Does stroke or silent infarct affect quality of life in adults with sickle cell disease?: A multi-centre study

Abstract Introduction: Stroke is a devastating complication of sickle cell disorders (SCD) causing major morbidity and mortality. Few data exist describing the impact of stroke and silent stroke on quality of life in SCD.The Sickle Natural History Study is a real-world multi-site study of adults with SCD in the UK collecting all standard-of-care clinical data. Historically, MRI/MRA brain was offered to all patients with neurological indications and to all HbSS patients in the last five years. MR imaging allows identification of those with “silent infarcts”, defined as ischaemic changes seen on an MRI scan in the absence of clinical stroke symptoms. Our study uses health-related QoL metrics to look at the link between self-reported QoL measures and stroke/silent infarct. Methods: Patients provided written consent to be included in this study. Only patients with a reported MRI scan and QoL scores were included in the analysis. MRI reports and clinical history allowed us to categorise patients into “stroke”, “silent stroke”, “normal”. Six QoL metrics were used to assess across six domains (ASCQ-Me pain impact, emotional impact, social impact, sleep impact, and PROMIS cognitive function and fatigue). Scores are based on an average score of 50 in a reference population, with higher scores representing better QoL (apart from PROMIS fatigue scores, where lower scores indicate better QoL). Univariate ANCOVA was used to assess the impact of MRI group on QoL metric scores while adjusting for confounders (age, sex, sickle genotype and transfusion therapy during follow up period). Adjusted means are used. Results: 317 had both an MRI scan and QoL scores to analyse. 194 patients (61.2%) were female. 225 patients (71%) had HbSS, 80 (25.2%) HbSC disease, 7 (2.2%) HbS/B+ thalassaemia, 3 (1%) HbS/B0 thalassaemia, 1 (0.3%) HbS/Lepore and 1 (0.3%) HbS/D Punjab. Of the patients included, 254 (80.1%) had a normal MRI, 27 (8.5%) had silent infarct and 36 (11.4%) had a history of stroke. After adjusting for confounders, there was no statistically significant difference in any QoL metric score between normal, silent and stroke groups. This includes mean cognitive function scores (normal 46.5, silent 48.2, stroke 47.3, p=0.623). When grouping patients with stroke or silent infarct together, this 'abnormal’ group reported on average less negative social impact compared with those with a normal MRI (normal group 48.3, abnormal group 52.0, p=0.045). They also reported, on average, slightly less impact of pain on quality of life than the normal group, which was of borderline statistical significance (normal group 50.3, abnormal group 53.2, p = 0.05). Other QoL measures were not found to have statistically significant differences between groups. Conclusion: We found no association between the presence or absence of silent infarcts or symptomatic stroke and self-reported QoL measures in our patient population, including self-reported cognitive function, after adjusting for age, sex, sickle genotype and transfusion therapy. We have demonstrated that patients with silent infarcts or symptomatic stroke do not report significantly different QoL measures when compared to those with a normal MRI scan in our patient population. We also found that patients with an abnormal MRI scan tended to report slightly better social functioning and less impact of pain than those with normal MRI scans, adjusting for the same factors as above; these findings are likely to be chance and notably all groups mean scores fall well within the population average scores for those scales. In our study population we had relatively small proportions of patients in the silent/stroke groups (19.9%). As a real-world study, missing data are likely not to be randomly missing but may be skewed towards those with significant cognitive impact or clinically disabling strokes. Conversely, those with a milder phenotype may be less likely to have had a surveillance MRI scan. Future longitudinal analysis will allow us to investigate whether the presence of stroke or silent infarct vs. normal imaging is associated with a decline in self-reported cognitive function over time.