Dr T. Matsuda, Center for Perinatal Medicine, Tohoku University Hospital, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-8574, Japan.
Objective To evaluate the safety of KUR-1246 as a tocolytic agent, we examined the effects of its long term infusion on respiratory and cardiovascular systems and general metabolism in pregnant sheep and their fetuses.
Design Animal experiment with chronically instrumented ewes and their fetuses.
Setting Center for animal experiments, Hokkaido University School of Medicine, Japan.
Sample Eight Suffolk ewes at 117 to 120 days of gestation.
Methods At 120–124 days of gestation, ewes (n= 4) were infused intravenously for 24 hours with KUR-1246 at 0.03 μg/kg/minute, a dose that completely inhibits oxytocin-induced uterine contractions in pregnant sheep. The controls received saline instead (n= 4). Statistical comparisons were carried out by repeated-measures ANOVA followed by Dunnett's test.
Main outcome measures Maternal and fetal values of heart rate, blood pressure, plasma electrolytes, glucose, insulin and non-esterified fatty acid levels, and blood gases and lactate level.
Results The maternal plasma levels of KUR-1246 increased and reached a plateau at 15 hours or later from the start of the infusion, whereas the fetal levels of it were below the lower limit of quantification (0.1 ng/mL) throughout the experiment. Significant differences over time between the ewes that had received with KUR-1246 and the controls were found for the following parameters: maternal heart rate, blood lactate, plasma glucose, and plasma insulin levels, and fetal plasma glucose and plasma insulin levels (P < 0.05). In the KUR-1246 treated ewes, significant changes from the pre-infusion value were detected in maternal blood lactate and fetal plasma glucose levels within 6 hours from the start of the infusion (P < 0.05). No significant differences were observed in other parameters in either ewes or fetuses.
Conclusion The physiologic changes induced by a 24-hour infusion of KUR-1246 were transient and considered to be within the compensatory capacity in both pregnant ewes and their fetuses, suggesting that KUR-1246 is a potentially safe tocolytic agent for use by long term infusion.
For the control of preterm labour, β2-adrenoceptor agonists such as ritodrine hydrochloride or terbutaline sulphate are currently used as representative tocolytic agents in clinical practice.1,2 Alongside their β2-stimulating effects, these drugs are commonly associated with cardiovascular complications, such as maternal tachycardia and pulmonary oedema, at clinical doses due mainly to the effect of concomitant β1-adrenoceptor stimulation.1–4 Therefore, their use in the management of persistent preterm labour at doses sufficient to arrest uterine contractions may be restricted in view of side effects. To try to resolve this problem, we developed a new β2-adrenoceptor agonist, named KUR-1246. Because this agent has a higher potency and selectivity for the β2-adrenoceptor than earlier compounds,5,6 it may be capable of arresting preterm labour in pregnant women with reduced cardiovascular side effects that occur via β1-adrenoceptor stimulation.
Prior to clinical trials to assess the tocolytic effect and safety of KUR-1246 in pregnant women, we performed animal experiments using pregnant sheep and their fetuses.7 First, we evaluated the inhibitory effects induced by cumulative intravenous administration of KUR-1246 on oxytocin-induced uterine contractions (over 100 Montevideo units) in pregnant sheep. Our results demonstrated the effectiveness of KUR-1246 as a tocolytic agent: (1) KUR-1246 suppressed oxytocin-induced uterine contractions by 60% at a dose of 0.003 μg/kg/minute, with complete inhibition being observed at doses of 0.03 μg/kg/minute or more; and (2) at 0.003 μg/kg/minute, there were no effects on the respiratory or cardiovascular system or on the general metabolic functions in either pregnant ewes or their fetuses.7 On this basis, we took the presumed clinical dosage of KUR-1246 to be within the range 0.003 to 0.03 μg/kg/minute. We postulated that the therapeutic protocol for KUR-1246 would be similar to that for ritodrine among current tocolytic medications8,9 (viz. initially, the dosage should be increased progressively until uterine quiescence is achieved, then a continuous infusion at the lowest effective concentration should be instituted to allow further fetal growth and maturation). Although we examined the effects of KUR-1246 during cumulative and short term infusion, its effectiveness and safety during long term continuous infusion have not been previously evaluated.
To assess the safety of KUR-1246 with long term administration, we infused the drug for 24 hours into pregnant ewes at a dose of 0.03 μg/kg/minute, which might be the maximal dosage in future clinical trials, while we analysed its effects on respiratory and cardiovascular systems and the general metabolism in pregnant sheep and their fetuses.
