Recovery after third-molar surgery: The effects of age and sex


In this study, we assessed the effects of age and sex on quality-of-life recovery after third-molar surgery.


Healthy subjects scheduled for removal of third molars were recruited at multiple sites for this study. Each patient was given a condition-specific instrument to be completed each postsurgery day for 14 days. Lifestyle and oral-function recovery were assessed by using a 5-point Likert-type scale. Recovery was defined as the number of days until the patient reported no or little trouble. Recovery from pain was defined as the number of days until no medications were taken. For each quality-of-life item, a Cox regression analysis was performed to assess the effects of age and sex on recovery after controlling for surgical-procedure variables.


Nine hundred fifty-eight subjects treated at 9 academic centers and 12 community practices were enrolled. Except for ability to open the mouth, recovery for all quality-of-life items for those 21 years or older significantly ( P <0.02) lagged behind recovery for younger subjects. Recovery for female subjects was significantly longer than for male subjects for all outcomes ( P <0.01).


Patients older than 21 and those who are female should be informed before removal of all 4 third molars that their oral function, lifestyle, and pain recovery will be prolonged compared with those who are younger and male.

Orthodontists often treat patients in the age range of 12 to 18 years. As a course of treatment is discussed with the patient, parents, or spouse, a recommendation concerning third molars should be made based on current data and the patient’s circumstances. Two questions are relevant. Should the third molars be removed? If the decision is third-molar removal, is it advantageous to the patient if the teeth are removed before treatment, even though the third molars might not be erupted, while the orthodontic appliances are in place, or soon after the appliances are removed? If a wait-and-see approach to third-molar management is adopted, will the patient comply with periodic monitoring of third molars for occlusal caries, periodontal inflammatory disease, and other less frequently occurring pathologies such as cysts or tumors?

Almost all young adults (95%) have at least 1 third molar, and three quarters have 4 third molars. According to Engstrom et al, third-molar crown development is usually complete by age 14 and root development by age 18, but third molars might or might not be in function at the occlusal plane (OP) by age 18. Failure of third molars to erupt to the OP has been attributed to a lack of space resulting from growth patterns of the jaw. Other authors have documented the difficulty in predicting before jaw growth is complete which third molars will erupt into function or remain impacted.

Many clinicians including orthodontists advise their patients to have third molars removed before or just after the teeth are visible in the mouth, particularly if the assumption is that the teeth will not erupt to the level of the OP. Clinicians and the public believe that recovery time after third-molar surgery is shorter at a younger age, although data are limited to support or refute the assumption. Bruce et al reported fewer clinical complications associated with third-molar surgery in younger patients. More older subjects had delayed healing and excessive pain after surgery. However, the reported data were not controlled for the complexity of the surgery, which was greater in older subjects. Chuang et al reported data from a cross-sectional analysis on subjects at least 25 years old having third-molar surgery. Increasing age in deciles was not significantly associated with increases in postsurgery inflammatory complications, wound infection, or alveolar osteitis, controlling for other explanatory variables such as the difficulty of the surgery. Phillips et al reported that, after third-molar surgery, delayed healing, wound infection, and alveolar osteitis were not associated with age at the time of surgery if these outcomes were controlled for the complexity of surgery. However, prolonged recovery as evaluated by quality of life (QOL) outcomes was significantly more likely for subjects older than 18 and for females patients.

Since active orthodontic treatment is often finished before the third molars erupt and jaw growth is complete, recommending treatment for third molars can be problematic. This study was designed to further assess the effects of age and sex on recovery of QOL outcomes, lifestyle, oral function, and pain after third-molar surgery in patients treated in either a community practice or an academic clinical center.

Material and methods

QOL after third-molar surgery and surgical data were obtained from subjects and clinicians at 21 clinical centers, 9 academic centers, and 12 community practices. Participating surgeons in community practices were fellows of the American Association of Oral and Maxillofacial Surgeons. In the academic centers, the surgeons were full-time faculty who were fellows of the American Association of Oral and Maxillofacial Surgeons or residents with at least 1 year of postdental school dentoalveolar surgical training. Patients scheduled for removal of all 4 third molars at these clinical sites were recruited between 1997 and 2009 for this clinical study approved by the institutional review board of the University of North Carolina.

Healthy subject-patients (ASA I and II), aged 14 to 40 years, were enrolled after consent for participation was obtained. Patients were excluded if they had a history of psychiatric treatment, were pregnant or lactating, or had radiographic evidence of severe periodontal disease. Presurgery baseline data were recorded, including reason for seeking third-molar removal and demographics (age, sex, ethnicity, education level). From the presurgery panoramic radiograph, mandibular third-molar position relative to the OP was categorized as 1 third molar below the OP or both below the OP. The standard surgery protocol consisted of common procedures in the United States: intravenous anesthesia, buccal access to third molars, and bone removal for the mandibular third molars by using rotary instrumentation. Bone removal was categorized as removal from 0 or 1 mandibular third molar or from both. Perioperative therapy with systemic or topical antibiotics or corticosteroids was administered at the surgeon’s discretion, but these data were not routinely recorded.

