The removal of the mandibular third molar (M3M) is one of the most common surgical procedures in the oral and maxillofacial area. In the immediate postoperative period, inflammatory complications such as swelling, pain, sensitivity, or trismus might occur from surgical damage to the adjacent hard and soft tissue^1,2. Preemptive approaches including cold application, postoperative instructions, and medications could prevent or reduce inflammatory complications after M3M removal³. Various medications have been employed to manage postoperative inflammatory complications through local or systemic administration^2-6.

Hyaluronidase (HUD) is an enzyme that degrades hyaluronic acid (HA), an intercellular base material of the connective tissue⁷, controls the binding force of intercellular space, and facilitates the absorption and diffusion of drugs⁸. In various clinical situations including ophthalmology, pain medicine, dermatology or otorhinolaryngology, HUD has been reported to be effective in reducing edema after surgery⁸, penetration of nerve block anesthetics⁹, and acceleration of wound healing¹⁰, and to have a reversal effect on HA filler¹¹. Submucosal injection of HUD also has been applied for supraglottic airway edema following tracheostomy¹². HUD injection was reported in clinical dentistry several decades ago¹³, but prospective controlled trials for HUD injection after M3M removal are lacking.

In the present study, we primarily aimed to evaluate the efficacy of submucosal HUD injection for postoperative sequelae (mouth opening limitations, facial swelling, and pain) immediately after surgical extraction of the M3M. Secondarily, we applied a patient-centered outcome questionnaire (PCOQ) for evaluation of the impact of HUD on the individual’s quality of life (QOL) after M3M surgery.

1. Study design and sample

This study was conducted as a double-blinded, randomized controlled trial from November 2018 through September 2019. A total of 48 participants were recruited, and each M3M site (left or right side) was randomly assigned to the experimental or control group. The inclusion and exclusion criteria are shown in Table 1. The Institutional Review Board (IRB) of Yonsei University Dental Hospital approved the study (IRB No. 2-2018-0021). All participants were informed about the study protocol and potential perioperative risks before obtaining their written informed consent. All procedures of the study involving human participants were in accordance with the Declaration of Helsinki (1964) and its later amendments or comparable 

2. Study protocol

Surgical extraction of M3M was performed at the Department of Oral and Maxillofacial Surgery of Yonsei University Dental Hospital. All surgeries were performed by a single surgeon (W.N.) under local anesthesia. Surgical difficulty index (DI) was based on tooth angulation by Winter¹⁴, ramus/occlusal relationship by Pell and Gregory¹⁵, and Pederson’s index¹⁶.(Table 2) Vital signs (systolic, diastolic blood pressure, and pulse rate) were verified before and after M3M surgery. Participants underwent consecutive M3M surgeries, which consisted of a first and second session, in order of random assignments. On the experimental side, the buccal mucoperiosteal flap was appropriately elevated and M3M was extracted with odontomy and ostectomy under local anesthesia (2% lidocaine hydrochloride with 1:100,000 epinephrine; Huons, Seongnam, Korea). The flap was adapted to the wound margin and sutured after submucosal injection of HUD 1,500 international units (IU)/mL (Hirax; BMI Korea, Jeju, Korea) A total of 1.0 mL was injected in the base of a mucoperiosteal flap, 0.5 mL in the mesial margin, and 0.5 mL in the distal margin. On the control side, a saline solution of 1.0 mL was administered as a placebo in the same manner. Every participant received oral medication including antibiotics (cefcapene pivoxil 100 mg three times per day), an analgesic (ibuprofen 200 mg three times per day) for three days after M3M surgery. For participants with allergies to antibiotics, alternative antibiotics (clindamycin 150 mg 4 times per day or ciprofloxacin 250 mg 2 times per day) were prescribed. If a participant suffered moderate pain (numeric rating scale [NRS] from 4 to 6) despite the administration of the drug, additional antipyretic analgesic (acetaminophen 650 mg) was recommended.

3. Data collection

Before and on days 2 and 7 after surgery (preOP, POD2, and POD7), postoperative sequelae including mouth opening and facial swelling were recorded with linear measurement. Mouth opening as a maximal interincisal distance was measured with a metal ruler. Facial swelling was measured with a tape measure and calculated as the sum of three linear measurements along reference points. The reference points included the tragus, a small cartilaginous prominence in front of the external auditory canal of the ear; oral commissure, corner of the mouth; pogonion, the most anterior point of the chin on the mandible in the midline); lateral canthus, outer corner of the eye; and gonion, the most inferior, posterior, and lateral point on the angle of the mandible.(Fig. 1) The amount of change of linear measurement from preOP to POD2 (∆POD2–preOP) and from preOP to POD7 (∆POD7–preOP) were calculated as the amount of facial swelling and compared between the control and experimental sides. The degree of pain was also measured via a NRS ranging from 0 to 10.

