Dexmedetomidine on renal ischemia-reperfusion injury in rats: assessment by means of NGAL and histology

Abstract

Background: Ischemic acute kidney injury is a common occurrence in the perioperative period and in critical patients admitted to intensive care units. The reestablishment of blood supply may worsen injury through the ischemia-reperfusion (I/R) mechanism. We investigated the effect of dexmedetomidine on the kidneys of rats subjected to an experimental I/R model. Methods: 34 rats anesthetized with isoflurane was undergone right nephrectomy and randomly assigned to four groups: Control C (saline solution); Dexmedetomidine D (dexmedetomidine); Sham S (saline solution); Sham with Dexmedetomidine SD (dexmedetomidine). The serum levels of neutrophil gelatinase-associated lipocalin (NGAL) were measured at time-points T1 (following stabilization), T2 (ischemia), T3 (reperfusion), T4 (12 h after of I/R). The kidneys were subjected to histological examination. Results: The NGAL levels were significantly higher at T4 compared with T1. Upon histological examination, the left kidneys in groups C and D exhibited a similar extent of cell injury. Conclusion: The levels of NGAL did not indicate either protection against or worsening of kidney injury. Histological examination for acute tubular necrosis showed that dexmedetomidine did not protect the kidneys from I/R.

• Histology
• ischemia
• NGAL
• rat
• reperfusion

Introduction

Acute kidney injury (AKI) is an abrupt reduction of kidney function and may occur following major surgery such as vascular, transplant or cardiothoracic procedures, as well as during clinical conditions such as sepsis.1 AKI involves important consequences such as longer hospitalization periods and increased morbidity and mortality.1 The main cause of AKI is ischemia-reperfusion (I/R) injury.2^,3 Reestablishment of blood supply to the kidneys following a period of deprivation results in edema, alteration of vascular permeability, parenchymal inflammatory infiltration, and tubular cell damage.4

Inflammation plays an important role in AKI caused by I/R. The damaged epithelium releases cytokines, which in turn recruit inflammatory cells. Neutrophils, macrophages, and natural killer cells cause direct epithelial damage and amplify the inflammatory response, resulting in apoptosis and/or necrosis.4

Dexmedetomidine is a drug with powerful and selective action on α2-adrenergic receptors5 that exhibits sedative and analgesic properties.6 There are reports of its protective effect against I/R in the brain7^,8 and heart9 of rats. With regard to the kidneys, one study using an I/R model in rats showed that the use of dexmedetomidine was associated with the presence of normal glomeruli and discrete edema in the tubular cells, thus indicating that this drug might reduce kidney injury.10 Another one, also employing an I/R model in rats, found a correlation between the use of dexmedetomidine and a reduction of serum urea and creatinine levels,11 whereas in other reduced serum creatinine levels was found, without protection from injury.12

The aim of the present study was to assess the effect of dexmedetomidine in an experimental I/R model in rats through histological examination and measurement of the serum levels of neutrophil gelatinase-associated lipocalin (NGAL), as an early biomarker for AKI.13^,14

Materials and methods

This study was approved by the institution’s animal experimentation ethics committee at Botucatu Medical School, São Paulo, Brazil. The preparation was based on a previous described method for renal I/R.15^,16

Animals were cared for according to the guidelines of the Council for International Organization of Medical Sciences (CIDMS) ethical code for animal experimentations. Animals fasted overnight prior to the experiment but were given free access to water.

Surgical preparation

A total of 34 Wistar rats were randomized with opaque-sealed envelopes to four groups. All the animals were anesthetized with 4% isoflurane and maintained under spontaneous ventilation through a mask with 1.5 to 3% isoflurane and 100% oxygen. A 24G catheter was inserted in the right internal jugular vein, and another catheter was placed in the left carotid artery. Lactated Ringer’s solution was administered at a rate of 3 mL/kg/h, and warm packs were placed below and above the animals to maintain a constant body temperature. Right nephrectomy was performed on all the animals. As described below, two groups were submitted to clamping of the left renal artery for 30 min (ischemia). The arterial clamp was then removed and reperfusion of the kidney was observed for 30 min (reperfusion).

• C – Control group (n = 10) – Normal saline + Ischemia + Reperfusion.

• D – Dexmedetomidine group (n = 8) – Dexmedetomidine (bolus of 3 µg/kg over 10 min, followed by infusion at a rate of 3 µg/kg/h) + Ischemia + Reperfusion.

• S – Sham group (n = 8) – Normal saline. This group was not subjected to ischemia-reperfusion maneuvers.

