| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Special Articles |
Departments of Surgery (J.A.S., R.U., M.A.Z.), Medicine (N.R.P., M.A.L.), and Pathology (M.A.L.) and the Robert Wood Johnson Clinical Scholars Program (J.A.S., N.R.P.), The Johns Hopkins University School of Medicine, and the Departments of Epidemiology (N.R.P.) and Health Policy and Management (N.R.P.), The Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland 21287
Address all correspondence and requests for reprints to: Martha A. Zeiger, M.D., Division of Endocrine and Oncologic Surgery, Carnegie 681, The Johns Hopkins Hospital, 600 North Wolfe Street, Baltimore, Maryland 21287-8611. E-mail: mzeiger{at}welchlink.welch.jhu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of 190 surgeons surveyed, 147 (77%) responded; 109 provided complete responses (57%). These surgeons spend 66% of their time in patient care and perform an average of 33 (range, 1–130) parathyroidectomies/yr. More than 72% of 1° HPT patients who underwent surgery were asymptomatic or minimally symptomatic. High volume surgeons (>50 cases/yr) had significantly lower thresholds for surgery with respect to abnormalities in preoperative creatinine clearance, bone densitometry changes, and levels of intact PTH and urinary calcium compared to their low volume colleagues (1–15 cases/yr). Overall reported surgical cure rates were 95.2% after primary operation and 82.7% after reoperation. Compared to high volume surgeons, low volume endocrine surgeons had significantly higher complication rates after primary operation (1.9% vs. 1.0% respectively; P < 0.01) and reoperation (3.8% vs. 1.5%; P < 0.001) as well as higher in-hospital mortality rates (1.0% vs. 0.04%; P < 0.05).
Endocrine surgeons operate on a large number of asymptomatic or minimally symptomatic 1° HPT patients. Even among a group of highly experienced surgeons who typically see patients after referral from endocrinologists, clinical outcomes and criteria for surgery vary widely and appear to be associated with surgeon experience. Their criteria for surgery diverge from NIH guidelines. These results implore the endocrine community to examine the evidential basis for decisions made in the management of 1° HPT.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A 1991 NIH Consensus Development Conference provided recommendations for the management of asymptomatic and minimally symptomatic 1° HPT (5), but there continues to be controversy regarding the appropriateness of criteria for medical therapy vs. surgical intervention. Thus, the optimal clinical management of these asymptomatic patients is not yet established. The relatively stable clinical course observed in these patients has challenged previous beliefs that 1° HPT is a progressive disease with osseous, renal, gastrointestinal, and neurological complications (6). However, recent data suggest that many of these asymptomatic patients actually have subtle neuropsychiatric impairments (7, 8) as well as progressive cardiovascular disease resulting in premature death (9, 10). At the same time, surgical care of 1° HPT is associated with marked increases in bone mass (11, 12) and improvement of many symptoms associated with 1° HPT (13, 14, 15). Moreover, surgery in the hands of experienced surgeons is rarely associated with complications (16, 17, 18). Although technically challenging, even reexploration for persistent 1° HPT and parathyroidectomy in the elderly (>70 yr of age) can yield surgical cure rates in excess of 94% (19, 20).
Variation in surgical practice patterns raises concern about increased cost and inconsistent quality of care. Understanding current practice patterns and the reasons for them is an important first step in defining indications for new management strategies as well as guiding appropriate and cost-effective management of these patients. The aim of this study was to assess physician decision-making regarding the surgical management of 1° HPT and, in particular, to measure adherence to the 1991 Consensus Conference statement. We examined endocrine surgeons’ thresholds for surgery and their relationship to surgeon experience. We also examined reported surgical outcomes, including complication rates and in-hospital mortality. Several factors may confound these assessments, so consideration was given to controlling for additional physician, patient, and practice characteristics.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A national, cross-sectional survey of all North American members of the American Association of Endocrine Surgeons (AAES) was conducted between May and July 1997. Surveys were distributed to members attending the society’s annual meeting and then by facsimile to nonattendees. Telephone calls were made as reminders to nonrespondents, and a second wave of surveys was sent to nonrespondents by facsimile. All responses were confidential. Unique numerical identifiers were used to track respondents and nonrespondents. These were removed after assembly of the database and before analysis.
