- Email: [email protected]
Duplex scan and histologic assessment of acute renal injury in a kidney-kidney crosstalk swine experimental model
Duplex scan and histologic assessment of acute renal injury in a kidney-kidney crosstalk swine experimental model Anna Paula W. Baptista Sincos, MD,a Angela Mazzeo, MS,b Igor Rafael Sincos, MD, PhD, MBA,a Felipe Coelho Neto, MD, MS,a Nelson Wolosker, MD, PhD,a Ricardo Aun, MD, PhD,a Katia R. M. Leite, MD, PhD,c Vitoria Penido de Paula,b and Oskar G. Kaufmann, MD, PhD,a São Paulo, Brazil
ABSTRACT Objective: The objective of this study was to identify the effect of two left renal vasculature occlusion strategies on the duplex ultrasound-assessed rheology and histology of the contralateral kidney. Methods: Pigs were randomly assigned to one of two groups: left renal artery-only clamping (A group, n ¼ 8) or left renal artery and vein clamping (AV group, n ¼ 9). Bilateral renal parenchymal biopsy specimens were taken every 10 minutes for 90 minutes. Duplex ultrasound resistive index (RI) and pulsatility index (PI) were measured. Mixed models with normal distribution and first-order autoregressive correlation structure and generalized estimating equation models were used. Results are presented as adjusted means with standard errors, estimated proportions with standard errors, and line plots with 95% confidence intervals. Results: RI and PI increased in the nonischemic kidney. In A group animals, RI values increased significantly (P < .01) after 30 minutes of ischemia and PI increased significantly (P < .04) from 30 to 60 minutes of ischemia. The number of histologic abnormalities was higher in A group than in AV group biopsy specimens. The percentage of lesions increased significantly after 10 minutes in A group nonischemic kidneys (P < .02) and between 50 and 80 minutes in AV group nonischemic kidneys (P < .01). Conclusions: Nonischemic kidneys were acutely affected by contralateral ischemia. Their function was more adversely affected by unilateral renal artery occlusion with preserved renal vein patency (A group). (J Vasc Surg 2017;-:1-8.) Clinical Relevance: Our finding that the nonischemic kidney experienced acute effects stemming from contralateral ischemia may help surgeons prevent acute kidney injury during vascular and urologic procedures. Moreover, simultaneous artery and vein clamping appeared to affect the occurrence of renal crosstalk, which was reflected in worsening perfusion, elevated resistive index and pulsatility index values, and a tendency for thrombi to form in control kidneys. In addition, renal clamping protocol affected the occurrence of histologic lesions.
Acute kidney injury (AKI) secondary to surgery may progress to definitive disease, depending on the severity and duration of the injury.1-3 The “safe” duration of clamping is controversial, and few studies have examined the acute behavior of the contralateral kidney during renal ischemia using duplex ultrasound scan parameters. Experimental studies have shown that AKI may be associated with the onset of lesions in distant organs, such as the liver, lungs, and brain.4 The renal damage caused by hypoxia triggers an inflammatory process that may
From the Division of Vascular Surgery, Department of Surgery, Hospital Israelita Albert Einsteina; the Albert Einstein Israeli Institute of Teaching and Research and Albert Einstein Israeli College of Health Sciencesb; and the Division of Pathology, Department of Anatomy, Clinics Hospital, University of São Paulo.c Author conflict of interest: none. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: Anna Paula W. Baptista Sincos, MD, Division of Vascular Surgery, Department of Surgery, Hospital Israelita Albert Einstein, Rua Tabapuã 82,
Conj
1101,
São
Paulo,
SP
04533-010,
Brazil
(e-mail:
draanna@
endovascularsp.com.br; [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.06.118
provoke a systemic reaction, resulting in pathophysiologic sequelae in distant organs in a process called kidney-organ crosstalk.4,5 However, the occurrence of kidney-kidney crosstalk and the potential damage of this inflammatory cascade to the contralateral kidney remain poorly understood.6-8 Importantly, there is no consensus in the literature on the relationship between increased AKI risk and clamping protocol. In fact, the advantages of using either renal artery-only clamping, maintaining venous flow and some renal perfusion, or artery and vein clamping, with total flow obstruction and reduced risk of bleeding, are not clearly defined.9,10 The use of vascular Doppler ultrasound for the analysis of renal perfusion and graft control is well established.11,12 Measurements of the resistive index (RI), pulsatility index (PI), and Doppler spectral pattern provide valuable information about renal parenchymal vascularity.11,13 Thus, several experimental animal models have been developed to detect changes in renal perfusion using Doppler ultrasound.14 In this study, we serially measured RI and PI to determine the effects of ischemia on the contralateral kidney and compared the relative impact of renal artery alone vs renal artery and vein clamping on renal crosstalk. 1
2
Journal of Vascular Surgery
Baptista Sincos et al
---
METHODS The study was conducted between January and December 2016 at the Center for Experimentation and Training in Surgery of the Albert Einstein Jewish Hospital, São Paulo, Brazil, which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. The study was approved by the Albert Einstein Jewish Hospital Animal Care and Use Committee. A completed Animal Research: Reporting In Vivo Experiments guidelines checklist is included in the Appendix (online only). Seventeen 3- to 4-month-old female Large White pigs weighing between 29 and 35 kg were randomly assigned to either a renal artery and vein (AV group, n ¼ 9) or artery-only (A group, n ¼ 8) clamping protocol. Ischemia-induced AKI was experimentally reproduced through either left artery and vein or artery-only clamping, with the right kidney serving as a nonischemic control. Renal biopsies and ultrasound assessment were performed bilaterally at time 0 and every 10 minutes up to 90 minutes of ischemia. Duplex ultrasound scan analysis and biopsies were also performed during kidney reperfusion, at 120 and 150 minutes into the surgical procedure. Preoperative anesthetic procedures. All pigs were preanesthetized intramuscularly with ketamine (10.0 mg/kg) and midazolam (0.25 mg/kg). The marginal ear vein was catheterized with a 22-gauge BD Insyte catheter (BD Infusion Therapy Systems Inc, Sandy, Utah) for venous access. Anesthesia was induced with thiopental (7 mg/kg), and fluid replacement was performed with a maintenance crystalloid solution at 10 mL/kg/h. Crystalloid solution at 1 to 2 mL/kg/h was administered in bolus form whenever necessary to maintain a mean blood pressure of at least 70 mm Hg. Size 7.0 Portex endotracheal tubes (Smiths Medical, Ashford, UK) were used, and inhalational anesthesia was achieved with 2% isoflurane with the ventilator set at a tidal volume of 10 mL/kg. Anesthesia was maintained with fentanyl (2.5 mg/kg). Continuous invasive arterial pressure monitoring was accomplished by arterial catheterization of the right femoral artery. The mean blood pressure in each group is presented in Table I. Blood pressure was maintained at a minimum of 70 mm Hg during the procedure. Animals with nonresponsive hypotension were excluded. Systemic heparin (150 units/kg) was administered before hilar clamping. Even though the use of heparin in urologic surgery is controversial, systemic heparinization was employed because of the susceptibility of pigs to thrombi.15,16 The use of vasoactive agents was not allowed, and hypotension was managed with volume replacement. After surgery, all pigs were humanely euthanized by a lethal dose of thiopental and 15 to 30 mg/kg of 19.1% potassium chloride.
2017
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Experimental swine model of unilateral renal ischemia Take Home Message: In vivo physiologic and histologic data support that significant unilateral kidney ischemia causes changes in the contralateral nonischemic kidney. Recommendation: The nonischemic kidney may suffer damage by unclear mechanisms when one kidney is critically ischemic.
