A, Degloving exposure of the maxilla; B, down fracture of the maxillary arch; and C, manner of 4-point fixation.
Cephalometric measurements. S indicates starting growth position; F, final growth position; P, predicted growth position; ΔX, distance of horizontal growth of the maxilla; and ΔY, distance of vertical growth of the maxilla.
Lowlicht RA, Jassin B, Kim M, Sasaki CT. Long-term Effects of Le Fort I Osteotomy for Resection of Juvenile Nasopharyngeal Angiofibroma on Maxillary Growth and Dental Sensation. Arch Otolaryngol Head Neck Surg. 2002;128(8):923-927. doi:10.1001/archotol.128.8.923
To analyze the long-term effects of the Le Fort I osteotomy approach for the resection of juvenile nasopharyngeal angiofibroma (JNA) on maxillary growth and dental sensation.
Prospective collection of structured data.
Tertiary care academic teaching hospital.
Between 1993 and 1998, 5 adolescents (aged 10-14 years) constituted the evaluable cohort among 14 patients who underwent Le Fort I osteotomy for JNA resection. Mean follow-up was 47.2 months.
The Le Fort I osteotomy approach was used to resect JNA. Cephalometric x-ray films were taken at various postoperative intervals to assess maxillary growth. The results were matched against age-correlated predictions from Dentofacial Planner software.
Main Outcome Measures
Horizontal and vertical maxillary growth were each measured anteriorly and posteriorly by comparing interval postoperative cephalometric x-ray films. Dental sensation was longitudinally evaluated by performing interval pulp testing postoperatively.
(1) Average vertical growth of the maxilla achieved 30% of predicted growth anteriorly (P = .02). (2) Average horizontal growth matched predicted growth in all patients. (3) All patients demonstrated long-term maxillary dental denervation.
Le Fort I osteotomy provides excellent surgical exposure for resection of JNA in the growing facial skeleton. Although it significantly affects vertical but not horizontal growth, its cosmetic effect is negligible. It also causes long-term dental denervation, which in most cases is undetected by patients.
THE LE FORT I osteotomy approach can provide an excellent method for the surgical extirpation of tumors involving the central skull base. The main advantages of this approach are the wide surgical exposure, the clear visibility of resection margins, and the lack of superficially visible scars.1
However, it is surprising how little information has been reported about the effects of this operation on maxillary growth. Normally, vertical growth of the maxilla occurs at the growth processes located on the attachments of the frontal and zygomatic bones and on the lower aspect of the alveolar process.2 The growth centers responsible for increasing the length of the maxilla are located adjacent to the attachment of the palatine bone and the maxillary tuberosities.3 Growth spurts of the maxilla occur between 10 and 12 years in girls and between 12 and 14 years in boys.4 However, this growth can continue into adulthood.5
Another consequence of a Le Fort I osteotomy is the potential loss of dental sensory function. With this approach, the infraorbital nerve is spared. However, the nasopalatine nerve, and the anterior, middle, and posterior superior alveolar nerves are cut.6 In addition, the greater palatine nerve may be disrupted from manipulation of the maxilla to increase surgical visualization. Although the Le Fort osteotomy is designed to avoid denervation of the soft tissues of the face, denervation of the teeth may be unavoidable.
The purpose of our study was to analyze the effects of Le Fort I osteotomy on the growth of the maxilla and on the long-term sensation of the teeth.
This study involved 5 boys aged 10 to 14 years constituting the evaluable cohort identified among 14 patients undergoing a Le Fort I osteotomy for the resection of juvenile nasopharyngeal angiofibroma (JNA) between 1993 and 1998 at Yale–New Haven Hospital, New Haven, Conn (Figure 1). Because the age period of greatest maxillary growth is from 8 to 16 years,8 we limited our study to patients in this age group, thus excluding 9 of 14 patients older than 16 years. The mean follow-up was 47.2 months. Preoperative embolization of the tumor feeding vessels was carried out in all patients. All operations were performed by 2 of us (C.T.S. and R.A.L.). The surgical technique involved sacrifice of the greater palatine arteries, application of rigid titanium fixation in all cases, and removal of impacted third molars. Maxillomandibular fixation was not applied postoperatively. All the patients had class I occlusion. One patient had preoperative and postoperative orthodontic therapy and another had postoperative orthodontic therapy.
