- Email: [email protected]
CLONIDINE TREATMENT FOR SHORT STATURE
1226
CLONIDINE TREATMENT FOR SHORT STATURE CARLO PINTOR SILVANO G. CELLA SANDRO LOCHE ROSA PUGGIONI ROBERTO CORDA VITTORIO LOCATELLI EUGENIO E. MÜLLER First Department of Paediatrics, University of Cagliari, and Department of Pharmacology, University of Milan,
Milan, Italy
pubertal children with constitutional growth delay (CGD) were treated with clonidine orally twice a day. In 25 of the children the height velocity rose on clonidine treatment, and in 21 of them by more than 2 cm/yr during the first 6 months of treatment (mean [SD] growth increment 4·4 [0·5] cm/yr). Of the 22 who were treated for 12 months the increment in height velocity was maintained in 13 (3·4 [0·4] cm/yr). Withdrawal of clonidine for 6 months did not stop the stimulatory effect of the drug on linear growth in 6 children, but in the other 8 children height velocities fell to pretreatment levels or
Summary
recognisable dysmorphic syndrome, or psychosocial disturbances, and no child had received long-term medication, human growth hormone therapy, or anabolic steroids. Informed written parental consent was received for the studies, which were also approved by the ad-hoc committee of the University of Cagliari Department of Paediatrics.
34
below. In a few children reinstitution of clonidine for 2-4 months resulted in a new increment in height velocity. A high height standard deviation score and low growth velocity before treatment were predictive of a good growth response to clonidine. Clonidine did not induce noticeable side-effects. It may be a useful form of therapy for children with CGD. Introduction
ADMINISTRATION of one of the analogues of the hypothalamic growth-hormone-releasing hormone (GHRH), the neuropeptide that specifically stimulates growth hormone (GH) release, to children with short stature evokes in most of them a sizeable GH response.l-4 This indicates the presence of functioning pituitary somatotrophs and the existence in these subjects of impaired hypothalamic GHRH synthesis and/or release. We previously reported that clonidine, an a-adrenergic agonist which is capable of effecting GHRH release,5—7 accelerated growth in 2 children with isolated GH deficiency and 4 children with constitutional growth delay (CGD) treated for 2 months.8 In the present study we report the effect of clonidine, given orally twice a day for up to 12 months, in 34 children with CGD. Patients and Methods
Treatment Clonidine (Catapresan, Boehringer Ingelheim, Florence) (0-1
mg/m2) was given orally in two doses, at 0800 h (0-033 mg/m2) and at bedtime (0-066 mg/m2). This treatment was continued in 34 children for 6 months and in 22 of them for up to 12 months. The 22 children were then withdrawn from clonidine for 6 months, during which time they were given a placebo. At the end of this period, clonidine was restarted in 6 children, and height velocity was re-evaluated 2-4 months later.
Tests Standard intravenous GHRH tests using 1 ug/kg bolus doses of GHRH (hpGRF-40, Bachem, Bubendorf) and oral clonidine tests (0-15 mg/m2) were done before and after 2, 4, 6, and 12 months of therapy. GH and somatomedin C (SM-C) determinations and provocative GHRH and clonidine tests were carried out 14 h after the last clonidine dose, except for the 12 month evaluation, which was done 3-10 days after clonidine withdrawal. Details of blood sampling after GHRH and clonidine have been published elsewhere."
Assays GH was measured by means of radioimmunoassay (reagents provided by CEA-IRE Sorin, Saluggia). The detection limit of the assay was 0-5 ng/ml, with an interassay coefficient of variation of 5-6% and an intra-assay variation of 4-8% at 10 ng. SM-C levels were measured with reagents provided by Nichols Institute Diagnostic (San Juan Capistrano, CA); in our laboratory, normal mean (SEM) values for prepubertal children are 1-25 (0-8) mU/ml for boys and 1 ’45 (0-4) mU/ml for girls. Pulse rate and supine blood pressure were monitored at the start of treatment, after 2, 4, 6, and 12 months of therapy, and also during the acute clonidine test; subjective and objective side-effects were recorded. Routine blood tests were carried out periodically.
Growth Measurements
Height was measured with a Harpenden stadiometer by the same (S. L. and R. P.), with an intra-assay variation (SD) of 0-12 cm and an interassay variation of 0 22 cm. Except where stated, the height velocities were calculated over a 6-month period. Height was expressed as standard deviation score (SDS) and height velocity (HV) in cm/yr. Bone age was assessed with the Tanner-Whitehouse 2 method. Subjects were categorised as responders (C-R) and non-responders (C-NR), according to the increment in HV recorded at 6 and/or 12 months, the C-R group having an increase of at least 2-0 cm/yr. This figure approximates to two trained observers
Patients 34 children (20 boys, 14 girls) with CGD were studied. Clinical and laboratory details of the children are shown in tables i and n. The children fulfilled Frazier’s criteria for CGD.1 In all cases there
25. Davie CE, Hutt SJ, Vincent E, Mason M. The family cleanliness scale. In: The young
child
strong familial pattern of childhood growth and adolescent development. Mean (SD) midparental height of the children was 158-6 (4-9) cm (mean height in Sardinian population 167 [5] cm’°). All these children had had a birthweight appropriate for gestational age and had a predicted adult height below the third percentile. There was no evidence of systemic disease, malnutrition, was a
home. London:
NFER/Nelson, 1983. Otto DA, Mushak P, Hicks RE. Separating the effects of lead and social factors on IQ. Environ Res 1985; 38: 144-54. 27. Winneke G, Kramer U, Brockhaus A, et al. Neuropsychological studies in children with elevated tooth-lead concentrations. Internat Arch Occup Environ Health 1983; at
26. Schroeder SR, Hawk B,
51: 231-52. 28. Lansdown R, Yule W, Urbanowicz M-A, Hunter J. The relationship between blood-lead concentrations, intelligence, attainment and behaviour in a school population: the second London study. Internat Arch Occup Environ Health 1986; 57: 225-35. 29. Bellinger DC, Needleman HL, Leviton A, Watemaux C, Rabinowitz MB, Nichols ML. Early sensory-motor development and prenatal exposure to lead. Neurobehav Toxicol Teratol 1984; 6: 387-402. 30. Pocock SJ, Ashby D, Smith MA. Lead exposure and children’s intellectual performance. Im J Epidemiol 1987; 16: 57-67. 31. Pocock SJ, Ashby D. Environmental lead and children’s intelligence: A review of recent epidemiological studies. Statistician 1985; 34: 31-44.
the
two
standard deviations
on
the variation that
occurs
in
whole-year velocities in normal children12 and has been previously used to define responsiveness to hGH13 and GHRH14 treatment in children with short
stature.
