A 57-yr-old man with no significant medical history was struck by an automobile moving at a high rate of speed, resulting in multiple injuries. According to responders at the scene, he was combative, with a Glasgow coma scale of 7, and was subsequently tracheally intubated and sedated. Upon arrival in the emergency department, his arterial blood pressure was 90 mm Hg. Central venous access was obtained via a subclavian line so that fluid resuscitation could be initiated. His initial trauma evaluation and radiographs revealed a fracture of the left tibia and fibula, multiple rib fractures, a diaphragmatic rupture with herniation of the abdominal contents into the thoracic cavity, and a splenic rupture. A computed tomographic (CT) scan of his head showed a small punctate hemorrhage within the posterior limb of the internal capsule on the right side, a small (8 mm) area of decreased attenuation in the left frontal vertex, also thought to be a hemorrhage, and bilateral frontal fluid collections (Fig. 1A). His neurological examination was notable for spontaneous eye-opening and his ability to move all four extremities, follow simple commands, and answer simple questions by showing one or two fingers. Verbal response could not be assessed in the presence of an endotracheal tube.
Because of the multiple operations and diagnostic procedures required within the first 48 h, the patient remained heavily sedated with propofol and morphine throughout much of this time, with breaks only for neurological examinations. His neurological status during these intervals remained unchanged from his initial examination. A follow-up head CT on hospital Day 2 revealed interval resolution of both hemorrhages. The patient remained tracheally intubated and on pressure support ventilation with pressure support values between 10 and 15 mm Hg.
The patient was weaned off propofol on hospital Day 3 but was found to be difficult to arouse as well as unable to move his left upper or lower extremity. He was sent for a repeat head CT (Fig. 1B) that showed a large infarcted area in the right frontal cortex. A carotid and vertebral artery angiogram obtained at that time revealed an area consistent with embolism in the right middle cerebral artery (MCA) distribution, occluding the superior anterior division of the right MCA but no evidence of dissection.
To help determine the source of the embolus, the patient underwent transesophageal echocardiogram (TEE), which demonstrated a persistent foramen ovale and a small right-to-left shunt at rest (Fig. 2) but no evidence of thrombus. Interestingly, auscultation of the patient’s chest revealed a prominent friction rub as well as a grade II/VI systolic murmur, neither of which were noticed on initial examination. Conversations with the patient’s family and primary care physicians did not reveal a previous diagnosis of an atrial septal defect or patent foramen ovale (PFO), and no mention was made of a murmur having been noted in any of the past 3 annual physical examinations. To complete the workup, lower extremity noninvasive ultrasound studies were performed and demonstrated no evidence of any lower extremity thrombus. Given his orthopedic injuries, there was concern for fat emboli, but a funduscopic examination performed at that time did not reveal any evidence of fatty deposits within the retina.
The patient was discharged from the intensive care unit (ICU) on hospital Day 31 after gradual improvement in his neurological status. He was able to move his left upper and lower extremities, continued to open his eyes to voice, and was able to nod yes and no to simple questions. Repeat head CTs, reviewed with staff neuroradiologists, showed no change in the size of his infarct.
Paradoxical embolism is a well-described entity and refers to an embolism that originates in the venous circulation and travels to the arterial circulation. Such an embolus may be from fat (from long-bone fractures), thrombus (such as deep venous thrombosis or a right atrial thrombus), or air (1). For such an embolism to enter the atrial circulation, there must be communication between the right and left sides of the heart, along with the proper pressure gradient to drive the embolus across. One substrate for such a communication is a PFO. PFO is a relatively common finding, with an incidence that has been estimated to be as frequent as 30%(2). As a result, PFO is a relatively common cause of stroke (3,4).
In the present case, it is most likely that the patient had had the PFO since birth, and the application of positive pressure ventilation along with systemic hypotension helped to create the proper conditions for a right to left shunt (5). It is also possible, although less likely, that the atrial septal defect (ASD) was generated traumatically given the nature of the injury to the patient, an initial cardiac enzyme increase, new electrocardiogram findings of nonspecific ST-elevations, right bundle branch block and right axis deviation, and the question of the new murmur found on hospital Day 3. However, there are only scattered case reports of traumatic ASD in the setting of blunt chest trauma (6–10), most often the result of motor vehicle accidents, and the echocardiogram results seemed more consistent with a congenital PFO.
The origin of embolism in this patient remains somewhat questionable. The early time course of the infarct makes it more likely that the patient suffered from either a fat or an air embolism rather than a thrombotic embolism. This is perhaps even more probable given the absence of any thrombus on both the lower extremity ultrasound study as well as the TEE. The fracture to the patient’s left tibia and fibula make fat embolism a potential etiology for the infarct. Fat embolism as a cause of paradoxical embolism is a rare entity that has been reported in the literature on a case report basis, usually as the result of orthopedic trauma or after surgery after total joint arthroplasty. (An excellent review of the cases of fat embolism can be found in reference (11) and the corresponding letter in reference (12).) Often, cases of fat embolism present as part of the fat embolism syndrome (FES). FES is characterized by the triad of hypoxia, petechiae, and neurological changes, all of which are usually manifest within 12–36 hours after the original injury. It has been estimated that patients with long-bone fractures have a risk as high as 20% of developing FES (11). In this case, the patient obviously had neurological changes but did not have a petechial rash, and his intubated state made it difficult to determine if he would have been hypoxic otherwise. Using the major and minor criteria suggested by Gurd and Wilson (13), our patient would not have qualified for having FES.
