocardial interstitial adenosine and tissue oxygenation has not yet been completely established. Therefore, the purpose of this study was to investigate the complex relationship between tissue myocardial oxygen tension (PtiO2) and interstitial myocardial adenosine and lactate concentrations by developing a new technique which combines a cardiac microdialysis probe and a Clark-type P2 electrode. The combined and the single microdialysis probes were implanted in the left ventricular myocardium of anesthetized pigs. The consequences of the combined use of microdialysis and P2 probes on myocardial PtiO2 and microdialysis performances against glucose were evaluated. A moderate but significant reduction in the relative recovery against glucose of the combined probe was observed when compared to that of the single microdialysis probe (42±2 v 32±1%, mean±...; n= 5; P<0.05), at 2 ll/min microdialysis probe perfusion flow. Similarly, myocardial oxygen enrichment, measured by the P2 electrode, was negligible when microdialysis probe perfusion flow was 2 ll/min. Systemic hypoxia (FiO2= 0.08) resulted in a significant decrease in PtiO2 from 30±4 to 11±2 mmHg, limited increase in coronary blood flow (CBF), and a significant increase in myocardial adenosine and lactate concentrations from 0.34±0.05 to 0.98±0.06 lmol/l and from 0.45±0.05 to 0.97±0.06 mmol/l respectively (P<0.05). Increasing the FiO2 to 0.3 restored the PtiO2 and hemodynamic parameters to baseline values with no changes in interstitial adenosine and lactate concentrations. Nevertheless, myocardial interstitial adenosine remained significantly higher than baseline values. In conclusion, this study demonstrates the ability of a combined probe to measure simultaneously regional myocardial PtiO2 and metabolite concentration during hypoxia. The hypoxia-induced increase in myocardial adenosine persists after correction of hypoxia. The physiological significance of this observation requires further studies. 2000 Academic Press
K W: Adenosine; Hypoxia; Myocardial oxygenation; PtiO2; Cardiac microdialysis.
∗ Research Fellow. Please address all correspondence to: Dan Ungureanu-Longrois, De´partement d’Anesthe´sie-Re´animation Chirurgicale, CHU de Brabois, Alle´e du Morvan, 54511 Vandoeuvre-les-Nancy, France. Tel: 33 (0) 3-83-15-41-66; Fax: 33 (0) 3-83-15-36-88; E-mail: d.longrois@ chu.nancy.fr
0022–2828/00/030493+12 $35.00/0
2000 Academic Press
494
E. M. Siaghy et al.
Introduction Previous studies have suggested that adenosine is one of the metabolic mediators that account for the changes in coronary blood flow during conditions of hypoxemia-triggered decreased oxygen supply and increased oxygen demand.1–3 Most of the evidence in support of the adenosine hypothesis for regulation of coronary blood flow (CBF) during hypoxemia comes from studies in which changes in cardiac tissue or coronary venous adenosine levels are correlated with changes in myocardial oxygen consumption and/or CBF.1,4–6 Nevertheless, controversies concerning the regulatory role of adenosine on CBF during conditions of hypoxemia still persist. For example, both an increase7 or no change8 in myocardial interstitial adenosine levels during hypoxia have been observed. One of the possible explanations for the difficulties in clearly establishing the regulatory role of adenosine during hypoxemia is that the release of adenosine mainly depends on myocardial oxygen availability which is more accurately reflected by regional tissue myocardial oxygen tension (PtiO2) than by CBF. Moreover, it is well recognized that cardiac tissue or coronary venous adenosine levels may not accurately reflect the myocardial interstitial adenosine concentration to which the coronary resistance vessels are exposed.2,3,7 Therefore, a technological approach that permits simultaneous measurements of both regional PtiO2 and myocardial interstitial adenosine concentration is still required. The purpose of the present study was to investigate the complex relationship between PtiO2 and interstitial myocardial adenosine and lactate concentrations by developing a new technique which combines a cardiac microdialysis probe and a Clark-type P2 electrode.
