- Email: [email protected]
Transsphenoidal adenomectomy for microprolactinomas: 10 to 20 years of follow-up
ELSEVIER
Featured Subject: Neoplasms & Diseases of the Pituitary
TRANSSPHENOIDAL ADENOMECTOMY FOR MICROPROLACTINOMAS: 10 TO 20 YEARS OF FOLLOW-UP Fadi Massoud, M.D., Omar Serri, M.D., Ph.D., Jules Hardy, M.D., Maurice Somma, M.D., and Hugues Beauregard, M.D. Services of Endocrinology and Neurosurgery, Notre-Dame Hospital, and University of Montreal, Montrkal, Que’bec, Canada
Massoud F, Serri 0, Hardy J, Somma M, Beauregard sphenoidal adenomectomy for microprolactinomas: years of follow-up. Surg Neurol 1996;45:341-6.
H. Trans10 to 20
KEY
WORDS
Microprolactinomas,tmnssphenoidaladenomectomy,relapse of hyperprolactinemia, tumor recurrence.
BACKGROUND
Transsphenoidal adenomectomy is an effective treatment for microprolactinomas. However, postoperative recurrence of hyperprolactinemia is not rare. This study was designed to evaluate the long-term outcome of women
with microprolactinomas da1 approach.
operated on by transsphenoi-
METHODS
We retrospectively studied 64 women with microprolactinomas who underwent transsphenoidal adenomectomy and were followed for 10 to 20 years. RESULTS
Postoperatively,
58 women (90%) had normal plasma pro-
lactin concentrations (<20 pg/L). After a mean of 3.3 years, during which the women were asymptomatic with normoprolactinemia, 25 (43%) had a relapse of hyper-
prolactinemia (~20 pg/L). However, their evolution varied. Fifteen women had symptomatic hyperprolactinemia. Computed tomography (CT) scans showed recurrent microadenomas in 2 women. The other 10 women had only hyperprolactinemia. Of these women, 5 had transient hyperprolactinemia (29 + 4 pg/L) for 5 years, after which prolactin declined to normal (13 ? 3 pg/L). The remaining five patients had elevated prolactin (31 5 3 pg/L) throughout the follow-up period (10 to 20 years). CT scan did not show recurrent adenomas in
these women. Thirty-three
women remained normopro-
lactinemic and asymptomatic for a mean period of 12 years (range, 10 to 20 years). CONCLUSIONS
In conclusion, most of the patients with late relapse of hyperprolactinemia have slight functional hyperprolactinemia and remain tumor recurrence.
asymptomatic
with no evidence
of
Address reprint requests to: Omar Serri, M.D., Ph.D., Unit& Metabolique, HBpital Notre-Dame, 1560, rue Sherbrooke Est, Montreal (Quebec), H2L 4Ml. Received June 7, 1995; accepted September 29, 1995. 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
P
rolactinoma represents the most frequent pituitary tumor. It is classified as either microprolactinoma (diameter,
342
Massoud et al
Surg Neurol 1996;45:341-6
prolactinomas [ 1,2,7,11-13,151. In most patients there was no radiologic evidence of tumor recurrence, and some patients were totally asymptomatic. The long-term clinical, biologic, and radiologic evolution of relapsing hyperprolactinemia is, however, unknown. The growth potential of microprolactinoma can be very slow, and the number of recurrences can increase with time. So far, the longest follow-up period after surgery for microprolactinomas does not extend beyond 10 years [ 171. For this reason, we studied 64 women with microprolactinomas, who were operated on by transsphenoida1 approach, and who were followed postoperatively for a period of at least 10, and up to 20, years.
MATERIALSANDMETHODS From a large series of 763 patients with prolactinomas operated on by Dr. Hardy in three university hospitals in Montreal, 215 women were treated at Notre Dame Hospital between 1970 and 1983. Of these, 64 women with microprolactinomas (who were regularly followed for a period of at least 10, and up to 20, years and with a mean of 12.3 + 2.8 years) are the subject of this report. All patients had biologic, surgical, and pathologic confirmation of the diagnosis of microprolactinoma. Measurements of plasma prolactin concentrations were made in the immediate postoperative period (lo-14 days), and annually thereafter. We also evaluated clinical evolution and the radiologic aspect of the pituitary fossa by computed tomography (CT) scans. The statistical analysis was done with the unpaired t-test (or the Wilcoxon’s rank sum test when variance was important) to compare average postoperative prolactin concentrations in the different subgroups of patients.
0
2
4
6
6
10
12
14
16
16
20
YEARS
Postoperative plasma prolactin concentrations five patients with transient hyperprolactinemia.
