Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible...

7
Clinical Endocrinology (1996) 44, 353–359 Apparent mineralocorticoid excess due to 11 -hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation S. Kitanaka, A. Tanae and I. Hibi Division of Endocrinology and Metabolism, National Children’s Hospital, Tokyo, Japan (Received 24 May 1995; returned for revision 29 September 1995; finally revised 12 October 1995; accepted 10 November 1995) Summary A Japanese boy with apparent mineralocorticoid excess (AME) is described. He was born with intra- uterine growth retardation (IUGR) and elevated serum level of creatine phosphokinase (CPK). He was studied at 2 years of age because of polyurea and polydipsia of one year’s duration and was found to have hypokalaemic alkalosis and sustained hyper- tension. His plasma renin activity and aldosterone levels were always low and his ratio of urinary tetra- hydrocortisol plus allo-tetrahydrocortisol to that of tetrahydrocortisone was very high. Therefore, AME due to 11 -hydroxysteroid dehydrogenase (11 -HSD) deficiency was diagnosed. He was successfully treated with a combination of spironolactone and nifedipine for at least 16 months. His blood pressure, plasma pH and serum potassium levels were normal- ized by this treatment, but serum CPK level remained high. We researched the birth records of previously reported AME cases and found that IUGR is a char- acteristic feature of AME. The mechanism by which IUGR occurs in AME is discussed and we speculate that 11 -HSD might be deficient in the placenta and/or fetal tissues, as well as in the kidney, in AME. An explanation for the elevated CPK could not be found. Apparent mineralocorticoid excess (AME) is a rare disease first described by Ulick et al. (1979) and characterized by malignant hypertension of early onset combined with hypokalaemic alkalosis. Cortisol binds to mineralocorticoid receptors with the same affinity as aldosterone, while cortisone does not. In AME, because of the inability to convert cortisol to inactive cortisone due to a congenital deficiency of 11 -hydroxysteroid dehydrogenase (11 -HSD) in kidney, cortisol binds to miner- alocorticoid receptors in the renal tubules and exerts excessive mineralocorticoid action despite the low levels of aldosterone and other mineralocorticoids. We report a Japanese boy with typical AME who was born with intrauterine growth retardation (IUGR) and showed signs and symptoms suggesting AME since the fetal period. The possible influence of 11 -HSD deficiency on fetal growth is discussed. Case report A 2-year-old Japanese boy, the second child of unrelated parents, was referred because of polyuria and polydipsia. He was delivered by Caesarean section because of fetal distress at 39 gestational weeks and IUGR was noted (weight 2532 g, 1 . 8 SD; length 44 . 5 cm, 2 . 6 SD). Since he had respiratory distress due to meconium aspiration syndrome and pneu- mothorax, he underwent artificial ventilation during the first week. At that time the attending neonatologist found that he had an unexplained metabolic alkalosis (pH 7 . 50 7 68, base excess 11 12 mmol/l) and hypokalaemia (2 7 4 3 mmol/ l), which necessitated potassium supplementation. He also had a high serum creatine phosphokinase (CPK) level (1000 2700 U/l, MM isozyme dominant) soon after birth and transient facial nerve palsy and slight hypertonicity were also noted. Blood pressure was normal and he showed gradual normalization of the hypokalaemic alkalosis by 2 months after birth. He was followed by a neurologist and his CPK level remained high despite the disappearance of hypertonicity. Psychomotor development was normal. Hypokalaemia (2 . 8 mmol/l) was detected at 11 months of age on routine investigation and polyuria and polydipsia were found at the same time. On admission, his blood pressure was very high and ranged from 160/90 mmHg at rest to 190/110 mmHg while awake. Body height and weight were 79 . 5 cm ( 1 . 9 SD) and 10 . 1 kg ( 1 . 3 SD) respectively. His general condition was good with normal muscle tonicity and strength. Blood examination revealed sustained hypokalaemia (2 1 2 6 mmol/l), metabolic alkalosis (pH 7 . 52, base excess 10 . 5 mmol/l), high CPK level (330 500 U/l, BB, 1%; MB, 10%; MM, 84%), which Case report 353 1996 Blackwell Science Ltd Correspondence: Dr Sachiko Kitanaka, Division of Endocrinology and Metabolism, National Children’s Hospital, 3-35-31 Taishido, Setagaya-ku, Tokyo 154, Japan.

Transcript of Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible...