Table I. Adjusted mean values and standard errors for the mean blood pressure in each group Clamping protocol Time, minutes
A group (n ¼ 8)
AV group (n ¼ 9)
Basal
101.4 (2.3)a
94.9 (2.2)
10
94.7 (2.0)b
93.6 (1.9)
20
a,b
87.9 (2.5)
94.7 (2.4)
30
80.9 (2.4)a,b
89.7 (2.3) 80.4 (2.4)b
40
78.1 (2.6)
50
78.1 (2.6)b
79.9 (2.4)b
60
b
70.6 (2.8)b
a,b
66.4 (3.2)b
b
72.4 (3.4)b
b
74.2 (4.0)b
70 80 90
73.1 (3.0) 76.6 (3.4)
76.1 (3.6)
72.9 (4.2)
A group, Artery-only clamping; AV group, artery and vein clamping. The data represent the adjusted mean values and standard errors calculated using generalized estimating equation models. a Significant difference between groups. b Significant difference from baseline.
Operative technique. The kidneys were accessed through a midline laparotomy from xiphoid to pubis. The kidneys were biopsied bilaterally at time 0. Next, left renal hilar clamping was achieved using bulldog forceps, and serial parenchymal biopsies were performed in both kidneys. Biopsy samples were then gently collected using a No. 11 ophthalmic scalpel blade, with no significant parenchymal injury or major blood loss. Reperfusion was initiated after 90 minutes. Ultrasound data. Ultrasound examination was performed using a Philips HD7 ultrasound system (Koninklijke Philips N.V., Amsterdam, The Netherlands) and multifrequency linear transducer (7-11 MHz) in twodimensional, color, pulsed, and power Doppler modes. Image registration was performed at time 0 to rule out any previous alterations. Power Doppler ultrasound was used whenever necessary to confirm the absence of flow. Ultrasound analysis. The RI and PI were calculated automatically by the ultrasound software (High Q
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
3
-
Fig 1. Adjusted means and 95% confidence intervals for the resistive index (RI) in the right kidney of pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols.
Automated Doppler analysis). An RI < 0.7 and PI < 1.2 were considered normal. Histology. All specimens collected for histologic examination were embedded in paraffin and processed using standard protocols. The specimens were cut to 3-mm sections and stained with hematoxylin and eosin. The slides were photographed under an optical microscope (Nikon, Tokyo, Japan), and the following histopathologic changes were analyzed using ImageJ software (National Institutes of Health, Bethesda, Md): degenerative changes in tubular cells, presence of pigmented cylinders, vascular congestion and edema, interstitial neutrophil infiltration, and interstitial hemorrhage. Lesions were scored as absent (score 0) or present (score 1). Statistical analysis. Categorical data are expressed as absolute frequencies and percentages, and continuous data are expressed as means with standard deviations and minimum-maximum values. The inferential analysis of ultrasound data was performed using mixed models with normal distribution and first-order autoregressive correlation structure. Results are presented as adjusted means with standard errors and line plots with 95% confidence intervals. Histologic lesions were analyzed using generalized estimating equation models, and results are presented as estimated proportions with standard errors and 95% confidence intervals. The P values obtained from the multiple comparisons between time points and clamping protocols were adjusted by the sequential Bonferroni method. The Fisher exact test was used to investigate the association between clamping protocol and thrombosis occurrence. All analyses were performed using SPSS Statistics
for Windows version 19.0 (IBM Corp, Armonk, NY). A P value < .05 was considered significant.
RESULTS Ultrasound analysis. Mixed model analysis of the effect of clamping protocol on RI showed a significant interaction between group (clamping protocol) and time (P < .001). In the A group (artery-only clamping), RI values increased significantly in the nonischemic kidney (relative to baseline values) after 30 minutes of ischemia. Conversely, no significant differences in RI values were detected across time in the AV group (artery and vein clamping), despite the increase in mean RI values from baseline (0.562) to 90 minutes (0.639). The differences in RI values between group A and AV nonischemic kidneys were significant at 30, 40, and 50 minutes of ischemia (Fig 1). Mixed model analysis of the effect of clamping protocol on PI showed a significant interaction between group (clamping protocol) and time (P ¼ .014). In the A group, PI values of nonischemic kidneys increased significantly (relative to baseline values) after 30 and 60 minutes of ischemia. Conversely, no significant differences in PI values were detected across time in the AV group, despite the increase in mean PI values from baseline (0.808) to 90 minutes (1.164). The differences in PI values between group A and AV nonischemic kidneys were significant at 30, 40, and 50 minutes of ischemia (Fig 2). After 90 minutes of ischemia, the renal clamp was released and renal reperfusion initiated. Renal artery thrombi were observed in nonischemic kidneys from three animals in the A group (37.5%) but were not detected in AV pigs. There was no significant difference in thrombosis occurrence between A and AV
4
Journal of Vascular Surgery
Baptista Sincos et al
---
2017
Fig 2. Adjusted means and 95% confidence intervals for the pulsatility index (PI) in the right kidney of pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols. Table II. Estimated percentage of lesions in renal biopsy specimens from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols Clamping protocol AV group Time, minutes Baseline
Nonischemic kidney
A group Ischemic kidney
Nonischemic kidney
Ischemic kidney
10.0 (3.5)
7.5 (4.9)
10.0 (7.1)
17.5 (6.6)
10
7.5 (3.4)
12.5 (4.9)
20.0 (3.5)a
32.5 (4.9)a
20
20.0 (5.0)
25.0 (5.9)a
30.0 (6.1)a
42.5 (6.6)a
30
27.5 (7.0)
32.5 (4.9)a
42.5 (7.4)a
45.0 (5.9)a
a
a
47.5 (7.0)a
45.0 (6.8)
a
55.0 (5.9)a
45.0 (6.8)
a
50.0 (7.1)a
40 50
27.5 (6.1)
37.5 (5.5) a
37.5 (6.6)
a
a
30.0 (5.0)
a
60
37.5 (8.2)
70
35.0 (8.5)a
47.5 (7.0)a
50.0 (8.7)a
50.0 (5.0)a
80
a
a
a
57.5 (4.2)a
a
52.5 (8.6)a
90
37.5 (5.5)
30.0 (9.4)
40.0 (7.1)
40.0 (5.0)
45.0 (7.7)
a
47.5 (7.0)
52.5 (7.9)
45.0 (8.5)
The data represent the estimated proportions and standard errors calculated using generalized estimating equation models. a Significant difference from baseline.
nonischemic kidneys (P ¼ .082), and no thrombi were detected in ischemic kidneys. Histologic analysis. Sections from eight animals in each group were examined. Samples from one animal in the AV group were excluded because of tissue damage during processing and storage. Renal lesions by type. The qualitative analysis of ischemic kidneys revealed vascular congestion and edema after 10 minutes of ischemia that were more prevalent in A group animals. Interstitial neutrophil infiltration was observed in four of eight biopsy specimens in the A group and two of eight biopsy specimens in AV
animals. Degenerative changes were more frequent after 40 minutes of ischemia and more prevalent in the A group. Interstitial hemorrhage was also more frequent in A group animals. Although rare, pigmented cylinders, which are proteins released by tubular renal cells during acute necrosis, were the only histopathologic change more prevalent in the AV group. Vascular congestion and edema were detected in nonischemic kidneys of the A and AV groups after 10 minutes. Interstitial neutrophil infiltration was observed after 30 minutes in both groups. Degenerative changes were more frequent after 40 minutes of ischemia and more prevalent in the A group. Interstitial hemorrhage was also more frequent in the A group and was detected in 50% of
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
5
-
Fig 3. Estimated percentage of lesions and 95% confidence intervals in renal biopsy specimens from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols.
