POSTHERPETIC neuralgia rarely presents as a motor deficit. 1–3Even more unusual is its presentation without a history of the typical rash in the distribution of the pain (zoster sine herpete). 4,5We present what we believe to be an extremely rare case of zoster sine herpete with significant motor involvement.
A 63-yr-old white man presented to the chronic pain clinic for evaluation of left-sided midabdominal and flank pain. The pain had begun approximately 2 months prior, and the patient had noticed a small “soft tissue mass” in his midabdominal area that had been gradually increasing in size. He complained of burning pain and hyperesthesia in the distribution of the bulge. A surgeon had been consulted for evaluation of what was initially thought to be a lipoma. The surgeon noted that the area had the appearance of a surgically denervated muscle, but the patient had never had surgery in this region. The patient was referred to the chronic pain clinic for further evaluation.
On physical examination, the abdominal wall was noted to be markedly asymmetric, with what appeared to be a soft mass in the left T10 distribution (fig. 1). The mass was slightly tender when palpated. No rash, vesicular eruptions, or cutaneous scarring was noted. The patient was asked to perform a Valsalva maneuver, which caused the affected area to become firm and to diminish in size. Hyperesthesia was noted in the left T10 distribution. The sensory examination was otherwise normal.
Abdominal computed tomography performed previously did not show any pathologic processes, including hematoma, lipoma, or other tumor masses. Further workup included thoracic spine magnetic resonance imaging, which revealed no evidence of nerve root impingement that might explain the thoracic nerve root symptoms. Serologic studies showed that the patient had positive anti-varicella-zoster virus (VZV) immunoglobulin G titers but was negative for anti-VZV immunoglobulin M. Electromyelography to evaluate the external abdominal oblique, internal abdominal oblique, rectus abdominus, and thoracic paravertebrals revealed membrane irritability in the form of positive waves and fibrillations in all of the muscle groups on the left, most prominently in the T10 distribution. Motor unit action potentials were present but were decreased in number, although the configuration and amplitude appeared essentially normal.
A presumed diagnosis of zoster sine herpete (shingles without cutaneous eruption) with thoracic motor involvement was made. The patient's soft tissue mass actually represented an area of thoracic intercostal and abdominal musculature with motor paralysis secondary to VZV. Treatment with a series of thoracic epidural steroid injections, gabapentin, and nortriptyline resulted in a significant resolution of his pain.
A physical examination at 6 months revealed that the patient's thoracic motor paralysis had almost completely resolved. He also reported nearly complete resolution of the burning pain and hyperesthesia. A follow-up electromyelogram performed 5 months after the initial study revealed motor unit action potentials that were essentially normal in overall number, amplitude, duration, and firing pattern. Membrane irritability in the T10 distribution was also dramatically improved.
Varicella-zoster virus is responsible for chickenpox, during the primary infection, and shingles, during reactivation episodes later in life. 6Following primary infection, the virus remains latent in the dorsal root ganglia of the nervous system. Its reactivation gives rise to shingles (herpes zoster), which is usually characterized by unilateral neuralgia followed by a vesiculobullous eruption in a dermatomal distribution. 6The incidence of shingles is known to be higher in the elderly (> 55 yr old), in the immunocompromised, and in those originally infected prior to 1 yr of age. 7
Rarely, patients experience shingles without the zosteriform rash (zoster sine herpete). Lewis 5described 120 patients that experienced “zoster-type” pain without a rash in a dermatomal distribution distant from a dermatome with a rash. There have been few well-documented cases of zoster sine herpete that have been virologically shown to be caused by VZV. 4
Virologic evidence of herpes zoster is difficult to obtain due to a lack of specificity, and because most cases are diagnosed based on clinical features. 6Positive anti-VZV immunoglobulin G antibodies represent previous exposure. Positive anti-VZV immunoglobulin M antibodies are suggestive, but not specific, for herpes zoster. 6Sundqvist 8showed that increases in anti-VZV immunoglobulin M, due to zoster, are transient. In a study of 25 patients, 21 were positive for VZV immunoglobulin M; however, only one patient was still positive 46 days after the illness. Therefore, it is not surprising that the immunoglobulin M serologic results were negative in this patient, since the delay in diagnosis resulted in the serologic results not being obtained until approximately 70 days after the onset of his illness. Dahl et al. 9reported that the specificity of clinical diagnosis of zoster (with a rash) is 95% and, therefore, laboratory confirmation is not always required. The specificity of clinical diagnosis in zoster sine herpete is unclear.