Statistical Analysis
Unpaired and paired Student’s t-tests, preceded by ANOVA where necessary, and the X2 test were used for statistical analysis. p < 0-05 (two-tailed) was taken to indicate a significant difference. Linear regression analysis was used to correlate the clinical and laboratory data. Discriminant analysis was applied to obtain HV and height SDS thresholds.’S
1227 results
After 6 months of clonidine therapy 25 of the 34 children showed an increase in height velocity (table l, nos 1-5, 7-22; table II, nos 25, 26, 30, 33), and in 21 this was greater than 2 cm/yr; 15 of the 21 C-R children (nos 1—4, 6-8, 10, 11, 14-16, 19, 21, 22) have now been treated for 12 months, and the increase in height velocity has been maintained in 13 of these. Patient 6, who was C-NR at 6 months, became C-R at 12 months of treatment, whereas the reverse was true for patients 2 and 10 (table 1). 7 of the C-NR children (nos 23, 26,27,29,30,31,32) have now been treated for 12 months, but none showed an increase in height velocity of 2 cm/yr at the end of the second 6-month period (table I). In 2 of the C-NR group (nos 23 and 27) a clear-cut growth deceleration was evident with clonidine. Withdrawal of clonidine for 6 months in 14 C-R children resulted in unimpaired or only slightly reduced growth velocity in 6 (nos 1,6,7,10,14,21), whereas in the remaining 8 children (nos 2, 3, 4, 8, 11, 15, 16, 22) growth velocity returned close to or below pretherapy values. In 6 of these 8 clonidine was restarted for 2—4 months, and a new increment in growth velocity was seen in 4 of them. Withdrawal of clonidine in C-NR children was associated in 2 cases (nos 23 and 26) with a rebound increment in growth velocity; there was no change in growth velocity in the other 4 (nos 27, 30, 31, 32) (table I). No alteration in the bone-age: chronological-age ratio,as assessed before and after 12 months’ treatment, was evident in the C-R children. Figs 1 and 2 show the pattern of height velocity before, during, and after clonidine treatment and after reinstitution of clonidine in individual C-R and C-NR children. Comparison of the pretreatment clinical indices in C-R and C-NR children (table I) revealed a significantly higher mean SDS and lower HV in the former than the latter (p < 0-05 and p<0-01, respectively). Mean bone age was lower in C-R than in C-NR children, but the difference did not reach statistical-significance. In the C-R children there was an inverse relation between pretreatment HV and the and a increase in HV at 12 months (r = 0-54, p<005) the and increase in relation between bone-age delay positive HV at 12 months (r = 0 60, p < 0-01). Discriminant analysis showed that a child could be classified with a confidence of 80% as C-R when his/her HV and SDS were 46 cm/yr and -2-5, respectively. Thus the probability of classifying as C-NR a child who may actually respond to clonidine is 20%. Growth-hormone Responses
to
Fig 1-Height velocities responders.
on
and off treatment in clonidine
Bars denote mean height velocity at the different time periods. For the sake of clarity SEs have been omitted. Numbers reported refer to individual case numbers.
23 children who were in pubertal stage Po, 4 (C-R nos 1, 10, 22 and C-NR no 27) had pubertal progression to stages P1 and P2. 4 children who were in pubertal stage P (C-R nos 14 and 21, C-NR nos 23 and 32) had pubertal progression to stages P2 and Ps. Of the 7 children who were in pubertal stage P2, 4 (C-R nos 2, 7, 16, C-NR no 31) had pubertal
progression to stages P3 and P4 (table 1). The frequency of pubertal changes during clonidine treatment was not significantly different between C-R and C-NR children 10r
GHRH and Clonidine
There were no significant differences between the two groups in the pretreatment biochemical indices (table n). In C-R children there was during treatment an initial rise in plasma GH (at 2 months) and SM-C levels (at 2 and 4 months). At 12 months there was a further increase in plasma GH but not in SM-C levels. Peak GH responses to GHRH and clonidine showed a similar trend: an initial increase (at 2 and 4 months) was followed by a slight decline at 6 months and then by another increase at 12 months. In C-NR children none of these indices was significantly altered during the course of clonidine treatment except for the peak GH response to clonidine at 2 and 12 months, which increased significantly (table II).
Pubertal Development
There was no significant difference in the frequency distribution of pubertal stages, as defined by Tanner et al,12 between C-R (Po = 15/22; Pl 2/22; P2 5/22) and C-NR =
=
(Po=8/12; Pl=2/12; P2=2/12) (X2=0.49 DF=2).
Of the
Fig 2--Height velocities
on
and off treatment in clonidine
non-
responders. Bars denote mean height velocity at the different time periods. For the sake of clarity SEs have been omitted. Numbers reported refer to individual case numbers.
1228 TABLE I-PATIENTS’ CLINICAL DETAILS AND RESPONSE IN HEIGHT VELOCITY TO CLONIDINE
(xz 9-09, DF 5). Determination of mean increase in HV in C-R children with and without pubertal progression gave values not statistically different (at 12 months, C-R without pubertal progression, 2-9 ± 0-3 cm/yr; C-R with pubertal progression, 3-8 ± 0-7 cm/yr). =
=
Side-effects None of the children had noticeable side-effects during 4 children presented with sedation and which diminished in 3, despite continued use of sleepiness, and discontinuation of the drug in 1. necessitated clonidine, 2 children had a skin rash and urticaria and had to be excluded from the study. Acute clonidine testing induced drowsiness or sleep and slight reduction (approximately 20 mm Hg) in systolic blood pressure both before and after chronic clonidine treatment. The results of routine blood tests remained normal in all patients during the course of clonidine treatment. treatment.
Discussion Clonidine is a potent GH secretagogue in both animals16,17 and man,18,19 and it acts via release of GHRH.5-7 Administration of GHRH evokes a GH response in most GH-deficient childrenl-4 and children with short stature,20 indicating a hypothalamic defect in GH release.
Our previously reported success with clonidine treatment in 4 children with CGD demonstrated that some children with short stature have a primary defect in the mechanism of GHRH release (but not synthesis).8 We have now shown that oral administration of clonidine to children with CGD promoted linear growth of more than 2 cm/yr in 21 of 34 children during 6 months and in 13 out of 22 children during 12 months of treatment. The effect of clonidine in children with CGD compares favourably with that of GH administered to children of short stature. 13,21 Withdrawal of clonidine after 12 months’ treatment resulted in deceleration of linear growth to pre-therapy or lower values in 8 C-R children, but the enhancing effect of clonidine on linear growth was maintained in the remaining 6 children. These findings are difficult to interpret; we suggest, however, that the children who continued to grow while off clonidine had less severe impairment of hypothalamic adrenergic neurotransmission than the remaining C-R children, in that pharmacological stimulation with clonidine was able to re-establish normal neurotransmitter function. This response is similar to the maintenance of increased height velocity seen in CGD children after withdrawal of anabolic steroids2z and fits in well with the transient impairment of GH secretion in some of these children.23 The children in our study are now off
1229 TABLE II-PATIENTS’ LABORATORY DETAILS DURING CLONIDINE TREATMENT
Results given as mean (SD). *p < 0-05 vs basal value (paired Student’s t-test). tp < 001 vs basal value (paired Student’s t-test). 34 patients were treated with clonidine for 6 months; 22 patients were treated for 12 months.
