Ann Thorac Surg 1999;68:22-28
© 1999 The Society of Thoracic Surgeons
Original Articles
Hyperdynamic circulation of arteriovenous fistula preconditions the heart and limits infarct size
Mahmoud A. Mahgoub, MDa,
Jian-Hau Guo, MDa,
Shi-Ping Gao, MDa,
Mohiuddin M. Taher, PhDa,
David D. Salter, MDa,
Andrew S. Wechsler, MDa,
Anwar S. Abd-Elfattah, PhDa
a Department of Surgery, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia, USA
Address reprint requests to Dr Mahgoub, Zagazig University-Egypt, PO Box 249 Zagazig, Egypt
e-mail: rumahgob{at}rusys.eg.net
 |
Abstract
|
|---|
Background. Chronic arteriovenous fistulae (AVF) create sustained hyperdynamic circulation. It is not known whether hyperdynamic circulation alters myocardial sensitivity to ischemia and reperfusion injury. We tested the hypothesis that AVF activate molecular responses that increase tolerance to infarction in dogs.
Methods. Twelve dogs were divided into two groups: 1) AVF group, where an AVF in the femoral region was done; and 2) sham-operated group (each n = 6). After 8 weeks, left ventricular performance was determined from stroke work/end-diastolic length relationship. Myocardial biopsy was obtained to determine heat-shock protein 70 and adenosine triphosphate (ATP) pool. Left anterior descending coronary artery was occluded for 90 minutes at 37°C, followed by 4 hours of reperfusion. Coronary blood flow was determined using different colored microspheres.
Results. The fistula group showed improvement of left ventricular performance (p = 0.03). The infarct size was significantly lower in the fistula group; it was 9.2 ± 2.0% in the fistula group versus 28.4 ± 5.2% in the sham group (p < 0.05). ATP depletion during ischemia was less in the fistula group (p = 0.02). Regional myocardial blood flow was significantly higher in the fistula group (p = 0.03).
Conclusions. Peripheral AVF improve the left ventricular performance, and decrease infarct size and ATP depletion. This protective effect is caused by the development of collaterals in the coronary circulation without expression of heat-shock protein 70.
|
Introduction
|
|---|
An arteriovenous fistula (AVF) is an abnormal communication between an arterial and a venous channel. AVF are usually characterized as either congenital or acquired. Congenital fistulae may result from failure of the embryonic vascular network to differentiate into arterial and venous systems. Most fistulae are acquired as a result of trauma, infection, neoplasm, and aneurysmal disease. The most common acquired fistulae are iaterogenic, as a result of ever-more frequent use of arterial catheters and needles. Fistulae are created for therapeutic causes such as dialysis and maintenance of patency of arterial graft [1].
The hemodynamic consequences associated with an AVF result from shunting of arterial blood into the low-resistance, high-capacitance venous channel(s), thus bypassing an intervening capillary bed. The magnitude of the local and systemic effects is dependent on the location, diameter, and number of vascular communications. In the feeding artery proximal to the fistula, there is an increase in the velocity and volume of blood flow. Although, normally, there is cessation of arterial blood flow during diastole, if a fistula is opened, antegrade flow is often continuous throughout, resulting in a fall in the peripheral resistance, a decrease in the proximal fistula artery pressure, and an increase in the proximal venous outflow (venous return). In the presence of large fistulae, left ventricular end-diastolic pressure gradually rises, leading, eventually, to high-output heart failure [2].
The cardiac effects of AVF include increased cardiac output, increased heart rate, cardiac dilation, decreased diastolic pressure with wide pulse pressure, and, to a lesser degree, hypertrophy of the heart. Cardiac output increases immediately after an AVF is opened, and peak flow is achieved within a few seconds. Cardiac output increases by an increase in stroke volume and/or an increase in heart rate; both mechanisms are operational in AVF, but an increase in stroke volume accounts for 80% to 90% of the rise in cardiac output. Although increasing the heart rate has little effect on cardiac output when venous return is normal, it does augment cardiac output in the presence of an AVF [3].
