Thiazide-Associated Hypercalcemia: Incidence and Association With Primary Hyperparathyroidism Over Two Decades

Abstract

Context:

Thiazide diuretics, the antihypertensive agent prescribed most frequently worldwide, are commonly associated with hypercalcemia. However, the epidemiology and clinical features are poorly understood.

Objective:

To update the incidence of thiazide-associated hypercalcemia and clarify its clinical features.

Patients and Methods:

In a population-based descriptive study, Olmsted County, Minnesota, residents with thiazide-associated hypercalcemia were identified through the Rochester Epidemiology Project and the Mayo Clinic Laboratory Information System from 2002–2010 and were added to the historical cohort beginning in 1992.

Main Outcome:

Incidence rates were adjusted to the 2010 United States white population.

Results:

Overall, 221 Olmsted County residents were identified with thiazide-associated hypercalcemia an average of 5.2 years after initiation of treatment. Subjects were older (mean age, 67 years) and primarily women (86.4%). The incidence of thiazide-associated hypercalcemia increased after 1997 and peaked in 2006 with an annual incidence of 20 per 100 000, compared to an overall rate of 12 per 100 000 in 1992–2010. Severe hypercalcemia was not observed in the cohort despite continuation of thiazide treatment in 62.4%. Of patients discontinuing thiazides, 71% continued to have hypercalcemia. Primary hyperparathyroidism was diagnosed in 53 patients (24%), including five patients who underwent parathyroidectomy without thiazide discontinuation.

Conclusions:

Many patients with thiazide-associated hypercalcemia have underlying primary hyperparathyroidism. Additionally, a sharp rise in thiazide-associated hypercalcemia incidence began in 1998, paralleling the increase observed in primary hyperparathyroidism in this community. Case ascertainment bias from targeted osteoporosis screening is the most likely explanation.

Over 65 million persons in the United States are affected by hypertension (1), with approximately 45.7 million patients utilizing antihypertensive therapy in 2013, almost twice the number in the next closest medication category (2). Thiazides are first-line treatment in essential hypertension and reduce mortality and cardiovascular events (1, 3). Hydrochlorothiazide is the most commonly prescribed antihypertensive medication worldwide, with over 50 million dispensed prescriptions for monotherapy in the United States alone in 2013 (2). In 2008, there were nearly 135 million prescriptions in the United States when combination products with hydrochlorothiazide, which comprise approximately two-thirds of total prescriptions, were included (2, 4).

Hypercalcemia associated with thiazide use is a well-known clinical entity. Thiazides exert their antihypertensive effect through an increase in sodium excretion by blocking the thiazide-sensitive NaCl transporter in the distal convoluted tubule, which is closely linked to calcium transport (5). Thiazides have several metabolic effects contributing to higher serum calcium levels, but increased renal tubular reabsorption of calcium resulting in reduced urine calcium excretion is the most likely cause (5–7). Serum calcium concentrations are increased independently of PTH levels (8). Conversely, loop diuretics induce natriuresis by inhibiting the Na-K-2Cl transporter in the thick ascending limb of the loop of Henle, causing increased urinary calcium losses and increased PTH (5, 8).

The epidemiology of thiazide-associated hypercalcemia is poorly understood. Based on prospective health screening measurements in the early 1970s in Sweden, Christensson et al (9) suggested that the prevalence of hypercalcemia in 20 to 63 year olds on thiazide diuretics was 1.9%—nearly three times higher than observed in same-aged subjects in the general population (0.6% prevalence). Most subjects with hypercalcemia on thiazides were older hypertensive women (90%) on treatment for 1–15 years. Withdrawal of thiazides resulted in persistent hypercalcemia in 75% after 1 year of follow-up. Subsequently, 70% of these subjects underwent parathyroidectomy for primary hyperparathyroidism (PHPT).

A second study from Sweden evaluating hypercalcemia after the introduction of automated chemistry panels in a single county hospital estimated that the prevalence of hypercalcemia in thiazide-treated adults was 0.4%, again approximately two to three times higher than was observed in the general population (10). However, when this study was conducted, chemistry panels were primarily performed on hospitalized patients, rather than the much larger ambulatory population. Even in this highly selected patient group, the hypercalcemia associated with thiazides was mild and predominantly in females (80%).

