Third Molar Coronectomy vs Total Removal in Second Molar Periodontal Healing

Abstract

Objectives

The objective of this randomised controlled trial was to compare the periodontal healing at the distal of the adjacent second molar after coronectomy or surgical removal and the surgical morbidities of the 2 techniques.

Methods

This is a pilot study of a split-mouth randomised clinical trial comparing the periodontal healing of the adjacent second molar after coronectomy or total removal of lower third molars, which was registered with the University of Hong Kong Clinical Trial Centre as HKUCTR-2948 on 21 February 2020. Cone beam computed tomography (CBCT) scans were taken preoperatively and at 6 months postoperatively. Changes in the bone level from the cemento-enamel junction, periodontal probing depth (PPD), and clinical attachment level (CAL) at the distal of the adjacent second molar were compared. Surgical morbidities in terms of pain, infection, neurologic deficit, root exposure, and need for second surgery for removal were assessed.

Results

Forty patients (22 male, 18 female) with mean age of 26.7 years were evaluated at 6 months postoperatively. Both coronectomy and total removal groups demonstrated statistically significant reduction in PPD and CAL. There was an alveolar bone gain of 1.28 ± 0.88 mm in the coronectomy group and 1.13 ± 0.73mm in the control group, which was statistically significant (P < .001 for both groups).

Conclusions

Early findings show the periodontal healing at the adjacent second molar of coronectomy was as good as that of total removal of lower third molar. There were no differences in terms of surgical morbidities between both groups.

Introduction

Coronectomy of the lower third molar is the surgical removal of the crown with intentional retention of the root. The initial aim of this technique was to reduce the risk of damaging the inferior alveolar nerve (IAN).1, 2, 3, 4 It is an alternative to the total removal of impacted third molars, which are a common occurrence.⁴ The adjacent second molar is often implicated by localised periodontitis caused by the presence of the lower third molar.⁵ Although periodontitis is a common indication for lower third molar surgery, some studies have shown that this may cause persistent postoperative periodontal defects.6, 7, 8, 9 To circumvent this problem, investigators have found solutions with varying degrees of success, including the use of regenerative materials such as autogenous bone,^10,11 allograft,^12,13 xenograft,¹⁴ bioactive glass,¹⁵ and resorbable and nonresorbabable membranes.16, 17, 18, 19, 20

Coronectomy was proven to be effective at preventing postoperative IAN deficit,¹ with complications such as pain and infection occurring at incidences similar to or lower than those of total removal.²¹ Previous prospective studies found that bone regeneration after third molar coronectomy was substantial and may be beneficial to the periodontal healing of the adjacent second molar.^22,23

This split-mouth, randomised controlled trial aims to compare the periodontal and bone regeneration at the adjacent second molar as well as the surgical morbidities of coronectomy and total removal of the lower third molars. This design would minimise bias and confounding factors, which could affect the accurate interpretation of outcomes. The initial postoperative 6-month results are presented in this paper.

Methods

Trial design

A split-mouth, randomised controlled trial was conducted at the Department of Oral and Maxillofacial Surgery from June 2020. This study followed the Declaration of Helsinki for research involving human participants, and written informed consent was obtained from all participants. Ethical approval was obtained from the Institutional Review Board (IRB) of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (IRB reference number UW20-083).

Inclusion and exclusion criteria

Patients with bilateral impacted lower third molars indicated for removal were screened. The inclusion criteria were as follows:

• 1.Both lower third molars were partially erupted or fully unerupted.
• 2.Both lower third molars were horizontally or mesio-angularly impacted.

Patients with one or more of the following criteria were excluded:

• 1.The lower third molar was associated with deep caries, pulpal pathology, or a cystic or neoplastic lesion.
• 2.The adjacent lower second molar was missing or unerupted.
• 3.The adjacent lower second molar was determined to have poor long-term prognosis.
• 4.The lower third molar had radiographic signs which showed high risk of inferior alveolar nerve injury.
• 5.Patients with factors which predisposed to a higher risk of infection, for example, smoking habit, diabetes mellitus, immunocompromising conditions, or undergoing immunosuppressive therapy.

