8/12/2019 1420500606_ftp

1/11

8/12/2019 1420500606_ftp

2/11

4 M . B A R RACEIs for myocardial protection poses lit tle concern tothe teratologist, since the vast majority of such use willbe in pati ents beyond the childbearing years. However,their use for mild or essential hypertension, and insuch conditions as the autoimmune diseases and dia-betes mellitus, could expose a considerable number ofwomen in the childbearing years, not a few of whomwill be pregnant. The use of ACEIs for the manage-ment of pregnancy-induced hypertension, by defini-tion, creates a certain a nd identifiable fetal exposure.The actual number of pregnancies exposed to ACEIscannot be determined from available data. Piper et al .('92) reported that of 106,813 Tennessee Medicaid pa-tients who delivered a live or stillborn infant during1983-1988, 19 were exposed to an ACEI. Given theexpansion of use of these agent s to the curren t time, itwould be reasonable to expect th at the number of fetalexposures will increase.

ADVERSE EFFECTSIn the enthusiasm to employ ACEIs as first lineagent s for treatment of hypertension and t o retard de-velopment of diabetic nephropathy, their safety some-times seems to be overstated. As wonderful as theyseem, ACEIs are not without adverse effects. As de-scribed below, because they suppress aldosterone secre-tion, they a re liable t o produce hyperkalemia, particu-larly in the face of sodium restriction, heart failure,diabetes, and coadministration of potassium-sparingdiuretics and nonsteroidal anti-inflammatory drugs(Williams, '88; Schlueter et al., '94). Their use in cer-tain circumstances, such as bilateral renal ar tery ste-nosis, can contribute to, not ameliora te, renal dysfunc-tion (Hricik et al., '83; Cooke and Debesse, '94). In somecases the dysfunction may be reversible (Wood et al.,'91) and in other s irreversible (Devoy et a l., '92). Per-haps the most common side effect of the ACEIs is a nirritating cough, which ironically has generated moreliterature than has the possibility of fetal damage.

As early as 1980, there were repor ts of fetal wastagein ACEI-exposed experimental animals (Broughton-Pipkin et al., '80, '82; Fe rri s and Weir, '82). In 1981, thefirst adverse outcome in a human pregnancy was re-ported (Duminy and Burger , '81). This was followed inshor t order by other cases implica ting both captopriland enalapril as potential fetotoxins (Guignard, '82;Boutroy et al., '84; Caraman e t al., '84; Fiocchi et al.,'84; Rothberg an d Lorenz, '84). Warnings about t he useof ACEIs in hum an pregnancy appeared in the lite ra-ture as early as 1985 (Lindheimer and Katz, '85). Onthe other hand, some recent reviews of the ACEIs makeno mention of the possibility of adverse fetal effects(Materson and Preston, '94; Pinkney and Yudkin, '94).Pryde et al. ('93) recently proposed the designationACE inhibitor fetopathy, summarized 29 affected

cases from the lite rature, and added 3 more. Two morecases have been reported since the review by Pryde et

al. (Piper e t al., '92'; Thorpe-Beeston et al., '93), and Ihave received at least part ial reports of five others frommedical colleagues and attorneys. It appears that thebulk of ACEI fetopathy cases have ensued after enal-april exposure, but cases associated with captopril an dlisinopril are recorded (Hanssens et al., '91; Pryde etal., '93). As pointed out by Brent and Beckman ('gl),there is no reason to assume tha t the other ACEIs ar eincapable of producing the same result. Similarly,many of the cases have involved coexposure to otherantihypertensive agents, and while combination the r-apy may pose more risk to t he f etus, monotherapy withan ACEI has produced the full-blown fetopathy.

The most commonly reported adverse effect of ACEIexposure of the fetus is that of second to third trimesteronset of oligohydramnios and growth restriction, fol-lowed by delivery of an infant whose neonatal course iscomplicated by prolonged and often profound hypoten-sion and anur ia (Rosa et a]., '89). The p resentation mayalso include the oligohydramnios deformation se-quence and its lethal component, pulmonary hypopla-sia (Guignard et al ., '81; Mehta and Modi, '89; Cunniffet al., '90; Pryde et al., '93). Although studies of thehistology of the kidneys in ACEI-exposed fetuses andinfants are few in number, they are consistent in the irdemonstration of renal tubular dysgenesis (Knott etal., '89; Cunniff et al., '90; Pryde et al., '93). Delayeddevelopment of the calvaria has been noted in enoughcases to consider it a part of the syndrome. Postnatalpersistence of a patent ductus arteriosus (PDA) hasalso been suggested to be pa rt of the syndrome, but it ishard to determine if th is finding is an effect of ACEIexposure.

The true rate s of adverse fetal effects from ACEI usein human pregnancy cannot be determined from avail-able information. Certainly, a number of exposed preg-nancies have resulted in no measurab le adverse effect(Ducret et al., '85; Kreft-Jais et al., '88; Piper et al.,'92). In th e largest published series there were 31preg-nancies exposed either to captopril (n 22) or enala-pril (n 9) (Kreft-Jais et al., '88).The reported adverseoutcomes included nine cases of intrauterine growthrestriction, three intrauterine deaths, and two infantswith PDA, of which one ultimately required ligation.The 22 other cases in this report were presumably un-affected.The adverse developmental effects of ACEIs are pur-posely designated ACEI fetopathy because as yet thereis no convincing evidence th at they cause harm in t hefirst trimest er of huma n gestation or its equivalent inlaboratory animals. Studies with captopril and ena la-pril in rats failed t o show malformations result ing from

lone case described by Piper et al. ('921, with microcephaly andencephalocele, was classified as an example of hypocalvaria: I believethi s infant's problems were probably unrelated to ACEI exposure.