With the approval of the Animal Care and Use Committee of Hokkaido University School of Medicine, this study was carried out from November 2000 to April 2001. Both the preparation and the protocol were essentially the same as those described in our previous studies.7,10 In brief, at 117 to 120 days of gestation, a total of eight Suffolk ewes with timed pregnancies underwent surgery while under anaesthesia (intrathecal tetracaine hydrochloride and intravenous ketamine hydrochloride). Five electrodes were fixed to the maternal trunk wall, and polyvinyl catheters were inserted into the maternal right jugular vein and carotid artery. After laparotomy and hysterotomy, catheters were placed in the fetal jugular vein, carotid artery and amniotic cavity. All fetal catheters were exteriorised through a small incision in the flank of the ewe. Throughout the study period, the ewes were unrestrained and housed in individual cages, with free access to water and food, and appropriate antibiotics were administered to the mother, fetus and into the amniotic cavity.
At least three days after the above surgery, the eight pregnant sheep were randomly divided into two groups: a KUR-1246 group (n= 4) and a control group (n= 4), and the experiment was conducted as follows: The ewes in the KUR-1246 group received an infusion of KUR-1246 at a dose of 0.03 μg/kg/minute through the jugular vein, which infusion was continued for 24 hours. KUR-1246 was diluted in isotonic saline and infused at a rate of 6.0 mL/hour by an infusion pump. In the control group, saline was infused instead in the same manner.
Until completion of the study, arterial and amniotic pressures and heart rate were measured continuously by using a pressure transducer and a polygraph, and recorded on a digital audio tape recorder and a personal computer. For sampling, maternal and fetal arterial blood and amniotic fluid were collected just before and at 3-hour intervals from the beginning of the KUR-1246 or saline infusion until the end of the experiment. After each fetal blood sampling, an equivalent volume of stored maternal heparinised blood was injected into the fetus through the venous catheter.
Maternal and fetal arterial pressure and heart rate were calculated as the averages of values obtained every 5 minutes at time points just before and 1, 3, 6, 9, 12, 15, 18, 21 and 24 hours after the start of the KUR-1246 or saline infusion. All fetal arterial pressure values were corrected for amniotic fluid pressure, and the heart rate was calculated from the pulse wave. Maternal electrocardiogram signals (lead V) were analysed offline by means of commercial software (SAECG, chart extension of PowerLab system; ADInstruments Japan, Nagoya, Japan).
Blood gas data (pH, Po2, Pco2 and base excess) and the blood lactate level were measured with a blood gas analyser (Blood Gas System 860; Bayer Diagnostics, Sudbury, UK). Blood gas data were corrected for maternal rectal temperature. Both the heparinised blood samples (5.0 mL) taken from the maternal or fetal carotid artery and the amniotic fluid samples (3.0 mL) were centrifuged immediately after collection, and the separated plasma and supernatants were then stored at −20°C until assayed. KUR-1246 levels in both maternal or fetal plasma and amniotic fluid were determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS; API-3000; PE-SCIEX, Ontario, Canada). Mass spectral acquisition was achieved in turbo spray positive ion mode for both KUR-1246 and internal standard (its derivative) by applying multiple reaction monitoring. The steady state plasma level and elimination half-life of KUR-1246 were determined by using a non-linear curve-fitting program Graphpad PRISM (Graphpad Software, California, USA) on the basis of a one-compartment model for intravenous constant infusion. Plasma electrolytes were analysed by using an auto-analyser (SERA-720; Horiba, Kyoto, Japan). Plasma glucose, insulin and non-esterified fatty acid levels were determined, using commercially available kits, by the glucose oxidase method11 (Glucose B-Test Wako, Wako Pure Chemical Industries, Osaka, Japan), radio-immunoassay12 (Rat Insulin RIA kit, LINCO Research, Missouri, USA) and the acyl-CoA oxidase method13 (NEFA C-Test Wako, Wako Pure Chemical Industries), respectively.
All values were expressed as means (SEM). For statistical analysis, SAS system version 6.12 (SAS Institute, North Carolina, USA) was used. A repeated-measures ANOVA was performed for the statistical comparison of changes over time in physiological parameters between the two groups. In brief, the interactions of variables (time and group) were judged by the F test of variance ratio (unbiased variance of group and time divided by that of sheep and times within groups). If significant differences were indicated between group and time, Dunnett's test was performed to test for significant changes from the pre-infusion value in the KUR-1246 group. Differences were considered significant at P < 0.05.
On the basis of continuous monitoring of the measured parameters, all ewes and fetuses were in good condition and suitable for physiological assessment. There was no evidence of intrauterine infections, preterm labour or fetal distress.
