Indomethacin and the fetal renal nonfunction syndrome

Acetylsalicylic acid interferes with embryonic kidney growth and development by a prostaglandin-independent mechanism

AIM

To evaluate the effects of the non-selective, non-steroidal anti-inflammatory drug (NSAID) acetylsalicylic acid (ASA), on ex vivo embryonic kidney growth and development.

METHODS

Pairs of fetal mouse kidneys at embryonic day 12.5 were cultured ex vivo in increasing concentrations of ASA (0.04-0.4 mg/mL) for up to 7 d. One organ from each pair was grown in control media and was used as the internal control for the experimental contralateral organ. In some experiments, organs were treated with ASA for 48 h and then transferred either to control media alone or control media containing 10 μmol/L prostaglandin E₂ (PGE₂) for a further 5 d. Fetal kidneys were additionally obtained from prostaglandin synthase 2 homozygous null or heterozygous (PTGS2^-/- and PTGS2^-/+) embryos and grown in culture. Kidney cross-sectional area was used to determine treatment effects on kidney growth. Whole-mount labelling to fluorescently detect laminin enabled crude determination of epithelial branching using confocal microscopy.

RESULTS

Increasing ASA concentration (0.1, 0.2 and 0.4 mg/mL) significantly inhibited metanephric growth (P < 0.05). After 7 d of culture, exposure to 0.2 mg/mL and 0.4 mg/mL reduced organ size to 53% and 23% of control organ size respectively (P < 0.01). Addition of 10 μmol/L PGE₂ to culture media after exposure to 0.2 mg/mL ASA for 48 h resulted in a return of growth area to control levels. Application of control media alone after cessation of ASA exposure showed no benefit on kidney growth. Despite the apparent recovery of growth area with 10 μmol/L PGE₂, no obvious renal tubular structures were formed. The number of epithelial tips generated after 48 h exposure to ASA was reduced by 40% (0.2 mg/mL; P < 0.05) and 47% (0.4 mg/mL; P < 0.01). Finally, growth of PTGS2^-/- and PTGS2^+/- kidneys in organ culture showed no differences, indicating that PTGS2 derived PGE₂ may at best have a minor role.

CONCLUSION

ASA reduces early renal growth and development but the role of prostaglandins in this may be minor.

Nonsteroidal antiinflammatory drugs, cyclooxygenase-2, and the kidneys

Nonsteroidal antiinflammatory drugs (NSAIDs) are one of the most commonly used classes of medications in the world, which function by inhibiting the cyclooxygenase (COX) enzymes and downregulating the inflammatory pathway. COX enzymes are constitutively expressed in the kidneys and function to maintain a homeostatic environment in terms of maintaining the glomerular filtration rate, blood pressure, sodium, water, and osmotic regulation. When the COX enzymes are inhibited by NSAIDs, a multitude of renal and vascular complications occur. This article aims to enlighten primary care physicians of the complications that arise with NSAIDs from a renal perspective and to present some management strategies.

Renal effects of ibuprofen for the treatment of patent ductus arteriosus in premature infants

In recent years ibuprofen has been proposed for the treatment of patent ductus arteriosus (PDA) as it has been proved to be equally as effective as indomethacin and shows fewer cerebral blood flow, intestinal and renal hemodynamic effects. A number of studies and several meta-analyses comparing both drugs are now available that debate whether indomethacin or ibuprofen should be used for PDA prophylaxis or closure. This review examines the available knowledge on the specific issue of the effects of ibuprofen on kidney function, as improved renal tolerance is a major argument in favor of its use in the routine treatment of PDA. There is sufficient evidence to consider that ibuprofen, at the currently proposed dosing regimen, has a similar efficacy to indomethacin but is better tolerated by the neonatal kidney when employed for the treatment of established PDA. However, adverse effects of ibuprofen have been evidenced both in trials on the use of ibuprofen for the prevention of PDA and of intraventricular hemorrhage-periventricular hemorrhage (IVH-PVH), and in experimental studies on a neonatal, anesthetized animal model. Thus ibuprofen, as with other cyclooxygenase (COX) inhibitors, may not be exempt from causing renal adverse effects, especially in circumstances when renal prostaglandin activation is maximal (i.e., when administrated early after birth, in more immature patients and in certain situations such as in the anesthetized rabbit). However, although the issue has been addressed extensively in the last decades, there is insufficient evidence that therapeutic intervention in PDA is beneficial in terms of mortality or clinically significant morbidity outcomes. Studies aimed at resolving this key issue are still needed.

