- Email: [email protected]
maternal ratio of nitroglycerin, and effects in newborns
Nitroglycerin application during cesarean delivery: Plasma levels, fetal/maternal ratio of nitroglycerin, and effects in newborns Matthias David, MD,a Matthias M. Walka, MD,b Bernhard Schmid, MD,d Pranav Sinha, MD,c Siegfried Veit, MD,e and Werner Lichtenegger, MDa Berlin and Neu-Ulm, Germany OBJECTIVE: We sought to investigate maternal and fetal nitroglycerin metabolization and to assess the clinical condition of neonates after intravenous nitroglycerin application during cesarean delivery. STUDY DESIGN: At the time of the uterine puncture incision, either 0.25 mg or 0.5 mg nitroglycerin or a physiologic sodium chloride solution was administered as an intravenous bolus. Plasma concentrations of nitroglycerin and its metabolites were measured in maternal venous blood and in umbilical blood samples taken immediately after cord clamping. Arterial blood pressure, pulse rates, and Apgar scores were recorded for the neonates 1, 5, and 10 minutes after birth. RESULTS: Sixty-two patients were included in the pharmacokinetic study. Median maternal plasma levels 1 and 5 minutes after injection of 0.5 mg nitroglycerin were 80 and 3.2 ng/mL, respectively; median maternal plasma levels 1 and 5 minutes after injection of 0.25 mg nitroglycerin were 38 and 1.2 ng/mL, respectively. In the umbilical vein 1 minute after application of 0.5 mg or 0.25 mg nitroglycerin, the plasma levels were 0.41 and 0.09 ng/mL, respectively, and in the umbilical artery they were 0.03 and 0.008 ng/mL, respectively. Circulatory parameters and Apgar scores in the neonates did not differ significantly from those found in the placebo group. CONCLUSION: The level of nitroglycerin in umbilical plasma was two to three orders of magnitude lower than that found in maternal plasma and clearly in a subtherapeutic range. There was no indication that prenatal application of nitroglycerin to facilitate obstetric management is hazardous for neonates. (Am J Obstet Gynecol 2000;182:955-61.)
Key words: Fetal-maternal plasma levels of nitroglycerin, glycerol trinitrate, cesarean delivery
Over the last decade, there have been several reports of successful obstetric use of nitroglycerin as a tocolytic. Nitroglycerin, which is also known as glycerol trinitrate, may be administered before, during, or after delivery and is well suited for use as a short-term tocolytic agent before external or internal change to extract a retained placenta or to correct a uterine inversion, as well as during cesarean delivery.1-4 In the clinical part of this study, two dosages of glycerol trinitrate (Nitro Pohl Infus; G. PohlBoskamp GmbH & Co, Hohenlockstedt, Germany) were From the Departments of Obstetrics and Gynecologya and Neonatologyb and the Institute for Clinical Chemistry and Biochemistry,c University Hospital Charité, Humboldt University, Berlin; A.A.I. Applied Analytical Industries Deutschland GmbH & Co, Neu-Ulmd; and the Department of Anesthesiology and Intensive Care, Friedrichshain Hospital, Berlin.e Received for publication March 15, 1999; revised September 14, 1999; accepted November 8, 1999. Reprint requests: Matthias David, MD, Charité, Campus VirchowKlinikum Klinik für Frauenheilkunde und Geburtshilfe, Augustenburger Platz 1 13353 Berlin, Germany. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/1/104365 doi:10.1067/mob.2000.104365
investigated for their clinical effect on uterine musculature relaxation during cesarean deliveries in comparison with a placebo. Glycerol trinitrate crosses the placenta in both animals and human subjects. In sheep a 1:25 ratio of arterial glycerol trinitrate concentrations between fetus and mother has been reported.5 Until now, however, there has been no information about the extent to which glycerol trinitrate passes from the maternal to the fetal circulation in human subjects. In the adult liver glycerol trinitrate is rapidly hydrolyzed by the glutathione system to 1,2- or 1,3-glyceryl dinitrate.6 The enzyme driving this process is glutathione S-transferase.6 However, the composition of isoenzymes in glutathione S-transferase is tissue dependent and changes with development. In adults α, θ, and µ subtypes predominate, but in the fetus and in neonates glutathione S-transferase is mainly placental. Because the kinetic properties of placental glutathione S-transferase differ from those of the adult isoforms,7 the question arises as to whether the fetus is able to metabolize significant amounts of glycerol trinitrate. Our study is the first 955
956 David et al
April 2000 Am J Obstet Gynecol
Fig 1. Nitroglycerin concentration in maternal plasma (0.25-mg group). y Axis is logarithmic.
Fig 2. Nitroglycerin concentration in maternal plasma (0.5-mg group). y Axis is logarithmic.
to determine the extent of fetomaternal glycerol trinitrate passage and fetal metabolization of glycerol trinitrate. The effect of glycerol trinitrate administration on the ease of fetal extraction in the study has already been reported.3
The pregnant women were assigned to treatment groups on the day surgery was scheduled according to a computer-generated list of random numbers that was kept in an area away from clinical care and was unavailable to surgeons and caregivers. The women were assigned to either the placebo group or one of the two nitroglycerin groups in the chronologic sequence of the cesarean deliveries. Both study drug and placebo were administered intravenously. The solutions were prepared by the responsible investigator 1 hour before cesarean delivery. All patients underwent standard induction of general anesthesia during the operation. Intravenous administration of 0.25 mg or 0.5 mg glycerol trinitrate (in 4.75 mL or 4.5 mL physiologic sodium chloride solution) or placebo (5 mL physiologic sodium chloride solution) was carried out at the time of the puncture incision in the lower uterine segment by the responsible investigator. The line was then flushed for 1 minute with 100 mL physiologic sodium chloride solution. Intravenous access was provided for application of anesthesia in all patients, either on the back of the right hand or on the right wrist. Venous blood was withdrawn from all patients in the glycerol trinitrate groups 1 and 5 minutes after administration of nitroglycerin through a second catheter fitted before the operation, usually in an antecubital vessel in the same arm. These samples were collected directly in precooled, specially laminated, glass vacuum tubes with no additional anticoagulant being used. Blood samples were centrifuged immediately in a refrigerated centrifuge, and the plasma was deep-frozen and stored at –80°C. In addition, after all deliveries in the glycerol trinitrate groups, samples of arterial and venous umbilical blood
Material and methods This double-blind, randomized, controlled study was carried out in the gynecology departments of Virchow Hospital and Charité Hospital (both part of Humboldt University, Berlin, Germany). Two doses of glycerol trinitrate, 0.25 mg and 0.5 mg, were investigated. Physiologic sodium chloride solution (0.9%) was used as a placebo control. A total of 97 patients scheduled for elective cesarean deliveries were included in the study during the period between June 1994 and July 1996. Inclusion criteria included maternal age of 18 to 45 years, gestational age of 34 to 42 completed weeks, singleton pregnancy, planned abdominal cesarean delivery irrespective of fetal position, and suitability for anesthesia as per American Society of Anesthesiologists status I and II. Exclusion criteria included pregnancy-induced hypertension, preeclampsia, heart disease or previous cardiac surgery, intolerance to nitrates, anemia, lung disease, hypotension, shock, alcohol abuse, serious systematic diseases like insulin-dependent diabetes mellitus, and intravenous tocolysis in the 48 hours before cesarean delivery. Before the start of the study, approval was obtained from the Ethics Committee of the Virchow Clinic, Berlin. Before the operation, all patients were fully informed of the nature and scope of the study, as well as the potential benefits and risks. All patients signed a declaration of consent.