therapy so that the duration of the stimulant effect of the drug can be determined. Restarting therapy for 2-4 months triggered a new growth increment in 4 of 6 children in whom clonidine withdrawal had dramatically decreased height velocity. The acceleration of linear growth in response to clonidine associated with an increase in related biochemical indices. This pattern was more evident at 2 and 4 months than at 6 months. This does not necessarily contradict the finding that linear growth was still considerably accelerated at 6 and 12 months; it may simply indicate that the overall stimulant action of clonidine on GH secretion was still operative at a stage when single biological indices denoted a state of initial refractoriness. The finding that levels of some of these indices rose again at 12 months does not negate this view, since at that time clonidine had been withdrawn for 3-10 days. It is tempting to speculate that subtle changes in the sensitivity of hypothalamic &agr;2-adrenoceptors acted upon by clonidine24 may be responsible for this behaviour. Were this the case, further work would be needed to establish the optimum dose, timing, and duration of clonidine treatment in children with CGD. The 12 C-NR children represent a heterogeneous group, since at the completion of 6 months’ therapy height velocity had decreased in 5, increased slightly in 2, and was maintained in 5. This pattern was in essence present at 12 months, when 7 of these children could be re-evaluated. Deceleration of height velocity with clonidine was a transient event in one case (no 23); in another (no 27) height velocity was still suppressed 6 months after clonidine withdrawal. In the C-NR children there was no significant change in growth-related biochemical indices. Some clinical features appear useful in identifying the children with CGD who will respond to clonidine; the responders had a clearly higher height SDS and lower height velocity and a greater (though not significant) delay in bone age than did non-responders. Similar observations have been made in short-stature children given hGH replacement therapy.13 In our study discriminant analysis showed that values of height velocity and height SDS of 4-6 cm/year and —25 cm/year, respectively, would predict responsiveness to clonidine in 80% of the children. There was no difference in the frequency distribution of pubertal stages between C-R and C-NR children or in the was
number of C-R and C-NR children undergoing pubertal progression while taking clonidine. Moreover, the increment in height velocity in C-R children was of similar extent irrespective of whether they had undergone pubertal progression. Overall these findings suggest that the acceleration of growth observed in the C-R children is not due to underlying spontaneous or clonidine-induced pubertal changes. Other neuroactive compounds have also been used in the treatment of short-stature children. Huseman and co-workers,25.26 who reported stimulation of linear growth in some hypopituitary children treated long term with levodopa or bromocriptine, ascribed GH deficiency and short stature in these subjects to hypothalamic dopaminergic dysfunction. However, in their studies the greater effectiveness of levodopa (the precursor of both dopamine and noradrenaline27) than bromocriptine (a specific dopamine agonist28) makes impairment of noradrenergic function more likely. Our findings with clonidine support this proposition. Whichever brain neurotransmitter is primarily involved, it seems very likely that many children with growth disorders have a primary neurosecretory dysfunction, and this makes the therapeutic use of neuroactive drugs sound. Administration of clonidine for 12 months stimulated linear growth in 13 of 22 children with CGD without eliciting noticeable side-effects. Responsiveness to clonidine was positively related to initial height SDS and inversely related to growth velocity. Withdrawal of clonidine showed in some children persistence of the stimulatory effect of the drug on height velocity, perhaps indicating that short-term clonidine treatment is valid in such cases. Finally, in a few children reinstitution of clonidine after withdrawal again stimulated growth. A longer follow-up study is warranted to determine the beneficial effect of this therapy on adult
height. Clearly, clonidine is potentially useful for the treatment of children with CGD, who represent about 25% of all children and adolescents.29 It also has the advantages of low cost, lack of effect on bone age, and suitability for oral administration.
Addendum Since submission of this manuscript a report has been published showing that clonidine administered for 12 months to 16 prepubertal children with CGD significantly increased GH levels, plasma somatomedin-C, and linear growth. 30 We thank Dr G. Reina for statistical analysis of the data and Miss Maria Lupo for secretarial assistance. Correspondence should be addressed to E. E. M., Department of Pharmacology, University of Milan, Via Vanvitelli 32, 20129 Milan, Italy. REFERENCES
MO, Cronin MJ. Growth hormone releasing factor: Clinical and basic studies. In: Müller EE, MacLeod RM, eds. Neuroendocrine perspective, vol 4. Amsterdam: Elsevier, 1985: 95-144. Grossman A, Savage MO, Lytras N, et al. Responses to analogues of growthhormone-releasing hormone in normal subjects, and in growth-hormone deficient children and young adults. Clin Endocrinol 1984; 21: 253-56. Pintor C, Fanni V, Loche S, et al. Synthetic hpGRF1-40 stimulates growth hormone and inhibits prolactin secretion in normal children and children with isolated growth hormone deficiency. Peptides 1983; 4: 929-33. Evain-Brion D. GRF European Multicenter Study. GH response to a single iv injection of synthetic 1-44 GRF in prepubertal children with growing failure. In: Müller EE, MacLeod RM, eds. Neuroendocrine perspective, vol 5. Amsterdam: Elsevier, 1986: 101-09. Eden S, Eriksson E, Martin JB, et al. Evidence for a growth hormone releasing factor mediating alpha-adrenergic influence on growth hormone secretion in the rat. Neuroendocrinology 1981; 33: 24-27.
1. Thorner
2.
3.
4.
5.
1230
POLYMORPHIC DNA MARKER ON X CHROMOSOME AND MANIC DEPRESSION
JULIEN MENDLEWICZ1
PHILIPPE SIMON2 SERGE SEVY1 FRANÇOIS CHARON1 HUGUETTE BROCAS2 SYLVIANE LEGROS1 GILBERT VASSART2,3
Department of Psychiatry,1 Interdisciplinary Research Institute,2 and Department of Clinical Chemistry,3 Free University of Brussels, Erasme Campus, Belgium
Heredity is vulnerability
important factor in manic depression. A genetic linkage has been demonstrated between manic depression and coagulation factor IX at Xq27 with a TaqI polymorphism at the F9 locus in DNA samples from peripheral leucocytes of manic depressive probands and relatives in 10 informative families. Statistical analysis of the pedigrees gave a maximum lod score of 3·10 at a recombination fraction of 0·11, demonstrating a linkage between a manic depressive locus and the F9 locus in the Summary
an
to
Xq27 region. Introduction HEREDITY is
important contributing factor to susceptibility manic-depressive illness (MDI). However, family, twin, and adoption studies have not elucidated the mode of transmission.1-4 Linkage analysis has rarely been applied to psychiatric conditions. Linkage studies with colour blindness and glucose-6-phosphate dehydrogenase (G6PD) deficiency as X-chromosomal markers have shown an
to
that
dominant X-linked gene may be involved in the genetic transmission of a subtype of MD 1. 1,5-12 The maximum lod reported for the MDI-G6PD linkage was 4 21 at a recombination fraction (8) of 0.06.10 However, apparent father-to-son transmission of the illness has been described in family studies,2,13 and data which do not support a linkage between colour blindness and bipolar illness have also been reported.14 Genetic heterogeneity, which has been postulated for manic depression,11,15,16 could a
6. Katakami H, Kato Y, Matsusnita N, et al. Effects of neonatal treatment with monosodium glutamate on growth hormone release induced by clonidine and prostaglandin E1 in conscious male rats. Neuroendocrinology 1984; 38: 1-5. 7. Miki N, Ono M, Shizume K. Evidence that opiatergic and &agr;-adrenergic mechanisms stimulate rat growth hormone release via growth-hormone-releasing factor.