8/12/2019 1420500606_ftp

3/11

ACE INHIBITORS 4 1high doses (Fujii et al., '85; Fujii and Nakatsuka, '85;Robertson et a l., '86; Al-Shabanah et al., 91;Al-Harbiet al., 92), although at the highest doses tested, fetalgrowth restriction, delayed ossification, and resorp-tions were noted in conjunction with evidence of ma-ternal toxicity. In humans, the case reported byThorpe-Beeston et al. ( 93) was found to have renalcystic dysplasia, a disorder distinct from the renal tu -bular dysgenesis found in other cases, but s till likely tobe a fetal rather th an a n embryonic effect. In the casereported by Duminy and Burger ('81) there was a ter-minal t ransverse deficiencyof a limb, again likely t o beof fetal origin. The second case mentioned by Piper etal. ( 92)was remarkable for microcephaly and enceph-alocele, the latter almost certainly a problem of firsttrimester origin, and I am aware of an enalapril expo-sure-associated case of anencephaly. These few dispar-ate anomalies associated with ACEI exposure do notconstitute sufficient evidence of true teratogenicity, al-though admittedly there is insufficient evidence t o beconfident of safety in the first trimester. Currently, therecommendation is to offer reassurance of unlikely firsttrimester effects, but to avoid second and third trimes-te r exposure, unless absolutely necessary for the moth-er's well being.

REVIEW O F RENIN-ANGIOTENSINSYSTEM (RAS)Angiotensin I is a decapeptide formed from the ac-tion of renin on its precursor, angiotensinogen (Fig. 1).Angiotensin I has no known biological action in hu-mans but is rapidly converted to the biologically active

octapeptide, angiotensin 11. This conversion is effectedby ACE, a zinc metalloprotease tha t catalyzes th e hy-drolysis of carboxy-terminal dipeptides from oligopep-tide substrates. The principal substrates are angioten-sin I and the nonapeptide, bradykinin, but othersubstrates are affected as well (Ehlers and Riordan,89). Angiotensin I1 is a potent vasoconstrictor, stimu-lates aldosterone secretion from the adrenal cortex,

and suppresses renin release by increasing sodium re-tention, whereby it closes a negative-feedback loop. Aparallel system involves the formation of bradykininand it s inactivation by kininase TI (which is identicalwith ACE). Angiotensin I1 raises blood pressure, andbradykinin, if not inactivated by the enzyme, lowersblood pressure.

ACE has two catalytic sites, one near the carboxylterminus and one near the amino acid terminus. Thesesites may have specific substrates, but angiotensin Iand bradykinin appear to be catalyzed equally a t bothsites (Johnston et al., 93). The ACE gene is develop-mentally regulated in a tissue-specific manner andplays a role in the regulation of renal function andgrowth (Yosipiv et al., 94). The ACE gene exhibitspolymorphism, marked by the insertion or deletion of a250 bp fragment in intron 16 of the gene (Rigat et al.,

Kininogen AngiotensinogenC-- eninaliikrein t

Arachidonic, Brad yk ini n\/ Angiotensi n IAcidKininase II Convertingnzyme ECF Expaniion

Na Retention?rostaglandin idoeteroneFig. 1. Schematic representation of the renin-angiotensin-bradyki-

n n system. ECF extracellular fluid .

'90). Homozygotes for the deletion (DD) show the high-est serum levels of ACE activity, while homozygotes forthe insertion (11)have the lowest levels. The prevalenceof the gene types varies in different populations (Lee,,941, and it has been reported that the deletion geno-type is less prevalent in diabetic patients with ne-phropathy than in those without nephropathy (Marreet al., 94).Angiotensin I1 receptors similarly have twotypes: type 1 (AT,) is found in a wide variety of tissuesin the adult and seems t o mediate all t he known effectson angiotensin I1 related to the RAS. This type of re-ceptor has two subtypes, ATlA and ATlB, which aresaid t o show few pharmacological differences and mayfunction as growth factors. While the type 2 (AT,) re-ceptor is the predominant type found in the gestationalday (GD) 19 ra t fetus, fetal adrenal cortex and renalglomeruli express both AT,, and AT,, mRNAs (Groneet al., '92; Shanmugam et al., 94).Their roles in fetaldevelopment are being investigated. What effect suchgenetic diversity, in mother or fetus, may have on theOccurrence of adverse fetal effects from ACEI exposureis unknown at thi s time, but should be explored.

The RAS becomes crucially important under condi-tions of low renal perfusion pressure (Hall et al., '77;Blythe, 83). In these conditions, which pertain to thefetus, angiotensin 11-mediated efferent arteriolar resis-tance is essential to the maintenance of glomerularfiltration (GFR) and production of urine (Rudolph etal., 71; Giugnard, '82). Activation of bradykinin bysuppression of angiotensin I1 can compromise GFR byvirtue of bradykinin's vasodilatory effect on the effer-ent arteriole (Kon et al., 93). In the adult, ACEIs in-crease renal blood flow. Usually this occurs withoutmodifying the GFR because vasodilation of efferent ar-terioles exceeds that of the afferent arterioles (Wil-liams, '88). In cases of renal artery stenosis, bilateral orunilateral in a solitary kidney, there is an inability t oincrease renal blood flow sufficiently to compensate forthe efferent arteriolar dilation and maintain glomeru-la r perfusion pressure (Hricik et al., '83).