Since the early 1970s, only one study has evaluated the epidemiology of thiazide-associated hypercalcemia (11). From 1991 to 2001, the overall age and sex-adjusted annual incidence of thiazide-associated hypercalcemia was 7.7 per 100 000 among residents of Olmsted County, Minnesota. The incidence increased after 1996, peaking at 16.2 per 100 000 in 1998. The affected subjects were older (mean age, 64 years) and predominantly women (94%), and were identified an average of 6 years after the initiation of thiazides. In patients who discontinued thiazide, 64% had persistent hypercalcemia. Patients in this cohort who were subsequently diagnosed with PHPT had higher serum calcium and PTH values.

The purpose of this study was to update our understanding of the incidence of thiazide-induced hypercalcemia in an era of routine calcium measurement, now spanning over two decades. In addition, we wanted to gain better understanding of its clinical features and relationship to PHPT. Finally, given the unique record system in this community (12), we hoped to provide guidance to clinicians in the evaluation and management of this disorder based on its natural history among unselected patients in the general population.

Patients and Methods

Most endocrinological care in this community is provided by the Mayo Clinic, which has maintained a common medical record with its two hospitals for over 100 years. The diagnoses and surgical procedures in these records are indexed, as are the medical records of other providers who serve the local population (13). After approval from the Institutional Review Boards of Mayo Clinic and the Olmsted Medical Center, we used this comprehensive medical records-linkage system (the Rochester Epidemiology Project) to identify all Olmsted County, Minnesota, residents with hypercalcemia from 2002 through 2010, including all patients with PHPT as described previously (14). In addition, all Olmsted County residents with serum calcium levels exceeding 10.1 mg/dL (2.52 mmol/L) at least twice between 2002 and 2010 were identified directly from Mayo's Laboratory Information System. For each potential case, the complete (inpatient and outpatient) community medical record was reviewed by one of the investigators (M.L.G.). Mayo Clinic records contain details of every inpatient hospitalization, outpatient clinic visit, emergency department and nursing home care, as well as all radiographic and pathology reports, including autopsies (12). This information was supplemented by that available from other providers for local residents, most notably the Olmsted Medical Center (13).

Patients were accepted as having thiazide-associated hypercalcemia if they met the following criteria: sustained hypercalcemia (serum calcium >10.1 mg/dL [2.52 mmol/L]) documented on two or more measurements with concomitant thiazide diuretic use for which no other cause (eg, pathologically or biochemically proven PHPT before initiation of thiazides, malignancy, family history of familial benign hypocalciuric hypercalcemia [FHH; OMIM 145980], creatinine level >2 mg/dL, or lithium therapy) was identified. Patients with hypercalcemia before thiazide initiation were excluded. Although the methods used to measure serum calcium levels changed over time, the normal range remained unchanged because instrumentation was calibrated against atomic absorption spectrophotometry according to certified references from the National Bureau of Standards. The date of diagnosis of thiazide-induced hypercalcemia was the date of the first elevated calcium level while on thiazides, consistent with prior criteria (11). Patients with thiazide-associated hypercalcemia who subsequently had biochemically proven PHPT must have met the inclusion criteria consistent with previous studies 3 months or longer after thiazide discontinuation (14). Serum intact PTH was measured by two-site immunochemiluminometric assay, which changed throughout the course of the study with normal ranges as follows: 1.0–5.2 pmol/L by manual bead immunoassay (in-house assay) in 1992 to December 2, 2003; 10–55 pg/mL (1.1–5.8 pmol/L) by Nichols Advantage (Nichols Institute Diagnostics) in December 3, 2003 to May 1, 2006; 10–67 pg/mL (1.1–7.1 pmol/L) by Diagnostics Products Corporation assay in March 31, 2005, to July 15, 2007; and 15–65 pg/mL (1.59–6.90 pmol/L) as performed on the Roche Cobas 6000 (Roche Diagnostics GmbH) since July 16, 2007. To qualify for the study, patients must have been residents of Olmsted County on the date the first elevated serum calcium was discovered. In addition, to assure comparability with our previous study (11), we established residency in Olmsted County for at least 1 year before the diagnosis of thiazide-associated hypercalcemia.

