FROM:
Chiropractic & Osteopathy 2009 (Sep 21); 17 (1): 9 ~ FULL TEXT
Donald R Murphy, Eric L Hurwitz, Jonathan K Gerrard, and Ronald Clary
BACKGROUND: It is commonly stated that nerve root pain should be expected to follow a specific dermatome and that this information is useful to make the diagnosis of radiculopathy. There is little evidence in the literature that confirms or denies this statement. The purpose of this study is to describe and discuss the diagnostic utility of the distribution of pain in patients with cervical and lumbar radicular pain.
METHODS: Pain drawings and descriptions were assessed in consecutive patients diagnosed with cervical or lumbar nerve root pain. These findings were compared with accepted dermatome maps to determine whether they tended to follow along the involved nerve root's dermatome.
RESULTS: Two hundred twenty-six nerve roots in 169 patients were assessed. Overall, pain related to cervical nerve roots was non-dermatomal in over two-thirds (69.7%) of cases. In the lumbar spine, the pain was non-dermatomal in just under two-thirds (64.1%) of cases. The majority of nerve root levels involved non-dermatomal pain patterns except C4 (60.0% dermatomal) and S1 (64.9% dermatomal). The sensitivity (SE) and specificity (SP) for dermatomal pattern of pain are low for all nerve root levels with the exception of the C4 level (Se 0.60, Sp 0.72) and S1 level (Se 0.65, Sp 0.80), although in the case of the C4 level, the number of subjects was small (n=5).
CONCLUSIONS: In most cases nerve root pain should not be expected to follow along a specific dermatome, and a dermatomal distribution of pain is not a useful historical factor in the diagnosis of radicular pain. The possible exception to this is the S1 nerve root, in which the pain does commonly follow the S1 dermatome.
From the FULL TEXT Article:
DISCUSSION
This study failed to find much support for the common notion that extremity pain that
arises from radiculopathy typically follows along a specific dermatome. In general, the
sensitivity (SE) and specificity (SP) of this finding were low, suggesting that this factor is not useful in making the diagnosis of radicular pain. The one exception is S1 radicular pain, in which a
dermatomal pattern of pain was found in nearly two-thirds of patients and the Se and Sp
were high enough (Se 0.65, Sp 0.80) to make this a useful finding in the diagnosis of S1
radiculopathy. In patients with C4 radicular pain, 60.0% had a dermatomal pattern and
the Se and Sp were also relatively high (Se 0.60, Sp 0.72), but there were only 5 subjects
with radicular pain at this level, so firm conclusions cannot be drawn.
This study does not allow firm conclusions to be drawn about the reason for the absence
of a dermatomal pattern of pain in most cases. One of the possibilities for this, however,
is that patients with nerve root pain may also have other sources of pain, such as the
intervertebral disk, dura mater or other tissues, that are producing a nociceptive, as
opposed to neurogenic, pain pattern [20]. Also, as Bove, et al [32] pointed out, it has
been demonstrated that spontaneous activity in neurons that innervate muscle or other
deep tissues can develop after nerve injury [45] or nerve inflammation [46]. If a portion
of the referred limb pain was arising from this spontaneous activity, the pattern of pain
would not be expected to follow a specific dermatome. Another possibility is that there
can be overlap between dermatomes, with one dermatome encompassing one or two
adjacent segments. [47, 48] So it may be possible for an individual with nerve root pain
to have a dermatomal distribution, but for this distribution to fail to precisely match the
pattern depicted in the classic dermatome maps. Finally, it is known that intense and/or
persistent nociceptive input can produce an expansion in the size of the receptive fields of
those dorsal horn cells that receive and project nociceptive signals from the periphery
[49]. As a result, these cells are capable of responding to input from a greater number of
incoming afferent fibers, leading to referral of pain that is perceived in a wider area than
would occur without this expansion. Nonetheless, none of these factors changes the
primary conclusion of this study, i.e., that the dermatome maps commonly used to
identify the expected pattern of radicular pain are not useful as a clinical diagnostic tool.
Finally, in a patient with conjoined nerve roots, which can be seen on imaging in
approximately 4% of individuals [50], the pain may follow the path of both nerve roots,
and thus not conform to the dermatome pattern of a single nerve root. None of the
patients in this sample had this anomaly, and beside this, multiple nerve root involvement
was considered in our analysis.
The findings of this study are consistent with those of other authors. Nitta, et al [51] used
selected nerve root block in 71 patients with lumbar radiculopathy and found that nerve
root pain at L4 and L5 commonly deviated from the classic dermatomal pattern, but that
at S1 typically followed the classic S1 distribution. Bove, et al [32] assessed 25 patients
diagnosed with lumbar radiculopathy to determine whether the pain was perceived as
“deep” or “on the skin”. In all cases the pain was reported to be “deep”, both at rest and
when evoked by performing a SLR [32]. They concluded that the diagnostic utility of
dermatomal maps should be questioned on the basis that in no case was pain described as
“on the skin”, which would be expected if the pain pattern was dermatomal in nature.