Recent work in our laboratory proved that chronic rapid cardiac pacing could decrease the infarct size in the ischemia/reperfusion animal model. In the latter study, limitation of infarct size was associated with expression of the protective heat-shock protein 70 (unpublished data). It is not known if the stress of the hyperdynamic circulation induced by the chronic AVF has a protective effect on the myocardium and what should be the mechanism of protection if any. In the clinical setting, it is not known whether the iaterogenic fistula made for hemodialysis in uremic patients has a protective effect on the myocardium or not. It is also not known whether other causes of hyperdynamic circulation such a pregnancy, anemia, aortic insufficiency, and thyrotoxicosis enhance protection against myocardial ischemia or not.
The purpose of this study is to investigate the effect of chronic peripheral AVF on cardiac performance and to determine whether hyperdynamic circulation stresses the heart in such a way that it becomes preconditioned against infarction.
|
Material and methods
|
|---|
All the animals were treated humanely in accordance with the United States Public Health Service standards as outlined in "Principles of Laboratory Animal Care," formulated by the National Society of Medical Research, and the "Guide for the Care and Use of Laboratory Animals," prepared by the National Academy of Sciences and published by the National Institutes of Health.
Sterile surgery
To establish a model of hyperdynamic circulation, 12 microfilaria-free dogs of either gender, weighing 19 to 26 kg, were randomly divided into two groups: a sham-operated group and an AVF group (each n = 6). Baseline echocardiography was performed on the animals. All the animals were anesthetized with intravenous injection of sodium pentobarbitol (30 mg/kg; Veterinary Laboratories, Lenexa, KS). Dogs were intubated, mechanically ventilated with a Bennet MAI respiratory (Puritan, Berkeley, CA), and heparinized by given porcine-based heparin (400 units/kg IV). All the animals were subjected to sterile surgery in the form of a 10-cm longitudinal skin incision in the medial aspect of the thigh. The femoral artery and femoral vein were exposed, the branches of the artery and the tributaries of the vein were ligated, and then the artery and the vein were clamped for 10 minutes. In the fistula group, a side-to-side femoro-femoral 1.5–2-cm arteriovenous fistula was created. After securing homeostasis, the skin was closed and the animal was allowed to recover. All animals received antibiotics in the form of cephazolin 1 g IV, intraoperatively, 2 days postoperatively, and then accordingly. Follow-up of the animals involved daily measurement of the pulse and respiratory rate, cleaning the wound with iodine solution, and palpating the thrill in the fistula group.
An animal was excluded from the study if the thrill was not palpable. Four weeks after the sterile surgery, M-mode and two-dimensional echocardiography was done for all the animals. Eight weeks after the sterile surgery, the animals were subjected to the acute study.
The acute study
Animals were reanesthetized, intubated, and ventilated. Left ventricular performance was assessed from the relationship between stroke work and end-diastolic dimension using a sensitive and load-independent index of contractility [4]. Intraventricular and carotid artery micromanometer-tipped catheters (Millar Instruments, Inc, Houston, TX) were used to monitor left ventricular and carotid pressure, respectively. Left ventricle (LV) dimensions were obtained with pulse transit sonomicrometry (Triton Technology, San Diego, CA). One pair of 2.5-mm-diameter LTZ-piezoelectric hemispheric crystals (Channel Industries, Santa Barbara, CA) was sewn onto the anterior and posterior myocardial wall along the minor axis of the LV. The epicardial diameter of the LV ranged from 40 to 66 mm. Analog data were digitized at 200 Hz and stored on the hard disk of a PC 386-MHz microcomputer. Subsequent analysis was done with the use of interactive software (Crunch program) developed in our laboratory. After complete instrumentation, the baseline preischemic left ventricular performance was assessed.
Assessment of adenine nucleotide pool intermediate metabolism
In a random fashion, a Transmural Tru-cut needle (Travenol Laboratories, Inc, Deerfield, IL) biopsy specimen (5–10 mg) was obtained preischemia, and 30, 60, and 90 minutes postischemia. Biopsy samples were immediately frozen and stored in liquid nitrogen. Purines in each specimen were extracted with 12% trichloroacetic acid (4°C) and neutralized. Denatured protein content from each biopsy was determined by the method of Lowery and associates [5]. Levels of myocardial adenosine triphosphate (ATP), adenosine diphosphate (ADP), adenodine monophosphate (AMP), adenosine, inosine, hypoxanthine, xanthine, and the oxidized form of nicotinamide adenine dinucleotide (NAD) were determined with high-performance liquid chromatography and expressed as nanomoles per milligram protein [6].
