| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EXTENSIVE CLINICAL EXPERIENCE |
Neuroendocrine Unit (L.N., A.D., R.Z., A.K.), Neurosurgical Service (B.S.), and Biostatistical Center (H.L.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Anne Klibanski, M.D., Neuroendocrine Unit, 55 Fruit Street, Bulfinch 457, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: aklibanski{at}partners.org.
| Abstract |
|---|
|
|
|---|
Objectives: Our objective was to identify the medical professional who first suspected acromegaly and the complaint prompting the diagnosis, and if these have changed. Additional goals were to assess the interval from symptom onset to diagnosis of acromegaly and to compare treatment trends over consecutive decades.
Design: This was a case-record retrospective study.
Setting: The study was performed in a neuroendocrine clinical center at a tertiary care center.
Subjects: A total of 100 patients (45 men and 55 women) with acromegaly referred from 1985–2005 was included in the study.
Results: Acral changes (24%) and headaches (20%) were most prevalent presenting symptoms prompting diagnosis. Eighteen percent reported no symptoms of acromegaly at diagnosis. The primary care physician most often initiated the evaluation (44%). Comorbidities were more prevalent in older patients (P = 0.001). The interval between symptom onset and diagnosis decreased, compared with previous reports. Radiation therapy was used less frequently in the decade after 1994 than in the prior (16 vs. 33%; P < 0.05).
Conclusions: The primary care doctor plays the major role in diagnosis of acromegaly. The increased use of brain magnetic resonance imaging may contribute to the many incidentally discovered cases and to the shortened time interval to diagnosis. Presumably due to the availability of new medical therapies, the use of radiation therapy has decreased.
| Introduction |
|---|
|
|
|---|
Although the clinical manifestations of acromegaly have been extensively reported, there is limited information pertaining to the type of medical professional who makes the initial diagnosis of acromegaly and minimal data available regarding the initial complaint that prompts medical evaluation. In addition, many new medical therapies have become available over the past 20 yr, including short acting and depot somatostatin analogs, cabergoline, and most recently, the GH receptor antagonist. The impact of these advances on previously used treatment modalities, particularly the use of radiotherapy (XRT), is not well established.
| Patients and Methods |
|---|
|
|
|---|
Consecutive records (n = 200) on patients with a diagnosis of acromegaly seen between 1985 and 2005 at the Neuroendocrine Clinical Center at Massachusetts General Hospital (MGH) were selected for review. Patients were included (n = 100) in the study if they had both a confirmed diagnosis of acromegaly and a complete medical record available. The diagnosis of acromegaly was based on clinical symptoms, the presence of a pituitary adenoma on magnetic resonance imaging (MRI) scan, biochemical confirmation, including nonsuppressed GH on an oral glucose tolerance test (OGTT) and/or normal IGF-I concentration, and in surgically treated patients, a pathologically confirmed GH secreting adenoma. The Partners Institutional Review Board approved the study.
Variables studied
The history data were obtained on uniform collection sheets based on review of findings from the patients complete MGH record as well as external records in patients who had prior evaluations at other centers. The interval of time between the onset of symptoms (based on patient report and prior medical records) and initial diagnosis (first record of elevated IGF-I or GH testing) was determined. Tumors were characterized as incidentally found if the patient presented with no symptoms attributable to acromegaly at diagnosis and, thus, by definition had 0-yr delay in diagnosis. Other variables evaluated included patient demographics (sex, age at diagnosis), main complaint leading to diagnosis, type of professional seen, past medical history, size of tumor (microadenoma or macroadenoma), therapeutic modalities used by referring or Center physicians, complications of therapy, IGF-I level at diagnosis, comorbidities at diagnosis (as determined by the review of prior outside medical records, as well as the history and physical examination recorded during the MGH Neuroendocrine Clinical Center visit), and therapeutic response (nadir suppression of GH during the OGTT and/or IGF-I concentration). The 76 patients who had surgery at MGH were seen preoperatively by a Neuroendocrine Clinical Center staff member. The 24 patients who did not have surgery at MGH were seen at different stages of their postoperative care (n = 20) or were receiving primary medical therapy (n = 4).
