Clinical Endocrinology (1992) 36,545-551
Adrenocortical dysfunction in paracoccidioidomycosis: comparison between plasma b-lipotrophinl adrenocorticotrophin levels and adrenocortical tests Ayrton C. Yorelra, Roberto Martinez, Margaret Castro and Luclla L. K. Ellas Division of Endocrinology and Infectious Diseases, Department of Medicine, Faculty of Medicine, 14049-RibeirBo Preto, SP,Brazil (Received 21 August 1991; returned for revision 9 October 1991; finally revised 26 November 1991; accepted 17 December 1991)
Summary OBJECTIVE Paracoccldloldomycosls is an important cause of Addlson’s disease In South America. We have carrled out an extensive and prospective study on paracoccidioldomycosls comparlng glucocortlcold, mineralocortlcoid and androgen function with adrenal regulators, ACTHIP-LPH and plasma renln activity (PRA). PATIENTS AND METHODS Forty-sevenmale patlents with active paracoccidioidomycosis were studied consecutively together with 20 healthy controls. On day 1, plasma aldosterone and PRA levels were measured In blood samples obtained from patients in the supine and erect posltlon. On day 2 at 0900 h, baseline plasma samples were taken for ACTH, 8-lipotrophin (/?-LPH), cortlsol, corticosterone, aldosterone, androstenedlone (&-A) and dehydroepiandrosteronesulphate (DHEA-S). ACTH 1-24 (250pg) was glven i.v. and blood samples for these steroid assays were taken at 1 and 2 hours. RESULTS Five patients (10%) had Addlson’s disease with hlgh basal plasma ACTH and p-LPH, and low cortisol levels after the ACTH test. In the remaining 42 patients, baseline ACTH and fl-LPH levels and plasma cortisol after ACTH were within the normal range. A high percentageof patients presented with reduced cortlcosterone (21YO of patients) and aldosterone (23%) secretion and increased PRA (31Yo).Plasma A,-A (19Yo)and DHEA-S (50%) levels were also reduced. CONCLUSIONS The frequency of Addison’s disease among our patients with paracoccidloidomycosls was 10%. In addltlon, a subset of patients presented wlth adrenal dysfunction detected by mineralocortlcold or androgen tests. In parallel to pathologlcal lesions a
Correspondence: Dr Ayrton C. Moreira, Department of Medicine, Faculty of Medicine 14049-RibeirZoPreto, SP,Brazil.
functional adaptation may occur durlng adrenal Involvement in paracaccldloldomycosis.
Paracoccidioidomycosis (South American blastomycosis) is the most common systemic mycosis in Latin America, mainly occurring in Brazil, Colombia and Venezuela (Londero, 1982). From its natural habitat, Paracoccidioides brasiliensis penetrates the host, usually through the lungs. Once established, the mycosis may progress rapidly (acute form) or slowly (chronic form) by lympho-haematogenic dissemination. The acute form is characterized by systemic lymphadenopathy, hepatosplenomegaly, bone marrow dysfunction, and great impairment of patient’s general condition, especially in young persons. The chronic form presents progressive pulmonary or extrapulmonary (mucocutaneous, lymph nodes, adrenal glands, intestine, bone, etc) manifestations. The patients are generally adult males and maintain a fair general condition (Franco et al., 1987). Infection with P. brasiliensisinvolving the adrenal glands has been recorded at rates between 44 and 80% of cases in autopsy series (Franco & Montenegro, 1982). Addison’s disease was first diagnosed in patients with paracoccidioidomycosis 30 years ago (Del Negro et al., 1961; Marsiglia & Pinto, 1966). In the last decade, four adrenal function studies in paracoccidioidomycosis have been reported (Del Negro ef al., 1980; Abad ef al., 1986; Tendrich et al., 1987; Colombo, 1989). These studies were usually focused on cortisol responses to synthetic ACTH administration and established graded severities of adrenal dysfunction in this disease ranging from Addison’s disease to limited cortisol production. The aldosterone (Aldo) response to ACTH has been described in one study and plasma dehydroepiandrosterone sulphate (DHEA-S) or ACTH levels were reported for a small number of patients. Recently studies in patients with autoimmune adrenalitis, the most common cause of Addison’s disease in developed countries, have shown elevated concentrations of plasma ACTH and/or plasma renin activity (PRA) as the most sensitive indices of adrenal failure (Leisti et al., 1983; Ketchum et al., 1984; Betterle et al., 1988). However, there are no detailed or extensive studies on plasma p-LPH and ACTH, on the PRA-aldosterone axis, and on adrenal androgens, compared with glucocorticoid secretion in para-
