ANATOMY AND PATHOLOGY
Compression, distortion, or stretching of the trigeminal roots by arteries, veins, vascular malformation, skull base bony anomalies, or slowly growing tumors can cause the pain of TN (
21). Compression of the nerve root by a blood vessel is often the cause in the majority of cases (
10,
30,
38,
74). Jannetta initially reported an offending blood vessel in 100% of his TN cases (including those found at reoperation). Subsequently, many series cited a high percentage of vascular compression as well, ranging from 79 to 97 (
10,
18,
30,
74,
77). By contrast, a tumor is found in 2 to 3% of cases (
13,
18,
74) and a small arteriovenous malformation in 0.5 to 2% (
18,
74).
Tumors, usually posterior fossa meningiomas or neuromas giving rise to symptoms indistinguishable from typical TN rather than numbness and atypical pain, are usually seen to distend the root, rather than invade it (
13). Direct infiltration of the nerve or ganglion tends to give rise to sensory loss and nonparoxysmal pain.
Approximately 2 to 4% of patients with TN have multiple sclerosis (MS), although it is rarely the first manifestation of the disease (
32,
69). MS should be considered in patients younger than 50, especially with history of bilateral TN. Very rarely, TN may accompany syringobulbia or develop after brainstem infarction (
81). In all, symptomatic TN is likely to explain approximately 5 to 7% of cases in unselected populations; the rest are due to vascular compression of the root or the cause remains unknown.
Pathophysiology
There is general acceptance that because no or only minimal sensory loss is encountered in TN, most of the trigeminal pathways must remain anatomically intact. On the other hand, the pain in its most stereotyped presentation (including remissions) is likely to reflect relatively limited neuronal dysfunction either in the trigeminal nerve or its central connections. Two points are important to note. First, there is compelling evidence from large case
series that simple decompression at the dorsal root entry zone renders most patients pain-free for years—in effect, cures them. Second, typical (or classical) features of TN are exceedingly rare in conditions that affect the trigeminal pathways outside the dorsal root entry zone. Even in MS-related TN, there is a plaque near the dorsal root entry zone, effectively disrupting the functions of the central terminals (
24).
The myelin at the dorsal root entry zone is derived from astrocytes and is considered less tolerant to compression than that produced in the peripheral nerve by Schwann cells. Neuropathologic studies of operative specimens from the dorsal root entry zone in patients with vascular compression and in patients with MS and no compression show significant demyelination and axons in direct opposition without intervening glial processes (
16,
46,
47,
66). This is thought to favor ephaptic transmission, that is, the transfer of nerve impulses from one set of fibers to another (
15,
46,
47,
59). Should this happen between fibers mediating tactile impulses and those mediating painful impulses from trigger zones, the result expectedly would be what is seen clinically, touch-evoked pain. From experimental nerve models it is hypothesized that affected axons acquire a state of hyperexcitability, which renders them capable of producing intense firing for seconds or minutes when stimulated (
15,
16,
46). Compression by a blood vessel in this regard is likely to be significant; intraoperative electrophysiologic studies showed delayed transmission in the affected trigeminal root, which normalized immediately after decompression (
43). It has also been suggested that secondary changes at the level of the ganglion cells (whether due to hitherto unknown mediators or secondary degenerative processes in ganglion cell somata) make these cells susceptible to crossexcitation (
15,
59,
65). If this were a synchronized event, it could lead to the simultaneous activation of several cells, producing the explosive pain pattern that TN represents (
15).
Although far from proven, this hypothesis explains why many treatments are effective in TN. Decompression removes the cause, neuroablative procedures interfere with the cell-to-cell cross-talk and hyperexcitability, and medication effectively prevents excessive firing. Many aspects of TN remain unknown, however, including reasons for spontaneous remissions, why compression in only a minority leads to TN, and the mechanism of TN in the small minority of patients without structural abnormality.
CLINICAL FEATURES
TN remains a clinical diagnosis, based on the patient’s description of pain. Crucial features include severe stabbing, or electric-shock-like pains, usually unilateral and located in the distribution of the trigeminal nerve. IHS diagnostic criteria (Revised International Classification of Headache Disorders [ICHD-II]) for TN are as follows:
A. Paroxysmal attacks of pain lasting from a fraction of a second to 2 minutes, affecting one or more divisions in the trigeminal nerve and fulfilling criteria B and C.
B. Pain has at least one of the following characteristics:
1. Intense, sharp, superficial, or stabbing
2. Precipitated from trigger areas or by trigger factors
C. Attacks are stereotyped in the individual patient.
D. There is no clinically evident neurologic deficit.
E. Not attributed to another condition
The IHS criteria of “symptomatic” TN are identical to those of “classical” TN, except for the demonstration of a nonvascular lesion (MS, tumor, etc.) in classical TN and the allowance of sensory impairment. Vascular compression, however, is considered part of the “classical” category.
Sometimes, TN is categorized either as “typical” and “atypical” irrespective of its cause, while pain arising from the trigeminal nerve but without all characteristics of TN and frequently demonstrating features common to any neuropathic pain is referred to as
painful trigeminal neuropathy. This practice has developed from observations that the treatment results are better in typical versus atypical cases, irrespective of the surgical method chosen, while most are contraindicated in trigeminal neuropathy (
7,
43,
84,
89).
