Assessment of the sinus lift operation by magnetic resonance imaging

British Journal of Oral and Maxillofacial Surgery (1999) 37, 285–289 © 1999 The British Association of Oral and Maxillofacial Surgeons

BRITISH JOURNAL OF ORAL

& M A X I L L O FA C I A L S U R G E RY

Assessment of the sinus lift operation by magnetic resonance imaging C.F. Gray, T.W. Redpath,* F.W. Smith,† R.T. Staff, R. Bainton‡ Department of Bio-Medical Physics, University of Aberdeen, Foresterhill, Aberdeen; *Department of Bio-Medical Physics and Bio-Engineering, University of Aberdeen, Foresterhill, Aberdeen; †MRI Centre, Woodend Hospital, Eday Road, Aberdeen; ‡Maxillofacial Department, Aberdeen Royal Infirmary, Foresterhill, Aberdeen, UK SUMMARY. Sinus lift operations are done to increase bone thickness before placement of osseointegrated implants in areas of the maxilla where there is insufficient bone between the lower border of the maxillary sinus mucosa and the oral mucoperiosteum. Full sectional information should be available to the surgeon before the sinus lift operation so that the outcome can be predicted, and to allow for assessment of the volume of bone graft required. It is also useful to assess the success of the procedure before placement of the implant, and to choose the implant length and orientation required for maximum stability. We describe the use of a 0.2 tesla magnetic resonance imaging (MRI) scanner to image a patient before and after sinus lift. MRI gave full sectional information of the bone on both occasions without using ionizing radiation. The quality of the images was good, and should be considered as an alternative to the accepted gold standard of computed tomography (CT), which exposes the patient to a high dose of radiation.

even if sophisticated software is used,10 dimensional accuracy is dependent on the positioning of the patient, which may generate errors.11 With MRI, we are able to image directly at any chosen plane without the need for reformatting. Mid-field MRI using a 1 tesla scanner12 and low-field MRI using a 0.2 tesla scanner13 have been described for assessment before an implant.

INTRODUCTION The technique of sinus lift before placement of an osseointegrated implant in areas of the maxilla where there is insufficient bone between the lower border of the mucosa of the maxillary sinus and the oral mucoperiosteum is well established.1 Full three-dimensional assessment of the region before the sinus lift operation and again before placement of the osseointegrated implants is advisable both to allow planning of the lift, and to allow assessment of the operation and selection and alignment of the optimum placement of the implant. Pre-implant assessment on panoramic radiographs gives only two-dimensional information, which may be inaccurate.2 Computed tomography (CT) is also used2–5 and, though it gives sectional information, patients are subjected to a radiation dose of roughly 150–300 times that of a panoramic radiograph.6 This is difficult to justify under current radiation protection guidelines for the UK if magnetic resonance imaging (MRI) is an effective alternative.7 MRI usespowerful magnets and radiowaves8 rather than ionizing radiation and, provided that certain guidelines are adhered to, is regarded as being without risk.9 One of the most important guidelines is that patients must be carefully screened for contraindications to MRI; these include the presence of cardiac pacemakers, some of the early heart valves, surgical clips (particularly in the brain), and shrapnel wounds (particularly in the eyes). Pregnant women should not be scanned during the first trimester; there is no confirmed hazard, but this is considered prudent.9 Sectional CT information of the oral regions is normally reformatted from axial sections to reduce the radiation dose to the eyes and thyroid gland but,

PATIENT AND METHODS The patient was examined both before and after a right maxillary sinus lift operation, using an ‘Open Viva’ low-field MR imager operating at 0.2 tesla (Siemens AG, Erlangen, Germany). This scanner is an open design, in which the patient lies under the upper pole of the magnet with the head in the system’s radiofrequency head coil (Fig. 1). Open scanners are better tolerated than conventional MR scanners in which patients are examined within a tunnel; this induces feelings of claustrophobia in a substantial minority of people. The patient was examined one week before the sinus lift operation and again using an identical protocol six weeks later, one week before placement of the implant. The sinus lift was done under general anaesthetic through a sulcus incision, and the buccal surface of the maxillary sinus was exposed. A 1.5-cm fenestration was made in the buccal bone with a rose-head bur, and the bone plate was dissected off the underlying intact sinus mucoperiosteum. The sinus mucoperiosteum was raised using a ball end burnisher, which minimizes the risk of perforating the sinus mucosa. A cancellous bone graft harvested from the hip by a 285

