Hemodynamic Changes During Laparoscopic Cholecystectomy Jean L. Joris, MD, Didier P. Noirot, MD, Marc J. Legrand, MD, Nicolas J. Jacquet, MD, and Maurice L. Lamy, MD,

Departments of Anesthesiology and Abdominal Surgery, CHU of Liege, Domaine du Sart Tilman, B-4000 Liege, Belgium

Hemodynamics during laparoscopic cholecystectomy under general anesthesia (isoflurane in N 2 0 / 0 2 (50%)) were investigated in 15 nonobese ASA Class I patients by using invasive hernodynamic monitoring including a flow-directed pulmonary artery catheter. During sur- gery, intraabdominal pressure was maintained auto- matically at 14 mm Hg by a C02 insufflator, and minute ventilation was controlled and adjusted to avoid hy- percapnia. Hemodynamics were measured before an- esthesia, after the induction of anesthesia, after tilting into 10" head-up position, 5 min, 15 min, and 30 min after peritoneal insufflation, and 30 min after exsuffla- tion. Induction of anesthesia decreased significantly mean arterial pressure and cardiac index (CI). Tilting the patient to the head-up position reduced cardiac pre-

load and caused further reduction of CI. Peritoneal in- sufflation resulted in a significant increase (235%) of mean arterial pressure, a significant reduction (220%) of CI, and a significant increase of systemic (265%) and pulmonary (290%) vascular resistances. The combined effect of anesthesia, head-up tilt, and peritoneal insuf- flation produced a 50% decrease in CI. Administration of increasing concentrations of isoflurane, via its vasodilatory activity, may have partially blunted these hemodynamic changes. These results demon- strate that laparoscopy for cholecystectomy in head-up position results in significant hemodynamic changes in healthy patients, particularly at the induction of pneumoperitoneum.

(Anesth Analg 1993;76:1067-71)

V arious pathologic gynecologic conditions have been diagnosed and treated by using laparos- copy for more than two decades. Endoscopic

surgery has been extended to gastrointestinal surgical procedures such as appendectomy, peritoneal adhesi- olysis, and cholecystectomy (1). The many benefits reported after laparoscopy (1-3) explain its increasing success and the efforts to introduce the endoscopic approach for other surgical procedures. However, peritoneal insufflation of C02 to create the pneumo- peritoneum necessary for laparoscopy induces intraoperative ventilatory (4-8) and hemodynamic (4-6,9,10) changes that complicate anesthetic manage- ment of laparoscopy. The position of the patient (head- down or head-up) required during these procedures also contributes to these changes (4,6-7).

The investigation of hemodynamics during gyneco- logic laparoscopy in the head-down position (6-7,9) revealed an increase (225%) of mean arterial blood pressure (MAP), a decrease (%15%) in cardiac output, and an increase (550%) of systemic vascular resistance

Accepted for publication December 11, 1992. Address correspondence and request for reprints to Dr. Jean Joris,

Department of Anesthesiology, CHU of Liege, Domaine du Sart Tilman, B-4000 Liege, Belgium.

during insufflation. Obese patients with potentially impaired cardiovascular status explored in one study (6) and the accuracy of the non-invasive technique (thoracic electrical bioimpedance) used in the other studies (9,101 raise questions (11,12). We investigated the hypothesis that peritoneal insufflation of car- bon dioxide for laparoscopic cholecystectomy in the head-up position would be associated with even greater decreases in cardiac output than insufflation for gynecologic laparoscopy in the head-down posi- tion. Hemodynamic changes during laparoscopic cholecystectomy were explored by using invasive monitoring (e.g., flow-directed pulmonary artery cath- eter) in non-obese patients without cardiorespiratory pathology.

