We read the review by Moore et al. (Reference Moore, Mander, Ames, Carne, Sanders and Watters2012), recently published in this journal, with great interest and compliment the authors for their thorough review of cognitive impairment and its relation to vitamin B12.

The authors identified only four intervention studies in neurological patients presenting with vitamin B12 deficiency. Although vitamin B12 is specifically associated with cerebrovascular pathology (Fassbender et al., Reference Fassbender, Mielke, Bertsch, Nafe, Froschen and Hennerici1999), none of these studies included patients with cerebrovascular pathology in particular. Additionally, the studies did not select for patients who were vitamin B12 deficient. Furthermore, cognition was assessed with the Mini-Mental State Examination only, and no assessments of depression were included (Moore et al., Reference Moore, Mander, Ames, Carne, Sanders and Watters2012). Recently, we found increased levels of fatigue and depression in lacunar stroke patients with vitamin B12 deficiency (Huijts et al., Reference Huijts, Duits, Staals and van Oostenbrugge2012). Possible effects of vitamin B12 supplementation on cognitive function and post-stroke fatigue and depression may significantly improve quality of life. Therefore, we performed a pilot study to determine the effect size of vitamin B12 supplementation on cognition. We included 14 first-ever lacunar stroke patients with a vitamin B12 deficiency presenting at our Neurology Department. Lacunar stroke was defined as an acute stroke syndrome with a small (<20 mm) ischemic lesion on acute brain magnetic resonance imaging in the brain stem, basal ganglia, or internal capsule, compatible with the occlusion of a single perforating small artery. We assessed cognitive function with an extensive neuropsychological test battery as well as fatigue (Checklist Individual Strength) (Vercoulen et al., Reference Vercoulen, Alberts and Bleijenberg1999) and depressive symptoms (Hospital Anxiety and Depression Scale-Depression subscale) (Zigmond and Snaith, Reference Zigmond and Snaith1983) by self-report. Supplementation (1 mg hydroxocobalamine) was administered for at least three months by their general practitioner. Although preliminary, significant differences between baseline assessment and follow-up were found for verbal learning scores (Rey Auditory Verbal Learning Test (Brand and Jolles, Reference Brand and Jolles1985)) (p = 0.008). This is in line with the recently published results of Walker et al. (Reference Walker2012). In addition, we calculated reliable change indices to determine if an individual's performance on a test has changed significantly after the intervention, while taking into account the measure's reliability (Jacobson and Truax, Reference Jacobson and Truax1991). In line with the group analyses we found a significant improvement in verbal learning for 6 out of 14 patients (43%). Overall, these individuals were younger (M = 56.0 vs. M = 65.5 years) and higher educated than those who remained stable (n = 6) or declined (n = 2), though not significant (p = 0.189 and p = 0.068, respectively). No effect was found for vitamin B12 supplementation on fatigue and depressive symptoms. One of the recommendations of Moore et al. (Reference Moore, Mander, Ames, Carne, Sanders and Watters2012) was the need for studies with more robust measures of cognition. We addressed this issue, as well as the recommendation to include depression measures. Additionally, we included patients with cerebrovascular pathology who are at risk of vitamin B12 deficiency (Fassbender et al., Reference Fassbender, Mielke, Bertsch, Nafe, Froschen and Hennerici1999) and have not been studied before in supplementation studies. One major limitation however is the small sample size. To provide future studies with a guideline for a minimal number of patients for a randomized controlled trial, we performed sample size estimations (Erdfelder et al., Reference Erdfelder, Faul and Buchner1996) based on our database. Expected differences between treatment and control group were set at 25%. Taking into account a drop-out rate of 20%, we estimated the minimal sample size to test a difference in memory function after supplementation to be 1,008 (504 per treatment arm). In order to detect a difference in fatigue and depression, 4,388 and 2,178 patients are needed, respectively. Assuming that 33% of lacunar stroke patients are vitamin B12 deficient (Huijts et al., Reference Huijts, Duits, Staals and van Oostenbrugge2012), 3,024, 13,164, and 6,534 lacunar stroke patients, respectively, will have to be screened in order to include these numbers. Considering these large sample sizes, future studies need a multi-center setting.