British Journal of Anaesthesia 1998; 81: 511514

Local anaesthetic infiltration for surgical exodontia of third molar

teeth: a double-blind study comparing bupivacaine infiltration with i.v.
ketorolac
D. J. MELLOR, A. H. MELLOR AND E. M. MCATEER

Summary
We studied 40 patients undergoing surgical
removal of at least two third molar teeth under
general anaesthesia. Patients were allocated
randomly to one of two groups: group B (n:20)
received bupivacaine up to 2 mg kg
91
, infiltrated
around the inferior alveolar nerves bilaterally,
and group K (n:20) received ketorolac 10 mg i.v.
at the start of surgery. There were no significant
differences between the two groups in postoper-
ative pain scores measured at 1 h using a visual
analogue scale. Group K had a significantly
lower incidence of side effects related to intra-
oral anaesthesia. Swallowing, speech and oral
continence were significantly better. Group K
scored higher for overall patient satisfaction,
measured using a visual analogue scale. We
failed to demonstrate any difference in early
postoperative recovery (coughing, laryngospasm,
stridor or arterial oxygen desaturation) between
the groups. We conclude that the use of 0.5%
bupivacaine infiltration was no more effective
than a single 10-mg injection of ketorolac while
giving rise to a higher rate of minor airway
complications and lower patient acceptability.
(Br. J. Anaesth. 1998; 81: 511514).
Keywords: anaesthetics local, bupivacaine; surgery, dental;
pain, postoperative; non-steroidal anti-inflammatory drugs;
analgesics non-opioid, ketorolac

Surgical extraction of impacted upper and lower third
molar teeth (wisdom teeth) is a procedure frequently
carried out on an outpatient, day-stay basis.
1
However,
pain after this procedure is frequently severe
1
and anal-
gesia is necessarily a balance between achieving ade-
quate pain relief while causing minimum side effects,
consistent with early discharge.
Several studies have demonstrated the efficacy of
opioids,
1
non-steroidal anti-inflammatory agents
26

and local anaesthetic infiltration
79
in the control of
postoperative dental pain. Each approach to postop-
erative analgesia has its own profile of benefits and
side effects. In particular, widespread use of intra-
oral local anaesthetics causes difficulty with speech
and swallowing and many feel this detracts from its
usefulness.
10
Anxiety has been expressed that wide-
spread intra-oral numbness may predispose to upper
airway complications in the immediate postoperative
period.
10

The aim of this study was to investigate the effi-
cacy, safety and patient tolerability of the widespread
local anaesthesia produced by bilateral inferior dental
nerve block compared with i.v. ketorolac.
Patients and methods
The study was approved by the Research Ethics
Committee of the United Leeds Teaching Hospitals
NHS Trust and all patients gave written consent. We
studied 40 patients, aged more than 18 yr, undergo-
ing at least two lower third molar teeth extractions
under general anaesthesia. Patients undergoing
removal of upper wisdom teeth or other dental
surgery were included only if simultaneously under-
going bilateral lower jaw surgery. In addition to ful-
filling our institutions criteria for suitability for
day-case surgery (body mass index :30 kg m
92
and
ASA grade I or II) patients were excluded if they had
a history of asthma or intolerance to non-steroidal
analgesics or local anaesthetic agents.
Patients were allocated randomly to one of two
groups by opening sealed envelopes in the anaes-
thetic room. There were 20 patients in group K
(ketorolac) and 20 in group B (bupivacaine). On
arrival in the anaesthetic room, electrocardiogram,
pulse oximeter and non-invasive arterial pressure
monitoring were instituted. After i.v. cannulation,
anaesthesia was induced with propofol 2.5 mg kg
91

i.v. followed by another 0.51 mg kg
91
if indicated
clinically, fentanyl 1 g kg
91
i.v. and mivacurium
0.2 mg kg
91
i.v.
The larynx was intubated using a size 6.5 or 7.0
cuffed nasotracheal tube and anaesthesia was main-
tained with 12% isoflurane and 70% nitrous oxide
in oxygen. A water moistened throat pack was
inserted by the anaesthetist.
All patients were operated on by the same surgeon
(A. H. M.). All local anaesthetic blocks were per-
formed by the surgeon who was experienced in per-
forming third molar extractions under both general
and local anaesthesia. Patients in group B received
0.5% bupivacaine i.v. to each side of the jaw, anaes-
thetizing the inferior dental nerve. In addition, any
other surgical areas (upper third molars) were infil-
trated with the same solution. A maximum dose of
bupivacaine 2 mg kg
91
was used. Patients in group K
received ketorolac 10 mg i.v. before the start of
surgery. Placebo (normal saline) was injected in the
D. J. MELLOR*, BSC, MB, BS, FRCA, A. H. MELLOR, BDS, FDSRCS(ED),
E. M. MCATEER, MB, BS, FFARCSI, Leeds General Infirmary, Great
George Street, Leeds LS1 3EX. Accepted for publication: April
17, 1998.
*Address for correspondence: Hospital for Sick Children,
Toronto, Canada M5G 1XA.
512 British Journal of Anaesthesia
same manner to the inferior dental nerves and other
surgical areas. Inferior dental nerve block was per-
formed bilaterally with the patients mouth opened
with a surgical prop. A 25-gauge needle was inserted
just medial to the anterior border of the mandible,
inside the mouth, at a point 1 cm above the occlusal
surface of the lower third molar. Keeping the syringe
parallel to the floor of the mouth and the syringe bar-
rel overlying the premolars of the opposite side, the
needle was advanced 2 cm until the mandible was
contacted, withdrawn slightly and 2 ml of solution
injected.
11
Surgery was performed using a dental burr
rather than the lingual split method.
12

