OUP user menu

Efficacy and Safety of Intravenous Immunoglobulin as Adjuvant Treatment for Refractory Neuropathic Pain. Results of an Open-Label, Multicenter Study

Stefano Jann MD, Ada Francia MD, Maria E. Fruguglietti MD, Luisa De Toni Franceschini MD, Roberto Sterzi MD
DOI: http://dx.doi.org/10.1111/j.1526-4637.2012.01478.x 1334-1341 First published online: 1 October 2012


Introduction. Neuropathic pain is frequently associated with many peripheral nervous system diseases and its successful treatment is an area of significant and critical unmet need.

Methods. Twenty adult outpatients of both sexes who had been suffering from painful polyneuropathy resistant to conventional therapies for at least 6 months and up to a maximum of 5 years and who reported severity of pain >60 units on a visual analog scale (VAS) at baseline were included in this open-label pilot study. Patients were randomly 1:1 allocated to receive adjuvant intravenous immunoglobulin (IVIG) (Flebogamma®, 2 g/kg) in addition to their regular therapy or to continue with the previous therapy (control group).

Results. The mean value of pain intensity (VAS) in the IVIG group dropped from 88 at baseline to 49 after the first week, and to 28 after 4 weeks, while values in the control group only slightly changed, from 85 to 78 after 1 week and to 75 after 4 weeks (P < 0.01). Almost 100% of patients reported strong/medium pain (Short Form McGill Pain Questionnaire) in both groups at baseline, while after 4–8 weeks, pain was reduced to moderate/light in 90% of patients in the IVIG group, whereas no improvement was reported in the control group (P < 0.01). In patients' quality of life, scores of the IVIG group (Short Form 36, Clinical Global Impression of Change, and Patient Global Impression of Change questionnaires) in all the follow-up were significantly higher than those of the control group (P < 0.01).

Conclusion. This unblinded pilot study showed a beneficial effect of IVIG on neuropathic pain intensity and quality of life in patients resistant to conventional treatments.

  • Intravenous Immunoglobulin
  • Neuropathic Pain
  • Peripheral Neuropathy
  • Refractory Pain
  • Randomized Trial


Neuropathic pain is defined as pain caused by a lesion or disease of the somatosensory system [1]. Neuropathic pain occurs in conditions such as diabetes, acquired immunodeficiency disease, physical injury, tumors, toxins, and vascular disorders [2]. The immune system has been described to play a major role in the manifestation of pain, with immune cells such as mast cells, neutrophils, macrophages, and T lymphocytes being involved in neuronal modulation of pain through release of mediators of inflammation [2,3]. In addition, immune cells have been implicated in the alteration of nociceptive processing characteristic of peripheral neuropathic pain [4].

Recently, some articles using peripheral blood mononuclear cell have described the overexpression of tumor necrosis factor (TNF) alpha and other pro-inflammatory cytokines in human painful neuropathies [5,6]. Thus, systemic interleukin (IL)-2 and TNF gene expression was about twofold higher in patients with painful neuropathy, compared with healthy control subjects. Conversely, gene expression of the anti-inflammatory cytokines IL-4 and IL-10 in those patients was reduced. Interestingly, these findings were independent of the underlying etiology, particularly with regard to inflammation.

Neuropathic pain is a daily challenge for many clinicians. Pharmacological treatment can result in a reduction of pain intensity in some patients, but this reduction can be insufficient and many patients have no relief at all [7,8]. Neuropathic pain is often managed with conventional symptomatic treatment including inappropriate use of nonsteroidal anti-inflammatory drugs (NSAIDs) as well as with drugs which can significantly compromise the normal activity of patients such as tricyclic antidepressant and anticonvulsant drugs [9,10]. Chronic neuropathic pain is known to be a persistent and exhausting pain that greatly impacts people's lives, making it difficult for them to perform daily tasks, it can affect a person's ability to work, and it can hinder sleep at night [11].

