Clinical outcomes and fiscal consequences of bilateral neck exploration for primary idiopathic hyperparathyroidism without preoperative radionuclide imaging or minimally invasive techniques☆☆☆

Patients and methods 

Diagnosis, treatment, and outcome data were prospectively collected for all patients with a diagnosis of primary idiopathic hyperparathyroidism in the endocrine surgical service at the university teaching hospitals located in Oxford, United Kingdom, during the 32-year period from 1968 through 2000. This dataset included all patients advised to undergo surgical exploration for this diagnosis, without exclusion, and without localization procedures performed before operation.

During the period when head and neck ultrasonography was being refined, a minority of patients underwent ultrasound imaging of their thyroid bed before referral. However, the findings obtained from these ultrasound imaging studies did not alter the intention to administer methylene blue at operation, or change the standard operative approach used in our patients throughout this series.

The operative strategy used throughout the study period involved full exploration of both sides of the neck, performed as an open operation under general anesthesia, as described previously by Nottingham et al9 and others. The surgical objectives were to identify all parathyroid tissue, and to remove any gland judged, on visual inspection by the operating surgeon, to be greater than twice normal size. If visual inspection confirmed normal gland size and appearance, to avoid damaging the blood supply, routine biopsy of normal parathyroid glands was not performed.

In cases where 3 or more glands were found to be enlarged the diagnosis of parathyroid hyperplasia was made and all but one-half of 1 gland (approximately 50 mg residual parathyroid tissue) was resected, protecting the vascular pedicle of this remaining gland. In cases where no single adenoma or fewer than 4 hyperplastic glands were identified, cervical thymectomy was performed. Patients found to have parathyroid carcinoma at the time of initial neck exploration were treated with en bloc tumor resection, ipsilateral thyroid lobectomy, central neck dissection, and biopsy of any suspicious masses in the lateral neck compartments.

Methylene blue was used to facilitate intraoperative recognition of parathyroid tissue in all patients, and was administered at a final concentration of 5 mg/kg in a total volume of 500 mL 5% dextrose in water, and was timed to infuse intravenously during the hour before incision. Our original reported experience with methylene blue localization30 has remained concordant with data subsequently reported by other authors.31

During the initial period of this series, intraoperative frozen-section biopsy was routinely performed for all patients. However, as our experience using methylene blue increasingly demonstrated its consistent ability to identify parathyroid tissue, frozen-section biopsy became increasingly reserved only for those patients in whom visual inspection of the parathyroid glands was equivocal for the diagnosis of adenoma; in the presence of multigland hyperplasia; or where texture, fixation to surrounding tissues, or lymph node involvement suggested parathyroid carcinoma. Cost calculations were on the basis of our original intent to treat all patients with intraoperative frozen-section biopsy.

Operative time, postanesthesia care unit (PACU) time, operative findings, frozen-section and permanent histopathology evaluations, albumin-corrected serum calcium levels at 3 and 12 month follow-up, and length of hospitalization were recorded. Time data (operative and PACU) are reported as the mean from the entire series. Unless otherwise indicated, the SEM was within 5% of the mean. In our series, we defined “cure” as biochemical confirmation that serum calcium and parathyroid hormone levels had returned to the normal reference ranges for our laboratory, measured at 1 year after operation.

Health care administration and costing in the United Kingdom are fundamentally different from those used in the United States and elsewhere, and these differences render it extremely difficult and unreliable to calculate per-procedure costs for our patient population in a manner easily translated to other health care systems. However, with consistent linear data for: (1) operative and recovery times; (2) patient length-of-stay; (3) use of radiology and pathology services; and (4) treatment outcomes we endeavored to calculate the costs for performance of this procedure. Our cost estimates were calculated using the identical use costs reported previously from a major, not-for-profit, university hospital system21, 29 and previously published average 1997 Medicare reimbursement rates for radiology and pathology professional fees.32 Methylene blue administration costs were derived from actual costs at the authors' US institution (University of Florida), calculated on the basis of an average dosage of 400 mg/patient, a 500 mL bag of 5% dextrose water, and an intravenous administration set, yielding an average cost of $39 per patient.

