Succes van pulpotomie bij kiezen met pulpitis met parodontitis gedetecteerd met behulp van AI en degenen zonder parodontitis

The dental pulp constitutes a distinct and intricate entity susceptible to various thermal, traumatic, microbial, and chemical stimuli. (1) As a result, the dentin-pulp complex experiences a sophisticated immune response, leading to the recruitment of immune cells and inflammation of dental pulp cells at the local level. (2,3,4) Historically, in instances of irreversible pulpitis, the conventional recommendation has been root canal treatment (RCT), based on the belief that the pulp lacked the ability to undergo recovery. (5,6)

Vital pulp therapy (VPT) is an approach focused on minimally invasive measures, aiming to seal the pulpal wound with a bioactive substance subsequent to the removal of infected pulpal tissues. (7,8) Procedures associated with VPT encompass direct and indirect pulp capping, partial pulpotomy, and full pulpotomy. (9) Typically, VPT interventions are carried out to safeguard the radicular pulp in both deciduous and adult immature teeth. (10,11) In accordance with the position statements released by the American Association of Endodontists (AAE) in 2021 and the European Society of Endodontology (ESE) in 2019, full and partial pulpotomy emerge as promising and more conservative alternatives to root canal treatment for managing mature permanent teeth afflicted with irreversible pulpitis and cariously exposed pulps. (10,12) This is also backed by literature since a lot of researches shows that performing full pulpotomy to manage mature permanent molars with irreversible pulpitis is highly successful. (30,31,32,33)

A full pulpotomy entails the complete removal of coronal pulp tissues, succeeded by the application of a biomaterial onto the remaining tissues at the root canal orifices. (12) The utilization of pulpotomy preserves the pulp mechanoreceptors, along with its defensive and developmental functions, including the formation of primary and secondary dentin.(12) Furthermore, pulpotomies are characterized by being less time-consuming, less invasive, and less intricate when compared to root canal treatment. Over the past two decades, numerous studies have indicated that pulpotomy can serve as an effective alternative to root canal treatment for managing irreversible pulpitis in mature permanent teeth.(13,14)

The AAE classification simplifies pulpitis into reversible and irreversible categories. However, the term "irreversible" is questioned as histological evidence shows no clear boundary for irreparability. In 2017, a new classification proposed by Wolters et al (15) eliminated the term "irreversible" and linked symptoms to VPT management strategies. Subsequent research by Careddu and Duncan in 2021 supported the new classification's potential prognostic benefit in partial pulpotomy.

Wolter's classification mainly entitles the following:

1. Initial Pulpitis:

   A heightened but not prolonged response to the cold test, insensitivity to percussion, and an absence of spontaneous pain. (15)

2. Mild Pulpitis:

   An intensified and extended reaction to cold, warm, and sweet stimuli lasting up to 20 seconds, potentially percussion-sensitive. Histologically, this suggests limited local inflammation confined to the crown pulp. (15)

3. Moderate Pulpitis:

   Evident symptoms, strong, heightened, and prolonged response to cold which can last up to several minutes, potentially percussion-sensitive, and spontaneous dull pain that can be partially or completely alleviated with pain medication. Histologically, this indicates extensive local inflammation confined to the crown pulp. (15)

4. Severe Pulpitis:

Severe spontaneous pain and distinct pain response to warm and cold stimuli, often sharp to throbbing, causing difficulty sleeping (exacerbated when lying down). Tooth sensitivity to touch and percussion is pronounced. Histologically, this suggests extensive local inflammation in the crown pulp, possibly extending into the root canals. (15)

Wolter et al indicated that initial and mild pulpitis will only require indirect pulp therapy. (15)

Traditionally, it was hypothesized that a tooth with a periapical radiolucency has a necrotic, infected pulp space. Thus, root canal treatment was the treatment of choice for these teeth.(2) However, it was proven that even in the presence of a large periapical radiolucency, portions of the radicular pulp tissues can maintain their vitality.(3) This suggests that in cases with apical periodontitis, if vital pulp tissues are evident after pulp exposure, pulpotomy can be a viable alternative treatment modality to root canal treatment.(4)