Met hods Fifteen fully informed patients scheduled for elective laparoscopic cholecystectomy gave consent to be in- cluded in this study after the approval of the ethics committee at our institution. Inclusion criteria were body weight less than 20% more than ideal weight, age between 18 and 70 yr, no acute cholecystitis, and no cardiorespiratory disease or medications. All patients were premedicated with hydroxyzine 50-75 mg by

01993 by the International Anesthesia Research Society 0003-2999/93/$5.00 Anesth Analg 1993;76:1067-71 1067

1068 JORIS ET AL. HEMODYNAMICS DURING LAPAROSCOPY

ANESTH ANALG 1993;76:1067-71

mouth 2 h before surgery and an intramuscular (IM) injection of midazolam, 5 mg, and atropine, 0.25 mg, just before transfer to the operating theater. General anesthesia was induced intravenously (IV) with sufen- tanil, 15-20 pg, thiopental, 5 mg-kg-I, and atracurium, 0.5 mg-kg-I. After tracheal intubation, general anes- thesia was maintained with nitrous oxide (50%) in ox- ygen and isoflurane. Isoflurane concentrations were adapted to keep the patient hemodynamically stable by an experienced anesthesiologist: mean arterial pressure (MAP) was not allowed to increase more than 20% above preinduction value. In addition to clinical signs, the anesthesiologist used the usual monitoring for this surgical procedure (i.e., noninvasive arterial blood pressure, electrocardiogram (ECG), heart rate, capnog- raphy, and pulse oximetry) but was unaware of the information derived from the invasive monitoring de- scribed below. A basal IV infusion (4 mL.kg-'.h-') of lactated Ringer's solution to compensate for fasting and intraoperative losses was given. Before the induc- tion of anesthesia, one radial artery was cannulated and a 7.5 F thermodilution Edwards Swan-Ganz@ cath- eter (Baxter) was introduced via the right internal jug- ular vein. During surgery, the expired end-tidal Pco2 (PETCO~) and isoflurane concentrations were monitored continuously by a Datex Capnomac@ sidestream ana- lyzer. Minute ventilation was controlled (Servo 900C; Siemens-Elema) and adjusted to keep the PETCO~ be- tween 30 and 40 mm Hg. During laparoscopy intra- abdominal pressure (IAP) was maintained automati- cally at 14 mm Hg by a C02 insufflator (Wolf 2154.201). The following variables were recorded on a Datex Car- diocap@" monitor: invasive arterial pressure, pulmo- nary artery pressure, right atrial pressure (RAP), pul- monary capillary wedge pressure (PCWP) measured with a Viggo-Spectramed transducer, and heart rate. Cardiac output (CO) was calculated by a SAT-1 com- puter (American Edwards Laboratories), and the av- erage of three measurements made at the end of expi- ration was considered. Systemic vascular resistance (SVR) and pulmonary vascular resistance (PVR) were calculated from the above variables as follows: SVR (dynes.~.cm-~) = (mean invasive arterial pressure - RAP).8O/CO; PVR (dyne~.s.cm-~) = (mean pulmo- nary artery pressure - PCWP).8O/CO. Data were col- lected before the induction of anesthesia (Tl), 10 min after the induction of anesthesia (T2), 10 min after tilt- ing into 10" head-up position (T3), 5 min (T4), 15 min (T5) and 30 min (T6) after the beginning of insufflation, and finally 30 min after exsufflation (T7). At T7, anes- thesia had been terminated for 215 min and the pa- tients, once again in the horizontal position, had been tracheally extubated and were breathing spontaneous- ly. The pressure transducer was located at the level of the right atrium and was moved after tilting to remain at this level. At the same time points, peak and plateau

Table 1. Patient Data

48.9 k 3.7 (20-69) Age, yr Sex ratio, M/F 4/11

Height, cm Duration of insufflation, rnin 62.9 ? 4.6 (35-105)

Weight, kg 66.8 f 3.0 (55-85) 165 ? 1 (152-175)

Results are mean ? \EM and (range)

airway pressures (Paw) were recorded on the Servo ventilator 900C (Siemens-Elema) and an arterial blood sample was drawn to measure arterial Pco2 (Paco2). Mean intrathoracic pressure was also measured in 12 patients by means of a latex esophageal balloon. The correct position of the balloon was confirmed by flu- oroscopy. Inasmuch as the pressure of the mediastinum on the balloon changes with patient position, pressures were recorded only after tilting to the head-up position (T3 to T6) and expressed as pressure change compared to T3 value.

Results are reported as mean t standard error (SEM). Data were analyzed by a one-way analysis of variance for repeated measures followed by Newman-Keuls test (13). Results were considered to be statistically signif- icant at the 5% critical level.

