Abstract
Psychiatric Lifetime Diagnoses are Associated with a Reduced Chance of Seizure Freedom after Temporal Lobe Surgery
Koch-Stoecker SC, Bien CG, Schulz R, May TW. Epilepsia 2017;58:983–993.
OBJECTIVE: To examine whether psychiatric comorbidity is a predictor of long-term seizure outcome following temporal lobe epilepsy surgery. METHODS: A sample of 434 adult patients who received temporal lobe resection to treat epilepsy between 1991 and 2009 and were psychiatrically assessed before surgery were followed for 2 years to assess seizure outcome. Stepwise multivariate logistic regression analyses were used to assess the impact of psychiatric variables on complete seizure freedom (Engel class IA), and freedom from disabling seizures (Engel class I). Lifetime histories of three psychiatric syndromes (PS: psychosis; depression; other) and five personality disorders (PD: DSM-IV Clusters A, B, and C; organic personality disorder; other) were considered as predictors, complemented by age at onset, duration of epilepsy, type of lesion (mesiotemporal sclerosis vs. other), and year of surgery. RESULTS: Seizure-freedom rates were significantly higher (p < 0.001) in patients with no history of PS or PD (N = 138; Engel class IA: 61.6%; Engel class I: 87.7%) than in those with any PS or PD (N = 296; Engel class IA: 39.5%; Engel class I: 58.8%). Particularly low seizure-freedom rates were found in patients with a diagnosis of psychosis (N = 32, Engel class IA: 21.9%; Engel class I: 40.6%), organic PD (N = 48, Engel class IA: 25.0%; Engel class I: 35.4%) or a double diagnosis of PS plus PD (N = 97; Engel class IA: 27.8%; Engel class I: 45.5%). No other variables emerged as significant risk factors in multivariate logistic regression analyses. SIGNIFICANCE: Patients with and without psychiatric comorbidities can benefit from temporal lobe epilepsy surgery; however, psychiatric comorbidities are negatively associated with postoperative seizure-freedom rates. Surgical outcome is related to the type and extent of preoperative psychiatric morbidity, which underscores the prognostic value of presurgical psychiatric evaluation. The data support the argument that there are common pathogenetic mechanisms underlying both epilepsy and psychiatric conditions.
Commentary
Have we reached a new age of psychosurgery? Unlikely, because at this point, surgically treating psychiatric illness is unsuccessful in all but a few clinical syndromes and unique circumstances. But we may have reached a new age of psychiatry and surgery, at least in understanding how psychiatric conditions complicate surgery. Although psychiatry and surgery mentioned in the same breath brings forth unpleasant recollections of flawed science and lack of reason, their mention today may be an intuitive response after considering the results of Koch-Stoecker et al. To be clear, psychosurgery was in no way the focus of this paper or this study. The impact of this paper is in vividly demonstrating the overlap between psychiatric illness and epilepsy in terms of surgical outcomes.
It is tempting, and perhaps necessary, to briefly revisit the history of psychiatry and surgery. In 1949, the neurologist Egas Moniz won the Nobel Prize for performing leucotomies. The outcomes were initially intriguing and yielded improvement in cases where there was little perceived hope for positive prognoses. Eventually the procedures were discredited because of the lack of psychiatric specificity and the imprecision of surgical techniques. Lack of documented baseline function may have also hampered assessment of the utility of the procedure, and surely prevented appreciation of nuanced outcomes, positive or negative, that may have been ascertained.
Surgery for temporal lobe epilepsy was evolving in roughly the same time frame, the 1960s and 1970s. In many ways, that history parallels the evolution of psychiatric neurosurgery, particularly with the presence of controversy. Contemplating brain surgery for people with epilepsy who had aggressive outbursts involved a serious ethical dilemma a half-century ago. A profound debate about how much epilepsy surgery could be construed as psychiatric neurosurgery, and vice versa was agonizingly considered by Mark and Neville in their JAMA editorial in 1973 (1).
The debate continues today, though the intersections of epilepsy surgery and psychiatric impact are better understood. Koch-Stoecker et al. present data from 434 patients who underwent temporal lobectomies over a 17-year period. Each patient had seizure localization in the temporal lobe, though had varying pathology ranging from mesial temporal sclerosis (MTS) or tumors, to nonlesional. Each patient received a psychiatric evaluation, which included a semistructured interview procedure that yielded “official” psychiatric diagnoses according to the Diagnostic and Statistical Manual of Mental Disorders (DSM). This in itself is remarkable given that psychiatric evaluation prior to epilepsy surgery is still uncommon in many surgery centers. Psychiatric syndromes (depression, psychosis, other) or personality disorders (clusters A, B, C, or organic) were not precisely specified, but instead were aggregated to account for changes in categoric criteria from differing DSM versions. Each patient received a follow-up evaluation 2 years later when Engel outcomes were established.
The results were profound. Surgical outcomes for epilepsy depended heavily upon the presence or absence of psychiatric syndromes and/or personality disorders identified prior to surgery. Other factors such as age of onset, duration of epilepsy, side of surgery, or presence of MTS were not associated with postsurgical seizure freedom. The best outcomes by far were in those with no history of psychiatric syndrome or personality disorder. The presence of both psychiatric syndrome and personality disorder was associated with a markedly worse outcome. Psychosis itself was a particularly strong predictor of a poor outcome for epilepsy. The supposition is clear: psychiatric illness may have a powerfully negative impact upon the efficacy of epilepsy surgery.
Ultimately, the blossoming paradigm of treating psychiatric illness in order to treat epilepsy appears to be reinforced. And with this data from Koch-Stoecker and colleagues, the paradigm extends to surgical treatment as well. What if psychiatric illnesses were identified and treated prior to the surgical procedures, or even within the 2-year follow-up period after surgery? Would the epilepsy outcome be different? We cannot answer these questions from this study, but clearly the psychiatric conditions matter. So it is curious to find that a recently published practice guideline outlining requirements for high-level epilepsy surgery centers did not insist upon inclusion of a psychiatrist on the treatment team. The only reference to psychiatric evaluation was that availability of “consultative expertise in multiple fields” was necessary (2). A more recent paper tracking surgical procedures performed and personnel involved in surgery centers did not even measure the presence of psychiatrists on treatment teams (3).
To be fair, questions still remain regarding the true impact of psychiatric illness upon epilepsy surgery outcomes. As the authors aptly point out, psychosocial stressors may be worse in patients with psychiatric illness. In theory, a relative lack of social support and potentially inconsistent treatment adherence may also contribute to a worse outcome. Even with the authors’ well-reasoned consideration of common pathogenetic mechanisms for epilepsy and psychiatric conditions, it is unclear if surgical procedures may actually have improved the psychiatric status in some cases. Follow-up psychiatric evaluations were not done at the 2-year mark, and literature reporting potential emergence of de novo psychiatric illness after surgery cannot be ignored (4, 5)
Although the data and analyses are robust, the fact that 358/792 eligible patients were not assessed is also notable. Even in this sophisticated center with an embedded psychiatrist, 45% of patients did not receive psychiatric baseline assessments. The reason was that the group only had one psychiatrist, who obviously could not be available 100% of the time. That perhaps is the most important lesson of this study, that psychiatric evaluation prior to epilepsy surgery is no longer optional. It is essential, and this point has been made repeatedly (6, 7). One can only hope that center directors and influential resource allocators will take heed. In terms of understanding and improving outcomes for persons with medically refractory epilepsy, psychiatry and surgery belong together, intuitively.