| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Articles |
Endocrine-Hypertension Division (S.T.H., S.M., J.S., G.E.-H.F.) and Psychiatry Division (A.L.S.), Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115
Address all correspondence and requests for reprints to: Ghada El-Hajj Fuleihan, M.D., M.P.H., Endocrine-Hypertension Division, Brigham and Women’s Hospital, Harvard Medical School, 221 Longwood Avenue, Boston, Massachusetts 02115. E-mail: gelhajjful{at}bics.bwh.harvard.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mallette et al. demonstrated increments from baseline in Cai levels at 3 months and in Cai and PTH levels and parathyroid gland volume at 103 months of Li therapy (12). In addition, administration of 600 mg Li to normal healthy volunteers was shown to increase iPTH levels acutely from 22–32 pg/mL without any change in Cai levels (13). These results suggest that Li may indeed exert a direct effect on the parathyroid gland.
In vitro, Li stimulates the release of iPTH from bovine and human parathyroid cells (14, 15) and increases the set-point for PTH release in response to Ca (14). In vivo, the administration of a graded Ca infusion to six healthy men, before and 5 days after Li, resulted in no changes in PTH suppressibility in response to Ca (16). However, a full characterization of the Ca-PTH axis was not possible because of the lack of a citrate infusion (16).
In this study, we have fully characterized the Cai-iPTH axis in patients taking Li and in control subjects to test the hypothesis that Li induces alterations in PTH dynamics that may not always be apparent under baseline conditions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Seven women taking chronic Li therapy for the treatment of bipolar affective disorder and seven age-matched control subjects were studied over 2 days. In both groups, four subjects received a Ca infusion on the first day, followed by the citrate infusion on the second day, whereas three subjects received the citrate infusion on the first day, followed by the Ca infusion on the second day. Because of the effect of the menstrual cycle on PTH levels (17), young women had to have spontaneous, regular cycles of 28–35 days, as documented by a three-month diary, to be eligible for the study. All premenopausal women (five in each study group) were studied during the early follicular phase of their menstrual cycle. In addition, two postmenopausal women, not taking hormone replacement therapy, were studied in each group. The subjects were on no other medication known to affect Ca and bone metabolism. Before enrollment, each subject underwent a physical examination and a laboratory evaluation that included a multichannel serum chemistry analysis and complete blood count with differential, an iPTH level, a TSH level, and a 25 hydroxyvitamin D [25(OH)D] level. Subjects with 25(OH)D levels below 15 ng/mL were excluded. To avoid any confounding effect of obesity on the PTH axis (18), all subjects had a body mass index (weight in kilograms divided by height in meters squared) of less than 30. The mean dietary Ca intake of all study subjects was evaluated with a Ca questionnaire generated by the General Clinical Research Center dietitian. The study was reviewed and approved by the Committee for the Protection of Human Subjects of the Brigham and Women’s Hospital. Informed written consent was obtained from each subject before participation.
Study design
The design of the protocol has been described in detail in previous publications examining PTH dynamics in healthy volunteers (19, 20). The protocol required two visits to the Ambulatory Clinical Center, one for a citrate and the other for a Ca infusion. For each visit, the subject arrived at 0800 h in a fasting state. An iv catheter was placed in a vein of each antecubital fossa and kept open with 5% dextrose in water. One iv line was used for blood sampling and the other for infusion of Ca or citrate. Samples for serum 1,25 dihydroxyvitamin D [1,25(OH)2 D] levels and chemistries, including blood urea nitrogen and creatinine (Cr), were obtained before the infusion started and on completion of the infusion. Serum 25 (OH)D levels were measured at the beginning of the infusion on the first day only.
