Hyperphosphatasemia (high serum ALP activity) is commonly detected in dogs with hepatic disease, hyperadrenocorticism, or bone disease. Although ALP activity is variable in specific disease conditions, studies1–3 indicate that the highest ALP activities are associated with hepatic cholestasis and spontaneous or iatrogenic hyperadrenocorticism, and milder increases are associated with skeletal disorders. Alkaline phosphatase measured during routine serum biochemical analysis includes several isoforms derived from a variety of tissues. In healthy dogs, circulating ALP predominantly comprises liver and bone isoforms.
Cholestasis in dogs causes synthesis and solubilization of membrane-associated ALP (ie, liver ALP isoform), which results in markedly increased serum ALP activity.4–6 Cholestasis can develop subsequent to obstructive hepatic or posthepatic disease, hepatic infections, compressive periportal lesions, or hepatocyte swelling, which leads to impaired canalicular bile flow4,6–8
Hyperphosphatasemia commonly results from high circulating concentrations of endogenous or exogenous corticosteroids; other drugs, particularly anticonvulsants, can also induce de novo synthesis of ALP.6,9–13 The increase in serum ALP activity in the former circumstance is primarily attributable to increased production of the corticosteroid ALP isoform, relative to production of other isoforms.3,4,6,13,14 Other causes of hyperphosphatasemia in adult dogs include increased osteoblastic activity associated with increased bone modeling (or remodeling) and lytic lesions of bone.4,6,14–18
Scottish Terriers reportedly have higher ALP activities than do other breeds.19 It has been suggested that high ALP activity in otherwise healthy Scottish Terriers may possibly be attributable to excessive activity of the bone ALP isoform, which is the mechanism of benign familial hyperphosphatasemia, a rare inherited condition in Siberian Huskies.19–21
Analysis of the results of another study2 indicated that the prevalence of disorders associated with hyperphosphatasemia (eg, hyperadrenocorticism, diabetes mellitus, pancreatitis, and hepatobiliary disease) is increased in Scottish Terriers, compared with the prevalence in other breeds. However, this association alone did not fully account for the higher mean serum ALP activity in reported cases.19,20 The purpose of the study described here was to determine potential causes of hyperphosphatasemia in apparently healthy Scottish Terriers and to describe clinical, clinicopathologic, and histologic findings in this population. Adrenal gland response to cortisol administration was also evaluated.
Materials and Methods
Animals—Thirty-four client-owned apparently healthy Scottish Terriers (males and females; ages, 1 to 11 years; 17 with and 17 without hyperphosphatasemia) were enrolled in the study. Dogs were considered to be healthy on the basis of results of physical examination, medical history review, a CBC, routine serum biochemical tests, and urinalysis. Dogs were excluded if they had any history of endocrine or hepatic disorders or if they had received any corticosteroid-containing medications or anticonvulsants within 6 months prior to the study. Some of the dogs were receiving commercial vitamin products, heartworm preventives, fish oils, and antihistamines, alone or in combination. The Virginia Tech Institutional Animal Care and Use Committee reviewed and approved all animal protocols used in this study. Written consent was obtained from the owners of all dogs prior to enrollment in the study.
Procedures—Clinicopathologic assays (except for hormone analyses and ALP isoform measurements), abdominal ultrasonographic examinations, and collection of liver biopsy specimens were conducted at the Virginia-Maryland Regional College of Veterinary Medicine. Two authors (DCR and JHK) performed physical examinations and obtained blood samples from 18 dogs evaluated and enrolled at other sites; these dogs did not undergo ultrasonographic examination, bile acids testing, urinalysis, hemostasis evaluations, or liver biopsy procedures.
Clinicopathologic assessment of blood obtained from various peripheral venipuncture sites (predominantly jugular vein) included analysis of a CBC; serum biochemical assays; partial thromboplastin, prothrombin, and buccal mucosal bleeding times; and activities of serum ALP isoformsa (bone, liver, and corticosteroid).22–26 Serum bile acids were measured in preprandial samples (obtained from dogs after food was with-held for 12 hours) and postprandial samples (obtained 2 hours after feeding). Urine samples were obtained via cystocentesis. In addition, measurement of serum cortisol, estradiol, androstenedione, progesterone, 17-hydroxyprogesterone, and aldosteroneb concentrations was performed for samples obtained immediately before and 1 hour after IM administration of cosyntropinc (a synthetic subunit of ACTH; 5 μg/kg [2.27 μg/lb]). Dogs were assigned to 1 of 2 groups on the basis of serum ALP activity (reference interval, 13 to 110 U/L). Dogs with hyperphosphatasemia had serum ALP activities greater than the reference interval; dogs without hyperphosphatasemia had serum ALP activities within the reference interval. Reference intervals for steroid hormones were previously established22,27 (Appendix).
Results of adrenal steroid hormone analyses for samples obtained before and after ACTH administration were used to further classify the dogs into 1 of 3 categories. Dogs that had all hormone concentrations within the reference intervals were classified as normal in regard to their adrenal gland function. Dogs that had serum cortisol concentrations above the reference interval after ACTH stimulation were classified as high cortisol; dogs that had ≥ 1 noncortisol adrenal steroid hormone concentration greater than the reference interval after ACTH stimulation were classified as high noncortisol.
Eleven dogs with and 5 without hyperphosphatasemia underwent complete ultrasonographic examination of the abdomen (including measurement of the width of both adrenal glands), which was performed by various attending board-certified veterinary radiologists on clinical duty at the time of examination. Urinalysis consisted of determination of urine specific gravity via refractometer and screening for the presence of protein, hemoglobin, bilirubin, glucose, and ketones via urinalysis stripsd and of casts, WBCs, and RBCs via light microscopy.e Anesthesia was induced with acepromazinef (0.025 to 0.050 mg/kg [0.011 to 0.023 mg/lb], IV) and oxymorphoneg (0.05 to 0.1 mg/kg [0.023 to 0.045 mg/lb], IV), and skin at the puncture site was prepared by clipping hair and cleansing with povidone-iodine solution. Multiple (3 to 4) specimens of liver tissue were obtained from these 16 dogs by use of an ultrasound-guided 14-gauge core biopsy needle.h One specimen was frozen at −30°C and submitted for copper analysis.i The remaining samples were placed in neutral-buffered 10% formalin for microscopic examination by a board-certified veterinary pathologist (RJP) who was unaware of the group assignment of the dogs.
