Hypertension, retinopathy, and acute kidney injury in dogs: A prospective study

Abstract Background Systemic hypertension (SH) is a potential complication of acute kidney injury (AKI) in dogs. Objective To describe the prevalence of SH and hypertensive retinopathy in dogs with AKI, to investigate the relationship between SH and severity of AKI and to assess possible factors associated with SH. Animals Fifty‐two dogs with AKI. Methods Prospective observational study of dogs presenting to a tertiary referral center that fulfilled the International Renal Interest Society (IRIS) guidelines for the diagnosis of AKI. Systolic blood pressure measurement, urine protein/creatinine ratio (UPCR), urine output, presence of hypertensive retinopathy and fluid overload (FO), survival to discharge and duration of hospitalization were subsequently assessed. The prevalence of SH was calculated and the relationship between SH and recorded factors was examined by nonparametric statistics. Results The prevalence of SH (≥160 mm Hg) on admission or during hospitalization was 75% (39/52) and in 56% (22/39) of cases this was severe (≥180 mm Hg). Sixteen percent (7/43) of dogs had evidence of hypertensive retinopathy and 77% (24/31) dogs had UPCR >0.5. Forty‐two percent (22/52) dogs had FO on admission or during hospitalization. There was no association between SH and IRIS AKI grade, oligo/anuria, survival to discharge, duration of hospitalization or proteinuria. Dogs with FO on presentation were more likely to be hypertensive at admission compared to dogs without FO (P = .02). Dogs that did not survive to discharge were more likely to have FO (P = .007). Conclusions and Clinical Importance Systemic hypertension is common in dogs with AKI. Systemic hypertension might be associated with FO, which itself is associated with nonsurvival. Monitoring for SH and FO is therefore warranted in dogs with AKI.


| INTRODUCTION
Acute kidney injury (AKI) is defined as an acute and abrupt decrease in kidney function resulting in abnormal glomerular filtration rate (GFR), tubular function and urine production. 1 Systemic hypertension (SH) is a potential complication of renal injury with variable occurrence in both dogs and cats with chronic kidney disease (CKD) and AKI. [2][3][4][5] Systemic hypertension (SH) can lead to target organ damage (TOD), which includes hypertensive retinopathy, hypertensive encephalopathy, left ventricular hypertrophy, and progression of kidney disease. Hypertensive retinopathy occurs in both cats and dogs with CKD and hypertensive retinopathy and left ventricular hypertrophy occurs in dogs with glomerular disease secondary to leishmaniasis. 3,6,7 The cause or causes of SH in kidney disease have yet to be fully elucidated but theories include impaired excretion of sodium and subsequent volume overload, excessive activation of the renin-angiotensin-aldosterone system, stimulation of the sympathetic nervous system via activation of chemosensitive afferent fibers and increase in systemic vascular resistance secondary to endothelial dysfunction. 8,9 In dogs with CKD, there is a positive correlation between systolic blood pressure (SBP) and degree of proteinuria and both SH and proteinuria are associated with disease progression and reduced survival time. 3,10 The prevalence of SH in CKD in various studies ranges between 9% and 93%. 11 Based on retrospective studies 81% to 87% of dogs with AKI have SH. 4 The literature on the effect of SH on the outcome of animals with AKI is sparse. Glomerular filtration rate is significantly reduced in hypertensive dogs compared to nonhypertensive dogs. 12 However, in AKI in cats presence of SH has no effect on survival. 5 The primary aims of this observational study were to describe the prevalence of SH and hypertensive retinopathy in dogs with AKI. The secondary aims were to investigate the relationship between SH and severity of AKI and to assess possible factors associated with SH. The hypotheses were that SH would be common in dogs with AKI, but ocular TOD would be less frequently detected. We further hypothesized that there would be no association between SH, severity of AKI or survival to discharge. International Renal Interest Society (IRIS) guidelines for the diagnosis of AKI; known access to nephrotoxins, serum creatinine increase >0.3 mg/dL over a 48 hour period or serum creatinine >1.6 mg/dL with 1 or more of the following criteria; evidence of renal tubular injury on urine analysis (renal glucosuria with normoglycemia, proteinuria with an inactive sediment, urinary casts, or both), imaging findings suggestive of AKI or oliguria (urine output <1 mL/kg/hr) over 6 hours. 