Materials and Methods Combined microdialysis-PO2 probe A flexible polarographic microcatheter Clark-type P2 probe (diameter 500 lm, length 20 mm, computer-supported LICOX system; GMS, Kiel-Mielkendorf, Germany) was combined with a concentric flexible microdialysis probe (membrane outer diameter 340 lm, length 20 mm, molecular weight cut-off (MWCO): 50 kD; Medicorp, Vandoeuvre, France). The sensitive areas of both the microdialysis probe and the P2 electrode were linked
using a metallic collar allowing their implantation within the same site through a fine custom-made guiding needle specially designed for this purpose. For correction of the myocardial PtiO2 measurements, the temperature within the myocardium was monitored and myocardial PtiO2 values were adjusted to myocardial temperature by means of a computer software (LICOX system; GMS, KielMielkendorf, Germany). Microdialysis probes were perfused with a Ringer solution (Fandre, Ludres, France) using a microinjection pump (Harvard Pump, Cambridge, MA, USA). An expanded view of the assembled combined P2/microdialysis probe is illustrated in Figure 1. Probes recovery performances, as determined in vitro, were 76±2% for adenosine and 56±2% for lactate.
Animal preparation Domestic 4–6-month-old pigs weighing 30–40 kg were chosen as experimental animals. All animals were housed and treated in accordance with accepted practices for humane care of laboratory animals. Animals were fasted overnight and premedicated with 10 mg/kg ketamine (Park-Davis, Courbevoie, France), and 0.2 mg/kg diazepam (Roche, Neuilly-Sur-Seine, France), given intramuscularly. Anesthesia was induced with 5 mg/ kg of thiopentone sodium (Specia, Rhone-PoulencRhorer, Paris, France) and maintained with a continuous infusion at a rate of 100–150 mg/h. This anesthetic regimen was continued during the entire study period in all animals. Neuromuscular blockade was obtained with pancuronium bromide (0.6 mg/kg, i.v.), with additional doses of 0.3 mg/ kg administered at 30 min intervals. The trachea was intubated with a cuffed endotracheal tube, and the lungs were mechanically ventilated with a mixture of oxygen and N2O at a minute volume of 150 ml/kg. Adjustments were made to maintain the PaCO2 in the range of 35–45 mmHg. A median sternotomy was performed and an aortic catheter was inserted via the right common carotid artery to measure mean arterial pressure (MAP). The electrocardiogram was monitored continuously. An electromagnetic flowmeter (Skalar, Stockholm, Sweden) was placed on the left anterior descending coronary artery (LAD) to measure coronary blood flow. A combined microdialysis-P2 probe and a single microdialysis probe as control were successively implanted into the free wall of the left ventricle as close as possible to the LAD as previously described.9 Briefly, the probes were inserted into a fine guiding needle which was then
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
495
Figure 1 Diagram showing the assembly and implantation of the combined microdialysis and Clark-type oxygen P2 electrode within the pig myocardium. The insert represents the active area of both microdialysis probe and Clark-type oxygen P2 electrode.
passed through the myocardium. As soon as the needle traversed the myocardium, it was removed and the probe was gently pulled back until the microdialysis membrane and/or active area of the P2 electrode were completely embedded within the myocardium, and then sutured into place. The microdialysis effluent was collected in plastic Eppendorf microtubes kept on ice during the entire collection period.
Experimental design At the end of the surgical preparation period the animals were allowed to stabilize for 2 h at FiO2 of
0.3. Baseline recordings of hemodynamic, myocardial PtiO2 and blood gas measurements were performed and microdialysate collection was started with 20 min sampling periods.
Consequences of microdialysis-PO2 probes coupling on microdialysis and PO2 electrode performances The consequences of spatial hindrance due to the presence of the P2 probe on microdialysis performance was assessed by comparing the in vivo relative recovery (Rrinvivo) of glucose of the single microdialysis probe and the combined probe. The in vivo relative recovery rate as a function of perfusion flow was determined. The probes were perfused at increasing flow from 0.5 to 12 ll/min. An
496
E. M. Siaghy et al.
equilibration period of at least 30 min was observed after each change in perfusion flow followed by three consecutive 20 min collection periods. Simultaneous plasma glucose concentration measurements were performed at the end of each sampling period. In vivo relative recovery rate was calculated according to the following formula: [Glucose]microdialysate , [Glucose]plasma
Rrinvivo=
based on the assumption that the concentration of glucose in the extracellular fluid of myocardial tissue is the same as that in plasma under steady state conditions.10,11 The consequences of oxygen enrichment of myocardial interstitial space due to the microdialysis probe perfusion with the Ringer solution containing 115 mmHg of O2 on P2 electrode measurements was assessed by measuring PtiO2 at increasing microdialysis probe perfusion flows (from 0 to 12 ll/ min). Repeated LAD occlusions were performed at each microdialysis probe perfusion flow to confirm the ability of the combined probe to rapidly detect an acute drop in PtiO2.