The updated cohort from 2002–2010 was combined with the previously identified Olmsted County thiazide-associated hypercalcemia cases from 1992 through 2001 (11). Incidence rates were based on the date of the initial elevated serum calcium level, consistent with previous studies (11). Age- and sex-specific incidence rates were estimated using the number of patients in each age (<45, 45–54, 55–64, 65–74, 75+ years) and sex group as the numerator, with corresponding age- and sex-specific person-years at risk as denominators (15). Person-years were estimated by decennial census data for the Olmsted County population with interpolation for intercensal years. The rates from our study, including previous reports, were directly age- and sex-adjusted to the U.S. white population in 2010. Ninety-five percent confidence intervals (CIs) for the incidence rates were calculated under the Poisson distribution. Categorical variables are presented as counts and percentages, whereas continuous data are summarized as means (standard deviation) or median (25–75% interquartile range). Independent t test was used to compare continuous variables, and χ² test was used to compare categorical variables.

Results

We identified 221 Olmsted County residents with thiazide-associated hypercalcemia (191 women, 30 men) during the 19-year study period, 1992–2010. The overall annual age- and sex-adjusted incidence was 12.2 (95% CI, 10.6–13.8) per 100 000 (Table 1). The incidence of thiazide-associated hypercalcemia increased after 1997, with an incidence of 19.3 and 19.7 per 100 000 person-years in 1998 and 1999, respectively, and a subsequent peak incidence in 2006 of 20.4 per 100 000 person-years (Figure 1). The annual incidence rate declined to 12.3 per 100 000 in 2010. The overall age-adjusted incidence was much higher in women (19.5 per 100 000 person-years; 95% CI, 16.7–22.3) than in men (3.6 per 100 000 person-years; 95% CI, 2.3–5.0; P < .001). The highest incidence rate was 76.3 per 100 000 person-years in 65- to 74-year-old women.

Table 1.

Incidence of Thiazide-Associated Hypercalcemia Among Olmsted County, Minnesota, Residents, 1992–2010, by Gender and Age Group

Age Group, y	Women		Men		Both Sexes
	n	Ratea	n	Ratea	n	Ratea
<45	4	0.5	3	0.4	7	0.4
45–54	18	10.9	3	1.9	21	6.5
55–64	57	51.1	12	11.3	80	31.7
65–74	57	76.3	6	9.3	71	45.2
≥75	55	66.0	6	12.6	73	46.5
Total	191	19.5 (16.7–22.3)b	30	3.6 (2.3–5.0)b	221	12.2 (10.6–13.8)c

^aIncidence per 100 000 person-years.

^bIncidence per 100 000 person-years directly age-adjusted to the United States white population in 2010.

^cIncidence per 100 000 person-years directly age- and sex-adjusted to the United States white population in 2010.

Figure 1.

Age-adjusted (to 2010 U.S. whites) incidence (per 100 000 person-years) of thiazide-associated hypercalcemia among Olmsted County, Minnesota, women (solid line) and men (dashed line), 1992–2010.

The mean age at diagnosis was 67.0 ± 11.3 years, and women comprised most cases (86.4%) (Table 2). The average serum calcium level before thiazide initiation was 9.68 ± 0.37 mg/dL [2.42 ± 0.09 mmol/L], as measured in 198 patients a mean of 26.2 months beforehand. Hypercalcemia was identified a mean 5.2 years (median, 3.9 years) after thiazide initiation (range, 4 days to 33 years). The average highest serum calcium was 10.72 mg/dL [2.67 mmol/L] (range, 10.2 to 12.5 mg/dL [2.5–3.1 mmol/L]). Most subjects were treated with thiazides for hypertension (94%). Serum PTH was measured in 121 patients after identification of hypercalcemia. In these patients, PTH was above normal in 41 subjects (34%); 50% of normal to the upper limit of normal in 52 (43%); <50% of normal to the lower limit of the normal in 26 (21%); and suppressed in two (2%). Neither of the patients with suppressed PTH levels had an identifiable secondary cause of their hypercalcemia.

