There has been a considerable increase in laparoscopic surgery during the last decade. It is predicted that in the next few years, laparoscopic techniques will be used in 50%–60% of all intraabdominal procedures (1). The induction of a pneumoperitoneum is necessary for most laparoscopic procedures. In general, pneumoperitoneum is created by insufflation of carbon dioxide (CO2) into the peritoneal space. A number of investigators have demonstrated significant alterations in the cardiovascular system after peritoneal insufflation of CO2 (2–4).
There is still controversy regarding the effects of pneumoperitoneum on splanchnic and hepatic perfusion. Several animal studies have demonstrated that insufflation of any gas may cause marked reduction in splanchnic and liver perfusion and may even induce intestinal ischemia associated with oxygen free radical production and bacterial translocation (5–9). Furthermore, approximately 50% of patients undergoing laparoscopic cholecystectomy react with a slight increase in liver enzymes and it has been speculated that this increase is attributable to impaired liver perfusion (10,11). Thus, intermittent deflation or even gas-free abdominal wall shift has been recommended in patients with compromised liver function or in critically ill patients (9). Conversely, Blobner et al. (12) have demonstrated in a pig model, that insufflation of CO2 with intraabdominal pressure (IAP) levels less than 16 mm Hg induces an increase in splanchnic perfusion. The increase in splanchnic perfusion was explained by the possible local vasodilative effects of CO2 on splanchnic vessels.
Currently very limited data on the effects of intraabdominally insufflated CO2 on splanchnic and hepatic perfusion in humans are available. The objective of the present study was to investigate the effects of insufflated CO2 on hepatic venous blood flow during laparoscopic surgery.
Eighteen consecutive patients undergoing laparoscopic cholecystectomy and 15 patients undergoing laparoscopic inguinal hernia repair were prospectively enrolled in this study. A group of 12 patients undergoing conventional hernia repair served as a control group. Exclusion criteria were atrial fibrillation or any other known cardiovascular disease as well as diseases of the esophagus or stomach. Patients with atrial fibrillation were excluded because blood flow measurements were performed over two cardiac cycles requiring a constant RR-interval. The study was approved by the ethics committee of the University Ulm. Written informed consent was obtained from all patients before inclusion in the study.
All patients received oral premedication with clorazepate dipotassium (Tranxilium®) (20 mg) in the evening and 1 h before induction of general anesthesia. In all patients standard clinical monitoring was applied, including continuous electrocardiogram (ECG), noninvasive oscillometric blood pressure, pulse oximetry, continuous airway pressure, and end-tidal CO2 partial pressure. Anesthesia was induced following a standardized regime of target controlled infusion of propofol (initial target plasma concentration, 3–4 μg/mL), continuous infusion of remifentanil (0.25–0.30 μg · kg−1 · min−1), and a bolus of mivacurium (0.25 mg/kg) (13). As routinely performed in our clinic, propofol was delivered via a commercially available syringe pump (Graseby 3500®; Graseby, Watford, UK), incorporating the Diprifusor® module to deliver propofol by target controlled infusion. After laryngoscopy and tracheal intubation total IV anesthesia was maintained with target controlled infusion of propofol (target plasma concentration, 2–3 μg/mL), continuous infusion of remifentanil (0.10–0.25 μg · kg−1 · min−1), and mivacurium (4–6 μg/kg/min). Inspired oxygen fraction was kept between 0.30 and 0.35 without use of nitrous oxide. All patients were ventilated on a pressure controlled mode with a positive end-expiratory pressure (PEEP) of 5 cm H2O and a respiratory rate of 8–10 breaths/min. End-tidal CO2 (ETco2) was kept constant between 35 and 40 mm Hg by adjusting respiratory rate and inspiratory pressure levels.
