0041-1337/03/7506-744/0

TRANSPLANTATION

Vol. 75, 744–749, No. 6, March 27, 2003 Printed in U.S.A.

Copyright © 2003 by Lippincott Williams & Wilkins, Inc.

Transplantation姞 ARTICLES MULTIPARAMETRIC MONITORING OF ISCHEMIA-REPERFUSION IN RAT KIDNEY: EFFECT OF ISCHEMIC PRECONDITIONING ´ ´ A. SOLA,1 L. PALACIOS,2 J. LOPEZ -MART´I,1 A. IVORRA,3 N. NOGUERA,3 R. GOMEZ ,3 R. VILLA,3 3 ´ J. AGUILO, AND G. HOTTER1 Background. Microelectrode technology is a promising tool for monitoring kidney ischemia and the changes induced by its therapeutic management. Ischemic preconditioning, that is, brief ischemic periods before sustained ischemia, has been shown to protect several organs, including the kidney, from ischemiareperfusion injury. We tested whether the effect of preconditioning could be appraised by real-time measurement of parameters representative of tissue hypoxia. Methods. In a sample of pentobarbital-anesthetized and mechanically ventilated rats, we studied the effect of renal ischemic preconditioning (10-min ischemia and 10-min reflow interval) on subsequent ischemiareperfusion (45 min and 60 min). Renal tissue electrical impedance, extracellular pH, and potassium concentration [Kⴙ] were measured continuously by implanted microelectrodes. Results. Ischemia induced an early, rapid rise in extracellular potassium and impedance module, followed by a phase of slower increase, whereas pH decreased rapidly, reaching a plateau. Preconditioning treatment did not cause significant changes in interstitial pH and [Kⴙ] but increased ischemic tissue impedance. During reperfusion, the three variables recovered progressively; however, after a decline, electrical impedance showed a clear postischemic increase. This rise was suppressed by preconditioning. Conclusions. Real-time measurement of any of the three parameters showed capability for early detection of ischemia. In contrast with findings in myocardial tissue, preconditioning in the kidney did not increase potassium cell loss during ischemia or improve ischemic acidosis or tissue impedance. Electrical im-

pedance increased for a second time during reperfusion, indicating the presence of a postischemic cellular edema; concealing this episode was the most noticeable effect of the preconditioning treatment.

Microelectrode monitoring of renal ischemia may provide a useful tool for both the rapid detection of ischemia and the changes induced by its therapeutic management. To test this capability, we used tissue microsensors to study the influence of ischemic preconditioning on kidney ischemia-reperfusion (I/R). Ischemic preconditioning (brief ischemic periods before protracted ischemia) has been shown to protect against hypoxic injury in a variety of organs (1). Because preconditioning is a nonpharmacologic treatment, it is expected to elicit moderate physiologic responses adequate to assess the microelectrode technique used. For monitoring, we selected three physical chemical tissue parameters representative of well-documented metabolic, ionic, and dielectric events occurring during ischemia: extracellular pH, potassium concentration, and electrical impedance. Tissue pH measurement is regularly used to measure tissue hypoxia (2). The use of surface pH and PCO2 electrodes has also been proposed to detect ischemia in the kidney (3, 4). In regard to extracellular [K⫹], early increases have been detected after the onset of renal ischemia as the result of loss of intracellular potassium ions (5). In addition, attenuation of intracellular potassium loss by means of K⫹ channel blockage has been shown to reduce hypoxic renal injury (5, 6). Finally, an increase in tissue bioimpedance has been considered a reliable indicator of anoxic cellular edema by the detection of the narrowing of extracellular space and closure of gap junctions (7). Ischemic cell swelling results from inhibition of energy metabolism (8), and it has been considered to be one of the major causes of renal ischemic damage (9). I/R injury is a well-known feature of renal dysfunction (10). Functional, metabolic, and morphologic methods have shown that ischemic preconditioning pretreatment protects the kidney from ischemic injury (11). Nevertheless, the effectiveness of this preconditioning has proved to be highly dependent on protocol approach (11–13), so it seems that tissue parameter monitoring may also help to assess the suitability of the strategy. Direct changes brought about by kidney ischemic preconditioning on the parameters that we measured have not been studied previously. However, by use of nuclear magnetic resonance, Cochrane et al. (11) showed an improved recovery of postischemic intracellular tissue pH after