Each patient had a reference lateral cephalometric x-ray film taken in a Weimer head frame at 90 kilovolt peak (kVp)/0.1125 milliampere second (mAs) within 1 week after surgery. The film was then repeated at a future interval varying from 8 to 108 months postoperatively. Each cephalogram was digitized by one of us (M.K.). Each image contained 66 points of data that were entered into Dentofacial Planner software (Dentofacial Software Inc, Toronto, Ontario). This software constructs a growth estimate curve for the location of the maxilla based on the age of the patient at the time of surgery, immediate postoperative skeletal landmarks, and the interval period.
The cephalometric growth data were analyzed in the following fashion. An axis was created parallel to the Frankfort horizontal and perpendicular to the Frankfort horizontal (Figure 2). The Frankfort horizontal is the distance from the superior aspect of the external auditory meatus (porion) to the inferior aspect of the orbital rim (the orbitale). Horizontal growth of the maxilla in an anterior fashion was defined as a positive right change along the x-axis. Vertical growth was defined as positive down change along the y-axis. The points to define maxillary position were the anterior nasal spine (ANS) and the posterior nasal spine (PNS). The ANS was defined as the point at which the anterior process reached a thickness of 3 mm, while the PNS was defined as the posterior end point of the hard palate (Figure 2).
Thus, spatial movement as a function of time of ANS defined anterior growth, and spatial movement as a function of time of PNS defined posterior growth. We then compared the actual growth of the maxilla as a percentage of growth predicted by the Dentofacial Planner software. The average of the actual growth was compared with the average predicted growth and the results were analyzed for statistical significance using a paired t test.
In the first week after surgery, the entire maxillary and mandibular dentition were tested electrically for pulp sensation using a monopolar Digitest Pulp Vitality Tester (model D626D; Parkell Electronics Division, Farmingdale, NY). The measurement technique used a constant current of 150 µA, as described previously by Matthews and Searle.7 Because surrounding soft tissue moisture contamination can introduce inaccuracies in pulp test data, each tooth tested was isolated from the saliva with soft tissue retraction, cotton rolls, and air drying. Current was applied incrementally until the patient was able to identify a stimulus. This test was then longitudinally repeated at a postoperative interval varying from 8 to 108 months postoperatively.
Table 1 summarizes the results of our study. For each patient, the follow-up interval is shown along with the actual measured maxillary growth and the predicted maxillary growth. The "S" point indicates the starting position of the ANS or PNS of the patient, as demonstrated in the cephalometric x-ray film taken 1 week after surgery. The "F" point indicates the final position of the ANS or PNS of the patient, as shown in the final cephalometric x-ray film taken at the indicated postoperative interval for each patient. The "P" point indicates the predicted position of the ANS or PNS of the patient, as shown in the cephalometric construct made for the patient from the Dentofacial Planner software. Thus, the distance S to F is the actual interval, and the distance S to P is the predicted interval growth. Based on historical data,8 all of the patients should have experienced adolescent growth during the interval.
For example, Table 1 shows that for patient 2, the horizontal growth of the maxilla (along the x-axis) is presented in the ANS (X) column. The ANS point moved 1 mm in the horizontal direction from the starting point to the finishing point (S-F distance). The predicted movement of the ANS point (S-P) was 1 mm. It appears that at 16 months postoperatively, this patient had exhibited predicted maxillary growth in the horizontal direction.
The vertical growth of the anterior maxilla (along the y-axis) is presented in the ANS-Y column. For patient 2, ANS moved 0 mm (S-F measurement). The predicted movement of the ANS point (S-P measurement) was 2 mm. Thus, there was 0% of predicted anterior maxillary vertical growth at 16 months postoperatively. Similar results were obtained by analyzing the movement of the PNS point in the PNS-X and PNS-Y columns.
Table 1 also summarizes the average growth data for all patients. It appears that all patients grew 100% of the predicted growth along the horizontal axis, ie, ANS (X) and PNS (X) are shown as 100% of predicted. However, growth of the maxilla along the vertical axis was not as complete. The average vertical growth of our patients along the anterior maxilla (as described in the ANS-Y column) was 30% of the predicted growth. This result was found to be statistically significant compared with the average expected growth using a paired t test (P = .02). For the posterior maxillary segment (PNS-Y), the average vertical growth was 27% of the predicted growth. This result was not found to be statistically significant (P = .06). Interestingly, the reduction in vertical maxillary growth was largely subjectively unnoticed in our patient cohort.
