*3.4. Heatmap Analysis*

In general, there was a distinct difference between the perio and endo groups. The perio group showed genera *Streptococcus, Veillonella, and Leptotrichia* (Figure 6A). Especially, *Fusobacterium* was found in a higher proportion in sample 36-2 compared to that in the other samples of the perio group. The endo group showed significant proportions of grampositive anaerobic cocci species, such as *Parvimonas micra* and *Peptostreptococcus stomatitis* (group) (Figure 6B).

**Figure 6. Heatmap analysis at the genus (A) and species (B) level with >3% frequency.** The perio group showed genera *Streptococcus, Veilonella, Leptotrichia, Prevotella,* and *Fusobacterium.* The endo group showed gram-positive anaerobic cocci species, such as *Parvimonas micra* and *Peptostreptococcus stomatitis*.

#### **4. Discussion**

The patient showed the stereotypic characteristics of a MIM patient. Based on our study in 2014 [1], the average age of 12 MIM patients at the initial visit was about 7.8 years (range 4–13 years) and it was the same with this patient. This seems to be due to the eruption of the MIM teeth including the permanent first molars. The patient in this study was female and the sex ratio was even in a previous study [1]. However, in the study by Kim et al. [8], there were 24 males out of 38, and in the study by Vargo et al. [3], there was a male predominance at 56.3%. The gender ratio is still controversial.

Interestingly, the patient had a medical history of systemic disease, preterm birth at nine months, and medication for ADHD. Reportedly, almost 95% of the MIM patients had diseases during infancy [8]. One of 12 in Lee's study [1], one of one in McCreedy's 2015 study [18], and four of 30 in Wright's 2016 study [19] were born preterm. In addition, meningitis, spina bifida, cerebral cyst, cephalohematoma, seizures, hydrocephalus, and other diseases were present in MIM patients [3,8,19]. According to a recent review, "the exact aetiology of this condition is unknown. However, the fact that root malformations are limited to isolated teeth suggests that a non-genetic, environmental factor related to past medical history could be the cause" [3]. The authors suggested a possible epigenetic association, which needs further investigation.

In this patient, all the permanent first molars, the deciduous molars, and the maxillary central incisors were MIM teeth. This case seemed more serious compared to those in previous studies [3,8,19]. MIM teeth occurred in the permanent first molar, and often only in the mandible or in the maxilla. In many cases, the primary second molar was affected, but rarely the primary first molar or the permanent second molar. In the future, a classification for MIM is needed. The prevalence of incisors affected by MIM was about half [1] or one-third [3].

Sampling was the most crucial part for identifying the microbiome by collecting plaque and obtaining accurate data. If the sampling is incorrect, the disease-related bacteria may not be accurately identified. In this study, the sampling method applied was the one used by researchers in previous studies [13–15]. The NGS method MiSeq was developed in 2011 by Illumina. The characteristics of MiSeq consists of minimization while maintaining the chemistry of HiSeq (Illumina) and is suitable for reading V3 and V4 of 16S rRNA applied to microbial community analysis [20].

A total of 17 samples were categorized into the perio and endo groups. The perio group had higher average OTUs and alpha diversity compared to the endo group. At the genus level, *Streptococcus, Veillonella*, and *Leptotrichia* were present in most of the perio group samples, but not in the endo group. These genera were usually found in healthy Korean preschool children [21].

Particularly, sample 36-2 with a clinically severe dentoalveolar abscess showed only the presence of the *Spirochetes* phylum. *Spirochetes* were present in subgingival plaques and elevated in advanced periodontal disease and endodontic infection [22]. *Treponema socranskii* and *Treponema medium* appeared at 1% frequency in sample 36-2. *Treponema* genera belonging to oral *Spirochetes* were isolated from the periodontal inflammation site [23].

In the heatmap, the perio group was subtly divided into three clusters: (1) 11-2, 11-1, Nor-2, 26-2, 26-1, Nor-1, and 36-1; (2) 16-2, 16-1, 46-3, 46-2, and 46-1; and (3) 36- 2. The first group included the MIM-incisor and non-MIM teeth, and the maxillary left MIM-molar. The incidence of *Leptotrichia, Prevotella,* and *Fusobacterium* was relatively high. The second group included the right MIM-molars with relatively high *Streptococcus* and *Veillonella*. Sample 36-2 showed a relatively large number of *Fusobacterium*. *Fusobacterium* is an absolute anaerobic gram-negative bacterium with a long filament, which is isolated from periodontal tissue and can aggregate with other oral bacteria [24]. It is one of the signs of acute dentoalveolar infection [15].

