<b><i>Background:</i></b> Spontaneous intracerebral hemorrhage (ICH) and the evolution of subsequent perihemorrhagic edema lead to midline shift (MLS), which can be assessed by transcranial duplex sonography (TDS). In this observational study, we monitored MLS with TDS in patients with supratentorial ICH up to day 14 after the ictus, and then correlated MLS with the outcome 6 months after hospital discharge. <b><i>Methods:</i></b> Sixty-eight patients with spontaneous ICH (volume >20 cm³) were admitted during a 1-year period between April 2009 and April 2010. Sixty-one patients fulfilled the inclusion criteria and were eligible for analysis. TDS to measure MLS was performed upon admission and then subsequently, using serial examinations in 24-hour intervals up to day 14. Statistical tests were used to determine cut-off values for functional outcome and mortality after 6 months. <b><i>Results:</i></b> The median National Institutes of Health Stroke Scale (NIHSS) score upon admission was 21 and the mean hematoma volume was 52 cm³. NIHSS score, functional outcome, hematoma volume and MLS were correlated in the examined patient cohort. ICH score upon admission, hematoma volume and the extent of MLS on days 1–14 were predictive of functional outcome and death. Values of MLS showed two peaks, the first between day 2 and day 5 and the second between day 12 and day 14, indicating that edema progresses not only during the acute but also during the subacute phase. Depending on the time point, an MLS of 4.5–7.5 mm or greater indicated an impending failure of conservative therapy. An MLS of 12 mm or greater at any time indicated mortality with a sensitivity of 69%, a specificity of 100% and positive and negative predictive values of 100 and 74%, respectively. <b><i>Conclusions:</i></b> MLS seems to be a crucial factor for outcome after ICH. Apart from the hematoma volume itself, edema adds to the intracranial pressure. To monitor MLS in early patient management after ICH, TDS is a useful noninvasive bedside alternative, avoiding increased radiation exposure and repeated transportation of critically ill patients. Cut-off values may help to reliably predict functional outcome and treatment failure in patients undergoing maximal neurointensive therapy.