In pregnancy-induced hypertension mean arterialblood pressure is correlated with serum ACE activity

8/12/2019 1420500606_ftp

4/11

4 2 M . B A R R(Li et al., '92) and is characterized by enhanced an-giotensin I1 sensitivity due to an increase in angioten-sin I1 receptor number (Graves et a l., '92). However,the RAS is known t o be both a circulating and a tissuehormonal system, and, of the two, tissue systems maybe the more important (Ehlers and Riordan, '89;Johnston et al., '92). ACE is widely distributed in thebody; it has been found in soluble form in body fluidsand in membrane-bound form in arterial endothelialcells within and outside the pulmonary circulation, inepithelial cells with brush borders (placenta, kidney,intestine, and choroid plexus), in neuroepithelial cells,and in the male genital tract (testis, prostate, and ep-ididymis) (Erdos and Skidgel, '87). Evidence is emerg-ing th at plasma ACE inhibition and hemodynamic re-sponses are separable (Lees et al., '92). Prolongedtreatment with an ACEI results in an increase ofplasma ACE but a decrease of ACE in the renal cortexand renal brush border (Michel et al., '93). This indi-cates that plasma and epithelial ACE are subject t olocal regulatory factors. The renal proximal convolutedtubule contains angiotensinogen, ACE, and angioten-sin I1 receptors (Moe et al., '93; Jackson et al., '91). Moeet al. ('93) found that renin could not be found in tu-bules from normal rats , but after enalapril administra-tion, renin was found. This was interpreted as evidenceof a tissue RAS that could generate angiotensin I1 tolocally regulate sodium absorption.It has been suggested that in part the effects ofACEIs are due to a locally acting, prostaglandin-depen-dent component to their hypotensive action. Linz et al. ,('93) reported that when bradykinin degradation wasinhibited by ramipril, formation of the endothelial au-tacoids, nitric oxide and prostacyclin, was enhanced,which they felt contributed to the beneficial effects ofthe drug. Similar conclusions were reached by others(Vanhoutte et al., '93; Busse et al., '93). On the otherhand, Gerber et al. ('93) found tha t while indomethacinreduced the urinary excretion of prostacyclin metabo-lite by more than 50%, it had no effect on the hypoten-sive effect of captopril or enalapril. They concludedthat neither ACEI had a significant prostacyclin-de-pendent component t o its hypotensive action, and Gan-sevoort et al. ('94) concluded that the antiproteinuricand renal hemodynamic effects of ACEIs were primar-ily due to interference in the RAS, rather than thebradykinin system. For now, the relation of the effectsof the ACEIs to bradykinin degradation and prosta-glandin production is not resolved. These two hor-mones may be acting a t the tissue level without corre-lated change in blood or excretion levels. Involvementof the prostaglandin system in ACEI fetopathy seemspossible, particularly given the number of reported in-stances of renal tubular dysgenesis (RTD)/fetal-neona-ta l anuria associated with nonsteroidal anti-inflamma-tory agent exposure (vide infra).Tissue kinin generation and degradation are coordi-nately regulated during intrauterine development,

while circulating angiotensin I1 and ACE activitychange reciprocally (Yosipiv et al., '94). It appears thatkinins and angiotensin I influence the development ofACE. It has been established that the RAS is active infetal life, when it plays an essential role in maintain-ing GFR under conditions of low renal perfusion pres-sure (Jelinek e t al., '86). At least in the lamb, it seemsthat the RAS i s more active in the fetus than in theneonate (Binder and Anderson, '92).Angiotensin I may also act on the growth of its tar-get tissues. Some cells, such as those of the adrenalcortex, show induction of cell division by angiotensin11.The mechanism by which angiotensin I1induces hy-perplasia of its target tissues is largely unknown butmay be from a direct action on proto-oncogene synthe-sis or an indirect action on growth factor secretion(Clauser et al., '92). As previously noted, the subtypesof angiotensin I1 receptors may function differentiallyas growth factors (Shanmugam et al. , '94).

ANIMAL STUDIESClassical studies of the ACEIs given during the pe-riod of organogenesis t o rats and rabbits have failed todemonstrate an increased incidence of malformation.

To emphasize a point made previously, lack of terato-genicity does not mean lack of fetotoxicity. The admin-istration of ACEIs to commonly used laboratory ani-mals, in late gestation and during lactation, in dosescomparable t o those used in human therapy, has beenassociated with significant fetal and neonatal wastage.These are fetotoxic not teratogenic effects. However, asdescribed below, none of these species is an especiallygood model for the ACEI fetopathy seen in humans. Sofar, no exploration of maternal and fetal ACEI effectsand physiology seems to have been done in nonhumanprimates which might be much better models of theeffects seen in humans; the development of such modelsshould be encouraged.

Rabbit modelThe adult rabbit is particularly sensitive t o ACEIs,showing a greater antihypertensive effect than doesthe ra t or dog. As measured by blockade of the acutepressor response t o exogenous angiotensin 1, enalapri-lat and captopril are of equal potency in pregnant andnonpregnant rabbits (Manson, personal communica-

tion). The administration of 2.5-5.0 mgikgiday of cap-topril t o pregnant rabbits yielded pregnancy loss ratesranging from 37 to 92% (Broughton-Pipken et al., '80,'82; Ferris and Weir, '82). The ED50 values for block-ade of the pressor response to angiotensin I by enala-prilat in the rabbit, ra t, and dog were 2.8, 5.1, and 6.4mgikg, respectively. Comparable ED50 values for cap-topril in these 3 species were 2.88, 26.1, and 80 mgikg,respectively (Manson, personal communication). Thebasis for this sensitivity in rabbits is not known.The late gestation rabbit fetus has been shown t o behighly sensitive t o the fetotoxic effects of ACEIs. When

8/12/2019 1420500606_ftp

5/11

ACE INHIBITORS 403administered in the human therapeutic range, ACEIsproduce fetal deaths in mid to late gestation, with apeak effect on GD 26 (Keith et al. , '82; Broughton-Pip-kin et al., '82; Minsker et al., '90). Maternal toxicity, asevidenced by elevation of blood urea nitrogen (BUN)and creatinine, and occasionally maternal death , oc-curs at the same treatm ent levels as those causing fetaldeaths. Fetal deaths observed in rabbits with ACEItreatment may be the result of decreased placental per-fusion commensurate with systemic hypotension in themother, rather than of a direct effect on fetal renalperfusion (Fe rri s and Weir, '82; Binder e t al., '89;Binder and Faber , '92). Captopril studies in the rabbitshowed no effect on amniotic fluid volume, amnioticflu id electrolyte composition, o r fetal plasma urea ni-trogen, creatinine, o r calcium levels (Manson, personalcommunication), When plasma volume was expandedby saline supplementation, maternal toxicity was re-duced but fetal deaths continued, particularly athigher dosage levels (ena lapril30 mg/kg/day) (Minskeret al., '90).

