Acid-base and electrolyte disturbances in patients with hypercalcemia.

ABSTRACT

Background. In the present study, we analyzed acid-base and electrolyte disturbances in hypercalcemic patients to determine the principal causes of hypercalcemia.

Methods. We studied a total of 76 hypercalcemic patients and 91 healthy individuals. Acid-base and electrolyte parameters were determined before any therapeutic intervention.

Results. Hyperparathyroidism and neoplasias were the most common causes of hypercalcemia. Hypercalcemic patients had increased serum urea and creatinine levels, a higher urea/creatinine ratio, and a higher rate of acid-base disorders, but lower serum albumin, potassium, chloride, phosphorus, and magnesium concentrations than those found in the control subjects. Notably, significant differences in acid-base balance and electrolyte concentrations were evident between patients with hyperparathyroidism and patients with cancer.

Conclusions. Primary hyperparathyroidism and neoplasia are the most common causes of hypercalcemia. A wide array of concurrent acid-base and electrolyte disorders may be evident in hypercalcemic patients. Differences in these laboratory parameters are helpful in diagnostic workup.

**********

THE VARIOUS acid-base and electrolyte abnormalities that have been occasionally described in hypercalcemic patients could be of help in distinguishing between patients with hypercalcemia related to hyperparathyroidism and those with hypercalcemia of nonparathyroid origin. (1-6) Moreover, information about the incidence and pathophysiology of concurrent acid-base and electrolyte disturbances in both serum and urine of hypercalcemic patients would be most relevant in clinical practice. We undertook the present study to further analyze the acid-base and electrolyte variables in hypercalcemia and their relationship to the principal causes of hypercalcemia.

SUBJECTS AND METHODS

The study was done at the Department of Internal Medicine of the University Hospital of Ioannina, Greece. Although the upper limit of normal for serum calcium in our hospital is 10.4 mg/dL, we decided to define hypercalcemia as a serum calcium value [greater than or equal to] 10.5 mg/dL after correction for abnormalities in the serum protein levels (namely albumin), to exclude cases of doubtful significance. All patients with hypercalcemia hospitalized over a period of 2 years were recruited for the study. The 76 patients (51 male, 25 female) ranged in age from 22 to 81 years, with a mean age of 57. Patients presenting with hypercalcemia readily explicable by a high serum albumin level had a further assessment and were excluded from additional study if the initial high values were not confirmed. Patients with a history of receiving drugs affecting acid-base balance and electrolyte parameters, such as diuretics (other than thiazides), vitamin supplements, antacids, cisplatin or other chemotherapeutic agents, and antibiotics (eg, aminoglycosides), were also excluded from the study. However, hypercalcemic patients receiving drugs that are causally related to the development of hypercalcemia, such as lithium, vitamin D, and thiazide diuretics, were included.

The control group consisted of 91 nonhypercalcemic subjects evaluated at the outpatient clinics of our department for elective purposes. All patients and control subjects had fasting venous blood obtained before any therapeutic intervention for the measurement of serum immunoreactive parathyroid hormone (iPTH). Arterial blood gas determinations were also done. Serum anion gap (SAG) was calculated as follows: SAG = [Na.sup.+] - ([Cl.sup.-] + [HCO.sup.-.sub.3] (normal range, 12 [+ or -] 2 mmol/L). (7) Glucose, urea, creatinine, uric acid, sodium, potassium, chloride, calcium, phosphorus, and magnesium levels were determined in a 2-hour fresh urine specimen.

In the calculation of the serum [C1.sup.-]/[PO.sup.3-.sub.4], ratio, phosphate ion levels were expressed as milligrams per deciliter. Standard formulas were used for the determination of the fractional excretion (FE) of electrolytes. Maximum tubular reabsorption rate for phosphate ([TM][PO.sup.3-.sub.4],/GFR) was calculated by the Walton and Bijvoet nomogram. (8)

Laboratory determinations were done by automated chemical analysis in our laboratory using an Olympus AU 560 analyzer. Specifically, urine and serum samples were analyzed using ion-sensitive electrodes for sodium, potassium, chloride, and calcium, and photometric assays for phosphorus and magnesium. The glutamate dehydrogenase method was used for the determination of urea levels, and a modification of the Jaffe method for creatinine. Serum total protein concentrations were measured by the Biuret method, and serum albumin by the bromocresol green method. Finally, serum parathyroid hormone levels were estimated by an immunoradiometric assay (CIS bio international, Cedex, France), which can determine the intact biologic chain of 84 amino acids of PTH in human serum. The intraassay and interassay coefficients of variation (CV) were 7.5% and 6.8%, respectively. Detection limit was 0.7 pg/mL (normal, 11 to 62 pg/mL). Arterial pH and P[co.sub.2] were determined using a pH blood gas analyzer, and serum bicarbonate wa s calculated from blood hydrogen and blood carbon dioxide tension according to the Henderson-Hasselbalch equation with an acidity exponent of 6.10 and a solubility coefficient of 0.0301. Serum and urine osmolality was assayed using a vapor pressure osmometer.