Survey development
The survey instrument, which inquired about surgeon and patient characteristics, thresholds for surgery, and clinical outcomes, was developed through the collaborative efforts of endocrine surgeons, endocrinologists, internists, and health services researchers. The study protocol was approved by the Joint Committee on Clinical Investigation at Johns Hopkins University; a preliminary instrument was piloted by facsimile to a nationally representative sample of members of the AAES in December 1996.
Survey content
The survey included seven general areas. First, we inquired
about respondents’ age, gender, and place of residence. Second, we
asked about the clinical presentation of patients with 1° HPT,
including their symptomatology, previous medical therapy, and
underlying pathology (single vs. double adenoma,
hyperplasia, or carcinoma). Third, we inquired about surgeons’
experience, including whether they had completed advanced training in
endocrine surgery, where and when they had trained, and how many
parathyroid and endocrine operations they had performed while in
surgical training. Fourth, we asked surgeons to describe their surgical
practice, including annual case load of parathyroidectomies, percentage
of their practice spent performing parathyroid and endocrine surgery,
and percentage of time spent with clinical, research, administrative,
and teaching activities. Fifth, we assessed surgeons’ thresholds for
surgery by asking them whether they would operate on a patient with
1° HPT based on the results of different levels of relevant
preoperative laboratory tests and clinical symptoms. In many instances,
these laboratory and clinical criteria were borrowed from the
recommendations made by the 1991 NIH Consensus Development Conference.
They were divided into low, medium, and high thresholds for surgery to
simplify the analysis (see Table 1
).
Separate thresholds were established for individual criteria, whereas
combinations of laboratory and clinical abnormalities were not
examined. Sixth, we asked surgeons to report their surgical outcomes,
including rates of eucalcemia, hypercalcemia, and hypocalcemia 6 months
postoperatively; minor and major complication rates; and in-hospital
mortality rates. Finally, we asked surgeons about the health care
environment in which they worked, including the degree to which managed
care had penetrated their practice.
|
Statistical analysis
The distributions of surgeons’ characteristics among volume
groups were compared using the 
2 statistic or Fisher’s
exact test (when appropriate) for categorical variables, including
thresholds for surgery, and ANOVA for continuous variables. Bivariate
analyses using linear regression were used to determine the association
between particular variables and our outcomes of interest. Multiple
linear regression was used to assess how outcomes differed among
surgeon volume groups after adjusting for possible confounding
variables such as surgeon’s gender, years in practice, geographic
location, and completion of advanced endocrine surgery training;
patient’s severity of illness as measured by their laboratory values
and clinical symptoms at presentation (asymptomatic, minimally
symptomatic, or frankly symptomatic); and managed care penetration.
As complication and in-hospital mortality rates as well as length of stay were skewed, a natural log transformation was performed to obtain a normal distribution. All statistical inferences pertaining to complication rates, death rates, and length of stay were based on the log-transformed data. All P values reported were the results of two-sided tests, and P > 0.05 was considered nonsignificant. Data were analyzed using STATA 5.0 (STATA release 5, College Station, TX).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of the 231 current members of the AAES, 24 corresponding members live overseas and did not receive questionnaires. Of the 207 North American members surveyed, 17 were excluded from further analysis: 10 were retired, and 7 no longer perform endocrine surgery or specialize in laboratory medicine or pathology. Of the remaining 190 members, 147 responded, yielding a response rate of 77.4%. There were 38 surveys that were incomplete, and these were not considered in the primary analysis (full-completion response rate, 57.4%). Overall, respondents appeared to be similar to nonrespondents in terms of gender (94.5% of nonrespondents were male) and geographic distribution, except that Western AAES members were underrepresented among respondents (7.6% vs. 16.7% of nonrespondents).
The demographic characteristics of the respondents are listed in Table 2. Briefly, the respondents were highly
experienced, with an average of nearly 20 yr in practice; highly
specialized, with an average of more than 40% of their practice in
parathyroid and endocrine surgery; and clinically active, with an
average of two thirds of their time spent in patient care. Nearly half
were in practice with more than 30% managed care penetration. The
overwhelming majority were male. On the average, male surgeons had been
in practice longer than their female colleagues (20.2 vs.
14.6 yr), and probably as a consequence of this were more active
clinically in parathyroid surgery (22.5% of the men vs.
