当前位置: 首页 > 期刊 > 《小儿科》 > 2005年第7期 > 正文
编号:11327351
Intravenously Administered Immunoglobulin in the Treatment of Childhood Guillain-Barré Syndrome: A Randomized Trial
http://www.100md.com 《小儿科》
     Division of Neuropediatrics and Muscular Disorders, Department of Pediatrics and Adolescent Medicine

    Institute of Medical Biometry and Informatics, University Hospital Freiburg, Freiburg, Germany

    ABSTRACT

    Objective. To determine the optimal treatment for childhood Guillain-Barré syndrome (GBS).

    Methods. We performed a randomized, multicenter study of GBS according to international diagnostic criteria. In study 1 (early treatment), children able to walk unaided for 5 meters were randomized for 1 g/kg intravenously administered immunoglobulin (IVIG) over 2 days or no treatment. The primary outcome measure was the degree of disability at nadir. In study 2 (treatment for severe GBS), children unable to walk 5 meters unaided were randomized for 1 g/kg IVIG over 2 days or 0.4 g/kg IVIG over 5 days. The primary outcome measure was the number of days needed to regain the ability to walk unaided. Children randomized for no treatment in study 1 could enter study 2 if loss of unaided walking occurred.

    Results. Ninety-five children with GBS were registered in 40 months. Twenty-one children were randomized in study 1 and 51 in study 2 (5 after deterioration in study 1). Twenty-eight children were not randomized for various reasons. Eleven of 21 patients in study 1 lost the ability to walk unassisted and 6 were bedridden, with no statistically significant difference between the children initially randomized for treatment versus no treatment. Recovery occurred faster in the group randomized for early treatment. In study 2, recovery did not differ significantly between the children treated for 2 days versus 5 days (median time to unaided walking: 19 days vs 13 days). Secondary transient deterioration in the disability score occurred more frequently in the group with the 2-day regimen than in the group treated for 5 days (5 of 23 patients vs 0 of 23 patients). Multivariate analysis with Cox regression showed that disease severity at the nadir was the only prognostic factor for recovery.

    Conclusions. Treatment with IVIG before loss of unaided walking did not give rise to a less severe course, but recovery occurred somewhat faster. However, given the small number of patients, the power of this conclusion is low. For treatment after loss of unaided walking, there was no significant difference in the effectiveness of 2 g/kg IVIG administered over 2 days versus 5 days. Early "relapses" occurred more frequently after the shorter treatment regimen.

    Key Words: Guillain-Barré syndrome immunoglobulins childhood disability

    Abbreviations: GBS, Guillain-Barré syndrome IVIG, intravenously administered immunoglobulin PE, plasma exchange CSF, cerebrospinal fluid CI, confidence interval ITT, intention-to-treat

    Guillain-Barré syndrome (GBS) is an acute, immune-mediated polyradiculoneuropathy that occurs less frequently among children than among adults. In 70% to 80% of cases, an acute infection is reported 3 to 6 weeks before the neurologic symptoms. This is thought to give rise to an abnormal immune stimulation. In the course of the disease, symmetric ascending paralysis leads frequently to the inability to walk and even to respiratory insufficiency. Research diagnostic criteria have been defined to permit comparative etiologic and therapeutic investigations.1–4

    For adults with severe GBS, plasma exchange (PE) is considered, on the basis of large placebo-controlled studies, the standard treatment, shortening the time required to discontinue ventilator use and to regain the ability to walk.5–7 Treatment with high-dose, intravenously administered immunoglobulin (IVIG) has also been shown, with substantial evidence, to shorten the time of recovery, similar to PE.8–11 Studies among children addressing the treatment of GBS with PE and IVIG demonstrated results similar to those among adults.12–17 However, those studies examined only small numbers of patients, usually with historical control subjects. Because of its lesser invasiveness and rarer complications, IVIG is preferred to PE by most authors for treating children.

    The optimal time of treatment and the best treatment schedule remain unknown. Although no studies have been performed on this topic, some physicians administer IVIG to children who are still able to walk in the hope of preventing a severe course of the disease. Furthermore, 3 case series18–20 and 1 study that compared retrospectively the treatment results for 2 hospitals21 gave rise to the opinion that the therapeutic effect of 2 g/kg body weight IVIG administered in 1 or 2 days might be superior to the usual schedule of 2 g/kg body weight distributed over 5 days.