of
of hyperprolactinemia. The other 10 women had only hyperprolactinemia. Of these women, 5 had slightly elevated plasma prolactin concentrations (29 +- 4 pg/L) for a mean period of 5 years (4-8 years), after which the concentrations declined to normal (13 f 3 pg/L) without any medication (Figure 1). The remaining 5 patients had similarly elevated plasma prolactin concentrations (31 + 3 pg/L) throughout the mean follow-up period of 12 years (range, lo-20 years) (Figure 2) and normal CT scans. The incidence of pregnancies in the two groups before the relapse of hyperprolactinemia was similar. Of the 58 patients initially cured, thirty-three remained asymptomatic with normal plasma prolactin concentrations for a mean period of 12 + 3 years (range, lo-20 years). The average postoperative plasma prolactin level of the patients in the different subgroups are presented in Table 1. Immediate postoperative plasma prolactin concentrations in women who remained in prolonged remission were significantly lower than those of patients who later presented with a
RESULTS In the immediate postoperative period, 58 patients (90%) had a normal plasma prolactin concentration (<20 pg/L). In these women, galactorrhea disappeared, and menstrual cycles were normalized. Twenty-seven of these patients had 40 spontaneous pregnancies. After a mean of 3.3 years (l-6 years), 25 patients had a relapse of hyperprolactinemia (220 kg/L). Hyperprolactinemia was associated with amenorrhea and/or galactorrhea in 15 of the 25 patients (average prolactin levels: 49 -C 4 pg/L). CT scans showed a recurrence of the microadenoma in 2 patients, 5 and 9 years respectively after the relapse
0
2
4
6
6
10
12
14
16
16
20
YEARS
Postoperative qlactinemia. tions of five
mean plasma prolactin
concentra-
patients
hyperpro-
with
persistent
Follow-up of Microprolactinomas
II
Surg Neurol 1996;45:341-6
Postoperative Plasma Prolactin Concentrations Patients with Microprolactinomas AVERAGE PROLACTIN w/L
PATIENTS
(1) Remissions (n = 32) (2) Symptomatic relapses (n = 15)
+
in
STANDARD DEWTION
4.9 10.3
+ r
3.3 3.9
12.2 5.2
+ +
6 5.2
Asymptomatic hyperprolactinemia
(3) Persistent (n = 5) (4) Transient (n = 5) *p < 0.05 For the following
comparisons:
1 vs. 2, 1 vs. 3, 2 vs. 4.
relapse of hyperprolactinemia. Similarly, the immediate postoperative plasma prolactin levels of patients who developed transient hyperprolactinemia were significantly lower than those of patients who later had symptomatic relapse of hyperprolactinemia.
DISCUSSION Our study confirms the high success rate of shortterm prolactin normalization after transsphenoidal adenomectomy in women with microprolactinomas [22]. It also confirms the high frequency of late relapses of hyperprolactinemia [15]. With 10 to 20 years of postoperative follow-up, we were able to determine the evolution of relapsing hyperprolactinemia, which tended often to be slight and asymptomatic. Hyperprolactinemia can occur as a transient event, lasting a few years and then disappearing. In these cases, there is no evidence of tumor recurrence, and no treatment is required. At variance, hyperprolactinemia can be moderate and symptomatic. In such cases, hyperprolactinemia does not regress spontaneously and can evolve to radiologic tumor recurrence as observed in two of our patients (one patient after 5, and the other after 9, years). In our study, the overall rate of relapse of hyperprolactinemia was 43%, which is higher than the rates generally reported in other studies [ 1,2,7,1113,181. This high rate can be due to the longer follow-up. We identified a case of relapse occurring as long as 10 years after surgery. The other studies with shorter follow-up periods may have underestimated the rate of relapse of hyperprolactinemia. Alternatively, different surgical techniques (selective microadenomectomy, as opposed to enlarged adenomectomy [3] or partial hypophysectomy [19]) can also explain different rates of relapsing hyperprolactinemia. Selective microadenomectomy aims at the removal of the adenoma with
343
preservation of normal pituitary tissue and, as a consequence, may possibly achieve only incomplete resection with a higher risk of relapse. However, a more aggressive resection can ensure the complete removal of the adenoma with an increased risk of alteration of other normal pituitary secretions. It should be stressed that most women with relapsing hyperprolactinemia remained asymptomatic. It is possible that some of these patients have a high-molecular-weight, biologically inactive prolactin as described in certain patients with hyperprolactinemia [5]. However, such a phenomenon is rare in prolactinomas. Another possibility is the existence of a critical level of hyperprolactinemia under which patients are asymptomatic. This observation has already been made in patients treated with bromocriptine or other dopaminergic agonists, with some patients achieving ovulatory menstrual cycles even when prolactin was lowered to supranormal values [ 81. This study confirms the prognostic value of immediate postoperative plasma prolactin concentration [15,22]. Women who later had relapse of hyperprolactinemia had an average postoperative plasma prolactin of 10 kg/L or more, whereas women with prolonged remission had an average plasma prolactin of approximately 5 pg/L.
CONCLUSION (1)Surgical resection
of microprolactinoma can achieve 90% normalization of plasma prolactin levels postoperatively. (21 Most patients with late relapse of hyperprolactinemia remain asymptomatic and do not show recurrence of the adenomas, 10 to 20 years following adenomectomy. Transsphenoidal adenomectomy remains a safe (3) and efficient therapeutic approach in women with microprolactinomas and can be recommended as an alternative to medical treatment, or for patients who are intolerant or resistant to dopamine agonists. The authors thank Femanda Janeiro for her secretarial assistance in the preparation of the manuscript.
REFERENCES 1. Buchfelder M, Fahlbusch R, Schotl W, Honegger J. Long-term follow-up results in hormonally active pituitary adenomas after primary successful transsphenoidal surgery. Acta Neurochir (Wien) 1991; 53(suppl):72-6.
