| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Cancer Epidemiology, Karolinska University Hospital (B.N., E.G.-K.), SE-17176 Stockholm; Research Center for Endocrinology and Metabolism, Sahlgrenska University Hospital (B.-Å.B.), SE-41345 Goteborg; and Department of Women and Child Health, St. Goran’s Hospital, Karolinska Institute (B.J.), SE-17177 Stockholm, Sweden
Address all correspondence and requests for reprints to: Prof. Bengt-Åke Bengtsson, Research Center for Endocrinology and Metabolism, Sahlgrenska University Hospital, SE-41345 Goteborg, Sweden.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The main weakness of the previous Swedish and United Kingdom studies (2, 3, 4) was the limited size of the cohorts. The objective of this study, therefore, was to analyze the incidence of PA and the mortality rates and causes of death of affected individuals in the whole of Sweden, which has a population of approximately 8.5 million.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Since the 1970s, the primary processing of cancer notifications has taken place at six regional cancer registries situated in Umea, Uppsala, Stockholm, Linkoping, Goteborg, and Lund. Each regional cancer registry supplies processed data to the Swedish Cancer Registry, which is responsible for coordination of the material. Notification forms for patients diagnosed between 1958 and 1984 are stored at the Swedish Cancer Registry, managed by the National Board of Health and Welfare. Since 1985, the original cancer notification forms have been filed and stored at the regional registries (oncology centers). According to the Swedish National Board of Health and Welfare (6), 98% of all diagnosed cases of tumor are calculated to be have been reported to the Cancer Registry.
Patients with pituitary tumors (code 195.3, ICD 7) diagnosed between 1958 and 1991 constituted the population studied. The original cancer notification forms for these patients were checked to determine whether they contained information on acromegaly or Cushing’s disease.
The Swedish Cancer Registry provided data for all individuals (n = 3321) with pituitary tumors. After exclusion of individuals with pituitary malignancies (n = 87), those who had died less than 1 month after diagnosis (n = 497), those with acromegaly (n = 352) or Cushing’s disease (n = 33), and those for whom the original notification form was missing (n = 73), 2279 individuals with PA remained for analysis (1010 women and 1269 men).
The diagnosis of PA was based on histological examination in 79.8% of cases, radiological examination in 11.7%, clinical examination in 2.3%, postmortem histological examination in 5.4%, and surgery without histological examination in 0.7%. The age-standardized incidence of PA per calendar year between 1958 and 1991 and the age-specific incidence of PA were calculated for men and women separately.
The expected numbers of deaths due to various diseases were calculated from annually published data from the Swedish National Central Bureau of Statistics. These data (for example, see Ref. 6) are specific for the cause of death, gender, calendar year, and age. The standardized mortality ratio (SMR) is defined as the ratio of the observed to the expected number of deaths.
Causes of death for the study population were obtained from the Swedish Cancer Registry. The primary cause of death was coded according to ICD 6–9. Vascular diseases were defined according to ICD9 codes 390–459. Codes 430–438 constitute deaths due to cerebrovascular diseases. Tumors were coded according to ICD7.
Statistical methods
The 95% confidence interval (CI) for disease-specific SMRs was calculated by treating the observed number as a Poisson variable and the expected number as fixed. The 95% CIs are given in parentheses after the SMR. Simple linear regression was performed on the incidence of PA in Sweden, standardized for age to the Swedish population in 1975.
Five-year excess mortality rates (EMRs) were calculated. The EMR in year 5 is defined as 1 minus the cumulative relative survival rate (CRSR) in year 5: EMR5 = 1 - CRSR5. For years 10 and 15 the EMR10 = (CRSR5 - CRSR10)/CRSR5, and the EMR15 = (CRSR10 - CRSR15)/CRSR10.
Multivariate survival analysis (Cox regression) (7), using the background factors age, calendar year, geographical region, and gender, was also carried out. The log-rank test was used to test survival differences between men and women. The standard statistical computer program SPSS (version 7.5, SPSS, Inc., Chicago, IL) was used for analyses.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mean age (±SD) at diagnosis was 52.3 ± 15.7 yr; mean ages for men and women were 53.8 ± 14.7 and 50.3 ± 16.6 yr, respectively. Women constituted 44.3% of the patients.
The mean follow-up time from diagnosis to death or until December 1991 was 10.6 ± 8.5 yr (men, 9.9 ± 8.2 yr; women, 11.6 ± 8.8 yr). The mean follow-up time for survivors was 11.9 ± 8.6 yr, and that for patients diseased before 1992 was 8.5 ± 7.9 yr.
Incidence
The mean age-standardized incidence for the diagnosis of PA
increased significantly during the study period, from approximately 6
cases/million inhabitants in 1958 to 11/million in 1991. This increase
was seen in both men and women (Fig. 1
).