Volume 182, Number 4 Am J Obstet Gynecol
were withdrawn immediately after the umbilical cord was cut. These blood samples were also handled as described above. Blood samples from the umbilical cord could not be taken at defined intervals after glycerol trinitrate application to the mother: Sampling times depended on the interval between application of glycerol trinitrate during uterine puncture incision and umbilical cord clamping after birth, which varied considerably. To allow a structured evaluation of the results, sampling times were grouped as follows: T1, interval between administration of glycerol trinitrate and cord clamping of 1 minute (defined as 30-90 seconds); T2, 2 minutes (91-150 seconds); T3, 3 minutes (151-210 seconds); and T4, 4 minutes (211-270 seconds). In both the 0.25-mg and 0.5-mg collectives, half the umbilical blood samples were in group T1. The concentrations of glycerol trinitrate and its metabolites, 1,2-glyceryl dinitrate and 1,3-glyceryl dinitrate, were measured in a reference laboratory (A.A.I. GmbH, Neu-Ulm, Germany). Suitable aliquots of the samples (standard sample volume, 1 mL) were prepared by fluid extraction and concentration. Analysis was done by means of a validated gas chromatography–mass spectroscopy method with a Trio 2000 GC/MC Quadrupol (Micromass, GB) system. After separation of the three components (glycerol trinitrate and 1,2-glyceryl dinitrate and 1,3-glyceryl dinitrate) by gas chromatography, the number of negative ions was determined after appropriate chemical ionization. The concentrations of all three analytes were measured simultaneously. The lower limit of quantitation of the method is 6 pg/mL for glycerol trinitrate and 25 pg/mL for 1,3-glyceryl dinitrate and 1,2glyceryl dinitrate. The upper limit is 2.4 ng/mL for glycerol trinitrate and between 3.0 and 9.0 ng/mL for the two metabolites. Samples with higher concentrations were diluted as necessary. The accuracy and precision of the method (both within and between series) was clearly better than ±10% over the entire range of concentrations. All infants delivered in the course of the study were examined by a neonatologist immediately after birth and were monitored closely during the postnatal adaptation phase. Arterial umbilical pH values were determined immediately after the umbilical cord was cut. Apgar scores, neonatal blood pressure, and pulse rates were documented 1, 5, and 10 minutes after birth. Blood pressure was measured by using oscillometry with a Dinamap 8100 (Critikon Inc, Norderstedt, Germany). Results Sixty-two patients were included in the pharmacokinetic evaluation: 30 of 32 patients received 0.25 mg glycerol trinitrate and 32 of 34 patients received 0.5 mg glycerol trinitrate. The remaining 4 patients receiving glycerol trinitrate were excluded from evaluation be-
David et al 957
Fig 3. Nitroglycerin concentrations 1 minute after intravenous bolus application to mother. y Axis is logarithmic.
cause of technical problems in blood sampling or preparation for analysis. Thirty-one patients received placebo. There were no significant differences between the 3 treatment groups with regard to age, maternal body habitus, height, weight, ethnicity, smoking status, duration of surgery, number of previous cesarean deliveries, fetal position, fetal size, gravidity, parity, gestational age, blood pressure, and pulse rate before induction of anesthesia and maternal hemoglobin, hematocrit, thrombocytes, prothrombin time, and partial thromboplastin time values on the day of delivery (Table I). The study population included only healthy pregnancies, as described by the inclusion and exclusion criteria. In the glycerol trinitrate groups, maternal blood samples were available for a total of 57 patients for both sampling times (1 and 5 minutes after application of glycerol trinitrate) and 5 patients for one of the sampling times. Plasma concentrations of glycerol trinitrate at each of the time points are shown in Figs 1 and 2. In some cases surgical or anesthetic requirements meant that sampling was delayed by approximately 1 minute. These cases have been included in Figs 1 and 2. Umbilical blood samples were analyzed from 59 of 62 births. In 55 cases samples were available from both the umbilical artery and the umbilical vein; in 4 cases samples were only available from the umbilical vein. Because plasma concentrations of glycerol trinitrate and its metabolites are dependent on time, meaningful comparison of maternal and umbilical blood samples is only possible where samples were collected at approximately the same time. In this study the ratio of the plasma concentrations between maternal and umbilical venous blood could only be determined for time T1 (n = 32). Box-and-whisker plots of the glycerol trinitrate concentrations at this time point are shown in Fig 3. One minute after application, the median glycerol trinitrate concentration in maternal venous plasma was approximately 500 times greater than that in umbilical blood. Although
958 David et al
April 2000 Am J Obstet Gynecol
Table I. Patient characteristics (mothers) 0.25-mg nitroglycerin group (n = 30)
0.5-mg nitroglycerin group (n = 32)
Placebo group (n = 31)
29 163 82.5
30 162 75.2
29 163 78.8
29 (96.7%) 0 1 0 39.3 40 12.1 37 115/70
31 (96.9%) 1 0 0 37.5 40 11.6 34.5 112.5/70
29 (93.5%) 1 0 1 38.4 39 12.1 36.5 135/79
Age (y) Height (cm) Weight (kg) Race White Asian Afro-Caribbean Oriental Primiparous (%) Gestational age (wk)40 Preoperative hemoglobin (g/dL) Preoperative hematocrit (mL/dL) Preoperative blood pressure, mm Hg (systemic/diastolic) Data are presented as median or number and percent.
