Ann Thorac Surg 2003;75:178-183
© 2003 The Society of Thoracic Surgeons
Original article: cardiovascular
Smooth muscle cell hypertrophy of renal cortex arteries with chronic continuous flow left ventricular assist
Shin’ichiro Kihara, MD, PhDa*,
Kenneth N. Litwak, DVM, PhDa,
Larry Nichols, MDb,
Philip Litwak, DVM, PhDa,
Marina V. Kameneva, PhDa,
Zhongjun Wu, PhDa,
Robert L. Kormos, MDa,
Bartley P. Griffith, MDc
a Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
b Department of Pathology, University of Pittsburgh, Pittsburgh, Pennsylvania, USA
c Department of Surgery, University of Maryland, Baltimore, Maryland, USA
Accepted for publication July 23, 2002.
* Address reprint requests to Dr Kihara, McGowan Institute for Regenerative Medicine, University of Pittsburgh, 3025 E Carson St, Pittsburgh, PA 15203, USA.
e-mail: kiharas{at}msx.upmc.edu
 |
Abstract
|
|---|
BACKGROUND: Pathophysiology of long-term continuous flow left ventricular assist is not well described. With many of these devices becoming available, it is important to examine for possible pathologic effects. In this study we examined the relationship between diminished pulsatility and pathologic changes in renal cortical arteries.
METHODS: Twenty-nine calves were implanted with various continuous flow left ventricular assist systems in a left ventricle-descending thoracic aorta bypass configuration. Pulsatility was quantified by pulse pressure and pulsatility index. Pathologic changes of the renal cortex arteries were described and evaluated by medial thickness, medial/vascular cross-sectional area ratio, and smooth muscle cell count, to quantify hypertrophy or hyperplasia. Seven calves, which underwent a sham-implant, were used as controls.
RESULTS: Systolic arterial pressure, pulse pressure, and pulsatility index were significantly lower and diastolic pressure was significantly higher than before implant in pump-implanted animals. Twenty-three of 29 pump-implanted calves (79.3%) had medial smooth muscle cell hypertrophy in renal cortex arteries, whereas none of sham-implanted calves had any abnormal lesions. When the pump-implanted calves were grouped according to the presence of smooth muscle cell hypertrophy, there was a clear trend toward lower pump flow rate in calves with lesions. Renal function was within the normal range in all calves.
CONCLUSIONS: There appears to be a relationship between smooth muscle cell hypertrophy in renal cortex arteries and continuous flow left ventricular assist. Furthermore, although the pathologic changes are likely multifactorial, these lesions appear to be related to lower pump assist rates.
|
Introduction
|
|---|
With the severe shortage of organ donors, implantable left ventricular assist systems (LVAS) are becoming a viable clinical alternative to heart transplantation, not only as a bridge to transplantation, but also as long-term implants. Although clinical trials of continuous flow LVAS have started [1–3], the pathophysiologic effects of these systems have not been well described. Diminished pulsatility produced by continuous flow LVAS is physiologically abnormal and may have effects on arterial morphology and vasoreactivity [4–6]. Nishimura and colleagues [5] have reported pathologic changes of aortic wall properties with long-term continuous flow LVAS. They describe smooth muscle cell (SMC) atrophy in the tunica media of the aortic wall, possibly caused by decreased mechanical stress. The effect of continuous flow LVAS support on medium-to-small sized arteries or arterioles needs to be reported. This study provides the evidence of pathologic changes in the renal cortical arteries after chronic, continuous flow LVAS implantation based on retrospective review. Furthermore, it demonstrates a possible relationship between pathologic changes and the hemodynamic characteristics of the pump flow.
|
Material and methods
|
|---|
Animals and device description
Twenty-nine male Jersey breed calves (55 to 87 kg) were implanted with a continuous flow LVAS through a left thoracotomy in left ventricle-descending thoracic aorta bypass fashion (implant). There were five pump types, both axial and centrifugal. Seven calves, which underwent a sham implant, were used as controls to exclude the effects of operation and postsurgical responses (control).
