Effect of Microvascular Decompression in Trigeminal Neuralgia Patients with or without Constant Pain
Tiril Sandell, M.D.
Department of Neurosurgery,
Division of Clinical Neuroscience,
The National Hospital (Rikshospitalet),
Oslo, Norway
Per Kristian Eide, M.D., Ph.D.
Department of Neurosurgery,
Division of Clinical Neuroscience,
The National Hospital (Rikshospitalet),
Oslo, Norway
Reprint requests:
Per Kristian Eide, M.D., Ph.D.,
Department of Neurosurgery,
Division of Clinical Neuroscience,
The National Hospital (Rikshospitalet),
N-0027 Oslo, Norway.
Email: ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■
Received, August 15, 2007.
Accepted, March 28, 2008.
OBJECTIVE: To study the effect of microvascular decompression (MVD) in trigeminal
neuralgia (TN) patients with or without constant pain.
METHODS: The study includes all first-time MVDs for facial pain performed by the
senior author (PKE) during the 6-year period from 1999 to 2005 in the Department of
Neurosurgery at the National Hospital. At the time of follow-up, pain relief was assessed
using a standard mail questionnaire; those patients still having residual pain were further
examined in the outpatient clinic or interviewed by phone.
RESULTS: The total study population includes 135 patients who underwent initial MVDs
(67% of MVDs for TN without constant pain and 33% of MVDs for TN with constant
pain). At the time of follow-up, the response rate was 95%, which provided us with
128 patients. The median observation period was 38 months (range, 12–87 mo). For
episodic pain, MVD caused complete (i.e., 100%) pain relief in 78% of TN patients
without constant pain and in 77% of TN patients with constant pain before MVD, and
a significant pain relief (i.e., worst pain marked as 0–3 cm on a 10-cm visual analog scale)
in 85 and 81%, respectively. For constant pain, MVD caused complete pain relief in
70% of the TN patients with constant pain before MVD, and significant pain relief in 77%.
CONCLUSION: In TN patients with constant pain before MVD, significant relief of
episodic and constant pain was observed in 81 and 77%, respectively. Hence, the presence
of constant pain should not prevent TN patients from being offered MVD.
KEY WORDS: Atypical trigeminal neuralgia, Constant pain, Microvascular decompression, Pain relief,
Trigeminal neuropathic pain, Typical trigeminal neuralgia
Neurosurgery 63:93–100, 2008 DOI: 10.1227/01.NEU.0000316856.32104.65 www.neurosurgery-online.com
Trigeminal neuralgia (TN) usually has a rather typical
clinical picture, with short-lasting episodic pain and
trigger points (4, 5, 8, 11), but it is recognized that TN
may include a constant pain that is burning, aching, and
throbbing in character (4, 5, 8, 12). TN with constant pain
(usually denoted atypical TN) represents a great diagnostic
challenge for the neurosurgeon because of uncertainty as to
whether the facial pain can be classified as TN or whether the
facial pain is trigeminal deafferentation pain caused by previous
neurodestructive procedures (4–7). Burchiel (4) has
made recent efforts in standardizing the terminology of facial
pains, differentiating between Type 1 TN (i.e., 50% episodic
pain) and Type 2 TN (i.e., 50% constant pain) in addition to
differentiating atypical TN from trigeminal neuropathic pain
(Table 1).
Although microvascular decompression (MVD) has a wellestablished
role in the treatment of typical TN without constant
pain, its role in treating TN with constant pain (atypical
TN) remains less clear. The presence of a constant pain component
has traditionally been considered a negative prognostic
factor for MVD (9, 14); however, one recent report indicates a
beneficial role of MVD even for atypical TN with constant pain
(12). Nevertheless, more data are needed to determine the role
of MVD in treating TN with constant pain.
For this purpose, we examined our material from firsttime
MVD procedures for facial pain during a 6-year period.
