Search for a command to run...
Despite improvements in our understanding of the pathophysiology of thrombotic thrombocytopenic purpura (TTP), little data exist regarding the long-term sequelae following a diagnosis of TTP. We present the results of a comprehensive evaluation of neurologic injury that included a magnetic resonance imaging (MRI), a neurocognitive testing, and an evaluation of health-related quality of life. Twenty-seven patients with a history of idiopathic TTP functioning normally in their activities of daily living were recruited from existing patient cohorts at both the Ohio State University (n = 12) (Columbus) and the University College London Hospitals (n = 15) (London, UK). Nine of 23 (39%) of the MRI studies were abnormal; 17/27 (63%) patients demonstrated neurocognitive impairment, particularly in tests of visual learning and memory. Health-related quality of life scores were also significantly lower than age- and gender-matched US norms for both the composite mental component score and physical component score. These data suggest that the prevalence of neurologic findings in TTP patients in remission is quite high and is largely undetected by routine clinical evaluations. Further longitudinal study will be required to define the risk for neurologic injury and the long-term prognosis in patients previously diagnosed with TTP. A total of 23 magnetic resonance imaging (MRI) exams were completed with 10 performed at the Ohio State University and 13 performed at the United College London Hospitals. Four patients did not undergo MRI testing due to either contraindications to MRI testing or patient refusal/inability to complete the study. Nine of 23 (39%) of the MRI studies were abnormal; seven of nine subjects with abnormal MRI scans showed mild, small vessel ischemic changes felt to be months to years in age. Two subjects had both large vessel infarctions and microvascular changes. Six of seven subjects with only small vessel ischemic changes had multiple lesions, with the majority being subcortical white matter abnormalities, consistent with lesions that would be expected to be seen in a microvascular thrombotic process. Cognitive function was assessed in all 27 convalescent subjects from both sites. Only 18/27 (67%) subjects were able to successfully complete the Groton Maze Learning Test; therefore, for the remaining nine subjects no score was able to be imputed for this testing domain. Nineteen of the 27 (70%) subjects showed performance <1 standard deviation (SD) below age-matched norms on one of four domains, 14 of 27 (52%) showed performance <2 SD below age-matched norms on one of four domains, and 11 of 27 (41%) showed performance <1 SD below age-matched norms on at least two of the four domains. From these data, 17 of 27 (63%) patients were classified as having cognitive impairment (≥2 domains <1 SD below age-matched controls, or one domain <2 SD below controls). The distribution of performance data for all tests was skewed heavily toward subnormal performance, tests of visual learning, and memory showing the greatest negative skew (see online supplement). SF-36v2 surveys were completed by all 27 subjects. In aggregate, the subjects scored significantly lower than age- and gender-matched US norms for both the mental component score (MCS) and physical component score (PCS). In addition, these data were also compared with more common disease states (anemia, depression, cancer) to provide a relevant comparison. The PCS in thrombotic thrombocytopenic purpura (TTP) patients is quite similar to the compared disease states, with the MCS significantly lower in TTP patients than in patients with anemia and cancer, and on par with patients diagnosed with depression (see online supplement). In the nine patients with abnormal MRIs, there was no significant difference in the presence of MRI abnormalities between the subjects studied within 1 year of their last acute episode of TTP and those greater than 1 year since their last acute episode. Although counterintuitive, there was a significantly higher rate of MRI abnormalities in patients with only one episode of TTP compared with those with greater than one episode of TTP (62 vs. 13%, P = 0.017). With respect to neurocognitive function, there was a significantly higher rate of impairment in patients whose most recent episode of TTP was <1 year compared with those >1 year from their last episode of TTP (73 vs. 31%, P = 0.035). However, there was no significant difference in the rate of cognitive impairment between patients with one versus more than one episode of TTP (38 vs. 55%, P = 0.401). The potential for neurologic injury as a result of an acute episode of TTP is widely known, but it has been recognized only recently that chronic neurocognitive deficits may be present and persist long after recovery from an acute episode of TTP [1]. Using complementary methodologies to evaluate neurologic injury and neurocognitive function, we have attempted to objectively document the prevalence and severity of neurologic abnormalities in a cohort of clinically stable TTP patients. Although these data would be strengthened by the inclusion of a control group, the relatively young median age of this cohort (45 years, range 18–62) and the lack of comorbid conditions in nearly all subjects suggest that there should be a low incidence of neurologic abnormalities independent of their diagnosis of TTP, supporting our hypothesis that these findings presented are likely related to their previous history of TTP. These data presented demonstrate a high rate of silent cognitive impairment (63%) and MRI abnormalities (39%) in a cohort of clinically stable without demonstrable acute neurologic signs or symptoms, suggesting that the true prevalence of neurologic injury in patients