Background The aim of this study was to assess the performance of a score-based diagnostic approach (SBDA) proposed in the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) 2012 guideline, and the usefulness of bone mineral density (BMD) measurement in SBDA as an objective finding in the diagnosis of celiac disease (CD). Methods The SBDA scores of 153 biopsy-proven celiac diagnosed children (derived from symptomatology, serology, human leukocyte antigen [HLA] analysis, histology) were calculated. Additionally, BMD Z scores obtained at diagnosis were also investigated. The diagnostic sensitivity of SBDA was tested in different scenarios in which low BMD was scored as a diagnostic finding. Results The mean age of children was 9.48 +/- 3.59 years and 54.2% were female. All patients scored >= 4, which is the minimum score to diagnose CD in SBDA. Mean BMD Z score in 142 of 153 patients was -2.70 +/- 1.16, and 73.9% of them were below -2. Moreover, different diagnostic scenarios without histology were tested. In one of them, BMD and HLA were not included and the sensitivity was 85.2%. In another one, low BMD was scored as an equivalent of malabsorption, HLA was not included and sensitivity was 97.2%. The sensitivities of these scenarios were significantly different (P = 0.001). Conclusion In the absence of both HLA and histology, accepting low BMD as an equivalent of malabsorption drastically increased the diagnostic sensitivity, while SBDA had limited success. Therefore, BMD might be useful when HLA and biopsy are not available.