Blood flow of the right and middle hepatic vein was assessed by use of multiplane transesophageal echocardiography (TEE). After induction of general anesthesia a 5.0/3.7 MHz multiplane transesophageal probe (Omniplane I, Hewlett-Packard Inc., Andover, MA) was introduced. The probe was connected to a Hewlett-Packard Sonos 5500 echocardiograph. For visualization of the hepatic veins the tip of the probe was advanced into the antrum of the stomach and flexed anteriorly. We (14) described in detail how to visualize the right and middle hepatic veins by multiplane TEE and how to acquire Doppler sonography curves. At the different times the diameter and the Doppler signal of the right and middle hepatic vein were obtained and stored on a magneto-optical disk. The Doppler signal was attained by pulsed wave Doppler technique (PW mode). The sample volume was placed in the center of the vessel at exactly the location that was used for measuring the diameter. Correction of the angle between the Doppler beam and the flow axis was performed for each Doppler measurement. In all measurements the angle between flow axis and Doppler beam was below 60 degrees. Both the vessel diameter and the Doppler signal were obtained at end-expiration. To verify the end of expiration an external breathing circuit pressure gauge (Fa. Draeger, Lübeck, Germany) was connected to the echocardiograph to visualize the airway pressure on the screen. Blood flow in the right and middle hepatic vein was calculated using the following formula:
where VTI = time velocity integral, π r2 = cross-sectional area of the vessel, HR = heart rate, and k = 0.7) (15,16). The VTI is the area under the Doppler curve over one cardiac cycle and represents the distance the blood travels during one cardiac cycle. The correction factor k is derived from an experimental study in pigs and takes into account that the blood flow is not flat but has a parabolic velocity profile (17).
Blood flow index of the right and middle hepatic veins was calculated by dividing the blood flow by the body surface area.
Echocardiographic measurements were analyzed off-line using the HP Sonos 5500 ultrasound system software. HR was derived from the ECG on the echocardiogram. Hepatic vein diameters were measured according to the leading edge to leading edge method. The VTI was determined by manual tracing of the outer shape of the Doppler curve and calculated by the integrated software of the echocardiograph. Two cardiac cycles were evaluated and averaged. All Doppler curves were evaluated by the same non-blinded observer (WS).
Measurements of the blood flow in the right and middle hepatic veins were performed at baseline (Tbase) and 5, 10, 20, 30, 40 min after insufflation of CO2 (T5, T10, T20, T30, T40) as well as 1 and 5 min after deflation (TE, TE5). At the same time, the hemodynamic variables (HR, mean arterial blood pressure) and ventilatory variables (minute ventilation, maximum inspiration pressure, PEEP, ETco2) were recorded. In all patients CO2 pneumoperitoneum was maintained at a pressure level of 12 mm Hg. For the control group opening and closure of the abdominal fascia was set equal with the beginning and end of pneumoperitoneum, respectively.
For assessment of the reproducibility of hepatic blood flow determination in 6 randomly selected patients the measurements at all 8 times were repeated more than 6 mo later by the same observer (WS) and independently by a second observer (RM). Thus assessment of intraobserver and interobserver reproducibility was based on 48 measurements. As accurate blood flow measurements depend on a constant RR-interval the variability of two consecutive RR-intervals in 187 measurements was evaluated.
All data are presented as median and 95% confidence intervals of the median unless otherwise stated. Intragroup changes in hemodynamic and echocardiographic variables over time were analyzed using analysis of variance on ranks for repeated measurements (Friedmann) and, if significant, Dunn's method was used to compare the variables with the baseline value. Intergroup differences between patients undergoing laparoscopic surgery and the control group at the different times were analyzed using the Mann-Whitney rank sum test. Statistical significance was assumed when P values were <0.05.
Reproducibility of hepatic blood flow measurements was analyzed by intraobserver and interobserver variability, simple regression analysis and bias analysis according to Bland and Altman (18). As a measure of intraobserver and interobserver variability, the mean percentage error was calculated as the absolute difference between two measurements divided by the mean of the two observed values. The mean difference and repeatability coefficient were calculated according to Bland and Altman. The repeatability coefficient is defined as two standard deviations of the differences between two repeated measurements (19).
Nine patients undergoing laparoscopic surgery (six cholecystectomy, three inguinal hernia repair) were excluded from analysis because of poor quality echocardiographic images after insufflation of CO2. Thus the final evaluation was based on 24 patients undergoing laparoscopic surgery (12 cholecystectomy, 12 inguinal hernia repair) and 12 patients undergoing conventional hernia repair (control group). The mean age of the laparoscopic group was 44 yr (range, 21–79 yr) and of the control Group 51 yr (range, 32–73 yr). The difference was not statistically significant (P = 0.17).