At the end of surgery, the operative time was
recorded as the total time from starting to ending
administration of isoflurane. In addition, difficulty of
surgery was graded IIII for each tooth removed
(table 1).
The patients lungs were ventilated with 100% oxy-
gen until spontaneous recovery from mivacurium neu-
romuscular block was adequate to extubate the
trachea. Extubation was performed in the left lateral,
head-down tilt position. It was performed by the
blinded observer who was responsible for grading the
quality of immediate postoperative recovery. Oxygen
was administered by variable performance mask for
10 min after spontaneous eye opening, and the patient
was discharged from hospital 2 h after recovery.
Assessments were made of the quality of recovery
and analgesia in the immediate postoperative period.
Quality of immediate recovery and presence of air-
way complications in the first 10 min after operation
were recorded by the same blinded observer and
recorded on a four-point scale. One hour after the
end of anaesthesia, pain was assessed using a visual
analogue scale from 0100 mm. At this time, assess-
ment was made of oral continence by observing the
patient swilling 30 ml of water in the mouth. Oral
continence and coughing were noted and graded.
Swallowing was tested next and graded according to
the patients degree of described difficulty. Speech
quality was assessed by the same blinded observer
who noted the presence or absence of speech slurring
when the patient read out a test sentence (a nursery
rhyme). Lastly, the patient was asked to grade satis-
faction with the whole surgical and anaesthetic expe-
rience on a visual analogue scale. The patient was
given identical written instructions to combine their
experiences of numbness, difficulty in speech and
swallowing, and amount of pain into a single satisfac-
tion score. All postoperative assessments were per-
formed by the same anaesthetist who was blinded to
the anaesthetic technique used. Blinding was
achieved by exclusion of the assessor from the anaes-
thetizing and operating areas. The rating scale used
by the blinded observer is shown in table 3. The
visual analogue scoring system was explained to
participating patients before anaesthesia.
Analgesia for all patients comprised codeine phos-
phate 30 mgparacetamol 500 mg combination
tablets (maximum eight per day). This postoperative
analgesic regimen was chosen as we felt it inadvisable
to prescribe an oral non-steroidal anti-inflammatory
agent without knowing which patients had received
ketorolac. Our normal practice is to prescribe
ibuprofen.
Statistical analysis of the results was performed
using an IBM personal computer with SPSS version
6.1 for Windows. Differences were assessed for sig-
nificance using Wilcoxons rank sum test for paired,
unmatched data, Students t test for parametric data
and Spearmans rank correlation test.
Results
There were 20 patients in each group. There were no
significant differences in operative time, operative
grade, number of teeth extracted or body weight
between the two groups (table 2). Grade of difficulty
of extraction of each tooth was recorded and the sum
used as an index of surgical trauma.
There was no significant difference in the incidence
of airway-related side effects between the groups
(table 3) during early recovery from anaesthesia and
surgery. However, no patient in group K suffered dif-
ficulty with swallowing, speech or oral continence at
1 h compared with 50% of patients in group B
(P:0.005 for both speech and swallowing and
P:0.01 for leakage of fluid).