Results 

During the study period, 688 patients were treated with 100% follow-up. There were no perioperative mortalities. No patients sustained injury to their recurrent laryngeal nerves, and there were no cases of postoperative hemorrhage, infection, or persistent symptomatic hypocalcemia. Wound cosmesis after operation, as assessed by the patient, was judged to be excellent with fewer than 2% of patients having keloid or dystrophic scar develop at their incision site.

On the basis of the assumption of a normal compliment of 4 parathyroid glands per patient, 2520 of 2752 (91.6%) glands were identified in our series without use of preoperative or intraoperative radionuclide imaging studies. Single gland adenomas were identified in 78.7% of patients, with 8 of these 542 patients having an adenoma in a supernumerary parathyroid gland. Twenty-two patients (3.2%) were found to have double adenomas, and multigland hyperplasia was identified in 98 patients (14.2%). Parathyroid carcinoma was diagnosed and treated during the primary operation in 10 patients (1.5%).

The mean operative time for the entire series of patients was 65 minutes. However, for patients in whom a single adenoma was removed in the presence of 3 or more normal parathyroid glands (78.7% of patients), the mean operative time decreased to 35 minutes. On completion of operation, average time spent in PACU before return to inpatient care was 30 minutes. All patients in our series were scheduled for routine postoperative admissions, with a planned stay of 48 hours, to provide prompt recognition and treatment if plasma calcium levels fell to symptomatic levels. During the course of the series, as a result of progressive bed shortages within our institution, postoperative stay was shortened to 24 hours. However, our cost calculations were on the basis of the original intent to treat the patient with a 48-hour inpatient stay.

Cure was achieved for 672 of the 688 patients (97.7%). Diagnosis, operative and recovery times, and outcome data are summarized in Table I.

Table I. Primary idiopathic hyperparathyroidism

	Diagnosis at operation
	Diagnosis	No. patients	Notes
	Single adenoma	542	8 in fifth gland
	Double adenoma	22
	Hyperplasia	98
	Carcinoma	10
	Full treatment at operation

	Outcomes after operation
	Category	No. patients	Percentage
	Perioperative mortality	None	None
	Complications	None	None
	Biochemical cure	672	97.7%
	Keloid formation	13	1.9%

Biochemical cure: normal serum calcium and PTH levels 1 year after operation.

A model for projecting the cost of treating patients using our operative approach was calculated by applying the model previously reported by Denham and Norman,21 and Greene et al29; average 1997 Medicare compensation32; and actual costs of methylene blue at our US institution (operating room time costs, $7.88/min; PACU costs, $2.25/min; hospital stay costs, $240/24 h; Frozen-section costs and pathologist professional fees, $84 and $60, respectively; sestamibi costs and radiologist professional fees, $305 and $124, respectively; and methylene blue, $39/patient). On the basis of our intent to treat, we assumed an inpatient hospital stay of 48 hours for all patients. Cost calculations did not include calculations for surgeon or anesthesiologist professional fees.

An average cost estimate of $1243 per patient was calculated for the entire study. However, in patients with a single adenoma, the mean cost for treatment decreased to $1107 per patient maintaining the original 48-hour inpatient care model. As bed shortages necessitated a decrease in postoperative stay to 24 hours, estimated costs were projected to decrease to $767 per patient with single adenomas. Cost estimates for patients treated using our operative approach are summarized in Table II.