Citrate infusion
Citrate (anticoagulant-citrate-dextrose USP Formula A [ACD-A], containing per 100 mL: 2.45 g dextrose, 2.2 g sodium citrate, and 0.7 g citric acid; Fenway Laboratories, Deerfield, IL) mixed in 5% dextrose water was administered via an IV infusion pump (Travenol, Deerfield, IL). Throughout the course of the infusion, blood pressure was monitored by an automated blood pressure recorder (adult/pediatric vital sign monitor, Critikon, Inc., Tampa, FL), and an EKG was obtained from a cardiac monitor (Physio Control Lifepak 7, Rowayton, CT) at each step before the infusion rate was increased. The study consisted of four 30-min pulse-step intervals of citrate infusion. In brief, a rapid 5-min infusion of citrate was followed by a slower infusion for 25 min. Progressively increasing rates of both the fast and slow infusions were used for three additional 30-min periods. The citrate dose was 42 mg citrate/mg/kg/hr, followed by 20 mg citrate/mg/kg/hr for the first 30-min interval; dosages for subsequent intervals were 70/33, 96/44, and 130/60 mg citrate/mg/kg/hr, respectively. iPTH levels, measured at 5 min of each step, reflect the effect of change in Cai on iPTH release, whereas the 30-min points, when Cai is more stable, reflect the effect of Cai concentration per se on iPTH dynamics (19). Samples for serum Cai and iPTH levels were collected anaerobically at 0, 5, 10, 20, and 30 min for each step of the infusion.
Ca infusion
Ca gluconate (Astra, Westboro, MA) was infused over three 30-min pulse-step intervals via an IV infusion pump by the same procedure as that used for the citrate. The doses for Ca for the fast/slow infusions were 2.4 mg/kg/hr followed by 0.75 mg/kg/hr in step 1; 3.4 mg/kg/hr followed by 1.25 mg/kg/hr in step 2; and 4.4 mg/kg/hr followed by 1.75 mg/kg/hr in step 3. Samples for serum Cai and iPTH levels were collected anaerobically at 0, 5, 10, 20, and 30 min for each step of the infusion.
Characterization of the Cai-iPTH axis
The following indices of parathyroid function were used to evaluate the relationship of iPTH to Cai: the maximal iPTH response to hypocalcemia, the minimal iPTH level after induced hypercalcemia, the set-point for iPTH, and the slope of the curve at the set-point. The set-point is defined as the serum Cai concentration at which the iPTH level is half maximal (21). Samples obtained at the 30-min time points of both the Ca and citrate infusions, when Cai levels were stable, were used to characterize the Cai-iPTH axis. We used GraphPad Prism software 1.0 (GraphPad Software, San Diego, CA) to fit sigmoidal curves to our data. This computer program generated the four variables mentioned above. In addition, data obtained from all time points of the Ca and citrate infusions also were compared between the two groups to further evaluate the effect of Li on Cai-iPTH dynamics.
Laboratory tests
Serum levels of Ca, phosphate, magnesium, and Cr were determined by the clinical chemistry laboratory by a calorimetric method with an Olympus AU-5061 analyzer (Olympus Corporation, Lake Success, NY). Blood for Cai was collected anaerobically and measured with an AVL 987-S electrolyte analyzer (AVL Scientific Corporation, Roswell, GA), which has a precision of 0.39% and an interassay precision of 1.7–2.5% for Cai levels between 1.12 and 1.48 mmol/L (normal range: 1.15–1.33 mmol/L, N = 57). Fasting urinary Ca and Cr levels were measured on both days before administration of the infusion. Fasting urinary Ca/Cr clearance ratios were calculated for both groups using the following formula: [UCa x SCr]/[SCa x UCr].
Serum iPTH was measured by the Allegro immunoradiometric assay [Nichols Institute, San Juan Capistrano, CA (22)]. The detection limit of the assay is 1.0 pg/mL (normal range 10–65 pg/mL) and the intra- and interassay CV percents are 1.7 and 6.5% at iPTH concentrations of 37.7 and 44.1 pg/mL, respectively. Serum 25(OH)D was measured by a competitive protein-binding assay (Incstar, Stillwater, MN) with the normal range 10–55 ng/mL. A radioreceptor assay kit (Nichols Institute Diagnostics), with the 1,25(OH)2D receptor from calf thymus, was used for the 1,25(OH)2D assay (normal range 15–65 pg/mL). For 25(OH)D, the intraassay CV% is between 8.7 and 8.9% at serum concentrations of 12 and 53 ng/mL, respectively, and the interassay CV% is 12% at a serum concentration of 49 ng/mL. For 1,25(OH)2D, the intraassay CV% is 8.7% at a concentration of 34 pg/mL, and the interassay CV% is 13.2% at a concentration of 43 pg/mL. Except for serum Cai samples, which were measured on the same day, all other serum samples were stored at -70 C. All samples from each patient (except for Cai) were run in duplicate and in the same assay.