Using the Students t test, there was no significant
difference between groups in VAS for pain (table 4).
Table 1 Grade of surgical difficulty in third molar exodontia.
Graded before operation from radiographs
I:No obvious hard tissue impaction. Tooth likely to elevate
without soft-tissue relieving incision.
II:Hard tissue impaction, requiring minimal amount of bone
removal.
III:Hard tissue impaction requiring moderate bone removal and
tooth division.
Table 3 Incidence of local anaesthetic-related side effects
Group K Group B
Early recovery airway complications
1:No untoward events. Saturation 994% 14 12
2:Minor coughing but saturation 994% 4 6
3:Major coughing or laryngospasm
with saturation 994%
2 1
4:Any situation leading to desaturation 0 1
Leakage from mouth at 1 h
1:No leakage 20 11
2:Minor leakage 0 9
3:Major leakage 0 0
4:Coughing 0 0
Difficulty in swallowing water at 1 h
1:No difficulty 20 10
2:Minor difficulty 0 9
3:Major difficulty 0 1
4:Coughing 0 0
Difficulty in speech at 1 h
1:Normal speech 20 10
2:Minor slurring noticeable 0 6
3:Major slurring of speech 0 4
4:Difficulty in understanding speech 0 0
Table 2 Patient characteristics (mean (SD) or median (range))
Group K Group B
Operative time (min) 30.6 (13.5) 31.8 (10.7)
Sum of extraction grade 6 (412) 6 (412)
No. of teeth extracted 4 (26) 4 (24)
Patient weight (kg) 70.1 (13.3) 69.2 (9.9)
Local anaesthetic infiltration for surgical exodontia of third molar teeth 513
However, the difference in patient satisfaction was
significant (P:0.01).
When pain and patient satisfaction were correlated
with operative time, difficulty of surgery and incidence
of side effects, the following correlations were signifi-
cant (P:0.01): duration of surgery correlated posi-
tively with pain scores; pain scores correlated inversely
with patient satisfaction; and patient satisfaction cor-
related inversely with incidence of speech impairment
and difficulty swallowing. There was no significant
correlation (P:0.01) between operative difficulty and
pain or number of teeth extracted and pain, or
between degree of pain and swallowing difficulty.
Discussion
In a recent survey,
13
it was demonstrated that there is
a dichotomy in anaesthetic practice for removal of
third molar teeth. Half of anaesthetists questioned
did not use local anaesthesia as part of their anaes-
thetic technique. Our aim was to demonstrate the
safety, efficacy and tolerability of local anaesthetic
when applied as a bilateral nerve block intra-orally.
The safety aspect of intra-oral local anaesthesia is
of particular concern to anaesthetists in the period
immediately after reversal and emergence from
anaesthesia. Any disruption of the protective laryn-
geal reflexes threatens airway contamination with
surgical debris and oral contents. One study
14
con-
cluded that local anaesthetic techniques are safe.
However, this series was conducted in patients
undergoing surgery with local anaesthesia under
conscious sedation with midazolam. Several series of
patients have been described who have undergone
tonsillectomy under general anaesthesia with local
anaesthetic supplementation.
15
This technique does
not appear to be associated with an increase in the
incidence of early postoperative airway problems and
this is in accordance with our findings. We noted no
significant increase in postoperative coughing, laryn-
gospasm or stridor between groups. It is impossible
from our sample size to conclude that this technique
is safe and a report of life-threatening upper airway
obstruction after the use of local anaesthesia intra-
orally for paediatric tonsillectomy is noted.
16

