PSEUDOHYPOPHOSPHATEMIA IN A PATIENT WITH MULTIPLE MYELOMA

Abstract

Objective:

To discuss the diagnosis and management of paraprotein interference in the setting of multiple myeloma (MM).

Methods:

We discuss the evaluation of hypophosphatemia in a patient with MM and present a review of the relevant literature.

Results:

Our patient, who had a history of MM, was found to have persistently undetectable serum phosphate which did not respond to aggressive phosphate replacement. His clinical condition was not consistent with severe phosphate depletion and hence paraprotein interference secondary to MM was suspected. Re-analyzation of samples on a different machine showed normal serum inorganic phosphate levels.

Conclusion:

Paraprotein interference from MM causing pseudohypophosphatemia can be overlooked and lead to unnecessary treatment. Recognition of this phenomenon is important to all clinicians, especially in light of potential complications of unnecessary treatment.

INTRODUCTION

Hypophosphatemia is an uncommon electrolyte imbalance in the general population. The prevalence of hypophosphatemia among hospitalized patients is estimated to be 0.2 to 2.2% (1). Severe hypophosphatemia is clinically important as it can lead to rhabdomyolysis, arrhythmias, respiratory failure, encephalopathy, hemolysis, and metabolic acidosis (2). The 3 major mechanisms of hypophosphatemia are internal redistribution, decreased intestinal absorption, and increased urinary excretion (1).

In the setting of multiple myeloma (MM), patients may have hypophosphatemia or hyperphosphatemia. True hypophosphatemia may be due to renal phosphate wasting secondary to proximal tubulopathy, also known as Fanconi syndrome (1). Hyperphosphatemia may commonly be seen in patients with advanced renal failure due to MM. We present a case that highlights profound pseudohypophosphatemia from paraprotein interference due to MM and subsequent iatrogenic hypocalcemia induced by aggressive phosphate repletion.

CASE REPORT

A 72-year-old male patient with a history of hypertension, hyperlipidemia, and recently diagnosed MM presented with acute back and abdominal pain. He complained of localized lower left back pain impairing his daily activity and generalized abdominal pain associated with constipation for several days. One day before admission, the patient had received his first cycle of chemotherapy which consisted of dexamethasone, lenalidomide, bortezomib, and intravenous zoledronic acid infusion. Home medications included acyclovir, amlodipine, aspirin, cyclobenzaprine, hydrocodone/acetaminophen, omeprazole, simvastatin, tamsulosin, and thiamine. Social history was pertinent for chronic alcoholism in the past, but abstinence for the last 6 months. Family history was non-contributory. The patient had no history of bowel resection, malabsorption, or chronic diarrhea.

Vital signs on admission were normal with temperature 98.8°F, heart rate 90 beats/minute, respiratory rate 18 breaths/minute, blood pressure 153/74 mm Hg, and oxygen saturation 97% on room air. Physical exam showed a well-nourished male, alert but in discomfort. On deep palpation, there were spots of mild tenderness along the lower left lateral back and in bilateral lower quadrants of the abdomen.

The complete metabolic panel on admission was significant for hyponatremia (sodium 127 mmol/L), hypercalcemia (total calcium 10.3 mg/dL, albumin 2.3 g/dL, albumin-corrected calcium 11.7 mg/dL), and high globulin 10 g/dL. Other electrolytes (potassium 3.6 mmol/L, chloride 93 mmol/L, bicarbonate 23.9 mmol/L, magnesium 1.6 mEq/L), urea nitrogen (10 mg/dL), and serum creatinine (0.87 mg/dL) were all normal. The complete blood count showed anemia (a significant drop in hemoglobin from baseline of 11 to 12 mg/dL to 8.9 mg/dL), normal white blood cell (6.2 × 10³ cells/μL) and normal platelets (168 × 10³ cells/μL).

Computerized tomography of the abdomen and pelvis was significant for mild wall thickening of the cecum and distended terminal ileum suggestive of colitis. The patient was admitted for suspected colitis and workup for acute back pain. Constipation and abdominal pain were attributed to opioid-induced constipation and bortezomib-induced ileus. The abdominal pain and constipation improved with a bowel regimen of multiple doses of senna, lactulose, and polyethylene glycol powder. Magnetic resonance imaging of the thoracic spine showed stable compression of T10 and a new 7-mm lesion in T8 and another 19-mm lesion in T12; hence acute back pain was attributed to fracture and new MM lesions in the thoracic spine.

On admission day 2, the patient was found to have an undetectable serum phosphate level of <0.5 mg/dL, the lowest reportable value at our institution. Aggressive phosphate repletion, both oral and intravenous, was initiated. On day 3, there was a further drop in hemoglobin from 9.8 g/dL to 6.8 g/dL, and the patient had respiratory distress requiring 4L of oxygen via nasal cannula, concerning for symptomatic severe hypophosphatemia.

After 2 days of phosphate replacement, laboratory test results showed hypocalcemia with serum calcium of 7.1 mg/dL and ionized calcium of 0.90 mmol/L (reference range is 1.15 to 1.29 mmol/L), low 25-hydroxyvitamin D of 20 ng/mL (reference range is 30 to 100 ng/mL), normal 1,25-dihydroxyvitamin D at 34 pg/mL (reference range is 18 to 72 pg/mL), and high parathyroid hormone of 421 pg/mL (reference range is 12 to 88 pg/mL) without baseline values to compare.

In the setting of MM and lack of clinical suspicion for phosphate depletion via gastrointestinal loss, acquired Fanconi syndrome was considered as the initial diagnosis. Twenty-four-hour urine sample showed high urinary phosphate at 2.15 g/day (reference range is 0.40 to 1.30 g/day), undetectable urinary calcium and glucose, and normal urinary uric acid 510 mg/day (reference range is 250 to 750 mg/day). It was not possible to calculate fractional excretion of phosphate due to undetectable serum phosphate.