Formalin-fixed, paraffin-embedded liver biopsy sections were stained with H&E, Snook reticulin stain, PAS, and PAS-amylase; selected samples were stained with Ziehl-Neelsen and Perl iron stain. The stained sections were evaluated histologically for features associated with vacuolar hepatopathy as described by Sepesy et al,26 including excessive hepatocellular cytoplasmic reticulation, hepatocytes with vacuolated cytoplasm, and hepatocellular swelling (determined on the basis of the size of cells and compression of sinusoids). A subjective scoring system was used to estimate the degree of vacuolar hepatopathy present in each section in accordance with 1 of 3 categories: no criteria (sections that did not have any of the described features), partial criteria (sections that had some but not all features), or all criteria (sections that had all features). A single slide per dog containing multiple (4 to 8) specimens of liver prepared from multiple core biopsy samples was examined and used to assign an average hepatopathy score for each dog. Severity (mild, moderate, or severe) and distribution (zone 1, periportal; zone 2, midzonal or midacinar; or zone 3, perivenular or periacinar) for the various histologic criteria were also subjectively evaluated.
Statistical analysis—A χ2 test of independence was used to compare the distribution frequencies of sex and vacuolar hepatopathy category between dogs with and without hyperphosphatasemia. Normal distributions of age, urine specific gravity, CBC results, serum biochemical analyses, ALP isoform activity and percentages, coagulation times, and bile acids and hormone concentrations were assessed via Anderson-Darling tests. Because the reference intervals for steroid hormones were sex dependent (Appendix), adrenal steroid hormone results were normalized by use of the following equation: normalized result = (analyte result − midvalue of the reference interval)/(reference interval range/2),28–33 where the reference interval range = (upper limit − lower limit) of the reference interval. The reference interval for any sex was −1 to 1 (no units) for each steroid hormone. A 2-sample unpaired Student t test (assuming unequal variance) was used to compare numeric data between dogs with and without hyperphosphatasemia. A Pearson correlation coefficient was determined between serum ALP activity and other numeric data that were significantly different between the 2 ALP groups or that had mean values outside the applicable reference interval. Data were analyzed by use of commercially available statistical software.j Values of P < 0.05 were accepted as significant.
Results
Physical examination did not reveal clinically relevant abnormalities in any dogs. Dogs with hyperphosphatasemia were significantly (P < 0.001) older than dogs without hyperphosphatasemia (mean ± SD age, 7.0 ± 0.60 and 2.6 ± 0.41 years, respectively). No association was detected between sex and ALP activity.
Mean values for plasma cholesterol concentration and alanine aminotransferase activity were significantly higher, and mean urine specific gravity was significantly lower, in dogs with hyperphosphatasemia, compared with values in dogs without hyperphosphatasemia (Table 1). Postprandial bile acids concentrations, prothrombin time, and partial thromboplastin time were within reference intervals for all dogs. In dogs with hyperphosphatasemia, activity of the corticosteroid ALP isoform was significantly increased (mean ± SE, 542.0 ± 178.0 U/L), compared with the activity in dogs without hyperphosphatasemia (14.4 ± 3.9 U/L). This isoform represented a significantly greater percentage of total ALP (mean, 69 ± 5.0%), compared with the percentage in dogs without hyperphosphatasemia (17 ± 3.8%).
Results of clinicopathologic analysis of serum and urine samples obtained from apparently healthy adult Scottish Terriers with and without hyperphosphatasemia.
| Test | Variable | Dogs with hyperphosphatasemia* | Dogs without hyperphosphatasemia† | P value‡ | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| RI | Mean | SE | No. < RI | No. > RI | Mean | SE | No. < RI | No. > RI | |||
| — | Age (y) | — | 7.00 | 0.60 | — | — | 2.59 | 0.41 | — | — | < 0.001 |
| Biochemical | ALT (U/L) | 17–66 | 87.00 | 19.00 | 0 | 7 | 42.20 | 10.00 | 0 | 1 | 0.048 |
| analysis | ALP (U/L) | 13–110 | 706.00 | 246.00 | 0 | 17 | 63.80 | 5.30 | 0 | 0 | 0.019 |
| Postprandial bile acids (Mmol/L) | < 30 | 6.59 | 1.70 | — | 0 | 1.78 | 0.39 | — | 0 | 0.020 | |
| Cholesterol (mg/dL) | 160–345 | 269.10 | 17.00 | 0 | 4 | 194.80 | 12.00 | 4 | 0 | 0.001 | |
| Urinalysis | Specific gravity | — | 1.018 | 0.002 | — | — | 1.037 | 0.004 | — | — | < 0.001 |
| ALP isoform | Corticosteroid (U/L) | 0–40 | 542.00 | 178.00 | 0 | 17 | 14.40 | 3.90 | 0 | 1 | 0.009 |
| panel | Liver (U/L) | ND | 199.00 | 88.00 | — | — | 31.20 | 3.30 | — | — | 0.074 |
| Bone (U/L) | ND | 96.00 | 42.00 | — | — | 26.60 | 2.80 | — | — | 0.119 | |
| Corticosteroid (%) | ND | 69.00 | 5.00 | — | — | 17.00 | 3.80 | — | — | < 0.001 | |
| Liver (%) | ND | 25.20 | 5.80 | — | — | 43.90 | 3.60 | — | — | 0.012 | |
| Bone (%) | ND | 9.77 | 1.00 | — | — | 39.10 | 3.60 | — | — | < 0.001 | |
Tests were performed on samples from 17 dogs/group, except for urinalysis, which was only performed on samples from 11 dogs with and 5 dogs without hyperphosphatasemia.
