This retrospective study involved 1249 panoramic radiographs of subjects aged 20 to 44 years represented by the three Malaysian ethnicities (Malay, Chinese and Indian) attending the Oral Health Center/SEGi University, Faculty of dentistry, Malaysia. Of the study population 918 OPGs were considered for the study. Patient’s details include age, gender and ethnicity were extracted from the patient’s clinical record and all details related to impaction were obtained from patient’s panoramic radiograph that was individually examined by two investigators. Patients without OPG and clinical data were excluded from the study. The ethical clearance was obtained from the university ethical committee.

The difficulty scores related to impacted 3Mm’s were recorded based on difficulty index proposed by Pederson [10], which include the sum of spatial relationship, depth and ramus relationship with respected to impacted 3Mms. This was further correlated with the patient clinical factors (age, gender and ethnicity respectively).

Data collected was entered into a spreadsheet (Excel 2010; Microsoft, US) and analysed subsequently using SPSS version 16.0. The prevalence and pattern of impacted third molars in relation to age, gender, ethnicity and type of impaction were assessed and displayed by frequency and percentage. p-value was assessed through Pearson Chi-square test and FISHER exact test.

Out of 1249 panoramic view, 918 (73.5%) were considered for the study showing 1812 impacted 3Mm (fully or partially impacted), along with complete patient’s data. The difference between the sexes regarding the impacted 3Mmwas found to be significant, women had more impacted molar i.e., 81.3% (n=746) than men 18.7% (n=172). The prevalence of impacted 3Mm was highest in the Chinese patients 63.4% (n=582) as compared to Malay 28.1% (n=258) and Indian patients 8.5% (n=78) in this study. The most frequent age group was 20-24 years showing 66% (n=606) of 3Mm impaction, whereas least frequent impaction i.e., 1.5% (n=14) was observed in 40-44 years age group [Table/Fig-1].

Based on the spatial relationship, 56% (1015) 3Mms showed mesioangular (MA) impaction, followed by 25% (n=453) 3Mms showing horizontal impaction, 13% (n=236) were vertically impacted and only 6% (n=108) were distoangular (DA) impacted [Table/Fig-1]. Result was significantly different as p-value was 0.023. Based on the depth index, 60.1% (n=1089) 3Mm were at the deep occlusal level, 30.3% (n=549) of 3Mm were at the medium occlusal level and 9.6% (n=174) of 3Mm were at the high occlusal level. Based on the ramus relationship or space available, 63% (n=1141) of 3Mm’s had no space, 34.2% (n=620) of 3Mm’s had reduced space and only 2.8% (n=51) of 3Mm’s had sufficient space [Table/Fig-1].

The level of difficulty encountered during extraction of impacted 3Mms was categorized into very difficult, moderately difficult and slightly difficult category and the numbers of cases falling under each category were 69% (n=632), 25% (n=232) and 6% (n=54) respectively. Further very difficult cases were highest in all the ethnic groups [Table/Fig-2]. To maintain the reader’s interest and to avoid ambiguity in the current study we will be discussing only the very difficult cases from here.

For the Malay ethnicity for very difficult cases, we found female scored higher prevalence of 76.2% (n=128) for impacted 3Mm’s in 20-24-year-old [Table/Fig-3,4 and 5]. Horizontal impaction was the highest among this ethnicity accounting for 46.4% (n=153) [Table/Fig-3,6]. Among the depth index, most of the impacted 3Mm’s were at deep occlusal level and p-value was found to be significant 84.5% (n=279) [Table/Fig-3,7]. An 81% (n=267) of the impacted 3Mm’s had no space with ramus [Table/Fig-3,8].

[Table/Fig-4]:

Illustrating the demography among Malaysian ethnicities

[Table/Fig-5]:

Illustrating the difficulty index distribution among age ranges

[Table/Fig-6]:

Illustrating the difficulty index distribution among spatial index

[Table/Fig-7]:

Illustrating the difficulty index distribution according to tooth impaction level

[Table/Fig-8]:

Illustrating the difficulty index distribution according to impacted distance from ramus

Among Chinese ethnics again female had higher prevalence of 83.9% (n=334) for very difficult cases [Table/Fig-4,9]. Age range was significantly high (74.9% (n=582) for 20-24-year-old with p-value 0.044 [Table/Fig-5,9]. MA impaction scored highest 41.7% (n=324), p=0.001 [Table/Fig-6,9]. For the depth level, deep occlusal was significant with p=.001 74.9% (n=582) [Table/Fig-7,9]. Most of very difficult cases were where no space with ramus was found significantly in 78.9% (n= 613), p=0.001 [Table/Fig-8,9].

The Indian ethnic group also showed that female scored higher prevalence of 81.8%(n=54) for very difficult impacted 3Mms in age group of 20-24 years old [Table/Fig-4,5,10]. MA impaction was highest 54.5% (n=76) [Table/Fig-6,10]. For the depth level, deep occlusal was highest 54.5% (n=76) and difficult index were significant where no space with ramus was observed in 87.9% (n=122) impacted 3Mm’sp=0.006 [Table/Fig-7,8,10].