Rat modelThe matura tion of the RAS in the fetal and neonatalrat kidney has been thoroughly studied (Ice et al., '88).Expression of the gene for the angiotensin I1 receptor

was not detected in the liver of newborn rats, but intheir kidneys gene expression was 2.5 times higherthan in the adult (Iwai et al., '91). Decreasing concen-tra tions of plasma renin from fetal, to newborn, to neo-natal rats are consistent with increased activity of theRAS in immature animals (Jelinek et al., '86). Reninwas first detected by immunostaining in the kidney onGD 19, when i t was located in the a rcuate and inter-lobular arteries. On GD 20, renin also appeared in th eafferent arteriole. With advancing age, detectable re-nin disappeared progressively from the arcuate and in-terlobular arte ries until i t was restricted primarily tothe afferent arteriole in the 20-day-old postnatal pupand finally only t o the juxtaglomerular apparatus inthe adult kidney. A similar bu t more rapid progressionof maturation has been observed in the mouse (re-viewed by Ice et al., '88). Studies of the synthesis ofrenin mRNA in th e r at have shown the same patternand timing (Gomez et al., '89).

The relevance of these findings to reproductive anddevelopmental toxicity studies of ACEIs in rat s is t ha tthe RAS develops late in gestation, beyond the timewhen treatment typically is administered in a segmentI1 ra t developmental toxicity study (GD 6-17). Not sur-prisingly, then, results from these studies have uni-formly failed to reveal any adverse effects on fetuses,with the exception of decrease in fetal body weight andossification at treatment levels th at also produce evi-dence of mater na l body weight decreases.

Interestingly, results from female fertility and lategestationilactation studies, in which ACEI exposure oc-curred during the critical period of fetal renin synthe-

sis (GD 19), have demonstrated adverse effects on ne-onates (Robertson et al., '86). At high treatment levels(approximately 30-300 mgikgiday for enalapril andlisinopril), pup deaths occurred during lactation. Atlower treatment levels (10 mg/kg/day), decreased pre-and postweaning body weight gain was observed. Sim-ilar findings have been obtained with other ACEIs,quinapril (Dostal et al., '91) and rentiapril (Cozens etal., '87).Late gestationilactation studies, with treatmentfrom GD 15 to th e end of lactat ion, would seem to pro-vide the most sensitive appraisal of ACEI effects on ra tpups. Kidneys from weanlings whose mothers receivedquinapril during la te ges tation (GD 15 on) and/or lac-tation had juxtaglomerular cell hypertrophy (Dostal etal., '91; Graziano et al., '93), although the latter au-thors reported they observed no morphological changesin the renal tubules and no adverse effects on renalfunction. Early postnatal treatment with the ACEIsenalapril an d captopril, and th e angiotensin receptor-1blocker losartan, but not the angiotensin receptor-2blocker PD-123319, in th e spontaneously hypertensivera t (SHR) and th e normotensive str ains WKY and WRproduced persistent, irreversible histopathological re-nal changes in adult life, long after th e tre atment hadstopped. These abnormalities were associated wi th im-paired urine concentrating ability (Friberg et al., '94).This should be looked at for other ACEIs. For t he mostpart, histopathology from such kidneys has either notbeen done or not been reported in the literature.

All in all, it does not appear that the rat is a suitablemodel for predicting or understanding ACEI fetopathyin humans. The RAS develops much l ater in gestationin rats than in humans, and in rats toxicity is mani-fested in neonates and weanlings, while in humans themajor manifestat ions are seen prenatally and at birth.Also, from the available data, it appears that the ra tplacenta is an effective barrier to ACEI penetrationinto the fetal compartment, although more data areneeded on this point.

Sheep modelThe chronically catheterized sheep serves as an ex-

cellent model to study the cardiovascular and renalchanges in the maternal and fetal compartments oc-curring dur ing pregnancy (Magness et al., '85; Hill an dLumbers, '88). When captopril, which crosses the sheepplacenta and blocks the fetal RAS, was administered inlate pregnancy (GD 119-133, term 147) a t 2.8-3.5mgikg IV to the ewe following exogenously adminis-tered angiotensin I, mater nal blood pressure was tran -siently reduced and returned to normal within 2 hr.Fet al blood pressure remained reduced for up to 2 days,and 7 of 8 lambs were stillborn (Broughton-Pipkin etal. , '82).

In a similar model, fetal renal function was studiedin sheep given captopril (GD 123-133) a t an infusionra te of 15 mg IV for 4 days, followed by a n int rafeta l

8/12/2019 1420500606_ftp

6/11

8/12/2019 1420500606_ftp

7/11

ACE INHIBITORS 4 5

Fig. 2. RTD n four different cases o f ACEI fetopathy; note particularly t he ductular ectasia, dilationof Bowman's spaces, and poor t o no differentiation of proximal convoluted tubules. x 20.