Dogs with hyperphosphatasemia consisted of 4 sexually intact females, 8 spayed females, 2 sexually intact males, and 3 neutered males.
Dogs without hyperphosphatasemia consisted of 1 sexually intact female, 7 spayed females, and 9 neutered males.
Mean group values were compared between dogs with and without hyperphosphatasemia; differences of P < 0.05 were accepted as significant.
ALT = Alanine aminotransferase. ND = Not determined. RI = reference interval. — = Not applicable.
Serum cortisol concentrations exceeded the reference intervals in some dogs with hyperphosphatasemia before (5/17 [29%]) and after (6/17 [35%]) ACTH administration (Tables 2 and 3). In dogs without hyperphosphatasemia, serum cortisol concentrations were within the reference interval before and after stimulation with ACTH. Mean normalized serum cortisol concentrations were significantly different between the 2 groups both before and after ACTH administration. However, the normalized mean change (Δ) in serum cortisol values (ie, the difference between concentrations measured before and after ACTH stimulation) was not significantly (P = 0.964) different between dogs with and without hyperphosphatasemia (0.33 and 0.34, respectively; normalized values did not have units), and normalized mean cortisol values were within the reference interval for both groups of dogs before and after ACTH stimulation.
Normalized serum hormone concentrations before and after ACTH stimulation in apparently healthy adult Scottish Terriers with and without hyperphosphatasemia.
| Analyte | Dogs with hyperphosphatasemia* | Dogs without hyperphosphatasemia† | P value‡ | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Mean | SE | No. < RI | No. > RI | Mean | SE | No. < RI | No. > RI | ||
| Cortisol | |||||||||
| Before ACTH | 0.49 | 0.20 | 0 | 5 | −0.30 | 0.09 | 0 | 0 | 0.002 |
| After ACTH | 0.82 | 0.15 | 0 | 6 | 0.04 | 0.11 | 0 | 0 | < 0.001 |
| Androstenedione | |||||||||
| Before ACTH | 2.00 | 0.58 | 0 | 9 | 0.26 | 0.42 | 1 | 2 | 0.023 |
| After ACTH | 0.03 | 0.22 | 1 | 3 | −0.39 | 0.15 | 2 | 0 | 0.130 |
| Estradiol | |||||||||
| Before ACTH | 0.81 | 0.25 | 0 | 6 | −0.15 | 0.17 | 3 | 1 | 0.004 |
| After ACTH | 0.68 | 0.23 | 0 | 7 | −0.08 | 0.14 | 1 | 1 | 0.011 |
| Progesterone | |||||||||
| Before ACTH | 1.05 | 0.60 | 0 | 5 | 1.77 | 1.70 | 0 | 3 | 0.695 |
| After ACTH | 2.11 | 0.36 | 0 | 12 | 2.77 | 0.70 | 0 | 12 | 0.413 |
| 17-hydroxyprogesterone | |||||||||
| Before ACTH | 1.57 | 0.49 | 0 | 9 | 0.87 | 0.43 | 0 | 3 | 0.294 |
| After ACTH | 2.20 | 0.53 | 0 | 12 | 2.53 | 0.69 | 0 | 10 | 0.778 |
| Aldosterone | |||||||||
| Before ACTH | −0.26 | 0.72 | 0 | 0 | −0.02 | 0.19 | 0 | 3 | 0.259 |
| After ACTH | 0.89 | 0.16 | 0 | 5 | 1.08 | 0.25 | 0 | 9 | 0.539 |
Analyte values and reference ranges were normalized for sex-specific reference ranges (Appendix) and have no associated units. Results were normalized by use of the following equation: normalized result = (analytic result − midvalue of the reference interval)/(reference interval range/2). Potential values ranged from −1 to 1. Normalized results can be converted to actual values by use of the following formula: true analyte value = (normalized result × [reference interval range/2]) + midvalue of the reference interval.
See Table 1 for remainder of key.
Summary of serum steroid hormone analysis after ACTH stimulation and vacuolar hepatopathy categories in apparently healthy adult Scottish Terriers with and without hyperphosphatasemia.
| Variable | Category | No. of dogs with hyperphosphatasemia* | No. of dogs without hyperphophatasemia† | Total |
|---|---|---|---|---|
| Serum steroid hormone values | All values within RIs | 0 | 2 | 2 |
| High cortisol | 6 | 0 | 6 | |
| High noncortisol | 11 | 15 | 26 | |
| Vacuolar hepatopathy assessment | 1 | 0 | 1 | 1 |
| 2 | 3 | 2 | 5 | |
| 3 | 8 | 2 | 10 |
For serum steroid hormone analysis, n = 34 dogs (17 dogs/group); for vacuolar hepatopathy analysis, n = 16 dogs (11 dogs with and 5 without hyperphosphatasemia). For hepatopathy analysis, formalin-fixed, paraffin-embedded core needle biopsy specimens of the liver were stained with H&E, Snook reticulin stain, PAS stain, and PAS-amylase stain; selected samples were stained with Ziehl-Neelsen and Perl iron stains. Stained sections were evaluated histologically for the following features associated with vacuolar hepatopathy: excessive hepatocellular cytoplasmic reticulation, hepatocytes with vacuolated cytoplasm, and hepatocellular swelling. Samples were categorized as follows: 1, met none of these criteria (histologically normal); 2, met ≤ 1, but not all criteria; or 3, met all criteria.
High cortisol = Serum cortisol concentration exceeded reference interval. High noncortisol = Concentration of ≤ 1 noncortisol steroid hormone exceeded reference interval; noncortisol steroid hormones assessed were androstenedione, estradiol, progesterone, 17-hydroxyprogesterone, and aldosterone.
See Table 1 for remainder of key.
