This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Diemunsch, P. A. Articles by Geny, B. Search for Related Content PubMed PubMed Citation Articles by Diemunsch, P. A. Articles by Geny, B. Related Collections Anesthetic Techniques

---

GENERAL ARTICLES

Calibrated Pneumoperitoneal Venting to Prevent N₂O Accumulation in the CO₂ Pneumoperitoneum During Laparoscopy with Inhaled Anesthesia: An Experimental Study in Pigs

Pierre A. Diemunsch, MD^*, Thomas Van Dorsselaer^*, Klaus D. Torp, MD^*, Roland Schaeffer, MD^*, and Bernard Geny, MD, PhD

*Département d’Anesthésiologie, I.R.C.A.D., Hôpitaux Universitaires, Strasbourg, France; and Institut de Physiologie, Faculté de Médecine, Strasbourg, France

Address correspondence and reprint requests to Pierre A. Diemunsch, MD, I.R.C.A.D., Hôpitaux Universitaires, 1, Place de l’Hôpital, 67000 Strasbourg, France.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Nitrous oxide (N₂O) accumulates in the CO₂ pneumoperitoneum during laparoscopy when N₂O is used as an adjuvant for inhaled anesthesia. This may worsen the consequences of gas embolism and introduce a fire risk. In this study, we quantified the pneumoperitoneal gas venting necessary to prevent significant contamination by inhaled N₂O. Four domestic pigs (26–30 kg) were anesthetized and ventilated with 66% N₂O in oxygen. A CO₂ pneumoperitoneum was insufflated and maintained at a pressure of 12 mm Hg. Each animal underwent three experimental conditions, in random sequence, for 70 min each: 1) no pneumoperitoneal leak, 2) leak of 2 L every 10 min (12 L/h), and 3) leak of 4 L every 10 min (24 L/h). Every 10 min, pneumoperitoneal gas samples were analyzed for fractions (FPn) of N₂O and CO₂. Without leaks, FPnN₂O increased continually and reached 29.58% ± 3.15% at 70 min. With leaks of 2 and 4 L every 10 min (12 and 24 L/h), FPnN₂O reached a plateau of <10% after 30 min. We conclude that calibrated pneumoperitoneal venting of 12 or 24 L/h is enough to prevent the constitution of potentially dangerous pneumoperitoneal gas mixtures if venting is constant.

IMPLICATIONS: External venting calibrated at four or eight initial pneumoperitoneal volumes per hour with compensation by fresh CO₂ is sufficient to prevent nitrous oxide buildup of more than 10% in the pneumoperitoneum during laparoscopy with inhaled general anesthesia if venting is constant.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
To provide the necessary surgical exposure during laparoscopic surgery and to allow for surgical manipulations, a pneumoperitoneum is created by insufflating gas into the peritoneal cavity. Several gases have been used to create the pneumoperitoneum, including nitrous oxide (N₂O). Because of concerns regarding intraabdominal fire and explosion hazards (1,2), CO₂ became the choice as the insufflating gas. The advantages of CO₂ as the gaseous medium result from its physical and physiologic properties; it is nontoxic, has excellent plasma solubility, and does not support combustion.

However, concerns have been raised that the advantages of the conventional CO₂ pneumoperitoneum could be compromised during general inhalation anesthesia with N₂O as an adjuvant inhaled anesthetic (3). We have previously studied the course of N₂O diffusion into a CO₂ pneumoperitoneum in a pig model (4) and demonstrated that levels of N₂O in excess of 29 vol% may indeed occur in the CO₂ pneumoperitoneum. Such levels of N₂O in a CO₂ atmosphere may be dangerous because they can support combustion of bowel gas (3). The risk of formation of gas embolism consisting of CO₂ and N₂O may also be of concern because the consequences of such embolization may differ from those with the same volume of CO₂ alone.

To avoid potentially fatal complications, N₂O could be eliminated from the inhaled mixture. An alternate solution is to partially vent the mixed-gas peritoneal atmosphere and to compensate for this loss by fresh plain CO₂ (without N₂O) from the insufflator to maintain the preset peritoneal pressure (5). The aim of this experimental study was to quantify, in the pig model we used previously, the amount of pneumoperitoneal gas that needs to be constantly renewed to prevent significant accumulation of N₂O in a CO₂ pneumoperitoneum.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
After institutional approval was obtained, four domestic pigs (26–30 kg) were subjected to the following three experimental conditions: 1) no pneumoperitoneal leak, 2) an external leak of 2 L every 10 min, and 3) an external leak of 4 L every 10 min. Each animal acted as its own control, undergoing the three conditions successively, in a random sequence determined by the Latin-square method. This design resulted in 12 experiments.