Page 1: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

Clinical Endocrinology (1996) 44, 353–359

Apparent mineralocorticoid excess due to11¯-hydroxysteroid dehydrogenase deficiency:a possible cause of intrauterine growth retardation

S. Kitanaka, A. Tanae and I. HibiDivision of Endocrinology and Metabolism, NationalChildren’s Hospital, Tokyo, Japan

(Received 24 May 1995; returned for revision 29 September1995; finally revised 12 October 1995; accepted 10 November1995)

Summary

A Japanese boy with apparent mineralocorticoidexcess (AME) is described. He was born with intra-uterine growth retardation (IUGR) and elevatedserum level of creatine phosphokinase (CPK). Hewas studied at 2 years of age because of polyureaand polydipsia of one year’s duration and was foundto have hypokalaemic alkalosis and sustained hyper-tension. His plasma renin activity and aldosteronelevels were always low and his ratio of urinary tetra-hydrocortisol plus allo-tetrahydrocortisol to that oftetrahydrocortisone was very high. Therefore, AMEdue to 11 ¯-hydroxysteroid dehydrogenase (11 ¯-HSD)deficiency was diagnosed. He was successfullytreated with a combination of spironolactone andnifedipine for at least 16 months. His blood pressure,plasma pH and serum potassium levels were normal-ized by this treatment, but serum CPK level remainedhigh.

We researched the birth records of previouslyreported AME cases and found that IUGR is a char-acteristic feature of AME. The mechanism by whichIUGR occurs in AME is discussed and we speculatethat 11¯-HSD might be deficient in the placenta and/orfetal tissues, as well as in the kidney, in AME. Anexplanation for the elevated CPK could not be found.

Apparent mineralocorticoid excess (AME) is a rare disease firstdescribed by Ulicket al.(1979) and characterized by malignanthypertension of early onset combined with hypokalaemicalkalosis. Cortisol binds to mineralocorticoid receptors with

the same affinity as aldosterone, while cortisone does not. InAME, because of the inability to convert cortisol to inactivecortisone due to a congenital deficiency of 11�-hydroxysteroiddehydrogenase (11�-HSD) in kidney, cortisol binds to miner-alocorticoid receptors in the renal tubules and exerts excessivemineralocorticoid action despite the low levels of aldosteroneand other mineralocorticoids. We report a Japanese boy withtypical AME who was born with intrauterine growth retardation(IUGR) and showed signs and symptoms suggesting AMEsince the fetal period. The possible influence of 11�-HSDdeficiency on fetal growth is discussed.

Case report

A 2-year-old Japanese boy, the second child of unrelatedparents, was referred because of polyuria and polydipsia. Hewas delivered by Caesarean section because of fetal distress at39 gestational weeks and IUGR was noted (weight 2532 g,ÿ1.8 SD; length 44.5 cm,ÿ2.6 SD). Since he had respiratorydistress due to meconium aspiration syndrome and pneu-mothorax, he underwent artificial ventilation during the firstweek. At that time the attending neonatologist found that he hadan unexplained metabolic alkalosis (pH 7.50� 7:68, baseexcess 11� 12 mmol/l) and hypokalaemia (2:7 � 4:3 mmol/l), which necessitated potassium supplementation. He also hada high serum creatine phosphokinase (CPK) level(1000� 2700 U/l, MM isozyme dominant) soon after birthand transient facial nerve palsy and slight hypertonicity werealso noted. Blood pressure was normal and he showed gradualnormalization of the hypokalaemic alkalosis by 2 months afterbirth. He was followed by a neurologist and his CPK levelremained high despite the disappearance of hypertonicity.Psychomotor development was normal. Hypokalaemia(2.8 mmol/l) was detected at 11 months of age on routineinvestigation and polyuria and polydipsia were found at thesame time.

On admission, his blood pressure was very high and rangedfrom 160/90 mmHg at rest to 190/110 mmHg while awake.Body height and weight were 79.5 cm (ÿ1.9 SD) and 10.1 kg(ÿ1.3 SD) respectively. His general condition was good withnormal muscle tonicity and strength. Blood examinationrevealed sustained hypokalaemia (2:1 � 2:6 mmol/l), metabolicalkalosis (pH 7.52, base excess 10.5 mmol/l), high CPK level(330� 500 U/l, BB, 1%; MB, 10%; MM, 84%), which

Case report

353# 1996 Blackwell Science Ltd

Correspondence: Dr Sachiko Kitanaka, Division of Endocrinologyand Metabolism, National Children’s Hospital, 3-35-31 Taishido,Setagaya-ku, Tokyo 154, Japan.