286

British Journal of Oral and Maxillofacial Surgery

Fig. 1 – The 0.2 tesla open scanner and the radiofrequency headcoil.

coffin lid technique was packed into the space between the raised sinus mucoperiosteum and the bony wall of the maxilla. The wound was closed primarily. Antibiotic prophylaxis was given before the procedure, and continued for five days. The patient was advised against nose-blowing or sneezing. At each examination session, the patient was first given a fast pilot scan with a low spatial resolution gradient echo sequence to check alignment. This takes 19 seconds, and gives images in the transaxial, sagittal, and coronal planes. An intravenous injection of paramagnetic gadolinium contrast (Magnevist, Schering AG, Berlin, Germany) was then given to increase the MRI signal from the mucosa, making it more visible as an area of high signal in a T1-weighted MRI image. The sagittal pilot (Fig. 2) was then used to plan seven high-resolution transaxial T1-weighted images, which give good a good picture of the relevant structures. One of these transaxial slices (Fig. 3) is then used to plan a further seven high-resolution slices at right angles to the maxilla through the right maxillary sinus (Fig. 4). Again using the transaxial set, we were able to plan a further set of quasisagittal images parallel to the maxilla, with both sides being pictured for comparison. The planned angulation of the sets is shown in Figure 5, and one of the resulting right-sided preoperative images in Figure 6. Both the left- and right-sided sets can be taken simultaneously, as the slices do not intersect over any area of importance. Each high-resolution fast spin-echo T1-weighted sequence takes five minutes 16 seconds, and can image a total of seven slices simultaneously. The image parameters were as follows: TR (time for repetition) = 620 ms; TE (time to echo) = 24 ms; field-of-view = 200 mm; matrix = 252×256 pixels; number of signal averages = 6; echo-train length=3; signal acquisition

Fig. 2 – The sagittal pilot image being used to set up the axial slices. The white lines indicate the planned position of each axial slice.

Fig. 3 – A high-resolution transaxial image being used to set up images at right angles to the maxillary arch on the patient’s right. The white lines indicate the planned position of each slice.

bandwidth = 130 Hz/pixel; slice thickness = 4.0 mm; and interslice gap = 0.4 mm. Measurements such as bone height may be made either with electronic callipers at the workstation or directly on the hard copy using the printed scale.

Assessment of the sinus lift operation by MRI

287

Fig. 5 – A high-resolution transaxial image being used to set up images parallel to the maxillary arch on both sides.

Fig. 4 – A preoperative high-resolution image at right angles to the right maxillary arch.

Fig. 7 – A preoperative high-resolution transaxial image clearly showing an intrasinus transverse septum on the patient’s right (arrow). A septum is also visible in the left maxillary sinus.

RESULTS

Fig. 6 – A preoperative high-resolution image parallel to the right maxillary arch.

From the initial set of preoperative images, we can clearly see the relevant dimensions of the maxillary sinus. From the high-resolution transaxial set, we can see the convoluted floor of the sinus together with a septal wall within each sinus (Fig. 7). The postoperative images in all three planes (Figs 8–10) confirm the increased vertical bone height, and

288

British Journal of Oral and Maxillofacial Surgery

A

B

Fig. 8 – Postoperative high-resolution transaxial images; (A) is at a higher level than (B), and the augmented bone is clearly seen on the patient’s right (arrows).

Fig. 9 – A postoperative high-resolution image at right angles to the maxillary arch on the patient’s right side, corresponding to the pre-operative scan shown in Figure 4. The augmented bone is clearly visible (arrow).

Fig. 10 – A postoperative high-resolution image parallel to the right maxillary arch, corresponding to the preoperative scan shown in Figure 6. The augmented bone is indicated (arrow).

make the final positioning of the dental implants predictable after cross-referencing. Table 1 gives measurements of the bone heights before and after operation

in four adjacent slices through the relevant area of the sinus, made from image sets taken at right angles to the maxilla.

