Advertisement

Parathyroidectomy for asymptomatic primary hyperparathyroidism: A revised cost-effectiveness analysis incorporating fracture risk reduction

Published:November 08, 2016DOI:https://doi.org/10.1016/j.surg.2016.06.062

      Background

      Recent data demonstrate decreased fracture risk after operation for asymptomatic primary hyperparathyroidism. We performed a revised cost-effectiveness analysis comparing parathyroidectomy versus observation while incorporating fracture risk reduction.

      Methods

      A Markov transition-state model was created comparing parathyroidectomy and guideline-based medical observation for a 60-year-old female patient with mild asymptomatic primary hyperparathyroidism. Costs were estimated using published Medicare reimbursement data. Treatment strategy outcomes, including risk of fracture, were identified by literature review. Quality adjustment factors were used to weight treatment outcomes. A threshold of $100,000/quality-adjusted life year was used to determine cost-effectiveness. Sensitivity analyses and Monte Carlo simulation were performed to examine the effect of uncertainty on the model.

      Results

      Parathyroidectomy was the dominant strategy (less costly and more effective) with an incremental cost savings of $1,721 and an incremental effectiveness of 0.185 quality-adjusted life years. Parathyroidectomy remained dominant when the relative risk reduction of fracture after operation was ≥14%, the cost of fracture was ≥$7,600, or the probability of recurrent laryngeal nerve injury was <12.5%. Monte Carlo simulation showed parathyroidectomy was cost-effective in 995/1,000 hypothetical patients.

      Conclusion

      When fracture risk reduction is considered, parathyroidectomy for mild asymptomatic primary hyperparathyroidism is the dominant strategy when compared to observation.
      Primary hyperparathyroidism (PHPT) is the most common cause of hypercalcemia and occurs in >135,000 patients in the United States each year.
      • Yeh M.W.
      • Ituarte P.H.
      • Zhou H.C.
      • Nishimoto S.
      • Liu I.L.
      • Harari A.
      • et al.
      Incidence and prevalence of primary hyperparathyroidism in a racially mixed population.
      Untreated PHPT is associated with bone loss and fracture.
      • Lundstam K.
      • Heck A.
      • Mollerup C.
      • Godang K.
      • Baranowski M.
      • Pernow Y.
      • et al.
      Effects of parathyroidectomy versus observation on the development of vertebral fractures in mild primary hyperparathyroidism.
      • Rubin M.R.
      • Bilezikian J.P.
      • McMahon D.J.
      • Jacobs T.
      • Shane E.
      • Siris E.
      • et al.
      The natural history of primary hyperparathyroidism with or without parathyroid surgery after 15 years.
      • Silverberg S.J.
      • Shane E.
      • Jacobs T.P.
      • Siris E.
      • Bilezikian J.P.
      A 10-year prospective study of primary hyperparathyroidism with or without parathyroid surgery.
      • Khosla S.
      • Melton 3rd, L.J.
      • Wermers R.A.
      • Crowson C.S.
      • O'Fallon W.
      • Riggs B.
      Primary hyperparathyroidism and the risk of fracture: a population-based study.
      • Vestergaard P.
      • Mosekilde L.
      Cohort study on effects of parathyroid surgery on multiple outcomes in primary hyperparathyroidism.
      While parathyroidectomy (PTX) is the only definitive therapy for PHPT, current consensus guidelines continue to identify a subset of asymptomatic PHPT patients with mild biochemical disease and age ≥50 years who are eligible for medical observation as an alternative to operation.
      • Bilezikian J.P.
      • Brandi M.L.
      • Eastell R.
      • Silverberg S.J.
      • Udelsman R.
      • Marcocci C.
      • et al.
      Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
      Several recent prospective and retrospective, population-based studies report a 24–31% relative risk reduction and a 4.8–11.34% absolute, 10-year risk reduction of fracture after PTX for PHPT.
      • Lundstam K.
      • Heck A.
      • Mollerup C.
      • Godang K.
      • Baranowski M.
      • Pernow Y.
      • et al.
      Effects of parathyroidectomy versus observation on the development of vertebral fractures in mild primary hyperparathyroidism.
      • Khosla S.
      • Melton 3rd, L.J.
      • Wermers R.A.
      • Crowson C.S.
      • O'Fallon W.
      • Riggs B.
      Primary hyperparathyroidism and the risk of fracture: a population-based study.
      • Vestergaard P.
      • Mosekilde L.
      Cohort study on effects of parathyroid surgery on multiple outcomes in primary hyperparathyroidism.
      • Yeh M.W.
      • Zhou H.
      • Adams A.L.
      • Ituarte P.H.
      • Li N.
      • Liu I.A.
      • et al.
      The relationship of parathyroidectomy and bisphosphonates with fracture risk in primary hyperparathyroidism: an observational study.
      These data need to be incorporated into the clinical decision-making process for offering an operation versus observation to patients with mild asymptomatic disease. To determine the optimal management strategy for these patients, the potential benefit of future fracture avoidance needs to be weighed against the upfront risks and costs of PTX.
      Previous studies have investigated the cost-effectiveness of PTX for observation-eligible patients with mild, asymptomatic PHPT
      • Zanocco K.
      • Angelos P.
      • Sturgeon C.
      Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sejean K.
      • Calmus S.
      • Durand-Zaleski I.
      • Bonnichon P.
      • Thomopoulos P.
      • Cormier C.
      • et al.
      Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
      ; however, no model has addressed the risk, quality-of-life (QOL) detriment, and cost of fragility fracture during medical observation. We incorporated these outcomes in a revised cost-effectiveness analysis comparing PTX versus observation in asymptomatic PHPT. We hypothesized that PTX is cost-effective for all patients with observation-eligible PHPT.

