July 2017 Case of the Month

Case Answer

Pseudohypoproteinemia secondary to hyperbilirubinemia

Gall bladder mucocele resulting in extrahepatic biliary obstruction

Explanation

The leukogram is consistent with corticosteroid-mediated stress, as well as mild inflammation. Stress is supported by the mild hyperglycemia. Inflammation is supported by the mild monocytosis (this also could be stress-mediated in the dog), mild hypoalbuminemia (negative acute phase response), and rouleaux formation noted on blood smear examination. Rouleaux formation appears as RBCs stacked like coins in a linear arrangement. Increased concentrations of fibrinogen or immunoglobulin as occurs with inflammation or antigenic stimulation, result in a decreased negative charge on the RBC surface and weaker electrostatic repulsive force (Question 1).¹

Question 2: Hyperfibrinogenemia (presumptive in this case) is one possible explanation for the discrepancy between the plasma total protein measured by refractometry (TP_ref) and the serum total protein measured by the biuret spectrophotometry assay (TP_biuret) on the chemistry analyzer. Fibrinogen is a positive acute phase protein synthesized by the liver and increases with inflammation. It is present in plasma, but not in serum due its utilization in clot formation. Therefore, in cases of inflammation, the plasma TP_ref is often greater than the serum TP_biuret. Another cause of a discrepancy is the presence of increased concentrations of non-protein soluble solids including glucose, urea, electrolytes, and lipoproteins, which increase the refractive index measured by a refractometer.^2,3 The refractive index is the bending of light as it passes through liquid.³ Although typically only gross lipemia (due to chylomicrons) is thought to increase the refractive index², this dog likely has increased VLDL, which can increase the refractive index. Therefore, the marked increase in cholesterol is likely contributing to the refractive index in this case. Although the TP_biuret is usually the more accurate measure of protein given that it is not affected by these non-protein solids, this case is an example where the TP_ref may be a more accurate estimate of the total protein. This is due to the negative interference of hyperbilirubinemia on the TP_biuret, which is the third reason for a discrepancy in this case. This interference was reported in a case of immune-mediated hemolytic anemia in a dog with marked hyperbilirubinnemia (81 mg/dL).⁴ In this case report, there was a persistent discrepancy between the TP_ref (7.8 and 7.0 g/dL; reference interval, 6-8 g/dL) and TP_biuret (3.7 and 3.0 g/dl; reference interval 4.7-6.8 g/dL) over the span of three days. This discrepancy was interpreted as beyond the difference normally seen between plasma and serum total protein concentrations as measured by these two methods. In addition, given that globulin concentrations are determined by the subtraction of the albumin concentration (as determined by the bromcresol green method on the chemistry analyzer)² from the TP_biuret, this also resulted in a marked hypoglobulinemia (0.6 and 0.4 g/dL, RI 2.2-3.6 g/dL) in the reported case. Hyperbilirubinemia, as tested in canine serum samples spiked with bilirubin up to a concentration of 40 mg/dL, has been shown to not interfere with the refractometric estimation of serum TP.³ However, similar testing documented an abrupt, near-linear decrease in TP_biuret starting at a bilirubin concentration of 9.9 mg/dL (see discussion for details of this interference).⁵

Question 3: The marked increases in ALT and AST activities indicate hepatocellular injury. The mild increase in CK activity also indicates muscle injury, which may be contributing to the increase in AST activity to some extent. The marked increases in ALP and GGT activities, along with the moderate hyperbilirubinemia (predominantly direct) are consistent with cholestasis. Given the marked increase in these cholestatic liver enzymes and dominant direct bilirubin component, a structural cholestasis from biliary obstruction is favored (e.g. bile cleaves GGT from enzyme surfaces, increases serum activity). Differential diagnoses for biliary obstruction include a neoplasm (e.g. biliary, pancreatic), gallbladder mucocele, and cholelithiasis. Stress is also a likely contributor to the increase in ALP activity via the liver and corticosteroid isoforms.  The hepatocellular injury may be secondary to the cholestasis, however a hepatotoxin (e.g. copper, blue-green algae) would also be a consideration in this case. The marked hypercholesterolemia can also be attributed to cholestasis, however it also raises the possibility of altered lipid metabolism due to an endocrinopathy (e.g. hyperadrenocorticism, hypothyroidism). This dog was, in fact, previously diagnosed with hypothyroidism.

Given the differential diagnoses, an abdominal ultrasound is recommended to evaluate for an obstructive lesion. An abdominal ultrasound was performed in this case, which revealed an enlarged and moderately thickened gallbladder filled with thick material, consistent with a gallbladder mucocele. The liver was also moderately enlarged confirming this physical examination finding.

Discussion

The biuret total protein assay is a 2-point-end bichromatic assay, meaning that at each of two points during the assay (before and after the reaction), absorbance readings are taken at two different wavelengths (primary and secondary) with our analyzer.⁵ In the reaction, copper (a component of one of the added reagents) complexes with peptide bonds in an alkaline solution creating a blue-purple product, which is measured spectrophotometrically.⁵ The color is proportional to the number of peptide bonds, and thus the sample’s protein concentration.^2,5 The primary absorbance reading is at 546nm (represents the total absorbance of the sample) and the secondary absorbance reading is at 700 nm (represents any absorbance due to an interfering substance, such as lipid).⁵ The net absorbance (primary wavelength absorbance reading – secondary wavelength absorbance reading) at each measurement point is determined.⁵ If there is protein in the sample, the second net absorbance will be greater than the first net absorbance. The greater the change, the more protein present.