(Fig. 3 . The sutures and fontanels are symmetricallyenlarged, an d in severe cases, the normally developingbra in is essentially unprotected by skul l and liable totrauma during labor and delivery. Although there arenow only a total of six cases in which this skull lesionis specifically described, I believe it is more commonbut unrecognized; it seems that many cases may beclassified as enlarged fontanels with widely split su-tures , but in instances in which hydrocephaly or mac-roencephaly are not present. Lacking age-specific mor-phometrics for the size of the calvarial bones, thedistinction of hypocalvaria in milder cases is admit-tedly a subjective assessment.The cause of the hypoplastic calvaria found withACEI exposure is unknown. Endochondral bone andmembranous bone grow and develop in entire ly differ-ent ways. Long bones develop i n a low oxygen environ-ment since nutrition takes place by diffusion throughthe cartilaginous epiphyses. Membranous bones, on t heother hand, have a high degree of vascularity, and a

high oxygen tension is required for their growth. Thepresumed fetal hypotension produced by ACEI expo-sure may result in hypoxic effects and th us a hypoplas-tic calvaria (Ba rr and Cohen, '91). A more remote pos-sibility is that inhibition of angiotensin I1 mayconcomitantly inhibit one of the var iety of growth fac-tors involved in calvar ial bone development.

Intrauterine growth restrictionThe status of the fetal RAS in pregnancies compli-cated by severe intrauterine growth restriction, and it spossible relationship to elevated fetoplacental vascularresistance, was explored by Kingdom et al. ( 93). In an

intrauterin e growth restriction group, cord venous an-giotensin I1was markedly elevated compared with con-trols, but angiotensin receptor concentration was notsignificantly altered, and no changes in angiotensinreceptor affinity were observed. This suggests thatthere i s augmentation of'the fetal RAS in a t least someforms of intrauterine growth restriction and that re-

8/12/2019 1420500606_ftp

8/11

4 6 M . B A R R

Fig. 3. Hypocalvaria. In this severe example, the fibrous tissuecomprising the fontanels and sutures has been removed, leaving onlythe calvarial bone, to demonstrate the diminutive size of the bones.(Reproduced from Barr and Cohen, '91, with permission of Wiley-Liss.)

sponsiveness of the fetoplacental vasculature to an-giotensin I1 is not diminished as might be expectedfrom the elevated plasma angiotensin I1 levels. An-giotensin I1may contribute to the increased fetoplacen-ta l vascular resistance observed in this disorder.Intrauter ine growth restriction (birth weight centile< l o ) has been observed in 21 of 27 cases of ACEIfetopathy for which birth weight information wasgiven (Pryde et al ., 93). In a single case (Pryde et al.,

'93: l),imb lengths were disproportionately short bymeasurement, although not to the degree that a diag-nosis of short-limbed dwarfism would be made. No dataon limb lengths are given for other cases. From mea-surement of limb lengths in other fetuses with pre-sumed hypotension, it is hypothesized that relativegrowth lag of the limbs is a marker of an adaptive shiftof circulatory pattern in the compromised fetus, withcirculation diminished to the periphery to preservecentral perfusion (unpublished observations). Thus itcould be predicted that a feature of ACEI fetopathywould be some shortening of the limbs.

PDAA final observation which appears in reviewing re-ports of neonatal outcomes in pregnancies exposed tothe ACEIs is the persistence of a PDA. While this may

be associated with th e high frequency of prematurity inthe cases reported, it is suggested that there appearsalso to be an increased need for surgical intervention.Out of seven PDA cases reported in association withACEI exposure, three required surgical ligation(Boutroy et al., '84; Caraman et al., '84; Plouin andTchobroutsky, '85; Kreft-Jais et al., '88).Given the pos-sible effect of the ACEIs on the fetal bradykinin-pros-

taglandin system, which would be expected t o increaseprostaglandin E and prostacyclin systemically and per-haps locally, the possibility that prenatal ACEIs mayinhibit ductal closure needs to be considered. Cer-tainly, more data are needed before any firm conclu-sion can be reached on this subject.

COULD IT BE THE MATERNAL DISEASEITSELF AND NOT THE DRUG?ACEI fetopathy infants have been born t o motherswhose hypertension was of widely varying etiology, in-cluding lupus erythematosus, renovascular hyperten-sion, nephrotic syndrome, glomerulonephritis, renal

transplantation, preeclampsia, and essential hyperten-sion (Pryde et al ., '93). Oligohydramnios, intrauterinegrowth restriction, and fetal distress are not uncom-mon complications of hypertensive pregnancies. How-ever, as discussed above, there is evidence that indi-cates the oligohydramnios, hypotension, and anur ia ofACEI fetopathy are truly drug-related rather than sim-ply phenomena of the underlying maternal disease pro-cess. The evolution of oligohydramnios followed by thedelivery of a neonate with prolonged hypotension andanuria has not been included in the well-described listof complications of maternal hypertension and its tra -ditional therapy, although one case of RTD has beenobserved under these circumstances (Martin et al., '92).With ACEI exposure, the probability is high that theoligohydramnios is in fact due t o a drug-related fetalhypotension and renal failure that persists into the neo-natal period. Evidence in support of this hypothesis isthe occurrence of profound hypotension and anuria insome neonates given low doses of ACEIs for treatmentof hypertension (Tack and Perlman, '88; Perlman andVolpe, '89; Scott and Purohit, '89; Wells et al., '90).Further, when ACE activity was measured in ACEI-exposed hypotensive neonates it was found t o be pro-foundly blunted and appeared t o normalize only afterdialysis removed the otherwise renally excreted drug(Schubiger et al., '88;Pryde et al., '93). There are reportsthat the onset of oligohydramnios was temporally re-lated to the initiation of ACEI therapy (Guignard et al .'81; Schubiger et a l., '88), and in one case the amnioticfluid volume increased toward normal after the ACEIwas discontinued (Broughton-Pipkin et al., '89).

