0021-972X/90/7101-0138$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 1 Printed in U.S.A.
Increased Monocyte Interleukin-1 Activity and Decreased Vertebral Bone Density in Patients with Fasting Idiopathic Hypercalciuria ROBERTO PACIFICI, MARCOS ROTHSTEIN, LEONARD RIFAS, KIN-HING W. LAU, DAVID J. BAYLINK, LOUIS V. AVIOLI, AND KEITH HRUSKA Divisions of Metabolism and Nephrology, The Jewish Hospital of St. Louis (M.R., K.H.), Washington University Medical Center, St. Louis, Missouri 63110; and the Department of Medicine, Pettis Veterans Administration Hospital and Loma Linda University (K-H. W.L., D.J.B.), Loma Linda, California 92357
ABSTRACT. Idiopathic hypercalciuria (IH) is a heterogeneous disorder frequently observed in patients with nephrolithiasis. At one extreme of its clinical spectrum is fasting hypercalciuria (FH), acondition characterized by increased bone resorption and turnover. In previous studies we have shown that monocytes from patients with high turnover osteoporosis and from women in early postmenopause elaborate increased amounts of interleukin-1 (IL-1), a cytokine that stimulates bone resorption in vitro and in uivo. Since IL-1 could also mediate the resorptive mechanism of FH and cause a clinically significant bone loss, we have studied the relationship of IH, vertebral mineral density, bone turnover, and monocyte IL-1 activity in 47 patients with absorptive hypercalciuria (AH), 23 with FH, and 38 nonhypercalciuric subjects with recurrent nephrolithiasis (controls). Vertebral mineral density, as measured by quantitative computer tomography, was decreased in each of the three patient groups, but was significantly lower in FH patients than in AH patients or control subjects. Twenty-four-hour total urinary hydroxyproline excretion was increased in FH patients com-
pared to that in AH patients or controls, but blood levels of osteocalcin were not. Monocytes from FH subjects yielded significantly more IL-1 (a + /?) activity than those from AH patients or controls; levels of IL-1 activity in monocytes of AH and control patients were similar. In IH subjects, significant correlations were found between IL-1 and hydroxyproline (r = 0.70; P < 0.0001), IL-1 and quantitative computer tomography values (r = —0.49; P < 0.005), and IL-1 and urinary calcium (r = -0.36; P < 0.05). Serum PTH levels were within normal limits in all subjects and were similar in the three study groups. 1,25Dihydroxyvitamin D3 levels, although higher in IH patients than in controls, were not significantly different in FH and AH subjects. Increased IL-1 activity and decreased vertebral mineral density are features of a subset of patients with IH. Although a cause-effect relationship remains to be established, increased monocytic IL-1 activity, rather than elevated PTH or 1,25dihydroxyvitamin D3 levels, could underlie the resorptive component of FH. (J Clin Endocrinol Metab 7 1 : 138-145,1990)
I
DIOPATHIC hypercalciuria (IH), a heterogeneous syndrome first described by Albright et al. (1), is the most common metabolic abnormality in patients with nephrolithiasis, occurring in approximately 50% of the patients with recurrent calcium stone formation (2). Calcium balance (3) and calcium absorption studies (46) have demonstrated that patients with IH have increased intestinal calcium absorption. The term absorptive hypercalciuria (AH) is frequently used to describe those patients with intestinal hyperabsorption of calcium who exhibit an abnormal response to an oral calcium load and excrete normal amounts of calcium after prolonged fasting (4). At the other extreme of the spectrum are patients who excrete increased amounts of calcium
after a calcium load as well as during the fasting state, pointing to the skeleton as an additional source of urinary calcium (7). Observations of increased bone turnover (8, 9) and a decrease in bone density of the appendicular skeleton in patients with fasting hypercalciuria (FH) (10-12) are consistent with the hypothesis that an increase in bone resorption contributes to some cases of hypercalciuria. Increased bone resorption in IH has been linked to both primary hyperparathyroidism (13) and a renal calcium leak, which, in turn, causes a tendency toward hypocalcemia and increased PTH secretion (14). Other studies have revealed, however, that FH patients generally exhibit normal or suppressed, rather than high, PTH values (15,16), suggesting that a PTH-independent mechanism accounts for the resorptive component of this form of hypercalciuria. Increased levels of serum 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] have been demonstrated in patients with
Received September 13, 1989. Address all correspondence and requests for reprints to: Roberto Pacifici, M.D., Division of Endocrinology and Metabolism, The Jewish Hospital of St. Louis, 216 South Kingshighway, St. Louis, Missouri 63110.