Table II. Cost analysis: Bilateral neck exploration with intraoperative methylene blue for treatment of primary hyperparathyroidism

	Category	Cost (all patients)	Cost (single adenoma)	Notes
	Operating room ($7.88/min)	$512	$276	Mean 65 vs 35 min
	Frozen section cost	$84	$84	Intent to treat 100% of patients
	Pathologist fee	$60	$60	Intent to treat 100% of patients
	PACU ($2.25/min)	$68	$68	Mean time 30 min
	Hospital stay	$480	$480	48-h intent to treat
	Methylene blue	$39	$39	Drug and administration costs
	Total	$1243	$1107

PACU, postanesthesia care unit (recovery room); methylene blue, 5 mg/kg intravenous administration in 500 mL 5% dextrose in water 1 hour before incision. Note: cost and fee references are on the basis of previously reported values derived from a major, not-for-profit, university hospital system21, 29 and previously published average 1997 Medicare reimbursement rates for radiology and pathology professional fees.32

Discussion 

This series summarizes more than 3 decades of our experience using the technique of bilateral neck exploration, systematic identification of all parathyroid tissue aided by intraoperative methylene blue, and definitive treatment of disease at the primary operation for treatment of patients with a diagnosis of primary hyperparathyroidism. The results reported in our series are by no means unique, and cure rates in excess of 96% have been achieved elsewhere by surgeons performing meticulous dissection and a systematic search for all parathyroid tissues.9, 33 However, we attribute our diagnostic and treatment successes and operative speed, in significant part, to our systematic approach and the ease with which parathyroid tissues are identified after staining with methylene blue. Our original experience using methylene blue30 continues to be confirmed by other reports25, 31, 34, 35, 36 indicating that its routine use speeds identification of the glands, and that it is safe and inexpensive.

Early experience with methylene blue infusion increasingly raised our confidence in its ability to identify parathyroid adenomas and hyperplasia, as confirmed by subsequent routine histopathology. This rendered routine frozen section unnecessary except for those rare instances where uptake of the vital dye was poor and visual inspection did not provide clear identification of abnormal glands. Frozen section was never relied on in the diagnosis of parathyroid carcinoma, which is much more dependent on the observations of an experienced surgeon with regard to parathyroid texture, evidence of local invasion, and nodal involvement. Thus, dispensing with frozen section had the merit of saving operating room time and conserving pathology resources.

The endocrine surgery practice in Europe contrasts with that in the United States. Patients identified with hypercalcemia and increased parathyroid hormone levels are typically referred directly from their primary care or generalist physicians to a surgical specialist. Thus, the surgeon has the first-hand responsibility to evaluate preoperative and intraoperative localization techniques, variations in surgical approach to this disease, and methods of evaluating treatment success in the intraoperative and postoperative periods.

During the first 2 decades of our experience, comparing our outcomes with other endocrine operation centers, the systematic approach and meticulous technique remained the unifying principals ensuring successful treatment of these patients. During the last 10 years, development of sestamibi imaging for parathyroid localization and preoperative neck ultrasound have not been shown to achieve greater cure rates in unselected patients. The experience with neck ultrasound in this series was limited to fewer than 30 patients, and in our hands this imaging modality identified only the largest and most easily visualized adenomas. Motivated by our concern for double adenomas and hyperplasia, preoperative ultrasound findings were not used to alter our approach of bilateral systematic exploration or the use of intraoperative methylene blue. Ultrasound did not, therefore, significantly impact on our operative times in these patients.

Development of sestamibi imaging for preoperative localization of parathyroid disease represented an exciting adjunct to our practice. Subsequent to its description in 1989,24 sestamibi localization has been widely evaluated in the treatment of patients with hyperparathyroidism.6, 11, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 24, 25, 27, 28, 29, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46 Our experience with preoperative sestamibi imaging has proven it to be valuable in patients undergoing reoperation for failed parathyroid operation and recurrent disease,47 particularly if supported by other localization studies such as MRI or selective venous parathyroid hormone sampling.

It is our impression that two recurrent themes have arisen from increasing experience with use of preoperative sestamibi imaging: First, that preoperative localization of parathyroid adenomas allow patients to undergo limited, and possibly less morbid, neck exploration; second, that imaging-directed parathyroid localization may extend surgical treatment of this disease to the hands of minimally invasive surgeons rather than the experienced endocrine surgeons who historically have provided patients the safest and most successful treatment outcomes.1, 2, 3, 4, 5, 6, 7, 8, 9, 10

