Introduction

The frequency of redo coronary bypass surgery is about 5-10% of coronary operations. The risks of the second operation are significantly higher than initial operation. An important element of risk is attributed to injuries to cardiac chambers, great vessels or bypass conduits which are adherent to the inner table of sternum (₁, ₂). Adhesions of the heart to the under surface of the sternum or surrounding structures not only increase the risk of rupture of the heart or major vessels but also makes operation technically difficult and prolongs operation time. Primary closure of the pericardium may decrease adhesion formation and reduces risk of re-sternotomy. But pericardial closure has adverse haemodynamic effects (₃,₄,₅,₆,₇,₈). Primary closure of the pericardium can lead to graft kinking and cardiac tamponade if they bleed post-operatively (₉). Bailey and colleagues reported that adhesions between the sternum and anterior surface of right ventricle significantly impair right ventricular function (₁₀). Primary closure of the pericardium is associated with decrease left ventricular diastolic function (₃, ₆, ₇). Cunningham and colleagues described peri- operative contraction of the pericardium as a potential cause of reduced left ventricular filling (₃). Daughters and colleagues observed release of pericardial sutures post-operatively improved cardiac output and stroke work index (₅).

These findings raise concerns regarding pericardial closure in patients with left ventricular dysfunction who require high preload to maintain cardiac output. In situations where autologous pericardium is used, primary pericardial closure may not be possible. To prevent the above mentioned complications or situations where primary pericardial closure is not possible, various other means to reconstruct pericardium with autologous fascia lata, heterologous pericardium, Dacron, expanded polytetrafluoroethylene (ePTFE), anti adhesive agents such as hyaluronic acid etc have been tried (₁₁,₁₂,₁₃). But Gallo and colleagues observed that reaction over the epicardial area facing the bovine pericardium graft greatly impeded the recognition of the coronary vessels and they also observed adhesions of the chest wall and lungs to the outer surface of the grafts (₁₁). Similarly Jacob et al observed in paediatric patients who received ePTFE grafts, a film of fine adhesions that made identification of coronary arteries difficult (₁₂). Other approaches such as parasternal incision, left or right thoracotomies have been reported to avoid injury during re-sternotomy. But sternotomy is still the best approach for multivessel disease and most of valve procedures.

So prevention of myocardial and graft injury is an important issue during re-sternotomy. We describe a technique to using a pericardial fat pad to cover cardiac chambers, bypass conduits and great vessels during primary heart operations.

Technique

Standard sternotomy is performed and pericardium is opened in inverted T shaped incision. In cases of coronary bypass operations after taking down Left Internal Mammary Artery (LIMA), the pericardial fat pad is mobilised from anterior surface of pericardium. We use diathermy to mobilise a tongue shaped piece of pericardial fat pad starting near diaphragmatic end on anterior surface of pericardium with the base near superior end of pericardium on the left side (Fig 1). Phrenic nerve lies posterior and we have never encountered any injury to it during dissection. During the standard cardiac operation the fat pad is tucked into the left pleural cavity away from operative field. For LIMA entry, the pericardium is slit open lateral to base of the fat pad. At the end of the procedure after checking for any bleeding, the pericardial fat pad is spread in front of heart. The margins of the fat pad and the cut margins of the pericardium are approximated with interrupted sutures (Fig 2). Thymus fat is approximated superiorly in the mid line. We found that most patients' fat pads are big enough to cover the right ventricle, great vessels and conduits. Total time taken for mobilisation of pericardial fat pad is less than five minutes.

Figure 1

Figure 1: Pericardial fat pad mobilized from anterior surface of pericardium.

Figure 2

Figure 2: Pericardial fat spread over the heart and sutured to cut margins of the pericardium.

Discussion

This procedure is done in all our patients except in patients with previous sternotomy, left anterolateral thoracotomy or history of mediastinal infection and chest radiation. We have not observed distortion of the LIMA graft or increased risk of cardiac tamponade. With the pericardium wide open at the entry of the LIMA, the course of the graft can be adjusted easily without fear of kinking. The technique is useful for skip LIMA or T grafts. The requirement of ionotrophs or diuretics was no different. There was no functional impairment of cardiac function and no increase post-operative bleeding observed. The Pericardial fat has no regular blood supply but receives numerous small branches supplying the pericardium which are mediastinal arteries, pericardiophrenic arteries, musculophrenic and pericardial branches of descending aorta. Since the blood supply to the pericardium and thus to the pericardial fat is from multiple sources, mobilising from one end does not interfere with viability. So for we have not come across necrotic fat tissue intra-op, post-operative or during re-operation. Apart from a little extra time we have not come across any shortcomings with this procedure.

So far we have performed re-operation in eight cases following fat pad procedure during primary operations. There were no dense adhesions and the heart was easy to mobilise. There are arbitrary grades of pericardial adhesions described in literature. But our scoring is: Grade 1= no adhesions, Grade 2= mild and could be dissected bluntly, Grade 3 = moderate and need some sharp dissection and Grade 4 = severe, unable to liberate without sharp dissection. Most of adhesions in territory of fat pad were Grade 2 and few areas were grade 3 requiring sharp dissection. None of the patients had the heart adherent to the under surface of sternum. There was no difference in visualizing the coronary arteries in the fat pad area and the rest of pericardium. It is our observation that after primary pericardial closure, right ventricle herniates through gaps in pericardium and adheres to sternum. It is not possible with the fat pad as it covers all of right ventricle. In dilated right or left ventricles where primary pericardial closure is not possible the fat pad provides tension free closure. It can be used without any risk of infection or mediastinal complications.

Pericardial fat pad is a good material to close the pericardium and it reduces adhesions of the heart to the inner table of the sternum. It is easy to separate and does not distort the grafts, and has no effect on ventricular function and maintains the barrier integrity of native pericardium. There is no extra cost involve in this procedure. We believe it to be a simple and safe procedure to facilitate re-sternotomy without effecting ventricular function or graft distortion.