Mineral trioxide aggregate improves healing response of periodontal tissue to injury in mice.

BACKGROUND AND OBJECTIVE: Mineral trioxide aggregate (MTA) is a biomaterial used in endodontic procedures as it exerts beneficial effects on regenerative processes. In this study, we evaluate the effect of MTA on healing of periodontal ligament (PDL) and surrounding tissue, following injury, in a transgenic mouse model and on the differentiation of murine mesenchymal progenitor cells in vitro. MATERIAL AND METHODS: We used an inducible Cre-loxP in vivo fate mapping approach to examine the effects of MTA on the contributions of descendants of cells expressing the αSMA-CreERT2 transgene (SMA9+ ) to the PDL and alveolar bone after experimental injury to the root furcation on the maxillary first molars. Col2.3GFP was used as a marker to identify mature osteoblasts, cementoblasts and PDL fibroblasts. The effects of MTA were examined 2, 17 and 30 days after injury and compared histologically with sealing using an adhesive system. The effects of two dilutions of medium conditioned with MTA on proliferation and differentiation of mesenchymal progenitor cells derived from bone marrow (BMSC) and periodontal ligament (PDLC) in vitro were examined using the PrestoBlue viability assay, alkaline phosphatase and Von Kossa staining. The expression of markers of differentiation was assessed using real-time PCR. RESULTS: Histological analyses showed better repair in teeth restored with MTA, as shown by greater expansion of SMA9+ progenitor cells and Col2.3GFP+ osteoblasts compared with control teeth. We also observed a positive effect on differentiation of SMA9+ progenitors into osteoblasts and cementoblasts in the apical region distant from the site of injury. The in vitro data showed that MTA-conditioned medium reduced cell viability and osteogenic differentiation in both PDLC and BMSC, indicated by reduced von Kossa staining and lower expression of osteocalcin and bone sialoprotein. In addition, cultures grown in the presence of MTA had marked decreases in SMA9+ and Col2.3GFP+ areas as compared with osteogenic medium, confirming reduced osteogenesis. CONCLUSION: MTA promotes regeneration of injured PDL and alveolar bone, reflected as contribution of progenitors (SMA9+ cells) into osteoblasts (Col2.3GFP+ cells). In vitro, MTA-conditioned medium fails to promote osteogenic differentiation of both PDLC and BMSC.

Gene-expression analysis of cementoblasts and osteoblasts.

BACKGROUND AND OBJECTIVE: Cementum and bone are similar mineralized tissues, but cementum accumulates much more slowly than bone, does not have vasculature or innervation and does not undergo remodeling. Despite these differences, there are no well-established markers to distinguish cementoblasts from other mature mineralizing cells such as osteoblasts and odontoblasts. The purpose of this study was to assess differences in gene expression between cementoblasts and osteoblasts using gene profiling of cell populations isolated directly from osteocalcin-green fluorescent protein (OC-GFP) transgenic mice. MATERIAL AND METHODS: OC-GFP reporter mice were used as they show labeling of cementoblasts, osteoblasts and odontoblasts, but not of periodontal ligament fibroblasts, within the periodontium. We sorted cells digested from the molar root surface to isolate OC-GFP(+) cementoblasts. Osteoblasts were isolated from calvarial digests. Microarray analysis was performed, and selected results were confirmed by real-time PCR and immunostaining or in situ hybridization. RESULTS: Microarray analysis identified 95 genes that were expressed at least two-fold higher in cementoblasts than in osteoblasts. Our analysis indicated that the Wnt signaling pathway was differentially regulated, as were genes related to skeletal development. Real-time PCR confirmed that expression of the Wnt inhibitors Wnt inhibitory factor 1 (Wif1) and secreted frizzled-related protein 1 (Sfrp1) was elevated in cementoblasts compared with osteoblasts, and Wif1 expression was localized to the apical root region. In addition, the transcription factor BARX homeobox 1 (Barx1) was expressed at higher levels in cementoblasts, and immunohistochemistry indicated that BARX1 was expressed in apical cementoblasts and cementocytes, but not in osteoblasts or odontoblasts. CONCLUSION: The OC-GFP mouse provides a good model for selectively isolating cementoblasts, and allowed for identification of differentially expressed genes between cementoblasts and osteoblasts.

αSMA-Expressing Perivascular Cells Represent Dental Pulp Progenitors In Vivo.

