A micro-computed tomographic evaluation of root canal isthmuses in maxillary and mandibular first molars of a Black South African population

1. Introduction

One possible common explanation for endodontic treatment failure is the complexity of the root canal anatomy.¹ One such a feature that increases this complexity is the presence of an isthmus within the root canal complex. Observation of literature especially micro-computed tomography (micro-CT)-focused studies has shown that even accessory root canal features such as isthmuses differ between population groups, which is not unexpected as human dental morphology reflects underlying genetic structure.^2-4 The high heterogeneity observed suggests that population/ethnicity, together with differences in sample selection and methodology, may contribute to the reported variability. Evidence regarding intrigate root canal morphology (such as accessory canals and complex configurations) in African populations remained limited until the publication of the first Black South African micro-CT study in 2024.⁵

An isthmus can be defined as a portal of communication connecting two root canals and has also been described as a corridor or transverse anastomosis.⁶ These communicating portals may contain infected pulpal tissue with causative microbes as long as ten years after treatment in failed endodontic cases.⁷ Because of their thin, often flat appearance, isthmuses are among the most difficult areas to clean and shape with current instruments. Mechanical preparation causes debris to accumulate within them often rendering contemporary irrigation regimes incomplete.^1,8

Conventional two-dimensional (2D) radiographs allow only a limited view of canal complexity. As a result, fine details frequently remain undetected.^1,9 High resolution imaging such as micro-CT enables a detailed, non-destructive investigation of complex internal morphology. This modality is often regarded as the reference standard for describing integrate root canal anatomy.^2,5,6,9,10

The aim of this study is to evaluate root canal isthmuses considering arch side, sex and age within the root canal system of Black South African first molars.

2. Materials and Methods

2.1. Study design

Permission to conduct the study was granted by the Research Ethics Committee of the Faculty of Health Sciences, University of Pretoria (298/2020). The study followed a quantitative, descriptive, cross-sectional and observational design, and the report followed the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines¹¹ (Figure 1).

2.2. Origin of scans

Human skulls were selected only where the age, sex and population affinity of the individual were known. Individuals were sourced from the Human Osteological Research Collection (HORC) at the Sefako Makgatho Health Sciences University (SMU) and the Pretoria Bone Collection (PBC) at the University of Pretoria (UP).¹² Population affinity in both collections is documented through self-identified ancestry supplied with each donation. Only individuals documented as Black South African were included. Family members consented to research use of the remains, after which the bodies formed part of the whole-body donation programme; where a third party donated, the Director-General granted approval. All bodies are protected by the National Health Act 61 of 2003.¹³

Scanning was completed at the South African Nuclear Energy Corporation (Necsa, Pelindaba, South Africa) using a Nikon XTH 225L industrial micro-focus X-ray CT system (Nikon Metrology, Leuven, Belgium). The settings used were 100 kV voltage, 100 µA current and a 2.0 s exposure time per projection. The X-ray spot size ranged between

0.001 and 0.003 mm (1–3 µm). The translation table has a rotational accuracy of 1/1000^th of a degree and a pixel-size capability of 200 µm × 200 µm. The PerkinElmer detector field of view was 400 mm × 400 mm; approximately 200 mm × 200 mm was used per maxilla or mandible.¹⁴

The 2D projections were reconstructed into three-dimensional (3D) volumes using Nikon CT Pro version 4.4.3 (Nikon Metrology). Isotropic voxel size ranged between 40.3 µm and 74.2 µm. Volumes were imported into Avizo 2019 (Thermo Fisher Scientific, Waltham, MA, USA) for post-acquisition processing.¹⁵

2.3. Segmentation, alignment and image acquisition

Each scan was opened in 3D in Avizo 2019 using the isosurface module within the software, and first molars were isolated through cropping and segmentation. Landmarks were placed on each tooth using the cemento-enamel junction (CEJ). The CEJ has been confirmed as a readily identifiable point of reference.¹⁶ A single landmark was placed at the highest occlusal point of each root on the buccal and palatal surfaces by following the CEJ, so that each molar carried three or four landmarks across the buccal and palatal/lingual surfaces.