138
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 09:48 For personal use only. No other uses without permission. . All rights reserved.
IL-1 ACTIVITY IN IH IH (17, 18), a finding that could explain the increased intestinal calcium absorption commonly seen in these patients. However, whether 1,25-(OH)2D3 plays a role in stimulating bone resorption and contributes to the pathogenesis of FH remains to be determined because 1,25(OH)2D3 levels are generally similar in FH and AH patients (19). In previous studies we have reported that monocytes from patients with high turnover osteoporosis produce higher levels of interleukin-1 (IL-1) than those from subjects with low turnover osteoporosis or from healthy controls (20). In addition, we have also shown that in osteoporotic men, IL-1 correlates with urinary calcium excretion (21). These findings suggest that IL-1, a potent in vitro and in vivo stimulator of bone resorption (2225), might underlie the resorptive process of FH. A chronic enhanced secretion of IL-1 could, in fact, stimulate bone resorption, increase fasting urinary calcium, and induce a clinically significant bone loss. This hypothesis was tested in this study; we found that FH subjects do in fact exhibit increased monocyte IL-1 activity, higher urinary hydroxyproline excretion, and decreased vertebral bone density compared to nonhypercalciuric patients and patients with AH.
Subjects and Methods Study subjects The study population consisted of 112 Caucasian adults with a history of recurrent passage of renal stones recruited consecutively over a 24-month period from patients of the Kidney Stone Center of St. Louis. All subjects had a history of at least 1 stone passage in the 12 months before the study and passage of at least 2 stones in all. All patients evaluated in the Kidney Stone Center were regarded as candidates for the study; only those who were not willing to participate and did not satisfy the inclusion/exclusion criteria (see below) were excluded. All gave their informed consent and were ambulatory and in good health. Except for IH, none had diseases that could affect calcium excretion, bone remodeling, and monocyte function. Excluded from the investigation were all subjects with a history of treatment with estrogen, progesterone, corticosteroids, anticonvulsants, sodium fluoride, bisphosphonates, calcitonin, cytotoxic drugs, and vitamin D in the 12 months before enrollment in the study. Also excluded were those who required daily intake of an antacid or antiinflammatory drugs. No one included in the study had a history of treatment with calcium supplements. Patients receiving specific treatment for IH (such as thiazide diuretics) were included; however, treatment was withheld (see below) for at least 2 weeks before and during the study. Protocol On visit 1 a pharmacological history and a dietary survey were obtained by a registered dietician. The patients were then instructed to discontinue any specific treatment for IH, such
139
as thiazide diuretics. Those patients who were not observing any dietary restriction were asked to remain on their customary diet. Those subjects who were on a low calcium diet were asked to resume a customary unrestricted diet. After at least 2 weeks, a period of time that has been shown to be sufficient for a change in calcium absorption to be induced (26), two 24-h urine collections were obtained from all patients. A registered dietician then provided the patients with a set of menus and instructed them to follow a gelatin-free 10-mmol calcium, 170260 mmol sodium diet for 11 days. On the tenth day of the low calcium diet, a third 24-h urine collection and a blood sample were obtained. Each urine sample was assayed for calcium, creatinine, phosphate, citrate, oxalate, uric acid, sodium, potassium, and magnesium. The blood sample was assayed for calcium, phosphorus, creatinine, blood urea nitrogen, uric acid, PTH, 1,25-(OH)2D3, and electrolytes. On the last day of the low calcium diet, a standard calcium tolerance test was performed, as described by Pak et al. (5) in the clinical research center of our institution. A 2-h urine sample was collected between 0800-1000 h after a 12-h overnight fast. At 1000 h, patients ingested 1 g elemental calcium in the form of liquid calcium gluconate (NeoCalglucon) mixed with 250 mL distilled water along with two pieces of dry toast