Extensiveness of the surgery was indicated by the length of the surgical procedure and the surgeon’s estimate of the degree of difficulty for each quadrant, on a scale of 1 (no difficulty) to 7 (maximal difficulty) with a possible subject total of 28. Length of surgery was categorized as ≤20, >20 to 30, >30 to 40, and >40 minutes. Degree of total difficulty was categorized as ≤9, >9 to 12, >12 to 16, and >16. The categorizations for OP position, bone removal, surgery time, and surgeon’s assessment were those used in the evaluation of prolonged clinical recovery.

After surgery, each patient was given a condition-specific QOL diary that had been developed, validated, and used in other clinical studies involving recovery after third-molar surgery. Each patient was asked to complete the 2-page diary each postsurgery day for 14 days. Recovery was organized into 3 QOL domains: lifestyle, oral function, and pain. The patient’s lifestyle and oral-function domains after surgery were assessed with a 5-point Likert-type scale with verbal anchors of “no trouble (1)” and “lots of trouble (5).” The impact of the surgery on the patient’s lifestyle included usual daily activities, social interactions, and recreation. Oral-function items assessed difficulty with mouth opening, chewing, and eating a regular diet. Recovery for lifestyle and oral function was defined as the number of days until the patient reported “no (1)” or “little trouble (2).” Based on the report of Snyder et al, the patients were asked to record whether medications, including over-the-counter ones, were taken for pain each day, and recovery from pain was defined as the number of days until no medications were taken.

Statistical analysis

Before analysis, age at surgery was stratified into 3 groups: <18, 18 to 21, and >21 years old. For each QOL item, unadjusted Kaplan-Meier estimates of the quartile values for recovery for the 3 age groups and for the sexes were calculated (version 9.1, Proc Lifetest SAS, Cary, NC) and a Cox regression time to event analysis (version 9.1, Proc PHREG SAS) was performed to assess the effects of age and sex on recovery. The hazard functions were calculated after controlling for the clinical and surgical predictor variables (OP position of the mandibular third molars, bone removal from the mandibular third molars, length of surgery, and surgical difficulty) identified by Phillips et al. The reference cells were female, less than 18 years of age, less than 20 minutes of surgery, bone removal on 0 or 1 mandibular third molar, 0 or 1 mandibular third molar below the OP, and total difficulty as assessed by the surgeon of 9 points or less. For all models, the proportional hazards assumption was assessed and considered appropriate. The level of significance was set at 0.05.


We enrolled 958 subjects who returned a completed 14-day postsurgery diary to the data center. A smaller proportion of subjects were treated in academic centers compared with community practices: 39% vs 61%, respectively. Forty-seven percent of the subjects were 21 years of age or older at the surgery (median, 24 years; interquartile range [IQR], 23-27 years); 31% were between 18 and 21 (median, 19; IQR, 18.5-20); and 21% were 18 or younger (median, 17; IQR, 16-17.5). More subjects were female (60%) and white (85%), although African Americans were well represented (8%). The subjects were well educated. Fifty-seven percent of all subjects and 74% of those 18 or more years old had at least some college education. A third of the subjects reported having third-molar symptoms before surgery that were sufficiently bothersome to prompt them to have the third molars removed “before it happens again.”

The median surgery time was the shortest for the youngest group and longest for the oldest group ( Table I ), although the average degrees of difficulty as estimated by the surgeon were highest for the youngest group and lowest for the oldest group. Both mandibular third molars were below the OP in 74% of the youngest subjects and required bone removal for access, compared with 49% of the oldest subjects. Both sexes were, on average, 21 years old at the surgery. Median surgery time and estimated degree of difficulty were slightly higher in the male subjects ( Table I ). The percentage of subjects with both mandibular third molars below the OP was slightly higher in the female subjects, but bone removal for access to the mandibular third molars was higher in the male subjects.