with a previous history of TTP is likely to be much greater than what would be predicted by the presence of symptoms alone. Although two subjects had residual neurologic deficits that dated back to their initial presentation, all patients studied were functioning normally, and either gainfully used or functioning normally in their activities of daily living at the time of their studies. None of the 27 patients demonstrated clinically apparent cognitive difficulties. The higher rate of neurocognitive deficits compared with MRI abnormalities suggests that neurocognitive testing may be a more sensitive marker of neurologic sequelae in TTP patients than imaging studies such as MRI. It is also interesting to note that four patients with MRI abnormalities did not demonstrate neurocognitive deficits. Although it is possible that the number or location of the lesions documented by MRI may not have been sufficient to result in demonstrable neurocognitive deficits in these four subjects, it is also possible that the abnormalities seen in these four patients may not be related to their previous acute episodes of TTP. Additional prospective study will be required to better understand the structure–function relationship between imaging abnormalities and functional deficits in patients with a previous history of TTP. In the only previous report in which patients were studied in convalescence from an acute episode of TTP, Fiorani et al. reported normal MRI studies in five TTP patients who recovered after initially presenting with neurologic signs and symptoms [2]. The MRIs performed in our study utilized diffusion-weighted imaging, a sensitive technique that can detect the microvascular lesions commonly seen in TTP patients. Given that abnormalities on diffusion-weighted imaging (DWI) are typically only seen after an acute injury for up to 2 weeks [3-5], the demonstration of new findings by DWI in this study could be evidence for new injury that developed in convalescence and likely not related to their previous acute episodes of TTP. This finding supports the hypothesis that patients with a history of TTP may be at risk for progressive neurologic injury in the absence of clinically overt TTP. The first published data reporting neurocognitive deficits in TTP patients were from the Oklahoma TTP-hemolytic uremic syndrome (HUS) registry who detailed their study of cognitive function in 24 patients with normal physical and mini-mental state examinations and no evidence of active TTP [1]. As a group, the patients performed significantly worse on four of the 11 cognitive domains tested compared with standardized data from normal individuals, with performance in 75% of subjects at least 1 SD below normal in one of these four domains. These data are consistent with the data presented here that demonstrated that 63% of subjects had neurocognitive impairment. This impairment was most frequent in the domains of visual learning, attention, and psychomotor function and suggests that in convalescent TTP patients there exists a persistent impairment in their ability to attend, learn, and make simple decisions quickly. The nature of the cognitive impairments presented in this study are also consistent with those observed by Kennedy et al. who also found that impairments were more frequent in the domains of psychomotor speed, attention/vigilance, and visual learning. These combined data suggest that chronic neurocognitive deficits in patients with a previous history of TTP are far more common than realized previously. The marked abnormalities in the Health-related quality of life (HRQoL) presented in this article are also consistent with previously published data by Lewis et al. [6]. Our cohort shows significant differences in all eight HRQoL domains and the combined physical and MCS compared with race- and gender-adjusted US norms, consistent with the published data from the Oklahoma TTP-HUS registry and suggest that patients with a previous history of TTP have a decreased quality of life. Although these data support the fact that patients with a previous history of TTP are at a markedly increased risk for chronic neurocognitive deficits, it is not clear if these deficits improve over time. In our cohort of patients, there was a significant difference in the prevalence of neurocognitive deficits in patients tested within 1 year of their last episode of TTP compared with those greater than 1 year from their last acute episode, suggesting that there may be improvement over time as patients get further away from their last acute episode of TTP. The data presented by Kennedy et al., however, suggest that abnormalities of cognitive function were not related to the time since their most recent TTP episode. Consistent with the data from the Oklahoma TTP-HUS registry, the risk for neurocognitive deficits in our cohort of patients was also not related to the number of previous TTP episodes. Ultimately, a prospective study with serial evaluations of neurocognitive function at defined time points will be required to definitively determine the risk factors for neurologic injury and the long-term prognosis of these deficits in TTP patients. Additional Supporting Information may be found in the online version of this article. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. Spero R. Cataland*, Marie A. Scully , James Paskavitz , Paul Maruff§, Leslie Witkoff*, Ming Jin¶, Nicholas Uva**, James C. Gilbert**, Haifeng M. Wu¶, * Department of Medicine, Ohio State University, Columbus, Ohio, Department of Haematology, University College London, London, United Kingdom, Perceptive Informatics, Inc., Waltham, Massachusetts, § CogState, Melbourne, Victoria, Australia, ¶ Department of Pathology, Ohio State University, Columbus, ** Archemix Corp., Boston, Massachusetts.