Cardiovascular Responses to Manual Dissection Associated With Trakhiatal Esophageal Resection

Krikor Yakoubian, MD, Bernard Bougeois, MD, Jean Marty, MD, Jean Pierre Marmuse, MD,

and Jean Marie Desmonts, MD

Marked cardiovascular responses to esophageal ma- significant increase in PCWP also occurred, whereas nipulation are commonly observed during trans- HR and SVR were unchanged. All parameters re- hiatal esophagectomy without thoracotomy. To delin- turned to control values once manipulation was eate the mechanism and the time course of these stopped. The simultaneous decreases in MAP and Cl hemodynamic changes, 10 patients undergoing this were clearly related to impaired venous return and/ surgical procedure were studied. Heart rate (HR), or ventricular ejection from manual compression. mean arterial blood pressure (MAP), pulmonary cap- From these findings, it can be assumed that long- illary wedge pressure (PCWP), cardiac output (CO), lasting alterations in hemodynamic status could be cardiac index (Cl), systemic vascular resistance (SVR), detrimental to patients with cardiac disease: there- and arteriovenous oxygen content difference (C[a- fore, careful intraoperative hemodynamic monitoring v]O,) were measured or calculated each time the is advisable. In addition, some patients with ad- surgeon’s hand entered the thorax to dissect the vanced cardiac dysfunction may not tolerate these esophagus. Significant and simultaneous decreases manipulations and may require an open thoracic in both MAP and Cl were observed in all cases during surgical procedure. manipulations (-46% from the baseline value). A 0 1990 by W.B. Saunders Company.

T RANSHIATAL esophagectomy without thoracotomy is commonly performed by

some surgeons in patients with benign or malig- nant disease.’ The esophagus is approached through both a left cervical and an abdominal incision, then blunt resection is performed follow- ing manual dissection of the esophagus by the transhiatal route. A gastric or colic tube is subsequently placed, tunneled through the poste- rior mediastinum, and then anastomosed to the residual part of the cervical esophagus.’ Advan- tages of this technique compared with an esoph- agectomy via a thoracotomy include: (1) the surgical procedure is faster and simpler; (2) need for postoperative ventilatory support is reduced; (3) postoperative pulmonary complications are lower; and (4) the risk of mediastinitis is mark- edly reduced when a leakage of the anastomosis occurs. However, marked arterial hypotension is commonly observed during manual dissection of the esophagus and at the time the stomach is brought up through the posterior mediastinum.lm4 These hemodynamic alterations might be ex-

petted to be deleterious in patients with cardiovas- cular disease, and postoperative myocardial in- farction has been reported to occur after this operation.1,3 In a series of 143 patients, Orringer and Orringer reported 11 deaths (8%), with a recent myocardial infarction present in three of these patients (27%).5 Similar findings were more recently reported by Marmuse, who ob- served a mortality rate of 10% in a series of 82 patients. In two of the eight deaths, a recent myocardial infarction was present. This investiga- tion was conducted in 10 patients undergoing esophagectomy for carcinoma through a trans- hiatal approach to further delineate the factors involved in the cardiovascular response to man- ual dissection of the esophagus.

MATERIALS AND METHODS

From the Dkpartement d’Anesthdsie et de RCani- mation Chirurgicale and the Service de Chirurgie Visckrale. Hbpital Bichat, Paris, France.

Address reprint requests to J.&i. Desmonts, MD, Dkpartment d’Anesthesie et de Rtanimation Chirurgicale, Hopital Bichat. 46 rue de Henri-Huchard 75878 Paris, Cedex 18. France.

Ten adult patients, ASA I or II, scheduled for transhiatal esophagectomy gave their informed consent to participate in this study, which was approved by the Institu- tional Review Board. The patients ranged in age from 41 to 80 years (mean * SD, 58.3 + 13.6), in weight from 38 to 80 kg (52.6 f 11.8), and in surface area from 1.31 to 2.0 mZ (1.57 + 0.2). All of them had total dysphagia to solid food. A history of coronary artery disease was noted in two patients, but none had congestive heart failure. Associated diseases present in the patients are listed in Table 1.

