Publications
APS Bulletin • Volume 12, Number 3, May/June 2002
Pain and Public Policy
Corey D. Fox, PhD, Department Editor
Predicting Prognosis of Abnormal Sensation After Trigeminal Nerve Injury
Yoshiki Imamura, DDS PhD; Shunji Shiiba, DDS PhD; Eiji Sakamoto, DDS; Shin-ichi Masumi, DDS PhD; Osamu Nakanishi DDS PhD; Gary J Bennett, PhD
Introduction
Patients occasionally complain of abnormal sensation such as pain, allodynia, and dysesthesia after oral and maxillofacial surgery and other dental procedures. In some cases, these abnormal sensations persist for a long time (Carmichael & McGowan, 1992; Delcanho, 1995). It is probable that these neurological complications are due to nerve damage. Some attempts have been reported to evaluate the severity of nerve damage, and recovery from it using quantitative sensory testing (QST) (Fredousi & MacGregor, 1985; Campbell, Shamaskin, & Harkins, 1987). However, good recovery in QST results does not always necessarily mean resolution of the patient’s complaints.
Evaluation of damaged nerves and prognosis of associated neurological symptoms should be standardized to properly evaluate the efficacy of treatment procedures. This study was conducted to develop methods to evaluate the severity of nerve damage and to see whether such an evaluation aided in the prediction of the duration of abnormal sensations after trigeminal nerve injury using self-reported symptoms and QST.
Methods
Self-reported symptoms (spontaneous pain, hypoesthesia, hyperalgesia, allodynia, and dysesthesia): Forty-nine cases where there was nerve injury as a consequence of dental treatment or oral and maxillofacial surgery were followed for 1 year. Patients who visited our clinic within 2 weeks after the injury were enrolled in this study. Symptoms were evaluated at the first visit and 1 year later. Spontaneous pain was defined as pain at rest without any kind of stimulation or physical movement. Hypoesthesia was defined as depressed sensation in the affected area versus a control area. Hyperalgesia was assessed to be present if the patient felt abnormally severe pain when he or she put an ice cube or hot water in the mouth. Allodynia was assessed as present if a painful sensation in the affected area occurred while brushing the teeth, shaving, applying make-up, or eating. Dysesthesia was assessed as present when a patient reported unpleasant, but not clearly painful, sensations in the situations described previously.
QST was applied in a second group to evaluate the severity of nerve damage more precisely. This group consisted of 37 patients who visited us within 2 weeks after injury. Each patient was evaluated at 2 weeks and 12 months after the injury. QST included touch perception threshold (TPT) using a series of von Frey hairs (Semmes-Weinstein aesthesiometer set, Stoelting, Chicago) and electric detection threshold (EDT) using a bipolar platinum hemispheric electrode (Trigeminal Electrode, MT Giken, Tokyo) and an electric stimulator (Sapphier, Medelec, London). Recovery of total sensation was evaluated by the patient 12 months after the injury.
All patients received methyl cobalamine prescription. Tricyclic antidepressants and anticonvulsants were administered to some patients who complained of dysesthesia, pain, hyperalgesia, and/or allodynia.
Statistical analyses
The Kruskal-Wallis statistic was used to evaluate self-reported recovery of total sensation groups. Wilcoxon’s signed-rank test was used for assessment of changes in TPT. Changes in EDT were assessed with the paired t-test.
Results
Spontaneous pain was observed in nine cases within 2 weeks after initial visit. Seven of the nine recovered completely within 1 year (Figure 1). Thirty-two patients complained of dysesthesia. Patients with allodynia and hyperalgesia within 2 weeks after initial visit showed significantly less recovery than did patients without these types of abnormal sensation. Fourteen cases that had had hypoesthesia with dysesthesia but not allodynia and/or hyperalgesia during the early post-injury period showed a good recovery, but 15 did not (Table 1). Dysesthesia was observed in all patients who had spontaneous pain after 1-year follow-up.
|
|
Deficits in the detection thresholds to the tactile and electrical stimuli recovered significantly 12 months after the injury (Figures 2 and 3). However, there was no relationship between the thresholds at this point and self-reported recovery of total sensation (Figure 4).
|
|
|
Severe deficits in the detection thresholds to tactile and electrical stimuli 2 weeks after the injury, almost certainly the result of relatively severe nerve injuries, were strongly associated with a long-time course of abnormal sensation.
Discussion
Nerve injury is not an uncommon complication of oral and maxillofacial surgery and other dental procedures (Carmichael & McGowan, 1992; Delcanho, 1995). However, diagnostic procedures to evaluate it and treatment strategies to remedy it have not been established. We investigated whether QST procedures (EDT and TPT) used shortly after the injury would be useful in evaluating damaged nerves and predicting the recovery of sensation and resolution of the patient’s complaints.
