Prevalence of Kidney Stones and Vertebral Fractures in Primary Hyperparathyroidism Using Imaging Technology

Abstract

Context:

The fourth International Workshop on the Management of Asymptomatic Primary Hyperparathyroidism (PHPT) has recently suggested that skeletal and renal imaging be routinely conducted. So far, no study has systematically assessed this issue.

Objective:

The objective was to evaluate the prevalence of kidney stones (KS) and vertebral fractures (VFs) in a cohort of patients with PHPT utilizing noninvasive imaging technology.

Design:

This was a prospective study evaluating patients consecutively diagnosed with PHPT in a single center over a 5-year period (2009–2013).

Setting:

The setting was a referral center.

Patients:

There were a total of 140 patients with PHPT (127 women [18 premenopausal and 109 postmenopausal] and 13 men; mean age, 63.2 ± 11 y).

Main Outcomes Measures:

Main outcome measures were the prevalence of KS by abdominal ultrasound, osteoporosis by dual-energy x-ray absorptiometry (DXA) (lumbar spine, femoral neck, total hip, and distal 1/3 radius), and VFs by vertebral spine x-ray, with attention to those categorized as symptomatic or asymptomatic.

Results:

Fifty-five percent of all subjects had KS by ultrasound, 62.9% had osteoporosis by T-score at any site, and 35.1% had VFs by x-ray. There was no difference in the incidence of VFs and densitometric osteoporosis between symptomatic and asymptomatic patients (VFs, 34.4 vs 34.7%; osteoporosis by DXA, 59.4 vs 65.8%), whereas more KS were detected in symptomatic (78%) than asymptomatic (35.5%). Twenty-two percent of patients classified as asymptomatic at baseline without osteoporosis by DXA were found to have KS and/or VFs.

Conclusions:

Nephrolithiasis and VFs are common in asymptomatic subjects with PHPT. The results provide evidence in support of recent recommendations that a more proactive approach be taken to detect silent bone and stone disease in asymptomatic PHPT.

The clinical profile of primary hyperparathyroidism (PHPT) has changed markedly over the past several decades. Specific signs and symptoms of the disease previously featured skeletal disorders (osteitis fibrosa cystica, bone cysts, and brown tumors of the long bones), nephrolithiasis, and nephrocalcinosis, as well as neuromuscular symptoms of hypercalcemia. These overt complications are no longer evident in most patients, a point best explained by the routine use of biochemical screening for serum calcium (sCa) in many countries (1). Nevertheless, skeletal involvement can be demonstrated when dual-energy x-ray absorptiometry (DXA) is used. Classically, lower bone mineral density (BMD) at the distal 1/3 radius (Rad), a cortical site, with relative preservation of BMD at the lumbar spine, a trabecular site, is seen (2). Although these data are in line with histomorphometric findings (3), they are not consistent with epidemiological reports demonstrating increased fracture risk at both cortical and trabecular sites (4–7). With reference to vertebral fractures (VFs) in patients with PHPT, prevalence varies widely across different reports (5–10). Our previous work demonstrated a high prevalence of morphometric VFs among PHPT patients, even among asymptomatic subjects (44%) (8).

With reference to renal involvement in PHPT, nephrolithiasis is still its most common complication, but kidney stones (KS) are less prevalent than they used to be (1). However, very few studies have focused on the frequency of KS through imaging technology in asymptomatic subjects (11–13). Retrospective data give a prevalence range of KS between 7 and 11% when abdominal ultrasound is utilized (11, 12). By spiral computed tomography (CT), Starup-Linde et al (13) reported the prevalence of 15.2 and 10.1% for nephrolithiasis and nephrocalcinosis, respectively.

The most recent guidelines on the management of asymptomatic PHPT suggest a more complete evaluation of the skeletal and renal systems, including imaging studies in any patient diagnosed with PHPT (1, 14).