The Avizo 2019 software was then able to compute a best-fit plane at the CEJ from these landmarks and used it to re-align the image stacks: applying identical steps to each scan minimised bias and oblique sectioning. Enamel, dentine and pulp were differentiated by using different colors within the watershed method (region-based semi-automatic segmentation).¹⁷ A procedure known as Masking or multi-slice editing⁹ corrected slices where minor dentinal cracks were noted or where the pulpal space extended beyond the roots.

2.4. Scan analysis

A format like that described by Jonker et al.^5,10,18 was followed during observation. Isthmus data were first collected by the main researcher, experienced in Avizo and endodontics. Each scan was allocated a unique code and data was captured anonymously, without knowledge of sex, arch side or age. The pulp of each scan was isolated, magnified and rotated for the best possible view of the pulpal complex. Brightness, contrast and sharpness were fixed at predefined calibration values, and the same parameters were used for every scan to allow standardisation. Capturing of data included the number and location of isthmuses by root third (coronal, midroot or apical) and isthmus type according to Fan et al.¹⁹ (Figure 2).

After data capturing was completed by the main researcher, a second researcher (a specialist in prosthodontics with endodontic experience) participated in the inter-observer reliability test. Before the main test, both researchers were calibrated on two micro-CT scans, unrelated to the reliability sample, by jointly reviewing images against the Fan et al.¹⁹ criteria; during calibration, explicit diagnostic thresholds (what constitutes as an isthmus, its region and its type) were agreed for a positive finding and each isthmus type. Any disagreements during calibration were resolved until a shared decision solution was established. During the main test the researchers recorded results independently. When results differed, the images of the tooth in question were re-evaluated and discussed until consensus was reached. The main researcher re-examined the same scans after approximately one week for the intra-observer assessment.^10,18

2.5. Inclusion criteria

Only high-resolution scans (no blurring or double imaging) of teeth from Black South Africans, with intact, fully developed roots and where the pulpal space could be accurately isolated, were selected.

2.6. Exclusion criteria

Three-rooted mandibular and fused-rooted maxillary teeth, immature apices (incomplete root formation), root fractures, coronal or radicular resorption, existing root canal treatments, extensive decay obscuring any root canal, metallic restorations, suboptimal-quality scans, and population groups other than Black South Africans were not considered.

2.7. Sample size

All available scans (n = 151) formed a convenience sample (Figure 1). From these scans, a total of 93 maxillary and 84 mandibular first molars (n = 177 total) from 87 individuals (n = 48 males; n = 39 females) were included after consideration of the inclusion and exclusion criteria. Both antimeres were eligible and therefore analysed separately: some individuals contributed only a single molar (n = 32) whilst the rest contributed more than one (n = 84). The sample had more teeth on the right side (n = 48 maxilla; n = 43 mandibula) compared to the left (n = 45 maxilla; n = 41 mandibula). More male molars (n = 47 maxillary; n = 46 mandibula) than female ones (n = 46 maxilla; n = 37 mandibula) were identified. The age of individuals ranged between 20 and 89 years.

2.8. Statistical analysis

Analysis was performed in R version 4.1.1 (R Core Team, Vienna, Austria). Associations between arch side, sex, age and isthmus type and the presence of isthmuses were tested with the chi-squared test of association, with significance set at p < 0.05. Because several subgroups, particularly the older age groups and rarer types such as Type IV, contained low or zero counts, the expected-frequency assumption was not met in every cell. As a result, the relevant p-values are reported for completeness but should be interpreted with caution.

To convey practical importance rather than relying on the p-value alone, the bias-adjusted Cramér’s V and its 95% confidence interval were calculated for every comparison. Given the many subgroup analyses across sex, age, arch side, root third and isthmus type, no formal multiple-comparison correction was applied. Instead, isolated

significant p-values accompanied by small effect sizes and confidence intervals spanning zero were considered for the purpose of generating hypothesis. Reliability was calculated using percentage agreement and Cohen’s kappa on approximately 20% of the sample (n = 20 randomly selected scans).^10,18 Random selection ensured that the chosen reliability sample reflected the anatomical variability present in the complete dataset.

