Secondary Logo

Journal Logo

Original Article

Closed-system anaesthesia for laparoscopic surgery: is there a risk for carbon monoxide intoxication?

Soro, M.; García-Pérez, M. L.; Ferrandis, R.; Aguilar, G.; Belda, F. J.

Author Information
European Journal of Anaesthesiology: June 2004 - Volume 21 - Issue 6 - p 483-488


Laparoscopic cholecystectomy has become the treatment of choice for cholelithiasis [1]. It has recently been demonstrated that the use of diathermy in a poor-in-oxygen environment, such as that produced during the carbon dioxide (CO2) pneumoperitoneum for laparoscopic surgery, causes a significant increase of intraperitoneal carbon monoxide (CO) concentration [2-4]. This produced a concomitant increase in carboxyhaemoglobin (COHb) concentration and the possibility for acute CO intoxication [2,5].

Carbon monoxide toxicity results from the displacement of oxygen from several haemoproteins, especially haemoglobin (Hb). The formation of COHb leads to a reduction in oxygen transport and a left-ward shift of the oxyhaemoglobin dissociation curve, which can result in tissue hypoxia even in the presence of high partial oxygen pressures.

Elimination of CO occurs through its displacement from Hb and its later excretion from the lungs as a gas without undergoing changes [6]. Our hypothesis is that the process of elimination may be impaired in patients anaesthetized with a closed system, where CO excreted through the lungs is accumulated in the breathing circuit and thus re-inhaled in increasing concentrations. This reinhalation of CO may increase the likelihood for intoxication.

The objective of our study was to ascertain the relationship between the intraperitoneal concentration of CO, the alveolar concentration of CO and the COHb blood concentrations in patients undergoing laparoscopic cholecystectomy during general anaesthesia when ventilated with a closed breathing system.


After approval of the experimental protocol by the hospital Ethics Committee, nine consecutive patients diagnosed with cholelithiasis and scheduled for laparoscopic cholecystectomy were enrolled in the study. Patients were two males and seven females with a mean age 57 ± 14 yr (range 45-74 yr), ASA I-III. We recorded a detailed history of preoperative factors that might influence COHb concentrations, such as smoking, exposure to other smokers, automobile use for more than 2 h a day and recent blood transfusion. All patients gave informed written consent to participate in the study.

Anaesthesia and mechanical ventilation of the lungs

Anaesthesia was induced with intravenous (i.v.) propofol 2 mg kg−1, fentanyl 3 μg kg−1 and rocuronium 0.5 mg kg−1, and maintained with propofol (10 mg kg−1 h−1, continuous i.v. infusion) and fentanyl (1 μg kg−1 i.v. bolus as required). Rocuronium in continuous i.v. infusion (0.5 mg kg−1 h−1) was used for muscle relaxation. After tracheal intubation, the patient was connected to a closed-system anaesthesia apparatus (Physioflex®; Dräger, Lübeck, Germany). The lungs were ventilated using 30% oxygen in air with a tidal volume of 8-9 mL kg−1. The respiratory rate was set to maintain normocapnia (PETCO2 4.2-5.0 kPa).

The Physioflex® uses a valveless high-flow (70 L min−1) closed circuit breathing system. The patient's lungs are ventilated by displacement of part of the circulating volume into the lungs by the action of up to four parallel moving membranes, which are powered by a secondary pressure. Exhalation is produced by pressure release in the circuit. The volume circulating is regulated by a computer, which use proportional integration and differentiation algorithms, depending on the membrane's position, before and after each ventilation cycle. Regulation of FiO2 is based on the oxygen loss and has the highest priority. The computer maintains FiO2 concentration and volume with 5 mL pulses of oxygen or the selected carrier gas. To avoid accumulation of foreign gases (nitrogen, water vapour, methane or acetone) the computer displays on a command for a flushing with fresh gas if the fraction of these gases exceeds 5%, or after a fixed period of 1 h [7].


