Quarterly Journal of Medicine, New Series 77, No. 284, pp. 1277-1285, December 1990
Incidence, Causes and Mechanism of Hypercalcaemia in a Hospital Population in Hong Kong C. C. SHEK, A. NATKUNAM, V. TSANG*, C. S. COCKRAMf and R. SWAMINATHAN From the Departments of Chemical Pathology, * Clinical Oncology and f Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, N.T., Hong Kong Accepted 23 May 1990
To determine the incidence and causes of hypercalcaemia in a hospital population in Hong Kong, all 29107 samples received in the laboratory in one year were analysed for plasma calcium and albumin, and samples with a plasma calcium concentration adjusted for albumin greater than 2.55 mmol/l were investigated. Plasma calcium > 2.55 mmol/l was found in 462 patients. Repeat samples were received from 302 of these and hypercalcaemia was confirmed in 183. The main causes of hypercalcaemia were malignancy (72.1 per cent), tuberculosis (6.0 per cent), and primary hyperparathyroidism (5.5 per cent). In the malignant hypercalcaemia group, carcinoma of lung was the most common (31.8 per cent) and carcinoma of breast was uncommon (3.0 per cent). Secondary deposits in bone were detected in 35 of the 122 solid tumours. In order to identify the mechanism of hypercalcaemia the contributions of renal tubular reabsorption and increased bone resorption to the plasma calcium concentration were calculated. Increased tubular reabsorption was the main contributor to hypercalcaemia in primary hyperparathyroidism and carcinoma of liver (none of whom had bony metastases) and it contributed significantly to hypercalcaemia in carcinoma of lung without bony metastases and carcinoma of oesophagus. We conclude that in Hong Kong (a) primary hyperparathyroidism is uncommon, (b) tuberculosis is an important cause and (c) humoral factors may be responsible for a relatively high proportion of cases of malignant hypercalcaemia. INTRODUCTION Several studies have shown that clinically unsuspected hypercalcaemia may often be detected [1-10]; however the prevalence has been reported to vary between 0.1 and 0.7 per cent [2, 11-13] in out-patients or the general population and between 3.0 and 3.6 per cent in hospital populations [4, 6]. Hyperparathyroidism is the commonest cause among patients outside of hospital, while malignancy is the most common cause in hospital patients, Address correspondence to Professor R. Swaminathan, Department of Chemical Pathology, Prince of Wales Hospital, Shatin, NT. Hong Kong. ©Oxford University Press 1990
Downloaded from by guest on December 8, 2015
SUMMARY
1278
C. C. Shek and others
hyperparathyroidism being second [5, 6]. These studies have been carried out in Caucasian populations and very few data are available from Asia. Our impression is that hyperparathyroidism is relatively uncommon among the Chinese population in Hong Kong. Therefore we have conducted a survey of the incidence and causes of hypercalcaemia among a hospital population of Hong Kong Chinese.
MATERIALS AND METHODS The study was carried out at the Prince of Wales Hospital, Shatin, Hong Kong between 1st March 1986 and 28th February 1987. This is a 1400 bed, regional hospital which covers all major specialties. The average admission rate was 146 patients per day during the study period and of these 46 per cent were admitted via the Accident and Emergency Department. All samples received in the laboratory for 'routine' investigations (electrolytes, renal function tests or liver function tests) from in-patients were analysed for plasma calcium and albumin concentrations. The plasma calcium concentration was corrected for albumin according to the formula [14]: corrected plasma calcium concentration (mmol/1) = measured plasma calcium concentration (mmol/l) + 0.025 [40 —albumin concentration (g/1)] Downloaded from by guest on December 8, 2015
After excluding renal, paediatric and obstetric patients, all samples with a corrected plasma calcium concentration above 2.55 mmol/1 were reanalysed and if the repeat result was also > 2.55 mmol/1, fasting blood and urine samples were requested for measurement of plasma concentrations of calcium, alkaline phosphatase, phosphate, parathyroid hormone (PTH), renal phosphate threshold concentration (TmP/GFR) and urine calcium, sodium and creatinine concentrations. On confirmation of hypercalcaemia (Ca2+ > 2.55 mmol/1 on second sample) one of the authors visited the ward and obtained more information about the patient and the underlying diagnosis. After discharge/death, the case