• SD – Sham and dexmedetomidine group (n = 8) – Dexmedetomidine (bolus of 3 µg/kg over 10 min, followed by infusion at a rate of 3 µg/kg/h). This group was not subjected to ischemia-reperfusion maneuvers.

The abdominal wall was closed, bupivacaine 0.25% was injected into the incisions and subcutaneous tramadol (2 mg/kg) was administrated post-operative for analgesia. The animals were woken and kept in a climate-controlled container for 12 h with food and water freely available. Twelve hours after, they were anesthetized again with isoflurane, subjected to left nephrectomy, and immediately sacrificed with intracardiac injection of sodium thiopental.

The mean arterial pressure (MAP), heart rate (HR), and body temperature were monitored and recorded at time-points T0 (at monitoring), T1 (immediately after the dissection of the cervical vessels), T2 (30 min after renal clamping), T3 (30 min after renal reperfusion) and T4 (following a 12-h period). Blood samples were collected to measure the NGAL levels at time-points T1, T2, T3 and T4.

Histologic preparation

The extracted right and left kidneys were placed separately in labeled flasks containing Duboscq-Brazil solution for 24 h. They were then stored until histological assessment in a solution of alcohol, picric acid, formalin and glacial acetic acid, before paraffin preparation. The sections were stained by hematoxylin and eosin (H&E). These sections were assessed by an experienced pathologist (M.A.D.), blind from the experimental groups, with regard to the percentage of tubular necrosis and severity of damage to tubular cell, according to a previous described score17 as shown in Table 1.

Table 1. Severity score of tubular cells damage and necrosis.

Grade	Type	Tubular necrosis (%)
0	No damage	Absent
1	Mild	<10%
2	Moderate	10 a 25%
3	Moderate to severe	25 a 50%
4	Severe	50 a 75%
5	Severe to very severe	>75%

Note: Adapted from Park et al., 2008.29

Statistical analysis

Statistical analysis was performed using the Stata/SE 9.0 software for Windows (Stata Corporation, College Station, TX). The measurements of body weight, temperature, serum NGAL, as well as the histological score were expressed as the median and 25–75% percentiles. The comparison of body weight, temperature and histological scores between groups at one time-point was performed by the Kruskal–Wallis test – Dunn post-hoc test. The comparisons between time-points for body weight, temperature and serum NGAL were performed by the Friedman test and the histological score by the Mann–Whitney test. The level of significance was established as p < 0.05.

Results

The weights (g) of the animals in groups C, D, S and SD were (mean) 439 g, 442 g, 436 g and 462 g, respectively (p = 0.42) showing homogeneity between then.

Body temperatures did not result in differences between the groups. Mean values from all groups ranged between 36.85 and 37.5 °C, within the physiological limits, as shown in Figure 1.

Figure 1. Average of the temperature per group and time-point (p = 0.21).

HRs had median values ranging from 227 to 250 (bpm), with a slight decreasing in D and SD groups without statistical significance.

The MAP was not different between the groups at T0. In all of them, the MAP was significantly lower at T1, T2 and T3 compared with T0, with statistical significance in all but S group, as shown in Table 2. The MAP decreased during subsequent measurements, most notably in groups D and SD (p < 0.001 and p = 0.006 respectively).

Table 2. Median and (25–75%) percentiles values of the mean arterial pressure (MAP) (mmHg) per group and time-point.

		Time-points
Group	T0^a	T1^a	T2^a	T3^a	p
C^b	101.0 (79.0–110.0)	97.5 (91.0–100.0)	80.5 (63.0–91.0)	87.5 (72.0–94.3)	0.02
D^b	80.5 (76.0–95.0)	66.5 (52.5–73.0)	52.0 (48.0–56.0)	56.0 (49.5–63.5)	<0.001
S^b	103.5 (88.5–111.0)	86.0 (77.0–99.0)	86.0 (74.0–95.5)	90.5 (75.0–100.5)	0.25
SD^b	100.5 (90.0–107.5)	67.5 (61.5–83.0)	69.0 (61.5–71.5)	68.5 (61.5–70.5)	0.006
p	0.18	0.004	<0.001	<0.001

Notes: C, control; D, dexmedetomidine; S, sham; SD, sham and dexmedetomidine.

T0: at monitoring. T1: at venoclysis. T2: 30 min after renal clamping. T3: 30 min after renal reperfusion. T4: following a 12-h period. p: p-value.

^aKruskal–Wallis analysis of variance – Dunn post-hoc test. Median (25–75% percentiles).

^bFriedman test.