and 300 cc apple juice. Urine was then collected between 1000-1400 h. Blood samples were taken at 0800, 1000, and 1400 h. Patients were allowed to drink 300 cc distilled water during the 0800-1400 h period to maintain adequate urine flow. Each urine sample collected during the 0800-1400 h period was assayed for calcium and creatinine. The blood samples collected during the test were assayed for calcium, creatinine, and PTH. In addition, IL-1 activity in the culture medium of peripheral blood monocytes, serum level of osteocalcin (BGP), and 24-h urinary excretion of total hydroxyproline (OHP) were measured in the last 46 (15 FH, 18 AH, and 13 controls) of the 108 patients enrolled in the study in samples obtained on the tenth day of the low calcium diet. To establish a normal range for IL-1 activity, this parameter was also measured in 20 healthy men and 10 healthy premenopausal women, aged 21-65 yr (mean ± SD, 43 ± 9.5 yr), with no history of kidney stones and no hypercalciuria who were ingesting a customary diet. Classification criteria Thirty-eight of the 112 subjects, aged 20-75 years (mean ± SE, 48.2 ± 2.0 yr; median, 46 yr), were classified as nonhypercalciuric stone-formers (controls), because on their customary calcium diet they had an average urinary calcium excretion not exceeding 0.1 mmol/kg BW for either sex and/or 6.23 mmol/ day for females and 7.48 mmol/day for males, and on the low calcium diet they demonstrated a normal fasting urinary calcium/creatinine ratio and a normal response to the oral calcium load. Twelve of these subjects had no metabolic abnormalities. The remaining 26 had abnormal urinary levels of uric acid, oxalate, or citrate or a combination of these abnormalities. The remaining 74 patients had an increased average calcium excretion (as defined above) while on their customary, calciumunrestricted diet. Forty-seven of these patients, aged 23-76 yr (mean ± SE, 41.2 ± 11.8 yr; median, 39), were regarded as having AH, because of a fasting calcium/creatinine ratio of 0.31
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 09:48 For personal use only. No other uses without permission. . All rights reserved.
PACIFICI ET AL.
140
mmol Ca/mmol creatinine or less, an increased urinary calcium response to an oral calcium tolerance test (postload calcium excretion, >0.56 mmol/mmol creatinine) and normal serum PTH levels. Twenty-three hypercalciuric subjects, aged 23-72 yr (mean ± SE, 46.8 ± 2.1 yr; median, 48), were classified as having FH because of a fasting calcium/creatinine ratio greater than 0.31 mmol Ca/mmol creatinine in the 2-h fasting sample (27). In addition, these patients had an increased urinary response to an oral calcium load and normal serum PTH levels. The remaining 4 patients had normal blood calcium levels, FH, and increased circulating PTH levels. These subjects were regarded as having renal hypercalciuria and were excluded from the study. Laboratory methods Biochemical assays. Blood and urinary calcium, phosphate, magnesium, creatinine, sodium, potassium, chloride, CO2, and uric acid were measured with standard atomic absorptiometry and photometric methods. Uric acid was measured by a uricase method (28), blood oxalate by the oxalate oxidase method (29), and PTH by the method of Hruska et al. (30), using CH9 antiserum, which binds to the COOH-terminal and midmolecule epitope of the protein. This assay does not distinguish between normal and suppressed values, but has been shown to separate normal from increased PTH values in most patients with normal renal function (31). 1,25-(OH)2D3 and total urinary OHP were measured by a radioreceptor method (32) and the colorimetric method of Prockop and Udenfriend (33), respectively. BGP was measured by RIA, using an antiserum that binds a midmolecule epitope of human osteocalcin (34). The normal range (mean ± 2 SD) for BGP with this assy is 2.7-13.5 ng/mL. Monocyte cultures. To assay IL-1, blood was drawn into EDTAcontaining tubes, and monocyte cultures were prepared as previously described (20, 35). Briefly, freshly drawn blood was fractionated on Ficoll/Hypaque, and the mononuclear cells were removed from the interface and washed twice with RPMI1640 medium. The medium was assayed for endotoxin by the limulus amoebocyte lysate assay. Endotoxin was not detected at the level of sensitivity of the assay (
Vol. 71, No. 1 Printed in U.S.A.
