The Effect of Propofol on Isolated Human Pregnant Uterine Muscle 

The study was approved by our institutional review board for research involving human subjects. Ten healthy patients scheduled for elective lower-segment cesarean section at full term who gave written consent were included in the study. None of the patients were in labor and all had epidural anesthesia. A standardized anesthetic technique was used with 20–25 ml 2% lidocaine containing 1:200,000 epinephrine and 100 [micro sign]g fentanyl.

Approximately 1 h after the induction of epidural anesthesia, a small segment of myometrium was excised from the upper incisional surface of the lower uterine segment after the neonate was delivered. Uterine muscle was easily recognized in the upper surface of the incision. A specimen was excised from the middle layer of the muscle. The myometrial segment was placed in Krebs solution within a cool tissue container and immediately sent to the laboratory.

Each specimen (n = 10) was dissected into three strips (measuring approximately 1 x 3 x 10 mm); the muscle fibers of these strips were oriented parallel to the longest dimension. These three strips were mounted in separate 15 ml-tissue baths containing Krebs solution. One end of the longest dimension of a muscle strip was connected to a hook that was fixed to the base of the bath. The other end of the strip was connected to another hook fixed to an extension of the lever arm of a force transducer. The initial length between the hooks was about 8 mm. The Krebs solution was composed of 118.2 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl², 1.2 mM MgSO⁴[middle dot] 7H²O, 25 mM NaHCO³, 1.2 mM NaH²PO⁴[middle dot] H²O, and 11 mM glucose. The bath solution was maintained at 37 [degree sign]C and continuously aerated with a gas mixture of 95% oxygen and 5% carbon dioxide. The pH ranged from 7.35 to 7.45.

The isometric tension of the myometrial strips was measured using a force displacement transducer (BGG-25 GMS; Kulite Semiconductor, Leonia, NJ), and the recording of traces were made on a computer (MacLab data recording system; Analog Digital Instruments, Castle Hill, Australia). An initial resting tension of 0.3 g was applied. The bath solution was flushed with fresh solution every 15 min. Approximately 90 min into the study (about 1.5 to 2 h from the time of muscle harvesting), muscle preparations developed rhythmic spontaneous contractions.

Muscle contraction activity was assessed by three parameters: resting tension, active tension, and contraction interval. The active tension was defined as an increase in tension during muscle contraction (the tension between peak tension and resting tension during contraction). The contraction interval was measured as the time between contractions. When the contractions became uniform in active tension and contraction interval for a period of 20 min, two of the three muscle preparations in which the contractions were similar in active tension and contraction interval were matched for experimentation.

One of the two muscle preparations was treated first with Diprivan (Zeneca Pharmaceuticals, Wilmington, DE), which was applied cumulatively to the bathing medium to achieve a final bath concentrations of 2.7 x 10^-6M (0.5 [micro sign]g/ml), 1.1 x 10^-5M (2 [micro sign]g/ml), and 5.5 x 10^-5M (10 [micro sign]g/ml) propofol. Approximately 20 min were allowed for observation at each application of the drug. The bath solution was drained thoroughly and replaced with fresh Krebs solution. Muscle contraction activity returned toward the control level. With new control recording of contractions in the buffer solution, Intralipid (Kabi Pharmacia, Clayton, NC), at equivalent concentrations to the doses of Diprivan, was applied to the bath to determine the potential effect of the fat emulsion on contractions (Figure 1).

In the other muscle preparation, the order of application of the experimental drugs to the bath was reversed as a Intralipid-washout-Diprivan sequence (Figure 2). At the end of the experiment, oxytocin (Fujisawa USA, Deerfield, IL) was applied to the bath in an incremental dose to determine whether oxytocin stimulates the muscle preparations (n = 10) in the presence of 10 [micro sign]g/ml Diprivan. A dose of 1.5 mU (100 [micro sign]U/ml in bath) was applied every 10 min so that the active tension returned to the control value measured in buffer solution.