Air embolism is another possible, and perhaps the most likely, etiology for the cerebrovascular accident (CVA) in our patient. Air embolism is a well known, if uncommon, complication of central venous catheterization, especially when central venous access is gained in an emergent situation (14,15). Several factors make air embolism more likely during central venous line (CVL) insertion: deep inspiration during insertion, hypovolemia, and upright positioning; some of these may have been present during catheterization of our patient. The incidence of air embolism after CVL insertion has been estimated to range from 1 in 3000 to as frequent as 1 in 47 in various reports (16). In a more controlled setting, it would be possible to insert the CVL under ultrasound guidance, thereby reducing the risk considerably; however, this is not likely to be feasible in the trauma patient given the need for rapid resuscitation. In addition, the rapid infusion of fluid boluses via rapid infusion systems may also be a potential etiology of air embolism in the polytrauma patient because air may be forced into the venous system. However, in the present circumstance, fluid boluses were administered via “wide open” flow from hanging bags of crystalloid, making it an unlikely etiology in our patient.
Two other aspects of this case deserve mention. First, there was a delay in diagnosis of our patient, given his need for continuous sedation for the multiple injuries and surgeries that took place over the first 48 hours of his hospitalization. He was given routine hourly or every-other-hourly neurological status checks throughout the time he was in the ICU, but he did spend a significant amount of time in and traveling to and from various procedures. It is possible that if motor examinations had been performed, even while the patient was sedated either in the procedure itself or while en route, some asymmetry in his motor examination may have been noted earlier. Of course, the decision to delay treatment to perform a motor examination will need to be made on a case-by-case basis, and with an already strong suspicion of neurological injury, it may be impractical, at best, and impossible, at worst, to perform a motor examination on a patient undergoing certain types of procedures.
In summary, we have presented a patient with a paradoxical embolism arising either from a fat or an air embolus. We surmise that our patient had an underlying PFO that was substrate for an embolic CVA in the setting of long bone fracture and CVL insertion in a patient undergoing positive pressure ventilation. The necessity for multiple interventions and diagnostic procedures in the first several hours in cases of polytrauma make routine neurological checks difficult to obtain and neurological status difficult to track. The combination of these factors creates a setting for the “perfect storm” of CVA in the polytrauma patient and should be considered in future trauma situations.
1. Corrin C. Paradoxical embolism. Br Heart J 1991; 26: 549.
2. Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1994; 59: 17–20.
3. Lamy C, Giannesini C, Zuber M, et al. Clinical and imaging findings in cryptogenic stroke patients with and without patent foramen ovale: the PFO-ASA Study. Stroke 2002; 33: 706–11.
4. Perloff JK. Ischemic neurologic complications of congenital heart disease in adults. UpToDate Online, version 10.3. http://www.uptodate.com
5. Lemaire F, Richalet JP, Carlet J, et al. Postoperative hypoxemia due to opening of a patent foramen ovale confirmed by a right atrium-left atrium pressure gradient during mechanical ventilation. Anesthesiology 1982; 57: 233–6.
6. Shapiro SB, Morris SE, Bull DA, Barton RG. Hypoxemia from an atrial septal defect 7 days after blunt thoracic trauma. J Cardiothorac Vasc Anesth 2000; 14: 56–8.
7. End A, Rodler S, Oturanlar D, et al. Elective surgery for blunt cardiac trauma. J Trauma 1994; 37: 798–802.
8. Baumgartel E. Cardiac rupture from blunt trauma with atrial septal defect. Arch Surg 1992; 127: 347–8.
9. Markewitz A, Klinner W. Atrial septum rupture following blunt thoracic injury within the scope of polytrauma. Chirurg 1989; 60: 694–7.
10. Rao G, Garvey J, Gupta M, Wisoff G. Atrial septal defect due to blunt thoracic trauma. J Trauma 1997; 17: 405–6.
11. Mellor A, Soni N. Fat embolism. Anaesthesia 2001; 56: 145–54.
12. Horowitz PE. Fat embolism [letter]. Anaesthesia 2002; 57: 818.
13. Gurd AR, Wilson RI. The fat embolism syndrome. J Bone Joint Surg Br 1974; 56B: 408–16.
14. Doblar DD, Hinkle JC, Fay ML, Condon BF. Air embolism associated with pulmonary artery catheter introducer kit. Anesthesiology 1982; 56: 307–9.
15. Orebaugh SL. Venous air embolism: clinical and experimental considerations. Crit Care Med 1992; 20: 1169–77.