Plate extrusion occurred through the skin of the lateral chin 6 months after reconstruction of a through-and-through oromandibular defect using a radial forearm free flap and a bridging mandibular reconstruction plate.
Size and shape comparison of a THORP (titanium hollow screw reconstruction plate) system (upper plate) vs the Leibinger Locking System (lower plate) shown from a lateral view (left) and profile height (right).
Left, Panoramic radiograph of a fractured titanium hollow-screw reconstruction plate reconstruction occurring through a screw hole, where the plate has minimal cross-sectional area. Right, Panoramic radiograph of a fractured Leibinger Locking System reconstruction occurring between 2 screw holes (arrow), where the plate has maximum cross-sectional area.
Blackwell KE, Lacombe V. The Bridging Lateral Mandibular Reconstruction Plate Revisited. Arch Otolaryngol Head Neck Surg. 1999;125(9):988–993. doi:10.1001/archotol.125.9.988
Lateral oromandibular reconstruction using a soft tissue free flap with a first-generation locking mandibular reconstruction plate (MRP) was rejected in a previous series by the senior author (K.E.B.) owing to a high incidence of delayed plate extrusion through the cheek skin.
To reexamine this method of reconstruction using a second-generation, low-profile MRP.
Patients and Design
A prospective case series of 27 patients with segmental defects of the lateral mandible after treatment of head and neck cancer.
An academic tertiary care referral center.
All patients had mandibular continuity restored using the Leibinger Locking System (Stryker Leibinger Inc, Kalamazoo, Mich) MRP. Associated soft tissue defects were repaired using radial forearm (n=22) or rectus abdominis (n=5) free flaps.
Main Outcome Measure
Incidence of hardware-related complications.
All microvascular flap transfers were successful. One patient experienced a plate fracture 9 months after reconstruction. Only 1 patient experienced external plate exposure, 6 months after undergoing reconstruction of a through-and-through defect. Reconstruction was successful in 25 (93%) of the cases after a median follow-up period of 19.5 months.
The high incidence of external plate exposure in patients undergoing lateral oromandibular reconstruction using soft tissue free flaps and first-generation locking MRPs may have resulted from a plate geometry that was prone to result in extrusion. After a similar length of follow-up, the incidence of reconstructive failure was reduced by using a low-profile, rounded-contour MRP. Final assessment of the durability of this technique will require long-term follow-up.
IN 1996, BLACKWELL et al1 reported a series of 14 patients who underwent lateral oromandibular reconstruction using soft tissue free flaps in conjunction with bridging titanium hollow screw reconstruction plate (THORP) mandibular reconstruction plates (MRPs). This method of reconstruction was rejected owing to a high incidence of delayed hardware-related complications. The incidence of hardware-related reconstructive failure was 29% in 14 patients after a median follow-up period of 16.5 months. One patient experienced fracture of a THORP MRP, while 3 patients suffered from hardware extrusion through the cheek skin. Analysis of this experience led to the conclusion that segmental defects of the lateral mandible were optimally reconstructed using a vascularized bone–containing free flap. Because of the critical role of the tongue for speech and swallowing and the complex 3-dimensional anatomy of the oropharynx, it was recognized that the soft tissue component of vascularized bone–containing free flaps would be inadequate to reconstruct the soft tissue component of some lateral oromandibular defects. In this situation, it would be necessary to carry out a vascularized bone–containing free flap to restore mandibular continuity and a simultaneous soft tissue free flap to reconstruct the soft tissues.
The THORP MRP was the first system of mandibular fixation that achieved an added degree of hardware stability by allowing for screw osteointegration and a mechanism to lock the screw heads to the screw holes of the plate. As a result of these properties, this first-generation locking MRP has achieved a very low incidence of hardware complications related to screw loosening.2 However, the THORP system has a high plate profile height, projecting 3.5 mm away from the underlying mandible or bone graft. In addition, the plate contour contains obtuse angles and straight edges. A second generation of locking MRPs has been introduced that has a low plate profile and smooth, rounded contour at the screw holes compared with the THORP MRP. The current study was undertaken to determine if the high incidence of THORP extrusion reported by Blackwell et al1 might be secondary to a plate geometry that was prone to extrusion. A follow-up series of patients underwent lateral oromandibular reconstruction using soft tissue free flaps in conjunction with a bridging, second-generation, low-profile locking MRP. The incidence of hardware-related complications was analyzed and compared with the previous series of patients.
