hyperparathyroidism: surgery decreases diabetes risk

 A recent article found that patients with primary hyperparathyroidism have reduced type II diabetes if they had parathyroidectomy over medical therapy, in this retrospective cohort study from Israel (see hyperparathyroid surgery dec diabetes in dropbox, or doi:10.1001/jamanetworkopen.2025.46222)

 

Details:

-- 50,199 patients who were diagnosed with primary hyperparathyroidism (pHPT) from 2000-2024 in the Clalit Health Services database, a nationwide databank of Israel’s largest healthcare organizations. These patients were compared to 150,265 controls who were matched for age, sex, and socioeconomic status. Those having secondary hyperparathyroidism were excluded

    -- the Clalit database includes information on diagnoses, hospitalizations, clinical visits, demographic characteristics, laboratory values, and medications. this database has a remarkably low patient turnover rate (1%), allowing for long-term follow-up

-- 6654 of these patients underwent parathyroidectomy, with follow-up as long as 15 years

-- mean age 64, 68% female, 90% Jewish/11% Arab

-- socioeconomic status low in 14%/medium in 63%/high in 23%

-- Charleston Comorbidity Index (a measure of medical co-morbidities, the higher number the worse): zero in 11%, 1-2 in 30%, at least 3 in 59%

-- baseline hypertension in 45% of the controls and 59% of those with pHPT, type II diabetes in 12% of controls and 18% of those with pHPT, coronary vascular disease in 17% of controls and 24% of those with pHPT, stroke in 8% of controls and 11% of those with pHPT

-- smoking 15%, BMI mean 28, calcium level mean 9.5 mg/dL, phosphorus level 3.5 mg/dL, PTH level 58 pg/mL in the control group and 145 in the pHPT group, urinary calcium level 127 mg/24h in the control group and 173 mg/24h in the pHPT group

  

-- Main outcomes: incident cases of of hypertension, type II diabetes (T2D), cardiovascular disease (CVD, including ischemic heart disease, myocardial infarction, and heart failure ), and strokes comparing those with pHPT vs controls, and also those with pHPT managed by parathyroidectomy vs nonsurgical management

    -- results were adjusted for Charleston Comorbidity Index, BMI, and ethnicity

  

Results:

-- comparing the patients in the pHPT versus matched controls, incident rate per 1000 person-years of follow-up, as compared to the control group, in the adjusted analysis:

    -- hypertension:

          -- pHTP: 22% higher, adjusted HR 1.22 (1.17-1.33), p<0.001

    -- T2D:

          -- pHTP: 7% higher, adjusted HR 1.07 (1.01-1.16), p=0.02

    -- CVD:

          -- pHTP: 28% higher, adjusted HR 1.28 (1.21-1.40), p<0.001

    -- stroke:

          -- pHTP: 22% higher, adjusted HR 1.22 (1.17-1.33), p<0.001

 

-- nonsurgical treatment  (n= 19,962 patients) versus parathyroidectomy (n= 6654 patients):

    -- baseline characteristics were well-balanced between these groups

    -- mean time from diagnosis to surgery was 879 days in those undergoing parathyroidectomies

-- outcomes after parathyroidectomy versus nonsurgical management, incidents per 1000 person-years:

    -- hypertension:

        -- 45.89 (42.99-49.00) versus 48.85 (47.32-50.43)

           -- adjusted HR: 1.19 (0.69-1.96), p=0.33, not statistically significant

    -- T2D:

        -- 10.77 (9.80 -11.83) versus 15.18 (14.58-15.81)

           -- adjusted HR: 0.56 (0.30-0.89), p=0.02, highly statistically significant decrease of 44% in those having surgery

    --CVD:

        -- 11.57 10.57-12.66) versus 12.49 (11.94 -13.07)

           -- adjusted HR: 0.93 (0.85 -1.02), p=0.92, not statistically significant

    -- stroke:

        -- 12.93 (11.92 -14.03) versus 14.12 (13.55-14.70)

           -- adjusted HR: 0.79 (0.50- 1.13), p=0.09, not statistically significant (but strong trend towards being significant

 

Commentary:

-- primary hyperparathyroidism (pHPT) is the 3rd most common endocrine disorder worldwide, typically associated with hypercalcemia and a range of systemic complications including high cardiovascular risk, hypertension, arrhythmias, and vascular calcification

-- studies have found that prolonged parathyroid (PTH) elevations are associated with vascular endothelial dysfunction and increased arterial stiffness (likely from the hypercalcemia affecting vascular smooth muscle function), insulin resistance, and left ventricular hypertrophy. these all are likely the cause for the increase in hypertension, T2D, and atherosclerotic events

   -- therefore, as expected, those patients in this study with pHPT did have more of these comorbidities at baseline than those in the control group,

   -- there have been many unknowns clinically in terms of the role of parathyroidectomy, though it does normalize PTH and calcium levels and can correct metabolic abnormalities. The data are less clear for improving cardiovascular risk, with variable results in prior studies.

   -- similar to the current study, one large study in Hong Kong with 3135 patients also found that surgical intervention was associated with 32% decreased risk of incident diabetes, and this was more evident in younger people and in those with more severe parathyroid hormone and calcium elevations: https://jamanetwork.com/journals/jamasurgery/fullarticle/2838017

 

-- this current study provided evidence that hyperparathyroidism itself was associated with significant increases in hypertension, type II diabetes, cardiovascular disease, and strokes.