Table II. Characteristics of neonates
Apgar score 1 min 5 min 10 min post partum Umbilical pH value (arterial) Systolic blood pressure (mm Hg)* 1 min 5 min 10 min post partum Diastolic blood pressure (mm Hg)* 1 min 5 min 10 min post partum Pulse (beats/min) 1 min 5 min 10 min post partum
0.25-mg nitroglycerin group (n = 30)
0.5-mg nitroglycerin group (n = 32)
Control group (placebo; n = 31)
9 (8-9) 10 (9-10) 10 (10-10) 7.36 (7.35-7.38)
9 (7-9) 10 (9-10) 10 (10-10) 7.32 (7.28-7.35)
9 (8-9) 10 (9-10) 10 (10-10) 7.32 (7.28-7.37)
63 (54-71) 68 (58-72.5) 62 (56-70)
67 (57-77) 67 (53.5-76.5) 60.5 (49-70.5)
62 (49.5-75) 64 (59-79) 62.5 (54-72) 38 (35-50) 43 (33-54) 36 (31-42) 163 (127-186) 162 (137-177) 153 (140-173)
43 (38-47) 37.5 (33-47) 37 (30-44)
41 (36.5-56) 40 (34-48.5) 36.5 (31.5-42.5)
158 (100-173) 161 (134-171) 153 (135-166)
140 (100-161) 153 (140-166) 151 (144-165.5)
Results shown are median and 25th to 75th percentile. There were no significant differences between treatment and placebo groups. *Mean blood pressures (95% range) for term neonates with birth weight corresponding to gestational age are as follows: systolic, 64 mm Hg (54-77 mm Hg); diastolic, 38 mm Hg (28-48 mm Hg).19
some of the maternal blood samples were taken 2 minutes after application of glycerol trinitrate, the number of cases in which this coincided with umbilical samples at time T2 and later was very small, and statistical analysis of the T2, T3, and T4 data was not supported by the software used. The median ratio of plasma concentrations between venous and arterial umbilical blood was 7.4:1 (95% confidence interval, 1.96-20.18) for 0.5 mg of glycerol trinitrate and 1:1 (95% confidence interval, 0.28-3.33) for 0.25 mg of glycerol trinitrate, irrespective of the dose applied and the interval between application of glycerol trinitrate and collection of umbilical blood (Fig 4). Similar ratios were obtained for the glycerol trinitrate
metabolites. The overall median ratio of plasma concentrations was 7.5:1 for 1,2-glyceryl dinitrate and 8.8:1 for 1,3-glyceryl dinitrate. The postnatal condition of the neonates was unaffected by the application of glycerol trinitrate. There were no significant differences between the 3 treatment groups with regard to median Apgar scores, median arterial umbilical pH values, blood pressure, and pulse rates (Table II). One minute after birth, 25th percentile Apgar scores and pulse rates in the 0.5-mg glycerol trinitrate group were lower than those found in the 0.25-mg glycerol trinitrate and placebo groups. Medians were similar, however, and differences were not significant. Pulse rates increased spontaneously in the neonates concerned, and
Volume 182, Number 4 Am J Obstet Gynecol
David et al 959
Fig 4. Ratio of nitroglycerin concentrations in umbilical vein and artery after intravenous bolus application to mother for 0.25- and 0.5-mg doses of glycerol trinitrate. y Axis is logarithmic.
5 minutes after birth, they were comparable in all 3 groups. Apart from routine tactile stimulation, no intervention was performed. We also inspected the data to establish whether abnormal findings in the neonates (Apgar score <7, arterial pH <7.20, and transfer to the neonatal department) correlated with higher levels of glycerol trinitrate in the umbilical cord. A subgroup consisting of those neonates with umbilical glycerol trinitrate levels in the upper quartile was compared with the remaining neonates. The Mann-Whitney U test was used for subgroup comparison. Significance was assumed at P < .05. Apgar scores and pH values in the upper-quartile subgroup did not differ significantly from those of the remaining neonates. Three of 62 infants delivered of mothers receiving glycerol trinitrate were transferred from the delivery room to the neonatal department compared with 1 of 31 in the placebo group. In the 0.5-mg glycerol trinitrate group a macrosomic term neonate born to a mother with insulin-dependent gestational diabetes had hypoglycemia and hypocalcemia. In the 0.25-mg glycerol trinitrate group a eutrophic term neonate with connate infection had an epileptic seizure and was mechanically ventilated for 4 days, and another eutrophic term neonate born after 36 weeks of gestation had slight hypoglycemia. None of these neonates was in the subgroup with higher levels of glycerol trinitrate nor were any of these events considered to be related to application of glycerol trinitrate. Comment In vitro investigations have demonstrated the dose-dependent inhibition of spontaneous and oxytocin-induced myometrial contractions by using glycerol trini-
trate,8 which suggests a potential clinical application in uterine relaxation or tocolysis. In recent years an increasing number of case reports and uncontrolled studies on the application of glycerol trinitrate in obstetrics have appeared in the literature.1, 2, 4, 9, 10 Doses between 0.05 mg and 1.85 mg glycerol trinitrate have been used successfully for various indications, both subpartum and postpartum, and doses have been applied intravenously, as patches, and as sublingual sprays.1, 2, 4 On intravenous application, drug action commences within 30 to 120 seconds,1 and the biologic half-life is only 2 to 2.5 minutes.11 Large intraindividual and interindividual variations of glycerol trinitrate plasma levels have been observed.11,12 In our study plasma concentrations of glycerol trinitrate were also scattered over a broad range. Glycerol trinitrate is metabolized to 1,2- or 1,3-glyceryl dinitrate by reductive hydrolysis and oxidation of glutathione with release of nitrogen dioxide.13 This process depends on the concentration of reduced glutathione as substrate and on the activity subtype of the catalyzing enzyme glutathione S-transferase. Data from animal experiments cannot be extrapolated to human subjects. Metabolization of glutathione is dependent not only on age but also on species. Compared with animals,14 human glutathione S-transferase activity is relatively high.15 In the fetus and the neonate the main isoform of glutathione S-transferase is placental glutathione S-transferase, which is not present in adults. The α and µ glutathione S-transferase subtypes are detectable in lower amounts than in adults, and the θ glutathione S-transferase subtype is not detectable perinatally. 7 Because of the postnatal increase in oxidized glu-
960 David et al