Table III. Estimated percentage of lesions in renal biopsy specimens of ischemic and control kidneys from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols regardless of clamping protocol Kidney Time, minutes
Nonischemic
Ischemic
Baseline
10.0 (4.0)
12.5 (4.3)
10
13.8 (2.9)a
22.5 (4.3)b
20
25.0 (4.1)b
33.8 (4.9)b
30
b
38.8 (4.1)b
b
42.5 (4.6)b
b
46.3 (4.9)b
b
b
40 50
35.0 (5.4)
33.8 (4.2)
37.5 (4.6)
60
41.3 (5.4)
70
42.5 (6.3)b
48.8 (4.3)b
80
b
51.3 (4.7)b
90
45.0 (5.2)
37.5 (6.6)
b
45.0 (5.2)
50.0 (5.6)
The data represent the estimated proportions and standard errors calculated using generalized estimating equation models. a Significant difference between nonischemic and ischemic kidneys. b Significant difference from baseline.
biopsy specimens after 70 minutes. Pigmented cylinders were not detected in the A group and were the only histopathologic change more prevalent in the AV group. Lesion biopsies. The quantitative analysis of the estimated percentage of lesions identified on histologic analysis is summarized in Table II and Fig 3. Ischemic kidney lesions by clamping protocol. The percentage of renal lesions increased significantly relative to baseline values after 20 minutes (25%) and 10 minutes (32.5%) in AV and A animals, respectively.
Nonischemic kidney lesions by clamping protocol. A significant increase in the percentage of lesions was also detected in nonischemic kidneys. The percentage of renal lesions increased significantly relative to baseline values only between 50 and 80 minutes (30%) in the AV group and after 10 minutes in A group animals. The percentage of lesions in renal biopsy specimens of ischemic and nonischemic kidneys increased significantly after 10 and 20 minutes, respectively, regardless of clamping protocol (Table III; Fig 4).
DISCUSSION In this study, we developed an experimental porcine model to study the effects of ischemia on the contralateral kidney and the effect of clamping protocol on renal crosstalk. Swine were used as the animal model because these animals are recognized as excellent biomedical models that closely replicate anatomic, physiologic, and immunologic conditions in humans.17 The duration of ischemia was set at 90 minutes in an attempt to more clearly demonstrate any possible differences between the two clamping protocols (artery-only and artery and vein clamping). We found a tendency of elevated RI and PI values in the nonischemic kidney during left hilar clamping, indicating that the renal perfusion changes experienced by the ischemic kidney were replicated in the contralateral kidney of animals in the artery-only (A) and artery and vein (AV) clamping groups. Mean RI and PI values were significantly higher than baseline values in the A group after 30 minutes of ischemia. This finding appears to reflect the increase in the number of histologic lesions in the A group after 10 to 20 minutes of ischemia. Thus,
6
Baptista Sincos et al
Journal of Vascular Surgery ---
2017
Fig 4. Estimated percentage of lesions and 95% confidence intervals in renal biopsy specimens of ischemic and nonischemic kidneys regardless of clamping protocol from pigs assigned to artery-only (A) and artery and vein (AV) clamping.
we believe that serial RI and PI measurements have predictive value to determine the onset of renal injury. In addition, thrombi were detected in three nonischemic renal arteries in the A group, which may also be indicative of remote organ crosstalk. These three thrombotic events were observed only after the 90 minutes of ischemia (reperfusion phase), when they affected the duplex ultrasound readings. The lack of difference for thrombosis rates between the A and AV groups (37.5% vs 0%) could be due to a type II statistical error. The remote organ dysfunction in AKI is induced by the release of humoral and cellular mediators into the bloodstream. Cytokines that are released during ischemia may play an important role in the pathogenesis of tissue injury. Tumor necrosis factor a is produced by tubular epithelial cells and activated leukocytes and induces the expression of adhesion molecules in endothelial cells, enhancing leukocyte infiltration. In addition, tumor necrosis factor a induces other inflammatory cytokines, such as interleukins 6, 8, and 10, which may affect other organ systems.18 This process is called organ crosstalk, a term derived from electronics denoting any signal or circuit that unintentionally affects another. Lane et al19 defined organ crosstalk in critical illness as “the effects of one malfunctioning organ on the function of another.” Several experimental studies have shown that renal ischemia-reperfusion-induced AKI can affect other organ systems.20 Fadillioglu et al21 found elevated levels of liver transaminases in rats that underwent warm kidney ischemia. Gardner et al4 found subtle histopathologic effects of AKI on liver, lung, brain, and kidney samples from pigs
using immunofluorescence. Because pigs can tolerate longer ischemia times,22 the biologically small effects observed in their study may be partially attributed to the short duration of renal ischemia (40 minutes). Histopathologic examination indicated a significant increase in the number of renal lesions of A group nonischemic kidneys after 10 to 20 minutes of ischemia but only after 50 to 60 minutes in AV group nonischemic kidneys. In addition, the number of lesions was higher in both A group ischemic and nonischemic kidneys between 10 and 80 minutes. This finding indicates that organ crosstalk can be affected by clamping protocol. In fact, in AV group animals, there appeared to be a renal protective effect in the impaired venous return as well as lack of cytokine release into the systemic circulation. Moreover, angiotensin II is released systemically from the ischemic kidney through the renal vein, upregulating angiotensin II production in the nonischemic kidney.7 Thus, because renal vein occlusion interrupts the angiotensin II pathway, this may explain why the AV group was associated with fewer flow and histologic effects on the contralateral, nonischemic kidney. In most studies showing necrosis after renal ischemia, kidney biopsy specimens were taken at least 24 hours after reperfusion.4 In this study, reperfusion biopsy specimens were taken only at 120 and 150 minutes because of the difficulty in keeping the animals alive without vasoactive drugs. After 150 minutes of surgery, on average, most animals were in an unresponsive, hypotensive state as a result of dehydration, hypothermia, and, possibly, anesthetic intolerance. However, the use of vasoactive agents was not allowed as it could interfere with the duplex ultrasound scan analysis. Thus, the short
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
7
-
period of reperfusion combined with the histologic technique used may have affected the diagnosis of dynamic mechanisms, such as necrosis and apoptosis. Bezerra et al14 developed an experimental model of renal ischemia injury in the collared peccary (Tayassu tajacu) and found a significantly higher number of lesions in ischemic kidneys than in nonischemic ones. In our study, the percentage of lesions was higher in ischemic kidneys but increased significantly over time in both ischemic and nonischemic kidneys. To the best of our knowledge, this is the first study to investigate the mechanism and acute effects of renal crosstalk using Doppler flowmetry analysis of the contralateral kidney. Even though no ultrasound signals would be generated during left renal clamping, Doppler ultrasound was used to ensure that no anomalous perfusion occurred in the ischemic kidney. During ischemia, there was a tendency for an increase in RI and PI values over time in the nonischemic kidney. The percentage of renal lesions increased less sharply in the AV group than in the A group over time in both ischemic and nonischemic kidneys. Thus, even though some studies propose using artery-only clamping to minimize the effects of ischemic injury23 as venous patency would lead to a minimal parenchymal perfusion, in our study we found a smaller tendency for renal lesions in animals from the AV group. Moreover, artery and vein clamping may minimize the risk of intraoperative bleeding and hemodynamic impairment.9,24 Renal Doppler ultrasound has been shown to have predictive value in AKI, even before the elevation in urinary nitrogen excretion and the change in urinary debt.11,12 The normal renal artery spectral Doppler waveform has a low RI that reflects the high perfusion of the renal bed.11 In this study, we serially measured RI and PI to detect worsening perfusion patterns over ischemia time. In the A group, RI and PI increased above normal values after 30 minutes of ischemia, and it is possible that the crosstalk between ischemic and contralateral kidneys started at that time. Even though the cause of renal dysfunction cannot be solely determined by an elevated RI, a normal RI is predictive of adequate renal function.13 Several studies have demonstrated that RI alterations on duplex ultrasound scan preceded the increase in serum creatinine concentration in the course of reversible acute renal failure, and they served as the basis for development of our experimental model.14,25-27 Limitations. The sample size was small, and despite its anatomic and immunologic similarities, the pig kidney does not completely replicate the biology of the human kidney.17 Moreover, for funding reasons, it was not possible to measure inflammatory cytokines, serum creatinine concentration, or glomerular function to demonstrate organ crosstalk.
Quantification of ischemia-induced changes of actin, phosphotyrosine, intercellular adhesion molecule 1, b-integrin, and urine biomarkers should be considered for further investigation.
