abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid...

dutan

ORIGINAL SCIENTIFIC ARTICLES

Abandoning Unilateral Parathyroidectomy:Why We Reversed Our Position after15,000 Parathyroid OperationsJames Norman, MD, FACS, FACE, Jose Lopez, MD, FACS, Douglas Politz, MD, FACS, FACE

BACKGROUND: Our group championed the techniques and benefits of unilateral parathyroidectomy. As ourexperience has matured, it seems this limited operation might be appropriate only occasionally.

METHODS: A single surgical group’s experience with 15,000 parathyroidectomies examined the ongoingdifferences between unilateral and bilateral techniques for 10-year failure/recurrence, multig-land removal, operative times, and length of stay.

RESULTS: With limited experience, 100% of operations were bilateral, decreasing to 32% by the 500th

operation (p � 0.001), and long-term failure rates increased to 6%. Failures were 11 times morelikely for unilateral explorations (p � 0.001 vs bilateral), causing gradual increases in bilateralexplorations to 97% at the 14,000th operation (p � 0.001). Ten-year cure rates are unchangedfor bilateral operations, and unilateral operations show continued slow recurrence rates of 5%(p � 0.001). Removal of more than one gland occurred 16 times more frequently when 4 glandswere analyzed (p � 0.001), increasing cure rates to the current 99.4% (p � 0.001). Of 1,060reoperations performed for failure at another institution, intraoperative parathyroid hormonelevels fell �50% in 22% of patients, yet a second adenoma was subsequently found. Operativetimes decreased with experience; bilateral operations taking only 5.9 minutes longer on average(22.3 vs 16.4 minutes; p � 0.001), which is 25 minutes less than unilateral at the 500th

operation (p � 0.001). By the 1,000th operation, incision size (2.5 � 0.2 cm), anesthesia, andhospital stay (1.6 hours) were identical for unilateral and bilateral procedures.

CONCLUSIONS: Regardless of surgical adjuncts (scanning, intraoperative parathyroid hormone), unilateral para-thyroidectomy will carry a 1-year failure rate of 3% to 5% and a 10-year recurrence rate of 4%to 6%. Allowing rapid analysis of all 4 glands through the same 1-inch incision has caused us toall but abandon unilateral parathyroidectomy. (J Am Coll Surg 2012;214:260–269. © 2012 by
the American College of Surgeons)
tttl

iarpttt

Primary hyperparathyroidism (pHPT) is a benign diseasewith malignant potential. Untreated, it is believed to carrya near 2-fold increase in development of several cancers (eg,breast, colon, prostate),1,2 and increases the risk of cardiac

isease, hypertension, and stroke by more than double,ltimately carrying a several-year decrease in life expec-ancy.3 Substantial if not severe osteoporosis will almostlways develop in affected patients, and 25% will get kid-ey stones. Besides the damage to many organ systems,

Disclosure Information: Nothing to disclose.Presented at the American College of Surgeons 97th Annual Clinical Con-gress, San Francisco, CA, October 2011.

Received October 24, 2011; Revised December 12, 2011; Accepted Decem-ber 14, 2011.From the Norman Parathyroid Center, Tampa, FL.Correspondence address: James Norman, MD, FACS, FACE, Norman Para-

ithyroid Center, 2400 Cypress Glen Dr, Wesley Chapel, FL 33544. email:[email protected]

260© 2012 by the American College of SurgeonsPublished by Elsevier Inc.

hese small parathyroid tumors typically cause chronic fa-igue, memory loss, and a host of other nonspecific symp-oms associated with a considerable decrease in quality ofife.4,5

Fortunately, pHPT is curable with removal of the sourceof excess parathyroid hormone (PTH). In about 76% ofpatients, the source is a single benign parathyroid tumor,and in the other 24% or so, there are 2, 3, or even 4overproducing glands that need to be removed—all or inpart.4 Because the number and location of the overproduc-ng glands are variable and usually not known before oper-tion, the surgical techniques for removing these parathy-oid tumors have historically included evaluation of all 4arathyroid glands to determine which of the glands needso be removed. As such, parathyroidectomy has necessi-ated general endotracheal anesthesia and bilateral explora-ion; an operation that typically includes a generous neck
ncision and several hours of operating time.
ISSN 1072-7515/12/$36.00doi:10.1016/j.jamcollsurg.2011.12.007

mailto:[email protected]

toghtrcre

261Vol. 214, No. 3, March 2012 Norman et al Abandoning Unilateral Parathyroidectomy

With the advent of the sestamibi scan in the early 1990scame the concept that a surgeon could know with highcertainty which patients harbored a single bad gland and itslocation. Armed with a localizing study, the operationcould be conducted with a unilateral approach, allowingfor a quicker, smaller operation. Our group was a principaladvocate for the unilateral approach, publishing numerousarticles between 1994 and 2005 touting the benefits of,and techniques used in, unilateral minimally invasiveparathyroidectomy.6-17 We published studies showing thathe vast majority of patients were best served by a unilateralperation if they had a localizing study showing a singleland, provided that some form of physiologic measure oformone production was performed in the operating roomo help assure that the source of the excess PTH had beenemoved. This method of unilateral operation has now be-ome common throughout the world, with expected cureates in the low to mid-90% range, depending on surgeonxperience.18-21

By 2005, volume of our practice had grown to �1,000parathyroid operations annually. With this volume camethe requirement of higher and higher cure rates, as a 3% to6% long-term failure rate causes considerably greaterfollow-up and management problems with such high sur-gical volumes. By 2010, our volume had exceeded 2,000parathyroidectomies per year, necessitating the very highestcure rates available. A confounding issue for our practicewas that many patients travel long distances to our center.A noncured patient who lives 3,000 miles away has differ-ent implications for our practice than that noncured pa-tient who lives within our community. Another confound-ing issue for our practice was the gradual but considerableincrease in referrals of patients with negative scans whowere not being afforded surgery by surgeons in their com-munity because they could not localize an adenoma preop-eratively. As our surgical volume increased, so too did thepercentage of complex and/or scan-negative patients.

