Utility of urinary alkaline phosphatase and γ-glutamyl transpeptidase in diagnosing acute kidney injury in dogs

The Veterinary Journal 220 (2017) 43–47

Contents lists available at ScienceDirect

The Veterinary Journal j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / t v j l

Utility of urinary alkaline phosphatase and γ-glutamyl transpeptidase in diagnosing acute kidney injury in dogs Ran Nivy *, Yochai Avital, Itamar Aroch, Gilad Segev Koret School of Veterinary Medicine, Hebrew University of Jerusalem, Rehovot 761001, Israel

A R T I C L E

I N F O

Article history: Accepted 15 December 2016 Keywords: Canine Renal failure Acute kidney injury Urinary alkaline phosphatase γ-Glutamyl transpeptidase

A B S T R A C T

The diagnostic utility of urinary alkaline phosphatase (uALP) and γ-glutamyl transpeptidase (uGGT) activities in naturally occurring acute kidney injury (AKI) was investigated in a heterogeneous group of dogs. The study included client-owned dogs with AKI (n = 32), chronic kidney disease (CKD, n = 13), lower urinary tract infection (LUTI, n = 15) and healthy controls (n = 24). uGGT and uALP activities were normalised to urinary creatinine (uCr) concentrations (uGGT/uCr and uALP/uCr, respectively). uALP/uCr and uGGT/uCr were positively and significantly correlated (r = 0.619, P < 0.001), and differed significantly (P ≤ 0.001) among groups, as well as between AKI and LUTI or CKD groups (P < 0.05), but not between the AKI and control groups. Areas under the receiver operator characteristics (ROC) curve for uALP/uCr and uGGT/uCr as predictors of AKI were 0.75 and 0.65, respectively, with optimal cut-off points showing poor to moderate sensitivity (59% for uALP/uCr and 79% for uGGT/uCr) and specificity (59% for uALP/uCr and 75% for uGGT/uCr). Higher cut-off points, with 90% specificity, showed low sensitivity (41% for both uALP/uCr and uGGT/uCr). In conclusion, uALP/uCr is superior to uGGT/uCr as a marker of AKI, but both uGGT/uCr and uALP/uCr have unsatisfactory discriminatory power for diagnosing naturally occurring AKI in dogs and therefore cannot be recommended as sole screening tests for canine AKI. However, both may serve as ancillary, confirmatory, biomarkers for detecting AKI in dogs if appropriate cut-off points with high specificities are used. © 2016 Elsevier Ltd. All rights reserved.

Introduction Acute kidney injury (AKI) in dogs is associated with high morbidity and mortality (Ross, 2011). Early recognition of AKI is pivotal for slowing and potentially reversing progression to overt renal failure (Cowgill and Langston, 2011; Ross, 2011; Kellum et al, 2013). Glomerular filtration rate (GFR) is considered to be the reference standard for assessing global kidney function (Von Hendy-Willson and Pressler, 2011). Due to technical aspects and limited availability of various methods for measuring GFR, serum creatinine concentration (sCr) commonly serves as a surrogate GFR marker (Braun et al., 2003; Von Hendy-Willson and Pressler, 2011). However, sCr has several shortcomings (Braun et al., 2003; Hokamp and Nabity, 2016): (1) sCr is affected by muscle mass and thus varies among healthy dogs of different breeds; (2) sCr is not expected to increase above the reference interval (RI) until 75% of kidney function is lost; (3) sCr does not represent the severity of renal damage until a steady state is reached; and (4) sCr reflects GFR changes, rather than tubular damage per se.

* Corresponding author. E-mail address: [email protected] (R. Nivy). http://dx.doi.org/10.1016/j.tvjl.2016.12.010 1090-0233/© 2016 Elsevier Ltd. All rights reserved.

