Article Text

Call for guidelines for monitoring renal function and haematological variables during intravenous infusion of immunoglobulin in neurological patients
  1. N P S BAJAJ,
  3. R BAHL,
  4. K STOTT,
  1. Departments of Neurology and Pharmacy, Hurstwood Park Neurological Centre (HPNC), Hurstwood Lane, West Sussex, UK
  1. Dr N P S Bajaj, Department of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London WC1 3BG, UK narinderbajaj{at}

Statistics from

Intravenous immunoglobulin (IVIg) is widely used in the treatment of some neurological conditions thought to have an underlying immune basis. Controlled studies of IVIg have demonstrated benefit in Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, and dermatomyositis. Treatment with IVIg has also been beneficial in myaesthenia gravis, multiple sclerosis, multifocal motor neuropathy with conduction block, polymyositis, Lambert-Eaton myaesthenic syndrome, stiff man syndrome, and Rasmussen's encephalitis.1

Various complications have been reported in the literature in association with IVIg therapy. These include headache, nausea, fever, rash, aches in the chest or limbs, anaphylaxis especially in association with IgA deficiency, leucopenia, neutropenia, autoimmune haemolysis, renal failure, thromboembolism, aseptic meningitis, and transmission of viral infections—for example, hepatitis C.1 The therapeutic dose of IVIg in the treatment of neurological disease has been empirically set at 2 g/kg, conventionally divided into five daily doses of 400 mg/kg, although some authors have shown that a 2 day infusion of 1 g/kg is not associated with any higher incidence of side effects than the 5 day infusion.2

Despite the widespread use of IVIg in neurological centres in the United Kingdom, to our knowledge there exists no consensus for advice either on monitoring haematological and renal function in patients pretreatment and post-treatment with IVIg, nor on the merits of shorter infusion periods of IVIg. Both of these factors have considerable cost implications for the National Health Service (NHS).

We have retrospectively examined the records of 21 patients admitted to a regional neurology centre (Hurstwood Park Neurological Centre), over an 18 month period. As several of these patients had multiple courses of IVIg treatment, the records contained 71 courses of treatment, although complete haematological data pretreatment and post-treatment was only available on 35 of these. The conditions treated included Guillain-Barré syndrome, chronic inflammatory demyelinating polyneuropathy, multifocal motor neuropathy with conduction block, and myaesthenia gravis. The average age of patients was 51 years (range 22–86 years). Eight patients had indications of possible renal dysfunction (based on one or more abnormal blood urea and creatinine concentration) within 6 months before the first course of treatment. All courses given were at 0.4 g/kg Sandoglobulin (Novartis )for 5 days. A course of treatment was stopped or delayed eight times out of 71 courses: the reasons stated included thombocytopenia (one patient: platelet count<100×109/l, normal range 150–400×109/l); leucopenia (three patients: white cell count<1.7×109/l, normal range 4–11×109/l), neutropenia (three patients: neutrophil count<1.5×109/l, normal range 2–7.5×109/l); and renal failure (one patient: sodium 134 mM, normal range 132–144 mM, potassium 5.6 mM, normal range 3.5–5.5 mM, urea 20 mM, normal range 2.5–6.7 mM, and creatinine 330 μM, normal range 60–120 μM]). Other adverse effects noted which did not delay or stop treatment included tachycardia (one patient), fall in haemoglobin (one), persistent pyrexia (one), nausea and vomiting (four), limb or chest pain (four), rigors (two), and headache (two). It was noted that there was no consensus on the level of leucopenia, neutropenia, raised urea, or creatinine at which treatment was discontinued among the six consultant neurologists at the centre. During a 5 day course, the white cell count was noted to fall below 4×109/l in 12 patients and the neutrophil count to below 2.0×109/l in eight patients. Urea and creatinine concentration only became abnormal in one patient (who developed renal failure requiring haemodialysis) whose renal function was mildly impaired before treatment (sodium 133 mM, potassium 4.6 mM, urea 7.0 mM, and creatinine 138 μM). In all patients with noted haematological derangement secondary to IVIg treatment, subsequent blood monitoring 14 days after stopping treatment showed a return to normal values.

The number of patients who had abnormal blood variables during this retrospective study suggests a need to establish guidelines for monitoring haematological and renal variables during IVIg therapy. Furthermore, given the cost implications of an abortive course of IVIg treatment, guidelines for a tolerable level of leucopenia, neutropenia, or urea/creatinine derangement during a standard 5 day course of IVIg also needs to be established. It may well be that neutropenia during IVIg treatment is transient and reversible, as suggested by this study, and if validated, there would be an argument against the need for regular haematological monitoring during IVIg therapy. This argument is supported by the findings of Koffman et al,3 who retrospectively reviewed the records of 46 patients with neuromuscular disease receiving standard courses of IVIg (Gamimune N (Bayer) 2 g/kg) compared with 23 patients given placebo infusions of dextrose in water. In this study leucopenia, neutropenia, and lymphopenia were noted in a large proportion of patients treated with IVIg. Despite this no significant side effects—for example, infection—resulted and all haematological derangements were transient.

Furthermore, consensus needs to be established on the effect of IVIg formulation and risk of sucrose nephropathy (sucrose is used as a stabiliser in some IVIg preparations). Some previous case reports have suggested a link between the use of high sucrose formulations of IVIg (for example, Sandoglobulin (Novartis)) and subsequent renal failure compared with the use of lower sugar or glycine based formulations.4 However, a recent report by Levy and Pusey5 comparing Vigam (BPL; 0.5 g sucrose/g immunoglobulin) and Sandoglobulin (Novartis: 1.76 g sucrose/g immunoglobulin) in various indications has shown no such correlation between concentration of sucrose stabiliser in IVIg and propensity to cause renal failure. In their study of 119 patients given 287 courses of IVIg, eight patients (6.7%) showed a deterioration in renal function regardless of preparation used. On this basis, they concluded that all patients given IVIg should have renal function monitored before, during, and 4–5 days after treatment. This should be compared with the study of Koffman et al,3 where none of the 46 patients given the same preparation of high dose IVIg had renal dysfunction.

Our current practice, based on the results of this audit and the literature available, is to check the renal function of all patients before IVIg therapy. Those in whom the renal function is mildly abnormal (normal sodium and potassium, urea 7–8 mM, and creatinine 120–150 μM) have their renal function monitored during and 5 days after IVIg treatment and are currently receiving low sucrose or no sucrose (Octogam(Octopharma)) IVIg formulations. Patients with more seriously impaired renal function are not being considered for IVIg therapy; alternative modes of treatment—for example, plasmapharesis—could be considered for this subgroup. Haematological function is also checked before IVIg therapy; if normal, no further monitoring is carried out during or after IVIg treatment. If there is evidence of mild leucopenia, neutropenia, or thrombocytopenia before IVIg, the full blood count is monitored on a daily basis during treatment and once more 5 days after treatment. Patients with more severe blood derangement (platelets<100×109/l, neutrophil count<1×109/l, and leucocyte count<2×109/l) are not being considered for IVIg therapy and again alternative modes of therapy would be considered.

A consensus statement on the recommended duration of treatment course (1–2 days v 5 days) and the requirements for blood monitoring during IVIg infusion will require further study and collaborative audit across the many neurological centres in the United Kingdom using this form of therapy. We think that the potential cost implications and side effect profile of IVIg justify a call for such a study.


We thank Professor Richard Hughes for his help in the preparation of this manuscript.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.