This work was supported by European Community (QLK6-CT2000 – 00064); Fondo de Investigaciones Sanitarias de la Seguridad Social (FISS 01/1691) and Ministerio de Ciencia y Tecnologia (SAF 2000 –3090-CE); EC (Microtrans IST99 –13047); and EC (ESPRIT-LTR-23485). 1 Department of Medical Bioanalysis, Instituto de Investigaciones Biome´ dicas, IIBB-CSIC, IDIBAPS. Barcelona, Spain. 2 Department of Physiology, Faculty of Biology, University of Barcelona, Spain. 3 Centro Nacional de Microelectro´ nica (CNM-CSIC), Campus UAB, Barcelona, Spain. Address correspondence to: Dr. G. Hotter, Department of Medical Bioanalysis, IIBB-CSIC-IDIBAPS, C/Rosello´ , 161, 7a planta, 08036 Barcelona, Spain. E-mail: [email protected]. Received 8 July 2002. Revision requested 30 September 2002. Accepted 13 November 2002. DOI: 10.1097/01.TP.0000054683.72223.2D 744

March 27, 2003

745

PALACIOS ET AL.

preconditioning. In addition, in myocardial tissue, preconditioning pretreatment has been shown to attenuate changes of electrical impedance during ischemia (14). Furthermore, it has been suggested that the protective role of ischemic preconditioning in the heart is mediated by adenosine triphosphate (ATP) K⫹-sensitive mitochondrial and cell membrane channels opening (15). This study evaluated the capability of implanted renal tissue microelectrodes for early detection of kidney ischemia and for monitoring changes caused by preconditioning pretreatment. We also explored the basis of the protective action of ischemic preconditioning. The results showed rapid changes in the three variables studied induced by kidney I/R. However, only the impedance values suggested a significant effect for preconditioning treatment, essentially during the reflow period.

portable unit that was suited to our experimental conditions and viable for future clinical application. ISEs (pH and K⫹) were calibrated with saline solutions and precision buffers, and time responses of these electrodes were less than 5 seconds in all circumstances. Data were not corrected for drift. A detailed report of the technical implementation and the characteristics of the impedance probes and related instrumentation is currently in preparation. An impedance microelectrode needle was inserted into the kidney by direct puncture. Bleeding was negligible. Effectively measuring impedance volume predominantly included the medullar region, because the inner electrode lay in the medulla and the outer one in the joint area between the medulla and cortex. ISE probes were inserted intrarenally at a depth of approximately 2 mm, with the help of a cutting edge needle allowing controlled deep puncture. Bleeding was minimal, and the tip of the electrodes lay in the boundary between the medulla and cortex. Electrodes were held by supports, and the insertion was maintained constant. Before the experiments, the electrodes were calibrated with adequate saline buffer systems.

MATERIALS AND METHODS

Statistics

Animals and Anesthesia

Data are expressed as mean ⫾ standard error of mean (SEM). Two-way repeated factor analysis of variance (ANOVA) with a single repeated factor (time) and a between factor (group) was applied to obtain the significance between groups, and Tukey’s multiple mean comparison test was used to assess differences from zero-time values within the group. A repeated-measures ANOVA was run for all data sets and, if significant, for ischemic and reperfusion intervals. Data plots were fitted to dual-line linear or logarithmic function. The statistical method described by Jones and Molitoris (16) was used to determine the breakpoints for ischemic individual trends. Unpaired or paired Student’s t test was performed to assess the differences between breakpoints and blood values.