o 1990 by W.B. Saunders Company. 0888-6296/90/0404~007803.00/0

All patients were premeditated with diazepam and atropine administered 45 to 90 minutes before surgery. Anesthesia was induced with thiopental, followed by succinyl- choline to facilitate endotracheal intubation. Mechanical ventilation was used in every patient. Anesthesia was main- tained with an O,-N,O mixture (40% to 60%), plus fentanyl or phenoperidine, and pancuronium was used as the muscle

458 Journal of Cardiorhoracic Anesthesia, Vol4, No 4 (August), 1990: pp 458-46 1

TRANSHIATAL ESOPHAGEAL RESECTION 459

Table 1. Associated Diseases

Associated Diseases

No. of Patients

Alcohol intake 6 Tobacco intake 7 Arteriopathy 1

Coronary artery disease 2

Hypertension 1 Chronic obstructive pulmonary disease 3

Malnutrition 3 Diabetes 2 Sepsis 1

relaxant. Radial arterial and pulmonary arterial catheters were inserted after induction. Heart rate (HR), mean arterial pressure (MAP), and pulmonary capillary wedge pressure (PCWP) were recorded, and cardiac output (CO) was measured with the thermodilution technique. Blood samples were withdrawn from the radial and pulmonary artery catheters for subsequent determination of arterial and mixed venous oxygen contents. Cardiac index (CI), systemic vascu- lar resistance (SVR), and arteriovenous oxygen content difference (C[a-v]O,) were calculated using standard formu- lae. Measurements were performed before, during, and after each time the surgeon’s hand entered the posterior mediasti- num to dissect the esophagus.

Total dissection time ranged from 5 to 15 minutes for each patient and required several manual manipulations. Twenty-seven manipulations were recorded from the whole group. For each patient, the lowest values of the MAP and other hemodynamic parameters were selected for the calcula-

CARDIAC OUTPUT 4.2 L/min 1.1 L/min

160

: ,I

’

100 1 . I I’! ‘;

tion of the mean values for the group. Data are presented as mean k SD. Statistical analysis was performed using an analysis of variance. A P < 0.05 was considered significant.

RESULTS

The duration of the surgical procedures ranged from 3 to 7 hours. The interval of time for intrathoracic dissection varied from 5 to 15 minutes, and each intrathoracic manipulation lasted between 30 and 90 seconds. The blood loss ranged from 440 to 3,230 mL. Three patients died during the postoperative period, two of them with a recent myocardial infarction. Both had documented coronary artery disease.

A typical recording of hemodynamic changes during intrathoracic manipulation is presented in Fig 1. Significant decreases in MAP and CI were observed at the time of each manipulation during intrathoracic esophageal resection in all cases. MAP decreases ranged from - 34.6% to - 70% of the basal value, with a mean decrease of - 52.2% (Fig 2). The decrease of CI occurred with the decrease in MAP and ranged from -32.9% to -79.2%, with a mean decrease value of -55% (Fig 2). At the same time, a significant increase in PCWP was ob- served (Fig 2). No significant changes in SVR

4.6 L/min r

AS? p 20 mm Hg

,a ;

TIME

IN OUT HEMODYNAMIC CHANGES OBSERVED DURING MANUAL DISSECTION

Fig 1. A typical recording of the hemodynamic events observed during intrathoracic esophageal resection. “In” represents the moment the surgeon’s hand entered the posterior mediastinum. “Out” represents the moment the surgeon’s hand was removed. Total time elapsed between the two points In and Out, was 55 seconds in this case.

460 YAKOUBIAN ET AL

mm Hg DISCUSSION

The present study permits assessment of the hemodynamic responses to intrathoracic man- ual dissection required for transhiatal esophageal resection. The decrease in blood pressure, ob- served in all cases during intrathoracic manipula- tion, was caused by a decrease in CI since SVR was unchanged. This was related to a decrease in preload, an impediment to ventricu- lar ejection, or both. The decrease in preload was caused by mechanical compression of the left or right atrium, producing an acute reduction in venous return with sudden decreases in CI and blood pressure. The increase in PCWP observed in almost all cases was the result of an increase in intrathoracic pressure produced by the surgeon. A second possible cause of the reduction in CO was an impediment to ventricular ejection by cardiac displacement.’ In this case, the increase in PCWP would be related to a temporary alteration of cardiac performance. The decreases in CI and blood pressure in either case were caused by the intrathoracic manipulations per- formed by the surgeon.