During the course of performing thousands of parathy-roidectomies, it became clear that there was no maneuverwe could perform preoperatively or intraoperatively thatwould assure us that all abnormal parathyroid tissue hadbeen removed—short of the direct evaluation of the otherglands. Patients who we were convinced were cured someyears earlier were showing up again with abnormal calcium

Abbreviations and Acronyms

IOPTH � intraoperative parathyroid hormonepHPT � primary hyperparathyroidismPTH � parathyroid hormone

and/or PTH levels dictating that the other side of the neck

be opened to remove another tumor. In retrospect, thesetumors were present at the first operation, but were smallerand producing less PTH, so removal of the large parathy-roid tumor gave the appearance that the patient was cured.These patients did not have recurrent disease; they werenever completely cured at their first operation. These pa-tients had persistent disease, it just was not completelyapparent for some months or years after the operation.

It slowly become obvious within our practice that theonly way to achieve a near 100% long-term cure rate was toevaluate the physiologic activity of all 4 parathyroid glandsin virtually all people at their first operation. As such, wegradually gave up on unilateral parathyroidectomy in al-most every instance by the end of 2006. This study outlinesthe progression of this concept and the long-term results ofunilateral vs bilateral parathyroidectomy in 15,000 pa-tients. Lessons learned here have important implicationsfor all surgeons performing parathyroidectomy, regardlessof their level of expertise.

METHODSStudy designA retrospective analysis was conducted of prospectively col-lected data on 15,060 consecutive patients undergoingparathyroidectomy by one surgical group during an 18-year period ending April 2011. All patients had primaryhyperparathyroidism; patients with secondary (renal) ortertiary (post-transplantation) hyperparathyroidism arenot included. All patients were selected for surgery basedon the biochemical diagnosis of pHPT as outlined in detailby our group recently.4 Preoperative scanning was never usedas a determinant of surgery vs no surgery. As outlined here, wedo not perform any scan before the morning of surgery.Other patients not included in this study were those whohad undescended parathyroid tumors necessitating anonstandard neck incision high in the neck and those withmediastinal tumors necessitating some form of thoracot-omy, because evaluation of all 4 parathyroid glands is notpossible during these operations. All patients signed a con-sent for review of their clinical data, which was collected ina nonidentifiable fashion in accordance with principlesoutlined in the Declaration of Helsinki and as required forour IRB approval.

Patient groupsFor purposes of comparisons, the 15,060 patients havebeen divided into 2 groups: group 1 consists of 14,000patients undergoing their first parathyroid operation(which was performed by the authors). This group wasused to establish short and long-term cure rates for unilat-
eral vs bilateral operations within our practice. This group

loptrg

Wpbip

262 Norman et al Abandoning Unilateral Parathyroidectomy J Am Coll Surg

was also used to track operative times for unilateral vs bi-lateral operations as the surgeons gained experience. Group2 consists of 1,065 patients referred to our practice forpersistent pHPT. The subgroup being examined consists of233 patients who had 1 parathyroid adenoma removed atanother institution (proven by pathology through size,weight, and histology) using a unilateral operation; all hada �50% drop in serum PTH at 20 minutes (or later) afterremoval of the parathyroid adenoma.

Preoperative localizationBefore referral to our center, 19% of the 14,000 first-operation patients were sent to us with one (or more) pre-operative scans that were read as positive; 65% had one ormore negative scans, and 16% were referred without anylocalizing scans performed. All 15,060 patients in thisstudy underwent planar sestamibi scanning the morning ofthe operation as the only localizing study. We do not scanpatients before the morning of the operation. Single-photon emission computed tomography scanning wasused occasionally for the first 1,000 or so patients and thenabandoned in the mid-1990s. Ultrasonography is not usedby our group and was obtained on �12 patients. Similarly,MRI was never used. CT scanning (4-dimensional andwith single-photon emission computed tomography�fusion as appropriate) is only used in our practice for pa-tients with tumors found to be in the deep mediastinum onplanar sestamibi. CT scanning in its various forms is notused for routine preoperative localization and thereforenone of the 15,060 patients in this study had CT scansperformed. Selective venous sampling has never been usedin our practice for any patient.

Determinant of unilateral vs bilateral explorationA number of variables have been used by our group todetermine good vs poor candidates for unilateral parathy-roidectomy. A recent publication of ours describes thesevariables, identifying 18 specific objective parameters thathad a statistically significant impact on our decision toundertake or avoid a unilateral operation.22 In summary,patients who had a unilateral exploration were required tohave a clearly positive, in-focus sestamibi scan with solitarylocalization of radioactivity that is clearly distinct from thethyroid (vague hot areas behind one thyroid lobe were notsufficient to warrant a unilateral approach). In addition,the remainder of the scan must be without any hint ofabnormality, patients must be older than 25 years of age,have no family history of pHPT or MEN syndromes, nogoiter or enlarged thyroid on either side of the neck, no historyof lithium use, no history of neck radiation, and no history of
pituitary, adrenal, or pancreatic disease.
Performance of unilateral explorationSince our 500th operation, our approach to unilateral ex-plorations is exactly as it is for a bilateral approach, both areperformed through a 1-inch centrally placed transverse in-cision (incisions are never placed off to one side). Whenperforming a unilateral exploration, our policy is to iden-tify and examine the ipsilateral parathyroid gland; by pro-tocol, failure to identify an ipsilateral normal gland dictatesthat the operation be converted in almost all cases to abilateral exploration. Removing a single adenoma and clos-ing the wound is never done because we want to clear oneside of the neck in case the patient is not cured, leaving thesecond operation to be performed on only the contralateralside, as described previously.16,18 If the second gland is en-arged, nondormant, or in any other way abnormal, theperation is converted to a bilateral exploration. In ourractice, and throughout the findings of this report, pa-ients who underwent a unilateral operation had a parathy-oid adenoma removed and the ipsilateral parathyroidland examined and determined to be normal.