The goal of this study was to examine the contribution of perivascular cells to odontoblasts during the development, growth, and repair of dentin using mouse molars as a model. We used an inducible, Cre-loxP in vivo fate-mapping approach to examine the contributions of the descendants of cells expressing the αSMA-CreERT2 transgene to the odontoblast lineage. In vivo lineage-tracing experiments in molars showed the contribution of αSMA-tdTomato+ cells to a small number of newly formed odontoblasts during primary dentinogenesis. Using an experimental pulp exposure model in molars to induce reparative dentinogenesis, we demonstrate the contribution of αSMA-tdTomato+ cells to cells secreting reparative dentin. Our results demonstrate that αSMA-tdTomato+ cells differentiated into Col2.3-GFP+ cells composed of both Dspp+ odontoblasts and Bsp+ osteoblasts. Our findings identify a population of mesenchymal progenitor cells capable of giving rise to a second generation of odontoblasts during reparative dentinogenesis. This population also makes a small contribution to odontoblasts during primary dentinogenesis.

Laparoscopic ventral hernia repair: a single center experience.

A retrospective chart review at the Carolinas Medical Center was performed on all patients who underwent laparoscopic ventral hernia repair (LVHR) from July 1998 through December 2003. LVHR was successfully completed in 270 of the 277 patients, or 98%, in whom it was attempted. Half of the patients (138/277) had at least one previous failed repair. The average defect measured 143.3 cm(2), and mesh was used in all repairs. The mean operating time was 168.3 min, mean blood loss was 50 cc, and average length of hospitalization was 3.0 days. Thirty-four complications occurred in 31 patients (11%). Only two mesh infections occurred (0.7%). At a mean follow-up period of 21 months, the rate of hernia recurrence was 4.7%. As experience grows and length of follow-up expands, LVHR may become the preferred approach for ventral hernia in difficult patients, especially obese patients and patients who have failed prior open repairs.

Evaluation of adhesion formation, mesh fixation strength, and hydroxyproline content after intraabdominal placement of polytetrafluoroethylene mesh secured using titanium spiral tacks, nitinol anchors, and polypropylene suture or polyglactin 910 suture.

BACKGROUND: The purpose of this study is to evaluate fixation methods for polytetrafluoroethylene (ePTFE) mesh with an in vivo model of laparoscopic ventral hernia repair. METHODS: In 40 New Zealand white rabbits, a 4 x 4-cm ePTFE mesh (n = 80, two per animal) was attached to an intact peritoneum with polyglactin 910 (PG 910) (n = 20) or polypropylene (PP) (n = 20) suture, titanium spiral tacks (TS) (n = 20), or nitinol anchors (NA) (n = 20). Mesh was harvested at 8 and 16 weeks for fixation strength testing, adhesion assessment, and collagen (hydroxyproline) content. Fixation strength on day 0 was determined with mesh attached to harvested abdominal wall. Statistical significance was determined as p < 0.05. RESULTS: There was no difference in fixation strength between PP (39.1 N) and PG 910 (40.0 N) sutures at time zero. At week 8, PP (25.7 N) was significantly stronger (p < 0.05) than PG 910 (11.4 N) suture, but not at week 16. The fixation strength of TS and NA (day 0, 15.4 vs 7.4 N; week 8, 17.5 vs 15.3 N; week 16, 19.1 vs 13.8 N) was not significantly different. Fixation with PP suture was significantly (p < 0.05) stronger than that with TS and NA at day 0 (39.1, 15.4, and 7.4 N, respectively) but not at weeks 8 or 16. The fixation strength of suture decreased significantly (p < 0.05) from day 0 to week 16 (PP: day 0 = 39.1 N, week 8 = 25.7 N, week 16 = 21.4 N; PG 910: day 0 = 40.0 N, week 8 = 11.4 N, week 16 = 12.8 N). The fixation strength of NA and TS did not change significantly (NA: day 0 = 7.4 N, week 8 = 15.3 N, week 16 = 13.8 N; TS: week 0 = 15.4 N, week 8 = 17.5 N, week 16 = 19.1 N). There were no differences in adhesion area based on fixation device used; however, there were more (p < 0.05) mesh samples using NA with adhesions compared to TS and adhesion tenacity was greater (p < 0.05) compared to that of TS, PP, and PG. Hydroxyproline content at weeks 8 and 16 was similar for all fixation devices. CONCLUSIONS: The initial fixation strength for nonabsorbable suture is significantly greater than that of the metallic fixation devices, but after 8 weeks there is no difference. Laparoscopic ventral hernia repair without transabdominal suture fixation may be predisposed to acute failure. The metallic devices have similar fixation strength, although the incidence of adhesions and tenacity of adhesions appear to be greater with the nitinol anchors. Since these devices have similar fixation strengths and most likely provide adequate supplementation to transabdominal sutures for mesh fixation after laparoscopic ventral hernia repair, their use should be based on other factors, such as their propensity for adhesions, ease of application, and cost.