Page 2: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

increased to 4133 U/l during a febrile episode, combined with ahigh aldolase level (22.5 U/l). Serum thyroid hormones,catecholamines, and renal function were normal. Ophthalmo-scopy revealed mild hypertensive change. Chest radiographshowed minimal cardiomegaly and echocardiography showedmild hypertrophy of the ventricular wall, but cardiac functionassessed by echocardiography was normal. His electromyo-gram and nerve conduction velocities were normal. His boneage was 2.4 years assessed by Tanner–Whitehouse 2 method.His father and mother were 156 and 152 cm in height,respectively. None of his family members had hypertensionfrom a young age.

Endocrine studies (Table 1)

Plasma renin activity was always low and did not respond toeither low sodium restriction or postural stimulation. Plasmaaldosterone level was also low, while the cortisol level wasnormal to high, and ACTH remained within the normal range.Plasma aldosterone showed poor responses to ACTH andcorticotrophin-releasing factor (CRF) stimulation, while deox-ycorticosterone, 11-deoxycortisol and corticosterone respondednormally to ACTH. Urinary aldosterone and serum cortisone(E) were undetectable.

On urinary steroid profile assessment by the gas chromato-graphy–mass spectrometry–selected ion monitoring method,tetrahydrocortisone (THE) excretion was extremely low, whiletetrahydrocortisol (THF) and alloTHF (5�THF) excretionswere within normal ranges. The (THF+5�THF)/THE ratio was43.7 and 17.4 on two determinations, which were markedly

higher than normal (Table 2). This abnormally high ratio wascompatible with a diagnosis of AME. His parents and his sisterunderwent the same urine analysis and a normal ratio of(THF+5�THF)/THE was found in all three.

Treatment (Fig. 1)

Spironolactone.The patient was initially treated with spirono-lactone, an aldosterone antagonist, which is reported to be mosteffective. The dose was gradually increased from 25 to 60 mg/day, and the serum potassium level normalized. However, tonormalize his blood pressure, a very high dose of 150 mg/day(15 mg/kg/day) was required, which caused elevation ofserum urea nitrogen, uric acid, and total cholesterol levels.Consequently, the dose of spironolactone was reduced to100 mg/day.

Dexamethasone.In order to normalize his blood pressure,dexamethasone was added to spironolactone treatment at dosesof 0.25, 0.5 and 1.0 mg. The final dose slightly reduced bloodpressure, but it was not suitable for long-term use because of itsside-effects. It should be noted that when the patient’s owncortisol secretion was suppressed by dexamethasone, plasmarenin activity and aldosterone levels increased.

Calcium antagonist vasodilator.For long-term control of bloodpressure, nifedipine was added to spironolactone. The maximaldose of 10 mg/day (1.0 mg/kg/day) was very effective, andreduced his blood pressure to normal.

354 S. Kitanaka et al.

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359

Table 1 Hormonal data

Daily profile rapid ACTH test CRF testBasal reference

Plasma/serum level (mean� SD) 600 h 1200 h 1800 h 2400 h 00 300 600 900 00 150 300 600

PRA (�g/l/h) 2:11� 1:14* 0.2 0.2 0.2 0.3ACTH (pmol/l) 4:05� 1:98* 1.2 1.9 1.3 5.0 5.3 15.0 14.4 8.0Cortisol (nmol/l) 337� 160* 679 317 309 834 701 894 933 867 635 665 742 781Aldosterone (pmol/l) 393� 209* 135 139 97 154 99 124 89 88 48 43 38 28DOC (pmol/l) 347� 114 152 60611-Deoxycortisol (nmol/l) 2:84� 0:45 1.8 4.7Corticosterone (nmol/l) 9:39� 2:14 11.4 70.2

Urine Reference range Patient

Aldosterone (nmol/24 h) 14–53 <3.6Cortisol (nmol/24 h) 55–250 58� 144

CRF, Corticotrophin releasing factor; PRA, plasma renin activity; DOC, deoxycorticosterone.Rapid ACTH test: ACTH 0.25 mg/m2 i.v.; CRF test: CRF 1.5�g/kg i.v.* Reference levels for normal age-matched children. Others are for normal adults.

Page 3: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

11¯-HSD deficiency and intrauterine growth retardation 355

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359

Fig. 1 Clinical course. a, The dotted lines in blood pressure indicate 90th percentile levels of systolic and diastolic blood pressure for age.b, Serum potassium level normalized with spironolactone 60 mg/day alone. However, a combination of nifedipine 10 mg/day and spironolactone100 mg/day was required to normalize his blood pressure. Plasma renin activity and aldosterone levels were always low but increased afterdexamethasone treatment. c, BUN, Blood urea nitrogen; d, PRA, plasma renin activity; e, P. aldo, plasma aldosterone.