    We performed a randomized, prospective, multicenter study addressing the following questions. Before the loss of unaided walking, can early treatment with IVIG reduce the subsequent severity of the disease After the loss of unaided walking, does treatment with 1 g/kg body weight IVIG over 2 days result in faster recovery of unaided walking than treatment with 0.4 g/kg body weight IVIG over 5 days

    METHODS

    Patients

    German, Swiss, and Austrian pediatric hospitals were invited through letters and through announcements at scientific meetings and on the website of the Gesellschaft für Neuropdiatrie to contribute patients to this study. Ninety hospitals expressed interest and received the study documents. Each hospital received 2 prepared randomization envelopes, for patients still able or not able to walk unaided. The envelopes were numbered and needed to be opened in that order. Treatment was to be initiated as soon as possible after the randomization envelope was opened. The treatment performance was open to physicians and patients. We planned no sham treatment because we considered the outcome criteria to be relatively robust. During the recruitment period of 40 months, 63 hospitals entered 1 patient into the study.

    Children and adolescents from the age of independent walking to 18 years were eligible. Inclusion and exclusion criteria were based on the published international research diagnostic criteria,2 including the results of cerebrospinal fluid (CSF) examinations. Testing of nerve conduction velocity (F-wave and electromyographic testing) and spinal magnetic resonance imaging were not mandatory, because many pediatric hospitals were not able to perform these investigations within a short time. Because of the intention to treat promptly and the well-known delay in the appearance of typical CSF and electrophysiologic findings, frequently patients were randomized and treated on the basis of the clinical suspicion of GBS only. If the diagnosis later proved to be incorrect, then the patient was removed from the study. An additional inclusion criterion was the informed consent of the parents and, as far as possible according to age, the patient. Additional exclusion criteria were known IgA deficiency, allergy to blood preparations, and severe cardiovascular diseases. The study was approved by the ethics committee of Freiburg University (register no. 140/98).

    Study Design

    Children who were still able to walk without aid for a distance of 5 m at inclusion in the study were to be randomized for 1 g/kg body weight IVIG over 2 days versus no immune treatment (early treatment study). The primary outcome measure for this group was the subsequent maximal disease severity.

    Children who were unable to walk 5 m unaided at inclusion in the study were to be randomized for 2 g/kg body weight IVIG over 2 days (1 g/kg body weight per day) versus 2 g/kg body weight IVIG over 5 days (0.4 g/kg body weight per day) (late treatment study). The primary outcome measure for this group was days from randomization needed to regain the ability to walk unaided. Children randomized for no treatment in the early treatment group could enter the late treatment group and be randomized again if loss of independent walking occurred.

    The IVIG brand was to be chosen individually according to availability in the local hospital pharmacies. At the time of the study, 9 different brands were available in the 3 countries, 3 of which were used for 63% of the patients.

    In both studies, as the main outcome measure, the functional ability of the patient during the course of the disease was scored with an ordinal disability score very similar to that used in published studies on GBS treatment (Table 1). Additional scores were used to monitor arm function, cranial nerve function, respiratory function, vegetative symptoms, and pain (Table 2). These scores were to be recorded prospectively on a standardized form on a daily schedule during the hospital stay or until independent walking was regained and at each outpatient appointment thereafter. If the scores changed between outpatient visits, then the date of this change was to be recorded as exactly as possible.

    To monitor side effects of IVIG treatment, daily examinations of vital signs and blood pressure and weekly examinations of blood counts, creatinine levels, and liver function, as well as urinalysis, were to be performed until 4 weeks after IVIG treatment. The liver function tests were to be repeated after 3 and 6 months. Accompanying this treatment study, a study was conducted on the cause and pathogenesis of GBS among children, including prospective testing for >30 infective agents, HLA alleles, and antiganglioside antibodies. Results of these investigations will be reported separately.