Increased Monocyte Interleukin-1 Activity and Decreased Vertebral Bone Density in Patients with Fasting Idiopathic Hypercalciuria ROBERTO PACIFICI, MARCOS ROTHSTEIN, LEONARD RIFAS, KIN-HING W. LAU, DAVID J. BAYLINK, LOUIS V. AVIOLI, AND KEITH HRUSKA Divisions of Metabolism and Nephrology, The Jewish Hospital of St. Louis (M.R., K.H.), Washington University Medical Center, St. Louis, Missouri 63110; and the Department of Medicine, Pettis Veterans Administration Hospital and Loma Linda University (K-H. W.L., D.J.B.), Loma Linda, California 92357
ABSTRACT. Idiopathic hypercalciuria (IH) is a heterogeneous disorder frequently observed in patients with nephrolithiasis. At one extreme of its clinical spectrum is fasting hypercalciuria (FH), acondition characterized by increased bone resorption and turnover. In previous studies we have shown that monocytes from patients with high turnover osteoporosis and from women in early postmenopause elaborate increased amounts of interleukin-1 (IL-1), a cytokine that stimulates bone resorption in vitro and in uivo. Since IL-1 could also mediate the resorptive mechanism of FH and cause a clinically significant bone loss, we have studied the relationship of IH, vertebral mineral density, bone turnover, and monocyte IL-1 activity in 47 patients with absorptive hypercalciuria (AH), 23 with FH, and 38 nonhypercalciuric subjects with recurrent nephrolithiasis (controls). Vertebral mineral density, as measured by quantitative computer tomography, was decreased in each of the three patient groups, but was significantly lower in FH patients than in AH patients or control subjects. Twenty-four-hour total urinary hydroxyproline excretion was increased in FH patients com-
pared to that in AH patients or controls, but blood levels of osteocalcin were not. Monocytes from FH subjects yielded significantly more IL-1 (a + /?) activity than those from AH patients or controls; levels of IL-1 activity in monocytes of AH and control patients were similar. In IH subjects, significant correlations were found between IL-1 and hydroxyproline (r = 0.70; P < 0.0001), IL-1 and quantitative computer tomography values (r = —0.49; P < 0.005), and IL-1 and urinary calcium (r = -0.36; P < 0.05). Serum PTH levels were within normal limits in all subjects and were similar in the three study groups. 1,25Dihydroxyvitamin D3 levels, although higher in IH patients than in controls, were not significantly different in FH and AH subjects. Increased IL-1 activity and decreased vertebral mineral density are features of a subset of patients with IH. Although a cause-effect relationship remains to be established, increased monocytic IL-1 activity, rather than elevated PTH or 1,25dihydroxyvitamin D3 levels, could underlie the resorptive component of FH. (J Clin Endocrinol Metab 7 1 : 138-145,1990)
I
DIOPATHIC hypercalciuria (IH), a heterogeneous syndrome first described by Albright et al. (1), is the most common metabolic abnormality in patients with nephrolithiasis, occurring in approximately 50% of the patients with recurrent calcium stone formation (2). Calcium balance (3) and calcium absorption studies (46) have demonstrated that patients with IH have increased intestinal calcium absorption. The term absorptive hypercalciuria (AH) is frequently used to describe those patients with intestinal hyperabsorption of calcium who exhibit an abnormal response to an oral calcium load and excrete normal amounts of calcium after prolonged fasting (4). At the other extreme of the spectrum are patients who excrete increased amounts of calcium
after a calcium load as well as during the fasting state, pointing to the skeleton as an additional source of urinary calcium (7). Observations of increased bone turnover (8, 9) and a decrease in bone density of the appendicular skeleton in patients with fasting hypercalciuria (FH) (10-12) are consistent with the hypothesis that an increase in bone resorption contributes to some cases of hypercalciuria. Increased bone resorption in IH has been linked to both primary hyperparathyroidism (13) and a renal calcium leak, which, in turn, causes a tendency toward hypocalcemia and increased PTH secretion (14). Other studies have revealed, however, that FH patients generally exhibit normal or suppressed, rather than high, PTH values (15,16), suggesting that a PTH-independent mechanism accounts for the resorptive component of this form of hypercalciuria. Increased levels of serum 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] have been demonstrated in patients with
Received September 13, 1989. Address all correspondence and requests for reprints to: Roberto Pacifici, M.D., Division of Endocrinology and Metabolism, The Jewish Hospital of St. Louis, 216 South Kingshighway, St. Louis, Missouri 63110.
