Objective: To evaluate the value of single-photon emission computed tomography / computed tomography (SPECT/CT) compared with planar bone scintigraphy (BS) in the assessment of necrotic bone fragments in patients with delayed bone healing or non-union after traumatic fractures. Methods: Retrospective evaluation of patients with traumatic fractures and suspected delayed healing or non-union and/or persistent pain or suspected infection who had undergone planar late phase BS and SPECT/CT between 2011 and 2018. On the BS and SPECT/CT images, a necrotic fragment was considered if there was an area of absent radiotracer uptake (photopenia) related to bone at the fracture site. Histology served as a reference standard (presence or absence of necrotic bone fragments). If histology was not available, intraoperative findings and combined clinical and imaging follow-up served as reference standards. Results: In 37 consecutive patients with traumatic fractures (femur (n = 18), tibia (11), humerus (6), radius (1), jaw (1)), necrotic bone fragments were suspected in 11 fractures (29.7%) on BS and in 16 fractures (43.2%) on SPECT/CT. 35 fractures (94.6%) had metallic implants. Histology showed necrotic fragments in 10/11 (90.9%) patients. For the detection of necrotic bone fragments, SPECT/CT showed sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of 100%, 95%, 94%, 100%, and 97%, respectively, and BS 67%, 95%, 91%, 81%, and 83%, respectively. SPECT/CT significantly outperformed BS with better area under curve (AUC) for SPECT/CT (0.9773) compared to planar imaging (0.8106) (p-value < 0.01). Conclusions: SPECT/CT is an accurate tool in the assessment of necrotic bone fragments in patients with delayed bone healing or non-union after traumatic bone fractures and is superior to conventional planar BS. Advances in knowledge: SPECT/CT is accurate and superior to planar BS in identification of necrotic bone fragments, responsible for delayed bone healing/non-union after fractures.