The results were expressed as means [+ or -] SD. Student's t test or the nonparametric Mann-Whitney test was used for statistical comparison of laboratory parameters between groups. Additionally, chi-square test with Yates correction was done to assess differences in percentages.

RESULTS

The causes of hypercalcemia in our patients are shown in Table 1. Hyperparathyroidism and neoplasia (primary sites are shown in Table 2) were the most common causes of increased serum calcium levels. Moreover, hypercalcemic patients exhibited a wide array of concurrent acid-base and electrolyte abnormalities (Table 3) in concert with the principal causes of hypercalcemia. There were no significant differences in the incidence of the various acid-base and electrolyte abnormalities between the two groups, except for metabolic alkalosis, which was more common in cancer patients (Table 3).

Respiratory alkalosis was observed in six patients. These patients had alkalemia with a mean pH of 7.51 (range, 7.47 to 7.53), a mean P[co.sub.2] of 34 mm Hg (range, 30 to 36 mm Hg), and a mean bicarbonate concentration of 19 mmol/L (range, 17 to 22 mmol/L). Four of them had cancer, with extensive bone metastases and intractable pain. One of the remaining two patients had thyrotoxicosis, with symptoms and signs of increased sympathetic activity, and the other was found to have sarcoidosis with pulmonary involvement and subsequent hypoxemia ([Po.sub.2] 52 mm Hg).

Four patients had metabolic acidosis with a mean pH of 7.32 (range, 7.30 to 7.35), a mean [Pco.sub.2] of 30 mm Hg (range, 28 to 32 mm Hg), and a mean bicarbonate concentration of 18.5 mmol/L (range, 16 to 22 mmol/L). Two patients with primary hyperparathyroidism and one patient with Addison's disease had a normal anion gap (range, 8 to 12 mmol/L) hyperchioremic metabolic acidosis, while one patient with neoplasia and renal failure (serum creatinine value, 6.4 mg/dL) had a high anion gap (20 mmol/L) metabolic acidosis.

Pure metabolic alkalosis was found in 9 patients. These patients had a mean pH of 7.52 (range, 7.49 to 7.54), a mean [Pco.sub.2] of 45 mm Hg (range, 43 to 48 mm Hg), and a mean bicarbonate concentration of 31 mmol/L (range, 28 to 34 mmol/L). Seven of these patients with generalized neoplasia had also evidence of volume depletion (increased [>40] ratio of urea/creatinine), mild hypokalemia (serum potassium value, 3.0 to 3.4 mmol/L), and hypochloremia (serum chloride value, 93 to 96 mmol/L) associated with decreased urine chloride excretion (<20 mmol/L). Pure metabolic alkalosis was also evident in the two patients presenting with diuretic-induced hypercalcemia. Two patients had metabolic alkalosis coexisting with a primary respiratory alkalosis. These patients had alkalemia with increased serum bicarbonate levels, though the measured [Pco.sub.2] was less than predicted by the appropriate respiratory compensation of metabolic alkalosis. (9) Specifically, these patients had an arterial pH of 7.52 and 7.54, a [Pc o.sub.2] of 40 and 41 mm Hg, and a bicarbonate concentration of 32 and 34 mmol/L, respectively. Both patients had neoplasia and severe hypercalcemia (serum calcium value >14 mg/dL) with upper gastrointestinal symptoms and recurrent episodes of vomiting.

Three patients had hyponatremia (serum sodium value, <135 mmol/L; range, 131 to 134 mmol/L). The first patient had generalized cancer with ascites, the second had diuretic-induced hypercalcemia, and the third, who presented with hypercalcemia, hyperkalemia, hyponatremia, and metabolic alkalosis, was found to have Addison's disease. Hypernatremia (serum sodium value >145 mmol/L; range, 146 to 151 mmol/L) was evident in 4 elderly cancer patients (>65 years old) with severe hypercalcemia (serum calcium value >14 mg/dL). They had increased insensible losses because of fever, tachypnea, and excessive sweating, as well as clinical and laboratory findings of extracellular volume depletion. In these patients, urine osmolality was low (<300 mOsm/kg; range, 252 to 294 mOsm/kg), which was not increased after vasopressin administration.