Remission was defined as normalization of IGF-I levels and/or appropriate suppression of GH during an OGTT according to established criteria at diagnosis (27, 28, 29, 30, 31). Patients were considered to have active disease if the IGF-I concentration was above the age and sex-adjusted normal range, and/or GH did not suppress adequately during an OGTT at least 3 months after a primary form of therapy. These variables were also compared between the two 10-yr periods (before and after 1995). Hypopituitarism was determined based on standard biochemical testing of pituitary function. The Cortrosyn stimulation test (Amphastar Pharmaceuticals, Inc., Rancho Cucamonga, CA) or insulin tolerance test was used to assess adrenal insufficiency. A low free T4 with normal or low TSH was considered to be central hypothyroidism. Low testosterone and normal or low gonadotropins in men represented central hypogonadism. Low estradiol and low FSH in women with amenorrhea or menstrual irregularity were considered evidence of hypogonadism in women.
Statistical methods
The data were analyzed using SAS software, version 8 (SAS Institute Inc., Cary, NC). For statistical analysis of results, the
2 test was used to compare proportions between groups. The Wilcoxons rank sum test or Kruskal-Wallis test was used depending on the number of groups being compared, to compare continuous measurement outcomes between the groups. All results were considered significant if two-sided P < 0.05, unless otherwise indicated. Data are expressed as mean ± SD unless otherwise indicated.
| Results |
|---|
|
|
|---|
Of the 100 patients, 55 were women, and 45 were men. The mean age was 41.2 ± 14.2 yr (range 10–80). The mean age at onset of symptoms leading to diagnosis was 38.7 ± 14.5 yr (range 8–78). The mean time to diagnosis was 2.5 ± 4.6 yr (median 1 yr, range 0–32). Excluding patients who were diagnosed incidentally, the mean time to diagnosis was 3.2 yr (±5.0) (median 2, n = 82). To account for changes in IGF-I assays over the two decades, a mean IGF-I factor was used (IGF-I/ upper limit of normal for assay), and at diagnosis was 2.39 ± 1.27 (range 0.55–8.50). Of tumors, 87% were macroadenomas, and 13% were microadenomas. Mean IGF-I factor at diagnosis was 2.4 ± 1.2 for microadenomas and 2.4 ± 1.3 for macroadenomas. Of patients, 25% were diagnosed before 1995 (period 1) and 75% during or after that year (period 2).
Main complaint leading to diagnosis
The most common problem leading to the diagnosis was acral changes (24%), followed by headaches (20%), amenorrhea (6% of total, 11% of women), dental changes (4%), carpal tunnel syndrome (4%), visual deficits (3%), sexual dysfunction (3%), arthralgias (2%), galactorrhea (2%), chest pain (2%), uncontrolled hypertension (2%), diabetes (1%), dizziness (1%), gynecomastia (1% of total, 2% of men), weakness (1%), and weight gain (1%). The most common initial complaint leading to diagnosis in both periods 1 and 2 was acral changes followed by headaches (Table 1
). Four of the 100 patients with acromegaly had multiple endocrine neoplasia 1. Among these patients, initial complaints leading to the diagnosis of acromegaly included galactorrhea (n = 1), dizziness (n = 1), chest pain (n = 1), and acral changes (n = 1).
|
Medical professionals
Table 2
shows the professional category of the health care provider who first recognized acromegaly. The primary care physician initiated the evaluation of acromegaly in 44% of the cases, an endocrinologist in 13%, emergency room physician in 10%, neurologist in 6%, dentist in 3%, and ophthalmologist in 3%. A pediatrician, urologist, ear nose and throat specialist, and orthopedist each made the diagnosis in 2% of patients. An obstetrician/gynecologist made the diagnosis in 2% of all patients and 3.6% of women. A podiatrist, rheumatologist, cardiologist, and pulmonologist each diagnosed 1% of patients. Seven percent were self-referred for diagnosis. In the emergency room setting (n = 10), the reasons for seeking emergency care were headaches (60%, n = 6), head trauma (20%, n = 2), and cardiac disease (20%, n = 2). The primary care physician played a major role in the diagnosis of acromegaly in both periods (47% in period 2 vs. 36% in period 1; P = 0.25; Table 2
). The order of frequency of professionals who first diagnosed acromegaly was comparable between the periods, although trends were found. The proportion of acromegaly cases diagnosed by an emergency room physician during the decade after 1995 (7%) was less than half that of the former decade (20%) (Table 2
). In addition, self-referred patients increased from 4% before 1995 to 8% after 1995, including one self-referred through the Internet in the latter group.