Consequences of changes in FiO2 on myocardial PtiO2, hemodynamic parameters and myocardial interstitial adenosine and lactate concentrations For these experiments, microdialysis probes were perfused at 2 ll/min flow rate. A 10 min equilibration period was observed following every FiO2 modification before microdialysis collection was started. During the hypoxemia period, FiO2 was lowered from 0.3 to 0.08 by replacing O2 with N2O. Following a 30 min hypoxemia period, the FiO2 was switched to 0.3 again. In order to achieve complete hemodynamic and metabolic recovery, a 1 h period with FiO2 at 0.3 was observed between the hypoxemia and hyperoxemia protocols. During the hyperoxemia period, FiO2 was raised from 0.3 to 0.8 and readjusted after 30 min to 0.3 again. Simultaneous measurements of myocardial PtiO2, hemodynamic recordings, microdialysis samples of adenosine and lactate and blood gas measurements were performed at each period.
Analytic procedures Arterial blood gas and pH determinations were performed on an ABL 4 automated blood gas analyzer (Radiometer A/S, Copenhagen, Denmark). In
order to determine the relative recovery of probes against glucose, serial arterial blood samples were drawn into chilled tubes at the end of each microdialysis sample collection period and centrifuged for 10 min at 2500 rpm. After centrifugation, the plasma was stored at −80°C until assay. Glucose in microdialysis and plasma samples and lactate in the myocardial microdialysis samples were analyzed using enzymatic kinetic methods (Glucose: Biotrol, Paris, France; Lactate: Kit 256773, Boehringer Mannheim GmbH Diagnostica, Mannheim, Germany). Adenosine in myocardial microdialysis samples was analyzed by high-performance liquid chromatography as previously described.12 Briefly, separation of compounds was achieved using a Symmetry C18-column (Symmetry, Waters, St Quentin-Yvelines, France) and a 1% (pH 5.3) to 25% (pH 5.58) methanol in 1 m KH2PO4 gradient. All peaks were detected at 254 nm and unknown samples were identified and quantified by comparing retention time and peak areas to known standards.
Statistical analysis Results are expressed as the mean±... Linear regression was performed to study the correlation between PaO2 or PtiO2 values and myocardial interstitial adenosine concentrations. Mann–Whitney U-test was performed to compare myocardial interstitial concentrations of adenosine and lactate during hypoxemia or hyperoxemia to normoxemia period. One-way analysis of variance with repeated measures was performed to compare the changes in PtiO2 as a function of perfusion flow, followed by Fisher test when statistical significance was detected. The criterion of significance was P<0.05.
Results Consequences of the combined use of microdialysis and PO2 probes on glucose relative recovery measurements and myocardial tissue PO2
Consequences on microdialysis probe glucose recovery Typical in vivo relative recovery results obtained with a 50 kD MWCO microdialysis membrane against glucose are displayed in Figure 2. As expected, the relative recovery rate was found to decrease in an exponential manner with increasing perfusion flow (n=5). A moderate but significant reduction in the recovery rate against glucose of
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
497
Figure 2 Relation between relative recovery of the combined and single microdialysis probes. Values are mean±... ∗ P<0.05 v single microdialysis probe for each microdialysis probe perfusion flow. Ρ, Single microdialysis probe; Ε, combined probe.
the combined probe was observed when compared to that of the single microdialysis probe ranging from 42±2% v 32±1% for the single and combined probe respectively (P=0.0017) at a 2 ll/min perfusion flow.
