Assessment of cardiovascular changes during laparoscopic hernia repair using oesophageal Doppler

Transcription

1 British Journal of Anaesthesia 1997; 78: Assessment of cardiovascular changes during laparoscopic hernia repair using oesophageal Doppler E. J. HAXBY, M. R. GRAY, C. RODRIGUEZ, D. NOTT, M. SPRINGALL AND M. MYTHEN Summary We have used an oesophageal Doppler to measure aortic blood flow velocity before, during and after induction of carbon dioxide pneumoperitoneum in 10 consecutive patients, mean age 58 yr, undergoing laparoscopic hernia repair. Derived values for stroke distance, minute distance and systemic vascular resistance showed considerable interpatient variation indicating unpredictable haemodynamic responses. Five minutes after insufflation of the abdomen there was a significant increase in mean arterial pressure from 8.5 to mm Hg (P 0.05) but both stroke distance and minute distance decreased significantly (mean 1.0 (SEM 1.4) cm to 9.0 (0.7) cm, P 0.05; and (8) cm min 1 to 596 (49) cm min 1, P 0.05; respectively) indicating a significant decrease in cardiac output. There was a corresponding increase in the index of systemic vascular resistance from 109 (747) to 079 (400) (P 0.05) which persisted after deflation of the abdomen. Oesophageal Doppler can provide continuous online haemodynamic data with a rapid response to acute changes and may have a role in non-invasive haemodynamic monitoring during laparoscopic procedures in older patients with cardiovascular disease. (Br. J. Anaesth. 1997; 78: ). Key words Surgery, laparoscopy. Cardiovascular system, effects. Measurement techniques, oesophageal Doppler. Minimally invasive laparoscopic surgery can reduce analgesic requirements and permit earlier postoperative mobilization than equivalent open procedures. 1 It has been advocated for high-risk patients with cardiac and pulmonary disease. But carbon dioxide pneumoperitoneum and the Trendelenburg position produce major haemodynamic and respiratory changes. 3 4 Studies during laparoscopic cholecystectomy have shown large variations in the magnitude of changes in cardiac output, systemic vascular resistance and mean arterial pressure. 5 6 Laparoscopic hernia repair is a recent addition to minimally invasive surgery. Many patients undergoing this type of surgery are older men with a high incidence of cardiac and respiratory impairment. Although studies in healthy gynaecological patients have shown only minor haemodynamic changes during laparoscopic procedures, this may not apply to laparoscopic hernia repair in older patients for whom little data are available. It has been suggested that invasive haemodynamic monitoring be used in patients with pre-existing cardiovascular disease undergoing laparoscopic surgery. 7 The aim of this study was to measure haemodynamic changes during laparoscopic hernia repair using the non-invasive technique of oesophageal Doppler. Aortic blood flow velocity and the derived haemodynamic indices of stroke distance, minute distance and systemic vascular resistance were recorded before, during and after induction of pneumoperitoneum. Patients and methods After obtaining Ethics Committee approval and informed consent, we studied 10 consecutive patients, ASA I III, undergoing elective laparoscopic inguinal hernia repair under general anaesthesia. Temazepam 0 mg orally was given as premedication h before operation. On arrival in the anaesthetic room heart rate and non-invasive arterial pressure (Dinamap) were recorded. Anaesthesia was induced using alfentanil 5 g kg 1, thiopentone sufficient to obtund the eyelash reflex ( 4 mg kg 1 ) and atracurium 0.5 mg kg 1, and the patient s trachea was intubated. Infusions of alfentanil 5 g kg 1 h 1 and atracurium 0.5 mg kg 1 h 1 were commenced. The lungs were ventilated with 70% nitrous oxide in oxygen. Ventilation was initially set at a tidal volume of 10 1 ml kg 1 with ventilatory frequency adjusted to maintain an end-tidal carbon dioxide partial pressure of kpa. Anaesthesia was maintained with isoflurane. The end-tidal concentration (Datex Capnomac Ultima, Datex, Helsinki, Sweden) was maintained constant at MAC in each patient until the last set of study variables had been recorded. All patients received 1 litre of Hartmann s E. J. HAXBY*, FRCA, C. RODRIGUEZ, FRCA, M. MYTHEN, MD, FRCA (Department of Anaesthesia); M. R. GRAY, MS, FRCS, D. NOTT, FRCS, M. SPRINGALL, CHM, FRCS (Department of Surgery); Charing Cross Hospital, Fulham Palace Road, London W6 8RF. Accepted for publication: January 10, *Address for correspondence: Department of Anaesthetics, St George s Hospital, Blackshaw Road, Tooting, London SW17 ORE.

