Pre-treatment with the angiotensin receptor 1 blocker losartan protects renal blood flow and oxygen delivery after propofol-induced hypotension in pigs

Hypotensive events are strongly correlated to the occurrence of perioperative acute kidney injury, but the underlying mechanisms for this are not completely elucidated. We hypothesised that anaesthesia-induced hypotension causes renal vasoconstriction and decreased oxygen delivery via angiotensin II-mediated renal vasoconstriction. Pigs were anaesthetised, surgically prepared and randomised to vehicle/losartan treatment (0.15 mg*kg−1). A deliberate reduction in arterial blood pressure was caused by infusion of propofol (30 mg*kg−1) for 10 min. Renal function and haemodynamics were recorded 60 min before and after hypotension. Propofol induced hypotension in all animals (p < 0.001). Renal blood flow (RBF) and renal oxygen delivery (RDO2) decreased significantly regardless of treatment but more so in vehicle-treated compared to losartan-treated (p = 0.001, p = 0.02, respectively). During recovery RBF and RDO2 improved to a greater extent in the losartan-treated compared to vehicle-treated (+ 28 ml*min−1, 95%CI 8–50 ml*min−1, p = 0.01 and + 3.1 ml*min−1, 95%CI 0.3–5.8 ml*min−1, p = 0.03, respectively). Sixty minutes after hypotension RBF and RDO2 remained depressed in vehicle-treated, as renal vascular resistance was still increased (p < 0.001). In losartan-treated animals RBF and RDO2 had normalised. Pre-treatment with losartan improved recovery of renal blood flow and renal oxygen delivery after propofol-induced hypotension, suggesting pronounced angiotensin II-mediated renal vasoconstriction during blood pressure reductions caused by anaesthesia.


Renal parameters.
Plasma creatinine, creatinine clearance, urine output and urinary sodium excretion did not change significantly as an effect of losartan treatment or propofol (Table 1).
PWP and RVP was not significantly changed during the experimental protocol. www.nature.com/scientificreports/

Discussion
This study was conducted to evaluate the acute effects of systemic Ang-II antagonism on renal perfusion, oxygenation and function after an anaesthesia-induced hypotensive event. The main finding was that losartan improved recovery of RBF and RDO 2 after a hypotensive dose of propofol. The data obtained indicates that RBF may be impaired for a considerable amount of time after hypotension even though usually monitored clinical parameters (i.e. MAP, HR, CO, plasma-lactate or urine output) remained unchanged or have normalised. Perioperative AKI has been described as an under-recognised problem with few available preventive treatment strategies 21 . In a majority of cases the underlying mechanism causing renal dysfunction is unknown 22,23 . Propofol is an agent frequently used to induce general anaesthesia. It is also a potent vasodilator which entails that the most common side-effect is hypotension 10 .
Ang-II causes powerful systemic vasoconstriction through Ang-II receptor 1 (AT 1 ) binding. This effect is regionally differentiated in that the renal circulation constricts more in comparison to other vascular beds 24 . Plasma levels of Ang-II are increased by hypotension and decreases RBF and glomerular filtration rate 25 . In this study we designed an experiment to investigate if renal perfusion was impaired due to Ang-II. The aim was to mimic a clinical anaesthesia induction resulting in an unwanted but transient hypotensive event. Even if hypotension is swiftly corrected the resulting potential increase in Ang-II may cause a persistent reduction in RBF that is difficult to monitor and treat in anaesthetised patients undergoing surgery. Patients that are extra susceptible Table 1. Renal Lactate (R-Lac), Renal Oxygen Consumption (RVO 2 ), plasma creatinine (P-Cr), creatinine clearance (Cr-Clearance), urine output and urinary sodium excretion (USE) in vehicle-and losartan-treated pigs during the treatment period and recovery period in vehicle-and losartan-treated. *Denotes within-subject effects: timepoint vs previous timepoint, † denotes between-subject effects at that timepoint.   27 . A clinical study showed that inhibition of Ang-II in diabetic CKD patients had beneficial effects on renal oxygenation 28 . Furthermore, the current study indicates that Ang-II may also be a culprit in acute settings of anaesthesia-induced hypotension. Continuous medication targeting Ang-II is common in large population of patients, such as CKD, diabetes and cardiovascular disease. These groups of patients also have higher risk for perioperative AKI. Prolonged renal vasoconstriction and decreased RDO 2 may provoke a mismatch in O 2 consumption and delivery ultimately leading to hypoxia 29,30 . In current clinical practice it is common procedure to withhold any ACEi/ARB's prior to surgery. The rationale is to avoid hypotension and cardiac events 31 . However, the preoperative management of ACEi/ARB's are extensively debated and currently the data for continuing or withholding preoperative treatment is not clear 20,32 . It was suggested over 20 years ago that hypertensive patients should continue Ang-II antagonistic treatment before surgery 33 . On the other hand, Brabant et al. demonstrated significantly higher risk for reduced arterial blood pressure and more treatment-resistant hypotensive events in patients continuing ARB's compared to other anti-hypertensives 34 . Interestingly, a recent multicentre study demonstrated no significant differences in AKI outcome in patients with ongoing ARB's/ACEi compared to patients withholding treatment. The potent vasoconstriction achieved by pharmacological treatment with Ang-II have also been demonstrated to have beneficial effects on blood pressure and reduced need of vasopressors during refractory shock 35 .
