Int. J. Oral Maxillofac. Surg. 2004; 33: 134–140 doi:10.1054/ijom.2003.0463, available online at http://www.sciencedirect.com
Clinical Paper Orthognathic Surgery
Recovery of nerve injury after mandibular sagittal split osteotomy. Diagnostic value of clinical and electrophysiologic tests in the follow-up
T. Teerijoki-Oksa1, S. K. Ja¨a¨skela¨inen2, K. Forssell1, H. Forssell1 Departments of 1Oral and Maxillofacial Surgery and 2Clinical Neurophysiology, Turku University Hospital, Lemminka¨isenkatu 2, 20520 Turku, Finland
T. Teerijoki-Oksa, S. K. Ja¨a¨skela¨inen, K. Forssell, H. Forssell: Recovery of nerve injury after mandibular sagittal split osteotomy. Diagnostic value of clinical and electrophysiologic tests in the follow-up. Int. J. Oral Maxillofac. Surg. 2004; 33: 134–140. 2003 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Abstract. The diagnostic value of several clinical, quantitative sensory tests (brush-stroke directional discrimination (BSD), touch detection threshold (TD), warm/cold (W/C) and sharp/blunt discrimination (S/B)), and electrophysiologic tests (mental nerve blink reflex (BR), nerve conduction study (NCS), cold (CDT), and warm (WDT) detection thresholds) in the recovery of inferior alveolar nerve (IAN) injury was evaluated in a prospective 1-year follow-up study of 20 patients after bilateral sagittal split osteotomy (BSSO). The subjective sensory alteration was assessed from patients’ drawings. The predictive values of different tests at 2 weeks were determined in relation to the subjective sensory recovery at 12 months. The most pronounced recovery of the nerve damage occurred during the first 3 months according to all measures used. After 3 months, the electrophysiologic tests, especially the NCS, indicated significant further improvement. Except for the TD test, all other clinical test results were normal already at 3 months postoperatively. At early and late controls, the NCS and the thermal quantitative sensory testing could best verify the subjective sensory alteration, and most accurately assess the degree of thick and thin fibre dysfunction. At 1 year, the nerve dysfunction, as revealed by the NCS, corresponded with the figures of sensory alteration reported by the patients (35% R, 40% L). The W/C, BSD, S/B and WDT tests had the best early positive predictive values. Electrophysiologic tests had higher negative predictive values compared to clinical tests.
Bilateral sagittal split osteotomy (BSSO) has become a standard procedure to correct mandibular deformities, despite the high risk of damage to the inferior alveolar nerve (IAN) associated with it. An accurate knowledge of the rate and 0901-5027/04/020134+07 $30.00/0
time of recovery of the nerve damage would be essential for the treating surgeons and for the patients. So far, however, the lack of standard and objective methods to evaluate the sensory disturbances after BSSO has resulted in greatly
Key words: inferior alveolar nerve; sensory tests; electrodiagnosis; nerve injury; sagittal split osteotomy; trigeminal nerve. Accepted for publication 9 June 2003
varying figures for permanent sensory disturbances, ranging from 0 to 72%21,29. The relation between the sensory test results and the subjective sensory disturbance of the patients has varied widely in many previous
2003 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Recovery of nerve injury after BSSO prospective studies assessing the recovery of IAN after BSSO3,6,10,14,27,31,32. In our previous study on intraoperative nerve damage during BSSO25, almost all electrophysiologic tests showed a clear association with the intraoperative nerve damage. Of the clinical sensory tests, the touch detection test (TD) showed the closest association with the degree of nerve injury at the end of the operation. In electrophysiologic diagnostics, the mental nerve blink reflex (BR) and the nerve conduction study (NCS) are used to evaluate the function of A beta sensory fibres2,17, and the quantitative thermal sensory tests (QST)28,29 to study the function of thin (A delta and C) fibres. According to our previous studies, the combined results of these tests exactly corresponded to the subjective sensory alteration reported by the patients25. Of the clinical sensory tests, the TD test is the most sensitive and clinically useful test, whereas other routinely used clinical tests like brush stroke directional discrimination (BSD), warm/cold discrimination (W/C), and sharp/blunt discrimination (S/B) are rather insensitive in detecting IAN injury after BSSO25. The aim of the present prospective study was to evaluate the recovery course of IAN injury after BSSO using a combination of clinical and electrophysiologic methods tested in our previous studies. The diagnostic value of different sensory tests at different follow-up times was evaluated, and the predictive values of the tests at 2 weeks postoperatively were determined in relation to the subjective sensory recovery at 12 months after BSSO operation.
