Inflammatory Brain Injury After Cardiopulmonary Bypass: Is It Real?

“For heart patients, … the spookiest, most Wellsian, most noncommonsensical weapon in the modern operative arsenal is the heart-lung machine. In my own efforts to understand the aftereffects of bypass surgery, now that I've lived with them for eight years, the heart-lung machine figures powerfully and ominously …. The heart-lung machine may not really be able to keep you there, but it does keep your body viable, breathing for it and circulating its blood.” “But the you that involved thought and personality, where did that go during the hours when the heart-lung machine was taking care of your basic biological functions? Your brain is not dead, but it has been neutralized, keeping only its own secret register of what is going on …. While the operation is happening you are neither really alive nor truly dead.” “Then there are the aftereffects of this noncommonsensical experience— life after such surgery will feel, for many, only somewhat like life. It may also feel somewhat like death—personality death, at least.” These were the reflections of Larry McMurtry1 in his autobiographical work Walter Benjamin at the Dairy Queen. Clearly, judging by this gripping and articulate example of his prose, Mr. McMurtry seems to have achieved full neurocognitive recovery after his 4-vessel on-pump coronary artery bypass graft (CABG) surgery. But can his description of the intellectual and personality-altering “aftereffects of the heart-lung machine,” in actuality be attributed to extracorporeal circulation? Opinion leaders in the cardiothoracic surgery and anesthesia communities have fed a media frenzy by suggesting that it can be.2 In their excellent review, Hogue et al.3 concluded that 30% to 65% of patients have evidence of cognitive decline 1 month after cardiac surgery. About 1 decade ago, experts reviewing adverse cerebral outcomes after cardiac surgery theorized that modalities modifying the systemic inflammatory response to cardiopulmonary bypass (CPB) might ameliorate both frank and subtle brain injury noted after cardiac surgery.4–6 But now, with nearly 2 decades of research on the subject, the association between CPB-induced inflammation and neurocognitive deficits is less than clear. In this issue of Anesthesia & Analgesia, Jungwirth et al.7 have published a well-controlled study in a rat model that fails to demonstrate a relationship between neurologic injury and the foreign surface area of CPB or donor blood used to prime the CPB pump. Despite an extracorporeal foreign surface area and blood prime volume that would be equivalent to exposing the average adult human to 15 CPB circuits and almost 8 L of blood (or approximately 23 total units of packed red blood cells combined with fresh frozen plasma), rats undergoing CPB with the human neonate-sized oxygenator and pump circuitry did not demonstrate differences in overall cognitive performance or long-term or short-term memory, compared with controls. This result is inconsistent with the theory that the inflammatory response induced by CPB and donor blood exposure is, in part, responsible for adverse neurologic outcomes.3–6 One explanation may be that the rats in the current study7 are young and otherwise healthy, with no chronic disease such as diabetes or atherosclerosis. Additionally, the surgical methods included minimal trauma and the bypass technique used fresh whole blood from heparinized donor rats, rather than stored aged units of packed red blood cells and fresh frozen plasma from a blood bank. Despite these limitations, this article serves as an opportunity to examine current knowledge regarding neurocognitive deterioration in the patient undergoing cardiac surgery, with consideration of the inflammatory response and alternate explanations for these unfortunate postoperative outcomes. Conventional wisdom suggests that neurocognitive decline after CPB is a result of an inflammatory response that is initiated by extracorporeal circulation. However, a review of the literature does not support this concept. For example, in an animal model that included both elderly rats and diabetic rats, de Lange et al.8 found no differences in short-term neurocognitive performance (8–14 days after surgery) in rats undergoing surgery with CPB, compared with those undergoing a sham operation. These investigators noted an increase in cytokine release (interleukin-6) after CPB in diabetic rats, but not in elderly rats. Westaby et al.9 showed, in humans, no association between maximal levels of inflammatory markers (complement C4a and C5b-9) and early or late neurocognitive function after CABG surgery with CPB. Furthermore, Parolari et al.10 demonstrated that postoperative levels of inflammatory markers, including interleukin-6, plasma tumor necrosis factor-α, C-reactive protein, and fibrinogen, differed little in patients undergoing CABG surgery with or without CPB. If the inflammatory response is not the primary driver of neurocognitive injury after CPB, the question must now be asked whether neurocognitive decline in adult cardiac surgical patients is related to the CPB pump in any way. To answer this question, we reviewed studies of both early (0–12 months) and late (1–6 years) neurocognitive outcomes. Neurocognitive outcomes in patients randomized to on-pump versus off-pump surgery have not shown significant differences in the incidence of early cognitive decline.10–17 In fact, neurocognitive outcomes in patients undergoing CABG surgery with CPB are similar to those of patients treated with percutaneous coronary intervention.18,19 Furthermore, the longitudinal neurocognitive performance of patients after CABG with CPB did not differ from that of comparable nonsurgical control subjects with coronary artery disease, at 3 months or 1 year after baseline testing.19 Finally, even healthy volunteers without coronary artery disease demonstrated an incidence of cognitive decline 3 to 4 months20,21 or 1 year12 after the initial assessment that was similar to patients undergoing CABG surgery. Although there is evidence that late cognitive decline does occur after cardiac surgery,22,23 and it has been suggested that it may be, in part, attributable to the use of CPB,22 the etiology for this phenomenon remains unclear. Studies comparing patients randomized to either on-pump or off-pump surgery have not found significant differences in the incidence or degree of late cognitive decline occurring 3 to 5 years after surgery.24,25 In addition, late cognitive decline occurring 5 to 6 years after CABG surgery did not differ in degree from longitudinal cognitive decline observed in patients of similar age either with coronary artery disease26 or without coronary artery disease.27 Perhaps decline in neuropsychological tests with time is related to progression of underlying cardiovascular disease or simply to natural aging.28,29 Thus, the recent literature supports the suggestion of Jungwirth et al.7 in the current issue of Anesthesia & Analgesia that “… other factors beyond CPB and inflammatory responses might contribute to adverse neurocognitive outcomes after cardiac surgery.” The national population is aging. An estimated 80,000,000 American adults have one or more types of cardiovascular disease and, of those, 38,100,000 are estimated to be age 60 years or older.30 In 2006, an estimated 7,095,000 inpatient cardiovascular operations and procedures were performed in the United States: 4.0 million in males and 3.1 million in females.30 More and more frequently, we will care for these patients undergoing cardiovascular procedures. In fact, neurocognitive impairment in many or most patients undergoing cardiac surgery may be preexisting, although subclinical.31,32 Likely, such preexisting impairment is related to combinations of a number of factors, including age,33–35 diabetes,36 cerebrovascular disease,34,35 limited social support,37,38 or depression.39,40 Although we claim no expertise in psychiatry, reading Mr. McMurtry's musings in Walter Benjamin at the Dairy Queen1 leads us to speculate that depression may have played a role in this great American writer's postoperative experience. In conclusion, we tend to agree with van Dijk and Kalkman,29 who propose a return to an approach that relies on overt clinical cognitive changes, rather than on variations in scores on neuropsychological tests. Most likely, we will learn that cognitive outcomes for patients needing cardiac surgery with CPB will depend little on the perfusion technique, but rather on the underlying physical and mental health of the patient.