In 2018, Taha et al. aimed to evaluate the outcome of full pulpotomy in mature permanent teeth using Biodentine. Eight of the cases exhibited preoperative periapical rarefaction, and seven of them showed improvement in the periapical index score. Similarly, in a 2016 randomized clinical trial, all of the seven cases with preoperative periapical radiolucencies showed complete resolution of the lesions at the end of the study.(5) Other earlier studies also reported success of pulpotomy procedures in cases with periapical involvement. (34,35,36) However, there is a growing need for further studies in order to determine the predictability of pulpotomy in teeth with apical radiolucencies, and to detect features that can act as negative prognostic factors in these cases.(4)

In 2013, the American Association of Endodontists classified the diagnosis of periapical condition into normal apical tissues, symptomatic apical periodontitis, asymptomatic apical periodontitis, chronic apical abscess, acute apical abscess, and condensing osteitis. The term "apical periodontitis" indicates that the apical periodontal tissues are inflamed. In symptomatic apical periodontitis, the tooth has a painful reaction to palpation, percussion, and biting. Radiographically, the tooth may have a normal periodontium or a periapical radiolucency depending on the stage of the periapical disease. On the other hand, teeth with asymptomatic apical periodontitis do not exhibit pain on palpation or percussion, and they are associated with a periapical radiolucency in the radiograph.(1)

It is well-recognized that the effectiveness of detecting radiographic apical periodontitis is contingent upon the proficiency of the operator(16). Therefore, it has become imperative to develop or employ Artificial Intelligence (AI)-based software solutions that can accurately diagnose periapical radiolucencies. Artificial intelligence (AI) is designed to mimic human intelligence and address specific challenges. AI contributes to the creation of algorithms that can learn from provided information and make predictions. Machine learning, a subset of AI, constructs algorithms based on data.(17) Among the earliest AI algorithms were neural networks (NNs). Deep learning neural networks are complex structures with multiple layers, while shallow learning neural networks are simpler with fewer layers. Convolutional neural networks (CNNs) are primarily used for analyzing intricate images.(17)

AI technology has arisen as a response to the need for machines to emulate human intelligence, facilitating the provision of more standardized results.(18) For instance, AI-enabled imaging applications in the medical domain exhibit autonomous object detection and image classification capabilities, with numerous studies demonstrating remarkable accuracy and promise in diagnosing pathologies(19), interpreting radiological data(20), and assessing dermatological conditions(21).

From a dental perspective, AI applications can be stratified into several domains, encompassing diagnosis, decision support, treatment planning, and prognosis prediction, with diagnostic capabilities being particularly prominent. AI stands poised to augment diagnostic precision and efficiency, thereby alleviating the professional burden on dentists, who increasingly rely on computerized decision support systems.(18)

Within the realm of endodontics, Anita Aminoshariae et al in 2021 mentioned that AI can currently assist in detection of peri-apical lesions, crown and root fractures, working length determination and even the morphology of root canals and the root canal systems.(23) A systematic review by Agata Ossowska et al in 2022 concluded that AI can also be used in predicting the ability of dental pulp stem cells to survive in some treatments and predicting the success of re-treatment of a failed root canal therapy (RCT).(24) However, it's important to note that AI tended to over-detect lesions while human operators tended to under-detect them.(22)

Duncan et al, in 2022, recommended the use of CBCT as it can detect periapical lesions better than 2-Dimensional X-Rays.(4)In another review by Issa et al in 2023, AI has exhibited an exceptional level of precision in detecting apical periodontitis within periapical radiographs, owing to its capacity to detect intricate patterns and attributes beyond the detection of human observation.(6)

This, in turn, positions it as an efficient auxiliary diagnostic instrument for dental professionals, simultaneously mitigating the clinical burden and improving the standard of care.(6)

The main goal of this research is to compare cases of moderate pulpitis without apical periodontitis in comparison to cases of severe pulpitis with AI-Detected apical periodontitis. Additionally, the study aims to evaluate the accuracy of a novel AI technology in identifying apical periodontitis from radiographs and to establish the correlation between the utility of AI as a dependable supplementary tool in the decision-making process when contemplating full pulpotomy for mature permanent molars.

In the paper by Duncan et al, even though it was proven that even in the presence of a large periapical radiolucency, portions of the radicular pulp tissues can maintain their vitality, more clinical studies are needed to assess the success rates of cases with severe irreversible pulpitis with apical periodontitis.

The study done by Anita Aminoshariae et al reported that AI can accurately detect peri-apical lesions. However, another study by Cantu et al reported that AI generally tends to over-detect lesions. Thus, it is imperative to compare AI findings to the operator's radiographical observations.