As shown in Table 1 insufflation of CO2 resulted in a significant increase in mean arterial blood pressure, whereas HR was not affected by insufflation of CO2. The control group revealed a slight but significant decrease in heart rate after baseline measurement. After induction of pneumoperitoneum a significantly higher inspiration pressure was necessary to maintain normocapnia. Despite a significant increase in ETco2 after insufflation of CO2, ETco2 was maintained within normal range throughout the study period (Table 1).
Five minutes after insufflation of CO2 the diameter of the middle hepatic vein was significantly reduced in comparison with the baseline value and in comparison with the control group (Table 2). The diameter of the right hepatic vein was less affected by the increased IAP with a tendency to decrease that did not reach statistical significance. CO2 pneumoperitoneum induced a significant increase in VTI of the right and middle hepatic vein. The increase in VTI remained constant during pneumoperitoneum and was significant in comparison to the control group. Conversely, in the control group the diameters and VTIs did not change over the entire study period (Table 2).
Insufflation of CO2 produced a marked and significant increase in blood flow index of the right hepatic vein (Table 3). The increase persisted during the entire period of pneumoperitoneum and was significant in comparison with the control group. In contrast the control group revealed a constant blood flow index in the right hepatic vein over the whole study period (Table 3).
After induction of pneumoperitoneum the increase in blood flow index of the middle hepatic vein was less pronounced but also reached significance (Table 3). Ten to 40 min after insufflation of CO2 the blood flow index in the middle hepatic vein was significantly higher in comparison with the baseline value. The control group showed a constant blood flow index in the middle hepatic vein. However, comparison between groups did not reveal significant differences at the different times.
As demonstrated in Table 4 the increase in hepatic blood flow could be observed in patients undergoing laparoscopic cholecystectomy as well as in patients undergoing laparoscopic inguinal hernia repair even if the increase in the middle hepatic vein did not reach significance in the cholecystectomy group (P = 0.054).
An example for the increase in blood flow of the middle hepatic vein after induction of CO2 pneumoperitoneum is shown in Figure 1.
Intraobserver and interobserver variability and results of the linear regression and bias analysis for assessment of the reproducibility of hepatic blood flow measurements are displayed in Table 5. The observed increase in blood flow of the right and middle hepatic veins was markedly larger than the repeatability coefficient, indicating that this observation was not attributable to a lack of reproducibility of the TEE technique. The mean variability of two consecutive RR-intervals was 1.1%.
The major finding of the present study is that CO2 pneumoperitoneum with an IAP of 12 mm Hg induces a significant increase in hepatic venous blood flow in healthy patients undergoing laparoscopic surgery. This finding is in contrast to the assumption that CO2 pneumoperitoneum may be harmful for liver function due to impaired perfusion, which was mainly based on experimental studies in animals (6–9).
The increase in hepatic blood flow was observed in both the right and middle hepatic veins. However, the increase was more pronounced in the right hepatic vein. This observation might be a result of different anatomic localizations of both veins. Whereas the middle hepatic vein was significantly compressed by the increased IAP, the diameter of the right hepatic vein remained almost unchanged.
There has been controversy regarding the effect of pneumoperitoneum on splanchnic and hepatic perfusion. Several animal studies have focused on this question. Junghans et al. (7) demonstrated in a pig model that an increased IAP after insufflation of any gas may induce a pressure-dependant reduction in splanchnic and hepatic perfusion. Eleftheriadas et al. (8) have shown that reduced splanchnic perfusion may be associated with intestinal ischemia, causing a bacterial translocation from the intestinal lumen to mesenteric lymphatic nodes, the liver and spleen, and oxygen free radical production in the spleen, liver, and lung. In contrast, Blobner et al. (12) recently demonstrated in a pig model that insufflation of CO2 at an IAP less than 16 mm Hg induces an increase in mesenteric artery and portal venous blood flow. As this effect was not observed during insufflation of air at the same IAP, the increase in the blood flow was explained by local vasodilative effects of CO2 on splanchnic vessels. However, a further increase in the IAP level more than 16 mm Hg was associated with a decrease in splanchnic perfusion.