A systematic evaluation of renal and skeletal involvement using imaging of the skeleton and the kidney in patients with PHPT has not yet been conducted. To this end, we evaluated the prevalence of KS and VFs in a cohort of patients consecutively diagnosed with PHPT over a 5-year period to define the clinical picture of the skeletal and renal disease associated with PHPT. Both symptomatic and asymptomatic patients were studied. The results provide evidence for the frequent presence of VFs and KS among patients believed to be asymptomatic. They support the recent guidelines that recommend a proactive approach to the evaluation of the skeleton and kidneys in patients with asymptomatic disease.

Patients and Methods

We studied 140 PHPT patients (127 women [18 premenopausal and 109 postmenopausal] and 13 men; mean age, 63.2 ± 11 y) consecutively referred to our Mineral Metabolism Unit between 2009 and 2013. The diagnosis of PHPT was made by conventional clinical and laboratory criteria, including a history of at least 1 year of hypercalcemia without evidence of a nonparathyroid etiology and elevated or inappropriately normal PTH levels.

After a baseline evaluation, patients were grouped as “symptomatic” or “asymptomatic” based on the presence or absence of at least one the following criteria: 1) symptomatic hypercalcemia, based on the patient's history and the presence of any of the following symptoms not ascribed to other concomitant conditions: polyuria, dehydration, nausea, vomiting, constipation, anorexia, fatigue, and neuromuscular symptoms (myopathy, depression, cognitive dysfunction, lethargy, coma); 2) history of fragility fractures in any of the following sites: vertebra, hip, distal radius, proximal humerus, and pelvis; and 3) history of nephrolithiasis or nephrocalcinosis (at least one episode of renal colic in the last 5 years and/or positive kidney imaging).

Patients had abdominal ultrasound performed by a skilled radiologist. Each ultrasonogram was performed with a low-to-medium frequency (3.5–5 MHz, depending on the physical characteristics of the subject) convex probe and the ultrasound scanner (Esaote MyLab 70 X Vision; Esaote). Ultrasonography was performed in the supine, right and left lateral decubitus positions. The presence, number, and position of stones were evaluated. Renal stones were detected by specific ultrasonographic signs, such as hyperechogeneity and posterior acoustic shadowing.

Standardized radiographs (x-rays) in both anteroposterior and left lateral projections of the thoracic and lumbar spine was obtained in 131 patients. The x-ray was centered at T8 and L3 levels and at a film-focus distance of 115 cm. VFs were defined using Genant's semiquantitative method (15). According to this method, a grade 1 (mild) fracture is a reduction in vertebral height of 25%; grade 2 (moderate), a reduction of 26–40%; and grade 3 (severe), a reduction of >40%. Nonvertebral fractures were recorded by history at the outset of the study. BMD by DXA (QDR-4500; Hologic Inc) was measured at the lumbar spine (L1–L4), femoral neck (FN), total hip (TH), and nondominant distal Rad in all patients. Fractured lumbar vertebrae were excluded from BMD measurement. The coefficients of variation (CVs) were 1.0% at lumbar spine, 1.5% at FN, 1.7% at TH, and 1.3% at the distal Rad (16).

In all patients, the following serum biochemical parameters were measured: total sCa, ionized calcium (Ca²⁺), phosphorus (P), creatinine (Cr), and PTH. Serum 25-hydroxyvitamin D [25(OH)D] levels were assessed in 94 patients. A 24-hour urine was also obtained routinely on all patients.

Serum Ca²⁺ was determined using an ion-specific electrode (Nova 8; Nova Biochemical); serum total 25(OH)D concentrations were measured by RIA (DiaSorin Inc); the intra- and interassay CVs were 8.1 and 10.2%, respectively (17). Serum PTH levels were assessed by immunoradiometric assay (N-tact PTH SP; DiaSorin Inc); the intra- and interassay CVs were 3 and 5.5%, respectively (18).

With the exception of Ca²⁺ measurement, blood samples were drawn and then stored at −70°C until assays were performed in one batch at the end of the study.

All patients gave written, informed consent before their inclusion in the study. The protocol was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments and was approved by the “Sapienza” University of Rome Ethics Committee.