10% of the women perform >50 parathyroidectomies/yr). This may be
explained by the relatively new trajectory for women in surgery. More
of the female endocrine surgeons completed advanced training (60% of
women vs. 22.8% of men), and more of their clinical
practice was restricted to endocrine and parathyroid surgery (60% for
women vs. 39% for men).
|
The majority of 1° HPT patients referred for surgery were asymptomatic (34.8%) or minimally symptomatic (37.4%); only 28.2% were frankly symptomatic. Most patients (81.8% of asymptomatic patients, 95.4% of minimally symptomatic patients, and 99.5% of frankly symptomatic patients) underwent parathyroidectomy. The majority of patients referred for surgery had been followed with laboratory surveillance and no medical therapy (69.8% of women and 91.2% of men). Sixty-two percent of referrals originated from endocrinologists, 21% from internists, 12% from family practitioners, and 5% from other surgeons. Parathyroid adenomas were found in 81.2% of patients at first operation and 70.3% at reoperation, whereas parathyroid hyperplasia was found in 14.9% of patients at first operation and 26.2% at reoperation.
Thresholds for surgery
There was wide variation among respondents about the criteria used
to decide whether to operate on 1° HPT patients. Table 3 shows the clinical and laboratory
abnormalities that, alone, endocrine surgeons reported would be
sufficient for recommending surgery. More than half of endocrine
surgeons were in agreement with the NIH Consensus Conference statement
regarding only four of the 12 laboratory or clinical findings for
patients aged 65 yr and older (see numbers in italics in
Table 3): one vertebral fracture (95.9%), one episode of
nephrolithiasis (85.4%), asymptomatic pancreatitis (65.5%), and a
minimal elevation of the ionized calcium level (62.8%).
|
There appears to be a consistent association between a surgeon’s
threshold for surgery and the number of parathyroid procedures the
surgeon performs each year. Overall, endocrine surgeons who performed
more parathyroidectomies had lower thresholds than their colleagues who
performed fewer cases. Figure 1, A–D,
shows that there was an association between volume of surgery performed
and threshold for surgery across a range of kidney and bone
manifestations of 1° HPT. This pattern was present for 8 of the 12
criteria used in younger patients (intact PTH, serum calcium, ionized
calcium, 24-h urinary calcium, creatinine clearance, bone densitometry
T-score, 6–12 month change in bone densitometry score, and
gastrointestinal symptoms) and was statistically significant for 5.
This phenomenon also was seen with older patients; lower thresholds
among higher volume surgeons were seen for 9 of 12 criteria (all of the
same criteria as for younger patients plus pancreatitis), but it was
statistically significant for only 2 (bone densitometry T-score and
gastrointestinal symptoms). No significant association was observed
between the thresholds for surgery and the degree of managed care
penetration.
|
Self-reported surgical success rates among survey respondents were
significantly higher for primary operations than for reoperations,
whereas complication rates were significantly higher for reoperations
(Table 4). Major complications were
defined as permanent recurrent laryngeal nerve injury, hematoma
requiring surgical evacuation, and perioperative myocardial infarction;
minor complications included wound infections and urinary retention.
In-hospital death rates were negligible and were comparable for primary
operations and reoperations. More than 95% of first time surgical
patients and 83% of reoperations were reported to have been cured of
1° HPT 6 months postoperatively. Minor complications, reported to be
about 3%, were nearly 3 times as common as major complications for
primary operations.
|
);
specifically, surgical cure rates and complication rates after primary
operations were associated with surgeon case load, whereas in cases of
reoperation only complication rates were associated with case load. As
surgeon case load increased, outcomes improved, and surgeons who
performed more than 50 parathyroidectomies/yr had significantly lower
major complication rates after primary operation than surgeons who
performed fewer than 15 parathyroidectomies/yr. This difference in
major complication rates was also striking after reoperation (1.48%
vs. 3.76%). There were significant differences among the
self-reported in-hospital mortality rates seen with the highest, medium
(15–49 cases/yr), and low volume surgeons (0.04% vs.
0.73% and 1.0%, respectively).