Consequences on PO2 electrode measurements Basal myocardial PtiO2 level measured in the absence of microdialysis probe perfusion (perfusion flow 0 ll/min) was 23±2 mmHg (n=5). As could be expected, perfusion of the microdialysis probe with Ringer solution containing a partial P2 pressure of 115 mmHg resulted in a progressive increase in myocardial PtiO2 values in relation to microdialysis perfusion flow. These were of 23±2 mmHg
at 0 ll/min and 64±7 mmHg at 12 ll/min. However, oxygen enrichment was negligible when microdialysis probe perfusion flow was 2 ll/min (Fig. 3). Regression analysis showed a significant linear correlation of PtiO2 with microdialysis perfusion flow (r=0.97). Moreover, when coronary occlusion was performed, myocardial PtiO2 values in the LAD region rapidly decreased from 23±2 mmHg to 2±4 mmHg at 0 ll/min perfusion flow, and from 30±2 to 6±2 mmHg at 2 ll/min perfusion flow (Fig. 4). Based on these results, a perfusion flow of 2 ll/min was used for all the subsequent in vivo experiments. Using this perfusion flow, we reached a compromise between the ability to detect a rapid change in interstitial metabolite concentration and a minimum oxygen enrichment
498
E. M. Siaghy et al.
Figure 3 Relation between myocardial tissue P2 (PtiO2) and perfusion speed of microdialysis probe. Values are mean±... ∗ P<0.05 v single microdialysis probe for each microdialysis probe perfusion flow.
of myocardial interstitial space due to the microdialysis probe perfusion.
Consequences of FiO2 changes on myocardial PtiO2 and myocardial interstitial adenosine and lactate concentrations
Myocardial PtiO2, blood gas and hemodynamic parameters Decreasing the FiO2 from 0.3 to 0.08 reduced arterial PaO2 from a baseline of 128±7 to 63±5 mmHg (0.16 FiO2) and 36±5 mmHg (0.08 FiO2), respectively. Arterial hypoxemia resulted in a significant increase in heart rate, a slight but not significant decrease of mean arterial pressure, and
an increase in CBF through the LAD, as an attempt to maintain oxygen delivery to the left ventricular myocardium close to its baseline value [Table 1(a)]. Under basal conditions (0.3 FiO2 and 2 ll/min perfusion flow of microdialysis probe), PtiO2 values were 30±4 mmHg within the left ventricular area during the control recording. No changes were noted during the subsequent control periods until the end of the experiments. During hypoxia, a significant decrease of myocardial PtiO2 in relation with FiO2 and arterial PaO2 was observed (Fig. 5). A significant linear correlation between myocardial PtiO2 and both arterial PaO2 (r=0.98) and FiO2 (r=0.97) was observed. During the hyperoxemia period, increasing FiO2 from 0.3 to 0.8 raised arterial PaO2 from a baseline of 128±7 mmHg to 170±9 mmHg (0.5 FiO2) and
499
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
Figure 4 Diagram showing the evolution of myocardial PtiO2 measured by combined probe at different microdialysis probe perfusion flow and the effect of LAD occlusion on myocardial PtiO2.
Table 1 Hemodynamic parameters at hypoxemia (a) and hyperoxemia (b) challenges. (a)
FiO2 0.3
HR (beats/min) MAP (mmHg) CBF (ml/min) PaO2 (mmHg)
94±2 98±2 49±8 128±7
0.16
0.08
116±6 95±7 62±7∗ 63±5
162±6∗ 75±3 85±11∗ 36±5
(b)
110±5 102±3 53±6 132±7
FiO2 0.3
HR (beats/min) MAP (mmHg) CBF (ml/min) PaO2 (mmHg)
0.3
110±5 102±3 53±6 132±7
0.5
0.8
92±2.74 107±8 44±6 170±9
88±2.42 98±5 42±8 271±12
0.3 105±6 96±4 44±7 130±5
Results are presented as mean±... ∗ P<0.05 as compared to values measures at FiO2 0.3. HR: heart rate, MAP: mean arterial pressure, CBF: coronary blood flow, PaO2: arterial P2.
271±12 mmHg (0.8 FiO2). No significant difference was observed in hemodynamic parameters (MAP, heart rate, CBF) when compared with control values. A significant increase in myocardial PtiO2 in relation with FiO2 and PaO2 was observed [Table 1(b)].