tathione caused by peroxidase activity, the question arises whether the glutathione capacity of the neonatal liver is sufficient for effective breakdown of a glycerol trinitrate bolus. In our study we were able to demonstrate that only a very small portion of the glycerol trinitrate administered to the mother crosses from the maternal to the fetal circulation: 1 minute after application of glycerol trinitrate, the median glycerol trinitrate concentration in the umbilical veins was two to three orders of magnitude lower than that found in maternal venous plasma. In addition, the marked difference in glycerol trinitrate concentrations between venous and arterial umbilical blood suggests that any glycerol trinitrate that does cross the placenta is rapidly metabolized. In animal experiments no significant changes have been reported in cardiovascular parameters or in the acid-base status in the fetus either during or after glycerol trinitrate infusion, and glycerol trinitrate has been found to have no teratogenic or embryotoxic effects.16 In human subjects glycerol trinitrate has been administered to pregnant women both intravenously to treat myocardial infarction and as patches for tocolysis in cases of cervical insufficiency and amniotic prolapse. Despite treatment lasting several days or weeks, no negative effects on the fetus were seen.17-19 de Rosayro et al5 suggested that the low umbilical level of glycerol trinitrate observed in sheep may be attributable to rapid metabolism in the maternal and fetal liver. It should also be mentioned, however, in addition to the above-described extensive hepatic first-pass effect, that spontaneous glycerol trinitrate hydrolysis has been found to occur in erythrocytes in healthy volunteers.12 After intravenous administration, considerable intraindividual and interindividual variation in glycerol trinitrate plasma levels has been observed. Adults receiving continuous infusion of 2.3-mg/h glycerol trinitrate (ie, 0.5 µg · kg–1 · min–1) had steady-state plasma concentrations of 1.6 to 5.2 ng/mL.12, 13 In the therapy for neonates after cardiac surgery, levels of 0.5 to 5 µg · kg–1 · min–1 (up to a maximum of 20 µg · kg–1 · min–1) glycerol trinitrate are administered by continuous infusion, and a steady-state glycerol trinitrate plasma concentration of 0.6 to 2.0 ng/mL would be expected. The median values observed in our study were appreciably lower: 0.09 ng/mL for 0.25 mg glycerol trinitrate and 0.49 ng/mL for 0.5 mg glycerol trinitrate (umbilical veins 1 minute after application of glycerol trinitrate). The general condition of the neonates was unaffected by glycerol trinitrate administration. Neonates born to mothers in the glycerol trinitrate groups did receive the drug through the umbilical vein. However, there were no significant differences from the placebo group in blood pressure or pulse in the first 10 minutes after birth. In the group receiving 0.5 mg glycerol trinitrate about 25% of
April 2000 Am J Obstet Gynecol
the newborns had somewhat lower pulse rates and correspondingly lower Apgar scores 1 minute after birth (Table II). This was neither clinically nor statistically significant. In none of the 3 newborns who were transferred to the neonatal department did a causal relationship with administration of glycerol trinitrate to the mother seem likely. The glycerol trinitrate and its metabolites that make their way into the fetal circulation do not seem to have any side effects on the cardiovascular system of the neonate in the period immediately after birth. The marked difference between venous and arterial levels of glycerol trinitrate and its metabolites in the umbilical cord indicates that the process of nitrate breakdown is already functioning well before birth. Because arteriovenous ratios in umbilical blood did not change with time, it appears that the neonate’s capacity for nitrate breakdown was not overloaded by the glycerol trinitrate bolus given to the mothers. We conclude that there is no evidence of major risk to the neonate from administration of an intravenous bolus of 0.25 or 0.5 mg glycerol trinitrate to the mother during a cesarean delivery. If an endangered or distressed fetus were to benefit from shortening the duration of a difficult cesarean delivery, from a pharmacokinetic point of view there would be no objection to application of glycerol trinitrate for uterus relaxation. REFERENCES
1. Riley ET, Flangan B, Cohen SE, Chitkara U. Intravenous nitroglycerin: a potent uterine relaxant for emergency obstetric procedures. Review of literature and report of three cases. Int J Obstet Anesth 1996;5:264-8. 2. Dufour P, Vinatier D, Puech F. The use of intravenous nitroglycerin for cervico-uterine relaxation: a review of the literature. Arch Gynecol Obstet 1997;261:1-7. 3. David M, Halle H, Lichtenegger W, Sinha P, Zimmermann T. Nitroglycerin to facilitate fetal extraction during cesarean delivery. Obstet Gynecol 1998;91:119-24. 4. Craig S, Dalton R, Tuck M, Brew F. Sublingual glyceryl trinitrate for uterine relaxation at caesarean section—a prospective trial. Aust N Z J Gynaecol 1998;38:34-9. 5. de Rosayro M, Nahrwold ML, Hill AB, Tait AR, Busch T, Kirsh MM. Plasma levels and cardiovascular effects of nitroglycerin in pregnant sheep. Can Anaesth Soc J 1980;27:560-4. 6. Hardman JG, Limbird LE. Goodmann & Gilman’s the pharmacological basis of therapeutics. 9th ed. New York: McGraw-Hill; 1995. p. 760-7. 7. Abramovitz M, Listowsky I. Developmental regulation of glutathione S-transferases. Xenobiotica 1988;18:1249-54. 8. Axemo P, Fu X, Lindberg B, Ulmsten U, Wessen A. Intravenous nitroglycerin for rapid uterine relaxation. Acta Obstet Gynecol Scand 1998;77:50-3. 9. Rowlands S, Trudinger B, Visva-Lingam S. Treatment of preterm cervical dilatation with glyceryl trinitrate, a nitric oxide donor. Aust N Z J Obstet Gynaecol 1996;36:377-81. 10. Mercier FJ, Dounas M, Bouaziz H, Lhuissier C, Benhamou D. Intravenous nitroglycerin to relieve intrapartum fetal distress related to uterine hyperactivity: a prospective observational study. Anesth Analg 1997;84:1117-20. 11. Glycerol trinitrate monograph. Bundesanzeiger (Federal Republic of Germany official monographs) 1990;43:1-4. 12. Noonan PK, Williams RL, Benet LZ. Dose dependent pharmacokinetics of nitroglycerin after multiple intravenous infusions
Volume 182, Number 4 Am J Obstet Gynecol
13.
14.
15.
