| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of Anesthesiology, Vascular Surgery, and Intensive Care Unit, Ambroise Paré University Hospital, Assistance Publique Hôpitaux de Paris, Boulogne, Cedex, France
Address correspondence and reprint requests to Antoine Vieillard-Baron, MD, PhD, Intensive Care Unit, Hôpital Ambroise Paré, 9 avenue Charles de Gaulle, 92104 Boulogne, Cedex, France. Address e-mail to antoine.vieillard-baron{at}apr.ap-hop-paris.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Mean arterial blood pressure is significantly increased with CO2 pneumoperitoneum (2,11,12), leading to left ventricular (LV) afterload increase (11). In this regard, a decrease in cardiac output observed during laparoscopic surgery may be related to cardiac dysfunction induced by an increased afterload. Moreover, this mechanism could be especially important in a population with frequent cardiovascular comorbidity.
Identifying mechanisms of cardiac output decrease during surgery is important because it definitely influences perioperative care management. For instance, administration of excessive intravascular volume to a patient with cardiac output decrease as a result of systolic dysfunction could be harmful. Conversely, administration of a vasodilator to a hypovolemic patient could be deleterious. Another issue is the frequent incidence of postoperative myocardial ischemia in patients scheduled for major aortic surgery (16). In this case, occurrence of systolic dysfunction related to peritoneal CO2 insufflation could indicate that caution should be exercised in the use of pneumoperitoneum for these patients. In other words, cardiac status should be precisely defined during the preoperative period and/or occurrence of cardiac dysfunction during pneumoperitoneum will necessitate more cardiac monitoring during the postoperative period.
A totally laparoscopic infra-renal aortic surgery was recently developed in our hospital (17) and the aim of our study was to establish the influence of intraperitoneal CO2 insufflation on cardiac preload and afterload during the procedure. For this, we used transesophageal echocardiography (TEE) to provide a complete and noninvasive analysis of cardiac function and blood volume status. We hypothesized that cardiac output decrease is related more to a cardiac workload increase than to a venous return impairment. Because the surgical procedure required that patients be placed in a lateral position, we also evaluated whether position could influence right ventricular (RV) function by decreasing respiratory system compliance.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
All usual cardiovascular drugs were taken with the anesthetic premedication, except for angiotensin-converting enzyme inhibitors and angiotensin receptor blockers. Before induction of anesthesia, 10 mL/kg Ringer's lactate solution was given. General anesthesia was induced with thiopental and maintained with isoflurane in air and oxygen mixture. Depth of anesthesia was adjusted to obtain a bispectral index between 40 and 60. Heart rate and systemic arterial blood pressure from an indwelling radial artery catheter and pulse oximetry were continuously recorded. The oxygen inspired fraction was continuously adjusted to obtain a Spo2 value more than 95%. After anesthesia induction, ventilatory variables were adjusted to keep end-tidal CO2 less than 34 mm Hg and were not modified throughout the study. Pneumoperitoneum pressure was adapted between 0 and 14 mm Hg with a variable-flow insufflator of CO2.
The surgical procedure has been previously described (18). Briefly, throughout the intraabdominal procedure, the patient was placed in right lateral and rotated decubitus position. After introduction of the endoscope and various trocars, a left retrocolic dissection was conducted and the small bowel and left mesocolon moved into the right part of the abdomen, thus exposing the infrarenal aorta.
TEE was performed with a Toshiba "CoreVision" model SSA-350A (Toshiba Corporation, Otawara-Shi, Japan) equipped with a multiplane 5-MHz transducer. The probe was inserted into the esophagus after patients were anesthetized and their lungs were mechanically ventilated. TEE images were recorded on videotape and reviewed for single-frame, stop-motion analysis by a blinded and trained observer unaware of the time of measurement. Under each set of experimental conditions, the TEE study was completed after 5 min of stabilization. Evaluations were first obtained in the supine position without pneumoperitoneum (Sp 0) and with a stable intraabdominal pressure of 14 mm Hg (Sp 14). Then, patients were turned in a strict right lateral decubitus position without pneumoperitoneum (Lp 0) and intraabdominal pressure was progressively increased to 7 mm Hg (Lp 7) and to 14 mm Hg (Lp 14).
A short-axis cross-sectional view of the LV at the mid-papillary muscle level was obtained by a transgastric approach. LV end-diastolic (ED) and end-systolic (ES) diameters (D) were measured from M-mode recording, permitting calculation of LV ED and ES volumes (V) using the formula of Teicholz: V = 7D3/(2.4 + D). LV stroke volume (LVSV) and LV ejection fraction (LVEF) were calculated using LVEDV and LVESV. Cardiac output was calculated by multiplying LVSV by heart rate. In the same view, LV wall thickness at end-systole (ThES) was also measured from M-mode recording. LV systolic wall stress (LVSWS) was calculated using the formula of Reichek et al. (19): LVSWS = 0.334 SAP (LVESD)/ThES (1 + ThES/LVESD), where SAP indicates systolic arterial blood pressure and LVESD is LV end-systolic diameter.