This series includes 15 men and 12 women who underwent segmental resection of the lateral mandible for squamous cell carcinoma of the oral cavity or oropharynx. Data were recorded prospectively using a personal computer spreadsheet database. The patients' ages at the time of treatment ranged from 32 to 88 years. Primary tumors were classified as stage II in 6 cases, stage III in 4 cases, and stage IV in 17 cases. Oromandibular defects were classified using a previously published classification.3
All surgical procedures were carried out after obtaining appropriate informed consent. The Leibinger Locking System (Stryker Leibinger Inc, Kalamazoo, Mich) MRP was used to bridge the segmental mandibular defects in all cases. The MRPs were applied to the buccal cortex of the remaining native mandible using standard techniques. An effort was made to use a minimum of 3 fixation screws in each mandibular segment, although short condylar segments would sometimes accommodate only 2 screws. Whenever oncologically feasible, the MRP was contoured and applied to the buccal cortex of the mandible before performing the segmental resection.
The indication for soft tissue free flap and bridging plate reconstruction was largely determined by the nature of the soft tissue defect in patients undergoing lateral segmental mandibulectomy. Patients in this series had complex or extensive soft tissue defects that in our judgment were poorly suited for reconstruction using the skin or muscle component of a fibula or iliac crest free flap. Soft tissue defects were reconstituted using radial forearm (n=22) or rectus abdominis (n=5) free flaps. Radial forearm flaps were selected for patients with lateral mandibulectomy defects associated with complex 3-dimensional soft tissue resections of the oropharynx, where the thin and pliable radial forearm flap was folded for reconstruction of the base of the tongue, lateral and posterior oropharynx, and the soft palate. Rectus abdominis flaps were often selected for soft tissue reconstruction in patients who underwent subtotal or total glossectomy.
Twenty-three patients (85%) in this series received high-dose, external-beam radiation therapy. Five patients received preoperative, and 17 patients, postoperative radiation therapy. One patient received both preoperative and postoperative radiation therapy.
Patient and tumor characteristics, defect classification, reconstructive method, and outcome are summarized in Table 1. Tumor was confined to the lingual aspect of the mandible in 22 of the 27 patients. One patient had tumor involving the masseter muscle that necessitated thinning of the cheek flap to achieve negative surgical margins. Four patients underwent through-and-through resection of mucosa, mandible, and skin to treat advanced cancers.
All free tissue transfers were successful, and there were no perioperative deaths. There were no serious early perioperative reconstructive complications (eg, total or partial flap necrosis, fistula, hematoma, or infection). Major perioperative medical complications occurred in 10 patients (37%). Three patients (11%) experienced delayed hardware-related reconstructive complications. One of the reconstructive complications did not require plate removal. This patient experienced malocclusion of her remaining teeth as a result of poor contouring of the MRP. This was corrected at a second surgery to recontour the plate 7 months after the initial reconstruction.
Two patients experienced hardware-related complications that required plate removal. One patient with intact contralateral maxillary and mandibular dentition and a long-standing history of tooth grinding experienced plate fracture 9 months after reconstruction. A second patient experienced delayed external plate exposure, with external extrusion of the MRP through the skin. She underwent reconstruction of a through-and-through oromandibular defect using a radial forearm free flap that was partially deepithelialized and folded to reconstruct the floor of the mouth and the external facial skin. After receiving postoperative radiation therapy, the plate extruded 6 months postoperatively through the portion of the free flap that was used to reconstruct the lateral chin skin (Figure 1). In both of these patients, plate removal resulted in swing of the mandibular remnant toward the side of the segmental mandibulectomy and some loss of cheek contour.
Four patients died of recurrent cancer during the study period, while 3 were alive with recurrent cancer. The remaining 20 patients were alive and free of evidence of recurrent disease at the end of follow-up. The follow-up period ranged from 6 to 32 months, with a median follow-up of 19.5 months. The MRP remained in place and functioning well in 25 (93%) of the 27 patients at the end of the follow-up period.