Figure 1 shows the changes in the maternal plasma concentration of KUR-1246 during the experiment. The level increased rapidly over the first 3 hours, and reached 5.03 (0.53) ng/mL by the end of the infusion. Pharmacokinetic analysis revealed the steady state plasma level and elimination half-life to be about 4.4 ng/mL and 3.3 hours, respectively. In contrast, the amniotic fluid levels of KUR-1246 remained at less than 5% of the maternal plasma level, while the fetal plasma levels in all samples were below the lower limit of quantification (0.1 ng/mL) throughout the course of infusion.
Among the maternal and fetal physiological parameters (heart rate, mean arterial pressure, plasma electrolytes, glucose, insulin, non-esterified fatty acid levels and blood pH, Po2, Pco2, base excess and lactate levels), significant differences over time between the two groups (ewes treated with KUR-1246 or saline) were found for the following parameters: maternal heart rate (P < 0.05), blood lactate (P < 0.05), plasma glucose (P < 0.05) and plasma insulin levels (P < 0.05), and fetal plasma glucose (P < 0.05) and plasma insulin levels (P < 0.05). In the KUR-1246 group, moreover, significant differences from the pre-infusion value were detected in maternal blood lactate (P < 0.05) and fetal plasma glucose levels (P < 0.05) within 6 hours from the start of the infusion. There were no significant differences in other parameters in either ewe or fetus. It should be noted that in the control group, there were no significant changes in the parameters from the pre-infusion values throughout the experiments in either ewe or fetus.
Figure 2 shows changes with time in maternal heart rate, mean arterial pressure and plasma potassium levels in the two groups. Maternal heart rate showed a rapid increase within 1 hour after the start of the KUR-1246 infusion, peaked at 3 hours [156.2 (5.1) bpm], and then gradually declined, but remained elevated, until the end of the infusion. Although the changes in heart rate over time exhibited a significant difference between the two groups (P < 0.05), no significant changes from the pre-infusion value [114.6 (15.1) bpm] were found at any time point. No significant differences over time in maternal mean arterial pressure and plasma potassium levels were found between the two groups, although plasma potassium levels tended to be decreased [from 3.9 (0.1) to 3.4 (0.2) mEq/L] by initiation of KUR-1246 infusion. No physical findings relating to maternal adverse effects (such as arrhythmia, ST-segment depression, T-wave flattening, cough, wheezing or tachypnoea) were evident in either group.
Figure 3 shows changes with time in maternal plasma glucose and insulin and blood lactate levels in the two groups. Changes over time of these parameters were all significantly different between them. The maternal blood lactate became significantly elevated [to 46.2 (7.2) mg/dL] within 6 hours from the start of the KUR-1246 infusion [compared with the pre-infusion value of 12.6 (2.8) mg/dL]. Within 6 hours, plasma levels of both maternal glucose and insulin were also raised, the maximum values being 126.0 (31.3) mg/dL and 2.3 (0.8) ng/mL, respectively. However, the values were not statistically different from the pre-infusion values [glucose: 61.4 (2.8) mg/dL, insulin: 0.5 (0.2) ng/mL].
Changes with time in fetal plasma glucose and insulin and blood lactate levels are shown in Fig. 4. The changes in plasma glucose and insulin levels over time exhibited a significant difference between the two groups (P < 0.05). The fetal plasma glucose level increased significantly [to 37.1 (8.1) mg/dL] at 3 hours after the start of drug infusion when compared with the pre-infusion value of 18.1 (0.8) mg/dL. The fetal plasma insulin level was elevated apparently within 6 hours and reached its maximum at 1.1 (0.4) ng/mL, but this value was not statistically different from the pre-infusion value [0.4 (0.1) ng/mL]. No apparent changes over time were found in fetal blood lactate level between the two groups.
First, let us consider the influence of KUR-1246 infusion on the pregnant ewes. Intravenous infusion of KUR-1246 induced an elevation in heart rate, with a maximal increase of approximately 40 bpm (36.3%) at 3 hours after the start of the infusion. The level then gradually declined but remained elevated until the end of the experiment. Several studies have intimated that the tachycardia induced by KUR-1246 was smaller than that induced by ritodrine or other β-stimulants, even though an equivalent inhibition of uterine contraction was achieved.5,7,14 Although our KUR-1246 infusion seemed to induce a prolonged tachycardia, the level was at no time significantly different from the pre-infusion value. Hence, we consider that any acute increment in heart rate induced by KUR-1246 at a dose under 0.03 μg/kg/minute would be compensated physiologically, suggesting that its long term administration would be associated with a low incidence of tachycardia.