Material and methods Twenty consecutive patients (8 women, 12 men) undergoing BSSO for correction of mandibular deformity were studied. All patients had mandibular retrognathia, and their mean age was 32.8 years (range 17–51 years). The Local Ethics Committee of the Turku University Hospital approved the study protocol. The clinical and electrophysiologic tests were performed preoperatively, and 2 weeks, 1 month, 3 months, 6 months and 12 months postoperatively on both sides. Surgical technique
The BSSOs were carried out using standard methods for this osteotomy26.
There were no total IAN transsections. The bone fragments were fixed together transbuccally with three bicortical titanium positional screws. Usually, a senior surgeon operated on the right side and a trainee on the left side of the patient. Other details of the surgical procedure and anaesthesia have been described earlier15,24. Subjective sensory disturbance
Patients’ subjective symptoms and discomfort associated with the sensory alterations within the IAN distribution were asked about at postoperative controls. For analyses, the patients were divided into two groups: those indicating any alteration of sensation on the symptom chart and those with normal subjective sensation. Clinical sensory tests
All clinical tests were done in a peaceful room while the patient was sitting with her/his eyes closed. The tests were performed bilaterally at midway between the corner of the mouth and midline on the mental skin region. To reduce bias, the investigator who performed the clinical sensory testing was blinded to the results of the electrophysiologic recordings and vice versa. The details of the testing have been previously reported25. The brush-stroke directional discrimination (BSD) test was done with a soft brush, swept along a 5 mm line to the right or to the left on the test area5, using ten pairs of strokes and a two-alternative forced choice design. The touch detection (TD) test was done using a Semmes–Weinstein Aesthesiometer (Stoelting Co., USA). The testing protocol was a simplified modification of that described by Essick5 1992. The highest preoperative threshold value, 3.22, was chosen for the upper normative limit. Warm/cold discrimination (W/C) was determined by touching the test area randomly with two small glass tubes, containing 15–20C (C) and 40–45C (W) water. Diameter of the glass tube was 10 mm. The patient had to decide whether the stimulus was warm or cold. Sharp/blunt discrimination (S/B) was tested by touching the test area randomly with a sharp or a blunt head of the mechanical probe. The patient had to decide whether the stimulus was sharp or blunt. All tests were repeated ten times and the results of the BSD, W/C and S/B
135
tests were deemed normal when d80% of the answers were correct. Electrophysiologic recordings
The recording of the mental nerve blink reflex (BR) has been described in detail earlier11,13,14. The BRs were elicited by electrical stimulation with a small bipolar surface electrode (Medtronic, Skovlunde, Denmark) at the mental nerve distribution. The BR responses were recorded with surface electrodes from the eye-closing muscles, simultaneously, on both sides. The latencies were measured to the onset of the R2i and R2c components on both sides. Out of eight responses, those with the shortest latencies were selected for analyses and compared to the reference values of our laboratory11. Quantitative sensory tests (QST)
The cold detection (CDT) and warm detection (WDT) thresholds were measured within the mental nerve distribution, and the classical method of limits7 was applied. The thresholds were measured with Thermotest apparatus (Somedic Sales AB, Ho¨rby, Sweden) equipped with a small, hand-held, rectangular probe (816 mm), especially constructed for stimulating the face area. For CDT and WDT the maximal temperature range was set at 10–50C, the baseline temperature at 30C, and the rate of temperature change at 1C/s. The amount of change (in C) from the baseline temperature to the CDT and WDT was used in further analyses, and compared to the reference values of our laboratory. Due to technical equipment failure, QST was not performed on altogether 14 out of 120 controls. Nerve conduction study (NCS)
The technique for orthodromic recording of NCV of the IAN has been described in detail elsewhere12,15. The recording needle electrode was placed in the vicinity of the oval foramen, and the electrical stimuli were given at the mental foramen. The onset latencies of the responses and the amplitudes from baseline to negative peak were determined from the recordings, and nerve conduction velocity was calculated. These variables were used in the analyses and compared with the reference values of our laboratory12 or normality. Due to the refusal of subjects (postoperative
136
Teerijoki-Oksa et al.