The susceptibility of prosthetic biomaterials to infection.

BACKGROUND: Despite the use of a sterile technique and the administration of prophylactic antibiotics during surgical procedures, mesh infection continues to complicate the use of biomaterials. The purpose of this study was to compare the susceptibility to infection of prosthetic biomaterials in a live-animal model. METHODS: The following seven prosthetic mesh biomaterials were used in this study. Expanded polytetrafluoroethylene (ePTFE) with silver/chlorhexidine (DM+), ePTFE (DM), porcine intestinal submucosa (S), polypropylene (M), ePTFE/polypropylene (X), hyaluronate/carboxymethylcellulose/polypropylene (SM), and human acellular dermal matrix (A). Lewis rats (n = 108) underwent creation of a single ventral hernia; 105 of them were repaired with a different mesh (2-cm2 piece). Twelve pieces of each mesh were inoculated at the time of hernia repair with 10(8) Staphylococcus aureus (n = 84). Three pieces of each mesh were placed without bacterial inoculation (n = 21). In three animals, no mesh was placed; instead, the peritoneum of the hernia defect was inoculated (n = 3). After 5 days, the animals were killed and the mesh was explanted (peritoneum for the nonmesh control). The mesh was vortex-washed and incubated in tryptic soy broth. Bacterial counts were determined using serial dilutions and spot plates and quantified in colony-forming units (CFU) per square centimeter of mesh present in the vortex wash fluid (wash count) and the soy broth (broth count). Data are presented as the mean log(10), with analysis of variance (ANOVA) and Tukey's test used to determine significance (p < 0.05). RESULTS: The DM+ material had no detectable live bacteria in the wash or broth counts in 10 of 12 tested samples (p = 0.05). Of the samples that showed bacterial growth, the peritoneum control group had a lower wash count than A (p = 0.05) and the lowest broth count of all the materials except for DM+ (p = 0.05). In addition, SM had a significantly lower wash count than A (p = 0.05), with no broth count difference. In regard to wash and broth counts, DM, M, X, SM, S, and A were no different (p = NS). CONCLUSIONS: The DM+ material was the least susceptible to infection. Impregnation with silver/chlorhexidine killed the inoculated bacteria, preventing their proliferation on the mesh surface. Other than DM+, native peritoneal tissue appears to be the least susceptible to infection. Silver/chlorhexidine appears to be an effective bactericidal agent for use with mesh biomaterials.

The laparoscopic repair of suprapubic ventral hernias.

BACKGROUND: The complexity of dissection and the close proximity of the hernia to bony, vascular, nerve, and urinary structures make the laparoscopic repair of suprapubic hernias (LRSPH) a formidable operation. We performed a prospective evaluation of the outcomes of patients undergoing LRSPH. METHODS: The study population comprised 36 patients undergoing LRSPH from July 1996 to January 2004. Patient demographics, hernia sizes, mesh types and sizes, perioperative outcomes, and recurrences were documented. After our early experience with this operation, the repair evolved to include transabdominal suture fixation to the pubic bone, Cooper's ligament, and above the iliopubic tract. RESULTS: There were 26 women and 10 men. They had a mean age of 55.9 years (range, 33-76) and a mean body mass index (BMI) of 31.0 kg/m2 (range, 22-67). Twenty-two (61%) of the repairs were for recurrent hernias, with an average of 2.3 previously failed open repairs each (range, 1-11). The mean hernia size was 191.4 cm2 (range, 20-768), and the average mesh size was 481.4 cm2 (range, 193-1,428). All repairs were performed with expanded polytetrafluoroethylene (ePTFE) mesh. Mean operating time was 178.7 min (range, 95-290). Mean blood loss was 40 cc (range, 20-100). One patient undergoing her fifth repair required conversion due to adhesions to a polypropylene mesh. Hospital stay averaged 2.4 days (range, 1-7). Mean follow-up was 21.1 months (range, 1-70). Complications (16.6%) included deep venous thrombosis (n = 1), prolonged pain for >6 weeks (n = 1), trocar site cellulitis (n = 1), ileus (n = 1), prolonged seroma (n = 1), and Clostridium difficile colitis (n = 1). Hernias recurred in two of our first nine patients, for an overall recurrence rate of 5.5%. Since we began using the technique of applying multiple sutures directly to the pubis and Cooper's ligament (in the subsequent 27 patients), no recurrences have been documented. CONCLUSIONS: Although technically demanding and time-consuming, the LRSPH is safe and technically feasible. Moreover, it results in a low recurrence rate and is applicable to large or multiply recurrent hernias. Transabdominal suture fixation to the bony and ligamentous structures produces a more durable hernia repair.