One ml (10 mg propofol) of Diprivan 1% was added to 24 ml normal saline to make a solution of 400 [micro sign]g/ml propofol (solution 1). Solution 1 (7.5 ml, 3,000 [micro sign]g propofol) was further diluted by adding 32.5 ml normal saline, making a solution of 75 [micro sign]g/ml propofol (solution 2). Solution 2 (1 ml, 75 [micro sign]g propofol) was diluted with 2 ml normal saline, producing a solution of 25 [micro sign]g/ml propofol (solution 3). Solution 3 (0.3 ml, 7.5 [micro sign]g propofol) was added to the bath (15 ml) to achieve 0.5 [micro sign]g/ml propofol in the bath. Solution 2 (0.3 ml, 22.5 [micro sign]g propofol) was applied to the bath to achieve a cumulative concentration of 2 [micro sign]g/ml propofol. The addition of solution 1 (0.3 ml, 120 [micro sign]g propofol) to the bath resulted in a cumulative concentration of 10 [micro sign]g/ml propofol. This volume (0.3 ml of normal saline) did not alter the tension of muscle preparations. Intralipid 10%, with a pH of 8.0, was diluted using the same method.

The mean values of resting tension, active tension, and contraction interval obtained from the two muscle preparations were used to represent a single specimen. The means of three successive muscle contractions measured in the buffer solution immediately before drug treatments were used as controls. Data were collected in 2.5 to 3 h.

All data are presented as mean +/- SD. Differences in the means of the resting tension, active tension, and contraction interval obtained from muscle preparations were analyzed by analysis of variance for repeated measures. If the F test showed an overall difference, comparisons of two means were performed with the two-tailed Student's t test for unpaired data. The probability values were then adjusted by Bonferroni correction. Probability values <0.05 were considered significant.

(Table 1) shows demographic data of the patients and indications for cesarean section. All muscle preparations in the study developed spontaneous contractions in buffer solution.

(Table 2) summarizes raw data of the resting tension, active tension, and contraction interval of muscle contractions when muscle preparations were treated with Diprivan. Diprivan at propofol concentrations of 0.5 and 2 [micro sign]g/ml did not significantly decrease the active tension (3.41 +/- 1.00 g [P >0.7, 4%] and 2.76 +/- 1.25 g [P >0.3, 23%], respectively) from the control value of 3.57 +/- 0.97 g. Diprivan at a concentration of 10 [micro sign]g/ml reduced the active tension to 1.95 +/- 1.28 g (P < 0.02, 45%) from the control value. When the mean values of the active tension produced by Diprivan were compared with those produced by Intralipid, the difference in the active tension was also significant between 10 [micro sign]g/ml Diprivan and Intralipid at an equivalent concentration (P < 0.04). There were no significant differences in the resting tension and contraction interval.

When muscle preparations were treated with Intralipid, there were no significant changes in muscle contractions as compared with the control values (Table 3).

Oxytocin with a dose of 140 +/- 50 [micro sign]U/ml increased the active tension depressed by Diprivan to the control tension levels in all muscle preparations.

In vascular smooth muscle studies, vascular specimens (rings) must be contracted by adding a vasoconstrictor to study vascular relaxation. Unlike vascular smooth muscle, uterine muscle has spontaneous rhythmic contractions. The tension developed by spontaneously contracting uterine muscle has been used as controls in in vitro studies. [9,10]The present study shows that although the vehicle solution (Intralipid) had no effects on muscle contractions, 10 [micro sign]g/ml Diprivan significantly reduced the active tension developed by contracting muscle. The decline in active tension was most likely produced by propofol and not by the fat emulsion. The return of the reduced active tension to the control tension levels with washout of Diprivan also suggests that the reduction in the active tension was caused by the drug and was unlikely to be caused by other factors. After washout of the initial drug, new control contractions were established before the other drug was added. The control values of muscle preparations treated with Intralipid differed slightly from those of muscle preparations treated with Diprivan with the washout.