The animals were premedicated with IM administration of ketamine (20 mg/kg) and azaperone (2 mg/kg). An IV catheter (22 gauge) was inserted into an auricular vein. IV induction was obtained with thiopental (10 mg/kg). Endotracheal intubation with a Portex 6-mm tube was facilitated by pancuronium (0.1 mg/kg). Anesthesia and neuromuscular blockade were maintained with isoflurane up to a maximum of 1.5 vol% end-tidal fraction (FET) and pancuronium (0.1 mg · kg^-1 · h^-1) IV, respectively. Ventilation was controlled with a Dräger Cicero ventilator (Dräger Médical, Antony, France), with an inspiratory fraction (FI) of 0.66 N₂O in oxygen at a fresh gas flow of 2.0 L/min. Minute ventilation was adjusted to keep the PETCO₂ between 35 and 45 mm Hg.

A Veress needle was inserted on the midline at the lower part of the umbilicus. Thirty minutes after the induction of anesthesia, after an equilibrium of N₂O was reached (as verified by FIN₂O = FETN₂O), a pneumoperitoneum was created by insufflation of CO₂ with a Striker insufflator (Striker, San Jose, CA). The time t₀ was defined as the time when an intraperitoneal pressure of 12 mm Hg was achieved. Intraperitoneal pressure was maintained at this level throughout the procedure by reinsufflation of CO₂. A second needle was inserted into the right flank to be used in Experimental Conditions 2 and 3 to create a measured pneumoperitoneal leak. Leakage was automatically compensated for by the insufflator with plain CO₂ to maintain the preset intraperitoneal pressure.

A pneumoperitoneal gas-sampling catheter was inserted percutaneously into the left flank, under endoscopic video guidance to keep the catheter tip at a distance from the insufflation port. Ten-milliliter samples of gas were taken at t₀ and every 10 min thereafter in gas-tight syringes and analyzed for pneumoperitoneal fractions (FPn) of N₂O and CO₂. In Experimental Conditions 2 and 3, the measured leak was created during the first 2 min of each 10-min period by aspirating the desired amount of gas with graduated syringes. The samples for gas analysis were taken during the last 2 min of each 10-min period. This procedure was repeated successively seven times under each experimental condition. At the end of each 70-min experiment (after the last gas sample was taken), the pneumoperitoneum was completely emptied. The following reinsufflation with plain CO₂ to the pressure of 12 mm Hg marked t₀ for the next experiment.

Continuous monitoring included electrocardiography and measurements of peritoneal pressure, FIO₂, oxygen saturation by pulse oximetry, PETCO₂, FIN₂O, FETN₂O, FIISOFLURANE, FETISOFLURANE, minute ventilation, and peak inspiratory pressure. The FPnN₂O and the FPnCO₂ were measured on site with a portable micro gas chromatograph (MGC) (P200; MTI Analytical Instruments, Fremont, CA). The results were verified with a capillary gas chromatograph coupled with a mass spectrometer (GC-MS) (MD 800 GC-MS; Fisons, Manchester, UK), according to the method we have previously described (4). The experimental data were shown in a two-dimensional scatter diagram.

In the experiments without external leaks, the equation for the development of FPnN₂O over time was determined by using the same growth model that we used previously for similar experimental conditions:

equation

where represents the value of FPnN₂O when t tends to infinity, i.e., FETN₂O = 66% in our setting.

The rate of uptake constant k is obtained after linearizing Equation 1:

equation

Linear regression analysis is then possible and estimates k, the 95% confidence interval of k, the P value of the F statistics of regression, and the Pearson coefficient of correlation.

However, in the experiments in which an external leak was created, the model used for experiments without leaks is inadequate because the boundary or maximum value of the FPnN₂O when t tends to infinity is unknown. We therefore determined the equations of the FPnN₂O over time by using nonlinear regression analysis. Results were compared by using mixed-design analysis of variance (ANOVA) with two fixed factors (leak type and time) and a random factor (the animal). Post hoc comparisons concerning leak type, time, and interaction between leak type and time were performed with the Newman-Keuls test and the method of contrasts.