16.
in healthy volunteers. J Pharmacokinet Biopharm 1985;13: 143-57. Fujii T, Adachi T, Usami Y, Tatematsu M, Hirano K. Variable glyceryl dinitrate formation as a function of glutathione S-transferase. Biol Pharm Bull 1996;19:1093-6. Gregus Z, Varga F, Schmelas A. Age-development and inducibility of hepatic glutathione S-transferase activities in mice, rats, rabbits and guinea pigs. Comp Biochem Physiol 1985;80:85-90. Pacifici GM, Franchi M, Colizzi C, Giuliani L, Rane A. Glutathione S-transferase in humans: development and tissue distribution. Arch Toxicol 1988;61:265-9. Oketani Y, Mitsuzono T, Ichikawa K, Itono Y, Gojo T, Gofuku M,
David et al 961
et al. Toxicological studies on nitroglycerin (nk-843)(6) teratological study in rabbits. Oyo Yakuri 1981;22:633-8. 17. Sheikh AU, Harper M. Myocardial infarction during pregnancy: management and outcome of two pregnancies. Am J Obstet Gynecol 1993;169:279-84. 18. Lees CC, Thompson C, Black RS, Campbell S, Danti L, Lojacono A, et al. Nitric oxide and preterm delivery: the GTN versus ritodrine trial. Prenat Neonat Med 1997;2(Suppl 1):10. 19. Versmold HT, Kitterman JA, Phibbs RH, Gregory GA, Tooley WH. Aortic blood pressure during the first 12 hours of life in infants with birth weight 610 and 4,220 grams. Pediatrics 1981;67:607-13.
Key words: Fetal-maternal plasma levels of nitroglycerin, glycerol trinitrate, cesarean delivery
Over the last decade, there have been several reports of successful obstetric use of nitroglycerin as a tocolytic. Nitroglycerin, which is also known as glycerol trinitrate, may be administered before, during, or after delivery and is well suited for use as a short-term tocolytic agent before external or internal change to extract a retained placenta or to correct a uterine inversion, as well as during cesarean delivery.1-4 In the clinical part of this study, two dosages of glycerol trinitrate (Nitro Pohl Infus; G. PohlBoskamp GmbH & Co, Hohenlockstedt, Germany) were From the Departments of Obstetrics and Gynecologya and Neonatologyb and the Institute for Clinical Chemistry and Biochemistry,c University Hospital Charité, Humboldt University, Berlin; A.A.I. Applied Analytical Industries Deutschland GmbH & Co, Neu-Ulmd; and the Department of Anesthesiology and Intensive Care, Friedrichshain Hospital, Berlin.e Received for publication March 15, 1999; revised September 14, 1999; accepted November 8, 1999. Reprint requests: Matthias David, MD, Charité, Campus VirchowKlinikum Klinik für Frauenheilkunde und Geburtshilfe, Augustenburger Platz 1 13353 Berlin, Germany. Copyright © 2000 by Mosby, Inc. 0002-9378/2000 $12.00 + 0 6/1/104365 doi:10.1067/mob.2000.104365
investigated for their clinical effect on uterine musculature relaxation during cesarean deliveries in comparison with a placebo. Glycerol trinitrate crosses the placenta in both animals and human subjects. In sheep a 1:25 ratio of arterial glycerol trinitrate concentrations between fetus and mother has been reported.5 Until now, however, there has been no information about the extent to which glycerol trinitrate passes from the maternal to the fetal circulation in human subjects. In the adult liver glycerol trinitrate is rapidly hydrolyzed by the glutathione system to 1,2- or 1,3-glyceryl dinitrate.6 The enzyme driving this process is glutathione S-transferase.6 However, the composition of isoenzymes in glutathione S-transferase is tissue dependent and changes with development. In adults α, θ, and µ subtypes predominate, but in the fetus and in neonates glutathione S-transferase is mainly placental. Because the kinetic properties of placental glutathione S-transferase differ from those of the adult isoforms,7 the question arises as to whether the fetus is able to metabolize significant amounts of glycerol trinitrate. Our study is the first 955
956 David et al
April 2000 Am J Obstet Gynecol
Fig 1. Nitroglycerin concentration in maternal plasma (0.25-mg group). y Axis is logarithmic.
Fig 2. Nitroglycerin concentration in maternal plasma (0.5-mg group). y Axis is logarithmic.
to determine the extent of fetomaternal glycerol trinitrate passage and fetal metabolization of glycerol trinitrate. The effect of glycerol trinitrate administration on the ease of fetal extraction in the study has already been reported.3
The pregnant women were assigned to treatment groups on the day surgery was scheduled according to a computer-generated list of random numbers that was kept in an area away from clinical care and was unavailable to surgeons and caregivers. The women were assigned to either the placebo group or one of the two nitroglycerin groups in the chronologic sequence of the cesarean deliveries. Both study drug and placebo were administered intravenously. The solutions were prepared by the responsible investigator 1 hour before cesarean delivery. All patients underwent standard induction of general anesthesia during the operation. Intravenous administration of 0.25 mg or 0.5 mg glycerol trinitrate (in 4.75 mL or 4.5 mL physiologic sodium chloride solution) or placebo (5 mL physiologic sodium chloride solution) was carried out at the time of the puncture incision in the lower uterine segment by the responsible investigator. The line was then flushed for 1 minute with 100 mL physiologic sodium chloride solution. Intravenous access was provided for application of anesthesia in all patients, either on the back of the right hand or on the right wrist. Venous blood was withdrawn from all patients in the glycerol trinitrate groups 1 and 5 minutes after administration of nitroglycerin through a second catheter fitted before the operation, usually in an antecubital vessel in the same arm. These samples were collected directly in precooled, specially laminated, glass vacuum tubes with no additional anticoagulant being used. Blood samples were centrifuged immediately in a refrigerated centrifuge, and the plasma was deep-frozen and stored at –80°C. In addition, after all deliveries in the glycerol trinitrate groups, samples of arterial and venous umbilical blood
Material and methods This double-blind, randomized, controlled study was carried out in the gynecology departments of Virchow Hospital and Charité Hospital (both part of Humboldt University, Berlin, Germany). Two doses of glycerol trinitrate, 0.25 mg and 0.5 mg, were investigated. Physiologic sodium chloride solution (0.9%) was used as a placebo control. A total of 97 patients scheduled for elective cesarean deliveries were included in the study during the period between June 1994 and July 1996. Inclusion criteria included maternal age of 18 to 45 years, gestational age of 34 to 42 completed weeks, singleton pregnancy, planned abdominal cesarean delivery irrespective of fetal position, and suitability for anesthesia as per American Society of Anesthesiologists status I and II. Exclusion criteria included pregnancy-induced hypertension, preeclampsia, heart disease or previous cardiac surgery, intolerance to nitrates, anemia, lung disease, hypotension, shock, alcohol abuse, serious systematic diseases like insulin-dependent diabetes mellitus, and intravenous tocolysis in the 48 hours before cesarean delivery. Before the start of the study, approval was obtained from the Ethics Committee of the Virchow Clinic, Berlin. Before the operation, all patients were fully informed of the nature and scope of the study, as well as the potential benefits and risks. All patients signed a declaration of consent.