Table 2 Urinary steroid profile

PatientNormal (1� 2y, n � 7) Sister

Cortisol metabolites (mg/gCr) day 1 day 2 mean (range) Father Mother (4y2m)

Tetrahydrocortisone (THE) 0.13 0.27 4.75 (2:00� 7:37) 1.33 1.43 1.81Allo-tetrahydrocortisone (5�THE) <0.03 0.05 0.63 (0:17� 0:93) 0.14 0.17 0.23Tetrahydrocortisol (THF) 1.59 0.97 2.00 (1:22� 2:81) 1.16 0.97 0.87Allo-tetrahydrocortisol (5�THF) 4.09 3.73 5.13 (0:69� 10:76) 1.65 1.96 2.0320�-Cortolone 0.02 0.02 1.35 (0:48� 2:00) 0.44 0.52 0.6120�-Cortolone 0.02 0.03 0.99 (0:44� 1:81) 0.34 0.36 0.3720�-Cortol 0.25 0.22 0.49 (0:25� 0:73) 0.19 0.26 0.2220�-Cortol 0.54 0.65 0.72 (0:25� 1:81) 0.70 0.65 0.25Cortisone <0.31 <0.08 (< 0:11� 0:49) 0.10 0.12 <0.19Cortisol <0.31 0.21 (< 0:11� 0:24) 0.05 <0.04 <0.19

(THF+5�THF)/THE ratio 43.7 17.4 1.56 (1:05� 2:85) 2.11 2.05 1.60

Page 4: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

Low sodium diet.There seems to be mild improvement of bloodpressure with low sodium diet, and his sodium intake wasrestricted to 1 g/day.

Finally, a combination of nifedipine 10 mg/day and spir-onolactone 100 mg/day, together with a low sodium diet,improved his hypertension and hypokalaemia without anysignificant side-effects.

Follow-up study

His blood pressure was maintained at a normal level for 16months with this combination treatment. Polyuria and poly-dipsia were ameliorated soon after normalization of the serumpotassium level. His height velocity was not influenced by thistreatment and his height remained aroundÿ2 SD, which may beappropriate judging from the heights of his parents.

IUGR among reported cases of AME

We surveyed case reports of AME for their gestational age,weight and length at birth. When such data were not described,we sent inquiries to the authors and made efforts to obtain thesedata. The results of this study are presented in Table 3. Birthweight was obtained for 18 AME patients, including the presentcase, and was below 2700 g in all except one, who was diag-nosed with AME at 21 years of age. SD score for birth weightcould be analysed in 5 cases and was underÿ1.1 in all five. SDscore for birth length could be analysed in only two casesincluding ours (ÿ2.9 SD andÿ2.6 SD) and was much lower thanthe SD score for birth weight in each patient, which suggestedthat the negative effects of AME on fetal growth may be morepredominant on height growth. P. M. Stewart mentioned that histhree cases of AME had IUGR with catch-up growth when serumpotassium was corrected (personal communication).

356 S. Kitanaka et al.

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359

Table 3 Neonatal and post-natal growth of AME: list of reported cases

Age at Growth retardation Gestational Birth Birthinvestigation at investigation weeks weight (g) height (cm) Reference

1 3y + 1870 Ulicket al. 19792 9y + 2000 Ulicket al. 19793 2y + Shackletonet al. 19804 3y + 41 2500 (ÿ1.8 SD)* Werderet al. 19745 5y9m Winter & McKenzie 19776 2y9m Shackletonet al. 19857 3y3m Phillipou 1978 (unpublished)8 1y9m 38 2170 (ÿ2.0 SD)* 43 (ÿ2.9 SD)* Fiselieret al. 19829 19y + Harincket al. 198410 5m 37 2360 (ÿ1.1 SD)* Honouret al. 198311 11y + Shackletonet al. 198512 3y9m + Shackletonet al. 198513 6y Shackletonet al. 198514 9y4m + full term 2100 Monderet al. 198615 4y ÿ full term 2600 Monderet al. 198616 8y + 36 2000 (ÿ1.4 SD)* Monderet al. 198617gy 21y ÿ normal normal Stewartet al. 198818 7y + Batistaet al. 198619 3y Peskovitz 1986 (unpublished)20

gy9y + Winter 1988 (unpublished)

21 3y2m ÿ full term 2330 Igarashiet al. 197922 2y0m ÿ 39 2532 (ÿ1.8 SD)* 44.5 (ÿ2.6 SD)* Our case 199623 2270 New (by personal communication)24 2640 New (by personal communication)25 2000 New (by personal communication)26 1960 New (by personal communication)27 neonatal growth retardation��� Stewart (by personal communication)28 neonatal growth retardation��� Stewart (by personal communication)29 neonatal growth retardation��� Stewart (by personal communication)

* SD score of birth weight and height for gestational week according to Japanese mean and standard deviation.ySiblings.