Physiological regulation of prostaglandins in the kidney

Cyclooxygenase-derived prostanoids exert complex and diverse functions within the kidney. The biological effect of each prostanoid is controlled at multiple levels, including (a) enzymatic reactions catalyzed sequentially by cyclooxygenase and prostanoid synthase for the synthesis of bioactive prostanoid and (b) the interaction with its receptors that mediate its functions. Cyclooxygenase-derived prostanoids act in an autocrine or a paracrine fashion and can serve as physiological buffers, protecting the kidney from excessive functional changes during physiological stress. Through these actions, prostanoids play important roles in maintaining renal function, body fluid homeostasis, and blood pressure. Renal cortical COX2-derived prostanoids, particularly PGI2 and PGE2, play critical roles in maintaining blood pressure and renal function in volume-contracted states. Renal medullary COX2-derived prostanoids appear to have an antihypertensive effect in individuals challenged with a high-salt diet. Loss of EP2 or IP receptor is associated with salt-sensitive hypertension. COX2 also plays a role in maintaining renal medullary interstitial cell viability in the hypertonic environment of the medulla. Cyclooxygenase-derived prostanoids also are involved in certain pathological processes. The cortical COX2-derived PGI2 participates in the pathogenesis of renal vascular hypertension through stimulating renal renin synthesis and release. COX-derived prostanoids also appear to be involved in the pathogenesis of diabetic nephropathy. COXs, prostanoid synthases, and prostanoid receptors should provide fruitful targets for intervention in the pharmacological treatment of renal disease.

Urinary prostaglandin E2 in the newborn and infant

Prostaglandin E(2) (PGE(2)) belongs to a family of biologically active lipids derived from the 20-carbon essential fatty acids. Renal PGE(2) is involved in the development of the kidney; it also contributes to regulate renal perfusion and glomerular filtration rate, and controls water and electrolyte balance. Furthermore, this mediator protects the kidney against excessive functional changes during the transition from fetal to extrauterine life, when it counteracts the vasoconstrictive effects of high levels of angiotensin II and other mediators. There is evidence that PGE(2) plays an important pathophysiological role in neonatal conditions of renal stress, and in congenital or acquired nephropaties. Thus, measurement of urinary PGE(2) as an index of renal synthesis of this primary prostaglandin may represent a non-invasive and sensitive method of investigating the homeostatic function of the kidney in early life. The aim of this literature review is to examine urinary PGE(2) as a non-invasive marker of renal homeostasis in the newborn and infant under both physiological and pathological conditions, or during treatments with widely used, potentially toxic drugs.

Prescription drugs and pregnancy

Prescribing drugs in pregnancy is an unusual risk-benefit situation. Drugs that may be of benefit or even life-saving to the mother can deform or kill the fetus. However, the risk to the fetus should not be exaggerated. There are only approximately 20 drugs or groups of drugs which are known to cause birth defects in humans. For one of these drugs to cause birth defects, a number of criteria must be fulfilled. The drug exposure must take place at a critical stage of pregnancy and the dose must be high enough to cause a threshold of exposure for an appropriate duration of time. For most of the known human teratogens, > 90% of pregnancies exposed during the first trimester result in normal offspring. Although only a few drugs are known to cause birth defects in humans, uncertainty about the safety of the majority may lead to underprescribing for pregnant women and women of childbearing age. Epidemiological studies of pregnancy outcome after specific drug exposures are often superficially reassuring, but most are severely limited in their power to detect adverse outcomes. Safety in animal studies may also be reassuring but species differences demand caution in this interpretation. Concerns about prescription drugs in the first trimester, when they can cause birth defects, are mostly quite different to concerns about use in the second and third trimesters. As the fetal organ systems mature, the fetus can be affected by the pharmacological activity of the drug in the same way as the mother. Many drugs have pharmacological effects on the fetus in the second and third trimesters but in most cases, they are well recognised and can be managed or avoided. The material presented in this paper is mostly concerned with the 'risks' associated with drugs in pregnancy. No attempt has been made to quantitate the possible benefits to the mother or fetus. Communicating the risk-benefit situation to the patient is always a challenge for physicians with limited time and sometimes limited knowledge. Fear of litigation is an unfortunate and an unwanted parameter in the assessment. Better knowledge of the parameters that determine teratogenicity may allow physicians to feel more confident in assessing the risks and benefits associated with prescribing in pregnancy.