During this period, the senior author (PKE) put special focus
on differentiating between TN with and without constant
pain during the preoperative assessment. Different from TN
with 100% episodic pain, TN with constant pain was characterized
by burning, aching, and throbbing pain in addition
to the episodic pain, which is comparable to the clinical picture
of trigeminal neuropathic pain (6, 7). MVD for TN with
constant pain was performed exclusively in patients in
whom the constant pain had developed from being exclusively
episodic.
PATIENTS AND METHODS
Patients
The study includes all patients (n 135) undergoing first-time MVD
for facial pain (90 without and 45 with constant pain) in this hospital by
the senior author (PKE) during the 6-year period from 1999 to 2005.
Among the 135 patients, 24 (17.8%) had received previous neurodestructive
procedures, namely percutaneous retrogasserian glycerol rhi-
rhizotomy
(PRGR) in 21 and peripheral alcohol in three. Patients with
multiple sclerosis were excluded from the study. Preoperative demographic
data were retrieved from patient records (Table 2).
TN with or without Constant Pain
The patients were categorized into the following categories depending
on the presence or absence of a constant pain component before
MVD and at follow-up:
- Type 1a: TN with 100% episodic pain characterized by exclusively
episodic pain lasting seconds to a few minutes, typically shooting
and electrical in character. The pain can be evoked, typically with
trigger points present. There are pain-free intervals. - Type 1b: TN with greater than 50% episodic pain characterized by
episodic pain lasting seconds to a few minutes, typically shooting and
electrical in character. In addition, constant pain is present day and
night with a burning, aching, and throbbing character. The episodic
pain component predominates over the constant pain component.
This facial pain is comparable to Type 1 according to Burchiel (4). - Type 2: TN with greater than 50% constant pain characterized by a
constant pain present day and night with a burning, aching, and
throbbing character. Although this pain is constant, its intensity is
fluctuating. The pain is spontaneous and typically nonevoked,
though episodic pain may be present. This facial pain is comparable
to Type 2 according to Burchiel (4). Although the clinical picture is
comparable to that of trigeminal neuropathic pain (6, 7), TN with
constant pain (typically denoted atypical TN) is not synonymous
with trigeminal neuropathic pain, nor are its causes, which include
previous surgery, trauma, or stroke. Hence, trigeminal neuropathic
pain should be differentiated from TN with constant pain (Type 2, or
atypical TN) (Table 1). It should be emphasized that our TN patients
with constant pain (Types 1b and 2) initially had TN without constant
pain (Type 1a). TN with constant pain (Types 1b and 2) may
also occur de novo.
Radiological Assessment of Neurovascular Compression
The presence of neurovascular conflict between the affected trigeminal
nerve and an offending vessel was assessed by preoperative magnetic
resonance angiography (MRA), using a three-dimensional timeof-
flight MRA sequence and a T2-weighted three-dimensional
sequence. The MRA findings were compared with the perioperative
observations during the MVD procedure (Table 3).
MVD
Indication for MVD was clinical and based on the patient history, i.e.,
the presence of TN with or without constant pain when constant pain
had evolved from exclusively episodic pain. This was noted in the
patient records. All patients underwent MVD in the park bench position.
With the use of a small retrosigmoid craniectomy, an opening in the
dura (1.5–2 cm) was made in the angle between the sigmoid and transverse
sinuses. Microsurgery on the trigeminal nerve was made via a lateral
cerebellar approach without using a retractor. Arteries were
removed from the nerve and kept away from the nerve with a piece of
Teflon (DuPont, Wilmington, DE) while compressing veins were coagulated
and divided. The least possible manipulation of the trigeminal
nerve was attempted. Watertight closure of the dura was performed
under a microscope. Perioperative and neurological complications of
surgery were recorded (Table 4).