Implant procedure
Anesthesia induction was with methohexital (10 mg/kg intravenously) after atropine premedication (30 to 45 mg subcutaneously). The animals were intubated endotracheally for mechanical ventilation and anesthesia was maintained with isoflurane and oxygen. A left thoracotomy through the fifth intercostal space was used for implant. For implant calves, the device was implanted in a left ventricle-descending thoracic aorta bypass configuration without cardiopulmonary bypass. The outflow graft was sewn to the descending thoracic aorta after systemic heparinization (1.5 mg/kg). Then, pledgetted braided Dacron sutures were placed around the left ventricular apex. The left ventricular apex was cored with a coring knife and the inflow cannula was inserted and fixed with the sutures. Ultrasonic flow probes (T206 series, Transonic Systems Inc, Ithaca, NY) were set on the pump outflow graft and pulmonary artery to provide measure of the pump flow and total cardiac output. Arterial pressure was obtained from a fluid-filled catheter implanted in the left carotid artery.
Control calves were also underwent thoracotomy. A short (2- to 4-cm) piece of Dacron graft with a blind end was anastomosed onto the descending thoracic aorta. The left ventricular apex was also cored in the same fashion as pump implant calves followed by being closed with two pledgetted Dacron sutures.
Pump operation mode and anticoagulation regime
For implant animals, all pumps were operated at a fixed rotational pump speed with a goal of keeping pump flow rate more than 60% of total cardiac output. All animals were given oral warfarin sodium daily starting 1 day before implant. The calves were given intravenous heparin sodium to keep activated coagulation time 1.5 to 2.0 times normal for the first 24 to 48 hours after implant. The heparin was discontinued when the oral warfarin sodium became effective with an international normalized ratio between 2.5 to 3.5. Control calves were also treated with the same anticoagulation regime.
Pump and hemodynamic data collection
Arterial pressure, pump speed, pump flow rate, and pulmonary artery flow were continuously monitored. Hemodynamic and pump data were collected before implant and weekly thereafter, at about the same time of the day (early afternoon) with a data acquisition system (WINDAQ: DATAQ Instruments Inc, Akron, OH) with the calves standing up. For implant animals, the end point of the study depended on the protocol for each pump (30 to 222 days after operation). Control calves were sacrificed 30 or 180 days after operation.
Hemodynamic data analysis
All hemodynamic and pump data were collected every 5 milliseconds for 20 seconds. Data in every cardiac cycle were analyzed and expressed as the mean value ± standard deviation for the 20-second time period of each measure. Pulsatility was quantified by pulse pressure and pulsatility index (pulse pressure/mean arterial pressure). When there was regurgitant flow in the pump flow wave (defined as pump flow <0 L/min), percent regurgitant flow volume ratio (regurgitant flow volume/forward flow volume) and percent regurgitant flow time (time of regurgitant flow per cardiac cycle) was calculated. These are expressed as mean calculated values from the data in each cardiac cycle in the 20-second data strip.
Renal function and pathologic study
Renal function was evaluated by blood urea nitrogen and creatinine in blood chemistry twice a week. The calves were euthanized at the end of their study with an overdose of methohexital (2.5 g intravenously) and the kidneys were explanted immediately after sacrifice. The tissue samples were fixed with 10% buffered formalin and embedded in methyl methacrylate and transverse sections cut at 4 µm. Sections were stained with hematoxylin and eosin for light microscopy, photographed and transferred to a personal computer for analysis. Vascular diameter, lumen diameter, medial thickness, vascular cross-sectional area, medial cross-sectional area, and smooth muscle cell count of the renal cortex arteries were measured according to the method of Amann and colleagues [7]. These variables were measured on 30 to 50 randomly selected arterioles in the renal cortex at a final magnification of 40:1 using planimetry software (Scion Image Beta 4.02, Scion Corporation, Frederick, MD). Arterial diameters were calculated as the mean of two opposite measurements in the direction of the minimal diameter, because the minimal lumen diameter is less affected by the sectioning angle than any other direction. Cross-sectional areas were also measured at the same site as medial thickness measurement, minimizing the effect of sectioning. Ratios of medial/vascular cross-sectional area, medial/vascular diameter, and medial cross-sectional area/SMC count were calculated to avoid the bias of actual arterial size [7].