Table 2.

Clinical and Laboratory Spectrum of Thiazide-Associated Hypercalcemia Among Olmsted County, Minnesota, Residents 1992–2010, Overall and in a Subset Later Found to Have PHPT

Characteristic	All Patients	PHPT Subset
Female gender	191 (86.4)	44 (83.0)
Age at onset of hypercalcemia, y	67.0 ± 11.3	67.8 ± 11.2
Serum calcium before thiazide use, mg/dL	9.68 ± 0.37	9.68 ± 0.43
Serum calcium before thiazide use, mmol/L	2.42 ± 0.09	2.42 ± 0.11
Maximum serum calcium on thiazides, mg/dL	10.72 ± 0.34	10.85 ± 0.40
Maximum serum calcium on thiazide, mmol/L	2.67 ± 0.08	2.71 ± 0.10
Years from thiazide start to hypercalcemia	5.2 ± 5.0	5.8 ± 7.2
PTH levels
Not measured	100 (45.2)	5 (9.4)
<Lower limit of normal	2 (0.9)	0 (0.0)
<50% of normal to lower limit of normal	26 (11.8)	5 (9.4)
50% of normal to upper limit of normal	52 (23.5)	20 (37.7)
>Upper limit of normal	41 (18.6)	23 (43.4)
Reason for thiazide usea
Hypertension	208 (94.1)	52 (98.1)
Edema	4 (1.8)	0 (0)
Hypercalciuria/nephrolithiasis	10 (4.5)	1 (1.9)

Data are expressed as mean ± SD or number (percentage).

^aSome patients had more than one reason for thiazide use.

Thiazide therapy was discontinued in 83 patients (38%) an average of 85 months (range, 1.6 to 397 months) after thiazide initiation (Figure 2). Serum calcium stayed persistently normal in 24 patients (29%) discontinuing thiazides, with a mean last serum calcium measurement 3.3 ± 3.4 years after the thiazide was stopped. Fifty-nine patients (71%) continued to have hypercalcemia despite stopping the thiazide. Overall, 138 patients (62%) continued their thiazide despite hypercalcemia, including five patients who underwent successful parathyroidectomy. In patients who continued thiazides, the mean serum calcium level at last measurement was 10.07 ± 0.48 mg/dL [2.51 ± 0.12 mmol/L] an average of 4.7 ± 3.4 years after detection of hypercalcemia, with continued hypercalcemia on the last measurement in 72 subjects (52.1%). The highest last measured serum calcium in those continuing thiazides was 11.1 mg/dL [2.77 mmol/L]. Patients continuing thiazides, excluding the five patients with PHPT who underwent parathyroidectomy, had similar clinical characteristics compared to those in whom the thiazide was discontinued (Table 3).

Figure 2.

Management and diagnostic profile of thiazide-associated hypercalcemia in Olmsted County, Minnesota, residents, 1992–2010.

Table 3.

Clinical and Laboratory Spectrum of Thiazide-Associated Hypercalcemia Among Olmsted County, Minnesota, Residents 1992–2010, Comparing Subsets Continuing and Discontinuing Thiazide