Clinical Assessment of Pain Relief
Follow-up assessment was performed by the first author (TS), who
was not involved in the management or surgery of the patients during
the study period. All patients received a questionnaire by mail;
the patients were asked to state whether facial pain had disappeared
completely (100%) without any signs of residual pain (and no medication
needed) or whether some facial pain was still present. Those
patients still having facial pain were seen at our outpatient clinic and
examined by the first author (TS). Those who were not able to meet
(e.g., because of age, travel distance, or other diseases) were interviewed
by phone. Clinical differentiation between residual TN with
or without constant pain was made. To assess the degree of residual
pain, a visual analog scale (VAS) of 0 to 10 cm was used to indicate
the patient’s worst intensity of residual pain (episodic and constant
pain components) (Tables 5 and 6). Pain relief was defined as complete
when facial pain had disappeared completely (100%) and as significant
when worst pain on the VAS was 0 to 3 cm. Complete pain relief
implied that no medication was taken.
Statistical Analysis
Statistical analysis was performed using PC-SPSS software version
12.0 (SPSS, Inc., Chicago, IL). Comparisons among groups (Types 1a,
1b, and 2) were performed using an independent-samples Student’s t
test for continuous data and χ2 test for categorized data. Significance
was accepted at the 0.05 level.
RESULTS
Patients and Pain Types
The patient data included a total of 135 first-time MVD procedures.
Demographic preoperative data are shown in Table 2.
There were no significant differences between TN patients
with or without constant pain regarding age, duration of pain
before MVD, or duration of follow-up after MVD (P 0.1;
independent-samples Student’s t test). Microvascular decompression
was performed for TN with 100% episodic pain (Type
1a) in 67% and for TN with constant pain (Types 1b and 2) in 33%.
Neurodestructive treatment had been performed previously
in 24 patients undergoing MVD procedures (PRGR in 21 and
peripheral alcohol in three), in 14 of 90 (16%) TN patients with
100% episodic pain (Type 1a), and in eight of 31 (26%) TN
patients with more than 50% constant pain (Type 2); hence, a
slight (nonsignificant) overrepresentation of previous neurodestructive
treatment in those TN patients with constant pain is
represented (Table 2).
Assessment of Neurovascular Conflict
MRA showed neurovascular contact in 97 of 135 trigeminal
nerves (72%) and no neurovascular contact in 28 (21%); preoperative
MRA was not available for 10 (7%) (Table 3).
Perioperative neurovascular compression was found in 129
MVD procedures (96%) (Table 3), which included a neurovascular
contact in 24 of the 28 MRA-negative observations (86%). A
single venous compression was observed in 25.8% of the TN
Type 2 group versus 15.6% in the TN Type 1a group, which did
not reach significance (P 0.08, Pearson χ2 test) (Table 3).
Complications of MVD
The perioperative and neurological complications of MVD
are presented in Table 4. Only one complication was encountered
in each patient. Acute perioperative complications (not
causing lasting complaints in any of the patients) were
observed in eight of 135 (6%) procedures. The worst complication
was in an 84-year-old woman who developed swelling in
her right cerebellar hemisphere that required a partial cerebellar
hemispherectomy after a few hours. Despite her age, she
recovered well, remaining pain-free. A 72-year-old man who
underwent an uneventful MVD was mobilized the first postoperative
day and recovered pain-free without any deficits, soon
returning to his home. Four weeks after surgery, without preceding
symptoms, he was found dead in his home; the cause of
death remains unknown.
At the time of follow-up, lasting problems were reported
after 18 (14%) of 128 MVDs. Five (4%) of 128 had persistent
facial numbness postoperatively; one of the five (with typical
TN) had had facial numbness before MVD. Six (4.7%) patients
had problems with dizziness. No patient had postoperative
hearing loss, although different degrees of subjective postoperative
hearing dysfunction were reported in seven of 128 (6%)
procedures.
Assessment of Residual Facial Pain
For assessment of residual facial pain after surgery, seven
patients were lost to follow-up (i.e., 95% response rate), leaving
a total of 128 initial MVD procedures. The median follow-up
period was 38 months (range, 12–87 mo).
Pain relief after MVD in relation to preoperative type of facial
pain is shown in Table 5. Among 85 MVD procedures in the TN
patients with 100% episodic pain (Type 1a), 66 (78%) patients
had complete (100%) relief from episodic pain, whereas an
additional six (7%) patients had significant pain relief (VAS,
1–3 cm) (see Table 5). We found no significant differences in
relief from episodic pain among patients who had 100%
episodic pain before MVD (Type 1a), had more than 50%
episodic pain (Type 1b), and had more than 50% constant pain
(P 0.79) (Table 5).
