Reversible Bitemporal Hemihypokinetic Pupil Without Hemianopia: A New Chiasmal Sign

ABSTRACT

Chiasmal syndromes present mostly with visual problems, such as changes in visual fields, decreased visual acuity, or dyschromatopsia (and classically without pupillary reflex defects). The prototypical bitemporal hemianopia upon visual field testing can easily suggest chiasmal compression due to sellar/suprasellar involvement. However, because of the complexity of the decussation of fibres at the optic chiasm and the presence of anatomical variants, unpredictable visual fields defects can be detected in chiasmal diseases. In some patients, especially in those who have undergone neurosurgical procedures, visual field examination and neuroimaging may not completely reflect the classical pattern of chiasmal visual loss. We describe a novel semiological sign, reporting a patient in which a pupillary bitemporal hemihypokinesia was not accompanied by hemianopia, with the phenomenon being abolished by surgical resection of the causative pituitary macroadenoma. In addition, this finding was an important tool in making the diagnosis.

Introduction

The presence of prototypical bitemporal hemianopia upon visual field testing often suggests chiasmal compression due to sellar/suprasellar involvement.¹ However, in some patients, especially in those who have undergone neurosurgical procedures (due to the traction/distortion of nerve fibres), visual field examination and neuroimaging may not completely reflect the classical pattern of chiasmal visual loss, making the diagnosis more challenging.²

Pupillary hemihypokinesia is defined as a diminished pupillary light reaction from the hemiretina with decreased vision, compared with the other seeing hemiretina, whereas pupillary hemiakinesia is the abolition of the pupillary light reaction from a blind hemiretina while it is maintained in the seeing portion.³ Traditionally, both hemihypokinesia and hemiakinesia are associated with the presence of a visual field deficit corresponding and proportional to the pupillary motor defect.^3,4

We describe a novel semiological sign, reporting a patient in which a pupillary bitemporal hemihypokinesia was not accompanied by hemianopia, with the phenomenon being abolished by surgical resection of the causative pituitary macroadenoma. In addition, this finding was an important tool regarding making the diagnosis.

Case report

A 38-year-old woman presented to the emergency room complaining of a two-day history of right retro-orbital pain, right progressive visual loss, and “seeing faded colours.”

Upon examination, the visual acuity was 0.8 in the right eye and 1.0 in the left eye. She had a right relative afferent pupillary defect (RAPD) and a right dyschromatopsia, with normal visual field testing confrontation and normal funduscopic findings.

She had a prior history of four partial trans-sphenoidal endoscopic resections of a non-functioning pituitary macroadenoma (without radiotherapy), the last one being 2 years earlier.

Brain magnetic resonance imaging (MRI), with good visualisation of the orbits, showed a pituitary macroadenoma with minimal compression of the optic chiasm and of the left optic nerve, with a slight increase in the macroadenoma dimensions compared with the previous examination (Figure 1). The cerebrospinal fluid was unremarkable, and auto-antibodies were negative (optic neuritis panel).

Figure 1.

  Brain magnetic resonance imaging of a pituitary macroadenoma during follow-up from 2018–2020. (a–c) T1-weighted post-gadolinium sagittal brain imaging: (a) one year before the last presentation to the emergency room (2018); (b) at the onset of symptoms (2019), measuring (2.7 x 2.4 × 2.2 cm); and (a) after the last subtotal resection (2020). (d–f) T1-weighted post-gadolinium coronal brain imaging: (d) one year before the last presentation to the emergency room (2018); (e) at the onset of symptoms (2019); and (f) after the last subtotal resection (2020). A slight enlargement from 2018 (b & e) to 2019 (a & d) were also noticed on both sagittal and coronal sequences. (g–i) Fluid attenuated inversion recovery (FLAIR) axial sequences: (g) one year before the last presentation to the emergency room (2018); (h) at the onset of symptoms (2019); and (i) after the last subtotal resection (2020). On the FLAIR axial slices (g & h), it is more difficult to see the increase in the chiasmal lesion compared to the coronal slices from 2018 (d) to 2019 (e); however, the asymmetrical compression of the left optic nerve is more noticeable.