The reductions in RBF and increase in RVR in vehicle-treated pigs were significant 60 min after propofol was discontinued, even though MAP had more or less recovered. Losartan attenuated the renal vasoconstriction and normalised renal blood flow. Although uncertain in this study as we did not measure AngII-levels in plasma, this was likely due to inhibition of excessive release of Ang-II caused by hypotension 36 . As a result, RDO 2 was reduced in the vehicle-treated pigs but greatly improved by losartan. It has also been shown that anaesthesia per se cause release of Ang-II 37 . Since losartan did not reduce MAP prior to hypotension the contribution of Ang-II to baseline blood pressure during anaesthesia in the current setting is likely small. The lack of pressure effect inhibiting systemic Ang-II is much like what is seen in normotensive conscious humans 38 . RVO 2 was decreased in the losartan-treated animals during the hypotensive event. Ang-II acts via the AT 1 -receptor to stimulate Na/K-ATPase and, in theory, thereby increasing renal oxygen consumption. This effect was not noted during the relatively short protocol executed in this study. Unfortunately oxygenation in renal tissue was not measured and it is possible that local hypoxia was present although not to the degree that it was reflected in renal vein blood gases. Furthermore, the subjects were healthy, young pigs with kidneys most likely more resistant to decreased renal blood flow than the average human patient undergoing general anaesthesia.
Other parameters for assessing renal function such as P-Cr, Cr-C, USE and urine output did not change during the course of the protocol. This is most likely due to the acute setting and that more profound renal impairment may be necessary for these markers to be altered in previously healthy individuals 39 .
Cardiovascular parameters were continuously monitored to ensure that systemic haemodynamics were not negatively affected by losartan and to identify possible mechanisms for changes in RBF. Losartan per se had no adverse effects on haemodynamics during the study protocol. Importantly, CBF remained unaffected by losartan during hypotension suggesting that losartan did not impair cerebral perfusion. Another crucial aspect is that Table 3. Central venous pressure (CVP), pulmonary artery pressure (PAP), pulmonary wedge pressure (PWP) and renal venous pressure (RVP) in vehicle-treated and losartan-treated pigs during baseline, treatment, hypotension and recovery. *Denotes within-subject effects: timepoint vs previous timepoint. www.nature.com/scientificreports/ losartan did not augment the hypotension caused by propofol. This is in agreement with the finding that MAP was affected similarly during hypovolaemia in anaesthetised pigs treated with either ACEi/ARB's or vehicle 40 . Limitations. All animals in this study arrived one at the time after a short transport and was assessed by experienced lab-technicians to be calm and non-stressed at time of anaesthesia induction. A mix of male and female pigs was used to eliminate gender variability. It would have been preferable to have blinded this study to rigorise the results, however due to practical reasons this was not feasible. The acute setting limits the information provided on possible long-lasting effects of Ang-II inhibition during anaesthesia. Furthermore, the relatively short duration of ischemia in young and healthy individuals also reduce the likelihood of renal structural damage; therefore no histological samples were collected. The purpose of this study was to demonstrate the physiological effects of Ang-II blockade. However, biomarkers such as urinary kidney injury marker-1 (KIM-1) or neutrophil gelatinase-associated lipocalin (NGAL) could have been analysed in order to detect mild renal injury. Plasma levels of Ang-II were not measured in this study and because of that we cannot definitely conclude that Ang-II levels were elevated in response to hypotension. However, it has been shown in similar animal models 36 . Finally, the present study uses a relatively small number of healthy pigs as a large animal model of the human situation. This is to enable research impossible to perform in patients, describing basic mechanisms of action. The results should be interpreted carefully and cannot be directly transferred to the clinic. Still, the prevention of renal vasoconstriction by losartan presented here merits future investigations of the renal effects of Ang-II inhibition in anesthetized human patients.