|
|
|
When sensitivity analyses were performed with the alternative classification of surgeons by parathyroidectomy volume (<10, 10–20, 21–50, and >50 cases/yr), the association between case load and outcomes was consistently present. For primary operations, the highest volume surgeons (>50 cases/yr) had significantly fewer complications than the lowest volume surgeons (1–10 cases/yr): 1.01% vs. 1.91% (P < 0.01); for reoperations, they had fewer major complications than the low medium (11–20 cases/yr) and the lowest volume surgeons: 1.48% vs. 2.7% and 3.8%, respectively (P < 0.05 and P < 0.001, respectively); and for in-hospital mortality, the high volume surgeons reported significantly fewer deaths than the high medium (21–50 cases/yr) and the low volume surgeons: 0.05% vs. 1.26% and 1.0% (P < 0.05). After adjustment for experience, differences remained significant for comparisons of complication rates after primary operation (P < 0.001) and reoperation (P < 0.01 vs. medium and P < 0.001 vs. low volume surgeons).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The potential impact of the variation in management strategies resulting from the controversy of medical vs. surgical therapy for patients with 1° HPT was last addressed in 1980, when a break-even analysis suggested that the cost of definitive diagnosis and surgical treatment of 1° HPT would be exceeded by the cost of 5.5 yr of medical follow-up (3). The economic impact of current practice patterns could be even more substantial today, given that the cost of medical surveillance is greater and the incidence and prevalence of 1° HPT have been increasing as a result of the routine determination of serum calcium by automated methods (26).
Based on our national survey of endocrine surgeons, there appears to be significant variation in physician decision-making regarding the surgical management of 1° HPT. This lack of consensus about optimal management strategies is particularly striking, given that survey respondents represent a group of highly experienced and clinically busy group of specialty endocrine surgeons. It suggests that this survey almost certainly underestimates the practice pattern variation that exists in the broader surgical community, where surgeons generally have less endocrine surgery expertise and less experience performing parathyroidectomies. It also does not account for any variation in practice patterns that might exist among endocrinologists, internists, and family practitioners in their decision to refer patients with 1° HPT to surgeons.
A notable finding of the study was the association of surgical volume, as measured by annual parathyroidectomy case load, with thresholds for surgery and short term clinical outcomes. Using administrative data, researchers have established an association between volume and clinical and/or economic outcomes for a variety of procedures, including parathyroidectomy (27), thyroidectomy (28), coronary artery bypass surgery, partial gastrectomies, colectomies, and cholecystectomies (29). Because we suspect that respondents would underestimate rather than overestimate departures from practice guidelines or clinical misadventures, the association between annual parathyroidectomy case load and self-reported practice patterns and outcomes should not be dismissed. However, it is impossible from a cross-sectional study such as this to prove causation, as temporal effects cannot be captured. Consequently, we are unable to determine whether surgeons with larger annual case loads have lower thresholds for operating on patients with 1° HPT because their rates of operative morbidity and mortality are lower than those of less busy surgeons (and their patients are more willing to have surgery because they incur less risk), or whether they have larger annual case loads because their thresholds for surgery are lower. The finding that there is an association between surgeon volume of parathyroidectomies and clinical outcomes raises the important question of whether there is unnecessary in-hospital morbidity and even mortality that might be avoided by directing patients with 1° HPT to more experienced parathyroid surgeons.
There are several possible limitations to this study. Because, as mentioned, the results are based on self-report, there is the potential for respondent reporting bias. In particular, conclusions regarding clinical outcomes must be treated with caution. Respondents were assured that results would be confidential and reported in aggregate so that individual respondents could not be identified. It is reassuring, however, that the outcomes determined in this study are similar to those previously reported by others (30, 31, 32). There is also the potential for selection bias among our sampling frame. The response rate, however, was 77%, and respondents and nonrespondents appeared to be similar. Finally, there is the question of whether our results are generalizable outside of the membership of the AAES. In favor of considering these data applicable to the majority of patients with 1° HPT, we learned from a national pilot survey of members of The Endocrine Society that endocrinologists refer 96% of their 1° HPT patients to general surgeons who specialize in endocrine surgery, 3% to otolaryngologists, and 1% to general surgeons (33). Further research about patterns of referral for patients with 1° HPT would be useful.
The results of this national survey suggest that endocrine surgeons are not complying with many of the practice guidelines outlined in the NIH consensus statement. It is possible that surgeons are not aware of the NIH guidelines. Alternatively, they could be aware of the guidelines but intentionally do not follow them; for example, surgeons might believe that there is a paucity of evidence (e.g. prospective randomized trials) supporting the consensus statement. This raises the question of whether the guidelines are obsolete and require reformulation, as third party payers might use published recommendations to guide their reimbursement strategies. For example, according to the published guidelines, all young patients with 1° HPT should undergo parathyroidectomy, as the effects of several decades of even mild untreated 1° HPT are still not well understood (2). In practice, however, endocrine surgeons appear to be employing guidelines for surgery based on laboratory and clinical criteria similar to those used for older patients. Overall, respondents had higher thresholds than recommended by the NIH for younger patients and lower thresholds than recommended for older patients.
