Introduction

Pain is a frequent and major consequence of spinal cord injury (SCI).1,2 Recent estimates of occurrence of pain in SCI patients ranged from 30% to in excess of 90%, with approximately 30% of this manifesting as neuropathic pain (NP).1,2,3 NP is an intractable form of pain resulting from nerve cell damage or axonal damage caused by a primary lesion (such as SCI) or dysfunction in the nervous system and may not require peripheral nociceptor activation.2,4 However, it frequently involves both peripheral and central sensitisation. NP is often described as paroxysmal, burning, stabbing, pulsing, electric shock-like or dysaesthesic – a spontaneous or evoked unpleasant abnormal sensation. Hyperalgesia, or heightened response to painful stimuli, may be present in the area of injury or in the surrounding area which is indicative of central sensitisation. Allodynia (pain in response to a non-painful stimulus such as a light touch) also indicates central sensitisation.4

While some forms of pain may be managed successfully with simple analgesics and adjunctive therapy (such as physiotherapy, transcutaneous electrical nerve stimulation (TENS)), NP is often more difficult to treat.1,2,5 Early diagnosis and treatment are preferable, as chronic NP is very difficult to treat. A realistic goal is to reduce the pain to an acceptable level for the patient, as total eradication is rarely possible.4

Tricyclic antidepressants, and anticonvulsants such as sodium valproate and carbamazepine, have been widely used in the treatment of NP. While these agents are often effective, adverse effects also frequently occur.4 Other agents used include membrane stabilisers such as mexiletine, benzodiazepines, and opioids.

Gabapentin is an anticonvulsant structurally related to the neurotransmitter gamma-amino butyric acid (GABA), however, it does not bind significantly to GABA receptors. The mechanisms of gabapentin's anticonvulsant and analgesic action remain unclear, but are believed to be different from other anticonvulsants. Gabapentin appears to be quite safe and tolerated with a low incidence of side effects.6

Promising results have been reported on use of gabapentin in the treatment of various NP syndromes. Two recent clinical trials have established the effectiveness of this agent for diabetic neuropathy and post-herpetic neuralgia.7,8

Methods

SCI patients prescribed gabapentin were identified from the pharmacy dispensing records and from manual searches through spinal office correspondence. Data were then retrieved from the medical records of those prescribed gabapentin for NP.

Pain was assessed prior to treatment (baseline), and at 1, 3 and 6 months during treatment, with a 10 cm visual analogue scale (Figure 1), which ranged from 0 (‘no pain’) to 10 (‘worst pain imaginable’) or by the documentation of a verbal description of pain. The verbal descriptions were based on patients' reports and were not standardised.

Figure 1
figure 1

Visual analogue scale

Full size image

Results

Forty-four patients were identified as being prescribed gabapentin for NP. Six of these patients had little or no further information other than the fact that therapy for two patients was subsidised by Veteran Affairs and another two by accident compensation.

Of the 38 patients with information, the mean age was about 47 years, with a range of 15–75 years. There were more males (n=28) than females (n=10). There were slightly more paraplegic (n=19) than tetraplegic patients (n=16) and almost three times more chronic (n=24), compared to acute (n=9), patients. (‘Acute’ in this situation is defined as less than 6 months from time of injury). Table 1 shows the breakdown in patient deomgraphics.

Table 1 Patient demographics
Full size table

There were various documentations of verbal reports of pain described by the patients. These reports ranged from constant, severe, shooting, burning pain, to constant tingling pain on movement and touching of ‘normal’ areas. Some examples of the verbal reports of pain documented prior to treatment with gabapentin are found in Table 2.

Table 2 Verbal reports of pain prior to treatment
Full size table

Information on previous medications used for NP was found for 32 of the 38 patients. On average, each patient took around four medications, and this ranged from one to eight. Table 3 contains a list of the types of medications taken by the patients.

Table 3 Previous medications used for NP
Full size table

The initial dose of gabapentin was 900 mg per day, usually as 300 mg three times a day. The median maintenance dose was 2400 mg daily, usually as 800 mg three times a day or 600 mg four times a day. The range was 900 mg to 4800 mg daily.

Based on visual analogue scores (VAS) and patients' verbal descriptions, 29 of the 38 patients (76%) had some improvement from gabapentin therapy. In total, there were eight reports of adverse effects. The majority of complaints were drowsiness, especially initially, dizziness, and somnolence. Nine patients ceased therapy – four of these because of adverse effects, in particular dizziness and somnolence. In the other five patients, gabapentin was deemed ineffective. In all nine cases, therapy was ceased within the first month (Table 4).

Table 4 Cessation of therapy
Full size table

Twenty-eight of the 38 patients originally identified had baseline VAS. Nineteen of these had scores at 1 month, 14 had scores at 3 months, and 16 had scores at 6 months of therapy. Of the 28 patients, 22 had more than one VAS (Table 5).

Table 5 Number of patients with VAS
Full size table

Prior to treatment, the mean baseline VAS for the 28 patients was 8.88. At 1 month of therapy, the mean score for 19 of these 28 patients was 5.47. From a sample of 14 and 16 patients, respectively, the mean score continued to drop at 3 and 6 months, although not as dramatically (Table 6). Eleven patients had data at all four measurement times and this data was analyzed using a one-way repeated measure analysis of variance (ANOVA). The mean pretreatment VAS was 8.86. At 1, 3 and 6 months of therapy, the score dropped to 5.23, 4.59 and 4.13, respectively. A statistically significant difference was found between times (F3,30=24.92, P<0.001) with a significant curvilinear trend (F1,10=23.63, P=0.001). Contrast testing indicated that the major difference lay between baseline and follow-up times with no significant differences apparent between 1, 3 and 6 months (Figure 2).

Table 6 Mean VAS over the four times of measurement
Full size table
Figure 2
figure 2

Visual analogue scores

Full size image

Apparent improvements were also noted with the verbal reports of pain following therapy with gabapentin. Comments such as ‘life/pain is liveable’ and ‘increase in quality of life’ were described by a number of patients. Table 7 lists some of the documented verbal reports relating to NP following treatment with gabapentin.

Table 7 Verbal reports following treatment with gabapentin
Full size table

Discussion

These results support the effectiveness of gabapentin for NP following SCI, especially when taking into consideration that many of these patients had trialled a number of other medications with limited success. The results also indicate that effectiveness was evident within the first month of therapy. In the nine patients where therapy was deemed ineffective, gabapentin was ceased within the first month. Most improvement occurred in the first month of treatment and there was a marginal continued improvement at 3 and 6 months of therapy.

Limitations of the results stem from the uncontrolled and non-standardised manner in which VAS and pain observations were made and obtained. In addition, not every patient had a complete set (four) of VAS. This was difficult to achieve given that it was not a formal clinical trial. Considerations that may confound results include whether NP is chronic or acute (as chronic NP may be even more difficult to treat). The type and location of NP may be an important factor, as may the level of SCI, and also whether the lesion is a complete or incomplete injury. Co-medications in particular can confound results, however, it was very difficult to retrospectively extract this information primarily due to the majority of these patients being out-patients and had their other medications attended to elsewhere.

In conclusion, although clinical controlled trials are required to confirm these results, our experience suggests that gabapentin offers an effective therapeutic alternative for the alleviation of NP following SCI.