Immune Western blotting
To determine the expression of heat-shock protein 70, the preischemic myocardial biopsy was homogenized in lysis buffer containing 5% SDS, 1% mercaptoethanol, and 0.1% p-methyl sulphonylflouride, and centrifuged for 5 minutes at 12000g. After protein determination, 10 µg of protein sample was loaded on a 45 SDS-polyacrylamide stocking gel for electrophoresis on a 12% separating gel, as described by Laemmli [7]. After SDS-PAGE, the gel was transferred to nitrocellulose membrane (Bio-Rad Laboratories, Los Angeles, CA) and then blocked with nonfat dry milk. The membranes were incubated with a mouse monoclonal antibody (1:1,000 dilution) then crossreacted with heat shock protein 70 (HSP70) (Stressgen Biotechnologies Corp, Victoria, Canada). The membrane was developed with enhanced chemoilluminscence and exposed to x-ray film for an appropriate time period [7].
Determination of regional myocardial blood flow (RMBF)
Measuring the RMBF with nonradioactive multiple-colored microspheres has been described elsewhere [8]. Multiple-colored microspheres (E-Z Trac, Los Angeles, CA) were used. They measured 11.9 ± 1.9 µm and numbered 2 x 107 microspheres/mL. A 0.4-mL amount of the stock suspension was diluted in 10 mL of 5% glucose solution and injected into the left atrium through a multiple-side-holed catheter passing through the left atrial appendage. Reference blood was withdrawn from the descending thoracic aorta by a plastic catheter passing through the nonoperated side femoral artery and advanced proximally to the descending aorta. The other side of the catheter was connected to a withdrawal pump (Model 915; Harvard Apparatus, Millis, MA). Blood was withdrawn at a rate of 9 mL/min starting 30 seconds before injecting the microspheres and ending 60 seconds after. At the end of the experiment, the heart was removed and cut transversely into 1-cm-thick slices; these slices were used to measure the infarct size and the RMBF alternately. Slices used to measure RMBF were divided into 12 sections; each section was further divided into an outer and an inner layer to yield 24 myocardial segments. Microspheres were extracted from the myocardial tissues by chemical and enzymatic digestion of the tissues; the myocardial segment was weighed, finely minced, and boiled with 2M potassium hydroxide (KOH). After cooling, it was centrifuged with Tissue Digestive Reagent II (E-Z Trac) until microspheres settled in the bottom of the tube. Microspheres were extracted from the blood in a similar way, except that blood was boiled with 2M NaOH. At this stage, RMBF was determined by one of two ways. The first method involved counting the microspheres by putting 0.05 mL of the tissue sediment and 0.10 mL of the blood sediment, respectively, in a Fucher-Rosenthal hemocytometer and counting under a light microscope at 200 magnification. RMBF can be calculated from the equation: RMBF (mL/min/g) = (Cm/Cr) x K x Qr/Wt. Cm represents the microsphere count in the myocardial sample; Qr is the withdrawal rate of the arterial blood; Cr is the number of microspheres in blood; K is the correction factor, which is 0.05/0.10 = 0.5 in this experiment; and Wt is the weight of the myocardial sample. The change of RMBF after ischemia was expressed as a percentage of postischemic RMBF to the preischemic value [8]. The second method used colored microspheres that were quantified by their dye content. The dye was extracted from the microspheres by adding N, N-dimethylformamide (DMF) (Sigma, St. Louis, MO). The dye content was measured by spectrophotometry at a wavelength 672 ± 10 nm for the blue dye and 530 ± 10 nm for the red dye [9]. The control we used was a nonmicrosphere-treated myocardial tissue that was digested, washed, and centrifuged in the same way with the addition of the same dose of DMF. Change of RMBF after ischemia was calculated as the percentage of the color of microspheres used after ischemia of the amount of color of microspheres used before ischemia. We have employed both techniques of measuring RMBF, and calculated the mean percentage ± standard error to confirm each other and delete any bias (see Comment).