In those TN patients with constant pain, 77% of the MVD
procedures caused complete relief of episodic pain. This finding
was observed in both the TN patients with more than 50%
episodic pain (Type 1b) and those with more than 50% constant
pain (Type 2) (Table 5). Significant relief from episodic pain was
observed in 81% in the Type 2 group. Complete freedom from
constant pain was observed after 11 (85%) of 13 MVDs in those
with more than 50% episodic pain (Type 1b) and after 19 (63%)
of 30 MVDs in those with more than 50% constant pain (Type
2) (P 0.001, Pearson χ2 test) (Table 5). Hence, freedom from
constant pain was less likely in those who had developed predominant
constant pain (Type 2) than in those with predominant
episodic pain in addition to constant pain (Type 1b).
We also separately analyzed the 24 patients who had previously
received neurotoxic treatment (PRGR in 21 and peripheral
alcohol in three patients). As shown in Table 6, pain relief
after MVD was not worse in this subgroup, as compared with
the total group (Table 5).
DISCUSSION
The main observation is that at a median 38 months after
MVD, significant relief of both episodic and constant pain was
found in a high proportion of TN patients with constant pain
before MVD (Types 1b and 2).
Classification of Facial Pain
Unfortunately, there are no universally accepted classifications
of facial pain. The term typical TN usually refers to TN
with only episodic pain and trigger points (11), whereas TN
with constant pain (usually denoted atypical) represents a
diagnostic challenge (4, 5, 8). According to our tradition, we
consider TN as atypical when it includes a component of constant
pain that is aching, burning, and throbbing in character;
the classification refers to the clinical picture rather than the
cause of facial pain (6, 7). TN can be considered a unique form
of neuropathic pain (5). The International Association for the
Study of Pain defines neuropathic pain as pain initiated or
caused by a primary lesion or dysfunction in the nervous system
(11). Hence, in our TN patients with constant pain (Types
1b and 2), the neurovascular compression can be considered
the primary lesion. This lesion is different from anatomic
lesions caused by, for example, stroke, tumor, surgery, or
trauma; in such cases, the term secondary TN is used (11).
Symptomatic TN refers to that caused by multiple sclerosis. It
should be emphasized that a neurodestructive procedure cannot
produce atypical TN (Types 1b and 2), but deafferentation
pain, which should not be treated as TN.
The Burchiel classification scheme for facial pains does not
use the terms typical and atypical TN (4, 8). As shown in Table
1, it distinguishes between Type 1 TN, with more than 50%
episodic pain and Type 2 TN, with more than 50% constant
pain. Trigeminal neuropathic pain is used exclusively for neuropathic
facial pain caused by trauma (4, 8). Other authorities
in the field, such as Sindou et al. (12), use the term atypical TN
for TN with constant pain. Because taxonomy still remains a
topic of discussion, we differentiate between TN with 100%
episodic pain (Type 1a), TN with more than 50% episodic pain
(Type 1b, comparable to Burchiel Type 1 TN), and TN with
more than 50% constant pain (Type 2, comparable to Burchiel
Type 2 TN).
Patients
In our study, MVD was performed for TN without constant
pain (Type 1a) in 67% and for TN with constant pain (Types 1b
and 2) in 33% of patients. These numbers are comparable with
the study by Sindou et al. (12) of 362 patients, consisting of
65.5% with typical TN and 35.5% with atypical TN (i.e., presence
of constant pain).
There are several reasons why a rather high proportion of TN
patients developed constant pain (Types 1b and 2) at the time
of MVD. For all MVD procedures, the median duration of pain
was 5 years. Development of constant pain may be a part of the
natural course of TN (5). Another possibility is that constant
pain developed in the course of previous destructive procedures
(6, 7); there was a slight, nonsignificant overrepresentation
of previous neurodestructive procedures in the Type 2 TN
group. In such cases, the atypical TN should be classified as
trigeminal deafferentation pain (see Table 1), because a neurodestructive
procedure cannot produce TN. The clinical examination
of our patients did not indicate deafferentation pain.