      Methods

      Reference case scenario

      The reference case was defined as a 60-year-old female patient with asymptomatic PHPT meeting current consensus guideline criteria for medical observation.
      • Bilezikian J.P.
      • Brandi M.L.
      • Eastell R.
      • Silverberg S.J.
      • Udelsman R.
      • Marcocci C.
      • et al.
      Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
      The age of 60 years was selected to maintain consistency with the previous model and falls within the reported average age range of observation-eligible PHPT patients.
      • Silverberg S.J.
      • Shane E.
      • Jacobs T.P.
      • Siris E.
      • Bilezikian J.P.
      A 10-year prospective study of primary hyperparathyroidism with or without parathyroid surgery.
      • Rao D.S.
      • Phillips E.R.
      • Divine G.W.
      • Talpos G.B.
      Randomized controlled clinical trial of surgery versus no surgery in patients with mild asymptomatic primary hyperparathyroidism.
      This reference patient was a healthy operative candidate for PTX via a cervical incision. The time horizon for the analysis was the patient's remaining life expectancy, as predicted by the current US Social Security System Actuarial Life Table.

      Decision model

      A previously constructed Markov stochastic cohort transition-state decision model comparing medical observation and PTX was revised to incorporate current cost and outcome data, including fracture risk, for the reference case.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      The cycle duration was 6 months. All modeling was performed with TreeAge Pro decision analysis software (TreeAge Software, Inc, Williamstown, MA). The major transition states for the medical observation and PTX strategies are shown in Fig 1. The event pathways and probabilities used in the model (Table I) were derived from literature review and current consensus conference recommendations for the medical and surgical management of asymptomatic PHPT.
      • Bilezikian J.P.
      • Brandi M.L.
      • Eastell R.
      • Silverberg S.J.
      • Udelsman R.
      • Marcocci C.
      • et al.
      Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
      Figure thumbnail gr1
      Fig 1Schematic of the major transition states in the Markov model.
      Table IModel assumptions
      VariableReference case valueRange of values used in Monte Carlo simulation distributionsSources
      Probabilities
       Annual risk of fracture in age-matched patient without PHPT3%1.5–4.5%
      • Khosla S.
      • Melton 3rd, L.J.
      • Wermers R.A.
      • Crowson C.S.
      • O'Fallon W.
      • Riggs B.
      Primary hyperparathyroidism and the risk of fracture: a population-based study.
      • Melton 3rd, L.J.
      • Crowson C.S.
      • O'Fallon W.M.
      Fracture incidence in Olmsted County, Minnesota: comparison of urban with rural rates and changes in urban rates over time.
       Annual risk of progression from asymptomatic PHPT to symptomatic PHPT1.6%0.8–2.4%
      • Rubin M.R.
      • Bilezikian J.P.
      • McMahon D.J.
      • Jacobs T.
      • Shane E.
      • Siris E.
      • et al.
      The natural history of primary hyperparathyroidism with or without parathyroid surgery after 15 years.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
       Annual risk of recurrence after curative PTX0.05%0.025–0.075%
      • Carneiro-Pla D.M.
      • Solorzano C.C.
      • Lew J.I.
      • Irvin 3rd, G.L.
      Long-term outcome of patients with intraoperative parathyroid level remaining above the normal range during parathyroidectomy.
      • Udelsman R.
      • Lin Z.
      • Donovan P.
      The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism.
       RLN injury after initial PTX0.5%0.25–0.75%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Udelsman R.
      • Lin Z.
      • Donovan P.
      The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism.
       RLN injury after reoperative PTX4%2–6%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Karakas E.
      • Muller H.H.
      • Schlosshauer T.
      • Rothmund M.
      • Bartsch D.K.
      Reoperations for primary hyperparathyroidism–improvement of outcome over two decades.
       Permanent hypoparathyroidism after initial PTX0.5%0.25–0.75%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Udelsman R.
      • Lin Z.
      • Donovan P.
      The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism.
       Permanent hypoparathyroidism after reoperative PTX1%0.5–1.5%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Karakas E.
      • Muller H.H.
      • Schlosshauer T.
      • Rothmund M.
      • Bartsch D.K.
      Reoperations for primary hyperparathyroidism–improvement of outcome over two decades.
       Persistent PHPT after initial PTX5%2.5–7.5%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
       Persistent PHPT after reoperative PTX10%5–15%
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Karakas E.
      • Muller H.H.
      • Schlosshauer T.
      • Rothmund M.
      • Bartsch D.K.
      Reoperations for primary hyperparathyroidism–improvement of outcome over two decades.
      Costs in US dollars
       Initial parathyroidectomy$5,702$2,851–8,553
       Reoperative parathyroidectomy$6,114$3,057–9,171
       RLN injury treatment$11,846$5,923–17,769
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
       Cost of fracture$16,281$8,140.50–24,421.50
      • Christensen L.
      • Iqbal S.
      • Macarios D.
      • Badamgarav E.
      • Harley C.
      Cost of fractures commonly associated with osteoporosis in a managed-care population.
       Annual cost of permanent hypoparathyroidism$876$438–1,314
       Annual cost to observe asymptomatic PHPT$278$139–417
      • Bilezikian J.P.
      • Brandi M.L.
      • Eastell R.
      • Silverberg S.J.
      • Udelsman R.
      • Marcocci C.
      • et al.
      Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
       Annual cost of calcimimetic therapy$20,192$10,096–30,288
      QOL adjustment factors
       Curative PTX1.001.00–1
      • Zanocco K.
      • Angelos P.
      • Sturgeon C.
      Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
       Asymptomatic hyperparathyroidism0.9870.97–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Burney R.E.
      • Jones K.R.
      • Christy B.
      • Thompson N.W.
      Health status improvement after surgical correction of primary hyperparathyroidism in patients with high and low preoperative calcium levels.