138
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 09:48 For personal use only. No other uses without permission. . All rights reserved.
IL-1 ACTIVITY IN IH IH (17, 18), a finding that could explain the increased intestinal calcium absorption commonly seen in these patients. However, whether 1,25-(OH)2D3 plays a role in stimulating bone resorption and contributes to the pathogenesis of FH remains to be determined because 1,25(OH)2D3 levels are generally similar in FH and AH patients (19). In previous studies we have reported that monocytes from patients with high turnover osteoporosis produce higher levels of interleukin-1 (IL-1) than those from subjects with low turnover osteoporosis or from healthy controls (20). In addition, we have also shown that in osteoporotic men, IL-1 correlates with urinary calcium excretion (21). These findings suggest that IL-1, a potent in vitro and in vivo stimulator of bone resorption (2225), might underlie the resorptive process of FH. A chronic enhanced secretion of IL-1 could, in fact, stimulate bone resorption, increase fasting urinary calcium, and induce a clinically significant bone loss. This hypothesis was tested in this study; we found that FH subjects do in fact exhibit increased monocyte IL-1 activity, higher urinary hydroxyproline excretion, and decreased vertebral bone density compared to nonhypercalciuric patients and patients with AH.
Subjects and Methods Study subjects The study population consisted of 112 Caucasian adults with a history of recurrent passage of renal stones recruited consecutively over a 24-month period from patients of the Kidney Stone Center of St. Louis. All subjects had a history of at least 1 stone passage in the 12 months before the study and passage of at least 2 stones in all. All patients evaluated in the Kidney Stone Center were regarded as candidates for the study; only those who were not willing to participate and did not satisfy the inclusion/exclusion criteria (see below) were excluded. All gave their informed consent and were ambulatory and in good health. Except for IH, none had diseases that could affect calcium excretion, bone remodeling, and monocyte function. Excluded from the investigation were all subjects with a history of treatment with estrogen, progesterone, corticosteroids, anticonvulsants, sodium fluoride, bisphosphonates, calcitonin, cytotoxic drugs, and vitamin D in the 12 months before enrollment in the study. Also excluded were those who required daily intake of an antacid or antiinflammatory drugs. No one included in the study had a history of treatment with calcium supplements. Patients receiving specific treatment for IH (such as thiazide diuretics) were included; however, treatment was withheld (see below) for at least 2 weeks before and during the study. Protocol On visit 1 a pharmacological history and a dietary survey were obtained by a registered dietician. The patients were then instructed to discontinue any specific treatment for IH, such
139
as thiazide diuretics. Those patients who were not observing any dietary restriction were asked to remain on their customary diet. Those subjects who were on a low calcium diet were asked to resume a customary unrestricted diet. After at least 2 weeks, a period of time that has been shown to be sufficient for a change in calcium absorption to be induced (26), two 24-h urine collections were obtained from all patients. A registered dietician then provided the patients with a set of menus and instructed them to follow a gelatin-free 10-mmol calcium, 170260 mmol sodium diet for 11 days. On the tenth day of the low calcium diet, a third 24-h urine collection and a blood sample were obtained. Each urine sample was assayed for calcium, creatinine, phosphate, citrate, oxalate, uric acid, sodium, potassium, and magnesium. The blood sample was assayed for calcium, phosphorus, creatinine, blood urea nitrogen, uric acid, PTH, 1,25-(OH)2D3, and electrolytes. On the last day of the low calcium diet, a standard calcium tolerance test was performed, as described by Pak et al. (5) in the clinical research center of our institution. A 2-h urine sample was collected between 0800-1000 h after a 12-h overnight fast. At 1000 h, patients ingested 1 g elemental calcium in the form of liquid calcium gluconate (NeoCalglucon) mixed with 250 mL distilled water along with two