Performance of bilateral explorationBilateral operations are conducted through the same inci-sion used for unilateral operations; the incision is not en-larged.4,16,18 Similarly, the type of anesthesia is unchanged.

e have never use any form of deep venous thrombosisrophylaxis nor have we used central lines, arterial lines, orladder catheters. The positioning of the patient is supinen a slightly reverse Trendelenburg position with a rolllaced transversely under the shoulders for neck extension.

AnesthesiaGeneral endotracheal anesthesia was routinely used for thefirst 250 or so patients. Since that time, we have used la-ryngeal masked airway as our method of airway manage-ment, combined with IV sedation using propofol andmidazolam as the primary agents. Approximately 2.6% ofpatients are intubated, typically because of severe morbidobesity, as reported by our group previously.23 We havenever used any form of nerve-monitoring endotrachealtube. We made a few attempts at the use of local anesthesia(field blocks) in the late 1990s and quickly abandoned thisapproach. We have not used any form of local anesthesia inthe past 13,000 patients.

Measurement of parathyroid glandhormone productionAs previously reported by our group,24,25 we determinethe physiologic activity of each parathyroid gland during theoperation as each gland is encountered, relying on the
gland’s metabolic activity rather than anatomic appearance

fs

magas

p6o

tw

et

1


or cellularity under a microscope to determine if it is ab-normal. This 3-second computer analysis determines theamount of PTH being produced (quantitatively) by themeasurement of contained gamma radioactivity (countsper second) compared against a standard curve of hormoneproduction (reported as pg/mL), taking into considerationthe volume of distribution of the patient and radioactivedecay (half-life) since injection. In patients with pHPT,parathyroid glands can be quickly determined to be normal(dormant), a parathyroid adenoma, a hyperplastic parathy-roid, or a clinically enlarged, nondormant parathyroid.24,25

By protocol, all glands that are nondormant are removed,as described.25 Use of this technology allows us to eliminaterozen-section analysis. The use of intraoperative PTH as-ay (IOPTH) has never been used by our group.

Histology/pathology/photographyRoutine use of frozen-section analysis was discontinuedafter approximately 750 patients, and by 2,000 patients ithad been used on �1% of all parathyroid operations (typ-ically reserved for occasional intrathyroid parathyroid tu-mors).25,26 Permanent histology, however, is obtained on allspecimens. Similarly, for the last 8,000 patients, all re-moved parathyroid tumors were documented by photog-raphy. Weights of removed parathyroid tissues (of all types)have been recorded for all removed parathyroid glandsfrom all 15,000� patients in this study.

Operative times and length of hospital stayOperative times are recorded in the electronic medical re-cord system by the operating room nurse. Similarly, lengthof stay in the recovery room before discharge to home isrecorded by the nursing staff in the medical record. Virtu-ally all patients in this study were discharged to home di-rectly from the recovery room. The postoperative length ofstay is determined by the nursing staff, which is at all timesunaware which patients had unilateral operations andthose that had bilateral operations.

Follow-up and analysisFollow-up is 100% at 6 months, 99.5% at 1 year, 99% at 2years, and 97.7% at 5 years and consists of serial calcium,ionized calcium, and PTH. Minimum follow-up is 6months. Mean follow-up is 5.9 � 3.3 years (range 6

onths to 18 years). Data are expressed as mean � SD andnalyzed using SPSS 11.0 (SPSS Inc). Differences betweenroups were assessed by independent t-test, chi-squarenalysis, and ANOVA as appropriate; p � 0.05 was con-
idered significant.
RESULTSDisease profilesThe average age in this study was 59.9 � 12.9 years (range9 to 105 years). Women constituted 75.1% and 24.9%were men. The 3-year mean preoperative serum calciumlevel was 10.9 � 0.6 mg/dL and the 3-year preoperativePTH was 105.8 � 48 pg/mL. Mean highest calcium levelof each patient was 11.4 � 0.5 mg/dL and mean highestPTH was 115 � 52 pg/mL. A very detailed look at thebiochemical profiles from 10,000 of these patients can befound in a recent publication from our group.4

Rates of unilateral vs bilateral operations over timeFigure 1 shows the rate of bilateral vs unilateral operationsduring 18 years and 14,000 patients who underwent theirfirst operation with the authors (solid line, left-hand verti-cal axis). Bilateral operations were the rule early in ourexperience, which quickly turned to be predominately uni-lateral by the 600th patient (p � 0.0001). At that time, the

ercentage of unilateral explorations was at its highest at8%. During the next 15 years, the percent of unilateralperations decreased at every 500th patient (p � 0.05). By

the 5,500th patient, our indications for unilateral explora-ion were becoming more select; only 20% of operationsere unilateral (p � 0.001 vs 4,000). This trend contin-

ued, by our 10,000th patient, �10% had a unilateral op-ration. Currently, 3.3% of all patients referred to our prac-ice undergo a unilateral parathyroidectomy (p � 0.0001

vs 1,000 through 10,000). Importantly, during this timethe rate of positive sestamibi scans increased from 74.8% to80.1% (p � 0.02; all scans interpreted by a single author,

Figure 1. Rate of unilateral parathyroidectomy and long-term curerates for a single surgical practice performing 14,000 first-timeoperations. Unilateral parathyroidectomy (left vertical axis) was per-formed in 68% of patients early in our experience, decreasing sig-nificantly at every 1,000th operation, reaching a low of 3.3% at the4,000th operation (p � 0.0001). The corresponding cure rate (right

vertical axis) was at its lowest when the unilateral operative rate washighest, increasing at every 1,000th operation as the rate of bilat-eral operations increased (p � 0.0001).