Readily available kidney function and tubular injury markers, including urine specific gravity, glycosuria and cylinduria, lack sensitivity or specificity (Cowgill and Langston, 2011). Other urinary biomarkers may serve as early indicators of tubular injury, even before GFR changes or azotaemia occur (Cobrin et al., 2013; De Loor et al., 2013; Hokamp and Nabity, 2016). Amongst these, urinary enzyme activities are specific indicators of renal tubular injury, since their molecular size precludes glomerular filtration and their urinary excretion increases following tubular injury (Clemo, 1998; D’Amico and Bazzi, 2003; Cobrin et al., 2013; De Loor et al., 2013). The diagnostic and prognostic value of urinary enzyme activities has been demonstrated in dogs and human beings with AKI (D’Amico and Bazzi, 2003; Cobrin et al., 2013; De Loor et al., 2013). Measurement of urinary alkaline phosphatase (uALP) and γ-glutamyl transpeptidase (uGGT) activities is simple, widely available and cost-efficient. Both are brush border enzymes located primarily in the metabolically active proximal renal tubule (Guder and Ross, 1984; Clemo, 1998). With their high molecular weight, the urinary activity of these enzymes is primarily considered to be of tubular origin rather than derived from the glomerular filtrate (Heiene et al., 1991; Clemo, 1998). Most studies of uGGT and uALP in dogs are limited to small cohorts of single aetiologies of AKI (Ellis et al., 1973; Adelman et al., 1979; Greco et al., 1985; De Schepper et al., 1989; Rao et al., 1990; Uechi et al., 1994a; Grauer et al., 1995;

44

R. Nivy et al. / The Veterinary Journal 220 (2017) 43–47

Rivers et al., 1996; Palacio et al., 1997; Clemo, 1998; Lobetti and Lambrechts, 2000; Lobetti and Joubert, 2000; Heiene et al., 2001; Raekallio et al., 2006; Palviainen et al., 2013; Ibba et al., 2016). Only the study by Heiene et al. (1991) evaluated their use in naturallyoccurring canine AKI. We investigated the clinical utility of uALP and uGGT in the diagnosis of naturally occurring AKI in a larger, more diverse, cohort of dogs. This study had two main aims: (1) to assess the utility of these biomarkers in differentiating AKI from other urinary tract diseases, specifically CKD, because of the prognostic implications of such a distinction; and (2) to serve as a preliminary study for further research regarding the use of these biomarkers in the recognition of early AKI, prior to the development of clinical signs and biochemical perturbations. Materials and methods Animals and study design The study was conducted at the Hebrew University Veterinary Teaching Hospital, Koret School of Veterinary Medicine, Israel, and approved by the Institutional Animal Care and Ethics Committee (approval number KSVM-VTH/20-2015; date of approval 2 November 2015). Client-owned dogs were prospectively enrolled, and divided into three groups: (1) AKI; (2) chronic kidney disease (CKD); and (3) lower urinary tract infection (LUTI). Controls included staff-owned dogs and dogs presented for elective neutering, which were deemed to be healthy on the basis of their history, physical examination and complete blood count findings. Azotaemia was an exclusion criterion. Diagnosis of LUTI was based on compatible clinical signs, and confirmed by urinalysis and a positive urine culture. Dogs were excluded from this group if they presented with azotaemia or ultrasonographic findings indicative of ascending urinary tract infection (Choi et al., 2010). CKD and AKI were diagnosed based on the International Renal Interest Society (IRIS) guidelines and grading system. The aetiology of AKI was recorded when identified. Collection of samples and laboratory methods Blood specimens for serum chemistry were collected as part of the basic clinical diagnostic work-up, in tubes containing no anticoagulant, with gel-separators. The blood was allowed to clot, then centrifuged, and the serum was separated and analysed within 60 min of collection. Urine samples were obtained as part of the basic clinical diagnostic work-up by cystocentesis and analysed within 30 min of collection. Biochemistry was determined using a wet chemistry analyser (Cobas Integra 400 Plus, Roche; 37 °C). uALP and uGGT activities were normalised to urinary creatinine concentration (uCr), i.e. uALP/uCr and uGGT/uCr, respectively.