Interstitial myocardial adenosine and lactate concentration As expected, following a rapid increase in microdialysate adenosine concentration due to probe implantation (data not shown), adenosine and lact-
500
E. M. Siaghy et al.
Figure 5 Relation between myocardial PtiO2 and FiO2. Values are mean±SEM. ∗ P<0.05 as compared to values measures at FiO2 0.3. There was a significant correlation between PtiO2 and FiO2 (r=0.97).
ate concentrations remained stable during the entire equilibration period. Decreasing the FiO2 from 0.3 to 0.08 resulted in a significant increase in myocardial interstitial adenosine and lactate concentrations [Table 2(a)]. Increasing the FiO2 from 0.08 to 0.3 restored the PtiO2 and hemodynamic parameters to baseline values. Nevertheless, myocardial interstitial adenosine concentration remained significantly higher then baseline values. Increasing FiO2 from 0.3 to 0.8 had no effect on myocardial interstitial adenosine and lactate concentrations in comparison with baseline values [Table 2(b)]. Finally, both myocardial interstitial adenosine and lactate concentrations were similar when meas-
ured with the single and the combined probe for the entire study period [Table 2(a,b)].
Discussion The main results of this study demonstrate that a technique which combined cardiac microdialysis with polarographic oxygen electrode permits simultaneous measurements of myocardial interstitial adenosine concentration and myocardial PtiO2. This study demonstrates that hypoxemia results in decreased myocardial oxygen availability as assessed by decreased PtiO2, leading to an increase in myocardial interstitial adenosine concentration and the
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
501
Table 2 Myocardial interstitial adenosine and lactate concentrations measured by the combined and single microdialysis probes at hypoxemia (a) and hyperoxemia (b) challenges. (a)
Adenosine (lmol/l) Single probe Combined probe Lactate (mmol/l) Single probe Combined probe
FiO2 0.3 (baseline)
0.16
0.08
0.3
0.34±0.05 0.29±0.05
0.69±0.11∗ 0.53±0.13∗
0.98±0.06∗ 0.80±0.10∗
0.85±0.15∗ 0.71±0.2∗
0.4±0.05 0.31±0.06
0.53±0.11 0.39±0.13
0.97±0.06∗ 0.7±0.1∗
0.73±0.2∗ 0.53±0.22
(b)
Adenosine (lmol/l) Single probe Combined probe Lactate (mmol/l) Single probe Combined probe
FiO2 0.3 (baseline)
0.5
0.8
0.3
0.22±0.05 0.19±0.07
0.15±0.05 0.18±0.06
0.18±0.09 0.18±0.07
0.15±0.08 0.13±0.05
0.34±0.04 0.30±0.03
0.36±0.09 0.32±0.06
0.34±0.05 0.22±0.05
0.33±0.05 0.29±0.08
Results are presented as mean±... ∗ P<0.05 as compared to baseline values measured at FiO2 0.3. HR: heart rate, MAP: mean arterial pressure, CBF: coronary blood flow, PaO2: arterial P2.
onset of myocardial anaerobic metabolism assessed by the increased myocardial lactate concentration. Adenosine is a potent vasodilatator whose concentration has been shown to increase in cardiac tissue in response to hypoxia.2,7,8 Much of the previous work used various experimental models,13–15 and most of the evidence for a role of adenosine came from studies showing that hypoxia-induced hyperemia was attenuated by adenosine deaminase,16 adenosine receptor blockade,15 or inhibition of adenosine production.7 In the present study, acute systemic hypoxia resulted in a three-fold increase of the myocardial interstitial adenosine and lactate concentrations. Lactate concentration has been employed as an indicator of onset of anaerobic metabolism.17–19 Our results are similar to those previously observed7 and consistent with the concept that during acute systemic hypoxia, myocardial interstitial adenosine plays a key role in stimulating coronary vasodilatation.7,12 It should be noted, however, that the observed sustained elevation in myocardial interstitial adenosine concentration during the recovery period is in conflict with the hypothesis that a significant correlation exists between hemodynamic parameters and interstitial adenosine release.7,20,21 Similar findings have been reported in our study of brain-dead pigs,12 and taken together these data
are consistent with more recent reports concerning the mechanism of local control of CBF. Adenosine is considered to exhibit a “retaliatory” effect when energy delivery does not match energy demand.22 The lack of correlation between hemodynamic events or coronary flow and adenosine release is usually regarded as a consequence of the multiplicity of the factors involved in CBF regulation.12 It has been suggested that other compounds interact with adenosine to mediate the coronary vasodilatation associated with hypoxia. For example, it has been demonstrated that adenosine stimulates ATP-sensives K+ channels in skeletal muscle to open with resulting release of K+ which acts as a vasodilatator.23 More recently, it has been shown that adenosine release elicited by hypoxia was dependent on nitric oxide production.24 Therefore, it will be important in the future to measure several putative regulators when investigating the mechanisms associated with hypoxia-induced vasodilatation. Nevertheless, the delayed decrease to baseline values of myocardial interstitial concentrations demonstrated in the present study after the return of FiO2 to 0.3 could have other explanations. This delay could be due to the length of the microdialysis sampling period used in the present study, which was 20 min, thus leading to an underestimation of
