29
Clinica Chimica Acta, 67 (1976) 29-34 0 Elsevier Scientific Publishing Company,
Amsterdam
-
Printed
in The Netherlands
CCA 7532
USE OF SMALL SAMPLES OF URINE TO MONITOR TESTICULAR RESPONSE TO HUMAN CHORIONIC GONADOTROPIN
MARY
G. METCALF
Medical
Unit,
(Received
and ESTHER
The Princess
August
Margaret
G. HUNT Hospital,
Christchurch,
2 (New Zealand)
11,1975)
summary The increased urinary excretion of androsterone and aetiocholanolone, 17-oxosteroids, and oestrogens which occurs after the administration of human chorionic gonadotropin to normal men, was followed by measuring steroid to creatinine ratios in small samples of urine. For androsterone + aetiocholanolone, values of the ratio correspond closely with 24 h output (correlation coefficients: 0.91, 0.92, and 0.91 for first morning, mid-morning, and mid-afternoon urine specimens). Comparable correlation coefficients for the 17-oxosteroids in urine were 0.90,0.95, and 0.91; and for oestrogens, 0.88, 0.87, and 0.86.
Introduction The validity of using small samples of urine to follow ovulatory changes has already been established [ 11. In this paper evidence is presented which suggests that a similar approach is possible when testicular function is to be evaluated. The surge in testosterone production which follows the administration of human chorionic gonadotropin (HCG) to normal men, is accompanied by corresponding increases in the urinary output of androsterone and aetiocholanolone (A + E), 17-oxosteroids, and oestrogens [2,3]. In what follows it is shown that this response may be monitored by measuring steroid to creatinine ratios in small samples of urine. The collection of 24 h specimens is unnecessary. Materials and methods Normal subjects
Men taking part in the study were hospital employees aged 18 to 45 years. Methods
Androsterone and aetiocholanolone
(A + E; 5a-androstan-3a-ol-17-one
+ 5/I-
androstan-3cu-ol-l7-one) were measured together by the gas chromatographic analysis of chloroform extracts of acid hydrolysed urine 141. The 17-0x0steroids in urine were measured as described by Drekter et al. [S] and modified by Hamburger 163, total oestrogens by fluorimetry [ 71, and creatinine by an autoanalyzer technique [8]. Interassay variability was 5.1% for the A f E assay (coefficient of variation), 7.1% for the 17-oxostcroids, 11.9% for the oestrogens, and 6.9% for the creatinine measurements. Sample collection Urine collections were commenced between 7.30 and 9.00 a.m. and continued for 24 h, with a mid-morning sample (passed between 9.30 and 11.30 a.m.), a mid-afternoon sample (2.30-4.30 p.m.) and the first morning specimen (6.00-7.30 a.m.) kept out of the main collection. The mid-morning samples ranged from 2.2 to 26.1% (mean, 10.8%) of the 24 h urine volume, mid-afternoon samples from 3.8 to 24.2% (mean 11.4%) and early morning samples from 8.1 to 57.7% (mean, 30.6%). A 24-h sample was reconstituted from each set of 4 divided samples. Steroid to creatinine concentrations were measured in the individual specimens and compared with the 24-h excretion rate. Test procedures Urine was collected from 7 men before (2 days), and during (3 days) adrenocortical suppression with dexamethasone (daily regime: 1 mg near mid-night, 0.5 mg at breakfast and 0.5 mg at lunch); and from 5 men during concurrent dexamethasone suppression and HCG stimulation (4500 units injected i.m., between 8 and 9 a.m. each day for 4 days). Dexamethasone suppression and specimen collection were continued for 3 days after the final injection of HCG. Results A highly significant correlation (P < 0.01) was observed between the 24-h output of each group of urinary steroids, and their concentration relative to creatinine in small samples of urine collected on the same day (Figs. l-3). For A f E and the 17-oxosteroids, the correspondence was improved if the statistical analysis was restricted to samples collected from the day preceding the first injection of HCG onwards (correlation coefficients between daily A + E output and the (A + E)/creatinine ratios in first morning, mid-morning and mid-afternoon samples = 0.91, 0.92, and 0.91, respectively; comparable values for 17oxosteroids = 0.90, 0.95, and 0.91; and for oestrogens = 0.88,0.87, and 0.86). The slopes of the regression lines relating steroid/creatinine ratios to steroid output were least for measurements made on the mid-morning specimens of urine (Figs. 1-3). Neither dexamethasone nor HCG altered significantly the rate at which creatinme appeared in the urine of the men tested. Creatinine excretion before treatment was 1.86 rt 0.21 g/24 h (mean + SD.), compared with 1.74 + 0.25 after two or more days on dexamethasone, and 1.76 + 0.23 g/24 h after three or more days stimulation with HCG. The types of response seen in a normal individual are summarised in Fig. 4. For all the men and all urine samples, the HCG-induced increase by the third
31
r=089 y i 1.361 * 1.41
y = 1.07x * ls3
0
5
to A.E
0 (mg)
10
5 I”
Creotlnlne(g)
mm,
0
5
10
Of unne
scimples
Fig. 1. Correlation between the amount of androsterone + aetioeholanoIone (A + E) in 24-h samples of urine, and the A + E to creatinine ratios in small samples collected during the same 24-h period. r. correlation coefficient calculated from 58 pairs of observations on 7 men. ~, urine collected from men in the basal state; .‘_, urine collected during dexamethasone suppression; . . urine collected during concurrent dexamethasone suppression and HCG stimulation.