Results The study included client-owned dogs with naturally-occurring AKI (n = 32 dogs), CKD (n = 13) or LUTI (n = 15), and healthy control dogs (n = 24), with median ages of 96 (range 7–168), 42 (range, 12– 162), 108 (range 24–180) and 10 (range 3–216) months, respectively. The median age of the control group was lower than for other groups (P < 0.05). Intact males and females constituted the majority of dogs in the AKI, CKD, LUTI and control groups (65% and 19%, 53% and 38%, 57% and 42%, and 8% and 79%, respectively). Median sCr concentrations and the IRIS classification of dogs in the AKI and CKD groups are shown in Table 1. Specific aetiologies were identified in 11/32 AKI cases and included pancreatitis, previous anaesthesia and snakebite (two cases of each), and pyometra, trauma, carprofen intoxication, cycad intoxication and leptospirosis (one each). uALP/uCr and uGGT/uCr varied considerably within groups (Fig. 1). Dogs with AKI had higher uALP/uCr, with a median of 0.01 and range of 0–6.39, expressed in (U/L)/(mg/dL), than dogs with CKD (median 0.01, range 0–0.31) or LUTI (median 0, range 0–0.07), but not control dogs (median 0.058, range 0.005–0.48). Dogs with AKI had higher uGGT/uCr, with a median of 0.40 and range of 0–31.4, expressed in (U/L)/(mg/dL), than dogs with CKD (median 0.025, range 0–2.2) or LUTI (median 0, range 0–2.1), but not control dogs (median 0.30, range 0.12–0.76). uALP/uCr and uGGT/uCr were significantly and positively correlated (r = 0.62, P < 0.001). The agreement between uALP/uCr and uGGT/uCr for each dog (when classified either as above or within/below reference interval) was 83%. Only dogs with AKI had uALP/uCr values above the reference interval, whilst increased uGGT/uCr was also noted in dogs with CKD (1/13 dogs) and LUTI (3/15 dogs), albeit not as commonly as in the AKI group (11/ 32 dogs). The ROC curve AUCs (95% CI) for uALP/uCr and uGGT/uCr as predictors of AKI were 0.75 (0.64–0.86) and 0.65 (0.53–0.78), respectively, with sensitivity and specificity for optimal cut-offs of 59% and 79%, and 59% and 75%, respectively. Cut-offs with 90% specificity had low sensitivities (41% for both; Table 2). ROC analyses for uALP and uGGT activities, not normalised to uCr, had inferior diagnostic performance as predictors of AKI (AUC 0.60 and 0.49, respectively). Discussion

Statistical analysis The distribution of continuous variables was assessed using the Shapiro–Wilk test. The non-parametric Kruskal–Wallis test was subsequently used to compare continuous variables among groups, with pair-wise Mann–Whitney U test post-hoc comparisons and Bonferroni’s correction. Spearman’s correlation test was used to investigate the association between uALP/uCr and uGGT/uCr. The reference interval was calculated based on the 2.5th to 97.5th inter-percentile ranges of the control group results. The uALP/uCr and uGGT/uCr results of each dog were later classified as above or within/below this proposed reference interval. The receiver operator characteristic (ROC) analysis, with its area under the curve (AUC) and 95% confidence interval (CI), was used to assess uALP/uCr and uGGT/uCr as predictors of AKI. The Youden index (J) was used to locate the optimal cut-off point (Perkins and Schisterman, 2006). All tests were two-tailed and a P value < 0.05 was considered to be significant. Statistical analyses were performed using SPSS 22.0.

This study investigated uGGT and uALP in a relatively large heterogeneous group of dogs with naturally acquired AKI. In contrast with a previous study (Heiene et al., 1991), uGGT/uCr and, to a lesser extent, uALP/uCr demonstrated unsatisfactory discrimination between dogs with AKI and dogs with other urinary tract conditions and healthy controls. Kidney damage is a continuum, with the initial injury progressing to kidney dysfunction and potentially culminating in kidney failure and overt azotaemia (Basile et al., 2012). Early diagnosis of AKI might allow timely intervention and improved prognosis (Basile et al., 2012). Since sCr is inadequate for detecting early AKI, urinary

Table 1 Median (range) serum creatinine concentrations among the study groups and the distribution of cases in the acute kidney injury and chronic kidney disease groups, based on International Renal Interest Society (IRIS) grading and staging guidelines. Acute kidney injury Median sCra (range) IRIS grade/stageb %

I 3

II 6

6.9 (1.5–22.0) III IV 25 38

V 28

1 8

Chronic kidney disease

Lower urinary tract infection

Control

4.4 (1.3–11.7) 2 3 23 31

0.66 (0.46–1.35) NA

0.7 (0.3–1.2) NA

4 38

NA, not applicable. a Serum creatinine (mg/dL). b IRIS staging of chronic kidney disease (http://iris-kidney.com/guidelines/staging.html; accessed 21 August 2016) and grading of acute kidney injury (http://www.iris-kidney.com/pdf/grading-of-acute-kidney-injury.pdf; accessed 21 August 2016).