Sample size determination, blinding, and randomisation

Previous studies on periodontal attachment of adjacent second molar after total removal of lower third molar have yielded contradicting results: The periodontal attachment change ranged from −2.3 mm to +1.9 mm, with more studies showing negative (increased periodontal probing depth [PPD] and clinical attachment level [CAL]) than unchanged or positive change.²⁴ It was therefore assumed that the change in PPD was 0 mm after total removal. One study conducted at this centre showed that periodontal attachment improved from 4.0 mm (SD = 1.5 mm) to 3.3 mm (SD = 1.2 mm) at 12 months after coronectomy.²⁵ With a type 1 error rate of 5% and 90% power, 51 participants were required in each group.

The randomisation list was generated by a research assistant using a computer programme. A patient with bilateral lower third molars was randomised with one side (left or right) entering the coronectomy (experimental) group and the contralateral side entering the total removal (control) group by simple randomisation. The allocation sequence was concealed from the surgeon in numbered, opaque, and sealed envelopes kept by the research assistant. The sealed envelope containing the allocation was opened by the research assistant after consent was obtained from the participant and the surgeon was notified. The participants and clinical outcome assessor were blinded to the allocation. Two radiographic examiners were recruited. They were blinded for preoperative assessment of the cone beam computed tomography (CBCT); however, it was impossible to blind the CBCT evaluation at 6 months postoperatively as the root of the tooth which had undergone coronectomy could be visualised.

Surgical procedure

The surgeries were performed under local anaesthesia by a single operator. The surgical procedures are described below.

Study group: coronectomy

The technique of coronectomy has been described in previous studies.^22,25, 26, 27, 28, 29 A 3-sided mucoperiosteal flap was raised. The lingual aspect was protected with a periosteal elevator. Buccal and distal bone were guttered. The lower third molar was decrowned along the cemento-enamel junction (CEJ) with a fissure bur. The root of the lower third molar was trimmed down to 3 mm below the bony margin. The distal aspect of the adjacent second molar was debrided with a hand scaler. Primary wound closure was done with resorbable polyglactin sutures (Vicryl, Ethicon, Inc., Johnson and Johnson Co.).

Control group: total removal

A 3-sided mucoperiosteal flap was raised. The lingual aspect was protected with a periosteal elevator. Buccal and distal bone were guttered. The lower third molar was sectioned and elevated. The distal aspect of the adjacent second molar was debrided with a hand scaler. Primary closure was done with resorbable polyglactin sutures (Vicryl, Ethicon, Inc., Johnson and Johnson Co.).

Postoperatively, both groups received analgesics (Panadol 1 g 4 times per day as needed and ibuprofen 400 mg 3 times a day as needed if not allergic) for 3 days and a 5-day course of antibiotics (amoxycillin 250 mg 3 times a day if not allergic). The sutures were removed at the second week postoperatively.

Outcome evaluation

The primary outcomes were the changes in PPD and CAL as indicators of periodontal attachment regeneration and changes in the bone level distance from the CEJ as an indicator of bone regeneration of adjacent lower second molars in the first postoperative 6 months. Secondary outcome measures were surgical morbidities including pain, infection, inferior alveolar nerve and lingual nerve deficit, root exposure and need for second surgery for removal, and the analysis of factors including age, sex, the angulation of third molar impaction, and their relationships with the changes in the periodontal parameters.

