Serum amyloid A reference intervals in raccoon dogs (Nyctereutes procyonoides koreensis): Implications for detecting inflammation

Abstract

Inflammation is a fundamental defense mechanism in animals, and its accurate detection is essential for diagnosis and treatment in veterinary medicine. Serum amyloid A (SAA), an acute-phase positive protein (APP), is widely used to detect inflammation in domestic animals. However, its application in wildlife, particularly raccoon dogs (Nyctereutes procyonoides koreensis), remains underexplored. This study aimed to establish a reference interval (RI) for SAA in raccoon dogs and explore its applicability as a biomarker of inflammation. A total of 83 blood samples from 43 raccoon dogs collected from the Korean Wildlife Rescue Center were categorized into three clinical groups: clinically healthy (CH), mild injury (M), and severe injury (S). The SAA concentrations were significantly higher in the S group than in the CH and M groups, whereas no significant difference was observed between the CH and M groups. The RI for SAA in CH raccoon dogs was determined to be 0.13–1.4 mg/L. Although SAA may have limited sensitivity for detecting mild or localized inflammation, the results support its use as a reliable marker of severe inflammation in raccoon dogs. Future studies should explore other APPs to improve the diagnostic accuracy of inflammation in wildlife health assessment.

INTRODUCTION

Inflammation is a vital defense mechanism in animals that protects them from infections, injuries, and diseases, and contributes to their overall health. Accurate monitoring of inflammation is particularly important in veterinary medicine because it facilitates the diagnosis of diseases and evaluation of severity, guides treatment strategies, and tracks recovery [7]. The availability of effective tools to monitor inflammation is crucial for veterinary diagnostics.

Serum amyloid A (SAA) is a reliable acute-phase protein (APP) primarily produced in the liver, and its levels rapidly increase in response to inflammatory stimuli. Pro-inflammatory cytokines, particularly interleukin (IL)-1 and IL-6, can cause SAA concentrations to increase by up to 1,000-fold within hours of the onset of inflammation. This elevation typically begins 8–12 hr after inflammation starts and peaks within 24–48 hr. The half-life of SAA is estimated to be 24–48 hr, and its concentration returns to undetectable levels once synthesis ceases [4, 10, 14]. These characteristics make SAA a valuable marker for tracking active phases of inflammation [1, 6].

Most studies on SAA have focused on domestic animals, and its diagnostic potential is well established. SAA is considered one of the most sensitive indicators of inflammation in equine medicine, outperforming conventional markers, such as fibrinogen levels and white blood cell counts, particularly in systemic inflammatory conditions. SAA levels in canines and bovines are clinically useful for diagnosing and monitoring inflammation caused by infections, trauma, and various inflammatory disorders. The clinical efficacy of SAA against these species has led to their widespread use in veterinary practice for specific diagnostic applications [1].

Although SAA is an effective biomarker for domesticated animals, its application in wildlife, especially in raccoon dogs (Nyctereutes procyonoides koreensis) remains largely unexplored. In South Korea, raccoon dogs are frequently admitted to wildlife rehabilitation centers and often present with inflammatory conditions. The population of this species has increased rapidly in recent years, driven by factors such as reduced predation and high adaptability, raising public health concerns regarding its potential role as a vector for zoonotic diseases [3, 9, 12].

This study aimed to establish a reference interval (RI) for SAA in raccoon dogs to provide a baseline for assessing inflammation in clinically healthy (CH) individuals. In addition, this study evaluated the potential of SAA as a biomarker of inflammation and examined its utility in the detection and monitoring of inflammatory processes. By defining a normative range of SAA concentrations in raccoon dogs, this study contributes to a more rigorous assessment of inflammatory conditions and lays the groundwork for future investigations. These findings support the advancement of disease management and public health surveillance in regions where raccoon dogs are prevalent, thereby contributing to broader wildlife health monitoring initiatives.

