Abstract
Objective measurements of duration of attacks have been performed in 8 (5 female and 3 male) patients suffering from primary first division (V-1) trigeminal neuralgia. The mean age of the patients was 67.5 ± 11.4 years, and the mean age at onset 64.0 ± 9.7 years. During the study the patients were off treatment. A total of 192 attacks were witnessed by the authors and exactly timed by a stop-watch. The duration of attacks ranged from 2 to 32 s, with a mean of 6.5 ± 6.1 s. The unweighted mean was 8.8 ± 5.7 s, with a range of 2.4-17.5 s. With the present data the duration of attacks of V-1 neuralgia has been exactly determined, and the clinical distinction of V-1 neuralgia from other shortlasting headaches, particularly from SUNCT, has been substantially clarified.
Introduction
From early observations (1, 2) to recently built diagnostic criteria (3), trigeminal neuralgia attacks have been regarded as lasting from a fraction of a second to a few minutes. For about three centuries such an estimation of duration of pain attacks – along with other symptoms and signs – seem to have sufficed for diagnosis. However, during the last years the clinical landscape has substantially changed, as several unilateral, shortlasting headaches with some similar features have been described. In fact, current descriptions with such a wide range of duration of trigeminal neuralgia attacks may pose diagnostic difficulties when comparing first division trigeminal neuralgia (V-1 neuralgia) with other shortlasting headaches in the same localization (3–7), namely, primary stabbing headache (PSH), shortlasting, unilateral, neuralgiform, headache attacks, with conjunctival injection and tearing (SUNCT), chronic paroxysmal hemicrania (CPH), and cluster headache.
SUNCT syndrome in particular is the main differential diagnosis of V-1 neuralgia (4, 6, 7). Duration of attacks and presence/absence of oculofacial autonomic accompaniments have been pivotal in distinguishing both disorders (8, 9). Nevertheless, contrary to what is the case in SUNCT (9), objective estimation of duration of attacks is lacking in V-1 neuralgia.
Since V-1 neuralgia is rare, objective assessment of the duration of attacks has been a difficult task. Fortunately, we have had the chance of collecting a considerable number of exactly timed V-1 neuralgia attacks that may add important clinical knowledge about trigeminal neuralgia, and also provide some important clues for the differential diagnosis of shortlasting unilateral headaches.
Subjects and methods
Through a 9-year period we have been collecting attacks of V-1 neuralgia. Eight patients (5 women and 3 men; mean age: 67.5 ± 11.4 years) with pure V-1 neuralgia (i.e. not extended to V-2 or V-3 territories) took part in this study. Patients no. 3 and 4 have been published elsewhere (8). The diagnosis was based on International Headache Society diagnostic criteria (10). Neurological and ophthalmological examinations were normal in all cases. Routine blood analysis including ESR and brain/orbital MRI did not show any abnormalities.
Six patients who had been successfully treated with carbamazepine (range of daily dose: 400–1200 mg) agreed to discontinue treatment before being studied in order to get full blown attacks. Treatment was tapered-off through 2 weeks and then a ‘washout’ time period of at least 10 days without treatment had to be accomplished. The patients came to the hospital when they noticed reappearance of attacks but not before the washout period had been completed. Once the measurements had been completed carbamazepine treatment was resumed. Two de novo patients (No. 4 and 6) were studied immediately after the diagnosis, so that treatment with carbamazepine was started after the timing of attacks. Also in these two patients carbamazepine proved to be effective.
For ethical reasons none of the patients was allowed to suffer attacks for more than 24 h. Studied attacks were precipitated by the usual precipitating mechanisms, thus reproducing patients’ natural situations during symptomatic periods. The attacks were timed with a stop-watch. For the sake of exactness, all the attacks were witnessed and timed by at least one of the authors. The patients were asked whether the objectively timed attacks were on average as the usual ones (before therapy was started) in terms of duration, topography, character and intensity of pain.
Patients gave written consent for the entire procedure, including withdrawal of treatment, delayed therapy onset, and precipitation of attacks. The ethical committee of Fundación Hospital Alcorcón approved the procedure.
The distribution of all time measurements was analysed making use of the SPSS statistical package (version 11.5). Finally, the duration of trigeminal neuralgia attacks in our series was compared with the duration of other shortlasting unilateral headaches that had already been objectively assessed.
Results
Attack duration was determined in a total of 192 attacks of V-1 neuralgia from 5 women and 3 men (Table 1), 49–79 years old (mean age: 67.5 ± 11.4 years). Age at onset ranged from 48 to 75 years (mean: 64.0 ± 9.7 years). The number of attacks per patient showed a great variability (3–51), owing to both individual differences in the flow of attacks and the ethical necessity of resuming/starting treatment within 24 h. The mean weighted duration of attacks was 6.5 ± 6.1 s, with a range of 2–32 s (Table 1). The mode was 2 s, and the median value was 4 s. The unweighted mean was 8.8 ± 5.7 s, with a range of 2.4–17.5 s (Table 1).