       Symptomatic hyperparathyroidism0.8970.79–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Burney R.E.
      • Jones K.R.
      • Christy B.
      • Thompson N.W.
      Health status improvement after surgical correction of primary hyperparathyroidism in patients with high and low preoperative calcium levels.
       Long-term hypoparathyroidism0.8940.79–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Vidal-Trecan G.M.
      • Stahl J.E.
      • Eckman M.H.
      Radioiodine or surgery for toxic thyroid adenoma: dissecting an important decision. A cost-effectiveness analysis.
       Curative PTX with RLN injury0.8910.78–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sejean K.
      • Calmus S.
      • Durand-Zaleski I.
      • Bonnichon P.
      • Thomopoulos P.
      • Cormier C.
      • et al.
      Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Spector B.C.
      • Netterville J.L.
      • Billante C.
      • Clary J.
      • Reinisch L.
      • Smith T.L.
      Quality-of-life assessment in patients with unilateral vocal cord paralysis.
       Asymptomatic PHPT with RLN damage0.8780.76–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Burney R.E.
      • Jones K.R.
      • Christy B.
      • Thompson N.W.
      Health status improvement after surgical correction of primary hyperparathyroidism in patients with high and low preoperative calcium levels.
      • Spector B.C.
      • Netterville J.L.
      • Billante C.
      • Clary J.
      • Reinisch L.
      • Smith T.L.
      Quality-of-life assessment in patients with unilateral vocal cord paralysis.
       Symptomatic PHPT and permanent RLN damage0.8770.75–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Burney R.E.
      • Jones K.R.
      • Christy B.
      • Thompson N.W.
      Health status improvement after surgical correction of primary hyperparathyroidism in patients with high and low preoperative calcium levels.
      • Spector B.C.
      • Netterville J.L.
      • Billante C.
      • Clary J.
      • Reinisch L.
      • Smith T.L.
      Quality-of-life assessment in patients with unilateral vocal cord paralysis.
       Fracture0.8150.63–1
      • Brazier J.E.
      • Green C.
      • Kanis J.A.
      Committee Of Scientific Advisors International Osteoporosis F
      A systematic review of health state utility values for osteoporosis-related conditions.
      • Tosteson A.N.
      • Gabriel S.E.
      • Grove M.R.
      • Moncur M.M.
      • Kneeland T.S.
      • Melton 3rd, L.J.
      Impact of hip and vertebral fractures on quality-adjusted life years.
       Permanent hypoparathyroidism and RLN injury0.7850.57–1
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sejean K.
      • Calmus S.
      • Durand-Zaleski I.
      • Bonnichon P.
      • Thomopoulos P.
      • Cormier C.
      • et al.
      Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
      • Sheldon D.G.
      • Lee F.T.
      • Neil N.J.
      • Ryan Jr., J.A.
      Surgical treatment of hyperparathyroidism improves health-related quality of life.
      • Spector B.C.
      • Netterville J.L.
      • Billante C.
      • Clary J.
      • Reinisch L.
      • Smith T.L.
      Quality-of-life assessment in patients with unilateral vocal cord paralysis.
      Time (y)
       Formal follow-up of asymptomatic disease105–15
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
       Reference patient remaining life expectancy2412–36
       Relative risk of fracture in asymptomatic PHPT30%0–60%
      • Lundstam K.
      • Heck A.
      • Mollerup C.
      • Godang K.
      • Baranowski M.
      • Pernow Y.
      • et al.
      Effects of parathyroidectomy versus observation on the development of vertebral fractures in mild primary hyperparathyroidism.
      • Khosla S.
      • Melton 3rd, L.J.
      • Wermers R.A.
      • Crowson C.S.
      • O'Fallon W.
      • Riggs B.
      Primary hyperparathyroidism and the risk of fracture: a population-based study.
      • Vestergaard P.
      • Mosekilde L.
      Cohort study on effects of parathyroid surgery on multiple outcomes in primary hyperparathyroidism.
      • Yeh M.W.
      • Zhou H.
      • Adams A.L.
      • Ituarte P.H.
      • Li N.
      • Liu I.A.
      • et al.
      The relationship of parathyroidectomy and bisphosphonates with fracture risk in primary hyperparathyroidism: an observational study.
       Discount rate3.0%0.015–0.045
      • Gold M.R.
      Cost-effectiveness in health and medicine.
       Health care cost inflation rate3.4%0.017–0.051
      RLN, Recurrent laryngeal nerve.
      Patients in the observation arm incurred the annual costs of formal guideline-recommended testing up to 10 years after diagnosis.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      If an observed patient developed a guideline-based indication for PTX at any time point, PTX was performed. Surgical outcomes, including complications, disease persistence, and recurrence requiring reoperation, were included in the model. The revised model assumed a 30% relative risk of fracture with observation compared to PTX.
      PTX was assumed to lower the annual risk of fracture from 3.9% to 3%, which was the annual risk of fracture in an age-matched patient without PHPT.
      • Khosla S.
      • Melton 3rd, L.J.
      • Wermers R.A.
      • Crowson C.S.
      • O'Fallon W.
      • Riggs B.
      Primary hyperparathyroidism and the risk of fracture: a population-based study.
      • Melton 3rd, L.J.
      • Crowson C.S.
      • O'Fallon W.M.
      Fracture incidence in Olmsted County, Minnesota: comparison of urban with rural rates and changes in urban rates over time.
      These probabilities of fracture were applied to both the PTX and observation strategies, as were the costs and QOL detriment associated with a fracture event. The strategy that produced the greatest quality-adjusted life expectancy (QALE) without exceeding an incremental cost-effectiveness ratio of $100,000 per quality-adjusted life year (QALY) was defined as optimal. A $100,000/QALY willingness-to-pay threshold for cost-effectiveness was selected based on current convention and allocation of heath care resources in the United States.
      • Braithwaite R.S.
      • Meltzer D.O.
      • King Jr., J.T.
      • Leslie D.
      • Roberts M.S.
      What does the value of modern medicine say about the $50,000 per quality-adjusted life-year decision rule?.
      • Neumann P.J.
      • Cohen J.T.
      • Weinstein M.C.
      Updating cost-effectiveness–the curious resilience of the $50,000-per-QALY threshold.

      Cost estimation

      All costs in the model were reported in 2015 US dollars and are listed in Table I. Direct costs of the medical observation and PTX strategies were estimated using the 2015 Medicare Prospective Payment System, average wholesale drug prices, and previously published cost estimates.
      • Zanocco K.
      • Angelos P.
      • Sturgeon C.
      Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Christensen L.
      • Iqbal S.
      • Macarios D.
      • Badamgarav E.
      • Harley C.
      Cost of fractures commonly associated with osteoporosis in a managed-care population.
      A third-party payer perspective was maintained for all costs. Annual observation costs included a physical examination, serum calcium and creatinine measurement, and bone density examination. An average expected, one-time cost of fracture of $16,281 was derived from literature reporting the costs of various types of fracture.
      • Christensen L.
      • Iqbal S.
      • Macarios D.
      • Badamgarav E.
      • Harley C.
      Cost of fractures commonly associated with osteoporosis in a managed-care population.
      A health care inflation rate of 3.4% was calculated from the mean of annual changes in the Consumer Price Index for Medical Care from 2005 to 2015. This inflation rate was applied to the future health care costs of the model and used to adjust older cost estimates to their 2015 values. A discount rate of 3% was applied to all future costs in the model.
      • Gold M.R.
      Cost-effectiveness in health and medicine.

      Effectiveness

      Effectiveness was measured by calculating QALEs for both strategies. Each of the possible treatment outcomes in the model was assigned a quality-adjustment factor based on literature review. The quality adjustment factor for fracture was simplified to an average utility of 0.815, which was derived from literature reporting the quality adjustment for various types of fracture.
      • Brazier J.E.
      • Green C.
      • Kanis J.A.
      Committee Of Scientific Advisors International Osteoporosis F
      A systematic review of health state utility values for osteoporosis-related conditions.
      • Tosteson A.N.
      • Gabriel S.E.
      • Grove M.R.
      • Moncur M.M.
      • Kneeland T.S.
      • Melton 3rd, L.J.
      Impact of hip and vertebral fractures on quality-adjusted life years.
      The methods of utility determination used by the primary sources are included in Supplemental Table I. Time elapsed while the reference patient was experiencing a particular outcome was multiplied by the quality adjustment factor of that outcome to yield effectiveness in QALYs. A 3% discount rate was also applied to all accumulated future QALYs.

      Sensitivity analysis

      Threshold analysis was performed on each variable in the model during 1-way sensitivity analysis to identify values where the PTX strategy became dominant (less costly and more effective) or cost-effective compared to observation. Threshold values for willingness-to-pay levels of $100,000/QALY and $150,000/QALY were calculated. A 3-way sensitivity analysis was performed to examine the combined effects of changes in the quality adjustment factor for asymptomatic PHPT and the cost and quality adjustment factor for fracture.
      Probabilistic sensitivity analysis using Monte Carlo simulation was performed, where triangular frequency distributions for each variable were simultaneously sampled during 1,000 consecutive iterations. The distributions were assigned a range of ±50% of the reference case estimate. In the case of utility, risk, or probability estimates, these variables were allowed to vary to the greatest extent, such that the reference case value was at the center of a range containing the boundary 0 or 1.

      Results

      Reference case

      PTX was the less costly and more effective strategy with an expected cost of $6,487 and an effectiveness of 17.54 QALYs. The observation strategy had an expected cost of $8,208 and an effectiveness of 17.35. Observation was $1,721 more costly than PTX and resulted in a loss of 0.19 QALYs. Observation was therefore dominated, because this strategy was both more costly and less effective than PTX. The median expected time to fracture was 17.5 years for the observation strategy and 23 years for the operation strategy.

      Sensitivity analysis

      The threshold conditions for cost-effectiveness and dominance of the PTX strategy during 1-way sensitivity analysis are shown in Table II. PTX remained the dominant strategy when the relative risk of fracture in observed, asymptomatic PHPT was >14%, the cost of fracture was >$7,600, the health care cost inflation rate was >1.5%, or the remaining life expectancy was >16 years. PTX was cost-effective until the remaining life expectancy was <3 years (Fig 2). Several variables returned no threshold values, because PTX was cost-effective and dominant for all possible values. The model was not sensitive to the rate of disease progression from asymptomatic PHPT to observation ineligibility or the complication rates for reoperative PTX.
      Table IIThreshold conditions for cost-effectiveness and dominance of the PTX strategy during one-way sensitivity analysis
      Variable$100,000/QALY threshold$150,000/QALY thresholdDominance threshold
      Probabilities
       Annual risk of fracture in age-matched patient without PHPTNoneNone>1.4%
       Annual risk of progression from asymptomatic PHPT to symptomatic PHPTNoneNoneNone
       Annual risk of recurrence after curative PTX<2.8%<3.4%<0.6%
       RLN injury after initial PTX<12.0%<11.9%<11.7%
       RLN injury after reoperative PTXNoneNoneNone
       Permanent hypoparathyroidism after initial PTX<11.9%<11.9%<9.8%
       Permanent hypoparathyroidism after reoperative PTXNoneNoneNone
       Persistent PHPT after initial PTXNoneNone<36%
       Persistent PHPT after reoperative PTXNoneNone<80%
      Costs in US dollars
       Initial parathyroidectomy<$36,000<$50,000<$8,300
       Reoperative parathyroidectomy<$430,000<$630,000<$42,000
       RLN injury treatment<$3.8 million<$5.6 million<$340,000
       Cost of fractureNoneNone>$7,600
       Annual cost of permanent hypoparathyroidism<$170,000<$245,000<$15,000
       Annual cost to observe asymptomatic PHPTNoneNone>$110
       Annual cost of calcimimetic therapy<$5.7 million<$8.3 million<$499,000
      QOL adjustment factors
       Asymptomatic hyperparathyroidismNoneNone<0.999
       Symptomatic hyperparathyroidismNoneNoneNone
       Long-term hypoparathyroidismNoneNoneNone
       Curative PTX with RLN injuryNoneNoneNone
       Asymptomatic PHPT with RLN damageNoneNoneNone
       Symptomatic PHPT and permanent RLN damageNoneNoneNone
       FractureNoneNoneNone
       Permanent hypoparathyroidism and RLN injuryNoneNoneNone
      Time (y)
       Formal follow-up of asymptomatic diseaseNoneNone>3
       Remaining life expectancy>3>2.5>16
       Relative risk of fracture in asymptomatic PHPTNoneNone>14%
       Discount rate<33%<50%<5%
       Health care cost inflation rateNoneNone>1.5%
      RLN, Recurrent laryngeal nerve.
      Figure thumbnail gr2
      Fig 2The incremental cost-effectiveness ratio for the parathyroidectomy strategy is displayed as a function of remaining life expectancy. The parathyroidectomy strategy is cost-effective when remaining life expectancy exceeds 3 years and dominant when life expectancy exceeds 16 years.
      The QOL adjustment factor for asymptomatic PHPT was the only quality adjustment variable that produced a threshold for dominance of the PTX strategy during 1-way sensitivity analysis. PTX was dominant only if the quality adjustment factor for asymptomatic PHPT was <0.999. PTX was cost-effective for all other possible quality adjustment factor values for asymptomatic PHPT.
      While the model was not generally sensitive to isolated changes in quality adjustment factors, 3-way sensitivity analysis of the quality adjustment factor for asymptomatic PHPT, quality adjustment factor for fracture, and cost of fracture demonstrated a larger combined effect of these 3 variables on the cost-effectiveness of the PTX strategy. As the cost of fracture increased, PTX was cost-effective in increasing combinations of quality adjustment factors for asymptomatic PHPT and fracture (Fig 3). When the cost of fracture was increased to 150% of the reference case assumption (from $16,281 to $24,422), the PTX strategy was either cost-effective or dominant in all quality adjustment combinations.
      Figure thumbnail gr3
      Fig 3Three-way sensitivity analysis examining the combined effects of the cost of fracture, the quality adjustment factor for fracture, and the quality adjustment factor for mild, asymptomatic PHPT on the cost-effectiveness of the PTX strategy. Intersecting quality-adjustment factors produce results where PTX dominates (light gray region), PTX is cost-effective (dark gray region), or observation is cost-effective (black region). As the cost of fracture increases (AC), PTX becomes cost-effective in increasing combinations of quality adjustment factors. When the cost of fracture is 150% of the reference case assumption (C), PTX is cost-effective or dominant in all quality-adjustment combinations.
      Monte Carlo simulation demonstrated PTX to be the optimal strategy in 995 (99.5%) of the iterations. PTX was the dominant strategy (both less costly and more effective) in 749 (74.9%) iterations. In an additional 246 (24.6%) iterations, PTX resulted in gains in QOL or cost savings that produced an incremental cost-effectiveness ratio of <$100,000/QALY. Among the remaining 5 (0.5%) cases in which observation was cost-effective, this strategy was less costly and less effective in 4 (0.4%) and more costly and more effective in 1 (0.1%) compared to PTX. All iterations from this simulation are plotted in Fig 4.
      Figure thumbnail gr4
      Fig 4All model assumptions were simultaneously varied across triangular frequency distributions in a 1,000-iteration Monte Carlo simulation. The incremental cost and incremental effectiveness of parathyroidectomy versus observation are plotted for each iteration. Points to the right of the dashed threshold line are iterations where parathyroidectomy was cost-effective (99.5%). Points to the left of the line are iterations where observation was cost-effective (0.5%). Parathyroidectomy was dominant in 74.9% of iterations. Observation was dominant in 0% of iterations. ICER, Incremental cost effectiveness ratio.