R. Nivy et al. / The Veterinary Journal 220 (2017) 43–47

45

Fig. 1. Box and whisker plots of urinary γ-glutamyl transpeptidase (uGGT)/urinary creatinine (uCr), in (U/L)/(mg/dL), and urinary alkaline phosphatase (uALP)/uCr, in (U/L)/(mg/dL) in healthy dogs (controls, n = 24), dogs with acute kidney injury (AKI, n = 32), dogs with chronic kidney disease (CKD, n = 13) and dogs with lower urinary tract infection (LUTI, n = 15). Boxes represent 25th to 75th interquartile ranges, solid horizontal lines through boxes represent medians and whiskers represent 5th and 95th percentiles. Outliers are plotted separately as dots.

tubular injury biomarkers have been investigated (Braun et al., 2003; Cobrin et al., 2013; De Loor et al., 2013). Low molecular weight proteins, as well as tubular enzymes and proteins, are believed to represent tubular dysfunction or injury due to inefficient resorption of the former or leakage of the latter (Clemo, 1998; D’Amico and Bazzi, 2003; De Loor et al., 2013). Urinary markers might prove superior to conventional laboratory analytes for early detection of tubular damage, as demonstrated in studies of aminoglycoside-induced AKI in dogs, where increased uGGT preceded changes in urine specific gravity and sCr by 3 and 5 days post-injury, respectively (Rivers et al., 1996). Notwithstanding their promise, urinary biomarkers have several limitations. Firstly, some assays have not been standardised and validated in dogs (Cobrin et al., 2013; De Loor et al., 2013). Secondly, the effect of extra-renal diseases on urinary biomarker levels has not always been thoroughly investigated; for example, the effect of albuminuria on urinary cystatin C concentration has not been adequately addressed in veterinary studies (Cobrin et al., 2013; De Loor et al., 2013). Similarly, neutrophil gelatinase-associated lipocalin was increased in dogs in intensive care units without overt AKI and in dogs with lower urinary tract diseases, raising questions about its specificity (Decavele et al., 2011; Cortellini et al., 2015). Finally, some urinary biomarkers are cost prohibitive and not commercially available, for example N-acetyl-β-d-glucosaminidase (NAG) and Kim1, and therefore cannot facilitate real time decision making (De Loor et al., 2013). Since measurement of uGGT and uALP is performed using routine serum ALP and GGT assays, it is readily available, cost efficient and

Table 2 Receiver operating characteristic (ROC) analyses for uGGT/uCr and uALP/uCr as predictors of naturally occurring acute kidney injury in dogs.

uGGT/uCra

uALP/uCrb

Cut-off point

Sensitivity (%)

Specificity (%)

AUC (95% confidence interval)

0.0008 0.3442* 0.7228 0.0124 0.0852* 0.2003

84 59 41 87 59 41

27 75 90 42 79 90

0.654 (0.526–0.783)

0.749 (0.640–0.858)

AUC, area under the curve. a Urinary γ-glutamyl transpeptidase/urinary creatinine: (U/L)/(mg/dL). b Urinary alkaline phosphatase/urinary creatinine: (U/L)/(mg/dL). * Optimal cut-off point, with least misclassifications.

