Several recent reports have described the minimally invasive plate osteosynthesis (MIPO) technique in the treatment of humeral shaft fractures by the anterior approach. The purpose of this study is to identify the danger zone for locking screw placement to avoid musculocutaneous nerve injury in the anterior compartment and radial nerve injury in the posterior compartment of the humerus relative to the humeral length. Eighteen arms of fresh cadavers were fixed with 10-hole locking compression plate (LCP) by anterior approach using the MIPO technique. Two locking screws on each end were fixed by the open technique; the rest of the screws were inserted percutaneously. The arms were dissected both anterior and posterior to identify musculocutaneous and radial nerve injuries. Humeral length with a simple palpable bony landmark was measured from the posterior tip of the acromion process to the lateral epicondyle. Damage or direct contact of the locking screws to the musculocutaneous or radial nerve was recorded, and the distance between the screws and the radial nerve was measured. The average humeral length was 29.71cm (99% confidence interval (CI): 28.54-30.86cm). The danger zone for the musculocutaneous nerve averaged 18.37% (99% CI: 17.06-19.60) to 42.67% (99% CI: 42.33-43.03) of the humeral length from the lateral epicondyle. The danger zone for the radial nerve averaged 36.35% (99% CI: 35.81-37.07) to 59.20% (99% CI: 59.00-59.46) of the humeral length, and the most dangerous screws that penetrated or touched the radial nerve lay 47.22% (99% CI: 45.27-49.17) to 53.21% (99% CI: 51.16-55.33) of the humeral length from the lateral epicondyle. An anteroposterior locking screw placed percutaneously endangered the musculocutaneous and radial nerves. From this cadaveric study, the danger zone for the musculocutaneous and radial nerves could be determined as a percentage of the humeral length. Since the zone with radial nerve injuries shows a large variation, this procedure should only be done by experienced surgeons.