Statistical analysis
All data collected by the data acquisition system or planimetry software were converted to fit Microsoft Excel 98 (Microsoft, Redmond, WA) and StatView 4.5 (SAS Institute Inc, Cary, NC) files for statistical analysis. All values were expressed as mean ± standard deviation. The Student’s t test and 
2 test was applied to test statistical significance of difference and p less than 0.05 was regarded to be statistically significant.
Humane animal care
All animals received humane care in accordance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research as well as with the "Guide for the Care and Use of Laboratory Animals" published by the National Institutes of Health (NIH publication 85-23, revised 1985) and the guidelines determined by the Institutional Animal Care and Use Committee of the University of Pittsburgh.
|
Results
|
|---|
Pump and hemodynamic data
At operation, there was no difference in hemodynamic variables between groups. Hemodynamic characteristics after operation are shown in Table 1.
In implant animals, systolic arterial pressure was significantly lower and diastolic pressure was significantly higher than control animals, but there was no difference in mean arterial pressure. Pulse pressure and pulsatility index were also lower in implant animals.
Pathologic study
There were two major pathologic changes in the renal cortex arteries in implant animals that were not found in any of control animals (Fig 1).
The major pathologic findings were as follows:
- SMC hypertrophy of tunica media (Fig 2):
23 of 29 calves (79.3%) had thickened walls of the arteries in juxtaglomerular area caused by SMC hypertrophy. They were more densely packed compared to glomerular arteries of control calves (Fig 1). Many arteries had decreased lumen area in these arterioles.
- Inflammatory cell infiltration of the arteries and the interstitium (Fig 3): 23 of 29 calves (79.3%) had inflammatory cell infiltration involving arterial walls and surrounding interstitium. The inflammatory cells were mainly eosinophils and lymphocytes. This lesion was only present in those animals with SMC lesions. Although there were macroscopically pale areas on the renal surface, we could find no evidence of infection or infarction in any animal in this series.
View larger version (136K):
[in this window]
[in a new window]
|
Fig 1. Renal cortex arteriole of a control calf. No pathologic lesion is observed. (Hematoxylin and eosin stain, magnification x40.)
|
|
View larger version (168K):
[in this window]
[in a new window]
|
Fig 2. Medial smooth muscle cell hypertrophy was seen in implant animals. (Hematoxylin and eosin stain, magnification x40.)
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Fig 3. Interstitial infiltration was observed in 23 of 29 implant animals. No lesion was seen in control calves. (Hematoxylin and eosin stain, magnification x40.)
|
|
Quantification of the SMC hypertrophy
There was no difference between implants and controls in vascular diameter, vascular cross-sectional area, and medial SMC count (Table 2).
Medial thickness, medial cross-sectional area, medial/vascular diameter ratio, and medial/vascular cross-sectional area ratio were significantly greater in Implants. Medial cross-sectional area per SMC was significantly greater in Implants, although there was no difference in SMC count, suggesting the SMC proliferation was due to hypertrophy not hyperplasia.
Implant calves were then divided into two groups according to the presence of SMC hypertrophy. Medial/vascular cross-sectional area ratio was used for grouping the animals because this variable is least affected by actual vascular diameter or medial thickness. Animals with a ratio greater than 1 standard deviation over the mean for control animals were classified as having SMC hypertrophy. Hemodynamic variables were then compared between these groups (Table 3).