NCS, two patients), these results were not available for all patients. Statistical methods
The frequencies of IAN damage as indicated by different tests were calculated for every follow-up point. The positive and negative predictive values of different tests at 2 weeks were determined in relation to the subjective sensory recovery at 12 months after BSSO using the fourfold table. Calculations were performed with the SAS software package (version 6.12, SAS Institute, Cary, NC, USA).
Results All patients had normal sensibility within the IAN distribution preoperatively. Figure 1a and 1b shows the frequencies of subjective sensory disturbances at 2 weeks, 1 month, 3 months, 6 months and 12 months after BSSO operation on the right and the left mental nerve distribution, as well as the frequencies of IAN injury as indicated by different electrophysiologic tests. Figure 2a and 2b shows the corresponding frequencies of IAN disturbances as indicated by different clinical sensory tests. At 2 weeks and 1 month postoperatively, most of the patients reported subjective sensory disturbance. The frequencies of IAN dysfunction varied from 12% to 100% depending on the test used. Almost all tests showed abnormal results more frequently on the left side compared to the right. The NCS indicated the highest frequencies of sensory dysfunction on both sides. Other electrophysiologic tests were less sensitive compared to the NCS. All clinical sensory tests, except for the TD test, showed lower frequencies of abnormal test results compared to the electrophysiologic tests. According to all measures used, the most pronounced recovery of the nerve damage occurred between 1 month and 3 months. During that time, the frequencies of abnormal findings in the electrophysiologic tests diminished, on average, by 13% on the right side, and by 33% on the left side. The corresponding improvement in the subjective sensory impairment was 15% on the right side, and 40% on the left side. After 3 months, the electrophysiologic tests indicated significant further improvement in several individual patients. The BR test was
Fig. 1. (a and b). Frequencies of IAN disturbances as documented subjectively and by electrophysiologic tests at different follow-up points after BSSO on the right (1a) and on the left (1b) sides. (BR=mental nerve blink reflex, CDT=cold detection threshold, WDT=warm detection threshold, NCS=nerve conduction study and subj.=subjective sensory disturbance.)
sensitive only up to 6 months, but the NCS showed further improvement up to 12 months on both sides. Similarly, the subjective sensory impairment continued to improve to some extent during the last
9 months. Except for the TD test, most clinical sensory tests were too insensitive to detect further improvement; almost all clinical tests were normal already at 3 months postoperatively.
Recovery of nerve injury after BSSO
137
cal TD test at 1 year after BSSO. Half of the patients reported normal sensation on both sides at the mental nerve distribution. However, 35% of the patients detected at least slight sensory disturbance on the right side, and 40% on the left side. The frequencies of nerve dysfunction as revealed by the NCS were equal to the figures for subjective sensory alteration on both sides. The predictive values of the clinical and electrophysiologic tests at 2 weeks, concerning the subjective nerve disturbance at 1 year, are presented in Table 1. The positive predictive values were best with the W/C (72.7%), WDT (66.7%), BSD (63.6%) and S/B (63.6%) tests. Electrophysiologic tests had higher negative predictive values compared to clinical tests, the probability of normal sensation after 1 year being 100% with a normal NCS at 2 weeks after nerve damage. Discussion
Fig. 2. (a and b). Frequencies of IAN disturbances as documented subjectively and by clinical sensory tests at different follow-up points after BSSO on the right (2a) and on the left (2b) sides. (BSD=brush-stroke directional discrimination, TD=touch detection threshold, W/C=warm/cold discrimination and S/B=sharp/blunt discrimination, subj.=subjective sensory disturbance.)