Laparoscopic cholecystectomy was performed by using a standard three-port technique [8]. After induction of general anaesthesia, the abdomen of each patient was insufflated with 100% CO2 until the intra-abdominal pressure was raised to 2 kPa. This pressure level was maintained throughout the procedure. The duration of electrocautery use was determined by start-stop cumulative arithmetic compilation. Factors that might influence the duration of the cautery were recorded as well as any complication that occurred during the procedure.


We measured COHb blood concentrations (as a percentage of the total Hb) and CO concentrations in alveolar and pneumoperitoneum gas (expressed as ppm), at the following four time points throughout surgery: immediately after the setting of the pneumoperitoneum before cautery use (baseline), and at 5, 15 and 30 min after starting with the electrocautery.

Blood samples

Blood samples were obtained from an arterial cannula placed in the radial artery following induction of anaesthesia. Blood concentrations of COHb were measured by using a previously calibrated co-oximeter (OSM3 Hemoximeter®; Radiometer, Copenhagen, Denmark). The manufacturer listed the accuracy of this instrument as ±1%.

Gas samples

Measurement of CO from the gas samples was performed by using a previously calibrated Micro CO-Meter® (Micro Medical, Rochester, England). The Micro CO-Meter® is a diagnostic device for measuring alveolar CO concentration. Measurements are obtained from a single expiration combined with a breath hold countdown timer. It detects CO concentrations between 1 and 600 ppm with a sensitivity of 1 ppm.

In order to determine the intraperitoneal concentrations of CO, gas samples from the pneumoperitoneum were obtained from the trocar which was positioned in the right lower abdomen by using a three-way stopcock connected by a silicone tubing with the Micro CO-Meter®. At the moment of sampling, the surgeon opened the tap, allowing the gas to flow towards the measuring device.

Alveolar CO concentrations were measured by gas sampling from the patient's endotracheal tube. At the moment of sampling, the endotracheal tube was clamped at end-inspiration, disconnected from the ventilation system Y-piece and connected to the Micro CO-Meter®. Then the clamp was removed so that the volume expired by the patient passed through the CO-Meter®, thus allowing measurement. Meanwhile controlled ventilation of the lungs was maintained by connecting the Y-piece to a test lung (with no residual volume); this way the ventilator detected just a minimal leak and continued to function normally. After the measurement was completed, the endotracheal tube of the patient was again connected to the Y-piece of the system to continue normal ventilation.

The elapsed time (in minutes) from the start of mechanical ventilation of the lungs to the moment in which the ventilator indicated the need of circuit flushing (indicating an excessive accumulation of foreign gases) was also recorded for each patient.

Statistical analyses

Descriptive statistics were used for data analysis and summary values in the text are expressed as mean ± standard error of the mean (SEM). In order to select the statistic test for comparing the means of related groups, we first determined the distribution type with the Kolgomorov-Smirnov test. The Friedman non-parametric test was used to determine differences between several groups. Afterwards we applied the Wilcoxon signed rank sum test to find the pairs with differences. Statistically significant differences were defined as P < 0.05.


We studied nine patients, seven females and two males, physical status ASA I-III, with a mean age of 57 ± 14 yr and a mean weight of 69 ± 12 kg. Factors possibly affecting the base COHb blood concentrations present in these patients were smoking (10-20 cigarettes per day) and living in an urban environment.

The mean duration of pneumoperitoneum was 42 ± 13 min (range 35-58 min). Mean cumulative electrocautery time was 2.4 ± 1.8 min (range 34 s to 4 min). Flushing of the ventilator circuit was required after a mean time of 39 ± 12 min after initiation of mechanical ventilation; therefore, flushing did not affect our results. The gallbladder showed minimal adhesions and inflammation in most patients and no complications were recorded. These results are shown in Table 1.

Table 1
Table 1:
General characteristics of the patients and intraoperative data (mean ± standard deviation (SD)).

Intraperitoneal CO concentrations increased from 0 ppm at baseline to 135 ± 212 (range 9-453) ppm at 5 min, to 481 ± 151 (range 1-592) ppm at 15 min and to 199 ± 111 (range 113-325) ppm at 30 min, after electrocautery. Statistically significant differences were found when comparing data from baseline to the minutes 5, 15 and 30 and from 15 to 30 min.

