Hemiballism-hemichorea (HC-HB) caused by non-ketotic hyperglycaemia was first reported by Bedwell in 1960. He described a 65 year old woman who developed ballistic movements in all four limbs during three episodes of hyperglycaemia(5). NKH is the most common metabolic cause of HC-HB and is often a reversible condition. HC-HB is a disease with a low prevalence and mainly described in case reports or case series involving individuals of Asian descent, females, and the elderly. The slightly increased incidence in the Asian population may suggest an underlying genetic predisposition(6). Besides an abnormal metabolic panel, it is associated with noticeable changes in neuroimaging. CT and MRI scans of several affected patients show hyperdensities and hyperintense signals respectively on the contralateral striatum(7). A meta-analysis of 53 cases conducted by Oh et al showed that 89% of the cases had hemichorea with contralateral putaminal hyperintensity and 11% had generalised chorea with bilateral putaminal changes on MRI(8). Findings on MRI include unilateral hyperintensity on T1-weighted images without the corresponding hyperintensities on T2-weighted images that characterize basal ganglia haemorrhage(9).
The pathophysiology of NKH-induced chorea is not clearly understood, with several competing hypotheses. The most commonly invoked theory suggests that depletion of gamma-aminobutyric acid (GABA) and acetylcholine, which are needed as an alternative energy source during nonketotic hyperglycaemia, leads to a decreased inhibitory signal to the thalamus resulting in hyperactive movement(1, 10). Another potential mechanism suggested, is acute putaminal dysfunction secondary to hyperglycaemic or hyperosmolar insult, associated with some degree of Wallerian degeneration of the internal white matter of the putamen have been considered to play a pathogenetic role in NKH patients with putaminal MRI abnormalities(11). As such, the rapid reversibility of the neurological signs with correction of the metabolic disturbance suggests a direct causal relation between hyperosmolar hyperglycaemia and the neurological signs.
Following recognition, NKH should be treated with aggressive glucose control, which typically leads to resolution of the hemichorea and imaging findings. Most patients with HC-HB show partial or complete resolution after treatment with neuroleptic drugs, and in most case reports there was complete resolution of symptoms at three months of follow up(11)(12)(6)(13). In the case series reported by Ryan et al, the treatment of six patients with typical or atypical neuroleptics was maintained throughout the follow up period of one month to five years. In four of these cases, mild chorea persisted on treatment(3).
Our patient was discharged on day four of admission with improved symptoms and optimisation of DM management. A review of the electronic records showed that he reported worsening chorea two weeks after discharge. After regular doses of haloperidol, he reported symptom resolution in his lower limb but persistent symptoms in his upper limb. This corroborates with our earlier literature review suggesting that some patients show partial resolution of their symptoms and require prolonged treatment.
In conclusion, although NKH is uncommonly caused by a dysfunction of glucose metabolism, we advise checking blood glucose in patients presenting with hyperkinesia, as the condition may rapidly resolve with hydration and resolution of the hyperglycaemia. It is also advisable to be cognisant that this movement disorder may be the initial presenting symptom of diabetes mellitus(14).
We would like to reiterate the importance of early recognition of this uncommon condition. In our case, thankfully, the mistaken first impression of treating the above patient for an intracranial haemorrhage did not result in adverse outcomes. Prompt diagnosis and appropriate glycaemic control would improve symptoms in such patients, which would in turn minimize discomfort as well as avoid unnecessary investigations and interventions.