pieces of dry toast and 300 cc apple juice. Urine was then collected between 1000-1400 h. Blood samples were taken at 0800, 1000, and 1400 h. Patients were allowed to drink 300 cc distilled water during the 0800-1400 h period to maintain adequate urine flow. Each urine sample collected during the 0800-1400 h period was assayed for calcium and creatinine. The blood samples collected during the test were assayed for calcium, creatinine, and PTH. In addition, IL-1 activity in the culture medium of peripheral blood monocytes, serum level of osteocalcin (BGP), and 24-h urinary excretion of total hydroxyproline (OHP) were measured in the last 46 (15 FH, 18 AH, and 13 controls) of the 108 patients enrolled in the study in samples obtained on the tenth day of the low calcium diet. To establish a normal range for IL-1 activity, this parameter was also measured in 20 healthy men and 10 healthy premenopausal women, aged 21-65 yr (mean ± SD, 43 ± 9.5 yr), with no history of kidney stones and no hypercalciuria who were ingesting a customary diet. Classification criteria Thirty-eight of the 112 subjects, aged 20-75 years (mean ± SE, 48.2 ± 2.0 yr; median, 46 yr), were classified as nonhypercalciuric stone-formers (controls), because on their customary calcium diet they had an average urinary calcium excretion not exceeding 0.1 mmol/kg BW for either sex and/or 6.23 mmol/ day for females and 7.48 mmol/day for males, and on the low calcium diet they demonstrated a normal fasting urinary calcium/creatinine ratio and a normal response to the oral calcium load. Twelve of these subjects had no metabolic abnormalities. The remaining 26 had abnormal urinary levels of uric acid, oxalate, or citrate or a combination of these abnormalities. The remaining 74 patients had an increased average calcium excretion (as defined above) while on their customary, calciumunrestricted diet. Forty-seven of these patients, aged 23-76 yr (mean ± SE, 41.2 ± 11.8 yr; median, 39), were regarded as having AH, because of a fasting calcium/creatinine ratio of 0.31
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 09:48 For personal use only. No other uses without permission. . All rights reserved.
PACIFICI ET AL.
140
mmol Ca/mmol creatinine or less, an increased urinary calcium response to an oral calcium tolerance test (postload calcium excretion, >0.56 mmol/mmol creatinine) and normal serum PTH levels. Twenty-three hypercalciuric subjects, aged 23-72 yr (mean ± SE, 46.8 ± 2.1 yr; median, 48), were classified as having FH because of a fasting calcium/creatinine ratio greater than 0.31 mmol Ca/mmol creatinine in the 2-h fasting sample (27). In addition, these patients had an increased urinary response to an oral calcium load and normal serum PTH levels. The remaining 4 patients had normal blood calcium levels, FH, and increased circulating PTH levels. These subjects were regarded as having renal hypercalciuria and were excluded from the study. Laboratory methods Biochemical assays. Blood and urinary calcium, phosphate, magnesium, creatinine, sodium, potassium, chloride, CO2, and uric acid were measured with standard atomic absorptiometry and photometric methods. Uric acid was measured by a uricase method (28), blood oxalate by the oxalate oxidase method (29), and PTH by the method of Hruska et al. (30), using CH9 antiserum, which binds to the COOH-terminal and midmolecule epitope of the protein. This assay does not distinguish between normal and suppressed values, but has been shown to separate normal from increased PTH values in most patients with normal renal function (31). 1,25-(OH)2D3 and total urinary OHP were measured by a radioreceptor method (32) and the colorimetric method of Prockop and Udenfriend (33), respectively. BGP was measured by RIA, using an antiserum that binds a midmolecule epitope of human osteocalcin (34). The normal range (mean ± 2 SD) for BGP with this assy is 2.7-13.5 ng/mL. Monocyte cultures. To assay IL-1, blood was drawn into EDTAcontaining tubes, and monocyte cultures were prepared as previously described (20, 35). Briefly, freshly drawn blood was fractionated on Ficoll/Hypaque, and the mononuclear cells were removed from the interface and washed twice with RPMI1640 medium. The medium was assayed for endotoxin by the limulus amoebocyte lysate assay. Endotoxin was not detected at the level of sensitivity of the assay (