    Statistical Methods

    Calculation of case numbers before the start of the study showed that the number of patients needed in both parts of the study to reach a sufficient power of 0.8 with an of .05 would be much higher than could be obtained in a reasonable time. Therefore, the study was planned as an explorative study and not a confirmative trial. Statistical evaluation was performed with the program package SPSS, version 11.0.1 (SPSS, Chicago, IL). After exclusion of cases with incorrect diagnoses (see below), evaluation was made on an intention-to-treat (ITT) basis. Because their initial clinical findings were indistinguishable from those of acute GBS, patients with chronic inflammatory demyelinating polyneuropathy were included in the ITT analysis. Univariate comparisons of nominal data were performed with the 2 test and those of ordinal and numerical data were performed with the Mann-Whitney test or through Kaplan-Meier analysis with the log-rank test. Multivariate survival analysis investigating the time measures of recovery was performed with Cox regression analysis with conditional backward elimination.

    RESULTS

    Patients

    During a period of 40 months (January 1999 through March 2002), 101 children were reported to the study. Five children were excluded because their disease was determined not to be polyradiculoneuritis (3 cases of transverse myelitis, 1 case of intraspinal bleeding, and 1 case of dissociative movement disorder). One child left the hospital very early and no data could be obtained. Therefore, 95 children fulfilling the diagnostic criteria for GBS could be included.

    Of the 95 children classified as having GBS, 53 were male and 42 were female. Their ages ranged from 12 months to 16.5 years, with a peak at 5 to 6 years (median: 6.2 years; 95% confidence interval [CI]: 1.4–15.9 years). At inclusion in the study, 26 children were able to walk without aid and 69 were not. Twenty-one children were randomized in the early treatment study, 5 of whom experienced deterioration and were randomized for a second time. Fifty-three children were randomized in the late treatment study, 2 of whom had to be excluded (1 because of withdrawal of parental consent and 1 because of insufficient data). Therefore, a total of 72 cases were treated according to the protocol. Because of logistic reasons, 28 additional patients could not be randomized (initiation of treatment before inclusion and randomization or no parental consent for randomization).

    Early Treatment Study

    Twenty-one children were randomized in the early treatment study. Unfortunately, fewer patients were randomized for no treatment than for treatment (7 for no treatment and 14 for treatment), most likely because of the small case number. One patient who was already unable to walk without help was randomized erroneously in this study; he was randomized for treatment, and we decided to include him in the ITT analysis. In the group randomized for no treatment, the median time from first symptoms to randomization was 9 days (95% CI: 4–22 days). In the group randomized for treatment, the median time from first symptoms to randomization was 7 days (95% CI: 2–19 days), and the median time from first symptoms to treatment was 7.5 days (95% CI: 2–19 days). The condition of 1 child randomized for IVIG deteriorated before treatment could be initiated. For 4 of the 7 patients randomized for no treatment, the ability for unaided walking was lost. In accordance with the protocol, the patients were randomized again and treated in the late treatment study. For these patients, the median time from first symptoms to IVIG treatment was 14 days (95% CI: 5–38 days). One of the patients in the later course was found to have relapsing, chronic, inflammatory demyelinating polyneuropathy.

    At the height of the disease, 11 of the 21 patients were not able to walk unaided and 6 were bedridden. One experienced respiratory insufficiency. The primary outcome parameter of maximal disability score did not differ between the treated and untreated children (Table 3). The groups did not differ in age, prior infections, and CSF findings. The median duration of progression was 10 days (95% CI: 6–38 days) in the group randomized for no treatment and 10 days (95% CI: 2–20 days) in the group randomized for treatment. Initial signs of improvement (referring to any of the scores) occurred earlier in the group randomized for IVIG than in the group randomized for no treatment, ie, a median of 4.5 days (95% CI: 2–14 days) vs 30 days (95% CI: 6–83 days) after randomization (Mann-Whitney test, P = .001). Improvement in the main disability score also differed significantly, ie, a median of 8 days (95% CI: 2–105 days) vs 32 days (95% CI: 6–83 days; Mann-Whitney test, P = .046). The median disability score 4 weeks after randomization was significantly lower in the group initially randomized for IVIG than in the group randomized for no treatment (Table 3). Mild side effects (allergic reaction, headaches, and hypertonus, 1 case each) were observed for 3 children treated with IVIG.

