Macroprolactinemia in women presenting with hyperandrogenic symptoms: Implications for the management of polycystic ovary syndrome
2004; Elsevier BV; Volume: 82; Issue: 6 Linguagem: Inglês
10.1016/j.fertnstert.2004.06.045
ISSN1556-5653
Autores Tópico(s)Ovarian cancer diagnosis and treatment
ResumoEight of 109 consecutive hyperandrogenic patients had increased serum PRL levels; macroprolactinemia was diagnosed in 4 patients by measuring PRL levels after precipitation of serum with polyethylene glycol, and after detecting macroprolactinemia, these 4 patients were diagnosed with polycystic ovary syndrome (PCOS). Therefore, macroprolactinemia must be ruled out in women presenting with hyperandrogenic symptoms and increased serum PRL concentrations to avoid misdiagnosis (hyperprolactinemia is an exclusion criterion for the diagnosis of PCOS), unnecessary diagnostic tests, and inappropriate use of dopaminergic agonists. Eight of 109 consecutive hyperandrogenic patients had increased serum PRL levels; macroprolactinemia was diagnosed in 4 patients by measuring PRL levels after precipitation of serum with polyethylene glycol, and after detecting macroprolactinemia, these 4 patients were diagnosed with polycystic ovary syndrome (PCOS). Therefore, macroprolactinemia must be ruled out in women presenting with hyperandrogenic symptoms and increased serum PRL concentrations to avoid misdiagnosis (hyperprolactinemia is an exclusion criterion for the diagnosis of PCOS), unnecessary diagnostic tests, and inappropriate use of dopaminergic agonists. At present, the diagnosis of the polycystic ovary syndrome (PCOS) requires exclusion of hyperprolactinemia (and of other etiologies such as nonclassic adrenal hyperplasia or androgen-secreting tumors) in women presenting with a combination of clinical or biochemical hyperandrogenism, menstrual dysfunction, and polycystic ovaries on ultrasound examination (1Zawadzki J.K. Dunaif A. Diagnostic criteria for polycystic ovary syndrome: towards a rational approach.in: Dunaif A. Givens J.R. Haseltine F.P. Merriam G.R. Polycystic ovary syndrome. Blackwell Scientific Publications, Boston1992: 377-384Google Scholar, 2Homburg R. What is polycystic ovarian syndrome? A proposal for a consensus on the definition and diagnosis of polycystic ovarian syndrome.Hum Reprod. 2002; 17: 2495-2499Crossref PubMed Scopus (95) Google Scholar, 3Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41–7.Google Scholar). However, hyperprolactinemia is not uncommon in women presenting with hyperandrogenic symptoms and menstrual disturbances (4Carmina E. Rosato F. Maggiore M. Gagliano A.M. Indovina D. Janni A. Prolactin secretion in polycystic ovary syndrome (PCO): correlation with the steroid pattern.Acta Endocrinol (Copenh). 1984; 105: 99-104PubMed Google Scholar, 5Luciano A.A. Chapler F.K. Sherman B.M. Hyperprolactinemia in polycystic ovary syndrome.Fertil Steril. 1984; 41: 719-725Abstract Full Text PDF PubMed Google Scholar, 6Milewicz A. Prolactin levels in the polycystic ovary syndrome.J Reprod Med. 1984; 29: 193-196PubMed Google Scholar), and although it is not clear how increased PRL levels might influence androgen excess, the diagnosis of PCOS cannot be sustained in these women. Furthermore, the presence of hyperprolactinemia in hyperandrogenic patients may lead to expensive diagnostic and therapeutic strategies, including pituitary imaging techniques and use of dopaminergic agonists. Macroprolactinemia occurs when the predominant form of PRL in serum is a 150- to 170-kDa complex (macroprolactin, or big big PRL), usually composed of PRL and an IgG autoantibody (7De Schepper J. Schiettecatte J. Velkeniers B. Blumenfeld Z. Shteinberg M. Devroey P. et al.Clinical and biological characterization of macroprolactinemia with and without prolactin–IgG complexes.Eur J Endocrinol. 2003; 149: 201-207Crossref PubMed Scopus (60) Google Scholar). Although macroprolactin exhibits limited bioactivity in vivo, it retains immunoreactivity. The clearance of macroprolactin is reduced, resulting in increased serum PRL values when measured with most immunoassays (8Smith T.P. Suliman A.M. Fahie-Wilson M.N. McKenna T.J. Gross variability in the detection of prolactin in sera containing big big prolactin (macroprolactin) by commercial immunoassays.J Clin Endocrinol Metab. 2002; 87: 5410-5415Crossref PubMed Scopus (160) Google Scholar). Cumulative experience indicates that macroprolactin accounts for as many as 25% of unselected hyperprolactinemic sera (8Smith T.P. Suliman A.M. Fahie-Wilson M.N. McKenna T.J. Gross variability in the detection of prolactin in sera containing big big prolactin (macroprolactin) by commercial immunoassays.J Clin Endocrinol Metab. 