Among the endo group, samples 16-4, 16-5, and 36-5 showed four phyla and 36-4 showed two phyla. *Saccharibacteria\_TM7* and *Proteobacteria* was not present as compared to the Perio group. At the genus level, *Parvimonas, Peptostreptococcus, Atopobium*, and *Dialister* were present in all the endo group samples. The anaerobic gram-positive cocci *Parvimonas* and *Peptostreptococcus* have been identified from caries dentin, infected pulp, and complex infections such as dental root canals, progressive periodontitis, and dental abscesses [15,25]. *Atopobium* was originally identified as an anaerobic lactic acid bacillus in a periodontal pocket. *Dialister* is a gram-negative bacterium, which is an absolute anaerobic and found in root canal infections and periodontal infections [25]. In addition, *Shuttleworthia, Bulleidia, Mogibacterium*, and *Olsenella*, identified at 1% frequency, are present in the root canals in endodontic-periodontal lesion [25]. Regarding species, the endo group showed significant numbers of gram-positive anaerobic cocci species, such as *Peptostreptococcus stomatitis* and *Parvimonas micra*.

According to the new classification of periodontal disease in 2018, general periodontitis is different from MIM complications because they usually start after middle-age [26]. In addition, early-onset periodontitis (EOP), localized juvenile periodontitis (LJP), or aggressive periodontitis (AP), categorized as periodontitis in the 2018 classification, were also different [27]. Although the location in the molar and incisor, and the sudden periodontal abscess were similar to MIM complications, MIM complications develop before puberty while EOP, LJP, and AP develop during puberty. AP was associated with *Aggregatibacter actinomyces*, but it was not identified in the MIM-complicated molar.

In the 2018 classification, "Other conditions affecting the periodontium" included periodontal abscess, endodontic-periodontal lesions, and localized tooth-related factors [28]. In the localized tooth-related factors, tooth anatomy such as cervical enamel projections, enamel pearls, and developmental grooves could enhance plaque retention [29,30]. Due to the inability to keep these areas clean, it was always accompanied by a poor prognosis despite adequate treatment. According to previous studies, the root malformation of MIM resembled cervical enamel projections, enamel pearls, and developmental grooves and cementum deformation were also confirmed [2]. Biofilm was present in the cervical lower part of the crown in the study by Witt et al. [31]. Periodontitis in MIM seemed to begin as localized tooth-related periodontitis and on further progression, it was assumed to be a periodontal abscess with a severe anatomic alteration.

Pulp necrosis occurs in all teeth affected by MIM regardless of the severity of the periodontitis. Pulp necrosis of MIM teeth could spontaneously occur due to the lack of nutrients and oxygen supply [32]. The anatomical characteristics of MIM seem to cause pulp necrosis and localized tooth-related periodontitis independently, and when combined, an endo-perio lesion might form. If a purulent bacterial infection was added to this, a periodontal abscess could occur [32]. There are two case reports of successful endodontic treatment of MIM teeth. Yue and Kim [9] successfully performed endodontic treatment of a MIM mandibular left permanent first molar of a 13-year-old boy and Byun et al. [10] treated MIM-suspected maxillary central incisors endodontically in a 12-year-old boy. Even if pulp treatment was performed, the prognosis is poor because the periodontal disease is difficult to treat. The long-term course of these cases should be determined.

Early diagnosis, the timely extraction of a MIM, and orthodontic treatment might be considered the treatment of choice, as suggested in previous studies [7,8,33–36]. This was because the prognosis for anatomical periodontitis is poor. Although there were two successful cases of endodontic treatment in the previously published papers [9,10], as long as the biofilm causing periodontal disease persists, as in the study by Witt et al. [31], it will cause recurring periodontitis. Rather, extraction of the MIM-molar at the proper time when the secondary molar can move mesially and considering orthodontic treatment may prove to be of long-term benefit to the patient [37].

There were limitations to this research. First, clinical tests including vitality tests, periodontal examination, and a periapical radiographic examination were not done at every visit. Second, microbiological analyses should be done in more MIM patients or more non-MIM patients for comparison.

#### **5. Conclusions**

The results of this study confirmed that the dentoalveolar infection of a MIM was different from localized juvenile periodontitis. It seemed that pulp necrosis and localized tooth-related periodontitis occurred spontaneously over time after the eruption of the MIM and developed into severe endo-perio lesions or a periodontal abscess. We carefully suggest that the treatment of choice was the extraction of the MIM at the appropriate time.

**Author Contributions:** Conceptualization, H.-S.L. and S.-C.C.; Methodology, H.-S.L.; software, H.J.K.; validation, K.L.; formal analysis, O.H.N.; investigation, S.-C.C.; resources, M.-S.K.; data duration, H.-S.L.; writing—original draft preparation, H.-S.L. and S.-C.C.; writing—review and editing, H.-S.L. and S.-C.C.; Supervision, project administration, funding acquisition, H.-S.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Basic Science Research Program of the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science, and Technology (NRF-2016R1C1B1015005 and 2020R1C1C1006937).

**Acknowledgments:** We thank to Je Seon Song and Soo-Hyun Kim for stating the MIM research together and Jung-Wook Kim for inspiring us.

**Conflicts of Interest:** The authors declare no conflict of interest.