Eight patients had hypokalemia (serum potassium value, <3.4 mmol/L; range 2.8 to 3.3 mmol/L), which was due to inappropriate kaliuresis (FE [K.sup.+] >6.4%, transtubular potassium gradient >2) in all but one patient. (10,11) The hypokalemic patients had higher serum calcium levels than the remaining normokalemic patients (15.4 [+ or -] 1.6 mg/dL vs 13.2 [+ or -] 1.9 mg/dL, P < .05). Additionally, about half of the patients with serum calcium levels >1.4 mg/dL were hypokalemic. The majority of patients with severe hypercalcemia (serum calcium value, >14 mg/dL) and normal renal function (serum creatinine value, <1.4 mg/dL) were hypokalemic. All hypokalemic patients also had alkalemia due to respiratory alkalosis, metabolic alkalosis, or mixed acid-base disorders. Four of the hypokalemic patients had hypomagnesemia. However, no evidence of other tubular dysfunction, such as glucosuria, uricosuria, or bicarbonaturia was noticed in patients with hypokalemia and inappropriate kaliuresis. Finally, hypokalemia with a ppropriate renal potassium conservation was found in the patient with severe thyrotoxicosis. Hyperkalemia (serum potassium value >5.3 mmol/L; range, 5.4 to 5.8 mmol/L) was found in two cancer patients with renal insufficiency, as well as in the patient with Addison's disease.

Hypomagnesemia (serum magnesium value <0.65 mmol/L; range, 0.48 to 0.62 mmol/L) was found in 8 patients and was accompanied by inappropriate urine magnesium levels (FE Mg ++ >4%). (12) Half of these patients also had hypokalemia. Hypomagnesemia patients had higher serum calcium levels than the remaining normomagnesemic patients (15.3 [+ or -] 1.3 vs 13.0 [+ or -] 1.2 mg/dL, P < 0.05).

Hypophosphatemia (serum phosphorus value <2.5 mg/dL; range, 1.8 to 2.4 mg/dL) was found in 9 patients. In 5 patients with primary hyperparathyroidism, hypophosphatemia was accompanied by inappropriate urine phosphorus levels, evidenced by an increased (>20%) FE [Po.sub.4], 3- and a decreased <2.7 mg/dL) [TmPO.sub.4], 3-/GFR. (8, 13) In 3 patients (2 with generalized cancer and 1 with leukemia), hypophosphatemia was found in association with severe respiratory alkalosis (arterial pH >7.50). Finally, in the patient with thyrotoxicosis, hypophosphatemia with appropriate renal phosphate conservation was evident along with severe respiratory alkalosis and hypokalemia.

Laboratory diagnostic test values for both patients and controls are shown in Table 4. Patients with hypercalcemia had higher serum urea and creatinine levels but lower serum albumin levels than controls. Serum urea/creatinine ratio, however, was also higher in hypercalcemic patients. Even though 19 hypercalcemic patients (7 with hyperparathyroidism and 12 with neoplasia) had increased serum creatinine levels on admission (1.6 to 6.4 mg/dL), there was a clear improvement in renal function after rehydration with normal saline. However, elevated serum creatinine levels persisted in 8 patients (range, 1.6 to 2.9 mg/dL). With regard to serum electrolytes, serum potassium, chloride, phosphorus, and magnesium levels were lower in patients than in controls, while no significant differences in the parameters of acid-base homeostasis were found between the two groups. Additionally, hypercalcemic patients had evidence of increased urine excretion of potassium, calcium, phosphorus, and magnesium.

Significant differences in the laboratory parameters were found between the two main groups of hypercalcemic patients (Table 5).

Specifically, patients with neoplasia exhibited higher values of serum urea, creatinine, urea/creatinine ratio, calcium, phosphorus, and bicarbonate, as well as higher arterial pH and [Pco.sub.2] levels, and lower values of serum albumin, chloride, magnesium, FE [PO.sub.4,.sup.3-], and urine calcium/creatinine ratio, as compared with hyperparathyroid patients. The difference in phosphorus levels between the two groups became clearer when patients with increased serum creatinine (>1.6 mg/dL) on admission were excluded (2.6 mg/dL versus 3.3 mg/dL, P<.01).

Serum chloride/phosphorus ratio was higher in hypercalcemic patients than in the control population (Table 4). Interestingly, serum chloride/phosphorus ratio was significantly lower in cancer patients than in patients with hyperparathyroidism. This difference was evident even after excluding patients with renal failure. There was a considerable overlap of data between the two groups. In fact, only 8 of 21 hyperparathyroid patients had a C1-/[PO.sub.4.,sup.3-] ratio >33, and only half of the cancer patients had a value <30, suggesting that these determinations cannot clearly differentiate between these two groups of hypercalcemic patients, as previously reported. (2)

Serum anion gap was not different between patients and controls. However, patients with hypercalcemia of parathyroid origin had significantly lower SAG values than patients with malignancy-associated hypercalcemia (Table 5), despite their higher serum albumin and lower serum calcium levels. This difference persisted, even though patients with increased serum creatinine on admission were excluded. A considerable overlap of SAG values between the two groups was evident. However, no cancer patient had a SAG value of 5 mEq/L or less, in contrast to 6 hyperparathyroid patients who had greater values. There was no correlation between SAG and serum calcium levels in either group or patients.

The PTH levels (reference range, 1.2 to 5.7 pmol/L) were increased in hyperparathyroid patients and decreased in patients with hypercalcemia of nonparathyroid origin (mean values, 19.2 [+ or -] 4.2 versus 1.9 [+ or -] 0.7 pmol/L, respectively; P< .01).