MATERIALS AND METHODS

Blood sample collection

A total of 83 blood samples were collected from 43 adult raccoon dogs housed in wildlife rescue centers in Jeonbuk (48 samples), Chungbuk (28 samples), and Jeonnam (12 samples), South Korea. Five individuals were sampled multiple times to monitor inflammation progression. The raccoon dogs were handled in accordance with standard protocols for the diagnosis and treatment of wildlife. Blood samples were aseptically collected from the jugular or cephalic veins for diagnostic and therapeutic purposes. To ensure safe handling, animals were physically restrained and muzzled. Serum was separated using a 5 mL SST tube (BD Vacutainer, Plymouth, UK). When immediate testing was not possible, the serum samples were stored at −20°C. Before the analysis, the frozen samples were thawed at room temperature. Based on the clinical evaluations, the samples were grouped into three categories: CH, mild injury (M), and severe injury (S). Patients in the S group had severe injuries, including acute fractures, major open wounds, and systemic inflammation resulting from infection. Raccoon dogs in the M group exhibited symptoms such as mild dehydration, moderate fatigue, and abrasions, which improved over time. Patients assigned to the CH group exhibited good health, normal appetite, and no clinical or laboratory abnormalities.

Assay precision and evaluation

Assay precision was evaluated on the same day (intra-assay) and on different days (inter-assay). This was performed by calculating the coefficient of variation (CV) from 20 repeated measurements of the three serum pools with low, medium, and high SAA concentrations. Serum pools were prepared by combining samples from several animals. All samples for testing were stored at −20°C and thawed before analysis. Given the absence of biological variation data for SAA in raccoon dogs, the acceptable CV limit was set at 12.4% based on human data [5].

The limit of detection (LoD) was calculated through repeated measurements over five days. A blank sample (distilled water) and a low-concentration SAA sample (7 mg/L) were tested 20 times. From these measurements, the limits of the blank (LoB) and LoD were calculated using the following formulae [13]:

LoB=mean of blank + 1.65 × SD of blank

LoD=LoB + 1.65 × SD of low concentration sample

To determine the limit of quantitation (LoQ), a low-concentration SAA sample (10 mg/L) was serially diluted with saline and measured 20 times over five days. The LoQ was defined as the lowest concentration at which the total observed error (TE_obs) was lower than the total allowable error (TE_a), which was set to 37.0% based on human SAA variation [6]. TE_obs was calculated using the formula TE=bias + 2 × SD, where bias is the difference between the measured and expected values.

SAA assay

The VET-SAA assay (Eiken Chemical Co., Tokyo, Japan) was performed using a Hitachi 3100 automatic biochemical analyzer (Hitachi Ltd., Tokyo, Japan). A 3-μL sample volume was used, following the manufacturer’s protocol. Calibration was performed using a standard traceable to the World Health Organization (WHO International Standard 92/680). Quality control materials (VET-SAA-QC-Low and VET-SAA-QC-High; Eiken Chemical Co., Tokyo, Japan) were used for all the assays. Samples exceeding the measurable range (5–200 mg/L) were diluted 1:3 with saline and reanalyzed.

Total protein and albumin assay

Total protein (TP) and albumin (Alb) concentrations were measured using Clinimate TP and Clinimate ALB assay kits (Sekisui Medical Co., Ltd., Tokyo, Japan) on a Hitachi 3100 automatic biochemical analyzer. Serum volumes of 5 μL and 3 μL were used for the TP and Alb assays, respectively, in accordance with the manufacturer’s instructions. Globulin (Glo) concentrations were calculated by subtracting albumin from the total protein values.

Monitoring inflammatory markers in a raccoon dog

A single adult raccoon dog that presented with a sarcoptic mange infestation and decubitus ulcers was admitted to the Jeonbuk Wildlife Center. Serial blood samples were collected over the 26-day treatment period. A total of nine blood samples were collected on different days (0, 1, 3, 4, 5, 7, 10, 13, 26), with the first sample (0) taken prior to treatment initiation. The raccoon dog was administered antibiotics (amoxicillin/clavulanic acid). SAA and C-reactive protein (CRP) levels were measured to evaluate the inflammatory response. CRP levels were analyzed using the VET Chroma System (ANIVET Inc., Chuncheon, South Korea). Temporal changes in SAA and CRP concentrations were monitored, and the results were graphically represented to illustrate biomarker trends throughout the treatment period.

Statistical analyses

The normality of the data distribution was assessed using the Shapiro–Wilk test, which is suitable for small sample sizes. SAA, Alb, and Glo concentrations, and albumin-to-globulin (A/G) ratios were compared among the clinical groups (CH, M, and S) using the Kruskal–Wallis test, followed by Dunn’s post hoc test for pairwise comparisons. Correlations between SAA and other variables, including Alb, Glo, and A/G ratio, were evaluated using Spearman’s rank correlation coefficient. Statistical significance was defined as P<0.05. Reference intervals (RIs) were established according to the American Society for Veterinary Clinical Pathology guidelines. The dataset was refined by identifying and removing outliers using Tukey’s method. All statistical analyses were performed using the R software (version 4.4.1; https://www.R-project.org).

