Background The Rice-Vannucci model of hypoxic-ischaemic encephalopathy (HIE) has been associated with a high degree of variability with respect to the development of cerebral infarction and infarct lesion volume. For this reason, we examined the occurrence of communicational blood flow within the common carotid (CCA), internal (ICA), and external (ECA) carotid arteries following CCA occlusion as a source of variability in the model. New method We propose a novel modification to the Rice-Vannucci model, whereby both the CCA and ECA are permanently ligated; mitigating communicational blood flow. Results Using magnetic resonance angiography, phase-contrast velocity encoding, and pulsed arterial spin labelling, the modified Rice-Vannucci model (CCA/ECA occlusion) was demonstrated to mitigate communicational blood flow, whilst significantly reducing ipsilateral hemispherical cerebral blood flow (CBF). Comparatively, the original Rice-Vannucci model (CCA occlusion) demonstrated anterograde and retrograde blood flow within the ICA and CCA, respectively, with a non-significant reduction in ipsilateral CBF. Furthermore, CCA/ECA occlusion plus hypoxia (8% O2/92% N2; 2.5 h) resulted in 100% of animals presenting with an infarct (vs 87%), significantly larger infarct volume at 48 h (18.5% versus 10.0%; p < 0.01), reduced standard deviation (± 10% versus ± 15%), and significantly worsened functional outcomes when compared to CCA occlusion plus hypoxia. Comparison with existing method We compared a modified Rice-Vannucci model (CCA/ECA occlusion ± hypoxia) to the commonly used Rice-Vannucci model (CCA occlusion ± hypoxia). Conclusion This study demonstrates that CCA/ECA occlusion in the Rice-Vannucci model of HIE reduces infarct volume variability by limiting communicational blood flow.