Back to Basics
Ilan Gabriely, M.D., James P. Leu, M.D., and Uriel S. Barzel, M.D.
A 41-year-old woman was brought by her husband to the emergency department with a history of 72 hours of epigastric pain, nausea, repeated vomiting, and altered mental status. Her blood calcium was found to be 18.9 mg per deciliter (4.7 mmol per liter).
Hypercalcemia of this degree is a medical emergency. The patient with severe hypercalcemia is invariably dehydrated, and the first line of treatment should be vigorous hydration with intravenous normal saline with close observation of blood electrolytes and renal function. Additional treatment measures would depend on the cause of the hypercalcemia, the history, and the result of the workup.
The patient’s husband reported that 4 days earlier they had returned from a vacation in Central America, where the patient had consumed “a lot of alcohol.” He noted that she was in her usual state of health until 3 days earlier, when she started having severe abdominal pain, followed by multiple episodes of vomiting. The patient had a history of peptic ulcer disease years before and still noted some occasional episodes of abdominal pain, which she had been treating with chewable antacid tablets containing calcium carbonate (Tums, GlaxoSmithKline). She also had a long history of chronic low back pain, which she treated with over-the-counter analgesics. A few months earlier, she had been told by her primary care physician that she had a mildly elevated blood calcium level of 10.9 mg per deciliter (2.7 mmol per liter).
Her physician was contacted and reported that her intact parathyroid hormone (PTH) blood level at her last examination was 33.9 pg per milliliter (normal range, 7 to 53). No further diagnostic studies were done. Medications at home included Tums as needed for abdominal pain and a multivitamin. Habitually, she had been drinking 2 to 3 shots of vodka and some wine daily. Her family history was negative for parathyroid disease, nephrolithiasis, and cancer.
The history of a measurable PTH level in the face of elevated blood calcium suggests that the patient may have primary hyperparathyroidism, the most common cause of hypercalcemia. The level of hypercalcemia in this case, however, would be extremely rare with primary hyperparathyroidism; if it occurred, such a “parathyroid crisis” might require emergency parathyroidectomy. Given her history of low back pain, imaging studies should be obtained to evaluate the patient for lytic lesions from a malignant process. She should also be evaluated for other causes of hypercalcemia, including hyperthyroidism, vitamin D overdose, neoplasia, and granulomatous diseases.
The patient’s reported alcohol use suggests the possibility that alcohol withdrawal may be contributing to her altered mental status, although the extreme hypercalcemia alone could account for her presentation. Close monitoring for withdrawal signs and symptoms is mandatory. Amylase and lipase should also be checked, given the possibility of alcohol-induced pancreatitis.
On physical examination, the blood pressure was 160/90 mm Hg, the heart rate 100 beats per minute, and the temperature 37.4°C. The patient was disoriented and intermittently writhing in pain. Her skin turgor was poor, and her oral mucosa was dry. The cardiac examination was significant only for tachycardia, and the pulmonary examination was normal. There was epigastric tenderness on palpation but no rebound tenderness. The patient was responsive to simple commands and was not tremulous. There were no other significant neurologic findings.
The blood urea nitrogen was 13 mg per deciliter (4.6 mmol per liter); creatinine, 1.1 mg per deciliter (97 µmol per liter); sodium, 137 mmol per liter; potassium, 3.2 mmol per liter; chloride, 81 mmol per liter; bicarbonate, 37.2 mmol per liter; glucose, 96 mg per deciliter (5.3 mmol per liter); phosphorus, 2.0 mg per deciliter (0.65 mmol per liter); and magnesium, 1.2 mg per deciliter (0.49 mmol per liter). Blood tests for intact PTH, PTH-related protein, and 25-hydroxyvitamin D were performed.
CLICK TO ENLARGE
The physical findings are consistent with the presence of a hypovolemic state. Other than the tachycardia, there are no other stigmata of alcohol withdrawal. However, it is unclear when the patient had her last drink, so prophylaxis for withdrawal may be warranted. Hypochloremia and alkalemia are probably due to persistent vomiting. The intravenous administration of normal saline at a rate of 200 ml per hour may reduce the level of blood calcium and provide sufficient chloride ion to allow for the correction of the hypochloremia and the alkalosis.
The intact PTH level was undetectable. The lipase level was 1848 U per liter (normal range, 7 to 60), and the amylase level was 1354 U per liter (normal range, 40 to 128) (Table 1). The patient was treated with intravenous saline at a rate of 200 ml per hour, and prophylaxis for alcohol withdrawal with intravenous lorazepam was started. Intravenous morphine was given intermittently to alleviate abdominal pain.
Before hyperparathyroidism can be ruled out, it is important to consider whether the “hook effect” may explain the undetectable PTH level. To identify an antigen, an immunoassay must contain an excess of the antibodies relative to the antigen. The hook effect refers to a situation in which the level of the circulating antigen is very high and thus the antibodies are fully bound, resulting in a severe underestimation of the hormone level. To overcome this problem, the assay must be carried out in multiple dilutions. Hypomagnesemia may impair the release of PTH and mask the presence of hyperparathyroidism. The elevation of amylase and lipase confirms the suspicion of acute pancreatitis.
The patient was treated with 2 g of intravenous magnesium gluconate. A repeat serum magnesium level was 1.7 mg per deciliter (0.70 mmol per liter). Intact PTH was again undetectable on repeated assays with dilutions. Computed tomography (CT) of the neck revealed no parathyroid masses. CT of the chest was normal. Abdominal CT showed an edematous pancreas with surrounding infiltration of the mesenteric fat, a finding consistent with acute pancreatitis. No bone lesions were noted on the chest or abdominal CT scans.
The patient’s blood calcium level fell to 11.8 mg per deciliter (3.0 mmol per liter) by day 2 and to 8.5 mg per deciliter (2.1 mmol per liter) by day 3 with intravenous hydration alone, and the serum creatinine level fell to 0.7 mg per deciliter (62 mmol per liter). Her mental status slowly improved.
The initial information that was obtained from the patient’s primary physician, which indicated a mildly elevated calcium level and an inappropriately “normal” PTH level, suggested mild primary hyperparathyroidism, but the currently suppressed intact PTH level appears inconsistent with this diagnosis. Insofar as primary hyperparathyroidism is associated with resetting of the calcium receptor to a higher set point, it is possible that the patient does have underlying primary hyperparathyroidism but that a second disorder has independently raised the calcium level and thus suppressed the PTH secretion.
The rapid decline in the serum calcium level with intravenous hydration indicates that there is no need to consider urgent parathyroid surgery. The normalization of the very high calcium level with hydration also makes a number of other causes of hypercalcemia (such as solid tumors, hematopoietic tumors, granulomatous disease, and cancer metastatic to bone) less likely. Once the patient’s mental status improves further, a careful history of her diet and supplements should be obtained and evaluated for possible vitamin D and calcium overdose.
The serum 25-hydroxyvitamin D level was 48 ng per deciliter (120 nmol per liter; normal range, 20 to 100 ng [50 to 250 nmol]). The level of PTH-related protein was normal. The serum calcium level remained normal after the intravenous fluids were discontinued. The patient’s mental status further improved, and there was a gradual improvement in the clinical and laboratory markers of pancreatitis.
The normal 25-hydroxyvitamin D level rules out vitamin D toxicity. The episode of acute pancreatitis could have been secondary to hypercalcemia or to alcohol excess.
In response to further questioning, the patient reported that during the days before admission she had consumed the contents of entire containers of Tums and a preparation containing sodium bicarbonate (Alka-Seltzer, Bayer) because of the severe abdominal pain. She did not consume any other medications or supplements.
Tums contains calcium carbonate, and Alka-Seltzer contains sodium bicarbonate. Hypercalcemia is an inevitable result of the intake of alkali and calcium in the context of dehydration and metabolic alkalosis. This history, together with the laboratory-test results, establishes the “milk alkali syndrome” as the primary diagnosis.