The serum concentration of at least 1 noncortisol steroid hormone was above the reference intervals after ACTH stimulation in 17 of 17 dogs with, and 15 of 17 dogs without, hyperphosphatasemia (Tables 2 and 3). Progesterone concentrations were increased in 12 of 17 dogs in both groups after ACTH administration, and 17-hydroxyprogesterone concentrations were increased in 12 of 17 dogs with hyperphosphatasemia and in 10 of 17 dogs without hyperphosphatasemia. Concentrations of androstenedione and estradiol before ACTH stimulation, and of estradiol after ACTH stimulation, were significantly greater in dogs with hyperphosphatasemia, compared with concentrations in dogs without hyperphosphatasemia.
Because the magnitude of hormone elevation may be important in interpretation of noncortisol adrenal hormone concentrations,34 a cutoff value of 2× the upper limit of the normalized reference interval was evaluated for each hormone. Progesterone concentrations exceeded this cutoff value after ACTH stimulation in 2 dogs with and 4 dogs without hyperphosphatasemia; the cutoff for 17-hydroxyprogesterone was exceeded in 4 dogs with and 5 dogs without hyperphosphatasemia. Examination of noncortisol steroid hormones in combination revealed that only 2 dogs without hyperphosphatasemia had concentrations of both progesterone and 17-hydroxyprogesterone above the cutoff value after ACTH stimulation. Mean concentrations of progesterone and 17-hydroxyprogesterone before and after ACTH stimulation, and mean concentration of androstenedione before ACTH administration, were above the respective reference intervals in dogs with hyperphosphatasemia. Significant positive correlations were identified for serum ALP activity with age, alanine aminotransferase activity, concentrations of serum cortisol and androstenedione after ACTH stimulation, serum activity of the corticosteroid ALP isoform, and percentage of ALP represented by the corticosteroid isoform (Table 4).
Correlation between numeric variables and serum ALP activity in apparently healthy Scottish Terriers with and without hyperphosphatasemia.*†
| Variable | P value | |
|---|---|---|
| Age | 0.528 | < 0.001 |
| ALT activity | 0.613 | < 0.001 |
| Cortisol concentration after ACTH stimulation | 0.337 | 0.050 |
| Androstenedione concentration after ACTH stimulation | 0.496 | 0.003 |
| Corticosteroid ALP isoform activity | 0.989 | < 0.001 |
| Corticosteroid ALP isoform percentage‡ | 0.356 | 0.039 |
Results are reported only for those variables that had a significant (P < 0.05) correlation with ALP activity.
Fraction (%) of total ALP represented by the corticosteroid ALP isoform.
See Table 1 for remainder of key.
In 3 of 11 dogs with and 1 of 5 dogs without hyperphosphatasemia, the liver was subjectively considered to be large during ultrasonographic examination; this finding was not detectable during abdominal palpation as part of the physical examination. Biliary tract findings and echogenicity of the liver parenchyma were unremarkable in the 16 examined dogs. Adrenal glands were determined to be of normal size (< 7 mm, measured ultrasonographically) in 15 of 16 dogs; a 4 × 4-mm cortical nodule was detected in 1 dog with hyperphosphatasemia that had also had high serum concentrations of noncortisol adrenal hormones.
In 11 of 11 dogs with hyperphosphatasemia and in 4 of 5 dogs without hyperphosphatasemia, histologic examination of stained sections from liver biopsy specimens identified cytomegaly with minimal displacement of nuclei in hepatocytes (Figure 1). Rarified cytoplasmic staining was also observed in these samples; this was accompanied by reticulated cytoplasm in most affected cells, but a few affected cells had vacuolated cytoplasm (Figure 2). The cytoplasmic change was diffusely distributed throughout the lobules in dogs with hyperphosphatasemia but was primarily observed within midacinar hepatocytes (zone 2) in dogs without hyperphosphatasemia. Other histologic features included minor amounts of lipofuscin in many of the hepatocytes, and small amounts of hemosiderin were occasionally seen in Kupffer cells. Cholangitis, unusual changes in portal cellularity, and canalicular bile were not observed in any samples. Hepatocyte cytoplasm was intensely stained with PAS but was virtually unstained in sections subjected to amylase digestion prior to PAS staining. Interpretation of the results of histologic observations (Table 3) for the majority of dogs with (11/11 examined) and without (4/5 examined) hyperphosphatasemia was moderate to severe hepatic cytoplasmic reticulation and vacuolation compatible with glycogen accumulation; this finding is commonly associated with glucocorticoid hepatopathy or glycogen storage diseases. Subjectively, severity and distribution of reticulation were greater in the hyperphosphatasemic dogs. Hepatic reticulation was the most common and prominent feature seen in those dogs meeting partial or full criteria (cytoplasmic reticulation, vacuolar cytoplasm, and cytomegaly) established for vacuolar hepatopathy categorization, and vacuolation was only detected in those meeting all criteria. The number of samples that met all the vacuolar hepatopathy criteria was overrepresented in the hyperphosphatasemic dogs, whereas the number that met partial criteria (1 or 2 of the described features) for vacuolar hepatopathy was significantly (P < 0.001) overrepresented in the dogs without hyperphosphatasemia. Only 1 of the dogs without hyperphosphatasemia was classified as having a normal liver biopsy sample. Copper concentrations were within the reference interval (liver dry weight: normal, 120 to 400 ppm; deficient, < 80 ppm; toxic, > 1,500 ppm) for biopsy specimens from all 16 dogs evaluated, and there was no significant (P = 0.319) difference between the 2 groups.