The CO in the exhaled gas was of 5.2 ± 3 (range 2-8) ppm at baseline and increased later to 26.8 ± 27 (range 3-66) ppm at 5 min, to 21.4 ± 25 (range 6-66) ppm at 15 min and to 7.5 ± 1 (range 6-9) ppm at 30 min after electrocautery. No significant differences were found for these values.

The COHb blood concentration at baseline was of 1.41 ± 0.4 (range 0.7-2.3) percent. It was of 1.30 ± 0.5 (range 0.4-3) percent at 5 min, 1.31 ± 0.5 (range 0.6-2.8) percent at 15 min and 1.28 ± 0.4 (range 0.5-1.8) percent at 30 min after electrocautery. These values were not significantly different. These results are shown in Figure 1.

Figure 1
Figure 1:
Intraperitoneal and alveolar (expired) CO and COHb concentrations in patients undergoing laparoscopic cholecystectomy during ventilation of the lungs with a closed anaesthesia breathing system. ♦: intraperitoneal CO; ●: expired CO; ▪: COHb.


Our study confirms that electrocautery produces significant increases of intraperitoneal CO during laparoscopic surgery. It also shows that CO diffusion from the peritoneum to the blood is negligible as it does not produce a rise in COHb concentration, nor a significant alveolar accumulation of CO, in spite of the absence of pulmonary elimination when using a closed system. This clearly shows that during the clinical conditions of laparoscopic cholecystectomy, the use of electrocautery does not imply a risk of CO intoxication. However, the results deserve further analysis.

Alveolar CO and COHb production

Several authors have assumed that peritoneal absorption of CO occurs during laparoscopic surgery, although in most cases no concomitant increase of COHb was found. Pulmonary CO elimination was indicated as the main reason for the absence of COHb increase. However, in our study, rebreathing of all the CO eliminated by the lungs resulted in no increase in COHb. To explain this, it is necessary to know which COHb value would be found in a patient exposed to a CO concentration similar to that of our study. We therefore applied the Haldane equation [9] which allows a prediction of the maximal COHb concentration produced according to the pulmonary partial pressures of O2 and CO: Equation (1)

In this equationM is the Haldane constant (value: 200-250). The alveolar CO concentrations found in our study at 5 and 15 min are circa 25 ppm. These figures are related to an alveolar CO concentration of 0.0025 kPa and a PCO of 0.2 kPa. The FiO2 of 0.3 in our study correlates with an alveolar PO2 of 22 kPa (the PCO2 being 5.3 kPa).

By substituting these values and an M value of 230 in equation (1), we obtain:

This result tells us that when the COHb value is in equilibrium, while the patient is breathing a gas mixture containing 25 ppm CO, the CO concentration would be 2.5%. This value is higher than the one found in our study: 1.3 ± 0.5%.

Tyuma [10] generalized the classical Koburn-Foster-Kane equation, which predicts the concentrations of COHb in relation to the inspired CO, the exposure time and other factors, such as blood volume and alveolar ventilation (VA). He proposed a formula that allows the calculation of the necessary time to reach an equilibrium between the inspired CO and COHb, that is, the time needed to reach the maximum value of COHb for that particular inspired CO concentration: Equation (2)

in which:

COHb = % COHb at the moment of equilibrium (COHbmax);

C = Oxygen (and CO) capacity of blood = Hb (g mL−1) × 1.39 (mL g−1) = 0.2 mL mL−1;

VB = body blood volume (assumed as 5000 mL for 70 kg body weight);

D = (1/DLCO) + (713/VA)

where DLCO is the pulmonary CO diffusion capacity (mL min−1 kPa−1).

For our patients with a mean body weight of 69 kg, a calculated VA of 4800 mL min−1 and taking a DLCO of 180 mL min−1 kPa−1, C · VB · D would result in a value of about 190. By substituting this in equation (2), the time required to reach a concentration of 2.5% COHb (calculated with the Haldane formula) when the inspired CO concentration is 25 ppm (PCO = 0.2 kPa), would be 264 min (4.4 h). Using the same equation of Tyuma (which was experimentally validated for any inspired CO concentration) in our study, we observed that the exposure of our patients to a pulmonary concentration of 25 ppm CO for a period of less than 30 min would result in maximal COHb blood concentrations of 0.7%. Obviously, the alveolar concentration of 7.5 ppm found at 30 min in our study would result in a much lower COHb value.

