Use this URL to cite or link to this record in EThOS:
Title: Malarial retinopathy and neurovascular injury in paediatric cerebral malaria
Author: MacCormick, I.
ISNI:       0000 0004 6060 0294
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Background Diseases of the brain are difficult to study because this organ is relatively inaccessible. Only one part of the central nervous system is available to direct, non-invasive observation – the retina. The concept of the retina as a window to the brain has created much interest in the retina as a source of potential markers of brain disease. Paediatric cerebral malaria is a severe neurological complication of infection with the parasite Plasmodium falciparum, which is responsible for death and disability in a significant number of children in sub-Saharan Africa. As with many neurological diseases, the precise mechanisms by which this infection causes damage to the brain remain unclear, and this hampers efforts to develop effective treatments. It may be that studying the retina in paediatric cerebral malaria could both illuminate pathogenesis specific to this disease, and also provide an illustration of how to approach retinal biomarkers in a new, and potentially more effective way. Methods I approached the aim of developing retinal features as markers of brain disease in paediatric cerebral malaria via several objectives. I made use of an existing clinical study to collect new retinal data from ophthalmoscopic examinations and fundus fluorescein angiograms from patients over three successive malaria seasons in Malawi, and added these to historical data obtained previously at the same site. I devised a new method for grading retinal images. I reviewed the biological plausibility of associations between retina and brain in cerebral malaria, and then considered analytical methods to interpret my retinal data effectively. Finally I estimated associations between retinal features, outcomes, and a radiological measure of brain swelling using combinations of regression models. Results My review of retinal and cerebral histopathology, vascular anatomy and physiology indicated that certain retinal and brain regions may be similarly prone to damage from sequestration as a result of interactions between aberrant rheology and microvascular geometry, such as branching patterns and arteriole to venule ratios. My review of evaluations of analogy and surrogacy suggested that biological similarities between retina and brain could be used to justify statistical evaluation of the amount of information the subject and object of the inference share about a common outcome, as used to assess surrogate end points for clinical trials. This kind of approach is able to address questions about whether a particular retinal feature is effectively equivalent to an analogous disease manifestation in the brain. I report analyses on three overlapping groups of subjects, all of whom had retinopathy positive cerebral malaria: children with admission ophthalmoscopy (n=817), children with admission fluorescein angiography (n=260), and children with admission angiography and MRI of the brain (n=134). Several retinal features are associated with death and longer time to recover consciousness in paediatric cerebral malaria. Broadly speaking, these features appear to reflect two processes: neurovascular sequestration (e.g. orange vessel discolouration and death), and neurovascular leakage (e.g. >5 sites of punctate leak and death). Respective adjusted odds ratios and 95% confidence intervals for these particular associations are: 2.88 (1.64-5.05); and 6.90 (1.52-31.3). Other related processes may also be important, such as ischaemia, which can be extensive. Associations between retina and brain are less clear, in part because of selection bias in the samples. Conclusions Neurovascular leak is important in fatal paediatric cerebral malaria, suggesting that fatal brain swelling may occur primarily as a result of vasogenic oedema. Other processes are also likely to be involved, particularly neurovascular sequestration, which is visible on retinal imaging as orange vessels or intravascular filling defects. Sequestration may plausibly cause leak through direct damage to tight junctions and by increasing transmural pressure secondary to venous congestion. Several types of retinal leakage are seen and some of these may represent re-perfusion rather than acute injury. Future work to investigate temporal changes in retinal signs may find clearer associations with radiological and clinical outcomes. The steps taken to evaluate retinal markers in cerebral malaria illustrate a more rigorous approach to retinal biomarkers in general, which can be applied to other neurological diseases.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available