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier without hyperphosphatasemia. Notice the zone 2 hepatocytes with rarified reticulated cytoplasm and central nuclei (most commonly observed in samples from healthy Scottish Terriers without hyperphosphatasemia); small foci of periportal lymphocytes are visible in the section. P = Portal area. L = Lymphocytes. 1 = Zone 1 hepatocytes. 2 = Zone 2 hepatocytes. H&E stain; bar = 50 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier without hyperphosphatasemia. Notice the zone 2 hepatocytes with rarified reticulated cytoplasm and central nuclei (most commonly observed in samples from healthy Scottish Terriers without hyperphosphatasemia); small foci of periportal lymphocytes are visible in the section. P = Portal area. L = Lymphocytes. 1 = Zone 1 hepatocytes. 2 = Zone 2 hepatocytes. H&E stain; bar = 50 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier without hyperphosphatasemia. Notice the zone 2 hepatocytes with rarified reticulated cytoplasm and central nuclei (most commonly observed in samples from healthy Scottish Terriers without hyperphosphatasemia); small foci of periportal lymphocytes are visible in the section. P = Portal area. L = Lymphocytes. 1 = Zone 1 hepatocytes. 2 = Zone 2 hepatocytes. H&E stain; bar = 50 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier with hyperphosphatasemia. Notice the diffusely swollen hepatocytes with rarified reticulated cytoplasm and central nuclei, and a few cells with more distinct cytoplasmic vacuolation (observed in samples from dogs with and without hyperphosphatasemia). V = Hepatocyte with distinct cytoplasmic vacuolation. R = Hepatocytes with reticulated cytoplasm. H&E stain; bar = 20 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier with hyperphosphatasemia. Notice the diffusely swollen hepatocytes with rarified reticulated cytoplasm and central nuclei, and a few cells with more distinct cytoplasmic vacuolation (observed in samples from dogs with and without hyperphosphatasemia). V = Hepatocyte with distinct cytoplasmic vacuolation. R = Hepatocytes with reticulated cytoplasm. H&E stain; bar = 20 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Photomicrograph of a section from a liver biopsy specimen obtained from a Scottish Terrier with hyperphosphatasemia. Notice the diffusely swollen hepatocytes with rarified reticulated cytoplasm and central nuclei, and a few cells with more distinct cytoplasmic vacuolation (observed in samples from dogs with and without hyperphosphatasemia). V = Hepatocyte with distinct cytoplasmic vacuolation. R = Hepatocytes with reticulated cytoplasm. H&E stain; bar = 20 μm.
Citation: Journal of the American Veterinary Medical Association 237, 2; 10.2460/javma.237.2.178
Discussion
Analysis of the results of the study reported here indicated a relationship between hyperphosphatasemia and excess serum adrenocortical steroid hormone concentrations in apparently healthy Scottish Terriers. This finding was reinforced by predominance of the corticosteroid ALP isoform in serum samples and by evidence of hepatocellular swelling and accumulation of glycogen (inferred by observation of widespread and common hepatic reticulation in H&E sections and determined to be compatible with glycogen by use of PAS stain) in fixed sections of liver biopsy specimens from hyperphosphatasemic dogs, results that are associated with hyperadrenocorticism.26 Other causes of elevated ALP activity were not apparent. There was no histologic evidence of hepatitis or copper-associated hepatotoxicosis. Liver function was determined to be normal on the basis of the results of bile acids test analysis, the lack of evidence of focal disease of the hepatic parenchyma or biliary system during ultrasonic examination of the abdomen, and the lack of substantial histologic abnormalities other than hepatocellular reticulation.
Concentrations of ≥ 1 adrenocortical hormone greater than the reference interval were detected in 32 of 34 Scottish Terriers in this study, whereas only 17 had ALP activity greater than the reference interval. Only 6 dogs had serum cortisol concentrations greater than the reference interval after ACTH stimulation; all of these had hyperphosphatasemia. However, 15 of 16 dogs from which liver biopsies were obtained (including dogs with [11/11] and without [4/5] hyperphosphatasemia) had histopathologic changes consistent with exposure to excess concentrations of circulating corticosteroids. Hepatocellular reticulation, vacuolation, or both have been reported in dogs without naturally occurring or iatrogenic hyperadrenocorticism; this was suggested to be the result of exposure to excess concentrations of glucocorticoids caused by stressful nonadrenal disease (ie, disease in organs or systems other than the adrenal glands).26 Neoplasia and acquired hepatobiliary disease were most commonly associated with hepatocellular vacuolation in a large retrospective study26 of vacuolar hepatopathy. Undetected illness was unlikely to be the cause of hyperphosphatasemia and hepatocellular reticulation in the Scottish Terriers described in the present study. The dogs did not have signs of disease according to the owners and were found to be free of nonadrenal disease via a thorough evaluation that included physical examination, extensive biochemical and hematologic testing, and ultrasonographic examination of the abdomen. However, similar testing performed again in the same dogs would allow more definitive conclusions to determine whether the biochemical and histopathologic changes are persistent and unchanged. In addition, determination of the cause or origin of the excess steroid hormone secretion would be important in elucidating the pathogenesis of this disorder.
The finding of adrenocortical hormone concentrations greater than the reference intervals in Scottish Terriers with and without hyperphosphatasemia leads to questions regarding the relationships among hormone excess, increased ALP activities, and hepatocellular reticulation. Some nonadrenal diseases are associated with activation of the hypothalamic-pituitary-adrenal axis, which results in excessive adrenal gland secretion of corticosteroids.35,36 Increased concentrations of 17-hydroxyprogesterone have been reported after administration of ACTH in dogs with nonadrenal illness, and the specificity of this hormone in the diagnosis of hyperadrenocorticism is much lower than that of cortisol.37,38 Although some investigators have reported the sensitivity of nonadrenal hormones to be less than that of cortisol, Ristic et al39 reported that 17-hydroxyprogesterone can be used to confirm a diagnosis of hyperadrenocorticism in dogs that do not have exaggerated cortisol responses to ACTH. It is possible that the high concentrations of noncortisol steroid hormones reported here are the result of an undetected nonadrenal disease in 32 of the 34 Scottish Terriers. However, changes in the liver consistent with exposure to steroid hormone excess were found in 4 of 5 dogs without hyperphosphatasemia from which biopsy specimens were collected; all 4 dogs had increased concentrations of noncortisol steroid hormones after ACTH stimulation.