The age-specific incidence peaked between 60–70 yr of age for both
sexes (Fig. 2). Between 1958–1979, the
average age and sex standardized annual reported incidence per yr in
Sweden of PAs was 7.13/million inhabitants; after 1980, it was 9.76
(P < 0.001).
|
|
The median survival time was 18.2 yr for men and 24.8 yr for women
(P < 0.001). The estimated 10-yr survival was 68.9%
for men and 76.4% for women (Fig. 3).
|
. During the first 10 yr after
diagnosis, the EMR was approximately 15% for both men and women.
Between 10–15 yr after diagnosis, the EMR increased to approximately
20%.
|
Mortality
In total, 842 (37%) of the original cohort had died by the time
of follow-up (325 women and 517 men). The overall SMR was 2.0 (CI,
1.9–2.2; Fig. 4a); the SMRs for
men and women were 1.9 (CI, 1.7–2.1) and 2.3 (CI, 2.1–2.5),
respectively.
|
The cause of death was known in 832 of the 842 deaths; the causes of
death of the patients are detailed in Table 2. Cardiovascular disease was the cause
of death in 346 patients. The mean age at death was 72.3 ± 9.0
yr. The overall SMR was 1.6 (CI, 1.4–1.7); the SMRs for men and women
were 1.4 (CI, 1.3–1.7) and 1.8 (CI, 1.5–2.1), respectively. In
patients between 40–69 yr of age, the overall SMR for cardiovascular
deaths was 3.6 (CI, 3.2–4.1; Fig. 4b
); the SMRs for men and
women were 3.1 (CI, 2.6–3.5) and 5.5 (CI, 4.3–6.6), respectively
(P < 0.001).
|
); the SMRs for men and women were 6.7 (CI,
4.8–8.7) and 7.3 (CI, 4.6–10.1), respectively. Excess mortality was also observed for tumors; the primary cause of death was given as tumor in 244 patients, and the overall SMR was 2.4 (CI, 2.1–2.7). After exclusion of patients in whom pituitary tumor was the primary cause of death (n = 95), the SMR decreased to 1.4 (CI, 1.21–1.7). No significant difference between the sexes was observed. The SMR for malign neoplasms was 1.5 (CI, 1.2–1.7). This decreased to a statistically insignificant level when the 38 cases of "other endocrine glands," that is pituitary-related diseases, were subtracted from the observed number (n = 148). However, 18 patients died from malignant tumors of the brain, with a SMR of 7.1 (CI, 4.2–11.3).
Endocrine diseases were the cause of death in 116 patients, with a SMR of 16.1 (CI, 13.2–19.0). In these 116 subjects, the cause of death was pituitary related in 106 patients; diabetes insipidus was the cause of death in 23 patients. Seven patients died as a result of diabetes mellitus, and 3 patients died as a result of adrenal insufficiency. Twenty-eight patients died from gastrointestinal diseases, with a SMR of 2.0 (CI, 1.3–2.9).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In contrast to usual observations in tumor epidemiology, the EMR increased after diagnosis for both men and women. During the first 10 yr after diagnosis, the EMR was approximately 15%; however, between 10–15 yr after diagnosis, the EMR increased to approximately 20%. Patients who were diagnosed more recently lived longer than those who were diagnosed earlier. These observations suggest that the causes of excess mortality were not linked to the PA per se, as most of them were probably removed by surgery, but were associated with other factors caused by the PA or its treatment, such as the development of hypopituitarism and the adequacy of its subsequent treatment.
The main finding from this retrospective registry study of 2279 patients with PA was a significant increase in overall mortality, mainly as a result of cardiovascular disease. The increased risk was significant for both cardiac and cerebrovascular deaths.
Changes in death rates in the general Swedish population during the 33-yr study period have been considered in the calculations. Expected mortality was obtained from cause-, sex-, calendar year-, and 5-yr age group-specific death rates for Sweden. Furthermore, we have no reason to believe that the patients differed from the general population with respect to variables such as consumption of alcohol, smoking habits, and ethnic or socio-economic background. If anything, there is a suspicion that hypopituitary patients smoke less than the general population (8).
Information obtained from the Swedish Cause of Death register has proved to be accurate for cardiovascular diseases (9). In the present study, however, it is conceivable that cardiovascular deaths have been underestimated, as there were 116 patients reported to have died from an endocrine disorder and 96 reported to have died from a benign or unspecified tumor. Such a suspicion is supported by the observations of Rosén and Bengtsson (2) and Bulow et al. (3), who reported that 4 of 19 and 15 of 17 patients, respectively, were classified as having died as a result of pituitary tumors when, in fact, they had died from cardiovascular causes.
Consistent with the results reported by Rosén and Bengtsson (2) and Bulow et al. (3), there was an increased risk of death from cardiovascular disease in the present study. The increase in cardiovascular deaths was most prominent in the 40- to 69-yr-old group. In contrast to the results presented by Bulow et al. (3), the increase in deaths from cerebrovascular causes did not differ significantly between the sexes, but, as shown previously (3), the increased cerebrovascular mortality was pronounced even in the younger patients.