DISCUSSION

In the present study, malignancy appears to have been the most common cause of hypercalcemia. More than 90% of our patients had either neoplasia or hyperparathyroidism, a finding that is in agreement with reports in the literature (Table 2) (14-16) Although a wide variety of tumors were seen in association with hypercalcemia, carcinomas of the breast and bronchus as well as hematologic neoplasias, such as multiple myeloma and leukemia, accounted for the majority of cases of tumor-related hypercalcemia (Table 2). (14, 15, 17)

Hypercalcemic patients exhibited a wide array of concurrent acid-base and electrolyte abnormalities (Table 3). Mild respiratory alkalosis, either pure or in combination with metabolic alkalosis, was primarily found in cancer patients, possibly because of pain-induced hyperventilation. It is of interest that pure metabolic alkalosis was found in a number of patients with neoplasia, who did not exhibit other common causes of metabolic alkalosis, such as vomiting or diuretic therapy. In these cases, in the presence of low PTH and volume contraction, hypercalcemia is commonly associated with increased bicarbonate reabsorption. (18, 19) Metabolic alkalosis is augmented by the coexistent hypovolemia, hypokalemia, and hypochloremia observed in these patients. In contrast, the metabolic acidosis associated with hyperchloremia and normal SAG observed in 2 patients with hyperparathyroidism was probably a consequence of direct inhibition of proximal tubular bicarbonate reabsorption by PTH. (20) Accordingly, patients wi th hyperparathyroidism tend to have lower arterial pH, [Pco.sub.2], and bicarbonate concentration than cancer patients. However, there was a great deal of overlap between these parameters in the two groups because of interrelated abnormalities. (1, 21) In fact, in experimental animals receiving long-term PTH, the resultant acid-base abnormality is metabolic alkalosis rather than metabolic acidosis. (22) Mild hyperchloremic metabolic acidosis (renal tubular acidosis type IV) was evident along with hyperkalemia and hyponatremia in the patient with Addison's disease.

Hyponatremia was rare in hypercalcemic patients, and it was related to well-known causes, namely hypovolemia, thiazide administration, and Addison's disease. Hypernatremia was found in a few patients with severe tumor-associated hypercalcemia due to a hypercalcemia-induced impaired renal concentrating ability resulting in a dilute urine with water loss in excess of sodium and potassium and to increased insensible losses. (23-25) Hypokalemia, mainly due to inappropriate kaliuresis, was noted in some patients showing the most severe degree of hypercalcemia. This increased urine potassium loss could be due to calcium-induced tubular damage, which may impair sodium reabsorption, leading to increased flow of water and sodium to the collecting tubules and subsequent potassium wasting. (5) Additionally, the coexistent hypomagnesemia (noticed in 4 hypokalemic patients) could have contributed to the increased renal potassium excretion. (26) Finally, the coexistent alkalemia could also have influenced potassium homeos tasis by promoting potassium entry into cells. Even though the prevalence of hypokalemia in our cohort was lower than that previously described in hypercalcemic patients, the degree and frequency of hypokalemia were greater at the higher serum calcium levels, as previously reported. (5) The presence of hypokalemia must always be considered when treating severe hypercalcemia; otherwise, vigorous use of diuretics may result in profound hypokalemia and tachyarrhythmias. (5) Hyperkalemia due to deterioration of renal function or Addison's disease was infrequently found in our cohort.

Disturbances of magnesium balance have not been adequately described in hypercalcemic patients. Hypomagnesemia was observed in a few patients presenting with severe hypercalcemia. This was mainly due to renal magnesium wasting resulting from inhibition of magnesium reabsorption in the proximal tubule and in the thick ascending loop of Henle, since calcium ions compete with magnesium ions at the basolateral membrane of the loop of Henle. (27) In the present study, the majority of hypomagnesemic patients had neoplasia, while hyperparathyroidism was found in only two patients with concurrent hypomagnesemia. It has been reported that in primary hyperparathyroidism, serum magnesium levels are usually normal, since the magnesiuric effect of hypercalcemia is counterbalanced by the enhancing effect of parathyroid hormone on magnesium reabsorption. (28)

Hypophosphatemia due to inappropriate phosphaturia was evident in some hyperparathyroid patients. Cancer patients exhibited higher serum phosphorus levels than patients with primary hyperparathyroidism, but only a minority of hyperparathyroid patients presented with decreased serum phosphorus levels. This normalization of serum phosphorus levels could be partly due to the increase of serum creatinine observed in some hyperparathyroid patients, since patients with hyperparathyroidism and hypercalcemia-induced renal dysfunction tend to have significantly higher serum phosphorus levels than the remaining hyperparathyroid patients. (28) In fact, when patients with elevated serum creatinine (> 1.6 mg/dL) on admission were excluded, hyper-parathyroid patients tended to show lower serum phosphorus levels. Furthermore, serum phosphorus is also sensitive to the dietary phosphate intake. (29) In addition to a decreased dietary intake, hypophosphatemia in cancer patients is aggravated by a phosphorus drive into cells d ue to respiratory alkalosis and to increased cell burden observed, especially in leukemic patients. (29-31) Therefore, it is suggested that the determination of serum phosphorus levels is of limited value in the differential diagnosis of hypercalcemia. (29-32)