A novel technique of lumbar hernia repair using bone anchor fixation.

Lumbar hernias are difficult to repair due to their proximity to bone and inadequate surrounding tissue to buttress the repair. We analyzed the outcome of patients undergoing a novel retromuscular lumbar hernia repair technique. The repair was performed in ten patients using a polypropylene or polytetrafluoroethylene mesh placed in an extraperitoneal, retromuscular position with at least 5 cm overlap of the hernia defect. The mesh was fixed with circumferential, transfascial, permanent sutures and inferiorly fixed to the iliac crest by suture bone anchors. Five hernias were recurrent, and five were incarcerated; seven were incisional hernias, and three were posttraumatic. Back and abdominal pain was the most common presenting symptom. Mean hernia size was 227 cm(2) (60-504) with a mesh size of 620 cm(2) (224-936). Mean operative time was 181 min (120-269), with a mean blood loss of 128 ml (50-200). Mean length of stay was 5.2 days (2-10), and morphine equivalent requirement was 200 mg (47-460). There were no postoperative complications or deaths. After a mean follow-up of 40 months (3-99) there have been no recurrences. Our sublay repair of lumbar hernias with permanent suture fixation is safe and to date has resulted in no recurrences. Suture bone anchors ensure secure fixation of the mesh to the iliac crest and may eliminate a common area of recurrence.

Responses of intradental nerves to chemical and osmotic stimulation of dentine in the cat.

Recordings have been made of the neural responses evoked by stimulation of dentine with solutions of NaCl, NH4Cl and dextrose. Stimulation of the outer dentine produced no response. From the inner dentine, a smaller number of impulses were recorded with solutions of NaCl than with corresponding concentrations of NH4Cl, but a much larger number than with solutions of dextrose. The discharge evoked by a solution increased in mean frequency and decreased in latency as the thickness of dentine was reduced. The responses suggest that the receptors were in the innermost dentine or the pulp and that they were excited by changes in extracellular fluid composition rather than by osmotic effects. The properties of the receptors appear to be different from those involved in pain from dentine in man.

Electrical stimulation of teeth.

The technical problems of stimulating intact teeth in clinical diagnosis and in pain research are discussed. Bipolar stimulation has no advantage over monopolar stimulation for most applications. A 100 V constant current stimulator is suitable for monopolar stimulation of healthy teeth with a large electrode at about threshold intensity, but, for monopolar stimulation up to 150 micronA and for bipolar stimulation, very much higher voltages are required and no suitable instrument is available. Circuits are given for isolated 100 V and 1kV constant current stimulators.

The thresholds of the jaw-opening reflex and trigeminal brainstem neurons to tooth-pulp stimulation in acutely and chronically prepared cats.