2 516 British Journal of Anaesthesia solution in the perioperative period. At the end of the procedure neuromuscular block was antagonized with neostigmine and glycopyrronium. Postoperative analgesia was left to the discretion of the individual anaesthetist. The same surgeon performed all operations in a standardized manner. Pneumoperitoneum was established by introduction of a Verres needle via a 1.-cm infra-umbilical incision. Two further ports were introduced laterally to facilitate surgery. The peritoneum was divided over the inguinal defect which was repaired with a mesh secured by staples. At the end of the procedure the pneumoperitoneum was released before removal of the ports and closure of the wounds. Individual patient and procedure details are shown in table 1. Patient numbers referred to in the text correspond to the numbers given in table 1. Eight patients underwent unilateral hernia repair and two patients (Nos and 8) underwent bilateral hernia repairs. Mean intra-abdominal pressure was maintained at less than 13 mm Hg at all times. CARDIOVASCULAR RECORDINGS After tracheal intubation, an oesophageal Doppler probe (ODM, Deltex, Chichester, UK) was inserted through the patient s mouth and positioned approximately cm from the teeth where well defined aortic blood flow signals could be detected. The probe was manipulated until the best signal was obtained. Five (M5) and 10 (M10) min after induction of anaesthesia, but before the first surgical stimulus, a set of baseline physiological variables were recorded: heart rate, arterial pressure, S p O, peak airway pressure, F I O, ventilatory frequency, Doppler stroke distance and minute distance. These measurements were then repeated at the times shown in table. On the oesophageal Doppler display, stroke distance is represented by the area under the flow velocity waveform and this represents the length a column of blood flows along the aorta with each ventricular systole. This is a linear index of left ventricular stroke volume. Minute distance is the product of stroke distance and heart rate and is a linear index of cardiac output. An index of systemic vascular resistance (SVRI) can be obtained by Table Times for data collection during laparoscopic hernia repair Times Description M5 5 min after induction (baseline) M10 10 min after induction I+1 1 min after abdominal insufflation I+3 3 min after abdominal insufflation I+5 5 min after abdominal insufflation T Immediately after Trendelenburg positioning T+5 5 min after Trendelenburg positioning T min after Trendelenburg positioning EDSX End of surgery PDEF After abdominal deflation dividing mean arterial pressure by minute distance. 8 Data were analysed using the Statview SE Graphics (Concepts Inc. Berkeley, CA, USA) statistical system. Statistical analysis was performed with a two-way analysis of variance for repeated measures and P 0.05 was considered significant. Results are expressed as mean (SEM). Results Five minutes after insufflation of the abdomen (I 5) there was a statistically significant increase in mean arterial pressure (MAP) from 8.5 to mm Hg (P 0.05) (fig. 1). Mean stroke distance (fig. ) and minute distance decreased significantly at this time from 1.0 (1.4) to 9.0 (0.7) cm (P 0.05) and from (8) to 596 (49) cm min 1 (P 0.05), respectively, indicating a significant decrease in cardiac output. There was a corresponding increase in the index of systemic vascular resistance from 109 (747) to 079 (400) (P 0.05) (fig. 3). Percentage changes from baseline (M5) in measured stroke distance and estimated SVRI are also shown for individual patients (figs 4, 5) and illustrate inter-patient variability. Pre-existing cardiovascular disease was present in patient Nos 1, 3, 8 and 9 (table 1). Peak airway pressure increased significantly from 19.4 (1.) to 5.7 (.0) cm H O (P 0.05) as did end-tidal carbon dioxide partial pressure from 4.3 (0.13) to 5.0 (0.1) kpa (P 0.05) (table 3). Mean arterial pressure remained significantly increased (P 0.05) throughout the procedure (fig. 1) as did end-tidal carbon dioxide partial pressure (P 0.05) Table 1 Patient characteristics and duration of surgery Patient No. Sex Age (yr) Weight (Kg) Previous history Medication ASA grade Duration of anaesthesia (min) 1 F 8 54 Ischaemic heart disease, myocardial infarction Atenolol III 55 M Prostatic carcinoma Goserelin II M Myocardial infarction, coronary artery bypass grafting Simvastatin III 70 4 M I 70 5 M I 65 6 M I 60 7 M I 80 8* M Hypertension Frusemide, II 100 amlodipine 9 M Atrial fibrillation Digoxin II M Asthma Salbutamol II 50 Mean