13 When available clinical records of the dogs were reviewed and dogs were excluded if there was any historical physical examination findings or previous clinicopathological data suggestive of CKD, including weight loss, polyuria, or both, and polydipsia greater than 4 weeks in duration, and previously documented azotemia with urine specific gravity <1.030. All dogs had renal ultrasound performed and dogs were excluded if diagnostic imaging findings were indicative of CKD. These included; small irregular kidneys and the presence of renal infarcts. Dogs were also excluded if they failed to have an initial SBP or serum creatinine concentration measurement, had a coexisting disease associated with SH (including hyperadrenocorticism, diabetes mellitus, pheochromocytoma), were on medication which could result in SH (glucocorticoids, ciclosporin, toceranib), or if they had been treated with antihypertensive drugs such as angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, beta blockers, calcium channel blockers, or acepromazine in the 48 hours before enrolment. Systolic blood pressure was measured within 12 hours of admission and at least once daily with Doppler sphygmomanometry. A standardized protocol adapted from the American College of Veterinary Internal Medicine (ACVIM) consensus guidelines was used; the dog was allowed to acclimatize for 5 to 10 minutes before placement of a cuff (with cuff width being 30%-40% of the circumference of the limb at the cuff site) on either the forelimb or hindlimb, at the discretion of the operator. The first measurement was discarded and a total of 5 to 7 consistent values were taken, discarding any SBP reading >20% than the other SBP measurements. The average of these readings was recorded. 11 Based on criteria in the ACVIM guidelines dogs were grouped based on their SBP and risk of TOD; normotensive (SBP < 140 mm Hg), prehypertensive (SBP 140-159 mm Hg), hypertensive (SBP 160-179 mm Hg) and severely hypertensive (SBP ≥180 mm Hg. 11 Systemic hypertension was defined as a SBP≥160 mm Hg. Where individual dogs had more than 1 series of blood pressure readings performed on a given day, a median daily SBP was used for analysis. Antihypertensive therapy was at the discretion of the attending clinician.

| MATERIALS AND METHODS
Fundic examination was performed once within 48 hours of admission by an American or European specialist in Ophthalmology or a supervised resident in training to assess for presence of hypertensive retinopathy. 11 Pupil dilatation was performed at the discretion of the ophthalmologist on a case-by-case basis. The ophthalmologists were unaware of the dog's SBP.
Urine output (UOP) was quantified either by measuring voided urine, weighing bedding, or urethral catheterization. Oliguria was defined as a urine output less than 1 mL/kg/hr for >6 hours. 2 Fluid overload (FO) was diagnosed by the attending clinicians who determined this based on a daily dog assessment including monitoring for acute weight gain, serous nasal discharge, chemosis, subcutaneous edema or detection of cavitatory fluid with ultrasonography. 14 Proteinuria was defined as UPCR >0.5 with an inactive sediment. 15 Hypoalbuminemia was defined as serum albumin concentration < 2.6 g/dL.
The IRIS grade of AKI, serum albumin concentration and UPCR on admission were recorded. SBP, UOP and evidence of FO were reviewed daily. Any change in AKI grade or SH classification during hospitalization was recorded, alongside the use of antihypertensive agents and any extracorporeal therapy. Hospitalization duration, survival to discharge, survival at 3 months and follow-up serum creatinine concentration and SBP where available were documented. The prevalence of SH and hypertensive retinopathy were calculated. Descriptive statistics were used to evaluate population characteristics. A Fisher's exact test was used to compare categorical data including number of dogs in each SBP category, AKI grade, presence of FO, proteinuria and hypoalbuminemia, and survival. A Bonferroni correction was used when comparing multiple categories. A Spearman's rank correlation coefficient was used assess for correlation between SBP, serum creatinine concentration and hospitalization duration. The level of statistical significance was set at P < .05 and adjusted as needed when using the Bonferroni correction.