CONCLUSIONS The nonischemic kidney experienced acute effects stemming from contralateral ischemia, and the hilar clamping protocol appeared to affect the occurrence of renal crosstalk, which was reflected in worsening perfusion, elevated RI and PI values, and a tendency for thrombi to form in nonischemic kidneys. The analysis of acute histologic findings and Doppler ultrasound parameters indicated that RI and PI measurements may have predictive value in AKI. We believe that intraoperative Doppler ultrasound can help guide surgeons in determining tolerable ischemia times and the probability of AKI onset. Nevertheless, further clinical and experimental research is required to fully understand ischemiareperfusion injury and renal crosstalk.
AUTHOR CONTRIBUTIONS Conception and design: AS, IS, FN, NW, RA, OK Analysis and interpretation: AS, IS, FN, OK, VP Data collection: AS, AM, KL, OK, VP Writing the article: AS, IS, FN, OK Critical revision of the article: AS, AM, IS, NW, RA, KL, OK, VP Final approval of the article: AS, AM, IS, FN, NW, RA, KL, OK, VP Statistical analysis: AS, IS, FN, KL, OK, VP Obtained funding: AS, AM, NW, RA, OK, VP Overall responsibility: AS
REFERENCES 1. Thompson RH, Frank I, Lohse CM, Saad IR, Fergany A, Zincke H, et al. The impact of ischemia time during open nephron sparing surgery on solitary kidneys: a multiinstitutional study. J Urol 2007;177:471-6. 2. Laven BA, Orvieto MA, Chuang MS, Ritch CR, Murray P, Harland RC, et al. Renal tolerance to prolonged warm ischemia time in a laparoscopic versus open surgery porcine model. J Urol 2004;172(Pt 1):2471-4. 3. Wahlberg E, DiMuzio PJ, Stoney RJ. Aortic clamping during elective operations for infrarenal disease: the influence of clamping time on renal function. J Vasc Surg 2002;36: 13-8. 4. Gardner DS, De Brot S, Dunford LJ, Grau-Roma L, Welham SJ, Fallman R, et al. Remote effects of acute kidney injury in a porcine model. Am J Physiol Renal Physiol 2016;310:259-71. 5. Doi K, Rabb H. Impact of acute kidney injury on distant organ function: recent findings and potential therapeutic targets. Kidney Int 2016;89:555-64. 6. Schmiedt CW, Mercurio A, Vandenplas M, McAnulty JF, Hurley DJ. Effects of renal autograft ischemic storage and reperfusion on intraoperative hemodynamic patterns and plasma renin concentrations in clinically normal cats undergoing renal autotransplantation and contralateral nephrectomy. Am J Vet Res 2010;71:1220-7.
8
Journal of Vascular Surgery
Baptista Sincos et al
---
7. Giani JF, Janjulia T, Taylor B, Bernstein EA, Shah K, Shen XZ, et al. Renal generation of angiotensin II and the pathogenesis of hypertension. Curr Hypertens Rep 2014;16:477. 8. Kobori H, Ozawa Y, Satou R, Katsurada A, Miyata K, Ohashi N, et al. Kidney-specific enhancement of ANG II stimulates endogenous intrarenal angiotensinogen in gene-targeted mice. Am J Physiol Renal Physiol 2007;293:F938-45. 9. Orvieto MA, Zorn KC, Mendiola F, Lyon MB, Mikhail AA, Gofrit ON, et al. Recovery of renal function after complete renal hilar versus artery alone clamping during open and laparoscopic surgery. J Urol 2007;177:2371-4. 10. Parekh DJ, Weinberg JM, Ercole B, Torkko KC, Hilton W, Bennett M, et al. Tolerance of the human kidney to isolated controlled ischemia. J Am Soc Nephrol 2013;24: 506-17. 11. Cano H, Castaneda DA, Patino N, Perez HC, Sanchez M, Lozano E, et al. Resistance index measured by Doppler ultrasound as a predictor of graft function after kidney transplantation. Transplant Proc 2014;46:2972-4. 12. Hansen KJ, Tribble RW, Reavis SW, Canzanello VJ, Craven TE, Plonk GW Jr, et al. Renal duplex sonography: evaluation of clinical utility. J Vasc Surg 1990;12:227-36. 13. Gao J, Ng A, Shih G, Goldstein M, Kapur S, Wang J, et al. Intrarenal color duplex ultrasonography: a window to vascular complications of renal transplants. J Ultrasound Med 2007;26:1403-18. 14. Bezerra Dde O, Feitosa ML, Almeida HM, Costa FA, Braga JF, Souza Fde A, et al. Collared pecary (Tayassu tajacu) as a new model of renal ischemic injury induced by clamping the renal artery. Acta Cir Bras 2014;29:560-72. 15. Sincos IR, da Silva ES, Belczak SQ, Baptista Sincos AP, de Lourdes Higuchi M, Gornati V, et al. Histologic analysis of stent graft oversizing in the thoracic aorta. J Vasc Surg 2013;58:1644-51.e4. 16. Sincos IR, Aun R, da Silva ES, Belczak S, Higuchi ML, Gornati VC, et al. Impact of stent-graft oversizing on the thoracic aorta: experimental study in a porcine model. J Endovasc Ther 2011;18:576-84. 17. Giraud S, Favreau F, Chatauret N, Thuillier R, Maiga S, Hauet T. Contribution of large pig for renal ischemiareperfusion and transplantation studies: the preclinical model. J Biomed Biotechnol 2011;1:532127. 18. Golab F, Kadkhodaee M, Zahmatkesh M, Hedayati M, Arab H, Schuster R, et al. Ischemic and non-ischemic
19.
20.
21.
22.
23.
24.
25.
26.
27.
2017
acute kidney injury cause hepatic damage. Kidney Int 2009;75:783-92. Lane K, Dixon JJ, MacPhee IA, Philips BJ. Renohepatic crosstalk: does acute kidney injury cause liver dysfunction? Nephrol Dial Transplant 2013;28:1634-47. Klein CL, Hoke TS, Fang WF, Altmann CJ, Douglas IS, Faubel S. Interleukin-6 mediates lung injury following ischemic acute kidney injury or bilateral nephrectomy. Kidney Int 2008;74:901-9. Fadillioglu E, Kurcer Z, Parlakpinar H, Iraz M, Cerbail G. Melatonin treatment against remote organ injury induced by renal ischemia reperfusion injury in diabetes mellitus. Arch Pharm Res 2008;31:705-12. Humphreys MR, Castle EP, Lohse CM, Sebo TJ, Leslie KO, Andrews PE. Renal ischemia time in laparoscopic surgery: an experimental study in a porcine model. Int J Urol 2009;16: 105-9. Tracy CR, Terrell JD, Francis RP, Wehner EF, Smith J, Litorja M, et al. Characterization of renal ischemia using DLP hyperspectral imaging: a pilot study comparing artery-only occlusion versus artery and vein occlusion. J Endourol 2010;24:321-5. Funahashi Y, Kato M, Yoshino Y, Fujita T, Sassa N, Gotoh M. Comparison of renal ischemic damage during laparoscopic partial nephrectomy with artery-vein and artery-only clamping. J Endourol 2014;28:306-11. Norris CS, Pfeiffer JS, Rittgers SE, Barnes RW. Noninvasive evaluation of renal artery stenosis and renovascular resistance. J Vasc Surg 1984;1:192-201. Luger-Hamer M, Barbiro-Michaely E, Sonn J, Mayevsky A. Renal viability evaluated by the multiprobe assembly: a unique tool for the assessment of renal ischemic injury. Nephron Clin Pract 2009;111:29-38. Yoon DY, Kim SH, Kim HD, Na DG, Goo JM, Choi HJ, et al. Doppler sonography in experimentally induced acute renal failure in rabbits: resistive index versus serum creatinine levels. Invest Radiol 1995;30:168-72.
Submitted Apr 3, 2017; accepted Jun 16, 2017.