If alveolar CO of only 25 ppm produces a COHb value below 0.7% when rebreathing of CO in the ventilatory circuit, it is reasonable that the majority of authors did not find significant variations in COHb when using an open ventilatory circuit [2,4,11]. The only study not in line with these findings is the one by Ott, which reports considerable increases of COHb [5].

We did not find any change in the concentrations of COHb, which remained at a mean concentration of 1.3 ± 0.5% (range 0.4-3%) from the start. This value may be considered as the physiological resulting from the normal breakdown of Hb (COHb around 1%) [12].

Accumulation of CO in closed systems

On the other hand, it remains uncertain why in our study CO does not accumulate in the lungs given the fact that it is not eliminated from the ventilatory circuit. It seems reasonable to suppose that the alveolar CO is in equilibrium with the COHb in the blood and with the peritoneal CO. This way if peritoneal CO does not accumulate because it is eliminated through the trocars, it would not accumulate in the lungs either. In our study the concentrations of expired CO further indicate that in the closed system the effect of rebreathing is not significant for CO liberated from COHb. Only in one other study the concentrations of alveolar CO have been measured in anaesthetized patients during gynaecological laparoscopic surgery using the same closed system as ourselves (Physioflex®) [13]. Carbon monoxide concentrations of 20-26 ppm were found in alveolar gas samples measured every 15 min during more than 100 min of surgery.

We found that pulmonary CO concentrations during laparoscopic surgery remained far below the safety limit for exposure to inhaled CO established by the Environmental Protection Agency (EPA) and the Occupational Safety and Health Administration (OSHA) even when using closed-system anaesthesia. According to the EPA, the maximum allowable 1 h exposure to ambient CO is 35 ppm, with a ceiling concentration of 200 ppm. The maximum allowable concentration of ambient CO set by the OSHA is 50 ppm for 8 h of exposure or 400 ppm for 15 min [14].

Intraperitoneal CO variations and diffusion into the blood

In our study the intraperitoneal CO concentration reached peak levels of 481 ± 151 ppm after 15 min of electrocautery; these concentrations, although they build up at different rates, are comparable to those found by other investigators using electrocautery for similar periods [2-5]. At 30 min (after the start of electrocautery) the intraperitoneal CO concentrations had decreased significantly to 199 ± 111 ppm. Knowing that no possibility of pulmonary CO elimination existed, we attributed the decrease in peritoneal CO concentration to the frequent intraperitoneal smoke flushing through the trocars in order to improve visibility for the surgeon. Moreover, the CO2 insufflation device maintained constant intra-abdominal pressure, by either bleeding gas from the system when the intra-abdominal pressure was greater than a preset value of 2 kPa, or adding up gas when the intra-abdominal pressure went down to a lower value. The diffusion of CO into the blood, which we assume that occurs, does not explain such sudden decreases of the intraperitoneal CO.

Pulmonary inhalation during 15 min of CO concentrations, equivalent to the measured peritoneal CO values, would produce COHb concentrations of 5% [10,15] which would increase to 10% if the inhalation was continued up to 30 min. However, the baseline COHb concentrations found in our study did not change, either at 15 or 30 min. This indicates that during the maintenance of the pneumoperitoneum the peritoneal CO diffusion is much lower than the pulmonary diffusion. Beebe concluded, just as we did, that the peritoneal membrane does not allow an efficient CO diffusion [4].

Several factors might explain this low effective diffusion of CO across the peritoneum during laparoscopic surgery. In the first place, it is possible that the peritoneal blood flow (which normally is about 1 L min−1) is much reduced, as the portal circulation is very sensitive to changes in the venous pressure [16], which in these patients under pneumoperitoneum, is at least 2 kPa. Secondly, the peritoneal atmosphere rich in CO2 causes a rightward shift of the COHb dissociation curve (which is identical to the one for HbO2) which will cause a reduction in the CO uptake of Hb in the peritoneum vessels [4]. Finally, the peritoneal membrane is thicker than the alveolar membrane.