potentially useful in guiding therapy in practice. A number of studies have shown marked increases in the urinary activity of one or both of these enzymes in experimental canine AKI models (Ellis et al., 1973; Adelman et al., 1979; Greco et al., 1985; Rao et al., 1990; Uechi et al., 1994a; Grauer et al., 1995; Rivers et al., 1996; Clemo, 1998; Lobetti and Lambrechts, 2000; Heiene et al., 2001; Palviainen et al., 2013; Ibba et al., 2016). In particular, uGGT has been investigated extensively in aminoglycoside-induced AKI in dogs, in which it has been shown to be a sensitive, early urinary biomarker (Greco et al., 1985; Rivers et al., 1996). However, experimental studies are mostly concerned with the early initiation and extension stages of AKI, whereas in the clinical setting dogs often present in the later stages of tubular injury, when severe GFR impairment and extensive tubular necrosis have already occurred (Malyusz and Braun, 1981; Greco et al., 1985; Rivers et al., 1996). In the clinical setting, limited information exists regarding the utility of uGGT and uALP for specific AKI aetiologies. In one study, elevations in uGGT, uALP and NAG activities were noted in 13/55 bitches with pyometra at presentation. Higher urinary enzyme activities were associated with proximal tubule histological changes, which normalised within 2 weeks post-operatively (Heiene et al., 2001). In another study, uGGT/uCr and uALP/uCr increased in dogs with European adder envenomation (Palviainen et al., 2013). uGGT activity is also increased in dogs with leishmaniasis and heartworm disease, in the absence of azotaemia (Uechi et al., 1994a; Palacio et al., 1997). Additional studies have investigated the effect of long-term carprofen administration, as well as anaesthesia and surgery, on uGGT/uCr and uALP/Ucr excretion in healthy dogs (Lobetti and Lambrechts, 2000; Lobetti and Joubert, 2000; Raekallio et al., 2006). However, these studies are limited either by the homogeneity of the cohort or by their experimental nature. In contrast, selecting a heterogeneous group of dogs with naturally occurring AKI in the present study more adequately reflects the utility of these biomarkers in the clinical setting, notwithstanding the greater variability it inherently entails. The study design, with the inclusion of dogs with CKD, dogs with LUTI and healthy control dogs, allows for testing the discriminatory ability and accuracy of both biomarkers in detecting AKI, but it excludes dogs with early tubular injury, in the absence of clinical signs and biochemical aberrations. This is a major limitation of the study, since these dogs constitute the main target population to be screened using sensitive biomarkers. To the best of the authors’ knowledge, only one study has previously investigated the use of uALP and uGGT in the diagnosis of AKI of

46

R. Nivy et al. / The Veterinary Journal 220 (2017) 43–47

various aetiologies compared to CKD and healthy dogs, but the cohort size was limited to 10 dogs with AKI (Heiene et al., 1991). A considerably larger number of dogs were presently enrolled in the AKI group, thereby improving the diversity of the aetiologies and the statistical power. Overall, the diagnostic utility of both biomarkers in detecting AKI was relatively low. Despite statistically significant differences in uALP/ uCr and uGGT/uCr among the different study groups, with the exception of the AKI and control groups, which did not differ, there was marked intra-group variability and significant inter-group overlap. Overlapping uGGT/uCr measurements between dogs with naturally occurring AKI or CKD and healthy control dogs has been reported previously by Heiene et al. (1991), but there was minimal uALP/uCr overlap between groups, in contrast with the present findings. These discrepancies could be due to several reasons. Firstly, lack of standardisation of assays might prevent comparison of results across studies. Secondly, conflicting data exist regarding diurnal fluctuations in enzymuria. Although spot uGGT/uCr measurement is correlated with its 24 h excretion (Gossett et al., 1987; Uechi et al., 1994b; Grauer et al., 1995), within-day fluctuations do occur (Gossett et al., 1987) and these might affect urinary enzyme excretion. To the best of the authors’ knowledge, such a correlation has not been reported for uALP in dogs. Thirdly, several in vitro and in vivo factors affect uALP and uGGT activity. Pyuria and haematuria, commonly seen in dogs with LUTI, may interfere with uGGT measurement (Clemo, 1998). Urine pH values < 5 may spuriously decrease uGGT and uALP activities (Jung et al., 1983). In one study, uGGT activity in normal dogs differed significantly when the data were analysed in relation to urine pH (<7 versus ≥ 7) (Brunker et al., 2009). Unfortunately, urine pH was unavailable for comparisons in the present study. ALP activity is present in the prostatic fluid and epididymis of dogs (Frenette et al., 1986), which might partly account for increased uALP activity in intact male dogs in the present study; male dogs constituted a major proportion of animals in the AKI, CKD and LUTI groups, but not the control group. Fourthly, although the proximal renal tubule is the most affected nephron segment in AKI, different aetiologies might affect different segments and thus may show different enzyme leakage patterns. Fifthly, the severity of injury is likely to be different among studies. Finally, enzyme inhibitors and excreted drugs might also interfere with their urinary activity (Clemo, 1998). Enzymuria might diminish with progression of tubular damage and with time, since tubular enzyme reserves may become depleted and enzyme secretion subsequently decreases, as demonstrated for uGGT in a murine model of AKI (Malyusz and Braun, 1981). This might partly account for the inconsistent elevations in uGGT and uALP activity in the AKI population in our study. On the other hand, transient kidney injury might only result in short term increases in uGGT activity (Lobetti and Lambrechts, 2000), and decreases in uALP and uGGT activities might reflect recovery from tubular injury (Clemo, 1998). Thus, the timing of uALP or uGGT measurement during different stages of tubular injury might affect their activity and sensitivity for detecting AKI. Although increased uGGT activity has been shown to persist for ≥5 days in aminoglycosideinduced AKI in dogs, experimental studies or studies of naturallyoccurring AKI with longitudinal monitoring of urinary enzyme activity over longer periods are scarce (Rivers et al., 1996). To overcome the effect of urine concentration and diuresis, uALP and uGGT activities were normalised to uCr and these normalised values demonstrated superior diagnostic performance compared to the non-normalised values. However, normalisation to uCr entails several limitations, including large inter-individual uCr excretion variations, which occur even in healthy dogs and are aggravated during unstable AKI (Rivers et al., 1996). Normalisation to a