SUMMARYOccasionally there is a drug whose record in preg-nancy is so frequently associated with adverse outcomeof so specific a pattern that it becomes clear that i ts usemust be restricted before scientific proof from epidemi-ological studies is obtained. I believe this to be the casewith the drug class of ACEIs. There are mammalianmodels suggesting substantial fetotoxicity in a dose-related fashion. There is a strong and consistent pat-tern t o the reported cases of ACEI-related adverseoutcomes: the syndrome of oligohydramnios-anuria,

8/12/2019 1420500606_ftp

9/11

ACE INHIBITORS 4 7neonatal hypotension, renal dysplasia, and hypocalva-ria is too specific in association with the use of thesedrugs to be ignored. There is a very plausible biologicmechanism to explain the relationship. The features ofACEI fetopathy suggest th at the underlying pathoge-netic mechanism is fetal hypotension, which may alsoresult from other exposures. Thus, while the fetopathymay not be truly specific to ACEIs, they are particu-larly liable to produce adverse fetal renal effects withtheir sequels (anuria-oligohydramnios, pulmonary hy-poplasia, growth res triction) and hypocalvaria.

RECOMMENDATIONSIt is advised that ACEIs not be prescribed t o preg-nant women. Based on current information, exposure

to ACEIs in the first trimes ter should not be consideredan indication for terminating a pregnancy. In suchcases, an alt ernative antihypertens ive regimen shouldbe substituted for the drug prior to ente ring the secondtrimes ter. If a woman is inadvertently on ACEI ther-apy in th e second or third trimes ters of pregnancy, sheshould be monitored for signs of fetal toxicity, includ-ing oligohydramnios, growth restriction, or fetal dis-tress. Although oligohydramnios has been observed t oreverse when the ACEI i s discontinued, a cautionarynote must be sounded. The signs of fetal toxicity maynot be detected until after irreversible damage to thefetus has occurred.

In the case of an ACEI-exposed fetus, at the time ofdelivery or before, the pediatricians should be notifiedof the potential for neonatal hypotension and anuriaand the possible need to attempt early dialysis t o re-move the otherwise renally excreted drug. However,the combination of hypotension and anur ia makes bothperitoneal and hemodialysis extremely difficult, andthe mortality rate is exceptionally high. Because thedamage th at can occur from these drugs is so severe, itis urged that their prescription to women of childbear-ing potential be limited and, if they must be used, t ha tthe women be warned about risks to the fetus and bemonitored very closely for the occurrence of pregnancy.

ACKNOWLEDGMENTSTwo highly esteemed colleagues, Dr. Je anne Manson

and Dr. Aileen Sedman, contributed helpful informa-tion, perceptive comments, and intelligent discussionduring the preparation of this review. For her role ingetting the product labeling of the ACEIs changed towarn of the fetal risks, Dr. Sedman warrants specialthanks . Thanks are also due to th e journal reviewersfor their cogent suggestions.

LITERATURE CITEDAl-Harbi, M.M., O.A. Al-Shabanah, N.M.A. Al-Gharably, and M.W.

Islam 1992) The effect of maternal administration of enalapril onfetal development in the rat. Res. Commun. Chem. Pathol. Phar-macol., 77t347-385.

Allanson, J.E., A.G.W. Hunte r, G.S. Mettler, and C. Jimenez 1992)

Renal tub ula r dysgenesis-A not uncommon autosomal recessivesyndrome: A review. Am. J. Med. Genet., 43.811-814.

Al-Shahanah, O.A. M.M. Al-Harhi, N.M.A. Al-Gharably, and N.W.Islam 1991) The effect of mat ernal administration of mptopril onfetal development in rat . Res.Commun. Chem. Pathol. Pharmacol.,73:221-230.

Bakhle, Y.S. 1968) Conversion of angiotensin I to angiotensin 11bycell-free extracts of dog lung. Nature, 220t919-921.

Bakris, G.L. (1993) Angiotensin-converting enzyme inhibitors andprogression of diabetic nephropathy. Ann. Int . Med., 118,643-644.Bar r, M., and M.M. Cohen 1991) ACE inhibitor fetopathy and hypo-

calvaria: The kidney-skull connection. Teratology, 44r485-495.Bavoux, F. (1992)Toxicite foetal des anti-inflammatoires non steroi-

diens. Presse Med., 21t1909-1912.Binder, N.D., and D.F. Anderson 1992) Plasma renin activity re-

sponses to graded decreases in re nal perfusion pressure in fetal andnewborn lambs. Am. J. Physiol., 262:R524-R529.

Binder, N.D., and J. J. Faber 1992) Effects of captopril on blood pres-sure, placental blood flow and u terine oxygen consumption in preg-nant rabbits. J. Pharmacol. Exp. Ther., 260r294-299.

Binder, N., J. Faber, and D. Anderson 1989)The effect of captopri l onblood pres sure, uter ine blood flow and oxygen consumption in thepregnant rabbit. Clin. Res., 37t6A.

Blythe, W.B. 1983) Captopril and renal autoregulation. N. Engl. J.Med., 308t390-391.Boutroy, M.J., P. Vent, B. Hunau lt de Ligny, an d A. Milton 1984)Captopril administration in pregnancy impairs fetal angiotensinconverting enzyme activity and neonatal adaptation. Lancet,2.935-936.

Brent, R.L nd D.A. Beckman 1991)Angiotensin-converting enzymeinhibitors, a n embryopathic class of drugs with unique properties:Information for clinical teratology counselors. Teratology, 43.543-546.