The observation that patients previously treated with a neurodestructive
procedure did not experience less pain relief than
the whole group may support this conclusion.
This report does not give a clear answer as to why 33% of the
TN patients had developed constant pain (Types 1b and 2). In
comparison with TN patients without constant pain (Type 1a),
TN patients with constant pain (Types 1b and 2) were not significantly
older, and their pain had not lasted significantly
longer (Table 2). Regarding their type of vascular compression,
there were no significant differences (Table 3). There only was a
slight, nonsignificant overrepresentation of previous neurodestructive
procedures (PRGR or peripheral alcohol) in those with
constant pain (Type 2) (Table 2). At the time of follow-up,
hypoesthesia was found in 2.4% of those without constant pain
(Type 1a), but in approximately 7% of those with constant pain
(Types 1b and 2) (Table 4). Another aspect not explored here is
possible differences in psychological profile.
Indications for MVD were based on pain history. In the preoperative
assessment, the senior author (PKE) strove to delineate
whether or not constant pain (Types 1b and 2) had developed
from exclusively episodic pain (Type 1a). It should be
stressed that MVD was performed only in TN patients in
whom constant pain had developed from first being exclusively
episodic. Hence, in the present 135 patients, facial pain,
after first being 100% episodic, had changed to include a constant
pain characterized by burning, aching, and throbbing,
present day and night. However, patients with a clinical picture
of trigeminal neuropathic pain without a history of facial pain
indicative of TN, but with a history of previous stroke, trauma,
or surgery, were not considered for MVD. Hence, no patients
with pain caused by a previous injury or deafferentation were
offered MVD. In this respect, it can be discussed whether the
facial pain in the 10 TN patients with constant pain who had
been treated previously with PRGR or alcohol qualified as
trigeminal deafferentation pain. Because our clinical examination
did not indicate that PRGR had caused deafferentation, we
decided to include these 10 patients. Nevertheless, in TN
patients who have previously received a neurodestructive procedure,
it can be a diagnostic challenge to determine whether
or not the constant pain represents deafferentation pain.
Neurovascular Contact
In our experience, the use of MRA revealed a neurovascular
contact in approximately 70% of the nerves. The success rate
may have been increased by the use of other magnetic resonance
imaging modalities. Nevertheless, these results underline
that the MVD indication should primarily be based on clinical
reasoning; in our practice, MRA is only used as an aid secondary
to the pain history.
Complications of Surgery
The present data show a rather favorable profile regarding
complications from MVD, which is in line with the reports of
previous series (10). In earlier reports, a greater complication
rate of MVD was reported (15). As expected, the complication
profile was similar among the different pain categories. A disadvantage
is that it was unclear whether hearing dysfunction
was present before surgery as well.
Pain Relief after Surgery Depending on Type of Pain
At a median of 38 months after MVD, we found that in TN
patients with 100% episodic pain (Type 1a) complete (i.e.,
100%) relief of pain was achieved in 78% and significant pain
relief (i.e., worst pain being 0–3 cm on a 10-cm VAS) was
achieved in 85%. These numbers are comparable with previous
reports (1–3, 13, 15), thus contributing to the extensive literature
that MVD causes lasting relief from episodic pain in a high
proportion of patients.
Among those 15% in whom MVD failed to relieve pain, constant
pain was present in 4.7%. This number could result from
the natural course of TN. It appears less likely that the observation
relates to the MVD procedure itself because least possible
manipulation of the trigeminal nerve was attempted.
Although the literature on MVD for TN with episodic pain is
extensive, less attention has been given to effects of MVD in TN
patients with constant pain (9, 14). The general view among
neurosurgeons has been that MVD is less effective for TN with
constant pain.
