Bulent Cuhaci

Unmeasured anions and mortality in the critically ill: The chicken or the egg?

I n this issue of Critical Care Medicine, Dr. Dworschak and colleagues (1) report that brief cardiac arrest, induced for testing implanted cardioverters/defibrillators, was associated with modest elevations of serum concentrations of biochemical central nervous system markers. These were neuron-specific enolase (NSE) and the glial protein S100. NSE is a metalloenzyme that catalyzes the dehydration of 2-phospho-D-glycerate to phosphoenolpyruvate in the glycolytic pathway; the isoenzyme is neuron specific (2). S100 is a calcium-binding protein that is involved in a large number of cellular activities, including protein phosphorylation, cyoskeletal assembly, transcription, and signal transduction; the S100B isoform is found in glial and Schwann cells of the nervous system (3). NSE and S100, when elevated in the serum or cerebrospinal fluid, have been proposed as markers of brain damage. A number of articles report a correlation of central nervous system damage with serum or cerebrospinal fluid concentrations of NSE or S100: marked elevations of serum NSE ( 33 ng/mL) and S100 ( 0.7 g/L), within the first 48 hrs after arrest or after focal ischemia, are usually associated with poor outcomes (4–7). Mild NSE or S100 elevations, however, do not always indicate deficits and are detectable before irreversible injury according to a study of transient focal ischemia in the gerbil (8). There is still controversy about the positive predictive value of S100 and NSE: a careful meta-analysis by Zandbergen et al. (9) revealed that even combining the studies involving NSE, S100, and other biochemical markers was not sufficiently accurate to provide prognostic information for nontreatment of comatose survivors of cardiac arrest. The findings of Dr. Dworschak and colleagues (1) provide evidence of an insult to the central nervous system, even though the elevations of S100 and NSE were mild in all but one patient. Although none of the patients had clinical neurologic abnormalities noted by nurses of physicians, more subtle abnormalities may have been missed. Although it seems that their results reflect a reversible insult to neurons and glia, it is also possible that central nervous system cell subpopulations may have died. What should be done? Cardiologists accept that it is necessary to test implanted cardioverters/defibrillators by causing ventricular tachycardia or fibrillation. (The testing is done in the cardiac catheter laboratory or in the operating room, usually while the patient is under general anesthetic.) More sensitive testing of the nervous system for dysfunction or damage from such brief arrest periods should be undertaken. Magnetic resonance scanning is prohibited because of the implanted cardioverter/defibrillator. Neuropsychologic testing before and after provide the most sensitive index. Electroencephalograms or more complex cognitive-evoked potential testing may also be tried. If these confirm that central nervous system damage occurs, steps to ameliorate ischemic damage should be undertaken. Some guidelines for the duration of arrest before vigorous resuscitative measures are instituted would be advisable. Because the cardiac arrest is artificially induced, there is an excellent opportunity for preemptive neuroprotection. The only strategy for neuroprotection from cardiac arrest that has been established is hypothermia (10, 11). This requires considerable effort and planning and is not without some risk. However, corticosteroids reduced NSE elevations when given 2 hrs before open-heart surgery compared with a control group that did not receive steroids (12). It is unclear whether this is true neuroprotection or just a membrane stabilizing effect preventing enzyme leakage during nonlethal ischemia. Nicardipine, a calcium channel antagonist, reduced cerebrospinal fluid NSE concentrations and infarction volume in a rat model of focal ischemia (13). Combination strategies may also be worth testing. Indeed, the brief cardiac arrest for cardioverter/defibrillator implantation may provide opportunities for further clinical trials in primary neuroprotective strategies. The reader is advised to stay tuned for follow-up studies of more detailed neurologic testing of implanted cardioverter/defibrillator patients. G. Bryan Young, MD, FRCPC Division of Neurology Sunnybrook and Women’s Toronto, Ontario, Canada