Characteristic	Thiazide Discontinuation	Thiazide Continuation Without Parathyroid Surgery	P Value
Female gender	73 (88.0)	114 (85.7)	NS
Age at onset of hypercalcemia, y	68.3 ± 11.2	66.4 ± 11.4	NS
Serum calcium before thiazide use, mg/dL	9.70 ± 0.35	9.67 ± 0.38	NS
Serum calcium before thiazide use, mmol/L	2.42 ± 0.09	2.41 ± 0.09
Maximum serum calcium on thiazides, mg/dL	10.79 ± 0.39	10.66 ± 0.25	NS
Maximum serum calcium on thiazides, mmol/L	2.69 ± 0.10	2.66 ± 0.06
PTH levels
Not measured	27 (32.5)	73 (54.9)	NS
<Lower limit of normal	0 (0.0)	2 (1.5)
<50% of normal to lower limit of normal	11 (13.3)	15 (11.3)
50% of normal to upper limit of normal	26 (31.3)	26 (19.5)
>Upper limit of normal	19 (22.9)	15 (11.3)
Years from thiazide start to hypercalcemia	6.2 ± 6.8	4.6 ± 3.5	NS
Reason for thiazide usea			NS
Hypertension	81 (97.6)	123 (92.5)
Edema	2 (2.4)	2 (1.5)
Hypercalciuria/nephrolithiasis	2 (2.4)	8 (6.0)

Abbreviation: NS, not significant. Data are expressed as mean ± SD or number (percentage).

^aSome patients had more than one reason for thiazide use.

PHPT was subsequently diagnosed in 53 patients (24%) with thiazide-associated hypercalcemia. Of the patients with PHPT, 23 (43%) had pathological confirmation; 23 (43%) had an inappropriate PTH in the setting of hypercalcemia; and seven (13%) had persistent hypercalcemia for 1 year or more after stopping the thiazide. Pathology revealed a parathyroid adenoma in 21 (91%) and multigland disease in two (9%). However, six patients had intermittent hypercalcemia postoperatively, including five patients with a single adenoma resected, suggesting that multigland disease was possibly present in seven individuals (30%). The mean age at onset of hypercalcemia in patients diagnosed with PHPT was 67.8 years, and 44 (83%) were women (Table 2). The mean maximum serum calcium in this subset was 10.85 ± 0.40 mg/dL [2.71 ± 0.10 mmol/L]. The mean time to identification of hypercalcemia after initiation of the thiazide in patients diagnosed with PHPT was 5.8 ± 7.2 years. Patients with PHPT had serum calcium levels above 11 mg/dL [2.74 mmol/L] in 26% compared to 10% of all patients on thiazides without formally diagnosed PHPT (P < .001).

Discussion

We have identified 221 Olmsted County, Minnesota, residents diagnosed with thiazide-associated hypercalcemia over the 19-year period, 1992–2010. Over this interval, an increased incidence of the condition began in 1998 and peaked in 2006. The subsequent incidence of thiazide-associated hypercalcemia remained above 1992–1997 rates despite a decline after 2006. The increase and subsequent decrease in the incidence of thiazide-associated hypercalcemia parallels the observed incidence of PHPT in Rochester (the central city of Olmsted County) during the same period (14). After an initial peak in 1974 that was linked to the introduction of automated serum chemistry panels, we identified a second sharp rise in the incidence of PHPT in 1998, which continued until 2007 when the rate also declined. However, the incidence of PHPT is likely higher in this group than was reported in Rochester residents generally because only 53 patients (24%) in our thiazide cohort met formal inclusion criteria for PHPT, given the requirement to have biochemical evidence 3 months or longer after thiazide discontinuation unless parathyroid surgery confirmed the diagnosis while remaining on thiazides.

Ascertainment bias, due to targeted screening for hypercalcemia as part of a secondary osteoporosis evaluation, is the most likely explanation for the increase in the incidence of thiazide-associated hypercalcemia since 1998. Specific treatments for osteoporosis first became available in the mid- 1990s. In addition, the National Osteoporosis Foundation's initial clinical practice guidelines for osteoporosis, as well as the Medicare Bone Mass Measurement Coverage Standardization Act that added coverage for bone mineral density (BMD) testing, both occurred in 1998 (16). Subsequently, Medicare reimbursement for BMD testing in clinics was reduced in 2007. The resulting attenuation in growth of BMD measurement (17) corresponds to the decline in both PHPT and thiazide-associated hypercalcemia incidence rates. We also identified a progressive increase in PTH testing at our institution from 1994–2008, whereas serum calcium measurement declined after a regulatory change in June 1996 that required individual orders rather than utilization of automated chemistry panels (14). However, PTH and calcium measurement were more common in women and at older ages—the same groups at risk for osteoporosis. Hence, the introduction of osteoporosis guidelines, availability of U.S. Food and Drug Administration-approved osteoporosis therapies, and changes in reimbursement for BMD testing may have significantly influenced the measurement of serum calcium and PTH in individuals at risk for both PHPT and thiazide-associated hypercalcemia.