Volume 182, Number 4 Am J Obstet Gynecol
were withdrawn immediately after the umbilical cord was cut. These blood samples were also handled as described above. Blood samples from the umbilical cord could not be taken at defined intervals after glycerol trinitrate application to the mother: Sampling times depended on the interval between application of glycerol trinitrate during uterine puncture incision and umbilical cord clamping after birth, which varied considerably. To allow a structured evaluation of the results, sampling times were grouped as follows: T1, interval between administration of glycerol trinitrate and cord clamping of 1 minute (defined as 30-90 seconds); T2, 2 minutes (91-150 seconds); T3, 3 minutes (151-210 seconds); and T4, 4 minutes (211-270 seconds). In both the 0.25-mg and 0.5-mg collectives, half the umbilical blood samples were in group T1. The concentrations of glycerol trinitrate and its metabolites, 1,2-glyceryl dinitrate and 1,3-glyceryl dinitrate, were measured in a reference laboratory (A.A.I. GmbH, Neu-Ulm, Germany). Suitable aliquots of the samples (standard sample volume, 1 mL) were prepared by fluid extraction and concentration. Analysis was done by means of a validated gas chromatography–mass spectroscopy method with a Trio 2000 GC/MC Quadrupol (Micromass, GB) system. After separation of the three components (glycerol trinitrate and 1,2-glyceryl dinitrate and 1,3-glyceryl dinitrate) by gas chromatography, the number of negative ions was determined after appropriate chemical ionization. The concentrations of all three analytes were measured simultaneously. The lower limit of quantitation of the method is 6 pg/mL for glycerol trinitrate and 25 pg/mL for 1,3-glyceryl dinitrate and 1,2glyceryl dinitrate. The upper limit is 2.4 ng/mL for glycerol trinitrate and between 3.0 and 9.0 ng/mL for the two metabolites. Samples with higher concentrations were diluted as necessary. The accuracy and precision of the method (both within and between series) was clearly better than ±10% over the entire range of concentrations. All infants delivered in the course of the study were examined by a neonatologist immediately after birth and were monitored closely during the postnatal adaptation phase. Arterial umbilical pH values were determined immediately after the umbilical cord was cut. Apgar scores, neonatal blood pressure, and pulse rates were documented 1, 5, and 10 minutes after birth. Blood pressure was measured by using oscillometry with a Dinamap 8100 (Critikon Inc, Norderstedt, Germany). Results Sixty-two patients were included in the pharmacokinetic evaluation: 30 of 32 patients received 0.25 mg glycerol trinitrate and 32 of 34 patients received 0.5 mg glycerol trinitrate. The remaining 4 patients receiving glycerol trinitrate were excluded from evaluation be-
David et al 957
Fig 3. Nitroglycerin concentrations 1 minute after intravenous bolus application to mother. y Axis is logarithmic.
cause of technical problems in blood sampling or preparation for analysis. Thirty-one patients received placebo. There were no significant differences between the 3 treatment groups with regard to age, maternal body habitus, height, weight, ethnicity, smoking status, duration of surgery, number of previous cesarean deliveries, fetal position, fetal size, gravidity, parity, gestational age, blood pressure, and pulse rate before induction of anesthesia and maternal hemoglobin, hematocrit, thrombocytes, prothrombin time, and partial thromboplastin time values on the day of delivery (Table I). The study population included only healthy pregnancies, as described by the inclusion and exclusion criteria. In the glycerol trinitrate groups, maternal blood samples were available for a total of 57 patients for both sampling times (1 and 5 minutes after application of glycerol trinitrate) and 5 patients for one of the sampling times. Plasma concentrations of glycerol trinitrate at each of the time points are shown in Figs 1 and 2. In some cases surgical or anesthetic requirements meant that sampling was delayed by approximately 1 minute. These cases have been included in Figs 1 and 2. Umbilical blood samples were analyzed from 59 of 62 births. In 55 cases samples were available from both the umbilical artery and the umbilical vein; in 4 cases samples were only available from the umbilical vein. Because plasma concentrations of glycerol trinitrate and its metabolites are dependent on time, meaningful comparison of maternal and umbilical blood samples is only possible where samples were collected at approximately the same time. In this study the ratio of the plasma concentrations between maternal and umbilical venous blood could only be determined for time T1 (n = 32). Box-and-whisker plots of the glycerol trinitrate concentrations at this time point are shown in Fig 3. One minute after application, the median glycerol trinitrate concentration in maternal venous plasma was approximately 500 times greater than that in umbilical blood. Although
958 David et al
April 2000 Am J Obstet Gynecol
Table I. Patient characteristics (mothers) 0.25-mg nitroglycerin group (n = 30)
0.5-mg nitroglycerin group (n = 32)
Placebo group (n = 31)
29 163 82.5
30 162 75.2
29 163 78.8
29 (96.7%) 0 1 0 39.3 40 12.1 37 115/70
31 (96.9%) 1 0 0 37.5 40 11.6 34.5 112.5/70
29 (93.5%) 1 0 1 38.4 39 12.1 36.5 135/79
Age (y) Height (cm) Weight (kg) Race White Asian Afro-Caribbean Oriental Primiparous (%) Gestational age (wk)40 Preoperative hemoglobin (g/dL) Preoperative hematocrit (mL/dL) Preoperative blood pressure, mm Hg (systemic/diastolic) Data are presented as median or number and percent.