Endocrinology 1984; 114: 1950-52. C, Cella SG, Corda R, et al. Clonidine accelerates growth in children with impaired growth hormone secretion. Lancet 1985; i: 1482-84. 9. Frazier SD. Abnormalities of growth. In: Collu R, Ducharme JR, Guyda H, eds. Pediatric endocrinology. New York: Raven Press, 1981: 167-202. 10. Solina G. Modalità di evoluzione di alcuni parametri antropometrici. Rassegna Medica Sarda 1981; 4: 427-35. 11. Pintor C, Puggioni R, Fanni V, et al. Growth-hormone releasing factor and clonidine in children with constitutional growth delay. Evidence for defective pituitary growth hormone reserve. J Endocrinol Invest 1984; 7: 253-56. 12. Tanner JM, Whitehouse RH, Hughes PCR, et al. Effect of human growth hormone treatment for 1 to 7 years on growth of 100 children. Arch Dis Child 1971; 46: 8. Pintor
745-82. 13. Van Vliet G, Styne DM, Kaplan SL, et al. Growth hormone treatment for short stature. N Engl J Med 1983; 309: 1016-22. 14. Ross JM, Tsagarakis S, Grossman A, et al. Treatment of growth-hormone deficiency with growth-hormone-release hormone. Lancet 1987; i: 5-8. 15. Cooley WW, Lohnes PR. Multivariate data analysis. New York: John Wiley, 1971. 16. Durand D, Martin JB, Brazeau P. Evidence for a role of &agr;-adrenergic mechanisms in regulation of episodic growth hormone secretion in the rat. Endocrinology 1977; 100: 722-28. 17. Lovinger R, Holland J, Kaplan M, et al. Pharmacological evidence for stimulation of growth hormone secretion by a central noradrenergic system in dogs. Neuroscience 1976; 1: 443-50. 18. Lal S, Tolis G, Martin JB, et al Effect of clonidine on growth hormone, prolactin, leuteinizing hormone, follicle-stimulating hormone, and thyroid-stimulating hormone in the serum of normal man. J Clin Endocrinol Metab 1975; 41: 827-32.
for these apparently contradictory findings. According to this model, only a subgroup of bipolar pedigrees will show close linkage to the X-chromosome and thus carry the X-linked gene, but not all bipolar illness can be X-linked. The increasing number of restriction fragment account
length polymorphisms (RFLP) permits linkage analysis with selected DNA probes. We provide here the first evidence for a genetic linkage between MDI and coagulation factor IX at Xq27. Patients and Methods Previous linkage studies have pointed to the subterminal region of the long arm of the X-chromosome as a possible site for the MDI 1ocus.1,5—12 Therefore a factor IX probe,17 known to hybridise to the Xq27 band,18 was used to test the hypothesis that the disease trait and a common RFLP at the F9 locus18,19 would co-segregate in informative pedigrees. The pedigrees of all consecutive admissions of unipolar or bipolar probands between April, 1983, and May, 1986, to the department of psychiatry, Erasme Hospital, were screened for possible linkage analysis. The patients were referred by private physicians, and from outpatient clinics and state and private
hospitals. Pedigree data
were obtained from the proband and available relatives at the time of admission. Personal interview with all available relatives after informed consent has proved more reliable than collecting family data from the proband only.20 Medical and social records about probands and relatives were used when available. All probands, spouses, and available relatives were separately examined by two investigators who used the schedule for affective disorders and schizophrenia21 according to the research diagnostic criteria.22 The family members were assessed blind with respect to the proband’s diagnosis. Disagreements were referred to the senior investigator (J.M.), who determined blind the final diagnoses. All diagnoses were made independently from knowledge of DNA marker analysis. Those subjects diagnosed as bipolar, unipolar, or cyclothymic (ie, primary affective disorders23) were considered affectively ill. Subjects with secondary depression or alcoholism were considered "well". We have previously shown that within a family unit containing an affectively ill proband, bipolar illness, unipolar illness, and cyclothymia are genetically related
phenotypes expressing the same genotype.4-6 Whole blood samples were collected by venepuncture into tubes containing EDTA, and DNA was extracted from leucocytes as described.’ 10 ttg DNA were cleaved by TaqI endonuclease.
19. Gil-Ad I, Topper E, Laron Z. Oral clonidine as a growth-hormone stimulation test. Lancet 1979; ii: 278-80. 20. Gelato MC, Malazowski S, Caruso-Nicoletti M, et al. Growth hormone (GH) responses to GH-releasing hormone during pubertal development in normal boys and girls: comparison to idiopathic short stature and GH deficiency. J Clin Endocrinol Metab 1986; 63: 174-79. 21. Gertner JM, Genel M, Gianfredi SP, et al. Prospective clinical trial of human growth hormone in short children without growth hormone deficiency. J Pediatr 1984; 104: 172-76. 22. Bettmann HK, Goldman HS, Abramowicz M, et al. Oxandrolone treatment of short stature: effect on predicted mature height. J Pediatr 1971; 79: 10-23. 23. Gourmelen M, Phan-Huu-Trung T, Girard F. Transient partial hGH deficiency in prepubertal children with delay of growth. Pediatr Res 1979; 13: 221-24. 24. Maura G, Bonanno G, Raiteri M. Chronic clonidine induces functional downregulation of presynaptic &agr;2-adrenoceptors regulating 3H-noradrenaline and 3H-5-hydroxytryptamine release in the rat brain. Eur J Pharmacol 1985; 112: 105-10. 25. Huseman CA, Hassing JM. Evidence of dopaminergic stimulation of growth velocity in some hypopituitary children. J Clin Endocrinol Metab 1984; 58: 419-25. 26. Huseman CA, Hassing JM, Sibilia MG. Endogenous dopaminergic dysfunction: a novel form of human growth hormone deficiency and short stature. J Clin Endocrinol Metab 1986; 62: 484-89 27. Andèn NE, Dahlstrôm A, Fuxe K, et al. Functional role of the nigro-neostriatal dopamine neurons. Acta Pharmacol Toxicol 1966; 24: 263-74. 28. Fluckiger E, Markstein R. Receptor pharmacology of ergot compounds. In: Tolis G, Stefanis C, Mountokalakis T, et al, eds. Prolactin and prolactinomas. New York: Raven Press, 1983:105-13. 29. Bierich JR, Enders H, Heinrich U, et al. Stunted growth with more or less normal appearance. Eur J Pediatr 1982; 139: 214-38. 30. Castro-Magana M, Angulo M, Fuentes B, et al. Effect of prolonged clonidine administration on growth hormone concentrations and rate of linear growth in children with constitutional growth delay. J Pediatr 1986; 109: 784-87.