Patients with thiazide-associated hypercalcemia have similar clinical characteristics to PHPT subjects in the community. Most patients with thiazide-associated hypercalcemia are older (mean age, 67 years) and female (86%), which compares to 77% females and a mean age of 60 years in Rochester PHPT patients in 1998–2010. Although the association between thiazide use and PHPT is not entirely clear, thiazides have been implicated as a cause of parathyroid gland enlargement (18). Interestingly, we noted that 30% of parathyroidectomy subjects may have multigland disease, which is 2-fold higher than is reported in sporadic PHPT. The relationship between thiazides and PHPT may also reflect a chance association with hypertension because both are more prevalent with older age. However, higher urinary calcium excretion and PTH levels have been reported in hypertensive patients (19). Hypercalcemia was identified an average of 5.2 years after thiazide initiation, which is later than would be expected if the hypercalcemia were primarily due to a thiazide effect upon renal tubular calcium resorption. Bolland et al (20) identified a mean increase of 0.8 mg/dL in albumin-adjusted serum calcium levels with thiazide use, but the increase in renal tubular calcium resorption and serum calcium was no longer seen after 2 years. Although never validated, some have suggested that a “thiazide challenge” may be a useful way to distinguish resorptive hypercalciuria (ie, PHPT) from a renal calcium leak causing secondary hyperparathyroidism in patients with normal renal function (21).

We estimate that 71% of patients with thiazide-associated hypercalcemia may have underlying PHPT based on continued hypercalcemia after stopping thiazides in our cohort. Factors that favor PHPT when hypercalcemia is observed while on thiazide therapy, in addition to hypercalcemia that is identified more than 2 years after thiazide initiation, include higher serum calcium and PTH levels at baseline. Patients with thiazide-associated hypercalcemia who were subsequently diagnosed with PHPT had higher maximum serum calcium levels (10.9 mg/dL [2.72 mmol/L]) compared to the overall cohort (10.7 mg/dL [2.67 mmol/L]). Patients diagnosed with PHPT also had PTH levels ≥ 50% of normal in 90% of those in whom it was measured, compared to 77% in the overall cohort.

If PHPT is the likely diagnosis in a patient with thiazide-associated hypercalcemia and PHPT surgical criteria are present (22), then parathyroidectomy may be recommended. However, it is important to consider the possibility of FHH before surgery, although thiazides can be associated with a reduction in urinary calcium excretion, thus complicating screening for this disorder. FHH is an autosomal dominant disorder with nearly 100% penetrance of hypercalcemia at all ages (23–25). Hence, prior documentation of normal serum calcium levels and the absence of hypercalcemia in family members would make the diagnosis of FHH unlikely in patients with thiazide-associated hypercalcemia. Patients with FHH are generally devoid of PHPT complications (eg, nephrolithiasis), and their risk of osteoporosis is similar to the general population. Also, PTH levels are elevated in only 5–25% of FHH subjects (23, 26). If a 24-hour urine calcium and creatinine measurement is needed in patients with thiazide-associated hypercalcemia, it can be performed after withholding diuretics for 3–5 days (26). It is notable that all five patients who had parathyroidectomy without thiazide discontinuation in this cohort had successful surgical outcomes.

Long-term discontinuation of hydrochlorothiazide may not be necessary in patients with thiazide-associated hypercalcemia. Indeed, we did not observe any serious sequelae related to hypercalcemia in these patients, despite continued use in 62%. Only 10% of the patients had maximal serum calcium levels >11 mg/dL [2.74 mmol/L], and no hypercalcemia emergencies were seen. Rare cases of thiazide-associated severe hypercalcemia have been reported, but they were generally associated with other factors contributing to hypercalcemia (27). Nonetheless, patients with preexisting hypercalcemia or other clinical conditions or medications putting them at risk for hypercalcemia may not be ideal candidates for thiazide diuretic use. In this study, we excluded patients who had hypercalcemia before thiazide initiation and did not address the use of thiazides in patients with preexisting hypercalcemia.