| RESULTS
Fifty-six dogs presented with AKI between July 2016 and November 2018, of which 4 were excluded; 1 with prior history of antihypertensive therapy and 3 with incomplete data. Fifty-two dogs were eligible for inclusion in the study. Of these dogs 49 were pure breeds; the most common pure breed was the Labrador (11/49) and 5/49 were sighthounds. Three out of 52 dogs were cross-breeds. Nineteen (37%) were male neutered, 18/52 (35%) female neutered, 7/52 (14%) female entire and 8/52 (15%) male entire. The median age and mean weight of dogs were 57 months (range 3-120) and 24.17 kg ±12.51, respectively. Dogs presented with a variety of causes of AKI (Table 1).
The median peak SBP at any time point was 175 mm Hg (range 90-300 mm Hg). The prevalence of SH at some point during hospitalization was 75% (39/52) which was severe in 56% cases (22/39) ( Table 2). Forty-three out of 52 dogs had a fundic examination during hospitalization and 21% (9/43) of these had evidence of ocular TOD.  dogs that had serum creatinine measured within the 3 month followup period after discharge had a serum creatinine above the upper limit of the laboratory defined value of >1.63 mg/dL. All of these dogs were hypertensive at some point during hospitalization. Four out of 5 dogs that had blood pressure measurements between 2 and 12 week after discharge were found to be hypertensive and on therapy.

| DISCUSSION
This prospective study shows that the prevalence of SH is high in dogs with community acquired AKI and appears to increase during hospitalization; 54% of dogs were hypertensive on presentation and 43% had severe hypertension. These values increased to 75% and 56% respectively during hospitalization. These results are similar to the currently reported SH occurrence rate of 81% to 87% in dogs with AKI during hospitaliszation. 4 There is therefore a clear requirement for frequent blood pressure monitoring of animals hospitalized with AKI.
Hypertensive retinopathy was detected in 16% dogs in this study.
This is less than reported in a study of all causes of hypertension, which report a prevalence of 62%. 16 However, the prevalence is similar to that previously described in dogs with CKD. 3 The low preva-   There was no association between the presence of SH and IRIS AKI grade or absolute serum creatinine concentration. This is similar to findings in cats with AKI and in dogs with cardiorenal syndrome. 5,24 These finding suggests SH can occur at all grades of AKI. Interestingly, canine studies have shown a reduction in GFR correlates with an increase in SBP. 10,12 In an experimental canine study, hypertensive dogs had a significantly reduced GFR and increase in renal tubular lesions and fibrosis compared to normotensive dogs in the weeks to months after AKI, suggesting SH has appreciable adverse effect on kidney structure and function 12 In the current study SH was treated at the clinician's discretion and hypertensive category reduced in around half of the dogs. This intervention could have masked the potential effect of SH on disease progression and outcome. Furthermore, the number of cases loss to follow-up was high and up sampling was not standardized reducing the study's power to detect the longterm effect of SH on kidney function. Finally, the IRIS guidelines used in this study for grading AKI are based on serum creatinine concentrations alone which is not considered an accurate indicator of GFR. 13,25 Future studies should therefore consider the routine use of commercial measures of GFR at standardized follow-up time points.
There was no association between SH at admission or development during hospitalization and survival nor between AKI grade and survival to discharge. The failure to detect an association between SH and survival to discharge is similar to that reported in cats with AKI. 5 Dogs enrolled had no pretreatment with antihypertensive agents and therefore we can reliably assess the effect of previously untreated SH on admission with disease severity and outcome. However, during hospitalization antihypertensive therapy was administered at the clinician's discretion and in around half of treated dogs their hypertension resolved. It is therefore difficult to elucidate the effect of persistent hypertension on outcome in this study. It is plausible that antihypertensive therapy masked the effect of SH on survival. The lack of association between severity of AKI and outcome is surprising and contrary to previous studies of AKI whereby increased AKI grade was associated with worsened outcome. 23,26,27 This difference could be explained by differences in causes of AKI in our population versus others and the over-reaching effect of etiology on outcome in AKI irrespective of absolute serum creatinine concentration. The study's small size, particularly the number of dogs in lower AKI grades, might also contribute to failure to detect an effect of AKI grade and outcome.