There is clinical evidence that intravenous immunoglobulins (IVIGs), an immune-modulating blood-derived product, can be effective in the treatment of pain associated with some neurological diseases such as Sjögren's syndrome, Guillain–Barré syndrome, complex regional pain syndrome [12], and chronic inflammatory demyelinating polyneuropathy [13–15]. The anti-inflammatory effect of IVIG has been the subject of considerable investigation. IVIG putative anti-inflammatory mechanisms include blockade of the Fc receptor, enhancement of antibody catabolism, and the suppression of pro-inflammatory cytokines [16]. With this background, it seems reasonable to consider a possible beneficial effect of IVIG in neuropathic pain and therefore clinical studies following this line of rationale are warranted. This is the first study where a highly purified, unmodified human IVIG was tested for pain intensity reduction and quality of life improvement in a cohort of patients with painful polyneuropathy resistant to conventional therapies.

Material and Methods

Aim and Study Design

An open-label, randomized, multicenter, phase II, pilot clinical study was conducted to assess the efficacy and safety of high-dose IVIG (Flebogamma®, Instituto Grifols, Barcelona, Spain) for the treatment of resistant neuropathic pain. The primary efficacy objective of the study was to assess the reduction of pain intensity after IVIG treatment, in addition to patient's regular drug therapy, according to a standard measurement scale. The secondary efficacy objective was to assess the impact of IVIG treatment on the patient's quality of life, as measured using standard questionnaires. Safety objective was the evaluation of adverse events (AEs) during IVIG treatment.

This investigator-initiated study was conducted under the principles of Good Clinical Practice and the Declaration of Helsinki (1964 and subsequent amendments). Two neurology centers in Italy participated in the study. The protocol was approved by the competent health authorities and by respective Ethics Committees. Patients' written informed consent was obtained prior to their participation into the study.

Recruited patients who met all inclusion criteria and no exclusion criteria were required to participate in four visits. After enrollment, patients were randomly allocated in a 1:1 proportion to receive IVIG in addition to their regular drug therapy or to continue with their previous therapy at their regular basis (control group). A centralized simple randomization was performed to prevent bias in allocation of patients in any of the two arms of treatment. Prior to the beginning of the study, a computer-generated randomization list was generated by an independent statistician who was no longer involved in the study. The investigators were blinded to the randomized sequence. Block randomization was used to balance the quantitative asymmetry of patients assigned to the groups both during and after enrollment. The investigators were not informed about the size of the blocks.

Patients started treatment at the first visit (basal) and then were followed up according to the following schedule: second visit at the end of IVIG treatment period (5 days); third visit at 4 weeks after the end of treatment; and fourth visit at 8 weeks after the end of treatment. Pain intensity and quality of life assessments were performed at every visit. Global clinical condition was assessed at third and fourth visits. AEs were monitored from visit 2 to visit 4. All data from the study were recorded by the investigators in electronic Case Report Forms.

Study Population

Subjects were recruited by physicians from those attending the neurology units at the Niguarda Hospital (Milan, Italy) and the University La Sapienza (Rome, Italy), where they were outpatients. Inclusion criteria for the study were the following: Adult patients of both sexes who have been suffering from painful polyneuropathy according to previously published definition criteria [17], resistant to conventional therapy for at least 6 months and up to a maximum of 5 years, and who reported baseline severity of pain >60 units (mm) on a visual analog scale (VAS) [18] at enrollment. All patients underwent a complete neurophysiological examination to exclude chronic inflammatory polyneuropathy (CIDP) and to confirm a polyneuropathy. Patients were defined treatment resistant if pain relief is <30% after treatment with NSAIDs, tricyclic antidepressant, antiepileptic, or opioid drugs at therapeutic dosages.

Exclusion criteria included the presence of other type of pain which may interact with the proper evaluation of painful polyneuropathy; having a serum IgA level <70 mg/dL; having untreated cardiovascular pathology, renal or hepatic failure, severe hematological alterations or a psychiatric disorder; having amputations; being a participant of other clinical study in the 30 days previous to the present study; and having any other condition that, at the physician judgment, could advise against participation of the patient in the study.

