Saliva is a biological fluid that contains a certain number of elements common to other body fluids such as blood or urine in a sufficiently large consistency to take into interest their assays for biological analysis. Interest in saliva assays is increasing during last years [1, 2], with various scientific reviews indicating their advantages as a diagnostic medium such as its collection is safe, non-invasive, inexpensive, and simple. In addition it can be sampled repeatedly without discomfort to the patient [3].

Therefore saliva, especially in human medicine, has gained considerable attention as a possible alternate fluid to blood and urinary analysis [1] with a growing scientific literature describing the use of salivary biomarkers to determine diabetic and cardiac related risk factors, among others [4]. In dogs, analytes such as C-reactive protein [5] or cortisol [6, 7] have been measured in saliva and were found to be suitable biomarkers of inflammation and stress respectively.

Creatine kinase (CK) and Aspartate aminotransferase (AST) are two enzymes found in cytoplasm of muscle cells. They are considered to be markers of muscle cell damage, and an increase in these enzymes in serum may be an indicator of muscle stress or damage due to exercise [8–10] or muscle injuries [11, 12]. CK is primarily found in skeletal muscle, myocardium, brain and intestine [13]; although its main activity in serum is released from skeletal muscle [9]. While serum AST is less muscle specific than CK, since it is also present in liver and its activity in serum can increase due to liver damage [14]. In humans CK has been measured in saliva and it was indicated that plasma could be its main source [15].

CK and AST have been measured in serum of dogs as a marker of muscle damage under different conditions, such as hemilaminectomy or ovariohysterectomy [16], exercise [10], and after anesthesia [17]. However, to the authors’ knowledge no studies about CK and AST in saliva in dogs have been published previously. The knowledge about the possibility of measurement of CK and AST could be of both veterinarian and human interest, since the dog is a widely used experimental model for research.

The objective of the present study was to assess if CK and AST activities could be measured in canine saliva, and to evaluate their possible changes after processes that cause muscle damage. For this purpose, commercially available automated spectrophotometric CK and AST assays were validated in canine saliva and these two enzymes were measured in healthy dogs and in a group of dogs with increased serum CK and AST due to different causes.

All experimental procedures were approved by the Local Ethical Committee of University of Murcia, and were performed in compliance with laws RD32/2007 and RD1201/2005 concerning animal experimentation in Spain.

CK in saliva showed intra- and inter-assay imprecision less than 1 and 16%, respectively, while AST in saliva showed intra- and inter-assay imprecision less than 1 and 7%, respectively (Table 1). Linearity under dilution resulted in coefficient of correlation close to 1 in both cases (Fig. 1). Limit of detection for salivary CK was 6.01 U/L (mean, 2.05; SD, 1.32) and for salivary AST was 2.11 U/L (mean, 0.82; SD, 0.43).

Assay	Comparison	Specimen	mean	Sd	CV (%)
CK, U/L	Intra	Specimen 1	64.6	0.12	0.2
		Specimen 2	293.5	1.46	0.5
	Inter	Specimen 1	49.1	7.26	14.8
		Specimen 2	182.8	27.74	15.2
AST, U/L	Intra	Specimen 1	91.8	0.36	0.4
		Specimen 2	639.3	3.19	0.5
	Inter	Specimen 1	84.7	5.75	6.8
		Specimen 2	619.1	15.47	2.5

Serum CK activity in healthy controls ranged between 57.9 and 239 U/L, whereas it was significantly higher in dogs with muscle damage (P < 0.001) with values between 319 and 78,650 U/L (Fig. 2a). Serum AST in healthy controls ranged between 23.8 and 36.9 U/L and dogs with muscle damage showed significantly higher activity in serum AST (P < 0.001) with values between 58.3 and 988 U/L (Fig. 2b).

Salivary CK in healthy controls ranged between 27 and 121 U/L, whereas dogs with muscle damage showed a significantly higher activity in saliva CK (P < 0.001) with values between 132 and 3862 U/L (Fig. 2c). AST in saliva in healthy controls ranged between 46.8 and 144 U/L, whereas dogs with muscle damage showed a significantly higher activity in saliva AST (P < 0.001) with values between 154 and 4340 U/L (Fig. 2d).