JN). During the time when our percentage of positive ses-

adomuc

i

t(c0b

pml2

9

rv0gro(


tamibi scans increased, our rate of unilateral explorationdecreased.

Cure rates over timeAlso shown on Figure 1 (dashed line, right-hand verticalxis) is the cure rate over time for all 14,000 patients un-ergoing their first operation at our center (mean durationf follow-up is 5.9 � 3.3 years; minimum follow-up of 6onths; range 6 months to 18 years). As the percentage of

nilateral operations decreases, the long-term cure rate in-reases (p � 0.001). Despite developing a very conservative

approach to unilateral operations, we could not achieve�98% cure rate until our indications for unilateral opera-tions were so strict that �13% of our operations wereunilateral.

Cure rates over time for bilateral vsunilateral operationsThe cure rates for 14,000 patients undergoing their firstoperation at our center were divided into those undergoinga bilateral vs unilateral parathyroid operations. Figure 2shows the cure rates over 10 years, demonstrating that thevast majority of failures in our practice come from thegroup of patients who underwent a unilateral operation(p � 0.001 vs bilateral at all time points). Although not allpatients undergoing a bilateral operation are cured, thosethat are believed to be cured at the time of operation con-tinue to be cured long term (p � 0.93 for 1-year vs 10-yearcure rates for patients undergoing bilateral exploration).Once cured after undergoing a bilateral operation withexamination of all 4 glands, a second de novo parathyroidtumor almost never develops, even when patients are fol-lowed for 15 years (p � 0.88 year 1 vs 15 years). The same

Figure 2. Ten-year cure rates after unilateral vs bilateral parathy-oidectomy. The cure rates of 14,000 patients undergoing unilaterals bilateral exploration demonstrate significant differences (p �.001) at all time points. Long-term cure rates for patients under-oing a bilateral operation do not change over time (p � 0.93), andecurrent disease develops by 10 years post surgery in at least 5%f those believed to be cured after undergoing a unilateral operationp � 0.001 at 2 years and beyond).

s not true for patients undergoing unilateral exploration; r

heir cure rates are significantly less at the time of surgeryp � 0.001 compared with bilateral), and their cure ratesontinue to drop for at least the subsequent 10 years (p �.0001 unilateral at time 1 year vs unilateral at 2 years andeyond).

Operative times for bilateral vsunilateral operationsOperative times for the authors to complete bilateral andunilateral parathyroid operations are shown in Figure 3 as afunction of surgeon experience. As surgeon experience in-creases, operative times decrease (p � 0.001), with most ofthe decreases seen in the first 5,000 patients, but continu-ing beyond 14,000 patients. Unilateral parathyroidectomyalso becomes faster with surgeon experience (p � 0.001).The difference in time required to perform a bilateral vsunilateral operation decreases significantly (p � 0.05) atevery 500th patient. By the 14,000th patient, a unilateral

arathyroidectomy took the authors a mean of 16.4 � 3.3inutes (range 7 to 39 minutes, mode 14 minutes). Simi-

arly, a bilateral parathyroid operation took a mean of2.3 � 7.4 minutes (range 12 to 91 minutes, mode 18

minutes). By the 13,000th operation, the mean time differ-ence between a unilateral and bilateral operation is 5.9 �3.7 minutes.

Postoperative stay for bilateral vsunilateral operationsOnly 212 of the 15,060 patients in this study were kept inthe hospital overnight (1.4%); 143 of them had a concom-itant thyroidectomy and the need for overnight stay wasnot related to the extent of the parathyroid operation, asreported previously.27 Of the most recent 13,000 patients,9.6% were discharged from the recovery room. Mean du-

Figure 3. Mean operative time for bilateral vs unilateral parathyroid-ectomy for a single surgeon during 14,000 operations. Operativetimes for both surgical approaches decreased as experience wasgained (p � 0.0001). Similarly, the difference in the time required toperform a bilateral vs unilateral operation decreased at every1000th operation (p � 0.01).

ation of hospital stay (from admission into the recovery

wo3tas

wdtrms

mp9

T3

s was

78�


room to discharge from the hospital, which is at the discre-tion of the recovery room nurses) was 98 � 14 minutes forunilateral explorations and 100 � 15 minutes for bilateraloperations (p � 0.87).

Causes of failures in 14,000 first-timeparathyroid operationsAt the 500th patient, the failure rate (noncure, as definedhere previously) for bilateral explorations was 2.8%. At the14,000th patient, the noncure rate for bilateral explorations

as 0.7% (p � 0.001). The cause of the failure in bilateralperations was the lack of finding any abnormal gland in1% of noncured patients (known to be noncures at theime of surgery), and the failure to find a second (or third)bnormal gland in 69% (known to be noncures only afterome days or weeks after the operation was concluded).

Within the unilateral group, the cause of the noncureas exclusively due to the failure to recognize multiglandisease. Because our protocol has always been to (attempto) examine the ipsilateral gland during a unilateral explo-ation, 97.8% of failures were due to missed second (orore uncommonly, third) adenoma on the contralateral

ide. At the 1,000th patient, the 5-year cure rate for unilat-eral operations was 93.8%. At the 10,000th patient, using a

uch more conservative selection criteria for unilateral ex-loration, the cure rate for a unilateral exploration was7.2% (p � 0.001 vs 1,000th patient). Regardless of how

conservative our application of the unilateral techniqueeven after 10,000 operations, we could not achieve cure

Table 1. Number of Parathyroid Glands Removed in Unilat

VariableUnilateral exploratio

(n � 3,000)

Single gland removed, % 96.9Two glands removed, % 3.1

hree glands removed, % 0.5 glands removed, % 0

The number of parathyroid glands removed during 3,000 unilateral operationoperations.

