| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Endocrinology and Metabolism, Department of Medicine, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, New York 10467
Address all correspondence and requests for reprints to: Martin I. Surks, M.D., Department of Medicine, Montefiore Medical Center, 111 East 210th Street, Bronx, New York 10467. E-mail: surks{at}aecom.yu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Early referral of patients with suspected thyroid nodules to an endocrinologist results in significant savings in both cost and patient’s time as well as increased precision of diagnosis.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
7%) are
malignant (1, 2, 3). The challenge to physicians in the evaluation of
thyroid nodules is to identify the few patients with thyroid malignancy
among the large majority of patients who have benign nodular disease.
The former always require surgical excision, whereas most patients with
benign nodular disease can be managed without surgery. Moreover, early
identification of thyroid malignancy in patients with thyroid nodules
should also improve outcome. A delay in diagnosis allows time for tumor
growth, and tumor size is correlated with tumor recurrence and
mortality (4). Thyroid nodules are identified during routine physical examination, discovered by the patient, or found incidentally during imaging procedures for nonthyroidal disorders. Once a palpable thyroid nodule has been detected, primary care physicians (PCP) generally order various diagnostic tests and imaging studies, and then refer patients for biopsy or surgery as needed. The most frequently ordered imaging studies for evaluation of thyroid nodules are the radionuclide scan and the thyroid sonogram. Most (80–85%) thyroid nodules trap radionuclide significantly less efficiently than surrounding normal thyroid tissue and are designated cold or nonfunctional nodules (5). The incidence of thyroid cancer in cold nodules is about 10%, whereas the incidence of thyroid cancer in functional lesions is less than 2% (5). Therefore, the positive predictive value for thyroid cancer of finding a cold nodule on radionuclide scan is less than 10%. Although thyroid sonography provides detailed anatomical information, it does not allow distinction between benign and malignant nodules. Moreover, thyroid sonography may show the presence of small (<1 cm in diameter) nodules that are not clinically relevant but result in further testing and cause the patient anxiety (6).
In contrast to the low positive predictive value of the radionuclide scan and the thyroid sonogram for thyroid cancer, fine needle aspiration biopsy (FNAB) has a sensitivity and specificity for thyroid cancer greater than 90% (7, 8, 9, 10, 11). Guidelines for evaluation of thyroid nodules have recently been published by the American Thyroid Association (12) and the American Association of Clinical Endocrinologists (13), in which referral to an endocrinologist for FNAB is the primary approach. As this recommendation, published by endocrinologists, may appear self-serving, we have carried out a study to evaluate the effect of referral to an endocrinologist, practicing according to these published guidelines, on the efficiency of diagnostic evaluation and management of patients with thyroid nodules. We collected data specifically focused on the cost, accuracy, and efficiency of evaluations pursued by PCPs, and in some instances surgeons, before referral to an endocrinologist and compared these parameters for the same patients as assessed by an endocrinologist.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The records of all patients referred to an endocrinologist (M.I.S.), the consulting endocrinologist (CE), during the interval of January 1, 1995, to December 31, 1996, were reviewed. Among 429 patients, 70 patients referred because of suspected thyroid nodules were identified, and their records were reviewed by one of us (R.O.). Pertinent information concerning evaluation of the thyroid abnormality by the PCP was tabulated. Recorded were age, sex, nature of referral (PCP, self-referred, second opinion for patients previously seen by a surgeon), reason for referral, description of the nodule including size and consistency or time since first detection of nodule, pertinent past history concerning radiation treatment, family history, and compressive symptoms such as dysphagia, dyspnea, and dysphonia. All laboratory studies ordered by the PCP were listed.
Data collection from the CE’s examination and evaluation
The data from history and physical examination obtained by the CE were then recorded as well as additional laboratory tests and procedures that were carried out to reach a diagnosis and develop a management plan. The CE practiced according to published guidelines for diagnosis and management of thyroid nodules (12, 13). The CE made a diagnosis and a management decision that used both data provided by the PCP and tests ordered as a result of endocrine consultation.
Assessment of patients by a blinded second endocrinologist
Limited information about each patient was then presented to a second endocrinologist (K.H.H.), designated the reviewing endocrinologist (RE), who also practiced according to published guidelines for diagnosis and management of thyroid nodules (11, 12). The RE was blinded to the diagnoses and management plans of the CE and PCP. Presented were the history of how the nodule was discovered; the presence of symptoms of thyrotoxicosis or hypothyroidism; whether the patient had pain and/or tenderness in the anterior neck, dysphagia, dysphonia, and dyspnea; and any history of radiation exposure. A description of the thyroid examination of the CE was provided as well as results of tests of thyroid function and antithyroid microsomal antibody titers, if available. Using the limited information that was presented about each patient, the RE indicated his plan of evaluation in sequential order. If tests or procedures requested by the RE had been available to or ordered by the CE, the results were then revealed to the RE. In this iterative manner, the RE made a diagnosis and developed a management plan using only salient history, description of physical examination of an experienced endocrinologist and essential laboratory findings.