Table I
Descriptive statistics for demographic and surgical characteristics of subjects
Age (y) Sex
All subjects
n = 958
n = 198
n = 290
n = 439
n = 571
n = 381
n % n % n % n % n % n %
Bone removal
Both 595 63 144 74 189 66 246 42 339 60 254 68
1 or none (missing 16 subjects) 347 37 51 26 98 34 190 56 222 40 122 32
Below OP
Both 541 61 139 74 184 69 206 49 338 63 202 57
1 or none (missing 64 subjects) 353 39 49 26 83 31 214 51 200 37 151 43
Median (IQR)
Age at surgery (y) 20.7 (18.2-24.2) 16.7 (15.9-17.5) 19.2 (18.5-20.0) 24.3 (22.7-27.2) 20.7 (18.2-24.1) 20.6 (18.2-24.3)
Total surgery time (min) 30 (20-40) 25 (20-35) 27 (20-35) 30 (20-40) 28 (20-40) 30 (20-40)
Surgeon’s estimate of difficulty (4-28 possible) 13 (9-16) 14.0 (10.0-17.0) 13.0 (10.0-16.0) 12.0 (9.0-16.0) 12.0 (9.0-16.0) 13.0 (10.0-17.0)
IQR, Interquartile range (P 75 – P 25 ).

Only mandibular third molars were considered.

The median number of days to recovery for all lifestyle and oral-function items was similar for all age groups. However, except for mouth opening, the distribution in the oldest group (≥21) was shifted by at least 1 day longer for recovery time, as indicated by the estimated time to recovery for 75% of the subjects in that group ( Table II ). The recovery from pain was prolonged by an average of 2.5 days in the oldest group.

Table II
Unadjusted Kaplan Meier median and 25 th and 75 th percentile values by age and by sex for the number of days to recovery defined by “little” or “no” interference with lifestyle (regular routine, social life, recreation, sleeping), oral function (eating, chewing, opening mouth), and the number of days until no pain medications were taken
All subjects
n = 958
median (P 25 -P 75 )
<18 years
n = 198
median (P 25 -P 75 )
18-21 years
n = 290
median (P 25 -P 75 )
≥21 years
n = 439
median (P 25 -P 75 )
n = 571
median (P 25 P 75 )
n = 381
median (P 25 -P 75 )
Regular routine 3 (1-5) 3 (1-4) 3 (1-4) 3 (2-5) 3 (2-5) 2 (1-4)
Social life 3 (1-5) 3 (1-4) 3 (1-4) 3 (2-5) 3 (2-5) 2 (1-4)
Recreation 3 (2-6) 4 (2-5) 3 (2-5) 4 (2-6) 4 (2-6) 3 (2-5)
Oral function
Eating 5 (3-7) 4 (3-6) 4 (3-6) 5 (3-8) 5 (3-8) 4 (3-6)
Chewing 5 (3-8) 5 (3-7) 5 (3-7) 6 (3.5-8) 6 (4-8) 5 (3-7)
Opening mouth 5 (3-7) 4 (3-7) 4 (3-6) 5 (3-7) 5 (3-7) 4 (3-6)
No pain medications 7 (4-10) 6 (4-9) 6 (4-9) 8.5 (5-12) 8 (5-12) 6 (4-9)
P 25 , represents the postsurgical day by which 25% of the subjects reported recovery.
Median , represents the postsurgical day by which 50% of the subjects reported recovery.
P 75 , represents the postsurgical day by which 75% of the subjects reported recovery.

The recovery function for all lifestyle items and all oral-function items, except mouth opening and pain, were significantly affected by age even after controlling for surgical procedural variables, tooth position, and sex ( Tables III and IV ). For all QOL outcomes, the oldest group, those at least 21 years of age at surgery, had a risk ratio significantly less than 1 relative to the youngest group (≤18), indicating that the likelihood of recovery on a given postsurgical day was significantly reduced for the subjects 21 or older. For example, on any day after surgery, the oldest group would be only 69% as likely to report recovery for regular routine and eating and only 52% as likely to report no medication for pain as the youngest group. The recovery functions for the youngest and middle groups (18-21 years) were not significantly different. Representative recovery curves for eating a regular diet, resuming a regular routine, and taking pain medications are shown in Figures 1 through 3 .

Table III
P values and adjusted risk ratios for the 3 lifestyle QOL items from the Cox regression analysis
Regular routine Social life Recreation
Variable P Risk ratio P Risk ratio P Risk ratio
Male <0.0001 1.416 0.0002 1.379 0.0067 1.254
Age 18-21 y 0.5764 0.935 0.5549 0.932 0.7570 1.036
Age >21 y 0.0009 0.685 0.0007 0.680 0.0460 0.806
The referent groups were female, age ≤18 years old, surgery time ≤20 minutes, bone removal on at most 1 mandibular molar, 1 or no mandibular molars below the OP, surgery difficulty, ≤9.
Only gold members can continue reading. Log In or Register to continue

Apr 13, 2017 | Posted by in Orthodontics | Comments Off on Recovery after third-molar surgery: The effects of age and sex
Premium Wordpress Themes by UFO Themes