A 3Mm’s are the most commonly impacted teeth in the oral cavity followed by maxillary third molar, maxillary canine and mandibular canine [12]. Surgical extraction of these impacted 3Mm is one of the most common oral-surgical procedures and is often associated with a varying degree of difficulty that can be related of pre-operative factors. Preoperative assessment of the difficulty of the 3Mm surgery is the most important factor that should be considered [13]. The present study showed the overall prevalence of 73.5% of impacted 3Mms in Malaysian ethnicity. This was found to be 63.4% in Chinese followed by 28.1% in Malays and 8.5% in Malay Indians. The score of difficulty level for impacted 3Mm surgery was highest in Chinese group (83.9%). Prevalence rate of impacted 3ms from other studies are 67.6% by Kanneppady et al., in Malaysians [7], Quek et al., reported a frequency of 68.6% in Singaporean Chinese [8] and Sandhu and Kaur have mentioned 76% prevalence rate among Indians in their study [9]. Al-Anqudi et al., Hassan, Hattab et al., and Reddy et al., have documented lower frequencies in studies done on Omanis (54.3%), Saudi Arabians (40.5%), Jordanians (47.4%) and Indians (27%), respectively [3–6]. This variation could be due to different criteria’s (their study includes both impacted maxillary and mandibular third molar), sample size and statistical methods used.

However, when comparison was done among the various ethnic groups, the current study findings were similar to the findings of Kanneppady et al., showing higher frequency of impacted 3Mm in Chinese [7]. This can be explained on the notion put forward by Keng and Foong who opined that Chinese have tooth-jaw size discrepancy and tapered arch form with wider inter-molar width which less favourably disposes to resorption of ramus [14]. So this leads to inadequate space for eruption of 3Ms eventually more chances of their impaction. We also agree with this fact.

Our study showed significantly higher frequency of impacted 3Ms in females (81.3%) in all the ethnic groups. This is in accordance with the studies of Kanneppady et al., Quek et al., Hashemipour et al., Hugoson and Kugelberg, Ma’aita and Alwrikat, and Kim et al., [7,8,15–18]. This could be explained as jaw growth ceases in females when 3Ms began to erupt whereas in men the jaw growth continues during the third molar eruption and thus provides more space for the tooth [8]. However, study from Saudi region revealed 3ms impactions were more prevalent in males [19]. Abu Alhaija et al., and Breik and Grubor had reported no sex predilection in 3Mms impaction [20,21]. The difficulty index score was highest for females in all the three ethnic groups.

The age range of the participants in the study was 20-44 years and the most frequent (66%) age group for impaction was 20-24 years in all the three ethnicities. This was in agreement with the study of Bokhari Syed et al., and Abu Alhaija et al., [19,20]. The difficulty index score was highest for 20-24 years of age group.

The most common spatial relationship observed was MA impaction in this study. However, in Malay ethnic group horizontal impaction was more common as compare to other two ethnic groups where MA impaction was mostly observed. Others have also reported that MA inclination was the most common for impacted 3Mms in their studies [5–8]. This can be explained on the basis of proportion of the root development, underdevelopment of the mesial root may lead to MA impaction, and overdevelopment of the mesial root may lead to DA impaction. Normal rotations of 3Ms were expected to occur from horizontal to MA and from MA to vertical. Failure of rotation from MA to vertical position is very common and may result in more MA impactions [8]. However, Reddy et al., reported vertical impaction to be the most common (50.0%) in the mandible [3].

The level of impaction is very important as it gives the idea of the depth and position of impacted tooth. This aids in proper pre-surgical planning and in very difficult cases the dentist would be cautious and already prepared to face the challenge if any. This study showed that 60% of the 3Mms were observed at the level C i.e., deepest level and highest score for level C among all ethnic groups were observed for impacted 3Mms in Malay (84.5%). Whereas studies of Quek et al., showed level B and Al-Anqudi et al., showed level A position of impacted 3ms respectively [6,8].

Another important variable to foresee eruption of 3Mms is the mesiodistal space between the distal surface of 2^nd molar and anterior ramus of the mandible. This study showed that 63% of the impacted 3Mms had no space. This has been reported that if mesiodistal space is as large as the mesiodistal width of the crown, the probability of its eruption is approximately 70%. Lack of space seems to be a major cause of abortive eruption; however, eruption cannot be guaranteed, despite adequate space available in the jaw [22,23].

We studied only the impacted 3Mms and only patients attending our institution health centre so more similar studies have to be carried out in other parts of Malaysia to substantiate our present findings.

The difficulty index (Pederson index) score was highest in Chinese ethnic group and even the operating surgeon felt that 3Mm surgery was more challenging in Chinese patients and took more operating time as compare to other two ethnic groups. Thus by predicting the difficulty index preoperatively for impacted 3m’s surgery it can prepare the dentist for the complications he/she is going to encounter at the time of the surgery, thereby avoiding or minimizing the postoperative complication for the ultimate benefit of the patient.