The efficacy of local anaesthetic infiltration has
been demonstrated previously.
710
The fact that it is
used widely as the sole form of anaesthesia for third
molar exodontia testifies to its effectiveness.
However, in our study we failed to demonstrate
improved analgesia compared with ketorolac 10 mg
i.v. We performed the local anaesthetic blocks while
patients were anaesthetized. After operation, we per-
formed no formal testing of the extent of the block. It
could be argued that our failure to achieve improved
analgesia was a technical problem in that the appro-
priate nerve bundles were not blocked adequately.
However, the high incidence of side effects, such as
speech impairment, swallowing difficulties and oral
incontinence in group B demonstrated the high suc-
cess rate in achieving neural block. Also, those with
a high score for side effects were in no less pain than
those with less side effects, suggesting that failure to
achieve adequate analgesia was not caused by failure
of the nerve blocks. While oral continence and speech
impairment do not appear to indicate an inability to
protect the patients airway and could be considered
clinically of little relevance, these problems are of
major concern to the patient who has difficulty
drinking and talking after operation. The correlation
between patient satisfaction and incidence of these
minor side effects is relevant.
Surgery for extraction of third molar teeth is fre-
quently performed under local anaesthesia alone.
Clearly, in this respect it is a successful technique.
However, surgeons find it difficult to assess intraop-
erative lingual nerve damage. When used as analgesia
for surgery performed under general anaesthesia, the
inability to immediately assess nerve damage after
operation could be viewed as a further disadvantage
of the technique.
The tolerability of local anaesthesia compared with
i.v. ketorolac was demonstrated by the highly signifi-
cant differences in side effects between groups K and
B. There were significant differences in patient satisfac-
tion between groups, with increased satisfaction in the
ketorolac group. There were no incidences of speech,
swallowing or oral continence problems in the ketoro-
lac group compared with more than 50% of patients in
the local anaesthesia group. We did not examine specif-
ically problems associated with the use of non-steroidal
anti-inflammatory drugs. These agents have a good
safety record in outpatient anaesthesia and the highly
significant results demonstrated here suggest that the
subjective and objective patient tolerability of ketorolac
was higher than that of bupivacaine infiltration.
In summary, we have demonstrated no significant
difference in analgesia between our two groups.
Instead, we demonstrated a significantly higher inci-
dence of minor airway-related side effects and a
significantly lower level of patient satisfaction in the
local anaesthesia group. We failed to demonstrate any
increase in early postoperative airway difficulties and,
together with other published studies, conclude that
the technique is probably safe. Local anaesthetic
infiltration may have a place with non-steroidal anti-
inflammatory agents, although we have not demon-
strated this here. Our study suggests that ketorolac
was a better tolerated, equally efficacious agent for
sole use.
References
1. Tighe KE, Gillan G, Korczac PKJ, Crossley AWA. An assess-
ment of morphine requirement after wisdom teeth extraction
under general anaesthesia. Anaesthesia 1995; 50: 816818.
2. Turner GA, Anson N, Williamson R. A comparison of intra-
muscular ketorolac with indomethacin suppositories in the
treatment of pain after oral surgery. Anaesthesia and Intensive
Care 1996; 26: 665668.
3. Wakeling HG, Barry PC, Butler PJ. Postoperative analgesia in
dental day case surgery. A comparison between Feldene
melt (Piroxicam) and diclofenac suppositories. Anaesthesia
1996; 51: 784786.
4. Kiersch TA, Halladay SC, Koschik M. A double-blind, ran-
domised study of naproxen sodium, ibuprofen and placebo in
postoperative dental pain. Clinical Therapeutics 1993; 15:
845854.
Table 4 Visual analogue scores (VAS) for pain and patient
satisfaction (mean (SD))
Group K Group B
VAS pain (cm) 2.5 (2.4) 3.3 (1.9)
VAS satisfaction (cm) 9.1 (0.72) 7.5 (2.0)
514 British Journal of Anaesthesia
5. Fricke J, Halladay SC, Bynum L, Francisco CA. Pain relief
after dental impaction surgery using ketorolac, hydrocodone
plus acetaminophen, or placebo. Clinical Therapeutics 1993;
15: 500509.
6. Walton GM, Rood JP, Snowdon AT, Rickwood D. Ketorolac
and diclofenac for postoperative pain relief following oral
surgery. British Journal of Oral and Maxillofacial Surgery 1993;
31: 158160.
7. Tordoff SG, Brossy M, Rowbotham DJ, James JN, James R,
Raphael JH. The effect of pre-incisional infiltration with ligno-
caine on post-operative pain after molar extraction under
general anaesthesia. Anaesthesia 1996; 51: 585587.
8. Chapman PJ, Ganendran A. Prolonged analgesia following
pre-operative bupivacaine neural blockade for oral surgery
performed under general anaesthesia. Journal of Oral and
Maxillofacial Surgery 1987; 45: 233235.
9. Wilson IH, Richmond MN, Strike PW. Regional analgesia
with bupivacaine in dental analgesia. British Journal of
Anaesthesia 1986; 58: 401405.
10. Tuffin JR, Cunliffe DR, Shaw SR. Do local analgesics injected
at the time of third molar removal under general anaesthesia
reduce significantly postoperative analgesic requirements?
A double blind controlled trial. British Journal of Oral and
Maxillofacial Surgery 1989; 27: 2732.
11. Dobson MB. Anaesthesia at the District Hospital. Geneva:
World Health Organization, 1988; 95.
12. Absi EG, Shepherd JP. A comparison of morbidity following
the removal of lower third molars by the lingual split and
surgical burr methods. International Journal of Oral and
Maxillofacial Surgery 1993; 22: 149153.
13. MacKenzie DW, Patel MF. Local anaesthesia for pain relief
following wisdom teeth extraction under general anaesthesia.
Anaesthesia 1996; 51: 200201.
14. Holland IS, Stassen LF. Bilateral block: is it safe and more
efficient during removal of third molars? British Journal of Oral
and Maxillofacial Surgery 1996; 34: 243247.
15. Johansen M, Harbo G, Illum P. Preincisional infiltration with
bupivacaine in tonsillectomy. Archives of Otolaryngology 1996;
122: 261263.
16. Bean-Lijewski JD. Glossopharyngeal nerve block for pain
relief after pediatric tonsillectomy: retrospective analysis and
two cases of life-threatening upper airway obstruction from
an interrupted trial. Anesthesia and Analgesia 1997; 84:
12321238.