Preoperative and postoperative assessment of the PPD and CAL were performed at 3 sites on the distal aspect of the adjacent lower second molars: the disto-lingual, mid-distal, and distal-buccal using a periodontal probe. A CBCT scan with a small field of view (FOV) was taken preoperatively and at 6 months postoperatively. The CBCT scanning was performed with a Planmeca ProMax 3D Mid (Planmeca Oy). All CBCT images were recorded at 90 kV and 5.6 mA using a preselected FOV with an exposure time of 4 seconds. The CBCT scans were taken in the ultra-low-dose mode. The FOV for the scans was (diameter × height): 4 × 5 cm (once, when 1 side, 1 jaw; twice, when 2 sides, 1 jaw). The data were reconstructed with slices at an interval of 0.5 mm, and a basic voxel size of 0.20 mm was used for evaluation. The datasets of the images were stored in DICOM format and imported to a dedicated software programme (Planmeca Romexis Version 4.4.0.R, Planmeca Oy). For the assessment of alveolar bone level, the second molar was set in an upright position in coronal and sagittal views. Then, the axial view was adjusted to the level of the pulpal floor, with the alveolar ridge set to vertical. In the sagittal view, a vertical line was drawn from the CEJ to the alveolar crest level. The measurement was done in 3 sites—disto-buccal, mid-distal, and disto-lingual—by shifting the image via sagittal view. By comparing the measurements of preoperative and postoperative images, the alveolar bone level change was calculated. This method of measurement is shown in Figure 1.

Fig.

Radiographic evaluation of bone regeneration distal to second molar. Supplementary: CONSORT flow diagram.

Postoperative symptoms and adverse events

The secondary outcomes were surgical morbidities including pain, infection rate, inferior alveolar nerve and lingual nerve deficit rate of the 2 groups, the root exposure rate, and the need of reoperation to remove the retained root in the coronectomy group at 6 months postoperatively. The visual analogue scale (VAS) was used to determine the pain and numbness score.

Statistical analysis

Data were analysed with the Statistical Package for Social Sciences (SPSS version 27.0 SPSS Inc.). Chi-squared test was used to determine whether there were any differences between the coronectomy and control group in terms of the side which was chosen to undergo surgery and the angulation of the tooth. Paired t test was used to assess the differences of the means between the 2 groups for pre- and postoperative PPD, CAL, and bone level distance from the CEJ of the distal of the adjacent lower second molar as well as the mean changes. A P value of .05 was considered statistically significant. Intraclass correlation coefficient was used to evaluate interexaminer agreement on the radiographic assessment.

Results

Fifty-two patients were recruited and underwent the surgical procedure. Forty patients attended the follow-up at 6 months postoperatively; 22 were male and 18 were female. Patient demographics are summarised in Table 1. The mean age was 26.7 years (SD = 4.6 years, range = 19.5 to 38.7 years) at the time of surgery. Further, 52.5% were 25 years old or younger, and 47.5% were older than 25 years. There were no significant differences between the coronectomy and the control group in terms of the side (left or right) which was chosen to undergo surgery (P = .654), and the angulation of the tooth (mesio-angular or horizontal) (P = 1.000). The follow-up time was 6 months. The Consolidated Standards of Reporting Trials (CONSORT) diagram is shown in the supplementary figure.

Table 1.

Patient demographics and details of the coronectomised lower third molars.

Variable	Class	Coronectomy group	Control group	P value
Age, y, mean (SD)	≤25 >25	26.7 (4.6) 52.5% (21) 47.5% (19)		N/A
Sex	Male	55% (22)		N/A
	Female	45% (18)
Tooth	38	47.5% (19)	52.5% (21)	.654
	48	52.5% (21)	47.5% (19)
Angulation	Mesio-angular	20% (8)	20% (8)	1.00
	Horizontal	80% (32)	80% (32)

N/A, not applicable.

The pre- and postoperative bone level distances from CEJ, PPD, and CAL are shown in Table 2. The crown of partially erupted third molar prevented accurate assessment of the mid-distal of the adjacent second molar during clinical examination; thus this value was not presented as it would lead to underestimation of the mean initial PPD and CAL. Both coronectomy and control groups showed statistically significant reductions in disto-buccal, disto-lingual, and mean values of bone level distance from CEJ, PPD, and CAL, indicating improvement in all the periodontal parameters. The intraclass correlation coefficient value was 0.916, which represents excellent repeatability. The amount of reduction for each parameter in both groups are shown in Table 3. The coronectomy group showed a slightly larger mean alveolar bone gain of 1.28 ± 0.88 mm as compared to the control group (1.13 ± 0.73 mm); however, this was not statistically significant (P = .375). Similarly, the coronectomy group showed slightly better improvements in mean PPD (1.43 ± 2.43 mm) and mean CAL (1.29 ± 2.31) compared to that of the control group (1.29 ± 1.94 mm and 1.16±1.90 mm, respectively), which were not statistically significant.