There were no significant differences in assist duration, mean, systolic or diastolic blood pressure between the two groups, but pulsatile pressure (p < 0.01) and pulsatility index (p < 0.01) were significantly higher in implant calves with SMC hypertrophy. Furthermore, there was a clean trend toward lower pump flow rate (p = 0.06) in animals with lesions. Overall incidence of pump regurgitant flow was higher in implant animals with SMC hypertrophy. The percent regurgitant volume ratio in implant animals with SMC hypertrophy was 13.9% ± 8.96% and percent regurgitant flow time ratio was 34.8% ± 2.26%
Renal function
Renal function was evaluated by blood chemistry twice a week. Blood urea nitrogen and serum creatinine level of Implant calves stayed within the normal range before and during this study and were not different from control calves.
|
Comment
|
|---|
Although clinical use of the continuous flow LVAS has started, few reports discuss long-term pathophysiologic consequence of continuous flow circulatory support [1–3]. In this study, we found SMC hypertrophy of the renal arteries with long-term continuous flow LVAS assistance. Pathologic changes of the aortic wall properties with continuous flow LVAS have been reported, suggesting that there is SMC atrophy in the tunica media of the aortic wall caused by diminished pulsatility, decreased systemic vascular resistance and contractility [4–6]. Although our results may seem incongruous from those described by Nishimura and colleagues [5], it should be remembered that those findings were in a large artery that is subjected to very different hemodynamic influences compared to the smaller arterioles.
Several causes of SMC proliferation or migration into intima of arteries have been described [8–10]. Generally, after an intimal injury occurred, there was SMC infiltration to the intima and proliferation as a healing process of the vessel wall. Previous reports described that there are many causes of intimal injury, for example, mechanical stress, ischemia, immune reaction, suggesting a multifactorial origin [11]. In support of early intimal injury, intimal hypertrophy was found in two calves implanted with continuous flow LVAS but terminated early due to pump failure (17 and 21 days after implant). Although certainly too small a number to draw any conclusions they are supportive of early intimal damage leading to a SMC response. Furthermore, as Litwak and colleagues [12] have reported that tissue perfusion in the lower half of the body in the calves with continuous flow LVAS decreased significantly after implant by 25% to 50% and remained at low levels after that although the total cardiac output was constant or even higher during a long-term period. Their data suggested that this sudden decrease of the tissue perfusion might be due to altered hemodynamics, resulting in malperfusion of major organs.
The question arises as to why 6 of 29 implant calves had no SMC hypertrophy of the renal cortex arteries. When implant calves were divided into two groups according to the presence of SMC hypertrophy, there was no significant difference in assist duration and arterial pressure, but pump-implanted animals with SMC hypertrophy had greater pulsatility and a trend toward lower pump flow rate compared to those without lesions. A larger pulsatility index is expected with a lower assist rate because the blood flow through the native aortic valve is greater; however, lower pump flow could make the pump more susceptible to regurgitant flow, as suggested by our results. Previously, we have demonstrated that this regurgitant flow may also be observed in the descending thoracic aorta distal to the outflow anastomosis when pump support is about 50%, although the overall pulsatility is almost normal [12]. These findings again suggest an underlying hemodynamic cause of the SMC hypertrophy, although much more work needs to be done to prove a firm connection.
The SMC hypertrophy in the tunica media seen in this study is similar to that described in renal arteries from hypertensive animals or patients [13–16]. These findings were seen with angiotensin II-elevated hypertension, such as renal hypertension, and occurred independently of the grade of hypertension [14]. Because this was a retrospective study, data from the renin-angiotensin-aldosterone system were not available for analysis; however, the renin-angiotensin-aldosterone system might be stimulated due to decreased renal blood flow, which may affect SMC hypertrophy in chronic phase. We intend to clarify this matter in further experiments.