Risk of persistent renal insufficiency in premature infants following the prenatal use of indomethacin for suppression of preterm labor

OBJECTIVE

To determine the risk of persistent renal insufficiency (RI) in premature infants following the use of antenatal indomethacin for suppression of preterm labor.

STUDY DESIGN

This population-based, retrospective review consisted of infants admitted during a 5-year period (1994-1999) to a tertiary referral neonatal intensive care nursery. Data were reviewed on 37 infants whose mothers received indomethacin for tocolysis, and on 37 matched controls. Renal insufficiency was defined as infant creatinine (Cr) > or = 1.5 mg/dl (133 micromol/l) for > or = 1 day.

RESULTS

Infants whose mothers had received indomethacin for tocolysis were more likely than matched controls to have RI (9 of 37 vs 2 of 37, p = 0.04). Among infants of indomethacin-treated mothers with elevated Cr, serum Cr remained > or = 1.5 mg/dl (133 micromol/l) for a median of 6 days and > 1.0 mg/dl (88 micromol/l) for a median of 19 days. The peak Cr and length of elevation were closely correlated (r(2) = 0.63, p < 0.0001).

CONCLUSION

Antenatal indomethacin can result in significant, prolonged RI in the infant. It may pose important risks to renal function and homeostasis in premature infants.

Renal pathology resulting from PGHS-2 gene ablation in DBA/B6 mice

Kidneys of prostaglandin H synthase-2 (PGHS-2) null mice fail to develop normally, leading to renal insufficiency. We have found that in a mixed DBA/B6 background, the lack of a functional PGHS-2 gene causes less severe renal pathology than was reported previously for PGHS-2 null mice in a B6 genetic background. The increase in blood urea nitrogen in the DBA/B6 strain of PGHS-2 null mice was significantly lower than reported for B6 PGHS-2 null mice (200% versus 270%). Cystic changes in DBA/B6 PGHS-2 null mice were also less severe. The DBA/B6 PGHS-2 null adult mice did not die from renal failure, unlike their B6/PGHS-2 counterparts that showed excessive neonatal and adult deaths. Therefore, DBA/B6 PGHS-2 null may be highly suitable to study the functional consequences of the lack of PGHS-2 in the kidney due to their less severe pathology and greater survival.

Cyclooxygenase 2 and the kidney

Cyclooxygenase metabolizes arachidonic acid to a family of bioactive fatty acids designated prostaglandins. Two isoforms of cyclooxygenase exist, designated COX1 and COX2. These isoforms are expressed in distinct but important areas of the kidney. COX1 predominates in vascular smooth muscle and collecting ducts, whereas COX2 predominates in the macula densa and nearby cells in the cortical thick ascending limb. COX2 is also highly expressed in medullary interstitial cells. Whereas COX1 expression does not exhibit dynamic regulation, COX2 expression is subject to regulation by several environmental conditions, including salt intake, water intake, medullary tonicity, growth factors, cytokines, and adrenal steroids. Recently, COX2-selective non-steroidal anti-inflammatory drugs have become widely available. Many of the renal effects of non-selective non-steroidal anti-inflammatory drugs (including sodium retention, decreased glomerular filtration rate, and effects on renin-angiotensin levels) appear to be mediated by the inhibition of COX2 rather than COX1. Therefore, in contrast to the gastrointestinal-sparing effects of COX2-selective non-steroidal anti-inflammatory drugs, when considering the kidney, the same caution must be applied when using COX2-selective inhibitors as has been used with traditional non-selective non-steroidal anti-inflammatory drugs.

NSAID-induced nephrotoxicity from the fetus to the child

In this review we report data available from the literature on the use of nonsteroidal anti-inflammatory drugs (NSAIDs) and the development of nephrotoxicity in the fetus, neonates and children. Up to the present day, several cases of severe and sometimes irreversible renal insufficiency have been described in neonates exposed to indomethacin prenatally or in the first days of life for treatment of patent ductus arteriosus (PDA). Until now, very few studies have been carried out on alternative treatments for PDA in preterm infants; ibuprofen has been shown to be as effective as indomethacin in closing the ductus in this patient group without affecting renal function. In children, NSAID-induced renal failure is a rare event and is usually reversible after discontinuation of the drug. However, caution should be taken when NSAIDs are administered to individuals with preexisting renal problems or with other potentially nephrotoxic drugs. In these situations, new approaches such as cyclo-oxygenase-2 selective inhibitors or prostanoid receptor selective antagonists could lead to alternative therapies for use in paediatrics.