Determination of unnecessary visits, tests, and costs
The number of visits to the PCP’s office and/or laboratories for various diagnostic tests before patients were referred to the CE was determined. Thyroid sonography; repeated determinations of serum thyroid hormones, TSH, and antithyroid antibodies; computerized axial tomography (CT scan); aspiration biopsy; and ultrasound-guided aspiration biopsy were assigned one visit each. A radionuclide scan required two visits. In some patients, thyroid sonography was performed at the same visit as the radionuclide scan. We then estimated the number of hours patients spent for visits to the PCP’s office and/or laboratories for these tests by assuming that each visit required 4 h in this metropolitan area, which included time for travel and waiting. The cost of various laboratory tests was calculated as the average of the cost per test for uninsured patients in the laboratories of Montefiore Medical Center, a private physician’s office laboratory, and a commercial laboratory in this area. The average charges were $135 for thyroid function tests (including serum T4, free T4 estimate, TSH, and autoantibodies), $340 for radionuclide uptake and scan, $230 for thyroid sonography, $225 for thyroid aspiration biopsy, and $500 for CT scan.
Unnecessary testing was defined as tests carried out before patients were referred to the CE that were not required by the RE to make a diagnosis or develop a management plan in accordance with published guidelines (11, 12). The cost of unnecessary tests was calculated, and the resultant number of hours spent by patients for unnecessary testing was determined.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All 70 patients, who were evaluated for thyroid nodules, were
included in this analysis. Their mean age was 51.7 ± 15.9 yr
(±SD; range, 15–78 yr; 11 males and 59 females; Table 1
). The time elapsed between
identification of a thyroid abnormality by the PCP and referral to the
endocrinologist varied widely: less than 1 month in 33%, 1–3 months
in 25%, 4–11 months in 13%, and more than 12 months in 29% (Fig. 1). Thirty-five patients (50%) were
referred directly by their PCPs, 25 patients (36%) had had thyroid
evaluations elsewhere but sought a second opinion, and 10
patients (14%) had seen a PCP in the past but were not directly
referred by that physician (Table 1). Thyroid abnormalities were
initially self-detected in 19 patients (27.1%), were detected by the
PCP in 49 patients (70%), and were incidentally discovered in only 2
patients (2.9%). Nodule size was described as less than 1 cm in 2
patients (2.9%), between 1–2 cm in 11 patients (15.7%), between 2–3
cm in 10 patients (14.3%), and more than 3 cm in 7 patients (10%). A
description of the thyroid abnormality by the PCP was not provided for
the remaining 40 patients (58.6%).
|
|
Before referral the 70 patients had the following diagnostic tests
performed: 114 thyroid function tests (T4 or
FT4 estimate, 60; TSH, 54) 17 thyroid autoantibody titers
(14 antithyroid microsomal antibody or anti-TPO antibody and 3
antithyroglobulin antibody determinations), 4 CT scans of the neck
and/or chest, 39 thyroid sonograms, 37 radionuclide uptakes and scans,
24 FNABs (performed by surgeons or other endocrinologists), and 3
ultrasound-guided FNABs (Table 2). We
determined that these diagnostic tests required 233 visits by patients
to a physician’s office or a laboratory (Fig. 2). Based on these evaluations, 8
patients had been referred for surgery before endocrine consultation
(Table 3).
|
|
|
The CE independently determined the following pertinent past history: only 4 patients had received radiation treatment, all before the age of 10 yr; 20 patients had a family history of thyroid disease; 4 had dysphagia; 3 had palpitations; and 2 had neck tenderness. On clinical examination, 67 patients were considered euthyroid, 2 were hyperthyroid, and 1 was hypothyroid. On neck examination, the CE noted that 47 patients had solitary thyroid nodules (67.1%), 4 patients had multinodular goiter (5.7%), 11 patients had diffusely enlarged thyroid glands (15.7%), and 1 patient had an extrathyroidal neck mass. Seven patients had normal neck examinations (10%).
After reviewing all data provided by the PCP, the CE ordered the
following diagnostic tests: 64 thyroid function tests (32, TSH; 32,
T4 or FT4 estimate), 30 determinations of
antithyroid antibodies, 6 radionuclide scans, no sonograms,
and 1 CT scan for the patient with an extrathyroidal neck mass. The
slides from previous FNABs in 12 patients were reviewed; a second FNAB
was performed in 2 of these patients (Table 2
). The CE ordered repeat
FNABs for the other 12 patients who had had an FNAB previously because
the cytology report indicated that the samples were inadequate for
diagnosis in 10 patients, and in 2 patients slides were unavailable for
review. In addition, the CE carried out 28 FNABs and ordered 2
ultrasound-guided FNABs.