Using an esophageal approach, we also obtained a four-chamber view of the cardiac cavities. From this view, pulsed-Doppler at the mitral annulus was performed and peak velocities of E and A waves were measured and expressed as E/A ratio (20). Velocity-time integral (VTIM) of mitral flow was also measured to estimate LV SV (21) and to corroborate LV SV variations measured from M-mode recording.
Doppler pulmonary VTI (VTIP) was recorded at the level of the RV outflow tract, together with pulmonary artery diameter (Dp), permitting calculation of RV stroke volume, as RVSV = VTIP x 
DP 2/4 (22). As previously reported (23), maximal RVSV at end-expiration and minimal RVSV at end-inspiration were collected, permitting calculation of decrease in RVSV related to tidal insufflation (
RVSV = maximal RVSV – minimal RVSV/maximal RVSV).
The superior vena cava (SVC) was examined from a long-axis view, using the two-dimensional view to direct the M-mode beam across the maximal diameter. Thus, we measured SVC diameter at end-expiration and at end-inspiration (24).
On the basis of a 15% decrease in LV SV, we calculated that 15 patients could test the null hypothesis at 0.05 significance with a power of 0.90. Statistical calculations were performed using the Statgraphics plus package (Manugistics, Rockville, MD). Data are expressed as means ± sd. Doppler echocardiographic measurements and clinical parameter changes under the five sets of experimental conditions were compared with repeated-measures analysis of variance and a Fisher's exact test for post hoc comparison. P values < 0.05 were considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
Pneumoperitoneum induced a significant increase in LVSWS (Fig. 1), associated with a marked increase in LV ESV. Significant decreases in LV EF (Fig. 1) and LV SV (Fig. 2) were also found. LV EDV did not change significantly, whereas a significant reduction in E/A ratio at the mitral annulus was observed (Fig. 3). Individual values of LV EF recorded during the preoperative period, at baseline (Sp0), and during pneumoperitoneum are reported in Figure 4. Abnormalities in the motion of the LV infero-posterior wall were observed in 2 patients with a history of myocardial infarction. New wall motion abnormalities were never noted during the study. We observed a significant increase in RVSV induced by pneumoperitoneum and right lateral position together, and RVSV at Lp 14 was significantly higher than at Sp 14. SVC diameter during respiration remained constant, with no change observed between end-inspiration and end-expiration. On the other hand, a significant increase in SVC diameter was observed related to pneumoperitoneum and body position. The intra- and inter-observer variabilities of our main echocardiographic measurements are reported in Table 3.
|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
We did not find an effect of pneumoperitoneum on venous conductance. Yet the significant increase in SVC diameter suggests a better venous conductance and a progressive increase in central venous pressure. Moreover, a significant change in SVC diameter related to tidal ventilation is effective in detecting hypovolemia (24) and in predicting cardiac output increase after blood volume expansion (27). The fact that SVC diameter was not influenced by tidal ventilation throughout the study suggests that intravascular volume status of all patients was adequate. Finally, LV preload, evaluated with LV EDV, was not modified during the study in our patients. However, LV preload decrease and caval vein zone 2 condition have been reported during peritoneal gas insufflation (3,4,9,10).
There are various explanations for the differences between our results and previous findings. In some studies, pneumoperitoneum was insufflated in a head-up tilt position (1,28), which induces a decrease in systemic venous return. In these cases, cardiac function was normalized after the patient's placement in a head-down tilt position (7). Impact of pneumoperitoneum on LV preload condition is highly dependent on preinsufflation blood volume status (6). In our study, all patients received 10 mL/kg before anesthesia induction and, thus, blood volume status was optimized before surgery.
Mean arterial blood pressure was increased by 25% with peritoneal CO2 insufflation. This was in accordance with values usually reported in healthy subjects (12). The association of arterial blood pressure and LV ESV increases strongly suggested an increased LV afterload produced by peritoneal CO2 insufflation. Indeed, we observed a significant increase in LV SWS, an index of LV afterload (29). In healthy patients, LV afterload increase did not affect LV systolic function (11–13) or produced only a transitory effect of no clinical relevance (14,15). By contrast, we observed impairment of LV systolic function, marked by a 25% decrease in LV EF. Thus, in the present study, LV systolic function was significantly affected by a significant increase in LV afterload as a result of peritoneal pressure increase with CO2 insufflation. Although all patients had a preoperative LV EF above 50%, 8 of them exhibited a decrease to less than 50% after peritoneal CO2 insufflation. Our specific population could partially explain discrepancies between previous studies in healthy patients and our results. Indeed, a majority of our patients had preexisting cardiovascular disease, i.e., chronic hypertension, and were probably extremely sensitive to LV afterload changes (1,30).