Data obtained from the MGC were compared with data from the GC-MS by using simple linear regression analysis, giving the coefficient of determination r². A P value of <0.05 was considered significant. The statistics were performed with StatView software (SAS Institute, Cary, NC) and Mathematica software (Wolfram Research, Champaign, IL).

	Results

Top Abstract Introduction Methods Results Discussion References

 
The study conditions during the 12 experiments are summarized in Table 1. The mean volume of CO₂ (±SD) necessary to create a 12 mm Hg pressure pneumoperitoneum was 3.27 ± 0.33 L. Figure 1 shows the evolution of the FPnN₂O over time. No N₂O was detectable at t₀ in any of the 12 experiments, including those performed as the second or third experiment in the same animal. Thus, complete emptying and reinsufflation of the pneumoperitoneum with plain CO₂ were achieved.

View larger version (25K): [in this window] [in a new window]   Figure 1. Time course of the nitrous oxide fraction in a CO2 pneumoperitoneum (FPnN2O) in four pigs under inhaled anesthesia, studied in a random cross-over design with no pneumoperitoneal leak and with pneumoperitoneal leaks of 2 L every 10 min (12 L/h) and 4 L every 10 min (24 L/h).

With a leak of 2 or 4 L at the beginning of each 10-min interval, the FPnN₂O increased only slowly and reached a plateau after 30 min, remaining <10% throughout the 70-min study period (Fig. 1). The curves were found to have the following equations: for the 2-L leak, FPnN₂O_(t) = 10.02(1 - 1.059exp^-0.0503t), sum of squares = 68.88; and for the 4-L leak, FPnN₂O_(t) = 7.04(1 - 1.058exp^-0.0530t), sum of squares = 23.65. No further increase in FPnN₂O occurred after min 30 for either leak size, and FPnN₂O values at Minute 30 and at Minute 70 did not differ from each other (P = 0.43 with the 2-L leak every 10 min and P = 0.25 with the 4-L leak every 10 min).

ANOVA confirmed a difference among experimental conditions in the evolution of FPnN₂O over time (P = 0.0001), and the difference in FPnN₂O values between the no-leak condition and the conditions with leaks increased throughout the experiment. In the analysis of FPnN₂O with the contrast method, the no-leak condition differed from the 4-L leak condition at 10 min (P = 0.01) and at 20 min (P = 0.0001) and differed from both leak conditions after 20 min (P = 0.0001 for the no-leak condition versus the 2-L leak at 30 min). At 70 min, FPnN₂O was 29.58% (±3.15%) with no leak, 9.02% (±2.20%) with a 2-L leak, and 6.57% (±0.33%) with a 4-L leak.

The two leak sizes did not differ from each other in FPnN₂O at 10 min, but a difference was observed at 20 min (P = 0.02). Thereafter, Fpnn₂o remained lower with the 4-L leak than with the 2-L leak.

The results obtained from the MGC correlated well with those from the GC-MS analysis, as expressed by the equation FPnN₂O (MGC) = 0.975 x FPnN₂O (GC-MS); r² = 0.990. Figure 2 represents the corresponding regression plot.

View larger version (23K): [in this window] [in a new window]   Figure 2. Correlation of direct measurements of the nitrous oxide fraction in a CO2 pneumoperitoneum (FPnN2O) performed in the operating room with a micro gas chromatograph (MGC) with the results obtained from the laboratory with a gas chromatograph coupled with a mass spectrometer reference device (GC-MS).

	Discussion

Top Abstract Introduction Methods Results Discussion References

The study period of 70 minutes was chosen after the results of a pilot study indicated that this duration was required to observe FPnN₂O values of 29% or more in our experimental model without leaks. The 70-minute period represents the initial, steep part of the growth curve, where the FPnN₂O value with time may also be represented by a simple linear regression model as FPnN₂O_(t) = 0.433t (r² = 0.984).

