In the present study, we established an antigen capture ELISA for measuring urine NGAL in dogs. Using this platform, urine NGAL was shown to act as an early predictive biomarker for acute kidney injury (AKI) after surgery. Recently, several studies have proposed that NGAL ought to be able to serve as a novel biomarker for predicting renal failure in humans [11, 13, 14]. However, up to the present, measurement of urine NGAL levels has not yet been included as a routine diagnosis for the prediction of renal function. Moreover, studies targeting NGAL in dogs have not been reported up to the present and as a result one might expect the detection reagent/kit to have limited availability. To be able to monitor NGAL levels in many individuals, it is important to develop a reliable and cost effective detection system.
The detection and measurement of NGAL level can be achieved by a range of methods including Western blot analysis and ELISA. So far, antibodies specific for canine NGAL are not commercially available; most of the commercial NGAL antibodies in use are specific for the human protein and cross reactivity with canine NGAL has been suggested based on gene homology and sequence similarity. Although a canine NGAL ELISA assay kit is commercially available, it is very expensive. In this context, the production of an antibody against canine NGAL is useful when developing ELISA and other assays for measuring NGAL levels in dogs. In current study, we have identified and expressed recombinant canine NGAL and used this protein to generate two different polyclonal antibodies. Compared with the results obtained from the commercial NGAL ELISA kit, we noticed that the overall NGAL concentrations of tested samples were measured as higher when our ELISA was used. The difference observed between these assays is probably, to a large degree, due to differences in the ELISA format, to higher antibody affinity, to better amplification of the detection signal, and to improved assay standardization. Nevertheless, the linear regression analysis indicated that results of in-house ELISA are highly correlated with the commercial kit, and that the NGAL concentration can be converted between these two systems using an equation. Hence, our in-house ELISA has the potential to be a useful tool when studying NGAL in dogs.
To our knowledge, this is the first study to investigate the role of NGAL in dogs with renal disease. Results of this study indicate that, as with humans, an increase in urine NGAL concentration is able to predict kidney injury in dogs after surgery. Compared with using serum creatinine for the diagnosis of AKI at 48 h post surgery, an increase in urine NGAL from the baseline is able to pinpoint AKI as early as 12 h after surgery in dogs.
Noticeably, the 2 h urine NGAL level has been found to be an effective biomarker for the prediction of AKI in several human medical studies [12–14]. In our study, the best time point for classifying dogs into those with and without AKI was much later at 12 h post surgery. As urinary catheterization of each dog after surgery cannot be accomplished, the collection of urine from dogs at 2 h post surgery is not routinely possible; therefore we may have missed the earliest point at which urine NGAL is elevated. Furthermore, due to limited clinical accessibility, urine could not be obtained from all the dogs at all time points and this resulted in missing data, which reduced the statistical power of our study. However, this preliminary study still clearly shows that urine NGAL level is able to act as a much earlier biomarker than serum creatinine when detecting post surgery AKI in dogs. Nonetheless, more research is needed to validate NGAL levels as a biomarker in dogs with various renal diseases and in dogs at different stages of disease progression.
Unlike previous human medical studies [12–14], serum NGAL was not as sensitive as urine NGAL in dogs. The reasons for this may involve the expression of NGAL occurring not only in the renal tubules but also in the respiratory tract, stomach and colon; furthermore, NGAL expression is also increased by acute bacterial infection and tissue injury . Thus NGAL accumulates separately in two pools, namely a systemic and a renal pool . During AKI, serum NGAL may also be derived from injured tissues other than the kidneys; these will both contribute to the systemic pool. Hence, serum NGAL concentration is not in these circumstances a specific indicator of disease with a renal origin. Additionally, studies of NGAL in human AKI have involved specific types of major surgery such as cardiac surgery ; however, in our study the types of surgery that the dogs underwent were diverse. The diversity of the surgery may help to obscure the serum NGAL results compared to the urine NGAL results. In contrast to serum NGAL, an increase in urine NGAL only occurs when the renal absorption ability of NGAL is disturbed by the kidney damage .
Despite the fact that development of post-operative AKI occurs most often after cardiac surgery in humans, it also occurs after human non-cardiac surgery [1, 26]. Although the mechanism of post-surgery AKI is well understood, some risk factors such as age, emergency status, high risk surgery, ischemic heart disease , liver disease, body mass index and chronic obstructive pulmonary disease have been reported to predispose patients to suffer renal insult after surgery . In the veterinary field, post-surgery AKI in dogs has been rarely reported, but this may be due to the lack of the early biomarker and a consensus standard. In our study, based on the AKIN criteria and the elevation of urine NGAL, AKI in dogs can now be recognized as early as 12 h post surgery.
The use of nephrotoxic medicines might have interfered with the results of this study . Nevertheless, we found that the percentage of patients using nephrotoxic medicine in the AKI group was not significantly higher than in No AKI groups and thus it would seem that this variable can be neglected as an influence on our results.
Several criteria have been established to increase the sensitivity of AKI detection at the early stage; however, these criteria are based on the loss of renal function. As an applicable tool for testing renal function, serum creatinine was used as the standard for AKI. However, unlike serum creatinine, urine NGAL is a real-time indicator of active kidney damage in humans  and can be used to examine the progressive kidney disease. Theoretically, the use of serum creatinine levels which was a renal functional marker, even with new criteria, cannot exclude AKI. However, before a more reliable standard is established, serum creatinine is the only alternative. In this study, the elevation of NGAL was found to be a suitable biomarker that is able to detect at an earlier time point than serum creatinine in dogs with postoperative AKI. The results indicated that, like in humans, urine NGAL can be applied as a useful biomarker for the detection of AKI in dogs. However, further studies are needed in order to understand the role of NGAL in various different kidney diseases.