Regarding the effects of pneumoperitoneum on splanchnic and liver perfusion in humans, only limited data are available at present. Two studies focused on gastric intramucosal pH (pHi) as an indicator of splanchnic perfusion with conflicting results. Although Koivusalo et al. (20) observed a decrease in pHi, Thaler et al. (21) did not find any changes in pHi during laparoscopic cholecystectomy. However, as neither local nor systemic CO2 resorption can be estimated during CO2 pneumoperitoneum a method based on measurement of local CO2-homeostasis may be profoundly influenced by use of CO2 insufflation and may therefore be inadequate to mirror changes in local perfusion.
Odeberg et al. (22) estimated hepatic blood flow by use of the indocyanine clearance technique in five patients undergoing laparoscopic cholecystectomy. They did not find any relevant changes in hepatic blood flow at IAP levels of 11–13 mm Hg.
Sato et al. (23) assessed blood flow in the middle hepatic vein by use of TEE and found a significant and time-dependent reduction in hepatic vein flow during laparoscopic cholecystectomy. In contrast, in the present study we observed a marked blood flow increase in the right hepatic vein and a moderate blood flow increase in the middle hepatic vein during CO2 pneumoperitoneum. We do not have a definite explanation for these discrepant findings. However, the anesthetic techniques used were quite different and may have affected the hemodynamic responses to CO2 pneumoperitoneum. Sato et al. combined continuous epidural anesthesia with general anesthesia using nitrous oxide and isoflurane. In the present study propofol and remifentanil were used for general anesthesia and no additional regional technique was applied. Activation of the sympathetic nervous and neurohumoral vasoactive system after CO2 insufflation is well recognized and responsible for the increase in systemic vascular resistance and arterial blood pressure (24). We assume that in the present study the increase in hepatic blood flow was a result of an increase in arterial blood pressure, even if this was only moderate, in combination with local vasodilative effects of CO2 on splanchnic vessels. By use of epidural anesthesia the activation of the vasoactive system after insufflation of CO2 may be profoundly inhibited. The missing increase in systemic blood pressure after CO2 insufflation in patients with epidural anesthesia may be one explanation for the different findings between the two studies.
Two previous studies did use TEE in an attempt to assess hepatic venous blood flow (23,25). Gardebäck et al. (25) used a biplane TEE probe to assess changes in hepatic blood flow during cardiopulmonary bypass in patients undergoing cardiac surgery. Because of the biplane technique they were not able to visualize the different hepatic veins in a long axis view. Sato et al. (23) for the first time used multiplane TEE for determination of blood flow in the middle hepatic vein during laparoscopic surgery as described above. We described in detail the visualization of the three hepatic veins and the assessment of Doppler sonography curves by TEE (14). We have demonstrated that assessment of blood flow in the left hepatic vein is not possible in most patients, as the angle of insonation is more than 60°. Thus one major limitation of TEE for assessment of hepatic blood flow is that TEE does not allow determination of total hepatic blood flow and is therefore limited to assess blood flow changes in the right and middle hepatic veins. In a pig model we have shown that changes in hepatic blood flow may be reliably and quantitatively detected by TEE (17). In the present study we found an intraobserver and interobserver variability of TEE-based hepatic blood flow measurements between 6.8% and 9.2%. This seems to be acceptable for blood flow measurements. By comparison, for thermodilution cardiac output measurements a variability of 5%–10% is well documented (26). Although TEE-based cardiac output determination is generally accepted, only rare data concerning intraobserver and interobserver variability are available. With respect to the variability of diameter and VTI measurements our results lie exactly within the range others found for the aortic valve area and aortic blood flow measurements in the context of cardiac output determination (27–29).
The formula used for Doppler sonographic calculation of blood flow has been applied for many clinical questions (15,16). Usually this formula is used without a correction factor k assuming that the velocity profile within the vessel is flat. This assumption seems to be accurate for the velocity profile within valve orifices, as several studies have demonstrated good agreement between Doppler echocardiographic and invasive determination of cardiac output (27–29). Within small vessels the velocity profile is more parabolic. Therefore using the formula without a correction factor for venous vessels will result in an overestimation of the true blood flow. For assessment of portal venous flow a correction factor of 0.57 has been described (30). From our experimental studies in pigs (17) we derived a correction factor of 0.7 for assessment of hepatic venous flow. In that study we were able to compare total hepatic blood flow measurements assessed by ultrasonic flow probes and by TEE in 14 pigs under different hemodynamic conditions. The calculation of the correction factor was based on 42 simultaneous measurements.