502
E. M. Siaghy et al.
peak values and overestimation of low values during dynamic processes such as the decrease in adenosine interstitial concentrations. This is one of the limitations of the microdialysis technique which gives information on a period of time. Another explanation could be related to the interaction between diazepam used to anesthetize the animals and the nucleoside transporter which might have decreased adenosine re-uptake.25 Cardiac microdialysis is an established technique for the study of metabolites in a clearly defined compartment, the interstitial space,26 the site at which many of the metabolites are either produced or exert their effects.27,28 Thus, the microdialysis technique provides information not readily obtained from other conventional methods, such as arteriovenous differences and tissues biopsies.2,3 The microdialysis technique uses the dialysis principle, therefore in vivo recovery rate is directly proportional to the size of the dialysis membrane area which defines the dialysis fraction volume.29 In the present study, glucose relative recovery rate obtained with the combined probe was significantly lower as compared to the single microdialysis probe at lower perfusion rate (32±2 v 42±1% at a 2 ll/min perfusion rate). This was related to the proximity of the P2 electrode which interferes with free diffusion of water and solutes through the microdialysis membrane, thus reducing microdialysis probe recovery. However, despite hindrance to free diffusion, the combined probe still provides a satisfactory probe recovery rate. Myocardial tissue oxygen tension is assumed to be the central parameter in local control of blood flow.30,31 There are numerous reports on oxygen measurement in biological systems and oxygen sensor fabrication, and new techniques for oxygen monitoring are spread throughout the scientific and engineering literature. Extensive reviews and books on this subject have been published. In the present study we used flexible polarographic catheters for continuous measurement of myocardial oxygen tension. It has been demonstrated under experimental and clinical conditions that this technique allows the determination of P2 profiles within human tumours or P2 distribution within the brain and the skeletal muscle of patients.32–34 The resulting P2 values are considered to reflect local tissue oxygenation, giving the net effect of local blood flow, microcirculatory flow distribution, and oxygen consumption.35 Some authors have advocated the use of surface rather than needle electrodes on the grounds that the latter can cause tissue damage.35,36 However, surface electrodes are susceptible to significant errors arising from in-
terfacial gas exchanges and their use is clearly limited in tissues such as muscle where the structure of the surface layer is known to differ significantly from the tissue beneath.37–39 The polarographic needle electrode provides clear advantages because it is representative of the transmural distribution of myocardial oxygen pressure.37 Furthermore, previous studies demonstrate that no histological evidence of significant tissue damage can be attributed to electrode placement.37 When an oxygen sensor is implanted directly into the myocardium, oxygen diffuses through the coated membrane toward the electrode. Therefore, the observed current is a direct reflection of tissue oxygen pressure. With the combined probe, the oxygen dissolved in the buffer used to perfuse the microdialysis probe diffuses through the implanted microdialysis membrane and subsequently enriches the microenvironment within the myocardial interstitial space. Consequently, as should be expected, the myocardial PtiO2 values obtained with perfused combined probe were higher but not significantly different at 2 ll/min perfusion rate as compared to those obtained at zero perfusion rate of microdialysis probe. Furthermore, tissue oxygen enrichment is linearly correlated with the perfusion rate of the microdialysis probe. The in vivo response time of the oxygen probe was tested by performing a brief LAD occlusion, the response of the P2 electrode reached 90% of the steady state value in less than 1 min. Moreover, a significant correlation was obtained for myocardial tissue P2 with arterial P2 and inspired oxygen fraction. All these results indicate a rapid response of the P2 electrode probe despite the close proximity of the microdialysis probe. Therefore, this study has demonstrated the ability of a combined probe to measure simultaneously regional myocardial interstitial tissue oxygenation and metabolite concentration. We present evidence that the close proximity of the dual probes causes no significant perturbation on P2 and microdialysis measurements under the experimental conditions of the study. Validation of this technique has demonstrated that myocardial interstitial adenosine concentration is increased as a result of decreased PtiO2 induced by hypoxemia. The technique described here will permit the study of the role of myocardial adenosine in other pathophysiological circumstances such as myocardial ischemia.