At 1 year postoperatively, the prevalence of nerve dysfunction varied from 0% to 35% on the right side, and from 0% to 41% on the left side, depending on
the test used. On both sides, the WDT test showed a slightly higher frequency of abnormality than the CDT test. The BR test was less sensitive than the clini-
All clinical and electrophysiologic tests used in the present study yielded recovery curves with a shape similar to that of the recovery of subjective sensory disturbance. Recovery of nerve damage was most marked during the first 3 months of follow-up, which is in line with the results reported in previous prospective 1-year follow-up studies on IAN recovery after BSSO1,10,27,31,32. The majority of nerve lesions in our patients appeared during the subperiosteal retraction on the medial side of the mandibular ramus24, and were thus probably mainly demyelinating lesions due to compression. Demyelinating nerve lesions generally recover during the first 4 months after injury20. Especially the NCS indicated further nerve regeneration even after 6 months. This corresponds to findings from other studies reporting improvement of sensory alteration up to 1 year after BSSO19,21,27,32. The frequency of subjective sensory disturbance at 12 months after BSSO found in the present study, 35% on the right side, and 40% on the left side, is in line with the results of other investigations when patients reporting ‘almost normal’ findings are included in the figures1,30, but also slightly lower figures have been reported (23%)6. In many studies on nerve recovery, subjective disturbances are found in a higher proportion than clinical testing methods have been able to detect3,18,32. This was true also in the present material, in which approximately one-third of the
138
Teerijoki-Oksa et al.
Table 1. Predictive values of clinical and electrophysiologic tests at two weeks, as regards the subjective sensory disturbance at one year. (ppv=positive predictive value, npv=negative predictive value, BSD=brush-stroke directional discrimination, TD=touch detection, W/C=warm/cold discrimination, S/B=sharp/blunt discrimination, BR=mental nerve blink reflex, WDT=warm discrimination threshold, CDT=cold discrimination threshold, NCS=nerve conduction study, ‘best’ values in bold) Tests
BSD
TD
W/C
S/B
BR
WDT
CDT
NCS
ppv % npv %
63.6 69.6
52.6 73.3
72.7 73.9
63.6 69.6
60.0 85.0
66.7 81.8
50.0 77.8
48.3 100
distributions with altered sensation showed normal clinical test results at 1 year. The electrophysiologic test methods used in the present study, e.g. NCS that is routinely used to diagnose peripheral nerve damage in general8,17, have only rarely been applied to the diagnosis of IAN damage after BSSO. The NCS, which in the present study appeared to be the most sensitive test method, has never earlier been used in the follow-up of IAN injuries, whereas the BR has been used in one earlier study14. The primarily high nerve injury frequencies recorded by the NCS and the BR tests indicated that thick myelinated A beta fibres were most often damaged during BSSO. This is in line with the fact that, as discussed earlier, the nerve lesions were probably more often of demyelinating type, as was also indicated by the quick recovery. The QST test has been used to study nerve damage after BSSO in a few earlier studies21,27,29. In prospective follow-up studies, QST has not earlier been applied using a small thermode suitable for evaluation of narrow distributions of the trigeminal nerve. A small thermode has been found to be more sensitive in detecting C-fibre dysfunction than a large one9, such as the 6 cm2 thermode used in the study by B J27. The relative insensitivity of a large thermode is due to the more effective central spatial summation with a larger probe. In addition, as the mental nerve distribution is small, a large thermode may inadvertently stimulate the normal adjacent nerve distributions, and thus lead to false negative results. The differences in the frequencies of nerve damage revealed by various electrophysiologic tests may be explained by their varying inability to assess all sensory nerve fibre populations. The BR and NCS tests are especially sensitive in detecting demyelinating injuries mainly affecting the large A beta afferents, whereas thermal QST is more sensitive in the detection of axonal damage to thin fibres28.
