This study sought to describe expected changes in a mirror-image prone electrocardiogram (ECG) compared with normal supine, including a range of cardiac conditions.Unwell COVID-19 patients are at risk of cardiac complications. Prone ventilation is recommended but poses practical challenges to acquisition of a 12-lead ECG. The effects of prone positioning on the ECG remain unknown.100 patients each underwent 3 ECGs: standard supine front (SF); prone position with precordial leads attached to front (PF); and prone with precordial leads attached to back in a mirror image to front (PB).Prone positioning was associated with QTc prolongation (PF 437 ± 32 ms vs. SF 432 ± 31 ms; p < 0.01; PB 436 ± 34 ms vs. SF 432 ± 31 ms; p = 0.02). In leads V to V on PB ECG, a qR morphology was present in 90% and changes in T-wave polarity in 84%. In patients with anterior ischemia, ST-segment changes in V to V on supine ECG were no longer visible on PB in 100% and replaced by an R-wave in V. Bundle branch block (BBB) remained detectable in 100% on PB, with left BBB appearing as right BBB on PB in 71% and QRS narrowing with qR in V for right BBB. ST-segment/T-wave changes in limb leads and arrhythmia detection were largely unaffected in PB.1As expected, the PB ECG is unreliable for the detection of anterior myocardial injury but remains useful for ST-segment/T-wave abnormalities in limb leads, BBB detection, and rhythm monitoring. The prone ECG is a useful screening tool with diagnostic utility in COVID-19 patients who require prone ventilation.