344
Massoud et al
Surg Neurol 1996;45:341-6
2. Ciccarelli E, Ghigo E, Miola C, Gandini G, Muller EE, Camanni F. Long-term follow-up of “cured” prolactinoma patients after successful adenomectomy. Clin Endocrinol 1990;32:583-92. 3. Grisoli F, Brue T, Graziani N, Costa R, Trouillas J, Begou D, Jaquet P. Enlarged adenomectomy for enclosed prolactinomas: a preliminary study of 26 cases. Acta Neurochir (Wien) 1990;103:92-8. 4. Hardy J, Beauregard H, Robert F. Prolactin-secreting pituitary adenomas: trassphenoidal microsurgical treatment. In: Robyn C, Harter M, eds. Progress in prolactin: physiology and pathology. Proceedings of the International Symposium on Prolactin, Nice, October 20-23, 1977. North Holland: Elsevier, 1978:36170. 5. Jackson RD, Wortsman J, Malarkey WB. Characterisation of a large molecular weight prolactin in women with idiopathic hyperprolactinemia and normal menses. J Clin Endocrinol Metab 1985;61:258-63. 6. Klibanski A, Neer RM, Beitins IZ, Ridgway EC, Zervas NT, McArthur JW. Decreased bone density in hyperprolactinemic women. N Engl J Med 1980;303:1511-4. 7. Maira G, Anile C, De Marinis L, Barbarino A. Prolactinsecreting adenomas: surgical results and long-term follow-up. Neurosurgery 1989;24:736-43. Endocri8. Molitch ME. Pathologic hyperprolactinemia. no1 Metab Clin North Am 1992;21:877-901. 9. Post KD, Biller BJ, Adelman LS, Molitch ME, Wolpert SM, Reichlin S. Selective transsphenoidal adenomectomy in women with galactorrhea-amenorrhea. JAMA 1979;242:158-62. 10. Randall RV, Laws ER, Abboud CF, Ebersold MJ, Kao PC, Scheithauer BW. Transsphenoidal microsurgical treatment of prolactin-producing pituitary adenomas: results in 100 patients. Mayo Clin Proc 1983;58:10821. 11. Rodman EF, Molitch ME, Post KD, Biller BJ, Reichlin S. Long-term follow-up of transsphenoidal selective adenomectomy for prolactinoma. JAMA 1984;252:921-4. 12. Scamoni C, Balzarini C, Crivelli G, Dorizzi A. Treatment and long-term follow-up results of prolactin secreting pituitary adenomas. J Neurosurg Sci 1991;35: 9-16. 13. Schlechte JA, Sherman BM, Chapler FK, Van Gilder J. Long-term follow-up of women with surgically treated prolactin-secreting pituitary tumors. J Clin Endocrino1 Metab 1986;62:1296-1301. 14. Schlechte J, Dolan K, Sherman B, Chapler F, Luciano A. The natural history of untreated hyperprolactinemia: a prospective analysis. J Clin Endocrinol Metab 1989;68:412-8. 15. Serri 0, Rasio E, Beauregard H, Hardy J, Somma M. Recurrence of hyperprolactinemia after selective transsphenoidal adenomectomy in women with prolactinoma. N Engl J Med 1983;309:280-3. 16. Thomson JA, Teasdale GM, Gordon D, McCruden D, Davies DL. Treatment of presumed prolactinoma by transsphenoidal operation: early and late results. Br Med J 1985;291:1550-3. 17. Thomson JA, Davies DL, McLaren EH, Teasdale GM. Ten year follow-up of microprolactinoma treated by transsphenoidal surgery. Br Med J 1994;309:1409-10. 18. Tucker HSG, Grubb SR, Wigand JP, Taylon A, Lankford HV, Blackard WG, Becker DP. Galactorrhea-amenor-
19.
20.
21. 22.
rhea syndrome: follow-up of 45 patients after pituitary tumor removal. Ann Intern Med 1981;94:302-7. Webster J, Page MD, Bevan TS, Richards SH, DouglasJones A, Scanlon MF. Low recurrence rate after partial hypophysectomy for prolactinoma: the predictive value of prolactin function tests. Clin Endocrinol (Oxf) 1992;36:35-44. Weiss MH, Teal J, Gott P, Wycoff R, Yadley R, Apuzzo ML, Giannotta FL, Kletzky 0, March C. Natural history of microprolactinomas: six-year follow-up. Neurosurgery 1983;12:180-3. Wilson CB. A decade of pituitary microsurgery: the Herbert Olivecrona Lecture. J Neurosurg 1984;61:814-33. Wilson CB. Role of surgery in the management of pituitary tumors. Neurosurg Clin North Am 199O;l: 139-59.
COMMENTARY This paper, from a distinguished group of investigators, provides valuable information on the outcome of patients who develop postoperative hyperprolactinemia following transsphenoidal surgery for microprolactinomas. Out of 64 women with microprolactinomas who underwent surgery and were followed for a period of 10 to 20 years, a total of 25 (43%) developed postoperative hyperprolactinemia. However, the value of their long-term follow-up is that it showed that in only two patients was this due to recurrent tumor. The message from this study is for clinicians not to assume that all patients who develop hyperprolactinemia following successful transsphenoidal surgery have recurrent tumor, which carries significant therapeutic implications. In my opinion, it is best to follow, rather than to aggressively treat, these patients. This is a valuable paper. I wish the authors had provided more discussion on possible mechanisms of postoperative hyperprolactinemia, which at present remains a mystery. However, a discussion of possible causes would have been of interest to the reader. George T. Tindall, M.D. Atlanta, Georgia The paper by Massoud and colleagues is a welcome extension of the previous study which appeared in the New England Journal of Medicine in 1983, concerning many of the same patients. Despite the deficiencies and the methodology of the retrospective clinical study published in the New England Journal, the results have been widely quoted in the endocrine and neurosurgical literature and have been utilized to argue against surgical therapy for prolactin-secreting microadenomas because of the
Follow-up of Microprolactinomas
alleged 40% “recurrence” rate that Massoud et al found in their study. Ever since the publication of the initial article, George Tindall, Jules Hardy, and others have maintained that recurrent modest hyperprolactinemia after surgical removal of a prolactin-secreting microadenoma did not necessarily mean that the tumor had been left behind. That argument is confirmed by the present study, which shows that on long-term review of 64 women with prolactinsecreting microadenomas surgically removed, only 2 of 25 patients who had recurrent hyperprolactinemia actually developed CT-detectable recurrent tumors during a follow-up period that ranged from 10 to 20 years. These findings are all the more compelling because the authors of this paper report a much higher rate of recurrent hyperprolactinemia, following initial remission after transsphenoidal microsurgery, than others who have presented similar data in patients with more consistent follow-up. The main drawback of the Montreal study is that it does not systematically follow all women surgically treated for prolactin microadenomas, but it has a