Anatomical atlases document classical safe corridors for the placement of transosseous fine wires through the calcaneum during circular frame external fixation. During this process, the posterior tibial neurovascular bundle (PTNVB) is placed at risk, though this has not been previously quantified. We describe a cadaveric study to investigate a safe technique for posterolateral to anteromedial fine wire insertion through the body of the calcaneum. A total of 20 embalmed cadaveric lower limbs were divided into two groups. Wires were inserted using two possible insertion points and at varying angles. In Group A, wires were inserted one-third along a line between the point of the heel and the tip of the lateral malleolus while in Group B, wires were inserted halfway along this line. Standard dissection techniques identified the structures at risk and the distance of wires from neurovascular structures was measured. The results from 19 limbs were subject to analysis. In Group A, no wires pierced the PTNVB. Wires were inserted a median 22.3 mm (range 4.7 to 39.6) from the PTNVB; two wires (4%) passed within 5 mm. In Group B, 24 (46%) wires passed within 5 mm of the PTNVB, with 11 wires piercing it. The median distance of wires from the PTNVB was 5.5 mm (range 0 to 30). A Mann-Whitney U test showed that this was significantly closer than in Group A (Hodges-Lehmann shift, 14.06 mm; 95% confidence interval (CI) 10.52 to 16.88; p < 0.0001). In Group B, with an increased angle of insertion there was greater risk to the PTNVB (rs = -0.80; p < 0.01). Insertion of wires using an entry point one-third along a line from the point of the heel to the tip of the lateral malleolus (Group A) appears to be the safer technique. An insertion angle of up to 30° to the coronal plane can be used without significant risk to the PTNVB. Insertion of wires halfway along a line from the point of the heel to the tip of the lateral malleolus (Group B) carried a significantly higher risk of injury to neurovascular structures and, if necessary, an angle of insertion parallel to the coronal plane should be used. Cite this article: Bone Joint J 2018;100-B:1054-9.