Treatment-resistant depression seems to respond better when medication decisions are made by referenced electroencephalogram than by the STAR*D algorithm, according to a 12-week randomized trial.
Among 114 patients randomized to the two treatment methods, those whose medication was guided by referenced-EEG (rEEG) improved significantly more than did those treated by Sequenced Treatment Alternatives to Relieve Depression (STAR*D) guidelines in both primary measures of depression, as well as in most of the secondary measurements, Dr. Charles DeBattista and his colleagues wrote in the January issue of Journal of Psychiatric Research (doi:10.1016/j.jpsychires.2010.05.009).
"These results warrant additional studies to determine the role of rEEG-guided psychopharmacology in the treatment of depression, and perhaps in other psychiatric disorders," wrote Dr. DeBattista of Stanford (Calif.) University and his coauthors. "If these results were confirmed, rEEG-guided pharmacotherapy would represent an easy, relatively inexpensive, predictive, objective office procedure that builds upon clinical judgment to guide antidepressant medication choice."
The rEEG relies on a large database built over an 18-year period. It now contains information on more than 1,800 patients who were followed for an average of 405 days. The database includes the patients’ unmedicated EEGs, along with outcomes of more than 17,000 medical trials. The result, according to Dr. DeBattista and his colleagues, is the availability of 74 EEG biomarkers that "provide a very large collection of outcomes data allowing calculation of statistical correlation between the biomarkers found in a patient’s EEG and the treatment response predictions to many medications, based on patients who had similar EEG biomarkers."
The STAR*D algorithm study (pdf), published in 2006, is a four-step treatment algorithm; most patients start at level 1 with citalopram. If they do not enter remission within 14 weeks, they either switch drugs or add another medication; this pattern can continue for four levels, ending with tranylcypromine or mirtazapine plus extended-release venlafaxine.
The trial by Dr. DeBattista and his colleagues randomized 114 patients with treatment-resistant depression. After undergoing a drug washout period consisting of five consecutive half-life days, patients in the active group were randomized to a treatment regimen based on rEEG, constructed after review of their baseline EEG.
Control patients who had failed only selective serotonin reuptake inhibitors received extended-release venlafaxine; those who had failed two or more classes of antidepressants received a medication regimen from steps 2-4 of STAR*D.
The primary end points were changes in the Quick Inventory of Depressive Symptomatology (pdf) (QIDS-SR16) and the Quality of Life Enjoyment and Satisfaction Questionnaire Short Form (pdf) (Q-LES-Q SF).
Secondary end points were changes in the Clinical Global Impression severity scale (pdf) (CGI-severity), the Clinical Global Impression improvement scale (pdf) (CGI-improvement), the Montgomery-Asberg Depression Rating Scale (MADRS), and the Patient Health Questionnaire-9 (PHQ-9).
The patients’ mean age was 44 years; most (63%) were women and white (66%). Overall, they had failed a mean of four previous central nervous system medications.
After 12 weeks, those in the rEEG group had improved significantly more than the control patients on the QIDS-16 (–7 vs. –4.5) and the Q-LES-Q-SF (18 vs. 9).
The rEEG group also improved significantly more in most of the secondary end points, including the CGI-improvement scores, the CGI-severity scores, and the final number of responders as measured by the MADRS system. However, the authors pointed out, the mean MADRS change from baseline to the study’s end was not significantly greater in the rEEG group than in the control groups.
The PHQ-9 end point also showed significant benefit in favor of rEEG, but in its remission measurement, rEEG was not significantly different from the control group (47% vs. 37%).
In a treatment curve, the authors noted that the groups began to separate as early as 2 weeks after the initiation of therapy, with the rEEG group continuing to improve more than the control group each week.
Treatment-related adverse events occurred in 45% of the rEEG group and 48% of the control groups – not significantly different. The most common in both groups were nausea/vomiting, fatigue, headache, anxiety, and insomnia. None of the adverse events or their rates was different from what could have been expected given the drugs employed, the authors noted.
Even in the absence of final remission superiority, the benefit of rEEG-guided treatment over STAR*D-guided treatment was clear and clinically relevant, the investigators said. "For example, the magnitude of improvement on the QIDS-SR16 response rate was 65% for rEEG compared with only 39% for the controls. Likewise the secondary measures on the Q-LES-Q-SF were also quite consistent in supporting the efficacy of the rEEG-guided therapy."