Propofol has been used extensively in patients because of its rapid plasma clearance, associated quick recovery time, and low incidence of postoperative nausea and vomiting. [11]Rapid awakening from anesthesia without emesis in the postoperative period would be advantageous in obstetric patients, who are at risk for acid aspiration in the perioperative period. Although propofol has not been approved for obstetric anesthesia, several clinical studies concerning the safety of propofol for induction of cesarean sections have been published. [1,12,13]As a result of blood loss with delivery of the fetus and placenta, propofol clearance was more rapid in patients undergoing cesarean section than in nonobstetric patients. [14]Maternal venous plasma concentrations of propofol are as high as 6 [micro sign]g/ml after an intravenous dose of 2.5 mg/kg propofol as an induction agent for cesarean section. [12]Plasma propofol concentrations at delivery, after a bolus induction dose of 2 mg/kg, ranged from 0.53 to 1.48 [micro sign]g/ml. [15]Distribution of propofol is rapid across the placenta, with an umbilical venous concentration/maternal venous concentration ratio of 0.65. [12] 

Propofol (2,6-diisopropylphenol) is a phenolic agent. Phenolic compounds have an inhibitory action on spontaneous phasic contractions of rat ileum in a dose-dependent manner [16]and also suppress cardiac muscle contractility. [17]Our previous study showed in in vitro tension recordings that phenol could suppress the active tension developed by spontaneously contracting myometrium. [18]Therefore, our results correlate with the findings of previous studies that phenolic agents may exert an inhibitory effect on the active tension developed by spontaneously contracting smooth muscle, including uterine smooth muscle.

The outer layer of myometrium obtained from nonpregnant women is more sensitive than the inner layer to the relaxant effect of vasoactive intestinal peptide. [9]Because muscle specimens in the current study were excised from the middle layer of the lower segment, further studies using the outer layer of the fundus may be needed to determine whether the fundus muscle response to propofol differs from that of the middle layer of the lower segment of the uterus.

Although the mechanism by which propofol relaxes smooth muscle tonus remains unclear, it has been suggested that the vasodilation may be caused by the Ca²⁺channel blocking effect of propofol on vascular smooth muscle [4,19]or by nitric oxide production and release from endothelial cells by propofol. [20]Increased sequestration of calcium in intracellular organelles by propofol is thought to cause airway smooth muscle relaxation. [8] 

Whether the decrease in muscle tension produced by propofol would affect uterine bleeding during cesarean section may warrant further study. However, given that >95% of propofol in blood is bound to plasma protein, [21]the free propofol concentration of 0.5 [micro sign]g/ml in vitro, which had no effect on the tension, would be comparable to 9–10 [micro sign]g/ml propofol in vivo. [4]Therefore propofol in clinically relevant concentrations would not decrease uterine tone and is unlikely to cause increased blood loss in patients who are undergoing cesarean section delivery. In these patients, use of propofol to induce and maintain anesthesia did not increase blood loss compared with patients who received thiamylal-isoflurane anesthesia. [22] 

The current study also suggests that oxytocin stimulates uterine muscle in the presence of propofol. Thus it is conceivable that the routine intravenous infusion of oxytocin during cesarean section delivery to stimulate uterine muscle contraction may negate an inhibitory effect of propofol on uterine muscle tone. However, further studies are required to determine whether the effect of propofol on uterine muscle contractions induced by oxytocin would be similar to its effect on spontaneous contractions.

In summary, 10 [micro sign]g/ml Diprivan significantly reduced the active tension developed by spontaneously contracting uterine muscle obtained from pregnant patients in isolated preparations. In the absence of significant effects of the fat emulsion (Intralipid) on active tension, the decline in the active tension was most likely caused by propofol. However, the propofol concentrations needed to significantly reduce the uterine muscle tone appear to be much greater than plasma free propofol concentrations reported by others during cesarean section delivery.