Herpes zoster usually affects sensory nerves; however, it can also damage motor neurons. The incidence of motor involvement with VZV reactivation has been estimated to be 0.5 2,10to 5%. 1,3The electromyelographic findings seen with thoracic motor paralysis are identical to those observed in our patient, with positive waves and fibrillations seen in the affected muscle groups. 11Documented cases most commonly have involved cranial nerves; however, there are also reports of abdominal, diaphragmatic, and lower extremity paralysis. 12–15Yaszay et al. 16describe unilateral left C5–C6 segmental paresis associated with a zoster rash in an otherwise healthy individual.
The true incidence of thoracic motor paralysis with herpes zoster is unknown. Thomas and Howard 3reviewed 1,210 patients with herpes zoster, 61 of whom were found to have zoster-induced segmental muscle weakness. The thoracic dermatome was involved in only two of these cases. Although the thoracic dermatome seems to be the most frequently affected in herpes zoster, it appears to have the lowest incidence of motor paralysis. 8This apparent low incidence may be misleading due to the fact that segmental weakness of intercostal and abdominal muscles is likely to pass unnoticed. 8,10
Motor paralysis secondary to herpes zoster appears to have a relatively good prognosis for recovery. Complete or nearly complete recovery has been reported in 50–70% of cases. 2,3,10Our patient demonstrated nearly complete resolution of his sensory symptoms and motor paralysis over a 5-month period, as evidenced by resolution of the abdominal wall asymmetry.
The differential diagnoses entertained in this case included soft tissue tumor mass (including lipoma or cyst), herniation of abdominal musculature, abdominal muscle hematoma, and lower motor neuron lesion. Tumor mass, hematoma, and muscle herniation were excluded by physical examination and abdominal computed tomography scan. Causes of lower motor neuron injury could include surgical injury, trauma, cutaneous nerve entrapment, or thoracic nerve root injury due to neural foraminal stenosis, compression fracture, or herniated nucleus pulposus. The patient had no history of surgery or trauma to the area, and magnetic resonance imaging revealed no evidence of thoracic nerve root injury. Cutaneous nerve entrapment is most commonly found in patients who have had previous surgery. 17Entrapment of these cutaneous nerves is typically noted as intermittent pain in the lateral rectus abdominis musculature with a single painful trigger point on palpation. 18A literature search revealed no descriptions of cutaneous nerve entrapment in the external or internal oblique musculature. As stated previously, electromyelographic evaluation of our patient revealed involvement of the internal and external oblique, rectus abdominus, and thoracic paravertebral musculature, making cutaneous nerve entrapment an unlikely diagnosis.
As discussed, diagnoses of zoster sine herpete are most often made clinically given the difficulty of obtaining serologic evidence. Clinical evidence suggests that our patient had both zoster sine herpete and thoracic motor paralysis secondary to herpes zoster. A literature search revealed that no similar cases have been reported. In this case, serologic evidence is consistent with, but not diagnostic of, herpes zoster. The clinical symptoms of pain and thoracic motor involvement, as documented by electromyelography, further strengthen the validity of the diagnosis. The patient's rapid recovery over 5 months is also consistent with the natural course of zoster motor paralysis. 2,3,10We believe that thoracic motor paralysis secondary to zoster sine herpete is a rare diagnosis. However, as mentioned previously, this diagnosis may be more common than is apparent due to the difficulty in diagnosing each occurrence. 8,10
The authors thank John C. Hendricks, M.D. (Associate Professor, Department of Surgery, Scott & White Memorial Hospital, Texas A&M University, Temple, Texas), for referral; John A. Schuchmann, M.D. (Associate Professor and Chairman, Department of Physical Medicine and Rehabilitation, Scott & White Memorial Hospital, Texas A&M University, Temple, Texas), for electromyelographic evaluation; and Jolene D. Bean-Lijewski, M.D. (Associate Professor, Department of Anesthesiology, Scott & White Memorial Hospital, Texas A&M University, Temple, Texas), and Jeff R. Gibson, M.D. (Assistant Professor, Department of Anesthesiology, Scott & White Memorial Hospital, Texas A&M University, Temple, Texas), for valuable comments in the preparation of this manuscript.