Electrophysiologic tests were generally more sensitive than clinical sensory tests in detecting nerve dysfunction in the present study. In addition, the nerve dysfunction frequencies revealed by these tests were closer to the reported subjective disturbances than those of the clinical tests—the only exception being the TD test. In our earlier study, too, the electrophysiologic test results correlated statistically significantly with the nerve injury at the end of the BSSO operation, whereas the clinical sensory tests, except again the TD test, did not25. The BSD, W/C and S/B are not sensitive enough to detect slight nerve disturbances25: these test methods revealed the severe nerve injuries when applied at the early stages of this follow-up study, but they were not able to detect the subjective sensory disturbances after 3 months. Similar results have been reported also by other authors1,27. The follow-up at different time points indicated that the BR test normalized earlier than the NCS. The BR test was useful up to 6 months, but after that it showed improvement only in some individual cases. Due to a central amplification factor of this polysynaptic reflex, a few intact A beta fibres in a damaged nerve trunk can elicit a normal blink reflex. The NCS gives most accurate information about the number of functioning axons and the conductive properties of the A beta fibres8,17. After substantial axonal nerve injury, the nerve conduction velocity can slow down permanently due to the Schwann cell regeneration having shorter internodal intervals than before injury, and it may never reach baseline22, as happened in many cases in our study. Thin fibre dysfunction as documented by the CDT and the WDT tests was less frequent than A beta afferent dysfunction verified by the NCS. One year after BSSO, the WDTs were more often abnormal than the CDTs on both sides, and the persistence of C fibre dysfunction was notable. The WDT test evaluates unmyelinated C fibres, and thus an
abnormal test result indicates axonal lesion with incomplete recovery as it has been shown that warm detection is especially dependent on C fibre innervation density4. This may be the reason for the slow normalization of the WDT test result. Our study indicated that clinical sensory tests measuring the function of thin nerve fibres (W/C and S/B tests) were clearly less sensitive at all follow-up points compared to the quantitative WDT and the CDT tests. The TD test showed similar frequencies of nerve dysfunction as the BR test during the follow up, and at 1 year, it was even slightly more sensitive than the BR in documenting the subjective sensory alteration. Semmes–Weinstein monofilaments have been used in several earlier follow-up studies on IAN recovery3,6,16,23, but the testing algorithms, and the time points used have varied, making comparison with the present results difficult. The TD test is a standardized, repeatable chair-side test method and, according to our previous findings, the only clinical sensory test to have good correlation with the intraoperatively verified IAN injury25. Although insensitive in late follow-up after 3 months, the clinical BSD, W/C and S/B tests, at early control time points, had the best positive predictive values as regards the sensory outcome at 1 year; an abnormality in these tests at 2 weeks best predicted the existence of subjective sensory disturbance 1 year postoperatively. Although the positive predictive values of these tests were only moderate, our results indicate that they have a role to play in early diagnostic evaluation, as an abnormal result in these tests is compatible with a more severe nerve injury and longer-lasting subjective sensory alteration. Z et al.33 also arrived at a similar conclusion on the usefulness of clinical sensory tests at predicting a nerve injury but, in line with the present findings, they found them to be unreliable in ‘ruling out’ a lesion, i.e. they show a high incidence of false-negative findings. All tests had higher negative predictive values than positive predictive values, and the electrophysiologic tests had higher values compared to the clinical tests. In particularly, normal NCS test results predicted normal subjective sensibility at 1 year correctly in all cases. In the only previous prospective study on the diagnostic value of electrophysiologic tests14, the positive (64%) and negative (87%) predictive values of the BR test were in good accordance