548
Clinical Endocrinology (1992) 38
A. C. Moreira et al.
coccidioidomycosis. On this basis we examined these adrenal function tests in a prospective study on patients with active paracoccidioidomycosis.
SP) was administered i.v. over 60 seconds at 0930 h. Blood samples for cortisol and aldosterone measurements were obtained 1 and 2 hours after the injection. Plasma corticosterone and &-A were measured in 1hour samples.
Patients and methods Subjects
Hormone assays
Forty-seven male patients (mean age 38.9 years, range 16-60 years, SD 12.8 years) with active paracoccidioidomycosis gave informed consent to take part in this study. The patients arrived consecutively at the Infectious Diseases Division of the University Hospital (Ribeiriio Preto, SP, Brazil) from April 1986 to August 1989. The clinical diagnosis of active paracoccidioidomycosiswas based on the presence of mucocutaneous lesions (74.5% of cases), lung disease (60%) or lymph node involvement (49%). The involvement of three or more organs was detected in 34% of cases. Fourteen of these patients had the acute form and 33 patients had the chronic form of paracoccidioidomycosis. The diagnosis was reaffirmed on the basis of serological test and microscopical observation and/or culture isolation of P . brasiliensis from sputum, mucocutaneous ulcer scrapings or biopsy material . (Restrepo et al., 1976). Twenty healthy male volunteers (mean age 36 years; range 20-53 years, SD 8-8 years) were also studied as a control group after giving informed consent.
Plasma ACTH and B-LPH levels were determined by RIA after extraction from plasma using silicic acid (Moreira et al., 1987; Castro 8c Moreira, 1990). The assay sensitivity was 2 pmol/l for ACTH and 5 pmol/l for @-LPH.The intra and inter-assay coefficients of variation (CVs) were 4.7 and 18.1% for ACTH and 6.8 and 20.9% for @-LPH,respectively. PRA was determined by RIA of generated angiotensin-I using commercial kits (Baxter-Travenol, Cambridge, MA). The assay sensitivity was 0.03 ng/l s. The details of plasma cortisol, corticosterone, aldosterone, DHEA-S and A4-A measurements have been published previously (Vieira et al., 1981; Moreira etal., 1990). All steroidassays hadintraassay CVs between 3 and 4% and interassay CVs between 7 and 16%. The assay sensitivity was 40 nmol/l for cortisol, 8 nmol/l for corticosterone, 86 pmol/l for aldosterone, 0.2 pmol/l for DHEA-S, and 0.3 nmol/l for &-A. All samples obtained during each test were analysed in duplicate in the same assay. Statistical analysis
Protocol
All studies were performed on the Internal Medicine ward of the University Hospital (Ribeiriio Preto, SP, Brazil) under fixed dietary conditions (sodium, 200 mmol/day; potassium, 63 mmol/day). The patients were not taking drugs, including ketoconazole, which would interfere with adrenal function at the time of the study. All tests were started after an overnight fast. An indwelling catheter was placed into a forearm vein 1 hour before each test. The study was carried out in this order: Postural stimulation After overnight recumbency, a baseline blood sample for aldosterone and PRA was obtained at 0800 h, and a second sample was obtained after the subject had been in an upright position for 2 hours. Baseline samples and ACTH test (a) Baseline samples. Three blood samples taken at 15minute intervals were obtained at 0900,091 5 and 0930 h for plasma ACTH, B-LPH and cortisol assays and at 0930 h for plasma corticosterone (B), aldosterone, DHEA-S and androstenedione (&-A) assays. (b) ACTH (0.25 mg tetracosactrin; Organon, SHo Paulo,
Baseline plasma ACTH, B-LPH and cortisol levels were calculated as the mean of the 0900, 0915 and 0930 h concentrations. In addition to the meanf SEM, the 5th and 95th percentiles (Ps-Pss) are also reported. A normal response of each steroid after ACTH administration was defined by the peak steroid value (May & Carey, 1985). Statistical analysis of the results was performed using the Wilcoxon rank sum test, Spearman's rank correlation procedure and two-way analysis of variance when appropriate. Results Evaluation of glucocorticoid function