CLONIDINE TREATMENT FOR SHORT STATURE CARLO PINTOR SILVANO G. CELLA SANDRO LOCHE ROSA PUGGIONI ROBERTO CORDA VITTORIO LOCATELLI EUGENIO E. MÜLLER First Department of Paediatrics, University of Cagliari, and Department of Pharmacology, University of Milan,
Milan, Italy
pubertal children with constitutional growth delay (CGD) were treated with clonidine orally twice a day. In 25 of the children the height velocity rose on clonidine treatment, and in 21 of them by more than 2 cm/yr during the first 6 months of treatment (mean [SD] growth increment 4·4 [0·5] cm/yr). Of the 22 who were treated for 12 months the increment in height velocity was maintained in 13 (3·4 [0·4] cm/yr). Withdrawal of clonidine for 6 months did not stop the stimulatory effect of the drug on linear growth in 6 children, but in the other 8 children height velocities fell to pretreatment levels or
Summary
recognisable dysmorphic syndrome, or psychosocial disturbances, and no child had received long-term medication, human growth hormone therapy, or anabolic steroids. Informed written parental consent was received for the studies, which were also approved by the ad-hoc committee of the University of Cagliari Department of Paediatrics.
34
below. In a few children reinstitution of clonidine for 2-4 months resulted in a new increment in height velocity. A high height standard deviation score and low growth velocity before treatment were predictive of a good growth response to clonidine. Clonidine did not induce noticeable side-effects. It may be a useful form of therapy for children with CGD. Introduction
ADMINISTRATION of one of the analogues of the hypothalamic growth-hormone-releasing hormone (GHRH), the neuropeptide that specifically stimulates growth hormone (GH) release, to children with short stature evokes in most of them a sizeable GH response.l-4 This indicates the presence of functioning pituitary somatotrophs and the existence in these subjects of impaired hypothalamic GHRH synthesis and/or release. We previously reported that clonidine, an a-adrenergic agonist which is capable of effecting GHRH release,5—7 accelerated growth in 2 children with isolated GH deficiency and 4 children with constitutional growth delay (CGD) treated for 2 months.8 In the present study we report the effect of clonidine, given orally twice a day for up to 12 months, in 34 children with CGD. Patients and Methods
Treatment Clonidine (Catapresan, Boehringer Ingelheim, Florence) (0-1
mg/m2) was given orally in two doses, at 0800 h (0-033 mg/m2) and at bedtime (0-066 mg/m2). This treatment was continued in 34 children for 6 months and in 22 of them for up to 12 months. The 22 children were then withdrawn from clonidine for 6 months, during which time they were given a placebo. At the end of this period, clonidine was restarted in 6 children, and height velocity was re-evaluated 2-4 months later.
Tests Standard intravenous GHRH tests using 1 ug/kg bolus doses of GHRH (hpGRF-40, Bachem, Bubendorf) and oral clonidine tests (0-15 mg/m2) were done before and after 2, 4, 6, and 12 months of therapy. GH and somatomedin C (SM-C) determinations and provocative GHRH and clonidine tests were carried out 14 h after the last clonidine dose, except for the 12 month evaluation, which was done 3-10 days after clonidine withdrawal. Details of blood sampling after GHRH and clonidine have been published elsewhere."
Assays GH was measured by means of radioimmunoassay (reagents provided by CEA-IRE Sorin, Saluggia). The detection limit of the assay was 0-5 ng/ml, with an interassay coefficient of variation of 5-6% and an intra-assay variation of 4-8% at 10 ng. SM-C levels were measured with reagents provided by Nichols Institute Diagnostic (San Juan Capistrano, CA); in our laboratory, normal mean (SEM) values for prepubertal children are 1-25 (0-8) mU/ml for boys and 1 ’45 (0-4) mU/ml for girls. Pulse rate and supine blood pressure were monitored at the start of treatment, after 2, 4, 6, and 12 months of therapy, and also during the acute clonidine test; subjective and objective side-effects were recorded. Routine blood tests were carried out periodically.
Growth Measurements
Height was measured with a Harpenden stadiometer by the same (S. L. and R. P.), with an intra-assay variation (SD) of 0-12 cm and an interassay variation of 0 22 cm. Except where stated, the height velocities were calculated over a 6-month period. Height was expressed as standard deviation score (SDS) and height velocity (HV) in cm/yr. Bone age was assessed with the Tanner-Whitehouse 2 method. Subjects were categorised as responders (C-R) and non-responders (C-NR), according to the increment in HV recorded at 6 and/or 12 months, the C-R group having an increase of at least 2-0 cm/yr. This figure approximates to two trained observers
Patients 34 children (20 boys, 14 girls) with CGD were studied. Clinical and laboratory details of the children are shown in tables i and n. The children fulfilled Frazier’s criteria for CGD.1 In all cases there
25. Davie CE, Hutt SJ, Vincent E, Mason M. The family cleanliness scale. In: The young
child
strong familial pattern of childhood growth and adolescent development. Mean (SD) midparental height of the children was 158-6 (4-9) cm (mean height in Sardinian population 167 [5] cm’°). All these children had had a birthweight appropriate for gestational age and had a predicted adult height below the third percentile. There was no evidence of systemic disease, malnutrition, was a
home. London:
NFER/Nelson, 1983. Otto DA, Mushak P, Hicks RE. Separating the effects of lead and social factors on IQ. Environ Res 1985; 38: 144-54. 27. Winneke G, Kramer U, Brockhaus A, et al. Neuropsychological studies in children with elevated tooth-lead concentrations. Internat Arch Occup Environ Health 1983; at
26. Schroeder SR, Hawk B,
51: 231-52. 28. Lansdown R, Yule W, Urbanowicz M-A, Hunter J. The relationship between blood-lead concentrations, intelligence, attainment and behaviour in a school population: the second London study. Internat Arch Occup Environ Health 1986; 57: 225-35. 29. Bellinger DC, Needleman HL, Leviton A, Watemaux C, Rabinowitz MB, Nichols ML. Early sensory-motor development and prenatal exposure to lead. Neurobehav Toxicol Teratol 1984; 6: 387-402. 30. Pocock SJ, Ashby D, Smith MA. Lead exposure and children’s intellectual performance. Im J Epidemiol 1987; 16: 57-67. 31. Pocock SJ, Ashby D. Environmental lead and children’s intelligence: A review of recent epidemiological studies. Statistician 1985; 34: 31-44.
the
two
standard deviations
on
the variation that
occurs
in
whole-year velocities in normal children12 and has been previously used to define responsiveness to hGH13 and GHRH14 treatment in children with short
stature.