The study was conducted under the supervision of our institution’s ethics commission and conformed to European Union guidelines for the handling and care of laboratory animals. Male Wistar rats (Ifa Credo, Spain), weighing 250 to 300 g, were anesthetized with sodium pentobarbital (50 mg/kg). Supplemental boluses were administered during each experiment. Polyethylene cannulas (PE-50, Clay Adams, Sparks, MD) were inserted through the left carotid artery into the aorta for blood sampling and saline infusion (1 mL/100 g/hr). A tracheotomy was performed, the trachea was intubated (polyethylene tubing, PE-240), and ventilation was maintained using a Harvard animal respirator. We kept PaCO2 values between 4.5 and 6 kPa using ventilatory control, whereas PaO2 was maintained between 15 and 20 kPa by adequate oxygen-air control of gas mixture supply. The abdominal area was covered with saline wet gauze at 37°C and a plastic cover to minimize dehydration of exposed tissues. Animals were permanently exposed to radiant heat, and deep abdominal and kidney surface temperatures were maintained at 36°C to 37°C (thermocouple probes KP1/45, Kane May, London UK). Arterial blood samples (0.15 mL) were obtained initially and during the reflow periods for gasometric control and measurement of arterial pH and plasma [K⫹] (BMS3 MK2 Blood Microsystem and EML100 Electrolyte Metabolite Laboratory, Radiometer, Denmark). Experimental Protocol To induce kidney I/R, laparotomy was performed, and the left renal pedicle was dissected and occluded with a nontraumatic microvascular clamp. Rats were randomized into three groups. A minimum of n⫽5 valid whole data set per parameter required six to eight animals per group as the result of fall-out of some measurements. The three groups were as follows: sham-operated controls (C); 45 min left renal ischemia, followed by 60 min reperfusion (I/R); and 45 min left renal ischemia, followed by 60 min reperfusion, preceded by ischemic preconditioning of 10 min of ischemia and 10 min of reperfusion (IP). Microelectrode Monitoring The monitoring assembly, based on Microsystems technology, was developed by the Centro Nacional de Microelectro´ nica. Probes included novel silicon needles (0.5⫻0.6-mm cross section) using a linear array of four planar electrodes (0.3⫻0.3 mm) for bioimpedance measurements at 1 kHz and commercial miniaturized ion-selective electrodes (ISEs) for pH and K⫹ activity measurement (Diamont General Development Corp., Ann Arbor, MI). Multifrequency analysis of impedance in rat kidney was performed previously, and 1 kHz was selected as a suitable frequency to measure extracellular space resistance properties. Single frequency determinations made it easier to develop an easy-to-handle,

RESULTS

Representative plots of individual experiments and mean value graphic analyses are shown for tissue pH values (Fig. 1A,B), interstitial potassium ion concentration (Fig. 2A,B), and impedance module (Fig. 3A,B). Baseline arterial plasma pH and [K⫹] measurements were used to set zero-time values of tissue pH and [K⫹]. Arterial blood gas composition was controlled, and no significant changes in acid-base status were found at any time during the experiments (Table 1). Ischemia Renal vascular occlusion induced immediate, rapid changes in the variables monitored: a decrease in pH and an increase in [K⫹] and impedance. Significant differences from baseline values were found for three variables (P⬍0.01) after 5 min of ischemia (Figs. 1B, 2B, 3B). Rapid initial changes in the three parameters were followed by a rate decrease as ischemia progressed. A biphasic tracing was apparent in almost all individual data plots (Figs. 1A, 2A, 3A), with a lower absolute value of slope in the second phase. In this second phase, pH, but not impedance or [K⫹], reached a plateau phase. Breakpoints were sharply defined for pH ischemic depiction (Fig. 1A) and were normally very apparent for impedance (Fig. 3A) and [K⫹] (Fig. 2A), with a single exception for each parameter in which a single logarithmic was not significantly improved by a two-line fit (Fig. 2A, plot I). The time breakpoints found (BP⫽mean⫾SEM) during ischemia, measured from the beginning of individual records (BPpH⫽30.6⫾2.8 min, BP[K⫹]⫽43.75⫾2.17 min, BPImpedance⫽36.9⫾3.55 min), showed the lowest mean val-

746

TRANSPLANTATION

FIGURE 1. Extracellular pH in the three groups. (A) Representative single experimental records: control (C), ischemiareperfusion (I/R) (I), and preconditioned plus I/R (P). Dualline logarithmic fit determines breakpoint for I and P plots. (B) pH mean values ⴞ standard error of mean (SEM). Significant differences given by analysis of variance (ANOVA test) between groups for the different time periods are indicated in figures; significant differences from baseline within group are discussed in the text.

ues for pH; the differences between BPpH and BP[K⫹] were significant (P⬍0.01). Reperfusion Vascular clamp release increased kidney pH, but the mean baseline levels were not recovered during the experiment

Vol. 75, No. 6

FIGURE 2. Extracellular potassium ion concentration ([Kⴙ]) in the three groups. (A) Representative single experimental records. Both a single logarithmic function for I plot and dual-line logarithmic fit determining a breakpoint for P plot are represented. (B) [Kⴙ] mean values ⴞ SEM. Significant differences given by ANOVA test between groups for the different time periods are indicated in the figures; significant differences from baseline within group are discussed in the text.