Vascularized bone–containing free flaps are a reliable and effective method of oromandibular reconstruction.4 Of the available donor sites, fibula free flaps, iliac crest free flaps, and scapula free flaps are selected frequently for mandibular reconstruction. However, the soft tissue components available in the fibula and iliac crest free flaps may be ill suited for reconstruction of certain composite oromandibular defects. In the case of the fibula flap, the skin paddle is attached to the fibula bone at the site of one or more cutaneous perforating branches of the peroneal vascular pedicle. This sometimes limits the mobility and arc of rotation of the skin paddle relative to the bone graft, making it difficult to use the skin component of the flap for reconstruction of complex 3-dimensional defects of the oropharynx that include portions of the tongue, lateral and posterior pharyngeal wall, and soft palate. In addition, the thickness of the subcutaneous fat in the leg often provides inadequate bulk for tongue reconstruction after total or subtotal glossectomy. The considerable bulk of the skin paddle of the iliac crest flap is better suited for reconstruction of substantial glossectomy defects. However, it may be too bulky to fold for reconstruction of soft tissues within the oropharynx. In addition, when an iliac crest flap is harvested from the ipsilateral hip for reconstruction of lateral mandibulectomy defects without osteotomy, the skin paddle is located on the lateral aspect of the reconstructed mandible after bone graft insetting. This anatomic relationship may limit the arc of rotation of the skin paddle for reconstruction of lateral oromandibular defects. Both the iliac crest and fibula flaps offer a second soft tissue component in the form of the internal oblique muscle and the soleus muscle.5,6 However, it is more difficult to achieve a watertight wound closure when using these muscle components for soft tissue repair, and the transferred muscle will undergo a considerable degree of denervation atrophy after insetting. The soft tissue components of the scapula system of flaps offers the most versatility of the vascularized bone–containing free flaps that are commonly used for oromandibular reconstruction. However, flap harvest requires intraoperative patient repositioning to the decubitus position for most cases of head and neck reconstruction, which eliminates the ability to decrease the length of the surgery by performing simultaneous tumor resection and free flap harvest.
To overcome the limitations of the soft tissue components of vascularized bone–containing free flaps for complex cases of oromandibular reconstruction, Wenig and Keller7 introduced the concept of using a soft tissue free flap that has attributes that are well suited for the defect in conjunction with a bridging mandibular reconstruction plate. Subsequent series of bridging mandibular reconstruction plates documented a high incidence of intraoral plate extrusion after reconstruction of anterior segmental mandibulectomy defects.8,9 However, these series reported a favorable outcome when free flaps and bridging plates were used to reconstruct lateral oromandibular defects. Schusterman and associates8 reported a 93% success rate in 14 patients with lateral mandibular resections who underwent reconstruction using soft tissue flaps and bridging MRPs. Boyd et al9 reported a 95% success rate in 20 patients who underwent lateral oromandibular reconstruction using fasciocutaneous free flaps in conjunction with bridging MRPs.9
In 1996, Blackwell et al1 reported a series of 14 patients who underwent lateral oromandibular reconstruction using soft tissue free flaps in conjunction with bridging THORP MRPs. While the incidence and nature of early postoperative reconstructive complications were acceptable, there was a 29% incidence of delayed reconstructive failure after a median follow-up period of 16.5 months. The most common cause of delayed reconstructive failure was external extrusion of the plate through the cheek skin, which occurred in 3 patients after postoperative intervals of 7, 15, and 15 months, respectively. It was postulated that delayed external plate extrusion occurred as a result of soft tissue contraction within the dead space medial to the MRP. The authors concluded that most simple lateral oromandibular defects were optimally reconstructed using a composite soft tissue–vascularized bone free flap, while some complex lateral defects would require transfer of 2 simultaneous free flaps for separate osseous and soft tissue reconstruction.
In the current series, the incidence of hardware-related reconstructive failure was reduced to 7% in 27 patients who underwent reconstruction using a second-generation, low-profile locking MRP after a similar median follow-up period of 19.5 months. Detailed analysis of the patient and defect characteristics failed to reveal discernible differences between the subjects of the 2 series that might account for the different outcomes. Tumor type and stage, defect classification, and flap selection were similar in the 2 series, and the incidence of radiation therapy in the previous series (100%) was only incrementally higher than in the current series (85%). The prevalence of through-and-through oromandibular defects, which is probably a risk factor for external plate extrusion, was higher in the current series than in the previous series.
In 1994, a second-generation locking MRP was introduced by Stryker Leibinger Inc. This system has a plate profile height of 2.8 mm, compared with a THORP system profile height of 3.5 mm. Perhaps more important than this difference in profile height, the Leibinger Locking System plate has a smooth, rounded contour compared with the geometry of the THORP system (Figure 2). More recently, additional second-generation locking MRPs have been introduced by Synthes U.S.A. (Paoli, Pa), Walter Lorenz Surgical Inc (Jacksonville, Fla), and KLS Martin, LP (Jacksonville, Fla). Most of these systems also offer a lower profile and rounded plate contour when compared with the THORP system. To our knowledge, the current series is the first to investigate the potential role of low-profile locking MRPs as a bridging plate system.