Baseline plasma cortisol, ACTH and B-LPH, and clinical features of Addison'sdisease.Figure 1 shows the basal plasma cortisol, ACTH and p-LPH levels obtained in 47 patient with paracoccidioidomycosis compared with control percentiles. Four patients (cases 4, 6, 18,43) had low or non-detectable cortisol values and high ACTH and B-LPH levels. These four patients had Addisonian pigmentation of buccal mucosa and/or skin. One patient (case 20) had a moderately high ACTH level (16 pmol/l) associated with an upper normal B-
547
Hypoadrenalism in paracoccidioidomycosis
Clinical Endocrinology (1992) 38
ti
‘I
x)o 200 I500
25
1250
O
I000 c
A
‘1
E .-2
L
T50
c 0
0
500
Plasma cortisol before and 1 and 2 hours after the ACTHI-24test and baseline plasma ACTH and I]-LPH in 47 patients with paracoccidioidomycosis. Cases 4 ( x ), 6 (A), 18 (O),20 (v)and 43 (v) had Addison’s disease while the remaining patients are represented by solid circles. The dashed lines or the hatched area limit the 5th and 95th percentiles for normal subjects. Fig. 1
250
0
LPH value (20 pmol/l), in contrast to borderline lower normal cortisol levels (190 nmol/l), and no pigmentary changes. In these five patients, there was a positive Spearman’s rank correlation between plasma ACTH and P-LPH levels in samples collected at the same time (r =0.57, n = 15, P
Adrenocortical dysfunction in paracoccidioidomycosis: comparison between plasma b-lipotrophinl adrenocorticotrophin levels and adrenocortical tests Ayrton C. Yorelra, Roberto Martinez, Margaret Castro and Luclla L. K. Ellas Division of Endocrinology and Infectious Diseases, Department of Medicine, Faculty of Medicine, 14049-RibeirBo Preto, SP,Brazil (Received 21 August 1991; returned for revision 9 October 1991; finally revised 26 November 1991; accepted 17 December 1991)
Summary OBJECTIVE Paracoccldloldomycosls is an important cause of Addlson’s disease In South America. We have carrled out an extensive and prospective study on paracoccidioldomycosls comparlng glucocortlcold, mineralocortlcoid and androgen function with adrenal regulators, ACTHIP-LPH and plasma renln activity (PRA). PATIENTS AND METHODS Forty-sevenmale patlents with active paracoccidioidomycosis were studied consecutively together with 20 healthy controls. On day 1, plasma aldosterone and PRA levels were measured In blood samples obtained from patients in the supine and erect posltlon. On day 2 at 0900 h, baseline plasma samples were taken for ACTH, 8-lipotrophin (/?-LPH), cortlsol, corticosterone, aldosterone, androstenedlone (&-A) and dehydroepiandrosteronesulphate (DHEA-S). ACTH 1-24 (250pg) was glven i.v. and blood samples for these steroid assays were taken at 1 and 2 hours. RESULTS Five patients (10%) had Addlson’s disease with hlgh basal plasma ACTH and p-LPH, and low cortisol levels after the ACTH test. In the remaining 42 patients, baseline ACTH and fl-LPH levels and plasma cortisol after ACTH were within the normal range. A high percentageof patients presented with reduced cortlcosterone (21YO of patients) and aldosterone (23%) secretion and increased PRA (31Yo).Plasma A,-A (19Yo)and DHEA-S (50%) levels were also reduced. CONCLUSIONS The frequency of Addison’s disease among our patients with paracoccidloidomycosls was 10%. In addltlon, a subset of patients presented wlth adrenal dysfunction detected by mineralocortlcold or androgen tests. In parallel to pathologlcal lesions a
Correspondence: Dr Ayrton C. Moreira, Department of Medicine, Faculty of Medicine 14049-RibeirZoPreto, SP,Brazil.
functional adaptation may occur durlng adrenal Involvement in paracaccldloldomycosis.