RESULTS

Assay performance and analytical characteristics of SAA

The intra- and inter-assay CVs for the SAA measurements were 1.87–4.23% and 1.76–8.16%, respectively (Table 1). Analytical sensitivity was determined by calculating the LoB at 0.87 mg/L (mean_blank: 0.56 mg/L; SD_blank: 0.19), while the LoD was established at 1.36 mg/L (SD_low concentration sample: 0.30 mg/L). Based on the TE_obs of 33.87%, the LoQ was set to 1.25 mg/L, which was lower than the TE_a of 37.0%.

Table 1. Assessment of intra- and inter-assay variations in serum amyloid A (SAA) measurement using VET-SAA in raccoon dog sera.

Comparison	No. of replicates	SAA (mg/L)		Coefficient of variation (%)
		Mean	SD
Intra-assay
Low	20	7.08	0.30	4.22
Moderate	20	21.57	0.40	1.87
High	20	88.30	2.31	2.62
Inter-assay
Low	20	3.03	0.25	8.16
Moderate	20	37.02	0.73	1.97
High	20	88.53	1.56	1.76

The terms low, moderate, and high refer to raccoon dog sera with varying SAA concentrations, whereas low, moderate, and high indicate low, moderate, and high SAA concentrations, respectively.

SAA concentrations across groups

SAA concentrations were significantly different among the groups. The CH group demonstrated a median SAA concentration of 0.8 mg/L (range was 0.1–1.5 mg/L). The M group presented a marginally elevated median concentration of 0.85 mg/L (range: 0.4–5.1 mg/L), whereas the S group displayed a markedly higher median SAA concentration of 139.2 mg/L (range: 22.2–341.8 mg/L), as shown in Table 2. Statistical analyses revealed significant differences in SAA levels between the CH and S groups (P<0.001) and between the M and S groups (P<0.01). However, no significant differences were observed between the CH and M groups (P>0.05), as shown in Fig. 1.

Table 2. Descriptive statistics for serum amyloid A (SAA), albumin, and globulin concentrations and the A/G ratio in raccoon dog samples.

Parameter	Group	n (individual)1	Q12	Median	3Q3	Range
SAA (mg/L)	CH	31 (23)	0.50	0.80	1	0.10–1.50
	M	16 (12)	0.58	0.85	2.35	0.40–5.10
	S	6 (6)	78.73	139.20	244.38	22.20–341.80
Albumin (g/dL)	CH	32 (24)	2.3	2.6	2.8	2.0–3.2
	M	16 (11)	2.6	2.8	2.9	2.2–3.6
	S	6 (6)	0.9	1.4	2.4	0.2–3.4
Globulin (g/dL)	CH	32 (24)	3.7	3.8	4.4	3.1–5.2
	M	15 (10)	4.0	4.4	4.8	3.4–5.2
	S	6 (6)	2.8	4.8	6.0	0.8–9.4
4A/G Ratio	CH	33 (25)	0.53	0.66	0.78	0.50–0.86
	M	15 (10)	0.58	0.64	0.69	0.38–0.82
	S	5 (5)	0.25	0.26	0.33	0.23–0.36

¹ n represents the total number of samples; the numbers in parentheses indicate the number of individual animals from which the samples were obtained. Multiple samples were collected from the same animal. ² Q1 (first quartile) is the value below which 25% of the data fall. ³ Q3 (third quartile) is the value below which 75% of the data fall. ⁴ A/G: albumin/globulin. The clinical groups: clinically healthy (CH), mild injury (M), and severe injury (S).

Fig. 1.

Concentrations of serum amyloid A (SAA), albumin, globulin, and the albumin-to-globulin (A/G) ratio in raccoon dogs. Box plots represent the median, interquartile range (25th–75th percentiles), and overall distribution of each parameter across three clinical groups: clinically healthy (CH), mild injury (M), and severe injury (S). Panel (A) displays the concentration of SAA, panel (B) shows albumin levels, panel (C) presents globulin concentrations, and panel (D) depicts the A/G ratio. Statistically significant differences among groups are indicated as follows: **P<0.01, ***P<0.001; ns, not significant.