Changes observed in the LV filling pattern, exhibited by a modification of mitral flow, i.e., a decrease in E/A ratio, demonstrated LV relaxation impairment (20). A progressive increase in SVC diameter suggested that blood return might be augmented from the abdominal cavity toward the thoracic cavity by the increase in abdominal pressure. Despite this potential augmentation, we did not observe any increase in LV diastolic dimensions, suggesting a high ventricular elastance. Moreover, a normal LV is unable to dilate abruptly in response to acute systolic dysfunction (31). Thus, LV diastolic function appeared to be impaired by pneumoperitoneum and was explained by increase in LV afterload (4).
Another factor that potentially can affect LV relaxation is a sudden increase in RV outflow impedance (32). Cyclic changes in RV SV during mechanical ventilation, including a decrease during pulmonary inflation and a progressive return to preinflation value during pulmonary deflation, have been found to result from cyclic changes in RV afterload resulting from cyclic changes in transpulmonary pressure (23). These respiratory variations are accentuated in hypovolemic patients (24). In the present study, amplitude of these respiratory changes was enlarged by an increase in abdominal pressure and posture changes from supine to lateral position. Although small changes were observed at baseline, RV SV decreased by 40% at each tidal ventilation when pneumoperitoneum was inflated with the patient in the lateral position. Thus, peritoneal pressure increase and positional changes may affect inspiratory RV ejection by way of hypovolemia and/or by RV afterload increase. We noted above the absence of preload effect. End-inspiratory airway pressure was progressively increased during peritoneal insufflation and posture changes. The fact that this increase occurred despite an unchanged tidal volume suggested that the compliance of lung and chest wall was reduced. Of course, chest wall compliance is reduced when a positive abdominal pressure is exerted on the diaphragmatic part of the chest wall. Such a compliance decrease, related to pneumoperitoneum and lateral position, has previously been described (33,34). But this mechanical effect would not change transpulmonary pressure because a reduced chest wall compliance would preferentially increase pleural pressure (23). In a previous report, Joris et al. (35) found a 9 mm Hg increase in intrathoracic pressure during a pneumoperitoneum of 14 mm Hg. However, reducing the apico-basal dimension of the lung by displacing the diaphragm toward the apex would also reduce functional residual capacity and decrease lung compliance (36). In the same manner, lateral position would also reduce functional residual capacity by modifying gas distribution (37) and thus would decrease lung compliance. Both resulted in an increased transpulmonary pressure for delivering the same tidal volume, responsible for an increase in RV afterload.
For ethical reasons, and as in many other studies exploring the effect of peritoneal CO2 insufflation, measurements were done according to the surgical timing. A limitation on the interpretation of our results is that the different sets of measurements were not randomized. Also, we could not exclude that our findings are not simply the result of hemodynamic changes during surgery. However, arterial blood pressure and cardiac variables were modified in a similar fashion during pneumoperitoneum and data recorded during periods without pneumoperitoneum (Sp0 and Lp0) were also comparable. All these observations indicate that peritoneal CO2 insufflation is probably the main factor in creating hemodynamic disturbances during laparoscopic surgery.
Our results demonstrate that laparoscopic surgery in patients undergoing abdominal aortic repair induces a significant increase in workload of the LV and RV. The first clinical implication is that fluid infusion is not a good option when a decrease in stroke volume is observed during this kind of surgery and that strict control of systemic blood pressure probably plays a key role. Also, these results suggest that cardiac events, such as perioperative myocardial ischemia, could be facilitated in patients who often present various cardiac comorbidities. Our observational study in a small number of patients was only designed to verify a hypothesis concerning hemodynamic variations induced by peritoneal CO2 insufflation. We cannot draw firm conclusions about the possible occurrence of postoperative cardiac events in relation to laparoscopic surgery. A larger study is required to answer this question.
In conclusion, laparoscopic surgery in patients undergoing abdominal aortic repair may produce a marked impairment in LV systolic function by way of increased LV afterload. Cardiac preload effect was never observed. This major hemodynamic dysfunction is also associated with enhancement of RV afterloading by mechanical ventilation.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C-C. Balick-Weber, P. Nicolas, M. Hedreville-Montout, P. Blanchet, and F. Stephan Respiratory and haemodynamic effects of volume-controlled vs pressure-controlled ventilation during laparoscopy: a cross-over study with echocardiographic assessment Br. J. Anaesth., September 1, 2007; 99(3): 429 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Cattano, A. Nicolini, M. Genovesi, and S. Buzzigoli Laparoscopic Cholecystectomy in an Adult with Complex-Corrected Congenital Heart Disease Anesth. Analg., January 1, 2007; 104(1): 242 - 243. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
