Off-Pump Bypass Surgery

Abstract

This is a review of 733 patients who underwent off-pump bypass surgery of the right coronary artery and left anterior descending coronary artery between 1969 and 1985.

Two hundred sixty-four patients underwent single bypass of the left anterior descending coronary artery, and 79 patients underwent single bypass of the right coronary artery. Both the left anterior descending and right coronary arteries were bypassed in 390 patients.

In contrast to the present-day use of mechanical devices to stabilize the target vessel, a 4-suture surgical technique was used for this purpose. This technique, which we illustrate, proved less cumbersome and made the graft anastomosis easy to perform.

Our early experience (1969 through 1972) in operating on 199 patients resulted in an operative mortality rate of 4.5% (9/199). From 1973 through 1985, improved patient selection and use of the left internal thoracic artery as the conduit of choice for bypass of the left anterior descending coronary artery reduced the operative mortality rate for 534 patients to 1.3% (7/534).

Routine postoperative angiograms were not performed; therefore, the graft patency rate is not available. However, an ongoing 34-year follow-up study of the 264 patients who underwent a single left anterior descending bypass showed the saphenous vein graft to be open in 64.3% (18/28) patients and the left internal thoracic graft in 92.2% (59/64) of patients studied. Seventy-four of the 264 patients in this study were still alive in 2003.

Off-pump coronary artery bypass surgery, which uses mechanical devices to dampen the movement of the beating heart, is now accepted.^1–3 This rebirth of interest in off-pump coronary bypass stimulated me to review our early experience with 733 consecutive patients in whom we used a simple 4-suture technique to stabilize the target vessel.

Patients

A daily log provided a list of consecutive off-pump bypass operations of right coronary arteries (RCA) and left anterior descending coronary arteries (LAD). Data recorded included the patients' dates of birth, addresses, types and dates of operation, operative deaths, autopsy findings, reoperations, and results of postoperative coronary angiograms. In addition, the patients' clinical charts, including operative notes, were reviewed for confirmation and for additional information.

Of the 733 patients who underwent surgery, 600 were men and 133 were women. Their average age was 54.6 years (range, 26 to 78 years). In 390 of the total group, both the RCA and the LAD were bypassed. In 264 of the patients, only the LAD was bypassed, and in the remaining 79 only the RCA was bypassed.

Surgical Technique

A midline sternal splitting incision was used to open the chest, and the heart was placed in a pericardial cradle. Laparotomy pads behind the heart displaced it anteriorly, thereby exposing the distal LAD. To expose the main RCA, myocardial traction sutures displaced the heart to the left and in a cephalad direction.

Four sutures were placed to stabilize and elevate a segment of the RCA (Fig. 1) or LAD (Fig. 2).⁴ Two were 3-0 silk sutures passed under the coronary artery 5 to 6 cm apart. These were placed widely and deeply enough to provide a cushion of myocardium between the sutures and the coronary artery (see inset in Fig. 2). This placement kept the sutures from direct contact with the coronary artery and thereby prevented any injury of the vessel when traction was applied to the sutures. The other 2 sutures were 4–0 silk sutures that were passed through epicardium along each side of the elevated coronary segment to provide additional stability. If coronary occlusion was not complete without excessive suture traction, a small bulldog clamp was applied.

Fig. 1 Stabilization of the right coronary artery by means of the 4-suture technique.

Fig. 2 Stabilization of the left anterior descending coronary artery by means of the 4-suture technique.

This 4-suture method for stabilizing a target vessel created a motionless, bloodless field, and a precise anastomosis of a saphenous vein graft (SVG) or left internal thoracic artery (LITA) to a linear arteriotomy was easily accomplished with either continuous or interrupted 6-0 or 7-0 Prolene sutures. The patient received 1.5 mg/kg of heparin before occlusion of the artery that was to be bypassed. This was reversed with protamine after coronary blood flow was reestablished.

Saphenous veins removed from the thigh through short transverse incisions provided bypass grafts for 469 RCAs and 193 LADs. The LITA, harvested by means of cautery, was used to bypass 461 LADs.

As a precaution, a heart-lung machine was connected to the 1st few patients, since there was concern that the displaced beating heart might not support circulation in the presence of coronary artery disease. However, we observed that the beating heart maintained circulation and that the prophylactic attachment of a heart-lung machine was not necessary. A rotating disc oxygenator was readily available on standby and was eventually used for 3 patients, one of whom died.

Results

Early Results (1969–1972). Table I shows the operative mortality, during the 1969 through 1972 period, for the 3 groups: single LAD bypass, single RCA bypass, and combined RCA and LAD bypasses. There were no deaths in the single LAD and single RCA groups, but 9 deaths occurred among the 120 patients who underwent bypass of both the RCA and the LAD.

TABLE I. Operative Mortality for Off-Pump Coronary Bypass Operations during 1969–1972

Postmortem examination in 8 of the 9 RCA–LAD patients who died revealed severe left main coronary artery disease in 3, diffuse triple-vessel disease in 2, and a recent posterior-inferior myocardial infarction in 3 who had required RCA endarterectomies. The operative mortality rate for all 3 groups during the period 1969 through 1972 was 4.5%

Later Results (1973–85). Starting in 1973, we changed the criteria for selecting operative candidates. Every effort was made to perform combined RCA–LAD bypass surgery only in patients whose disease was limited to those 2 vessels. In addition, during the 1973 through 1985 period, the decision was made to choose the LITA, when possible, for bypass of the LAD (see Fig. 3). Of the 534 patients who underwent surgery during this period, 7 died, for a reduced mortality rate of 1.3% (Table II).

Fig. 3 Comparative use, per annum, of saphenous vein grafts and left internal thoracic artery grafts for bypass of left anterior descending coronary arteries.

LITA = left internal thoracic artery; SVG = saphenous vein graft

TABLE II. Operative Mortality for Off-Pump Coronary Bypass Operations during 1973–1985

Of the 2 deaths among the 264 patients who underwent a single LAD bypass, one developed irreversible ventricular tachycardia 2 hours after surgery. The autopsy showed a patent LITA graft to the LAD and severe obstruction of the left main coronary artery. The 2nd death was due to a dissecting aortic aneurysm with an intimal tear 2 cm proximal to the SVG–aortic anastomosis. The 1 death after a single RCA bypass was due to pulmonary insufficiency.

Because this series of off-pump operations was not part of a clinical study, routine postoperative coronary angiograms were not performed, and, as a result, information about postoperative graft patency is not available. Only patients with recurrent symptoms underwent repeat angiograms, which were frequently performed many years after surgery. As a part of an ongoing 34-year follow-up study of the 264 patients who underwent a single LAD bypass, a review of repeat angiograms showed that the SVG was patent in 64.3% (18/28) of the patients studied and the LITA in 92.2% (59/64). Seventy-four of the 264 patients were still alive in 2003.

Discussion

Initially, coronary artery surgery with the heart beating was performed only on completely blocked arteries. In 1958, Longmire and colleagues⁵ reported having performed endarterectomies with the heart beating, in order to relieve total obstruction of the RCA or the LAD. In 1969, Kolessov⁶ reported having used LITA grafts, without a pump, to bypass 5 completely blocked LADs and 1 blocked circumflex coronary artery.

Because most of the coronary vessels that were treated in our series were not totally occluded and were functioning, the question arose of whether or not the occlusion time required to complete a graft anastomosis would cause myocardial damage. As it happened, cardiac function did not change during the short time (less than 15 minutes) required to perform a routine graft anastomosis to a vessel with a lumen. The postoperative electrocardiogram showed no evidence of myocardial ischemia.

Green's group,⁷ in 1970, reported on their pioneering work in which they sutured a LITA to a distal LAD while the patient was on the pump. Our series of operations shows that a LITA can be anastomosed to the LAD on the beating heart without a pump. For us, the LITA became the bypass vessel of choice for the LAD, beginning in 1973. From 1973 through 1985, 87% (183/210) of patients in the single LAD bypass group received a LITA shunt. This fortunate decision was made on the basis of studies showing that a LITA inserted into a left ventricular myocardial tunnel (Vineberg and McMillan⁸) developed intramyocardial collateral vessels and remained a patent conduit for many years. It has since been shown that the LITA graft remains open longer than a vein graft and accounts in large part for longer patient survival.^9–11

The difference between recent off-pump coronary artery surgery and our early experience lies in the technique of stabilizing the target vessel. Present-day mechanical devices dampen cardiac motion either by holding the myocardium by means of small suction cups or by applying downward pressure on each side of the coronary segment. These stabilizing instruments do not necessarily occlude the target vessel; therefore, bleeding occurs. A bloodless field is attained with a CO₂ blower, an intracoronary occluder, a shunt, or a bulldog clamp.

In contrast, the 4-suture technique used in this series is simple. Applying traction to sutures that have been passed appropriately under the target segment occludes the artery without injuring it, providing an elevated, motionless target vessel and a bloodless field for easy anastomosis of a bypass graft. Mechanical devices are not needed for bypassing coronary arteries on the front of the heart. However, mechanical devices could be a useful aid to hold the heart when using this 4-suture technique to bypass the circumflex artery and its branches.

Our experience demonstrated that the selection of vessels to be bypassed with the heart beating was critical. During the first 3 years (1969–1972), we operated on patients with extensive disease that involved vessels other than the target vessels. During the 1973–1985 period, better selection of patients and use of the LITA for bypassing the LAD lowered our operative mortality rate to 1.3% of the 534 consecutive patients who underwent surgery.

Conclusion

Our early experience with off-pump surgery for treating coronary disease shows that the RCA and LAD can be bypassed safely using a simple traction suture technique that provides a nonmoving, obstructed coronary segment for an accurate, easily performed graft anastomosis.

Acknowledgment

I extend special thanks to my secretary Ms Nancy Negosian, who maintained the accurate daily log of operative procedures that made this study possible, and to my son Jay L. Ankeney, Jr., who helped prepare the text.