References 1. E SW, K RM, B AN, R R, B RL. Measurements of coronary plasma and peripheral
Myocardial Microdialysis and Sensitive Oxygen Electrode Study in Pigs
2.
3.
4. 5.
6. 7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
infusate adenosine concentrations during exercise in conscious dog: Relationship to myocardial oxygen consumption and coronary blood flow. J Mol Cell Cardiol 1983; 15: 673–683. ML DA, S LI, L U. Systemic hypoxia elevates skeletal muscle interstitial adenosine levels in humans. Circulation 1998; 98: 1990–1992. V W DG, W J, S J, W RJ, L RD, M RM J et al. Cardiac microdialysis to estimate interstitial adenosine and coronary blood flow. Am J Physiol 1990; 258: H1642–H1649. B RM, K RM, E SW, R R. Adenosine in the local regulation of blood flow: A brief overview. Fed Proc 1983; 42: 3136–3142. E SW, S CW, K RM, G JM, R R, B RM. Adenosine and metabolic regulation of coronary blood flow in dogs with renal hypertension. Hypertension 1983b; 5: 943–950. O RA, B R. Metabolic control of coronary blood flow. Prog Cardiovasc Dis 1987; 29: 369–387. W T, S J, M RM, V W DG. Changes in interstitial adenosine during hypoxia: Relationship to oxygen supply:demand imbalance, and effects of adenosine deaminase. Cardiovasc Res 1994; 28: 1320–1325. D UK, J E, K H. Interstitial transudate concentration of adenosine and inosine in rat and guinea pig hearts. Am J Physiol 1988; 254: H1125–H1132. M PM, B B, J Y, P G, C JP, A K et al. In vivo monitoring of myocardial interstitial neuropeptide tyrosine release following brain death. Transplant Proc 1993; 25: 3167–3168. H Y, M T, K C, H Y, Y N, T M. In vivo calibration of microdialysis probe by use of endogenous glucose as an internal recovery marker: Measurement of skin distribution of tranilast in rats. J Pharma Pharmacol 1998; 50: 621–626. P LJ, K JK, B J. Microdialysis of the interstitial water space in human skin in vivo: Quantitative measurement of cutaneous glucose concentrations. J Invest Dermatol 1992; 99: 357– 360. H-D P, S EM, D Y, U-L D, R JP, B B et al. Increase in myocardial interstitial adenosine and net lactate production in brain-dead pigs: An in vivo microdialysis study. Transplantation 1998; 66: 1278– 1284. H JP, M GP, B RM. Myocardial adenosine formation during hypoxia: Effects of ecto5′-nucleotidase inhibition. J Mol Cell Cardiol 1992; 24: 295–303. H SC, F EO. Adrenergic blockade blunts adenosine concentration and coronary vasodilation during hypoxia. Circ Res 1992; 70: 1203–1216. S J, B H. Assessment of vasoactive metabolites released from the isolated guinea pig during heart hypoxia and beta-adrenergic stimulation. Basic Res Cardiol 1981; 76: 365–368. M GF, D HF, J CE. Adenosine deaminase attenuates canine coronary vasodilation during systemic hypoxia. Am J Physiol 1986; 250: H579–H583.