The most important observation of this study is that MVD
for TN patients with constant pain (Types 1b and 2) resulted in
complete relief of episodic pain in 77% and significant pain
relief in 81%. Complete (100%) relief of constant pain was
observed in 85% of the group with more than 50% episodic
pain (Type 1b) and 63% of the group with more than 50% constant
pain (Type 2) before MVD; significant relief of constant
pain (i.e., worst constant pain being 0–3 cm on a 10-cm VAS)
was observed in 71% of the Type 2 group. With a median observation
period of 38 months, these numbers are high. The percentages
are comparable to the results of Sindou et al. (12),
who reported relief of atypical TN after MVD. However, the
results do not support earlier data that the presence of constant
pain represents a negative prognostic factor for MVD (14).
The most important reason for differences among data probably
relates to the selection of patients. The present data apply
only to TN patients who have developed constant pain (Types
1b and 2) after initially having episodic pain (Type 1a). We
would expect the results of MVD to diminish dramatically if
patients with trigeminal neuropathic pain, e.g., caused by stroke,
trauma, or surgery, or deafferentation after a neurodestructive
procedure, were included among those receiving MVD.
Thus, the major lesson from this data is that in one-third of
TN patients, episodic facial pain had developed into a constant
facial pain of burning, aching, and throbbing character that
was continuously present, day and night, in addition to their
episodic pain (Types 1b and 2). After MVD, both the episodic
and the constant pain components disappeared in a high proportion
of patients. It should be noted, however, that the chance
of being free from constant pain was significantly lower when
constant pain had become predominant (Type 2) as compared
with a situation in which episodic pain still was predominant
(Type 1b) (Table 5).
CONCLUSION
Among TN patients who have developed constant pain in
the course of their disease, MVD produces relief of both
episodic and constant pain components in a significant proportion.
The data support a role for MVD for TN even after
episodic pain has developed into constant pain. It is important
to stress that, in our TN patients with constant pain (Types 1b
and 2), the TN had evolved from exclusively episodic (Type 1a).
Hence, a careful delineation of the pain history is crucial in the
preoperative assessment.
REFERENCES
- Barker FG 2nd, Jannetta PJ, Bissonette DJ, Jho HD: Trigeminal numbness and
tic relief after microvascular decompression for typical trigeminal neuralgia.
Neurosurgery 40:39–45, 1997. - Barker FG 2nd, Jannetta PJ, Bissonette DJ, Larkins MV, Jho HD: The long-term
outcome of microvascular decompression for trigeminal neuralgia. N Engl J
Med 334:1077–1083, 1996. - Broggi G, Ferroli P, Franzini A, Servello D, Dones I: Microvascular decompression
for trigeminal neuralgia: Comments on a series of 250 cases, including
10 patients with multiple sclerosis. J Neurol Neurosurg Psychiatry
68:59–64, 2000. - Burchiel KJ: A new classification for facial pain. Neurosurgery 53:1164–1167,
- Burchiel KJ, Slavin KV: On the natural history of trigeminal neuralgia.
Neurosurgery 46:152–155, 2000. - Eide PK, Rabben T: Trigeminal neuropathic pain: Pathophysiological mechanisms
examined by quantitative assessment of abnormal pain and sensory
perception. Neurosurgery 43:1103–1110, 1998. - Eide PK, Stubhaug A: Relief of trigeminal neuralgia after percutaneous retrogasserian
glycerol rhizolysis is dependent on normalization of abnormal temporal
summation of pain, without general impairment of sensory perception.
Neurosurgery 43:462–474, 1998. - Eller JL, Raslan AM, Burchiel KJ: Trigeminal neuralgia: Definition and classification.
Neurosurg Focus 18:E3, 2005. - Frazier CH, Russell EC: Neuralgia of the face, an analysis of 754 cases with
relation to pain and other sensory phenomena before and after operation.
Arch Neurol Psychol 11:557–563, 2003. - Kalkanis SN, Eskandar EN, Carter BS, Barker FG 2nd: Microvascular decompression
surgery in the United States, 1996 to 2000: Mortality rates, morbidity
rates, and the effects of hospital and surgeon volumes. Neurosurgery
52:1251–1262, 2003. - Merskey H, Bogduk N: Classification of Chronic Pain. Seattle, IASP Press, 1994.