3 Oesophageal Doppler in laparoscopic hernia repair 517 Figure 1 Mean arterial pressure (MAP) (mean, SEM; n 10) during laparoscopic hernia repair. Times are those shown in table. *P Figure 4 Percentage change in oesophageal Doppler stroke distance ( %SD) from baseline (time M5) for 10 patients during laparoscopic hernia repair. Times are those shown in table. Patient numbers correspond to data shown in table. Patients with ischaemic heart disease (IHD), hypertension (HT) or atrial fibrillation (AF) are indicated. Figure Oesophageal Doppler stroke distance (mean, SEM; n 10) during laparoscopic hernia repair. Times are those shown in table. *P Figure 5 Percentage change in systemic vascular resistance index ( %SVRI) from baseline for 10 patients during laparoscopic hernia repair. Times are those shown in table. Patient numbers correspond to data shown in table. Patients with ischaemic heart disease (IHD), hypertension (HT) or atrial fibrillation (AF) are indicated. Figure 3 Index of systemic vascular resistance (SVRI) (mean, SEM; n 10) during laparoscopic hernia repair. Times are those shown in table. *P (table 3). During Trendelenburg positioning both minute distance and stroke distance started to increase towards pre-inflation values but SVRI remained significantly increased (P 0.05) compared with baseline values, even after deflation of the abdomen. There were no significant changes in heart rate and oxygen saturation throughout the procedure. Patient No. 8 had a profound decrease in MAP on induction of anaesthesia (from 117 to 60 mm Hg) which was treated with ephedrine 6 mg. MAP increased progressively after this but did not reach pre-induction levels. Patient Nos 1 and 8 had episodes of multifocal ventricular ectopics which resolved without treatment and without new ECG

4 518 British Journal of Anaesthesia Table 3 Heart rate (HR), minute distance (MD), oxygen saturation ( S p O ), end tidal carbon dioxide partial pressure ( P E CO ) and peak airway pressure (PAP) during laparoscopic hernia repair (mean (SEM)). Times are those shown in table. *P 0.05 Time Preop. M5 M10 I+1 I+3 I+5 T T+5 T+30 EDSX PDEF HR (beat min 1 ) (.9) (4.0) (3.8) (4.9) (5.5) (6.0) (5.1) (4.7) (5.1) (4.1) MD (cm) * 601.1* 595.6* 66.7* (8.1) (8.1) (50.1) (47.3) (49.) (57.) (69.9) (89.9) (7.0) (85.8) S p O (%) P E CO (kpa) * 5.4* 5.8* 5.9* 6.* (0.09) (0.13) (0.09) (0.13) (0.09) (0.09) (0.13) (0.5) (0.3) (0.3) PAP (cm H O) * 5.1* 5.7* 7.1* 7.4* 7.4* 7.3* 0.4 (1.) (1.4) (.3) (.3) (.0) (.1) (.) (.) (.0) (1.5) changes after operation. Patient No. 7 had an episode of hypotension h after operation associated with bradycardia which was treated successfully with fluids and glycopyrronium and he also made an uneventful recovery. Discussion We have used oesophageal Doppler for non-invasive haemodynamic monitoring during laparoscopic hernia repair. Despite maintaining a relatively low intra-abdominal pressure ( 13 mm Hg) throughout the procedure, significant haemodynamic changes were seen. Group trends were similar to those described during pneumoperitoneum by other investigators: a relatively stable heart rate in association with an increase in MAP, decrease in indices of cardiac output and increase in the index of systemic vascular resistance during insufflation of the abdomen with carbon dioxide. 9 Inter-patient variation in cardiovascular response to insufflation of the abdomen with carbon dioxide and assumption of the Trendelenburg position was marked. This degree of variation in our small group of 10 patients suggests the need for close monitoring during laparoscopic procedures in this patient population. Ten consecutive patients undergoing intraperitoneal laparoscopic inguinal hernia repair were selected regardless of age and medical status. Four of the group had coexistent cardiovascular disease and the changes seen in these patients were more marked than in those patients without cardiovascular disease, particularly the increase in SVRI. This subgroup with cardiovascular disease also showed greater haemodynamic variability in response to induction of anaesthesia and insufflation of the abdomen. It is not clear if this was related to underlying cardiovascular insufficiency or to cardiac medication. The anaesthetic technique was selected to provide cardiovascular stability but systemic arterial pressure decreased in all patients after induction of anaesthesia and it may have been preferable to wait longer before recording baseline variables. We made no attempt to control the fluids administered, although no patient received more than 15 ml kg 1 of Hartmann s solution. The oesophageal Doppler has been used extensively to monitor cardiac output and other variables in healthy subjects and critically ill patients. 