Albumin and globulin concentrations and A/G ratio

The Alb concentration was the highest in the M group, with a median of 2.8 g/dL (range: 2.2–3.6 g/dL), followed by the CH group with a median of 2.6 g/dL (range: 2–3.2 g/dL). The lowest Alb concentration was observed in the S group, with a median of 1.4 g/dL (range: 0.2–3.4 g/dL). No statistically significant differences were observed between the groups (P>0.05). Furthermore, the Glo concentration was the highest in the S group, with a median of 4.55 g/dL (range: 0.8–9.4 g/dL), followed by the M group with a median of 4.4 g/dL (range: 3.4–5.2 g/dL). The lowest Glo concentration was observed in the CH group with a median of 3.8 g/dL (range: 3.1–5.2 g/dL). No statistically significant differences were observed between the groups (P>0.05). Regarding the A/G ratio, the CH group exhibited the highest median with 0.66 (range: 0.5–0.86), followed by the M group with a median of 0.64 (range: 0.38–0.82) and the S group with a median of 0.26 (range: 0.23–0.36). Statistical analyses identified significant differences in the A/G ratio between the CH and S groups (P<0.001), and between the M and S groups (P<0.01). However, no significant differences were observed between the CH and M groups (P>0.05; Fig. 1 and Table 2).

Correlations among the SAA, albumin, and globulin concentrations and the A/G ratio

Correlation analysis revealed a significant relationship between SAA levels and other protein biomarkers. A weak negative correlation was observed between SAA and Alb (r= −0.333, P<0.01), whereas the correlation between SAA and Glo was weak and not statistically significant (r=0.077, P>0.05). A weak negative correlation was also found between SAA and A/G ratio (r= −0.272, P<0.05). Alb and Glo concentrations exhibited moderate inverse correlations (r= −0.437, P<0.001). Alb demonstrated a strong positive correlation with the A/G ratio (r=0.878, P<0.001), whereas Glo and the A/G ratio were strongly negatively correlated (r= −0.787, P<0.001).

Monitoring of inflammatory markers in a treated raccoon dog

Serial measurements of SAA and CRP levels were conducted over 26 days in a raccoon dog with sarcoptic mange and decubitus ulcers admitted to the Jeonbuk Wildlife Center. During the treatment period, the SAA concentrations exhibited marked fluctuations, peaking on day 2 (153.4 mg/L) and gradually declining by day 26 to levels approaching the upper reference limit established for CH raccoon dogs (1.4 mg/L). CRP concentrations showed less pronounced fluctuations than SAA concentrations, exhibiting a moderate peak on day two and a steady decline throughout the treatment period. Temporal trends in SAA and CRP concentrations illustrated the dynamic response of these APPs to inflammation and clinical recovery. Notably, the SAA levels were substantially elevated during the early treatment stages, indicating systemic inflammation. Both markers progressively declined as the clinical symptoms improved. These results suggested that SAA may function as a sensitive biomarker for monitoring inflammation and recovery in raccoon dogs undergoing treatment (Fig. 2).

Fig. 2.

Temporal changes in serum amyloid A (SAA) and C-reactive protein (CRP) concentrations during treatment of a raccoon dog with sarcoptic mange and decubitus ulcers. The graph depicts SAA (blue) and CRP (light blue) concentrations measured at multiple time points throughout the treatment period. The x-axis indicates the number of days since treatment initiation, and the y-axis represents biomarker concentrations (mg/L).

Establishment of an SAA RI for raccoon dogs

The RI for SAA in CH raccoon dogs was established based on a parametric analysis, as the data exhibited a normal distribution (P=0.304, Shapiro–Wilk test). The mean SAA concentration was 0.76 ± 0.33 mg/L, with a median value of 0.8 mg/L and a range of 0.1–1.5 mg/L. The calculated reference limits for SAA in this population were 0.13–1.40 mg/L (Table 3).

Table 3. Reference intervals for serum amyloid A (SAA) and biochemical analytes in healthy raccoon dogs.