Electrical stimuli were applied to tooth-pulp in cats and the thresholds of the jaw-opening reflex and of neurons in the trigeminal sensory nuclei were determined. The effects of the method of preparation of the animal for stereotaxic recording were determined by making observations on animals set up in one of three ways: acutely in the usual manner; chronically, three to five days before recording; and acutely with precautions to minimize nociceptive input to the central nervous system. The threshold of the jaw-opening reflex increased progressively during the setting up of the normal, acute preparations and at the time brainstem recording began was significantly higher in these than in either the chronic or low-trauma acute preparations. Previous studies have shown that the increase in threshold is maintained for several hours and is not due to the effects of the anaesthetic. In normal acute preparations, few units (27/154) were found that had thresholds below 50 microA, 0.1 ms, whereas many units were encountered that responded to such a stimulus in chronic (147/152) and low-trauma acute (99/127) animals. In the chronic and in low-trauma acute preparations, there was no significant difference between the thresholds of the units in the main sensory trigeminal nucleus and spinal subnucleus oralis compared with those in subnucleus caudalis. Thus the preparation of an animal for stereotaxic recording can cause a severe and long-lasting depression in the excitability of neurons in the trigeminal sensory nuclei and an increase in the threshold of the jaw-opening reflex. This effect will have influenced the results of previous studies on the responses evoked in central neurons by stimulation of tooth-pulp, and may have similarly affected recordings from other regions.

Intrapulpal nerve stimulation in the rat.

Electrical stimulation of the rat incisor pulp has been frequently used as a method to study the mechanisms of pain. However, several of the techniques used to stimulate nociceptive nerves in these teeth also tend to excite non-nociceptive nerves in adjacent tissues. This report deals with the development and verification of a procedure which selectively stimulates pulpal nerves over a wide range of stimulus intensities. The methods consisted of placing electrodes in labial cavities of each of the two lower rat incisors, 1 mm distal to the level of the crest of the interdental papilla. Each cavity had a diameter of 0.5 mm and a depth of about three-quarters fo the crown diameter. Each electrode consisted of a Ag/AgCl filling compacted onto the end of a 0.1 mm silver wire. Constant current stimuli were passed between the two electrodes. EMG was monitored to detect reflex activation of the digastric muscle in response to intradental or extradental afferent excitation. The response consisted of two components. One was a long latency potential (LLP) (37 ms latency) (89 microA ave TH); the other was a short latency potential (SLP) (12 ms latency) (650 microA min TH). In all animals that LLP was abolished by bilateral sectioning of the tooth pulp 7 mm below the level of the labial gingival margin. The SLP was not abolished. This would indicate that the LLP was due to stimulation of pulpal fibers, and the SLP to excitation of extrapulpal fibers. This may prove to be a helpful method for nociceptive researchers to easily verify that they are actually stimulating pulpal afferents.

Methods for recording the jaw-opening reflex to tooth-pulp stimulation in awake cats.

Techniques are described for use in awake, unrestrained cats which enable recordings to be made from the digastric muscle, electrical stimuli to be applied to the teeth, and intravenous injections to be made via an indwelling cannula. A headpiece was fixed to the skull of the animal and leads were passed subcutaneously from it to electrodes in the muscle and the teeth. A silicone rubber cannula was inserted into the external jugular vein and connected to an injection port in the headpiece. The headpiece incorporated a miniature 9-way connector which was connected to the electrode leads.

Do primary afferents from tooth-pulp cross the midline?

Compound action potentials were recorded from the pulps of canine teeth in cats during stimulation of the inferior dental nerve, trigeminal ganglion or brain stem on one side. Recordings were also made from the inferior dental nerve while stimulating the canine pulps. No evidence for pulpal fibres crossing the midline was found.

Sensory and reflex responses to tooth pulp stimulation in man.

Experiments have been carried out to investigate whether all tooth pulp afferent nerves are capable of producing pain. Monopolar and bipolar stimuli were applied to teeth in human subjects and sensory thresholds determined. EMGs were recorded from the masseter and the anterior digastric muscles. With stimuli up to three times the sensory threshold, no response could be detected in the digastric but at, or just above, the sensory threshold, inhibitory effects were produced in masseter muscle. The latency of the muscle response with bipolar stimulation was 18-22 msec. There was no evidence of stimulus spread to nerves outside the teeth. Bipolar and monopolar stimulation both produced the same sensation but this was not described as painful. It is concluded that some pulpal afferent nerves may not be capable of producing pain, and that the sensory and reflex responses at threshold were probably produced by the same fibres.

The effects of anaesthetics and remote noxious stimuli on the jaw-opening reflex evoked by tooth-pulp stimulation in the cat.