Because of the controversial effect of Li on bone mass (8, 23, 24), bone mineral density (BMD) of the spine (L1-L4), hip (femoral neck, trochanter), and forearm (distal 1/3 radius) was measured in all subjects, with dual-energy X-ray absorptiometry, QDR 2000 (quantitative digital radiography, Hologic, Inc, Waltham, MA). BMD was expressed in both groups as Z scores, that is, the number of SD, compared with an age-matched normative database supplied by the bone densitometer software.
Statistical methods
Baseline demographics were analyzed with a two-tailed t test. Summary measures for both study groups, including functions of the maximal responsiveness, minimal responsiveness, set-point, and slope derived from the sigmoidal curve, were compared by t test. Data from both infusions included multiple measures (i.e., Cai and iPTH) sampled repeatedly, over time, that were analyzed with use of repeated-measure ANOVA. The results were expressed as mean ± SEM, unless otherwise indicated. Significance was indicated for P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As shown in Table 1
, there was no
significant difference in the average age, body mass index, or dietary
Ca intake between the two study groups. The average duration of Li use
was 5.23 ± 4.0 yr (mean ± SD). The mean Li
level for six patients, obtained within 3 months of the study and while
on a stable Li dose for at least 6 months, was 0.63 ± 0.08 mmol/L
(normal range is 0.5 to 1.3 mmol/L). Mean baseline serum Ca, 25(OH)D,
and 1,25(OH)2D levels were not significantly different
between the two groups. However, serum Cai and iPTH levels
were slightly, but significantly, higher in the Li-treated group (Table 1). Fasting urinary Ca levels, but not urinary Ca/Cr clearance ratios,
were significantly lower in the Li-treated patients, compared with the
controls, P = 0.046 (Table 1).
|
The set-point is defined as the Cai concentration at
which iPTH levels are half-maximally inhibited. Rapid changes in serum
Cai levels can shift the sigmoidal curve, relating
Cai to iPTH levels, to the right, thereby elevating the
set-point (19). Therefore, to evaluate the effect of changing
Cai concentration per se rather than the effect
of the rate of change in Cai on iPTH levels, we used the
30-min points from both the citrate and Ca infusions to plot the
sigmoidal curve. The 30-min time points reflect stable Cai
levels following the increases in infusion rates. The set-points were
calculated using two methods. In the first method, the 30-min time
points for both the citrate and Ca infusions were averaged for each
study group, and two separate sigmoidal curves were fitted to the
average data points for each study group, as shown in Fig. 1. The set-points derived from this
method were 5.10 ± 0.01 mg/dL for the Li-treated patients and
4.86 ± 0.02 mg/dL for the controls. In the second method, a
sigmoidal curve was fitted to each individual’s data to derive an
individual set-point. The individuals’ set-points were then averaged
to calculate a mean set-point for each study group. The set-point with
this second method was 5.08 ± 0.04 mg/dL for the Li-treated
patients and 4.88 ± 0.04 mg/dL for the normal subjects,
P = 0.004. According to the second method, the average
slopes of the sigmoidal curves in the Li group and the control group
were -12.9 ± 1.9 and -12.2 ± 1.5, respectively,
P = 0.768. The maximal and minimal iPTH responses
derived from the sigmoidal curves did not differ between the two study
groups; they were 80.4 ± 5.0 pg/mL and 11 ± 2.12 pg/mL in
the Li-treated group, compared with 79.9 ± 12.2 pg/mL and
6.2 ± 3.3 pg/mL in the control group, respectively.