(Fig. 1B). [K⫹] decreased rapidly; this was the only parameter in the ischemic group that returned to baseline values (P⬎0.05) during the reflow period (Fig. 2B). Postischemic impedance decreased sharply, reaching a minimum within the first 10 min of reperfusion, but then increased for the second time, reaching new peak values after approximately 20 min of reperfusion, before a slow decrease (Fig. 3A,B).

March 27, 2003

747

PALACIOS ET AL.

ischemia (Fig. 3B). Furthermore, during reperfusion, preconditioning significantly decreased impedance values, eliminating the postischemic increase of impedance found in nontreated rats (Fig. 3A,B).

DISCUSSION

Understanding the effects of I/R injury is important and increasingly relevant with the use of marginal organs, particularly organs from non-heart-beating donors. The use of microelectrodes to monitor the effects of ischemia is clearly an important way of studying the critical events that take place and, possibly, establishing which organs are likely to remain viable. Although experiments have shown that pH and PCO2 electrodes placed at the surface of the kidney are able to detect renal cortical ischemia (3, 4), no current method exists for reliable monitoring of kidney ischemia. The use of microsensor technology for real-time measurement of intraorgan tissue parameters may lead to possible clinical applications in the near future, as smaller probes and more advanced telemetric monitoring technology become available. This article discusses the changes induced by kidney I/R and ischemic preconditioning treatment on three renal tissue parameters, which all register changes soon after the onset of ischemia and thus may be good candidates for detecting ischemia and monitoring its therapeutic management. Model and Methodologic Aspects

FIGURE 3. Impedance module values in the three groups. (A) Representative single experimental record. Both a logarithmic and linear dual-line functions, determining breakpoints, are fit, respectively, for I and P plots. (B) Relative impedance module (referred to baseline) in the three groups. Mean ⴞ SEM. Significant differences given by ANOVA test between groups for the different time periods are indicated in figures; significant differences from baseline within group are discussed in the text.

Effect of Preconditioning Ischemic preconditioning did not significantly influence pH and [K⫹] values during I/R (Figs. 1B and 2B). Slopes of the curves fitted to individual data and time breakpoints found (BPpH⫽35.7⫾1.65 min, BP[K⫹]⫽38.2⫾3.39 min, BPImpedance⫽40.33⫾0.88 min) did not significantly change by preconditioning. Nevertheless, preconditioned kidneys reached higher impedance mean values during protracted

The ischemic protocol we used is a common model of ischemic acute renal failure (11). The preconditioning schedule was established after analyzing the results of previous studies, which showed its protective effect (11, 12). However, different experimental studies of the protective effect of ischemic preconditioning on kidney I/R injury and renal function have produced discordant results (11, 13), and it has been suggested that a critical factor for success is the preconditioning schedule (11, 12). Microelectrode sensors were placed close to the corticomedullar junction, the area in which the most pronounced injury has been found in I/R. To implant the microelectrodes, a small puncture wound was made in the kidney, but bleeding was zero or minimal, responses to vascular maneuvers were rapid, and the variables measured were stabilized quickly. Renal microangiography has shown a well-preserved microvasculature around silastic coils implanted after a puncture wound in the kidney cortex and the medulla for gas-tension measurement (17). Furthermore, although insertion of microdialysis probes in rat kidney analysis showed transient changes in interstitial metabolites, attributed to puncture injury, values returned to normal levels within minutes (18). The impedance probes used in most past studies (19) are excessively invasive for intraorgan small-volume measurement, as is required in the rat kidney. In this study, these problems were overcome by the use of a novel miniature silicon needle. Unlike ISEs, which measure tissue properties at the electrode tips, bioimpedance properties are measured in intact tissue in a volume corresponding to the electrical field generated around the electrodes (7). In small organs such as the rat kidney, because the electric field covers a high relative volume, the measurements are representative of homogeneous tissue properties.