selected cohort of only 64 of a potential 215 women treated for microprolactinoma. The selection criteria were not mentioned in either paper, and we do not have information regarding the fate of the women who were not included. Recent studies of prolactin-secreting microadenomas treated by transsphenoidal surgery show an initial remission rate in 80%-90% of patients. The incidence of subsequent hyperprolactinemia in studies that more thoroughly report the follow-up of the patients, such as a recent review by John Thompson of Graham Teasdale’s series from Glasgow, reveals an initial normalization rate of 86.5%, a long-term (greater than 10 years) sustained remission in 74%, and a recurrence of hyperprolactine mia in 13% [2]. Our own careful follow-up study of the patients treated at the Mayo Clinic revealed a 24% incidence of recurrence of hyperprolactinemia in a group of patients followed for a mean of 5?4 years [l]. A common feature of these follow-up studies is that despite the incidence of recurrent hyperprolactinemia, the actual diagnosis of the recurrent pituitary tumor is quite uncommon. The current report leaves us with a number of unanswered questions. Massoud et al’s follow-up study includes 64 women, 58 of whom had “normal” prolactin levels following surgery. The fate of the remaining 6 women is not further discussed. Of the 58 women who had initial remissions, their clinical features are not presented and we do not know how many of them had resumption of ovulatory menses or resolution of galactorrhea. Among the 25 women
Surg Neural 1996:45:341-6
345
who had recurrent hyperprolactinemia, it is stated that 10 women had “only hyperprolactinemia,” but the extent of this abnormality and the reproducibility on serial laboratory studies is not mentioned. It certainly is possible that these abnormal values might simply represent normal fluctuations that one sees in prolactin levels. The assessment of whether or not a recurrent tumor developed is based on CT evidence, and obviously it would have been preferable to base this particular determination on high-resolution magnetic resonance imaging scans. Hyperprolactinemia alone may not be the best way to judge the success of treatment for prolactinsecreting microadenomas. The important issues are whether the elevation in prolactin has a pathologic effect interfering with the menstrual cycle, fertility, or producing galactorrhea, and whether the elevations detected and reported are consistent findings that have a significant impact on the quality of life of the patient. The bottom line in this study is one that should be encouraging to most neurosurgeons involved in the surgery of pituitary tumors. It appears that the long-term outlook for patients who require surgical management of prolactinomas is better than we have been led to be lieve, and surgery for these lesions should be viewed not as a failure, but as a useful adjunct to an intelligently monitored program for treatment. Edward
R. Laws, Jr., M.D., FACS Mary Lee Vance, M.D.
Charlottesville, Virginia 1 Laws ER, Ebersold MJ, Piepgras DC, Abboud CF, Randall RM, Scheithauer BW. The role of surgery in the management of prolactinoma in prolactin. In: MacLeod RM, Thorner MO, Scapagnini U, eds. Prolactin, basic and clinical correlates. New York: Springer-Verlag, 1985:849-53. 2. Thomsen JA, Davies DL, McLaren EH, Teasdale GM. Ten-year follow-up of microprolactinoma treated by transsphenoidal surgery. Br Med J 1994;309:1409-10.
AUTHORS REPLY Dr. Tindall suggests that we provide more discussion of the possible mechanisms of postoperative hyperprolactinemia. A relapse of hyperprolactinemia after surgical removal of a microprolactinoma is, in some instances, the consequence of regrowth of tumor remnants. Alternatively, a hypothalamic-pituitary dysfunction mayoccurbecauseofdecreaseddeliveryofprolactininhibiting factors due to a scarring process at the site of the tumor cavity after surgery. However, in other instances, different mechanisms may be more
346
Massoudet al
Sur Neurol 199t ;45:341-6
likely. Functional hyperprolactinemia must be considered independent of the surgical procedure. Hyperprolactinemia may be the consequence of certain drugs, stress, or primary hypothyroidism. Some reports have documented the presence of a large molecular-weight prolactin with reduced bioactivity as the major immunoreactive component of prolactin in women with sustained hyperprolactinemia and norma ovulatory menses. Transient and fluctuating hyperprolactinemia has aiso been reported in women with infe~ility and short-luteal phase and in women with various degrees of the amenorrhea-galactorrhea syndrome, probably as a result of imbalance between low progesterone and unopposed estrogen levels. Drs. laws and Vance question several points of the methodoio~ of our study. The long-term follow-up of all patients with microprolactinomas operated by Dr. Hardy during the 70s was unfortunately impossible because most of these patients were referred from abroad. We therefore studied, without further selection, all of the women who, firstly, had proven microprolactino-
mas and, secondly, were followed for at least 10 years after surgery (22 patients were included in our previous report) [15]. Six women remained hyperprolactinemic and symptomatic following surgery. These patients were treated successfully with bromocriptine. All 58 women described with initial remissions had resumption of regular menstrual cycles and resolution of galactorrhea. Although ovulation was not documented by plasma progesterone levels in many patients, 27 had 40 spontaneous pregnancies. Plasma prolactin concentrations depicted at each time interval (Figures 1 and 2) are, for each woman, the mean of at least two values on different days. In some cases, the mean plasma prolactin concentration was calculated from six samples in a 24-hour period. The magnitude of hyperprolactinemia is shown in these figures. Omar Serri, M.D., Ph.D. Jules Hardy, M.D. congeal,
Quebec, Canada
Featured Subject: Neoplasms & Diseases of the Pituitary
TRANSSPHENOIDAL ADENOMECTOMY FOR MICROPROLACTINOMAS: 10 TO 20 YEARS OF FOLLOW-UP Fadi Massoud, M.D., Omar Serri, M.D., Ph.D., Jules Hardy, M.D., Maurice Somma, M.D., and Hugues Beauregard, M.D. Services of Endocrinology and Neurosurgery, Notre-Dame Hospital, and University of Montreal, Montrkal, Que’bec, Canada
Massoud F, Serri 0, Hardy J, Somma M, Beauregard sphenoidal adenomectomy for microprolactinomas: years of follow-up. Surg Neurol 1996;45:341-6.