Table 2. Findings at Second Parathyroid Operation after �

Drop in PTH, %Patients

n %

50�59 49 2160�69 81 350�79 61 260�89 30 1390 12 5

A total of 233 patients were referred for a second operation following the rem
minutes post adenoma resection.PTH, parathyroid hormone.
rates equal to that seen with a bilateral approach (p �0.001).

Number of glands removedThe number of parathyroid glands removed during themost recent 3,000 bilateral operations is compared with themost recent 3,000 unilateral operations in Table 1 (to elim-inate surgeon experience as a variable). The number ofabnormal glands removed when performing a bilateral op-eration is dramatically higher than during a unilateral op-eration, even for experienced surgeons using very conser-vative criteria for which patients are selected for unilateralexploration. When all 4 parathyroid glands are examinedand their physiologic activity determined, we remove morethan one gland in 24.7% of patients. In patients with aclearly positive sestamibi scan (meeting our criteria for aunilateral exploration) who undergo bilateral exploration,we remove more than one gland in 19.6% of patients.

Second adenomas removed after a 50% or moredrop in intraoperative PTHTable 2 shows the percent drop in PTH measured during233 unilateral operations performed at another institution,which were subsequently referred to our center for reopera-tive surgery because of noncure. All patients in this grouphad a �50% drop in PTH at 20 minutes (or more) postex-cision of an adenoma (required to be in this study group)with all but 9 (96%) having the IOPTH level fall into thenormal range. Before their first operation, these 233 pa-

s Bilateral OperationsBilateral exploration

(n � 3,000) p Value

75.3 �0.000116.7 �0.00015.1 �0.0012.9 0.03

dramatically lower than the number of glands removed during 3,000 bilateral

Drop in Parathyroid Hormone during First OperationFindings at second operation

1 additionaladenoma

2 additionaladenomas p Value

39 10 0.00675 6 �0.00158 3 �0.00128 2 0.00512 0 0.11

f one parathyroid adenoma and a greater than 50% drop in PTH levels at 20

eral vn

50%

oval o

(agdooIAaotpc0

sdf

faptairfeorcrp

tsr�ctlp

pcStcp

ppwut


tients had mean highest serum calcium and PTH levels of11.3 � 0.4 mg/dL, and PTH of 121 � 28 pg/mL, respec-tively. By 1-year postoperation, the mean serum calciumand PTH levels were 11.0 � 0.3 mg/dL and 94 � 29pg/mL. Mean time between first operation and referral toour center for a second operation was 2.1 � 0.8 yearsrange 2 weeks to 8.5 years). Table 2 also shows the findingst the second operation where the other 3 parathyroidlands were evaluated. The drop in PTH values �50%uring the first operation did not prove that there were nother glands, nor was it predictive in any way of how manyther abnormal glands were present. Even patients withOPTH levels falling �90% had missed second adenomas.s seen in Table 2, the higher the percentage of PTH dropfter removal of an adenoma, the higher the chance thatnly one additional bad gland would be present if the pa-ient was not cured. After their second operation, all 233atients were cured (as defined here) with the mean cal-ium and PTH levels 1-year postoperatively being 9.6 �.1 mg/dL and 32 � 7 pg/mL, respectively.

DISCUSSIONAlthough our group was a dominant driving force for uni-lateral parathyroidectomy through the 1990s and early2000s, we have all but abandoned this approach. As wehave tweaked and manipulated the indications and tech-niques used in parathyroid surgery for many thousands ofpatients, we have concluded that the promises of unilateralexplorations do not live up to the hype for the long term.This is especially true when all patients with pHPT aregiven the opportunity to have a curative operation, even iftheir localizing scans are negative. We believe there is a rolefor unilateral exploration, however, its role at high-volumecenters or centers of excellence should be minimized.

The underlying precept of a unilateral parathyroid op-eration is the conviction that some form of preoperative orintraoperative testing can be trusted to rule out any addi-tional abnormal, overproductive parathyroid gland(s). Ourexperience documented in this report shows that no matterwhat the imaging techniques (and the expertise of the teamperforming the scans), the status of all 4 glands cannot beknown preoperatively. Additionally, we also provide evi-dence that there is no intraoperative measure or test thatcan be performed, short of examining all 4 parathyroidglands, which can assure no other abnormal, overproduc-ing parathyroid gland is present. Our experience, alongwith the first-hand experience of many others,28-32 hashown that IOPTH assays cannot be used to definitivelyetermine the status of other parathyroid glands and there-
ore, cannot assure cure. It has become apparent after per-
orming �17,000 parathyroid operations; the only way tochieve 10-year cure rates �94.5% (even by the most ex-erienced parathyroid surgeons), is to examine all 4 para-hyroid glands in nearly all patients, avoiding the allure andttraction of the unilateral approach. The recent literatures full of reports from numerous groups touting high cureates with unilateral operations using IOPTH assays. Un-ortunately, these patients are highly selected in nearly ev-ry case, with patients having a negative scan (or scans) notperated on and therefore not included in the analysis. Theeader is encouraged to examine the operative selectionriteria in all these reports, as nearly 20% of our referralsesult from other surgeons refusing to operate without aositive scan.The ultimate test for any approach to parathyroidec-

omy is long-term cure of the disease. Our group uses thetrictest definition of cure, ie, serum calcium level mustemain �10.0 mg/dL and the serum PTH must remain