Study Product and Dose

Flebogamma® (Instituto Grifols) is a liquid IVIG solution obtained from human plasma. Viral inactivation is accomplished by pasteurization, which has been proven effective in inactivating both enveloped and nonenveloped viruses, and by 8% polyethyleneglycol precipitation step. The finished product is a highly pure product containing 5 g human normal immunoglobulin and 5 g D-sorbitol in 100 mL of water for injection.

IVIG group patients received 0.4 g/kg/day of Flebogamma® for 5 consecutive days (total dose 2 g/kg) in addition to their regular drug therapy. This dose of IVIG is considered adequate for indications requiring immunomodulation [19].

Efficacy and Safety Assessments

Pain intensity was the primary end point, measured at every visit using the VAS as a standard tool for quantitatively rating of pain from 0 mm (no pain) to 100 mm (maximal pain). Taking into consideration that, as this pilot study was not blinded, a reduction in pain intensity of 20% in the in the group of patients treated with IVIG could be attributable to conditioning or placebo-like effect [20], as well as that resistance to conventional treatment was defined if pain relief was <30%, it was then believed that a real effect of IVIG treatment would be evidenced by a reduction of 50% in pain intensity. Therefore, for this study, it was hypothesized that pain intensity rate would drop down from 80 mm (±10 mm in range) expected at recruitment to 40 mm (±20 mm in range) after 5 days in the group of patients treated with IVIG, while rate would remain similar to basal in the control group. Additionally to VAS, the patient's pain experience was identified as secondary variable by means of the Short Form McGill Pain Questionnaire (SF-MPQ) [21] according to a qualitative pain intensity scale with the following range: unbearable, strong, medium, moderate, light, and no pain.

For assessment of other secondary end points, quality of life was evaluated by means of standard questionnaires. The Short Form 36 (SF-36) questionnaire [22] was employed at every visit for rating both physical and mental health status with scores in a range 0–100. Global clinical condition was rated at visits 3 and 4 by means of the Clinical Global Impression of Change (CGI-C) and the Patient Global Impression of Change (PGI-C) [23]. These tests included a qualitative health improvement scale with the following range: much better, moderately better, slightly better, no change, slightly worse, moderately worse, and much worse.

For safety assessment, all AEs occurring to a participating subject from the first IVIG infusion to the end of the study period, whether communicated spontaneously by the patient or observed by the investigator, were registered by the investigator from visit 2 (end of therapy period) to visit 4 (end of study). An AE was defined as an undesirable experience occurring to an individual following administration of a medicinal product, regardless whether it was related or not to the product [24]. AEs were categorized as expected or unexpected and graded according to their intensity (mild, moderate, and severe) and seriousness (serious or nonserious). Finally, the cause–effect relationship between the study product and the AE was classified in five categories: definite, probable, possible, conditional, and doubtful [24].

Statistical Analysis

All patients who fulfilled the inclusion criteria and none of the exclusion criteria, treated with at least one infusion of IVIG, and with at least one follow-up efficacy assessment were included in the efficacy analysis. All AEs reported were included in the safety analysis.

For quantitative variables, mean, standard deviation of the mean, minimum and maximum, and 95% confidence interval of the mean were used. For qualitative variables, absolute or relative frequency in percentage is presented. Primary and secondary parameters were analyzed following a covariance analysis model, being the basal values of analyzed parameters as the covariate, and being parameters after 5 days of treatment as the dependent variable. The arm with treatment was the grouping factor. A second analysis was performed following the repeated-measures analysis of variance, being the treatment as the grouping factor and the basal values, after 5 day, after 1 month, and after 2 month, as the repeated measures (independent variables). Multiple comparisons intragroup and between groups were evaluated respective of basal values and after 5 days by using the Bonferroni test. Pain intensity measured by SF-MPQ between groups was analyzed by means of Friedman's test, and the Mann–Whitney's U-test was employed in intragroup evaluation of relief time. The general clinical condition was analyzed by means of a linear trend test. Incidence of AEs was evaluated by means of the Fisher's exact t-test. Statistical significance was set at P ≤ 0.05. Analysis was performed using the statistical package BMDP Dynamic version 2009 (BMDP Statistical software; Los Angeles, CA, USA).