Previous studies indicate that the threshold of the jaw opening reflex (JOR) evoked by tooth-pulp stimulation is much lower in cats subjected to minimal surgical trauma and a short period of anaesthesia than in animals prepared for stereotaxic recording from the brainstem. Experiments have been carried out to determine whether the higher JOR thresholds observed in the latter group of cats could be attributed to the duration of the anaesthesia or the greater surgical trauma to which they were subjected. The effects on the JOR evoked by tooth-pulp stimulation of brief episodes of noxious and high intensity electrical stimulation of other tissues have been studied in anaesthetized cats. In lightly anaesthetized, control animals, the reflex threshold was usually below 100 microA, 0.1 ms and maintained anaesthesia did not affect this. Alphaxalone/alphadolone, methohexitone and alpha-chloralose produced similar results. Noxious or high intensity electrical stimuli applied to a paw, a pinna or the scalp caused either no change or a decrease in the JOR threshold of cats lightly anaesthetized with alphaxalone/alphadolone. With deeper anaesthesia, these same conditioning stimuli caused a maintained increase in JOR threshold which could be reversed by decreasing the anaesthetic dose. The results suggest that the high threshold of the JOR observed in earlier experiments was not due to anaesthesia but may have been caused by trauma.

Thresholds to electrical stimulation of nerves in cat canine tooth-pulp with A beta-, A delta- and C-fibre conduction velocities.

The thresholds to electrical stimulation and the conduction velocities of nerve fibres supplying the pulp of cat canine teeth have been determined. Compound action potentials and the responses of 80 single units were recorded from the inferior alveolar nerve. The properties of 4 types of coronal stimulating electrode were compared. In the single unit studies separate estimates of intradental and extradental conduction velocity of the fibres were obtained by stimulating the pulp in the crown and in the root of the tooth. The units had extradental conduction velocities ranging from 57.7 to 0.9 m . s-1 and therefore included A beta-, A delta- and C-fibres. The thresholds of the units to coronal stimulation ranged from 7 microA, 0.1 ms to 805 microA, 1.0 ms.

Distribution of P2X3 receptors in the rat trigeminal ganglion after inferior alveolar nerve injury.

The ATP-gated cation channel receptor P2X3 is associated with nociceptive primary sensory neurons. We have, using immunohistochemistry, examined the expression of P2X3 in rat trigeminal ganglia 4-22 days after ligation/section or chronic constriction of the mandibular inferior alveolar nerve. In the normal trigeminal ganglion the anti-P2X3 receptor antibody labeled 37-58% of all neurons. Double labeling demonstrated that about 70-95% of the small neurons that bind the isolectin I-B4 displayed P2X3-immunoreactivity, and that about 40% of larger RT97-positive nerve cells were P2X3 receptor-immunoreactive. At 4 and 10 days after inferior alveolar nerve injury, the proportion of P2X3-immunoreactive neurons had increased to about 65% (range 52-78%). Examinations at the injury sites showed an intense P2X3 receptor-immunoreactivity in nerve endings. At longer survival stages the proportion of P2X3 receptor-positive sensory neurons had returned to control values. These results show that the P2X3 receptor is transiently upregulated and anterogradely transported in trigeminal primary sensory neurons after nerve injury. Since the receptor is accumulated in injured nerve endings, it may be associated with abnormal impulse propagation from these sites.

The vascularity of dental pulp in cats.

The fraction of the volume of the coronal pulp of cat canines that is occupied by blood vessels was estimated by measurement of the cross-sectional areas of all the vessels in a complete transverse section of the pulp from each of four teeth. The sections were taken 0.5 mm from the pulp cornu. Overall, 14.4% of the area of the pulp was occupied by vessels. In the core of the pulp, the average value was 42.9%, and superficially, near the odontoblast layer, it was between 5 and 10%. The average capillary density was 1402/mm2, which is higher than in most other tissues. Laser Doppler flow meters can be used for recording blood flow from the coronal pulp of intact teeth, but these instruments are linear only if the moving blood cells occupy no more than 1% of the tissue volume. The present results suggest that this figure is exceeded in pulp.

Three-point alignment of zygoma fractures with miniplate fixation.

Over the years, depression of the malar eminence has been one of the most persistent complications of zygoma fractures, regardless of the fixation technique used. The use of a sublabial incision for direct visualization and alignment of the fracture lines at the infraorbital rim and lateral maxillary buttress is described. When coupled with single miniplate fixation of the zygoma across the frontozygomatic suture, the sublabial approach results in improved reduction of the fractures with rigid stabilization. Our experience with this approach, as well as with two-point wire fixation and with miniplate fixation without sublabial exposure, is reviewed.