3. Results

3.1. Sample characteristics and examiner agreement

Of the 177 first molars studied, 91 were from the right side of the mandible or maxilla (n = 91/177, 51.4%) and 86 from the left (n = 86/177, 48.6%). Ninety-four teeth belonged to males (n = 94/177, 53.1%) and 83 to females (n = 83/177, 46.9%). Ages ranged between 20 and 89 years. Intra- and inter-observer percentage agreement was calculated as 96.73% and 97.78% respectively for maxillary teeth and 94.12% and 96.08% respectively for mandibular teeth. When chance agreement was considered using Cohen’s kappa, the corresponding overall values were κ = 0.90 and κ = 0.95 for the maxillary teeth and κ = 0.75 and κ = 0.94 for the mandibular teeth, indicating substantial to almost perfect agreement according to the Landis and Koch classification²⁰ (Table 1).

One item namely the mandibular mesial-root isthmus returned as κ = 0.00 despite 94.12% agreement (a recognised kappa artefact arising when a finding is almost universally present). As a result, the item was excluded from the mean kappa. Importantly, the near-universal mesial-root prevalence should be interpreted with this limitation in mind: chance agreement is therefore a possibility rather than an observed agreement.

3.2. Mandibular first molars

3.2.1. Sex, arch side and age

Isthmuses were more commonly found in the mesial (M) root than in the distal (D) root. In the M root they were relatively evenly distributed by arch side, being present in 97.67% of teeth on the right (n = 42/43) and 95.12% on the left (n = 39/41). In the D root they were less common, at 69.77% on the right (n = 30/43) and 78.05% on the left (n = 32/41). By sex, M-root isthmuses were present in all male teeth (n = 47/47, 100%) and in most female teeth (n = 34/37, 91.89%), and Type III was the most common type in both arches and both sexes (Table 2). A similar pattern was seen in the D root (males n = 36/47, 76.60%; females n = 26/37, 70.27%). The apical region was the most common root third for isthmuses in both roots (Table 2).

Two statistically significant associations emerged for the M root, but their effect sizes and confidence intervals qualify how they should be interpreted. Coronal isthmuses in the M root were more prevalent on the right (100.00%) than on the left (78.38%) (p = 0.027), although the bias-adjusted Cramér’s V of 0.233 (95% CI 0.000–1.000) indicates only a small association (confidence interval includes zero). Apical isthmuses in the M root were more prevalent in males (95.35%) than in females (78.05%) (p = 0.003); this was the single most robust association in the study, with an adjusted V of 0.351 (95% CI 0.151–1.000) whose lower bound excluded zero, denoting a small-to-moderate and reasonably stable effect. No statistically significant relationship was noted for the D root by sex or side.

Across age groups, the total number of teeth with isthmuses, and particularly apical isthmuses, tended to increase with age in both roots, reaching significance in the D root. The overall presence of D-root isthmuses differed significantly across age (p = 0.004; adjusted V = 0.390, 95% CI 0.000–1.000), as did the apical D-root subgroup (p = 0.027; adjusted V = 0.298, 95% CI 0.000–1.000). In the coronal region of the M root the prevalence was more stable, at roughly 83–90% of teeth across the age range.

Considering isthmus type, Type III became increasingly more prevalent with age with the highest incidence in the 60–69-year group for the M root and in the 40–49-year group for the D root. Also, Type II isthmuses declined, and Type I disappeared in the older M-root groups. Both these types were rarely found in the D root. Type IV was rare throughout both roots. A statistically significance of p = 0.002 for Type IV in the D root was noted (Table 3). However, this finding arises from a table in which every cell count is zero. It can be speculated that this might be a statistical artefact rather than a meaningful difference. Finally, it is important to note that the older age groups (especially the 70+ group, n = 3) contained very few teeth, all age-related comparisons should be regarded as exploratory rather than statistical impact.