Table 2.

Preoperative and postoperative bone level distance from the CEJ, PPD, and CAL.

Variable	Coronectomy Group			Control Group
	Preoperative	Postoperative	P value	Preoperative	Postoperative	P value
Bone level from CEJ
Disto-buccal	5.46 (1.98)	4.20 (1.59)	<.001	5.38 (1.63)	4.31 (1.62)	<.001
Disto-lingual	5.23 (2.10)	3.93 (1.75)	<.001	5.83 (1.62)	4.08 (1.54)	<.001
Mean	5.50 (1.92)	4.22 (1.58)	<.001	5.52 (1.59)	4.39 (1.47)	<.001
PPD
Disto-buccal	4.83 (2.00)	3.60 (1.88)	.010	4.90 (2.00)	3.63 (1.58)	.006
Disto-lingual	4.53 (1.95)	2.90 (1.34)	<.001	4.45 (1.52)	3.15 (1.47)	.001
Mean	4.68 (1.60)	3.25 (1.49)	<.001	4.68 (1.38)	3.39 (1.17)	<.001
CAL
Disto-buccal	5.03 (1.94)	4.00 (1.91)	.022	5.13 (2.15)	3.98 (1.66)	.013
Disto-lingual	4.63 (1.97)	3.08 (1.44)	<.001	4.53 (1.49)	3.35 (1.67)	.003
Mean	4.83 (1.58)	3.54 (1.51)	.001	4.83 (1.38)	3.66 (1.29)	<.001

Values are mean (SD).

CEJ, cemento-enamel junction; PPD, periodontal probing depth; CAL, clinical attachment level.

Table 3.

Reduction in bone level distance from CEJ, PPD, and CAL after 6 months.

Variable	Coronectomy group	Control group	P value
Bone level distance from CEJ
Disto-buccal	1.26 (1.09)	1.07 (0.75)	.307
Disto-lingual	1.31 (1.19)	1.30 (1.15)	.986
Mean	1.28 (0.88)	1.13 (0.73)	.375
PPD
Disto-buccal	1.22 (2.88)	1.28 (2.76)	.919
Disto-lingual	1.63 (2.59)	1.30 (2.33)	.527
Mean	1.43 (2.43)	1.29 (1.94)	.702
CAL
Disto-buccal	1.03 (2.71)	1.15 (2.81)	.797
Disto-lingual	1.55 (2.62)	1.18 (2.36)	.459
Mean	1.29 (2.31)	1.16 (1.90)	.708

Values are mean (SD).

CEJ, cemento-enamel junction; PPD, periodontal probing depth; CAL, clinical attachment level.

Subgroup analyses were performed to assess the impact of a patient's sex, age, and angulation of the lower third molar on changes in the periodontal parameters for both groups, with results shown in Table 4. There was no statistically significant difference in the mean reduction of all the parameters for patients aged 25 years old or younger or older than 25 years in both coronectomy and control groups. In the coronectomy group, males showed a greater amount of alveolar bone gain (1.59 ± 0.87 mm) compared with females (0.91 ± 0.76 mm), which was statistically significant (P = .014). For males, the coronectomy group also showed greater alveolar bone gain than the control group (1.13 ± 0.58 mm; P = .035). In the coronectomy group, there was larger alveolar bone gain when the tooth was horizontal (1.42 ± 0.92 mm) as compared to mesio-angular (0.73 ± 0.40 mm; P = .004).

Table 4.

Bone levels, PPD, and CAL (mm) changes at the distal aspect of the adjacent second molar, according to sex, age, and angulation of the third molar impaction.