In all pigs, the leaks were fixed at 12 or 24 L/h. The body weights of the animals were 26.0, 29.7, 29.4, and 30.0 kg. For pigs in this range of weight (28.8 ± 1.9 kg), the usual volume needed to create a 12 mm Hg pressure pneumoperitoneum is approximately 3 L (3.27 ± 0.33 L in our 12 experiments in 4 pigs). The leak sizes were chosen to represent a fourfold and an eightfold turnover of the volume initially insufflated into the peritoneum. This corresponded to a rate of 0.4 L · h^-1 · kg^-1 for the 12 L/h leak and 0.8 L · h^-1 · kg^-1 for the 24 L/h leak.

Creating a 4-L leak every 10 minutes, i.e., leakage at an overall rate of 24 L/h, prevents accumulation of N₂O in the pneumoperitoneum. A 2-L leak every 10 minutes achieves a similar end result. However, a difference in FPnN₂O from the no-leak condition is achieved only after 20 minutes with a 2-L leak, whereas a difference is already achieved after 10 minutes with a 4-L leak.

In this study, the leaks were repetitively created for 2-minute periods at 10-minute intervals. The aim of this procedure was to approach the conditions of a constant leak. In contrast, long periods of tight pneumoperitoneum with intermittent brief high-flow leaks occur typically when instruments are removed after a long period of precise work without external gas leak. The difference between a constant low-flow leak and a sporadic high-flow leak seems important. A 24 L/h leak (with peritoneal flushing with 24 L of plain CO₂) during a few minutes at the end of each hour would not prevent a potentially dangerous increase in FPnN₂O. In our experimental model, maintaining the pneumoperitoneum without a leak for almost one hour resulted in an increase of 26.12% ± 3.21% (Fig. 1, "no leak" curve). The same vented volume released in 4-L fractions every 10 minutes kept the N₂O fraction constantly <10% (6.55% ± 0.30% after one hour; P = 0.0001 for the comparison with 26.12% ± 3.2%) (Fig. 1, 4-L leak curve).

For combustion to occur, an ignition source, a combustible material, and an atmosphere supporting combustion need to exist simultaneously. Intraabdominal explosions and fires occurred in the 1970s (1,2) and were reported as late as the mid 1990s (6,7), often with serious consequences for the patient. The cautery device can serve as an ignition source, with bowel gas or introduced foreign bodies as combustible material and various gases as the supporting atmosphere. N₂O contains 2.5 times more oxygen than air and supports combustion. Since the switch from N₂O to CO₂ as the insufflating gas for creation of a pneumoperitoneum, intraabdominal combustion has not been of great concern. The assumption that the effects of inhaled N₂O on a CO₂ pneumoperitoneum were minor was supported by measurements during short procedures (8) or single measurements at the end of the operation (9), which found only small concentrations of N₂O in the CO₂ pneumoperitoneum. However, measurements at a single point or over short periods of time may not reflect adequately the peaks that can occur intraoperatively after prolonged periods without pneumoperitoneal gas turnover.

The presence of a mixed-gas (CO₂/N₂O) pneumoperitoneum can also be of concern when gas embolization occurs. This concern and its clinical implications are currently under investigation.

In our pig model, pneumoperitoneal CO₂ renewal at a rate of 4 L every 10 minutes resulted in an N₂O concentration of <10% (approximately 6%), which would eliminate the concern for combustion. Because the volume of a 12 mm Hg pneumoperitoneum in this experimental model is of the same order as that in the human adult (3 L), a pneumoperitoneal leak of the same order of magnitude (24 L/h) may provide similar results in both situations. This hypothesis needs to be verified in humans. In laparoscopic surgery, the gas flow through the insufflator is typically 10- to 30-fold greater than this rate of CO₂ renewal during the exchange of instruments or during more or less arbitrary venting that occurs in the course of the operation. However, this may not prevent the buildup of potentially harmful concentrations of N₂O during the parts of the procedure that do not involve pneumoperitoneal gas loss.

If the data obtained in this experiment are confirmed in the patient setting, CO₂ renewal at a fresh gas flow of only 28 L/h (24 L/h to replace the constantly vented pneumoperitoneal gas plus 3 to 4 L/h to compensate for uptake of CO₂ from the pneumoperitoneum into the circulation) (10) would be enough to maintain the intraperitoneal gas composition at safe levels. At this rate of gas flow, the risk of the patient’s having hypothermia because of unnecessarily high rates of intraperitoneal gas flow can be minimized. [Core temperature decreases 0.3°C with each 50 L of insufflated CO₂ (11).] In this study, the FPnN₂O did not reach 10% during the observation period with a leak of either 2 or 4 L every 10 min. Thus, this rate of flow may be considered adequate for the sake of security. Unintentional leaks around or through trocars are less prominent with newer materials and techniques and should not be relied on as a substitute for a calibrated leak.