When data from healthy controls and those with muscle damage were pooled, positive correlations were observed between serum and salivary CK (r = 0.579; P = 0.001) and between serum and salivary AST (r = 0.674; P < 0.001).

Serum activities of CK and AST reflect damage of muscle and are useful markers to determine the condition of muscle tissue both in dogs [10, 13, 23, 24] and humans [25, 26]. In particular, increases in CK following physical activity has been associated with muscle damage in both racing and hunting dogs [9] and similar findings have been described in humans [8, 25].

The assays used in our study showed an adequate intra-assay imprecision and accuracy to measure the activity of both enzymes in saliva. Therefore CK and AST can be measured in saliva in dogs with automated spectrophotometric assays available for routine use in clinical pathology laboratories. The inter-assay imprecision was higher than 15% for CK, that is considered as the acceptable limit [27]; this fact could be related with the stability of the enzyme under storage and further studies should be made to elucidate the reason. In case of using assays from a different manufacturer than those used in our study, an optimization of specimen volume and analytical validation of the assay would be recommended.

Our study reports for the first time the existence of CK and AST activities in saliva of dogs. In healthy dogs of our study CK showed lower values in saliva compared to serum which agrees with previous reports made in humans [15]. The lower values of CK in saliva could be due to the existence of a limit in the transfer of CK from plasma to saliva as it has been suggested in human for CK and other analytes such as NT-ProBNP [15, 28]. This situation seems not occur in the case of AST, that showed higher values in saliva than in serum. Additional studies should be performed to elucidate the reasons of the higher values of AST in saliva compared with serum in dogs.

Dogs with muscle damage showed significant increases in CK and AST in saliva and there was no overlap in the values of these enzymes between healthy dogs and dogs with muscle damage. In dogs with muscle damage the values of CK in saliva did not reach the values in serum (upper value in saliva of 3862 versus 78,650 U/L in serum) and again a limit in the transfer of CK from plasma to serum could be suggested to explain this fact. On the other hand in the case of AST values in saliva of dogs with muscle damage were higher than in serum (upper value in saliva of 4340 versus 988 U/L in serum), and further studies should be undertaken to elucidate the mechanisms that can produce this effect.

Significant correlations were obtained between salivary and serum activities of CK and AST in dogs and this would indicate a possible pass of the enzymes from serum to saliva. However these correlations were moderate, similarly to what has been found in humans [15]. One cause of this moderate correlation could be due to the different dynamics after a muscle damage that the enzymes can have in serum and saliva. In addition, from the analytical point of view it should be considered the different kinetics that the enzymatic reactions had in serum and saliva, although the sample dilution and further dilution in the reagents would likely minimize the matrix effect. Further studies about the process allowing muscle enzymes to filter from plasma to saliva, as well factors that could increase or decrease this transfer, would be recommended. In addition the possibility of enzyme synthesis by the salivary glands should be explored. This would allow a better understanding of the clinical significance of muscle enzymes in saliva and their correlation with serum.

Saliva analysis could be a potential suitable alternative to serum due to its non-invasive collection, which could be convenient in cases where dogs are reluctant to blood sampling or in young dogs of small breeds. However, it is important to point out some limitations that the measurements of CK and AST in saliva could have from a practical point of view. The first one is that in some cases it can not be obtained enough volume for analysis, as it have been previously described for salivary cortisol measurements in dogs [29]. The second one is that the presence of gingivitis and/or dental plaque is known to produce increases in salivary CK and AST in humans [20], and therefore the interpretation of these enzymes in saliva of dogs with these diseases should be made with caution.

This is a pilot study and further studies with a larger population of dogs should be made to confirm these findings. In addition although in our study the specimens have been analyzed in the first hour after collection, it would be of interest to evaluate the influence of different storage conditions in the stability of these enzymes in saliva. Furthermore, additional studies could be of interest to evaluate the behavior of the different CK isoenzymes in saliva and also the dynamics of CK and AST in serum and saliva after an experimental induced muscle damage as well in dogs with and without periodontal disease and gingivitis.

Results of the present study indicate that CK and AST activities can be measured in canine saliva with commercially available spectrophotometric assays. In addition these enzymes show higher values in saliva of dogs with muscle damage compared with healthy dogs and their values have a moderate correlation with those of serum.