65 pg/mL. Although some authors would argue thatured patients can often have elevated PTH levels for someime after surgery,33,34 we rarely see this phenomenonargely because of our aggressive calcium supplementationostoperatively.35,36 Therefore, any patients with persistent

elevations of PTH postoperatively (�75 pg/mL) are con-sidered noncured by our group until proven otherwise.These uncommon patients are followed closely for yearsuntil the final disposition is known. Similarly, after evalu-ating tens of thousands of patients with pHPT, we havelearned that adults older than age 30 years with normalparathyroid function should have calcium levels in the 9s,not in the 10s (regardless of the upper limit of normal,which includes children and teens).4 That is, frequent or

ersistent calcium levels (in an adult) �10.0 mg/dL indi-ate that a parathyroid tumor is present in nearly all cases.eeing calcium levels in the 10s in any postoperative pa-ient indicates to us that the patient is not cured (unless theorresponding PTH is �15 pg/mL) and additional over-roducing parathyroid tissue is presumed to be present.Figure 1 illustrates the progress of our thoughts about

arathyroid surgery and our quest of many thousands ofatients to find a way to perform a unilateral operation thatould yield a long-term cure rate of �99%. To avoid fail-res, we are developing protocols that dictate when an at-empted unilateral operation be converted to bilateral.22

Even with this 11% conversion rate of good candidates forunilateral operation to a bilateral operation, we could notget our 6-month cure rates �96.5% or our 3-year curerates �95%. Our analysis showed that virtually all non-cures came from patients who had a unilateral operation.As illustrated in Figure 1, as our experience grew we con-
tinuously decreased the indications for unilateral explora-

aanittm

secspa

sttocaltppsg


tion, and our rate of bilateral exploration increased. By the11,000th operation, we had decreased the indications forunilateral exploration so dramatically that we essentiallygave up on this approach altogether. It is important toremember that we use no qualifying test to determine whois operated on and who is not. That is, this report includesall types of pHPT from normocalcemic pHPT to severehypercalcemic pHPT, all of whom are operated on withoutthe requirement for any type of positive localizing study.The only patients we consider for a unilateral approachwould be those with special circumstances, such as follow-ing a previous neck dissection for cancer, or a patient whohas undergone a previous thyroidectomy.

Figure 2 shows a dramatic decrease in total operativetimes, factored heavily into our decisions to eliminate uni-lateral operations. Through thousands of operations welearned a number of techniques that allowed us to performbilateral operations faster than we could perform unilateraloperations some years earlier. As the time differential be-tween a unilateral and bilateral operation decreased to amatter of 6 minutes or so, the rewards of performing aunilateral exploration became less attractive. Clearly, themost important aspect of rapid parathyroid surgery hasbecome the use of the gamma probe as the single determi-nant of parathyroid gland activity and, therefore, the ulti-mate decision maker of whether a parathyroid gland isabnormal and removed or normal and left in place. This3-second analysis completely eliminates frozen-sectionanalysis and the time that it requires, and eliminates thevague, unhelpful information it typically provides.37 Thectual surgical techniques we have learned are the subject ofn upcoming series of technical articles but, as stated inumerous publications, we do not put the gamma probe

nto the wound as is commonly thought—we do not usehe probe to find parathyroid tumors. The probe is not aumor-detecting tool, but rather it is an ex vivo hormone-easuring device.24,25

Probably the single most important finding of this studyis that patients believed to be cured after a unilateral explo-ration can turn out to be only “better” and not cured, ascure rates after unilateral exploration continue to decreasefor at least 10 years. Long-term cure rates are affected pri-marily by the following 2 variables: experience of the sur-geon and extent of the operation. In this study, one surgeon(JN) participated directly in �98% of the operations, sourgeon expertise is eliminated as a variable except as thexperience increases with time. The difference in long-termure rates between unilateral and bilateral explorations ishown in Table 1 and is due to the number of abnormalarathyroid glands removed when all 4 parathyroid glands
re evaluated. We have come to the conclusion that there is
imply no mechanism to assess the functional status ofhese other glands without physically doing so. Althoughhere might be some controversy about our nomenclaturef these additional abnormal glands that are removed (be-ause we do not use histology and all historical reports ofbnormal parathyroid glands use histology and gland cel-ularity as the backbone of their nomenclature system),here is no doubt that these glands are enlarged and over-roducing hormone. In fact, our removal of more than onearathyroid gland in nearly 25% of cases is in line (orlightly higher) than that seen with other experiencedroups when all 4 glands are assessed.28,29 Importantly, the

literature is now full of reports of patients cured after theirunilateral parathyroid operation, yet the follow-up innearly all of these reports is �1 year. This study demon-strates that 1-year cure rates are meaningless in patientsundergoing unilateral parathyroidectomy.

Unilateral parathyroidectomy with the removal of oneparathyroid tumor followed by IOPTH assay to determinea 50% drop in PTH as an indicator of cure has becomepopular during the past decade because it allows surgeonsof various experience levels to perform a successful opera-tion on a patient with a positive scan. However widespreadthis method is, it remains controversial and discounted bysome experts because of its associated false-positive andfalse-negative rates.29 Removal of one parathyroid tumor(typically without proving the ipsilateral gland is normal)flies in the face of what is known about the number ofabnormal glands removed when all 4 glands are assessed.Most surgical reports throughout the history of this diseaseshow that somewhere between 20% and 25% of patientshave more than one parathyroid gland removed when all 4glands are examined, which is in line with the nearly 25%of patients who have more than one gland removed in thecurrent report (with bilateral exploration). Other studieshave used IOPTH assays after adenoma removal and beforeevaluation of all 4 glands and have shown that the IOPTHassay underestimates the presence of additional parathy-roid tumors in 16% of cases.29 Although the current studydoes not take into account all of the unilateral parathyroidoperations performed by surgeons who were associatedwith failures sent to our center—and the overall cure rate isnot known—it does include the largest cohort of uncuredparathyroid patients who had a �50% drop in PTH dur-ing their operation. Using IOPTH during the operationappears to simply tell the surgeon that they removed aparathyroid tumor that is overproducing PTH—a fact thatthe surgeon usually already knew. That is, a drop inIOPTH of �50% and into the normal range tells the sur-geon that the removed gland is a parathyroid adenoma that
is making a large amount of PTH. It does not necessarily

1

1

1

1

1

1

1


provide comment on the status of the other 3 glands. Itseems clear that a drop in serum PTH by 50% (or even90%) simply cannot definitively determine that all of theother glands are dormant (normal).