Broughton-Pipkin, F . , and C.P. Wallace 1986)The effect of enalaprilMK-4211,an angiotensin-converting enzyme inhibitor, on th e con-scious pregnant ewe and her foetus. Br.J. Pharmacol., 87t533-542.

Broughton-Pipkin,F., S R urne r, and E.M. Symonds(1980)Possiblerisk with captopril during pregnancy: Some animal data. Lancet,2t1256.

Broughton-Pipkin, F . , E.M. Symonds, and S.R. Turner 1982) Theeffect of captopril (SQ 14,225)upon mother and fetus in th e chron-ically cannulated ewe and in the pregnant rabbit. J . Physiol., .323t415-422.

Broughton-Pipkin, F., P.N. Baker, and E.M. Symonds 11989) An-giotensin-converting enzyme inhibitors in pregnancy. Lancet,2r96-97.

Buderus, S., B. Thomas, H. Fahnenstich, and S. Kowulewski 1993)Renal failure in two p reterm infants: Toxic effect of pren atal ma-ternal indomethacin treatment. Br. J. Obstet. Gynaecol., 100:97-98.

Busse, R., I. Fleming, and M. Hecker 1993) Endothelium-derivedbradykinin: Implications for angiotens in-converting enzyme inhib-itor therapy. J. Cardiovasc. Pharmacol., 221Suppl. 51rS31S36.

Carama n, P.L., A. Miton, B. Hurau lt de Ligny, et al . (1984)Grossessessous captopril. Therapie, 39t59-62.

Clauser, E., C. Bihoreau, C. Monnot, et al. 1992) Angiotensin-11, agrowth factor. Diabete Metabol., 18:129-136.Cooke, H.M., and A. Debesse 11994) Angiotensin-converting enzymeinhihitor-induced rena l dysfunction: Recommendations for preven-tion. Int. J. Clin. Pharmacol. Ther., 32:65 70.

Cozens, D.D., S.J. Barton, R. Clark , E.W. Hughes, J.M. Offer, and Y.Yamamoto 119871 Reproductive toxicity s tudies of rentiapril. Arz.ForschJDrug Res., 37:164-169.

Cunniff, C., K.L. ones, K. Phillipson, S. Short, and J. Wujek 1990)Oligohydramnios and renal tubul ar malformation associated withmaternal enalapril use. Am. J. Obstet. Gynecol., 162.187-189.

Cushman, D.W., and M.A. Ondetti 1991) History of the design ofcaptopril and related inhibitors of angiotensin-converting enzyme.Hypertension, 17r589-592.

Devoy, M.A.B., C.R.V. Tomson, M.E. Edmunds , J . Feehally, and J.

8/12/2019 1420500606_ftp

10/11

8/12/2019 1420500606_ftp

11/11

ACE INHIBITORS 4 9The effects of a converting enzyme inhibitor icaptopril) and an-giotensin-I1on fetal r enal function. Br. J. Pharmacol., I IU:821-827.

Magness, R., K. Osei-Boaten, M. Mitchell, and C. Rosenfeld (1985) Invitro prostacyclin production by ovine uterine and systemic arter-ies: Effects of angiotensin 11. J . Cl in. Inves t., 76t2206-2212.

Marre, M., P. Bernadet, Y. Gallois, et al. (1994) Relationships be-tween angiotensin I converting enzyme gene polymorphism, plasmalevels, and diabetic retinal and renal complications. Diabetes, 43:

Martin, R.A., K.L. Jones, A. Mendoza, M. Barr, and K. Benirschke(1992)Effect of ACE inhibition on th e fetal kidney: Decreased renalblood flow. Teratology, 46r317-321.

Materson, B.J., an d R.A. Preston (1994) Angiotensin-converting en-zyme inhibitors in hypertension. Arch. Int. Med., I54:513-523.

Mehta, N., and N. Modi (1989) ACE inhibitors in pregnancy. Lancet,2:96.

Michel, R., D. Stephan, M. Grima, M. Barthelmebs, and J.L. Imbs(1993) Effects of one-hour and one-week tre atm ent with ramipril onplasma and renal brush border angiotensin-converting enzyme inthe rat . Eur. J . Pharmacol., 242:237-243.

Minsker, D.H., W.J. Bagdon, J.S. MacDonald, R.T. Robertson, andD.L. Bokelman 1990)Maternotoxicity and fetotoxicity of an an-giotensin-converting enzyme inhibitor, enalapril, i n rabbits. Fund.Appl. Toxicol., 14r461-470.

Moe, O.W., K. Ujiie, R.A. Star , et al . (1993) Renin expression in renalproximal tubule. J. Clin. Invest., 9 1 774-779.Per lma n, J.M., and J.J. Volpe (1989) Neurologic complications of cap-topri l trea tme nt of neona tal hypertension. Pedia trics , 83:47-52.Pinkney, J.H., and J.S.Yudkin (1994)Antihypertensive drugs: Issuesbeyond blood pressure control. Prog. Cardiovasc. Dis., 36:397-415.

Piper, J.M., W.A. Ray, and F.W. Rosa (1992) Pregnancy outcome fol-lowing exposure to angiotensin-converting enzyme inhibitors. Ob-stet. Gynecol., 80:429-432.

Plouin, P.F., and C. Tchobroutsky (1985) Inhibition de l'enzyme deconversion de l'angiotensine au cows de la grossesse humaine:Quinze cas. Presse Med., 14t2175-2178.

Pryde , P.G., A.B. Sedman, C.E. Nugen l, and M. Barr (1993) Angioten-sin-converting enzyme inhibito r fetopathy. d . Am. SOC.Nephrol.,3t1575-1582.