Page 5: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

Discussion

Ulick et al. (1979) first described AME in two children withhypertension, hypokalaemia and alkalosis, suppressed plasmarenin activity and aldosterone levels. The pattern of urinarysteroids excretion indicated impaired conversion of cortisol tocortisone suggesting a deficiency of 11�-hydroxysteroidhydrogenase (11�-HSD).

Mineralocorticoid receptor (type I receptor) is non-selectiveand does not distinguish cortisol, corticosterone, and aldo-sterone in anin vitro analysis (Krozowski & Funder, 1993;Edwardset al., 1988). The mechanism by which mineralo-corticoid receptor selects aldosteronein vivo despite the factthat cortisol is present in the circulation at 100–1000 times theconcentration of aldosterone, is explained by 11�-HSDconverting cortisol to inactive cortisone. In the absence of11�-HSD activity, cortisol can access the mineralocorticoidreceptor and act as a mineralocorticoid. Deficiency of 11�-HSDactivity may occur congenitally as AME or may be induced as aresult of pharmacological inhibition by liquorice intoxication.AME differs from Liddle syndrome in that it responds tospironolactone and in abnormal secretion of urinary steroids. Inrare instances, the reduced metabolic clearance of cortisol inAME may be due not primarily to 11�-HSD deficiency but to adefect in ring-A reduction which is described as a type 2 variant(Ulick et al., 1990; 1992).

The human 11�-HSD cDNA was first cloned by Tanninet al.(1991) by screening a human testicular cDNA library with a rat11�-HSD cDNA obtained from liver microsomes. However,subsequent analyses of the gene in four AME patients werenormal (Nikkila et al., 1993). 11�-HSD is present in mosttissues and 11�-HSD-mediated control of local steroid levels iscrucial for both homeostasis and survival and it seems to haveorgan-specific functions (Monder, 1991). Stewartet al. (1994)found that human kidney 11�-HSD differed from the clonedtype 1 isoform. In contrast to the low-affinity, nicotinamideadenine dinucleotide phosphate(NADP)-dependent type 1enzyme, this type 2 isoform was high-affinity and NAD-dependent. Type 1 11�-HSD cDNA was recently cloned(Albiston et al., 1994) and a point mutation in this gene hassince been reported in an AME patient (Wilsonet al., 1995).Genetic analysis of our present case is now under way.

To our knowledge, there have been no more than 30 reportedcases of AME due to congenital 11�-HSD deficiency(Shackleton & Stewart, 1990). There are two sibling casessuggesting the hereditary nature of this disease, but urinarysteroid analysis of both parents of the affected siblings werenormal even after ACTH stimulation (Shcakletonet al., 1995).Urinary steroid analyses of both parents and a sister of ourpatient were also normal. AME cases are found among all racesand there is one Japanese case that was reported prior to the first

description of AME as dexamethasone responsive mineralo-corticoid excess and was proven to be AME on furtherinvestigation (Igarashiet al., 1979). This case presented withpolyuria and polydipsia at 9 months and paralysis of theextremities at 2 years 6 months. Spironolactone was effective,but she was successfully treated with dexamethasone alone. Shewent through normal puberty and normal pregnancy and isalive.

Typical symptoms of AME include failure to thrive, polyuriaand polydipsia due to hypokalaemia-induced nephrogenicdiabetes insipidus. The onset of these symptoms is usuallyfrom late in the first year to the third year. The youngest andmost severe case was a boy presenting at 6 weeks withsubarachnoid haemorrhage who died at 5 months (Honouretal., 1983), and there was also a case in which the patient hadpolydipsia, polyuria and failure to thrive from birth (Shackletonet al., 1985). As in most cases, our case presented withpolydipsia and polyuria at one year of age, although he hadhypokalaemia and alkalosis during his neonatal period. There isanother reported case in which hypokalaemia and alkalosisremained even after surgery for pyloric stenosis at 3 weeks andwas recognized to be hypertensive at 2 years 9 months(Shackletonet al., 1985). In normal infants younger than 3months, steroids containing an 11-carbonyl group (cortisone)are dominant, so there might have been some influence from thelack of relatively high 11�-HSD activity during the neonatalperiod. Although the main symptoms present later in life, AMEmight have already been present at birth.

