We present a prospective randomized experimental study comparing bone regeneration obtained in 60 post-traumatic frontal sinuses obliterated with either calvarial bone dust ( n = 30, group I) or calvarial bone and demineralized bone matrix (DBM; n = 30, group II). Radiological follow-up included high-resolution computed tomography with quantitative micro-density analysis in Hounsfield units (HU), together with a volumetric evaluation of the ossification at 6 and 24 months after surgical treatment. Epidemiological information and potential drawbacks were analysed. Bone volume and density data (HU) for the regenerated areas were subjected to statistical analysis at 6 and 24 months for both groups. Results were compared with reference values obtained from frontal and temporal bone in every patient. Complications developed for 10% of operated sinuses. The resulting bone formation (HU) in group I patients was significantly better than that obtained in group II. Ossification progressed in a statistically significant manner in both groups when compared at 6 and 24 months postoperatively. The use of DBM as a biomaterial associated with calvarial bone dust for sinus obliteration shows long-term safe results, similar to autogenous bone, but with a lower final bone density.
One of the main objectives when treating a frontal sinus fracture is to create a ‘safe sinus’, minimizing patient morbidity. The basic principles of frontal sinus obliteration have been developed over the last 50 years. Traditionally, autogenous fat and temporalis muscle have been the materials commonly used for this purpose. Nevertheless, the best long-term results have been described with the use of autogenous bone. Autogenous bone grafting is considered the gold standard for sinus obliteration; however the availability of autogenous bone is limited and the procedure is associated with donor site morbidity. An ideal biomaterial that is able to overcome these drawbacks has not yet been identified.
Although the combining of autogenous bone grafts with biomaterials has been described in multiple studies, no relevant advantages have been demonstrated to date. Nevertheless, different mixtures and proportions may be used, taking advantage of the various osteogenic, osteoinductive, and osteoconductive properties of the biomaterials currently available. The demineralized bone matrix (DBM) DBX (Musculoskeletal Transplant Foundation, Edison, NJ, USA) is one such commercially available preparation (DePuy Synthes) and contains demineralized processed human cadaveric bone combined with sodium hyaluronate.
The objective of this study was to compare the ossification obtained when obliterating the post-traumatic frontal sinus with either autogenous bone or a mixture of autogenous bone and a ‘bone substitute’ (DBM). Long-term radiological follow-up with high-resolution computed tomography (HRCT) was used to assess the quality of ossification and to detect possible complications.
Materials and methods
A prospective randomized experimental study of 60 post-traumatic frontal sinuses was performed. All patients received surgical treatment (performed by the same team, always including the first author) between January 2004 and January 2008. We included a group of patients treated with calvarial bone shavings plus DBM who have been reported previously in this journal. All patients were treated and controlled as established in the protocol approved by the clinical research ethics committee of the study hospital, respecting the ethical standards applied to this type of study. This was a single-blind study; the patients did not know to which study group they had been assigned. The frontal sinus obliteration was performed in every patient using a subcranial approach, and the bone density of the regenerated bone was analysed long-term.
The inclusion criteria were as follows: (1) Patient with a frontal sinus fracture suitable for obliteration in accordance with the treatment algorithm proposed by Rohrich and Hollier. (2) Medically stable patient able to undergo the proposed surgical procedure; this procedure was usually done during the first week after the trauma. (3) Surgical treatment for every patient included a subcranial approach.
The surgical protocol included a coronal zigzag incision, access to the sinus by temporary removal of the anterior wall fracture fragments, removal of the sinus mucosa (including sinus wall drilling), naso-frontal duct obliteration with a bone graft, and sinus cavity filling with either calvarial bone shavings (obtained with a bone harvesting scraper device), or calvarial bone plus DBM. Bilateral fractured sinuses were treated independently; if the bony septum was incomplete or fractured, a calvarial bone graft was used to reconstruct the septum. Patients with bilateral fractures received the same treatment on both sides. An anterior sinus wall reconstruction was performed afterwards, with either a bone fragment reduction and osteosynthesis (with titanium miniplates and screws) or titanium mesh placement, or a combination of the two.
Two study groups were created in a randomized manner. Group I comprised 30 sinuses obliterated with calvarial bone shavings. Group II comprised 30 sinuses obliterated with a mixture of calvarial bone shavings and DBM, 50% each.
Calvarial bone harvesting was performed using a bone scraper (Autogenous, distributed by Mozo Grau Ltd., Spain) under saline solution irrigation, obtaining high quality bone shavings from the outer calvarial and diploic space with minimal donor site morbidity ( Fig. 1 ).
For group II, an average of 5 ml of DBM mixed with a similar amount of calvarial bone shavings was used for the obliteration of a single frontal sinus ( Fig. 2 ).
Standard postoperative short-term follow-up was done for every patient with clinical and radiological (CT scan) assessments.
Long-term patient follow-up was done in every case by means of clinical assessment at 6, 12, and 24 months and imaging. HRCT was performed at 6 months after surgery, the main study objective being a quantitative bone density analysis comparing the results obtained in the two study groups and a comparison with the frontal (R1) and temporal bone (R2) densities of every patient. Postoperative complications were also screened at this time. HRCT was done again at 24 months after the primary surgery, with the same study objectives as for the 6-month HRCT scan, comparing the results and relating them to complications. An independent CT workstation, the Brilliance Extended Workspace, was used (Philips Healthcare). The image display matrix was 1.280 × 1.024 pixels. Additional software used included ‘Image Fusion’ and ‘Bone Mineral Analysis’ (Philips Healthcare).
Axial, sagittal, and coronal images were generated using multi-planar reconstruction (MPR) in order to calculate the volume of every obliterated sinus and the corresponding average Hounsfield units (HU) – numeric information contained in each pixel related to tissue density. All processed images were used in DICOM (Digital Imaging and Communications in Medicine) format.
Analysis of the relationship between quantitative and qualitative variables was performed for two categories with the Student’s t -test. Non-parametric tests were used if needed. Associations among quantitative variables were assessed using the Spearman correlation coefficient. All results were stratified according to the study group and sinus side. All results were presented with a level of confidence of 95% ( P ≤ 0.05). The data analysis and graphics were generated using Stata statistical software (StataCorp., College Station, TX, USA).
A total of 60 frontal sinuses were treated in 39 trauma patients; 33 of the patients were males and six were females. The mean patient age was 36 years (range 15–76 years). The average length of clinical follow-up was 36 months (range 24–72 months). Seventeen males presented bilateral fractures and 16 a unilateral fracture, for a total of 50 frontal sinuses treated. Four females presented bilateral fractures and two a unilateral fracture, for a total of 10 sinuses treated. Only six females were included in the study, with three assigned to each study group.
The registered data were used to assess bone density (in HU) for both study groups at 6 months ( Table 1 ) and at 24 months ( Table 2 ). New bone formation (HU) in group I (autogenous calvarial bone) was found to be significantly better than that obtained in group II (bone + DBM), at both 6 and 24 months.
|H 0 : HU 6 months (group I) = HU 6 months (group II); z = 6.372; P = 0.0000|