Sirs: The early criteria for brain death (BD) included the absence of all nervous system functions (i. e. cerebral and spinal death) [1]. However, the demonstration of spinal reflexes in brain-dead patients [2, 5-8, 12, 13] led to revision of the criteria and acceptance of these reflexes as compatible with the diagnosis of BD [10]. Although the first studies noted that the Babinski sign was not observed in BD [7, 8], reviews still state that it may be present [10, 15]. We studied the characteristics of the plantar reflex in BD.
MethodsWe performed a prospective study of 144 patients with the diagnosis of BD [3]. To elicit the plantar reflex, one of the authors (GRdF) applied a tactile stimulus to the lateral plantar surface and transverse arch [4]. Responses were classified as: 1) great toe flexion; 2) the Babinski sign; 3) other components (other than the Babinski sign) of the flexion synergy (e. g. triple flexion reflex); 4) any combination of the above responses; and 5) no movement of both great toes. Since the boundary between the "undulating toe flexion sign" [9] and great toe flexion is often not very clear, both movements were classified as great toe flexion.The characteristics of patients with and without the plantar reflex were compared with Fisher's exact test, the chi-square test, and Student's t test. Multivariate analysis was performed including age (cutoff point 45 years), sex, cause of BD, head trauma, and heart rate (cutoff point 100 beats per minute).
ResultsSeventy-nine (55 %) of the 144 patients retained plantar reflexes (Table): in 61 (42 %), the reflex was classified as either flexion or an undulating toe flexion; in 13 (9 %), one or more components of the flexion synergy other than the Babinski sign were observed; and in 5 (3 %) there was a combination of the above reflexes. No patient had the Babinski sign. Patients with plantar reflexes were older and had a lower heart rate and a lower frequency of head trauma as the cause of BD (25 % vs. 46 %, p = 0.01) than patients without plantar reflexes. After multivariate analysis, head trauma was still inversely associated with the presence of plantar reflexes (OR 0.32, 95 % CI 0.14-0.73, p = 0.007).
DiscussionIn this large series, 55 % of the patients had plantar reflexes but none showed the Babinski sign. In most patients with the flexion response, the movement was slower and of lower amplitude than usual. Although the Babinski sign is one component of the flexion synergy, the Babinski sign was not present even when one or all other components of the flexion synergy were seen. Because BD is comparable to transection at the C1 level [7, 14], we refer to Riddoch's study of reflexes after spinal cord section [11], in which he mentions: that "a very interesting movement of the toes was evoked, consisting of simple flexion of the great toe". This was the most common early reflex in complete spinal section and he did not consider it the normal flexion reflex. van Gijn [14] recorded this reaction in three brain-dead patients and called it "a paradox...