The finding of adrenal glands with normal sizes in all but 1 dog with hyperphosphatasemia is surprising in light of other evidence of excessive adrenocortical function. Adrenomegaly has been detected during abdominal ultrasonographic examination in 77% to 80% of dogs with pituitary-dependent hyperadrenocorticism40,41 and would therefore be expected in most dogs with hyperphosphatasemia in the study reported here. The most likely explanation for this finding would be that the increases of adrenocortical steroid hormones in these dogs were mild and associated with subclinical adrenal gland enlargement or that the steroid hormones originated in a site other than the adrenal cortex. The small increases in circulating hormone concentrations above reference intervals, lack of detectable clinical signs, and exaggerated response to ACTH administration determined in the present study make a diagnosis of subclinical hyperadrenocorticism most likely.
Another consideration would be that Scottish Terriers may typically have exaggerated adrenocortical hormone secretion and response to ACTH such that the hormone results are not abnormal for this breed, although as previously discussed, the histopathologic findings argue against this. It is possible that Scottish Terriers have unique adrenocortical function or perhaps have a hypothalamic-pituitary-adrenal response that is sensitive to mild and undetected nonadrenal disease or to the stress of travel and hospitalization. Although a breed-specific pattern of unstimulated hormone concentrations and response to ACTH administration could be another explanation, the authors are not aware that any unique pattern of adrenocortical hormone secretion specific to a single breed has been reported. The reference intervals we used to evaluate the results of adrenal gland function tests were established at the reference laboratory, which performed the adrenal assays in accordance with standard methods.22 We have no reason to assume these ranges were not applicable to our study population, nor are we aware of the percentage of dogs of breeds other than the Scottish Terrier in which all adrenal hormone concentrations are within the reference interval. However, we can infer from the standard statistical approach used to establish these ranges that there would be a probability of only 2.5% that a healthy dog would have a serum cortisol concentration above the reference interval after ACTH stimulation and a probability of 12.5% of that dog having ≥ 1 of the 5 other noncortisol steroid hormones similarly increased. Those percentages of false-positive results are notably smaller than the percentages of positive results after ACTH stimulation in the study reported here: 35% of dogs with hyperphosphatasemia had high serum cortisol concentrations, and 100% of dogs with and 88% of dogs without hyperphosphatasemia had high concentrations of ≥ 1 noncortisol steroid hormone.
Scottish Terriers with ALP activity within the reference interval were selected as control animals for dogs of the same breed that had hyperphosphatasemia. Unfortunately, we were unable to identify a sufficient number of Scottish Terriers without hyperphosphatasemia that had ages similar to those of the hyperphosphatasemic dogs. Although the failure to have an age-matched control group of dogs with ALP activity within the reference interval is a shortcoming of the study, it may indicate that the abnormalities found in Scottish Terriers with hyperphosphatasemia are part of a progressive disorder. The finding of similar, but less severe, histologic characteristics and adrenal gland test abnormalities in the group of dogs without hyperphosphatasemia supports this hypothesis. Investigators in other studies14,42 found that the fractions of liver-associated and corticosteroid-associated ALP isoforms increased with age in healthy dogs of various breeds; however, they failed to detect an increase in total serum ALP activity. In contrast, Nestor et al19 reported a positive association of ALP activity with age in Scottish Terriers, which was similar to findings reported in humans.43
Additionally, the positive correlation of ALP activity with serum cortisol concentration in the present study provided further evidence that hyperphosphatasemia might be a progressive disorder in this breed. Increased basal cortisol concentrations have been detected in geriatric dogs.42,43 This could account for some of the differences detected in hormone concentrations between dogs with and without hyperphosphatasemia in the present study (ie, in this study, dogs with hyperphosphatasemia were significantly older than dogs without hyperphosphatasemia). The finding in another study43 that the magnitude of changes in serum cortisol concentrations in response to ACTH administration was not different between young and geriatric dogs was consistent with the findings between the 2 groups of Scottish Terriers in the present study. Longitudinal studies that include clinicopathologic testing during an interval of several years will be necessary to determine if the histologic, hormonal, and biochemical abnormalities we observed are indeed progressive in this breed.
Another limitation of the present study was the lack of liver biopsy specimens from several of the dogs without hyperphosphatasemia. These were apparently healthy client-owned dogs without detectable changes in serum ALP activity, and many owners were reluctant to allow their dogs to undergo this procedure.
The lack of clinical signs of adrenal corticosteroid excess in dogs in the study reported here may be attributable to the fact that cortisol concentrations were only mildly increased, compared with the reference interval. In the dogs without hyperphosphatasemia, cortisol concentrations after ACTH administration remained within the reference interval in all samples. Additionally, the concentration of either progesterone or 17-hydroxyprogesterone was increased > 2× the upper limit of the reference intervals in 7 of 17 dogs with and 6 of 17 dogs without hyperphosphatasemia, which indicated mild excesses (< 2× the upper limit) of these hormones in most of the dogs. The chronic nature and gradual onset of hyperadrenocorticism could have resulted in owners' failure to notice subtle signs of hormone excess in these study dogs. For example, the significantly lower urine specific gravity (mean, 1.018) in dogs with hyperphosphatasemia, compared with the urine specific gravity (mean, 1.037) of dogs without hyperphosphatasemia, supports some effect of excess adrenal steroid hormone and could indicate some degree of polyuria not reported by the owners.
The syndrome of hyperphosphatasemia, hepatocellular reticulation, and serum concentrations of adrenocortical steroid hormones above the reference interval in Scottish Terriers appears to be a benign condition. None of the dogs in the present study had clinical or biochemical evidence of liver dysfunction, and none of the examinations resulted in findings consistent with clinically important hyperadrenocorticism. Additional studies are necessary to determine whether the abnormalities described here could become more severe with time and eventually result in clinically overt hyperadrenocorticism or hepatic dysfunction. Testing of Scottish Terriers with hyperphosphatasemia for hyperadrenocorticism, including measurement of noncortisol adrenocortical hormones, should be considered prior to pursuing more invasive procedures, such as a liver biopsy.
ABBREVIATIONS
| ALP | Alkaline phosphatase |
| PAS | Periodic acid-Schiff |
Veterinary Diagnostic Laboratory, College of Veterinary Medicine, University of Illinois, Urbana, Ill.