Parameter	Unit	n1	Mean ± SD	Median	Min.–Max.	Distribution2	Method3	LRL4	URL5
SAA	mg/L	31 (23)	0.76 ± 0.33	0.8	0.1–1.5	G	P	0.1 (0.0–0.3)	1.4 (1.2–1.6)
Total protein	g/L	27 (20)	6.49 ± 0.38	6.5	5.7–7.2	G	P	5.7 (5.5–5.9)	7.2 (7.0–7.4)
Albumin	g/dL	33 (25)	2.57 ± 0.38	2.6	1.5–3.2	G	P	1.8 (1.6–2.0)	3.3 (3.1–3.5)
Globulin	g/dL	32 (24)	4.00 ± 0.56	3.8	3.1–5.2	NG	NP	3.1 (2.9–3.1)	5.2 (5.2–5.4)
6A/G ratio	-	33 (25)	0.67 ± 0.13	0.66	0.50–0.86	NG	NP	0.50 (0.49–0.50)	0.86 (0.86–0.89)

¹ n represents the total number of samples; numbers in parentheses indicate individual animals. Multiple samples may have originated from the same animal. ² G: Gaussian distribution; NG: non-Gaussian distribution. ³ P: parametric method; NP: nonparametric method. ⁴ LRL: lower reference limit (numbers in parentheses represent 90% confidence intervals). ⁵ URL: upper reference limit (numbers in parentheses represent 90% confidence intervals). ⁶ A/G: albumin/globulin.

DISCUSSION

The primary objective of this study was to establish an RI for SAA concentration in raccoon dogs and to evaluate its potential as an APP for monitoring inflammation. The results showed that SAA levels in healthy raccoon dogs ranged from 0.13 to 1.40 mg/L, with a median of 0.80 mg/L. SAA concentrations were significantly higher in raccoon dogs with severe injuries such as fractures or infections. This is consistent with findings in other species, where SAA serves as a reliable indicator of systemic inflammation [2, 16]. However, animals with minor or localized injuries did not exhibit significantly increased SAA levels, suggesting that SAA primarily reflects severe or systemic inflammatory responses, rather than localized tissue damage. Similar patterns have been observed in felines, where SAA levels do not consistently increase in cases of mild inflammation [15].

A case analysis of a raccoon dog with sarcoptic mange and decubitus ulcers further highlighted the distinct inflammatory responses of different APPs. CRP levels peaked at 500 mg/L on day 1, whereas SAA levels reached 299.8 mg/L on day 7. This delayed and less pronounced response of SAA compared to that of CRP suggests its differential roles in inflammation. Although CRP may react more rapidly to acute injuries or inflammatory triggers, SAA may better reflect ongoing systemic inflammation. However, owing to infrequent sampling over a 26-day period, the observed APP fluctuations may not fully capture their kinetics. Thus, frequent sampling is essential for the accurate characterization of these dynamics.

These findings underscore the utility and limitations of SAA as an inflammatory biomarker in raccoon dogs. Although the SAA reliably indicates severe systemic inflammation, its limited response to minor injuries suggests that relying on a single APP may not provide a comprehensive assessment of inflammatory conditions. To improve diagnostic accuracy, both human and veterinary medicine have proposed a multi-APP approach that combines positive and negative APPs, commonly referred to as the APP index [8]. This method may improve monitoring of inflammation in raccoon dogs, particularly when SAA alone does not fully reflect the complexity of immune responses. Given that inflammatory diseases such as pneumonia and enteritis may not present with visible clinical signs, monitoring APPs can help detect the underlying inflammation in rescued raccoon dogs. However, further research is required to determine the diagnostic accuracy of SAA and CRP levels under such conditions.

Several methodological considerations must be acknowledged. A key limitation of this study was the repeated sampling of individuals, which may introduce a potential bias from autocorrelation, although this is common in wildlife research. Statistical techniques such as mixed-effects models can address these dependencies and increasing the diversity of sampled individuals strengthens the robustness of the findings [11]. In addition, factors such as age, sex, and seasonal variations may influence APP levels and warrant further investigation to refine the RI for different physiological conditions.

In conclusion, this study established an RI for SAA in raccoon dogs and demonstrated its utility as a marker of systemic inflammation. However, its limited sensitivity to mild or localized inflammation highlights the need for a broader approach that incorporates multiple APPs. Future research should focus on characterizing the interactions between different APPs, refining diagnostic thresholds, and improving sampling methodologies to enhance the clinical utility of inflammatory biomarkers in raccoon dogs. These efforts will contribute to the improvement of diagnostic strategies for wildlife medicine and provide valuable insights into the inflammatory processes in this species.

CONFLICT OF INTEREST

The authors declare no financial or personal relationships that could influence the content of this article.