Experimental design
In the acute study, all the animals were anesthetized, ventilated, and the heart function was monitored and measured with the Millar transducers and the crystals for assessing the baseline LV function. Animals were heparinized, then 0.4 mL of blue microspheres dissolved in 10 mL 5% glucose was injected in the left atrium through a multiple-holed catheter passing through the left atrial appendage. Reference atrial blood was withdrawn through a descending aortic catheter introduced through the nonoperated femoral artery; blood was withdrawn at a rate of 9 mL/min for 90 seconds, beginning 30 seconds before the injection of the microspheres. Preischemic transmural biopsy was taken to determine the baseline of purines and to check for the heat-shock protein 70. The LAD artery was occluded below the origin of the major diagonal branches for 90 minutes at 37°C followed by reperfusion for 4 hours. Twenty minutes after ischemia, the same dose of the red microspheres was injected and the reference blood was withdrawn. A transmural biopsy was taken every 30 minutes after ischemia to check for the rate of ATP depletion. After 4 hours of reperfusion, the LAD artery was religated, the heart was cut, and 10 mL Evans blue dye 1% was injected in the right and left coronary arteries. The heart was cut transversely into 1-cm-thick slices to measure the infarct size and the RMBF alternately. Slices chosen to measure the infarct size were stained with 1% triphenyl tetrazolium chloride (TCC) for 20 minutes, and then 10% formalin was added. The percentage of the infarct size to the area at risk, and the percentage of the area at risk to the LV, were measured in a computer program in our laboratory. Slices chosen to measure RMBF were rinsed thoroughly with distilled water and saved to be used later.
Statistical analysis
Data were presented as mean plus or minus the standard error of the mean. Sequential measurements were compared with repeated measure analysis of variance with GraphPad Prism software (Grand Pad Software Inc, San Diego, CA). Differences were considered significant if p < 0.05.
|
Results
|
|---|
Echocardiographic data
M-mode and two-dimensional echocardiography of the minor axis of the left ventricle showed no significant difference between the two groups before creating the fistula (data not shown).
Four weeks postoperatively, there was a significant difference between the two groups, as shown in Table 1. Stroke volume was 48.67 ± 7.66 mL in the fistula group and 37.12 ± 3.40 mL in the sham group (p = 0.07; not significant [NS]). Cardiac output was significantly increased in the fistula group (6.46 ± 0.54 L/min vs 3.73 ± 0.54 L/min in the sham group; p = 0.001). Heart rate was also significantly different between the two groups (138.17 ± 8.24 beats/min in the fistula group vs 102.33 ± 3.51 in the sham group; p = 0.003).
The end-diastolic volume was 80.66 ± 7.15 mL in the fistula group versus 63.94 ± 4.72 mL in the sham group (p = 0.07; NS). There has been no concomitant increase in the end-diastolic intraventricular diameter (4.17 ± 0.2 cm in the fistula group vs 4.13 = 0.16 cm in the sham group), which indicates an increase in the end-diastolic pressure in the fistula group. However, the left ventricular function was preserved as the ejection fraction was 0.60 ± 0.05 in the fistula group versus 0.57 ± 0.02 in the sham group (p = 0.61; NS), and the fraction shortening was 31.95 ± 2.72 in the fistula group versus 30.32 ± 0.88 in the sham group (p = 0.35; NS).
Intraoperative left ventricular performance
Table 2 depicts left ventricular performance in the sham and AVF groups. There was a significant increase in the left ventricular end-diastolic pressure in the fistula group (19.4 ± 6.30 mm Hg vs 5.91 ± 1.01 mm Hg in the sham group; p = 0.02). The systolic carotid artery pressure was 94.80 ± 3.09 mm Hg in the fistula group versus 110.07 ± 1.80 mm Hg in the sham group (p = 0.03). The arterial diastolic pressure showed a significant decrease in the fistula group (62.19 ± 3.56 mm Hg vs 85.73 ± 1.81 mm Hg in the sham group; p = 0.04). The mean arterial pressure was 77.91 ± 1.66 mm Hg versus 92.69 ± 1.96 mm Hg in the sham group (p = 0.03). Figure 1 shows a significant improvement in the preischemic load-independent left ventricular performance (SW/EDD) in the fistula group (83.7 ± 7.8 dyn/cm2 x 103 vs 64.7 ± 7.2 dyn/cm2 x 103 in the sham group; p = 0.03).