- Sindou M, Leston J, Howeidy T, Decullier E, Chapuis F: Micro-vascular
decompression for primary trigeminal neuralgia (typical or atypical). Long
term effectiveness on pain; prospective study with survival analysis in a consecutive
series of 362 patients. Acta Neurochir (Wien) 148:1235–1245, 2006. - Slettebø H, Eide PK: Aprospective study of microvascular decompression for
trigeminal neuralgia. Acta Neurochir (Wien) 139:421–425, 1997. - Szapiro J Jr, Sindou M, Szapiro J: Prognostic factors in microvascular decompression
for trigeminal neuralgia. Neurosurgery 17:920–929, 1985. - Taha JM, Tew JM Jr: Comparison of surgical treatments for trigeminal neuralgia:
Reevaluation of radiofrequency rhizotomy. Neurosurgery 38:865–871, 1996.
COMMENTS
Microvascular decompression (MVD) is a highly effective, well-documented
treatment for the lancinating pain of trigeminal neuralgia
(TN). However, there is less agreement regarding the effectiveness of
MVD for “atypical†TN, in which constant rather than episodic pain
predominates. In this study, Sandell and Eide evaluated results in two
groups of patients: those with typical, lancinating pain and those with
constant pain. Their results strongly suggest that MVD should be considered
for a subset of patients with atypical TN.
Patients were divided into two groups: those with brief, lancinating,
episodic pain (Type 1 or typical), and those with constant, burning
pain with episodic exacerbations (Type 2 or atypical); 135 patients
underwent MVD, 67% of whom were classified as Type 1 and 33% of
whom were Type 2. In the Type 2, atypical pain group, complete relief
of the episodic component of the pain was demonstrated in 77%, and
complete relief of constant pain was achieved in 70%. Significant pain
relief was achieved in more than 80% of patients, regardless of their
classification.
This report demonstrates the importance of diagnostic precision in
predicting outcome from neurosurgical interventions. In this series, all
patients with Type 2 TN originally started out with Type 1 TN. Given
the good results in this patient group, an initial typical presentation
appears to be an important diagnostic criterion for success of MVD.
Continued refinement of our ability to recognize facial pain syndromes
as distinct entities will lead to further improvement in outcomes. Not
all facial pain is due to TN!
Jaimie M. Henderson
Stanford, California
The diagnosis of TN is a clinical diagnosis based on a careful history.
There are no definitive tests to verify the diagnosis. The failure of
MVD in the management of facial pain other than that from TN makes
this clinical diagnosis extremely important for the neurosurgeon.
Traditionally, neurosurgeons have reserved MVD for patients with the
typical symptoms of TN neuralgia and have avoided MVD when the
presentation is atypical. In the current publication, Sandell and Eide
have analyzed the results of MVD in a series of 135 patients undergoing
the procedure, one-third of whom underwent surgery for atypical
TN. Interestingly and importantly, 70% of patients with atypical TN
achieved complete pain relief, and 77% achieved significant pain relief.
These results are quite comparable to the results attained in the group
of patients with typical TN.
It is extremely important that the reader note the definition of atypical
TN used in this study. These are not patients with atypical facial
pain or patients with trigeminal neuropathic pain. In the preoperative
assessment of his patients, the senior author was quite careful to determine
whether or not atypical pain had developed in the course of more
typical TN. MVD was reserved exclusively for patients in whom an
atypical TN developed in the course of more typical TN. Therefore,
these data apply only to patients with TN who have developed atypical
symptoms (constant facial pain of burning, aching, and throbbing
character) after initially having typical TN. The results suggest that
this group of patients may also benefit from MVD and should be considered
for treatment.
Daniel L. Barrow
Atlanta, Georgia
Sandell and Eide commented that the presence of constant pain creates
concern over the presence of deafferentation pain. I am less
convinced that this represents a major dilemma, as deafferentation pain
generally results from the ablative procedure, whereas constant pain
present before the ablation can easily be recognized as a component of
the original syndrome. In the present study, only 10 patients presented
with constant pain and a history of a previous ablative procedure.