748

TRANSPLANTATION

Vol. 75, No. 6

TABLE 1. Baseline and final arterial blood gases and acid-base values (meanⴞSEM) in the three rat groups Group

C I/R P⫹I/R

Time

pH

PCO2 KPa

[HCO3⫺] mEq/L

Base excess mEq/L

PO2 KPa

Baseline Final Baseline Final Baseline Final

7.372⫾0.020 7.402⫾0.037 7.380⫾0.016 7.385⫾0.013 7.388⫾0.008 7.365⫾0.025

5.39⫾0.020 5.03⫾0.38 5.46⫾0.26 5.23⫾0.32 4.97⫾0.32 5.95⫾0.28

23.5⫾0.6 23.2⫾0.8 24.2⫾1.0 23.0⫾1.1 23.4⫾0.7 24.2⫾1.1

⫺1.2⫾1.10 ⫺1.0⫾1.35 ⫺0.8⫾0.90 ⫺1.5⫾0.90 ⫺1.8⫾0.49 ⫺1.5⫾1.19

17.41⫾0.69 16.83⫾0.79 18.29⫾0.45 17.94⫾0.31 17.99⫾0.58 17.65⫾0.49

SEM, standard error of mean; I/R, ischemia-reperfusion; C, control; P, preconditioned.

Changes Monitored During Ischemia-Reperfusion Each variable monitored during ischemia showed an immediate change, followed by a decreased rate change with time. Whereas impedance and [K⫹] continued to increase during late ischemia, pH reached a minimal plateau phase. This difference in evolution over time can be attributed to the different underlying mechanisms of the ischemia-induced changes. Lack of oxygen causes intracellular pH to decrease as the result of glycolytic lactic acid generation and ATP depletion processes (2). The ischemic pH breakpoint probably indicates the end of these active metabolic processes, because beyond it, pH reaches a plateau phase. Metabolic influence also seems apparent in the evolution of [K⫹] and impedance, because the time breakpoints for the three parameters are close. However, [K⫹] and impedance changes, beyond the respective breakpoints, indicate passive ion and water shifts occurring after ATP depletion (8). Significant tissue cellular ATP depletion has been found within the first minutes of renal ischemia (20), and in other studies, ATP and intracellular pH values remained stable beyond 25 min of ischemia (11). Our results showing extracellular pH evolution are in agreement with these changes. Among the factors that may contribute to rapid increase of interstitial potassium after the onset of ischemia, dysfunction of the sodium-potassium pump as the result of ATP depletion (8), activation of potassium channels (5), and water shifts to intracellular compartment (8) should be considered. In the kidney, as in other tissues, one of the earliest ischemic events is a loss of intracellular K⫹ (5). Renal ischemia increases the open probability of the K⫹ ATP channels independently of pump activity (21). Our results indicate that the activation of potassium channels is an early response to hypoxia, and that K⫹ efflux must be independent of active K⫹ uptake, because the extracellular K⫹ rate increase is still pronounced in the late ischemic phase (Fig. 2A,B). During ischemia, we found a progressive increase of bioimpedance measured at low frequency. This rise is considered to reflect the occurrence of an anoxic edema as the result of cell swelling, which leads to a reduction of extracellular space, an increase in extracellular resistance, and cell-to-cell uncoupling (7, 22) in organs with gap junctions like the kidney (23). Tissue edema has been reported to be one of the major causes of renal ischemic damage (9, 24). The prompt, rapid onset of intrarenal impedance changes, with a maximum rate of increase after renal artery occlusion, contrasts with the sequential changes described in myocardial impedance (14). Our study found a second increase in impedance values during reperfusion. This increase may indicate the occurrence of intracellular edema during reflow, probably linked with the no-reflow phenomenon, that is, the absence