Replacement therapy with cortisone acetate, thyroid hormones, and sex hormones has been used for more than 40 yr in the treatment of hypopituitarism. GH is usually the first hormone to decrease in the development of hypopituitarism. During the last 10 yr, the consequences of untreated growth hormone deficiency in adults have been delineated (10), and a number of cardiovascular risk factors, such as visceral obesity, high/low density lipoprotein cholesterol ratio, and blood pressure, have been found to improve after GH replacement therapy (11, 12). In addition cardiac function has been reported to improve after GH treatment (13, 14). Imperfect conventional replacement therapy and untreated GH deficiency might therefore explain the increased mortality due to cardiovascular disorders found in the present study.
The increased incidence of deaths resulting from malignant brain tumors is in contrast to that in previous studies (2, 3). In addition, there was a nonsignificant increase in leukemia. It is conceivable that several patients were treated with pituitary irradiation after surgery, which could explain the increase in these types of tumors (15).
The increased SMR for gastrointestinal diseases is surprising. Hypothetically, patients with acute gastrointestinal disorders in need of acute surgical care might not have been properly treated with corticosteroids in the event of concomitant ACTH deficiency.
In the present study we have not analyzed the incidence of pituitary failure for practical reasons. However, we can assume that most patients suffered from macroadenomas, which are associated with a high incidence of concomitant pituitary failure (16). In addition, 80% of the patients were operated upon, which suggests that most of the tumors were macroadenomas.
In conclusion, in this large cohort of subjects with pituitary adenoma reported to the Swedish Cancer Registry between 1958 and 1991, the incidence of pituitary adenomas was found to increase. In addition, excess mortality was found in the study population, mainly due to increased cardiovascular mortality.
|
Footnotes |
|---|
Received March 16, 1999.
Revised November 23, 1999.
Accepted December 15, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Stochholm, C. H. Gravholt, T. Laursen, P. Laurberg, M. Andersen, L. O. Kristensen, U. Feldt-Rasmussen, J. S. Christiansen, M. Frydenberg, and A. Green Mortality and GH deficiency: a nationwide study Eur. J. Endocrinol., July 1, 2007; 157(1): 9 - 18. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. M. Dekkers, N. R. Biermasz, A. M. Pereira, F. Roelfsema, M. O. van Aken, J. H. C. Voormolen, and J. A. Romijn Mortality in Patients Treated for Cushing's Disease Is Increased, Compared with Patients Treated for Nonfunctioning Pituitary Macroadenoma J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 976 - 981. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Ferrante, M. Ferraroni, T. Castrignano, L. Menicatti, M. Anagni, G. Reimondo, P. Del Monte, D. Bernasconi, P. Loli, M. Faustini-Fustini, et al. Non-functioning pituitary adenoma database: a useful resource to improve the clinical management of pituitary tumors Eur. J. Endocrinol., December 1, 2006; 155(6): 823 - 829. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. F. Daly, M. Rixhon, C. Adam, A. Dempegioti, M. A. Tichomirowa, and A. Beckers High Prevalence of Pituitary Adenomas: A Cross-Sectional Study in the Province of Liege, Belgium J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 4769 - 4775. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Molitch, D. R. Clemmons, S. Malozowski, G. R. Merriam, S. M. Shalet, M. L. Vance, and for The Endocrine Society's Clinical Guidelines Su Evaluation and Treatment of Adult Growth Hormone Deficiency: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1621 - 1634. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V K B Prabhakar and S M Shalet Aetiology, diagnosis, and management of hypopituitarism in adult life. Postgrad. Med. J., April 1, 2006; 82(966): 259 - 266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. S. Dumont, E. C. Nemergut II, J. A. Jane Jr, and E. R. Laws Jr Postoperative Care Following Pituitary Surgery J Intensive Care Med, May 1, 2005; 20(3): 127 - 140. [Abstract] [PDF]
|
|
|
|
|
|
K. K. Miller, B. M. K. Biller, J. G. Lipman, G. Bradwin, N. Rifai, and A. Klibanski Truncal Adiposity, Relative Growth Hormone Deficiency, and Cardiovascular Risk J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 768 - 774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Svensson, B.-A. Bengtsson, T. Rosen, A. Oden, and G. Johannsson Malignant Disease and Cardiovascular Morbidity in Hypopituitary Adults with or without Growth Hormone Replacement Therapy J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3306 - 3312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Twickler, M. J. M. Cramer, G. M. Dallinga-Thie, M. J. Chapman, D. W. Erkelens, and H. P. F. Koppeschaar Adult-Onset Growth Hormone Deficiency: Relation of Postprandial Dyslipidemia to Premature Atherosclerosis J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2479 - 2488. [Full Text] [PDF]
|
|
|
|
|
|
T. B. Twickler, G. M. Dallinga-Thie, F. L. J. Visseren, W. R. de Vries, D. W. Erkelens, and H. P. F. Koppeschaar Induction of Postprandial Inflammatory Response in Adult Onset Growth Hormone Deficiency Is Related to Plasma Remnant-Like Particle-Cholesterol Concentration J. Clin. Endocrinol. Metab., March 1, 2003; 88(3): 1228 - 1233. [Abstract] [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 |