      Discussion

      This revised cost-effectiveness analysis demonstrates that PTX is the dominant management strategy for many patients with mild, asymptomatic PHPT, because operative management at the time of diagnosis produced a greater QALE at a lower cost compared with medical observation. Based on current life expectancy estimates, PTX was dominant for patients <70 years old and remained cost-effective for patients >90 years old.
      In previous models of asymptomatic PHPT, PTX was shown to be cost-effective by producing increased QOL with an acceptable additional cost, ranging from $721/QALY to $4,778/QALY.
      • Zanocco K.
      • Angelos P.
      • Sturgeon C.
      Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
      • Zanocco K.
      • Sturgeon C.
      How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
      • Sejean K.
      • Calmus S.
      • Durand-Zaleski I.
      • Bonnichon P.
      • Thomopoulos P.
      • Cormier C.
      • et al.
      Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
      When the fracture risk reduction of operation is considered, PTX shifts from being merely cost-effective to a state of dominance with increased QOL and cost savings of $1,721 compared with observation. Similarly, the conclusion of previous models was dependent on the assumption that “asymptomatic” PHPT is in fact symptomatic and produces diminished QOL.
      In our previous model, the quality adjustment for asymptomatic PHPT needed to be <0.998 for PTX to be cost-effective.
      • Zanocco K.
      • Angelos P.
      • Sturgeon C.
      Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
      A similar analysis conducted within the French health care system required the quality adjustment factor for asymptomatic disease to be <0.999.
      • Sejean K.
      • Calmus S.
      • Durand-Zaleski I.
      • Bonnichon P.
      • Thomopoulos P.
      • Cormier C.
      • et al.
      Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
      Although there is mounting evidence that subjective QOL does in fact improve after operation in seemingly asymptomatic patients, the concept of truly symptom-free PHPT remains part of the current consensus guidelines.
      • Bilezikian J.P.
      • Brandi M.L.
      • Eastell R.
      • Silverberg S.J.
      • Udelsman R.
      • Marcocci C.
      • et al.
      Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
      • Zanocco K.
      • Butt Z.
      • Kaltman D.
      • Elaraj D.
      • Cella D.
      • Holl J.L.
      • et al.
      Improvement in patient-reported physical and mental health after parathyroidectomy for primary hyperparathyroidism.
      The present work demonstrates that, once fracture is considered, minor differences in subjective symptoms become insignificant in relation to the objective cost and QOL effects of fracture.
      Several limitations are apparent in this study. The known complications of operations deemed to be of low likelihood or significance (wound infection, hematoma requiring operative evacuation, operative mortality, hypertrophic cervical scar, temporary recurrent laryngeal nerve injury, and temporary hypoparathyroidism) were omitted to maintain computational simplicity. More costly and morbid operation to excise ectopic glands was not considered. The absence of these events from the model introduced bias favoring the PTX strategy. Given the strong dominance of PTX compared to observation, these simplifications were unlikely to change the model's conclusions.
      The cost and QOL impact of nephrolithiasis during observation of asymptomatic PHPT were omitted due to lack of primary data, introducing bias favoring the observation strategy. The model also favored observation by assuming that patients who progress to symptomatic disease immediately underwent PTX without experiencing the diminished QOL of symptomatic PHPT. Observed patients were followed for 10 years to maintain consistency with our previous analysis. Longer follow-up would have increased the costs of observation. In addition, the model only allowed one lifetime fracture event. In reality, the occurrence of an initial fracture would likely increase the probability of additional future fractures,
      • Drew S.
      • Judge A.
      • Cooper C.
      • Javaid M.K.
      • Farmer A.
      • Gooberman-Hill R.
      Secondary prevention of fractures after hip fracture: a qualitative study of effective service delivery.
      adding additional costs and QOL detriment to the observation strategy. Removing these biases would have also strengthened the dominance of the operation strategy and would not have changed the model's conclusions.
      Other simplifications of the model introduced imprecision without clearly biasing either strategy. The third-party payer perspective was used to maintain consistency with our previous model. A more complete analysis from the societal perspective would have included transportation time and lost patient work productivity associated with medical and operative encounters. These costs would be accumulated through both the PTX and observation strategies, and their inclusion would be unlikely to appreciably change the model's results.
      While focused PTX and 4-gland exploration techniques are both performed, no distinction was made between these approaches in the model. This assumption is valid, because operative outcomes are similar and Medicare reimburses identically for the 2 techniques. The occurrence of a fragility fracture and the operative complications of recurrent laryngeal nerve injury and permanent hypoparathyroidism all have a wide spectrum of severity and treatment costs; however, these were all assigned one treatment cost and QOL adjustment. We believe the uncertainty associated with these estimates was adequately addressed with sensitivity analysis.
      In conclusion, in our revised model, PTX is less costly and more effective (ie, dominant) compared to medical observation in patients with asymptomatic PHPT who are <70 years old. PTX remains cost-effective for all other surgical candidates with asymptomatic disease. To improve QOL and lower healthcare costs, definitive operative management should be offered to all patients with PHPT.