It has been reported that β-adrenoceptor stimulation causes hypotension, pulmonary oedema and hypokalaemia, as well as the concomitant tachycardia. In fact, diastolic hypotension, acute pulmonary oedema and flattening of the T-wave were reported to occur with both ritodrine and terbutaline.2–4 However, in the present study, no apparent hypotension, respiration failure, arrhythmia or depression of the ST–T segment was observed in the KUR-1246 group; although the maternal plasma potassium decreased transiently, but with no statistical significance. These results suggest that for long term infusion, KUR-1246 is safer than other β-mimetics.
On the other hand, elevation of maternal plasma glucose and insulin and blood lactate levels was evident at 3 to 6 hours from the beginning of the KUR-1246 infusion. Because such increments were also seen when ritodrine or terbutaline was continuously administered to pregnant women or sheep,15–17 we consider these metabolic changes to be common when β-stimulants are used as tocolytic agents. However, these changes induced by KUR-1246 infusion were transient, recovered within 12 hours and had no effects on blood gas parameters. Therefore, any elevations were not so severe as to cause acidaemia and subsequent respiratory compensation.
In the present study, infusion of KUR-1246 did not significantly affect the maternal plasma non-esterified fatty acid levels. However, in a previous study (on pregnant ewes), we observed a significant increase in plasma non-esterified fatty acid levels during KUR-1246 infusion at the same dosage as used in the present study, although this increment declined rapidly and disappeared at 2 hours after the end of infusion.7 Therefore, although we cannot rule out a possible increase in the maternal plasma non-esterified fatty acid levels within the first 3 hours of infusion, we consider that KUR-1246 is likely to have little influence on the plasma non-esterified fatty acid level, because no prolonged elevation of it was observed during long term infusion of the drug. However, further studies will be required to determine the detail of effects on plasma non-esterified fatty acid level.
Now, let us turn to the influence of KUR-1246 infusion on the fetus. The maternal infusion of KUR-1246 resulted in transient increases in fetal plasma glucose and insulin levels, as a consequence of transient maternal hyperglycaemia, with no influence on the other measured parameters. It has been reported that maternal hyperglycaemia can induce a state of accelerated oxygen consumption, leading to lactic acidosis, in fetuses.18,19 In the present study, indeed, KUR-1246 infusion resulted in transient elevation of the fetal plasma glucose level (<40 mg/dL); but the fetal arterial pH, Po2 and lactate levels remained unaltered throughout the infusion, whereas the maternal blood lactate level was transiently elevated. It was also reported that in fetal lambs, the degree of lactic acidaemia was related to the magnitude of hyperglycaemia (over 150 mg/dL) and/or hypoxemia20–22 and maternal levels of lactate, which is known to pass the placenta.23 Therefore, we consider that the extent of maternal hyperglycaemia and lactate elevation induced by prolonged KUR-1246 infusion was not so severe as to cause fetal lactic acidosis, thus suggesting that its long term application could be allowed in view of the reduced risk to fetal metabolic parameters.
In addition, even though maternal infusion was continued for 24 hours, KUR-1246 was unquantifiable in the fetal plasma throughout the experiments. This low permeability of KUR-1246 across the sheep placenta was probably due to the structure of the placenta,24 as we discussed in great detail in the previous report,7 which finding further supports its long term application with less adverse effects on the fetus. However, previous studies revealed that the mean feto-maternal ratio of plasma ritodrine levels in pregnant women25 was apparently higher than that in pregnant sheep26 (about 25%vs <10%). Moreover, direct infusion of ritodrine into fetal lambs led to the development of fetal hypoxaemia and lactic acidaemia in response to hyperglycaemia.27 Considering the anatomic differences in placental structure between sheep and humans,28 further studies will be required to determine the effects of KUR-1246 on fetuses.
To evaluate the safety of KUR-1246, we administered it intravenously for 24 hours at a constant dose of 0.03 μg/kg/minute, presumed to be the maximal dosage in future clinical trials, and examined its effects on pregnant sheep and their fetuses. Long term infusion of KUR-1246 caused some cardiovascular and metabolic perturbations in ewes and fetuses, but these were transient and apparently compensated physiologically. Judging from these results, KUR-1246 may be a potentially safe tocolytic agent when administered to humans by prolonged infusion.
The authors would like to thank Professor Kunihiko Kobayashi, MD, Department of Pediatrics, Hokkaido University School of Medicine, for critical review and suggestions regarding this manuscript; Masuo Akahane, PhD and Masami Kojima, PhD for critical suggestions about this experiment; and Satoru Okajima, MD, Soromon Kataoka, MD and Keiko Ueda, MD, for technical assistance.
Potential conflicts of interest
KUR-1246 was originally synthesised and developed by Kissei Pharmaceutical, who employed Dr Murata and covered the costs this study, which was performed at Hokkaido University.