2002; 87: 5410-5415Crossref PubMed Scopus (160) Google Scholar). Suliman et al. (9Suliman A.M. Smith T.P. Gibney J. McKenna T.J. Frequent misdiagnosis and mismanagement of hyperprolactinemic patients before the introduction of macroprolactin screening: application of a new strict laboratory definition of macroprolactinemia.Clin Chem. 2003; 49: 1504-1509Crossref PubMed Scopus (162) Google Scholar) recently showed that macroprolactinemia is a significant cause of misdiagnosis, unnecessary investigations, and inappropriate treatment. In their series of subjects retrospectively identified as having macroprolactinemia, 93% underwent pituitary imaging and 87% had received treatment with dopaminergic agonists. Given that hyperprolactinemia is an exclusion criterion for the diagnosis of PCOS, the presence of macroprolactinemia might interfere with the correct diagnosis and management of hyperandrogenic patients. The increased serum PRL levels would incorrectly exclude PCOS. Symptoms, such as menstrual disturbances, may be attributed to hyperprolactinemia, leading to inadequate diagnostic and therapeutic strategies. I have conducted a retrospective evaluation of the diagnostic and therapeutic consequences of macroprolactinemia in a series of 109 consecutive hyperandrogenic patients, of whom 8 presented with increased serum PRL levels above the upper limit of normal (24 ng/mL) on at least two different occasions. A clinical history was taken and a complete physical examination was performed, including the presence or absence of galactorrhea in the women presenting with hyperprolactinemia. None of these patients was taking any drug at the initial visit that might have increased serum PRL levels, including oral contraceptives (OC). Hirsutism was defined by a modified Ferriman-Gallwey score >7 (10Hatch R. Rosenfield R.L. Kim M.H. Tredway D. Hirsutism: implications, etiology, and management.Am J Obstet Gynecol. 1981; 140: 815-830Abstract Full Text PDF PubMed Scopus (824) Google Scholar). Evidence for oligo-ovulation was provided by chronic oligomenorrhea, by luteal phase P <4 ng/mL, or by basal body temperature (BBT) charts. Hormonal studies were performed between days 5 and 10 of the menstrual cycle. To avoid a stress-related increase in serum PRL levels at venipuncture, serum samples were obtained 15–30 minutes after placing an indwelling intravenous line in a forearm vein, between 8 and 9 AM and after a 12-hour overnight fast. The Hospital Ethics Committee approved the study, and informed consent was obtained from every patient. Serum was assayed for serum PRL levels using an automated immunochemiluminescence assay (Immulite, Diagnostic Products Corporation, Los Angeles, CA), with 6.2% and 8.5% intra-assay and interassay coefficients of variation. In addition, a complete hormone profile, which included total and calculated free T (11Vermeulen A. Verdonck L. Kaufman J.M. A critical evaluation of simple methods for the estimation of free testosterone in serum.J Clin Endocrinol Metab. 1999; 84: 3666-3672Crossref PubMed Google Scholar) and DHEAS concentrations among other determinations, was obtained as described elsewhere (12Escobar-Morreale H.F. Lasuncion M.A. Sancho J. Treatment of hirsutism with ethinyl estradiol-desogestrel contraceptive pills has beneficial effects on the lipid profile and improves insulin sensitivity.Fertil Steril. 2000; 74: 816-819Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar). In the 8 women with increased serum PRL levels, serum was also assayed after precipitation with polyethylene glycol (PEG) as previously described by Smith et al. (8Smith T.P. Suliman A.M. Fahie-Wilson M.N. McKenna T.J. Gross variability in the detection of prolactin in sera containing big big prolactin (macroprolactin) by commercial immunoassays.J Clin Endocrinol Metab. 