Cyclooxygenase-2-selective inhibitors impair glomerulogenesis and renal cortical development

BACKGROUND

Antenatal exposure to nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with renal dysgenesis in humans.

METHODS

These studies characterized cyclooxygenase-2 (COX-2) versus COX-1-selective inhibition on nephrogenesis in the rodent using histomorphometry, immunohistology, and in situ hybridization.

RESULTS

Administration of a COX-2-selective inhibitor (SC58236), started during pregnancy until weaning, significantly impaired development of the renal cortex and reduced glomerular diameter in both mice and rats. An identical phenotype was demonstrated in COX-2 -/- mice. In contrast to its effects on the developing kidney, a COX-2 inhibitor had no effect on glomerular volume in adult mice. This effect was specific for COX-2 because maternal administration of a COX-1-selective inhibitor (SC58560) did not affect renal development despite significantly inhibiting gastric mucosal prostaglandin E2 (PGE2) synthesis in pups. The expression of COX-2 immunoreactivity peaked in the first postnatal week and was localized to S-shaped bodies and the macula densa in the cortex. Treatment with a COX-2 inhibitor during this period (from postnatal day 0 to day 21) severely reduced glomerular diameter, whereas treatment limited to pregnancy did not affect glomerular size.

CONCLUSION

These data demonstrate an important role for COX-2 activity in nephrogenesis in the rodent, and define a specific time period of susceptibility to these effects.

Cyclooxygenase-2-selective inhibitors impair glomerulogenesis and renal cortical development

BACKGROUND

Antenatal exposure to nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with renal dysgenesis in humans.

METHODS

These studies characterized cyclooxygenase-2 (COX-2) versus COX-1-selective inhibition on nephrogenesis in the rodent using histomorphometry, immunohistology, and in situ hybridization.

RESULTS

Administration of a COX-2-selective inhibitor (SC58236), started during pregnancy until weaning, significantly impaired development of the renal cortex and reduced glomerular diameter in both mice and rats. An identical phenotype was demonstrated in COX-2 -/- mice. In contrast to its effects on the developing kidney, a COX-2 inhibitor had no effect on glomerular volume in adult mice. This effect was specific for COX-2 because maternal administration of a COX-1-selective inhibitor (SC58560) did not affect renal development despite significantly inhibiting gastric mucosal prostaglandin E2 (PGE2) synthesis in pups. The expression of COX-2 immunoreactivity peaked in the first postnatal week and was localized to S-shaped bodies and the macula densa in the cortex. Treatment with a COX-2 inhibitor during this period (from postnatal day 0 to day 21) severely reduced glomerular diameter, whereas treatment limited to pregnancy did not affect glomerular size.

CONCLUSION

These data demonstrate an important role for COX-2 activity in nephrogenesis in the rodent, and define a specific time period of susceptibility to these effects.

Renal tubular dysgenesis in twin-twin transfusion syndrome

In twin-twin transfusion syndrome (TTTS), the disparity in circulation is reflected in discordant fetal growth, urine output, and amniotic fluid accumulation. The effect of uneven shunting of the growth factor and nutrient-rich vasculature on development and differentiation of the kidney has not been well studied. We analyzed renal tubular growth and differentiation in 25 fetal autopsies with TTTS (13 donors and 12 recipients, including 9 sibling pairs) between 18 and 33 weeks gestation. Immunohistochemical markers for fumarylacetoacetate hydrolase (FAH), Leu-M1, and Lotus tetragonolobus (LTA) were used to identify proximal convoluted tubules, and epithelial membrane antigen (EMA) was used to demonstrate distal convoluted and collecting tubules. FAH appeared to be more specific and reliable than either Leu-M1 or LTA in the identification of proximal tubules. Donors tended to demonstrate a paucity of proximal tubules with crowding of glomeruli characteristic of renal tubular dysgenesis (RTD). The degree of dysgenesis was greater in later gestations and associated with more severe growth restriction. Donors in TTTS are at risk for the development of RTD. Several authors suggest ischemia as the underlying cause of "acquired" RTD. However, in this setting there is no evidence of cell death or necrosis, and we suggest that hypoperfusion leading to decreased glomerular filtration is the underlying etiology, with the severity of RTD related to the degree of shunting.

Clinical utility of amniotic fluid volume assessment

Abnormal amniotic fluid volume is associated with increased maternal risk and perinatal morbidity and mortality. Until the advent of ultrasonography, the invasive nature of amniotic fluid volume assessment limited its clinical utility. Refinements in quantifying the noninvasive sonographic assessment of oligohydramnios and hydramnios have improved the ability of clinicians to identify at-risk pregnancies. This article reviews the available methods of amniotic fluid volume assessment and outlines a comprehensive approach to sonographic screening and monitoring.