On the other hand, thiazides are inexpensive and effective antihypertensive agents with proven cardiovascular benefit. This is especially of concern because some have reported increased cardiovascular disease in patients with PHPT (28). In addition, thiazides offer a possible reduction in fracture risk (29), lessen the likelihood of nephrolithiasis (5), and may theoretically reduce the progression of PHPT if present (11). It is also possible that thiazides may be beneficial in PHPT patients with renal hypercalciuria, which has been associated with persistent hyperparathyroidism after parathyroidectomy and parathyroid adenomas in the context of preceding chronic secondary hyperparathyroidism (30–32). Additionally, subjects with renal hypercalciuria may be more likely to develop hypercalcemia several years after initiation of thiazides in association with multigland parathyroid disease.

A potential concern is vitamin D and calcium supplementation in thiazide-treated patients, which could be associated with increased hypercalcemia via increased intestinal absorption of calcium. However, only limited case reports have described hypercalcemia with concomitant use of thiazides and vitamin D supplements (33). In a post hoc analysis of black subjects on hydrochlorothiazide who were randomized to placebo or vitamin D3 up to 4000 IU daily, only one of the 83 subjects had hypercalcemia at 3 months (34). Moreover, vitamin D repletion in patients with vitamin D deficiency and PHPT has been shown to be safe and effective in restoring satisfactory vitamin D levels without increasing serum or urine calcium while reducing PTH levels (35). PTH secretion can also be effectively modified by the serum calcium level (36) and oral calcium intake (37, 38). Hence, in general, use of typical vitamin D and calcium supplements in patients who are also on thiazides does not appear to be associated with significant problems related to hypercalcemia.

Our study has expected limitations given its retrospective design. Thus, we had no prospective measurements of serum calcium or other calcium-related laboratory tests before or after thiazide initiation, making it difficult to precisely delineate when hypercalcemia occurs after thiazide initiation. We also did not have routine assessment of PHPT-related complications such as osteoporosis or heart disease. Additionally, the population of Olmsted County is primarily white, which limits the application of study results to more ethnically diverse populations (13). This is important because PHPT appears to be more common among blacks compared to other races (39). Black individuals also have higher rates of hypertension and are commonly prescribed thiazide diuretics due to their effectiveness in low-renin patient groups (40). We were unable to confidently assess thiazide prescriptions at our institution due to changes in the prescription ordering system during the period of interest. Finally, we could neither evaluate dietary calcium intake nor calcium and vitamin supplementation in these patients.

Conclusion

There was an increase in the incidence of thiazide-associated hypercalcemia that began in 1998, peaked in 2006, and subsequently declined thereafter. These changes correspond to similar changes observed in PHPT incidence rates in this community during the same period. Practice variations throughout these years, including the introduction of osteoporosis screening guidelines, new osteoporosis therapies, and modified Medicare reimbursement paradigms for BMD testing have likely resulted in case ascertainment bias due to targeted screening in patients being evaluated for osteoporosis. We estimate that as many as 71% of subjects with thiazide-associated hypercalcemia may have underlying PHPT based on the persistence of hypercalcemia in those who discontinue the medication. The most appropriate management strategy for these patients remains unclear. Discontinuation of thiazides may not be necessary in all patients given the mild and nonprogressive nature of the hypercalcemia that is observed and the potential benefits that are conferred to patients who continue treatment. Additionally, patients with thiazide-associated hypercalcemia who have clinical features suggestive of PHPT may benefit from evaluation for PHPT complications rather than discontinuation of the thiazide. However, randomized trials of continuation vs discontinuation of thiazides would be required to elucidate the best management recommendations for patients with thiazide-associated hypercalcemia.