Interstitial nephritis was only found in the Implant calves with lesions. The pathogenesis of this finding is unclear, but possible etiologies could include immunologically mediated reaction to the pump, medication, hemodynamic changes, or to the ischemia. All pumps were made of titanium, which is a well-known biocompatible material and commonly used for implantable medical devices. Furthermore, the same antibiotics and warfarin sodium were used for all animals, including controls.
There are several reports that this pathology may be caused by ischemia secondary to hypertension-induced vascular changes [17, 18]. Truong and colleagues [19] developed a sophisticated theory about the relationship between chronic renal ischemia and interstitial nephritis. They reported that the ischemic kidney displays interstitial chronic inflammatory infiltrates. Ischemia alone can cause a constellation of changes fulfilling the accepted features of chronic interstitial nephritis and may well be the final common pathway for chronic interstitial nephritis of diverse etiologies. This report suggests that the interstitial nephritis seen in implant calves with lesions is likely related to ischemia.
As stated before, the cause of SMC hypertrophy seen in this study is likely multifactorial, and could be due to any combination of hemodynamic, hormonal, and mechanical changes. In this study, we examined the hemodynamic changes. Furthermore, the mechanism of these pathologic findings could be different in the short term and the long term. Our theory is that the changes in the short term are mostly dependent on drastic changes in hemodynamics just before and after operation, and those in the long term may be strongly affected by hormonal changes including the renin-angiotensin-aldosterone system. This may be the largest animal study with continuous flow LVAS, however, even this is too small to perform the multivariate analysis necessary to clarify which factor is the most likely responsible for the lesions.
A major limitation to applying this model to the actual heart failure patient with LVAS is that these animals have hearts with normal cardiac function. Flow characteristics are totally different in severe heart failure patients, especially in the early postoperative period. Indeed, the conditions that we have shown would likely occur if there was restored cardiac function over time in patients with long-term assist or destination therapy.
Finally, it is important to note that there were no differences in this study between any of the five pumps. This suggests the findings of this study are related to continuous flow LVAS support as a whole, rather than a specific pump. However, with careful attention paid to pump control and renal function, our data suggest that development of these lesions may be avoided. Further research to clarify these issues will need to be done as patients begin to receive continuous flow LVAS for longer periods of time.
In conclusion, continuous flow left ventricular assist has a high incidence of SMC hypertrophy in renal cortex arteries. These pathologic changes are likely multifactorial; however, they may be partly due to initial ischemic injury caused by renal malperfusion associated with lower pump flow rate.
|
Acknowledgments
|
|---|
This work was supported in part by a grant from the National Institutes of Health, 1R01 HL64950–01. We express our gratitude to Dr Kristine Krotec, Mary J. Watach, and Lisa Gordon for their excellent technical assistance.