The present study and current classification schemes focus on constant
versus episodic pain as the major indicator of “atypia.†However,
patients may also present with pain outside the trigeminal distribution
and psychological overlay. Thus, I tend to view tic as a continuum
from classic (no psychological overlay, within the Vth nerve distribution,
and purely episodic) to highly atypical. Indeed this continuum
extends from atypical TN to atypical facial pain.
Ultimately, the more atypical a case of TN is, the less apt I am to do
MVD without radiological evidence of significant vascular compression.
On the other hand, I am extremely hesitant to perform ablative
procedures on patients with less typical symptoms. This bias comes
from the fact that ablative approaches to constant neuropathic pain
throughout the body tend to have poor durability and can cause deafferentation
pain. Thus, my approach has been to proceed with neuromodulation
approaches including peripheral nerve, ganglion, motor
cortex, and deep brain stimulation in patients with predominantly
constant pain (atypical TN) in whom I cannot find a vessel on imaging.
If I can find a vessel, I will always perform an MVD before
embarking on neuromodulation, because the former procedure has
the potential for cure, whereas neuromodulatory approaches generally
produce only partial relief. Moreover, the latter require indwelling
devices with all of the associated morbidity. Finally, the durability of
facial pain relief with neuromodulation is unclear. Put simply, whether
or not MVD is as good for constant pain or episodic pain, I would still
use it for atypical pain if imaging is suggestive of a vessel. Although
the study by Sandell and Eide will not change my protocol, it provides
encouraging data to support MVD in discussions with patients with
Type 2 TN.
The authors note that constant pain may occur de novo, but all of
their patients with Type 1b/2 TN reported that the syndrome started
with purely episodic pain (Type 1a). This feature may be an important
clue to implicate a vascular etiology and explain why MVD is so successful
in these patients. When discussing MVD with patients with
Type 2 TN, it will be important not to quote these results to those who
presented originally with a predominantly constant pain syndrome.
Finally, the present study convincingly contradicts the popular notion
that the constant pain component of TN syndromes derives from
venous compression.
Nicholas M. Boulis
Atlanta, Georgia
It is conventional wisdom that the paroxysmal component of TN
pain is best relieved by MVD surgery. A common corollary of this
postulate is that patients with significant constant pain as part of
their facial pain syndrome do not fare as well with MVD. Some surgeons
go as far as not to operate on such patients for fear that their
atypical features imply a worse clinical outcome. In this article,
Sandell and Eide challenge this conventional wisdom with their experience
in treating patients with such pain. In this retrospective review
of a series of 135 patients, they compared the results of patients with
less than 50% constant pain with those whose constant pain is more
than 50% of the total. Those patients with pure paroxysmal pain were
also analyzed as part of this study. The group as a whole was treated
in a similar fashion, by a single surgeon, at a single center. Follow-up
ranged between 1 and 7 years. They found that the relief of both
paroxysmal and constant pain occurs with high frequency after MVD
and does not seem to depend on the degree of preoperative constant
pain reported by the patient.
The message of this article is that MVD should not be denied merely
because of the presence or even degree of constant pain in the symptom
complex. And although Sandell and Eide provide good information
to support this message, it should be remembered that the data
may be confounded by a number of factors. Given that this is a retrospective
study, for example, it is quite likely that baseline patient characteristics
in the three groups are not comparable. Moreover, the widely
disparate follow-up period may yield comparisons between groups
followed for different periods of time. Finally, it is important to remember
that because entry into this study was not randomized or controlled,
the number of patients in each group varied widely. It is difficult,
therefore, to make definitive conclusions about the effect of
constant pain on results after MVD.
Despite the limitations inherent in a retrospective study, the authors
here do a good job of examining the role of nonparoxysmal pain on the
results of surgical treatment. Their data certainly show that constant
atypical components of trigeminal neuralgia may not be such a poor
prognostic factor after all.
Oren Sagher
Ann