pre-defined urine dilution, or 24 h urine collection, might overcome these limitations. Most AKI cases enrolled in clinical studies, as well as in the present study, either have azotaemic, IRIS grade II–IV AKI or nonazotaemic IRIS grade I AKI, with compatible clinical signs, imaging findings, oliguria/anuria and documentation of progressive sCr increases, thereby negating the need for urinary biomarkers (Cobrin et al., 2013; De Loor et al., 2013). In other cases, stored urine samples, collected prior to development of azotaemia, were tested, but those samples were retrospectively retrieved only after these dogs had developed azotaemic AKI, thereby introducing a selection bias (Nabity et al., 2012; Segev et al., 2013). To assess the sensitivity and specificity of a biomarker in detecting early AKI prior to development of clinical signs, azotaemia or decreased urine production, urine samples should be collected prospectively from dogs at risk of developing AKI, while these animals are followed and monitored for development of AKI. This has been applied in the assessment of anaesthesia-related AKI in healthy dogs (Lobetti and Lambrechts, 2000; Lobetti and Joubert, 2000). Therefore, the present study in dogs with naturally-acquired AKI should be regarded as preliminary, providing a platform for future studies evaluating the potential utility of biomarkers in screening for early, occult, AKI in the clinical setting. Future studies should also investigate whether combining several urinary biomarkers improves their diagnostic performance, as demonstrated in human medicine (Han et al., 2009). The number of CKD and LUTI cases in this study was limited, thereby decreasing the statistical power. Additionally, the control dogs were younger than affected dogs, although the impact of age on urinary enzyme excretion is unclear. In human beings, both uGGT and uALP excretion increases with age (Trachtenberg and Barregard, 2007), but similar information in dogs is unavailable. In bitches with pyometra, there was a poor correlation between age and both uALP and uGGT (Heiene et al., 2001). Conclusions uGGT/uCr and uALP/uCr demonstrate unsatisfactory discriminatory power for diagnosing AKI in dogs and are unsuitable screening tests for AKI. uALP/uCr is superior to uGGT/uCr as a marker for AKI. Since measurement of both parameters is readily available and cost efficient, uGGT/uCr and uALP/uCr may serve as ancillary, confirmatory biomarkers for establishing a diagnosis of AKI in dogs, if appropriate, high specificity cut-off points are selected. Conflict of interest statement None of the authors of this paper has a financial or personal involvement with other people or organisations that could inappropriately influence or bias the content of the paper. Acknowledgements Preliminary results were presented as an Abstract at the 26th European College of Veterinary Internal Medicine – Companion Animals (ECVIM-CA) Congress, Goteborg, Sweden, 8–10 September, 2016. References Adelman, R.D., Spangler, W.L., Beasom, F., Ishizaki, G., Conzelman, G.M., 1979. Furosemide enhancement of experimental gentamicin nephrotoxicity: Comparison of functional and morphological changes with activities of urinary enzymes. Journal of Infectious Diseases 140, 342–352. Basile, D.P., Anderson, M.D., Sutton, T.A., 2012. Pathophysiology of acute kidney injury. Comparative Physiology 2, 1303–1353. Braun, J.P., Lefebvre, H.P., Watson, A.D., 2003. Creatinine in the dog: A review. Veterinary Clinical Pathology 32, 162–179.