Paracoccidioidomycosis (South American blastomycosis) is the most common systemic mycosis in Latin America, mainly occurring in Brazil, Colombia and Venezuela (Londero, 1982). From its natural habitat, Paracoccidioides brasiliensis penetrates the host, usually through the lungs. Once established, the mycosis may progress rapidly (acute form) or slowly (chronic form) by lympho-haematogenic dissemination. The acute form is characterized by systemic lymphadenopathy, hepatosplenomegaly, bone marrow dysfunction, and great impairment of patient’s general condition, especially in young persons. The chronic form presents progressive pulmonary or extrapulmonary (mucocutaneous, lymph nodes, adrenal glands, intestine, bone, etc) manifestations. The patients are generally adult males and maintain a fair general condition (Franco et al., 1987). Infection with P. brasiliensisinvolving the adrenal glands has been recorded at rates between 44 and 80% of cases in autopsy series (Franco & Montenegro, 1982). Addison’s disease was first diagnosed in patients with paracoccidioidomycosis 30 years ago (Del Negro et al., 1961; Marsiglia & Pinto, 1966). In the last decade, four adrenal function studies in paracoccidioidomycosis have been reported (Del Negro ef al., 1980; Abad ef al., 1986; Tendrich et al., 1987; Colombo, 1989). These studies were usually focused on cortisol responses to synthetic ACTH administration and established graded severities of adrenal dysfunction in this disease ranging from Addison’s disease to limited cortisol production. The aldosterone (Aldo) response to ACTH has been described in one study and plasma dehydroepiandrosterone sulphate (DHEA-S) or ACTH levels were reported for a small number of patients. Recently studies in patients with autoimmune adrenalitis, the most common cause of Addison’s disease in developed countries, have shown elevated concentrations of plasma ACTH and/or plasma renin activity (PRA) as the most sensitive indices of adrenal failure (Leisti et al., 1983; Ketchum et al., 1984; Betterle et al., 1988). However, there are no detailed or extensive studies on plasma p-LPH and ACTH, on the PRA-aldosterone axis, and on adrenal androgens, compared with glucocorticoid secretion in para-
548
Clinical Endocrinology (1992) 38
A. C. Moreira et al.
coccidioidomycosis. On this basis we examined these adrenal function tests in a prospective study on patients with active paracoccidioidomycosis.
SP) was administered i.v. over 60 seconds at 0930 h. Blood samples for cortisol and aldosterone measurements were obtained 1 and 2 hours after the injection. Plasma corticosterone and &-A were measured in 1hour samples.
Patients and methods Subjects
Hormone assays
Forty-seven male patients (mean age 38.9 years, range 16-60 years, SD 12.8 years) with active paracoccidioidomycosis gave informed consent to take part in this study. The patients arrived consecutively at the Infectious Diseases Division of the University Hospital (Ribeiriio Preto, SP, Brazil) from April 1986 to August 1989. The clinical diagnosis of active paracoccidioidomycosiswas based on the presence of mucocutaneous lesions (74.5% of cases), lung disease (60%) or lymph node involvement (49%). The involvement of three or more organs was detected in 34% of cases. Fourteen of these patients had the acute form and 33 patients had the chronic form of paracoccidioidomycosis. The diagnosis was reaffirmed on the basis of serological test and microscopical observation and/or culture isolation of P . brasiliensis from sputum, mucocutaneous ulcer scrapings or biopsy material . (Restrepo et al., 1976). Twenty healthy male volunteers (mean age 36 years; range 20-53 years, SD 8-8 years) were also studied as a control group after giving informed consent.