Table II. Characteristics of neonates
Apgar score 1 min 5 min 10 min post partum Umbilical pH value (arterial) Systolic blood pressure (mm Hg)* 1 min 5 min 10 min post partum Diastolic blood pressure (mm Hg)* 1 min 5 min 10 min post partum Pulse (beats/min) 1 min 5 min 10 min post partum
0.25-mg nitroglycerin group (n = 30)
0.5-mg nitroglycerin group (n = 32)
Control group (placebo; n = 31)
9 (8-9) 10 (9-10) 10 (10-10) 7.36 (7.35-7.38)
9 (7-9) 10 (9-10) 10 (10-10) 7.32 (7.28-7.35)
9 (8-9) 10 (9-10) 10 (10-10) 7.32 (7.28-7.37)
63 (54-71) 68 (58-72.5) 62 (56-70)
67 (57-77) 67 (53.5-76.5) 60.5 (49-70.5)
62 (49.5-75) 64 (59-79) 62.5 (54-72) 38 (35-50) 43 (33-54) 36 (31-42) 163 (127-186) 162 (137-177) 153 (140-173)
43 (38-47) 37.5 (33-47) 37 (30-44)
41 (36.5-56) 40 (34-48.5) 36.5 (31.5-42.5)
158 (100-173) 161 (134-171) 153 (135-166)
140 (100-161) 153 (140-166) 151 (144-165.5)
Results shown are median and 25th to 75th percentile. There were no significant differences between treatment and placebo groups. *Mean blood pressures (95% range) for term neonates with birth weight corresponding to gestational age are as follows: systolic, 64 mm Hg (54-77 mm Hg); diastolic, 38 mm Hg (28-48 mm Hg).19
some of the maternal blood samples were taken 2 minutes after application of glycerol trinitrate, the number of cases in which this coincided with umbilical samples at time T2 and later was very small, and statistical analysis of the T2, T3, and T4 data was not supported by the software used. The median ratio of plasma concentrations between venous and arterial umbilical blood was 7.4:1 (95% confidence interval, 1.96-20.18) for 0.5 mg of glycerol trinitrate and 1:1 (95% confidence interval, 0.28-3.33) for 0.25 mg of glycerol trinitrate, irrespective of the dose applied and the interval between application of glycerol trinitrate and collection of umbilical blood (Fig 4). Similar ratios were obtained for the glycerol trinitrate
metabolites. The overall median ratio of plasma concentrations was 7.5:1 for 1,2-glyceryl dinitrate and 8.8:1 for 1,3-glyceryl dinitrate. The postnatal condition of the neonates was unaffected by the application of glycerol trinitrate. There were no significant differences between the 3 treatment groups with regard to median Apgar scores, median arterial umbilical pH values, blood pressure, and pulse rates (Table II). One minute after birth, 25th percentile Apgar scores and pulse rates in the 0.5-mg glycerol trinitrate group were lower than those found in the 0.25-mg glycerol trinitrate and placebo groups. Medians were similar, however, and differences were not significant. Pulse rates increased spontaneously in the neonates concerned, and
Volume 182, Number 4 Am J Obstet Gynecol
David et al 959
Fig 4. Ratio of nitroglycerin concentrations in umbilical vein and artery after intravenous bolus application to mother for 0.25- and 0.5-mg doses of glycerol trinitrate. y Axis is logarithmic.
5 minutes after birth, they were comparable in all 3 groups. Apart from routine tactile stimulation, no intervention was performed. We also inspected the data to establish whether abnormal findings in the neonates (Apgar score <7, arterial pH <7.20, and transfer to the neonatal department) correlated with higher levels of glycerol trinitrate in the umbilical cord. A subgroup consisting of those neonates with umbilical glycerol trinitrate levels in the upper quartile was compared with the remaining neonates. The Mann-Whitney U test was used for subgroup comparison. Significance was assumed at P < .05. Apgar scores and pH values in the upper-quartile subgroup did not differ significantly from those of the remaining neonates. Three of 62 infants delivered of mothers receiving glycerol trinitrate were transferred from the delivery room to the neonatal department compared with 1 of 31 in the placebo group. In the 0.5-mg glycerol trinitrate group a macrosomic term neonate born to a mother with insulin-dependent gestational diabetes had hypoglycemia and hypocalcemia. In the 0.25-mg glycerol trinitrate group a eutrophic term neonate with connate infection had an epileptic seizure and was mechanically ventilated for 4 days, and another eutrophic term neonate born after 36 weeks of gestation had slight hypoglycemia. None of these neonates was in the subgroup with higher levels of glycerol trinitrate nor were any of these events considered to be related to application of glycerol trinitrate. Comment In vitro investigations have demonstrated the dose-dependent inhibition of spontaneous and oxytocin-induced myometrial contractions by using glycerol trini-
trate,8 which suggests a potential clinical application in uterine relaxation or tocolysis. In recent years an increasing number of case reports and uncontrolled studies on the application of glycerol trinitrate in obstetrics have appeared in the literature.1, 2, 4, 9, 10 Doses between 0.05 mg and 1.85 mg glycerol trinitrate have been used successfully for various indications, both subpartum and postpartum, and doses have been applied intravenously, as patches, and as sublingual sprays.1, 2, 4 On intravenous application, drug action commences within 30 to 120 seconds,1 and the biologic half-life is only 2 to 2.5 minutes.11 Large intraindividual and interindividual variations of glycerol trinitrate plasma levels have been observed.11,12 In our study plasma concentrations of glycerol trinitrate were also scattered over a broad range. Glycerol trinitrate is metabolized to 1,2- or 1,3-glyceryl dinitrate by reductive hydrolysis and oxidation of glutathione with release of nitrogen dioxide.13 This process depends on the concentration of reduced glutathione as substrate and on the activity subtype of the catalyzing enzyme glutathione S-transferase. Data from animal experiments cannot be extrapolated to human subjects. Metabolization of glutathione is dependent not only on age but also on species. Compared with animals,14 human glutathione S-transferase activity is relatively high.15 In the fetus and the neonate the main isoform of glutathione S-transferase is placental glutathione S-transferase, which is not present in adults. The α and µ glutathione S-transferase subtypes are detectable in lower amounts than in adults, and the θ glutathione S-transferase subtype is not detectable perinatally. 7 Because of the postnatal increase in oxidized glu-
960 David et al