3.3. Maxillary first molars

3.3.1. Sex, arch side, root thirds and age

The highest prevalence of isthmuses was found in the mesio-buccal (MB) root. By sex, MB-root findings were similar between males (n = 41/47, 87.23%) and females (n = 41/46, 89.13%), whereas the DB and P roots showed a higher prevalence in males (DB: n = 29/47, 61.70%; P: n = 22/47, 46.80%) than in females (DB: n = 18/46, 39.13%; P: n = 15/46, 32.61%) (Table 4). As in the mandible, the most common type was Type III in both arches and both sexes, and the apical region was found to be the most common root third.

The prevalence was similar for both arches (right: n = 43/48, 89.58%; left: n = 39/45, 86.67%). In contrast, in the disto-buccal (DB) root, isthmuses were more commonly found on the right (n = 28/48, 58.33%) than on the left (n = 19/45, 42.22%), with an overall prevalence of 50.54% (47/93) for the DB root. The palatal (P) root showed a higher prevalence on the left (n = 22/45, 48.89%) than on the right (n = 15/48, 31.25%).

No statistically significant difference was noted between arch sides in the maxilla. Considering the effect of sex, a significant difference was found in the DB root. Isthmuses were more prevalent in males (p = 0.047). However, the adjusted Cramér’s V of 0.201 (95% CI 0.000–1.000) indicates a small association whose confidence interval includes zero. As a result, this finding could perhaps be considered suggestive at most.

Within age groups, the midroot region of the DB root was more likely to contain an isthmus in the 40–49-year range (p = 0.015; adjusted V = 0.334, 95% CI 0.000–1.000), and a Type IV isthmus in the MB root also reached significance across age (p = 0.026; adjusted V = 0.271, 95% CI 0.000–1.000). Same as before, the tables had several empty or single-observation cells possibly impacting the analysis outcome (Table 5). For the maxillary roots (Table 5), isthmus prevalence in the apical region tended to increase slightly in older individuals in particular 50–59 years of age, while coronal and midroot isthmuses were comparatively uncommon and, in the MB root, decreased with age.

4. Discussion

Without adequate knowledge of the root canal complex, clinical outcomes and long-term tooth survival may be jeopardised,¹ and the root canal isthmus is among the factors contributing to endodontic failure. Over the years, researchers have investigated root and canal morphology using either conventional 2D radiographs,²¹ in vitro macroscopic and clinical investigations,²² clearing and staining techniques,²³ and 3D modalities such as cone-beam computed tomography (CBCT)²⁴ and micro-CT,^4,10 the last often considered by many as the gold standard because of its high ex vivo resolution.^2,5,6,9 As no previous study on root canal isthmuses in the first molars of Black South Africans was identified, the present study aimed to address this gap and contribute towards the limited micro-CT literature pool describing isthmus morphology across populations.

Reported isthmus prevalences vary widely between populations ranging between approximately 10%²⁵ to 100%²⁶. For mandibular first molars, the present study calculated values between 91.89% and 100% in the M root and 70.27% to 76.60% in the D root. These values are within range, they are on the higher end of the scale. The near-universal occurrence of mesial-root isthmuses compared with the considerably lower prevalence in distal roots represents one of the principal

anatomical findings of this study and highlights the substantially greater complexity of the mesial root canal system. In comparison, Estrela et al.²⁷ reported 87.9% in Brazilians but arch side or sex were not investigated. Also, the investigators used CBCT whose lower resolution may hamper the identification of finer communications.⁹ This could in part explain the slightly lower value reported by these authors. Importantly, a methodological rather than purely population-based explanation is plausible.

In the current study, isthmus prevalence was found to be generally higher in males compared to females and higher on the right arch than the left. Shakeri et al.,²⁸ using CBCT to investigate mandibular first molars of Iranians also found a higher prevalence in males, though their comparisons were not statistically significant and not specified by root region. In contrast, male predominance of isthmuses in the apical third of the M root reached significance (p = 0.003) in the present study. The effect-size confidence interval excluded zero (adjusted V = 0.351, 95% CI 0.151–1.000) illustrating higher significance compared to other findings. A functional explanation has been proposed: greater masticatory forces in males, and on the preferred (often right) chewing side, might result in higher rates of adaptive dentine deposition and root canal structural change.^29,30 However, this reasoning remains speculative, as no biomechanical measurements were made and a hypothesis was suggested for future research.

M-root isthmuses in the mandibular sample of teeth of the current study were most prevalent apically (males n = 47/47, 100.00%; females n = 29/37, 78.38%; right n = 38/43, 88.37%; left n = 38/41, 92.68%). A similar finding was reported by Fan et al.¹⁹ in the same root of Chinese individuals (86%; n = 60/70). However, an accurate comparison cannot be made as these authors did not report on sex or arch side.

Findings from the current investigation would suggest that isthmuses can be present in any root and at any level. Looking at maxillary first molars, the MB root had the highest prevalence (88.13%) which is comparable to the findings of Estrela et al.²⁷ (93.5% reported in the same teeth of Brazilians) where the authors also used micro-CT. In contrast, a lower prevalence of 68.18% was reported by Spagnuolo and co-workers³¹ in an Italian sample but this finding should be interpreted with caution due to small sample size (n = 22).³²

The markedly higher DB- and P-root prevalences found in the present investigation (approximately 50.5% and 39.7% respectively) compared to some earlier reports deserve further comment. In particular, the relatively high prevalence of palatal-root isthmuses deserves attention because the palatal root has traditionally been regarded as anatomically less complex than the buccal roots. Findings from available literature on the prevalence of isthmuses in the DB and P roots of maxillary first molars are scarce but in one Japanese study, the authors recorded only 6–10% (DB root) and 3–6% (P root).³³ Importantly, these Japanese investigators only evaluated the apical 6

mm in comparison to the present study where the entire root was assessed. This important difference in root lengths that were examined between the two studies highlight the difference between methodologies rather than a true population difference. Also, isthmuses are not distributed uniformly along the root, restricting observations to the apical portion inevitably excludes coronal and mid-root communications, particularly in the DB and palatal roots where isthmuses may occur at multiple levels. Consequently, whole-root analysis is expected to produce higher overall prevalence estimates.

The consistent dominance of the Type III (mixed) isthmus configuration across nearly all categories and roots was seen in the present study. This type of isthmus is one where separate and partial communications co-exist along the root (Figure 2). The predominance supports the view this type of isthmus is not a single discrete structure but rather a longitudinally variable web of connections whose completeness changes along the length of the root. Naturally, such a configuration would become more common when, for example, two main root canals in a root converge apically. This phenomenon would explain the higher observed apical concentration of Type III isthmuses. Similarly, earlier work found a similarly high Type III prevalence.³⁴ The Type II isthmus was also relatively common in mesial roots of the current investigation in comparison to other work, a difference that may again reflect classification thresholds and root length examined rather than true biological divergence.

The age-related increase in D-root isthmuses (p = 0.004 overall; p = 0.027 apically) is biologically plausible but should be interpreted with care. Continued secondary dentine deposition and canal narrowing with age can alter internal morphology.³⁵ Therefore, classifying an isthmus in older individuals may reflect age-related remodelling rather than a genuinely higher developmental prevalence. Also, a cross-sectional sample derived from skeletal remains cannot clearly separate a true cohort difference from this remodelling effect. In addition, the modest statistically significant effect sizes (adjusted V 0.29–0.39, both with confidence intervals reaching zero) and small older-age subgroups reinforce the fact that results should be interpreted with caution. The age-related finding of this paper should be best framed as a suggested hypothesis about age-related morphological changes to be further explored in longitudinal or larger samples.

The current study revealed several results that reached nominal significance while carrying only small effect sizes, and no multiple-comparison correction was applied across the many subgroup analyses. In incidences where the adjusted Cramér’s V confidence interval spanned zero (for all apart from the male apical M-root association) the difference should be treated as exploratory considering the ex vivo design, a single population group and sparse cells in several tables. In summary, the data formulated from this investigation support an interpretation centred on anatomical awareness and treatment planning rather than direct, quantified improvements in outcome.

Accurate comparison with some earlier studies proved difficult. For example, in one Brazilian sample the canal’s exact point of origin for evaluating isthmuses was not specified, allowing the possibility of subjectivity during assessment, whereas the present study used defined landmarks as described by Jonker et al.⁵ Other variation stems from methodology, geographical setting and population.^18,23,28 Micro-CT literature on isthmus prevalence across adult age groups is particularly scarce. Liu et al.³⁶ studied maxillary first molars in nine- to twelve-year-old individuals and, similar to the present study, the highest prevalence was noted in the MB root. These authors used CBCT, a younger age range of individuals and only the MB root was considered. Once again, these variables make accurate comparison challenging.

The present study has several limitations. Firstly, dried skulls were used, and the presence of micro-cracks though preservation-related change could potentially interfere with observations. However, accurate identification of isthmuses was maximised using multi-slice editing.⁹ Secondly whole skeletonised maxillae and mandibulae were scanned to preserve the natural anatomical relationships and avoid extraction-induced damage to the roots, thereby providing a more representative assessment of root canal anatomy for instance to allow inter-arch comparison. However, scanning intact jaws may require a larger field of view than isolated teeth, which can influence achievable image resolution (40–74 µm voxel size) and the detection of very fine anatomical structures. Although even 75–80 µm is reported to give sufficient diagnostic accuracy,³⁷ individually extracted teeth can be scanned at much smaller voxel sizes (<10–30 µm).⁹ These methodological differences should be considered when comparing the present findings with studies based on individually scanned extracted teeth. It is also important to note that because of the relatively wide range of voxel sizes used in the current investigation, its influence on detection was examined by comparing the highest- against the lowest-resolution scans. No systematic difference was noted. Thirdly, potential clustering is possible where one individual contributed teeth to more than one subgroup. Also, the fact that within each arch most individuals contributed a single molar of the relevant type can limit but not entirely remove statistical non-independence. By analysing arches separately and unanimously the effect of clustering was reduced to a minimum. Finally, as a convenience sampling method was used on a single South African population without direct comparison to others, this sample may not fully represent the wider population. Black South Africans are a heterogeneous group consisting of several ethnic groups and where admixture may be present but is not recorded in the metadata.³⁸ Caution should therefore be exercised when generalising the reported prevalence estimates beyond the sampled collections, despite similar or smaller samples having been used elsewhere.³⁹

5. Conclusion

Root canal isthmuses were found in all roots of the mandibular and maxillary first molars examined in this Black South African sample. Prevalence varied between teeth and roots, but isthmuses occurred in every region (coronal, midroot and apical), most commonly apically. Prevalence was higher in males in the mandibular molars, whereas the maxillary distribution between sexes was more even except in the DB root. The Type III isthmus was the most common type across arch side, sex and age, and prevalence showed a variable, generally increasing trend with age. Except for the male apical M-root association, the significant differences were of small magnitude and should be regarded as a calculated hypothesis for future investigations.

Rather than claiming a direct outcome improvement when these individuals are endodontically treated, the findings of this study serve as anatomical evidence that may support case assessment, disinfection strategies and endodontic education within the South African context. Black South Africans is the largest population group in South Africa (approximately 79%).⁴⁰ The findings can, although more cautiously, contribute towards a wider global context given increasing population mobility.³⁴ As micro-CT studies across populations remain scarce and few report prevalence separately by arch, sex and age, future work might include multicentre studies comparing matched samples using a common protocol and root length; longitudinal or age-related designs distinguishing developmental prevalence from age-related remodelling; scanning of individually extracted teeth at the highest possible resolution and studies linking isthmus presence to measurable treatment outcomes.

References

1. Vertucci FJ. Root canal morphology and its relationship to endodontic procedures. Endod Topics. 2005;10(1):3-29.
2. Versiani MA, Ordinola-Zapata R, Keleş A, Alcin H, Bramante CM, Pécora JD, et al. Middle mesial canals in mandibular first molars: a micro-CT study in different populations. Arch Oral Biol. 2016;61:130-137.
3. Rathmann H, Reyes-Centeno H. Testing the utility of dental morphological trait combinations for inferring human neutral genetic variation. Proc Natl Acad Sci U S A. 2020;117(20):10769-10777.
4. AL-Rammahi HM, Chai WL, Nabhan MS, Ahmed HMA. Root and canal anatomy of mandibular first molars using micro-computed tomography: a systematic review. BMC Oral Health. 2023;23:339.
5. Jonker CH, Lambourn G, Oettlé AC, Foschi F, Theye C, L’Abbé EN. A new methodology to determine the orifice for root canal configurations in first permanent molar root and canal morphologies using micro-computed tomography. J Clin Med. 2024;13(1):71.
6. Keleş A, Keskin C. A micro-computed tomographic study of band-shaped root canal isthmuses, having their floor in the apical third of mesial roots of mandibular first molars. Int Endod J. 2018;51(2):240-246.
7. Carr GB, Schwartz RS, Schaudinn C, Gorur A, Costerton JW. Ultrastructural examination of failed molar retreatment with secondary apical periodontitis: an examination of endodontic biofilms in an endodontic retreatment failure. J Endod. 2009;35(9):1303-1309.
8. Leoni GB, Versiani MA, Silva-Sousa YT, Bruniera JFB, Pécora JD, Sousa-Neto MD. Ex vivo evaluation of four final irrigation protocols on the removal of hard-tissue debris from the mesial root canal system of mandibular first molars. Int Endod J. 2017;50(4):398-406.
9. Ahmed HMA. A critical analysis of laboratory and clinical research methods to study root and canal anatomy. Int Endod J. 2022;55(Suppl 2):229-280.
10. Jonker CH, Foschi F, L’Abbé EN, Oettlé AC. A micro-computed tomographic evaluation of root canal morphology of mandibular first molars in a Black South African subpopulation. J Clin Med. 2025;14(7):2301.
11. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Int J Surg. 2014;12(12):1495-1499.
12. L’Abbé EN, Alblas A, Ackermann JHP, Bam L, Beaudet A, Bothma PN, et al. Bakeng se Afrika: digital skeletal repository: advancing biological anthropology and medical research in South Africa. In: Shapiro L, ed. How to Use 3D Printing Innovations and Digital Storage to Democratize Anatomy Education. 1st ed. Cham: Springer Nature; 2024:63-75.
13. Republic of South Africa. National Health Act 61 of 2003. Pretoria: Government Printer; 2004. Available at: https://www.gov.za/documents/acts/national-health-act-61-2003-23-jul-2004. Accessed January 14, 2026.
14. Hoffman JW, De Beer FC. Characteristics of the micro-focus X-ray tomography facility (MIXRAD) at Necsa in South Africa. Paper presented at: 18th World Conference on Nondestructive Testing; April 16-20, 2012; Durban, South Africa.
15. Westenberger P. Avizo — three-dimensional visualization framework. In: Geoinformatics 2008 — Data to Knowledge. Reston (VA): US Geological Survey; 2008:13-14.
16. Krasner P, Rankow HJ. Anatomy of the pulp-chamber floor. J Endod. 2004;30(1):5-16.
17. Roerdink JBTM, Meijster A. The watershed transform: definitions, algorithms and parallelization strategies. Fundam Inform. 2000;41(1-2):187-228.
18. Jonker CH, L’Abbé EN, van der Vyver PJ, Zahra D, Oettlé AC. A micro-computed tomographic evaluation of maxillary first molar root canal morphology in Black South Africans. J Oral Sci. 2024;66(3):151-156.
19. Fan B, Pan Y, Gao Y, Fang F, Wu Q, Gutmann JL. Three-dimensional morphologic analysis of isthmuses in the mesial roots of mandibular molars. J Endod. 2010;36(11):1866-1869.
20. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159-174.
21. Walker RT. Root form and canal anatomy of mandibular first molars in a southern Chinese population. Endod Dent Traumatol. 1988;4(1):19-22.