|
An evaluation of the data obtained from all time-points revealed
that the Cai levels were higher in the Li-treated patients
than in the controls throughout the citrate infusion, P
= 0.0008 (two-way ANOVA, Fig. 2A). Both
study groups had sharp increments in iPTH levels within the first 10
min of each step that were not significantly different between the two
groups (two-way ANOVA, Fig. 2B).
|
As in the citrate infusion, an evaluation of the data obtained
from all time-points revealed that Li-treated patients had
significantly higher serum Cai levels than the controls,
P = 0.018 (two-way ANOVA, Fig. 3A). However, iPTH levels did not differ
significantly between the two study groups in response to the Ca
infusion, (Fig. 3B). However, our study did not have the power to
detect differences in PTH suppressibility of the magnitude
observed.
|
The BMD, expressed as Z scores, of the lumbar spine in the Li-treated patients was slightly higher than that of the controls (Z = 0.73 ± 0.4 and Z = -0.32 ± 0.3 respectively, P = 0.08). In the Li and the control group, the Z scores for the femoral neck were -0.09 ± 0.4 and -0.19 ± 0.6, respectively; for the trochanter, -0.13 ± 0.4 and -0.14 ± 0.4, respectively; and for the distal 1/3 radius, 0.42 ± 0.3 and -0.15 ± 0.21, respectively. There were no differences between the two study groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Stimuli for elevated iPTH levels include low serum Ca, 25(OH)D levels, and/or 1,25(OH)2D levels. Despite comparable baseline total Ca and vitamin D metabolite levels in the two study groups, serum Cai was higher in patients on Li. If parathyroid dynamics were entirely normal, such elevations in Cai levels would be expected to result in lower, rather than higher, iPTH levels as observed in our study. Other conditions leading to elevated iPTH levels include mild renal failure or renal hypercalciuria (25, 26). Renal function in the Li-treated patients, as determined by serum Cr levels, was normal, and fasting urinary Ca/Cr clearance ratios were comparable with the ratio calculated in the control group. However, the lower urinary Ca, but not Ca/Cr clearance ratio, in the Li-treated subjects, despite comparable serum Ca between the two groups, raises the possibility of an effect of Li on renal function (5, 27).
The Li-induced shift in the iPTH set-point to the right that has previously been demonstrated in vitro (14) and is now documented in vivo in our studies suggests a direct effect of this cation on the parathyroid gland. This hypothesis is further confirmed by the acute increase in PTH levels in response to Li administration, both in vitro and in vivo (13, 14, 15). The mechanisms of Li-induced changes at the cellular level have been studied extensively (28). Considerable evidence has accumulated over the last few years, identifying key components of signal transduction pathways (G proteins, adenyl cyclase, protein kinase C) and gene expression in the central nervous system as targets for the molecular effects of Li (28). Racke et al. evaluated the locus of action of Li in parathyroid cells and demonstrated that Li blunted cation (Ca and magnesium)-induced changes in cytosolic Ca but had no effect on Ca-induced hydrolysis of inositol phosphates (29, 30). These findings suggest that the locus of action of Li is distal to the calcium receptor (CaR), although its precise site remains unclear. The observed alteration in PTH dynamics and the reduction in Ca excretion in patients on Li therapy could be caused by an effect of Li on the CaR signal transduction system of the parathyroid gland and kidney and can be likened to familial hypocalciuric hypercalcemia (FHH) (31). Using a Ca infusion, Khosla et al. suggested a shift in the Ca/PTH curve to the right in patients with FHH. However, because a hypocalcemic stimulus was not applied, a full characterization of the PTH/Ca axis was not possible (32).