Population Characteristics

A total of 32 subjects were assessed for eligibility of which eight did not meet inclusion criteria and four declined to participate in the study. The 20 patients who met the inclusion criteria were enrolled and randomized 1:1 in the control (conventional therapy) and the IVIG groups. One patient ascribed to the control group gave up before the first basal visit, so 19 patients started the study. One patient in the IVIG treatment group dropped out after visit 1 because of an AE, although apparently not related to the study product, as described later in the safety results. Finally, 18 patients ended the study. Figure 1 shows the flow of subjects through the study. Also in the IVIG treatment group, one patient showed inconsistent pain severity results at visit 1 (VAS pain intensity of 36 mm) despite being enrolled (inclusion criteria required >60 mm of pain intensity). Therefore, it was considered a violation of inclusion criteria and the patient was excluded for efficacy analysis. The patient, however, was not excluded from the safety assessments.

Figure 1

Flow of patients through the study. IVIG = intravenous immunoglobulin.

Of the population starting the study, 13 patients were men (eight in the IVIG group and five in the control group) and six were women (two in the IVIG group and four in the control group). Average age was 66.5 ± 7.5 years (66.3 ± 5.6 in the IVIG group and 66.8 ± 9.6 in the control group; total range 48–81). There were no statistical differences between the two groups according to age and sex. IgA level in the control and IVIG group were >140 mg/dL and >160 mg/dL, respectively). Patients' diseases associated with neuropathic pain are shown in Table 1.

View this table:
Table 1

Patients' diseases associated with neuropathic pain

IVIG GroupControl Group
Diabetic neuropathy57
Chemotherapy-induced painful neuropathy11
Cryoglobulinemic (HCV-related) painful neuropathy11
Idiopathic painful polyneuropathy30
  • HCV = hepatitis C virus; IVIG = intravenous immunoglobulin.

Conventional therapy followed by patients in the control group was as follows: six patients took anticonvulsant drugs (four took pregabalin, one took gabapentin, and one took oxcarbazepine); one patient took a combination of anticonvulsant drug (gabapentin) and antidepressant (duloxetine); one patient took opioid analgesic (oxycodone); and one patient took a combination of analgesic opioid (oxycodone) and acetaminophen. All patients followed their regular prescribed dosage.

Conventional therapy followed by patients in the IVIG group was as follows: seven patients took antiepileptic drugs (six took pregabalin and one oxcarbazepine), two patients took a combination of pregabalin and duloxetine, and one patient took a combination of pregabalin, duloxetine, and oxycodone. None of these patients modified this therapy during the study.

Primary Efficacy End Point

At visit 1 (basal), pain intensity evaluated according to VAS was similar in both the IVIG-treated group and the conventional treatment group (88.0 ± 13.2 mm ranging 61–100, and 85.0 ± 11.5 mm ranging 67–100, respectively). Figure 2 shows the progression of VAS values during the study in both treatment groups. After 5 days of treatment (visit 2), values in the IVIG group significantly dropped to 49.6 ± 13.0 mm (P < 0.01), but were only slightly lower in the conventional treatment group (78.3 ± 8.5 mm). At visit 3, values continued dropping to 28.8 ± 15.2 mm in the IVIG group and remained in the same range at visit 4 (both P < 0.01 with respect to visits 1 and 2). By contrast, in the conventional treatment group, VAS values measured at visits 3 and 4 were very similar to that of visit 2. Considering the comparison between the two groups, the reduction of VAS values along visits in the IVIG group was significantly lower than that of the conventional treatment group (P < 0.01).

Figure 2

Pain intensity evaluated according to a visual analog scale (VAS) at visit 1 (baseline), visit 2 (5 days), visit 3 (30 days), and visit 4 (60 days) in patients treated with intravenous immunoglobulin (IVIG; N = 8) or with conventional therapy (C.T.; N = 9). * P < 0.01 in visits 2, 3, and 4 vs baseline, in visits 3 and 4 vs visit 2, and in IVIG vs C.T.