In 1998, Denham and Norman21 presented a paper at the American College of Surgeons Clinical Congress with subsequent publications20 purporting to demonstrate the cost-effectiveness of preoperative sestamibi imaging for use in all patients with idiopathic primary hyperparathyroidism. These articles, together with others,13, 17, 19, 37 challenged the accepted surgical gold standard of bilateral neck exploration and identification of all parathyroid tissue without preoperative radionuclide localization.9, 33, 38 The key message to emerge from these reports, and in some instances subsequent publications targeted at prospective patients by the Internet, was that preoperative sestamibi imaging offered a quicker, cheaper, and more effective way of curing idiopathic hyperparathyroidism when compared with traditional bilateral neck exploration without preoperative imaging. A single study also alluded to the additional benefits of decreased pain and improved cosmesis.20

On the basis of these reports, we believed it appropriate to compare our experience treating this disease with the cost and outcome data presented in MIP articles to determine whether the scan-directed MIP approach would prove superior to those obtained in our practice. Understanding that the intrinsic differences in health care costing in the United Kingdom render direct cost comparisons unreliable, we elected to apply the identical costing schema previously reported.21, 32 This schema was used for subsequent standard neck exploration versus scan-directed MIP cost comparisons29 and we applied it to the preoperative and postoperative care, and pathology and radiology use in our series. Applying this costing schema to our series yielded an average cost of $1107 for each patient with a parathyroid single adenoma compared with $1123, reported by Denham and Norman,21 per patient undergoing scan-directed MIP for the same diagnosis.

However, careful scrutiny of these reports generated a number of significant additional concerns for their claims. First, the reports imply that the outcome data are like being compared with like, whereas the outcome data compare highly selected (scan-directed MIP) with unselected (standard neck exploration) patients. In the original Denham and Norman report,21 all patients underwent preoperative dual-phase sestamibi imaging, with the time period and total number of patients presenting with primary idiopathic hyperparathyroidism unreported. Twenty-two percent of these patients were found to be unsuitable for scan-directed unilateral operation because sestamibi imaging did not demonstrate uptake, did not demonstrate single gland disease in an easily accessible site for unilateral approach, or showed multiple gland disease. This subset represents the most challenging and time-consuming patients for endocrine surgeons to cure, and were excluded from analysis. These exclusion data have borne up in subsequent similar studies, at rates greater than 15%.17

Extrapolating from the above exclusion data, the total number of patients reviewed from the 4 series quoted (1 unpublished) in the Denham and Norman21 report appears to be 68. The authors acknowledge the additional costs of failed sestamibi imaging increases the average patient cost to $1263. Although this adjustment begins to address the issues of missed patients considered for scan-directed parathyroid operation, it falls short of the mark as it cannot fully account for all patients with multigland disease (17.4% in our series) and it does not include additional costs for sestamibi required for intraoperative gamma probe localization, advocated by the authors in a subsequent publication.19

In addition, the high cure rates asserted by this report were not supported by precise data,21 and subsequent reports may also be criticized for follow up intervals considered inadequate to define “cure.”17, 19, 20 Operative times for bilateral exploration for primary idiopathic hyperparathyroidism in excess of 2 hours, as quoted above,21 represent, we believe, surgeon inexperience or failure to use a systematic strategy for the identification of all parathyroid tissues, accessible by the neck, within 60 minutes.

The inevitable economics of health care worldwide have forced surgeons to adopt a policy of ambulatory operation for increasing numbers of procedures, and this trend was increasingly applied to patients throughout our series. Although potential downsides for patients undergoing ambulatory parathyroid operation include the loss of convenient and reliable biochemical monitoring after operation that may lead to morbidity, and loss of valuable teaching resources in academic centers, several initial studies,17, 28, 41 including our own experience, suggest that it is at least safe.

On the basis of our series data, careful evaluation of the above-cited reports, and our experience with shortening hospital stays after parathyroid operation, we believe that an alternate, and perhaps more appropriate, cost comparison between our standard bilateral neck exploration and scan-directed MIP is shown in Table III.