    Late Treatment Study

    Fifty-one children were subjected to ITT analysis in the late treatment study. Twenty-five were randomized for the 2-day IVIG schedule (group A) and 26 for the 5-day IVIG schedule (group B). The median time from first symptoms to randomization was 6 days (95% CI: 0–18 days) in group A and 6.5 days (95% CI: 1–37 days) in group B. The median time from first symptoms to treatment was 7 days (95% CI: 0–18 days) in group A and 7 days (95% CI: 1–38 days) in group B. Treatment was usually administered at or shortly before the nadir of the disease. The median duration of disease progression was 7 days (95% CI: 1–18 days) in group A and 7 days (95% CI: 1–38 days) in group B. There was no significant difference in the disability scores between the 2 groups at inclusion in the study, at the height of the disease, or 4 weeks after randomization (Table 4). Two patients later were found to be suffering from chronic inflammatory demyelinating polyneuropathy (1 in group A and 1 in group B), 2 experienced a GBS relapse after many months (1 in group A and 1 in group B), and 1 suffered from GBS with concurrent myelitis (group B).

    The primary outcome, namely, time to regain unaided walking after randomization, did not differ significantly between the groups (median: 19 days in group A and 13 days in group B; log-rank test, P = .94) (Fig 1). The median time to improve 1 point on the disability score was 5 days in both groups. However, early transient deterioration in the disability score of 1 point occurred more frequently in group A than in the group treated over 5 days (5 of 23 patients vs 0 of 23 patients; Fisher's exact test, P = .049). The duration of the progressive phase was longer for patients who later experienced secondary deterioration (median: 13 vs 6 days; Mann-Whitney test, P = .046).

    Multivariate analysis was performed with Cox regression analysis with conditional backward elimination. Age, gender, CSF protein level, treatment protocol, maximal disability score, and the occurrence of secondary deterioration were entered as independent variables. The results showed no treatment schedule influence, but the highly significant influence of the disability score on the primary outcome measure of days to regain unaided walking was revealed. A higher disability score was correlated clearly with a slower recovery. The time needed to improve 1 point on the disability score also depended only on the maximal disease severity, with the same independent variables (Table 5).

    Side effects such as allergic reactions (n = 3), headaches and nausea (n = 3), fever (n = 2), leukopenia (n = 2), and proteinuria (n = 1) were reported for 18% and 20% of the patients in group A and group B, respectively. However, these side effects were minor and did not interfere with treatment.

    DISCUSSION

    In this prospective multicenter study on childhood GBS, we tried to include as many patients as possible, with the aim of obtaining valid data. A polio surveillance study (acute flaccid paralysis surveillance) was performed in Germany concurrently with our trial. During the same time period, 108 children with GBS were reported to the acute flaccid paralysis surveillance study, 37 of whom were also participating in our study. Fifty-eight German patients were included in our study but were not reported to the acute flaccid paralysis surveillance study (K. Beyrer, personal communication). Therefore, we are aware of 166 German children with GBS for the 40-month period. We estimate that 50% to 60% of the German childhood GBS cases were included in our study.

    It was not the aim of this study to investigate whether IVIG in childhood GBS is superior to no treatment. We are aware that, to date, this question regarding children has been addressed only in case series with historical control subjects and in one very small controlled study with a sequential design.15–17,19 The results of those studies paralleled the findings of well-designed controlled studies among adults.5–11,22 However, frequently the clinical course of childhood GBS is less severe than that of GBS among adults and usually recovery is complete.16,23 Results and treatment indications for adults thus cannot be applied directly to children. Nevertheless, the performance of a large randomized study among children with GBS, comparing IVIG with no treatment, was not feasible for both ethical and practical reasons, in light of the data available. Instead, we decided to study the emerging, but not yet scientifically investigated, use of IVIG for children still able to walk without aids and to answer the question of whether the therapeutic effect of a highly concentrated administration of IVIG is superior to the conventional dosing schedule, as inferred from some published observations.18,20,21