Regarding evaluation of patient's pain experience by means of the SF-MPQ, at basal visit, most patients reported strong or medium pain in both the IVIG and the conventional therapy group. As shown in Figure 3, at subsequent visits, the percentage of patients reporting moderate or light pain in the IVIG group progressively increased, reaching statistical significance in visits 3 and 4 with respect to basal values (P < 0.01). By contrast, no improvement was observed in the conventional therapy group. Comparison of treatments showed statistically significant differences at all the follow-up visits in favor of the IVIG group (P < 0.01).

Figure 3

Pain experience evaluated with the Short Form McGill Pain Questionnaire (SF-MPQ) at visit 1 (baseline), visit 2 (5 days), visit 3 (30 days), and visit 4 (60 days) in patients treated with intravenous immunoglobulin (IVIG; N = 8) or with conventional therapy (C.T.; N = 9). * P < 0.01 in visits 3 and 4 vs baseline; § P < 0.01 in IVIG vs C.T.

Secondary Efficacy End Points

Results of the SF-36 questionnaire showed that basal scores were similar for both the IVIG group and the conventional therapy group (25.63 ± 4.24 and 28.67 ± 4.44, respectively regarding physical health, and 20.88 ± 5.96 and 21.78 ± 3.49, respectively, regarding mental health). During the follow-up, patients of the IVIG group scored significantly higher in all visits with respect to basal values in both physical and mental health status (P < 0.01), up to doubling scores at visit 3. Conversely, scores of patients in the conventional therapy group remained always almost identical along the study. Moreover, comparison of groups showed that scores of the IVIG group in all the follow-up visits were significantly higher than those of the conventional therapy group (P < 0.01). These data are summarized in Figure 4.

Figure 4

Quality of life evaluated with the Short Form 36 (SF-36) questionnaire at visit 1 (baseline), visit 2 (5 days), visit 3 (30 days), and visit 4 (60 days) in patients treated with intravenous immunoglobulin (IVIG; N = 8) or with conventional therapy (C.T.; N = 9). * P < 0.01 in visits 2, 3, and 4 vs baseline and in IVIG vs C.T.

With regard to rating results of CGI-C and PGI-C tests, there was a concordance of both physician and patient opinion in considering a positive trend to improvement of the global clinical condition associated with the IVIG treatment. As seen in Figures 5 and 6, 100% of patients in the IVIG group were rated to be in a better condition in both CGI-C and PGI-C tests, whereas most patients in the conventional therapy group felt no change or being worse. Results were statistically significant at both visits 3 and 4 (P < 0.01).

Figure 5

Clinical condition evaluated with the Clinical Global Impression of Change (CGI-C) at visit 3 (30 days) and visit 4 (60 days) in patients treated with intravenous immunoglobulin (IVIG; N = 8) or with conventional therapy (C.T.; N = 9). * P < 0.01 in IVIG vs C.T.

Figure 6

Clinical condition evaluated with the Patient Global Impression of Change (PGI-C) at visit 3 (30 days) and visit 4 (60 days) in patients treated with intravenous immunoglobulin (IVIG; N = 8) or with conventional therapy (C.T.; N = 9). * P < 0.01 in IVIG vs C.T.


One patient in the IVIG group reported a hypertensive peak event after 4 days of treatment. Although the AE was considered mild, not serious, and apparently not related to the study product (previously, the patient complained many other similar episodes), the patient voluntarily dropped out from the study. No other AE was registered.


IVIG has been used with success in the treatment of pain associated with neurological diseases and with neuropathies of diverse etiology [12,15,25–27], including diabetic neuropathy [14,28] possibly through a reduction of neuroinflammation, a condition that is present in neuropathic pain [16]. This, being the first multicenter, prospective trial in which IVIG (Flebogamma®) was used for the treatment of patients with painful polyneuropathy resistant to conventional therapies, the study was not considered for a blinded and placebo-controlled design, although it was strengthen by randomization into IVIG group and no intervention group. Under these conditions, results showed that IVIG at the given dose significantly reduced the intensity of pain in the study period, concomitant to a positive impact in the improvement of patient's quality of life and global clinical condition.

