|
|
||||||||
Ann Thorac Surg 1998;65:993-998
© 1998 The Society of Thoracic Surgeons
a Department of Cardiac Surgery, Royal Infirmary, Glasgow, Scotland, United Kingdom
b Department of Pathology, Royal Infirmary, Glasgow, Scotland, United Kingdom
c Department of Nuclear Medicine, Royal Infirmary, Glasgow, Scotland, United Kingdom
Accepted for publication October 21, 1997.
Address reprint requests to Dr Wheatley, Dept of Cardiac Surgery, Royal Infirmary, 10 Alexandra Parade, Glasgow G31 2ER, Scotland
| Abstract |
|---|
|
|
|---|
Methods. Kidneys were harvested from each of four groups of 6 pigs. Group A were anesthetized and heparinized only. The remaining three groups underwent cardiopulmonary bypass at 28°C, group B for 30 minutes and groups C and D for 120 minutes; group D had an additional 30 minutes of normothermic perfusion at the end of the experiment. Renal cortical blood flow was measured using radiolabeled microspheres. Microvascular morphology was defined by corrosion casting and scanning electron microscopy.
Results. In group A, renal vascular resistance was 61 ± 5.1 mm Hg · mL-1 · min-1. This value decreased to 28 ± 7.8 in group B and 25 ± 4.0 in group C (p < 0.05), and increased in group D to 40 ± 4.1 (p < 0.05 versus groups A, B, and C). Cortical thickness, as measured by microvascular casts in groups A, B, and C, was 33, 34, and 31 mm, respectively, with equal distribution of the resin to the superficial and deep cortex but was significantly reduced in group D to 22 mm (p < 0.05 versus groups A, B, and C), with failure of the resin to fill the superficial cortical layer. Diameters of glomeruli as seen on the casts were 111 ± 10.38 µm in group A, 100 ± 9.24 µm in group B, and 82 ± 4.4 µm in group C (p < 0.05 group A versus group C). The glomeruli from group D were still significantly smaller than group A (93 ± 10.35 µm, p < 0.05). Mean glomerular capillary diameters were 4.65 ± 0.26 µm in group A, 3.9 ± 0.16 µm in group B, 3.6 ± 0.19 µm in group C, and 3.65 ± 0.3 µm in group D (p < 0.05 group A versus groups B, C, and D).
Conclusions. Hypothermic nonpulsatile cardiopulmonary bypass decreased renal vascular resistance, but the superficial and deep layers of the cortex were perfused equally. Glomeruli were reduced in size because of capillary narrowing. This was consistent with diversion of blood through bypass channels. With restoration of normothermia, underperfusion of the superficial cortex occurred, with potential for damage to these nephrons during the increased metabolic demands of rewarming.
| Introduction |
|---|
|
|
|---|
The anatomic site of this injury has so far evaded detection [5, 6]. The use of pulsatile, as opposed to nonpulsatile, perfusion may improve distribution of intrarenal blood flow, with preferential perfusion of parts of the cortex, suggesting that changes in the microcirculation may be of importance in this pathology [7]. The superiority of increased pump flow rate as opposed to vasoconstriction as a means of increasing renal perfusion has been demonstrated recently in pigs [8]. In this study, vasoconstriction despite producing increased systemic arterial pressure did not increase renal perfusion.
The effects of CPB on the blood vessels of the kidney were studied in a porcine model using CPB, hypothermia, and rewarming. Blood flow within the different layers of the renal cortex was measured using radioactive microspheres. Microvascular corrosion casting was used to delineate the exact distribution of intrarenal blood flow. Scanning electron microscopy of the resin casts was used to elucidate changes at the glomerular level. This technique has been used extensively in the past to reproduce the state of the renal microvasculature in many species [9, 10].