    The number of patients in our early treatment study with children still able to walk unaided clearly is too small for valid conclusions. Comparison is also made difficult by the skewed randomization, with only 7 children being randomized for no treatment. Eleven patients experienced additional deterioration (5 of 7 randomized for no treatment and 6 of 14 randomized for IVIG), but the difference between the patient groups was not significant. Only 6 of 21 patients lost the ability to walk completely, and 1 required ventilation. This course was much milder than could be expected from an earlier study with an unselected group of children with GBS, of whom 60% became bedridden and 20% required ventilation.16 Rather than attributing this mild course to the treatment with IVIG, we think that most of the children who were still able to walk unaided at the time of diagnosis of GBS represented the mild end of the disease severity spectrum. Several reports have addressed the mild variant of GBS. In a small case series of 12 adults, no differences were observed, compared with the more severe cases, regarding age, gender, preceding illnesses, clinical symptoms, duration of progression, or electrophysiologic findings.24 However, in a large epidemiologic survey in the Netherlands, the mild courses were associated with younger age, male gender, and shorter time to reach the nadir.4

    We found that early treatment with IVIG in mild cases did not change disease severity but did give rise to a faster onset of improvement. The French Cooperative Group25 treated adults with mild GBS (defined as being able to walk with or without aid but not to run or being able to stand up unaided) randomly with 0 or 2 PE treatments. As for our children, treatment gave rise to a faster onset of motor recovery. Fewer treated patients experienced deterioration, but the number of ventilation-treated patients and the time required to walk unassisted did not differ significantly between the treated group and the untreated group. Forty-four percent of those patients were not able to walk unaided at the time of randomization; therefore, that sample was affected more severely than were our children in the early treatment study.

    We were not able to show that early IVIG treatment could ameliorate subsequent disease severity. However, the statistical power of this finding is low because of the small patient number, and we cannot prove that early IVIG treatment does not improve the course. Many patients who are still able to walk independently at diagnosis show a mild course, without requiring intensive care, and 15 of our 21 patients were still able to walk at the nadir. Therefore, the faster onset of improvement shown for our early-treated patients and by the French group does not have the same clinical impact as findings for patients who are unable to walk or require ventilation.

    In our late treatment study, we could not demonstrate faster recovery among children treated with 2 g/kg body weight IVIG in 2 days, compared with the more protracted regimen. To our knowledge, only 1 study comparing 2 different IVIG schedules among adults with severe GBS has been published. Raphael et al26 compared treatment with 0.4 g/kg body weight IVIG for 3 days with the same daily dose administered for 6 days among adults unsuitable for PE. Those authors noted better effects with the longer schedule, particularly for severely ill and ventilation-treated patients. However, that study is unlike ours because it compared different IVIG doses, whereas the total dose administered in our study was the same in both groups.

    Rather unexpectedly, we observed significantly more early transient relapses among children treated for 2 days, compared with 5 days, although the 2 groups had received the same total IVIG dose. Those early transient relapses must not be confused with the relapses among the children with chronic inflammatory demyelinating polyneuropathy and "relapsing GBS," which occurred much later. Early transient relapses or "treatment-related fluctuations" have been observed after IVIG treatment and after PE. A frequency of 10% was reported,27 but rates can be as high as 30% to 70%.28,29 Ongoing disease activity concurrent with fading immunoglobulin concentrations has been discussed as being causative, but no influence of an earlier onset of treatment could be demonstrated.29 We noted associations between the occurrence of early relapses and not only the treatment schedule but also the duration of the previous disease progression. It could be speculated that, after the faster administration of IVIG over only 2 days, the immunoglobulins could be more rapidly cleared from the serum, giving rise to an earlier fading of action while the pathophysiologic processes of the disease are still ongoing. However, we are not able to prove this hypothesis because of the lack of serial serum immunoglobulin measurements.