Statistical analysis

Descriptive statistics are expressed as mean ± SD. The percentage of patients with KS, VFs, and osteoporosis at BMD was calculated in the entire cohort and according to their status as symptomatic or asymptomatic. Osteoporosis was defined by T-score ≤2.5 at any of the three skeletal sites. Differences in anthropometric, demographic, biochemical, and densitometric data between the two groups were assessed by unpaired t test. Comparisons between categorical variables were assessed through Fisher's exact and χ² tests. Ordinal variables were tested by using the Jonckheere-Terpstra test. All statistical tests were performed at the two-sided P < .05 significance level. Statistical analysis was performed using SAS version 9.2 (SAS Institute, Inc).

Results

The characteristics of this PHPT cohort are described in Table 1.

Table 1.

Characteristics of the Cohort of PHPT Patients (n = 140)

Characteristics	PHPT	Normal Range
Age, y	63.1 ± 11	—
Weight, kg	66 ± 11	—
Height, cm	161 ± 8	—
BMI, kg/m2	25.5 ± 4	18.5–24.99
Time since menopause, y	16.9 ± 10.5	—
sCa, mg/dL	11.2 ± 0.9	8.4–10.0
Ca2+, mmol/L	1.45 ± 0.13	1.17–1.33
P, mg/dL	2.9 ± 0.6	2.8–4.6
Cr, mg/dL	0.8 ± 0.3	0.7–1.2
PTH, pg/mL	110 ± 95	13–54
25(OH)D, ng/mL	28 ± 15	≥30
uCa, mg/24 h	291 ± 147.8	4 mg × kg/24 h
L1–L4 BMD, g/cm2	0.835 ± 0.150	—
T-score	−2.019 ± 1.302
FN BMD, g/cm2	0.645 ± 0.108	—
T-score	−1.949 ± 0.849
TH BMD, g/cm2	0.762 ± 0.140	—
T-score	−1.527 ± 1.151
Rad BMD, g/cm2	0.571 ± 0.109	—
T-score	−2.122 ± 1.415
Osteoporosis, %a	62.9	—

Abbreviation: uCa, urinary calcium. Data are expressed as mean ± SD. Dashes mean the normal range is not applicable for that specific item.

^aEvaluated by BMD.

Table 2 shows the baseline characteristics that led to the classification of symptomatic PHPT. A history of KS was obtained in 36.4% of all symptomatic patients, whereas a history of fractures was noted in 14.2%. Only 1.4% of patients were symptomatic of hypercalcemia (two patients with polyuria). An overall prevalence of symptoms was 45.7%. The percentages are not additive because some patients were symptomatic in more than one area. With the use of ultrasound and x-rays, the cohort showed a different breakdown, with 55% of all subjects showing evidence of KS (77 of 140); 16.4% of patients had bilateral KS (Table 3), and 35.1% had VFs. The incidence of KS and VFs was then evaluated as a function of the original classification as symptomatic or asymptomatic (Table 3). There was no difference in the incidence of VFs between the symptomatic and asymptomatic cohorts (34.4 vs 34.7%). Although more KS were detected in symptomatic (78%) than asymptomatic (35.5%) patients, nevertheless the incidence of KS among asymptomatic patients is impressive.

Table 2.

Prevalence of the Criteria Used for Classifying PHPT Patients as Symptomatic (as per Baseline Clinical Evaluation) in the Entire Cohort (n = 140)

Criteria	Prevalence, %
History of nephrolithiasis/nephrocalcinosis	36.4
History of VF	5
History of hip fracture	0.7
History of distal radius fracture	7.8
History of proximal humerus fracture	0.7
History of pelvis fracture	0
Hypercalcemia-related symptoms	1.4

Table 3.