|
References
|
|---|
- Wieselthaler G.M., Schima H., Dworschak M., et al. First experiences with outpatient care of patients with implanted axial flow pumps. Artif Organs 2001;25:331-335.[Medline]
- Maher T.R., Butler K.C., Poirier V.L., Gernes D.B. Heartmate left ventricular assist devices: a multigeneration of implanted blood pumps. Artif Organs 2001;25:422-426.[Medline]
- Frazier OH, Myers TJ, Jarvik RK, et al. Research and development of an implantable, axial-flow left ventricular assist device: the Jarvik 2000 Heart. Ann Thorac Surg 2001;71:S125–32; S144–6
- Nishimura T., Tatsumi E., Nishinaka T., et al. Diminished vasoconstrictive function caused by long-term nonpulsatile left heart bypass. Artif Organs 1999;23:722-726.[Medline]
- Nishimura T., Tatsumi E., Taenaka Y., et al. Effects of long-term nonpulsatile left heart bypass on the mechanical properties of the aortic wall. Asaio J 1999;45:455-459.[Medline]
- Nishinaka T., Tatsumi E., Nishimura T., et al. Cardiac autonomic nervous function during long-term nonpulsatile left heart bypass. Artif Organs 1999;23:500-503.[Medline]
- Amann K., Gharehbaghi H., Stephen S., Mall G. Hypertrophy and hyperplasia of smooth muscle cells of small intramyocardial arteries in spontaneously hypertensive rats. Hypertension 1995;25:124-131.[Abstract/Free Full Text]
- Sinclair R.A., Antonovych T.T., Mostofi F.K. Renal proliferative arteriopathies and associated glomerular changes: a light and electron microscopic study. Hum Pathol 1976;7:565-588.[Medline]
- Ruggenenti P., Remuzzi G. Malignant vascular disease of the kidney: nature of the lesions, mediators of disease progression, and the case for bilateral nephrectomy. Am J Kidney Dis 1996;27:459-475.[Medline]
- Bochaton-Piallat M.L., Gabbiani F., Redard M., Desmouliere A., Gabbiani G. Apoptosis participates in cellularity regulation during rat aortic intimal thickening. Am J Pathol 1995;146:1059-1064.[Abstract]
- Ceiler D.L., Nelissen-Vrancken H.J., Smits J.F., De Mey J.G. Pressure but not angiotensin II-induced increases in wall mass or tone influences static and dynamic aortic mechanics. J Hypertens 1999;17:1109-1116.[Medline]
- Litwak KN, Kihara S, Kameneva MV, et al. Effects of continuous flow left ventricular assist device support on skin tissue perfusion and aortic hemodynamics. ASAIO J 2002 (in press)
- Bevan R.D., Eggena P., Hume W.R., Van Marthens E., Bevan J.A. Transient and persistent changes in rabbit blood vessels associated with maintained elevation in arterial pressure. Hypertension 1980;2:63-72.[Abstract/Free Full Text]
- Johnson R.J., Alpers C.E., Yoshimura A., et al. Renal injury from angiotensin II-mediated hypertension. Hypertension 1992;19:464-474.[Abstract/Free Full Text]
- Nag S., Robertson D.M., Dinsdale H.B. Morphological changes in spontaneously hypertensive rats. Acta Neuropathol 1980;52:27-34.[Medline]
- Wiener J., Loud A.V., Giacomelli F., Anversa P. Morphometric analysis of hypertension-induced hypertrophy of rat thoracic aorta. Am J Pathol 1977;88:619-633.[Abstract]
- Heptinstall RH. Essential hypertension. In Heptinstall RH, ed. Pathology of the kidney, 3rd ed. Boston/Toronto: Little, Brown and Company 1983:181–246
- Grone H.J., Helmchen U. Impairment and recovery of the clipped kidney in two kidney, one clip hypertensive rats during and after antihypertensive therapy. Lab Invest 1986;54:645-655.[Medline]
- Truong L.D., Farhood A., Tasby J., Gillum D. Experimental chronic renal ischemia: morphologic and immunologic studies. Kidney Int 1992;41:1676-1689.[Medline]
This article has been cited by other articles:
|
|
|
|
 
C. Ootaki, M. Yamashita, Y. Ootaki, K. Kamohara, S. Weber, R. S. Klatte, W. A. Smith, A. L. Massiello, S. N. Emancipator, L. A.R. Golding, et al.
Reduced Pulsatility Induces Periarteritis in Kidney: Role of the Local Renin-Angiotensin System
J. Thorac. Cardiovasc. Surg.,
July 1, 2008;
136(1):
150 - 158.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Schenk, S. Weber, D. Ing, V. Luangphakdy, R. S. Klatte, C. R. Flick, J.-F. Chen, M. W. Kopcak Jr, Y. Ootaki, K. Kamohara, et al.
MagScrew Total Artificial Heart In Vivo Performance Above 200 Beats Per Minute
Ann. Thorac. Surg.,
April 1, 2005;
79(4):
1378 - 1383.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