    The disability score at the height of the disease was the only significant prognostic factor for recovery among our patients. Furthermore, electrophysiologic findings indicating axonal damage might have been expected to be of prognostic significance, according to the literature.30–32 Electrophysiologic investigations could not be performed for all patients in this multicenter study. However, when reported, distinct slowing in nerve conduction velocity was usually indicative of acute demyelinating inflammatory polyneuropathy. A diagnosis of acute motor axonal neuropathy was considered for only 3 children with reduced compound motor potential amplitude and normal nerve conduction velocity after disease duration of >10 days. Axonal damage has been associated with Campylobacter jejuni infections, with a worse prognosis than for GBS caused by other agents.33,34 C jejuni was found to be the cause of the disease preceding GBS for only 2 of our randomized patients (results not shown). As in this study, the Campylobacter-associated axonal type of GBS among children has been observed only rarely in Europe. It more frequently occurs in less developed countries and under poor sanitary conditions.31,32

    The IVIG side effects among our patients were mild but not infrequent. Allergic reactions, transient liver dysfunction, hypertonus, aseptic meningitis, proteinuria, renal dysfunction, and increased serum viscosity have been reported for adults and children treated for neurologic and hematologic diseases.35–38 Although more side effects might have been expected with the faster treatment regimen, the number and type of side effects did not differ between the groups in our study.

    Contrary to what we expected on the basis of the literature, our randomized study revealed that treatment with 2 g/kg body weight IVIG within 2 days did not yield better results than distributing the same total dose over 5 days. The 2 schedules were equally well tolerated, with only mild transient (but not infrequent) side effects. Early relapses occurred more frequently with the 2-day regimen than with the 5-day regimen.

    ACKNOWLEDGMENTS

    The study was supported by (in alphabetical order) Bayer Vital GmbH, Centeon Pharma GmbH, Grifols Deutschland GmbH, Immuno GmbH/Baxter Health Care, Novartis Pharma GmbH, and Octapharma GmbH.

    We thank the following colleagues who contributed patients to the study: H. Giesen, Schweinfurt, Germany; G. Jorch, Bielefeld, Germany; A. Knust, Siegen, Germany; R. Keimer, Stuttgart, Germany; R. Heyer, Hannover, Germany; A. Hler, München, Germany; A. Hübner, Dresden, Germany; D. Dralle, Giessen, Germany; M. Rister, Koblenz, Germany; B. Rodeck, Osnabrück, Germany; G. Kluger, Vogtareuth, Germany; K. Brockmann, Gttingen, Germany; H.-J. Christen, Hannover, Germany; M. Teufel, Bblingen, Germany; K. Grohmann, Berlin, Germany; T. Voit, Essen, Germany; M. Straburg, Würzburg, Germany; P. Borusiak, Wuppertal-Barmen, Germany; D. Rating, Heidelberg, Germany; U. Stephani, Kiel, Germany; M. Dohrmann, Marburg, Germany; K. John, Hamm, Germany; M. Schchtele, Karlsruhe, Germany; S. Knig, Mannheim, Germany; A. Fiedler, Regensburg, Germany; J. Penzien, Augsburg, Germany; M. Plein, Remscheid, Germany; J. Sperner, Lübeck, Germany; B. Püst, Hamburg, Germany; E.-M. Feldmann, Neunkirchen, Germany; M. Fedorczak, Lrrach, Germany; J. Oppermann, Cottbus, Germany; H. Strehl, Dortmund, Germany; H.-P. Weber, Lüdenscheid, Germany; P. Navratil, Coesfeld, Germany; H.-L. Spohr, Berlin, Germany; D.I. Zafeiriou, Thessalonike, Greece; H. Schweder, Gppingen, Germany; C. Dominick, Ludwigshafen, Germany; W. Jost, Homburg, Germany; E. Boltshauser and A. Capone-Mori, Zürich, Switzerland; R. Eichhorn, Ludwigsburg, Germany; M. Schirrmacher, Vechta, Germany; F. Tegtmeyer, Kassel, Germany; O. Weber, Basel, Switzerland; K. Mahler, Wittlich, Germany; U. Gruber-Sedlmayr, Graz, Austria; R. Rossi, Berlin, Germany; A. van Baalen, Berlin, Germany; J. Blanke, Meppen, Germany; C. Parlasca, Oberhausen, Germany; K. Raab, Nürnberg, Germany; F. Aksu, Datteln, Germany; M. Meusers, Herdecke, Germany; B. Hebing, Detmold, Germany; B. Seger-Fritz, Hanau, Germany; H. Bode, Ulm, Germany; M. Lüchtrath, Mnchengladbach, Germany; H. Hergerder, Kiel, Germany; C. Thiels, Bochum, Germany; B. Tpke, Aalen, Germany; and S. Lübke, Flensburg, Germany.