Statistical Analysis
Unpaired and paired Student’s t-tests, preceded by ANOVA where necessary, and the X2 test were used for statistical analysis. p < 0-05 (two-tailed) was taken to indicate a significant difference. Linear regression analysis was used to correlate the clinical and laboratory data. Discriminant analysis was applied to obtain HV and height SDS thresholds.’S
1227 results
After 6 months of clonidine therapy 25 of the 34 children showed an increase in height velocity (table l, nos 1-5, 7-22; table II, nos 25, 26, 30, 33), and in 21 this was greater than 2 cm/yr; 15 of the 21 C-R children (nos 1—4, 6-8, 10, 11, 14-16, 19, 21, 22) have now been treated for 12 months, and the increase in height velocity has been maintained in 13 of these. Patient 6, who was C-NR at 6 months, became C-R at 12 months of treatment, whereas the reverse was true for patients 2 and 10 (table 1). 7 of the C-NR children (nos 23, 26,27,29,30,31,32) have now been treated for 12 months, but none showed an increase in height velocity of 2 cm/yr at the end of the second 6-month period (table I). In 2 of the C-NR group (nos 23 and 27) a clear-cut growth deceleration was evident with clonidine. Withdrawal of clonidine for 6 months in 14 C-R children resulted in unimpaired or only slightly reduced growth velocity in 6 (nos 1,6,7,10,14,21), whereas in the remaining 8 children (nos 2, 3, 4, 8, 11, 15, 16, 22) growth velocity returned close to or below pretherapy values. In 6 of these 8 clonidine was restarted for 2—4 months, and a new increment in growth velocity was seen in 4 of them. Withdrawal of clonidine in C-NR children was associated in 2 cases (nos 23 and 26) with a rebound increment in growth velocity; there was no change in growth velocity in the other 4 (nos 27, 30, 31, 32) (table I). No alteration in the bone-age: chronological-age ratio,as assessed before and after 12 months’ treatment, was evident in the C-R children. Figs 1 and 2 show the pattern of height velocity before, during, and after clonidine treatment and after reinstitution of clonidine in individual C-R and C-NR children. Comparison of the pretreatment clinical indices in C-R and C-NR children (table I) revealed a significantly higher mean SDS and lower HV in the former than the latter (p < 0-05 and p<0-01, respectively). Mean bone age was lower in C-R than in C-NR children, but the difference did not reach statistical-significance. In the C-R children there was an inverse relation between pretreatment HV and the and a increase in HV at 12 months (r = 0-54, p<005) the and increase in relation between bone-age delay positive HV at 12 months (r = 0 60, p < 0-01). Discriminant analysis showed that a child could be classified with a confidence of 80% as C-R when his/her HV and SDS were 46 cm/yr and -2-5, respectively. Thus the probability of classifying as C-NR a child who may actually respond to clonidine is 20%. Growth-hormone Responses
to
Fig 1-Height velocities responders.
on
and off treatment in clonidine
Bars denote mean height velocity at the different time periods. For the sake of clarity SEs have been omitted. Numbers reported refer to individual case numbers.
23 children who were in pubertal stage Po, 4 (C-R nos 1, 10, 22 and C-NR no 27) had pubertal progression to stages P1 and P2. 4 children who were in pubertal stage P (C-R nos 14 and 21, C-NR nos 23 and 32) had pubertal progression to stages P2 and Ps. Of the 7 children who were in pubertal stage P2, 4 (C-R nos 2, 7, 16, C-NR no 31) had pubertal
progression to stages P3 and P4 (table 1). The frequency of pubertal changes during clonidine treatment was not significantly different between C-R and C-NR children 10r
GHRH and Clonidine
There were no significant differences between the two groups in the pretreatment biochemical indices (table n). In C-R children there was during treatment an initial rise in plasma GH (at 2 months) and SM-C levels (at 2 and 4 months). At 12 months there was a further increase in plasma GH but not in SM-C levels. Peak GH responses to GHRH and clonidine showed a similar trend: an initial increase (at 2 and 4 months) was followed by a slight decline at 6 months and then by another increase at 12 months. In C-NR children none of these indices was significantly altered during the course of clonidine treatment except for the peak GH response to clonidine at 2 and 12 months, which increased significantly (table II).
Pubertal Development
There was no significant difference in the frequency distribution of pubertal stages, as defined by Tanner et al,12 between C-R (Po = 15/22; Pl 2/22; P2 5/22) and C-NR =
=
(Po=8/12; Pl=2/12; P2=2/12) (X2=0.49 DF=2).
Of the
Fig 2--Height velocities
on
and off treatment in clonidine
non-
responders. Bars denote mean height velocity at the different time periods. For the sake of clarity SEs have been omitted. Numbers reported refer to individual case numbers.
1228 TABLE I-PATIENTS’ CLINICAL DETAILS AND RESPONSE IN HEIGHT VELOCITY TO CLONIDINE
(xz 9-09, DF 5). Determination of mean increase in HV in C-R children with and without pubertal progression gave values not statistically different (at 12 months, C-R without pubertal progression, 2-9 ± 0-3 cm/yr; C-R with pubertal progression, 3-8 ± 0-7 cm/yr). =
=
Side-effects None of the children had noticeable side-effects during 4 children presented with sedation and which diminished in 3, despite continued use of sleepiness, and discontinuation of the drug in 1. necessitated clonidine, 2 children had a skin rash and urticaria and had to be excluded from the study. Acute clonidine testing induced drowsiness or sleep and slight reduction (approximately 20 mm Hg) in systolic blood pressure both before and after chronic clonidine treatment. The results of routine blood tests remained normal in all patients during the course of clonidine treatment. treatment.
Discussion Clonidine is a potent GH secretagogue in both animals16,17 and man,18,19 and it acts via release of GHRH.5-7 Administration of GHRH evokes a GH response in most GH-deficient childrenl-4 and children with short stature,20 indicating a hypothalamic defect in GH release.
Our previously reported success with clonidine treatment in 4 children with CGD demonstrated that some children with short stature have a primary defect in the mechanism of GHRH release (but not synthesis).8 We have now shown that oral administration of clonidine to children with CGD promoted linear growth of more than 2 cm/yr in 21 of 34 children during 6 months and in 13 out of 22 children during 12 months of treatment. The effect of clonidine in children with CGD compares favourably with that of GH administered to children of short stature. 13,21 Withdrawal of clonidine after 12 months’ treatment resulted in deceleration of linear growth to pre-therapy or lower values in 8 C-R children, but the enhancing effect of clonidine on linear growth was maintained in the remaining 6 children. These findings are difficult to interpret; we suggest, however, that the children who continued to grow while off clonidine had less severe impairment of hypothalamic adrenergic neurotransmission than the remaining C-R children, in that pharmacological stimulation with clonidine was able to re-establish normal neurotransmitter function. This response is similar to the maintenance of increased height velocity seen in CGD children after withdrawal of anabolic steroids2z and fits in well with the transient impairment of GH secretion in some of these children.23 The children in our study are now off
1229 TABLE II-PATIENTS’ LABORATORY DETAILS DURING CLONIDINE TREATMENT
Results given as mean (SD). *p < 0-05 vs basal value (paired Student’s t-test). tp < 001 vs basal value (paired Student’s t-test). 34 patients were treated with clonidine for 6 months; 22 patients were treated for 12 months.