day of stimulation was never less than 50% of the pre-HCG level. This increase averaged 128% when the A + E assay was used to follow response, 107% for the 17-oxosteroids and 281% for the oestrogens (Table I). It changed little when HCG stimulation was continued for 4 days (mean response by the fourth day for A + E, 140%; 17-oxosteroids, 133%; oestrogens 301%).
17”Oxostero8ds (mg) Creotlnine (gj
I” 5rn1(311samples
0,
“rlne
Fig. 2. Correlation between the 24-h excretion rate of urinary 17-oxosteroids, and the l’l-oxosteroid to creatinine ratios in small samples of urine collected during the same 24-h period. For symbols see legend to Fig. 1.
First
0 0
Fig.
y = 1 611.1.34 sample I 1 1 10 15 20
morning I 5
3. Correlation
in small
between
samples
of urine
the
24-h
collected
excretion
during
rate
of total
the samt? 24-h
orstrogens,
period.
For
and
symbols
oestrogen
to creatinine
see legend
to
Fig.
ratios
I.
Discussion There is little doubt that the response of men to HCG may be measured by means of observations made on small samples of urine. Either A + E, 17-0x0steroid, or oestrogen to creatinine ratios, may be used for this purpose. In each case, values of the ratio correlated to a significant degree with daily steroid output (Figs. l-3). The time at which the small urine samples are collected is important. The TABLE
I
RESPONSE WITH
OF
HCG:
STEROIDS
AND
CREATININE
IN
Parameter Urine
5
DEXAMETHASONE-SUPPRESSED.
COMPARISON
BETWEEN
OESTROGENS SMALL
THE
NORMAL %
EXCRETED
SAMPLES
OF
measured:
URINE
EACH
24
IN h,
24 h
MEN THE
AND
COLLECTED
Steroid
sample:
INCREASE
ON
TO
THEIR THE
Steroid/creatinine ~___~
3
DAYS
AMOUNT
OF
INCREASE
SAME
STIMULATION A + E.
DAY.
(9)
First
Mid-
Mid-
morning
morning
afternoon
116
118
_____ Assay mean
A + E (me)
observed 17-Oxosteroids
(mg)
(pg)
118
observed D. amount
excreted
the
H. amount
excreted
the third
day
129-453
immediately day
of
HCG
54
74-217 259
312 range
7 2-2 122
72-202
range
mean
~_____
149
81-222
mea” observed
Oestrogens
132 range
preceding stimulation.
92-358 the first
injection
95-164
87-184
98
88
63-146
50-121
308
245
142-507
144-391
of HCG:
17-0X0-
RELATIVE
TO
33
Dl
Hl Day
of
H2
H3
H4
D2
test
Fig. 4. Effect of HCG (4500 I.U. each day) on the A + E. i7-oxosteroids and oestrogens in the urine of a man aged 20 years: comparison between 24-h excretion rates (block diagram) and steroid to creatinine ratios in mid-afternoon (A. -A), first morning (o v) and mid-morning IO- - - - - -0) samples of urine. The mid-morning samples were collected on the day following HCG injection. B. basal state; Dl, 3rd day of dexamethasone suppression; Hl, H2, H3, H4 1st. 2nd. 3rd and 4th days of HCG stimulation; D2, dexamethasone continued after period of HCG stimulation.