Unilateral parathyroidectomy has 2 major benefits thatcannot be denied. First, any patient who is not cured aftersuch an approach is able to undergo a simple, noncomplexsecond operation that is performed in virgin tissues. Sec-ondly, the unilateral approach allows many surgeons with alack of specific training in parathyroid surgery to take careof this large body of patients without the need for patientsto travel to a center of excellence or be forced to undergo alarge exploratory operation that has lower chances of excel-lent outcomes in the hands of inexperienced surgeons.38,39

Surgeons using the unilateral approach to parathyroidsurgery (regardless of the hormone adjuncts used intraop-eratively) must understand that the long-term cure ratefrom this technique will rarely exceed 95% and that �5%of patients who are believed to be cured at the time ofsurgery will have a recurrence in the ensuing 10 years.Patients (and their referring doctors) should be made awarethat they have a considerable chance of requiring a secondoperation and that closer long-term follow-up is necessarymore than for patients who are cured after having all 4parathyroid glands assessed. This phenomenon is illus-trated in Figure 2, where patients believed to be cured aftera 4-gland operation are almost always cured long-term, butpatients who are believed to be cured after a unilateraloperation must be followed much longer because they canpresent with persistent/recurrent disease for at least 10years.

Unfortunately, the quest for unilateral operations has atremendous downside that is rarely discussed: patients areincreasingly denied curative surgery simply because theirdoctors cannot locate an offending adenoma on a preoper-ative scan. The phenomenon of “watching the disease untilthe scan becomes positive” is not in the best interest of ourpatients and was the topic of a recent editorial by our groupstating quite simply that “the quest for a mini-unilateraloperation has gone too far.”40 The current report demon-strates that patients with negative localizing studies canhave a minimally invasive operation. In fact, virtually all ofthe last 8,000 patients at our center have had the exact sameoperation regardless of scan findings—we simply do notcare if their scans are positive or negative. This report alsomakes it clear that an operation does not have to be unilat-eral to be minimally invasive. Unilateral parathyroidec-tomy is not the panacea and it should not be the goal ofsurgeons performing parathyroid surgery. In fact, unilateral
parathyroidectomy in our practice is essentially dead.
Author ContributionsStudy conception and design: Norman, PolitzAcquisition of data: Norman, Lopez, PolitzAnalysis and interpretation of data: Norman, Lopez, PolitzDrafting of manuscript: Norman, PolitzCritical revision: Norman, Lopez, Politz

REFERENCES

1. Almquist M, Manjer J, Bondeson L, Bondeson AG. Serum calciumand breast cancer risk: results from a prospective cohort study of7,847 women. Cancer Causes Control 2007;18:595–602.

2. Norenstedt S, Granath F, Ekbom A, et al. Breast cancer associ-ated with primary hyperparathyroidism: a nested case controlstudy. Clin Epidemiol 2011;25:103–106.

3. Piovesan A, Molineri N, Casasso F, et al. Left ventricular hyper-trophy in primary hyperparathyroidism. Effects of successfulparathyroidectomy. Clin Endocrinol 1999;50:321–328.

4. Norman J, Goodman A, Politz D. Calcium, parathyroid hor-mone, and vitamin D in patients with primary hyperparathy-roidism: normograms developed from 10,000 cases. EndocrPract 2011;17:384–394.

5. Espiritu RP, Kearns AE, Vickers KS, et al. Depression in primaryhyperparathyroidism: prevalence and benefit of surgery. J ClinEndocrinol Metab 2011 Sep 14. [Epub ahead of print].

6. Norman J, Albrink M. Minimally invasive videoscopic parathy-roidectomy: a feasibility study in dogs and humans. J Laparoen-dosc Adv Surg Techn 1997;7:301–306.

7. Norman J, Chheda H. Minimally invasive parathyroidectomyfacilitated by intraoperative nuclear mapping. Surgery 1997;122:998–1004.

8. Norman J. The technique of intraoperative nuclear mapping tofacilitate minimally invasive parathyroidectomy. Cancer Con-trol 1997;4:500–504.

9. Norman J, Chheda H, Farrell C. Minimally invasive parathyroid-ectomy: reducing operative time and potential complications whiledecreasing potential complications. Am Surg 1998;5:391–396.

0. Denham D, Norman J. Cost-effectiveness of preoperative sesta-mibi scan for primary hyperparathyroidism is dependent solelyupon surgeon’s choice of operative procedure. J Am Coll Surg1998;186:293–304.

1. Norman J, Denham D. Minimally invasive radioguided para-thyroidectomy in the reoperative neck. Surgery 1998;124:1088–1093.

2. Murphy C, Norman J. The 20 percent rule: a simple instanta-neous radioactivity measurement defines cure and allows elimi-nation of frozen section and hormone assays during parathy-roidectomy. Surgery 1999;126:1023–1029.

3. Costello D, Norman J. Minimally invasive radioguided parathy-roidectomy. Surg Oncol Clin N Am 1999;8:555–564.

4. Norman JG. Minimally invasive radioguided parathyroidectomy:an endocrine surgeon’s prospective. J Nucl Med 1999;39:15N–24N.