The NGAL levels exhibited a considerable increase at T4 compared with the initial (T1). This increase was greater in groups C and D compared with S and SD, which had a significant difference (p = 0.001), as shown in Table 3.

Table 3. Median and (25–75%) percentiles values of neutrophil gelatinase-associated lipocalin (NGAL) (ng/mL) per group and time-point.

		Time-points
Group	T1^a	T2^a	T3^a	T4^a	p
C^b	10.8 (2.4–11.9)	11.7 (10.6–15.1)	13.9 (10.8–16.4)	400 (288.6–400)	0.024
D^b	48.8 (38.4–66.1)	81.5 (42.2–95.9)	60.1 (42.0–91.1)	398.3 (381.4–398.5)	<0.001
S^b	13.3 (10.5–14.8)	10.5 (10.2–12.4)	8.9 (7.4–11.9)	127.3 (25.5–173.3)	0.01
SD^b	29.4 (0.0–71.4)	29.3 (5.5–73.50)	31.6 (8.9–68.6)	167.3 (164.7–177.6)	0.003
p	0.007	0.001	0.002	0.001

Notes: C, control; D, dexmedetomidine; S, sham; SD, sham and dexmedetomidine.

T0: at monitoring. T1: at venoclysis. T2: 30 min after renal clamping. T3: 30 min after renal reperfusion. T4: following a 12-h period. p: p-value.

^aKruskal–Wallis analysis of variance – Dunn post-hoc test. Median (25–75% percentiles).

^bFriedman test.

The kidneys were assessed according to the tubular necrosis score as described above. Left kidneys in groups C and D had median score values of 3 (moderate to severe tubular cell damage) while the Sham groups, as expected, exhibited no damage (Table 4). Examples are shown in Figure 2.

Figure 2. (A) A left kidney section (H&E) in S group with a grade 0 severity score (no tubular cell damage). (B) A left kidney section (H&E) in C group with a grade 3 severity score (moderate to severe damage with 25–50% of tubular necrosis). Arrows from left to right: pyknosis, karyolysis and karyorrhexis. Images increased by 400×.

Table 4. Kidney histology.

	Kidney
Groups	Left (LK)^a	Right (RK)^a	p
C^b	3 (2–3.75)	0 (0–0)	0.004
D^b	3 (1–3.75)	0 (0–0)	0.002
S^b	0 (0–0)	0 (0–0)	–
SD^b	0 (0–0)	0 (0–0)	–
P	0.0001	0.39

Notes: C, control; D, dexmedetomidine; S, sham; SD, sham and dexmedetomidine. p: p-value. Tubular necrosis score per group shown as median and (25–75%) percentiles values.

^aKruskal–Wallis analysis of variance – Dunn post-hoc test. Median (25–75% percentiles).

^bMann–Whitney test.

Discussion

AKI due to ischemia-reperfusion injury is a complex and dynamic process characterized by a reduction of the glomerular filtration rate, tubular cell damage and necrosis, glomerular injury and signs of tubular obstruction by cell debris.4

The influence of body temperature on the I/R process has already been reported.18^,19 Under these conditions, hyperthermia exerts a deleterious effect that worse the renal injury, whereas hypothermia reduces damage and protects the kidneys. In the present study, the animals’ body temperature was affected by several external factors. The parenteral solutions were administered at room temperature, the abdominal cavity was opened, and ventilation was not controlled by means of gas re-inhalation and warming. These factors all result in heat loss. However, the animal temperature was kept as constant as possible by placing warm packs above and below the animals, while rectal temperature was monitored. Differences in temperature were not observed among the groups. Although in SD group the temperature at time-point T0 was lower compared with T1 and T2 (p = 0.005), the body temperature remained within the normal range at all the analyzed time-points; withdrawing this factor as a possible bias in the study.

The HR was slight decreased in groups D and SD after the administration of dexmedetomidine was started. Therefore, the actual HR of the animals may have been higher. The reduction in the HR did not influence the study results, as it was not associated with important hemodynamic changes nor did it induce significant alterations in renal perfusion, as previously discussed.

