Suspected functional visual loss is commonly encountered in routine clinical practice and often presents a challenge to the ophthalmologist. The main role of electrophysiology is to provide objective evidence of normal retinal and/or intracranial visual pathway function in the presence of subjective reports that suggest otherwise. In some patients, it is not possible to distinguish between malingering and hysteria; in others, there is little doubt. A clinical review of functional visual loss has addressed the diagnostic distinctions.²⁰ The term nonorganic visual loss is preferred, particularly if volitional aspects are uncertain.

It was commented in the first edition of this volume that although the clinical value of electrophysiology in the diagnosis of nonorganic visual loss was widely accepted, there were relatively few reports in the literature.¹⁵ That remains the case. The first report appears to be that of Potts and Nagaya,²⁵ who reported that patients with hysterical amblyopia had normal foveal visual evoked potentials (VEP) to a 0.06-degree flashing red stimulus, whereas patients with strabismic amblyopia showed diminished or absent responses. The use of diffuse flash stimulation was also described by Arden's group,¹ and those authors cautioned against the unequivocal acceptance of electrophysiological criteria, citing one case of almost certain hysterical amblyopia in which scotopic flash VEP (FVEP) anomalies were observed. It is the case, however, that functional overlay superimposed on underlying organic dysfunction may be encountered. Subsequent reports, though mostly anecdotal, confirmed the finding of normal FVEPs in nonorganic visual loss.^2,3,11,22 The use of the FVEP in assisting disclose the nonorganic basis of suspect symptoms following trauma was also described.²¹

However, although the use of a luminance stimulus may be satisfactory in the evaluation of hysterical total blindness, which is not usually difficult to ascertain clinically, a contrast stimulus is necessary to evaluate less severe reported deficits. Halliday¹⁹ first described the use of the pattern-reversal VEP (PVEP) in hysterical visual loss, reporting normal, symmetrical PVEPs in both eyes of patients despite markedly asymmetrical visual acuity. The technique was thought to be most useful in unilateral visual loss in which the good eye acts as a control. It was also stressed that a normal PVEP, although strongly suggestive of nonorganic visual loss, does not preclude the existence of some organic disease.

The early studies using flashed-pattern VEPs^12,13,26 found that the maximum response amplitude occurred with check sizes of 10–30 minutes of arc. Later studies with pattern reversal came to similar conclusions for small fields but further suggested an interrelationship between check size and field size.^8,23,24 Small checks and fields are better for foveal stimulation, large checks and large fields for more peripheral retina.

The clinical management of patients with nonorganic visual loss would be facilitated if a simple relationship between VEP measurements and visual acuity existed that enabled an objective assessment of acuity. Unfortunately, most scaling methods that use pattern reversal perform relatively poorly as predictors of acuity,⁷ although Halliday and McDonald¹⁰ suggested that a well-formed pattern-reversal VEP is incompatible with a visual acuity of ∼6/36 or worse. Despite the problems, attempts to assess acuity objectively in patients who are suspected of nonorganic visual loss have been made. Wildberger³² reported the findings in two groups of patients: 17 “malingerers” (mostly schoolgirls) with no signs of an organic lesion and 10 patients (accident or disease) with signs but marked overlay. The findings in the patients were compared with those obtained in normal subjects to the same four check sizes in relation to insertion of graded orthoptic filters intended to reduce the acuity. The VEPs were easily able to detect malingering in the second group of patients, in whom acuities were usually claimed to be markedly reduced, but were not sensitive in the first group with milder claimed reductions. The VEP was assessed with the offset amplitude of the P100 component (P100–N135); in this author's laboratories, the onset amplitude of the pattern-reversal VEP (N75–P100) seems more sensitive. It is probably advisable routinely to measure both parameters. A recent publication³³ reported the use of pattern-reversal VEP in 72 patients with functional visual loss, suggesting that the discrepancy between acuity estimated by VEP and the best performed visual acuity was less than three lines on a Snellen chart in 88% of patients.

The use of pattern-onset stimulation, rather than pattern reversal, was previously described to yield improved results (G. B. Arden, personal communication; see also Holder¹⁵). This has been confirmed in the authors' laboratories (V. McBain, et al., unpublished data). It is recognized that the pattern-onset or appearance VEP (PaVEP) is more susceptible to blur than is the pattern-reversal VEP. This enables a more direct relationship to be established between visual acuity and the presence or absence of a response to small check sizes of varying contrast. In addition, the spatiotemporal tuning function of the PaVEP is simpler and correlates closely with contrast sensitivity,^17,19,24 and the responses are often of larger amplitude than in reversal VEPs.²⁴ Furthermore, if a short (e.g., 40ms) appearance time is used, it is difficult for the patient to defocus the pattern voluntarily.¹⁸ PaVEP results are generally superior to those obtained with reversal when the patient is deliberately trying to influence the results.