Throughout hospitalization the patient remained alert, awake, and maintained fair appetite. Further workup of anemia and respiratory distress revealed anemia of chronic disease and significant pulmonary atelectasis due to pain and bed rest. Mild hyponatremia was attributed to syndrome of inappropriate secretion of antidiuretic hormone due to pain and was corrected with fluid restriction.

Within 5 days of hospitalization, the serum phosphate remained persistently undetectable despite aggressive replacement (a total phosphate dose of 9,750 mg; oral 9,000 mg and 750 mg intravenous). While there was no improvement in hypophosphatemia despite aggressive phosphate repletion, serum calcium steadily dropped from 10.3 mg/dL to 6.7 mg/dL (Fig. 1). As the clinical condition did not match with the profound refractory hypophosphatemia, paraprotein interference with serum phosphate assay was suspected.

Fig. 1.

The patient’s true serum phosphate and calcium values throughout the hospitalization. The serum phosphate was falsely decreased due to paraprotein interference. Given the aggressive phosphate repletion in days 1 through 5, the patient developed hypocalcemia due to phosphate binding. The patient’s phosphate and calcium normalized after phosphate repletion was discontinued.

Initial serum samples showed persistent hypophosphatemia using the UniCel DxC 800 Synchron Clinical System (Beckman Coulter, Inc., Brea, CA) were sent out for reanalysis at Quest diagnostic lab utilizing the Olympus AU5800 (Beckman Coulter). These tests showed normal serum phosphate values of 3.1 mg/dL and 2.8 mg/dL. Serial dilution of the serum sample on admission day 5 showed a true hyperphosphatemia (6 mg/dL with 10 times dilution and 5.6 mg/dL with 4 times dilution), correlating well with the serum phosphate level of 6.1 mg/dL from Quest diagnostic lab.

Once pseudohypophosphatemia due to paraprotein interference was suspected, all phosphate replacements were stopped and calcium carbonate and calcitriol were judiciously supplemented. Serum calcium improved to normal (9.2 mg/dL) within 2 days of discontinuing phosphate replacement (Fig. 1). The patient was discharged after his clinical condition improved. It was cautioned by the laboratory that future phosphate measurements for this patient should be sent out or analyzed with sample dilution.

DISCUSSION

Hypophosphatemia can be acute or chronic. Acute hypophosphatemia with phosphate depletion can be seen among critically ill patients and may lead to significant complications such as anemia, generalized weakness, respiratory failure, seizures, and encephalopathy. Because of hypophosphatemia and phosphaturia, renal phosphate wasting from a proximal renal tubulopathy such as acquired Fanconi syndrome was considered. The absence of glucose and uric acid wasting in the urine made Fanconi syndrome unlikely. The acute worsening of anemia and respiratory distress initially misled us to consider as a true symptomatic severe hypophosphatemia.

However, the persistently undetectable serum phosphate values despite aggressive repletion did not correlate with the clinical picture of our patient who was well-nourished, had normal oral intake, and no evidence of phosphate loss. Mildly elevated urinary phosphate can be explained by homeostatic response of renal phosphate excretion to excessive phosphate replacement. The hypocalcemia may be due to binding of inorganic phosphate with serum calcium secondary to aggressive phosphate replacement and also the effect of Zoledronic acid infusion which was given just before the admission. High parathyroid hormone can be explained by secondary hyperparathyroidism due to hypocalcemia and vitamin D deficiency.

In tests run on the DxC 800 analyzer, paraproteins can cause sample precipitation when mixed with reagent resulting in undetectable serum phosphate values. An accurate result can be achieved by dilution with normal saline solution or using a protein-free filtrate prepared with trichloroacetic acid. The use of surfactant in the AU5800 analyzer eliminates the need to prepare a protein-free filtrate and minimizes paraprotein interference. Therefore, the AU5800 analyzer was able to provide the correct serum phosphate values. The possibility of analyzer malfunction was ruled out by having normal quality control measures on the days of testing and no such low serum phosphate values were reported in samples from all other patients.

In our case, hypophosphatemia was a false measurement due to paraprotein interference. This finding has been reported in other case reports (3–6). There is no well-defined mechanism. Some postulated mechanisms for paraprotein interference include IgG interference with formation or stabilization of phosphomolybdate in automatic auto-analyzers, or increased binding of inorganic phosphate (3). Paraprotein interference in MM can cause not only pseudohypophosphatemia but also pseudohyperphosphatemia (7–10).

Other plasma cell dyscrasias and lymphoreticular malignancies associated with abnormal immunoglobulin synthesis such as Waldenstrom macroglobulinemia have also been identified in paraprotein interference (11). Paraproteins may also interfere with measurements of lipids, bilirubin, calcium, and sodium (12,13). When paraprotein interference is suspected, blood samples should be deproteinized, ultra-filtrated, or diluted prior to analysis or should be run on a different automated assay known to have less interference by paraproteins.

CONCLUSION

The differential diagnosis for hypophosphatemia is broad, and phosphate-wasting renal tubulopathies such as acquired Fanconi syndrome in MM is a possible differential diagnosis. However, pseudohypophosphatemia due to paraprotein interference with laboratory assays is also possible. Even though such a phenomenon does not routinely occur in every MM patient, this should be considered whenever the clinical picture does not correlate with the laboratory result. The unnecessary phosphate replacement in such patients can result in complications, such as the hypocalcemia seen in our patient. Further mechanistic studies should be conducted to characterize paraprotein interference.

Abbreviation

MM

multiple myeloma