The above-proposed Li-induced alterations in Ca sensing, and therefore PTH dynamics, could conceivably result in hyperparathyroidism with four-gland hyperplasia. However, two reviews evaluating the pathology of the parathyroid glands in patients with hyperparathyroidism have yielded conflicting results (4, 33). In the review of Mallette et al., the pathology was predominantly that of adenomas (4). A more recent histopathologic retrospective study demonstrated that the predominant pathology was that of parathyroid hyperplasia in five out of six patients treated with Li, whereas an adenoma was found in ten out of eleven patients on tricylic antidepressant medication (33). It would seem that Li unmasks frank hyperparathyroidism in only a subset of patients predisposed for that condition, apriori.
The effect of the Li-induced hyperparathyroidism on the skeleton remains controversial (8, 23, 24). Spine, hip, and distal 1/3 radius Z scores in both the Li-treated patients and the controls were within the range expected for age in all sites studied. These results are, however, based on a small number of subjects studied cross-sectionally. Yet, similarly to FHH (34, 35), patients on Li may have normal bone density despite slightly higher iPTH levels, possibly caused by a lowering of urinary Ca excretion (4, 23).
In conclusion, our study documents a clear alteration in PTH dynamics in patients on chronic Li therapy, even though their baseline total Ca and iPTH levels were well within the normal range. This may reflect generalized altered Ca sensing, not only at the level of the parathyroid gland, but also of the kidney.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received February 26, 1997.
Revised May 6, 1997.
Accepted May 29, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. J. Felsenfeld, M. Rodriguez, and E. Aguilera-Tejero Dynamics of Parathyroid Hormone Secretion in Health and Secondary Hyperparathyroidism Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1283 - 1305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Livingstone and H. Rampes Lithium: a review of its metabolic adverse effects J Psychopharmacol, May 1, 2006; 20(3): 347 - 355. [Abstract] [PDF]
|
|
|
|
 |
|
 P. Clement-Lacroix, M. Ai, F. Morvan, S. Roman-Roman, B. Vayssiere, C. Belleville, K. Estrera, M. L. Warman, R. Baron, and G. Rawadi Lrp5-independent activation of Wnt signaling by lithium chloride increases bone formation and bone mass in mice PNAS, November 29, 2005; 102(48): 17406 - 17411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Caskey and T. M. Pickett Disturbed calcium metabolism in a patient with bipolar disorder and impaired renal function Nephrol. Dial. Transplant., August 1, 2005; 20(8): 1752 - 1755. [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Strewler A 64-Year-Old Woman With Primary Hyperparathyroidism JAMA, April 13, 2005; 293(14): 1772 - 1779. [Full Text] [PDF]
|
|
|
|
 |
|
 J.-C. Souberbielle, E. Lawson-Body, B. Hammadi, E. Sarfati, A. Kahan, and C. Cormier The Use in Clinical Practice of Parathyroid Hormone Normative Values Established in Vitamin D-Sufficient Subjects J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 3501 - 3504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M.A. Hirschfeld, C. L. Bowden, M. J. Gitlin, P. E. Keck, R. H. Perlis, T. Suppes, M. E. Thase, and K. D. Wagner Practice Guideline for the Treatment of Patients With Bipolar Disorder (Revision) Focus, January 1, 2003; 1(1): 64 - 110. [Full Text] [PDF]
|
|
|
|
 |
|
 E. Smith, G. A. Coetzee, and B. Frenkel Glucocorticoids Inhibit Cell Cycle Progression in Differentiating Osteoblasts via Glycogen Synthase Kinase-3beta J. Biol. Chem., May 10, 2002; 277(20): 18191 - 18197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Haden, E. M. Brown, A. L. Stoll, J. Scott, and G. El-Hajj Fuleihan The Effect of Lithium on Calcium-Induced Changes in Adrenocorticotrophin Levels J. Clin. Endocrinol. Metab., January 1, 1999; 84(1): 198 - 200. [Abstract] [Full Text]
|
|
|
|
|
|
T. W. L. Mak, C.-C. Shek, C.-C. Chow, Y.-K. Wing, and S. Lee Effects of Lithium Therapy on Bone Mineral Metabolism: A Two-Year Prospective Longitudinal Study J. Clin. Endocrinol. Metab., November 1, 1998; 83(11): 3857 - 3859. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