Table III. Alternate cost analysis: Standard bilateral neck exploration verus scan-directed minimally invasive parathyroidectomy for single adenoma

	Category	Cost (standad exploration)	Cost (scan-directed MIP)	Notes
	Operating room ($7.88/min)	$276	$362	Mean 35 vs 46 min
	Frozen section cost	$84	$84
	Pathologist fee	$60	$60
	PACU ($2.25/min)	$68	$68	Mean time 30 min
	Hospital stay	$240	$120	24-h vs 2-h short stay
	Methylene blue	$39	$0	Drug and administration cost
	Sestamibi	$0	$305	Preoperative dose only
	Radiologist fee	$0	$124	Preoperative study only
	Sestamibi (failed scan)	$0	$60	19.5% of patients
	Radiologist fee (failed scan)	$0	$24	19.5% of patients
	Additional cost for patients without single adenoma	$242	$346	19.5% of patients
	Total	$1109	$1553
	Additional costs
	Sestamibi	$0	$305	Intraoperative gamma probe
	Grand total	$1109	$1858

PACU, postanesthesia care unit (recovery room); methylene blue, 5 mg/kg intravenous administration in 500 mL 5% dextrose in water 1 hour before incision. Note: cost and fee references are on the basis of previously reported values derived from a major, not-for-profit, university hospital system21, 29 and previously published average 1997 Medicare reimbursement rates for radiology and pathology professional fees.32

No attempt was made in our series to assess the pain experienced by patients treated with standard Kocher cervical incision. Our standard surgical practice reflects our institution's long-standing interest in pre-emptive analgesia and our subjective observations that patients' scars are barely detectable when we examine them during their 1-year follow-up. Assertions that pain and cosmesis after scan-directed minimally invasive parathyroidectomy are improved20 are unsupported by objective data. Symmetrical scars, particularly on the face, head, and neck, have long been held by plastic and reconstructive surgeons as aesthetically more pleasing,48 and those authors who advocate placing incisions only over the area localized by sestamibi, are creating, albeit short, asymmetric scars.

Advances in the use of radioisotopes selectively taken up by the parathyroid glands, single-photon emission computed tomography scanning, and hand-held gamma probes have broadened the applicability of minimally invasive surgical techniques to parathyroid operation. Many surgeons, including those with inadequate training and experience in endocrine operation, appear to find these advances irresistible despite their considerable cost, and the absence of any prospective randomized trials demonstrating benefit for patients with primary idiopathic hyperparathyroidism. Readers may debate that a formal prospective randomized trial will be required to fully address this question, however, a properly designed trial will require skilled and experienced endocrine surgeons performing both techniques in a randomized fashion, and it is difficult to conceptualize such a trial created that is truly free of operator bias. Furthermore, some have suggested that an experienced endocrine surgeon could apply scan-directed minimally invasive parathyroid operation with even greater speed than their standard bilateral exploration. Although this may be a theoretical possibility, using our series data, completely eliminating operation, PACU, and inpatient stay, would only decrease treatment costs for scan-directed MIP to equivalence with standard exploration, whereas functionally blinding the surgeon to the status of the remaining parathyroid glands.

It is our opinion, from this and many other excellent series worldwide with near-identical results, that patients with primary idiopathic hyperparathyroidism are best served when a full, bilateral neck exploration is performed by an experienced endocrine surgeon. We believe that this approach offers patients the best opportunity for definitive cure during their first operation with minimal risk for morbidity and at minimal cost. Sestamibi imaging should, in our opinion, be reserved for patients with recurrent hyperparathyroidism or a failed initial neck exploration. Support for preoperative sestamibi imaging for all patients with primary idiopathic hyperparathyroidism should be set aside, as the costs are higher, effects on cure rates unproven, and benefits to individual patients marginal, even if eligible for minimally invasive operation. As endocrine surgeons are presented with increasing numbers of patients with the diagnosis of idiopathic primary hyperparathyroidism, we believe that the costs saved by maintaining the gold standard of bilateral neck exploration could be better applied to organized and rigorous training for the next generation of skilled and experienced endocrine surgeons. They will be needed to maintain successful outcomes for all patients, not just those with unequivocal disease on preoperative imaging scans.