Treatment of symptomatic polyhydramnios with indomethacin

Polyhydramnios carries a poor fetal prognosis with an expected neonatal death rate of nearly 30%. Approximately one-fourth of this perinatal mortality is a result of the effects of prematurity. The poor outcome with usual management of polyhydramnios led us to introduce the therapeutic use of prostaglandin synthetase inhibitors so as to decrease amniotic fluid volume and to prevent premature labor. Twenty-two women (20 singleton and two twin pregnancies) were included in a retrospective study from 1983 to 1992. Indomethacin was given at a dose of 3 mg/kg/day. Treatment was started at 28.2 +/- 3.8 weeks of amenorrhea and discontinued after 35 weeks. We observed a significant effect of indomethacin on amniotic fluid volume and avoided severe preterm delivery in all patients. Mean gestational age at birth was 37.5 +/- 1.6 weeks of amenorrhea (range 35.5-39 weeks). We did not observe any maternal or neonatal side effects of indomethacin therapy. However, we reported three neonatal deaths out of 24 infants: two related to undiagnosed fetal anomalies (one Nager syndrome and one cerebral malformation) and one related to umbilical cord entanglement in a monoamniotic twin pregnancy. Since our first report, several open studies supporting our data have been published. However, although indomethacin appears to be effective in the treatment of polyhydramnios, our goal is to analyze efficacy and side effects, so as to define conditions of clinical use.

Persistent anuria, neonatal death, and renal microcystic lesions after prenatal exposure to indomethacin

OBJECTIVE

The purpose of this study is to emphasize the high risk of renal failure and severe morphologic changes related to prolonged prenatal exposure to indomethacin.

STUDY DESIGN

Referred renal specimens from six anuric neonates exposed in utero to indomethacin were studied. Clinical charts were retrospectively reviewed. Indomethacin dosages varied from 150 to 400 mg daily, and the drug was given for a 2- to 11-week period, until birth.

RESULTS

All infants died in anuria, 4 of them after 7 to 39 days on peritoneal dialysis. In 5 infants cystic dilatations of superficial nephrons were associated with ischemic changes of the deep cortex. By immunohistochemical analysis intrarenal renin content was increased in 4 of 5 patients.

CONCLUSION

Long-term indomethacin treatment during pregnancy may lead to the development of renal failure and irreversible renal damage with cystic dilatation of developing nephrons in an exposed fetus. Prior stimulation of the renin-angiotensin system may favor this complication.

Pregnancy in women with end-stage renal disease: treatment of anemia and premature labor

There is little experience with the use of various therapies in the end-stage renal disease patient who becomes pregnant. Erythropoietin for the treatment of anemia has become part of the standard treatment regimen of dialysis patients, but experience with its use in pregnancy is limited. We report five cases of its use in dialysis patients during pregnancy. We found no evidence that it crossed the placenta or that it made blood pressure control more difficult. We found that patients required a higher dose of erythropoietin to maintain hematocrit levels than they had before pregnancy. Another therapy involves the treatment for premature labor, which is the most common cause of pregnancy loss in dialysis patients. Two of our patients were successfully treated with indomethacin for premature labor. Both drugs are useful tools in the management of pregnant dialysis patients.

NSAIDs: maternal and fetal considerations

Nonsteroidal anti-inflammatory drugs (NSAIDs) gained popularity in the late 1970s. Inhibition of prostaglandin synthesis with indomethacin has been reported to be effective for prevention of labor and for treatment for symptomatic polyhydramnios. Concern about its possible constrictive effect on the fetal ductus arteriosus has limited its use in pregnancy. Maternal indomethacin therapy has also been associated with reduction in urine production in the fetus and with oligohydramnios. Obstetricians have discouraged pregnant women from taking analgesic doses of aspirin, mainly because of the availability of paracetamol (acetaminophen), which causes less gastric irritation, but also because of fear of maternal and fetal hemorrhage and of possible premature closure of the ductus. These fears largely derive from studies on patients taking large doses and from extrapolation from other NSAIDs. The likelihood that treatment with 60-75 mg/day of aspirin markedly reduces the incidence of preeclampsia and fetal intrauterine growth retardation makes it important to reexamine its use. This review describes the pharmacology and pharmacokinetics of aspirin with particular reference to pregnancy and considers teratogenesis, prolongation of pregnancy and labor, maternal bleeding, fetal and neonatal bleeding, possible effects on the ductus arteriosus and pulmonary circulation, and possible nonspecific effects on intelligence and breast feeding and acute toxicity in the neonate.

