Objective: Virtual autopsy is a non-invasive/minimally invasive method for conducting an autopsy, with the assistance of imaging techniques. We aim to review the benefits of virtual autopsy in detecting pathologies in the paediatric population. Method: The procedure adhered to Institute of Medicine and Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines. Seven databases including MEDLINE and SCOPUS were searched for articles published 2010–2020 worldwide in English. A narrative synthesis of the findings of included studies was carried out to discuss and summarize the results of the review. Results: From 686 studies on paediatric deaths, 23 met selection/quality criteria. Virtual autopsy was better than conventional autopsy in detecting skeletal lesions and bullet trajectory, thus a crucial tool in the investigation of traumatic and firearm deaths. Virtual autopsy was superior to conventional autopsy in identifying the point of bleeding in postoperative deaths and objectively quantifying air/fluid in body cavities. Virtual autopsy was a useful adjunct for detecting pulmonary thrombo-embolism, foreign body aspiration, drowning and metastatic malignancies. The use of non-contrast imaging in investigating natural paediatric deaths did not offer more information than conventional autopsy. Misinterpretation of normal post-mortem changes as pathological findings was another disadvantage of virtual autopsy leading to erroneous conclusions. But accuracy may be improved with contrast enhancement and post-mortem magnetic resonance imaging. Conclusion: Virtual autopsy is a crucial tool in the investigation of traumatic and firearm deaths in the paediatric population. Virtual autopsy will be useful as an adjunct to conventional autopsy in asphyxial deaths, stillbirths and decomposed bodies. Virtual autopsy has limited value in differentiating antemortem and post-mortem changes with the added risk of misinterpretations, therefore should be used with caution in natural deaths.