or reduction of vascular reflow after relief of kidney ischemia (24). Decreases in renal blood flow of 30% and 50% have been measured in rats and dogs, respectively, after 30 min of reperfusion after 30 and 60 min of renal artery clamping (25, 9). Ischemic Preconditioning Influence on Ischemia-Reperfusion Measured Parameters The evolution of monitored intrarenal parameters supports the idea that the mechanisms underlying the protective actions of ischemic preconditioning vary according to organ (1), particularly between heart and kidney (11). In our study, ischemic preconditioning did not reduce ischemic acidosis, as has been reported for myocardial ischemia (26). Although severe intracellular acidosis is considered a cause of injury in ischemic myocardial tissue (11), acidic tissue pH during I/R may have a protective effect in kidney and other tissues (27). Measuring intracellular pH (pHi) by means of nuclear magnetic resonance spectroscopy, Cochrane et al. (11) also found that pHi decreased to a similar degree in preconditioned and control kidneys during ischemia; however, pHi recovery during reflow improved in preconditioned kidneys. Our result, the absence of any improvement in tissue pH recovery in preconditioned kidneys during reperfusion, does not necessarily contradict Cochrane et al.’s finding. Tissue-implanted microelectrodes measure extracellular pH, and the improved recovery of intracellular pH in preconditioned kidneys with no changes in extracellular pH could be attributed to the enhanced activity of membrane acid-base equivalent exchangers in the postischemic period. We also report that preconditioning did not increase extracellular potassium during ischemia. Therefore, our results do not favor the idea of a protective pathway of preconditioning in the kidney through the opening of ATP-sensitive potassium channels, as has been proposed for myocardial tissue (15, 28, 29). In the ischemic kidney, it has been postulated that the activation of potassium channels contributes to hypoxic injury in proximal tubules (5). Furthermore, contrary to what has been suggested for myocardial tissue, attenuation of potassium loss during hypoxia by channel blockage protects against hypoxia-induced injury (5, 6). Surprisingly, we found higher tissue impedance values in the late ischemic phase of preconditioned kidneys. In contrast with its effect on myocardial tissue, ischemic preconditioning did not postpone or attenuate the ischemic tissue resistivity increase in kidney (14, 15). These results indicate that the preventive effect of kidney ischemic preconditioning cannot be attributed to a reduction of hypoxic intracellular edema. Indeed, kidney preconditioning may be effective despite the increased bioimpedance values found. In this re-

March 27, 2003

PALACIOS ET AL.

gard, the results of a study on differential effects of swelling and anoxia (60 min warm ischemia) on kidney function showed that nonanoxic edema is much less damaging than anoxia, and that ischemic injury in kidney was not the simple consequence of the spatial disruption of cell architecture (9). The most interesting effect of ischemic preconditioning found here was a significant reduction of postischemic tissue impedance. This may indicate that preconditioning decreases cellular edema during the reflow period. Further information supports this (25), showing that ischemic preconditioning protects against I/R-mediated kidney injury by improving renal blood flow during reperfusion. The same action, described in other tissues such as the small bowel, has been associated with an improvement of tissue injury and metabolism (30). CONCLUSION

Our results show that intrarenal microelectrode real-time measurement of any of the three variables studied is a suitable technique for early detection of ischemia in the kidney. Tissue-impedance monitoring using a single microprobe showed the occurrence of reperfusion cellular edema. The improvement in reflow tissue bioimpedance was the most noticeable beneficial effect of ischemic preconditioning. Some physiologic conclusions of interest, regarding the actions of preconditioning, can be drawn from the negative results obtained. Evidence is given to suggest that, unlike the heart, the protective action of preconditioning in the kidney does not seem to be mediated by the opening of potassium channels, decreased tissue acidosis, or improved tissue impedance properties during the ischemic period. REFERENCES 1. Ishida T, Yarimizu K, Gute DC, et al. Mechanisms of ischemic preconditioning. Shock 1997; 8: 86. 2. Fiddian-Green RG. Gastric intramucosal pH, tissue oxygenation and acidbase balance. Br J Anaesth 1995; 74: 591. 3. Dmochowski JR, Couch NP, Kempf RA, et al. Electrometric surface pH of the ischemic kidney and the effect of hypothermia. J Surg Res 1966; 6: 45. 4. Tønnessen TI, Kvarstein G. PCO2 electrodes at the surface of the kidney detect ischaemia. Acta Anaesthesiol Scand 1996; 40: 510. 5. Reeves WB, Shah SV. Activation of potassium channels contributes to hypoxic injury in proximal tubules. J Clin Invest 1994; 94: 2289. 6. Peters SM, Tijsen MJ, Bindels RJ, et al. Protection against hypoxic injury of rat proximal tubules by felodipine via a calcium-independent mechanism. Pflugers Arch 1995; 431: 20. 7. Gersing E. Impedance spectroscopy on living tissue for determination of the state of organs. Bioel Bioenerg 1998; 45: 145. 8. Leaf A. Regulation of intracellular fluid volume and disease. Am J Med 1970; 49: 291.

