Mild primary hyperparathyroidism: a misnomer?

Keywords::

• calcium
• cancer
• cardiovascular disease
• fracture
• parathyroid

In primary hyperparathyroidism (PHPT) treatment is traditionally offered to symptomatic patients, including those with proven end-organ damage (osteoporosis, renal failure or renal stones), individuals with severe hypercalcemia (serum calcium >2.85 mmol/l), and/or patients under 50 years of age, who in essence are considered to be at high risk of developing complications. The vast majority of patients (>80% [1]) do not fit these criteria, often being referred to as having ‘mild’ PHPT, and as such are simply monitored for the development of complications, while receiving no specific treatment.

Progression of hypercalcemia

How many of these patients progress towards more severe hypercalcemia? In the Columbia University (NY, USA) natural history study of PHPT, approximately 50% of 121 patients followed up did not initially undergo parathyroidectomy. In the group not undergoing surgery, 27% had ‘progressed’ to surgery by 10 years, and 37% by 15 years [1,2], although the actual percentage developing progressive hypercalcemia was not stated. Furthermore, the number of patients with data at follow-up was small, and some of the original ‘non-operated’ cohort had met contemporary surgical criteria at baseline.

The Parathyroid Epidemiology and Audit Research Study (PEARS) from Tayside, Scotland (UK) identified individuals with ‘mild’ PHPT in whom the serum calcium increased by 0.2 mmol/l or reached a concentration of 2.9 mmol/l. Although over 10 years 13% fulfilled these criteria, it was only sustained in 2% of patients for over 6 months [3]. This is similar to the 1% of patients progressing to meet NIH biochemical criteria for hypercalcemia in a Scandinavian registry study [4]. These last two studies investigated biochemical progression, while the Columbia University study looked at a combined end point of biochemistry, symptoms and outcomes, without differentiating between them. Another difference between the US and the Scottish and Swedish studies is that serum vitamin D concentrations may be lower in the North European studies, thereby blunting progression of hypercalcemia. Overall, the proportion of patients who progress seems to be low.

Interestingly, meeting NIH surgical criteria at baseline did not predict clinical or biochemical progression in the Columbia University study [2]. It thus seems sensible to consider what the meaningful clinical outcomes are for patients with PHPT, and how they relate to baseline indices – if at all.

Bone disease

The main concern for bone health in PHPT relates to low bone mineral density (BMD) and fracture risk. Although bone density may be stable for approximately 8 years, by 10–15 years of follow up BMD declined by 10 and 35% in the femoral neck and distal radius, respectively, in the Columbia University study [2]. By contrast, sustained improvements in BMD were seen at all three sites examined in the cohort undergoing parathyroidectomy. Furthermore, data from randomized controlled trials have shown improvements in BMD as early as 1–2 years after parathyroidectomy in ‘mild’ parathyroidectomy at one or more sites [5].

Using record linkage of hospital admission data, the PEARS study found an increase in (admissions with) fractures in ‘mild’ PHPT [6], and there was a nonsignificant trend towards reduced fracture rates in 200 patients who had undergone parathyroidectomy. A previous cohort study demonstrated a reduction in fracture rates of approximately 30% post parathyroidectomy after adjusting for age, sex and history of previous fracture [7]. Although this was not specifically a cohort of ‘mild’ PHPT patients, the effect on fracture risk remained in those without a prior fracture, possibly representing those with milder disease.

Renal disease

Primary hyperparathyroidism is associated with renal stones and renal failure. Parathyroidectomy seems to be effective at preventing renal stone recurrence. In the Columbia University study, 20 patients with PHPT had renal stones at baseline and were followed up for 10 years [1]. Of the 12 patients undergoing parathyroidectomy, none developed further stones, while six of eight patients not undergoing parathyroidectomy had further renal stones. The PEARS study showed a decreased rate of renal stone formation following parathyroidectomy in 200 patients undergoing parathyroidectomy for all forms of PHPT (although mainly in those meeting NIH criteria for parathyroidectomy) [3]. In addition, the PEARS study demonstrated a fivefold increased risk of renal stone formation in patients not meeting the NIH criteria for surgery – so-called ‘mild’ PHPT [6]. There were no data to examine the impact of parathyroidectomy in such patients, but a significant reduction in the proportion of patients developing chronic renal failure after parathyroidectomy was demonstrated [3].