2002; 87: 5410-5415Crossref PubMed Scopus (160) Google Scholar). In brief, 200 μL of serum was mixed with 200 μL of a 25% (wt/vol) solution of 6,000 kDa PEG (Merck-Schuchardt, Art. 807491, Hohenbrunn, Germany) in phosphate buffered saline (PBS) at pH 7.4. Samples were mixed for 1 minute in a vortex mixer and incubated at room temperature for 10 minutes and immediately centrifuged for 30 minutes at 1,800 × g. The supernatant was separated and assayed for PRL as described. The result was multiplied by 2 to compensate for the 1:1 dilution during the first step. An untreated aliquot of each serum was run within the same assay. A decrease ≥40% in the serum PRL level after precipitation with PEG was considered indicative of macroprolactinemia. The clinical and biochemical characteristics of the eight hyperandrogenic patients presenting with hyperprolactinemia, and their diagnostic and therapeutic management before and after ruling out macroprolactinemia, are summarized in Table 1.TABLE 1Clinical and biochemical characteristics of the eight hyperandrogenic women presenting with increased serum PRL concentrations, and diagnostic and therapeutic strategies applied before and after testing for macroprolactinemia.Patient No.12345678Age (yr)3123222918182925BMI (kg/m2)59.823.529.336.722.523.123.417.7Hirsutism score8956191369AcneNoNoYesNoYesNoNoNoAndrogenic alopeciaNoNoNoYesNoNoNoNoOligo-ovulationYesYesYesYesNoYesYesYesGalactorrheaNoNoNoNoNoNoNoNoTotal T (ng/dL)84.551.136.659.466.173.581.081.2Free T (ng/dL)1.60.90.71.51.01.11.31.1DHEAS (ng/mL)34202490162022502660229067063160Ultrasound scanNormalNormalPCOPCONPNormalPCONormalPRL (ng/mL)49.790.537.539.841.629.351.433.8PRL–PEG (ng/mL)11.511.421.415.538.628.646.027.4Pituitary MRI scanNormalNormalNormalMicroadenomaNormalNPNormalNPInitial diagnosisH/IHPH/IHPAcne/IHPProlactinomaH/IHPH/IHPAH/IHPNCCAH/IHPFertility desiredNoNoNoYesNoNoNoNoInitial treatmentBiliopancreatic diversionCAB + OCPOCPBCP + MetforminOCPOCPCAB + DXMDXMFinal diagnosisPCOS/MPPCOS/MPPCOS/MPPCOS/MPH/IHPH/IHPAH/IHPNCCAH/IHPPresent treatmentSustained weight-lossOCPOCPMetforminOCPOCPDXMDXMMentrual cycle after treatmentRegularRegularaTreated with OCP.RegularaTreated with OCP.RegularRegularaTreated with OCP.RegularaTreated with OCP.RegularRegularThe upper limits of the normal range were as follows: total T, 62 ng/dL; free T, 1 ng/dL; DHEAS, 3,500 ng/mL; PRL, 24 ng/mL. AH = adrenal hyperandrogenism; BCP = bromocriptine; CAB = cabergoline; DXM = low-dose dexamethasone; H = hirsutism; IHP = idiopathic hyperprolactinemia; MP = macroprolactinemia; NCCAH = nonclassic congenital adrenal hyperplasia; NP = not performed; OCP = oral contraceptive pill; PCO = ultrasonographic polycystic ovaries.a Treated with OCP. Open table in a new tab The upper limits of the normal range were as follows: total T, 62 ng/dL; free T, 1 ng/dL; DHEAS, 3,500 ng/mL; PRL, 24 ng/mL. AH = adrenal hyperandrogenism; BCP = bromocriptine; CAB = cabergoline; DXM = low-dose dexamethasone; H = hirsutism; IHP = idiopathic hyperprolactinemia; MP = macroprolactinemia; NCCAH = nonclassic congenital adrenal hyperplasia; NP = not performed; OCP = oral contraceptive pill; PCO = ultrasonographic polycystic ovaries. In four of these women, macroprolactinemia was the actual cause of the increased serum PRL concentrations detected initially, considering that PRL levels decreased ≥40% after precipitation of serum with PEG, to values within the normal range. Before macroprolactinemia was diagnosed, pituitary magnetic resonance imaging (MRI) scans were performed. The pituitary gland was normal in three of these women, who were diagnosed initially as having idiopathic hyperprolactinemia and hirsutism or acne. The fourth patient had a 4-mm image suggesting the presence of a pituitary microadenoma. She was diagnosed with microprolactinoma and treated with bromocriptine and metformin. The latter was used because she was markedly insulin resistant and was having difficulties in conceiving. Of the remaining three macroprolactinemic patients, one patient had extreme obesity and was treated with biliopancreatic diversion. After pronounced weight loss (27% of her initial weight), ovulation was restored and her serum androgen concentrations were within the normal range, yet serum PRL levels remained increased. Another patient was treated with cabergoline and OC, and the other patient received OC alone, with close monitoring of the serum PRL concentrations, which remained mostly unchanged. After macroprolactinemia was detected, these four women were diagnosed with PCOS, and treatment with dopaminergic agonists was stopped in the two patients previously treated with these drugs. At present, two of the patients are ovulating regularly (the patient who underwent biliopancreatic diversion and the patient treated with metformin), and two patients have regular menstrual cycles while treated with OC. Four hyperandrogenic women presented with increased serum PRL values, but did not have macroprolactinemia. In two