    FOOTNOTES

    Accepted Nov 8, 2004.

    No conflict of interest declared.

    REFERENCES

    Asbury A. New concepts of Guillain-Barré syndrome. J Child Neurol. 2000;15 :183 –191

    Asbury AK, Cornblath DR. Assessment of current diagnostic criteria for Guillain-Barré syndrome. Ann Neurol. 1990;27 :S21 –S24

    Royden-Jones H. Childhood Guillain-Barré syndrome: clinical presentation, diagnosis, and therapy. J Child Neurol. 1996;11 :4 –12

    van Koningsveld R, van Doorn PA, Schmitz PIM, et al. Mild forms of Guillain-Barré syndrome in an epidemiologic survey in the Netherlands. Neurology. 2000;54 :620 –625

    Guillain-Barré Syndrome Study Group. Plasmapheresis and acute Guillain-Barré syndrome. Neurology. 1985;35 :1096 –1104

    French Cooperative Group on Plasma Exchange and Guillain-Barré Syndrome. Efficiency of plasma exchange in Guillain-Barré syndrome: role of replacement fluids. Ann Neurol. 1987;22 :753 –761

    Raphael JC, Chevret S, Hughes RAC, Annane D. Plasma exchange for Guillain-Barré syndrome. Cochrane Database Syst Rev. 2002;(2):CD001798

    Bril V, Ilse WK, Pearce R, et al. Pilot trial of immunoglobulin versus plasma exchange in patients with Guillain-Barré syndrome. Neurology. 1996;46 :100 –103

    Hughes RA, Raphael JC, Swan AV, van Doorn PA. Intravenous immunoglobulin for Guillain-Barré syndrome. Cochrane Database Syst Rev. 2004;(1):CD002063

    Plasma Exchange/Sandoglobulin Gullain-Barré Syndrome Trial Group. A randomized trial comparing intravenous immune globulins and plasma exchange in Guillain-Barré syndrome. Lancet. 1997;349 :225 –230

    van der Meche FG, Schmitz PI. A randomized trial comparing intravenous immune globulin and plasma exchange in Guillain-Barré syndrome: Dutch Guillain-Barré Study Group. N Engl J Med. 1992;326 :1123 –1129

    Epstein MA, Sladky JT. The role of plasmapheresis in childhood Guillain-Barré syndrome. Ann Neurol. 1990;28 :65 –69

    Kathri BO, Flamini R, Baruah JK, et al. Plasmapheresis with acute inflammatory polyneuropathy. Pediatr Neurol. 1990;6 :17 –19

    Jansen PW, Perkin RM, Ashwal S. Guillain-Barré syndrome in childhood: natural course and efficacy of plasmapheresis. Pediatr Neurol. 1993;9 :16 –20

    Vajsar J, Sloane A, Wood E, Murphy EG. Plasmapheresis vs intravenous immunoglobulin treatment in childhood Guillain-Barré syndrome. Arch Pediatr Adolesc Med. 1994;148 :1210 –1212

    Korinthenberg R, Schulte Mnting J. Natural history and treatment effects in Guillain-Barré syndrome: a multicentre study. Arch Dis Child. 1996;74 :281 –287

    Abd-Allah SA, Jansen PW, Ashwal S, Perkin RM. Intravenous immunoglobulin as therapy for pediatric Guillain-Barré syndrome. J Child Neurol. 1997;12 :376 –380

    McGhee B, Jarjour IT. Single-dose intravenous immune globulin for treatment of Guillain-Barré syndrome. Am J Hosp Pharm. 1994;51 :97 –99

    Gürses N, Uysal S, Cetinkaya F, et al. Intravenous immunoglobulin treatment in children with Guillain-Barré syndrome. Scand J Infect Dis. 1995;27 :241 –243