| Material and methods |
|---|
|
|
|---|
The CPB was set up between bicaval cannulas for venous drainage (Gambro 36/51 whistle tip) and an aortic cannula for arterial return (Bardic 16F catheter). The electrocardiogram and arterial and venous pressures were continuously monitored with a Sirecust 404 monitor and polygraph. The bypass circuit comprised a Safe II membrane oxygenator (Polystan, Wilford, England) and a Stockert-Shiley roller pump (Stockert-Shiley, Midhurst, England). One liter of crystalloid prime was used and flows were maintained at 80 mL · kg-1 · min-1 at 28°C. In groups C and D, the aorta and pulmonary artery were cross-clamped and 10 mL/kg body weight cold St. Thomas crystalloid cardioplegia was instilled into the aortic root every 30 minutes. In group D, the animals were rewarmed for the final 30 minutes to 36°C. At the end of the experiment, the animals were killed by 10-mg bolus doses of morphine and midazolam followed by exsanguination into the venous reservoir. This study was approved by the University of Glasgow ethical committee. All animals received humane care in compliance with the European convention on animal care and in accordance with the guidelines of "The Animals (Scientific Procedures) Act" 1986.
Measurement of tissue blood flow
Tissue perfusion was measured using 15-µm microspheres, radiolabeled with cobalt or chromium. The 1.0 to 1.5 x 106 microspheres were agitated on a roller to ensure complete mixing and then injected rapidly. Blood was withdrawn from the femoral arterial line using a controlled rate aspiration pump set at 7.5 mL/min commencing 30 seconds before and continuing for 1 minute after the injection. The sample was divided into counting vials and radionuclide activity assessed.
Regional blood flow per 100 grams of tissue (RBF) was calculated according to the formula:
, where Ct is the activity per gram of tissue sample, WR is the rate of pump withdrawal (7.5 mL/min), and Cr is the activity in the reference blood sample. Renal vascular resistance (RVR), expressed as mm Hg · mL-1 · min per 100 g, was calculated as:
.
Corrosion casting and scanning electron microscopy
After excision of the kidneys, the superior polar renal artery was catheterized with a 5F cannula and flushed with heparinized saline until free of blood. Batsons No. 17 methylmethacrylate monomer solution (1:1.25, Batsons monomer to methylmethacrylate; Park Scientific, Northampton, England) was injected at controlled pressure of 50 to 70 mm Hg (measured on the inlet side) until it was seen to be flowing freely from the renal veins. The preparation was left at room temperature for 2 hours to allow polymerization before being bisected coronally to visualize the intrarenal anatomy. The blocks were immersed in alternate baths of 20% potassium hydroxide and water every 48 hours, until completely macerated of all organic material. Any residual fat adherent to the resin skeleton was dissolved by detergent solution.
The vascular casts were mounted on aluminium blocks with silver paste, coated in a gold/palladium Polaron SE 5000 sputter-coater (Electron Microscopy Services, London, England) and viewed under a scanning electron microscope (Jeol UK Ltd, Welwyn Garden, England). The working distance was standardized at 48 mm and the accelerating voltage at 5 keV. An overview was obtained of the distribution of renal blood flow by scanning at x10 magnification, and measuring the cortical width from the arcuate artery to the edge of the cortical vasculature. Specimens were also viewed at a neutral angle and at 15 degrees clockwise and counterclockwise under a magnification of x400 to x700 in a random blinded manner. Afferent arterioles were followed to their glomeruli and subsequent transformation into the efferent arterioles and peritubular capillary plexus. Bypass channels were defined as large diameter vessels directly linking afferent arteriole and efferent arteriole or capillary plexus, without forming a glomerulus. Glomeruli were scanned onto an image analysis system and the following parameters measured using a digitizing pad (Fig 1): polar diameter, distance parallel to the vascular pole; transverse diameter, the distance perpendicular to PD; afferent arteriole, the diameter 10 µm proximal to the vascular pole; and glomerular capillaries, average diameter of 10 vessels per glomerulus.