	Reduction in bone level distance from CEJ			PPD reduction			CAL reduction
	Coronectomy group	Control	P value	Coronectomy group	Control	P value	Coronectomy group	Control	P value
Sex Male Female P value	1.59 (0.87) 0.91 (0.76) .014	1.13 (0.58) 1.12 (0.90) .983	.035 .461	1.45 (2.38) 1.39 (2.55) .933	1.18 (2.21) 1.42 (1.62) .709	.530 .964	1.42 (2.43) 1.25 (2.22) .927	1.07 (2.17) 1.28 (1.56) .733	.544 .961
Age, y ≤25 >25 P value	1.33 (0.89) 1.23 (0.90) .730	1.25 (0.50) 1.02 (0.89) .327	.644 .459	1.00 (2.17) 1.81 (2.63) .298	1.39 (1.90) 1.19 (2.03) .745	.407 .249	0.92 (2.30) 1.62 (2.32) .346	1.18 (1.81) 1.14 (2.02) .946	.604 .289
Angulation Horizontal Mesio-angular P value	1.42 (0.92) 0.73 (0.40) .004	1.24 (0.73) 0.68 (0.54) .050	.387 .810	1.40 (2.56) 1.56 (1.94) .861	1.09 (2.10) 2.06 (0.86) .212	.614 .520	1.17 (2.46) 1.75 (1.58) .533	0.97 (2.05) 1.94 (0.82) .201	.721 .772

Values are mean (SD).

CEJ, cemento-enamel junction; PPD, periodontal probing depth; CAL, clinical attachment level.

No patients from either group presented with pain, swelling, infection, or inferior alveolar nerve and lingual nerve deficit at both 2 weeks and 6 months postoperatively. No patients from the coronectomy group presented with root exposure at 6 months postoperatively.

Discussion

The key findings were as follows: (1) Both coronectomy and control groups resulted in modest improvement in bone level distance from CEJ, PPD, and CAL at 6 months; (2) there was no statistically significant difference between both groups in the improvement of these periodontal parameters at 6 months; and (3) age was not a factor affecting periodontal parameters in either group.

Some studies found that lower third molar surgery resulted in persistent postoperative periodontal defects.6, 7, 8, 9 Whilst regenerative periodontal therapy may produce improvement in periodontal parameters,³⁰ some authors found that there was no predictable benefit³¹ or evidence that any regenerative procedure was superior.³⁰ Furthermore, these materials may not be affordable for the average patient. Vignudelli et al²³ found that the gain in alveolar bone after coronectomy was comparable to that of total removal plus regenerative periodontal therapy rather than that of total removal without other treatment. Healthy periodontium at the adjacent second molar was maintained when these patients were followed up for 3 years.³² However, there was no control group in this study. In addition, periapical radiographs were used to assess bone levels, which may cause flawed measurements due to image distortion. A previous study using CBCT for evaluation found that with coronectomy, more than 3 mm of bone was regenerated, reducing the bone level distance from CEJ from around 5 mm to 2 mm. However, this observation was made at a much longer follow-up time of 54 to 108 months postoperatively.²² This pilot study was the first randomised controlled trial comparing coronectomy and total removal of the lower third molar with a split mouth design, which minimises bias and confounding. The use of CBCT with a standardised protocol to evaluate bone levels allowed for a 3-dimensional assessment which was reproducible, with excellent interexaminer agreement. Initial results showed that in the intermediate term, although the coronectomy group showed a very slight benefit over total removal group for all the periodontal parameters, these differences were not statistically significant. Previous authors postulated that the improvement in periodontal parameters could be due to root migration, which brings tissue growth from the periodontal ligament²² in a way similar to orthodontic extrusion.³³ Another hypothesis was that coronectomy may result in less bone removal intraoperatively.²³ Leung and Cheung²⁶ found that the root tended to migrate within the first 2 years postoperatively. Therefore, a longer follow-up would be needed to assess the impact of coronectomy on periodontal parameters.