      Supplementary data

      References

        • Yeh M.W.
        • Ituarte P.H.
        • Zhou H.C.
        • Nishimoto S.
        • Liu I.L.
        • Harari A.
        • et al.
        Incidence and prevalence of primary hyperparathyroidism in a racially mixed population.
        J Clin Endocrinol Metab. 2013; 98: 1122-1129
        • Lundstam K.
        • Heck A.
        • Mollerup C.
        • Godang K.
        • Baranowski M.
        • Pernow Y.
        • et al.
        Effects of parathyroidectomy versus observation on the development of vertebral fractures in mild primary hyperparathyroidism.
        J Clin Endocrinol Metab. 2015; 100: 1359-1367
        • Rubin M.R.
        • Bilezikian J.P.
        • McMahon D.J.
        • Jacobs T.
        • Shane E.
        • Siris E.
        • et al.
        The natural history of primary hyperparathyroidism with or without parathyroid surgery after 15 years.
        J Clin Endocrinol Metab. 2008; 93: 3462-3470
        • Silverberg S.J.
        • Shane E.
        • Jacobs T.P.
        • Siris E.
        • Bilezikian J.P.
        A 10-year prospective study of primary hyperparathyroidism with or without parathyroid surgery.
        N Engl J Med. 1999; 341: 1249-1255
        • Khosla S.
        • Melton 3rd, L.J.
        • Wermers R.A.
        • Crowson C.S.
        • O'Fallon W.
        • Riggs B.
        Primary hyperparathyroidism and the risk of fracture: a population-based study.
        J Bone Miner Res. 1999; 14: 1700-1707
        • Vestergaard P.
        • Mosekilde L.
        Cohort study on effects of parathyroid surgery on multiple outcomes in primary hyperparathyroidism.
        BMJ. 2003; 327: 530-534
        • Bilezikian J.P.
        • Brandi M.L.
        • Eastell R.
        • Silverberg S.J.
        • Udelsman R.
        • Marcocci C.
        • et al.
        Guidelines for the management of asymptomatic primary hyperparathyroidism: summary statement from the Fourth International Workshop.
        J Clin Endocrinol Metab. 2014; 99: 3561-3569
        • Yeh M.W.
        • Zhou H.
        • Adams A.L.
        • Ituarte P.H.
        • Li N.
        • Liu I.A.
        • et al.
        The relationship of parathyroidectomy and bisphosphonates with fracture risk in primary hyperparathyroidism: an observational study.
        Ann Intern Med. 2016; 164: 715-723
        • Zanocco K.
        • Angelos P.
        • Sturgeon C.
        Cost-effectiveness analysis of parathyroidectomy for asymptomatic primary hyperparathyroidism.
        Surgery. 2006; 140 (discussion 81-2): 874-881
        • Zanocco K.
        • Sturgeon C.
        How should age at diagnosis impact treatment strategy in asymptomatic primary hyperparathyroidism? A cost-effectiveness analysis.
        Surgery. 2008; 144: 290-298
        • Sejean K.
        • Calmus S.
        • Durand-Zaleski I.
        • Bonnichon P.
        • Thomopoulos P.
        • Cormier C.
        • et al.
        Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis.
        Eur J Endocrinol. 2005; 153: 915-927
        • Rao D.S.
        • Phillips E.R.
        • Divine G.W.
        • Talpos G.B.
        Randomized controlled clinical trial of surgery versus no surgery in patients with mild asymptomatic primary hyperparathyroidism.
        J Clin Endocrinol Metab. 2004; 89: 5415-5422
      1. Actuarial life table [Internet]. US Social Security Administration, Woodlawn (MD)2016 (Available from: https://www.ssa.gov/oact/STATS/table4c6.html)
        • Melton 3rd, L.J.
        • Crowson C.S.
        • O'Fallon W.M.
        Fracture incidence in Olmsted County, Minnesota: comparison of urban with rural rates and changes in urban rates over time.
        Osteoporos Int. 1999; 9: 29-37
        • Braithwaite R.S.
        • Meltzer D.O.
        • King Jr., J.T.
        • Leslie D.
        • Roberts M.S.
        What does the value of modern medicine say about the $50,000 per quality-adjusted life-year decision rule?.
        Med Care. 2008; 46: 349-356
        • Neumann P.J.
        • Cohen J.T.
        • Weinstein M.C.
        Updating cost-effectiveness–the curious resilience of the $50,000-per-QALY threshold.
        N Engl J Med. 2014; 371: 796-797
      2. Medicare [Internet]. Centers for Medicare and Medicaid Services, Baltimore (MD)2016 (Available from: https://www.cms.gov/Medicare/Medicare.html)
      3. Red Book Online [Internet]. Truven Health Analytics Inc, Ann Arbor (MI)2016 (Available from:https://www.micromedexsolutions.com)
        • Christensen L.
        • Iqbal S.
        • Macarios D.
        • Badamgarav E.
        • Harley C.
        Cost of fractures commonly associated with osteoporosis in a managed-care population.
        J Med Econ. 2010; 13: 302-313
      4. Bureau of Labor Statistics Data. Available from: http://data.bls.gov/timeseries/CUUR0000SAM?output_view=pct_12mths.