Additional material for this article may be found online at www.jvascsurg.org.
Journal of Vascular Surgery Volume
-,
Number
-
APPENDIX (online only).
Baptista Sincos et al
8.e1
8.e2
Baptista Sincos et al
Journal of Vascular Surgery ---
2017
ABSTRACT Objective: The objective of this study was to identify the effect of two left renal vasculature occlusion strategies on the duplex ultrasound-assessed rheology and histology of the contralateral kidney. Methods: Pigs were randomly assigned to one of two groups: left renal artery-only clamping (A group, n ¼ 8) or left renal artery and vein clamping (AV group, n ¼ 9). Bilateral renal parenchymal biopsy specimens were taken every 10 minutes for 90 minutes. Duplex ultrasound resistive index (RI) and pulsatility index (PI) were measured. Mixed models with normal distribution and first-order autoregressive correlation structure and generalized estimating equation models were used. Results are presented as adjusted means with standard errors, estimated proportions with standard errors, and line plots with 95% confidence intervals. Results: RI and PI increased in the nonischemic kidney. In A group animals, RI values increased significantly (P < .01) after 30 minutes of ischemia and PI increased significantly (P < .04) from 30 to 60 minutes of ischemia. The number of histologic abnormalities was higher in A group than in AV group biopsy specimens. The percentage of lesions increased significantly after 10 minutes in A group nonischemic kidneys (P < .02) and between 50 and 80 minutes in AV group nonischemic kidneys (P < .01). Conclusions: Nonischemic kidneys were acutely affected by contralateral ischemia. Their function was more adversely affected by unilateral renal artery occlusion with preserved renal vein patency (A group). (J Vasc Surg 2017;-:1-8.) Clinical Relevance: Our finding that the nonischemic kidney experienced acute effects stemming from contralateral ischemia may help surgeons prevent acute kidney injury during vascular and urologic procedures. Moreover, simultaneous artery and vein clamping appeared to affect the occurrence of renal crosstalk, which was reflected in worsening perfusion, elevated resistive index and pulsatility index values, and a tendency for thrombi to form in control kidneys. In addition, renal clamping protocol affected the occurrence of histologic lesions.
Acute kidney injury (AKI) secondary to surgery may progress to definitive disease, depending on the severity and duration of the injury.1-3 The “safe” duration of clamping is controversial, and few studies have examined the acute behavior of the contralateral kidney during renal ischemia using duplex ultrasound scan parameters. Experimental studies have shown that AKI may be associated with the onset of lesions in distant organs, such as the liver, lungs, and brain.4 The renal damage caused by hypoxia triggers an inflammatory process that may
From the Division of Vascular Surgery, Department of Surgery, Hospital Israelita Albert Einsteina; the Albert Einstein Israeli Institute of Teaching and Research and Albert Einstein Israeli College of Health Sciencesb; and the Division of Pathology, Department of Anatomy, Clinics Hospital, University of São Paulo.c Author conflict of interest: none. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: Anna Paula W. Baptista Sincos, MD, Division of Vascular Surgery, Department of Surgery, Hospital Israelita Albert Einstein, Rua Tabapuã 82,
Conj
1101,
São
Paulo,
SP
04533-010,
Brazil
(e-mail:
draanna@
endovascularsp.com.br; [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2017 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2017.06.118
provoke a systemic reaction, resulting in pathophysiologic sequelae in distant organs in a process called kidney-organ crosstalk.4,5 However, the occurrence of kidney-kidney crosstalk and the potential damage of this inflammatory cascade to the contralateral kidney remain poorly understood.6-8 Importantly, there is no consensus in the literature on the relationship between increased AKI risk and clamping protocol. In fact, the advantages of using either renal artery-only clamping, maintaining venous flow and some renal perfusion, or artery and vein clamping, with total flow obstruction and reduced risk of bleeding, are not clearly defined.9,10 The use of vascular Doppler ultrasound for the analysis of renal perfusion and graft control is well established.11,12 Measurements of the resistive index (RI), pulsatility index (PI), and Doppler spectral pattern provide valuable information about renal parenchymal vascularity.11,13 Thus, several experimental animal models have been developed to detect changes in renal perfusion using Doppler ultrasound.14 In this study, we serially measured RI and PI to determine the effects of ischemia on the contralateral kidney and compared the relative impact of renal artery alone vs renal artery and vein clamping on renal crosstalk. 1
2
Journal of Vascular Surgery
Baptista Sincos et al
---
METHODS The study was conducted between January and December 2016 at the Center for Experimentation and Training in Surgery of the Albert Einstein Jewish Hospital, São Paulo, Brazil, which is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. The study was approved by the Albert Einstein Jewish Hospital Animal Care and Use Committee. A completed Animal Research: Reporting In Vivo Experiments guidelines checklist is included in the Appendix (online only). Seventeen 3- to 4-month-old female Large White pigs weighing between 29 and 35 kg were randomly assigned to either a renal artery and vein (AV group, n ¼ 9) or artery-only (A group, n ¼ 8) clamping protocol. Ischemia-induced AKI was experimentally reproduced through either left artery and vein or artery-only clamping, with the right kidney serving as a nonischemic control. Renal biopsies and ultrasound assessment were performed bilaterally at time 0 and every 10 minutes up to 90 minutes of ischemia. Duplex ultrasound scan analysis and biopsies were also performed during kidney reperfusion, at 120 and 150 minutes into the surgical procedure. Preoperative anesthetic procedures. All pigs were preanesthetized intramuscularly with ketamine (10.0 mg/kg) and midazolam (0.25 mg/kg). The marginal ear vein was catheterized with a 22-gauge BD Insyte catheter (BD Infusion Therapy Systems Inc, Sandy, Utah) for venous access. Anesthesia was induced with thiopental (7 mg/kg), and fluid replacement was performed with a maintenance crystalloid solution at 10 mL/kg/h. Crystalloid solution at 1 to 2 mL/kg/h was administered in bolus form whenever necessary to maintain a mean blood pressure of at least 70 mm Hg. Size 7.0 Portex endotracheal tubes (Smiths Medical, Ashford, UK) were used, and inhalational anesthesia was achieved with 2% isoflurane with the ventilator set at a tidal volume of 10 mL/kg. Anesthesia was maintained with fentanyl (2.5 mg/kg). Continuous invasive arterial pressure monitoring was accomplished by arterial catheterization of the right femoral artery. The mean blood pressure in each group is presented in Table I. Blood pressure was maintained at a minimum of 70 mm Hg during the procedure. Animals with nonresponsive hypotension were excluded. Systemic heparin (150 units/kg) was administered before hilar clamping. Even though the use of heparin in urologic surgery is controversial, systemic heparinization was employed because of the susceptibility of pigs to thrombi.15,16 The use of vasoactive agents was not allowed, and hypotension was managed with volume replacement. After surgery, all pigs were humanely euthanized by a lethal dose of thiopental and 15 to 30 mg/kg of 19.1% potassium chloride.
2017
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Experimental swine model of unilateral renal ischemia Take Home Message: In vivo physiologic and histologic data support that significant unilateral kidney ischemia causes changes in the contralateral nonischemic kidney. Recommendation: The nonischemic kidney may suffer damage by unclear mechanisms when one kidney is critically ischemic.
Table I. Adjusted mean values and standard errors for the mean blood pressure in each group Clamping protocol Time, minutes
A group (n ¼ 8)
AV group (n ¼ 9)
Basal
101.4 (2.3)a
94.9 (2.2)
10
94.7 (2.0)b
93.6 (1.9)
20
a,b
87.9 (2.5)
94.7 (2.4)
30
80.9 (2.4)a,b
89.7 (2.3) 80.4 (2.4)b
40
78.1 (2.6)
50
78.1 (2.6)b
79.9 (2.4)b
60
b
70.6 (2.8)b
a,b
66.4 (3.2)b
b
72.4 (3.4)b
b
74.2 (4.0)b
70 80 90
73.1 (3.0) 76.6 (3.4)
76.1 (3.6)
72.9 (4.2)
A group, Artery-only clamping; AV group, artery and vein clamping. The data represent the adjusted mean values and standard errors calculated using generalized estimating equation models. a Significant difference between groups. b Significant difference from baseline.