Regarding primary efficacy end point, there was no pain reduction in the control group, as it was hypothesized as patients were refractory to conventional therapy. In the group of patients treated with IVIG, mean pain intensity rate as measured by VAS dropped down to around 50 mm after 5 days. This fell inside the 40 ± 20 mm margin that was hypothesized as the primary objective. Furthermore, the reduction of pain rate was even more apparent in the following visits up to the end of the study after 8 weeks. As expected, in the control group, pain rate remained similar to baseline values. The observed pain relief results are in agreement with those from other studies on painful conditions treated with IVIG at similar doses in which VAS was employed [15,29]. It is possible, however, that similar effects could be achieved with lower doses such as those that are effective in the treatment of pain associated with other neurological diseases [12,13]. As a complement to VAS studies, in this report, the SF-MPQ was employed for the clinical assessment of pain [30]. Our SF-MPQ results confirmed and correlated well to those obtained by VAS, although they did not add additional information about patients' neuropathic pain experience.

Secondary efficacy end points were related to changes in patient's quality of life. Only patients treated with IVIG reported a marked improvement in both physical and mental health as detected by the SF-36 questionnaire. Moreover, the SF-36 scores associated with health improvement consistently followed a pattern similar to that observed in the reduction of pain intensity. Results of the global clinical condition rating through CGI-C and PGI-C tests in the third and fourth visits confirmed that patient's quality of life was significantly improved, with a high degree of concordance between patient's and physician's impression.

The good safety profile of IVIG is already supported by clinical literature, including those patients with neuroimmunological disorders [31,32]. In the present study, only one AE of minor importance and not related to the study product was observed, and study withdrawal was not a physician's advice but a patient's decision. Although AEs, such as headache or nausea are relatively common to IVIG administration, previous unblinded studies using IVIG for pain treatment reported AEs rates in accordance with ours. Hence, Tamburin and Zanette [29] observed no AEs in five IVIG-treated patients with diabetic lumbosacral radiculoplexus neuropathy, while Goebel [27] reported AEs in 19% of 130 patients with chronic pain syndrome, both studies with a dose regime similar to ours. Therefore, in our unblinded study with 10 patients, it was plausible to expect one AE in one subject. Nevertheless, the relatively small sample size of our pilot study does not allow drawing definite conclusions on the safety of IVIG in this kind of patients.

Statistically significant changes from baseline were seen consistently in primary and secondary efficacy outcome measures at all visits. However, several limitations should be considered when interpreting study results. Firstly, this was a no-placebo, open study and neither the investigators nor patients were blinded and therefore resulted in favoring better outcomes evaluation in the IVIG-treated patient group. Patients in the IVIG group would have a positive attitude toward the new treatment and, similarly, a potential negative attitude of patients being randomized to no intervention at all might be possible. Both attitudes were a potential source of bias. For instance, it should be noted both the near absence of any pain relief in the control group, which is in contrast with a recent report [33], and the apparent high percentage of responders in comparison with other reports [12,27]. The unblinded nature of our study and its associated influence on the attitude of patients could explain these supposed discrepancies. The second limitation is related to generalization of results. As it was a limited cohort study, results can not be generalized to all individuals with refractory neuropathic pain. The wide group of pathologies with associated neuropathic pain in our sample of patients partially helped to reduce this limitation. Another possible criticism is that the two arms of this study are not matched for the peripheral nerve disease. It was not done because the aim of the study was to reduce neuropathic pain and not to treat painful neuropathies. The original idea was that IVIG can be a symptomatic therapy and not only a disease-modifying treatment. Lastly, our follow-up period was 8 weeks, which only allowed short-term outcome assessments.

In summary, although definitive conclusions cannot be drawn because of the pilot nature of this study, IVIG had a significant beneficial impact on neuropathic pain severity in patients resistant to conventional treatments. Moreover, patients significantly improved their quality of life. Significant side effects were not observed along the follow-up. We are working to develop a multicenter, double-blind, placebo-controlled study.


The authors are grateful to Grifols for providing the medicinal product and independent statistician services. Dr. Jordi Bozzo at Grifols is acknowledged for his writing assistance for the preparation of this manuscript, under the direction of the authors.


View Abstract