Depletion of ATP during ischemia
Figure 2 shows that 30 minutes of LAD artery occlusion at 37°C caused 26.7% depletion of ATP from the preischemic value in the fistula group versus 39.1% in the sham group. However, 60 and 90 minutes of LAD artery occlusion caused a loss of 45.1% and 57% of the preischemic ATP values, respectively, in the fistula group versus 76.3% and 88.3% loss, respectively, in the sham group (p = 0.02).
Expression of heat-shock protein
Immunoblot analysis and densitometry revealed that inducible heat-shock protein 70 was not expressed in the fistula or in the sham myocardial biopsies.
Infarct size
After LAD artery occlusion for 90 minutes and reperfusion for 4 hours, there was a significant decrease in the infarct size (measured as a percentage of the area at risk) in the fistula group (9.19 ± 2.02% vs 28.44 ± 5.28% in the sham group; p = 0.02), and the area at risk as a percentage of the left ventricle was 38.30 ± 8.28% in the fistula group versus 42.80 ± 2.88% in the sham group (p = 0.07 NS) (Fig 3).
View larger version (16K):
[in this window]
[in a new window]
|
Fig 3. The percentage of the infarct size of the risk area, and the risk area of the left ventricle in the fistula and sham groups.
|
|
Regional myocardial blood flow
In each myocardial segment, whether from the ischemic or nonischemic region, the percentage of the blood flow after regional ischemia (measured from the number of the blue microspheres or the amount of the blue dye) of the amount of blood flow before the regional ischemia (measured as the number of the red microspheres or the amount of the red dye) was calculated. The mean values (± standard error) were calculated for all the outer segments and all the inner segments (from the ischemic plus nonischemic regions), to give an idea about the effect of regional ischemia on the blood flow in the whole subepicardium and the whole subendocardium in both groups. In the fistula group, the mean percentage (blood flow after LAD ligation/blood flow before LAD ligation) was significantly higher in the outer and inner layers compared with the corresponding values of the sham group (p value between groups < 0.05) (Fig 4).
View larger version (18K):
[in this window]
[in a new window]
|
Fig 4. Regional myocardial blood flow in the subepicardium and subendocardium in the fistula and sham groups during LAD occlusion (described as percentages of the preischemic values).
|
|
|
Comment
|
|---|
AVF constitute the most powerful stimuli to collateral development. The extent of the collateral bed far exceeds that associated with an atherosclerotic occlusion of a comparable artery. The mechanism of the development of the collaterals is not well understood, however, some investigators have explained it as a function of increased velocity of blood flow [10].
The rationale of this study was to investigate whether the hyperdynamic circulation induced by AVF has a myocardial protective effect in the form of a decrease in the infarct size in the ischemia/reperfusion animal model and/or a decrease in the ATP depletion during ischemia. The study aimed, also, to investigate what mechanism of protection was employed, if any. Does the increased velocity of the blood flow and/or increased blood volume induced by AVF develop collaterals in the coronary circulation as in the peripheral circulation and re-distribute the myocardial blood flow to limit the infarct size if one vessel if ligated? Does the increased venous return and/or heart rate stress the myocardium and produce the protective heat-shock protein 70?
Hemodynamic data
Shunting of blood from the high-pressure artery to the low-pressure vein causes the local and systemic hemodynamic changes of AVF. AVF cause a decrease in the total peripheral resistance and, consequently, the arterial blood pressure. Systolic arterial pressure is maintained within normal limits by the increase in the cardiac output, which results from the increase in the stroke volume and/or heart rate; therefore, the diastolic pressure is that which is decreased, causing a wide pulse pressure. In the presence of large chronic fistulae, left ventricular end-diastolic pressure gradually rises until the heart fails [11]. Cardiac failure rapidly develops when there is a massive leak from the arterial to the venous side of the circulation; this is particularly true with large aorto-caval fistulae, which are highly fatal. On the other hand, failure may never occur or may be delayed for many years if the fistula is small. In human beings, the situation is more complex; children with healthy hearts tolerate the increased circulatory overload for prolonged periods without lapsing to cardiac failure. Infants with massive congenital fistulae of the brain or liver are an exception to this rule, as they may cause life-threatening cardiac failure, necessitating early operative intervention. In adults, the development of cardiac failure depends not only on the site and size of the fistula, but also on the presence of preexisting coronary or myocardial disease [12]. Increased stroke volume alone accounts for 80% to 90% of the rise in cardiac output associated with an acute or chronic AVF. The elevated stroke volume has been attributed to the Frank-Starling mechanism, which is initiated by a slight rise in arterial pressure; it may also reflect an increase in myocardial contractility that develops in response to increased sympathetic adrenergic outflow and elevated levels of circulating catecholamines [13]. In our study, M-mode and two-dimensional echocardiography 4 weeks after making the fistula, and the hemodynamic study during the acute surgery, showed a significant increase in the cardiac output, heart rate, end-diastolic pressure, wall thickness, and the systolic slope (SW/EDD) in the fistula group compared with the sham group. These studies also, showed a significant decrease in the arterial diastolic pressure in the fistula group compared with the sham group.
