A patient presented with cardiogenic shock, and a left ventricular assist device (LVAD) was inserted. Transthoracic echocardiography demonstrated air in the left side of the heart whenever the patient coughed. A chest x-ray and computed tomography of the chest did not reveal a pneumothorax. Air appeared to enter the left atrium (LA) around the LA cannula. The working diagnosis was a right pleural tear with intermittent passage of air from the pleural space (supra-atmospheric during coughing) into the LA (around the cannula), which contained a large-bore LVAD cannula on -40 mm Hg suction. This was confirmed by inserting a right-sided chest drain, and no air appeared during coughing. This case demonstrates an unusual complication of LVAD use (systemic air embolization) and highlights the point of variable physiology during dynamic versus static imaging procedures.
A 52-year-old man presented with an acute anterior ST-segment elevation myocardial infarction complicated by acute pulmonary edema. Urgent double-vessel percutaneous coronary intervention was performed, with stenting of left anterior descending and left circumflex artery stenoses. The patient required intubation and ventilation, an intra-aortic balloon pump, and inotropic support to manage cardiogenic pulmonary edema. Twelve hours later, persistent cardiogenic shock necessitated insertion of a left ventricular assist device (LVAD). A Thoratec LVAD (Thoratec Corp, Pleasanton, CA) was implanted, with the left-sided heart inflow cannula placed in the left atrium (LA) (on suction with a negative pressure of -40 mm Hg) and the left-sided heart outflow cannula attached as an end-to-side anastomosis to the ascending aorta. A left atrial, rather than a left ventricular inflow cannula was inserted because of a recent large anteroapical myocardial infarction (with consequent poor tissue quality) precluding stable suturing and positioning of an apical left ventricular cannula.
Six weeks later, during a routine transthoracic echocardiogram (TTE), it was observed that air bubbles would appear in the left side of the heart during coughing, initially in the LA (see Video 1 as an example of air entering the left side of the heart while coughing). These were not present during normal respiration. Note the stagnant flow in the left ventricle and absence of aortic valve opening (the left ventricle has been “bypassed” because of the LA positioning of the LVAD inflow cannula).
A transesophageal echocardiogram was performed under conscious sedation, but no air entered the left side of the heart during coughing. However, a mobile thrombus was noted in the inflow cannula in the left side of the heart, along with intermittent LA wall suck-down into the inflow cannula, with resultant LA wall hematoma and dissection and an associated localized pericardial effusion ( Figures 1 and 2 ; Videos 2 and 3 ).
A chest x-ray and computed tomography scan of the chest revealed a right pleural effusion with no evidence of a pneumothorax. The LVAD was functioning correctly, and there was no evidence of air within the LVAD circuit.
The patient then developed a focal neurologic deficit with a receptive and expressive dysphasia, right homonymous hemianopia, and reduced conscious state (Glasgow Coma Scale of 13). A computed tomography scan of the brain was normal. A repeat TTE demonstrated extensive entry of air into the left side of the heart during coughing, with the entry site appearing around the left-sided heart inflow cannula in the LA ( Video 1 ).
A working diagnosis of a pleural tear was made, with coughing inducing extravasation of air into the pleural space and the air tracking around into the open pericardial space and LA. This was confirmed by the insertion of a chest drain into the right side and repeating the TTE. The chest drain bubbled when the patient coughed, confirming the presence of an air leak. With the chest drain inserted, no air appeared in the left side of the heart during vigorous coughing. The pleural tear healed, and no further air subsequently entered the left side of the heart after removal of the chest drain.
Because there had been no evidence of myocardial recovery, the LVAD insertion was planned as a bridge to transplantation. The high operative risk precluded inserting a longer-term ventricular assist device or repositioning the LVAD inflow cannula. Because of associated significant acute medical issues (including coagulopathy, renal failure, acute respiratory distress syndrome, sepsis, and focal neurologic deficits caused by cerebral embolism), the patient died 64 days after LVAD insertion.
Discussion
This case demonstrates an unusual complication of LVAD use. Air embolism is a known risk with ventricular assist device use because of potential anatomic (connection of the circulation to the atmosphere via ventricular assist device cannulae or occasionally an open chest) and mechanical (device malfunction) entry points for air entry. However, it is a rare complication, with only 2 case reports of air embolism associated with LVAD use caused by device malfunction or entry of air into the systemic circulation during a pump exchange. After review of the literature, to our knowledge, this is the first case report of systemic air embolization through the LA wall cannula site from an intermittent pleural air leak.
By linking the presence of bubbles in the left side of the heart only with coughing and the localization of bubble entry into the LA around the cannula site on TTE, it was hypothesized that there was an air leak on the surface of the lung, with the subsequent passage of bubbles into the subatmospheric pressure of the LA. In this case, it was assumed that there may have been a mild degree of trauma to the surface of the right lung during insertion of the left-sided heart inflow cannula. A flap acting as a 1-way valve or the pleural effusion may have acted as a barrier to the appearance of a clinically or radiologically apparent pneumothorax. Contributing to the absence of air on radiologic imaging would have been the decompressive effect of the drainage of air into the LA.
Of note, this case also demonstrates that pathology on imaging may be a dynamic rather than a static process. A resting TTE with no coughing would have missed this pathologic finding, as demonstrated by the normal resting chest radiologic imaging. Frequent coughing during the computed tomography scan of the chest may have resulted in the detection of air on the chest imaging.
During respiration, the normal pressure within the pleural space is negative (-5 to -8 mm Hg). However, with coughing or forced expiration, the intrapleural pressure may increase significantly, up to +30 mm Hg. As in this case, flow of a gas (air) in this environment is from an area of high pressure (intrapleural space during coughing) to an area of low pressure (LA, being drained by a large-bore cannula on negative suction at -40 mm Hg). Once there had been a leak of air into the pleural space with coughing, this would have then tracked around the pleural space into the pericardial space (where the pericardium was open) and into the LA around the cuff of the LA inflow cannula.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
