Neurology -- Correspondence for Hughes et al., 61 (6) 736-740

Correspondence published:

Practice parameter: Immunotherapy for Guillain–Barré syndrome: Report of the Quality Standards Subco: Norman Latov (19 November 2003)

Reply to Latov: Richard AC Hughes (19 November 2003)

Practice parameter: Immunotherapy for Guillain–Barré syndrome: Report of the Quality Standards Subco 19 November 2003

Norman Latov,
Cornell University / The Neuropathy Association
The Peripheral Neuropathy Center, 635 Madison Ave, Suite 400, New York, N.Y. 10022
Send Correspondence to journal:
Re: Practice parameter: Immunotherapy for Guillain–Barré syndrome: Report of the Quality Standards Subco

nol2002{at}med.cornell.edu Norman Latov

The report by the Quality Standards Subcommittee or the American Academy of Neurology [1] is a commendable review of the literature. The conclusions limiting recommendation for therapy with plasmapheresis or IVIg to “nonambulatory adult patients with GBS,” or as “treatment options for children with severe GBS,” are questionable. Because these recommendations are “evidence based,” they raise questions regarding the suitability of basing complex medical decision-making or practice guidelines on such criteria.
“Evidence based medicine,” purports to extend guidelines suitable for therapeutic trials, to all forms of medical knowledge. Current practices that do not meet “evidence based” criteria, are considered suspect or even invalid. This is despite the fact that most of what we know or do today, beginning with the neurological examination, is based on demonstrated efficacy, proven through reproducibility and predictability over many years of practice, and by multiple investigators.
Practice guidelines that are “evidence based” are consequently more restrictive than those based on current clinical practice, and do not consider clinical judgment, or even common sense, as a basis for decision making. In the case of GBS, as example, a more reasonable approach might favor instituting treatment sooner than recommended, to prevent loss of ambulation. That particular issue has not been addressed, but given the complexity and expense of evidence based research [2], requiring that every aspect of a particular therapy be so tested is unrealistic, and would severely limit our options, to the detriment of patient care.
Basing the strength of recommendation on the quality of evidence is also fundamentally problematic. In practice, we make recommendations according to the best available evidence, even if it is not perfect. For example, a practicing neurologist would recommend plasmapheresis or IVIg therapy to children with GBS as frequently and emphatically as to adults, despite the fact that the evidence is not as good in children.
An “evidence based” analysis typically begins with hundreds of publications, which are quickly whittled to a handful that meet class III criteria or better. None of the papers by the founders of modern neurology, including Guillain and Barre themselves, would make the first cut. It is an embarrassing exercise, with little appreciation or respect for our brilliant heritage and hard-won knowledge. It reminds one of another “Cultural Revolution”, which probably also started out as somebody’s notion of a good idea, but with disastrous consequence.
References
1. Hughes RAC, Wijdicks EFM, Barohn R, Benson E, Cornblath DR, Hahn AF, Meythaler JM, Miller RG, Sladky JT, Stevens JC. Practice parameters: Immunotherapy for Guillain-Barre syndrome. Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2003; 61: 736 -740.
2. Caplan LR. Evidence based medicine: concerns of a clinical neurologist. J Neurol Neurosurg Psychiat. 2001: 71: 569-576.
Disclosure: Dr. N. Latov has received honoraria for consulting on topics related to diagnosis and therapy of neuroimmune diseases from Athena Diagnostics, Aventis Behring, Bayer Pharmaceuticals, Biogen, Pfizer, Quest Diagnostics, Wyeth Pharmaceuticals, and ZLB Bioplasma.

Reply to Latov 19 November 2003

Richard AC Hughes,
Professor of Neurology
Department of Clinical Neurosciences, Guy's King's and St Thomas School of Medicine London SE1 1UL
Send Correspondence to journal:
Re: Reply to Latov

richard.a.hughes{at}kcl.ac.uk Richard AC Hughes

Dr Latov’s diatribe against evidence-based medicine in general and the Guillain-Barré syndrome (GBS) practice parameter in particular ignores the anonymous adage “Rules are for the obedience of the inexperienced and the guidance of wise men.” When effect sizes are only moderate, which applies to intravenous immunoglobulin and plasma exchange in GBS, it is not possible for even the wisest and most experienced physician to determine whether a patient’s recovery has occurred because of or in spite of treatment.
The rigorous collection of evidence allowed the practice parameter confident statements about some scenarios such as treatment of early severe GBS in adults. Extrapolation of the results to help make decisions about populations, such as children and those with mild disease, has to be done with caution. The risks [1] of these interventions may not be justified in those with an excellent prognosis without treatment such as those who are still able to walk after eight days. [2]
Unsurprisingly, Guillain and Barré made no contribution to ascertaining the value of specific treatments for GBS. Guillain did make recommendations about supportive care but was reluctant to admit that his syndrome could ever be fatal. My expectation is that the founding fathers of Neurology would have embraced evidence-based practice and contributed to the process of collecting that evidence. One wishes that there were more situations when effect sizes are so large that randomized trials are not necessary. The danger is that we are still ignoring treatments because effect sizes are only moderate and we have not undertaken appropriate systematic reviews. A striking example is the failure of the profession to introduce thrombolytic treatment for myocardial infarction until the late 1980s although the evidence of a significant effect had been available ten years earlier. [3] It is unlikely that neurologists are much wiser than cardiologists.
References
1. Dalakas MC, Clark WM. Strokes, thromboembolic events, and IVIg: Rare incidents blemish an excellent safety record. Neurology 2003; 60(11):1736-1737.
2. Green DM, Ropper AH. Mild Guillain-Barré syndrome. Arch Neurol 2001; 58(7):1098-1101.
3. Antman EM, Lau J, Kupelnick B, Mosteller F, Chalmers TC. A comparison of results of meta-analyses of randomized control trials and recommendations of clinical experts. Treatments for myocardial infarction. JAMA 1992; 268(2):240-248.