We read the response of colleagues Sakthiswary & Das (Reference Sakthiswary and Das2011) to our article ‘Location and progression of cerebral small-vessel disease and atrophy, and depressive symptom profiles: the Second Manifestations of ARTerial disease (SMART)-Medea study’ (Grool et al. Reference Grool, van der Graaf, Mali, Witkamp, Vincken and Geerlings2011b) with great interest. We regret if Sakthiswary & Das (Reference Sakthiswary and Das2011) found the discussion of our methodological section incomplete, and would hereby like to react to their comments.

First, in our study we adjusted our analyses of brain changes with depressive symptoms for the most common potential confounders, including age, sex, education, Mini-Mental State Examination (MMSE) score, life-style and vascular risk factors, physical functioning and antidepressant use. In the SMART-Medea study, blood investigations included renal and thyroid function tests, whereas full blood count was only available for part of the study population. Since the effect estimates and significance levels did not change after we additionally adjusted for renal and thyroid function tests, we do not expect that uraemia, hypothyroidism or hyperthyroidism accounted for the observed relations between brain changes and depressive symptoms in our study.

Second, subcortical infarcts on magnetic resonance imaging (MRI) were divided into lacunar infarcts (3–15 mm) and large subcortical infarcts (>15 mm). Although a grey area of uncertainty around the cut-off size of 15 mm for lacunar infarcts exists (Lodder, Reference Lodder2007), it is frequently used in the literature. Also, it has been recently argued that counting cavities >15 mm as lacunes should be done with caution, since many acute lacunar lesions will decrease in size over time (Potter et al. Reference Potter, Marlborough and Wardlaw2011).

Third, the Patient Health Questionnaire-9 (PHQ-9) is a brief and commonly used measure for assessing the severity of depressive symptoms, and has been well-validated in the general population (Kroenke et al. Reference Kroenke, Spitzer and Williams2001) and in patients with coronary heart disease (Thombs et al. Reference Thombs, Ziegelstein and Whooley2008). We therefore do not share the opinion that a method asking the respondent about the number of days that symptoms are present would necessarily be preferable. Further, our population had, on average, normal age-adjusted cognitive performance and few subjects in our sample scored below the normal range of global cognitive function; also, crude scores for immediate and delayed recall, mental flexibility, attention, processing speed and intelligence were previously shown to be comparable with age- and education-adjusted scores in another study that included independently living men (Muller et al. Reference Muller, Grobbee, Aleman, Bots and van der Schouw2007; Grool et al. Reference Grool, van der Graaf, Mali and Geerlings2011a). We therefore do not expect that the use of a self-report measure such as the PHQ-9 to assess subjective symptoms of anhedonia and concentration problems in this population will have led to unreliable information.