Ravid, M., H. Savin, I. Ju tr in , T. Bental, B. Katz, and M. Lishner(1993) Long-term stabilizing effect of angiotensin-convert ing en-zyme inhibition on plasma creatinine and on proteinuria in nor-motensive type-I1 diabetic pat ien ts. Ann. Int. Med. 118.577-581.

Restaino, I., B.S. Kapla n, P. Kaplan, H.K. Rosenberg , C. Witzleben,and N. Roberts (1991) Renal dysgenesis in a monozygotic twin:Association with in utero exposure to indomethacin. Am. J. Med.Genet., 39.252-257.

Rigat, B., C. Hubert, F. Alhenc-Gelas, F. Cambien, P. Corvol, and F.Soubrier (1990) An insertion deletion polymorphism in angiotensinI converting enzyme gene accounting for half the variance of serumenzyme levels. J. Clin. Invest., 861343-1346.

Robertson, R.T., D.H. Minsker, and D.L. Bokelman (1986) MK-421(enalapril maleate): Late gestation and lactation study in r ats. Jpn .Pharmacol. Ther., 14:43-55.

Elodicio, J.L., and L.M. Ruilope (1993) Angiotensin-converting en-zyme inhibitors in the t reatm ent of mild arter ial hypertension.Clin. Exp. Hyper tens., 15:1277-1289.

Rosa, F.W., L.A. Bosco, C. Fossum-Graham, J.B. Mils tien, M. Dreis,and J . Creamer (1989) Neonatal anuria with mater nal angiotensin-converting enzyme inhibition. Obstet. Gynecol., 74:371-374.

384-388.

Rothberg, A.D., and R. Lorenz (1984) Can captopril cause fetal andneonatal renal failure? Pediatr. Pharmacol., 4t189-192.

Rudolph, A.M., M.A. Heyman, K.A.W. Teramo, C.T. Barrett, andN.C.R. Raiha (1971) Study on the circulation of the previable hu-man fetus. Pediatr. Res., 5t452-465.

Sassano, P., G. Chatellier, E. Billaud, F. Alhenc-Gelas, P. Corvol, andJ. Menard (1987) Treatment of mild to moderate hypertension withor without t he converting enzyme inhibitor enalapril: Resul ts of asix-month double-blind tr ial . Am. J . Med., 83t227-235.

Schlueter, W., T. Keilani, and D.C. Batlle (1994) Tissue renin an-giotensin systems: Theoretical implications for t he development ofhyperkalemia using angiotensin-converting enzyme inhibitors.Am. J. Med. Sci., .107~Suppl. ):S81486.

Schubiger,G. ,G . Flury, and J. Nussberger (19881 Enalapril for preg-nancy-induced hypertension: Acute renal failure in the neonate.Ann. Int. Med., 108:215-216.

Scott, A.A., and D.M. Purohit (1989) Neonatal renal failure: A com-plication of materna l antihypertensive therapy. Am. J. Obstet. Gy-necol., 160t1223-1224.

Shanmugam, S., C. Monnot, P. Corvol, and J.M. Gasc (1994) Distr i-bution of type 1angiotensin I1receptor subtype messenger RNAs inthe rat fetus. Hypertension, 23:137-141.

Simeoni,U., J. Messer,P.Weisburd, J . Haddad, and L . Willard (1989)Neonatal renal dysfunction and intrauterine exposure to prosta-glandin synthesis inhib itors. Eu r. J . Pediatr., I48:371-373.Tack, E.D., and J.M. P erlm an 17988) Renal failu re in sick hyperten-sive premature infants receiving captopril ther apy. J . Pediatr.,112:805-810.

Thorpe-Beeston, J.G ., N.A. Armar, M. Dancy, G.W. Cochrane, G.Ryan, a nd C.H. Rodeck (19931 Pregnancy and ACE inhibitors. Rr.J.Obstet. Gynaecol., IUU:692-693.

Thurston, H. (1992)Angiotensin-converting enzyme inhibition a s 1st-line treatment for hypertension. Clin. Exp. Pharmacol. Physiol.,Suppl. 19t67-71.

Todd, P.A., and R.C. Heel (1986) Enalapril: A review of its pharma-codynamic properties, and therapeut ic use in hypertension and con-gestive heart failure. Drugs, 3It198-248.

Vanhoutte, P.M., C.M. Boulanger, S.C. Illiano, T. Nagao, M. Vidal,and J.V. Mombouli (1993) Endothelium-dependent effects of con-verting-enzyme inhibitors. J . Cardiovasc. Pharmacol.,2XSuppl.51:SlO-Sl6.Viberti, G. ,C.E. Mogensen, L.C. Groop, et al. (1994) Effect of captoprilon progression t o clinical proteinuria in patients with insulin-de-pendent diabetes-mellitus and microalbuminuria. JAMA, 271275-279.

Vidt, D.G., E.L. Bravo, and F.M. Fouad (1982) Captopril. N. Eng l. J.Med., 306:214-219.

Voyer, L.E., R. Dru t, and J. H. Mendez (1994) Fetal renal maldevel-opment with oligohydramnios following mat ernal use of piroxicam.Pediatr. Nephrol., 8t592-594.

Wells, T.G., T.E. Bunchman, a nd G.L. Kea rns (1990) Treatm ent ofneonatal hypertension with enalapril. J. Pediatr., 7:664-667.

Williams, G.H. (1988)Converting enzyme inhibitors i n t he t reatmentof hypertension. N. Engl. J. Med., 319:1517-1524.Wood,E.G., T.E. Bunchman, and R.E. Lynch (1991) Captopril- induced

reversible acute renal failure in an infant with coarctation of theaort a. Pediatrics, 88:816-818.Yosipiv, I.V., S. Dipp, and S.S. Eldahr (1994) Ontogeny of somaticangiotensin-converting enzyme. Hypertension, 23t369-374.