therapy so that the duration of the stimulant effect of the drug can be determined. Restarting therapy for 2-4 months triggered a new growth increment in 4 of 6 children in whom clonidine withdrawal had dramatically decreased height velocity. The acceleration of linear growth in response to clonidine associated with an increase in related biochemical indices. This pattern was more evident at 2 and 4 months than at 6 months. This does not necessarily contradict the finding that linear growth was still considerably accelerated at 6 and 12 months; it may simply indicate that the overall stimulant action of clonidine on GH secretion was still operative at a stage when single biological indices denoted a state of initial refractoriness. The finding that levels of some of these indices rose again at 12 months does not negate this view, since at that time clonidine had been withdrawn for 3-10 days. It is tempting to speculate that subtle changes in the sensitivity of hypothalamic &agr;2-adrenoceptors acted upon by clonidine24 may be responsible for this behaviour. Were this the case, further work would be needed to establish the optimum dose, timing, and duration of clonidine treatment in children with CGD. The 12 C-NR children represent a heterogeneous group, since at the completion of 6 months’ therapy height velocity had decreased in 5, increased slightly in 2, and was maintained in 5. This pattern was in essence present at 12 months, when 7 of these children could be re-evaluated. Deceleration of height velocity with clonidine was a transient event in one case (no 23); in another (no 27) height velocity was still suppressed 6 months after clonidine withdrawal. In the C-NR children there was no significant change in growth-related biochemical indices. Some clinical features appear useful in identifying the children with CGD who will respond to clonidine; the responders had a clearly higher height SDS and lower height velocity and a greater (though not significant) delay in bone age than did non-responders. Similar observations have been made in short-stature children given hGH replacement therapy.13 In our study discriminant analysis showed that values of height velocity and height SDS of 4-6 cm/year and —25 cm/year, respectively, would predict responsiveness to clonidine in 80% of the children. There was no difference in the frequency distribution of pubertal stages between C-R and C-NR children or in the was
number of C-R and C-NR children undergoing pubertal progression while taking clonidine. Moreover, the increment in height velocity in C-R children was of similar extent irrespective of whether they had undergone pubertal progression. Overall these findings suggest that the acceleration of growth observed in the C-R children is not due to underlying spontaneous or clonidine-induced pubertal changes. Other neuroactive compounds have also been used in the treatment of short-stature children. Huseman and co-workers,25.26 who reported stimulation of linear growth in some hypopituitary children treated long term with levodopa or bromocriptine, ascribed GH deficiency and short stature in these subjects to hypothalamic dopaminergic dysfunction. However, in their studies the greater effectiveness of levodopa (the precursor of both dopamine and noradrenaline27) than bromocriptine (a specific dopamine agonist28) makes impairment of noradrenergic function more likely. Our findings with clonidine support this proposition. Whichever brain neurotransmitter is primarily involved, it seems very likely that many children with growth disorders have a primary neurosecretory dysfunction, and this makes the therapeutic use of neuroactive drugs sound. Administration of clonidine for 12 months stimulated linear growth in 13 of 22 children with CGD without eliciting noticeable side-effects. Responsiveness to clonidine was positively related to initial height SDS and inversely related to growth velocity. Withdrawal of clonidine showed in some children persistence of the stimulatory effect of the drug on height velocity, perhaps indicating that short-term clonidine treatment is valid in such cases. Finally, in a few children reinstitution of clonidine after withdrawal again stimulated growth. A longer follow-up study is warranted to determine the beneficial effect of this therapy on adult
height. Clearly, clonidine is potentially useful for the treatment of children with CGD, who represent about 25% of all children and adolescents.29 It also has the advantages of low cost, lack of effect on bone age, and suitability for oral administration.
Addendum Since submission of this manuscript a report has been published showing that clonidine administered for 12 months to 16 prepubertal children with CGD significantly increased GH levels, plasma somatomedin-C, and linear growth. 30 We thank Dr G. Reina for statistical analysis of the data and Miss Maria Lupo for secretarial assistance. Correspondence should be addressed to E. E. M., Department of Pharmacology, University of Milan, Via Vanvitelli 32, 20129 Milan, Italy. REFERENCES
MO, Cronin MJ. Growth hormone releasing factor: Clinical and basic studies. In: Müller EE, MacLeod RM, eds. Neuroendocrine perspective, vol 4. Amsterdam: Elsevier, 1985: 95-144. Grossman A, Savage MO, Lytras N, et al. Responses to analogues of growthhormone-releasing hormone in normal subjects, and in growth-hormone deficient children and young adults. Clin Endocrinol 1984; 21: 253-56. Pintor C, Fanni V, Loche S, et al. Synthetic hpGRF1-40 stimulates growth hormone and inhibits prolactin secretion in normal children and children with isolated growth hormone deficiency. Peptides 1983; 4: 929-33. Evain-Brion D. GRF European Multicenter Study. GH response to a single iv injection of synthetic 1-44 GRF in prepubertal children with growing failure. In: Müller EE, MacLeod RM, eds. Neuroendocrine perspective, vol 5. Amsterdam: Elsevier, 1986: 101-09. Eden S, Eriksson E, Martin JB, et al. Evidence for a growth hormone releasing factor mediating alpha-adrenergic influence on growth hormone secretion in the rat. Neuroendocrinology 1981; 33: 24-27.
1. Thorner
2.
3.
4.
5.
1230
POLYMORPHIC DNA MARKER ON X CHROMOSOME AND MANIC DEPRESSION
JULIEN MENDLEWICZ1
PHILIPPE SIMON2 SERGE SEVY1 FRANÇOIS CHARON1 HUGUETTE BROCAS2 SYLVIANE LEGROS1 GILBERT VASSART2,3
Department of Psychiatry,1 Interdisciplinary Research Institute,2 and Department of Clinical Chemistry,3 Free University of Brussels, Erasme Campus, Belgium
Heredity is vulnerability
important factor in manic depression. A genetic linkage has been demonstrated between manic depression and coagulation factor IX at Xq27 with a TaqI polymorphism at the F9 locus in DNA samples from peripheral leucocytes of manic depressive probands and relatives in 10 informative families. Statistical analysis of the pedigrees gave a maximum lod score of 3·10 at a recombination fraction of 0·11, demonstrating a linkage between a manic depressive locus and the F9 locus in the Summary
an
to
Xq27 region. Introduction HEREDITY is
important contributing factor to susceptibility manic-depressive illness (MDI). However, family, twin, and adoption studies have not elucidated the mode of transmission.1-4 Linkage analysis has rarely been applied to psychiatric conditions. Linkage studies with colour blindness and glucose-6-phosphate dehydrogenase (G6PD) deficiency as X-chromosomal markers have shown an
to
that
dominant X-linked gene may be involved in the genetic transmission of a subtype of MD 1. 1,5-12 The maximum lod reported for the MDI-G6PD linkage was 4 21 at a recombination fraction (8) of 0.06.10 However, apparent father-to-son transmission of the illness has been described in family studies,2,13 and data which do not support a linkage between colour blindness and bipolar illness have also been reported.14 Genetic heterogeneity, which has been postulated for manic depression,11,15,16 could a
6. Katakami H, Kato Y, Matsusnita N, et al. Effects of neonatal treatment with monosodium glutamate on growth hormone release induced by clonidine and prostaglandin E1 in conscious male rats. Neuroendocrinology 1984; 38: 1-5. 7. Miki N, Ono M, Shizume K. Evidence that opiatergic and &agr;-adrenergic mechanisms stimulate rat growth hormone release via growth-hormone-releasing factor.