Frequent observation of IUGR among AME patients furthersupports prenatal onset of AME. IUGR may not be due tointrauterine hypokalaemia because hypokalaemia of the fetusmight be corrected by simple diffusion through the placenta,and also because patients with neonatal Bartter’s syndrome arenot always associated with IUGR. Placental 11�-HSD is knownto protect the fetus from exposure to excessively highconcentration of cortisol in maternal blood (5� 10 timeshigher than that in the fetus) by converting cortisol to inactivecortisone (Murphyet al., 1974; Beitinset al., 1973). Prenatalexposure of the fetus to glucocorticoids retards intrauterinegrowth (Reinischet al., 1978). In rats, placental 11�-HSDactivity correlated positively with birth weight (Benediktssonetal., 1993). The same positive correlation between placental11�-HSD and birth weight in humans has been recentlyreported (Stewartet al., 1995). Furthermore, human placental11�-HSD was found to be a NAD-dependent isoform (type 2)identical with human renal 11�-HSD (Brownet al., 1993). Inaddition, this type 2 11�-HSD isoform activity is ubiquitouslydistributed in a variety of human mid-gestational fetal tissues(Stewartet al., 1994; 1995). We therefore strongly suspect thatnot only renal but also placental and/or fetal 11�-HSD activityis deficient in AME patients and that their deficiency is the

11¯-HSD deficiency and intrauterine growth retardation 357

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359

Page 6: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

cause of intrauterine growth retardation. Although it has beenreported that excessive fetal exposure to glucocorticoids mayincrease the incidence of stillbirth and teratogenicity (Warrell& Taylor, 1968; Bongiovanni & McPadden, 1960), nomalformations have been reported in AME cases. Conse-quently, we propose that neonates with intrauterine growthretardation accompanied by idiopathic hypokalaemia andalkalosis need to be followed carefully with monitoring ofblood pressure, taking AME into consideration.

There were some interesting findings in the adrenal hormonedata in our case. First, the basal aldosterone level was low anddid not respond either to the rapid ACTH test, as reported inother cases (Fieselieret al., 1982; Batistaet al., 1986), or toCRF test before treatment. These data suggest that the excessmineralocorticoid action suppressed aldosterone secretion tosuch a degree that ACTH stimulation could not elicit theresponse. Moreover, even after treatment when the serumpotassium level and blood pressure normalized, aldosterone didnot display a normal response to ACTH stimulation (baseline72.3 pmol/l; 60 minutes 80.6 pmol/l). It was strongly suspectedthat even after the symptoms of mineralocorticoid excess wereameliorated, subclinical mineralocorticoid excess state mightremain. Secondly, when we used dexamathasone, whichsuppressed the patient’s own cortisol secretion, his plasmarenin activity and aldosterone rose to normal levels. Thissuggests that the feedback system of the renin–angiotensin–aldosterone axis was preserved. Serum and urinary cortisolmeasurements indicated that dexamethasone suppressed thecortisol level by no more than 80% of the normal level. Thisimplies, in consideration of the fact that cortisol is present in thecirculation at 100–1000 times the concentration of aldosterone,that not all cortisol acts as mineralocorticoid in AME. Thedefect in 11�-HSD may be partial or there may be othermechanisms distinguishing aldosterone from glucocorticoids.The possibility that mineralocorticoid receptor plays an activerole in the mechanism of aldosterone selectivity in addition to11�-HSD has recently been suggested (Lombeset al., 1994).

Our case is also interesting in that the patient sustained a highlevel of creatine phosphokinase from birth, even after normal-ization of blood pressure and serum potassium levels bytreatment. To our knowledge, there is no AME case in theliterature that is associated with a high creatine phosphokinaselevel. Muscle biopsy was not performed because the patient hadno muscle-related symptoms. Currently, the aetiology of thishigh creatine phosphokinase level and its relation with AMEremain unknown.

In summary, we reported a case of apparent mineralo-corticoid excess who was born with intrauterine growthretardation and had hypokalaemia and alkalosis as a neonate.High CPK level was noted from birth and persisted even aftertreatment. We found that intrauterine growth retardation was a

characteristic of apparent mineralocorticoid excess and wespeculate that type 2 11�-hydroxysteroid dehydrogenase mightbe deficient in the placenta and/or fetal tissues as well as in thekidney.

Acknowledgements

The authors thank Dr K. Honma, Clinical Laboratories, KeioUniversity Hospital, for kindly assaying the urinary steroidprofiles, and Drs M. I. New, E. Werder and P. M. Stewart forproviding us with the birth records of their cases. We arealso grateful to Dr C. Kitanaka for critically reading themanuscript.

References

Albiston, A.L., Obeyesekere, V.R., Smith, R.E. & Krozowski, Z.S.(1994) Cloning and tissue distribution of the human 11�-hydro-xysteroid dehydrogenase type 2 enzyme.Molecular and CellularEndocrinology, 105,R11–R17.