The present study contains several limitations. First, we did not measure cardiac output. Therefore we cannot relate the increase in hepatic blood flow to changes of cardiac output. As it was our primary goal to obtain accurate hepatic blood flow measurements, we did not want TEE probe manipulation during the repeated measurements. Thus we decided not to measure cardiac output by TEE. The invasive procedure of cardiac output measurements by a pulmonary artery catheter did not seem to be justifiable. However, the influence of pneumoperitoneum on cardiac output was investigated in several previous studies and no relevant changes in healthy adults were found (31,32).
Second, induction of pneumoperitoneum was associated with a reduction in TEE image quality in some patients. Therefore, 9 of 33 patients undergoing laparoscopic surgery had to be excluded from analysis.
Third, because of the respiratory shift of the liver it is not possible to determine the Doppler signal throughout the entire respiratory cycle without loss of the hepatic vein by the Doppler sample volume. Therefore in the present study we calculated the blood flow from Doppler signals obtained at the end of exhalation. As the hepatic venous blood flow may decrease during inspiration, calculated hepatic blood flow could be overestimated by this method.
Fourth, we did not specifically investigate the effects of changes in body position, although in patients undergoing laparoscopic inguinal hernia repair a slight Trendelenburg was used and in patients undergoing laparoscopic cholecystectomy a reverse Trendelenburg was used. However, during the first measurement after insufflation of CO2 all patients were still in a supine position. Furthermore the increase in hepatic blood flow was observed in both groups.
In summary we have demonstrated that establishment of a pneumoperitoneum with CO2 at low IAP levels is associated with an increase in hepatic blood flow in healthy patients.
1. Soper NJ. Laparoscopic general surgery: past, present and future. Surgery 1993;113:1–3.
2. Wahba RWM, Beique F, Kleiman SJ. Cardiopulmonary function and laparoscopic cholecystectomy. Can J Anaesth 1995;42:51–63.
3. Joris JL, Noirot DP, Legrand MJ, et al. Hemodynamic changes during laparoscopic cholecystectomy. Anesth Analg 1993;76:1067–71.
4. Harris SN, Ballantyne GH, Luther MA, Perrino AC. Alterations of cardiovascular performance during laparoscopic colectomy: a combined hemodynamic and echocardiographic analysis. Anesth Analg 1996;83:482–7.
5. Ishizaki Y, Bandai Y, Shimomura K, et al. Safe intraabdominal pressure of carbon dioxide pneumoperitoneum during laparoscopic surgery. Surgery 1993;114:549–54.
6. Diebel LN, Wilson RF, Dulchavasky SA, Saxe J. Effect of increased intraabdominal pressure on hepatic arterial, portal venous, and hepatic microcirculatory blood flow. J Trauma 1992;33:279–82.
7. Junghans T, Böhm B, Gründel K, et al. Does pneumoperitoneum with different gases, body positions, and intraperitoneal pressures influence renal and hepatic blood flow? Surgery 1997;121:206–11.
8. Eleftheriadas E, Kotzampassi K, Papanotas K. Gut ischemia, oxidative stress, and bacterial translocation in elevated abdominal pressure in rats. World J Surg 1996;20:11–6.
9. Richter S, Olinger A, Hildebrand U, et al. Loss of physiologic hepatic blood flow control (“hepatic arterial buffer response”) during CO2
-pneumoperitoneum in the rat. Anesth Analg 2001;93:872–7.
10. Saber AA, Laraja RD, Nalbandian HI, et al. Changes in liver functions tests after laparoscopic cholecystectomy: not so rare, not always ominous. Am Surg 2000;66:699–702.
11. Andrei VE, Schein M, Margolis M, et al. Liver enzymes are commonly elevated following laparoscopic cholecystectomy: is elevated intra-abdominal pressure the cause? Dig Surg 1998;15:256–9.
12. Blobner M, Bogdanski R, Kochs E, et al. Effects of intraabdominally insufflated carbon dioxide and elevated intraabdominal pressure on splanchnic circulation. Anesthesiology 1998;89:475–82.