Operative technique. The kidneys were accessed through a midline laparotomy from xiphoid to pubis. The kidneys were biopsied bilaterally at time 0. Next, left renal hilar clamping was achieved using bulldog forceps, and serial parenchymal biopsies were performed in both kidneys. Biopsy samples were then gently collected using a No. 11 ophthalmic scalpel blade, with no significant parenchymal injury or major blood loss. Reperfusion was initiated after 90 minutes. Ultrasound data. Ultrasound examination was performed using a Philips HD7 ultrasound system (Koninklijke Philips N.V., Amsterdam, The Netherlands) and multifrequency linear transducer (7-11 MHz) in twodimensional, color, pulsed, and power Doppler modes. Image registration was performed at time 0 to rule out any previous alterations. Power Doppler ultrasound was used whenever necessary to confirm the absence of flow. Ultrasound analysis. The RI and PI were calculated automatically by the ultrasound software (High Q
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
3
-
Fig 1. Adjusted means and 95% confidence intervals for the resistive index (RI) in the right kidney of pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols.
Automated Doppler analysis). An RI < 0.7 and PI < 1.2 were considered normal. Histology. All specimens collected for histologic examination were embedded in paraffin and processed using standard protocols. The specimens were cut to 3-mm sections and stained with hematoxylin and eosin. The slides were photographed under an optical microscope (Nikon, Tokyo, Japan), and the following histopathologic changes were analyzed using ImageJ software (National Institutes of Health, Bethesda, Md): degenerative changes in tubular cells, presence of pigmented cylinders, vascular congestion and edema, interstitial neutrophil infiltration, and interstitial hemorrhage. Lesions were scored as absent (score 0) or present (score 1). Statistical analysis. Categorical data are expressed as absolute frequencies and percentages, and continuous data are expressed as means with standard deviations and minimum-maximum values. The inferential analysis of ultrasound data was performed using mixed models with normal distribution and first-order autoregressive correlation structure. Results are presented as adjusted means with standard errors and line plots with 95% confidence intervals. Histologic lesions were analyzed using generalized estimating equation models, and results are presented as estimated proportions with standard errors and 95% confidence intervals. The P values obtained from the multiple comparisons between time points and clamping protocols were adjusted by the sequential Bonferroni method. The Fisher exact test was used to investigate the association between clamping protocol and thrombosis occurrence. All analyses were performed using SPSS Statistics
for Windows version 19.0 (IBM Corp, Armonk, NY). A P value < .05 was considered significant.
RESULTS Ultrasound analysis. Mixed model analysis of the effect of clamping protocol on RI showed a significant interaction between group (clamping protocol) and time (P < .001). In the A group (artery-only clamping), RI values increased significantly in the nonischemic kidney (relative to baseline values) after 30 minutes of ischemia. Conversely, no significant differences in RI values were detected across time in the AV group (artery and vein clamping), despite the increase in mean RI values from baseline (0.562) to 90 minutes (0.639). The differences in RI values between group A and AV nonischemic kidneys were significant at 30, 40, and 50 minutes of ischemia (Fig 1). Mixed model analysis of the effect of clamping protocol on PI showed a significant interaction between group (clamping protocol) and time (P ¼ .014). In the A group, PI values of nonischemic kidneys increased significantly (relative to baseline values) after 30 and 60 minutes of ischemia. Conversely, no significant differences in PI values were detected across time in the AV group, despite the increase in mean PI values from baseline (0.808) to 90 minutes (1.164). The differences in PI values between group A and AV nonischemic kidneys were significant at 30, 40, and 50 minutes of ischemia (Fig 2). After 90 minutes of ischemia, the renal clamp was released and renal reperfusion initiated. Renal artery thrombi were observed in nonischemic kidneys from three animals in the A group (37.5%) but were not detected in AV pigs. There was no significant difference in thrombosis occurrence between A and AV
4
Journal of Vascular Surgery
Baptista Sincos et al
---
2017
Fig 2. Adjusted means and 95% confidence intervals for the pulsatility index (PI) in the right kidney of pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols. Table II. Estimated percentage of lesions in renal biopsy specimens from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols Clamping protocol AV group Time, minutes Baseline
Nonischemic kidney
A group Ischemic kidney
Nonischemic kidney
Ischemic kidney
10.0 (3.5)
7.5 (4.9)
10.0 (7.1)
17.5 (6.6)
10
7.5 (3.4)
12.5 (4.9)
20.0 (3.5)a
32.5 (4.9)a
20
20.0 (5.0)
25.0 (5.9)a
30.0 (6.1)a
42.5 (6.6)a
30
27.5 (7.0)
32.5 (4.9)a
42.5 (7.4)a
45.0 (5.9)a
a
a
47.5 (7.0)a
45.0 (6.8)
a
55.0 (5.9)a
45.0 (6.8)
a
50.0 (7.1)a
40 50
27.5 (6.1)
37.5 (5.5) a
37.5 (6.6)
a
a
30.0 (5.0)
a
60
37.5 (8.2)
70
35.0 (8.5)a
47.5 (7.0)a
50.0 (8.7)a
50.0 (5.0)a
80
a
a
a
57.5 (4.2)a
a
52.5 (8.6)a
90
37.5 (5.5)
30.0 (9.4)
40.0 (7.1)
40.0 (5.0)
45.0 (7.7)
a
47.5 (7.0)
52.5 (7.9)
45.0 (8.5)
The data represent the estimated proportions and standard errors calculated using generalized estimating equation models. a Significant difference from baseline.
nonischemic kidneys (P ¼ .082), and no thrombi were detected in ischemic kidneys. Histologic analysis. Sections from eight animals in each group were examined. Samples from one animal in the AV group were excluded because of tissue damage during processing and storage. Renal lesions by type. The qualitative analysis of ischemic kidneys revealed vascular congestion and edema after 10 minutes of ischemia that were more prevalent in A group animals. Interstitial neutrophil infiltration was observed in four of eight biopsy specimens in the A group and two of eight biopsy specimens in AV
animals. Degenerative changes were more frequent after 40 minutes of ischemia and more prevalent in the A group. Interstitial hemorrhage was also more frequent in A group animals. Although rare, pigmented cylinders, which are proteins released by tubular renal cells during acute necrosis, were the only histopathologic change more prevalent in the AV group. Vascular congestion and edema were detected in nonischemic kidneys of the A and AV groups after 10 minutes. Interstitial neutrophil infiltration was observed after 30 minutes in both groups. Degenerative changes were more frequent after 40 minutes of ischemia and more prevalent in the A group. Interstitial hemorrhage was also more frequent in the A group and was detected in 50% of
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
5
-
Fig 3. Estimated percentage of lesions and 95% confidence intervals in renal biopsy specimens from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols.
Table III. Estimated percentage of lesions in renal biopsy specimens of ischemic and control kidneys from pigs assigned to the artery-only (A) and artery and vein (AV) clamping protocols regardless of clamping protocol Kidney Time, minutes
Nonischemic
Ischemic
Baseline
10.0 (4.0)
12.5 (4.3)
10
13.8 (2.9)a
22.5 (4.3)b
20
25.0 (4.1)b
33.8 (4.9)b
30
b
38.8 (4.1)b
b
42.5 (4.6)b
b
46.3 (4.9)b
b
b
40 50
35.0 (5.4)
33.8 (4.2)
37.5 (4.6)
60
41.3 (5.4)
70
42.5 (6.3)b
48.8 (4.3)b
80
b
51.3 (4.7)b
90
45.0 (5.2)
37.5 (6.6)
b
45.0 (5.2)
50.0 (5.6)
The data represent the estimated proportions and standard errors calculated using generalized estimating equation models. a Significant difference between nonischemic and ischemic kidneys. b Significant difference from baseline.
biopsy specimens after 70 minutes. Pigmented cylinders were not detected in the A group and were the only histopathologic change more prevalent in the AV group. Lesion biopsies. The quantitative analysis of the estimated percentage of lesions identified on histologic analysis is summarized in Table II and Fig 3. Ischemic kidney lesions by clamping protocol. The percentage of renal lesions increased significantly relative to baseline values after 20 minutes (25%) and 10 minutes (32.5%) in AV and A animals, respectively.