notes were re-examined and clinical information rechecked. In patients with malignancy, bony metastasis was considered to be present when bony deposits were detected on bone scanning or on radiological examination. These investigations were done only when necessary for clinical management. Primary hyperparathyroidism was diagnosed on the basis of hypercalcaemia with elevated PTH and no other underlying cause. Plasma concentrations of calcium, albumin, phosphate, alkaline phosphatase activity and urine concentrations of phosphate, creatinine, calcium and sodium were measured on an American Monitor Parallel Analyser (American Monitor Co, Indiancope, USA). Plasma PTH was measured by a commercial assay ('Incstar', Immuno-Nuclear Corporation, USA). Plasma wasfirstextracted by an antibody linked to sepharose beads and this antibody binds to the intact molecule and N-terminal fragments of PTH. After washing the beads with saline, the intact and N-terminal fragments were eluted with hydrochloric acid and the intact PTH in the eluate was assayed by radioimmunoassay using an antibody directed at the midmolecule. The between batch coefficient of variation was 9 per cent at a concentration of 21 pmol/1. The detection limit of the assay is 2 pmol/1. The upper limit for PTH in healthy subjects was found to be 6 pmol/1 (n = 20). TmP/GFR was estimated by the method and nomogram of Walton and Bijvoet [15]. The range for TmP/GFR found in 25 healthy Chinese subjects was 0.80-1.35 mmol/1. Urine calcium excretion rate was calculated in relation to glomerular filtration rate [16] and this was corrected for sodium excretion based on the relationship between Ca/Cr and
Hypercalcaemia in a Hospital Population
1279
Na/Cr ratios [17]. Tubular maximum for calcium reabsorption was also calculated and corrected for sodium excretion as described by Need et al. [18]. The contributions of renal tubular reabsorption, bone resorption and renal function to the hypercalcaemia were assessed using the method described by Nordin [19] and Percival et al. [20] based on the relationship between fasting serum calcium concentration and calcium excretion [16]. RESULTS
TABLE 1. Malignancy associated with hypercalcaemia Primary site
Number (%)*
Number (%)t of cases with bony metastasis
Bronchus (lung) Liver Oesophagus Haematological Nasopharynx Colorectal Bladder Breast Biliary tract Cervix Stomach Others Unknown
42(31.8) 13(10.2) 13(10.2) 10(7.6) 9 (6.8) 5 (3.8) 5 (3.8) 4 (3.0) 4 (3.0) 3 (2.3) 3 (2.3) 13(9.8) 8(6.1)
12(29) 0(0) 2(15) 4(44) 1 (20) 1(20) 4(100) 0(0) 1(3) 0(0) 5(38) 5(63)
* Calculated as percentages of all malignancies, t Calculated as percentages for that primary site.
Downloaded from by guest on December 8, 2015
During this study period, there were 30 878 admissions, excluding paediatric, obstetric and renal cases, and 29 107 blood samples (approximately 97 per cent of these admissions) were received. Samples from 462 patients had a corrected plasma calcium concentration greater than 2.55 mmol/1 (1.58 percent). A second (fasting) sample was received from 302 of these patients and hypercalcaemia was confirmed in 183 (60.6 per cent), a figure similar to that reported in the literature [6, 9, 17, 20]. In the remaining 119 patients, the plasma calcium concentration on the first sample was 2.63 (±0.01) mmol/1 and was below the cut off value (2.55 mmol/1) on the second (fasting) sample. The estimated incidence of hypercalcaemia in this population was approximately 0.6 per cent. The majority of the patients (76.5 per cent) had values below 3.00 mmol/1. There were six (3.3 per cent) cases with a plasma calcium higher than 4.0 mmol/1. Hypercalcaemia was not suspected clinically in 118 (64.5 per cent) of these patients. The most common cause of hypercalcaemia was malignancy (132 patients, 72.1 per cent) followed by tuberculosis (11 patients, 6.0 per cent). Primary hyperparathyroidism was found in 10 patients (5.5 per cent). Immobilization and thyrotoxicosis were identified as the causes in three and two patients, respectively. In 25 patients (13.7 per cent) the cause of the hypercalcaemia could not be identified. The most common site of primary tumour in the 132 patients with malignancy was the bronchus, followed by liver, oesophagus and nasopharynx (Table 1). Carcinoma of the breast accounted for only 3.0 per cent of cases, and 10 (7.6 per cent) patients had haematological malignancies (including lymphoma and myeloma). Bony secondaries were
C. C. Shek and others
1280 s.o-
4.5-
Ion
•
4.0-
c i
J
3.5-
j
