ABSTRACT
BACKGROUND: Pharmacological treatment of depression is currently led by the trial and error principle mainly because of lack of reliable biomarkers. Earlier findings suggest that baseline alpha power and asymmetry could differentiate between responders and non-responders to specific antidepressants. AIM: The current study investigated quantitative electroencephalographic (QEEG) measures before and early in treatment as potential response predictors to various antidepressants in a naturalistic sample of depressed patients. We were aiming at developing markers for early prediction of treatment response based on different QEEG measures. MATERIALS AND METHODS: EEG data from 25 depressed subjects were acquired at baseline and after one week of treatment. Mean and total alpha powers were calculated at eight electrode sites F3, F4, C3, C4, P3, P4, O1, O2. Response to treatment was defined as 50% decrease in MADRS score at week 4. RESULTS: Mean P3 alpha predicted response with sensitivity and specificity of 80%, positive and negative predictive values of 92.31% and 71.43%, respectively. The combined model of response prediction using mean baseline P3 alpha and mean week 1 C4 alpha values correctly identified 80% of the cases with sensitivity of 84.62%, and specificity of 71.43%. CONCLUSIONS: Simple QEEG measures (alpha power) acquired before initiation of antidepressant treatment could be useful in outcome prediction with an overall accuracy of about 80%. These findings add to the growing body of evidence that alpha power might be developed as a reliable biomarker for the prediction of antidepressant response.
Subject(s)
Antidepressive Agents/therapeutic use , Depressive Disorder, Major/diagnostic imaging , Depressive Disorder, Major/drug therapy , Electroencephalography/methods , Adult , Area Under Curve , Bulgaria , Cohort Studies , Depressive Disorder, Major/diagnosis , Female , Humans , Male , Middle Aged , Pilot Projects , Predictive Value of Tests , ROC Curve , Risk Assessment , Severity of Illness Index , Statistics, Nonparametric , Treatment OutcomeABSTRACT
Continuous positive airway pressure (CPAP) improves autonomic activity in patients with chronic heart failure (CHF) and central sleep apnoea (CSA), but its effect on heart rate variability (HRV) during therapy has not been reported. We hypothesized that CPAP may decrease HRV, despite its beneficial effects on sympathetic overactivation, due to the expected stabilization of breathing. Sixty-seven CHF patients underwent polysomnography (PSG). Ten of them presented with CSA (age 66.1±8.5 years, apnoea-hypopnea index [AHI]=57.6±23.3, central AHI [cAHI]=41.6±24.6 [mean±SD]) and were subjected to a second PSG with manual CPAP titration. Beat-to-beat heart intervals for a 6-hour period of sleep were extracted from each recording and HRV was analysed. CPAP significantly reduced AHI (AHI=23.1±18.3 P=.004). Standard deviation of normal-normal interbeat interval (SDNN) (61.5±29.0 vs 49.5±19.3 ms, P=.021), root mean square of successive differences (RMSSD) (21.8±9.2 vs 16.4±7.1 ms, P=.042), total power (lnTP=7.8±1.1 vs 7.4±0.8 ms2 , P=.037), low frequency power (lnLF=5.5±1.5 vs 5.0±1.4 ms2 , P=.003) and high frequency power (lnHF=4.6±1.0 vs 4.0±1.0 ms2 , P=.024) were decreased. There was a strong correlation between the decrease in AHI and the decrease in lnHF (Spearman's ρ=.782). CPAP leads to a decrease in spectral and time domain parameters of HRV during therapy in CHF patients with CSA. These changes are best explained by the effect which CPAP-influenced breathing pattern and lowered AHI exert on HRV.
Subject(s)
Continuous Positive Airway Pressure/methods , Heart Failure/therapy , Heart Rate/physiology , Sleep Apnea, Central/therapy , Aged , Chronic Disease , Continuous Positive Airway Pressure/trends , Female , Heart Failure/diagnosis , Heart Failure/physiopathology , Humans , Male , Middle Aged , Polysomnography/methods , Polysomnography/trends , Sleep Apnea, Central/diagnosis , Sleep Apnea, Central/physiopathology , Treatment OutcomeABSTRACT
Chronic heart failure (CHF) is a major health problem associated with increased mortality, despite modern treatment options. Central sleep apnea (CSA)/Cheyne-Stokes breathing (CSB) is a common and yet largely under-diagnosed co-morbidity, adding significantly to the poor prognosis in CHF because of a number of acute and chronic effects, including intermittent hypoxia, sympathetic overactivation, disturbed sleep architecture and impaired physical tolerance. It is characterized by repetitive periods of crescendo-decrescendo ventilatory pattern, alternating with central apneas and hypopneas. The pathogenesis of CSA/CSB is based on the concept of loop gain, comprising three major components: controller gain, plant gain and feedback gain. Laboratory polysomnography, being the golden standard for diagnosing sleep-disordered breathing (SDB) at present, is a costly and highly specialized procedure unable to meet the vast diagnostic demand. Unlike obstructive sleep apnea, CSA/CSB has a low clinical profile. Therefore, a reliable predictive system is needed for identifying CHF patients who are most likely to suffer from CSA/CSB, optimizing polysomnography use. The candidate predictors should be standardized, easily accessible and low-priced in order to be applied in daily medical routine. The present review focuses on a pathophysiological approach to the selection of some predictors based on parameters reflecting the etiology, the pathogenesis and the consequences of CSA/CSB in CHF.