|
Comorbidities present at diagnosis were diabetes mellitus (15%), hypertension (25%), coronary heart disease (5%), or sleep apnea (1%). Of patients, 40% had multiple comorbidities. Of patients, 8% had a malignancy, including papillary thyroid cancer (n = 3), breast cancer (n = 2), chronic lymphocytic leukemia (n = 1), prostate cancer (n = 1), and colon cancer (n = 1). More patients were diagnosed before than after age 40 (53 vs. 47%; P = 0.016). The number of years of delay in diagnosis did not predict the presence of comorbidities at diagnosis in this population. Patients with one or more comorbidities at diagnosis, including diabetes, heart disease, hypertension, or sleep apnea, did not differ significantly in the number of years of delay in diagnosis compared with those who did not have these comorbidities at diagnosis (mean 2.1 ± 3.9 yr vs. 2.8 ± 5.0 yr; P = 0.25). Patients with one or more comorbidities at diagnosis did not differ significantly in gender or mean IGF-I factor compared with those who did not have these comorbidities at diagnosis. However, the age at diagnosis was predictive of the presence of a comorbidity at diagnosis, i.e. older patients were more likely to present with comorbidities. The mean age at diagnosis was 48.0 ± 14.8 yr in those with a comorbidity compared with 36.8 ± 11.8 yr in those with no comorbidity (P < 0.001). Of patients older than 55 yr at diagnosis, 76.5% had at least one comorbidity compared with 32.5% of patients younger than 55 yr (P < 0.001).
Treatment modalities and outcomes
Surgical treatment occurred at multiple surgical centers because many patients were referred for endocrine management so that surgical outcomes represent results from many different surgeons. Of patients, 46% had transsphenoidal surgery (TSS) alone, 28% had TSS plus medical therapy, 12% had TSS plus XRT, 8% had TSS plus XRT plus medical therapy, and only 4% were treated with primary medical therapy. The medications included cabergoline, somatostatin analogs, and pegvisomant, used alone or in combination. The four patients treated with medical therapy alone were all receiving somatostatin analogs, and all were controlled based on GH suppression tests and/ or IGF-I levels in the normal range. One was receiving a combination somatostatin analog and pegvisomant. Among the 40 patients using medical therapy alone or medical therapy with surgery and/or XRT, 21 had normalized IGF-I and/or GH levels.
One percent had a craniotomy, and 1% refused treatment. Surgical complications such as cerebral spinal fluid leak (n = 1), infection (n = 1), or bleeding (n = 1) were rare, occurring in 3.2% of the 95 surgically treated patients. Overall, 51% of patients surgically treated at multiple surgical centers were cured. The preoperative mean IGF-I factor was not significantly different between cured and not cured after surgery [2.29 ± 1.39 vs. 2.50 ± 1.1 (±SD); P = 0.16]. Of patients with microadenomas and those with macroadenomas, 75% and 48%, respectively, were cured. The remaining patients had residual disease and required adjunctive treatment for control. Among all patients, including those treated with radiation, panhypopituitarism occurred in 11% and transient diabetes insipidus in 3%. Women were more likely to have pituitary dysfunction (P = 0.0007), with panhypopituitarism in seven out of 11 and transient diabetes insipidus in three out of three.
Changes in therapeutic modalities over time
The proportion of microadenomas and macroadenomas did not significantly differ between the two time periods (12 vs.13%, 87 vs. 88%; P = 0.86). Trends in treatment modalities used changed during the periods. The overall use of TSS was similar during study periods before 1995 and after (96 vs. 93%). There was no significant difference in the use of TSS as the sole treatment (49%, n = 37) in period 2 compared with period 1 (36%, n = 9). However, radiation therapy was used much less frequently in period 2 (16% in period 2, n = 12 vs. 33% in period 1, n = 8, P < 0.05) using one sided z-test (Fig. 1
).
|
| Discussion |
|---|
|
|
|---|
In our series the most prevalent presenting complaints leading to diagnosis were acral change (24%) and headaches (20%). Most other series describe all disease-related clinical manifestations, rather than specifically reporting the presenting complaint prompting the diagnosis. Acral changes and headaches are among the most prevalent symptoms, with a frequency up to 100 and 87%, respectively (25). Thus, it is expected that these symptoms were the most prevalent complaints leading to diagnosis.
In a series of 310 patients taken from Klijn et al. (32) and Nabarro (7), Molitch (18) reported symptoms at presentation, including menstrual disturbances in 13%, acral changes in 11%, headaches in 8%, carpal tunnel syndrome in 6%, and chance (detected by physicians, dentist, or x-rays) in 40%. Despite the common finding of rheumatological symptoms in acromegaly (33), this was a very uncommon (2%) complaint leading to the diagnosis in this series. Of note, in the last decade, the second most common reason leading to the diagnosis of acromegaly was an incidental finding rather than a sign or symptom of the disease, and this may reflect the increased use of head MRI as part of the assessment of many types of complaints other than headache and routine use after head trauma. Most patients with an incidentally found tumor had a preexisting disease unrelated to acromegaly that necessitated physician visits (74%), and this may have increased contact with a health professional and facilitated identification of acromegaly.