Effective treatment for peptic ulcer disease was first introduced by Bertram Sippy1 in 1915. The “Sippy regimen” included hourly ingestion of milk and cream (and the gradual addition of eggs and cooked cereal) for 10 days, combined with the ingestion of alkaline powders. Although noncurative, this regimen provided some symptomatic relief.2 However, later reports showed serious toxicity associated with this regimen, including renal failure, alkalosis, and hypercalcemia, with normalization of all measures once the treatment was withdrawn.3 Over the next several decades, the milk alkali syndrome was frequently described, mostly in men with peptic ulcer disease who were receiving treatment with large amounts of calcium from milk and absorbable alkali.4 It proved fatal in some patients who had protracted vomiting due to secondary pyloric obstruction and who presented with hypovolemia, renal failure, alkalosis, and hypercalcemia. With the advent in recent years of better treatment options for peptic ulcer disease, the prevalence of the milk alkali syndrome has greatly declined.5
During the past 15 years, the milk alkali syndrome has been reported in patients without a history of peptic ulcer disease, most commonly in women taking calcium supplements at doses above the recommended range of 1200 to 1500 mg of elemental calcium daily for the prevention and treatment of osteoporosis.5 Furthermore, calcium has been added to many over-the-counter products and supplements, such as fast-acting antacids, vitamin preparations, juices, and even acetaminophen, which has provided multiple opportunities for inadvertent excessive intake of calcium by consumers.
The pathogenesis of the milk alkali syndrome involves a reduction in the ability of the kidney to excrete excess calcium. This reduction is secondary to a decrease in the glomerular filtration rate (due to renal vasoconstriction and hypovolemia) and to a significant increase in tubular reabsorption of calcium (secondary to metabolic alkalosis).6 The plasma PTH level decreases with the rise in the serum calcium level. Thus, the constellation of excess oral intake of calcium and milk, plus impaired renal function, may result in PTH suppression, hypercalcemia, and hyperphosphatemia. In cases of chronic ingestion of excessive alkaline calcium preparations and milk, metastatic calcifications and occasionally nephrocalcinosis may occur,7 whereas such complications are not expected with acute milk alkali syndrome, as in the present case.
The most important clue to the diagnosis of the milk alkali syndrome is a history of excessive calcium and alkali intake. Many patients do not consider over-the-counter preparations as medications and therefore do not report taking calcium supplements. Conversely, physicians may fail to identify certain over-the-counter preparations or supplements as a significant source of calcium. Hence, a detailed history attending to diet and supplement intake is critical.
Hypercalcemia, metabolic alkalosis, and impaired renal function are classic laboratory findings in patients with the milk alkali syndrome.8 Hyperphosphatemia, hypophosphaturia, and hypercalciuria may also be present, depending on the cause of the milk alkali syndrome. (For example, hyperphosphatemia is common in patients who have ingested excessive amounts of milk and calcium carbonate, whereas hyperphosphatemia does not generally develop in patients with an excessive intake of calcium carbonate alone.)
The documentation of low levels of intact PTH in patients with hypercalcemia is helpful in the diagnosis of the milk alkali syndrome,9 since this finding indicates that primary hyperparathyroidism cannot explain the hypercalcemia. When parathyroid function is normal, the plasma PTH level varies inversely with the plasma calcium level. Once excess calcium intake is stopped and the serum calcium level normalizes, plasma PTH may rebound. Reports indicate that the rebound in PTH level occurs rapidly (within hours) when excess calcium intake is stopped and the patient is vigorously hydrated and that the intact PTH level reaches a peak at 7 days.10 Intact PTH levels may be transiently elevated during this phase in patients who have an abrupt decrease (“overcorrection”) in blood calcium levels. Thus, it is critical that any intact PTH level be correlated with simultaneous blood calcium levels.
In conclusion, the milk alkali syndrome, which was commonly seen four decades ago in men, seems to be increasingly prevalent, probably because of the increased use of over-the-counter calcium preparations and supplements by women.10 A basic tenet of medical care is that a complete history is often the key to diagnosis. This is clearly true for the present case, in which the correct diagnosis depended on a history of excess calcium and alkali intake in a dehydrated and alkalotic patient, supported by the documentation of a normal 25-hydroxyvitamin D level and a suppressed level of intact PTH.