Recovery of nerve injury after BSSO with the present figures for the same test at the same time point (60% and 85%, respectively). On the basis of these results, the following diagnostic procedure in the follow-up of BSSO patients’ nerve dysfunction can be recommended. At early controls within 3 months after the operation, clinical sensory testing should include the diagnostically most sensitive clinical test, the TD test (when possible, with preoperative baseline measurement) and the W/C discrimination test that evaluates the thin fibre function, and has the best positive predictive value. During the early postoperative period, NCS and thermal QST can best verify the subjective sensory alteration, and most accurately assess the degree of thick and thin fibre dysfunction, and thus are highly recommended when available. At late control points, the clinical sensory tests do not offer sensitive enough tools for the detection of residual nerve dysfunction or for the follow-up of nerve recovery, with the possible exception of the TD test. The NCS is sensitive to nerve dysfunction even at 1 year after BSSO, as is the QST, especially the WDT test is a good tool for late nerve injury diagnosis within the IAN distribution. Thus, these tests should always be included in the evaluation of patients with late sensory complaints. Acknowledgments. The study was supported by grants from Turku University Hospital and the Finnish Dental Society.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
References 1. B J, A P, I S. Sensibility following sagittal split osteotomy in the mandible. A prospective clinical study. Plast Reconstr Surg 1998: 102: 325–333. 2. C G, A R, I M, I P, R A, M M. Mandibular nerve involvement in diabetic polyneuropathy and chronic inflammatory demyelinating polyneuropathy. Muscle Nerve 1998: 21: 1673–1679. 3. C LL, T BD, C GM, B RA, K SD, R JD. A comparison of questionnaire versus monofilament assessment of neurosensory deficit. J Oral Maxillofac Surg 1996: 54: 454–459. 4. D PJ, Z I, G DA, J D, K JL, OB PC. Cool, warm, and heat-pain detection thresholds: Testing methods and inferences about anatomic distribution of
15.
16.
17.
18.
receptors. Neurology 1993: 43: 1500– 1508. E GK. Comprehensive clinical evaluation of perioral sensory function. Oral Maxillofac Surg Clin N Am 1992: 4: 503–526. F KL, H TJ, P KJ, B MJ. Neurosensory recovery following the mandibular bilateral sagittal split osteotomy. J Oral Maxillofac Surg 1995: 53: 1300–1306. F H, L U, S WG. Method for quantitative estimation of thermal thresholds in patients. J Neurol Neurosurg Psychiatry 1976: 39: 1071–1075. H CM J. Cranial nerve monitoring. Syllabus for course F, annual meeting of the American Association of Electrodiagnostic Medicine, Orlando, FL, October 1998: 23–31. H MJ, S B, A FB, K EH, N B. Quantitative thermal perception testing in adults. J Clin Neurophysiol 1999: 16: 462–471. J DL, W LM. Intraoperative recording of trigeminal evoked potentials during orthognathic surgery. Int J Adult Orthod Orthognath Surg 1990: 5: 167– 174. J¨ ¨ ¨ SK. Blink reflex with stimulation of the mental nerve. Methodology, reference values, and some clinical vignettes. Acta Neurol Scand 1995: 91: 477–482. J¨ ¨ ¨ SK. A new technique for recording sensory conduction velocity of the inferior alveolar nerve. Muscle Nerve 1999: 22: 455–459. J¨ ¨ ¨ SK, P JK. Clinical application of the blink reflex with stimulation of the mental nerve in lesion of the inferior alveolar nerve. Neurology 1994: 44: 2356–2361. J¨ ¨ ¨ SK, P JK, L R. The mental blink reflex in the diagnosis of lesions of the inferior alveolar nerve following orthognathic surgery of the mandible. Br J Oral Maxillofac Surg 1996: 34: 87–95. J¨ ¨ ¨ SK, T-O T, F K, V¨ ¨ K, P JK, F H. Intraoperative monitoring of the inferior alveolar nerve during mandibular sagittal-split osteotomy. Muscle Nerve 2000: 23: 368–375. K ND, B SB, S DP. Recovery of neurosensory function following orthognathic surgery. J Oral Maxillofac Surg 1990: 48: 124–134. K J. Anatomy and physiology of the peripheral nerve. In: Electrodiagnosis in diseases of the nerve and muscle: principles and practice. New York: Oxford University Press 2001: 63–90. L JI, G-M OT. Sensory impairment following sagittal split osteotomy for correction of mandibular retrognathism. Int J Adult Orthod Orthognath Surg 1991: 6: 161–167.
