The Pericardium: The Bag That Can Save or Strangle the Heart

One bag, three layers

Say "pericardium" and you have named three membranes, not one. The outermost is the fibrous pericardium: a cone of dense irregular collagen with its apex above and its base below, and the defining property that it is essentially inelastic — it will remodel and stretch over weeks, but it will not stretch in an hour. It is tethered on every side, which is exactly why the heart never swings loose in the chest. Below, its base is fused to the central tendon of the diaphragm, so firmly that the two cannot be separated by dissection. In front, the sternopericardial ligaments run from it to the back of the sternum. Above, at the apex of the cone, it does not simply stop but blends with the adventitia of the great vessels — the ascending aorta, the pulmonary trunk, the superior vena cava and the pulmonary veins — so the sac is sealed by being continuous with the coats of the vessels leaving it. Laterally it is covered by the mediastinal pleura, and the phrenic nerves with their accompanying vessels descend in that plane, plastered to its sides. The whole structure occupies the middle mediastinum, which is described in the mediastinum.

Inside the fibrous sac lies the serous pericardium, and the single most important thing to grasp about it is that it is not two membranes but one continuous sheet folded back on itself, exactly like the arrangement in the pleura. The parietal layer lines the inner surface of the fibrous pericardium and is fused to it. Where the great vessels enter and leave, that lining turns sharply — reflects — and runs back down over the surface of the heart itself as the visceral layer, better known as the epicardium. Between the two lies the pericardial cavity: not an empty space but a potential space, a film containing roughly 15 to 50 mL of clear serous fluid that lets the beating heart glide against its own bag some hundred thousand times a day without friction. The epicardium is more than a lining; beneath it runs the subepicardial fat in which the coronary arteries and cardiac veins travel before they dive into the muscle.

Two sinuses left behind by the folding

The reflections are not random: the arteries leave together, the veins arrive together, and two gaps are left between them. The serous pericardium reflects around the great vessels in two separate groups. One tube of reflection surrounds the arterial pair — the ascending aorta and the pulmonary trunk — which leave the heart side by side, wrapped in a common sleeve. A second, larger and more irregular reflection surrounds the venous group: the superior and inferior venae cavae and the four pulmonary veins, arranged in an inverted-U shape. Because these two reflections are separate, a passage is left between them. That passage is the transverse pericardial sinus: bounded in front by the ascending aorta and pulmonary trunk, and behind by the superior vena cava and the anterior surface of the atria. You can pass a finger through it from one side of the heart to the other. For the cardiac surgeon this is not a curiosity but a working tool — passing a ligature or a clamp through the transverse sinus encircles both great arteries at once, which is precisely how the arterial outflow is controlled when a patient goes onto cardiopulmonary bypass. The vessels it surrounds are described in the aorta and great vessels.

The second space is the oblique pericardial sinus, and it is quite different: not a through-passage but a blind cul-de-sac. It lies behind the left atrium, and its walls are formed by the venous reflection itself — the four pulmonary veins to either side and the inferior vena cava below, with the sac closed above. Slide a hand upwards behind the heart from below and it comes to a dead stop in this recess, like a hand entering a closed pocket. Both sinuses are developmental leftovers: in the embryo the heart tube is slung from the back of the pericardial cavity by a fold called the dorsal mesocardium, and when the middle of that fold breaks down the transverse sinus is what remains of the hole it left. Clinically the oblique sinus matters because it lies immediately behind the left atrium described in the heart chambers — so an effusion can collect there silently, and a transoesophageal probe sitting in the oesophagus looks straight through the back of the left atrium into it.

Nerves and vessels: why pericardial pain lands on the shoulder

The bag borrows its nerve from a muscle in the floor of the chest — and inherits that muscle's dermatomes. The fibrous pericardium and the parietal layer of the serous pericardium are supplied by the phrenic nerve, root value C3, C4 and C5 — the same nerve that supplies the diaphragm to which the sac is fused, running down on each side of the pericardium between it and the mediastinal pleura. This is a somatic nerve, and the consequence is immediate: pericardial pain is sharp, well localized in quality, and it refers to the dermatomes of C3–C5 — the supraclavicular region, the root of the neck, and above all the tip of the shoulder. A patient with pericarditis who says the pain shoots into the left shoulder and up the side of the neck is not describing anything mysterious; they are describing the sensory territory of the nerve their pericardium shares with their diaphragm. There are also vagal and sympathetic contributions of uncertain significance. The visceral layer — the epicardium on the heart itself — is essentially insensitive to pain, which is why the pain of pericarditis and the pain of myocardial ischaemia feel so different: one is somatic and sharp, the other visceral, dull and crushing.

The blood supply follows the same company. The principal artery is the pericardiacophrenic artery, a slender branch of the internal thoracic artery that descends alongside the phrenic nerve on the lateral surface of the sac all the way to the diaphragm — nerve and artery travelling together as a named neurovascular pair. It is joined by the musculophrenic artery (the internal thoracic's other terminal branch), by small pericardial branches directly from the descending thoracic aorta, and by twigs from the bronchial, oesophageal and superior phrenic arteries. Venous blood returns through the pericardiacophrenic veins into the internal thoracic veins, and through small pericardial veins into the azygos system. The visceral layer is not supplied by any of these: the epicardium takes its blood from the coronary arteries, because it belongs to the heart, not to the bag.

The pressure–volume curve: 150 mL versus a litre

The same sac, the same fluid — and two completely different outcomes, decided entirely by speed. The pericardium has a small amount of slack built in: the cavity has a reserve volume, a little unused space it can give up before its walls come under tension at all. Fill that reserve slowly and nothing happens to the pressure inside. Exceed it, and the curve turns almost vertical — because the collagen has run out of give, each further millilitre now raises intrapericardial pressure steeply. That is the entire explanation for the clinical paradox. A rapidly accumulating effusion — a stab wound, a ruptured ventricular free wall after a myocardial infarction, an ascending aortic dissection leaking backwards into the sac — blows through the reserve volume in seconds, and as little as 100 to 200 mL takes the pressure above the filling pressure of the right heart. A slowly accumulating effusion, over weeks, gives the fibrous pericardium time to remodel and creep outwards, so a litre or even more can be tolerated with an oddly comfortable patient and only a huge globular heart on the chest film.

Once intrapericardial pressure exceeds the pressure inside the chambers during diastole, the thin-walled right atrium and right ventricle collapse inwards, venous return is obstructed, filling falls, and with it stroke volume and cardiac output. The clinical signature is Beck's triad: hypotension, distended neck veins from the dammed-up venous return, and muffled heart sounds because fluid is a poor conductor. Two more signs complete the picture. Pulsus paradoxus is an exaggeration of a normal phenomenon — during inspiration blood is drawn into the right heart, which bulges the septum leftwards and, inside a rigid sac, can only fill at the left ventricle's expense; a fall in systolic pressure of more than 10 mmHg on inspiration is the result, and in severe cases the radial pulse disappears with each breath in. Electrical alternans on the ECG is the strangest of all: the QRS complexes alternate in height beat to beat, because the heart is literally swinging back and forth in a pool of fluid, changing its position relative to the electrodes with every contraction.

Draining the bag, and when the bag itself falls ill

Pericardiocentesis is the anatomy of the sac turned into a procedure. The classic subxiphoid approach puts the needle in the angle between the xiphoid process and the left costal margin, advanced at about thirty to forty-five degrees to the skin and aimed at the tip of the left shoulder. Every element of that description is anatomical: entering below the costal margin avoids the internal thoracic vessels running a centimetre or so lateral to the sternal edge; the subxiphoid route stays inferior to the pleural reflections so the needle does not cross a pleural cavity and create a pneumothorax; and the aim toward the left shoulder follows the axis of the heart into the largest, most dependent part of the effusion. In modern practice the needle is guided by echocardiography and often placed from an apical or parasternal window instead, because ultrasound shows exactly where the fluid actually is. The surface landmarks are set out in thoracic surface anatomy and procedures.

Inflame the two serous layers and they stop sliding — they start scraping. Acute pericarditis produces a pain that is retrosternal and sharp, worse on lying flat and on deep inspiration, and characteristically relieved by sitting up and leaning forward — a posture so consistent that patients adopt it before anyone tells them to, because it lifts the heart away from the inflamed parietal layer. On auscultation there may be a pericardial friction rub, a superficial scratching sound like walking on fresh snow, best heard at the left sternal edge with the patient leaning forward in expiration. The ECG is distinctive: widespread, concave-upwards "saddle-shaped" ST elevation across many leads with PR-segment depression, and no reciprocal changes — quite unlike the territorial, convex ST elevation of an infarct. Causes run from viral infection (the commonest) to tuberculosis, uraemia, malignancy, autoimmune disease and radiation; treatment is usually high-dose anti-inflammatory drugs with colchicine rather than anything aimed at the heart itself.

Chronic inflammation ends in a different disease altogether. In constrictive pericarditis the sac becomes thickened, fibrosed and sometimes calcified — classically after tuberculosis, and now also after cardiac surgery or radiotherapy — until the heart is encased in a rigid shell. Filling stops abruptly in early diastole as the ventricle hits that shell, producing an early diastolic "pericardial knock", a raised JVP that paradoxically rises further on inspiration (Kussmaul's sign), and a congested liver with ascites in a patient whose ventricles contract perfectly well. The definitive treatment is surgical: the pericardium is stripped away entirely, and life without it is essentially normal — proof that the sac is a useful accessory rather than a necessity. Two more entries belong here. Dressler's syndrome is an autoimmune pericarditis appearing two to six weeks after a myocardial infarction, when antigens released from necrotic myocardium provoke an immune response — a late complication to keep in mind alongside the acute management covered in acute coronary syndrome. And in cardiac surgery the pericardium is deliberately opened and then left open or only loosely approximated, precisely so that postoperative bleeding or oedema cannot reproduce the tamponade the surgeon has just spent hours preventing.