-- in addition, it found that of the 6654 patients who had a parathyroidectomy, surgery was associated with significantly reduced incidence of type II diabetes but without clear benefits for hypertension, CVD, or stroke

-- which all means that hyperparathyroidism is not simply a disease of calcium and phosphate perversions, but does extend to important cardiometabolic concerns

    -- ie, those with pHPT who do not have surgical parathyroidectomy seemed to have an increased risk of type II diabetes  

-- which strongly suggests that all patients with pHPT should be monitored for incident T2D as well as kidney stones (and encouraged to have intake of large amounts of water) and osteoporosis (and having regular bone density determinations)

-- one question is why was there no effect of parathyroidectomy on hypertension or cardiovascular disease . This is unclear from the study but might be the result of long-term exposure to the high levels of PTH and hypercalcemia leading to significant and perhaps nonreversible vascular disease. In addition, in this study there was any 879-day lag between hyperparathyroidism diagnosis and surgery. To the extent that long-term exposure to high PTH levels is the culprit, this would argue strongly that the surgical intervention should be done as soon as possible after diagnosis

-- another question is what should the screening be for people with hyperparathyroidism

    -- is there value in routine cardiovascular valuations? Should we be checking coronary artery calcium scores, since there may well be increased vascular calcification (not found in all studies, but was found in https://www.surgjournal.com/article/S0039-6060(19)30612-9/abstract )? Coronary CT angiography? There was an almost significant decrease and cerebrovascular events. Should that be assessed routinely? What would be the best way? And, of course, we need clear data that shows that these tests matter, and that parathyroidectomy helps

 

-- medical therapy in patients not either desiring or appropriate for surgical intervention is somewhat complex, but includes the following considerations:

    -- correction of the hypercalcemia: stopping medications that increase calcium such as lithium or thiazide diuretics (and perhaps including low-dose torsemide starting at 5 mg, which provides the same hypertension benefits as thiazides but actually leads to increased calcium excretion and somewhat lower serum calcium levels)

    -- in patients with severe hypercalcemia, cinacalcet can decrease serum calcium levels by activating the calcium censoring receptor in the parathyroid gland and decreasing PTH secretion, typically lowering calcium in the range of 1 to 2 mg/dL and parathyroid hormone by about 30 pg/mL, though there is not much difference in bone density or 24-hour urinary calcium-to-creatinine ratio (though there was a study which showed much more significant lowering of the PTH level)

    -- osteoporosis: increases of bone mineral density with bisphosphonates are about the same as with parathyroidectomy, though BMDs should be repeated regularly to make sure that bone density does not decrease too much in those not having surgery. Little data on fracture outcome with alendronate in hyperparathyroidism, and in fact though there are increases in bone density at the hip and lumbar spine, there is not much change at the radius. denosumab is an alternative patients who do not tolerate bisphosphonates, though serum calcium levels are unaffected (alendronate can lead to a small and often temporary decreases in calcium levels)

        -- at least in rats (no human studies i have found), bisphosphonates do seem to decrease calcium levels some and decrease arterial calcification (https://www.ahajournals.org/doi/10.1161/01.atv.21.5.817)

    -- raloxifene can also decrease calcium modestly, but long-term outcomes in larger numbers of patients are required

    -- in terms of calcium and vitamin D, patient should have their vitamin D level checked and probably should have levels between 30-50 ng/mL (which may reduce the PTH level in those with low 25(OHD levels)); in patients with higher levels of urinary calcium excretion, vitamin D level should be at a lower level to avoid kidney stones. Calcium should still be taken since a low calcium diet can increase PTH secretion and aggravate bone disease (this can be complex: since for a high serum calcitriol level, a higher calcium intake can exacerbate hypercalcemia and hypercalciuria, so it may be reasonable to restrict calcium to <800 mg/day and check on its effects on serum/urine calcium). 

 

Limitations:

-- the study was based on a retrospective cohort being followed in the largest Israeli healthcare database, but this raises issues of the accuracy of the coding within the database and the generalizability of results to other patient groups with different diets and other baseline medical and social conditions

-- as a retrospective cohort study, we cannot eliminate the possible possibility of residual confounding, which renders the results as an association and not clear causality.

    -- for example, there may well be important medical or social conditions that were not assessed. And the medical conditions in this study were not stratified by degree of their potential impact (e.g. were there levels of baseline blood pressure etc that affected the results differently??). And were there important differences between patients who elected surgery vs not; a difference in patients would distort results (hence the need for a randomized controlled trial)

    -- and, given the quite high prevalence of cardiovascular disease in the population in the absence of hyperparathyroidism, there might be a significant effect on the results for this outcome independent of hyperparathyroidism. we do not have any information of how the nonsurgical patients were treated (eg,statins?), or what their response to the treatment was in terms of PTH levels, calcium, and phosphate responses. We do not know these items even for those who had surgical intervention

-- and, there was no consistency evident in the nonsurgical management of hyperparathyroidism. were they taking diets/meds that would change their calcium or vitamin D levels during the study?? how did these values change during the followup? Unfortunately, unlike hyperaldosteronism, we do not have a targeted medication such as we have for hyperaldosteronism, where spironolactone prevents the binding of aldosterone to its receptors and thereby limits aldosterone effect (ie pHTP does not have a specific medication to decrease its metabolic effects)

 

So, this study does bring up a few issues:

-- there does seem to be an advantage of parathyroid surgery, especially early in the disease in younger people as well as in those with more biochemical changes from severe hyperparathyroidism

-- this study did not truly evaluate nonsurgical therapy vs surgical therapy (we need an RCT for that), especially since we have no information on what the nonsurgical therapy was nor the biological outcomes of that therapy. but this study in conjunction with the other study noted above does raise more profoundly the potential importance of surgical intervention especially in decreasing the incidence of type II diabetes. 

-- and it raises the issue of whether we should monitor people with hyperparathyroidism for cardiovascular disease as well as the usual monitoring we do already for kidney stones and osteoporosis. We do need studies to suggest which of the cardiovascular monitoring studies makes sense if any do

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them