5. Norman J, Murphy C, Chheda H.The false positive parathyroidsestamibi: a real or perceived problem and a case for radioguidedparathyroidectomy. Ann Surg 2000;231:31–37.

6. Murphy C. Minimally Invasive radioguided parathyroidectomy(MIRP). In: Whitman E, ed. Operative Techniques in General
Surgery. New York: Saunders; 2001:66–79.

1

1

2

2

2

2

2

2

2

2

2

2


17. Gallagher S, Denham D, Norman J. The impact of minimallyinvasive parathyroid surgery on the way endocrinologists treathyperparathyroidism. Surgery 2003;134:910–917.

8. Norman J, Rubello D, Giuliano A, Mariani G. Minimally inva-sive radioguided parathyroidectomy in primary hyperparathy-roidism. In: Mariani G, Strauss HW, Guiliano A, eds. Radiogu-ided Surgery: A Comprehensive Team Approach. New York:Springer; 2008:226–232.

9. Norman J. Minimal parathyroid surgery: recent trends becom-ing standard of care yielding smaller, more successful operationsat lower cost. Otolaryngol Clin N Am 2004;37:683–688.

0. Greene AB, Butler RS, McIntyre S, et al. National trends inparathyroid surgery from 1998 to 2008: a decade of change.J Am Coll Surg 2009;209:332–343.

1. Richards ML, Thompson GB, Farley DR, Grant CS. An opti-mal algorithm for intraoperative parathyroid hormone monitor-ing. Arch Surg 2011;146:280–285.

2. Norman J, Politz D. Prospective study in 3,000 consecutiveparathyroid operations demonstrates 18 objective factors thatinfluence the decision for unilateral versus bilateral surgical ap-proach. J Am Coll Surg 2010;211:244–249.

3. Norman J, Aronson K. Outpatient parathyroid surgery and dif-ferences seen in the morbidly obese patient. Otolaryngol HeadNeck Surg 2007;136:282–286.

4. Norman J, Politz D. Measuring individual parathyroid glandhormone production in real-time during radioguided parathy-roidectomy. Experience in over 8000 operations. Minerva En-docrinol 2008;33:147–157.

5. Norman J, Politz D. 5000 parathyroid operations without fro-zen section or PTH assays: measuring individual parathyroidgland hormone production in real-time. Ann Surg Oncol 2009;16:656–666.

6. Goodman A, Politz D, Lopez J, Norman J. Intrathyroid para-thyroid adenoma: incidence and location—the case against thy-roid lobectomy. Otolaryngol Head Neck Surg 2011;144:867–871.

7. Norman J, Politz D. Safety of immediate discharge followingparathyroidectomy: a prospective study in 3000 consecutive pa-tients. Endocr Pract 2007;13:105–113.

8. Siperstein A, Berber E, Mackey R, et al. Prospective evaluationof sestamibi scan, ultrasonography, and rapid PTH to predictthe success of limited exploration for sporadic primary hyper-parathyroidism. Surgery 2004;136:872–880.

9. Siperstein A, Berber E, Barbosa GF, et al. Predicting the success
of limited exploration for primary hyperparathyroidism using
ultrasound, sestamibi, and intraoperative parathyroid hormone:analysis of 1158 cases. Ann Surg 2008;248:420–428.

30. Chiu B, Sturgeon C, Angelos P. Which intraoperative parathy-roid hormone assay criterion best predicts operative success? Astudy of 352 consecutive patients. Arch Surg 2006;141:483–487.

31. Karakousis GC, Han D, Kelz RR, et al. Interpretation of intra-operative PTH changes in patients with multi-glandular pri-mary hyperparathyroidism (pHPT). Surgery 2007;142:845–850.

32. Kebebew E, Hwang J, Reiff E, et al. Predictors of single-gland vsmultigland parathyroid disease in primary hyperparathyroid-ism: a simple and accurate scoring model. Arch Surg 2006;141:777–782.

33. Ning L, Sippel R, Schaefer S, Chen H. What is the clinicalsignificance of an elevated parathyroid hormone level after cu-rative surgery for primary hyperparathyroidism? Ann Surg2009;249:469–472.

34. Biskobing DM. Significance of elevated parathyroid hormoneafter parathyroidectomy. Endocr Pract 2010;16:112–117.

35. Vasher M, Goodman A, Politz D, Norman J. Postoperative cal-cium requirements in 6000 patients undergoing outpatientparathyroidectomy: easily avoiding symptomatic hypocalcemia.J Am Coll Surg 2010;211:49–54.

36. Press D, Politz D, Lopez J, Norman J. The effect of vitamin Dlevels on postoperative calcium requirements, symptomatic hy-pocalcemia, and PTH levels in following parathyroidectomy forprimary hyperparathyroidism. Surgery 2011;150:1061–1068.

37. Carneiro-Pla DM, Romaguera R, Nadji M, et al. Does histopa-thology predict parathyroid hypersecretion and influence cor-rectly the extent of parathyroidectomy in patients with sporadicprimary hyperparathyroidism? Surgery 2007;142:930–935.

38. Mitchell J, Milas M, Barbosa G, et al. Avoidable reoperations forthyroid and parathyroid surgery: effect of hospital volume. Sur-gery 2008;144:899–906.

39. Soon PS, Yeh MW, Sywak MS, et al. Minimally invasive para-thyroidectomy using the lateral focused mini-incision approach:is there a learning curve for surgeons experienced in the openprocedure? J Am Coll Surg 2007;204:91–95.

40. Norman J. Controversies in parathyroid surgery: the questfor a “mini” unilateral operation seems to have gone too far.J Surg Oncol 2012;105:1–3. doi: 10.1002/jso.22040. Epub
2011 Aug 31.

abandoning unilateral parathyroidectomy: why we reversed our position after 15,000 parathyroid...

Documents