Results Demographic characteristics of the patients are repre- sented in Table 1. After the induction of anesthesia, MAP and cardiac index (CI) decreased (Table 2). After tilting to the head-up position, RAP and PCWP de- creased and a further reduction of CI and MAP was observed. From T1 to T3, a linear relationship linked MAP and CI (MAP = 16.9 CI + 36.7; Y = 0.94, P < 0.05). After peritoneal insufflation (T4), this relationship changed. Indeed, whereas MAP increased (35% ), CI significantly decreased (20%). Peritoneal insufflation also resulted in a significant increase of systemic and pulmonary vascular resistances (SVR, 65% and PVR, 90%) (Table 2). RAP and PCWP increased significantly during insufflation and did not change significantly from T4 to T6. Deepening of anesthesia by the admin- istration of increasing concentrations of isoflurane (Table 3) allowed partial correction of the changes of CI, SVR, and PVR which, however, remained significantly different from preoperative values during CO, insuf- flation (Table 2). No significant changes of heart rate and Paco2 were noted during surgery (Tables 2 and 3). The combined effects of anesthesia, the head-up tilt, and increased intraabdominal pressure produced a sig- nificant decrease (50%) of CI (Table 2). After insuffla- tion, peak and plateau Paw as well as intrathoracic pressure significantly increased (Table 3). Thirty min- utes after exsufflation, all hemodynamic variables had returned to preoperative values.

ANESTH ANALG 1993;76:3067-71

JORIS ET AL. 1069 HEMODYNAMICS DURING LAPAROSCOPY

Table 2. Hemodynamic Changes During Laparoscopic Cholecystectomy

Hemod ynamic changes*

MAP, mm Hg HR, beats/ min- ' RAP, mm Hg PCWP, mm Hg CI, L.rnin-'.rn-, SVR, dynes.s.cm-' PVR, dynes.s.cm-'

T1 before

induction

96 f 13 79 f 4 8 f 1 9 f 1

3.6 * 0.1 1139 f 49 229 f 21

T2 after

induction

T3 head-up

T4 T5 T6

Pneumoperitoneum

5 min 15 min 30 min

T7 after

surgery

87 f 3 83 + 4 8 + 1 9 f 1

2.7 f 0.2t 1389 + 108 292 f 31

71 f 2 t 76 t 4 5 + 1 7 f 1

2.2 * 0.2t 1452 k 111 272 f 28

97 f 5$ 102 + 5$ 93 f 5$

10 f 2$ 11 f 1$ 10 f 1$ 79 f 4 81 + 5 81 f5

14 f 2t,$ 14 f It,$ 12 f It,$ 1.8 f O. l t , $ 2.4 t O . l t 2.3 f 0.2t

2367 f 251t,$ 1777 f 146t 1676 f 120t 521 + 55t,$ 423 f 36tJ 397 f 35t,$

105 f 5$ 78 f 4 8 + 1 9 + 1

3.4 f 0.3 1365 f 96 291 f 3 2

Hemodynamic changes were measured or calculated before the induction of anesthesia (Tl), 10 min after the induction of anesthesia (T2), 10 min after tilting

Results are mean ? SEM. * Abbreviations used are: MAP, mean arterial pressure; HR, heart rate; RAP, right atrial pressure; PCWP, pulmonary capillary wedge pressure; CI, cardiac index;

SVR, systemic vascular resistance; PVR, pulmonary vascular resistance. t P < 0.05 as compared with TI; $ P < 0.05 as compared with T3.

into 10" head-up position (T3), 5 min (T4), 15 min (TS), and 30 min (T6) after the beginning of insufflation, and 30 min after exsufflation (T7).

Table 3. Ventilatory Changes During Laparoscopic Cholecystectomy

T4 T5 T6 T1 T2 T3 Pneumoperitoneum T7

Ventilatory before after head-up after changes* induction induction 5 min 15 min 30 min surgery

Paco', mm Hg 41.2 f 0.9 36.3 f 1.4t 34.1 f 0.7t 35.8 t 0.9t 39.1 f 1.0 37.5 f 0.8 43.2 * 1.2$ Pao,, mm Hg 84 f 3 243 + 12t 216 f 13t 234 f 9 t 230 f llt 222 f 10t 82.2k 3.1 Isoflurane, % 0.46 f 0.03 0.38 + 0.04 0.84 f 0.05$ 1.02 f 0.084 0.90 f 0.8$ AITP, mm Hg 0 9.4 f 1.1$ 9.6 f 1.0$ 8.7 +- 0.8$ Peak Paw, cm H 2 0 14.5 f 1.1 14.9 f 1.3 21.8 f 1.9$ 22.6 t 1.2$ 22.7 f 1.3$ Plateau Paw, cm H 2 0 10.6 * 1.2 1.2 t 1.3 17.9 + 1.8$ 18.4 f 1.5$ 18.4 + 1.2$