H. Trans10 to 20
KEY
WORDS
Microprolactinomas,tmnssphenoidaladenomectomy,relapse of hyperprolactinemia, tumor recurrence.
BACKGROUND
Transsphenoidal adenomectomy is an effective treatment for microprolactinomas. However, postoperative recurrence of hyperprolactinemia is not rare. This study was designed to evaluate the long-term outcome of women
with microprolactinomas da1 approach.
operated on by transsphenoi-
METHODS
We retrospectively studied 64 women with microprolactinomas who underwent transsphenoidal adenomectomy and were followed for 10 to 20 years. RESULTS
Postoperatively,
58 women (90%) had normal plasma pro-
lactin concentrations (<20 pg/L). After a mean of 3.3 years, during which the women were asymptomatic with normoprolactinemia, 25 (43%) had a relapse of hyper-
prolactinemia (~20 pg/L). However, their evolution varied. Fifteen women had symptomatic hyperprolactinemia. Computed tomography (CT) scans showed recurrent microadenomas in 2 women. The other 10 women had only hyperprolactinemia. Of these women, 5 had transient hyperprolactinemia (29 + 4 pg/L) for 5 years, after which prolactin declined to normal (13 ? 3 pg/L). The remaining five patients had elevated prolactin (31 5 3 pg/L) throughout the follow-up period (10 to 20 years). CT scan did not show recurrent adenomas in
these women. Thirty-three
women remained normopro-
lactinemic and asymptomatic for a mean period of 12 years (range, 10 to 20 years). CONCLUSIONS
In conclusion, most of the patients with late relapse of hyperprolactinemia have slight functional hyperprolactinemia and remain tumor recurrence.
asymptomatic
with no evidence
of
Address reprint requests to: Omar Serri, M.D., Ph.D., Unit& Metabolique, HBpital Notre-Dame, 1560, rue Sherbrooke Est, Montreal (Quebec), H2L 4Ml. Received June 7, 1995; accepted September 29, 1995. 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
P
rolactinoma represents the most frequent pituitary tumor. It is classified as either microprolactinoma (diameter,
342
Massoud et al
Surg Neurol 1996;45:341-6
prolactinomas [ 1,2,7,11-13,151. In most patients there was no radiologic evidence of tumor recurrence, and some patients were totally asymptomatic. The long-term clinical, biologic, and radiologic evolution of relapsing hyperprolactinemia is, however, unknown. The growth potential of microprolactinoma can be very slow, and the number of recurrences can increase with time. So far, the longest follow-up period after surgery for microprolactinomas does not extend beyond 10 years [ 171. For this reason, we studied 64 women with microprolactinomas, who were operated on by transsphenoida1 approach, and who were followed postoperatively for a period of at least 10, and up to 20, years.
MATERIALSANDMETHODS From a large series of 763 patients with prolactinomas operated on by Dr. Hardy in three university hospitals in Montreal, 215 women were treated at Notre Dame Hospital between 1970 and 1983. Of these, 64 women with microprolactinomas (who were regularly followed for a period of at least 10, and up to 20, years and with a mean of 12.3 + 2.8 years) are the subject of this report. All patients had biologic, surgical, and pathologic confirmation of the diagnosis of microprolactinoma. Measurements of plasma prolactin concentrations were made in the immediate postoperative period (lo-14 days), and annually thereafter. We also evaluated clinical evolution and the radiologic aspect of the pituitary fossa by computed tomography (CT) scans. The statistical analysis was done with the unpaired t-test (or the Wilcoxon’s rank sum test when variance was important) to compare average postoperative prolactin concentrations in the different subgroups of patients.
0
2
4
6
6
10
12
14
16
16
20
YEARS
Postoperative plasma prolactin concentrations five patients with transient hyperprolactinemia.