Technical factors in the recording of patients who are suspected of hysteria or malingering are of paramount importance but receive little attention. The patient may fail to fixate, may attempt to defocus, may attempt prolonged eye closure during blinking, and so on. PVEP changes have been reported under such conditions.^5,28,29 Direct observation of the patient, with the patient aware of such monitoring, will often result in improved compliance (figure 51.1). Careful observation of both the raw electroencephalographic (EEG) input and the developing average is advisable; the appearance of an alpha rhythm in the ongoing EEG may indicate a failure in concentration. Equally, a tendency for the P100 component to broaden or increase in latency in the acquired average suggests that accommodation or fixation is unsatisfactory. Verbal commands to the patient to concentrate and attend to the fixation mark (or the center of the screen if perception of the fixation mark is denied) may be beneficial. If all perception is denied with one eye, fixation can be obtained by using the better eye, and an instruction to “try to keep your eyes still” will surprisingly often produce good results following occlusion of this eye and prompt commencement of stimulation. It may be necessary to stop averaging after fewer sweeps than usual to prevent waveform deterioration. A tendency for the P100 component to sharpen and remain of stable or slightly reducing latency during acquisition of the average only occurs with good patient compliance; observation of a broadening of the peak and/or increase in P100 latency suggests poor patient compliance and demands intervention by the recordist. A subjective report of stimulus perception should be obtained from the patient in all cases. Marked discrepancies between the subjective reports and the objective electrophysiological findings can be useful in alerting the examiner to nonorganic visual loss, particularly if marked interocular perceptual asymmetries are unaccompanied by VEP asymmetries.

Figure 51.1.  

VEP and PERG findings in a 32-year-old female with reduced right visual acuity following trauma. Litigation was pending. Left eye findings (6/5) are normal. Ophthalmic examination was normal. Right eye PVEP was normal when the patient was aware that she was under direct observation [RE(o)] but delayed when she thought that she was unobserved [RE(u)].

It is advisable to record the PERG. The PERG P50 component is very susceptible to deterioration with poor compliance, and a normal PERG can only be obtained with satisfactory fixation, accommodation, and so on. Factors that affect the PERG are described elsewhere in this volume (see chapter 22). Equally, cases of mild maculopathy, with an unequivocally abnormal PERG, may sometimes have a PVEP within the normal range (Holder, unpublished observations); a normal PVEP should not therefore be presumed to preclude mild macular dysfunction. Rover and Bach²⁷ report the use of simultaneous recording of the PERG and PVEP to reveal malingering. Their patients complained of marked acuity loss, but only a few representative cases were discussed, and no quantitative patient data were presented. In particular, no mention was made as to whether their cases were bilateral or unilateral. A normal PERG and PVEP indicated malingering; a normal PERG and an abnormal PVEP indicated a “lesion of the visual pathway”; and an abnormal PERG and an abnormal PVEP indicated blurred image, poor cooperation, or retinal dysfunction.

Simultaneous PERG and PVEP can be particularly useful in unilateral cases; PVEP interpretation in a patient who is deliberately attempting to influence the results can be substantially improved by knowledge of the retinal response simultaneously recorded to the same stimulus. Binocular registration of the PERG, with the “good” eye enabling fixation, will usually reveal whether the PERG from the “bad” eye is abnormal or not (see figure 51.1). Significant macular dysfunction is excluded if the PERG from the bad eye is normal with binocular stimulation. If there is a unilateral PERG abnormality in the bad eye, the nature of the abnormality will indicate either ganglion cell/optic nerve or more distal dysfunction depending on whether the N95 or P50 component of the PERG is affected¹⁶ (see chapter 22).

Routine ERG should also be performed; ERGs can be markedly abnormal in retinal dysfunction with no or minimal fundus abnormality but constricted visual fields, and field loss is often a feature of nonorganic visual loss. PERG and PVEP findings may be normal in such conditions if the maculae are spared. However, because normal PVEPs may be found in patients with cortical blindness, particular care should be taken before making the diagnosis of non-organic visual loss if the symptoms suggest possible cortical dysfunction. Celesia's group⁶ studied a 72-year-old woman with bilateral destruction of area 17 and attributed the presence of normal VEPs to conduction in extrageniculocalcarine pathways. Bodis-Wollner et a1.⁴ reported normal VEPs in a 6-year-old boy with destruction of areas 18 and 19 but preservation of area 17. Similarly, retrochiasmal lesions may not give a PVEP abnormality even with a demonstrable field defect (see chapter 78). Evaluation of the P300 component may circumvent such problems.³⁰

To conclude, electrodiagnostic testing is invaluable in the detection or confirmation of nonorganic visual loss, particularly if it is unilateral, so one eye can be judged against the other. Although the reader is reminded of the warning of Halliday⁹ that normal electrophysiology does not preclude the presence of some underlying organic disease, the demonstration of normal electrophysiology can be reassuring, both for the clinician and, if the patient is a child, for concerned parents. Particular caution must be exercised if there is a possibility of cortical dysfunction. The most challenging patients are probably those with functional overlay superimposed on genuine underlying organic dysfunction. It is essential that an accurate history is taken and comprehensive ophthalmic and neurological examination performed; particular techniques for revealing functional deficit have been described in full elsewhere.³¹ Electrophysiological examination is always advisable if there is any doubt. The objective nature of electrodiagnostic testing may not only demonstrate normal visual pathway function in patients whose symptoms suggest otherwise, but also reveal the presence of organic dysfunction in a patient with a presumed diagnosis of nonorganic visual loss (figure 51.2).

Figure 51.2.  

Electrophysiological findings in a 43-year-old female with a one-month history of sudden painless visual loss in the right eye. Ophthalmic examination was normal. The patient had a mother and sister who had previously had optic neuritis as part of multiple sclerosis, and it was suspected that the visual loss was nonorganic. The PVEPs from both right and left eyes are delayed; the lack of relative afferent pupillary defect presumably relating to the subclinical demyelination in the asymptomatic left optic nerve. Note the mild relative N95 reduction in the PERG from the right eye compared with the left.