In all patients, there was no evidence of sensation as recorded by electrical stimulation to all the maxillary teeth at the postoperative interval period. However, pinprick sensation was intact in the adjacent palatal mucosa, and dental numbness went largely unnoticed. These results are summarized in Table 2.
Although there are numerous studies of the effects of plating in long bone growth, few specifically address the growth effect across maxillary suture lines. There are 2 studies on the effect of Le Fort I osteotomies on maxillary growth. However, both of these studies were made on patients undergoing surgical repositioning of the maxilla for correction of dentofacial deformities. A 1997 study showed normal maxillary growth in patients undergoing a Le Fort I osteotomy for excessive vertical maxillary growth.9 In addition, these patients underwent concomitant orthodontic therapy. The method of evaluation involved measurement of the presurgical and postsurgical spatial change in ANS and PNS in a lateral cephalogram. The surgical group was compared with unoperated-on matched controls from a data bank similar to that of the Dentofacial Planner software. The mean age of the patients was 14.5 years and the mean postsurgical follow-up was 25 months. In our study only 2 of the 5 patients underwent postsurgical orthodontics, and our mean follow-up was 47.2 months.
An earlier study that had examined the maxillary growth of subjects (aged 13-17 years) who had undergone surgical repositioning of the maxilla with interosseous wiring and intermaxillary fixation found that there was an increase of the vertical dimension of the maxilla after surgery.10 However, no comparison with a control group was made. Patients in our series differed from patients in previous studies in that rigid fixation plates across the horizontal osteotomy were used to stabilize the maxilla. Because JNA is located posterior to the maxilla, it is unlikely that the growth of the tumor played a role in disturbing maxillary growth centers. We believe the effect on vertical maxillary growth is therefore surgically related.
All of our patients exhibited denervation of the maxillary teeth in the long-term postoperative period. However, a study by Al-Din et al11 that looked at the sensory nerve disturbances that occur after Le Fort I osteotomy showed contrasting results. They reported that of patients who had a positive response to an electric pulp tester preoperatively, 78% regained sensitivity to pulp testing after 6 months. Other studies of sensation in teeth following Le Fort I osteotomy also indicate that there are no long-term sensory sequelae in the teeth after this procedure. A study by De Jongh and coworkers12 in 1986 compared electric pulp testing of 10 patients after Le Fort I osteotomy with 10 matched controls who did not have any surgery. He found that 14 months after surgery, 71% of the teeth tested were responsive to electric pulp stimulation. The figure in the control group was 93%. Other studies by Kahnberg and Engstrom13 and Tajima14 also demonstrated that most patients who underwent Le Fort I osteotomy recovered teeth sensation to pulp testing 6 to 18 months after surgery. In contrast, all the patients in our study showed long-term dental denervation following the Le Fort I osteotomy—explained in part by our necessity to section both anterior and posterior superior alveolar nerves.
In examining palatal sensation, Al-Din et al11 found that the return of fine touch and pinprick sensation of the palate was affected by whether the greater palatine nerve was divided during surgery. That all of the patients in our study were able to regain pinprick sensation in the postoperative period confirms this explanation since palatine nerves were stretched by the down fracture but anatomically preserved.
Despite the important advantages of surgical exposure, the clear visibility of resection margins, and the avoidance of visible facial scars, the Le Fort I osteotomy approach comes with certain postoperative consequences. This study demonstrates that patients who undergo Le Fort I osteotomy for JNA resection have long-term dental denervation, although, for the most part, this was unnoticed by our patients. These results contrast to those of previous studies in patients undergoing Le Fort I osteotomy for skeletal repositioning (orthognathics).
We also confirm that Le Fort I osteotomy performed with metal rigid fixation is unlikely to have an impact on horizontal maxillary growth. However, we demonstrate that there is a statistically significant decrease of anterior vertical maxillary growth. As all of our patients developed a functional occlusion, this effect was not clinically significant, nor was the effect on vertical maxillary growth noticeable to the patients or their parents.
Accepted for publication January 9, 2002.
This study was supported in part by the McFadden Endowment, the Harmon Endowment, and the Mirikitani Endowment.
This study was presented at the Eastern Section meeting of the Triological Society, Pittsburgh, Pa, January 29, 2000.
Corresponding author: Roger A. Lowlicht, DDS, 185 Maple Ave, North Haven, CT 06473.