Endocrinology 1984; 114: 1950-52. C, Cella SG, Corda R, et al. Clonidine accelerates growth in children with impaired growth hormone secretion. Lancet 1985; i: 1482-84. 9. Frazier SD. Abnormalities of growth. In: Collu R, Ducharme JR, Guyda H, eds. Pediatric endocrinology. New York: Raven Press, 1981: 167-202. 10. Solina G. Modalità di evoluzione di alcuni parametri antropometrici. Rassegna Medica Sarda 1981; 4: 427-35. 11. Pintor C, Puggioni R, Fanni V, et al. Growth-hormone releasing factor and clonidine in children with constitutional growth delay. Evidence for defective pituitary growth hormone reserve. J Endocrinol Invest 1984; 7: 253-56. 12. Tanner JM, Whitehouse RH, Hughes PCR, et al. Effect of human growth hormone treatment for 1 to 7 years on growth of 100 children. Arch Dis Child 1971; 46: 8. Pintor
745-82. 13. Van Vliet G, Styne DM, Kaplan SL, et al. Growth hormone treatment for short stature. N Engl J Med 1983; 309: 1016-22. 14. Ross JM, Tsagarakis S, Grossman A, et al. Treatment of growth-hormone deficiency with growth-hormone-release hormone. Lancet 1987; i: 5-8. 15. Cooley WW, Lohnes PR. Multivariate data analysis. New York: John Wiley, 1971. 16. Durand D, Martin JB, Brazeau P. Evidence for a role of &agr;-adrenergic mechanisms in regulation of episodic growth hormone secretion in the rat. Endocrinology 1977; 100: 722-28. 17. Lovinger R, Holland J, Kaplan M, et al. Pharmacological evidence for stimulation of growth hormone secretion by a central noradrenergic system in dogs. Neuroscience 1976; 1: 443-50. 18. Lal S, Tolis G, Martin JB, et al Effect of clonidine on growth hormone, prolactin, leuteinizing hormone, follicle-stimulating hormone, and thyroid-stimulating hormone in the serum of normal man. J Clin Endocrinol Metab 1975; 41: 827-32.
for these apparently contradictory findings. According to this model, only a subgroup of bipolar pedigrees will show close linkage to the X-chromosome and thus carry the X-linked gene, but not all bipolar illness can be X-linked. The increasing number of restriction fragment account
length polymorphisms (RFLP) permits linkage analysis with selected DNA probes. We provide here the first evidence for a genetic linkage between MDI and coagulation factor IX at Xq27. Patients and Methods Previous linkage studies have pointed to the subterminal region of the long arm of the X-chromosome as a possible site for the MDI 1ocus.1,5—12 Therefore a factor IX probe,17 known to hybridise to the Xq27 band,18 was used to test the hypothesis that the disease trait and a common RFLP at the F9 locus18,19 would co-segregate in informative pedigrees. The pedigrees of all consecutive admissions of unipolar or bipolar probands between April, 1983, and May, 1986, to the department of psychiatry, Erasme Hospital, were screened for possible linkage analysis. The patients were referred by private physicians, and from outpatient clinics and state and private
hospitals. Pedigree data
were obtained from the proband and available relatives at the time of admission. Personal interview with all available relatives after informed consent has proved more reliable than collecting family data from the proband only.20 Medical and social records about probands and relatives were used when available. All probands, spouses, and available relatives were separately examined by two investigators who used the schedule for affective disorders and schizophrenia21 according to the research diagnostic criteria.22 The family members were assessed blind with respect to the proband’s diagnosis. Disagreements were referred to the senior investigator (J.M.), who determined blind the final diagnoses. All diagnoses were made independently from knowledge of DNA marker analysis. Those subjects diagnosed as bipolar, unipolar, or cyclothymic (ie, primary affective disorders23) were considered affectively ill. Subjects with secondary depression or alcoholism were considered "well". We have previously shown that within a family unit containing an affectively ill proband, bipolar illness, unipolar illness, and cyclothymia are genetically related
phenotypes expressing the same genotype.4-6 Whole blood samples were collected by venepuncture into tubes containing EDTA, and DNA was extracted from leucocytes as described.’ 10 ttg DNA were cleaved by TaqI endonuclease.
19. Gil-Ad I, Topper E, Laron Z. Oral clonidine as a growth-hormone stimulation test. Lancet 1979; ii: 278-80. 20. Gelato MC, Malazowski S, Caruso-Nicoletti M, et al. Growth hormone (GH) responses to GH-releasing hormone during pubertal development in normal boys and girls: comparison to idiopathic short stature and GH deficiency. J Clin Endocrinol Metab 1986; 63: 174-79. 21. Gertner JM, Genel M, Gianfredi SP, et al. Prospective clinical trial of human growth hormone in short children without growth hormone deficiency. J Pediatr 1984; 104: 172-76. 22. Bettmann HK, Goldman HS, Abramowicz M, et al. Oxandrolone treatment of short stature: effect on predicted mature height. J Pediatr 1971; 79: 10-23. 23. Gourmelen M, Phan-Huu-Trung T, Girard F. Transient partial hGH deficiency in prepubertal children with delay of growth. Pediatr Res 1979; 13: 221-24. 24. Maura G, Bonanno G, Raiteri M. Chronic clonidine induces functional downregulation of presynaptic &agr;2-adrenoceptors regulating 3H-noradrenaline and 3H-5-hydroxytryptamine release in the rat brain. Eur J Pharmacol 1985; 112: 105-10. 25. Huseman CA, Hassing JM. Evidence of dopaminergic stimulation of growth velocity in some hypopituitary children. J Clin Endocrinol Metab 1984; 58: 419-25. 26. Huseman CA, Hassing JM, Sibilia MG. Endogenous dopaminergic dysfunction: a novel form of human growth hormone deficiency and short stature. J Clin Endocrinol Metab 1986; 62: 484-89 27. Andèn NE, Dahlstrôm A, Fuxe K, et al. Functional role of the nigro-neostriatal dopamine neurons. Acta Pharmacol Toxicol 1966; 24: 263-74. 28. Fluckiger E, Markstein R. Receptor pharmacology of ergot compounds. In: Tolis G, Stefanis C, Mountokalakis T, et al, eds. Prolactin and prolactinomas. New York: Raven Press, 1983:105-13. 29. Bierich JR, Enders H, Heinrich U, et al. Stunted growth with more or less normal appearance. Eur J Pediatr 1982; 139: 214-38. 30. Castro-Magana M, Angulo M, Fuentes B, et al. Effect of prolonged clonidine administration on growth hormone concentrations and rate of linear growth in children with constitutional growth delay. J Pediatr 1986; 109: 784-87.