Cardiovascular disease

Primary hyperparathyroidism appears to have many overlapping features with the so-called metabolic syndrome, being associated with hypertension, dyslipidemia, endothelial dysfunction, insulin resistance and diabetes [6,8–10]. There is good evidence from European studies supporting this link with cardiovascular morbidity and mortality in more severe PHPT/hypercalcemia [9,11], although this finding was not replicated in a US study from Rochester, MN, USA [12].

Numerous studies examining surrogate markers of cardiovascular disease or cardiovascular risk in ‘mild’ PHPT have produced conflicting results, perhaps reflecting a lack of adequately powered studies, or patients being studied at differing stages of the disease process. For example, while some studies demonstrate a positive effect of parathyroidectomy on left ventricular hypertrophy [13], others have found no improvement in left ventricular hypertrophy [14] or endothelial function [15].

The PEARS study reported increased standardized morbidity and mortality ratios for both cardiovascular and cerebrovascular disease from a cohort of 1683 patients meeting the criteria for ‘mild’ PHPT. After adjusting for potential confounders, an increase in all-cause mortality, fatal (2.2-fold) and non-fatal (3.6-fold) cardiovascular events was observed in individuals with ‘mild’ PHPT [6].

However, it is not clear whether parathyroidectomy leads to an improvement in cardiovascular risk factors. Parathyroidectomy is thought to have no influence on hypertension [16], although a recent study demonstrated a small improvement in blood pressure and triglycerides in mild PHPT [17]. Previous studies, although underpowered, demonstrated no benefit of parathyroidectomy on cardiovascular mortality [4], which may reflect the lack of improvement in risk factors. The EVOLVE trial has been established to see if cinacalcet reduces cardiovascular events in patients receiving dialysis who have secondary tertiary hyperparathyroidism [18].

Cancer

More recently, studies have shown a link between PHPT and cancer. The PEARS study indicated a 2.3-fold increased risk of death from any cancer, and a 2.1-fold increased risk of developing cancer in a population-based study of 1424 patients with PHPT not achieving NIH criteria for surgery [6]. In a larger population-based study of all patients with PHPT (n = 9782), the standard incidence ratio of any cancer was 1.4 compared with the general population [19]. There was a notable increase in cases of breast, colon, renal and cutaneous squamous cell carcinoma. However, the rate of cancer was similar 15 years after parathyroidectomy, indicative of causal disassociation, suggesting a common etiological factor rather than a direct effect of PHPT. Intriguingly, the authors suggested either a common genetic predisposition or an acquired inability to withstand environmental influences.

Conclusion

Observational data tend to suggest an increased risk of fractures, renal stones, renal failure, cardiovascular disease and cancer at all stages of PHPT. There is an improvement in BMD and probably renal stones after parathyroidectomy at all biochemical stages of PHPT, but with no evidence of benefit from parathyroidectomy for cardiovascular disease or cancer. Thus, the risks associated with milder and more severe forms of PHPT appear to be similar. So-called ‘mild’ PHPT appears to be an insidious disease, possibly with fewer symptoms but with similar long-term consequences. The analogy with diabetes is very striking, and the term ‘mild’ diabetes has long-since disappeared. Perhaps now is the time to abandon the term ‘mild’ PHPT and call it ‘insidious’ PHPT, or even drop the prefix altogether. However, what should be done for patients with PHPT who do not reach the NIH criteria for surgery remains unclear. There is no good evidence that intervening in such patients is beneficial. A randomized controlled trial is required to answer this question, using either surgery or a medical intervention such as cinacalcet as the active intervention. However, even if this did show benefit, other interventions may be equally efficacious, such as bisphosphonate treatment and/or aggressive cardiovascular risk factor management. Questions remain about the management of PHPT, especially if outcomes such as the association with cancer, are actually not causal. There is circumstantial evidence that links with cardiovascular end points may also be in causal disassociation, as parathyroidectomy does not reverse hypertension [16] or insulin resistance [20,21]. As a result we should be cautious about the possible effect of parathyroidectomy on cardiovascular end points, but this further increases the need for a randomized controlled trial.

Financial & competing interests disclosure

Graham Leese has had a research grant from Amgen. The authors would like to acknowledge the financial support of the Chief Scientist Office, Scotland, UK. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.