effect of HCG on the steroids in a mid-morning sample of urine collected less than two hours after injection is not the same as that on the steroids in a first morning specimen collected some 20 h later, and for each group of steroids, the slope of the regression lines relating steroidlcreatinine ratios and 24-h output was least for the mid-morning samples (Figs. l-3). It is for this reason that HCG response should be measured in samples collected at approximately the same time each day. The correspondence between steroid output and steroid to creatinine ratios was closest for urine specimens passed during the period of testicular stimulation. In these circumstances the fit was better for the A + E and 17-oxosteroid 0.91-0.95) than for the oestrogen assay assays (correlation coefficients, (0.86-0.88). Changes in oestrogen output however, gave a more sensitive index of response. The mean oestrogen increment by the third day of stimulation was
almost three times the basal level; this compared with the much smaller increments observed for A + E and the 17-oxosteroids (Table I). The main drawback to the use of small samples of urine for following HCG response is the requirement that the concentration of creatinine be measured in every specimen. In practice this requirement is rarely a hindrance. The advent of autoan~yser techniques has meant that creatinine meas~ements on large numbers of specimens are readily avaiiable. Acknowledgements This project was supported by grants from the New Zealand Medical Reseach Council and the North Canterbury Hospital Board. Our thanks are due to Miss V. Pugh who did some of the steroid measurements, to the staff of the autoanalyzer laboratory at Christchurch Hospital who undertook the creatinine estimations, and to Mrs. R. Spinks and Mr. L. Wilkinson who prepared the script. References 1 2 3 4 5
?Artcalf. M.G. (1973) Am. J. Obstet. Grnec. 117. 1041 i?icl.fafusy, F:., Palmer, K.Y. (1967) Eur. Rev. Endocrinol., Suypl. 2,337 Mstealf. M.G. and Cowlcs, R..J. (1975) Clin. Endocrinol. in press Metcalf. M.G. (1970) Clin. Biothcm. 3, 271 .J. Clin. EndoIIrckter. I.J.. Hrisfer, A., S&m. K.G., Stern, S., Pearson. S. and McCavack. T.H. (1952) crinol. 12, 55 6 Hamburger. C. (1952) Acta Endocrinol. 9. 129 7 Bruwn, J.B.. MacNaughtan. C.. Smith, M.A. and Smvth. B. (1968) J. Endocrinol. 40. 175 8 Chasson, A.L. Grady, I4.d. and Stanley, M.A. (1961) Am. J. Clin. Pathol. 35, 83
Clinica Chimica Acta, 67 (1976) 29-34 0 Elsevier Scientific Publishing Company,
Amsterdam
-
Printed
in The Netherlands
CCA 7532
USE OF SMALL SAMPLES OF URINE TO MONITOR TESTICULAR RESPONSE TO HUMAN CHORIONIC GONADOTROPIN
MARY
G. METCALF
Medical
Unit,
(Received
and ESTHER
The Princess
August
Margaret
G. HUNT Hospital,
Christchurch,
2 (New Zealand)
11,1975)
summary The increased urinary excretion of androsterone and aetiocholanolone, 17-oxosteroids, and oestrogens which occurs after the administration of human chorionic gonadotropin to normal men, was followed by measuring steroid to creatinine ratios in small samples of urine. For androsterone + aetiocholanolone, values of the ratio correspond closely with 24 h output (correlation coefficients: 0.91, 0.92, and 0.91 for first morning, mid-morning, and mid-afternoon urine specimens). Comparable correlation coefficients for the 17-oxosteroids in urine were 0.90,0.95, and 0.91; and for oestrogens, 0.88, 0.87, and 0.86.