Clinical Endocrinology Service, University of Tennessee, Knoxville, Tenn.
Cortrosyn, Amphastar Pharmaceuticals, Rancho Cucamonga, Calif.
Multistix Reagent Strips, Siemens Healthcare Diagnostics, Deerfield, Ill.
Eclipse E4000, Nikon Inc, Melville, NY.
Atravet, Wyeth-Ayerst Laboratories, Guelph, ON, Canada.
Numorphan, Endo Pharmaceutical Inc, Chadds Ford, Pa.
Tru-Cut, Baxter Healthcare, Deerfield, Ill.
Colorado Veterinary Diagnostic Laboratories, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colo.
Minitab, version 15, Minitab, State College, Pa.
References
- 1.
Syakalima M, Takiguchi M & Yasuda J, et al. The canine alkaline phosphatases: a review of the isoenzymes in serum, analytical methods and their diagnostic application. Jpn J Vet Res 1998;46:3–11.
- 2.↑
Syakalima M, Takiguchi M & Yasuda J, et al. Separation and quantification of corticosteroid-induced, bone and liver alkaline phosphatase isoenzymes in canine serum. Zentralbl Veterinarmed A 1997;44:603–610.
- 3.
Solter PF, Hoffmann WE & Chambers MD, et al. Hepatic total 3α-hydroxy bile acids concentration and enzyme activities in prednisone-treated dogs. Am J Vet Res 1994;55:1086–1092.
- 4.
Stockham SL, Scott MA. Enzymes. In: Fundamentals of veterinary clinical pathology. Ames, Iowa: Iowa State Press, 2002;433–460.
- 5.
Stockham SL, Scott MA. Liver function. In: Fundamentals of veterinary clinical pathology. Ames, Iowa: Iowa State Press, 2002;461–486.
- 6.
Fernandez NJ, Kidney BA. Alkaline phosphatase: beyond the liver. Vet Clin Pathol 2007;36:223–233.
- 7.
Hoffmann G, van den Ingh TS & Bode P, et al. Copper-associated chronic hepatitis in Labrador Retrievers. J Vet Intern Med 2006;20:856–861.
- 8.
Shih JL, Keating JH & Freeman LM, et al. Chronic hepatitis in Labrador Retrievers: clinical presentation and prognostic factors. J Vet Intern Med 2007;21:33–39.
- 9.
Sanecki RK, Hoffmann WE & Gelberg HB, et al. Subcellular location of corticosteroid-induced alkaline phosphatase in canine hepatocytes. Vet Pathol 1987;24:296–301.
- 10.
Barroga EF, Kadosawa T & Okumura M, et al. Influence of vitamin D and retinoids on the induction of functional differentiation in vitro of canine osteosarcoma clonal cells. Vet J 2000;159:186–193.
- 11.
Muller PB, Taboada J & Hosgood G, et al. Effects of long-term phenobarbital treatment on the liver in dogs. J Vet Intern Med 2000;14:165–171.
- 12.
Wiedmeyer CE, Solter PE, Hoffmann WE. Alkaline phosphatase expression in tissues from glucocorticoid-treated dogs. Am J Vet Res 2002;63:1083–1088.
- 13.
Bunch SE. Hepatotoxicity associated with pharmacologic agents in dogs and cats. Vet Clin North Am Small Anim Pract 1993;23:659–670.
- 14.
Syakalima M, Takiguchi M & Yasuda J, et al. The age dependent levels of serum ALP isoenzymes and the diagnostic significance of corticosteroid-induced ALP during long-term glucocorticoid treatment. J Vet Med Sci 1997;59:905–909.
- 15.
Allen MJ, Hoffmann WE & Richardson DC, et al. Serum markers of bone metabolism in dogs. Am J Vet Res 1998;59:250–254.
- 16.
Barger A, Graca R & Bailey K, et al. Use of alkaline phosphatase staining to differentiate canine osteosarcoma from other vimentin-positive tumors. Vet Pathol 2005;42:161–165.
- 17.
Komnenou A, Karayannopoulou M & Polizopoulou ZS, et al. Correlation of serum alkaline phosphatase activity with the healing process of long bone fractures in dogs. Vet Clin Pathol 2005;34:35–38.
- 18.
Nozaki K, Kadosawa T & Nishimura R, et al. 1,25-Dihydroxyvitamin D3, recombinant human transforming growth factor-beta 1, and recombinant human bone morphogenetic protein-2 induce in vitro differentiation of canine osteosarcoma cells. J Vet Med Sci 1999;61:649–656.
- 19.↑
Nestor DD, Holan KM & Johnson CA, et al. Serum alkaline phosphatase activity in Scottish Terriers versus dogs of other breeds. J Am Vet Med Assoc 2006;228:222–224.
- 20.
Gallagher AE, Panciera DL, Panciera RJ. Hyperphosphatasemia in Scottish Terriers: 7 cases. J Vet Intern Med 2006;20:418–421.
- 21.
Lawler DF, Keltner DG & Hoffman WE, et al. Benign familial hyperphosphatasemia in Siberian Huskies. Am J Vet Res 1996;57:612–617.
- 22.↑
Frank LA, Hnilica KA, Oliver JW. Adrenal steroid hormone concentrations in dogs with hair cycle arrest (Alopecia X) before and during treatment with melatonin and mitotane. Vet Dermatol 2004;15:278–284.
- 23.
Oliver JW. Adrenal function testing in animals. Mod Vet Pract 1981;62:145–147.
- 24.
Hoffmann WE, Sanecki RK, Dorner JL. A technique for automated quantification of canine glucocorticoid-induced isoenzyme of alkaline phosphatase. Vet Clin Pathol 1988;17:66–70.
- 25.
Sanecki RK, Hoffmann WE & Hansen R, et al. Quantification of bone alkaline phosphatase in canine serum. Vet Clin Pathol 1993;22:17–23.