So, how does bilirubin interfere with this assay? In the previously mentioned spiking study, the primary wavelength absorbance measurements before and after the reaction increased with increasing bilirubin concentrations (starting at 9.9 mg/dL).⁵ This is because bilirubin has a maximum spectral absorption near 455nm, and therefore absorbs the primary wavelength light (546 nm), along with the intended protein in the sample.⁵ After the reaction, there was also an increase in the secondary wavelength absorbance readings.⁵ This cannot be explained by bilirubin, as it does not absorb the secondary wavelength light of 700 nm. It is hypothesized that biliverdin, which has a maximum spectral absorbance near 700 nm, is the interfering substance (this hypothesis has not been confirmed).^4,5 Bilirubin can be oxidized to biliverdin in an alkaline environment, which is the pH of the biuret reaction.^4,5 To understand how this would cause a false decrease in the final calculated absorbance value and thus the sample’s total protein concentration, see the simplified equation below (the dilution factor taking into account the addition of a reagent has been removed):

Final calculated absorbance ₌ (1λ_after – 2λ_after) – (1λ_before – 2λ_before)

1λ_after is the primary wavelength measurement after the reaction

2λ_after is the secondary wavelength measurement after the reaction

1λ_before is the primary wavelength measurement before the reaction

2λ_before is the secondary wavelength measurement before the reaction

Therefore, 1λ_after and 1λ_before are both falsely increased by the bilirubin. The 1λ_after is still greater than 1λ_before due to the presence of protein in the sample. 2λ_after is falsely increased, presumably by biliverdin production in the reaction, however 2λ_before is unchanged. Therefore, with progressively higher concentrations of bilirubin, the first term of the equation is progressively lower while the second term is progressively higher, resulting in a net false decrease.

Follow-up:
The dog was hospitalized overnight in preparation for surgery the next morning. A cholecystectomy was performed without complications. The dog recovered well from the surgery and re-check chemistry panels during hospitalization showed decreasing liver enzyme activities and bilirubin and cholesterol concentrations. The dog was discharged to the care of the owners three days post-operatively.

Test	Presentation	Day of surgery	1 day post-op	2 days post-op	Reference interval
ALT (U/L)	2698	1918	852	579	20-98
AST (U/L)	564	153	127	57	14-51
Alkaline phosphatase (U/L)	3590	2896	1383	1086	17-111
GGT (U/L)	309	264	130	86	0-6
Total bilirubin (mg/dL)	16.3	5.5	0.9	0.5	0.0-0.2
Direct bilirubin (mg/dL)	14.1	4.1	0.6	0.3	0.0-0.1
Indirect bilirubin (mg/dL)	2.2	1.4	0.3	0.2	0.0-0.2
Cholesterol (mg/dL)	782	570	310	261	138-332
CK (U/L)	1478	434	2132	425	48-261

The gall bladder was submitted for histopathologic examination, which confirmed the diagnosis of a gall bladder mucocele with mild, diffuse papillary hyperplasia. Surgical biopsies of the liver revealed multifocal, random, suppurative hepatitis with small foci of hepatocellular necrosis, nodular hyperplasia, portal and centrilobular lymphoplasmacytic infiltrates, and portal vein hypoperfusion. The pattern of hepatitis suggested an embolic showering of bacteria to the liver. However, aerobic and anaerobic bacterial cultures of bile were negative. The dog was discharged with multiple antibiotics to provide broad spectrum coverage for the possibility of bacteremia. Nodular hyperplasia is a common incidental finding in the liver of middle-aged to older dogs, and may have been a mild contributor to the increased ALP in this case. The portal lymphoplasmacytic infiltrates suggested an underlying component of inflammatory bowel disease. Abnormal splanchnic blood flow (most commonly due to portal vein compression) could have resulted in acquired portal vein hypoperfusion in this case.

References

1. Harvey JW. Veterinary Hematology: A Diagnostic Guide and Color Atlas. St. Louis: Elsevier; 2012.
2. Stockham SL, Scott MA. Fundamentals of Veterinary Clinical Pathology. Second. Ames, Iowa: Blackwell Publishing; 2008.
3. Gupta A, Stockham SL. Refractometric total protein concentrations in icteric serum from dogs. J Am Vet Med Assoc. 2014;244(1):63-67.
4. Garner BC, Priest H, Smith J. Pseudo-hypoproteinemia in a hyperbilirubinemic dog with immune-mediated hemolytic anemia. Vet Clin Pathol. 2014;43(2):266-269.
5. Gupta A, Stockham SL. Negative interference of icteric serum on a bichromatic biuret total protein assay. Vet Clin Pathol. 2014;43(3):422-427.

Authored by: Dr. Ashleigh Newman