'Arterial Pcoz (Pam2) and Poz (Path), expired end-tidal concentration of isoflurane, and peak and plateau airway pressure (Paw) were measured 10 min after the induction of anesthesia (T2), 10 min after tilting into 10' head-up position (T3), 5 min (T4), 15 min (T5). and 30 min (T6) after the beginning of insufflation. Pacoz and Pa02 also were measured before the induction of anesthesia (TI) and 30 min after exsufflation (T7). Change in intrathoracic pressure (AITP) as compared to T3 were calculated during pneumoperitoneum.

Results are mean ? SEM. t P < 0.05 as compared with T1; $ P < 0.05 as compared with T3.

Discussion This study demonstrates that peritoneal C02 insuffla- tion to an intraabdominal pressure of 14 mm Hg, nec- essary for laparoscopic cholecystectomy, induces major hemodynamic changes in healthy non-obese patients without cardiac disease. These significant disturbances are characterized by an increase in MAP, SVR, and PVR, and a decrease of CI. CI markedly decreased to as much as 50% of preoperative values 5 min after the beginning of insufflation. The hemodynamic changes observed in this study complement the results of recent studies dur- ing gynecologic laparoscopy in the head-down posi- tion (6,9,10). By using an invasive monitoring in pa- tients in the head-up position, we observed a more pronounced increase of SVR and PVR.

The pathophysiology of these changes remains un- clear. Before insufflation, SVR did not change signifi- cantly and a linear relationship linked MAP and CI. Anesthetic induction drugs depressed the myocardium and reduced CI and MAP. Tilting the patient to the

head-up position reduced venous return as evidenced by the fall of RAP and PCWP. Consequently, CI and MAP further decreased. After peritoneal insufflation, this relationship changed. Indeed, MAP at T4, although similar to that at T1, was associated with significantly lower CI. Only an increase of SVR can explain the in- crease in MAP pressure observed after insufflation de- spite the significant reduction of CI. Increased venous resistance (14) and compression of the abdominal aorta may coqtribute to the increase in cardiac afterload (5,9,10). Measurement of the radial-femoral arterial pressure gradient before and during peritoneal insuf- flation could help to confirm this hypothesis. It is un- likely that the increases in MAP and SVR are entirely related to mechanical factors. First, the correction of these changes was gradual and took several minutes when the IAP was released suddenly at the end of the operation (6,10,15). Moreover, since in our study the IAP was maintained automatically at 14 mm Hg during laparoscopy, the compression of the abdominal aorta

1070 JORIS ET AL. HEMODYNAMICS DURING LAPAROSCOPY

ANESTH ANALG 1993;761067-71

did not change and the disturbance of venous return remained constant, as evidenced by unchanged RAP and PCWP from T4 to T6. Despite this steady state, SVR decreased nevertheless. The partial correction of the hemodynamic variables occurring between T4 and T6 may be attributed to the vasodilating properties of iso- flurane (16) and/or to the natural history of the he- modynamic changes. These changes, however, per- sisted for more than 20 min in other studies (5,17,18). All these observations suggest the involvement of other mechanism(s) such as the release of humoral factors. Potential mediators of the increase in SVR are cate- cholamines, prostaglandins, the renin-angiotensin sys- tem, and vasopressin (10,19,20). Plasma noradrenaline levels and plasma renin activity did not change during pneumoperitoneum for diagnostic laparoscopy in pa- tients breathing spontaneously (20). On the other hand, a fivefold increase of vasopressin was observed in 60% of patients when IAP was increased to +10 mm Hg (20). Increases in plasma vasopressin levels were correlated with changes in IAP, intrathoracic pressure (ITP), and transmural RAP (20). The increasing pulmonary resis- tance described in this study is another potential stim- ulator of vasopressin release (20). Therefore, both me- chanical and putative humoral phenomena could contribute to the increase in SVR.