13. Vuyk J, Mertens MJ, Olofsen E, et al. Propofol anesthesia and rational opioid selection: determination of optimal EC50-EC95 propofol-opioid concentrations that assure adequate anesthesia and a rapid return of consciousness. Anesthesiology 1997;87:1549–62.
14. Meierhenrich R, Gauss A, Georgieff M, Schütz W. Use of multi-plane transoesophageal echocardiography in the visualization of the main hepatic veins and acquisition of Doppler sonography curves. Comparison with the transabdominal approach. Br J Anaesth 2001;87:711–7.
15. Feigenbaum H. Hemodynamic information derived from echocardiography. In: Feigenbaum H, ed. Echocardiography. Philadelphia: Lea & Febiger, 1994:181–4.
16. Gill RW. Measurement of blood flow by ultrasound: accuracy and sources of error. Ultrasound Med Biol 1985;11:625–41.
17. Schütz W, Meierhenrich R, Traeger K, et al. Is it feasible to monitor total hepatic blood flow by use of transesophageal echography? An experimental study in pigs. Intensive Care Med 2001;27:580–5.
18. Bland JM, Altman DG. Statistical methods of assessing agreement between two methods of clinical measurement. Lancet 1986;I:307–10.
19. British Standards Institution. Precision of test methods I: Guide for the determination and reproducibility for a standard test method (BS 5497, part1). London: BSI, 1979.
20. Koivusalo AM, Kellokumpu I, Ristkari S, Lindgren L. Splanchnic and renal deterioration during and after laparoscopic cholecystectomy: a comparison of the carbon dioxide pneumoperitoneum and the abdominal wall lift method. Anesth Analg 1997;85:886–91.
21. Thaler W, Frey L, Marzoli P, Messmer K. Assessment of splanchnic tissue oxygenation by gastric tonometry in patients undergoing laparoscopic and open cholecystectomy. Br J Surgery 1996;83:620–4.
22. Odeberg S, Ljungqvist O, Sollevi A. Pneumoperitoneum for laparoscopic cholecystectomy is not associated with compromised splanchnic circulation. Eur J Surg 1998;164:843–8.
23. Sato K, Kawamura T, Wakusawa R. Hepatic blood flow and function in elderly patients undergoing laparoscopic cholecystectomy. Anesth Analg 2000;90:1198–202.
24. Koivusalo AM, Lindgren L. Effects of carbon dioxide pneumoperitoneum for laparoscopic cholecystectomy. Acta Anaesthesiol Scand 2000;44:834–41.
25. Gardeback M, Settergren G, Brodin LA. Hepatic blood flow and right ventricular function during cardiac surgery assessed by transesophageal echocardiography. J Cardiothorac and Vasc Anesth 1996;10:318–22.
26. Stevens JH, Raffin TA, Mihm FG, et al. Thermodilution cardiac output measurement: effects of respiratory cycle on its reproducibility. JAMA 1985;253:2240–2.
27. Perrino AC, Harris SN, Luther MA. Intraoperative determination of cardiac output using multiplane transesophageal echocardiography: a comparison to thermodilution. Anesthesiology 1998;89:350–7.
28. Feinberg MS, Hopkins WE, Davila-Roman VG, Barzilai B. Multiplane transesophageal echocardiographic Doppler imaging accurately determines cardiac output measurements in critically ill patients. Chest 1995;107:769–73.
29. Descorps-Declere A, Smail N, Vigue B, et al. Transgastric, pulsed Doppler echocardiographic determination of cardiac output. Intensive Care Med 1996;22:34–8.
30. Moriyasu F, Ban N, Nishida O, et al. Clinical application of an ultrasonic Duplex system in quantitative measurement of portal blood flow. J Clin Ultrasound 1986;14:579–88.
31. Gannedahl P, Odeberg S, Brodin LA, Sollevi A. Effects of posture and pneumoperitoneum during anaesthesia on the indices of left ventricular filling. Acta Anaesthesiol Scand 1996;40:160–6.
32. Bäcklund M, Kellokumpu I, Scheinin T, et al. Effect of temperature of insufflated CO2
during and after prolonged laparoscopic surgery. Surg Endosc 1998;12:1126–130.