Myocardial preconditioning and arteriovenous fistula
This study showed, for the first time, elements of myocardial protection in the form of a significant decrease in the infarct size in the ischemia/reperfusion animal model of AVF. Also, a significant decrease in the rate of ATP depletion during ischemia in the fistula group was found without expression of the protective heat-shock protein 70.
Measurement of regional myocardial blood flow
In 1969, Domenech and associates [14] first validated the use of radioactive microspheres for the measurement of RMBF. However, because of the precautionary measures needed to minimize radiation exposure, use of radioactive microspheres is restricted to specially licensed laboratories. To avoid the hazards of these limitations, Hale and associates [15] proposed a method of measuring RMBF with nonradioactive colored microspheres. This technique depends on the tissue digestion by a combination of enzymatic and chemical methods, trapping the microspheres within a given sample and then counting them in a hemocytometer using light microscopy [15]. However, the limitations to the use of this method were summarized in the difficulty to differentiate and count different colored microspheres, and that it is time consuming because of the tedious counting of individual microspheres [9]. Kowallik and associates [9] developed a technique of measuring the RMBF by extracting the dye from the microspheres and quantitative estimate of the dye using the spectrophotometry under specific wavelength. We have employed both techniques to verify each other.
Normally, the subendocardium exhibits greater oxygen extraction, coronary blood flow, and oxygen consumption than the subepicardium [16]. We calculated the percentage of the RMBF after ligating the LAD artery to the preischemic values in all the outer and all the inner myocardial segments in both animal groups. The mean percentage was found to be significantly higher in both the outer and inner segments in the fistula group compared with the sham group; ie, both the whole subendocardium and the whole subepicardium were better perfused during regional ischemia than the corresponding regions of the sham group. This indicates the presence of good collaterals in the coronary circulation in the fistula group: enough to partially perfuse the ischemic area and decrease the infarct size and the rate of ATP depletion during ischemia. Therefore, the development of collaterals in the fistula group can explain the cardioprotective effect of the AVF even without the expression of inducible heat-shock protein 70.
Blood volume is often increased in the presence of a chronic arteriovenous fistula. In one series, the increase ranged from 200 to more than 1,000 mL/m2 of body surface [17]. The excess blood is accommodated in all parts of the expanded fistulous circuit, including proximal arteries and veins, cardiac chambers, central veins, and collateral vessels. Further investigations are required to investigate whether the increased blood volume is the only factor employed in increasing the myocardial vascularity, or if other mediators help to open collaterals in the coronary circulation.
During pregnancy, which is another form of the hyperdynamic circulation, some investigators tried to investigate the cause of the maternal systemic vasodilation and the low placental resistance to blood flow during pregnancy. Sladek and associates [18] claimed these changes were a result of the increased nitric oxide during pregnancy. They reported that the uterine arteries had increased nitric oxide synthase activity, protein expression, and 3', 5' cyclic monophosphate production during pregnancy. However, a controversy still exists over the placental level of nitric oxide synthase during pre-eclampsia, which was proved to be within normal, not decreased, levels [19]. Thanda and associates [20] reported that nitric oxide synthase activity increased in early pregnancy in the rat placenta, decreased with the progression of pregnancy, and became the lowest in the term placenta. Therefore, if nitric oxide is involved in the hyperdynamic circulation induced by early pregnancy, further investigations are required to determine the role of nitric oxide and/or other mediators in the arteriovenous fistula and other causes of hyperdynamic circulation.
|
Acknowledgments
|
|---|
We acknowledge the help of Dr Michel S. Hess, Professor of Cardiology, Medical College of Virginia, for his advice throughout the work. Supported in part by National Institutes of Health Grant HL5-1090 (ASA), and National Institutes of Health Training Grant 07537 (MSH).
|
References
|
|---|
-
Binak K., Regan T.J., Christensen R.C. Arteriovenous fistula; hemodynamic effect of occlusion and exercise. Am Heart J 1960;60:495-502.