- 26.↑
Sepesy LM, Center SA & Randolph JF, et al. Vacuolar hepatopathy in dogs: 336 cases (1993–2005). J Am Vet Med Assoc 2006;229:246–252.
- 27.
Frank LA, Rohrbach BW & Bailey EM, et al. Steroid hormone concentration profiles in healthy intact and neutered dogs before and after cosyntropin administration. Domest Anim Endocrinol 2003;24:43–57.
- 28.
Fujita A, Sato JR & Rodrigues Lde O, et al. Evaluating different methods of microarray data normalization. BMC Bioinformatics [serial online]. 2006;7:469. Available at: www.biomedcentral.com/bmcbioinformatics.
- 29.
Ni TT, Lemon WJ & Shyr Y, et al. Use of normalization methods for analysis of microarrays containing a high degree of gene effects. BMC Bioinformatics 2008;9:505. Available at: www.biomedcentral.com/bmcbioinformatics.
- 30.
Park T, Yi SG & Kang SH, et al. Evaluation of normalization methods for microarray data. BMC Bioinformatics [serial online]. 2003;4:33. Available at: www.biomedcentral.com/bmcbioinformatics.
- 31.
Reverter A, Barris W & McWilliam S, et al. Validation of alternative methods of data normalization in gene co-expression studies. Bioinformatics 2005;21:1112–1120.
- 32.
Wu W, Xing EP & Myers C, et al. Evaluation of normalization methods for cDNA microarray data by k-NN classification. BMC Bioinformatics [serial online]. 2005;6:191. Available at: www.biomedcentral.com/bmcbioinformatics.
- 33.
Xie Y, Jeong KS & Pan W, et al. A case study on choosing normalization methods and test statistics for two-channel microarray data. Comp Funct Genomics 2004;5:432–444.
- 34.↑
Scott-Moncrieff J. Atypical and subclinical hyperadrenocorticism. In: Bonagura J, Twedt D, eds. Kirk's current veterinary therapy XIV. St Louis: Elsevier Saunders, 2009;219–224.
- 35.
Behrend EN, Kemppainen RJ & Boozer AL, et al. Serum 17-α-hydroxyprogesterone and corticosterone concentrations in dogs with nonadrenal neoplasia and dogs with suspected hyperadrenocorticism. J Am Vet Med Assoc 2005;227:1762–1767.
- 36.
Kaplan AJ, Peterson ME, Kemppainen RJ. Effects of disease on the results of diagnostic tests for use in detecting hyperadrenocorticism in dogs. J Am Vet Med Assoc 1995;207:445–451.
- 37.
Chapman PS, Mooney CT & Ede J, et al. Evaluation of the basal and post-adrenocorticotrophic hormone serum concentrations of 17-hydroxyprogesterone for the diagnosis of hyperadrenocorticism in dogs. Vet Rec 2003;153:771–775.
- 38.
Sieber-Ruckstuhl NS, Boretti FS & Wenger M, et al. Evaluation of cortisol precursors for the diagnosis of pituitary-dependent hypercortisolism in dogs. Vet Rec 2008;162:673–678.
- 39.↑
Ristic JM, Ramsey IK & Heath EM, et al. The use of 17-hydroxyprogesterone in the diagnosis of canine hyperadrenocorticism. J Vet Intern Med 2002;16:433–439.
- 40.
Grooters AM, Biller DS & Theisen SK, et al. Ultrasonographic characteristics of the adrenal glands in dogs with pituitarydependent hyperadrenocorticism: comparison with normal dogs. J Vet Intern Med 1996;10:110–115.
- 41.
Barthez PY, Nyland TG, Feldman EC. Ultrasonographic evaluation of the adrenal glands in dogs. J Am Vet Med Assoc 1995;207:1180–1183.
- 42.
Strasser A, Niedermuller H & Hofecker G, et al. The effect of aging on laboratory values in dogs. Zentralbl Veterinarmed A 1993;40:720–730.
- 43.↑
Heino AE, Jokipii SG. Serum alkaline phosphatase levels in the aged. Ann Med Intern Fenn 1962;51:105–109.
Appendix
Sex-specific reference intervals22,27 for serum adrenal steroid hormone concentrations in dogs before and after ACTH stimulation.
| Analyte | Before ACTH | After ACTH | ||||||
|---|---|---|---|---|---|---|---|---|
| Sexually intact female | Spayed female | Sexually intact male | Neutered male | Sexually intact female | Spayed female | Sexually intact male | Neutered male | |
| Cortisol (ng/mL) | 4.0–59.9 | 2.1–58.8 | 2.5–56.7 | 2.0–56.5 | 66.7–174.8 | 65.0–174.6 | 70.8–108.5 | 70.6–151.2 |
| Androstenedione (ng/mL) | 1.9–11.9 | 0.1–5.7 | 2.1–42.6 | 0.1–3.6 | 3.8–42.1 | 2.7–39.7 | 6.8–79.2 | 2.4–29.0 |
| Estradiol (pg/mL) | 31.5–65.4 | 30.8–69.9 | 30.5–66.6 | 23.1–65.1 | 31.8–63.1 | 27.9–69.2 | 30.0–65.6 | 23.3–69.4 |
| Progesterone (ng/mL) | 0.03–2.16 | 0.01–0.49 | 0.02–0.40 | 0.01–0.17 | 0.33–4.33 | 0.10–1.50 | 0.55–1.70 | 0.22–1.45 |
| 17-hydroxyprogesterone (ng/mL) | 0.05–0.69 | 0.01–0.77 | 0.06–0.84 | 0.01–0.22 | 0.68–4.44 | 0.40–1.62 | 0.37–2.87 | 0.25–2.63 |
| Aldosterone (pg/mL) | 3.5–139.9 | 3.5–139.9 | 3.5–139.9 | 3.5–139.9 | 72.9–396.5 | 72.9–398.5 | 72.9–398.5 | 72.9–398.5 |
For later statistical analysis, reference intervals for steroid hormones were normalized (range, −1.0 to 1.0).
