Leigh syndrome: historical perspective and clinical variations.

Abstract

Leigh syndrome (LS) has emerged as the quintessential neuropathological signature of a mitochondrial encephalopathy. Did D. Leigh ever imagine this possibility in 1951 when he described the neuropathology of a young infant who died after 6 weeks of devastating neurological symptoms [9]? It is unlikely. But to his credit he recognized the histopathological overlap between his case and Wernicke encephalopathy; and he speculated about the role of an altered thiamine-dependent process as the basis for the similarity. Not far off the mark as we would learn over the next several decades. Leigh’s case showed sparing of the mamillary bodies and the putamen. The former observation has remained true but the latter appears now to be an exception. The difficulty in defining LS as a clinical entity followed subsequent observations by noted neuropathologists who reported the histopathology of LS in all age groups with a diversity of clinical prodromes and phenotypes. This debate was perhaps best epitomized by Crosby and Chou [2] when they described ragged-red fibers (RRF) in LS. At the time, this report engendered a vigorous debate as to whether the patient had LS or Kearns–Sayre syndrome (KSS). The current view of this historic controversy is the fact that both LS and KSS are mitochondrial diseases. RRF are rarely seen in LS, and are the rule in KSS. Montpetit et al. [11] described the clinical findings in a series of LS cases and concluded that the condition was mainly a subcortical brainstem syndrome. For the reasons stated earlier, the disease phenotype was not clearly defined. Insights into the biochemical basis for LS emerged in the 1960s. Worsley et al. [13] described two siblings with lactic acidosis. At autopsy, LS was confirmed. This report also represented the first description of familial lactic acidosis. In 1968, Hommes et al. described pyruvate carboxylase (PC) deficiency associated with LS. This association later was disputed by several authors. On the other hand, LS has been described with biotinidase deficiency, a form of multiple carboxylase deficiency. Willems et al. [12] described a young child with cytochrome c oxidase (COX) deficiency and LS, viewed by many at that time as an implausible correlate. In 1979, De Vivo et al. described an infant with LS and defective activation of the pyruvate dehydrogenase complex [3]. It was presumed that the basic defect involved the pyruvate dehydrogenase-specific phosphatase. Two other siblings, a male and female, also had LS suggesting autosomal recessive inheritance. The importance of COX deficiency in LS was emphasized by DiMauro et al. in 1987 when five patients were described with generalized partial COX deficiency and LS [4]. Numerous reports have since appeared linking COX deficiency and LS (Table 1). In 1990, Fujii et al. [7] described two children with complex I deficiency and LS, and Morris et al. [10] emphasized the deficiency of complex I as a common correlate to LS. In 1990, Holt et al. described a new mitochondrial disease associated with heteroplasmy and maternal inheritance [8]. This condition later was referred to as the NARP syndrome (neuropathy, ataxia and