HUD has already been approved for use, and the side effects or safety risks were minimal. However, if an adverse drug reaction occurred, the participant would be eliminated from the clinical trial and treatment of the adverse reaction prioritized. At any time during the clinical trial, participants could voluntarily withdraw for any reason, and investigators also had the authority to exclude participants from a clinical trial. A completion of the clinical trial for each patient was defined as accomplishing all processes including clinical procedures and examination as specified in the clinical trial protocol.

4. Questionnaire

A PCOQ was completed by participants at POD2 and POD7. Each participant was instructed to complete a total of 13 questions which were based on the OHIP (Oral Health Impact Profile) by Slade and Spencer¹⁷. Each item was scored as follows: never, 1; hardly ever, 2; occasionally, 3; fairly often, 4; and very often, 5.(Table 3)

5. Statistical analysis

Statistical analysis was performed with the final data obtained by distortion correction. A descriptive analysis was used for each variable and surgical difficulty between the control and experimental sides was compared using the Wilcoxon signed rank test. Statistical significance was analyzed for differences in linear measurement values or postoperative sequelae (mouth opening, facial swelling, and pain) between control and experimental sides. Analysis of variance using repeated measures was used to compare the outcome variables. All data acquisition and analyses were performed with Microsoft Excel, Microsoft 365 (Microsoft, Redmond, WA, USA) and IBM SPSS Statistics software for Windows (ver. 22.0; IBM, Armonk, NY, USA). A P-value of <0.05 was considered statistically significant.

A total of 36 patients (17 males and 19 females) who had fulfilled the follow-up visits and completed the PCOQ were included in the final data analysis.(Fig. 2) The mean age of patients was 24.7±5.35 years (range, 18-42 years). There was no significant difference in DI of M3M between the control and experimental sides (P=0.099).

1. Mouth opening

There was no statistical difference in baseline value of mouth opening between the control and experimental sides at preOP (P=0.713). After surgery, the experimental side demonstrated less change in the mouth opening than the control side at POD2 and POD7 (P=0.028 and 0.001).(Fig. 3. A)

2. Facial swelling

There was no statistical difference in baseline facial swelling between the control and experimental sides at preOP, POD2, and POD7 (P=0.061, 0.409, and 0.352). The experimental side demonstrated significantly less change at POD2 (∆POD2–preOP) and POD7 (∆POD7–preOP) based on preOP than the control side (P=0.027). Also, ∆POD7–preOP showed less difference between the control and experimental sides compared with ∆POD2–preOP.(Fig. 3. B)

3. Pain

The mean postoperative pain score in the control and experimental sides was 4.26±2.30 and 1.90±1.45 at POD2 and 2.15±1.35 and 0.83±1.05 at POD7, respectively. The experimental side showed a significantly lower pain score than the control side at POD2 and POD7 (P=0.000 and 0.001).(Fig. 3. C)

4. PCOQ

At POD2, the experimental side showed a significantly lower score than the control side, indicating less discomfort after HUD injection in Q1 (pain), Q3 (swelling), Q8 (pronunciation), and Q12 (sleep) (P<0.05).(Fig. 4. A) At POD7, the experimental group showed significantly less discomfort after HUD injection in Q1 (pain) and Q7 (food intake) (P<0.05).(Fig. 4. B)

5. Adverse effects

Vital signs verified before and after M3M surgery and HUD injection were within normal range. HUD related adverse events were not observed in the control arm.

Previous studies assessed various modalities including systemic or local corticosteroids^3,18,19, natural substance², platelet concentrates^20,21, and adjuvant laser therapy²² for reducing postoperative swelling and trismus or pain after surgical extraction of M3M. For several decades in the area of dentistry or maxillofacial surgery, HUD, has been investigated in only a few studies regarding postoperative swelling after guided bone regeneration (GBR) with simultaneous dental implant placement²³ or efficacy of anesthesia for irreversible pulpitis⁹. HUD was effective in reducing swelling at the site of GBR in maxilla²³ and for increased duration of block anesthesia of the inferior alveolar nerve with lidocaine⁹. In the present study, the local administration of HUD was evaluated in M3M surgery, which is one of the most common procedures in oral and maxillofacial surgery. We also estimated the QOL of patients during the postoperative period via PCOQ.