Generally, the key to the right diagnosis is careful history taking that includes both the patient’s spontaneous description of the pain and a semistructured practitioner’s interview, probing the specific pain characteristics. Not all patients volunteer that their pain is “electric-shock-like,” “lightninglike,” lancinating, or stabbing, and many will insist that it is exceedingly variable and lasts well beyond the 2 minutes allowed by the IHS criterion. To the patient, the experience of the intense paroxysm may be more relevant than differentiating the duration of a single stab. A very poorly controlled trigeminal pain may be described as continuous by the patient attempting to highlight the relentlessness of the rapidly recurring paroxysms.
Paroxysm trigger factors are mechanical by nature, although rarely, extreme cold, smell, or taste has been implicated (
73). Sometimes, patients develop methods of avoiding pain triggers, including avoiding eating and drinking, which leads to dehydration in severe cases.
The pain of TN is almost always unilateral. Even in rare bilateral cases, the two sides react independently to various stimuli and the attacks come unsynchronized. Paroxysms are frequently followed by refractory periods, up to minutes (
40), during which time the previous stimulus, if repeated, fails to provoke an attack.
TN has a tendency for weeks- or even months-long remissions, especially in the early phase of the condition. Not all TN starts with the classical features.
Pretrigeminal neuralgia, an intermittent unilateral trigeminal pain lacking some of the characteristics of TN pain, can occur (
22). The pain, however, responds to carbamazepine and later develops into a definite form of TN. Conversely, it has been suggested that with time, many patients with a classical presentation develop other, less paroxysmal pain, in effect making “typical” cases “atypical,” with poorer treatment outcomes (
11). If true, this will affect timing of treatment, which is discussed below.
Several other facial pain conditions bear similarities to TN and are diagnosed on the basis of history. This underlines the importance of obtaining a detailed pain description of the quality of pain (
Table 127-1). There are several caveats that one should be aware of. About 10% of patients will not respond to carbamazepine, and up to 40% do not demonstrate trigger areas on clinical examination (
72). Occasionally, patients describe redness and swelling of the face after a severe attack (
60), and when the first division is involved, a distinction from SUNCT syndrome (short-lasting, unilateral, neuralgiform headache with conjunctival injection and tearing) can be difficult (
27). Although several groups have shown sensory changes in TN (
53,
58), most unoperated patients show only minimal changes with conventional bedside testing. Indeed, substantial sensory loss should raise the possibility of a symptomatic TN, and investigations should be directed accordingly.
INVESTIGATIONS
Advanced imaging techniques form the backbone of etiologic assessment and also provide helpful information for the neurosurgeon in case operative treatment is being considered. Whenever possible, each patient with a new diagnosis of TN should undergo magnetic resonance imaging (MRI), preferably using 3-D reconstruction techniques to (a) rule out other causes than vascular compression (MS, tumor) and (b) assess the relationship of the nerve to the neighboring blood vessels. Whole-brain T2-weighted images are the minimum with suspected MS.
Several groups have shown that the identification of a treatable cause—compression of the nerve at the root entry zone by an overlying blood vessel—is possible with considerable reliability. In a pioneering study, Meaney and coworkers (1995) used 3-D fast imaging with steady-state precession (FISP) reconstruction images to demonstrate blood vessels as high signal intensity structures around the nerve in any orientation (
52). While arteries were easily identifiable, veins could be properly visualized only after enhancement with intravenous gadolinium. The method was validated in 50 patients with 55 symptomatic nerves (5 patients had bilateral TN) who underwent posterior fossa exploration. In the 52 explorations carried out, neurovascular contact was confirmed in all of the 49 cases suggested by 3-D FISP. A further case with negative preoperative imaging was shown to lack any vascular contact at operation. There were two false-negative scans, and no false positives, leading to a sensitivity of 100% and a specificity of 96%. A similar study using identical imaging parameters confirmed these results (
61).
Akimoto et al. combined 3-D FISP with 3-D constructive interference in steady state (CISS) to improve the visualization of the trigeminal nerve (
1). This combined imaging technique improves the visualization of structures in the cerebrospinal fluid space due to its use of heavily T2-weighted source images. In 24 consecutive patients who underwent microvascular decompression (MVD), there was excellent agreement between the preoperative imaging and operative diagnosis in all but one, and a compression was found by an undiagnosed vessel in this patient (
1). Even if this is only an observational study without radiologic blinding, the results are impressive. Routine MRI techniques, although less time consuming, do not produce images accurate enough to determine the relationship of the nerve and blood vessels in its vicinity.
MRI provides the clinician with the anatomic substrate of clinically diagnosed TN but does not enable one to diagnose TN because vascular contacts are reported in 8% of asymptomatic nerves (
52). In other words, the functional significance of a contact between a vessel and the trigeminal nerve is unclear in patients who do not have TN clinically. Also, false-negative findings remain a possibility, as small arteries (diameter <1 mm), arachnoidal thickenings, and similar less common causes are not detected using MRI (
74). Despite these shortcomings, 3-D reconstruction MRI in trigeminal neuralgia, when used in conjunction with critical assessment of the quality of the pain, provides a rare opportunity for a far more precise diagnosis than is the norm in painful conditions involving the head and face.
Other methods, such as quantitative sensory testing and laser-evoked potentials, provide limited diagnostic benefit, mainly in quantifying afferent dysfunction, and do not have a therapeutic role, either in decision making or follow-up.