of
of hyperprolactinemia. The other 10 women had only hyperprolactinemia. Of these women, 5 had slightly elevated plasma prolactin concentrations (29 +- 4 pg/L) for a mean period of 5 years (4-8 years), after which the concentrations declined to normal (13 f 3 pg/L) without any medication (Figure 1). The remaining 5 patients had similarly elevated plasma prolactin concentrations (31 + 3 pg/L) throughout the mean follow-up period of 12 years (range, lo-20 years) (Figure 2) and normal CT scans. The incidence of pregnancies in the two groups before the relapse of hyperprolactinemia was similar. Of the 58 patients initially cured, thirty-three remained asymptomatic with normal plasma prolactin concentrations for a mean period of 12 + 3 years (range, lo-20 years). The average postoperative plasma prolactin level of the patients in the different subgroups are presented in Table 1. Immediate postoperative plasma prolactin concentrations in women who remained in prolonged remission were significantly lower than those of patients who later presented with a
RESULTS In the immediate postoperative period, 58 patients (90%) had a normal plasma prolactin concentration (<20 pg/L). In these women, galactorrhea disappeared, and menstrual cycles were normalized. Twenty-seven of these patients had 40 spontaneous pregnancies. After a mean of 3.3 years (l-6 years), 25 patients had a relapse of hyperprolactinemia (220 kg/L). Hyperprolactinemia was associated with amenorrhea and/or galactorrhea in 15 of the 25 patients (average prolactin levels: 49 -C 4 pg/L). CT scans showed a recurrence of the microadenoma in 2 patients, 5 and 9 years respectively after the relapse
0
2
4
6
6
10
12
14
16
16
20
YEARS
Postoperative qlactinemia. tions of five
mean plasma prolactin
concentra-
patients
hyperpro-
with
persistent
Follow-up of Microprolactinomas
II
Surg Neurol 1996;45:341-6
Postoperative Plasma Prolactin Concentrations Patients with Microprolactinomas AVERAGE PROLACTIN w/L
PATIENTS
(1) Remissions (n = 32) (2) Symptomatic relapses (n = 15)
+
in
STANDARD DEWTION
4.9 10.3
+ r
3.3 3.9
12.2 5.2
+ +
6 5.2
Asymptomatic hyperprolactinemia
(3) Persistent (n = 5) (4) Transient (n = 5) *p < 0.05 For the following
comparisons:
1 vs. 2, 1 vs. 3, 2 vs. 4.
relapse of hyperprolactinemia. Similarly, the immediate postoperative plasma prolactin levels of patients who developed transient hyperprolactinemia were significantly lower than those of patients who later had symptomatic relapse of hyperprolactinemia.
DISCUSSION Our study confirms the high success rate of shortterm prolactin normalization after transsphenoidal adenomectomy in women with microprolactinomas [22]. It also confirms the high frequency of late relapses of hyperprolactinemia [15]. With 10 to 20 years of postoperative follow-up, we were able to determine the evolution of relapsing hyperprolactinemia, which tended often to be slight and asymptomatic. Hyperprolactinemia can occur as a transient event, lasting a few years and then disappearing. In these cases, there is no evidence of tumor recurrence, and no treatment is required. At variance, hyperprolactinemia can be moderate and symptomatic. In such cases, hyperprolactinemia does not regress spontaneously and can evolve to radiologic tumor recurrence as observed in two of our patients (one patient after 5, and the other after 9, years). In our study, the overall rate of relapse of hyperprolactinemia was 43%, which is higher than the rates generally reported in other studies [ 1,2,7,1113,181. This high rate can be due to the longer follow-up. We identified a case of relapse occurring as long as 10 years after surgery. The other studies with shorter follow-up periods may have underestimated the rate of relapse of hyperprolactinemia. Alternatively, different surgical techniques (selective microadenomectomy, as opposed to enlarged adenomectomy [3] or partial hypophysectomy [19]) can also explain different rates of relapsing hyperprolactinemia. Selective microadenomectomy aims at the removal of the adenoma with
343
preservation of normal pituitary tissue and, as a consequence, may possibly achieve only incomplete resection with a higher risk of relapse. However, a more aggressive resection can ensure the complete removal of the adenoma with an increased risk of alteration of other normal pituitary secretions. It should be stressed that most women with relapsing hyperprolactinemia remained asymptomatic. It is possible that some of these patients have a high-molecular-weight, biologically inactive prolactin as described in certain patients with hyperprolactinemia [5]. However, such a phenomenon is rare in prolactinomas. Another possibility is the existence of a critical level of hyperprolactinemia under which patients are asymptomatic. This observation has already been made in patients treated with bromocriptine or other dopaminergic agonists, with some patients achieving ovulatory menstrual cycles even when prolactin was lowered to supranormal values [ 81. This study confirms the prognostic value of immediate postoperative plasma prolactin concentration [15,22]. Women who later had relapse of hyperprolactinemia had an average postoperative plasma prolactin of 10 kg/L or more, whereas women with prolonged remission had an average plasma prolactin of approximately 5 pg/L.
CONCLUSION (1)Surgical resection
of microprolactinoma can achieve 90% normalization of plasma prolactin levels postoperatively. (21 Most patients with late relapse of hyperprolactinemia remain asymptomatic and do not show recurrence of the adenomas, 10 to 20 years following adenomectomy. Transsphenoidal adenomectomy remains a safe (3) and efficient therapeutic approach in women with microprolactinomas and can be recommended as an alternative to medical treatment, or for patients who are intolerant or resistant to dopamine agonists. The authors thank Femanda Janeiro for her secretarial assistance in the preparation of the manuscript.
REFERENCES 1. Buchfelder M, Fahlbusch R, Schotl W, Honegger J. Long-term follow-up results in hormonally active pituitary adenomas after primary successful transsphenoidal surgery. Acta Neurochir (Wien) 1991; 53(suppl):72-6.
