Diagnostic Criteria for CIDP 1975-2001 (EFNS/PNS)

Carol L. Koski, M.D. and Richard A. Lewis, M.D., Principal Authors

The first description of a series of chronic inflammatory demyelinating polyneuropathy (CIDP) patients was published in 1975. The clinical characteristics of CIDP included:

• a progressive course of disease over 6 months
• usually slowed nerve conduction velocities and occurrence of conduction block
• spinal fluid albumino-cytological disassociation
• nerve biopsy demonstrating segmental demyelination
• remyelination subperineural or endoneurial edema
• perivasular inflammation.

Exclusion criteria in this early proposal were associated diseases, monoclonal gammopathy, and evidence of heredity neuropathy. (Dyck, 1975)

The clinical characteristic of this descriptive proposal served as the basis for a formalized set of CIDP criteria devised in 1988 (Barohn, 1989) which reduced disease progression time to 2 months, a time frame that stands today. The 1988 criteria were devised primarily to enable clinicians to make a timely diagnosis of CIDP and to not delay treatment for those patients whose clinical symptoms and laboratory features did not fall into the definite group. (Saperstein, 2001)

Later CIDP criteria elaborated and/or modified the original criteria. Subsequent criteria have retained specific aspects presented in earlier guidelines in certain areas, while in other areas subsequent criteria have proposed or implemented changes, modifications, or presented more detailed descriptions.

The major areas the early guidelines addressed included:

• the clinical features of CIDP
• the electrodiagnostic features of CIDP
• the laboratory features to use in the evaluation process, and
• various requirements for the different diagnostic categories.

There have been a number of guidelines and criteria developed to assist the clinician and researcher in diagnosing CIDP. All the criteria are designed to utilize clinical, electrodiagnostic, laboratory and pathologic features to determine whether the patient has "definite, probable or possible" CIDP. The earlier criteria tended to require nerve biopsy but this is now recommended for supportive evidence. The American Academy of Neurology ( AAN ) criteria, created as research criteria, are considered the least sensitive primarily because of the conduction study criteria as well as the requirement for nerve biopsy and spinal fluid examination to substantiate a definite diagnosis of CIDP. However, the AAN definition for conduction block is actually more liberal than most, and the clinical requirements defined by AAN are less than most.

More than 10 published criteria have been designed to explicitly define the parameters of conduction changes that reflect a demyelinating neuropathy, not just CIDP but also Guillain Barre Syndrome (GBS) , Multifocal Motor Neuropathy (MNN) and other demyelinating neuropathies. They differ in the degree of slowing, the number of abnormalities required, the definition of conduction block to determine definite, probable and possible demyelination. A comparison study of 10 criteria in patients with GBS, CIDP, ALS and diabetic neuropathy revealed sensitivities ranging from 39 to 89% for CIDP. One set identified an ALS patient with demyelinating features and 8 of the criteria overlapped with diabetic neuropathy. The three most sensitive to CIDP each overlapped with diabetic neuropathy in over half. Based on this the authors devised another set of criteria but could only come up with 75% sensitivity while avoiding overlaps with diabetic neuropathy ( van den Bergh, 2004). Despite the inability to obtain ideal sensitivity and specificity these criteria can be helpful to the clinician and are important for clinical trials when uniformity of diagnosis is critical.

Table 1 compares the diagnostic criteria published in 1998 (Barohn, 1998) with those published by the American Academy of Neurology (AAN) Ad hoc subcommittee in 1991 (Ad Hoc Committee, 1991), criteria published in 200l (Saperstein, 2001), and the more recent Inflammatory Neuropathy Cause and Treatment (INCAT) criteria.

Table 1. Diagnostic Criteria for CIDP: 1988-2001

	Barohn et al., 1988	AAN Ad Hoc Subcommittee, 1991	Saperstein, 2001	INCAT, 2001
Mandatory clinical features
Pattern of clinical involvement	Symmetric, proximal + distal Weakness	Motor and/or sensory dysfunction involving more than one limb	Major: symmetric, proximal + distal weakness Minor: exclusively distal weakness or sensory loss	Motor and sensory dysfunction in > 1 extremity; significant disability in arm or leg function
Reflexes	Areflexia or hyporeflexia	Areflexia or hyporeflexia	Areflexia or hyporeflexia	Areflexia or hyporeflexia
Time course	-	-	At least 2 mos.	At least 2 mos: stable or worsening clinical condition
Laboratory features
Electrodiagnostic studies	Motor conduction velocity reduced <70% of LLN	3 of 4 criteria, CV in 1 nerve slow NCV, 2 nerves, á DL, 2 nerves, á F wave - 2 nerves	2 of 4 AAEM criteria	CD or TD in 2 nerves and á DL, F wave or NCV in 1 other nerve; if no CB at least combination of á DL, F wave or NCV in 3 nerves
CSF studies	CSF protein >45 mg/dL	Mandatory: cell count <10/mm3† negative VDRL Supportive: elevated protein	Mandatory: protein >45 mg/dL Supportive: cell count < 10/mm3†	Supportive but not mandatory
Nerve biopsy	Predominant features of demyelination§ Inflammation	Unequivocal evidence of demyelination and remyelination	Predominant features of demyelination§ Inflammation (not required)	Supportive but not mandatory
Requirements for diagnostic categories
Definite	Clinical, electrodiagnostic, CSF, and biopsy	Clinical, electrodiagnostic, CSF‡, and biopsy	Clinical major, electrodiagnostic, and CSF (biopsy supportive but not mandatory)	Clinical and electrodiagnostic
Probable	Clinical and 2 out of 3 laboratory features	Clinical, electrodiagnostic, and CSF‡	Clinical major, electrodiagnostic or CSF, and biopsy	-
Possible	Clinical and 1 out of 3 laboratory features	Clinical and electrodiagnostic	Clinical major and 1 out of 3 Clinical minor and 2 out of 3	-

† CSF cell count > 10/mm3 (or >50/mm3 if HIV seropositive) should prompt evaluation for HIV, Lyme disease, and lymphomatous or leukemic infiltration of nerve roots.

‡ AAN criteria require only a normal CSF count and a negative VDRL (elevated protein is not a mandatory feature).

§ To include segmental demyelination, remyelination, and onion-bulb formation.