Age was proposed as a factor affecting periodontal healing as older individuals were more frequently found to have periodontal defects following third molar surgery.^8,9,34, 35, 36 Kugelberg et al⁸ found that patients older than 25 years had reduced periodontal attachment gain at the adjacent second molar after lower third molar surgery. In this study, there was no statistically significant difference in the mean reduction of all the parameters for patients aged 25 years or younger or older than 25 years in both groups. This finding agreed with the studies which found that bone regeneration after coronectomy was not affected by age.^22,32

Male patients had a greater reduction of bone level distance from CEJ (1.59 ± 0.87 mm) as compared to female patients (0.91 ± 0.76 mm) in the coronectomy group. Whilst a previous study found that male patients had more bone regeneration, this was attributed to a worse preoperative periodontal condition.²² In this study, the baseline values of bone level distance from CEJ were similar for male (5.76 ± 1.99 mm) and female patients (5.19 ± 1.82 mm), with no significant difference between the 2 groups (P = .355). Coronectomy seemed to be beneficial in male patients, as this group had a larger reduction of bone level distance from CEJ compared to the control group (1.13 ± 0.58 mm; P = .035). There was no significant difference in baseline values between the coronectomy group and the control group (5.73 ± 1.74; P = .927). Whilst males may have greater bone regeneration in the coronectomy group, it must be noted that the difference is small and may not be clinically significant. It remains to be seen in longer follow-up whether this difference would increase and whether there are any possible reasons for this finding.

Third molars which were horizontally impacted showed greater bone regeneration after coronectomy (1.42 ± 0.92 mm) than those which were mesio-angularly impacted (0.73 ± 0.40 mm; P = .004). These results are contradictory to a previous coronectomy study that found that the direction of impaction had no impact on bone regeneration.²² Other studies had also found that third molars with mesio-angular impaction^34,37,38 had higher PPD reductions at 6 months postoperatively. Pons-Vicente et al³⁹ found that 95.8% of lower third molars with 50 to 90 degrees angulation were associated with preoperative vertical bone loss of more than 4 mm, which is true for only 47.4% of those with less angulated (30 to 50 degrees) third molars. In the coronectomy group of this study, third molars that were horizontally impacted had a higher baseline bone level distance from CEJ (5.75 ± 1.69 mm) compared to mesio-angularly impacted third molars (4.50 ± 2.55 mm); however, this difference was not statistically significant (P = .098). It must be noted that in this study, the proportion of mesio-angular third molars compared to horizontal third molars was small.

Regarding surgical morbidities, no patients from either group presented with pain, swelling, infection, inferior alveolar nerve or lingual nerve deficit at both 2 weeks and 6 months postoperatively. No patients from the coronectomy group presented with root exposure at 6 months. Previous studies had shown that coronectomy was safe in the short and long term.²⁶

One of the main limitations of this pilot study is the large number of dropouts. This study was carried out during the COVID-19 pandemic, and the main reasons cited for dropout were inability to contact patient, COVID-19 infection at the time of scheduled follow-up, emigration, and time conflict with work.

Conclusions

Coronectomy resulted in improvement in bone level distance from CEJ, PPD, and CAL which was as good as that of total removal of the lower third molar at 6 months postoperatively. There was no statistically significant difference between both groups in terms of the improvement in these periodontal parameters. Male patients appeared to have a greater amount of bony regeneration in the coronectomy group. Age was not a factor in determining the improvement of periodontal parameters in either the coronectomy or the control group. The final results would be presented after longer follow-up.

Funding

This study is supported by the General Research Fund (Project Ref No: 17119520), Research Grant Council, Government of the HKSAR. The sponsors had no involvement in the study design, collection, analysis, and interpretation of data; in the writing of the manuscript; or in the decision to submit the manuscript for publication.

Author Contributions

Si Ling Pang: conceptualisation, data curation, formal analysis, investigation, methodology, project administration, writing–original draft, writing–review and editing. Wai Kan Andy Yeung: conceptualisation, investigation, methodology, project administration. Kuo Feng Hung: data curation, formal analysis. Liuling Hui: data curation, formal analysis. Hui Zhen Jasmine Chung: data curation, methodology. Yiu Yan Leung: conceptualisation, funding acquisition, supervision, writing–review and editing. All authors reviewed the manuscript. All authors have viewed and agreed to this submission.

Conflict of interest

None disclosed.