The mechanism of the decrease of CI may be mul- tifactorial. CO depends on venous return, myocardial contractility, and cardiac afterload (14). In animal stud- ies, inferior vena caval flow declined with increases of IAP of 10 mm Hg or more (5,17,18). Femoral vein pres- sures increased roughly in parallel with the increases of IAP consequent to pooling of blood in the peripheral circulation (4,5). The subsequent decline in venous re- turn closely paralleled the decrease in CO (5,171. Par- adoxically, we and others (4-6) observed increased car- diac filling pressures (RAP, PCWP) during peritoneal insufflation. However, consequent to the increase in IAP, ITP also increased (5,8,20) (Table 3). The increase of minute ventilation necessary to avoid hypercapnia (6-8) associated with a decrease of thoracopulmonary compliance secondary to the elevation of the dia- phragm (9) results in an increase of peak and plateau Paw during mechanical ventilation. This increase in Paw (Table 3) probably contributes to the increased ITP. Transmural RAP (i.e., the pressure within the right atria minus the extracardiac pressure), which should be used rather than directly measured RAP as an indicator of venous return to the heart in case of elevated IAP, did not change very much (5,18). The observation of a sig- nificant decrease in stroke volume, despite only a slight or no decrease of transmural RAP, suggests a shift of the ventricular function curve (Frank-Starling’s law) to the right, perhaps secondary to the increase in afterload (5,18). The increase in SVR is most likely the cause of the decreased CO (5,10,18) rather than from an

increased sympathetic activity in response to the de- creased CO (9). Indeed, vasodilation by the adminis- tration of increasing concentrations of isoflurane im- proved CO although RAP did not change. Finally, the concentrations of isoflurane used in this study do not depress the normal myocardium (16). The effect of in- creased IAP and ITP on myocardial contractility can not, however, be determined from this study. Trans- esophageal echocardiography should be helpful in as- sessing contractility and venous return. Thus, during laparoscopy, the decrease in CO can be explained by a reduction in venous return and/or an increase of SVR. It seems that the normal heart, which tolerates an in- crease of afterload very easily in normal conditions, becomes sensitive to changes in afterload much like a decompensated heart (21), when this normal heart has to face the artificial conditions of pneumoperitoneum under general anesthesia.

These results indicate the need for caution in patients with impaired cardiac function, anemia, or hypovole- mia scheduled for laparoscopy. Such situations may oc- cur during laparoscopy in patients with ruptured ec- topic pregnancies, or diagnostic laparoscopy in pa- tients with blunt and penetrating injuries to the abdo- men and in patients with peritonitis. These results bear a particular significance in cases of laparoscopy in older patients, such as for gastrointestinal surgical proce- dures; these patients are more likely to have known or latent cardiac disease. Hemodynamic consequences of pneumoperitoneum have not been explored in patients with cardiac disease. However, indirect data suggest that hemodynamics may be more altered in these pa- tients than in ASA Class I patients (22). In all these cases, it would seem prudent to reduce the rate of in- sufflation and limit abdominal inflating pressures to a minimum. The trend to readily propose the laparo- scopic approach for patients with impaired cardiac function because of easier and smoother postoperative recovery should be tempered by the risks related to the intraoperative hemod ynamic changes induced by peri- toneal insufflation. The anesthetic management of lap- aroscopy is beyond the scope of this paper. However, anesthetics with vasodilating action should be favored, and anesthetics that directly depress the heart should be avoided. Pure vasodilator and cardiotonic drugs may be necessary in patients with compromised car- diac function. Finally, it should be noted that the usual intraoperative cardiovascular monitoring (blood pres- sure, heart rate, capnography, pulse oximetry) gives no accurate information on the increase of SVR and the reduction of CO.

In conclusion, these results highlight the fact that laparoscopy induces significant hemod ynamic changes even in healthy patients and creates increases of SVR and PVR, an increase of MAP, and a reduction of CO. These disturbances could be mediated both

ANESTH ANALG 1993;76:1067-71

JORIS ET AL. 1071 HEMODYNAMICS DURING LAPAROSCOPY

mechanically and humorally. Whereas these cardiovas- cular changes should not be hazardous in healthy pa- tients, special care and monitoring are mandatory for patients with impaired cardiac function. In these pa- tients postoperative benefits of laparoscopy should be balanced against intraoperative risks.

