Arterial to End-tidal Carbon Dioxide Pressure Difference during Laparoscopic Surgery in Pregnancy

Eight healthy pregnant woman at 17, 20, 23, 23, 23, 24, 26, and 30 weeks of gestation scheduled for laparoscopic cholecystectomy for recurrent gallstone cholecystitis in pregnancy were studied. The study protocol was approved by the hospital ethical committee, and all patients provided written informed consent. Fetal heart rate was evaluated by ultrasound before induction of general anesthesia. After treatment with sodium citrate (30 ml administered orally) and denitrogenation of the lungs with 100% oxygen, a rapid sequence induction was performed with sodium pentathol and succinylcholine, and the trachea was intubated with a 7.0-mm endotracheal tube. General anesthesia was maintained with desflurane in an air–oxygen mixture, fentanyl and cisatracurium. Left uterine displacement was assured during anesthesia by placing a rolled bed sheet under the right lumbar region. Intraoperative maternal monitoring included continuous electrocardiogram, pulse oximetry, capnography, oropharyngeal temperature, and noninvasive blood pressure measurements. The Ohmeda capnograph (5250 RGM; Ohmeda, Englewood, CO) was calibrated before the commencement of each anesthesia. After induction of general anesthesia, a 20-gauge arterial cannula was introduced into the radial artery, observing sterile precautions. Initial pulmonary ventilation was adjusted to maintain Pet^CO2around 32 mmHg. After obtaining arterial blood for baseline blood gas measurements, peritoneal insufflation of carbon dioxide limited to peak inflation pressure of 15 mmHg was used. Pulmonary ventilation was adjusted to maintain Pet^CO2around 32 mmHg by increasing both the respiratory rate (from 10 to 12 breaths/min) and tidal volume as is currently the practice at our institution (inspiratory time:expiratory time ratio 1:2). Additional samples for measurements of arterial blood gas tensions were obtained during and after the end of carbon dioxide insufflation (approximately every 10 min) and in the postanesthesia care units. Arterial blood gas analysis was performed at 37°C using a Nova blood gas analyzer (Stat Profile Ultra; Nova Biomedical, Waltham, MA) after two-point calibration, and the results were corrected to the patient’s temperature. 15Intravenous ephedrine (10-mg bolus doses) was used if the systolic blood pressure decreased below 20% of the baseline measurements. Fetal heart rate was reassessed 10 min after the patient emerged from general anesthesia and continued into the recovery room until discharge. One-way analysis of variance was used to determine if the difference between the variables in each phase of laparoscopic surgery was statistically significant. A P  value < 0.05 was considered significant. The differences in variables that were significant according to analysis of variance were examined using the Student t  test for paired observations.

There was no significant differences in mean Pet^CO2, Pa^CO2, Pa^CO2–Pet^CO2, p  H, bicarbonate (HCO³), and base excess–deficit during various phases of laparoscopy (table 1) except in minute volume and peak inspiratory pressures (P < 0.01, analysis of variance). The peak inspiratory pressures and minute volume were higher (P < 0.05, paired t  test) during insufflation compared with before or after carbon dioxide pneumoperitoneum. The mean Pa^O2value observed during insufflation (fraction of inspired oxygen = 40%) was 245 mmHg (SD, 55), and in the postoperative period was 150 mmHg (SD, 36). The duration of anesthesia was 83.7 min (SD, 13.5; range, 70–110 min), and insufflation period was 36.3 min (SD, 11.6; range, 25–60 min). The mean oropharyngeal temperature was 35.5°C (SD, 0.3; range, 35.1–36°C), and the mean carbon dioxide abdominal insufflation pressure was 11.7 mmHg (SD, 1.3). All parturients had uneventful progress of pregnancy except one patient (30 weeks’ gestation) who had mild uterine contractions without cervical effacement in the immediate postoperative period that required magnesium therapy for 12 h.

Our results in eight pregnant patients show that there were no significant differences in Pa^CO2–Pet^CO2during carbon dioxide pneumoperitoneum as compared with preinsufflation values during laparoscopic surgery. This is in contrast to results obtained by Cruz et al.  13(nine pregnant ewes) and Hunter et al.  14(four pregnant ewes), in which the difference increased by a mean of 10 mmHg during insufflation. Among the 28 observations obtained in eight patients during carbon dioxide insufflation in our study, the highest Pa^CO2–Pet^CO2observed was 5.1 mmHg as compared with 16 and 25 mmHg observed by Cruz et al.  13and Hunter et al.  14in ewes. The results observed in our study are consistent with our earlier observations of derived Pa^CO2–Pet^CO2of 6–7 mmHg from transcutaneous P^CO2monitoring in laparoscopic surgery in a parturient. 4Therefore, the physiologic consequences of pneumoperitoneum are different in humans from those in pregnant ewes, as seen by a lowered Pa^CO2–Pet^CO2during insufflation in pregnant patients.

Although the sheep model is used for obstetric research, there could be physiologic differences between the two species. For example, the preinsufflation Pa^CO2–Pet^CO2in pregnant ewes ranges from 6 to 15 mmHg, 13,14whereas it is lower in pregnant humans (0.6 mmHg; range, −2.5 to 5.1 mmHg; occasionally Pet^CO2exceeds Pa^CO2). 16,17In our study, the values varied from 0 to 5 mmHg, similar to those reported in pregnant subjects undergoing cesarean section and in women during postpartum sterilization. 16,17The smaller preinsufflation Pa^CO2–Pet^CO2in humans translates into a lower preinflation alveolar dead space in pregnant patients than in pregnant ewes. 18A lower preinsufflation alveolar dead space in pregnant patients probably results in a smaller subsequent changes in alveolar dead space during carbon dioxide insufflation, and thus to a smaller Pa^CO2–Pet^CO2in pregnant patients than in pregnant ewes.

Amos et al.  12suggested that fetal loss in their series (four of seven fetal deaths, three during first postoperative week and another 4 weeks postoperatively) could have been a result of prolonged respiratory acidosis despite maintaining Pet^CO2in the low to mid 30s. 12However, they did not have direct information on hypercarbia or acidosis. It is important to note that parturients in their series had other conditions, including such risk factors as pancreatitis and perforated appendix, which are generally believed to increase the risk of fetal loss and could have contributed to increased fetal mortality. 12 

Because of the relatively infrequent nature of laparoscopic surgery in pregnancy, it is difficult to accumulate a large number of patients, as surgery is preferably delayed until the conclusion of pregnancy. However, our observations in this prospective study suggest that capnography is adequate to guide ventilation during carbon dioxide insufflation in pregnant patients. Assuming that the maximum Pa^CO2–Pet^CO2of approximately 7 mmHg (as observed in our study) could occur during anesthesia, a Pet^CO2of 32 mmHg should allow Pa^CO2in the range not greater than Pa^CO2that is usually encountered in the postoperative period after laparoscopic surgery in pregnant patients. At our institution, the outcome has been successful in 23 patients (retrospective study, 10 subjects, 1991–1995 5; case report, 1 subject, 1997 4; present prospective study, 8 subjects, 1998–1999; and 4 subjects outside of these studies) who underwent laparoscopic surgery during pregnancy.

The authors thank Leslie Gilbertson, M.D., Mercedes Concepcion, M.D., Vladmir Formanek, M.D., Stan Lee-son, M.D., and William Camann, M.D., for facilitating this study.