Characteristics of PHPT Patients Classified as Symptomatic and Asymptomatic (as per Baseline Clinical Evaluation)

Characteristics	Symptomatic	Asymptomatic	P
n	64	76
Age, y	62.2 ± 11	63.9 ± 12	NS
Weight, kg	68.2 ± 12	64.2 ± 10	<.05a
Height, cm	162.7 ± 9	159.6 ± 7	<.05a
BMI, kg/m2	25.8 ± 4	25.2 ± 4	NS
Time since menopause, y	17.4 ± 11	16.5 ± 10	NS
sCa, mg/dL	11.3 ± 0.9	11 ± 0.8	NS
Ca2+, mmol/L	1.47 ± 0.13	1.43 ± 0.12	NS
P, mg/dL	2.8 ± 0.5	2.9 ± 0.6	NS
Cr, mg/dL	0.8 ± 0.2	0.8 ± 0.3	NS
PTH, pg/mL	115 ± 88	106 ± 100	NS
25(OH)D, ng/mL	26.6 ± 16	29.5 ± 15	NS
uCa, mg/24 h	294.5 ± 138	288.3 ± 157	NS
L1–L4 BMD, g/cm2	0.844 ± 0.146	0.826 ± 0.153	NS
T-score	−1.9 ± 1.3	−2 ± 1.3
FN BMD, g/cm2	0.642 ± 0.114	0.648 ± 0.103	NS
T-score	−1.9 ± 0.9	−1.9 ± 0.7
TH BMD, g/cm2	0.757 ± 0.133	0.766 ± 0.147	NS
T-score	−1.6 ± 1.4	−1.4 ± 0.9
Rad BMD, g/cm2	0.589 ± 0.096	0.557 ± 0.117	NS
T-score	−1.9 ± 1.5	−2.3 ± 1.3
Osteoporosis, %c	59.4	65.8	NS
KS, %	78.1	35.5	<.0001b
VFs, %	34.4	34.7	NS

Abbreviations: NS, nonsignificant; uCa, urinary calcium. Data are expressed as mean ± SD.

^at test.

^bχ² test.

^cEvaluated by BMD.

There was also no difference in the number of VFs between the two groups (data not shown).

Interestingly, 17 of 76 (22.4%) patients classified as asymptomatic at baseline without osteoporosis by DXA were found to have KS and/or VFs.

Osteoporosis by T-score at any site was noted in 62.9% of all study subjects (Table 1). As shown in Table 3, there was no difference is the prevalence of osteoporosis by DXA between the symptomatic (59.4%) and asymptomatic subjects (65.8%).

Among patients with VFs, only 39.1% had osteoporosis by DXA at L1–L4.

Table 4 summarizes the characteristics of individuals who had KS only or had osteoporosis by bone density or fractures. The mean age of patients with only KS (58.9 ± 14.2 y) was significantly younger (P < .05) than that of patients with only fractures or osteoporosis by T-score (65.7 ± 9.5 y). Those two groups also differed by body weight (71.8 ± 12.7 vs 64.1 ± 9.9 kg; P < .01), height (165 ± 9.7 vs 159.5 ± 7.5 cm; P < .02, respectively), and BMD at the three sites (P < .0001 for all). Interestingly, the KS group had essentially normal bone density at all sites. No differences in biochemical parameters were found by comparing patients with only KS and patients with only fractures and/or osteoporosis. Results were confirmed after adjusting for age, weight, and height. Stone formers had significantly higher sCa levels compared to non-stone formers (11.3 ± 1 vs 11.0 ± 0.7 mg/dL, respectively; P < .05), whereas urinary calcium levels did not differ between patients with and without KS (294 ± 146 vs 287.7 ± 152 mg/24 h).

Table 4.

Characteristics of Individuals Who Either Had KS Only or Had Osteoporosis by Bone Density or a Fracture

Parameter	Only KS	Only OPa and/or Fractureb	Pc
n	22	45
Age, y	58.9 ± 14.2	65.7 ± 9.5	<.05
Height, cm	165 ± 9.7	159.5 ± 161.6	<.02
Weight, kg	71.8 ± 12.7	64.1 ± 9.9	P <.02
L1–L4 T-score	−0.8 ± 0.8	−2.6 ± 1	P <.0001
FN T-score	−1.1 ± 0.6	−2.3 ± 0.7	P <.0001
TH T-score	−0.7 ± 0.9	−1.8 ± 0.8	P <.0001
Rad T-score	−0.7 ± 0.8	−2.7 ± 1.1	P <.0001

Abbreviations: OP, osteoporosis; LS, lumbar spine. Data are expressed as mean ± SD.

