“There is a high base rate of cognitive impairment in adults with temporal lobe epilepsy (TLE)—as high as 80% in some studies,” Robyn Busch, PhD, ABPP, notes. “Recent research has demonstrated substantial variability in the cognitive patterns observed among adults with TLE, even within focal epilepsy syndromes that appear to be relatively homogenous in other respects. TLE is also associated with high rates of depression and anxiety, again with variability across patients.”
However, it is unclear how psychiatric symptoms relate to variations in cognitive function among patients with TLE and whether they difer across cognitive phenotypes, according to Dr. Busch.
For a study published in Epilepsia, Dr. Busch and colleagues examined the relationship between cognitive phenotypes and psychiatric symptomatology in patients with TLE. Enrolled adults with drug-resistant TLE underwent a neuropsychological evaluation that included at least two measures from five cognitive domains to derive International Classification of Cognitive Disorders in Epilepsy (IC-CoDE)-based cognitive phenotypes, which include intact, single domain impairment, bidomain impairment, and generalized impairment. The study team also screened participants for depression and anxiety. The researchers obtained psychiatric history and medication data from EHRs and used logistic regression models to assess the relationship between IC-CoDE phenotypes and psychiatric variables.
Cognitive Phenotype Influenced by Multiple Factors
The study included 826 patients (mean age, 40.3; 55.3% women). Median age at epilepsy onset was 20, and median epilepsy duration was 15 years.
Dr. Busch and colleagues found a significant association between the Beck Depression Inventory— Second Edition (BDI-2) with distribution of cognitive phenotype following adjustment for covariates, such that a higher score on the BDI-2 was associated with more severe cognitive phenotypes. With each 5-point increase in BDI-2 raw score, the odds of a more severe cognitive phenotype increased by 12.3% (OR=1.123; 95% CI=1.052-1.198; Holm-corrected P=0.007). Based on a raw BDI2 score cutof of 11, patients who were depressed had 1.76 times greater odds of a more severe cognitive phenotype compared with patients who were not depressed (OR=1.762; 95% CI, 1.357-2.291; Holm-corrected P<0.001).
Additional factors that increased the odds of a more severe cognitive phenotype included:
- > Number of psychotropic medications (P<0.05)
- > Number of antiseizure medications (P<0.001)
- > Use of antiseizure medications with moodworsening efects (P=0.005)
- > History of a psychiatric diagnosis (P<0.05)
The researchers did not identify a strong relationship between anxiety and cognitive phenotype.
“Our results show that patients with moderate or severe depression according to the BDI-2 had about two times higher odds of having a worse cognitive phenotype compared with patients with minimal depression,” Dr. Busch notes (Figure). “Similarly, patients taking a higher number of psychotropic and antiseizure medications had increasingly worse cognitive phenotypes (ie, more global cognitive impairment).”
Implications for Treatment of TLE
The findings have important implications for treating patients, Dr. Busch continues.
“Specifically, our data suggest that clinicians should consider pharmaceutical load and medication profiles when treating patients with TLE, particularly in those with cognitive difficulties. With respect to antiseizure medications, those known to be associated with mood-worsening efects were strongly related to cognitive phenotype, while those with mood-enhancing qualities were not associated with cognitive profile.”
There are also implications for surgical decisionmaking, given that preoperative cognitive performance is one of the most important factors for predicting postoperative outcome. “Providers must be careful not to overestimate cognitive impairment or misattribute global impairments to brain pathology without considering the potential impact of mood or medication efects,” Dr. Busch says.
She advises clinicians to screen patients with TLE for mood symptoms and examine a range of treatment options for those with comorbid depression, particularly as part of the surgical workup, because these patients may benefit from treatment before surgery.
“Longitudinal studies that incorporate neuroimaging findings are needed to further our understanding of the complex relationships between cognition, mood, and seizures and to determine whether nonpharmacologic treatment of mood symptoms alters cognitive phenotype,” Dr. Busch says. “We also hope to address whether this pattern holds in patients with other epilepsy syndromes and in pediatric samples, as well as across cultures and treatment settings.”