Evidence indicates that the serum chloride/phosphorus ratio could help distinguish between patients with hyperparathyroidism-related hypercalcemia and those with hypercalcemia of nonparathyroid origin. (2,3) In accordance, a significantly higher ratio in hyperparathyroid vs cancer patients was evident in our study. (24,29,32) Nevertheless, there was a considerable overlap in the ratio values between the two groups due to the previously-mentioned overlap in serum phosphorus levels and to the already described complex influences affecting acid-base status and therefore serum chloride levels. (4,29,32)

Our findings did not substantiate that hypercalcemia may be associated with a lowering of SAG, as has been previously suggested. (33) In fact, SAG was slightly, yet insignificantly lower in hypercalcemic patients than in controls. However, SAG was lower in hyperparathyroid vs cancer patients. This observation was in agreement with previously published data. (4) The difference in SAG values between the two groups was evident despite the higher average calcium and lower albumin levels in patients with malignancy-associated hypercalcemia. Additionally, in hyperparathyroid patients, the average serum calcium level was 3.1 mg/dL (1.5 mEq/L) higher than that of the controls. Thus, their average elevation of serum calcium above normal was considerably less than their average reduction of SAG below normal (3.5 mEq/L). The reduction of SAG values in these patients, however, could not be totally attributed to the increase in serum calcium concentration. The etiology for disproportion between the low SAG values and the elevation in serum calcium levels in hyperparathyroid patients, as well as the difference in SAG values between the two groups of hypercalcemic patients, is multifactorial. Cancer patients exhibited higher serum creatinine, and this relatively mild degree of renal insufficiency would be expected to raise SAG slightly. (34,35) However, the difference in SAG values between the two groups remained even when the patients with increased serum creatinine on admission were excluded. In addition, lower serum phosphorus concentration observed in patients with hyperparathyroidism may interfere with the observed decrease in SAG levels. Finally, acid-base balance disturbances could have contributed to the observed discrepancies, since metabolic acidosis (seen in some hyperparathyroid patients) tend to decrease SAG values, while metabolic alkalosis (seen in some cancer patients) have the opposite effect. (7) Even though the diagnostic usefulness of SAG in an individual patient is limited, it has been reported that a SAG v alue [less than or equal to]5 mEq/L in a hypercalcemic patient may be a helpful clue against malignancy. This is substantiated in our study, since cancer patients had SAG values >5 mEq/L. The lack of a significant correlation between SAG and serum calcium levels, together with the multifactorial influence on SAG values (as described), limit the usefulness of SAG as a diagnostic tool in the differential diagnosis of hypercalcemia.

In conclusion, primary hyperparathyroidism and neoplasia appear to be the most common causes of hypercalcemia. A wide array of concurrent acid-base and electrolyte disorders may be evident in hypercalcemic patients. Differences in simple laboratory tests, such as serum and urine electrolytes, as well as arterial blood gas determination may provide valuable clues as to further diagnostic evaluation.

TABLE 1

Causes of Hypercalcemia in the Study Population

                                 No. of
            Cause               Patients

Neoplasia                          47
Primary hyperparathyroidism        21
Thyrotoxicosis                      1
Overdosage of vitamin D             1
Lithium-induced hypercalcemia       1
Sarcoidosis                         2
Thiazide-induced hypercalcemia      2
Addison's disease                   1
                                  ----
  Total                            76

TABLE 2

Diagnosis of Malignancy-Related Hypercalcemia in the Study Population

    Primary        No. of
      Site        Patients

Lung                 11
Multiple myeloma     14
Lymphoma              4
Breast                6
Kidney                3
Colon                 2
Sarcoma               1
Pancreas              1
Undefined             5
                    ----
  Total              47

TABLE 3

Concurrent Acid-Base and Electrolyte Abnormalities in Patients With
Hypercalcemia

                       Hypercalcemic  Hyperparathyroid   Cancer
                         Patients         Patients      Patients
Abnormality              (n = 76)         (n = 21)      (n = 47)

Respiratory alkalosis        6               -             4
Metabolic acidosis           4               2             1
Metabolic alkalosis+        11               -             9
Hyponatremia                 3               -             1
Hypernatremia                4               -             4
Hypokalemia                  8               2             5
Hyperkalemia                 3               -             2
Hypomagnesemia               8               2             6
Hypophosphatemia             9               5             3


                          P
Abnormality            Value *

Respiratory alkalosis    NS
Metabolic acidosis       NS
Metabolic alkalosis+     .05
Hyponatremia             NS
Hypernatremia            NS
Hypokalemia              NS
Hyperkalemia             NS
Hypomagnesemia           NS
Hypophosphatemia         NS

* Difference between hyperparathyroid and cancer patients.