References 1 . Dubois F, Icard P, Bertholet G, Levard H. Coelioscopic cholecys-

tectomy. Ann Surg 1990;160:485-7. 2. Grace PA, Quereshi A, Colenian J, et al. Reduced postoperative

hospitalization after laparoscopic cholecystectomy. Br J Surg

3. Joris J, Cigarini I, Legrand M, et al. Metabolic and respiratory changes following cholecystectomy performed via laparotomy or laparoscopy. Br J Anaesth 1992;69:341-5.

4. Kelman GR, Swapp GH, Smith I, et al. Cardiac output and ar- terial blood-gas tension during laparoscopy. Br J Anaesth 1972; 44:115542.

5. Ivankovich AD, Miletich DJ, Albrecht RF, et al. Cardiovascular effects of intraperitoneal insufflation with carbon dioxide and nitrous oxide in the dog. Anesthesiology 1975;42:281-7.

6. Schoeffler P, Haberer JP, Manhes H, et al. Repercussions circu- latoires et ventilatoires de la coelioscopie chez l'obese. Ann Fr Anesth Reanim 1984;3:10-5.

7. Ciofolo MJ, Clergue F, Seebacher J, et al. Ventilatory effects of laparoscopy under epidural anesthesia. Anesth Analg 1990;70: 357-61.

8. Joris J, Ledoux D, Honore P, Lamy M. Ventilatory effects of C02 insufflation during laparoscopic cholecystectomy. Anesthesiology 1991;75(Suppl):AI 21.

9. Johannsen G, Andersen M, Juhl B. The effect of general anaes- thesia on the haemodynamic events during laparoscopy with C02-insufflation. Acta Anaesthesiol Scand 1989;33:132-6.

10. Torrielli R, Cesarini M, Winnock S, et al. Modifications hemody- namiques durant la coelioscopie: etude menee par bioimpedance electrique thoracique. Can J Anaesth 1990;374&51.

1991 ;78:160-2.

11. Preiser JC, Daper A, Parquier JN, et al. Transthoracic electrical bioimpedance versus thermodilution technique for cardiac out- put measurement during mechanical ventilation. Intensive Care Med 1989;15:221-3.

12. Jewkes C, Sear JW, Verhoeff F, et al. Non-invasive measurement of cardiac output by thoracic electrical bioimpedance: a study of the reproducibility and comparison with thermodilution. Br J Anaesth 1991;67788-94.

13. Winer 8. Statistical principles in experimental design. New York McGraw-Hill, 1976.

14. Guyton AC. Textbook of medical physiology. 6th ed. Phila- delphia: WB Saunders Co, 1981.

15. Marshall RL, Jebson PJR, Davie IT, Scott DB. Circulatory effects of peritoneal insufflation with nitrous oxide. Br J Anaesth 1972; 44: 11 83-7.

16. Lowenstein E, Reiz S. Effects of inhalation anesthetics on sys- temic hemodynamics and the coronary circulation. In: Kaplan JA, ed. Cardiac anesthesia. 2nd ed. Philadelphia: WB Saunders Co, 19873-35.

17. Richardson JD, Trinkle JK. Hemodynamic and respiratory alter- ations with increased intra-abdominal pressure. J Surg Res 1976; 20401-4.

18. Diamant M, Benumof JL, Saidman LJ. Hemodynamics of in- creased intra-abdominal pressure. Anesthesiology 1978;48:23-7.

19. Punnonen R, Viinamaki 0. Vasopressin release during laparoscopy: role of increased intraabdominal pressure. Lancet 1982;1:175-6.

20. Solis-Hermzo ]A, Moreno D, Gonzalez A, et al. Effect of intratho- racic pressure on plasma arginine vasopressin levels. Gastroenterology 1991;101:607-17.

21. Braunwald E. Pathophysiology of heart failure. In: Braumwald E, ed. Heart disease: a textbook of cardio-vascular medicine. 4th ed. Philadelphia: WB Saunders Co, 1992393-418.

22. Wittgen CW, Andrus CH, Fitzgerald SD, et al. Analysis of the hemodynamic and ventilatory effects of laparoscopic cholecys- tectomy. Arch Surg 1991;126:997-1001.