Our results revealed no relationship between QST results and the patient’s satisfaction with recovery 1 year after the injury. We think it is likely that one reason for absence of any relationship is that our tests were not capturing the essence of the symptoms. We often saw patients hesitate when rating their sensation with a numeric score or with a VAS rating when the affected region was stimulated 1 year after the injury (data are not shown). It is likely that patients experienced hypoesthesia, dysesthesia, or allodynia; but the QST result is only the threshold for a sensation, not its quality. Thus, QST data may reveal recovery from hypoesthesia without yielding any information about the quality of the restored sensation. A patient’s satisfaction with recovery probably had little to do with recovery from hypoesthesia per se, but rather on whether dysesthesia, allodynia, and/or hyperalgesia were present. It thus appears likely that we will need a multidimensional evaluation for the assessment of nerve injury and its consequences, where both thresholds and qualitative phenomena are investigated.
Some self-reported symptoms and QST data showed a close relationship with abnormal sensations of long duration. Patients who just complained of hypoesthesia but no other kinds of abnormal sensation early after the injury had a better prognosis than those who had dysesthesia, allodynia, and hyperalgesia soon after injury.
It is possible that hypoesthesia after mild nerve injury is due to neurapraxia rather than degeneration (Seddon, 1954), and the hypoesthesia is thus transient. In more severe injuries, the hypoesthesia may be due to axonal degeneration, and phenomena associated with degeneration may be necessary to produce dysesthesia, allodynia, and hyperalgesia (Sommer & Myers, 1995; Ramer, French, & Bisby, 1997). In animals, hyperalgesia resolves when the injured nerve regenerates (Myers, Heckman, & Rodriguez, 1996; Lindenlaub & Sommer, 2000). Dysesthesia, allodynia, and hyperalgesia may persist if regeneration is impaired or incomplete.
The initial time of QST examination in this study was 2 weeks after the injury. At this time, degeneration is largely complete and regeneration has just begun. QST data from this early period assesses the severity of sensory loss (i.e., hypoesthesia). Measuring the severity of hypoesthesia during this period may allow us to estimate the magnitude of degeneration, which is likely to be correlated to the risk of incomplete regeneration and thus the risk of persistent abnormal sensation. However, unidimensional QST data is of little help in evaluating the abnormal sensations seen long after injury.
Conclusions
We can roughly predict the appearance of persistent abnormal sensation using self-reported symptoms and QST data obtained within 2 weeks of nerve injury. Patients with hypoesthesia but no other symptoms who showed mild deficits in detection thresholds with QST procedures tended to recover satisfactorily. These results suggest that persistent abnormal sensation is associated with severity of nerve injury.
References
Campbell, R.L., Shamaskin, R.G., & Harkins, S.W. (1987). Assessment of recovery from injury to inferior alveolar and mental nerves. Oral Surgery, Oral Medicine, and Oral Pathology, 64, 519–526.
Carmichael, F.A., & McGowan, D.A. (1992). Incidence of nerve damage following third molar removal: A West of Scotland Oral Surgery Research Group study. The British Journal of Oral & Maxillofacial Surgery, 30, 78–82.
Delcanho, R.E. (1995). Neuropathic implications of prosthodontic treatment. The Journal of Prosthetic Dentistry, 73, 146–152.
Fredousi, A.M., & MacGregor, A.J. (1985). The response of the peripheral branches of the trigeminal nerve to trauma. International Journal of Oral Surgery, 14, 41–46.
Lindenlaub, T., & Sommer, C. (2000). Partial sciatic nerve transection as a model of neuropathic pain: A qualitative and quantitative neuropathological study. Pain, 89, 97–106.
Myers, R.R., Heckman, H.M., & Rodriguez, M. (1996). Reduced hyperalgesia in nerve-injured WLD mice: Relationship to nerve fiber phagocytosis, axonal degeneration, and regeneration in normal mice. Experimental Neurology, 141, 94–101.
Ramer, M.S., French, G.D., & Bisby, M.A. (1997). Wallerian degeneration is required for both neuropathic pain and sympathetic sprouting into the DRG. Pain, 72, 71–78.
Seddon, H.J. (1954). Peripheral nerve injuries. Nerve injury committee. Medical Research Council. Council Special Report, 282, H. M. Stationery Office, London.
Sommer, C., & Myers, R.R. (1995). Neurotransmitters in the spinal cord dorsal horn in a model of painful neuropathy and in nerve crush. Acta Neuropathologica, 90, 478–485.