Assessment of the sinus lift operation by MRI

289

Table 1 – Vertical bone height in the right maxilla measured before and after operation from four adjacent slices, from image sets taken at right angles to the maxillary sinus through the augmented region Slice number 1 2 3 4

Preoperative height (mm)

Postoperative height (mm)

12 11 14 14

30 29 28 20

DISCUSSION MRI provided full sectional information about the sinus both before and after operation. Such scanning is useful because it allows planning of the sinus lift operation itself, as well as planning the subsequent placement of the implant. The vertical bone heights were easy to measure. The quality of the scans is sufficiently good to facilitate estimation of the volume of defect to be filled, by reference to the parallel and right-angled scans. This was not done in this case, and would require further work to verify. For complex cases with few reference points, a surgical/imaging template with gadolinium reference markers should be used to aid interpretation, a technique which we have used when using MRI to plan placement of implants.12–13 For ease of orientation and to avoid errors, each film should display the pilot image and the position of the images given. CT gives a substantial dose of radiation to the patient, so we think that scanning both before and after the operation would result in excessive radiation. MRI does not use ionizing radiation, so it should be considered as an alternative to CT, at least for the scan before sinus lift. However, we think that MRI can also be used for assessment for the implant.12–13 Although this is a separate issue, if it is correct, it implies that MRI could be used to plan the whole procedure, from sinus lift to placement of the implant. Any possible geometric distortion from old ferromagnetic dentine pins is limited, as any artefact is extremely localized, and implant sites are in edentulous regions.12–13 Modern pins are less likely to cause this effect, as they are less likely to be strongly ferromagnetic. However, measurements of distance are likely to be invalid near these artefacts and should not be attempted in affected areas.12–13 The small geometric distortion caused by disturbance of the magnetic field at air/tissue interfaces does not seem to cause appreciable error, but further work is needed to confirm this.13 Should it prove to do so, low-field MRI scanners such as the one that we used are likely to give proportionally lower levels of error than mid-field and high-field scanners.13 References 1. Boyne PJ, James RA. Grafting of the maxillary sinus floor with autogenous marrow and bone. J Oral Surg 1980; 38: 613–616. 2. Jeffcoat M, Jeffcoat RL, Reddy MS, Berland L. Planning interactive implant treatment with 3-D computed tomography. J Am Dent Assoc 1991; 122: 40–44.

3. Schwarz MS, Rothman SLG, Chafetz N, Rhodes M. Computed tomography in dental implantation surgery. Dent Clin North Am 1989; 33: 555–597. 4. Lacan A, Étienne D. Scanner en implantologie dentaire. Ann Radiol (Paris) 1990; 33: 418–422. 5. Imhof K. Dental CT: a new program for planning and evaluating implantations of the jaw. Electromedica 1992; 60: 26–29. 6. National Radiological Protection Board. Guidelines on radiology standards for primary dental care. Documents of the National Radiological Protection Board 1994; 5: 1–57. 7. National Radiological Protection Board. Protection of the patient in x-ray computed tomography. Documents of the National Radiological Protection Board 1992; 3: 1–16. 8. Stark DD, Bradley WG, eds. Magnetic Resonance Imaging. St Louis: CV Mosby, 1988. 9. National Radiological Protection Board. Board statement on clinical magnetic resonance diagnostic procedures. Documents of the National Radiological Protection Board 1991; 2: 1–29. 10. Casselman JW, Deryckère F, Robert Y, Declercq C, Neyt L. Dentascan: programme de reconstruction tomodensitométrique utilisé pour l’évaluation anatomique du mandible et du maxillaire dans le bilan pré-opératoire des implants dentaires. Ann Radiol (Paris) 1990; 33: 408–417. 11. Kohavi D, Bar-Ziv J, Marmary Y. Effect of axial plane deviation on cross-sectional height in reformatted computed tomography of the mandible. Dentomaxillofac Radiol 1997; 26: 189–191. 12. Gray CF, Redpath TW, Smith FW. Pre-surgical dental implant assessment by magnetic resonance imaging. J Oral Implantol 1996; 22: 147–153. 13. Gray CF, Redpath TW, Smith FW. Low-field magnetic resonance imaging for implant dentistry. Dentomaxillofac Radiol 1998; 27: 225–229.

The Authors Crawford F. Gray BDS MSc DGDP (UK) Research Fellow Roger T. Staff PhD Senior Medical Physicist Department of Bio-Medical Physics Thomas W. Redpath PhD, FInstP Department of Bio-Medical Physics and Bio-Engineering University of Aberdeen Foresterhill Aberdeen Francis W. Smith MD, FRCR Consultant Radiologist MRI Centre Woodend Hospital Eday Road Aberdeen Roger Bainton FDS FRCS Consultant Oral and Maxillofacial Surgeon Maxillofacial Department Aberdeen Royal Infirmary Foresterhill Aberdeen, UK Correspondence and requests for offprints to: Dr Thomas W. Redpath, Senior Lecturer, Department of Bio-Medical Physics and Bio-Engineering, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, UK Paper received 24 July 1998 Accepted 5 March 1999