+ Primary or in combination with respiratory alkalosis.

TABLE 4

Laboratory Parameters in Patients and Control Subjects

                                  Hypercalcemic Patients
     Particular                          (n = 76)

Serum urea (mg/dL)                    67 [+ or -] 46
Serum creatinine (mg/dL)             1.8 [+ or -] 1.5
Serum urea/creatinine ratio           37 [+ or -] 22
Serum albumin (g/dL)                 3.9 [+ or -] 0.5
Serum sodium (mmol/L)                142 [+ or -] 3
Serum potassium (mmol/L)               4 [+ or -] 0.7
Serum chloride (mmol/L                99 [+ or -] 5
Serum calcium (mg/dL)               13.5 [+ or -] 2.2
Serum phosphorus (mg/dL)            3.35 [+ or -] 1
Serum magnesium (mmol/L)            0.76 [+ or -] 0.1
Arterial pH                         7.38 [+ or -] 0.04
Arterial [Pco.sub.2] (mm Hg)          37 [+ or -] 6
Serum bicarbonate (mmol/L)            26 [+ or -] 7
Serum anion gap (mmol/L)             9.0 [+ or -] 4
Serum chloride/phosphorus             31 [+ or -] 10
FE * Na + (%)                        1.8 [+ or -] 0.6
FE K + (%)                            23 [+ or -] 26
FE Cl- (%)                           3.2 [+ or -] 0.9
FE CA ++ (%) +                       4.2 [+ or -] 4.6
Urine calcium/creatinine (mg/mg)    0.27 [+ or -] 0.15
FE [PO.sub.4], 3-                     29 [+ or -] 23
[TmPO.sub.4], 3-/GFR (mg/dL)        2.56 [+ or -] 0.32
FE Mg ++ (%)                          10 [+ or -] 10

                                   Control Subjects     P
     Particular                        (n = 91)       Value

Serum urea (mg/dL)                  25 [+ or -] 11    < .005
Serum creatinine (mg/dL)           0.8 [+ or -] 0.3   < .005
Serum urea/creatinine ratio         29 [+ or -] 6     < .01
Serum albumin (g/dL)               4.2 [+ or -] 0.6   < .05
Serum sodium (mmol/L)              139 [+ or -] 4        NS
Serum potassium (mmol/L)           4.4 [+ or -] 0.5   < .05
Serum chloride (mmol/L)            104 [+ or -] 3     < .05
Serum calcium (mg/dL)              9.1 [+ or -] 0.6   < .001
Serum phosphorus (mg/DL)           3.9 [+ or -] 0.9   < .05
Serum magnesium (mmol/L)           0.9 [+ or -] 0.1   < .05
Arterial pH                       7.39 [+ or -] 0.05     NS
Arterial [Pco.sub.2] (mm Hg)        39 [+ or -] 4        NS
Serum bicarbonate (mmol/L)          25 [+ or -] 3        NS
Serum anion gap (mmol/L)          10.0 [+ or -] 3        NS
Serum chloride/phosphorus           26 [+ or -] 9     < .05
FE * Na + (%)                      1.6 [+ or -] 0.5      NS
FE K + (%)                          16 [+ or -] 4     < .05
FE Cl- (%)                         2.7 [+ or -] 1.1      NS
FE CA ++ (%) +                     2.1 [+ or -] 0.4   < .01
Urine calcium/creatinine (mg/mg)  0.16 [+ or -] 0.02  < .01
FE [PO.sub.4], 3-                   18 [+ or -] 4     < .01
[TmPO.sub.4], 3-/GFR (mg/dL)      2.88 [+ or -] 0.12  < .05
FE Mg ++ (%)                       5.2 [+ or -] 3     < .05

A 2-hour urine sample was used for the determination of fractional
excretions.

* Fractional excretion.

+ Total serum calcium was used for the determination.

TABLE 5

Laboratory Parameters in Patients With Primary Hyperparathyroidism and
Neoplasia

                                 Patients With       Patients With
                              Hyperparathyroidism      Neoplasia
Parameter                          (n = 21)             (n = 47)

