Pain in Guillain-Barre syndrome

Although pain is acknowledged as a common symptom in several retrospective reports on Guillain-Barre syndrome (GBS), ^[1-7] the various pain syndromes seen in GBS often go unrecognized and may be undertreated. Uncontrolled pain can occur despite compassionate and supportive measures such as the use of air mattresses, careful turning of patients and positioning of limbs, and the use of padding over the elbows and knees to prevent pressure palsies. ^[8] In the only previously published prospective study of pain in GBS, Ropper and Shahani ^[9] detailed the clinical features and problems of pain in 29 consecutive patients. They described low back and proximal leg pain of muscular origin in 55% of their patients early in the illness and in 72% at some time over the first month. However, there was very little description of other pain syndromes; there was no formal assessment of pain intensity, pain relief, or disability; and follow-up was limited. We report in a prospective longitudinal survey the character, intensity, and frequency of pain in GBS and the response to treatment.

Methods.

The clinical series consisted of 55 consecutive adult patients seen at the University of Western Ontario, London, Ontario, or the University of Calgary, Calgary, Alberta, and diagnosed with GBS according to defined criteria. ^[10] Four other patients were excluded from analysis because they turned out to have subacute ^[11] or chronic ^[12] inflammatory demyelinating polyradiculopathy. Patients were evaluated within 48 hours of admission and at 2, 4, 8, 16, and 24 weeks. Telephone follow-up was obtained when possible for those patients who were unable to keep outpatient appointments. Patients were asked to describe the character of their pain independent of paresthesia and other sensory symptoms. ^[13] Pain intensity was assessed using a linear Visual Analogue Scale ([VAS] 0 to 10 cm) and the Present Pain Intensity of the McGill Pain Questionnaire. ^[14] The latter is a categorical scale of pain consisting of the words none, mild, discomforting, distressing, horrible, and excruciating. VAS for pain intensity on admission was correlated with the Disability Grading Scale (Table 1) ^[15] on admission and at every evaluation time. Treatment was noted, and the response to pain intervention was evaluated using a linear VAS (0 to 10 cm) for pain relief.

Values are expressed as means +/- SD. Correlations between pain intensity and disability were performed by the Spearman correlation test in which r is the correlation coefficient. Results were considered significant at p < 0.05.

Results.

Patient demographics and clinical characteristics.

Fifty-five consecutive patients with GBS (34 men, 21 women) were followed for a mean of 20.7 weeks +/- 7.1 (SD). Their mean age was 50.6 years. One patient had preexisting diabetes mellitus, and one had mild renal failure. Primary treatment for GBS included plasmapheresis in 30 (54.5%), intravenous immunoglobulin in 13 (23.6%), and both in 8 (14.5%). Thirteen patients (23.6%) required intubation and mechanical ventilation. On admission the mean Disability Grading Scale score was 6.1 +/- 2.0, with 49 patients (89.1%) being moderately disabled (grade 3) or worse. Three patients died during the course of the study-two respirator-dependent patients died of pulmonary sepsis at the ages of 78 and 81, and one patient with Emery-Dreifuss muscular dystrophy died of cardiac complications independent of GBS. At 24 weeks in 45 assessable patients, the mean Disability Grading Scale score was 1.9 +/- 1.8, with 13 patients (28.9%) grade 3 or worse.

Pain was reported by 47 patients (85.5%) on admission. Pain preceded weakness by a mean of 6.1 days in 16 patients, weakness preceded pain by a mean of 5.3 days in 14, and both symptoms appeared simultaneously in seven. Ten patients were unable to report the temporal pattern of pain and weakness. An additional two patients reported new-onset pain during follow-up, so that 49 patients (89.1%) experienced pain attributable to GBS at some time during the course of their illness.

