A rare case of HIV encephalopathy presenting with an isolated cerebellar syndrome

Case presentation

We present the case of a 44-year-old gentleman diagnosed with HIV-1 in 2008. He transferred to our care in 2013 on tenofovir disproxil fumarate/emtricitabine, darunavir and ritonavir. Unfortunately, he disengaged from care in 2013. He re-presented to hospital in November 2018 with Pneumocystis jiroveci Pneumonia with a CD4 count of 15 cells/mcl (2%) and HIV viral load of 517,000 copies/ml and was restarted on tenofovir disproxil fumarate/emtricitabine and raltegravir. By October 2019 he was virally suppressed and later switched to Symtuza© (darunavir/cobicistat/tenofovir alafenamide fumarate/emtricitabine fixed dose combination) for simplification.

In March 2020, he remained virally supressed (viral load < 50 copies/ml) and had good adherence to his antiretroviral therapy (ART). However, three weeks later he was admitted to hospital with a 1-week history of rapidly progressive unsteadiness of gait, preceded by a two-week prodrome of headache, tinnitus, bilateral tremor and visual disturbance. On examination, he had signs of an isolated cerebellar syndrome, with fully preserved conscious level, bilateral gaze-evoked nystagmus, and profound truncal and appendicular ataxia of all four limbs. Baseline blood tests showed normal inflammatory markers: white cell count (WCC) 3.8 × 10⁹/L (normal range: 4–11) and CRP < 1 (normal range: <5). HIV viral load was 243 copies/ml and CD4 count 147 cells/mcl (12%). MRI brain revealed abnormal expansion and T2/FLAIR signal abnormality bilaterally in the middle cerebellar peduncles, in addition to numerous white matter foci bilaterally in the cerebral hemispheres, with one focus in the left cerebellar white matter. Cerebrospinal fluid (CSF) analysis showed a red cell count 4 × 10⁶/L, WCC 10 × 10⁶/L, protein 0.46 g/L (0.05–0.45), glucose 3.8 mmol/L (serum 5.4 mmol/L). CSF culture showed no growth. There was no evidence of opportunistic infections from CSF sampling – negative results were obtained for JC virus, BK virus, toxoplasma, herpes simplex virus, varicella zoster virus (VZV), Epstein-Barr virus (EBV), cryptococcal antigen, cytomegalovirus, TB culture and 16S ribosomal RNA PCR. The only positive CSF result was a raised HIV viral load of 1160 copies/ml, which was disproportionately high compared to the plasma viral load. Unfortunately, genotypic resistance testing failed repeatedly on serum and CSF viral load samples. A diagnosis of HIV encephalopathy was made and oral zidovudine added to his Symtuza©. Of note, prior to the CSF viral load result, neurology colleagues advised commencing intravenous methylprednisolone due to diagnostic uncertainty and progression of neurological symptoms. He received just one dose of methylprednisolone before the CSF viral load result was obtained and intensification of his zidovudine regimen favoured instead. Two weeks after the addition of zidovudine, his neurological symptoms had markedly improved, he was mobile and self-caring. He was discharged home on Symtuza© and oral zidovudine and with community neurorehabilitation.

After four weeks, zidovudine was stopped due to side-effects and his antiretroviral regimen switched to Biktarvy© (bictegravir/tenofovir alafenamide fumarate/emtricitabine fixed dose combination) and Rezolsta© darunavir/cobicistat fixed dose combination as it was a low pill-burden regimen, robust and with less potential for CNS side-effects. Three-months later, a repeat MRI brain showed complete resolution of the cerebellar peduncle changes, albeit with persistence of the multiple foci high signal abnormalities in the rest of the cerebral white matter. By October 2020, the only abnormal neurological sign on examination was a mild intention tremor of both upper limbs. In November 2020, a repeat LP showed normal CSF constituents and negative HIV PCR.

Discussion

Cerebellar syndrome is an important manifestation of HIV, particularly in the pre-ART era. Cerebellar dysfunction in HIV is typically attributed to opportunistic infections such as toxoplasmosis, VZV and CMV, but is also identified in progressive multifocal leukoencephalopathy, autoimmune disease and primary CNS lymphoma.^1,2 There are cases of cerebellar atrophy in the absence of one of these identifiable causes and could be related to the direct neurotoxic effects of the HIV, although CSF viral loads were not documented.^2–6

Our patient had no evidence of opportunistic infections clinically, radiologically or on CSF analysis. CSF PCR for JC virus was negative, and although a brain biopsy was not performed to definitively exclude JC-virus associated granule cell neuronopathy, the subsequent rapid clinical recovery was not in keeping with this pathology.

In the setting of a disproportionately high CSF compared to plasma HIV viral load, reasonable exclusion of other pathologies, and suggestive MRI abnormalities, in conjunction with a very good clinical response to intensification of ART regimen to optimise CSF penetration, we conclude that this case represents HIV encephalopathy manifesting clinically as an isolated cerebellar syndrome.

HIV encephalopathy remains an important complication of HIV, even in patients who are virally suppressed. ⁷ It is recognised as a HIV-associated neurocognitive disorder and classically presents with defeclts in attention, memory and executive functioning. HIV viral activation of macrophages, which cross the blood-brain barrier, causes the release of pro-inflammatory cytokines and chemokines within the brain. This, coupled with HIV viral proteins, cause neuronal damage and volume loss.^8,9 Damage is commonly recognised in subcortical structures, however, it has been demonstrated in the cerebellum.^10–13 Interestingly, damage to the cerebellum can be an early feature of HIV encephalopathy. ⁸ Additionally, ocular motor disorders can highlight early HIV encephalopathy in the form of cerebellar involvement. ¹⁴ Importantly, damage to the cerebellum may not be reversible, unless early treatment using ART with good CNS penetration are initiated. ⁸ However, it is prudent to note that there are conflicting data regarding the overall CNS benefits of individual antiretrovirals. Whilst the clinical penetration score (CPE) has been developed to rank antiretroviral penetration into the CNS, the British HIV Association (BHIVA) recommends that the CPE score should not influence choice of agent for patients with neurocognitive disorder.^15,16

Conclusion

Cerebellar syndrome in people living with HIV should trigger a thorough diagnostic work-up. We recommend that clinicians be mindful of HIV encephalopathy as a cause of cerebellar dysfunction, as prompt initiation of ART with good CNS penetration can reverse symptoms.