We found that during laparoscopic surgery in adults with a mean electrocautery time use of 2.4 min, no significant increase in COHb was produced, even in absence of pulmonary CO removal (closed-system anaesthesia). The most probable hypothesis for this to happen is a low peritoneal CO absorption. Under clinical conditions, the CO accumulated in the lungs (in equilibrium with the peritoneal CO) does not cause COHb concentration to increase. Furthermore, even when using a closed system, pulmonary CO levels remain far below the safety limits established for CO exposure. Thus we can state with certainty that no CO intoxication will be produced under clinical conditions as long as electrocautery is used for reasonable periods of time and the intraperitoneal gas is regularly vented.


1. Cunningham AJ, Brull SJ. Laparoscopic cholecystectomy: anesthetic implications. Anesth Analg 1993; 76: 1120-1133.
2. Esper E, Russell TE, Coy B, Duke III BE, Max MH, Coil JA. Transperitoneal absorption of thermocautery-induced carbon monoxide formation during laparoscopic cholecystectomy. Surg Laparosc Endosc 1994; 4: 333-335.
3. Wu JS, Monk T, Luttmann DR, Meininger TA, Soper NJ. Production and systemic absorption of toxic byproducts of tissue combustion during laparoscopic cholecystectomy. J Gastrointest Surg 1998; 2: 399-405.
4. Beebe DS, Swica H, Carlson N, Palahniuk RJ, Goodale RL. High levels of carbon monoxide are produced by electrocautery of tissue during laparoscopic cholecystectomy. Anesth Analg 1993; 77: 338-341.
5. Ott DE. Carboxyhaemoglobin due to peritoneal smoke absorption from laser tissue combustion at laparoscopy. J Clin Laser Med Surg 1998; 16: 309-315.
6. Meredith T, Vale A. Carbon monoxide poisoning. BMJ 1988; 296: 71-79.
7. Rendell-Baker L. Future directions in anesthesia apparatus. In: Eisenkraft JB, ed. Anesthesia Equipment. Principles and Applications. Chicago, USA: Mosby-Yearbook Inc, 1993: 691-694.
8. Rosin LD. Laparoscopic cholecystectomy. In: Zinner MJ, ed. Maingot's Abdominal Operations. Connecticut, New York, USA: Appleton & Lange, 1997: 1855-1865.
9. Cotes JE. Exchange of gas in the lung. In: Cotes JE, ed. Lung Function, 5th edn. Oxford, UK: Blackwell Scientific, 1993: 263-298.
10. Tyuma Y, Ueda Y, Imaizumi K, Kosaka H. Prediction of the carboxyhemoglobin levels during and after carbon monoxide exposures in various animal species. Jap J Physiol 1981; 31: 131-143.
11. Nezhat C, Seidman DS, Vreman HJ, Stevenson DK, Nezhat F, Nezhat C. The risk of carbon monoxide poisoning after prolonged laparoscopic surgery. Obstet Gynecol 1996; 88: 771-774.
12. Cathleen C, Toby L. Poisoning. In: Shoemaker WC, Ayres S, Grenvik A, Holbrook PR, Thompson WL, eds. Textbook of Critical Care, 3rd edn. Pennsylvania, USA: Saunders, 1995: 1186-1210.
13. Versichelen L, Rolly G, Vermeulen H. Accumulation of foreign gases during closed-system anaesthesia. Br J Anaesth 1996; 76: 668-672.
14. Moon R. Cause of poisoning relation to halogenated agents still not clear. Anesth Patient Safety Found Newslett 1994; 9: 13-16.
15. Peterson JE, Stewart RD. Absorption and elimination of carbon monoxide by inactive young men. Arch Environ Health 1970; 21: 165-171.
16. Strunin I. The splanchnic, hepatic and portal circulations. In: Prys-Roberts C, ed. The Circulation in Anaesthesia. Oxford, UK: Blackwell Scientific, 1980: 241-251.


© 2004 European Academy of Anaesthesiology