Unfortunately, the subjects in the present study were not asked about the superficial vs.
deep location of their pain, so no confirmation of the finding of Bove, et al [32] could be
made. However, it is significant that the conclusions regarding the diagnostic utility of
dermatome maps were the same in these two studies. Ljunggren, et al [52] assessed 77
subjects with “lumbago sciatica” secondary to herniated disk and found some similarity
in the pain location between patients with L5 those with S1 radiculopathy, but specific
dermatomal maps were not used in this comparison. Anderberg, et al [33] found no
relationship between the distribution of pain and the level of cervical radicular pain as
determined by selective nerve root block.
The dermatome pattern for the S1 nerve root that is most commonly described in the
literature involves the posterolateral thigh and leg and the lateral foot. This study found
that this pattern of pain was seen in 65% of patients with S1 radicular pain. Thus, a
dermatomal pain pattern may be useful diagnostically in patients with S1 nerve root pain.
However, it should be noted that no patients who did not have radiculopathy were
included in these data. It is known that the lower extremity referred pain pattern of
somatic structures innervated by the S1 segment also commonly follows the classic S1
dermatome [53]. In addition, the study did not query subjects as to whether their pain
was perceived as deep or superficial. Further work, specifically which assesses how
common it is for patients with other pain sources to report pain that follows a similar
pattern as that of S1 radiculopathy, is required to clarify this.
For patients with radiculopathy at levels other than S1, the patient’s description and
drawing of the pain pattern does not appear to be a useful piece of diagnostic
information. Clinicians should not expect the pain from radiculopathy at levels other
than S1 to follow along a specific dermatome.
Scapular pain was present in approximately half the patients with cervical radicular pain.
There was a trend toward increased likelihood of the presence of scapular pain relative to
nerve root level, suggesting that the lower the cervical nerve root of involvement, the
greater the likelihood of the presence of scapular pain. However, the small sample size
does not allow definitive conclusions to be drawn about this. It is not clear whether the
scapular pain arises from the nerve root itself or from other sources of pain in these
patients. However, it is interesting that a strong majority (78.8%) of those patients who
reported scapular pain had HD, with or without LCS. The commonness of scapular pain
in patients with HD may suggest that the scapular pain may arise from referred pain from
the disk itself, rather than arising from the nerve root. Slipman, et al [54] assessed the
referred pain patterns of 41 patients undergoing provocative discography in the cervical
spine. They found that the scapula area was one of the most common areas of referred
pain in these patients, and was reported most commonly by patients with concordant pain
provoked by injection of the C4-5 through C6-7 levels. This is consistent with the
findings presented here that scapular pain was most common in patients with nerve root
pain from C5, C6 and C7 which, in those cases in which HD was present, would involve
the C4-5 through C6-7 levels. However, additional work in the area of sources of
referred scapular pain is required before firm conclusions can be drawn. In addition,
because of the low Se and SP, the presence of scapular pain is not useful for the purpose
of diagnosing nerve root pain per se. Further work is needed to determine the diagnostic
utility of the presence of scapular in diagnosing disk pain.
The majority of patients described the quality of their pain as either “aching” or “sharp”.
Far fewer described the pain as “burning”. There were no significant differences
between nerve root levels with regard to pain description. The Se and Sp for “aching”
and “sharp” pain descriptions were low, suggesting that these descriptions are of little
diagnostic value in identifying nerve root pain. It appears from the data presented here
that the description of “burning” pain is highly specific (Sp 0.86-1.00) for the presence
of radicular pain, however, given the low number of positive responses to this
description, these high estimates of specificity are likely an artifact of the study
population and require confirmation in other clinical populations.
One potential weakness of this study is its retrospective nature. However, this may also
be seen as a strength in that the description of each patient’s pain pattern was recorded by
the examining clinician in the manner that is normally carried out in clinical practice,
rather than as part of a research project on the dermatomal or non-dermatomal nature of
nerve root pain. Thus, clinician bias regarding the expected pain pattern was not a factor
in this recording.
CONCLUSION
It is concluded from the data presented here that in most cases nerve root pain should not
be expected to follow along a specific dermatome, at least as described by commonly
used dermatomal maps, and a dermatomal distribution of pain is not a useful historical
factor in the diagnosis of radiculopathy. The exception to this is S1 radicular pain, in
which the pain does commonly follow the S1 dermatome. Scapular pain is common in
patients with cervical radicular pain, particularly those whose nerve root pain is related to
HD, and may represent referred pain from the disk itself. The quality of pain is generally
an insensitive and non-specific finding in patients with nerve root pain.