^aOsteoporosis, defined as T-score <2.5 at any site; evaluated by DXA.

^bVFs and/or non-VFs.

^ct test.

Subjects who fractured differed from those who had not fractured by age (68.1 ± 8 vs 60.8 ± 11 y; P < .0001), years since menopause (19.1 ± 9 vs 15.2 ± 11 y; P < .05), FN T-score (−2.2 ± 0.8 vs −1.8 ± 0.9; P < .03), TH T-score (−1.9 ± 1.3 vs −1.3 ± 2; P < .03), and distal Rad T-score (−2.5 ± 1.2 vs −1.9 ± 1.5; P < .03). No difference was found in L1–L4 T-score and in biochemical parameters between patients who had or had not fractured (data not shown).

Discussion

The asymptomatic form of PHPT is its most common presentation in Western countries, with a prevalence ranging from 72.7 to 95% (9, 19). Geographical differences in this pattern, as well as in the prevalence of the associated complications represent an intriguing area of research (20–22). Our study showed a higher prevalence of symptoms than has recently been reported. The higher prevalence of symptomatic disease could be due to a selection bias because of the nature of our referral center that could have attracted more symptomatic individuals. Patients are indeed frequently referred to our unit for reduced BMD and/or fragility fractures. Nevertheless, one-fourth of the patients presented with no complications at baseline. They were referred only for the findings of high sCa and/or high PTH levels. The point of this study, however, was not to ascertain the incidence of symptomatic vs asymptomatic PHPT, but rather to assess the likelihood that when measures such as ultrasound used to identify KS and vertebral x-rays to identify VFs are employed, a much greater than expected incidence of KS and VFs will be found.

At the Third International Workshop on the Management of Asymptomatic Primary Hyperparathyroidism (23), routine evaluation of KS and VFs was not recommended in the assessment of asymptomatic patients. The Fourth International Workshop has recently revisited this point and has published revised recommendations that include routine abdominal and vertebral imaging in the initial evaluation of patients (1, 14). In this context, our study is the first to provide evidence for this recommendation. We found an unexpectedly high incidence of KS and VFs among both our asymptomatic and symptomatic patients.

Our data showed a 55% overall prevalence of nephrolithiasis in the entire cohort of 140 patients, significantly higher than the few previous studies reporting findings by abdominal ultrasound or CT (11–13). Because this finding actually appears closer to some “historical” reports (21, 24), one could argue that it is explained, at least in part, by the high percentage of symptomatic patients in our cohort. In this context, the observation that the great majority of symptomatic patients had a previous clinical history (within 5 y) of nephrolithiasis (Table 2) is consistent with data reporting it as the most common overt complication of the disease (1). Nevertheless, when the data were analyzed by dividing patients according to the presence or absence of symptoms, KS was also seen in a high percentage of asymptomatic patients (35.5%). The results also should be evaluated with regard to the background prevalence of silent nephrolithiasis in the general population. Retrospective studies have reported a prevalence of incidental KS of 7.8–8.6% in patients who underwent routine abdominal ultrasound or CT (25, 26). In PHPT, recent data report an 11.35% prevalence of silent KS (11). The higher prevalence in our cohort could be partially explained by the different experimental design of the two studies (retrospective vs observational), as well as by the lack of stratification for metabolic risk factors for nephrolithiasis other than those typically related to PHPT, such as serum and urinary PH, anion gap, and citraturia, in both cohorts. In this context, we investigated biochemical features that could have predisposed this population to a higher risk of KS. Only sCa showed slightly higher levels in patients with KS compared to non-stone formers. This finding is in accordance with the study of Cassibba et al (11) and appears to confirm the idea that more active parathyroid disease is associated with greater likelihood of stone formation. Mean urinary calcium levels, however, did not show any significant difference between patients with and without KS. Mechanisms beyond the urinary calcium load must account for nephrolithiasis in PHPT. The risk of nephrolithiasis has been associated with other biochemical and local urinary factors in PHPT (27). Moreover, genetic studies describe an association between KS formation and calcium-sensing receptor gene polymorphisms in both normocalcemic and PHPT patients. These putative genetic factors remain to be elucidated (28, 29). Scillitani et al (29) reported a significant association between calcium-sensing receptor AGQ haplotype and KS in PHPT patients independent of hypercalciuria. Our results are limited by the single 24-hour urine calcium value that was obtained per subject. Vitamin D insufficiency could limit the calcium absorption in some patients with stones.