tathione caused by peroxidase activity, the question arises whether the glutathione capacity of the neonatal liver is sufficient for effective breakdown of a glycerol trinitrate bolus. In our study we were able to demonstrate that only a very small portion of the glycerol trinitrate administered to the mother crosses from the maternal to the fetal circulation: 1 minute after application of glycerol trinitrate, the median glycerol trinitrate concentration in the umbilical veins was two to three orders of magnitude lower than that found in maternal venous plasma. In addition, the marked difference in glycerol trinitrate concentrations between venous and arterial umbilical blood suggests that any glycerol trinitrate that does cross the placenta is rapidly metabolized. In animal experiments no significant changes have been reported in cardiovascular parameters or in the acid-base status in the fetus either during or after glycerol trinitrate infusion, and glycerol trinitrate has been found to have no teratogenic or embryotoxic effects.16 In human subjects glycerol trinitrate has been administered to pregnant women both intravenously to treat myocardial infarction and as patches for tocolysis in cases of cervical insufficiency and amniotic prolapse. Despite treatment lasting several days or weeks, no negative effects on the fetus were seen.17-19 de Rosayro et al5 suggested that the low umbilical level of glycerol trinitrate observed in sheep may be attributable to rapid metabolism in the maternal and fetal liver. It should also be mentioned, however, in addition to the above-described extensive hepatic first-pass effect, that spontaneous glycerol trinitrate hydrolysis has been found to occur in erythrocytes in healthy volunteers.12 After intravenous administration, considerable intraindividual and interindividual variation in glycerol trinitrate plasma levels has been observed. Adults receiving continuous infusion of 2.3-mg/h glycerol trinitrate (ie, 0.5 µg · kg–1 · min–1) had steady-state plasma concentrations of 1.6 to 5.2 ng/mL.12, 13 In the therapy for neonates after cardiac surgery, levels of 0.5 to 5 µg · kg–1 · min–1 (up to a maximum of 20 µg · kg–1 · min–1) glycerol trinitrate are administered by continuous infusion, and a steady-state glycerol trinitrate plasma concentration of 0.6 to 2.0 ng/mL would be expected. The median values observed in our study were appreciably lower: 0.09 ng/mL for 0.25 mg glycerol trinitrate and 0.49 ng/mL for 0.5 mg glycerol trinitrate (umbilical veins 1 minute after application of glycerol trinitrate). The general condition of the neonates was unaffected by glycerol trinitrate administration. Neonates born to mothers in the glycerol trinitrate groups did receive the drug through the umbilical vein. However, there were no significant differences from the placebo group in blood pressure or pulse in the first 10 minutes after birth. In the group receiving 0.5 mg glycerol trinitrate about 25% of
April 2000 Am J Obstet Gynecol
the newborns had somewhat lower pulse rates and correspondingly lower Apgar scores 1 minute after birth (Table II). This was neither clinically nor statistically significant. In none of the 3 newborns who were transferred to the neonatal department did a causal relationship with administration of glycerol trinitrate to the mother seem likely. The glycerol trinitrate and its metabolites that make their way into the fetal circulation do not seem to have any side effects on the cardiovascular system of the neonate in the period immediately after birth. The marked difference between venous and arterial levels of glycerol trinitrate and its metabolites in the umbilical cord indicates that the process of nitrate breakdown is already functioning well before birth. Because arteriovenous ratios in umbilical blood did not change with time, it appears that the neonate’s capacity for nitrate breakdown was not overloaded by the glycerol trinitrate bolus given to the mothers. We conclude that there is no evidence of major risk to the neonate from administration of an intravenous bolus of 0.25 or 0.5 mg glycerol trinitrate to the mother during a cesarean delivery. If an endangered or distressed fetus were to benefit from shortening the duration of a difficult cesarean delivery, from a pharmacokinetic point of view there would be no objection to application of glycerol trinitrate for uterus relaxation. REFERENCES
1. Riley ET, Flangan B, Cohen SE, Chitkara U. Intravenous nitroglycerin: a potent uterine relaxant for emergency obstetric procedures. Review of literature and report of three cases. Int J Obstet Anesth 1996;5:264-8. 2. Dufour P, Vinatier D, Puech F. The use of intravenous nitroglycerin for cervico-uterine relaxation: a review of the literature. Arch Gynecol Obstet 1997;261:1-7. 3. David M, Halle H, Lichtenegger W, Sinha P, Zimmermann T. Nitroglycerin to facilitate fetal extraction during cesarean delivery. Obstet Gynecol 1998;91:119-24. 4. Craig S, Dalton R, Tuck M, Brew F. Sublingual glyceryl trinitrate for uterine relaxation at caesarean section—a prospective trial. Aust N Z J Gynaecol 1998;38:34-9. 5. de Rosayro M, Nahrwold ML, Hill AB, Tait AR, Busch T, Kirsh MM. Plasma levels and cardiovascular effects of nitroglycerin in pregnant sheep. Can Anaesth Soc J 1980;27:560-4. 6. Hardman JG, Limbird LE. Goodmann & Gilman’s the pharmacological basis of therapeutics. 9th ed. New York: McGraw-Hill; 1995. p. 760-7. 7. Abramovitz M, Listowsky I. Developmental regulation of glutathione S-transferases. Xenobiotica 1988;18:1249-54. 8. Axemo P, Fu X, Lindberg B, Ulmsten U, Wessen A. Intravenous nitroglycerin for rapid uterine relaxation. Acta Obstet Gynecol Scand 1998;77:50-3. 9. Rowlands S, Trudinger B, Visva-Lingam S. Treatment of preterm cervical dilatation with glyceryl trinitrate, a nitric oxide donor. Aust N Z J Obstet Gynaecol 1996;36:377-81. 10. Mercier FJ, Dounas M, Bouaziz H, Lhuissier C, Benhamou D. Intravenous nitroglycerin to relieve intrapartum fetal distress related to uterine hyperactivity: a prospective observational study. Anesth Analg 1997;84:1117-20. 11. Glycerol trinitrate monograph. Bundesanzeiger (Federal Republic of Germany official monographs) 1990;43:1-4. 12. Noonan PK, Williams RL, Benet LZ. Dose dependent pharmacokinetics of nitroglycerin after multiple intravenous infusions
Volume 182, Number 4 Am J Obstet Gynecol
13.
14.
15.
16.
in healthy volunteers. J Pharmacokinet Biopharm 1985;13: 143-57. Fujii T, Adachi T, Usami Y, Tatematsu M, Hirano K. Variable glyceryl dinitrate formation as a function of glutathione S-transferase. Biol Pharm Bull 1996;19:1093-6. Gregus Z, Varga F, Schmelas A. Age-development and inducibility of hepatic glutathione S-transferase activities in mice, rats, rabbits and guinea pigs. Comp Biochem Physiol 1985;80:85-90. Pacifici GM, Franchi M, Colizzi C, Giuliani L, Rane A. Glutathione S-transferase in humans: development and tissue distribution. Arch Toxicol 1988;61:265-9. Oketani Y, Mitsuzono T, Ichikawa K, Itono Y, Gojo T, Gofuku M,
David et al 961
et al. Toxicological studies on nitroglycerin (nk-843)(6) teratological study in rabbits. Oyo Yakuri 1981;22:633-8. 17. Sheikh AU, Harper M. Myocardial infarction during pregnancy: management and outcome of two pregnancies. Am J Obstet Gynecol 1993;169:279-84. 18. Lees CC, Thompson C, Black RS, Campbell S, Danti L, Lojacono A, et al. Nitric oxide and preterm delivery: the GTN versus ritodrine trial. Prenat Neonat Med 1997;2(Suppl 1):10. 19. Versmold HT, Kitterman JA, Phibbs RH, Gregory GA, Tooley WH. Aortic blood pressure during the first 12 hours of life in infants with birth weight 610 and 4,220 grams. Pediatrics 1981;67:607-13.