Nonischemic kidney lesions by clamping protocol. A significant increase in the percentage of lesions was also detected in nonischemic kidneys. The percentage of renal lesions increased significantly relative to baseline values only between 50 and 80 minutes (30%) in the AV group and after 10 minutes in A group animals. The percentage of lesions in renal biopsy specimens of ischemic and nonischemic kidneys increased significantly after 10 and 20 minutes, respectively, regardless of clamping protocol (Table III; Fig 4).
DISCUSSION In this study, we developed an experimental porcine model to study the effects of ischemia on the contralateral kidney and the effect of clamping protocol on renal crosstalk. Swine were used as the animal model because these animals are recognized as excellent biomedical models that closely replicate anatomic, physiologic, and immunologic conditions in humans.17 The duration of ischemia was set at 90 minutes in an attempt to more clearly demonstrate any possible differences between the two clamping protocols (artery-only and artery and vein clamping). We found a tendency of elevated RI and PI values in the nonischemic kidney during left hilar clamping, indicating that the renal perfusion changes experienced by the ischemic kidney were replicated in the contralateral kidney of animals in the artery-only (A) and artery and vein (AV) clamping groups. Mean RI and PI values were significantly higher than baseline values in the A group after 30 minutes of ischemia. This finding appears to reflect the increase in the number of histologic lesions in the A group after 10 to 20 minutes of ischemia. Thus,
6
Baptista Sincos et al
Journal of Vascular Surgery ---
2017
Fig 4. Estimated percentage of lesions and 95% confidence intervals in renal biopsy specimens of ischemic and nonischemic kidneys regardless of clamping protocol from pigs assigned to artery-only (A) and artery and vein (AV) clamping.
we believe that serial RI and PI measurements have predictive value to determine the onset of renal injury. In addition, thrombi were detected in three nonischemic renal arteries in the A group, which may also be indicative of remote organ crosstalk. These three thrombotic events were observed only after the 90 minutes of ischemia (reperfusion phase), when they affected the duplex ultrasound readings. The lack of difference for thrombosis rates between the A and AV groups (37.5% vs 0%) could be due to a type II statistical error. The remote organ dysfunction in AKI is induced by the release of humoral and cellular mediators into the bloodstream. Cytokines that are released during ischemia may play an important role in the pathogenesis of tissue injury. Tumor necrosis factor a is produced by tubular epithelial cells and activated leukocytes and induces the expression of adhesion molecules in endothelial cells, enhancing leukocyte infiltration. In addition, tumor necrosis factor a induces other inflammatory cytokines, such as interleukins 6, 8, and 10, which may affect other organ systems.18 This process is called organ crosstalk, a term derived from electronics denoting any signal or circuit that unintentionally affects another. Lane et al19 defined organ crosstalk in critical illness as “the effects of one malfunctioning organ on the function of another.” Several experimental studies have shown that renal ischemia-reperfusion-induced AKI can affect other organ systems.20 Fadillioglu et al21 found elevated levels of liver transaminases in rats that underwent warm kidney ischemia. Gardner et al4 found subtle histopathologic effects of AKI on liver, lung, brain, and kidney samples from pigs
using immunofluorescence. Because pigs can tolerate longer ischemia times,22 the biologically small effects observed in their study may be partially attributed to the short duration of renal ischemia (40 minutes). Histopathologic examination indicated a significant increase in the number of renal lesions of A group nonischemic kidneys after 10 to 20 minutes of ischemia but only after 50 to 60 minutes in AV group nonischemic kidneys. In addition, the number of lesions was higher in both A group ischemic and nonischemic kidneys between 10 and 80 minutes. This finding indicates that organ crosstalk can be affected by clamping protocol. In fact, in AV group animals, there appeared to be a renal protective effect in the impaired venous return as well as lack of cytokine release into the systemic circulation. Moreover, angiotensin II is released systemically from the ischemic kidney through the renal vein, upregulating angiotensin II production in the nonischemic kidney.7 Thus, because renal vein occlusion interrupts the angiotensin II pathway, this may explain why the AV group was associated with fewer flow and histologic effects on the contralateral, nonischemic kidney. In most studies showing necrosis after renal ischemia, kidney biopsy specimens were taken at least 24 hours after reperfusion.4 In this study, reperfusion biopsy specimens were taken only at 120 and 150 minutes because of the difficulty in keeping the animals alive without vasoactive drugs. After 150 minutes of surgery, on average, most animals were in an unresponsive, hypotensive state as a result of dehydration, hypothermia, and, possibly, anesthetic intolerance. However, the use of vasoactive agents was not allowed as it could interfere with the duplex ultrasound scan analysis. Thus, the short
Journal of Vascular Surgery Volume
-,
Number
Baptista Sincos et al
7
-
period of reperfusion combined with the histologic technique used may have affected the diagnosis of dynamic mechanisms, such as necrosis and apoptosis. Bezerra et al14 developed an experimental model of renal ischemia injury in the collared peccary (Tayassu tajacu) and found a significantly higher number of lesions in ischemic kidneys than in nonischemic ones. In our study, the percentage of lesions was higher in ischemic kidneys but increased significantly over time in both ischemic and nonischemic kidneys. To the best of our knowledge, this is the first study to investigate the mechanism and acute effects of renal crosstalk using Doppler flowmetry analysis of the contralateral kidney. Even though no ultrasound signals would be generated during left renal clamping, Doppler ultrasound was used to ensure that no anomalous perfusion occurred in the ischemic kidney. During ischemia, there was a tendency for an increase in RI and PI values over time in the nonischemic kidney. The percentage of renal lesions increased less sharply in the AV group than in the A group over time in both ischemic and nonischemic kidneys. Thus, even though some studies propose using artery-only clamping to minimize the effects of ischemic injury23 as venous patency would lead to a minimal parenchymal perfusion, in our study we found a smaller tendency for renal lesions in animals from the AV group. Moreover, artery and vein clamping may minimize the risk of intraoperative bleeding and hemodynamic impairment.9,24 Renal Doppler ultrasound has been shown to have predictive value in AKI, even before the elevation in urinary nitrogen excretion and the change in urinary debt.11,12 The normal renal artery spectral Doppler waveform has a low RI that reflects the high perfusion of the renal bed.11 In this study, we serially measured RI and PI to detect worsening perfusion patterns over ischemia time. In the A group, RI and PI increased above normal values after 30 minutes of ischemia, and it is possible that the crosstalk between ischemic and contralateral kidneys started at that time. Even though the cause of renal dysfunction cannot be solely determined by an elevated RI, a normal RI is predictive of adequate renal function.13 Several studies have demonstrated that RI alterations on duplex ultrasound scan preceded the increase in serum creatinine concentration in the course of reversible acute renal failure, and they served as the basis for development of our experimental model.14,25-27 Limitations. The sample size was small, and despite its anatomic and immunologic similarities, the pig kidney does not completely replicate the biology of the human kidney.17 Moreover, for funding reasons, it was not possible to measure inflammatory cytokines, serum creatinine concentration, or glomerular function to demonstrate organ crosstalk.
Quantification of ischemia-induced changes of actin, phosphotyrosine, intercellular adhesion molecule 1, b-integrin, and urine biomarkers should be considered for further investigation.