clu
E
m O
3.0-
2.5^
4
:
•iHl • /
;
t
i
i. —
Incidence, Causes and Mechanism of Hypercalcaemia in a Hospital Population in Hong Kong C. C. SHEK, A. NATKUNAM, V. TSANG*, C. S. COCKRAMf and R. SWAMINATHAN From the Departments of Chemical Pathology, * Clinical Oncology and f Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, N.T., Hong Kong Accepted 23 May 1990
To determine the incidence and causes of hypercalcaemia in a hospital population in Hong Kong, all 29107 samples received in the laboratory in one year were analysed for plasma calcium and albumin, and samples with a plasma calcium concentration adjusted for albumin greater than 2.55 mmol/l were investigated. Plasma calcium > 2.55 mmol/l was found in 462 patients. Repeat samples were received from 302 of these and hypercalcaemia was confirmed in 183. The main causes of hypercalcaemia were malignancy (72.1 per cent), tuberculosis (6.0 per cent), and primary hyperparathyroidism (5.5 per cent). In the malignant hypercalcaemia group, carcinoma of lung was the most common (31.8 per cent) and carcinoma of breast was uncommon (3.0 per cent). Secondary deposits in bone were detected in 35 of the 122 solid tumours. In order to identify the mechanism of hypercalcaemia the contributions of renal tubular reabsorption and increased bone resorption to the plasma calcium concentration were calculated. Increased tubular reabsorption was the main contributor to hypercalcaemia in primary hyperparathyroidism and carcinoma of liver (none of whom had bony metastases) and it contributed significantly to hypercalcaemia in carcinoma of lung without bony metastases and carcinoma of oesophagus. We conclude that in Hong Kong (a) primary hyperparathyroidism is uncommon, (b) tuberculosis is an important cause and (c) humoral factors may be responsible for a relatively high proportion of cases of malignant hypercalcaemia. INTRODUCTION Several studies have shown that clinically unsuspected hypercalcaemia may often be detected [1-10]; however the prevalence has been reported to vary between 0.1 and 0.7 per cent [2, 11-13] in out-patients or the general population and between 3.0 and 3.6 per cent in hospital populations [4, 6]. Hyperparathyroidism is the commonest cause among patients outside of hospital, while malignancy is the most common cause in hospital patients, Address correspondence to Professor R. Swaminathan, Department of Chemical Pathology, Prince of Wales Hospital, Shatin, NT. Hong Kong. ©Oxford University Press 1990
Downloaded from by guest on December 8, 2015
SUMMARY
1278
C. C. Shek and others
hyperparathyroidism being second [5, 6]. These studies have been carried out in Caucasian populations and very few data are available from Asia. Our impression is that hyperparathyroidism is relatively uncommon among the Chinese population in Hong Kong. Therefore we have conducted a survey of the incidence and causes of hypercalcaemia among a hospital population of Hong Kong Chinese.
MATERIALS AND METHODS The study was carried out at the Prince of Wales Hospital, Shatin, Hong Kong between 1st March 1986 and 28th February 1987. This is a 1400 bed, regional hospital which covers all major specialties. The average admission rate was 146 patients per day during the study period and of these 46 per cent were admitted via the Accident and Emergency Department. All samples received in the laboratory for 'routine' investigations (electrolytes, renal function tests or liver function tests) from in-patients were analysed for plasma calcium and albumin concentrations. The plasma calcium concentration was corrected for albumin according to the formula [14]: corrected plasma calcium concentration (mmol/1) = measured plasma calcium concentration (mmol/l) + 0.025 [40 —albumin concentration (g/1)] Downloaded from by guest on December 8, 2015
After excluding renal, paediatric and obstetric patients, all samples with a corrected plasma calcium concentration above 2.55 mmol/1 were reanalysed and if the repeat result was also > 2.55 mmol/1, fasting blood and urine samples were requested for measurement of plasma concentrations of calcium, alkaline phosphatase, phosphate, parathyroid hormone (PTH), renal phosphate threshold concentration (TmP/GFR) and urine calcium, sodium and creatinine concentrations. On confirmation of hypercalcaemia (Ca2+ > 2.55 mmol/1 on second sample) one of the authors visited the ward and obtained more information about the patient and the underlying diagnosis. After discharge/death, the case notes were re-examined and clinical