It is important to note that in this series, the diagnosis was most commonly suspected by the primary care physician (44%), which reinforces the critical role these professionals play in making this diagnosis. Few educational resources dealing with GH excess are devoted to primary care and emergency room physicians, although this study shows that these physicians play a major role in an early identification of acromegaly. These findings suggest that resources should be directed toward increased awareness of the disease and its diagnosis in this provider group. Self or relative referrals reflect an increased awareness regarding the disease by the general population during the last decade.
Comorbidities in acromegaly have been well described (34, 35, 36). In their series, Jadresic et al. (35) reported a frequency of 34% for cardiac problems, 32% for hypertension, and 27% for diabetes mellitus. In the large series by Ezzat et al. (36), 50% had hypertension, and 30% had diabetes. A lower than expected proportion of the patients reported in this study had diabetes mellitus (15%), hypertension (25%), or heart disease (5%). Cancer prevalence was also slightly lower in this study (8%) than in the prevalence rate reported based on collective prior series of 10.9% (34). The rate of papillary thyroid cancer at 3% is consistent with recent reports and supports a higher prevalence of thyroid cancer associated with acromegaly than expected in the general population, but ascertainment bias may contribute because these patients were more likely to be seen by physicians and specifically endocrinologists. Although only a single patient had documented sleep apnea at presentation, this likely greatly underestimates the true presence of this condition because formal sleep tests were not routinely preformed to detect this. The decreased morbidity associated with acromegaly in this cohort compared with previously reported series may reflect the shorter time to diagnosis reported in our series compared with others and supports the importance of early diagnosis in minimizing comorbidities (34, 35, 36). The presence of any of these comorbidities was not related to tumor size, presenting complaint, the type of professional making the diagnosis, or the time from onset of symptoms until diagnosis. Although the reported duration of disease did not predict comorbidities, an older age at diagnosis was clearly associated with a higher rate of comorbidity. Because older patients are more likely to have other diseases, age may be a confounder, and causality between later diagnosis and higher morbidity cannot be established. Because most patients were diagnosed early in this cohort, the power to evaluate the impact of prolonged diagnostic delay on the development of comorbidities was limited.
TSS is typically the primary therapy of acromegaly (9, 27) and, as such, was the most common form of therapy among our study population (94%). The surgical results reported here reflect the outcomes from multiple referring physicians across many centers where a dedicated pituitary surgeon may not have been available. Surgical cure rates reported from single centers, including our own, confirm that an experienced pituitary surgeon leads to increased surgical remission rates with fewer complications (14, 20, 28, 37, 38, 39, 40, 41, 42). Radiation therapy was used less frequently after 1994. This decrease in the use of radiation likely reflects the greater availability of medical treatment options (43, 44, 45, 46, 47, 48). Overall, almost 50% of women and 35% of men required medical therapy (octreotide analogs, dopamine agonists, or a GH receptor antagonist) to control their disease after surgery.
In summary, an analysis of records from a tertiary care referral center over two decades showed that the primary care provider and the emergency room physicians are key in identifying patients with acromegaly. In the last two decades, incidental findings, rather than symptoms of acromegaly, were the second most likely reason for the disease to be recognized, and, in the latter decade only, brain MRI done for other reasons led to the discovery of many such cases. In recent practice, the time from the onset of disease to diagnosis of acromegaly is shorter than previously reported, likely related to improvements in diagnostic techniques and disease awareness. In this cohort, with an overall shorter time to diagnosis, the comorbidity rate was also lower than in prior reports. Although the reported duration of disease did not predict comorbidities, a later age at diagnosis was clearly associated with a higher rate of comorbidity. Finally, in the past decade, radiation therapy continues to be used but less commonly than in the preceding decade, and this may be attributable to advances in pharmacological therapies for acromegaly.
| Footnotes |
|---|
First Published Online April 1, 2008
1 L.N. and A.D. contributed equally to this work. ![]()
Abbreviations: MGH, Massachusetts General Hospital; MRI, magnetic resonance imaging; OGTT, oral glucose tolerance test; TSS, transsphenoidal surgery; XRT, radiotherapy.
Received September 24, 2007.
Accepted March 21, 2008.
| References |
|---|
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |