This cross-sectional study was performed on 662 maxillary and mandibular dental casts of 331 patients (over 9000 teeth). The patients were selected randomly from Iranian patients attending the Orthodontic Department and two private orthodontic clinics in Ahvaz, Iran. For collecting the data, all the available patients’ records, as well as their archival casts and radiographs, were consecutively evaluated until reaching the desired sample size. A total of 809 patient records (along with their casts and radiographs) were evaluated. The inclusion criteria were being of Iranian descent, 12–35 years old, and having a complete permanent dentition (except the third molars) with no more than 2 extractions. The exclusion criteria were patients with any systemic diseases affecting the teeth or syndromes, cleft lips or palates, any earlier histories of orthodontic, prosthodontic, or surgical treatments. Also excluded were patients who did not have all the permanent teeth completely (except cases of hypodontia, cases of sporadic excluded teeth, cases of one or two extracted teeth, and also except the third molars), or patients with more than two extracted teeth, and a lack of complete eruption of more than two of the existing permanent teeth (including the second molars). The other exclusion criteria were single teeth with visible restorations, caries, crown fractures, or veneers (or a history of them), and teeth that had not been fully erupted. Cases with poor cast quality or a lack of panoramic radiographs or lateral cephalograms in the patient file would be excluded. Information regarding the patients’ age, sex, and type of skeletal malocclusion (Angle classes I, II, and III) was recorded from their files, radiographs, and casts. The data collection was performed from 2018 to 2020. No patient was exposed to X-rays for this research, and all the used radiographs were archival and taken merely for treatment purposes. Protocol ethics were approved by the Research Committee of the University (ethical code: U-98142) under the Helsinki Declaration [13,14,15].

The sample size was pre-determined as 267 patients using the following formula with conservative parameters: \(n=(}^* p*(1- p) )/(}^)\) where Z = 1.96, p (prevalence) = 0.5 (as the most conservative prevalence, i.e., the prevalence yielding the maximum sample size within this formula), and d (precision) = 0.06 (as a conservative precision). The sample size was augmented to 331 patients to ensure greater precision. After obtaining the data, the average prevalence of the traits/anomalies was 20.65%; the highest prevalence was 82.18%. For these average and maximum prevalence rates, sample sizes of 175 and 157 cases would suffice respectively, indicating that the current sample size of 331 cases was adequately large.

All archival dental casts had been poured with white dental stone for orthodontic application. All 4 quarters of each patient were examined carefully by two observers (an experienced orthodontist and a trained dentist), twice each. They tried to identify the 60 traits/anomalies mentioned in Table 1 and Figs. 1 to 13 (25 dental traits that might appear in 60 teeth) [6, 16] as well as supernumerary teeth (hyperdontia), microdontia (totaling 62 traits/anomalies), and crowding. Microdontia was defined as noticeably small but normally shaped teeth [16].

A shovel-shaped incisor. Figures 1 to 13 were created by taking screenshots of an educational Android program with direct permission from the developer and owner (3D Tooth Anatomy 1.0.3; Universal Hospital LP, Richmond, Virginia, USA; developed by Dr Rami Ammoun, assistant professor at Virginia Commonwealth University, Richmond, Virginia, USA) and editing some of the screenshots using 2D image editing software (Photoshop, Adobe, San Jose, California, USA)

An incisor with a talon cusp

The Carabelli cusp is visible on the palatal surface of the mesiopalatal cusp

The fifth cusp on the distal side of a maxillary first molar (arrow). The Carabelli cusp is as well visible on the mesiopalatal cusp

The hypoconulid absence: The upper tooth is a mandibular first molar with all common cusps; the lower one is a mandibular first molar without the hypoconulid cusp

The hypocone absence: The upper image shows a maxillary first molar with 4 cusps; the lower one shows a maxillary second molar without the hypocone cusp

The deflecting wrinkle (arrow) on the mesiolingual occlusal surface of a mandibular molar

The mesial ridge (yellow arrow) and distal accessory ridge (orange arrow) of the maxillary canine

The distal trigonid crest connecting the mesiobuccal and mesiolingual ridges (orange arrow), the sixth cusp on the distal side (blue arrow), and the tuberculum intermedium on the lingual side (purple arrow)

Accessory cusps on the mesial and distal marginal ridges

The mesial and distal accessory ridges

An accessory cusp on the lingual side of a 3-cusp premolar

Of these 62 traits and anomalies, 44 were found (Table 2); the prevalence of the rest of them was 0%. For hyperdontia diagnosis, also panoramic radiographs were evaluated for possible impacted supernumerary teeth. The data spreadsheets were evaluated by the two observers as well as a third evaluator to find any inconsistencies (which were rather rare). Any cases of inconsistency were re-evaluated on dental casts by both observers. The third evaluator did not check the dental casts.

The results were collected at two levels: (A) at the quarter level (hemi-mandible/hemi-maxilla), which showed each nonmetric trait/anomaly in each quarter; (B) at the patient level, which showed each trait/abnormality plus crowding and Angle classes in individual patients.

Thirty casts were reevaluated by one of the observers about one year after the original assessment, and the interrater and intrarater agreements were calculated to be high or excellent for all the found traits and anomalies (Kappa > 0.6, P < 0.05). Descriptive statistics and Wilson 95% confidence intervals (CIs) were calculated for prevalence rates. The ages of men and women were compared using an unpaired t-test. Data were also summarized for quarters. Associations between the presence of nonmetric traits besides hyperdontia/microdontia with genders, left/right sides, and skeletal malocclusions (Angle classes I, II, and III) were assessed using a chi-square test. Correlations across dental traits were assessed using a Spearman correlation coefficient. The level of significance was set at 0.05.