-
Sabiston D.C., Jr, Therlin E.O., Gregg D.E. Physiological studies in experimental high output cardiac failure produced by aorto-caval fistula. Surg Forum 1955;6:233-241.
-
Nakano J. Effect of arteriovenous fistula on the cardiovascular dynamics. Jpn Heart J 1971;12:392-401.[Medline]
-
Spratt J.A., Tyson G.S., Glower D.D., et al. The end-systolic pressure volume relationship in conscious dogs. Circulation 1987;75:1295-1309.[Abstract/Free Full Text]
-
Lowery O.H., Rosenbrough N.J., Farr A.L., Randall R.J. Protein measurement with Folin phenol reagent. J Biol Chem 1951;193:265-275.[Free Full Text]
-
Abd-Elfattah A.S., Wechsler A.S. Superiority of HPLC to assay for enzymes regulating adenine nucleosidase pool intermediates metabolism. J Liq Chrom 1987;10:2653-2694.
-
Laemmli U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-685.[Medline]
-
Ishikawa K., Ogawa I., Shimzu M., et al. The importance of good endocardial reflow immediately after reperfusion for myocardial salvage in dogs. Jpn Circ J 1991;55:983-993.[Medline]
-
Kowallik P., Schulz R., Guth B.D., et al. Measurement of regional myocardial blood flow with multiple colored microspheres. Circulation 1991;83:974-982.[Abstract/Free Full Text]
-
Lie M., Sejested O.M., Kiil F. Local regulation of vascular cross section during changes in femoral arterial blood flow in dogs. Circ Res 1970;27:727-734.[Abstract/Free Full Text]
-
Ingebrigtsen R., Krag J., Lerand S. Velocity and flow of blood in the proximal femoral artery proximal to an experimental arteriovenous fistula. Acta Chir Scand 1962;124:45-53.
-
Dorney E.R. Peripheral arteriovenous fistula of fifty-seven years duration with refractory heart failure. Am Heart J 1957;54:778-783.
-
Nakano J., Deschryver C. Effects of arteriovenous fistula on systemic and pulmonary circulation. Am J Phys 1964;207:1319-1326.
-
Domenech R.J., Hoffman J.I.E., Nobel M.I., Saunders K.B., Henson J.R., Subijanto S. Total and regional coronary blood flow measured by radioactive microspheres in conscious and anaesthetized dogs. Circ Res 1969;25:581-596.[Abstract/Free Full Text]
-
Hale S.L., Alker K.J., Kloner R.A. Evaluation of non-radioactive, colored microspheres for measurement of regional myocardial blood flow in dogs. Circulation 1988;78:428-434.[Abstract/Free Full Text]
-
Briden K.L., Weiss H.R. Effect of moderate arteriovenous shunt on regional extraction, blood flow, and oxygen consumption in the dog heart. Cardiovasc Res 1981;15:206-212.[Medline]
-
Warren J., Elkins D.C., Nickerson J.L. The blood volume in patients with arteriovenous fistulae. J Clin Invest 1951;39:220-227.
-
Sladek S.M., Mogness R.R., Conradk P. Nitric oxide and pregnancy. Am J Physiol 1997;272:441-465.
-
Weiner C.P., Knowles R.G., Moncada S. Induction of nitric oxide synthases early in pregnancy. Am J Obst Gyn 1994;171:838-843.[Medline]
-
Thanda M., Saheki S., Kitagawa H., Yano J., Masturo S. Gestational changes in nitric oxide synthase activity in the rat placenta. J Obst Gyn Res 1996;22:267-273.
Accepted for publication December 26, 1998.
Top
Abstract
Introduction