139
19. N K, U K, M N, T S, Y E, O H. The assessment of trigeminal sensory nerve paraesthesia after bilateral sagittal split osteotomy: modified somatosensory evoked potentials recording method. J Craniomaxillofac Surg 1997: 25: 97– 101. 20. R LR. Traumatic injury to peripheral nerves. Muscle Nerve 2000: 23: 863–873. 21. S-M S, K H, O R, N FW. A prospective electromyographic and computer- aided thermal sensitivity assessment of nerve lesions after sagittal split osteotomy and Le Fort I osteotomy. J Oral Maxillofac Surg 2001: 59: 128–139. doi: 10.1053/ joms.2001.20480. 22. S ME. Regeneration of peripheral nerve. In: Sumner AJ, ed.: The physiology of peripheral nerve disease. Philadelphia: WB, Saunders Company 1980: 358–431. 23. T Y, N H, M H, F M, A JL, T H, U E, T T, Y GY. A clinical analysis of the recovery from sensory disturbance after sagittal splitting ramus osteotomy using a SemmesWeinstein pressure aesthesiometer. Bull Tokyo Dent Coll 1998: 39: 189–197. 24. T-O T, J¨ ¨ ¨ S, F K, F, H, V¨ ¨ K, T T, V A. Risk factors of nerve injury during mandibular sagittal split osteotomy. Int J Oral Maxillofac Surg 2002: 31: 33–39. doi:10.1054/ ijom.2001.0157. 25. T-O T, J¨ ¨ ¨ S, F K, V A, F H. An evaluation of clinical and electrophysiologic tests in nerve injury diagnosis after mandibular sagittal split ostetomy. Int J Oral Maxillofac Surg 2003: 32: 15–23. doi:10.1054/ijom.2002.0325. 26. T BC, W RP J. Mandibular ramus surgery. In: Proffit WR, White RP, Jr, eds: Surgical-Orthodontic Treatment. St. Louis: Mosby 1991: 264– 282. 27. V B R, J KA. Psychophysical study of the mechanisms of sensory recovery following nerve injury in humans. Brain 1994: 117: 149–167. 28. V R, O JL. Quantitative somatosensory thermotest. A key method for functional evaluation of small caliber afferent channels. Brain 1992: 115: 893– 913. 29. W A. On inferior alveolar nerve function after sagittal split osteotomy of the mandible. Thesis. Stockholm: 1999: 11. 30. Y L, K J, L P, O K. Prediction of recovery from neurosensory deficit after bilateral sagittal split osteotomy. Oral Surg Oral Med Oral Path Oral Radiol Endod 2000: 90: 275–281.
140
Teerijoki-Oksa et al.
31. Y L, K J, O K. Comparison of different tests assessing neurosensory disturbances after bilateral sagittal split osteotomy. Int J Oral Maxillofac Surg 1998: 27: 417–421. 32. Y T, N T, K T, M N, N T, S H, H K. Impairment of the inferior
alveolar nerve after sagittal split osteotomy. J Craniomaxillofac Surg 1989: 17: 271–278. 33. Z JR, M RA, G JM, M M, D LF. The accuracy of clinical neurosensory testing for nerve injury diagnosis. J Oral Maxillofac Surg 1998: 56: 2–8.
Address: Tuija Teerijoki-Oksa Savonlinna Central Hospital Keskussairaalantie 6 57120 Savonlinna Finland Tel.: +358-15-5817082 E-mail:
[email protected]