Duration of V-1 neuralgia attacks
Seven patients considered that timed attacks were similar to regular ones (before treatment), both as regards to duration, and intensity/character of pain. Patient no. 4 estimated that attacks experienced during the procedure were longer than the usual ones.
When compared to other shortlasting unilateral headaches in orbital or periorbital localization, V-1 neuralgia attacks had longer duration than PSH paroxysms, but shorter than SUNCT, CPH or cluster headache (Table 2). The figure shows the distribution of attacks in the present series of V-1 neuralgia (n = 192), together with the distribution of attacks in a previously reported series of SUNCT (n = 348) that was studied with similar methods (9). Attack duration in both disorders was significantly different (Student’s t-test, P < 0.001). The range of duration for V-1 neuralgia attacks was 2–32 s, and the range for SUNCT was 5–250 s. Most V-1 neuralgia events (82.3%) lasted up to 10 s, while only 2.6% of SUNCT events were so brief. Conversely, there was just a single V-1 neuralgia attack (0.5%) that lasted more than 30 s, whereas the duration of most SUNCT attacks (66.4%) exceeded that point (Fig. 1).
Duration of attacks in V-1 neuralgia (░) and SUNCT (□). Graph showing the duration of attacks in the present series of V-1 neuralgia (n = 192) and a former series of SUNCT (n = 348) by Pareja et al. (9). For each time interval the number of attacks is expressed as a percentage.
Objective duration of primary unilateral, shortlasting, headaches with orbital/periorbital localization
Objective measurements of PSH have not been performed. However, with attacks usually lasting c. 1 second it can be assumed that patients’ own estimations should be similar to objective timing.
Median of daytime cluster attacks: 37 min; median of nocturnal cluster attacks: 39 min.
Discussion
The duration of trigeminal neuralgia attacks shows a great variability among different descriptions that can be found in the literature (11–16), but this is the first time that objective measurements have been performed. A comparison of our objective data with previous reports indicates that the range of duration of trigeminal neuralgia attacks has generally been overestimated. There are several possible reasons for such overestimates. On the one hand, patients may have overestimated the length of solitary attacks due to their atrocious intensity. On the other hand, patients may have had difficulties in distinguishing between a single attack, a volley of paroxysms, and attacks followed by a relatively prolonged aftermath. With objective measurements such possible sources of error have been minimized.
However, our results after objective timing of V-1 neuralgia attacks do find a parallel in the estimations of the patients themselves as described by Sjaastad et al. (7). These authors collected a series of 19 patients and reported an attack duration of just a few seconds. It should be mentioned that although attacks could last up to 30 s, the ultrashort attacks lasting 5 s or less predominated, which was considered characteristic of V-1 neuralgia.
According to our data, exact timing of attacks of V-1 neuralgia rendered a mean of 6.5 s with a range of 2–32 s. Getting such ultrabrief attacks has not been a surprise, since a strong modulation of pain seems to occur in genuine neuralgias. Pathogenically, the offset of attacks may be as important as the triggering and the abrupt onset of pain. In fact, trigeminal neuralgia is characterized by both ultrashort duration of attacks and the presence of refractory periods, and these features indicate that pain is strongly modulated.
In our study, a possible, residual influence of carbamazepine on the flow and/or duration of V-1 neuralgia attacks are remote. First of all, the washout period (10 days) by far exceeded the plasma mean life of the drug, which in chronic treatment has been estimated as 15 ± 5 h (17). It should be mentioned that after a washout period of>3 times the mean life of a given drug no clinical effect is obtained. Secondly, the patients considered that measured attacks were similar to the ones previously experienced before treatment was started.
We have focused on duration of V-1 neuralgia because of the obvious diagnostic implications when other orbital or periorbital shorlasting headaches must be ruled out. Whether this finding holds true for V-2 and V-3 neuralgia must await further evidences, but admittedly there should be no substantial differences in the duration of the attacks stemming from any of the three branches of the same trigeminal nerve.
Our study has corroborated previous estimations that V-1 neuralgia attacks are much shorter than those of SUNCT (6, 7). With the present data the differential diagnosis of both disorders has been clarified to a large extent (Table 2, Fig. 1). The attack duration gap is even higher when comparing the reported measurements of attacks of CPH (18) and cluster headache (19) with those of V-1 neuralgia. Therefore, there should be little problems in differentiating V-1 neuralgia from SUNCT, CPH or CH from a temporal point of view.
Only paroxysms of PSH are shorter than those of trigeminal neuralgia. In accordance with the patients’ own estimations, the usual duration of PSH is 1–3 s (20–22), with most stabs lasting just 1 second (3, 10, 20, 21). Obviously the shortest V-1 attacks may overlap with the longest stabbing paroxysms. Nevertheless, under ordinary circumstances both disorders are easily distinguishable, even when their temporal aspects are regarded.
In conclusion, our study has ultimately clarified the duration of V-1 neuralgia attacks. These findings inherently carry important clues for the differential diagnosis of V-1 neuralgia vs. other shortlasting headaches. From present and previous evidences, the distinction between V-1 neuralgia and SUNCT is finally clear-cut.