503
17. A H, T J, H M, N I, S I. Implication of myocardial lactate metabolism during coronary artery bypass grafting (see comments). Cardiovasc Surg 1997; 5: 210–215. 18. N M, F Y, K T, Y T, T K, U M et al. Changes in myocardial lactate metabolism during ramp exercise in patients with effort angina and microvascular angina. JCJ 1996; 60: 876–888. 19. R GM, K JC, A S, P WI, H P, C P et al. Failure to demonstrate myocardial ischaemia in patients with angina and normal coronary arteries. Evaluation by continuous coronary sinus pH monitoring and lactate metabolism (see comments). Eur Heart J 1996; 17: 1175–1180. 20. E SW, K RM, R R, B RM. Comments on “Pericardial superfusion to measure cardiac interstitial adenosine concentration” (letter). Am J Physiol 1983; 245: H1085–H1086. 21. M WL, B L, B A, F DH, R R, B RM. Canine myocardial adenosine and lactate production, oxygen consumption, and coronary blood flow during stellate ganglia stimulation. Circ Res 1979; 45: 708–718. 22. M P, P JD, S LL. Regulation of extracellular adenosine production by ectonucleotidases of adult rat ventricular myocytes. Am J Physiol 1992; 263: H40–H47. 23. M JM, T T, T L. A link between adenosine, ATP-sensitive K+ channels, potassium and muscle vasodilatation in the rat in systemic hypoxia. J Physiol 1993; 472: 1–9. 24. S MR, M JM. Studies on the roles of ATP, adenosine and nitric oxide in mediating muscle vasodilatation induced in the rat by acute systemic hypoxia. J Physiol 1996; 495: 553–560. 25. S TW. Actions of benzodiazepines and the benzodiazepine antagonist flumazenil may involve adenosine [letter]. J Neurol Sci 1999; 163: 199–201. 26. M PM, B B, J Y, S A, C JP, P G et al. Microdialysis in the estimation of interstitial myocardial neuropeptide Y release. Regul Pept 1993; 49: 81–90. 27. M T, L H, Z W, R Z, G O. Penetration of rifampicin into the brain tissue and cerebral extracellular space of rats. J Antimicrob Chemother 1993; 31: 731–737. 28. S L. Drug distribution studies with microdialysis: I. Tissue dependent difference in recovery between caffeine and theophylline. Life Sci 1991; 49: 1835–1842. 29. B H, D J, S A, D NH. Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J Neurochem 1984; 43: 1369–1374. 30. M W, S S, F H. Intramyocardial oxygen pressure and coronary blood flow during experimental coronary stenosis. Adv Exp Med Biol 1984; 169: 341–350. 31. S JA, D J. Theoretical analysis of coronary blood flow and tissue oxygen pressurecontrol. Adv Exp Med Biol 1993; 346: 189–195. 32. B P, W¨ S, K T, W
504
E. M. Siaghy et al.
K. Skeletal muscle partial pressure of oxygen in patients with sepsis. Crit Care Med 1994; 22: 640– 650. 33. B P, W P, V D G, N CA, A M, R TC et al. Preservation of regional myocardial function and myocardial oxygen tension during acute ischemia in pigs: Comparison of selective synchronized suction and retroinfusion of coronary veins to synchronized coronary venous retroperfusion. J Am Coll Cardiol 1994; 23: 459–469. 34. D EM, Z A, B R, W JD, F PP, Y HF. Correlations between brain tissue oxygen tension, carbon dioxide tension, pH, and cerebral blood flow—A better way of monitoring the severely injured brain? Surg Neurol 1998; 49: 650–654. 35. C PF, H H, C M, B H, H-
36. 37.
38. 39.
J, V B et al. Left ventricular surface tissue oxygen pressures determined by oxygen sensitive multiwire electrodes in pigs. Cardiovasc Res 1991; 25: 207–216. L N. Skeletal and cardiac muscle oxygenation. Adv Exp Med Biol 1985; 191: 37–43. G AR, E PJ, M S, O’H D, P KH, G CJ et al. Measurements of oxygenation and perfusion in skeletal muscle using multiple microelectrodes. J Muscle Res Cell Motil 1997; 18: 149–159. G G, L N, A AL, M C. Relationship of venous P2 to muscle O2 during hypoxemia. J Appl Physiol. 1989; 67: 1093–1099. O S, L N. Data acquisition and information processing in MDO oxygen electrode measurement of tissue oxygen pressure. Acta Anaesthesiol Scand 1980; 24: 161–165.