8 Good agreement has been shown with thermodilution techniques. 10 In the oesophageal position, excellent signals are obtained from blood travelling through the adjacent descending aorta thereby providing continuous beat-by-beat monitoring with little artefact. The probe is easy to position and signals can be interpreted without specialist training. Waveform changes frequently preceded changes in arterial pressure or heart rate and appeared to provide earlier warning of haemodynamic deterioration. Disadvantages of oesophageal Doppler include a small risk of injury to the pharynx and oesophagus and a requirement for anaesthesia so that prolonged postoperative monitoring may be difficult. Although probe position mostly remains stable, repositioning is occasionally required. Haemodynamic measures using this technique depend on several assumptions. It is assumed that the angle of incidence of the ultrasound beam to the direction of blood flow is the same as that of the probe face to the long axis of the instrument and that blood flow in the descending aorta remains in fixed proportion to total left ventricular output over wide ranges of flow, pressure and temperature. 7 Under resting conditions the validity of these assumptions has been supported by experimental data, but during insufflation of the abdomen, the distribution of aortic blood flow may vary and a greater percentage of cardiac output may be diverted to the head and upper body as regional afterload is altered. Pulmonary artery catheterization has been used for haemodynamic monitoring during laparoscopic procedures. 11 This technique permits more precise calculation of left ventricular preload, afterload and oxygen delivery, which may be useful in high-risk cases. The pulmonary artery catheter may also be inserted before induction of anaesthesia and retained after operation for prolonged monitoring. These advantages must be weighed against a low but appreciable risk of morbidity which may not be appropriate for the majority of patients. Oesophageal Doppler appears to offer a safer alternative with the added benefit of continuous online display which is particularly suitable for monitoring acute haemodynamic changes. There is no doubt that insufflation of the abdomen with carbon dioxide produces measurable effects on the cardiovascular system but the clinical significance of these is uncertain. Insufficient morbidity and mortality data are available to determine if invasive monitoring affects outcome in high-risk

5 Oesophageal Doppler in laparoscopic hernia repair 519 patients undergoing laparoscopic surgery. Extending the scope of laparoscopic techniques to older groups of patients at higher risk of cardiovascular complications requires reappraisal of haemodynamic monitoring during anaesthesia. References 1. Stoker DL, Spiegelhalter DJ, Singh R, Wellwood JM. Laparoscopic versus open inguinal hernia repair: randomised prospective trial. Lancet 1994; 343: Dubois F, Berthelot G, Levard H. Laproscopic cholecystectomy: historical perspective and personal experience. Surgical Laparoscopy and Endoscopy 1991; 1: Safran DB, Orlando R. Physiologic effects of pneumoperitoneum. American Journal of Surgery 1994; 167: Wilcox S, Vandam LD. Alas, poor Trendelenburg and his position! A critique of its uses and effectiveness. Anesthesia and Analgesia 1988; 67: Wahba R, Beique F, Kleiman S. Cardiopulmonary function and laparoscopic cholecystectorhy. Canadian Journal of Anaesthesia 1995; 4: Wiitgen C, Andrus C, Fitzgerald S, Baudendistel L, Dahms T, Kaminski D. Analysis of the haemodynamic and ventilatory effects of laparoscopic cholecystectomy. Archives of Surgery 1991; 16: Feig BW, Berger DH, Dupuis JF, Dougherty T, Johnston DA, Gross RJ, Ota DM. Haemodynamic effects of CO abdominal insufflation during laparoscopy in high risk patients. Anesthesia and Analgesia 1994; 78: S Singer M. Esophageal Doppler monitoring of aortic blood flow: beat by beat cardiac output monitoring. In: Royston D, Feeley TW, eds. International Anesthesiology Clinics, Monitoring in Anaesthesiology: Current Standards and Newer Techniques 1993; 3: No Wolf JS, Stoller ML. The physiology of laparoscopy: basic principles, complications and other considerations. Journal of Urology 1994; 15: Singer M, Clarke J, Bennett ED. Continuous hemodynamic monitoring by esophageal Doppler. Critical Care Medicine 1989; 17: Odeberg S, Ljungqvist O, Svenberg T, Gannedahl P, Backdahl M, von Rosen A, Sollevi A. Haemodynamic effects of pneumoperitoneum and the influence of posture during anaesthesia for laparoscopic surgery. Acta Anaesthesiologica Scandinavica 1994: 38: 76 7.