Many studies have reported the risk factors for developing MRONJ and have determined the following: age, sex, antiresorptive medications, duration of therapy, corticosteroid use, anemia, diabetes, cancer type, tooth extraction, and periodontal disease 5. As CT is essential for evaluating the extent of lesions and determining treatment methods, we considered that identifying the relationship between CT image findings and prognosis is indispensable for appropriate management. In the present study, we focused on pre-treatment findings, including various clinical symptoms and CT image findings. Moreover, we investigated which could be risk factors for poor prognosis in patients with MRONJ.
Since the AAOMS first published a position paper, the treatment strategy for MRONJ has remained controversial owing to the lack of a consensus on the treatment methods. The goals of both conservative and surgical treatments are the same: control of pain or secondary infection and prevention of extension or development of the disease 5. Numerous reports have emphasized the efficacy of surgical treatment 14-18. Additionally, recent studies have reported that conservative therapy alone results in residual disease, and extensive surgery can be performed even at the early stages of MRONJ 19,20. In a systematic review, the cure rates of patients who underwent conservative treatment, minimal debridement, and extensive surgery were 33%, 74%, and 84%, respectively 14. Similarly, in the present study, the cure rates were 50%, 64%, and 100%, respectively, and conservative treatment showed a lower recovery rate than the other treatment methods. In addition, multivariate analysis identified conservative treatment alone as a risk factor for poor prognosis. As shown in table 4, conservative treatment alone had a worse cure rate than surgical intervention at every stage, which is consistent with previous studies 14,21,22. Despite the low recovery rate, conservative treatment remains the main option for patients who refuse to undergo surgery or whose general condition is insufficient to tolerate surgery. Varoni et al. reported that antibiotic agents could potentially promote sequestration and the efficacy of combination therapy, including medication and surgical treatment 23. However conservative treatment only provides temporary remission and is less likely to be curative management, it is still indispensable for a significant pool of patients with MRONJ.
Osteosclerosis is the most common feature in patients with MRONJ 24. It appears to be a side effect of antiresorptive or antiangiogenic agents or a response to inflammation or infection. Previous studies described the positive relationship between the degree of sclerosis and the progression of the disease 9,25. In the present study, any relation between osteosclerosis and the clinical stage or prognosis was found. However, some cases showed sclerotic change in entire of jawbone which made deciding resection area extremelly difficult. We suggested that osteoslerosis can be a major obstacle in determining the treatment strategy of MRONJ.
Osteolysis, which results from the destruction of bone tissue and indicates disease progression, is also frequently observed in patients with MRONJ. In the position paper, the following feature is an important criterion for classifying patients as stage 3 or below: osteolysis extending to the inferior border of the mandible or sinus floor 5. Since the infection occurs at the mucosal defect site and resorbs the bone superficially, osteolytic changes are commonly observed in the cortical bone 26. Cortical perforation was observed as osteolysis progressed. Bone disruption tends to occur at the buccal site compared with the lingual bone 11. Further progression of osteolysis results in a periosteal reaction, which is new bone formation caused by inflammation or irritation of the periosteal membrane 27,28. Few studies have reported the relationship between CT findings and treatment outcome, and osteolysis and periosteal reaction were previously determined as significant risk factors for incomplete recovery after surgical treatment 12,29,30. In the present study, the rate of patients with osteolysis increased with MRONJ progression (Table 5). The cure rates among these patients by treatment method were as follows: conservative treatment, 25% (8/32); minimal debridement, 40.6% (13/32); extensive surgery, 100% (4/4). The cure rates in patients with lingual cortical perforation and periosteal reaction were 22.5% (9/40) and 11.4% (4/35), respectively. These imaging findings are the result of osteolysis progression, and multivariate analysis showed osteolysis as a significant risk factor for poor prognosis. We suggest that complete removal of the osteolytic region should be set as a treatment goal to improve the recovery rate. Immediate therapeutic intervention is required, because the healing rate worsens as osteolysis progresses. Furthermore, the cure rate of patients with periosteal reactions who underwent surgical treatment was 23.5% (4/17), while no patients recovered with conservative treatment. The healing rate was low; however, we concluded that surgical intervention is the only method to cure MRONJ with periosteal reaction. Extensive surgery should be performed in cases with severe pain or intractable infection which significantly underimine the quality of life (QOL).
The multivariate analysis also identified non-sequestration as a risk factor for a poor prognosis. Reduced bone turnover may result in sequestration, defined as necrotic bone separation from the surrounding bone 31. In CT images, it is observed as a non-continuous bone fragment surrounded by osteolytic change 32. In cases of sequestrum, conservative surgical approaches, such as sequestrectomy, are indicated 33. Since it is minimally invasive and has a better cure rate than conservative treatment, sequestrectomy is the most commonly performed surgical method 34,35. In the present study, we performed minimal debridement which include sequestrectomy and removing necrotic soft tissue. The success rate of patients with non-sequestration was only 39.4% (13/33), whereas patients with sequestration showed a high recovery rate of 90.3% (28/31). Because the boundary between the lesion and the viable bone is indistinct, we recommend avoiding minimal debridement in cases without obvious sequestration. It is crucial to recognize that even if symptoms are in temporary remission after the removal of necrotic tissues, the disease is prone to recurrence from the residual regions. Extensive surgery is required in cases with less sequestration expectation for completely removing the lesions and dramatical QOL improvement. However, as it is less invasive and has no limitation of adaptation, with a high recovery rate, minimal debridement should be aggressively performed in patients with sequestration. For appropriate treatment strategies, unraveling the pathophysiology of MRONJ, determining the antibiotic period for sequestration, and determining the time for therapeutic intervention are indispensable.
This study was limited by its retrospective, non-matched design, which meant that other risk factors, such as oral hygiene indices, could not be examined. Additionally, a multivariate analysis was performed to decrease the effect of confounding factors as much as possible, and selection bias could not be completely excluded. A large-scale prospective cohort study is needed to further evaluate the prognosis in patients with MRONJ.
In conclusion, we successfully demonstrated the associations between various risk factors and the prognosis in patients with MRONJ. Regardless of pre-treatment clinical symptoms, patients who received conservative treatment alone showed a statistically poor prognosis. In the CT imaging findings, osteolysis and non-sequestration were risk factors for poor prognosis.