Conclusion
The present study demonstrate that pre-treatment with losartan prevent renal vasoconstriction and improve recovery of renal blood flow and oxygen delivery after anaesthesia-induced hypotension.

Methods
Ethical approval and animals. All methods were carried out in accordance to relevant guidelines and regulations. Ethical approval for this study (Dnr. 5.8.18-02325/2019) was provided by the Uppsala Animal Ethics Board of the Swedish Board of Agriculture (Jordbruksverket), Sweden on March 29, 2019. Male and female Norwegian Landrace breed/Hampshire/Yorkshire pigs (24 ± 2 kg, 3-4 months old) were purchased from a local farmer in Uppsala, Sweden, to eliminate gender variability. The pigs were born on the farm and housed in group in large cages with water and food ad libitum. A total of 22 pigs were used; 11 vehicle-treated and 11 losartantreated. Pigs arrived at the laboratory at 8.00 am (a 20-min truck ride in cage) one/two at the time and were randomised to treatment/control group by an 'every other' method.
Anaesthesia. At the time of arrival, the pigs were weighed and sedated with an intramuscular injection of tiletamine-zolazepam (Zoletil 6 mg*kg −1 ) and xylazine (Rompun 2.2 mg*kg −1 ). After 3 min, pigs were tracheostomised and put under mechanical ventilation and given ketamine (Ketaminol 20 mg) and morphine (20 mg) in a peripheral vein in the ear. Pentobarbital (8 mg*kg -1 *h −1 ) and morphine (0.26 mg*kg −1 *h −1 ) dissolved in a glucose/sodium-solution (sodium chloride 2.5 mg*ml −1 and glucose 25 mg*ml −1 ) was given for maintenance of anaesthesia and rocuronium (Esmeron 2,5 mg*kg −1 *h −1 ) for muscle relaxation. After finished experimental protocol, pigs were euthanised with potassium chloride. Surgical preparation. Peripheral vein catheters were placed in each ear for heated (38 °C) infusion of anaesthesia and Ringer's acetate (15 mg*kg −1 *h −1 for the first hour followed by 10 mg*kg −1 *h −1 ). Mechanical ventilation was set to achieve arterial PCO 2 4.5-5.5 kPa with FiO 2 0.30. Tidal volume (TV) was started at 10 ml*kg -1 , respiratory rate (RR) 25 and positive end-expiratory pressure (PEEP) at 5 cm H 2 O, if needed TV/ RR was adjusted to maintain arterial PCO 2 4.5-5.5 kPa. The carotid artery on the right side was catheterised with a single lumen catheter for continuous monitoring of MAP and blood sampling. The jugular vein on the right side was catheterised with a 3 lm-catheter for cardiac output, fluid infusion and continuous monitoring of central venous pressure (CVP). Furthermore, a balloon-tipped pulmonary artery catheter (7.5F Swan-Ganz, Edwards Lifesciences, Irvine, CA) was placed into the right jugular vein and advanced into the pulmonary artery for monitoring of pulmonary arterial pressure (PAP), pulmonary wedge pressure (PWP) and blood sampling. The location of the PA-catheter was confirmed by assessing the pressure-curve on the monitor derived from the tip of the catheter. After the right atrium was reached, the catheter was carefully advanced into the pulmonary artery. To confirm correct placement, the balloon was inflated to obtain a pulmonary wedge pressure curve on the monitor. A flow probe (FSB-series 3, Transonic, Ithaca, NY) was placed around the carotid artery on the left side for monitoring of carotid blood flow (CBF). The left jugular vein was catheterised with a single lumen catheter for blood gas analysis. A suprapubic catheter (Foley no. 8. A Datex-Ohmeda S/5 monitor, Madison, WI) was placed in the bladder for urine collection. The pigs were then turned to lay on their right side. A 10 cm incision was made from rib to hip to locate and dissect the left kidney. A flow probe (FSB-series 4) was placed around the renal artery for continuous monitoring of RBF. A single lumen catheter was then placed into the renal vein for monitoring of renal vein pressure (RVP) and blood sampling. All incisions were closed with sutures (Prolene 3.0). After surgical preparation, the pigs were allowed to recover for 45 min before the experiment commenced.