The premature ventricular contractions ob- served in 50% of the patients during manual dissection probably were secondary to mechani- cal stimulation of the heart, but could also be due to decreased myocardial oxygenation following the decreased CI and MAP. However, no sign of myocardial ischemia was observed in any of the patients in the V, ECG lead. The decrease in CO

100

80

60

40

20

0 BEFORE DURING AFTER TIME

L/min/m2

zo

1.0

0.0 BEFORE DURING AFTER rlME

mm Hg

10

t-l BEFORE DURING AFTER TIME

Fig 2. Changes in mean arterial pressure (MAP), cardiac index (Cl), and pulmonary capillary wedge pressure (PCWPl during intrathoracic manipulations. ??P < 0.06 versus before.

and HR were observed (Table 2). However, 5 of 10 patients exhibited multiple ventricular extra- systoles at least once during manipulations. A significant increase in the C(a-v)O, was also observed at this time (Table 2); arterial blood gases remained unchanged.

All parameters tended to return to the baseline values once the surgeon discontinued the manipulation. However, the time to return to baseline varied among patients.

Table 2. Heart Rate (HR], Systemic Vascular Resistance

(SVRI, Arterial Blood Gases, Arteriovenous 0, Difference (C[a-v]O,], and Pulmonary Shunt fQs/Qt) Before. During, and After lntrathoracic Manual Manipulations fn = 27)

Basal

During lntrathwacic After Manipulation Manipulation*

HR (beats/min) 89 + 15 93 + 23 93 * 14

SVR (dyne . s - cm-7 1990 L 720 2020 + 850 1660 f 640

PH 7.43 f 0.06 7.45 i 0.06 - PaO, (mm Hg) 106 f 24 95 f 25 PaCO, (mm Hg) 34 f 8 32 f B -

C(a-v)O, (mL/ 100 mL) 5.8 f 0.9 8.9+1.0t -

Qs/Qt (%) 13 + 3 9*3 -

*Arterial and venous blood samplings were not performed

after manipulation. fP < 0.05.

TRANSHIATAL ESOPHAGEAL RESECTION 461

also produced the widening of the arteriovenous oxygen content difference.

Periods of hypotension are not likely to be deleterious in healthy patients, but may be dan- gerous in patients with coronary artery disease or in the elderly. The incidence of postoperative myocardial infarction is significantly higher in patients with critical coronary stenosis who have marked decreases in driving pressure.6 The 30% mortality in the present series is quite high, but only 10 patients were included. The ages of the patients developing lethal complications were 42, 70, and 80 years. Marmuse has recently re- ported a survey of 82 patients undergoing trans- hiatal esophagectomy with a mortality rate of 2.8% below 75 years of age and 43% above 75 years, which is consistent with this present study. Although this study could not determine whether esophagectomy without thoracotomy is safer than

esophagectomy with thoracotomy, it draws atten- tion to the consequences of hypotension that might be implicated in the postoperative morbid- ity of patients with coronary artery disease.’

From these findings, it is recommended that extensive hemodynamic monitoring be used in all patients undergoing this surgical proce- dure. The continuous monitoring of blood pres- sure with a radial arterial catheter is necessary since the decrease in blood pressure was associ- ated with a decrease in CI in all patients. Continuous monitoring of mixed venous oxygen saturation could also be used to augment intermit- tent measurements of CO. To minimize the consequences of hypotension, it is suggested that oxygenation be optimized during manual manip- ulations, blood volume be increased, and the duration of each manipulation be kept as short as possible.

REFERENCES

1. Orringer MB, Sloan H: Esophagectomy without thoracotomy. J Thorac Cardiovasc Surg 76:643-654, 1978

4. Marmuse JP: Transhiatal esophagectomy: Techni- cal procedure. J Chir (Paris) 125:585-592,1988

2. Terz JJ, Beatty JP, Kokal WA, Wagman LD: Transhiatal esophagectomy. Am J Surg 154:42-48, 1987

3. Tryzelaar JF, Neptune WB, Ellis Jr FH: Esoph- agectomy without thoracotomy for carcinoma of the esopha- gus. Am J Surg 143:486-489, 1982

5. Orringer MB, Orringer JS: Esophagectomy with- out thoracotomy: A dangerous operation? J Thorac Cardio- vast Surg 85:72-80, 1983

6. Rao TLK, Jacobs KH, El Etr A: Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 59:499-505, 1983