344
Massoud et al
Surg Neurol 1996;45:341-6
2. Ciccarelli E, Ghigo E, Miola C, Gandini G, Muller EE, Camanni F. Long-term follow-up of “cured” prolactinoma patients after successful adenomectomy. Clin Endocrinol 1990;32:583-92. 3. Grisoli F, Brue T, Graziani N, Costa R, Trouillas J, Begou D, Jaquet P. Enlarged adenomectomy for enclosed prolactinomas: a preliminary study of 26 cases. Acta Neurochir (Wien) 1990;103:92-8. 4. Hardy J, Beauregard H, Robert F. Prolactin-secreting pituitary adenomas: trassphenoidal microsurgical treatment. In: Robyn C, Harter M, eds. Progress in prolactin: physiology and pathology. Proceedings of the International Symposium on Prolactin, Nice, October 20-23, 1977. North Holland: Elsevier, 1978:36170. 5. Jackson RD, Wortsman J, Malarkey WB. Characterisation of a large molecular weight prolactin in women with idiopathic hyperprolactinemia and normal menses. J Clin Endocrinol Metab 1985;61:258-63. 6. Klibanski A, Neer RM, Beitins IZ, Ridgway EC, Zervas NT, McArthur JW. Decreased bone density in hyperprolactinemic women. N Engl J Med 1980;303:1511-4. 7. Maira G, Anile C, De Marinis L, Barbarino A. Prolactinsecreting adenomas: surgical results and long-term follow-up. Neurosurgery 1989;24:736-43. Endocri8. Molitch ME. Pathologic hyperprolactinemia. no1 Metab Clin North Am 1992;21:877-901. 9. Post KD, Biller BJ, Adelman LS, Molitch ME, Wolpert SM, Reichlin S. Selective transsphenoidal adenomectomy in women with galactorrhea-amenorrhea. JAMA 1979;242:158-62. 10. Randall RV, Laws ER, Abboud CF, Ebersold MJ, Kao PC, Scheithauer BW. Transsphenoidal microsurgical treatment of prolactin-producing pituitary adenomas: results in 100 patients. Mayo Clin Proc 1983;58:10821. 11. Rodman EF, Molitch ME, Post KD, Biller BJ, Reichlin S. Long-term follow-up of transsphenoidal selective adenomectomy for prolactinoma. JAMA 1984;252:921-4. 12. Scamoni C, Balzarini C, Crivelli G, Dorizzi A. Treatment and long-term follow-up results of prolactin secreting pituitary adenomas. J Neurosurg Sci 1991;35: 9-16. 13. Schlechte JA, Sherman BM, Chapler FK, Van Gilder J. Long-term follow-up of women with surgically treated prolactin-secreting pituitary tumors. J Clin Endocrino1 Metab 1986;62:1296-1301. 14. Schlechte J, Dolan K, Sherman B, Chapler F, Luciano A. The natural history of untreated hyperprolactinemia: a prospective analysis. J Clin Endocrinol Metab 1989;68:412-8. 15. Serri 0, Rasio E, Beauregard H, Hardy J, Somma M. Recurrence of hyperprolactinemia after selective transsphenoidal adenomectomy in women with prolactinoma. N Engl J Med 1983;309:280-3. 16. Thomson JA, Teasdale GM, Gordon D, McCruden D, Davies DL. Treatment of presumed prolactinoma by transsphenoidal operation: early and late results. Br Med J 1985;291:1550-3. 17. Thomson JA, Davies DL, McLaren EH, Teasdale GM. Ten year follow-up of microprolactinoma treated by transsphenoidal surgery. Br Med J 1994;309:1409-10. 18. Tucker HSG, Grubb SR, Wigand JP, Taylon A, Lankford HV, Blackard WG, Becker DP. Galactorrhea-amenor-
19.
20.
21. 22.
rhea syndrome: follow-up of 45 patients after pituitary tumor removal. Ann Intern Med 1981;94:302-7. Webster J, Page MD, Bevan TS, Richards SH, DouglasJones A, Scanlon MF. Low recurrence rate after partial hypophysectomy for prolactinoma: the predictive value of prolactin function tests. Clin Endocrinol (Oxf) 1992;36:35-44. Weiss MH, Teal J, Gott P, Wycoff R, Yadley R, Apuzzo ML, Giannotta FL, Kletzky 0, March C. Natural history of microprolactinomas: six-year follow-up. Neurosurgery 1983;12:180-3. Wilson CB. A decade of pituitary microsurgery: the Herbert Olivecrona Lecture. J Neurosurg 1984;61:814-33. Wilson CB. Role of surgery in the management of pituitary tumors. Neurosurg Clin North Am 199O;l: 139-59.
COMMENTARY This paper, from a distinguished group of investigators, provides valuable information on the outcome of patients who develop postoperative hyperprolactinemia following transsphenoidal surgery for microprolactinomas. Out of 64 women with microprolactinomas who underwent surgery and were followed for a period of 10 to 20 years, a total of 25 (43%) developed postoperative hyperprolactinemia. However, the value of their long-term follow-up is that it showed that in only two patients was this due to recurrent tumor. The message from this study is for clinicians not to assume that all patients who develop hyperprolactinemia following successful transsphenoidal surgery have recurrent tumor, which carries significant therapeutic implications. In my opinion, it is best to follow, rather than to aggressively treat, these patients. This is a valuable paper. I wish the authors had provided more discussion on possible mechanisms of postoperative hyperprolactinemia, which at present remains a mystery. However, a discussion of possible causes would have been of interest to the reader. George T. Tindall, M.D. Atlanta, Georgia The paper by Massoud and colleagues is a welcome extension of the previous study which appeared in the New England Journal of Medicine in 1983, concerning many of the same patients. Despite the deficiencies and the methodology of the retrospective clinical study published in the New England Journal, the results have been widely quoted in the endocrine and neurosurgical literature and have been utilized to argue against surgical therapy for prolactin-secreting microadenomas because of the
Follow-up of Microprolactinomas