On admission, mean pain intensity (VAS) was 4.7 +/- 3.3. This represents moderate pain despite the fact that many patients had already been given acetaminophen or codeine by the referring physicians or house staff. Pain that was either distressing, horrible, or excruciating was described by 26 patients (47.3%). This subgroup represented a mean pain intensity (VAS) of 7.0 +/- 2.0. Mean pain intensity for the entire group is shown in Figure 1. Pain intensity on admission correlated poorly with neurologic disability on admission (r = 0.26, p = 0.06) and throughout the period of study (r < 0.20, p > 0.10).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Mean pain intensity (+SD) over time in 55 patients with GBS.

Discussion.

The most striking finding of this study was that back and leg pain of a moderate to severe nature was a presenting symptom in almost half our patients with GBS. Recognition of this problem is especially important in a disease that may immobilize patients and require intubation so that patients are unable to communicate their distress. However, we did not find a correlation between pain on admission and subsequent disability, unlike a previous suggestion ^[9]; patients with significant pain do not necessarily progress to severe paralysis or have a poor prognosis.

The etiology of back and leg pain in GBS is likely multifactorial. During the acute stage, muscles that are becoming paralyzed are commonly painful. Although we did not routinely measure serum creatinine kinase, Ropper and Shahani ^[9] found that this enzyme was consistently elevated in patients with painful muscles. Our finding that acute back and leg pain was commonly exacerbated by straight leg raising provides indirect evidence that traction on inflamed nerve roots ^[2] could be responsible for some of the pain. Sciatica in GBS can lead to a misdiagnosis of acute radiculopathy caused by herniated disk. ^[9,17] However, pain exacerbation by nerve root traction maneuvers is so characteristic of GBS that it actually aids in the diagnosis of this syndrome. Irritation of the nervi nervorum, which innervates nerve trunks, ^[18] may also refer pain to the paraspinal region via the dorsal primary rami.

Although none of our patients received corticosteroids, methylprednisolone may relieve severe muscle pain in GBS, ^[19] and this may be due to its anti-inflammatory effects at the nerve root level. During the recovery phase of GBS, muscle and joint pain is common but may not be as severe as in the acute stage. We speculate that persistent muscle and joint pain may be related to mechanical factors; one report suggests that weakness of the paraspinal musculature can lead to poor posture and abnormal mechanical forces on the spine that can prolong back pain. ^[20] Although a program of passive and active-assisted exercises is crucial to prevent contractures and regain strength, range-of-motion exercises are frequently painful in the setting of weak muscles and stiff joints.

We relied frequently on oral and parenteral opioid analgesics to treat somatic back and extremity pain; this strategy was effective in most patients. Several patients placed on intravenous morphine infusions in the intensive care unit later described this intervention as being vital to their recovery. This is in contrast to the report of Ng et al. ^[7] who stated that opioid analgesics had limited value in ventilated patients. Starting doses of sustained-release oral morphine were in the range of 30 to 60 mg bid. Starting doses of intravenous morphine infusion were usually in the range of 1 to 3 mg/hr and reached a maximum of 7 mg/hr.

Opioid drugs were generally well tolerated. Respiratory depression is always a concern in non-ventilated GBS patients, but careful dose titration and screening of vital capacity and respiratory rate minimizes the risk. Psychological dependence or addiction was never an issue during opioid withdrawal. None of our patients had a history of drug abuse, and in such a population the risk of iatrogenic addition is extremely low. ^[21] Opioid analgesics probably contributed to constipation in our patients despite the use of an aggressive bowel regimen involving bowel stimulants and stool softeners. We did not employ epidural morphine, but it would be an acceptable alternative to systemic opioids in ventilated patients with primarily low back and leg pain in the acute phase of the disease. ^[22,23] Spinal opioids can produce profound analgesia in small doses without motor, sensory, or autonomic effects. ^[24]

In summary, patients with GBS require close monitoring for pain and aggressive use of analgesic agents for those with moderate to severe pain.

Acknowledgments

We thank Sue Kimber, RN, and Peggy Vandervoort, RN, for clinical assistance and data collection and Anthony Iezzi, PhD, for statistical support.