Abstract Background Major depressive disorder (MDD) is the main cause of disability worldwide, its outcome is poor, and its underlying mechanisms deserve a better understanding. Recently, peripheral acetyl- l -carnitine (ALC) has been shown to be lower in patients with major depressive episodes (MDEs) than in controls. l -Carnitine is involved in mitochondrial function and ALC is its short-chain acetyl-ester. Our first aim was to compare the plasma levels of l -carnitine and ALC, and the l -carnitine/ALC ratio in patients with a current MDE and healthy controls (HCs). Our second aim was to assess their changes after antidepressant treatment. Methods l -Carnitine and ALC levels and the carnitine/ALC ratio were measured in 460 patients with an MDE in a context of MDD and in 893 HCs. Depressed patients were re-assessed after 3 and 6 months of antidepressant treatment for biology and clinical outcome. Results As compared to HC, depressed patients had lower ALC levels ( p < 0.00001), higher l -carnitine levels ( p < 0.00001) and higher l -carnitine/ALC ratios ( p < 0.00001). ALC levels increased [coefficient: 0.18; 95% confidence interval (CI) 0.12–0.24; p < 0.00001], and l -carnitine levels (coefficient: −0.58; 95% CI −0.75 to −0.41; p < 0.00001) and l -carnitine/ALC ratios (coefficient: −0.41; 95% CI −0.47 to −0.34; p < 0.00001), decreased after treatment. These parameters were completely restored after 6 months of antidepressant. Moreover, the baseline l -carnitine/ALC ratio predicted remission after 3 months of treatment (odds ratio = 1.14; 95% CI 1.03–1.27; p = 0.015). Conclusions Our data suggest a decreased mitochondrial metabolism of l -carnitine into ALC during MDE. This decreased mitochondrial metabolism is restored after a 6-month antidepressant treatment. Moreover, the magnitude of mitochondrial dysfunction may predict remission after 3 months of antidepressant treatment. New strategies targeting mitochondria should be explored to improve treatments of MDD.