Plasma ACTH and B-LPH levels were determined by RIA after extraction from plasma using silicic acid (Moreira et al., 1987; Castro 8c Moreira, 1990). The assay sensitivity was 2 pmol/l for ACTH and 5 pmol/l for @-LPH.The intra and inter-assay coefficients of variation (CVs) were 4.7 and 18.1% for ACTH and 6.8 and 20.9% for @-LPH,respectively. PRA was determined by RIA of generated angiotensin-I using commercial kits (Baxter-Travenol, Cambridge, MA). The assay sensitivity was 0.03 ng/l s. The details of plasma cortisol, corticosterone, aldosterone, DHEA-S and A4-A measurements have been published previously (Vieira et al., 1981; Moreira etal., 1990). All steroidassays hadintraassay CVs between 3 and 4% and interassay CVs between 7 and 16%. The assay sensitivity was 40 nmol/l for cortisol, 8 nmol/l for corticosterone, 86 pmol/l for aldosterone, 0.2 pmol/l for DHEA-S, and 0.3 nmol/l for &-A. All samples obtained during each test were analysed in duplicate in the same assay. Statistical analysis
Protocol
All studies were performed on the Internal Medicine ward of the University Hospital (Ribeiriio Preto, SP, Brazil) under fixed dietary conditions (sodium, 200 mmol/day; potassium, 63 mmol/day). The patients were not taking drugs, including ketoconazole, which would interfere with adrenal function at the time of the study. All tests were started after an overnight fast. An indwelling catheter was placed into a forearm vein 1 hour before each test. The study was carried out in this order: Postural stimulation After overnight recumbency, a baseline blood sample for aldosterone and PRA was obtained at 0800 h, and a second sample was obtained after the subject had been in an upright position for 2 hours. Baseline samples and ACTH test (a) Baseline samples. Three blood samples taken at 15minute intervals were obtained at 0900,091 5 and 0930 h for plasma ACTH, B-LPH and cortisol assays and at 0930 h for plasma corticosterone (B), aldosterone, DHEA-S and androstenedione (&-A) assays. (b) ACTH (0.25 mg tetracosactrin; Organon, SHo Paulo,
Baseline plasma ACTH, B-LPH and cortisol levels were calculated as the mean of the 0900, 0915 and 0930 h concentrations. In addition to the meanf SEM, the 5th and 95th percentiles (Ps-Pss) are also reported. A normal response of each steroid after ACTH administration was defined by the peak steroid value (May & Carey, 1985). Statistical analysis of the results was performed using the Wilcoxon rank sum test, Spearman's rank correlation procedure and two-way analysis of variance when appropriate. Results Evaluation of glucocorticoid function
Baseline plasma cortisol, ACTH and B-LPH, and clinical features of Addison'sdisease.Figure 1 shows the basal plasma cortisol, ACTH and p-LPH levels obtained in 47 patient with paracoccidioidomycosis compared with control percentiles. Four patients (cases 4, 6, 18,43) had low or non-detectable cortisol values and high ACTH and B-LPH levels. These four patients had Addisonian pigmentation of buccal mucosa and/or skin. One patient (case 20) had a moderately high ACTH level (16 pmol/l) associated with an upper normal B-
547
Hypoadrenalism in paracoccidioidomycosis
Clinical Endocrinology (1992) 38
ti
‘I
x)o 200 I500
25
1250
O
I000 c
A
‘1
E .-2
L
T50
c 0
0
500
Plasma cortisol before and 1 and 2 hours after the ACTHI-24test and baseline plasma ACTH and I]-LPH in 47 patients with paracoccidioidomycosis. Cases 4 ( x ), 6 (A), 18 (O),20 (v)and 43 (v) had Addison’s disease while the remaining patients are represented by solid circles. The dashed lines or the hatched area limit the 5th and 95th percentiles for normal subjects. Fig. 1
250
0
LPH value (20 pmol/l), in contrast to borderline lower normal cortisol levels (190 nmol/l), and no pigmentary changes. In these five patients, there was a positive Spearman’s rank correlation between plasma ACTH and P-LPH levels in samples collected at the same time (r =0.57, n = 15, P