Results of CE’s evaluation
Of the eight patients referred for surgery before endocrine
referral (Table 3; patients 1–8), the CE recommended immediate surgery
for one (patient 3), and in another patient (patient 6), surgery was
recommended only if the lesion failed to decrease significantly in size
after 3-month treatment with a TSH-suppressive dose of thyroid hormone.
The other six patients (patients 1, 2, 4, 5, 7, and 8) were considered
to have benign disease.
In another 7 patients (patients 9–15), the CE recommended surgery for 5 (patients 11–15) and for the 2 others (patients 9 and 10) if the lesion persisted after 3 months of TSH suppression therapy using thyroid hormone. Of the 9 patients recommended for surgery, 4 had papillary carcinoma with a tumor more than 3 cm diameter (patients 3 and 11–13). Twelve of the remaining 55 patients, not referred for surgery, were recalled for follow-up neck examinations after more than 3 months had elapsed; 8 received no treatment or follow-up evaluation. Thirty-seven patients were treated with L-T4 or L-T3 for hypothyroidism, euthyroid goiter, or Hashimoto’s disease or were observed for 3 months to determine whether a thyroid nodule would decrease in size. One patient was referred back to the PCP for further evaluation of an extrathyroidal metastatic carcinoma.
RE’s proposed evaluation and management plan
Pertinent history, the results of neck examination by the CE,
thyroid function tests, and antithyroid antibody titers, if available,
were then presented to the RE. The RE ordered 71 thyroid function tests
(33 TSH; 38 FT4 estimates), 48 determinations of
antithyroid antibodies, 1 thyroid sonogram, and 6 radionuclide scans
(Table 2). In addition, he recommended FNABs in 43 patients.
Based on these data, the RE recommended observation with
follow-up in 14 patients, radionuclide treatment of a toxic adenoma in
3 patients, and L-T4 treatment with follow-up
in 35 patients; 9 patients required no treatment or follow-up. All of
the 6 patients who had been referred for surgery by the CE (Table 3;
patients 3 and 11–15) were also referred for surgery by the RE. Three
additional patients were referred for surgery by the RE (Table 3;
patients 6, 9, and 10). Among these were 2 patients for whom the CE had
recommended surgery if the lesion did not significantly decrease in
size after 3-month treatment with thyroid hormones; for the third, he
recommended a second aspiration biopsy if the lesion did not decrease
in size after L-T3 treatment. Of the
8 patients referred for surgery before endocrine referral, only 2
(Table 3; patients 3 and 6) were recommended for surgery by the RE.
Five of the remaining 6 patients had received recommendations for
surgery by surgeons; only 1 of the 5 (patient 3) had had FNAB. Their
diagnoses were colloid nodule (2 patients), colloid goiter,
multinodular goiter, benign cyst, and toxic adenoma. Based on the
limited information presented to the RE and the additional tests that
he requested, concordance between the RE and the CE in evaluation and
treatment plan for all 70 patients was 87.1% and 94.3%,
respectively.
Unnecessary testing and costs
Unnecessary testing was defined as laboratory tests and/or procedures ordered before endocrine consultation but not required by the RE for evaluation and management. These included 153 visits to the PCP’s office or a laboratory, 44 sets of tests, 32 radionuclide scans, 39 sonograms and 3 CT scans. We estimated the direct cost of unnecessary testing was $27,290 (average, $390/patient), and there were also 30 unnecessary repeated physician office visits. In addition to the direct cost of these tests, we estimated that patients spent an average of 8.7 h for unnecessary testing and office visits in our metropolitan area.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We showed that the mode of thyroid nodule evaluation pursued by PCPs resulted in an average unnecessary monetary loss of $390/patient for imaging studies and laboratory tests that were not required for clinical decision-making as well as the cost of 30 unnecessary visits to the physician’s office. These evaluations also resulted in an average loss of 8.7 h of each patient’s time according to our estimates. Moreover, the mean 8.7 h spent per patient for unnecessary tests often resulted in a loss of 2–4 half-days from work or other activities, causing additional economic loss to both the patient and the employer. Further, among the 70 patients referred to the CE, 42% were referred more than 3 months after nodule detection. As these patients were undergoing an evaluation for possible malignancy, the length of time before patients were referred to the CE was troubling.