R. Nivy et al. / The Veterinary Journal 220 (2017) 43–47

Brunker, J.D., Ponzio, N.M., Payton, M.E., 2009. Indices of urine N-acetyl-β-Dglucosaminidase and γ-glutamyl transpeptidase activities in clinically normal adult dogs. American Journal of Veterinary Research 70, 297–301. Choi, J., Jang, J., Choi, H., Kim, H., Yoon, J., 2010. Ultrasonographic features of pyonephrosis in dogs. Veterinary Radiology and Ultrasound 51, 548–553. Clemo, F.A., 1998. Urinary enzyme evaluation of nephrotoxicity in the dog. Toxicologic Pathology 26, 29–32. Cobrin, A.R., Blois, S.L., Kruth, S.A., Abrams-Ogg, A.C., Dewey, C., 2013. Biomarkers in the assessment of acute and chronic kidney diseases in the dog and cat. Journal of Small Animal Practice 54, 647–655. Cortellini, S., Pelligand, L., Syme, H., Chang, Y.M., Adamantos, S., 2015. Neutrophil gelatinase-associated lipocalin in dogs with sepsis undergoing emergency laparotomy: A prospective case-control study. Journal of Veterinary Internal Medicine 29, 1595–1602. Cowgill, L.D., Langston, C., 2011. Acute kidney insufficiency. In: Bartges, J., Polzin, D. (Eds.), Nephrology and Urology of Small Animals. Wiley-Blackwell, West Sussex, UK, pp. 472–523. D’Amico, G., Bazzi, C., 2003. Urinary protein and enzyme excretion as markers of tubular damage. Current Opinion in Nephrology and Hypertension 12, 639–643. De Loor, J., Daminet, S., Smets, P., Maddens, B., Meyer, E., 2013. Urinary biomarkers for acute kidney injury in dogs. Journal of Veterinary Internal Medicine 27, 998–1010. De Schepper, J., De Cock, I., Capiau, E., 1989. Urinary γ-glutamyl transferase and the degree of renal dysfunction in 75 bitches with pyometra. Research in Veterinary Science 46, 396–400. Decavele, A.S., Dhondt, L., De Buyzere, M.L., Delanghe, J.R., 2011. Increased urinary neutrophil gelatinase associated lipocalin in urinary tract infections and leukocyturia. Clinical Chemistry and Laboratory Medicine 49, 999–1003. Ellis, B.G., Price, R.G., Topham, J.C., 1973. The effect of tubular damage by mercuric chloride on kidney function and some urinary enzymes in the dog. ChemicoBiological Interactions 7, 101–113. Frenette, G., Dube, J.Y., Tremblay, R.R., 1986. Origin of alkaline phosphatase of canine seminal plasma. Archives of Andrology 16, 235–241. Gossett, K.A., Turnwald, G.H., Kearney, M.T., Greco, D.S., Cleghorn, B., 1987. Evaluation of γ-glutamyl transpeptidase-to-creatinine ratio from spot samples of urine supernatant, as an indicator of urinary enzyme excretion in dogs. American Journal of Veterinary Research 48, 455–457. Grauer, G.F., Greco, D.S., Behrend, E.N., Mani, I., Fettman, M.J., Allen, T.A., 1995. Estimation of quantitative enzymuria in dogs with gentamicin-induced nephrotoxicosis using urine enzyme/creatinine ratios from spot urine samples. Journal of Veterinary Internal Medicine 9, 324–327. Greco, D.S., Turnwald, G.H., Adams, R., Gossett, K.A., Kearney, M., Casey, H., 1985. Urinary γ-glutamyl transpeptidase activity in dogs with gentamicin-induced nephrotoxicity. American Journal of Veterinary Research 46, 2332–2335. Guder, W.G., Ross, B.D., 1984. Enzyme distribution along the nephron. Kidney International 26, 101–111. Han, W.K., Wagener, G., Zhu, Y., Wang, S., Lee, H.T., 2009. Urinary biomarkers in the early detection of acute kidney injury after cardiac surgery. Clinical Journal of the American Society of Nephrology 4, 873–882. Heiene, R., Biewenga, W.J., Koeman, J.P., 1991. Urinary alkaline phosphatase and 7-glutamyl transferase as indicators of acute renal damage in dogs. Journal of Small Animal Practice 32, 521–524. Heiene, R., Moe, L., Molmen, G., 2001. Calculation of urinary enzyme excretion, with renal structure and function in dogs with pyometra. Research in Veterinary Science 70, 129–137.