The present study revealed a significant reduction of postoperative discomfort, including limitation of the mouth opening, facial swelling, and pain, at POD2 and POD7 on the experimental side. Like previous studies regarding the local administration of other drugs including submucosal dexmethasone or methhyprednisolone¹⁹, trismus was significantly reduced on the HUD side. Several studies have reported that injection of HUD could alleviate postoperative or posttraumatic facial swelling. Kwoen et al.²³ reported that postoperative swelling after GBR with simultaneous dental implant placement was lower in the HUD group, and this result was more effective in the maxilla. In a clinical study of closed reduction surgery after nasal bone fracture, the duration from injury to surgery was significantly lower in the HUD group than the control group, leading to improved patient satisfaction resulting from earlier discharge and return to their daily life²⁴. Unlike postoperative swelling, few studies have presented data on postoperative pain related to HUD after oral and maxillofacial surgery. Contrary to the results of the present study, Kwoen et al.²³ reported that no statistically significant difference was found between HUD and control groups. Due to ethical reasons related to clinical study, the evaluation of pain might be difficult because the participants were instructed to take an analgesic as needed for unbearable pain.

In the PCOQ, participants reported a significant improvement on the experimental side in speaking and sleeping at POD2 and in food intake at POD7 after M3M surgery. Due to spontaneous improvement of postoperative morbidities over time, the amount of difference between the control and experimental sides decreased at POD7. However, split-mouth design could induce a bias for the interpretation of PCOQ²⁵. For instance, the patient’s experience of M3M surgery in the first session could affect the response to M3M surgery in subsequent sessions. Contamination or spilling from one side (the experimental or control side) could affect the other side (the experimental or control side). Considering the limitations of the split-mouth randomized controlled trial design, a careful interpretation of the study results is required.

The mechanism of HUD action has been suggested in several studies. By degrading HA in the extracellular matrix (ECM), HUD reduces HA viscosity and increases membrane permeability. Disruption of the ECM barrier increases tissue diffusion and resorption of excess fluids into the systemic circulation⁷. In vivo and clinical studies have documented the consequent anti-edema effect of HUD^12,26,27. By resolving ground substance of the connective tissue, HUD was shown to be effective in early treatment of hematoma and fibrosis^28,29. Furthermore, HUD regulates inflammatory response by decreasing the infiltration of inflammatory mediators such as neutrophils to the inflammatory site⁸ or increasing anti-inflammatory cytokines including TNFα (tumor necrosis factor alpha), interleukin (IL)-1α, IL-4, and IL-10, thus promoting wound healing¹⁰.

Allergic reaction to HUD is uncommon³⁰. In this study, no participants showed any adverse effects after HUD injection. However, several cases of HUD hypersensitivity have been documented, being characterized by erythematous edema associated with immediate and delayed hypersensitivity reactions^31,32. Since most medical HUD is made of proteins derived from bovine, ovine, and caprine sources, the risk of hypersensitivity should be considered before HUD administration in patients allergic to bovine collagen and bee stings³³. To screen the adverse reactions, a preliminary skin test of HUD with 3 IU has been recommended^11,31,34. In addition to allergic reaction, areas where local infection such as acute pericoronitis or chronic abscess exist, HUD is also not indicated because it may facilitate spread of the infection.

A limitation of this study was lack of positive control employing conventional modalities such as local or systemic administration of corticosteroids. Considering the low cost and easy access to conventional anti-inflammatory agents, the benefits of HUD should be comparatively analyzed. Further studies are required before wide clinical use, including comparison of the efficacy between HUD and conventional anti-inflammatory agents and also standardization of HUD concentration.

Despite the limitations of the present study, local injection of HUD could be an efficient modality in reducing postoperative sequelae, suggesting that HUD could be an effective preemptive modality for M3M surgery patients when conventional anti-inflammatory agents are restricted due to adverse effects.

W.N. conceived the idea and S.L. designed the study. H.K. and S.L. reviewed the medical records and contributed to data acquisition. S.L. analyzed the data and prepared the manuscript. W.N. provided guidance for all aspects of the study and critical revision of the article. All authors reviewed and approved the final version of the manuscript.

The IRB of Yonsei University Dental Hospital approved the study (IRB No. 2-2018-0021). The written informed consent was obtained from all participants.

Written informed consent was obtained from the patients for publication of this article and accompanying images.

No potential conflict of interest relevant to this article was reported.