|
| Results |
|---|
|
|
|---|
|
|
|
|
|
|
|
|
| Comment |
|---|
|
|
|---|
The finding of reduced renal vascular resistance during nonpulsatile perfusion at 28°C may have two explanations. First, there may be uncontrolled dilatation of afferent and efferent arterioles as compensation for the nonpulsatile low pressure arterial waveform and hypothermia. Andersson and colleagues [5] demonstrated renal blood flow (measured by thermodilution) to be correlated more strongly with pump flow rates than with systemic arterial pressure. This suggests that control of the renal circulation during CPB may be at the level of the bypass pump, rather than at a vascular level. Although the systemic arterial pressures maintained in our study were low, the addition of vasoconstrictors to elevate these may not have altered perfusion patterns as recently demonstrated by ODwyer and colleagues [8].
Second, a state of fixed vasodilatation can be hypothesized in the afferent and efferent arterioles resulting in unimpeded flow through all cortical glomeruli, with filtration pressures across the glomerular capillary membrane directly dependent on systemic perfusion pressure. This has been shown previously to be the case by Urzua [13] and Badner [16] and their colleagues. They reported a direct relationship of both urine output and glomerular filtration rate with systemic perfusion pressure. This may also explain the failure of urine output during CPB to predict postoperative renal function [12]. One can speculate on the presence of a "functional" circulation maintaining glomerular filtration and urine output, and a "nutritional" circulation maintaining tissue perfusion and viability. Although the two may normally be identical, they may become separate entities during the abnormal hemodynamics of hypothermic CPB. With rewarming in group D, despite nonpulsatile perfusion, the renal vascular resistance increased again and may be a reflection of the reduction in renal vascular bed available at this stage, as the cortical cast thickness in these kidneys was reduced. This was primarily attributable to decreased blood flow through the superficial cortical nephrons, although blood flow through the juxtamedullary nephrons was preserved. The increase in mean size of the glomeruli seen in group D kidneys may be attributable to increased interstitial edema, which may play a part in increasing vascular resistance. Although this has not been demonstrated directly, the finding of unchanged mean capillary size in the casts of group D glomeruli would indicate an increase in the extravascular space as the major contributor to the larger glomeruli. It can only be speculated that this is attributable to rewarming and not the extra period of bypass as a warm bypass group was not included in the study.
The strong correlation of renal vascular resistance with hemoglobin levels would be expected to occur due to the well recognized effect of hemodilution on blood viscosity and tissue perfusion [17]. Although the degree of hemodilution achieved during this study was excessive, the confounding effects of blood transfusion were avoided. Our results would also indicate a major role for hypothermia, although the hemoglobin levels remained low in group D, the RVR was higher than in groups B and C. It may be that some degree of vascular smooth muscle action may be regained after return to normothermia, allowing the renal vascular resistance to increase despite the low systemic pressures and reach the level seen in group D animals. There may be some heterogeneity in this effect among different nephrons, some displaying preglomerular vasoconstriction and others remaining dilated [18, 19]. The potential for ischemic damage to some poorly perfused nephrons during this stage of increased metabolic demand is clearly apparent. This may be the crucial time when manipulations aimed at improving renal cortical perfusion, such as dopamine and mannitol, could be instituted. Further studies using a warm bypass group may clarify whether it is the hypothermia or the length of bypass that sets the stage for the poor perfusion of the superficial cortex during subsequent rewarming.
An alternative explanation for the low RVR during CPB may be that thoroughfare vascular channels exist within the kidneys allowing blood to bypass the high resistance glomeruli partially or completely. The afferent arterioles of hibernating animals whose renal blood flow can be reduced dramatically during winter sleep are known to be controlled by sphincters in the preglomerular vessels [20]. This may be analogous to the conditions found during hypothermic CPB. The maintenance of the distribution of perfusion to the entire thickness of the cortex seen in the corrosion casts of the hypothermic bypass groups B and C, compared with group A, would favor flow through these "aglomerular" bypass channels as a possible explanation for the reduced renal vascular resistance.