It is surprising that there is so much variation in practice even after the decision has been made to refer patients with suspected 1° HPT for surgery. Whether patients referred for parathyroidectomy actually undergo an operation may depend to a large extent on the experience of their surgeons. This finding raises the question of whether the threshold for referral for surgery used by endocrinologists, internists, and family practitioners differs from the thresholds for operating employed by surgeons. The results of this study implore the endocrine community to examine the evidential basis for decisions made in the management of 1° HPT. They have important implications for patients, referring physicians who must decide where to send their 1° HPT patients for surgery, surgeons, and third party payers who are responsible for setting medical coverage and reimbursement policies.
|
Acknowledgments |
|---|
Received February 27, 1998.
Revised April 13, 1998.
Accepted April 20, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. A. Johnson, M. E. Tublin, and J. B. Ogilvie Parathyroid Imaging: Technique and Role in the Preoperative Evaluation of Primary Hyperparathyroidism Am. J. Roentgenol., June 1, 2007; 188(6): 1706 - 1715. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Sejean, S. Calmus, I. Durand-Zaleski, P. Bonnichon, P. Thomopoulos, C. Cormier, P. Legmann, B. Richard, X. Y Bertagna, and G. M Vidal-Trecan Surgery versus medical follow-up in patients with asymptomatic primary hyperparathyroidism: a decision analysis Eur. J. Endocrinol., December 1, 2005; 153(6): 915 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. J. Strewler A 64-Year-Old Woman With Primary Hyperparathyroidism JAMA, April 13, 2005; 293(14): 1772 - 1779. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Morrison, W. Farrar, J. Kneile, N. Williams, Y. Liu-Stratton, A. Bakaletz, M. A. Aldred, and C. Eng Molecular Classification of Parathyroid Neoplasia by Gene Expression Profiling Am. J. Pathol., August 1, 2004; 165(2): 565 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-Y. Lo, W.-F. Chan, A. W. C. Kung, K.-Y. Lam, S. C. F. Tam, and K. S. L. Lam Surgical Treatment for Primary Hyperparathyroidism in Hong Kong: Changes in Clinical Pattern Over 3 Decades Arch Surg, January 1, 2004; 139(1): 77 - 82. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. Barker II, A. Klibanski, and B. Swearingen Transsphenoidal Surgery for Pituitary Tumors in the United States, 1996-2000: Mortality, Morbidity, and the Effects of Hospital and Surgeon Volume J. Clin. Endocrinol. Metab., October 1, 2003; 88(10): 4709 - 4719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kebebew, Q.-Y. Duh, and O. H. Clark Parathyroidectomy for Primary Hyperparathyroidism in Octogenarians and Nonagenarians: A Plea for Early Surgical Referral Arch Surg, August 1, 2003; 138(8): 867 - 871. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. C. Chow, W. B. Chan, J. K. Y. Li, N. N. Chan, M. H. M. Chan, G. T. C. Ko, K. W. Lo, and C. S. Cockram Oral Alendronate Increases Bone Mineral Density in Postmenopausal Women with Primary Hyperparathyroidism J. Clin. Endocrinol. Metab., February 1, 2003; 88(2): 581 - 587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. N. Fahy, R. J. Bold, L. Beckett, and P. D. Schneider Modern Parathyroid Surgery: A Cost-benefit Analysis of Localizing Strategies Arch Surg, August 1, 2002; 137(8): 917 - 923. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. B. Inabnet III, C. K. Kim, R. S. Haber, and R. A. Lopchinsky Radioguidance Is Not Necessary During Parathyroidectomy Arch Surg, August 1, 2002; 137(8): 967 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Okamoto, T. Kamo, and T. Obara Outcome Study of Psychological Distress and Nonspecific Symptoms in Patients With Mild Primary Hyperparathyroidism Arch Surg, July 1, 2002; 137(7): 779 - 783. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Hindie, D. Melliere, C. Jeanguillaume, P. Urena, C. deLabriolle-Vaylet, and L. Perlemuter Unilateral Surgery for Primary Hyperparathyroidism on the Basis of Technetium Tc 99m Sestamibi and Iodine 123 Subtraction Scanning Arch Surg, December 1, 2000; 135(12): 1461 - 1468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Siminoski Asymptomatic hyperparathyroidism: Is the pendulum swinging back? Can. Med. Assoc. J., July 1, 2000; 163(2): 173 - 175. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