    Shahar E, Shorer Z, Roifman CM, et al. Immune globulins are effective in severe pediatric Guillain-Barré syndrome. Pediatr Neurol. 1997;16 :32 –36

    Kanra G, Ozon A, Vajsar J, et al. Intravenous immunoglobulin treatment in children with Guillain-Barré syndrome. Eur J Paediatr Neurol. 1997;1 :7 –12

    Haupt WF, Rosenow F, van der Ven C, et al. Sequential treatment of Guillain-Barré syndrome with extracorporeal elimination and intravenous immunoglobulin. J Neurol Sci. 1996;137 :145 –149

    Delanoe C, Sebire G, Landrieu P, et al. Acute inflammatory demyelinating polyradiculoneuropathy in children: clinical and electrodiagnostic studies. Ann Neurol. 1998;44 :350 –356

    Green D, Ropper AH. Mild Guillain-Barré syndrome. Arch Neurol. 2001;58 :1098 –1101

    French Cooperative Group on Plasma Exchange in Guillain-Barré Syndrome. Appropriate number of plasma exchanges in Guillain-Barré syndrome. Ann Neurol. 1997;41 :298 –306

    Raphael JC, Chevret S, Harboun M, et al. Intravenous immune globulins in patients with Guillain-Barré syndrome and contraindications to plasma exchange: 3 days versus 6 days. J Neurol Neurosurg Psychiatry. 2001;71 :235 –238

    Kleyweg RP, van der Meche FG. Treatment related fluctuations in Guillain-Barré syndrome after high-dose immunoglobulins or plasma-exchange. J Neurol Neurosurg Psychiatry. 1991;54 :957 –960

    Irani DN, Cornblath DR, Chaudhry V, et al. Relapse in Guillain-Barré syndrome after treatment with human immune globulin. Neurology. 1993;43 :872 –875

    Romano JG, Rotta FT, Potter P, et al. Relapses in the Guillain-Barré syndrome after treatment with intravenous immune globulin or plasma exchange. Muscle Nerve. 1998;21 :1327 –1330

    Mc Khann GM, Griffin JW, Cornblath DR, et al. Plasmapheresis and Guillain-Barré syndrome: analysis of prognostic factors and the effect of plasmapheresis. Ann Neurol. 1988;23 :347 –353

    Reisin RC, Pociecha J, Rodriguez E, et al. Severe Guillain-Barré syndrome in childhood treated with human immune globulin. Pediatr Neurol. 1996;14 :308 –312

    Paradiso G, Tripoli J, Galicchio S, Fejerman N. Epidemiological, clinical, and electrodiagnostic findings in childhood Guillain-Barré syndrome: a reappraisal. Ann Neurol. 1999;46 :701 –707

    Rees JH, Soudain SE, Gregson NA, Hughes RAC. Campylobacter jejuni infection and Guillain-Barré syndrome. N Engl J Med. 1995;333 :1374 –1379

    Ogawara K, Kuwabara S, Mori M, et al. Axonal Guillain-Barré syndrome: relation to anti-ganglioside antibodies and Campylobacter jejuni infection in Japan. Ann Neurol. 2000;48 :624 –631

    Watson JDG, Gibson J, Joshua DE, Kronenberg H. Aseptic meningitis associated with high dose intravenous immunoglobulin therapy. J Neurol Neurosurg Psychiatry. 1991;54 :275 –276

    Oomes PG, van der Meche FG, Kleyweg RP. Liver function disturbances in Guillain-Barré syndrome: a prospective longitudinal study in 100 patients: Dutch Guillain-Barré Study Group. Neurology. 1996;46 :96 –100

    Warrier I, Bussel JB, Valdez L, et al. Safety and efficacy of low-dose intravenous immune globuline (IVIG) treatment for infants and children with immune thrombocytopenic purpura. J Pediatr Hematol Oncol. 1997;19 :197 –201

    Stangel M, Kiefer R, Pette M, et al. Side effects of intravenous immunoglobulins in neurological autoimmune disorders: a prospective study. J Neurol. 2003;250 :818 –821(Rudolf Korinthenberg, MD,)