22. Baratto-Filho F, Fariniuk LF, Ferreira EL, Pecora JD, Cruz-Filho AM, Sousa-Neto MD. Clinical and macroscopic study of maxillary molars with two palatal roots. Int Endod J. 2002;35(9):796-801.
23. Rwenyonyi CM, Kutesa AM, Muwazi LM, Buwembo W. Root and canal morphology of maxillary first and second permanent molar teeth in a Ugandan population. Int Endod J. 2007;40(9):679-683.
24. Martins JNR, Marques D, Silva EJNL, Caramês J, Versiani MA. Worldwide analyses of maxillary first molar second mesiobuccal prevalence: a multicenter cone-beam computed tomographic study. J Endod. 2018;44(11):1641-1649.e1.
25. Keleş A, Keskin C, Alqawasmi R, Versiani MA. Micro-computed tomographic analysis of the mesial root of mandibular first molars with bifid apex. Arch Oral Biol. 2020;117:104792.
26. Harris SP, Bowles WR, Fok A, McClanahan SB. An anatomic investigation of the mandibular first molar using micro-computed tomography. J Endod. 2013;39(11):1374-1378.
27. Estrela C, Rabelo LE, de Souza JB, Alencar AH, Estrela CR, Sousa Neto MD, et al. Frequency of root canal isthmi in human permanent teeth determined by cone-beam computed tomography. J Endod. 2015;41(9):1535-1539.
28. Shakeri F, Azizi H, Razian F, Haghanifar S, Hoshyari N. Prevalence of mid-mesial canal and isthmuses in mandibular first and second molars in an Iranian population. J Res Dent Maxillofac Sci. 2019;4(1):30-35.
29. Khamnei S, Sadat-Ebrahimi SR, Salarilak S, Oskoee SS, Houshyar Y, Shakouri SK, et al. Manifestation of hemispheric laterality in chewing side preference and handedness. Bioimpacts. 2019;9(3):189-193.
30. Youssef RE, Throckmorton GS, Ellis E 3rd, Sinn DP. Comparison of habitual masticatory patterns in men and women using a custom computer program. J Prosthet Dent. 1997;78(2):179-186.
31. Spagnuolo G, Ametrano G, D’Antò V, Formisano A, Simeone M, Riccitiello F, et al. Microcomputed tomography analysis of mesiobuccal orifices and major apical foramen in first maxillary molars. Open Dent J. 2012;6:118-125.
32. Best AM, Lang TA, Greenberg BL, Gunsolley JC, Ioannidou E; Task Force on Design and Analysis in Oral Health Research. The Oral Health Statistics Guidelines for Reporting Observational Studies and Clinical Trials in Oral Health Research: explanation and elaboration. J Oral Maxillofac Surg. 2024;82(11):1475-1493.
33. Matsunaga S, Yamada M, Kasahara N, Kasahara M, Odaka K, Fujii R, et al. Tooth root cross-section variations of significance for endodontic microsurgery and predicted risk of concealed canal isthmus based on cross-sectional morphology: three-dimensional morphological analysis of Japanese maxillary first molars using micro-CT. J Hard Tissue Biol. 2019;28(2):153-158.
34. Natanasabapathy VA, Buvaneshwari S, Sneha S, Vasudevan A, Mahendran SS, Namasivayam A, et al. Prevalence and morphology of root canal isthmus in human permanent teeth using micro-computed tomography: a systematic review. Saudi Endod J. 2021;11(2):142-153.
35. Alak SG, Keleş A, Keskin C, Martins JNR, Versiani MA. Age-related changes in the morphology of the root canal system of mandibular first molars: a micro-CT study. Clin Oral Investig. 2023;27(8):4667-4675.
36. Liu YJ, Yang WD, Wang WM, Zhu YN, Lin ZT, Zhu M. Relationship between canal morphology and isthmus in mesio-buccal roots of maxillary first molars in 9- to 12-year-old children: an in-vivo cone-beam computed tomography analysis. Arch Oral Biol. 2020;112:104645.
37. Bai B, Tang Y, Wu Y, Pei F, Zhu Q, Zhu P, et al. Ex vivo detection of mandibular incisors’ root canal morphology using cone-beam computed tomography with four different voxel sizes and micro-computed tomography. BMC Oral Health. 2023;23(1):656.
38. Sapo O. Estimating ancestry among South African ethnic groups [Master’s thesis]. South Africa: University of Pretoria; 2021.
39. Versiani MA, Ordinola-Zapata R. The Root Canal Dentition in Permanent Dentition. 1st ed. Heidelberg, Germany: Springer; 2018:89-240.
40. Statistics South Africa. Census 2011: Census in brief. Pretoria: Statistics South Africa; 2012.