Serum urea (mg/dL)              60 [+ or -] 51       72 [+ or -] 45
Serum creatinine (mg/dL)       1.4 [+ or -] 1.1     1.9 [+ or -] 1.6
Serum urea/creatinine ratio     33 [+ or -] 8        43 [+ or -] 9
Serum albumin (g/dL)           4.1 [+ or -] 0.4     3.8 [+ or -] 0.5
Serum sodium (mmol/L)          137 [+ or -] 3       138 [+ or -] 2
Serum potassium (mmol/L)       4.2 [+ or -] 0.7     4.0 [+ or -] 0.6
Serum chloride (mmol/L)        107 [+ or -] 4       102 [+ or -] 5
Serum calcium (mg/dL)           12 [+ or -] 1.5    13.8 [+ or -] 2.4
Serum phosphorus (mg/dL)       3.1 [+ or -] 1       3.5 [+ or -] 1
Serum magnesium (mmol/L)      0.82 [+ or -] 0.08   0.75 [+ or -] 0.09
Serum PTH (pg/mg)              124 [+ or -] 31       10 [+ or -] 6
Arterial pH                   7.36 [+ or -] 0.04   7.42 [+ or -] 0.03
Arterial [Pco.sub.2] (mm Hg)    35 [+ or -] 6      39.5 [+ or -] 5
Serum bicarbonate (mmol/L)      22 [+ or -] 6        27 [+ or -] 6
Serum anion gap (mmol/L)         7 [+ or -] 5       9.5 [+ or -] 3
Serum chloride/phosphorus       35 [+ or -] 9        29 [+ or -] 7
FE * [Na.sup.+] (%)            1.7 [+ or -] 0.5     1.7 [+ or -] 0.7
FE [K.sup.+] (%)                23 [+ or -] 16       24 [+ or -] 17
FE [Cl.sup.-](%)               3.1 [+ or -] 0.9     3.3 [+ or -] 1
FE [Ca.sup.++] (%) +           4.3 [+ or -] 4.2     4.2 [+ or -] 4.8
Urine calcium/creatinine      0.29 [+ or -] 0.12   0.20 [+ or -] 0.09
 (mg/ng)
FE [PO.sub.4,.sup.3-] (%)       31 [+ or -] 22       26 [+ or -] 18
Tm[PO.sub.4,.sup.3-]/GFR      2.51 [+ or -] 0.29   2.54 [+ or -] 0.28
 (mg/dL)
FE [Mg.sup.++] (%)              10 [+ or -] 11       10 [+ or -] 9


                                P
Parameter                     Value

Serum urea (mg/dL)            <.05
Serum creatinine (mg/dL)      <.05
Serum urea/creatinine ratio   <.05
Serum albumin (g/dL)          <.05
Serum sodium (mmol/L)           NS
Serum potassium (mmol/L)        NS
Serum chloride (mmol/L)       <.05
Serum calcium (mg/dL)         <.05
Serum phosphorus (mg/dL)      <.05
Serum magnesium (mmol/L)      <.05
Serum PTH (pg/mg)             <.001
Arterial pH                   <.05
Arterial [Pco.sub.2] (mm Hg)  <.05
Serum bicarbonate (mmol/L)    <.05
Serum anion gap (mmol/L)      <.05
Serum chloride/phosphorus     <.05
FE * [Na.sup.+] (%)             NS
FE [K.sup.+] (%)                NS
FE [Cl.sup.-](%)                NS
FE [Ca.sup.++] (%) +            NS
Urine calcium/creatinine      <.05
 (mg/ng)
FE [PO.sub.4,.sup.3-] (%)     <.05
Tm[PO.sub.4,.sup.3-]/GFR        NS
 (mg/dL)
FE [Mg.sup.++] (%)              NS

A 2-hour urine sample was used for the determination of fractional
excretions.

* Fractional excretion.

+ Total serum calcium was used for the determination.

References

(1.) Perez GO, Oster JR: Acid-base pathophysiology in endocrine diseases. Miner Electrolyte Metab 1985; 11:192-204

(2.) Palmer FJ, Nelson JC, Bacchus H: The chloride-phosphate ratio in hypercalcemia. Ann Intern Med 1974; 80:200-204

(3.) wills MR: Value of plasma chloride concentration and acid-base status in the differential diagnosis of hyperparathyroidism from other causes of hypercalcemia J Clin Pathol 1971; 24:219-227

(4.) Oster JR, Gutierrez R, Schlessinger FB, et al: Effect of hypercalcemia on the anion gap. Nephron 1990; 55:164-169

(5.) Aldinger KA, Samaan NA: Hypokalemia with hypercalcemia. prevalence and significance in treatment. Ann Intern Med 1977; 87:571-573

(6.) Mitnick P, Greenberg A, Goffman T, et al: Effects of two models of hypercalcemia on renal acid base metabolism. Kidney Int 1982; 21:613-620

(7.) Emmett M, Narins RG: Clinical use of the anion gap. Medicine 1977; 56:38-54

(8.) Walton RJ, Bijvoet OLM: Nomogram for derivation of renal threshold phosphate concentration. Lancet 1975; 2:309-310

(9.) Javaheri S, Shore NS, Rose BD, et al: Compensatory hypo-ventilation in metabolic alkalosis. Chest 1982; 81:296-301

(10.) Elisaf M, Siamopoulos KG: Fractional excretion of potassium in normal subjects and in hypokalemic patients. Postgrad Med J 1995; 71:211-212

