Several studies have investigated fatigue in the general population, in primary care facilities as well as in patients with fatigue-related physical diseases, but only marginally in patients with Major Depressive Disorder (MDD). Therefore, the investigation of correlates of depression-related fatigue is highly warranted and expected to facilitate the implementation of effective fatigue-specific treatment strategies. Depressed patients often suffer from comorbid anxiety disorders (CADs) or subthreshold anxiety symptoms. This study aimed to investigate the independent correlation of the severity of fatigue in female patients with MDD with the presence, number and type of CADs. We studied 70 consecutive female MDD patients (48.6% inpatients), aged 23-65 years (mean 48.2±10.6 years), currently in a Major Depressive Episode [17-item Hamilton Depression Rating Scale (HDRS) score≥17] and free of other fatigue-associated conditions. Diagnostic assessments were made with the short structured DSM-IV-based MINI version 5.0.0. Reported fatigue was assessed with the 14-item Chalder Fatigue Questionnaire (FQ). Correlations between the FQ score and age, inpatient status, HDRS score, presence and number of CADs were calculated. Then, stepwise multiple regression analyses were performed, with the FQ score as the dependent variable,so as to isolate independent predictors of the severity of fatigue. 92.9% of patients had clinically significant fatigue. 62.9% were suffering from at least one CAD (38.6% met criteria for one CAD,21.4% for two and 2.9% for three). 51.4% were diagnosed with generalized anxiety disorder (GAD),25.7% with panic disorder and/or agoraphobia (PD/AP), 17.1% with social anxiety disorder and 7.1%with obsessive-compulsive disorder. The FQ score was significantly correlated with the HDRS score(r=0.406, p<0.001), the presence of any CAD(s) (rho=0.4, p=0.001), the number of CADs (rho=0.393,p=0.001), the presence of GAD (rho=0.421, p<0.001) and the presence of PD/AP (rho=0.252, p=0.035).In multiple regression analyses, the presence and number of CADs and the presence of comorbid GAD turned out as significant independent predictors of the FQ score along with the HDRS score.The severity of fatigue in female MDD patients is independently correlated with the presence and number of CADs and, in specific, comorbid GAD. Our findings imply that: (1) this effect might in part account for greater impairment/disability and adverse prognosis for MDD with CADs; (2) high levels of fatigue, putatively clustering with anxiety symptoms, may be a marker of severity and anxiety disorders comorbidity for MDD and may define an "anxious-fatigued" subtype/phenotype in this population; (3) medications and psychotherapies for the management of severe depression-related fatigue should also target CADs.