Experimental protocol. Pigs were block-randomised to no treatment (vehicle, n = 11) or losartan treatment (n = 11). Baseline recordings for 5 min were performed before treatment was started. An intravenous infusion of losartan was commenced with a bolus (0.2 mg*kg −1 ) followed by a continuous infusion (0.15 mg*kg −1 *h −1 )  41 . Vehicle-treated pigs received no additional infusion. The experiment was divided into four consecutive periods; baseline (5 min before drug administration), treatment (60 min with drug infusion), hypotension (10 min with propofol infusion) and recovery (60 min). Hypotension was induced by continuous infusion of propofol (30 mg*kg −1 ). At every timepoint (Fig. 3), blood was sampled for blood-gas status and analysis of creatinine, sodium and potassium. Vital parameters were noted by reading of the mechanical ventilator, the monitor for pressures and the Transonic flow-meter. In addition, arterial blood was also collected halfway through the treatment and recovery periods. Urine was continuously collected only over the 60-min periods (treatment and recovery). Blood and urine samples were analysed by the in-house hospital laboratory at the Uppsala University Hospital (Clinical Chemistry Laboratory) to evaluate plasma and urinary creatinine (enzymatic method) and sodium/ potassium levels (flame photometry). Blood gases were sampled from the carotid artery, renal vein, jugular vein and pulmonary artery to evaluate pO 2 , PCO 2 , SO 2 , haemoglobin, lactate and electrolyte levels (Radiometer Medical ApS, ABL800 FLEX, Brönshöj, Denmark). Parameters that were registered and calculated for results are MAP, RBF, renal vascular resistance (RVR), RDO 2 , CBF, cardiac output (CO), heart rate (HR), renal venous oxygen saturation (RVSO 2 ), renal venous lactate (RLac), renal oxygen consumption (RVO 2 ), renal oxygen extraction rate (RO 2 ER), creatinine clearance (Cr-C), plasma creatinine (P-Cr), urine output and urinary sodium excretion (USE). The cardiovascular pressures CVP, PAP, PWP was only monitored to confirm heart and pulmonary function.
Statistical analysis. All data are displayed as mean ± 95% confidence interval. The software Statistica (Stat-Soft, Uppsala, Sweden) was used for statistical analysis. The repeated measurements analysis of variance with 2 levels of between-subject factors and 4 levels of within-subject factors (RM 2 × 4 ANOVA) was used. Corrections for multiple comparisons were made with specific contrasts as posthoc analysis.

Data availability
All data generated or analysed during this study are included in this published article (and its Supplementary  Information files). Pigs were sedated, surgically prepared and then allowed to recover. Baseline measurements were performed before infusion of either losartan or vehicle. The infusion continued for 60 min before hypotension was induced by administering propofol (30 mg*kg −1 ) continuously over 10 min. After the infusion of propofol was discontinued the animals were followed for an additional 60 min. The protocol was then ended, and pigs were euthanised with intravenous bolus of potassium chloride.