alleged 40% “recurrence” rate that Massoud et al found in their study. Ever since the publication of the initial article, George Tindall, Jules Hardy, and others have maintained that recurrent modest hyperprolactinemia after surgical removal of a prolactin-secreting microadenoma did not necessarily mean that the tumor had been left behind. That argument is confirmed by the present study, which shows that on long-term review of 64 women with prolactinsecreting microadenomas surgically removed, only 2 of 25 patients who had recurrent hyperprolactinemia actually developed CT-detectable recurrent tumors during a follow-up period that ranged from 10 to 20 years. These findings are all the more compelling because the authors of this paper report a much higher rate of recurrent hyperprolactinemia, following initial remission after transsphenoidal microsurgery, than others who have presented similar data in patients with more consistent follow-up. The main drawback of the Montreal study is that it does not systematically follow all women surgically treated for prolactin microadenomas, but it has a selected cohort of only 64 of a potential 215 women treated for microprolactinoma. The selection criteria were not mentioned in either paper, and we do not have information regarding the fate of the women who were not included. Recent studies of prolactin-secreting microadenomas treated by transsphenoidal surgery show an initial remission rate in 80%-90% of patients. The incidence of subsequent hyperprolactinemia in studies that more thoroughly report the follow-up of the patients, such as a recent review by John Thompson of Graham Teasdale’s series from Glasgow, reveals an initial normalization rate of 86.5%, a long-term (greater than 10 years) sustained remission in 74%, and a recurrence of hyperprolactine mia in 13% [2]. Our own careful follow-up study of the patients treated at the Mayo Clinic revealed a 24% incidence of recurrence of hyperprolactinemia in a group of patients followed for a mean of 5?4 years [l]. A common feature of these follow-up studies is that despite the incidence of recurrent hyperprolactinemia, the actual diagnosis of the recurrent pituitary tumor is quite uncommon. The current report leaves us with a number of unanswered questions. Massoud et al’s follow-up study includes 64 women, 58 of whom had “normal” prolactin levels following surgery. The fate of the remaining 6 women is not further discussed. Of the 58 women who had initial remissions, their clinical features are not presented and we do not know how many of them had resumption of ovulatory menses or resolution of galactorrhea. Among the 25 women
Surg Neural 1996:45:341-6
345
who had recurrent hyperprolactinemia, it is stated that 10 women had “only hyperprolactinemia,” but the extent of this abnormality and the reproducibility on serial laboratory studies is not mentioned. It certainly is possible that these abnormal values might simply represent normal fluctuations that one sees in prolactin levels. The assessment of whether or not a recurrent tumor developed is based on CT evidence, and obviously it would have been preferable to base this particular determination on high-resolution magnetic resonance imaging scans. Hyperprolactinemia alone may not be the best way to judge the success of treatment for prolactinsecreting microadenomas. The important issues are whether the elevation in prolactin has a pathologic effect interfering with the menstrual cycle, fertility, or producing galactorrhea, and whether the elevations detected and reported are consistent findings that have a significant impact on the quality of life of the patient. The bottom line in this study is one that should be encouraging to most neurosurgeons involved in the surgery of pituitary tumors. It appears that the long-term outlook for patients who require surgical management of prolactinomas is better than we have been led to be lieve, and surgery for these lesions should be viewed not as a failure, but as a useful adjunct to an intelligently monitored program for treatment. Edward
R. Laws, Jr., M.D., FACS Mary Lee Vance, M.D.
Charlottesville, Virginia 1 Laws ER, Ebersold MJ, Piepgras DC, Abboud CF, Randall RM, Scheithauer BW. The role of surgery in the management of prolactinoma in prolactin. In: MacLeod RM, Thorner MO, Scapagnini U, eds. Prolactin, basic and clinical correlates. New York: Springer-Verlag, 1985:849-53. 2. Thomsen JA, Davies DL, McLaren EH, Teasdale GM. Ten-year follow-up of microprolactinoma treated by transsphenoidal surgery. Br Med J 1994;309:1409-10.
AUTHORS REPLY Dr. Tindall suggests that we provide more discussion of the possible mechanisms of postoperative hyperprolactinemia. A relapse of hyperprolactinemia after surgical removal of a microprolactinoma is, in some instances, the consequence of regrowth of tumor remnants. Alternatively, a hypothalamic-pituitary dysfunction mayoccurbecauseofdecreaseddeliveryofprolactininhibiting factors due to a scarring process at the site of the tumor cavity after surgery. However, in other instances, different mechanisms may be more
346
Massoudet al
Sur Neurol 199t ;45:341-6
likely. Functional hyperprolactinemia must be considered independent of the surgical procedure. Hyperprolactinemia may be the consequence of certain drugs, stress, or primary hypothyroidism. Some reports have documented the presence of a large molecular-weight prolactin with reduced bioactivity as the major immunoreactive component of prolactin in women with sustained hyperprolactinemia and norma ovulatory menses. Transient and fluctuating hyperprolactinemia has aiso been reported in women with infe~ility and short-luteal phase and in women with various degrees of the amenorrhea-galactorrhea syndrome, probably as a result of imbalance between low progesterone and unopposed estrogen levels. Drs. laws and Vance question several points of the methodoio~ of our study. The long-term follow-up of all patients with microprolactinomas operated by Dr. Hardy during the 70s was unfortunately impossible because most of these patients were referred from abroad. We therefore studied, without further selection, all of the women who, firstly, had proven microprolactino-
mas and, secondly, were followed for at least 10 years after surgery (22 patients were included in our previous report) [15]. Six women remained hyperprolactinemic and symptomatic following surgery. These patients were treated successfully with bromocriptine. All 58 women described with initial remissions had resumption of regular menstrual cycles and resolution of galactorrhea. Although ovulation was not documented by plasma progesterone levels in many patients, 27 had 40 spontaneous pregnancies. Plasma prolactin concentrations depicted at each time interval (Figures 1 and 2) are, for each woman, the mean of at least two values on different days. In some cases, the mean plasma prolactin concentration was calculated from six samples in a 24-hour period. The magnitude of hyperprolactinemia is shown in these figures. Omar Serri, M.D., Ph.D. Jules Hardy, M.D. congeal,
Quebec, Canada