The use of dexmedetomidine reduces the MAP5; however, its effect is biphasic. When used at low doses that induce lower plasma concentrations, dexmedetomidine acts on pre-synaptic α2 receptors, causing arterial hypotension due to central sympatholysis and inhibition of peripheral neurotransmission. Higher doses induce a hypertensive response due to the activation of post-synaptic α2 receptors of the vascular smooth muscle. Such doses are also associated with dose-dependent bradycardia due to a reduction of the sympathetic tonus and increased vagal activity of the baroreflex.6^,20

Arterial blood pressure plays an important role in renal perfusion.21 In the present study, the MAP decreased in all the groups at T1, T2 and T3 compared with T0. The MAP reduction was greater in groups D and SD, in which dexmedetomidine was administered. All the MAP values remained within the parameters of the renal perfusion self-regulatory mechanism, which maintains kidney perfusion constant when the MAP ranges from 60 to 180 mmHg, except for time-points T2 and T3 in D group which the MAP fell slightly below the curve of renal self-regulation during the renal artery clamping stage and during the kidney reperfusion stage. This condition might induce renal hypoperfusion and thus prolong the stage of ischemia, regardless of how; analysis of the results did not unveiled ischemia on that group, concluding that the MAP reduction did not influence the final outcome.

NGAL is an early marker of kidney injury and is used as a measure of tubular stress.13 It is primarily synthesized in the kidneys (distal nephron) in response to kidney injury and eliminated through the urine.22 NGAL is also synthesized systemically in response to kidney injury, filtered by the glomeruli, reabsorbed by the proximal tubular epithelium, and eliminated at a low level through the urine.14

In the present study, the NGAL levels at time-point T4 were higher in the C group compared with S and SD and higher in D group compared with SD. The comparison between all groups was statistical distinct. These findings indicate that dexmedetomidine did not exert a protective effect on the renal function of rats subjected to the I/R model. In the literature, studies that evaluated the effect of dexmedetomidine on renal I/R injury differ in their methods and findings. In these, the drug has been administered via either the peritoneal or intravenous route. Additionally, the drug has been given at doses ranging from 25 µg/kg to 100 µg/kg via the intraperitoneal route and as 1 µg/kg intravenous bolus over 10 min followed by continuous infusion at a rate of 1 µg/kg/h. Kidney injury was reduced when dexmedetomidine was administered via the intraperitoneal route at the above mentioned doses.10^,11 However, this protective effect was not observed in a study where it was administered as continuous intravenous infusion.12 In the present research, dexmedetomidine was administered as an intravenous bolus of 3 µg/kg over 10 min, followed by continuous infusion at a rate of 3 µg/kg/h. The infusion doses used were determined based on the equation recommend by the FDA (U.S. Food and Drug Administration).23 This equation transforms the drug dose in rats (RD) into the human equivalent dose (HED) expressed as mg/kg or µg/kg. This equation was used to make the doses used in the experimental model as close as possible those used in clinical practice. After using FDA equation and to facilitate handling of the infusion pump, the rat dose of a bolus of 3 µg/kg administered over 10 min (which corresponds to a bolus of 0.48 µg/kg in humans), followed by continuous infusion of 3 µg/kg/h (which corresponds to an infusion of 0.48 µg/kg/h in humans) was selected. We sought to establish whether the total administered dose or the route of administration might influence the study results.

Acute tubular necrosis (ATN) is the most common form of AKI, and although it is characterized by reduction in renal function, it is not always attended by evident morphological damage. The histological assessment performed in the present study looked for evidence of ATN. As no damage was found in the right kidney of any animal, this was used as a control for the analysis of the left kidney and served to ensure the reliability of the experiment. The left kidneys of the animals in groups SD and S did not exhibit any degree of injury upon histological examination, whereas those in groups C and D exhibited similar degrees of injury, approximately 25–50% necrosis, which is defined as moderate to severe injury. These findings indicate that dexmedetomidine did not exert a protective effect against I/R injury and are in agreement with the results of the serum NGAL levels, on which dexmedetomidine also did not have any effect.

This study is an experimental model for renal I/R investigation and has limitations. A single kidney model was utilized,15^,16 since there were right nephrectomies to prevent hyper activation of the renin-angiotensin-aldosterone system and severe vasoconstriction on the contralateral kidney.24 There are still uncertainty about the best methods for testing and several drugs under investigation to prevent I/R lesions, like Allopurinol,22 hence, some long way before appropriate human studies. For instance, the optimal ischemic injury time is not well established in the literature. The models of renal ischemia in rats range from 20 to 60 min.25^,26 Another example of the uncertain methodology is the variety of dexmedetomidine doses. We started with a low dose of 3 µg/kg/h but there are reports of up to 20 µg/kg/h in rats.27

Albeit we did not have the expected results in nephroprotection with Dexmedetomidine, it is important to report those negative results in the literature, preventing selection bias on future studies and reviews.28^,29

Conclusion

In the present study, the use of dexmedetomidine did not correlate with kidney protection against ischemia-reperfusion injury, as assessed by histological examination and measurement of NGAL.

Declaration of interest

All the authors declare no competing interests.