Appropriate and efficacious testing
Radionuclide thyroid scans and thyroid sonography are relatively expensive imaging procedures that were the main contributors to the high cost of unnecessary testing in our patients. As both of these tests have low specificity for predicting thyroid cancer, the FNAB is still required for diagnosis in most patients. Therefore, the cost saving achieved by early referral of patients to an endocrinologist who practices according to guidelines recommending limited use of these imaging modalities is understandable. The number of patients for whom these tests would provide sufficient confidence in a diagnosis to preclude referral to an endocrinologist is small. The majority of patients studied by these procedures would still require FNAB and appropriate recommendation for therapy. It is clear that considerable savings can be achieved by early referral to an endocrinologist who practices according to published guidelines (12, 13).
Eight patients, five of whom had had FNABs performed before endocrine consultation, had been advised to have surgery before referral to the endocrinologist. However, only two were referred to surgery by the CE and RE; the remaining patients were determined to have benign disease that did not require surgery. Thus, in six of eight patients, a recommendation for surgery was reversed by the endocrinologist. This suggests that not only are evaluations performed by PCPs more costly and more time consuming than if patients were referred to endocrinologists early during the evaluation, but they are also more likely to lead to recommendations for unnecessary surgery. Additionally, performing surgery on this group of eight patients would not have identified three of the four patients who subsequently were shown to have papillary thyroid carcinoma.
Difficulty in diagnosis of thyroid nodules
Among the 70 patients referred for thyroid nodules, 7 were found to have normal thyroid examinations by the CE. Laboratory tests ordered for this group of patients included 4 thyroid sonograms, 3 radionuclide scans, and 1 CT scan. It is possible that some of these patients may have had transient abnormalities of the thyroid gland, such as focal thyroiditis, which may have regressed in the interval between evaluation by the PCP and endocrine consultation. More likely in our experience, these patients could have been spared testing performed to evaluate thyroid nodules that were never present. A recent report indicated that the specificity of assessing whether patients had goiter was poor (14). There is, therefore, a high rate of false positives in determining the presence of thyroid abnormalities on physical examination. Common reasons for false positive thyroid findings in our experience include prominent thyroid and/or cricoid cartilage, hypertrophied and tense sternocleidomastoid and strap muscles of the neck, and obesity. In addition, false positive thyroid examinations probably result in unnecessary testing, usually thyroid sonography and radionuclide scans, in many patients who are not eventually referred to an endocrinologist. In our geographical area, fees for endocrinology consultation ($250, usually significantly discounted by insurance and managed care companies) are generally smaller than those for either thyroid sonography or radionuclide scans. Therefore, when the PCP suspects thyroid abnormalities, endocrine evaluation is more economical and efficient in detecting and diagnosing clinically relevant thyroid nodules than are imaging studies. Moreover, these findings highlight the need for improved teaching of thyroid examination techniques in medical schools and postdoctoral training programs.
Current health care policy in the United States is moving toward a patient care paradigm that may limit access to specialists. This paradigm may or may not result in high quality medical care at reduced costs for patients who have disorders in some medical specialties. However, our data suggest that early referral of patients to an endocrinologist when thyroid nodular disease is suspected would actually result in significant savings of money spent for unnecessary diagnostic testing, a decreased loss of patients’ time from work or other activities, a decrease in length of time that patients with benign disease harbor concern about the possibility of thyroid malignancy, and fewer surgeries in patients with benign disease.
Our data indicate that accuracy and efficiency are better achieved by early referral of patients suspected of having thyroid nodular disease to an endocrinologist. We suggest that the principal role of the physician who detects thyroid nodules is to determine whether the patient has hypothyroidism or hyperthyroidism, and evaluate the patient for thyroid autoimmunity. Subsequent evaluations are more effectively and efficiently performed by an endocrinologist.
Received April 30, 1998.
Revised July 7, 1998.
Accepted July 14, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E. Papini, R. Guglielmi, A. Bianchini, A. Crescenzi, S. Taccogna, F. Nardi, C. Panunzi, R. Rinaldi, V. Toscano, and C. M. Pacella Risk of Malignancy in Nonpalpable Thyroid Nodules: Predictive Value of Ultrasound and Color-Doppler Features J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 1941 - 1946. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Marqusee, P. R. Larsen, and S. J. Mandel Ultrasonography in Management of Nodular Thyroid Disease Ann Intern Med, September 4, 2001; 135(5): 384 - 384. [Full Text] [PDF]
|
|
|
|
|
|
E. L. Mazzaferri and R. T. Kloos Current Approaches to Primary Therapy for Papillary and Follicular Thyroid Cancer J. Clin. Endocrinol. Metab., April 1, 2001; 86(4): 1447 - 1463. [Full Text]
|
|
|
|
|
|
F. N. Bennedbæk and L. Hegedüs Management of the Solitary Thyroid Nodule: Results of a North American Survey J. Clin. Endocrinol. Metab., July 1, 2000; 85(7): 2493 - 2498. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