Kidney imaging, either by x-ray, ultrasound, or CT, is now recommended in the clinical assessment of any PHPT patients to look for nephrolithiasis. Notwithstanding the lower sensitivity of ultrasound compared to CT, recent data suggest that ultrasonography should be used as an initial imaging test in patients at risk for nephrolithiasis, particularly considering the lower amount of radiation patients are exposed to, compared to CT (30). The need for further abdominal imaging is a matter of professional judgment (30).

Data on skeletal complications in our cohort showed the presence of osteoporosis by DXA in most of the patients, as well as a 35.1% prevalence of morphometric VFs. The data further confirm the critical involvement of the trabecular compartment in the bone disease associated with PHPT (5, 8, 9). Interestingly, only a minority of patients with VFs had osteoporosis by T-score at lumbar spine, meaning that factors other than reduced areal BMD are likely to account for VFs in PHPT. This is further supported by the absence of any difference in lumbar spine T-score between those who did or did not fracture. The finding is in line with recent reports demonstrating the presence of microarchitectural trabecular abnormalities captured by high-resolution technology and trabecular bone score in PHPT (10, 31).

Our data link fracture risk in PHPT to well-known risk factors such as age and time since menopause (5, 9, 10, 32). These factors were not associated with KS incidence.

Consistent with previous reports (11, 21), our results did not reveal any differences in the metabolic profile of patients with nephrolithiasis vs patients with skeletal complications. Because patients with nephrolithiasis were found to be younger than patients with the skeletal complications, different mechanisms accounting for the pathogenesis of such complications could be hypothesized and reflect the possible presence of a different clinical profile of PHPT patients. Normal BMD among stone formers supports this possibility.

Although not explicitly stated in guidelines, the standard of care after parathyroidectomy is to monitor BMD every 1–2 years. With regard to KS, because urinary calcium is not likely to be elevated in those with KS, there is no standard of care to measure urinary calcium after parathyroidectomy. Moreover, most patients who were stone formers before parathyroidectomy do not form stones after parathyroidectomy (33, 34).

Our study has some limitations. First, patients were enrolled from a specialized center, thus implying a possible selection bias. Second, we did not have access to control groups of subjects with KS or VFs without PHPT. Thus, we cannot be certain of factors that might be specific to PHPT insofar as these complications are concerned. Finally, although we did not evaluate patients based upon their vitamin D levels, we did ascertain that there was no difference in 25(OH)D levels when stone formers were compared to non-stone formers.

To our knowledge, this is the first study that reports the prevalence of target organ complications among so-called “asymptomatic patients” when noninvasive imaging modalities are used routinely. The results demonstrating that nephrolithiasis and VFs are readily detected in asymptomatic subjects with PHPT support the recent recommendations of the Fourth International Workshop on the Management of Asymptomatic PHPT, namely, to be more proactive with regard to detecting silent bone and stone disease (1, 14). Our results emphasize the need for a thorough evaluation of the kidneys and the skeleton in asymptomatic patients, before classifying them as asymptomatic (1, 14). With the use of imaging technologies, many asymptomatic patients will be found to have significant skeletal and renal disease and, because KS and VFs are included among the 2014 recommendations for surgery in asymptomatic patients (14), this point should be carefully considered in future research focused on the costs of a disease that, when successfully managed from a surgical point of view, can be permanently cured.