Introduction The validity of using small samples of urine to follow ovulatory changes has already been established [ 11. In this paper evidence is presented which suggests that a similar approach is possible when testicular function is to be evaluated. The surge in testosterone production which follows the administration of human chorionic gonadotropin (HCG) to normal men, is accompanied by corresponding increases in the urinary output of androsterone and aetiocholanolone (A + E), 17-oxosteroids, and oestrogens [2,3]. In what follows it is shown that this response may be monitored by measuring steroid to creatinine ratios in small samples of urine. The collection of 24 h specimens is unnecessary. Materials and methods Normal subjects
Men taking part in the study were hospital employees aged 18 to 45 years. Methods
Androsterone and aetiocholanolone
(A + E; 5a-androstan-3a-ol-17-one
+ 5/I-
androstan-3cu-ol-l7-one) were measured together by the gas chromatographic analysis of chloroform extracts of acid hydrolysed urine 141. The 17-0x0steroids in urine were measured as described by Drekter et al. [S] and modified by Hamburger 163, total oestrogens by fluorimetry [ 71, and creatinine by an autoanalyzer technique [8]. Interassay variability was 5.1% for the A f E assay (coefficient of variation), 7.1% for the 17-oxostcroids, 11.9% for the oestrogens, and 6.9% for the creatinine measurements. Sample collection Urine collections were commenced between 7.30 and 9.00 a.m. and continued for 24 h, with a mid-morning sample (passed between 9.30 and 11.30 a.m.), a mid-afternoon sample (2.30-4.30 p.m.) and the first morning specimen (6.00-7.30 a.m.) kept out of the main collection. The mid-morning samples ranged from 2.2 to 26.1% (mean, 10.8%) of the 24 h urine volume, mid-afternoon samples from 3.8 to 24.2% (mean 11.4%) and early morning samples from 8.1 to 57.7% (mean, 30.6%). A 24-h sample was reconstituted from each set of 4 divided samples. Steroid to creatinine concentrations were measured in the individual specimens and compared with the 24-h excretion rate. Test procedures Urine was collected from 7 men before (2 days), and during (3 days) adrenocortical suppression with dexamethasone (daily regime: 1 mg near mid-night, 0.5 mg at breakfast and 0.5 mg at lunch); and from 5 men during concurrent dexamethasone suppression and HCG stimulation (4500 units injected i.m., between 8 and 9 a.m. each day for 4 days). Dexamethasone suppression and specimen collection were continued for 3 days after the final injection of HCG. Results A highly significant correlation (P < 0.01) was observed between the 24-h output of each group of urinary steroids, and their concentration relative to creatinine in small samples of urine collected on the same day (Figs. l-3). For A f E and the 17-oxosteroids, the correspondence was improved if the statistical analysis was restricted to samples collected from the day preceding the first injection of HCG onwards (correlation coefficients between daily A + E output and the (A + E)/creatinine ratios in first morning, mid-morning and mid-afternoon samples = 0.91, 0.92, and 0.91, respectively; comparable values for 17oxosteroids = 0.90, 0.95, and 0.91; and for oestrogens = 0.88,0.87, and 0.86). The slopes of the regression lines relating steroid/creatinine ratios to steroid output were least for measurements made on the mid-morning specimens of urine (Figs. 1-3). Neither dexamethasone nor HCG altered significantly the rate at which creatinme appeared in the urine of the men tested. Creatinine excretion before treatment was 1.86 rt 0.21 g/24 h (mean + SD.), compared with 1.74 + 0.25 after two or more days on dexamethasone, and 1.76 + 0.23 g/24 h after three or more days stimulation with HCG. The types of response seen in a normal individual are summarised in Fig. 4. For all the men and all urine samples, the HCG-induced increase by the third
31
r=089 y i 1.361 * 1.41
y = 1.07x * ls3
0
5
to A.E
0 (mg)
10
5 I”
Creotlnlne(g)
mm,
0
5
10
Of unne
scimples
Fig. 1. Correlation between the amount of androsterone + aetioeholanoIone (A + E) in 24-h samples of urine, and the A + E to creatinine ratios in small samples collected during the same 24-h period. r. correlation coefficient calculated from 58 pairs of observations on 7 men. ~, urine collected from men in the basal state; .‘_, urine collected during dexamethasone suppression; . . urine collected during concurrent dexamethasone suppression and HCG stimulation.
day of stimulation was never less than 50% of the pre-HCG level. This increase averaged 128% when the A + E assay was used to follow response, 107% for the 17-oxosteroids and 281% for the oestrogens (Table I). It changed little when HCG stimulation was continued for 4 days (mean response by the fourth day for A + E, 140%; 17-oxosteroids, 133%; oestrogens 301%).
17”Oxostero8ds (mg) Creotlnine (gj
I” 5rn1(311samples
0,
“rlne
Fig. 2. Correlation between the 24-h excretion rate of urinary 17-oxosteroids, and the l’l-oxosteroid to creatinine ratios in small samples of urine collected during the same 24-h period. For symbols see legend to Fig. 1.
First
0 0
Fig.
y = 1 611.1.34 sample I 1 1 10 15 20
morning I 5
3. Correlation
in small
between
samples
of urine
the
24-h
collected
excretion
during
rate
of total
the samt? 24-h
orstrogens,
period.
For
and
symbols
oestrogen
to creatinine
see legend
to
Fig.
ratios
I.