(11.) Ethier JH, Kamel KS, Magner PO, et al: The transtubular potassium concentration in patients with hypokalemia and hyperkalemia. Am J Kidney Dis 1990; 15:309-315

(12.) Bettinelli A, Bianchetti MG, Borella P, et al: Genetic heterogeneity in tubular hypomagnesemia-hypokalemia with hypocalcuria (Gitelman's syndrome). Kidney Int 1995; 47:547-551

(13.) Narins RG, Jones ER, Stom MC, et al: Diagnostic strategies in disorders of fluid, electrolyte and acid-base homeostasis. Am J Med 1982; 72:496-520

(14.) Fisken RA, Heath DA, Somers S, et al: Hypercalcemia in hospital patients. clinical and diagnostic aspects. Lancet 1981; 1:202

(15.) Fisken RA, Heath DA, Bold AM: Hypercalcemia--a hospital survey. Q J Med 1980; 196:405-418

(16.) Bourke E, Delaney V: Assessment of hypocalcemia and hypercalcemia. Clin Lab Med 1993; 13:157-181

(17.) Nussbaum SR: Pathophysiology and management of severe hypercalcemia. Endocrinol Metab Clin North Am 1993; 22:343-362

(18.) Heinemann HO: Metabolic alkalosis in patients with hypercalcemia. Metabolism 1965; 14:1137-1152

(19.) Hulter HN, Sebastian A, Toto RD, et al: Renal and systemic acid-base effects of the chronic administration of hypercalcemia-producing agents: calcitriol, PTH, and intravenous calcium. Kidney Int 1982; 21:445-458

(20.) Coe FL: Magnitude of metabolic acidosis in primary hyperparathyroidism. Arch Intern Med 1974; 134:262-265

(21.) Hulter HN: Effects and interrelationships of PTH, [Ca.sup.++], vitamin D, and P in acid-base homeostasis. Am J Physiol 1985; 248:F739-F752

(22.) Hulter HN, Peterson JC: Acid-base homeostasis during chronic PTH excess in humans. Kidney Int 1985; 28:187-192

(23.) Zeffren JL, Heinemann HO: Reversible defect in renal concentrating mechanism in patients with hypercalcemia. Am J Med 1982; 33:54-63

(24.) Rosen S, Greenfeld Z, Bernheim J, et al: Hypercalcemic nephropathy: chronic disease with predominant medullary inner stripe injury. Kidney Int 1990; 37:1067-1075

(25.) Galla JH, Booker BB, Luke RG: Role of the loop segment in the urinary concentrating defect of hypercalcemia. Kidney Int 1986; 29:977-982

(26.) Solomon R: The relationship between disorders of [K.sup.+] and [Mg.sup.++] homeostasis. Semin Nephrol 1987; 2:253-262

(27.) Al-Ghamdi SMG, Cameron EC, Sutton RAL: Magnesium deficency: pathophysiology and clinical overview. Am J Kidney Dis 1994; 24:737-752

(28.) Sutton RAL: Plasma magnesium concentration in primary hyperparathyroidism. BMJ 1970; 11:529-533

(29.) Wong ET, Freier EF: The differential diagnosis of hypercalcemia. an algorithm for more effective use of laboratory tests. JAMA 1982; 247:75-80

(30.) Mallette L: The hypercalcemias. Semin Nephrol 1992; 12:159- 160

(31.) Stewart AF, Horst R, Feftos LJ, et al: Biochemical evaluation of patients with cancer associated hypercalcemia: evidence for humoral and non-humoral groups. N Engl J Med 1980; 303:1377-1383

(32.) Boyd JC, Ladenson JH: Value of laboratory tests in the differential diagnosis of hypercalcemia. Am J Med 1984; 77:863-872

(33.) Gabow PA: Disorders associated with an altered anion gap. Kidney Int 1985; 27:472-483

(34.) Widner B, Gerhardt RE, Harrington JT: Serum electrolyte and acid-base composition: the influence of graded degrees of chronic renal failure. Arch Intern Med 1979; 139:238-247

(35.) Hakim RM, Lazarus JM: Biochemical parameters in chronic renal failure. Am J Kidney Dis 1988; 11:238-247

RELATED ARTICLE: KEY POINTS

* Acid-base and electrolyte parameters were evaluated in 76 patients with hypercalcemia (serum calcium > 10.5 mg/dL) before any therapeutic intervention.

* Hyperparathyroidism and neoplasias (breast, bronchus, and hematologic) were the most common causes of hypercalcemia.

* Hypercalcemic patients had higher serum creatinine levels, but lower serum albumin, potassium, chloride, phosphorus, and magnesium levels than those in healthy control subjects.

* Significant differences in the laboratory parameters were also evident between the two main patient groups.

From the Department of Internal Medicine, University of Ioannina Medical School, Ioannina, Greece.

Reprint requests to Moses S. Elisaf, MD, University of Ioannina Medical School, Department of Internal Medicine, GR 45110, Ioannina, Greece.