CONCLUSIONS The nonischemic kidney experienced acute effects stemming from contralateral ischemia, and the hilar clamping protocol appeared to affect the occurrence of renal crosstalk, which was reflected in worsening perfusion, elevated RI and PI values, and a tendency for thrombi to form in nonischemic kidneys. The analysis of acute histologic findings and Doppler ultrasound parameters indicated that RI and PI measurements may have predictive value in AKI. We believe that intraoperative Doppler ultrasound can help guide surgeons in determining tolerable ischemia times and the probability of AKI onset. Nevertheless, further clinical and experimental research is required to fully understand ischemiareperfusion injury and renal crosstalk.
AUTHOR CONTRIBUTIONS Conception and design: AS, IS, FN, NW, RA, OK Analysis and interpretation: AS, IS, FN, OK, VP Data collection: AS, AM, KL, OK, VP Writing the article: AS, IS, FN, OK Critical revision of the article: AS, AM, IS, NW, RA, KL, OK, VP Final approval of the article: AS, AM, IS, FN, NW, RA, KL, OK, VP Statistical analysis: AS, IS, FN, KL, OK, VP Obtained funding: AS, AM, NW, RA, OK, VP Overall responsibility: AS
REFERENCES 1. Thompson RH, Frank I, Lohse CM, Saad IR, Fergany A, Zincke H, et al. The impact of ischemia time during open nephron sparing surgery on solitary kidneys: a multiinstitutional study. J Urol 2007;177:471-6. 2. Laven BA, Orvieto MA, Chuang MS, Ritch CR, Murray P, Harland RC, et al. Renal tolerance to prolonged warm ischemia time in a laparoscopic versus open surgery porcine model. J Urol 2004;172(Pt 1):2471-4. 3. Wahlberg E, DiMuzio PJ, Stoney RJ. Aortic clamping during elective operations for infrarenal disease: the influence of clamping time on renal function. J Vasc Surg 2002;36: 13-8. 4. Gardner DS, De Brot S, Dunford LJ, Grau-Roma L, Welham SJ, Fallman R, et al. Remote effects of acute kidney injury in a porcine model. Am J Physiol Renal Physiol 2016;310:259-71. 5. Doi K, Rabb H. Impact of acute kidney injury on distant organ function: recent findings and potential therapeutic targets. Kidney Int 2016;89:555-64. 6. Schmiedt CW, Mercurio A, Vandenplas M, McAnulty JF, Hurley DJ. Effects of renal autograft ischemic storage and reperfusion on intraoperative hemodynamic patterns and plasma renin concentrations in clinically normal cats undergoing renal autotransplantation and contralateral nephrectomy. Am J Vet Res 2010;71:1220-7.
8
Journal of Vascular Surgery
Baptista Sincos et al
---
7. Giani JF, Janjulia T, Taylor B, Bernstein EA, Shah K, Shen XZ, et al. Renal generation of angiotensin II and the pathogenesis of hypertension. Curr Hypertens Rep 2014;16:477. 8. Kobori H, Ozawa Y, Satou R, Katsurada A, Miyata K, Ohashi N, et al. Kidney-specific enhancement of ANG II stimulates endogenous intrarenal angiotensinogen in gene-targeted mice. Am J Physiol Renal Physiol 2007;293:F938-45. 9. Orvieto MA, Zorn KC, Mendiola F, Lyon MB, Mikhail AA, Gofrit ON, et al. Recovery of renal function after complete renal hilar versus artery alone clamping during open and laparoscopic surgery. J Urol 2007;177:2371-4. 10. Parekh DJ, Weinberg JM, Ercole B, Torkko KC, Hilton W, Bennett M, et al. Tolerance of the human kidney to isolated controlled ischemia. J Am Soc Nephrol 2013;24: 506-17. 11. Cano H, Castaneda DA, Patino N, Perez HC, Sanchez M, Lozano E, et al. Resistance index measured by Doppler ultrasound as a predictor of graft function after kidney transplantation. Transplant Proc 2014;46:2972-4. 12. Hansen KJ, Tribble RW, Reavis SW, Canzanello VJ, Craven TE, Plonk GW Jr, et al. Renal duplex sonography: evaluation of clinical utility. J Vasc Surg 1990;12:227-36. 13. Gao J, Ng A, Shih G, Goldstein M, Kapur S, Wang J, et al. Intrarenal color duplex ultrasonography: a window to vascular complications of renal transplants. J Ultrasound Med 2007;26:1403-18. 14. Bezerra Dde O, Feitosa ML, Almeida HM, Costa FA, Braga JF, Souza Fde A, et al. Collared pecary (Tayassu tajacu) as a new model of renal ischemic injury induced by clamping the renal artery. Acta Cir Bras 2014;29:560-72. 15. Sincos IR, da Silva ES, Belczak SQ, Baptista Sincos AP, de Lourdes Higuchi M, Gornati V, et al. Histologic analysis of stent graft oversizing in the thoracic aorta. J Vasc Surg 2013;58:1644-51.e4. 16. Sincos IR, Aun R, da Silva ES, Belczak S, Higuchi ML, Gornati VC, et al. Impact of stent-graft oversizing on the thoracic aorta: experimental study in a porcine model. J Endovasc Ther 2011;18:576-84. 17. Giraud S, Favreau F, Chatauret N, Thuillier R, Maiga S, Hauet T. Contribution of large pig for renal ischemiareperfusion and transplantation studies: the preclinical model. J Biomed Biotechnol 2011;1:532127. 18. Golab F, Kadkhodaee M, Zahmatkesh M, Hedayati M, Arab H, Schuster R, et al. Ischemic and non-ischemic
19.
20.
21.
22.
23.
24.
25.
26.
27.
2017
acute kidney injury cause hepatic damage. Kidney Int 2009;75:783-92. Lane K, Dixon JJ, MacPhee IA, Philips BJ. Renohepatic crosstalk: does acute kidney injury cause liver dysfunction? Nephrol Dial Transplant 2013;28:1634-47. Klein CL, Hoke TS, Fang WF, Altmann CJ, Douglas IS, Faubel S. Interleukin-6 mediates lung injury following ischemic acute kidney injury or bilateral nephrectomy. Kidney Int 2008;74:901-9. Fadillioglu E, Kurcer Z, Parlakpinar H, Iraz M, Cerbail G. Melatonin treatment against remote organ injury induced by renal ischemia reperfusion injury in diabetes mellitus. Arch Pharm Res 2008;31:705-12. Humphreys MR, Castle EP, Lohse CM, Sebo TJ, Leslie KO, Andrews PE. Renal ischemia time in laparoscopic surgery: an experimental study in a porcine model. Int J Urol 2009;16: 105-9. Tracy CR, Terrell JD, Francis RP, Wehner EF, Smith J, Litorja M, et al. Characterization of renal ischemia using DLP hyperspectral imaging: a pilot study comparing artery-only occlusion versus artery and vein occlusion. J Endourol 2010;24:321-5. Funahashi Y, Kato M, Yoshino Y, Fujita T, Sassa N, Gotoh M. Comparison of renal ischemic damage during laparoscopic partial nephrectomy with artery-vein and artery-only clamping. J Endourol 2014;28:306-11. Norris CS, Pfeiffer JS, Rittgers SE, Barnes RW. Noninvasive evaluation of renal artery stenosis and renovascular resistance. J Vasc Surg 1984;1:192-201. Luger-Hamer M, Barbiro-Michaely E, Sonn J, Mayevsky A. Renal viability evaluated by the multiprobe assembly: a unique tool for the assessment of renal ischemic injury. Nephron Clin Pract 2009;111:29-38. Yoon DY, Kim SH, Kim HD, Na DG, Goo JM, Choi HJ, et al. Doppler sonography in experimentally induced acute renal failure in rabbits: resistive index versus serum creatinine levels. Invest Radiol 1995;30:168-72.
Submitted Apr 3, 2017; accepted Jun 16, 2017.
Additional material for this article may be found online at www.jvascsurg.org.
Journal of Vascular Surgery Volume
-,
Number
-
APPENDIX (online only).
Baptista Sincos et al
8.e1
8.e2
Baptista Sincos et al
Journal of Vascular Surgery ---
2017