749

9. Jamart J, Lambotte L. Differential effects of swelling and anoxia on kidney function and its consequences on the mechanism of action of intracellular organ preservation solutions. Transplantation 1982; 34: 176. 10. Ueda N, Kaushal GP, Shah SV. Apoptotic mechanisms in acute renal failure. Am J Med 2000; 108: 403. 11. Cochrane J, Williams BT, Banerjee A, et al. Ischemic preconditioning attenuates functional, metabolic and morphologic injury from ischemic acute renal failure in the rat. Ren Fail 1999; 21: 135. 12. Riera M, Herrero I, Torras J, et al. Ischemic preconditioning improves postischemic acute renal failure. Transplant Proc 1999; 31: 2346. 13. Islam CF, Mathie RT, Dinneen MD, et al. Ischaemia-reperfusion injury in the rat kidney: the effect of preconditioning. Br J Urol 1997; 79: 842. 14. Cinca J, Warren M, Carreno A, et al. Changes in myocardial electrical impedance induced by coronary artery occlusion in pigs with and without preconditioning: correlation with local ST-segment potential and ventricular arrhythmias. Circulation 1997; 96: 3079. 15. Tan HL, Mazon P, Verberne HJ, et al. Ischemic preconditioning delays ischemia induced cellular electrical uncoupling in rabbit myocardium by activation of ATP sensitive potassium channels. Cardiovasc Res 1993; 27: 644. 16. Jones RH, Molitoris BA. A statistical method for determining the breakpoint of two lines. Anal Biochem 1984; 141: 287. 17. Nelimarkka O, Niinikoski J. Renal venous oxygen tension as an indicator of tissue hypoxia in hemorrhagic shock. Crit Care Med 1986; 14: 128. 18. Eklund T, Wahlberg J, Ungersted U, et al. Interstitial lactate, inosine and hypoxanthine in rat kidney during normothermic ischaemia and recirculation. Acta Physiol Scand 1991; 143: 279. 19. Suzuki H, Robinson MK, Rounds JD, et al. Glutathione deficiency accentuates hepatocellular fluid accumulation after ischemia-reperfusion. J Surg Res 1994; 57: 632. 20. Molitoris BA, Chan LK, Shapiro JD, et al. Loss of epithelial polarity: a novel hypothesis for reduced proximal tubule Na⫹ transport following ischemic injury. J Membr Biol 1989; 107: 119. 21. Quast U. ATP-sensitive K⫹ channels in the kidney. Naunyn Schmiedebergs Arch Pharmacol 1996; 354: 213. 22. Grimnes S, Martinsen OG. Bioimpedance and bioelectricity basics. London, Academic Press 2000. 23. Hillis GS, Duthie LA, Mlynski R, et al. The expression of connexin 43 in human kidney and culture cells. Nephron 1997; 75: 458. 24. Flores J, DiBona DR, Beck CH, et al. The role of cell swelling in ischemic renal damage and the protective effect of hypertonic solute. J Clin Invest 1972; 51: 118. 25. Ogawa T, Nussler A, Tuzuner E, et al. Contribution of nitric oxide to the protective effects of ischemic preconditioning in ischemia-reperfused rat kidneys. J Lab Clin Med 2001; 138: 50. 26. Asimakis GK, Inners-McBride K, Medillin G, et al. Ischemic preconditioning attenuates acidosis and post-ischemic dysfunction in isolated rat heart. Am J Physiol 1992; 263: H887. 27. Shanley PF, Shapiro JI, Chan L, et al. Acidosis and hypoxic medullary injury in the isolated perfused kidney. Kidney Int 1988; 34: 791. 28. Cleveland JC Jr., Meldrum DR, Rowland RT, et al. Adenosine preconditioning of human myocardium is dependent upon ATP-sensitive K⫹ channel. J Mol Cell Cardiol 1997; 29: 175. 29. Cohen MV, Baines CP, Downey JM. Ischemic preconditioning: from adenosine receptor of KATP channel. Annu Rev Physiol 2000; 62: 79. 30. Sola A, Rosello-Catafau J, Alfaro V, et al. Modification of glyceraldehyde3-phosphate dehydrogenase in response to nitric oxide in intestinal preconditioning. Transplantation 1999; 15: 1446.