Urinary and cardiovascular responses to indomethacin infusion in the ovine fetus

OBJECTIVE

Our objective was to explore the urinary and cardiovascular responses of the near-term ovine fetus to plasma indomethacin levels similar to those in the human neonate undergoing indomethacin therapy.

STUDY DESIGN

Chronically catheterized ovine fetuses between 125 and 139 days of gestation were studied. After a 1-hour control period we gave a bolus of 0.35 mg/kg estimated fetal weight of indomethacin into a fetal vein, followed by a 0.017 mg/kg/min continuous infusion for 5 hours (n = 9). Results were compared with a vehicle-infusion-only group (n = 10).

RESULTS

During the first 3 hours of indomethacin infusion, fetal urinary output was increased by an average of 84.9% +/- 55.6% (analysis of variance, p less than 0.01). Urinary osmolality and sodium and chloride concentrations underwent sustained increases throughout the infusion period (p less than 0.001). Sodium excretion increased by 212% +/- 111% (p less than 0.05). Fetal arterial and venous pressures increased (p less than 0.001), and the change in urinary flow correlated positively with the change in arterial pressure (R = 0.55, p = 0.014). Fetal heart rate increased by 10% +/- 4% 1 hour after the bolus and remained elevated throughout the remainder of the infusion relative to vehicle-infused animals (p less than 0.001). Vehicle infusion had no effect on any fetal variable.

CONCLUSIONS

This study does not support the hypothesis that indomethacin acutely reduces urinary flow rate in the late-gestation ovine fetus. Further, the observed urinary flow increases may be mediated in part by a pressure diuresis.

Oligohydramnios associated with prostaglandin synthetase inhibitors in preterm labour

The use of prostaglandin synthetase inhibitors (indomethacin and ibuprofen) for greater than 72 h in women with preterm labour (n = 67) was significantly (P less than 0.001) associated with ultrasound-recorded oligohydramnios compared with the use of ritodrine or magnesium sulphate (control group, n = 67). Oligohydramnios developed in 26 of 37 women (70%) treated with indomethacin and 8 of 30 women treated with ibuprofen (27%) (P less than 0.01). Oligohydramnios developed in only two control subjects and one of these pregnancies was associated with intrauterine growth retardation. All 34 cases of oligohydramnios in the treatment group resolved after stopping medication, whereas the two in the control group did not resolve. No instances of renal failure, premature closure of the ductus arteriosus, pulmonary hypertension, or bleeding disorders were noted in any of the treated or control infants.

Oligohydramnios, renal insufficiency, and ileal perforation in preterm infants after intrauterine exposure to indomethacin

Three preterm infants exposed antenatally to indomethacin developed a characteristic syndrome consisting of edema and hydrops with a bleeding disorder at birth, oliguric renal failure during the first 3 postnatal days, and acute pneumoperitoneum resulting from localized ileal perforation(s) at the end of the first week of life. Despite the value of indomethacin for arresting preterm labor, the physician must take into account the potential hazards of drug toxicity.

Treatment of polyhydramnios with prostaglandin synthetase inhibitor (indomethacin)

Polyhydramnios, which is caused by an excess of amniotic fluid, was diagnosed uniformly in eight patients by ultrasound examination and clinical symptoms, namely, premature uterine contraction, abdominal pain, and respiratory discomfort that resulted from excessive pressure on the diaphragm. Fetal anomalies incompatible with extrauterine existence were excluded by serum alpha-fetoprotein levels, repeated sonography, amniocentesis, and chromosomal analyses. Patients were treated with 2.2 to 3.0 mg of indomethacin/kg body weight/day. The treatment was started at an average gestational age of 24 +/- 0.5 weeks and continued for 2 to 11 weeks. All patients had significant improvement with a simultaneous reduction in amniotic fluid volume, fundal height, and umbilical perimeter. All patients were delivered satisfactorily at an average of 39 weeks' gestation. The fetuses were within the normal body weight range of 2750 to 3600 gm and showed normal development during the neonatal period of 2 to 6 months. Before delivery without indomethacin, the polyhydramnios reformed. Adverse effects or intolerance to the drug was avoided by either a reduction in the dose or cessation of indomethacin therapy. Our studies suggest that indomethacin therapy is an effective way to manage pregnancy complicated by polyhydramnios.