Batista, M.C., Mendonc˛a, B.B., Kater, C.E., Arnhold, I.J.P., Rocha, A.,Nicolau, W. & Bloise, W. (1986) Spironolactone-reversible ricketsassociated with 11�-hydroxysteroid dehydrogenase deficiencysyndrome.Journal of Pediatrics, 109,989–993.

Beitins, I.Z., Bayard, F., Ances, I.G., Kowarski, A. & Migeon, C.J.(1973) The metabolic clearance rate, blood production, interconver-sion and transplacental passage of cortisol and cortisone in pregnancynear term.Pediatric Research, 7, 509–519.

Benediktsson, R., Lindsay, R.S., Noble, J., Seckl, J.R. & Edwards,C.R.W. (1993) Glucocorticoid exposure in utero: new model foradult hypertension.Lancet, 341,339–341.

Bongiovanni, A.M. & McPadden, A.J. (1960) Steroids duringpregnancy and possible fetal consequences.Fertility and Sterility,11, 181–186.

Brown, R.W., Chapman, K.E., Edwards, C.R.W. & Seckl, J.R. (1993)Human placental 11�-hydroxysteroid dehydrogenase: evidence forpartial purification of a distinct NAD-dependent isoform.Endocrinology, 132,2614–2621.

Edwards, C.R.W., Stewart, P.M., Burt, D., Brett, L., McIntyre, M.A.,Sutanto, W., de Kloet, E.R. & Monder, C. (1988) Local-ization of11�-hydroxysteroid dehydrogenase-tissue specific protector of themineralocorticoid receptor.Lancet, 2, 986–989.

Fieselier, T.J.W., Otten, B.J., Monnens, L.A.H., Honour, J.W. & vanMunster, P.J.J. (1982) Low-renin, low-aldosterone hypertension andabnormal cortisol metabolism in a 19-month-old child.HormoneResearch, 16, 107–114.

Harinck, H.I.J., van Brummelen, P., van Seters, A.P. & Moolenaat, A.J.(1984) Apparent mineralocorticoid excess and deficient 11�-oxidation of cortisol in a young female.Clinical Endocrinology,21, 505–514.

Honour, J.W., Dillor, M.J., Levin, M. & Shah, V. (1983) Fatal, low-renin hypertension associated with a disturbance of cortisolmetabolism.Archives of Diseases in Childhood, 58, 1018–1020.

Igarashi, Y., Egi, S., Takehiro, A., Ohzeki, T. & Kawaguchi, H. (1979)Studies on the metabolic abnormality of cortisol and corticosterone ina case of dexamethasone responsive mineralocorticoid excess.FoliaEndocrinologica Japonica, 55, 1341–1357 (In Japanese).

358 S. Kitanaka et al.

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359

Page 7: Apparent mineralocorticoid excess due to 11β-hydroxysteroid dehydrogenase deficiency: a possible cause of intrauterine growth retardation

Krozowski, Z.S. & Funder, J.W. (1983) Renal mineralocorticoidreceptors and hippocampal corticosterone-binding species haveidentical intrinsic steroid specificity.Proceedings of the NationalAcademy of Sciences, USA, 80, 6056–6060.

Lombes, M., Kenouch, S., Souque, A., Farman, N. & Rafestin-Oblin,M.E. (1994) The mineralocorticoid receptor discriminates aldoster-one from glucocorticoids independently of the 11�-hydroxysteroiddehydrogenase.Endocrinology, 135,834–840.

Monder, C. (1991) Corticosteroids, receptors, and the organ-specificfunctions of 11�-hydroxysteroid dehydrogenase.FASEB Journal, 5,3047–3054.

Monder, C., Shackleton, C.H.L., Bradlow, H.L., New, M.I., Stoner, E.,Iohan, F. & Lakshmi, V. (1986) The syndrome of apparentmineralocorticoid excess: its association with 11�-dehydrogenaseand 5�-reductase deficiency and some consequences for corticoster-oid metabolism.Journal of Clinical Endocrinology and Metabolism,63, 550–557.

Murphy, B.E.P., Clark, S.J., Donald, I.R., Pinsky, M. & Vedady, D.(1974) Conversion of maternal cortisol to cortisone during placentaltransfer to the human fetus.American Journal of Obstetrics andGynecology, 118,538–541.

Nikkila, H., Tannin, G.M., New, M.I., Taylors, N.F., Kalaitzoglou, G.,Monder, C. & White, P.C. (1993) Defects in the HSD11gene encoding 11�-hydroxysteroid dehydrogenase are not found inpatients with apparent mineralocorticoid excess or 11-oxoreductasedeficiency.Journal of Clinical Endocrinology and Metabolism, 77,687–691.