The existence of four anatomically distinct aglomerular pathways was first suggested by Cassellas and colleagues in 1979 and may act as bypass pathways through the mesangium after blockage of the glomerular capillaries [2123]. These channels may exist in all kidneys as a potential source of nutritional perfusion during periods of altered hemodynamics or embolic occlusion that may occur during CPB. This divergence in nutritional and functional perfusion may explain the failure of high systemic arterial pressure during CPB to improve subsequent renal function [12]. It may be that increasing arterial blood pressure increases flow through the glomeruli, thus increasing filtration rates. Low pressures may continue to allow flow through the nutritional circulation, while increasing efferent arteriolar tone to maintain filtration.
The changes in the sizes of the glomeruli would again support the hypothesis of blood flow diversion through bypass pathways during hypothermic CPB. If all blood flow were to pass through the glomeruli, then the decrease in RVR would be expected to increase the sizes of the glomerular capillaries to accommodate the increased flow. If the glomerular capillaries were bypassed, however, their volume would not be increased.
Renal dysfunction after CPB is a common occurrence. The reversibility of this dysfunction would favor a vascular change as an explanation. Although renal vascular resistance decreases during hypothermic CPB, the increased flow may pass through channels other than through the glomeruli. This could be analogous to hibernating animals where, although renal perfusion and viability are maintained, little or no urine is produced during winter sleep. With return of normothermia (and pulsatile perfusion), these changes may gradually reverse, thereby allowing return of normal renal function. During the critical phase of rewarming, however, nephrons with incomplete return of perfusion, particularly in the superficial cortex, may be at risk of ischemic damage. Under conditions of prolonged hypoxia or hypotension associated with low cardiac output, however, this damage may be compounded and lead to clinically apparent and permanent dysfunction.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. A. J. M. Huybregts, R. de Vroege, E. K. Jansen, A. W. van Schijndel, H. M. T. Christiaans, and W. van Oeveren The Association of Hemodilution and Transfusion of Red Blood Cells with Biochemical Markers of Splanchnic and Renal Injury During Cardiopulmonary Bypass Anesth. Analg., August 1, 2009; 109(2): 331 - 339. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Maier, W. R. Hasibeder, C. Hengl, W. Pajk, B. Schwarz, J. Margreiter, H. Ulmer, J. Engl, and H. Knotzer Effects of phenylephrine on the sublingual microcirculation during cardiopulmonary bypass Br. J. Anaesth., April 1, 2009; 102(4): 485 - 491. [Abstract] [Full Text] [PDF] |
||||
![]() |
H Vermeer, S Teerenstra, R. de Sevaux, H. van Swieten, and P. Weerwind The effect of hemodilution during normothermic cardiac surgery on renal physiology and function: a review Perfusion, November 1, 2008; 23(6): 329 - 338. [Abstract] [PDF] |
||||
![]() |
R. B. Yates and M. Stafford-Smith The genetic determinants of renal impairment following cardiac surgery. Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2006; 10(4): 314 - 326. [Abstract] [PDF] |
||||
![]() |
S. W. van den Dool, M. N. Wasser, J. W. de Fijter, J. Hoekstra, and R. J. van der Geest Functional Renal Volume: Quantitative Analysis at Gadolinium-enhanced MR Angiography--Feasibility Study in Healthy Potential Kidney Donors Radiology, July 1, 2005; 236(1): 189 - 195. [Abstract] [Full Text] [PDF] |
||||
![]() |
S Dittrich, A Schuth, H Aurich, J vonLoeper, C Grosse-Siestrup, and P E Lange Haemodilution improves organ function during normothermic cardiopulmonary bypass: investigations in isolated perfused pig kidneys Perfusion, May 1, 2000; 15(3): 225 - 229. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |