Hypotension, a Pericardial Effusion, and a Dilated Right Ventricle at 03:10

Each finding needs to be weighed separately, because they do not all point the same way.

The effusion: 1.6 cm anterior to the right ventricle in diastole is a moderate effusion. On its own it is not enough to explain shock. In a patient six days after arthroplasty and on a direct oral anticoagulant, a haemorrhagic pericardial effusion is plausible, though a small reactive effusion after major surgery is also common and usually harmless.

Right atrial systolic collapse is sensitive for tamponade but not specific. It can occur with any moderate effusion, and it is more likely when right-sided pressures are low, which is the opposite of this patient. What would raise my confidence is right ventricular diastolic collapse, and, more than either, respirophasic variation in mitral inflow above about 25% and tricuspid inflow above about 40%, plus diastolic bowing of the interventricular septum into the left ventricle.

The right ventricle: a dilated, hypokinetic free wall with apical sparing (McConnell's sign) is a pattern associated with acute pulmonary embolism and acute cor pulmonale. It is not specific, but in a patient six days after arthroplasty, with clear lungs, hypoxaemia and a plethoric IVC, the pretest probability of pulmonary embolism is high. Absent B-lines support a non-cardiogenic cause of the hypoxia.

The plethoric IVC fits both tamponade and right heart failure from pulmonary embolism.

On balance, the combination of a dilated, hypokinetic right ventricle with apical sparing, clear lungs, hypoxaemia, postoperative status and tachycardia makes pulmonary embolism the more likely driver. The effusion is probably an incidental or reactive finding until the inflow Doppler says otherwise. The honest answer is that both can coexist, and the next few minutes should be spent proving which is dominant rather than choosing by gestalt.

I would add the following, in this order, because each is fast and each changes management.

Pulsed-wave Doppler of mitral and tricuspid inflow across the respiratory cycle, from the apical four-chamber view. A mitral inflow variation of more than about 25% and a tricuspid variation of more than about 40% support tamponade; blunted or absent variation argues against it. This is the most useful discriminator when RA collapse is the only tamponade sign.

Right ventricular inflow and outflow timing, looking for diastolic collapse of the RV free wall and for a dilated RV that is under pressure.

A subcostal and parasternal short-axis view to see whether the interventricular septum flattens or bows in diastole or in systole. Systolic flattening with a D-shaped left ventricle favours pressure-overloaded right heart from pulmonary embolism. Leftward septal bowing throughout diastole favours tamponade.

A tricuspid regurgitation jet for an estimated right ventricular systolic pressure, and a look at the pulmonary artery acceleration time; a short acceleration time with a mid-systolic notch supports pulmonary hypertension from acute embolism.

Lung and pleural assessment: absent sliding with a small effusion would raise the possibility of a small embolic infarct.

A quick compression ultrasound of both legs, particularly the left popliteal vein. A demonstrable deep vein thrombosis in this setting would substantially raise the probability of pulmonary embolism and is quick to obtain.

What I would not rely on: electrical alternans, which is insensitive, and the size of the effusion alone, which does not determine tamponade physiology.

This is the judgement call, and both of the nurse's options carry serious risk.

Thrombolysis is the more dangerous reflex here. Major surgery within the previous three weeks is a strong relative contraindication, and this patient is six days out from a knee arthroplasty, on an anticoagulant, with a fresh prosthesis and a wound. Systemic thrombolysis in that setting carries a substantial risk of major bleeding, including into the pericardial space if the effusion is in fact haemorrhagic. It should not be given empirically on a bedside impression in a hospital without cardiac surgical backup.

Pericardiocentesis is equally premature without right-sided diastolic collapse and documented inflow variation. Needling a moderate effusion in a patient on rivaroxaban, without a confirmed haemodynamic indication, exposes him to a preventable bleed.

What I would do: stabilise and confirm. Give cautious fluid resuscitation with frequent reassessment, because the right ventricle is failing and fluid loading can worsen septal shift; start vasopressor support to maintain coronary perfusion; correct the anticoagulant effect only if a bleeding source is identified, and discuss reversal with the local protocol, because reversing rivaroxaban in a patient with suspected pulmonary embolism removes his protection. Complete the inflow Doppler and leg vein study. Then obtain CT pulmonary angiography, accepting that single-phase technique is less sensitive for small peripheral emboli but adequate for a central embolus in a patient with this degree of right heart strain. Speak to the cardiologist and the on-call intensivist by phone now, and arrange transfer regardless of the CT result, because this patient needs a level of care this hospital cannot provide.

If right ventricular diastolic collapse and significant inflow variation do appear while I am scanning, the decision reverses: echo-guided pericardiocentesis with reversal of anticoagulation and immediate transfer becomes the priority, and thrombolysis remains off the table.