In the current series of low-profile bridging MRPs, there was only 1 case of external plate extrusion, and this occurred after reconstruction of a through-and-through defect. It is possible that the low-profile height and smooth contour of the second-generation locking MRP contributed to a low incidence of plate extrusion through the cheek skin when compared with the first-generation locking MRP. If this is the case, then it is likely that external plate extrusions previously reported in patients with bridging THORP MRPs resulted from a plate geometry that was prone to extrude rather than from soft tissue contraction within the dead space medial to a bridging plate. Indeed, experience with the THORP system for fixation of vascularized bone grafts indicates that there is a notable incidence of plate extrusion even when the plate is used in conjunction with a vascularized bone graft.4 Defects were described in the current series and in the previous series by Blackwell et al1 using the most detailed method of oromandibular defect classification reported in the literature,4 and this analysis failed to reveal significant differences between the 2 groups of patients. However, we recognize the possibility that the decreased incidence of plate extrusion seen in the current series when compared with the previous series of patients who underwent THORP reconstruction may be the result of unidentified variables between the 2 series of patients. For instance, evidence suggests that the length of the segmental mandibular defect and the volume of the resection specimen have an impact on the likelihood of failure after bridging plate reconstruction.10 These factors were not considered in the current analysis.
The incidence of plate fracture was not increased when using a low-profile locking MRP in the current series compared with previous series of bridging plates using the larger THORP system. Possible explanations for why a smaller plate is not more prone to fracture may relate to differences in the titanium alloys used in the 2 plating systems or to differences in plate geometry. When a bending force is applied to a THORP MRP, there is a tendency for plate deformation to occur at the screw holes. Plate deformation occurs primarily between the screw holes when contouring a Leibinger Locking System MRP. In general, plate fractures tend to occur at these sites of plate deformation due to metal fatigue. A THORP plate fracture usually occurs through a screw hole where the plate has a minimum cross-sectional area, while the Leibinger Locking System plate fracture that occurred in the current series occurred between 2 screw holes, where the plate has a maximum cross-sectional area (Figure 3). Regardless of intrinsic plate strength, all patients who are able to chew using remaining maxillary and mandibular dentition are at risk to eventually experience fracture of a bridging MRP. Therefore, longer follow-up will be necessary before the durability of bridging Leibinger Locking System MRPs can be assessed.
A series by Cordeiro and Hildalgo11 reported that the average operative time for patients who underwent oromandibular reconstruction using soft tissue free flaps and bridging MRPs was 721 minutes, which falls within the range of most osteocutaneous flap oromandibular reconstructions. Therefore, the soft tissue free flap with bridging plate option should not be selected for cases of lateral oromandibular reconstruction in an attempt to decrease surgical time or perioperative medical morbidity in patients who are acceptable candidates to undergo a vascularized bone–containing free flap on the basis of defect analysis. Indeed, the incidence of major medical complications in the current series was 39%, compared with a 17% incidence of major medical complications seen during the senior author's (K.E.B.) career experience with 119 microvascular head and neck reconstructions.12 However, the increased incidence of medical complications in the current series probably reflects our bias to limit the complexity of reconstruction in a patient population with substantial preoperative comorbidity who are at high risk to experience complications.
Most of the patients in this series would have required 2 simultaneous free flaps to achieve an osseous reconstruction with a good-quality soft tissue reconstruction. Performance of 2 simultaneous free flaps adds approximately 2 hours to the length of surgery when compared with performance of a single free flap.13 It is doubtful that this incremental increased length of surgery would result in a significantly increased incidence of medical complications. However, use of 2 simultaneous free flaps results in increased donor site morbidity; there is no doubt that performance of 2 simultaneous free flaps adds to the complexity of reconstruction when compared with using a single soft tissue free flap with a bridging plate. The increased morbidity resulting from the complexity of simultaneous free flap transfer is reflected in the 7% incidence of total flap failure reported in 43 cases of dual free flap oromandibular reconstruction described in the literature,13- 16 compared with the 100% flap survival achieved in the current series. In addition, Nakatsuka et al15 reported a 41% incidence of early perioperative reconstructive complications in a series of dual free flap oromandibular reconstruction, while there were no cases of significant perioperative reconstructive complications in the current series.
Reconstruction was successful in 93% of 27 cases of lateral oromandibular reconstruction using soft tissue free flaps in conjunction with second-generation bridging locking MRPs. The high incidence of delayed external plate extrusion seen previously with THORP plates used in this manner was reduced by using a low-profile, rounded-contour plate in the current series, and use of a smaller plating system did not result in an increased incidence of plate fracture after a similar length of follow-up. Vascularized bone–containing free flaps remain the preferred method for reconstruction of most lateral oromandibular defects. However, preliminary analysis of the current series indicates that a soft tissue free flap used with a low-profile bridging plate might suffice in cases where a significant soft tissue resection would otherwise require transfer of 2 simultaneous free flaps. Final assessment of the durability of this technique will require long-term follow-up.
Accepted for publication May 14, 1999.
Reprints: Keith E. Blackwell, MD, Box 951624, UCLA Medical Center, Los Angeles, CA 90095-1624.