47

Hokamp, J.A., Nabity, M.B., 2016. Renal biomarkers in domestic species. Veterinary Clinical Pathology 45, 28–56. Ibba, F., Mangiagalli, G., Paltrinieri, S., 2016. Urinary γ-glutamyl transferase (GGT) as a marker of tubular proteinuria in dogs with canine leishmaniasis, using sodium dodecylsulphate (SDS) electrophoresis as a reference method. The Veterinary Journal 210, 89–91. Jung, K., Pergande, M., Schreiber, G., Schroder, K., 1983. Stability of enzymes in urine at 37°C. Clinica Chimica Acta 131, 185–191. Kellum, J.A., Lameire, N., KDIGO AKI Guideline Work Group, 2013. Diagnosis, evaluation, and management of acute kidney injury: A KDIGO summary (Part 1). Critical Care 17, 204. Lobetti, R., Lambrechts, N., 2000. Effects of general anesthesia and surgery on renal function in healthy dogs. American Journal of Veterinary Research 61, 121–124. Lobetti, R.G., Joubert, K.E., 2000. Effect of administration of nonsteroidal antiinflammatory drugs before surgery on renal function in clinically normal dogs. American Journal of Veterinary Research 61, 1501–1507. Malyusz, M., Braun, D., 1981. Enzymuria (the output of γ-glutamyl-transpeptidase and of N-acetyl-β-D-glucosaminidase) in the course of experimental renovascular hypertension. Enzyme 26, 32–42. Nabity, M.B., Lees, G.E., Cianciolo, R., Boggess, M.M., Steiner, J.M., Suchodolski, J.S., 2012. Urinary biomarkers of renal disease in dogs with X-linked hereditary nephropathy. Journal of Veterinary Internal Medicine 26, 282–293. Palacio, J., Liste, F., Gascon, M., 1997. Enzymuria as an index of renal damage in canine leishmaniasis. Veterinary Record 140, 477–480. Palviainen, M., Raekallio, M., Vainionpaa, M., Lahtinen, H., Vainio, O., 2013. Evaluation of renal impairment in dogs after envenomation by the common European adder (Vipera berus berus). The Veterinary Journal 198, 723–724. Perkins, N.J., Schisterman, E.F., 2006. The inconsistency of “optimal” cutpoints obtained using two criteria based on the receiver operating characteristic curve. American Journal of Epidemiology 163, 670–675. Raekallio, M.R., Hielm-Bjorkman, A.K., Kejonen, J., Salonen, H.M., Sankari, S.M., 2006. Evaluation of adverse effects of long-term orally administered carprofen in dogs. Journal of the American Veterinary Medical Association 228, 876–880. Rao, P.S., Cavanagh, D.M., Fiorica, J.V., Spaziani, E., 1990. Endotoxin-induced alterations in renal function with particular reference to tubular enzyme activity. Circulatory Shock 31, 333–342. Rivers, B.J., Walter, P.A., O’Brien, T.D., King, V.L., Polzin, D.J., 1996. Evaluation of urine γ-glutamyl transpeptidase-to-creatinine ratio as a diagnostic tool in an experimental model of aminoglycoside-induced acute renal failure in the dog. Journal of the American Animal Hospital Association 32, 323–336. Ross, L., 2011. Acute kidney injury in dogs and cats. Veterinary clinics of North America: Small Animal Practice 41, 1–14. Segev, G., Palm, C., LeRoy, B., Cowgill, L.D., Westropp, J.L., 2013. Evaluation of neutrophil gelatinase-associated lipocalin as a marker of kidney injury in dogs. Journal of Veterinary Internal Medicine 27, 1362–1367. Trachtenberg, F., Barregard, L., 2007. The effect of age, sex, and race on urinary markers of kidney damage in children. American Journal of Kidney Diseases 50, 938–945. Uechi, M., Nogami, Y., Terui, H., Nakayama, T., Ishikawa, R., Wakao, Y., Takahashi, M., 1994a. Evaluation of urinary enzymes in dogs with early renal disorder. Journal of Veterinary Medical Science 56, 555–556. Uechi, M., Terui, H., Nakayama, T., Mishina, M., Wakao, Y., Takahashi, M., 1994b. Circadian variation of urinary enzymes in the dog. Journal of Veterinary Medical Science 56, 849–854. Von Hendy-Willson, V.E., Pressler, B.M., 2011. An overview of glomerular filtration rate testing in dogs and cats. The Veterinary Journal 188, 156–165.