Reinisch, J.M., Simon, N.G., Karow, W.G. & Gandelman, R. (1978)Prenatal exposure to prednisone in humans and animals retards intra-uterine growth.Science, 202,436–438.

Shackleton, C.H.L., Honour, J.W., Dillon, M.J., Chantler, C. & Jones,R.W.A. (1980) Hypertension in a four-year-old child: gas chromato-graphic and mass spectrometric evidence for deficient hepaticmetabolism of steroids.Journal of Clinical Endocrinology andMetabolism, 50, 786–792.

Shackleton, C.H.L., Rodriguez, J., Arteaga, E., Lopez, J.M. & Winter,J.S.D. (1985) Congenital 11�-hydroxysteroid dehydrogenase defi-ciency associated with juvenile hypertension: Corticosteroid meta-bolite profiles of four patients and their families.ClinicalEndocrinology, 22, 701–712.

Shackleton, C.H.L. & Stewart, P.M. (1990) The hypertension ofapparent mineralocorticoid excess syndrome. InEndocrine Hyper-tension(ed. E.G. Biglieri & J.C. Melby), pp. 155–173. Raven Press,New York.

Stewart, P.M., Corrie, J.E.T., Shackleton, C.H.L. & Edwards, C.R.W.

(1988) Syndrome of apparent mineralocorticoid excess: a defect inthe cortisol-cortisone shuttle.Journal of Clinical Investigations, 82,340–349.

Stewart, P.M., Murry, B.A. & Mason, J.I. (1994) Type 2 11�-hydroxysteroid dehydrogenase in human fetal tissues.Journal ofClinical Endocrinology and Metabolism, 78, 1529–1532.

Stewart, P.M., Murray, B.A. & Mason, J.I. (1994) Human kidney 11�-hydroxysteroid dehydrogenase is a high affinity nicotinamideadenine dinucleotide-dependent enzyme and differs from thecloned type I isoform.Journal of Clinical Endocrinology andMetabolism, 79, 480–484.

Stewart, P.M., Rogerson, F.M. & Mason, J.I. (1995) Type 2 11�-hydroxysteroid dehydrogenase messenger ribonucleic acidand activity in human placenta and fetal membranes: its relation-ship to birth weight and putative role in fetal adrenal steroido-genesis.Journal of Clinical Endocrinology and Metabolism, 80,885–890.

Tannin, G.M., Agarwal, A.K., Monder, C., New, M.I. & White, P.C.(1991) The human gene for 11�-hydroxysteroid dehydrogenase.Journal of Biological Chemistry, 266,16653–16658.

Ulick, S., Levine, L.S., Gunczler, P., Zanconato, G., Ramirez, L., Rauh,W., Rosler, A., Bradlow, H. & New, M.I. (1979) A syndrome ofapparent mineralocorticoid excess associated with defects in theperipheral metabolism of cortisol.Journal of Clinical Endocrinologyand Metabolism, 49, 757–764.

Ulick, S., Tedde, R. & Mantero, F. (1990) Pathogenesis of the type 2variant of the syndrome of apparent mineralocorticoid excess.Journalof Clinical Endocrinology and Metabolism, 70, 200–206.

Ulick, S., Tedde, R. & Wang, J.Z. (1992) Defective ring A reduction ofcortisol as the major metabolism error in the syndrome of apparentmineralocorticoid excess.Journal of Clinical Endocrinology andMetabolism, 74, 593–599.

Warrell, D.W. & Taylor, R. (1968) Outcome for the foetus of mothersreceiving prednisolone during pregnancy.Lancet, 1, 117–118.

Werder, E., Zachmann, M., Vo¨llmin, J.A. Veyrat, R. & Prader, A.(1974) Unusual steroid excretion in a child with low reninhypertension.Research in Steroids, 6, 385–389.

Wilson, R.C., Krozowski, Z.S., Li, K., Obeyesekere, V.R., Razzaghy-Azar, M., Harbison, M.D., Wei, J.Q., Shackleton, C.H.L., Funder,J.W. & New, M.I. (1995) A mutation in the HSD11B2 gene in afamily with apparent mineralocorticoid excess.Journal of ClinicalEndocrinology and Metabolism, 80, 2263–2266.

Winter, J.S.D. & McKenzie, J.K. (1977) A syndrome of low-reninhypertension in children. InJuvenile Hypertension(eds M.I. New &L.S. Levine), pp. 123–131. Raven Press, New York.

11¯-HSD deficiency and intrauterine growth retardation 359

# 1996 Blackwell Science Ltd,Clinical Endocrinology, 44, 353–359