Finally, another solution to the problem of making the peritoneal atmosphere safe would be to constantly monitor its composition. However, the infrared technology used in most capnographs is not sufficient for such monitoring. Furthermore, reference GC-MS devices are expensive, cumbersome, and difficult to operate without extensive training. Portable MGCs are easy to use and adequately sized to fit on an anesthesia machine. In the range tested and for the gases used in our study, the MGC was as accurate as the GC-MS, which requires a much more involved procedure. Use of the MGC could lead to various and more precise on-line analysis of various gases in patient care areas. Besides N₂O, the detection and quantification of the other gases normally confined to the digestive tract (i.e., hydrogen and CH₄) may be of particular interest. We observed peaks in intraperitoneal levels of hydrogen and CH₄ after deliberate opening of the colon in pigs. This observation could be applied to the chemical detection of accidental digestive perforations in humans by monitoring the composition of the pneumoperitoneum.

In conclusion, our experimental model indicates that external venting of the pneumoperitoneum with a calibrated leak of 12 or 24 L/h is enough to prevent the buildup of gases that increase the risk of embolization or explosion. Such a small turnover is efficient only if the external leak is calibrated to approach constant conditions over time and does not occur as high-flow episodes separated by long periods of tight pneumoperitoneum, during which N₂O may accumulate in the CO₂ peritoneal atmosphere. These findings need to be confirmed for longer observation periods and in the human clinical setting.

	Footnotes

 
Presented in part at the annual meeting of the American Society of Anesthesiologists, San Francisco, CA, October 14–18, 2000.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Gunatilake DE. Case report: fatal intraperitoneal explosion during electrocoagulation via laparoscopy. Int J Gynaecol Obstet 1978; 15: 353–7.[Medline]
2. El-Kady AA, Abd-El-Razek M. Intraperitoneal explosion during female sterilization by laparoscopic electrocoagulation: a case report. Int J Gynaecol Obstet 1976; 14: 487–8.[Medline]
3. Neuman GG, Sidebotham G, Negoianu E, et al. Laparoscopy explosion hazards with nitrous oxide. Anesthesiology 1993; 78: 875–9.[Web of Science][Medline]
4. Diemunsch PA, Torp KD, Van Dorsselaer T, et al. Nitrous oxide fraction in the carbon dioxide pneumoperitoneum during laparoscopy under general inhaled anesthesia in pigs. Anesth Analg 2000; 90: 951–3.[Abstract/Free Full Text]
5. Diemunsch PA, Doumbia S, Leroy J. N₂O usage in laparoscopic cases [letter]. Anesth Analg 2000; 91: 1307.[Free Full Text]
6. Use of wrong gas in laparoscopic insufflator causes fire. Health Devices 1994; 23: 55–6.[Medline]
7. Greilich PE, Greilich NB, Froelich EG. Intraabdominal fire during laparoscopic cholecystectomy. Anesthesiology 1995; 83: 871–4.[Web of Science][Medline]
8. Carmeron AE, Dear GL, Pocock TJ, Tennant RW. Gas exchange in abdominal cavity during laparoscopy. J R Soc Med 1983; 76: 1015–8.[Abstract]
9. Weenig CS. N₂O usage in laparoscopic cases. Anesth Analg 2000; 91: 1306–7.[Free Full Text]
10. Joris JL. Anesthesia for laparoscopic surgery: an atlas of regional anesthesia procedures. In: Miller RD, ed. Anesthesia. 5th ed. Philadelphia: Churchill Livingstone, 2000: 2003.
11. Ott DE. Laparoscopic hypothermia. J Laparoendosc Surg 1991; 1: 127–31.[Medline]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Diemunsch, P. A. Articles by Geny, B. Search for Related Content PubMed PubMed Citation Articles by Diemunsch, P. A. Articles by Geny, B. Related Collections Anesthetic Techniques