of these patients pituitary MRI scans depicted normal pituitary glands. In the remaining two patients, pituitary MRI scans were not performed, given that the increase in their serum PRL concentrations were very mild and did not increase after treatment with OC. These women were diagnosed with idiopathic hyperprolactinemia, associated with hirsutism in two patients, with adrenal hyperandrogenism in one woman (because of lack of suppression of serum androgens during triptorelin-induced gonadal suppression), and with nonclassic congenital adrenal hyperplasia in the last patient (because of ACTH-stimulated 17-hydroxyprogesterone levels above 10 ng/mL). At present, the patients with hirsutism and true idiopathic hyperprolactinemia are being treated with OC and their serum PRL levels remain stable, but slightly above the normal range. The patients with idiopathic hyperprolactinemia and adrenal hyperandrogenism or nonclassic congenital adrenal hyperplasia are under low-dose dexamethasone treatment, and have regular menstrual cycles despite mildly increased serum PRL levels. The association of hyperandrogenism and hyperprolactinemia has been recognized for decades, yet the mechanisms underlying this association remain largely unknown. Among other hypotheses, the increase in circulating estrone (E1) frequently found in hyperandrogenic women might stimulate PRL secretion (4Carmina E. Rosato F. Maggiore M. Gagliano A.M. Indovina D. Janni A. Prolactin secretion in polycystic ovary syndrome (PCO): correlation with the steroid pattern.Acta Endocrinol (Copenh). 1984; 105: 99-104PubMed Google Scholar), yet hyperprolactinemia might also result from an altered opioid and dopaminergic tone at the hypothalamus (5Luciano A.A. Chapler F.K. Sherman B.M. Hyperprolactinemia in polycystic ovary syndrome.Fertil Steril. 1984; 41: 719-725Abstract Full Text PDF PubMed Google Scholar). More recent studies suggest that hyperandrogenism and hyperprolactinemia have independent origins in these patients, and that older reports possibly overestimated the prevalence of hyperprolactinemia in hyperandrogenic women (13Bracero N. Zacur H.A. Polycystic ovary syndrome and hyperprolactinemia.Obstet Gynecol Clin North Am. 2001; 28: 77-84Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). However, the management of hyperandrogenic patients presenting with hyperprolactinemia is especially difficult, because there are no precise guidelines for the diagnosis and treatment. In some of these patients, the clinical and biochemical picture resembles that of PCOS, but current criteria for the diagnosis of PCOS require exclusion of hyperprolactinemia (1Zawadzki J.K. Dunaif A. Diagnostic criteria for polycystic ovary syndrome: towards a rational approach.in: Dunaif A. Givens J.R. Haseltine F.P. Merriam G.R. Polycystic ovary syndrome. Blackwell Scientific Publications, Boston1992: 377-384Google Scholar, 2Homburg R. What is polycystic ovarian syndrome? A proposal for a consensus on the definition and diagnosis of polycystic ovarian syndrome.Hum Reprod. 2002; 17: 2495-2499Crossref PubMed Scopus (95) Google Scholar, 3Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41–7.Google Scholar). I suspected that some of the hyperandrogenic hyperprolactinemic patients in my series actually had PCOS, and that their increased serum PRL levels were caused by macroprolactinemia. The results presented here not only confirm this suspicion, but also illustrate some of the consequences that this misdiagnosis had for the management of these women. Thus, for all these women I had requested pituitary MRI scans, which were actually unnecessary. I had treated two of these patients with dopaminergic agonists, including the one patient whom I diagnosed with microprolactinoma because of a pituitary image suggestive of microadenoma and increased serum PRL levels caused by macroprolactinemia. Moreover, had this woman conceived, I would have ordered serial campimetric evaluations during pregnancy, evaluations that would also have been unnecessary. Fortunately, these practices were not associated with any undesirable side effects in these women. In conclusion, it appears important to rule out macroprolactinemia in women presenting with hyperandrogenic symptoms and increased serum PRL concentrations to avoid misdiagnosis, unnecessary diagnostic tests, and inappropriate use of dopaminergic agonists. The author thanks Dr. José I. Botella-Carretero, M.D., Genoveva González, and M. Paz Muñoz for their technical help.
Referência(s)