information rechecked. In patients with malignancy, bony metastasis was considered to be present when bony deposits were detected on bone scanning or on radiological examination. These investigations were done only when necessary for clinical management. Primary hyperparathyroidism was diagnosed on the basis of hypercalcaemia with elevated PTH and no other underlying cause. Plasma concentrations of calcium, albumin, phosphate, alkaline phosphatase activity and urine concentrations of phosphate, creatinine, calcium and sodium were measured on an American Monitor Parallel Analyser (American Monitor Co, Indiancope, USA). Plasma PTH was measured by a commercial assay ('Incstar', Immuno-Nuclear Corporation, USA). Plasma wasfirstextracted by an antibody linked to sepharose beads and this antibody binds to the intact molecule and N-terminal fragments of PTH. After washing the beads with saline, the intact and N-terminal fragments were eluted with hydrochloric acid and the intact PTH in the eluate was assayed by radioimmunoassay using an antibody directed at the midmolecule. The between batch coefficient of variation was 9 per cent at a concentration of 21 pmol/1. The detection limit of the assay is 2 pmol/1. The upper limit for PTH in healthy subjects was found to be 6 pmol/1 (n = 20). TmP/GFR was estimated by the method and nomogram of Walton and Bijvoet [15]. The range for TmP/GFR found in 25 healthy Chinese subjects was 0.80-1.35 mmol/1. Urine calcium excretion rate was calculated in relation to glomerular filtration rate [16] and this was corrected for sodium excretion based on the relationship between Ca/Cr and
Hypercalcaemia in a Hospital Population
1279
Na/Cr ratios [17]. Tubular maximum for calcium reabsorption was also calculated and corrected for sodium excretion as described by Need et al. [18]. The contributions of renal tubular reabsorption, bone resorption and renal function to the hypercalcaemia were assessed using the method described by Nordin [19] and Percival et al. [20] based on the relationship between fasting serum calcium concentration and calcium excretion [16]. RESULTS
TABLE 1. Malignancy associated with hypercalcaemia Primary site
Number (%)*
Number (%)t of cases with bony metastasis
Bronchus (lung) Liver Oesophagus Haematological Nasopharynx Colorectal Bladder Breast Biliary tract Cervix Stomach Others Unknown
42(31.8) 13(10.2) 13(10.2) 10(7.6) 9 (6.8) 5 (3.8) 5 (3.8) 4 (3.0) 4 (3.0) 3 (2.3) 3 (2.3) 13(9.8) 8(6.1)
12(29) 0(0) 2(15) 4(44) 1 (20) 1(20) 4(100) 0(0) 1(3) 0(0) 5(38) 5(63)
* Calculated as percentages of all malignancies, t Calculated as percentages for that primary site.
Downloaded from by guest on December 8, 2015
During this study period, there were 30 878 admissions, excluding paediatric, obstetric and renal cases, and 29 107 blood samples (approximately 97 per cent of these admissions) were received. Samples from 462 patients had a corrected plasma calcium concentration greater than 2.55 mmol/1 (1.58 percent). A second (fasting) sample was received from 302 of these patients and hypercalcaemia was confirmed in 183 (60.6 per cent), a figure similar to that reported in the literature [6, 9, 17, 20]. In the remaining 119 patients, the plasma calcium concentration on the first sample was 2.63 (±0.01) mmol/1 and was below the cut off value (2.55 mmol/1) on the second (fasting) sample. The estimated incidence of hypercalcaemia in this population was approximately 0.6 per cent. The majority of the patients (76.5 per cent) had values below 3.00 mmol/1. There were six (3.3 per cent) cases with a plasma calcium higher than 4.0 mmol/1. Hypercalcaemia was not suspected clinically in 118 (64.5 per cent) of these patients. The most common cause of hypercalcaemia was malignancy (132 patients, 72.1 per cent) followed by tuberculosis (11 patients, 6.0 per cent). Primary hyperparathyroidism was found in 10 patients (5.5 per cent). Immobilization and thyrotoxicosis were identified as the causes in three and two patients, respectively. In 25 patients (13.7 per cent) the cause of the hypercalcaemia could not be identified. The most common site of primary tumour in the 132 patients with malignancy was the bronchus, followed by liver, oesophagus and nasopharynx (Table 1). Carcinoma of the breast accounted for only 3.0 per cent of cases, and 10 (7.6 per cent) patients had haematological malignancies (including lymphoma and myeloma). Bony secondaries were
C. C. Shek and others
1280 s.o-
4.5-
Ion
•
4.0-
c i
J
3.5-
j
clu
E
m O
3.0-
2.5^
4
:
•iHl • /
;
t
i
i. —