Discussion There is little doubt that the response of men to HCG may be measured by means of observations made on small samples of urine. Either A + E, 17-0x0steroid, or oestrogen to creatinine ratios, may be used for this purpose. In each case, values of the ratio correlated to a significant degree with daily steroid output (Figs. l-3). The time at which the small urine samples are collected is important. The TABLE
I
RESPONSE WITH
OF
HCG:
STEROIDS
AND
CREATININE
IN
Parameter Urine
5
DEXAMETHASONE-SUPPRESSED.
COMPARISON
BETWEEN
OESTROGENS SMALL
THE
NORMAL %
EXCRETED
SAMPLES
OF
measured:
URINE
EACH
24
IN h,
24 h
MEN THE
AND
COLLECTED
Steroid
sample:
INCREASE
ON
TO
THEIR THE
Steroid/creatinine ~___~
3
DAYS
AMOUNT
OF
INCREASE
SAME
STIMULATION A + E.
DAY.
(9)
First
Mid-
Mid-
morning
morning
afternoon
116
118
_____ Assay mean
A + E (me)
observed 17-Oxosteroids
(mg)
(pg)
118
observed D. amount
excreted
the
H. amount
excreted
the third
day
129-453
immediately day
of
HCG
54
74-217 259
312 range
7 2-2 122
72-202
range
mean
~_____
149
81-222
mea” observed
Oestrogens
132 range
preceding stimulation.
92-358 the first
injection
95-164
87-184
98
88
63-146
50-121
308
245
142-507
144-391
of HCG:
17-0X0-
RELATIVE
TO
33
Dl
Hl Day
of
H2
H3
H4
D2
test
Fig. 4. Effect of HCG (4500 I.U. each day) on the A + E. i7-oxosteroids and oestrogens in the urine of a man aged 20 years: comparison between 24-h excretion rates (block diagram) and steroid to creatinine ratios in mid-afternoon (A. -A), first morning (o v) and mid-morning IO- - - - - -0) samples of urine. The mid-morning samples were collected on the day following HCG injection. B. basal state; Dl, 3rd day of dexamethasone suppression; Hl, H2, H3, H4 1st. 2nd. 3rd and 4th days of HCG stimulation; D2, dexamethasone continued after period of HCG stimulation.
effect of HCG on the steroids in a mid-morning sample of urine collected less than two hours after injection is not the same as that on the steroids in a first morning specimen collected some 20 h later, and for each group of steroids, the slope of the regression lines relating steroidlcreatinine ratios and 24-h output was least for the mid-morning samples (Figs. l-3). It is for this reason that HCG response should be measured in samples collected at approximately the same time each day. The correspondence between steroid output and steroid to creatinine ratios was closest for urine specimens passed during the period of testicular stimulation. In these circumstances the fit was better for the A + E and 17-oxosteroid 0.91-0.95) than for the oestrogen assay assays (correlation coefficients, (0.86-0.88). Changes in oestrogen output however, gave a more sensitive index of response. The mean oestrogen increment by the third day of stimulation was
almost three times the basal level; this compared with the much smaller increments observed for A + E and the 17-oxosteroids (Table I). The main drawback to the use of small samples of urine for following HCG response is the requirement that the concentration of creatinine be measured in every specimen. In practice this requirement is rarely a hindrance. The advent of autoan~yser techniques has meant that creatinine meas~ements on large numbers of specimens are readily avaiiable. Acknowledgements This project was supported by grants from the New Zealand Medical Reseach Council and the North Canterbury Hospital Board. Our thanks are due to Miss V. Pugh who did some of the steroid measurements, to the staff of the autoanalyzer laboratory at Christchurch Hospital who undertook the creatinine estimations, and to Mrs. R. Spinks and Mr. L. Wilkinson who prepared the script. References 1 2 3 4 5
?Artcalf. M.G. (1973) Am. J. Obstet. Grnec. 117. 1041 i?icl.fafusy, F:., Palmer, K.Y. (1967) Eur. Rev. Endocrinol., Suypl. 2,337 Mstealf. M.G. and Cowlcs, R..J. (1975) Clin. Endocrinol. in press Metcalf. M.G. (1970) Clin. Biothcm. 3, 271 .J. Clin. EndoIIrckter. I.J.. Hrisfer, A., S&m. K.G., Stern, S., Pearson. S. and McCavack. T.H. (1952) crinol. 12, 55 6 Hamburger. C. (1952) Acta Endocrinol. 9. 129 7 Bruwn, J.B.. MacNaughtan. C.. Smith, M.A. and Smvth. B. (1968) J. Endocrinol. 40. 175 8 Chasson, A.L. Grady, I4.d. and Stanley, M.A. (1961) Am. J. Clin. Pathol. 35, 83
