Early diagnosis of jaw osteomyelitis by easy digitalized panoramic analysis

Abstract

Osteomyelitis is an intraosseous inflammatory disease and it is characterized by progressive inflammatory osteoclasia and ossification. Its most typical pathogenesis is bacterial infection and it may be induced by trauma, radiation or specific drug as well. As treatment of osteomyelitis, early treatment by antibiotics with accurate diagnosis is the best way. Development of medical science and increased oral sanitation contributed to reducing prevalence of maxillary osteomyelitis for the past couple of decades. Recently, a new type of maxillary osteomyelitis being induced by drug called bisphosphonate related osteonecrosis of the jaw (BRONJ) or osteoradionecrosis which secondarily induced by radiotherapy is represented. Taking panoramic radiograph is a routine process for patients who visited dental clinic at the first time. Possibility of quantitative analysis is on the rise to assist interpretation of osteomyelitis in this study. The objective is to perform early diagnosis of osteomyelitis on digital panoramic radiograph using basic function provided by PACS that is a program showing radiographic image. This study was performed by targeting total 95 patients whose symptom was confirmed as osteomyelitis under clinical, radiologic, pathological diagnosis at SNUDH oral & maxillofacial surgery for total 11 years from 2008 to 2017. Above patients with osteomyelitis were classified under clinical, radiological and pathological diagnosis and it was divided into 5 categories including osteoradionecrosis (ORN), bisphosphonate-related osteonecrosis of jaw (BRONJ, suppurative/sclerosing type) and bacterial osteomyelitis, (suppurative/sclerosing type). The photographic density in a certain area in digital panoramic radiograph was measured by using measure area rectangle that is one of basic function in INFINITT PACS®(INFINITT Healthcare, Seoul, Korea) being used at SNUDH among PACS. Photographic density of both sides was compared by dividing digital panoramic radiograph of the patients with osteomyelitis based on median line. Randomly sampled 117 persons who did not receive diagnosis of osteomyelitis by visiting SNUDH oral & maxillofacial surgery from 2008 to 2017 were targeted as control group (WNL). Conditional inference tree, which is one type of decision making tree was being generated with program R as a statistical analysis. Statistical significance was measured in comparison among suppurative type and sclerosing type of BRONJ and bacterial osteomyelitis by t-test with SPSS. In conditional inference tree being generated by obtained data, in case that average value difference exceeded 54.49 and min value difference was fewer than 54.49 and bigger than 12.81 and min value difference exceeded 39, such result is also suspicious of osteomyelitis. Along with this result, the fact that disease could be correctly classified based on probability of 88.1% as a result. There is no difference in photographic density value of BRONJ and bacterial osteomyelitis. This result means that it is unable to classify BRONJ and bacterial osteomyelitis by analyzing panoramic radiograph quantitatively as existing researches. Significance of this study is that it would be reasonable to measure photographic density by basic function only being used in PACS and use its data as assistant means while diagnosing osteomyelitis.