Of those dogs which survived and for which information was available, 50% were azotemic on 1 or more occasion after discharge.
All of these dogs were hypertensive at some point during hospitalization. Inferences on the effect of SH on development of CKD cannot be made given the small number of dogs, but this supports the need for regular monitoring of dogs after an episode of AKI. Only 5 dogs had follow-up blood pressure, of which 4/5 were persistently hypertensive and on antihypertensive therapy. Monitoring blood pressure after discharge is particularly important considering human AKI dogs are more likely to develop hypertension during follow-up. 28 Despite the frequent occurrence of SH in dogs with kidney disease the pathogenesis is poorly understood. Suggested etiologies include volume impaired excretion of sodium leading to volume overload, excessive activation of the renin angiotensin-aldosterone system, stimulation of the sympathetic nervous system secondary to activation of chemosensitive afferent neurons by local ischemia and inflammation, reduced bioavailability of the vasodilator nitric oxide and increased production of the vasoconstrictor endothelin. 8,9 In this study the presence of SH at admission was associated with the presence of FO at admission suggesting a role for volume and sodium excess in the pathogenesis of hypertension in this population. It is unclear why this relationship did not persist during hospitalization but the authors hypothesize it is related to the duration of FO; during hospitalization daily frequent monitoring of weight encourages early detection of FO. This might prompt the de-escalation of fluid therapy before the development of clinically detectable volume overload. In human dialysis, dogs' volume status, particularly those with FO are associated with both predialysis and postdialysis blood pressure.
Blood pressure falls during hemodialysis with fluid removal and the decrease in blood pressure is greatest with larger amounts of fluid removal and with higher ultrafiltration rates. 29  Situational hypertension, an increase in blood pressure as a result of adrenergic stimulation during situations of stress or anxiety, is difficult to control in any clinical study. 11 Situational hypertension has been documented in hospitalized dogs, in particular greyhounds. 31,33,34 Furthermore, greyhounds have significantly higher SBP than other breeds. 35 In this study, there were 5 sighthounds and 3/5 were considered hypertensive, of which 2 had SBP over 180 mm Hg. Although in the majority of studies the increase in blood pressure in the hospital was marginal, values up to 200 mm Hg are reported and therefore the presence of situational hypertension could lead to misclassification of dogs in all hypertensive categories. 35,36 Despite hospital protocols being in place to minimize the effect of stress on the blood pressure readings and following ACVIM guidelines 11 to measure blood pressure, the prevalence of SH in this study might have been increased by dogs with solely situational hypertension. Continuous blood pressure monitoring would be required to fully mediate the effect of situational hypertension.
Doppler sphygmomanometery is a recognized indirect measure of blood pressure. However, indirect blood pressure measurements have been shown to underestimate SBP in the hypertensive dog and therefore using this methodology might have underestimated the true prevalence of SH. 37 The gold standard technique for measurement of blood pressure is direct arterial catheterization, but is not practical in this subset of dogs. Telemetric blood pressure measurement could be useful both during hospitalization and after discharge to reliably detect SH, TOD and the effect of treatment in dogs with AKI. 38 In summary, SH is common in dogs with AKI. However, hypertensive retinopathy appears to be uncommon. Systemic hypertension can occur at all grades of AKI and therefore SBP should be monitored in all dogs with AKI irrespective of severity. Systemic hypertension on admission does not appear to affect outcome, however, in light of the routine use of antihypertensives the effect of persistent SH on outcome cannot be fully elucidated from this study. The relationship between FO and SH needs to be further explored but considering FO was associated with worsened outcome measures should be taken to monitor for, and prevent FO in dogs with AKI.

The authors acknowledge the Queen Mother Hospital for Animals
Ophthalmology team for their help with this study.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
Authors declare no IACUC or other approval was needed.