        • Gold M.R.
        Cost-effectiveness in health and medicine.
        Oxford University Press, New York1996 (p. xxiii. 425 p)
        • Brazier J.E.
        • Green C.
        • Kanis J.A.
        • Committee Of Scientific Advisors International Osteoporosis F
        A systematic review of health state utility values for osteoporosis-related conditions.
        Osteoporos Int. 2002; 13: 768-776
        • Tosteson A.N.
        • Gabriel S.E.
        • Grove M.R.
        • Moncur M.M.
        • Kneeland T.S.
        • Melton 3rd, L.J.
        Impact of hip and vertebral fractures on quality-adjusted life years.
        Osteoporos Int. 2001; 12: 1042-1049
        • Zanocco K.
        • Butt Z.
        • Kaltman D.
        • Elaraj D.
        • Cella D.
        • Holl J.L.
        • et al.
        Improvement in patient-reported physical and mental health after parathyroidectomy for primary hyperparathyroidism.
        Surgery. 2015; 158: 837-845
        • Drew S.
        • Judge A.
        • Cooper C.
        • Javaid M.K.
        • Farmer A.
        • Gooberman-Hill R.
        Secondary prevention of fractures after hip fracture: a qualitative study of effective service delivery.
        Osteoporos Int. 2016; 27: 1719-1727
        • Carneiro-Pla D.M.
        • Solorzano C.C.
        • Lew J.I.
        • Irvin 3rd, G.L.
        Long-term outcome of patients with intraoperative parathyroid level remaining above the normal range during parathyroidectomy.
        Surgery. 2008; 144 (discussion 93-4): 989-993
        • Udelsman R.
        • Lin Z.
        • Donovan P.
        The superiority of minimally invasive parathyroidectomy based on 1650 consecutive patients with primary hyperparathyroidism.
        Ann Surg. 2011; 253: 585-591
        • Karakas E.
        • Muller H.H.
        • Schlosshauer T.
        • Rothmund M.
        • Bartsch D.K.
        Reoperations for primary hyperparathyroidism–improvement of outcome over two decades.
        Langenbecks Arch Surg. 2013; 398: 99-106
        • Sheldon D.G.
        • Lee F.T.
        • Neil N.J.
        • Ryan Jr., J.A.
        Surgical treatment of hyperparathyroidism improves health-related quality of life.
        Arch Surg. 2002; 137 (discussion 6-8): 1022-1026
        • Burney R.E.
        • Jones K.R.
        • Christy B.
        • Thompson N.W.
        Health status improvement after surgical correction of primary hyperparathyroidism in patients with high and low preoperative calcium levels.
        Surgery. 1999; 125: 608-614
        • Vidal-Trecan G.M.
        • Stahl J.E.
        • Eckman M.H.
        Radioiodine or surgery for toxic thyroid adenoma: dissecting an important decision. A cost-effectiveness analysis.
        Thyroid. 2004; 14: 933-945
        • Spector B.C.
        • Netterville J.L.
        • Billante C.
        • Clary J.
        • Reinisch L.
        • Smith T.L.
        Quality-of-life assessment in patients with unilateral vocal cord paralysis.
        Otolaryngol Head Neck Surg. 2001; 125: 176-182