The automated perimetry test (APT) demonstrated a mild superior temporal reduction in sensitivity in the field of the left eye and a superior temporal scotoma with diffuse loss of sensitivity in the right eye (Figure 2). Optical coherence tomography (OCT) revealed borderline loss of fibre layers in the nasal, temporal, and superior fields of the right eye. Although the mean value in the nasal field was normal in the left eye, on the retinal nerve fibre layer (RFNL) thickness map there was a focal borderline loss of volume in the lower-nasal (LN) fibres (Figure 3). However, it was not possible to differentiate acute from chronic findings as we did not have any previous OCT imaging to compare with.

Figure 2.

Automated perimetry testing at the onset of symptoms (2019). (a) Humphrey Visual Field Analyser greyscale maps; (b) thresholds values; and (c) pattern deviation maps. Left eye: there is mild superior temporal reduction in sensitivity. Right eye: there is a superior temporal scotoma with diffuse loss of sensitivity. Abbreviations: LE = left eye; RE = right eye.

Figure 3.

Optical coherence tomography (OCT) at the onset of symptoms (2019) using the Optoviu device, software version 2018.0.0.18. (a) OCT imaging of the retinal nerve fibre layer (RNFL) thickness. In the left eye the average values are normal but on the RNFL thickness map there is focal borderline loss of volume in the lower-nasal fibres. In the right eye the average values reveal a borderline loss of volume in the nasal, temporal, and superior fibres, but on the RNFL thickness map there is predominant thinning of the superior nasal fibres, where the value dips into the abnormal range. (c) OCT imaging of the ganglion cell complex (GCC). In the left eye it is normal (although the analysis is not fully centred on the fovea). In the right eye there is predominant thinning in the nasal region. The analyses of the RNFL and the GCC are discussed in detail in the text (discussion section).

GCC = ganglion cell complex; ONH = optic nerve head; OD = right eye; OS = left eye; RNFL = retinal nerve fibre layer

Thus, a right optic neuropathy of uncertain aetiology was considered (probably compressive, but it was not possible to completely rule out an inflammatory demyelinating cause), so a 5 day intravenous methylprednisolone was administered, leading to an 80% improvement in the patient’s symptoms. However, 1 week later she returned due to recurrence of her symptoms in the right eye including retro-orbital pain, loss of vision, and dyschromatopsia. A more thorough pupil examination was performed by other professionals, and it was noted that the pupil finding was not simply an RAPD. When hemifield light stimulation of each eye was performed individually (while occluding the non-tested eye) with a beam flashlight, a peculiar pupil response was noted: right nasal hemiretina light stimulation produced a poor pupil contraction, whereas stimulation of the temporal hemiretina resulted in better pupil contraction (online supplement Video 1). In the left eye, the difference was not as significant as in the right eye but it was also visible (online supplement Video 2).

An orbital MRI was performed and showed no lesions in the optic nerves. The clinical scenario was explained by the simultaneous compression of the central optic chiasm (binasal fibres) in both optic nerves, the right eye being more affected than the left. The patient underwent a new transsphenoidal resection of the adenoma, without any complications. She had normal visual acuity and pupillary light responses when re-evaluated 10 months after surgery (online supplement Videos 1 and 2).

Discussion

In this paper we describe a novel chiasmal sign in which a pupillary bitemporal hemihypokinesia was not accompanied by hemianopia and the phenomenon was abolished by surgical treatment.

Although the patient had a pituitary macroadenoma, the presentation here was acute, with pain, marked asymmetrical involvement, dyschromatopsia, an RAPD, and a mild decrease in visual acuity, which is not commonly seen in chiasmal syndromes.^1,2,5,6 This led to the suspicion of an inflammatory demyelinating cause. In our patient, the response to corticosteroid therapy probably occurred due to a temporary reduction in local oedema/inflammation/compression of the right optic nerve, with recurrence of the visual symptoms soon after corticosteroid withdrawal.

In the presence of bitemporal hemihypokinesia, light stimulation that does not respect the vertical meridian may elicit an RAPD, confounding the diagnosis.^1,6,7

Additionally, the swinging flashlight test done between the two hemiretinas (in each eye) or between the two eyes could elicit an RAPD in our case, since there was an asymmetry between the hemiretinas in each eye separately and between the two eyes.

The APT demonstrated a pronounced asymmetrical visual field deficit, and the OCT revealed a borderline loss of fibre layers in the nasal, temporal, and superior fields of the right eye, as well as a focal borderline loss of the LN fibres in the left eye (which could explain the mild superior temporal reduction in sensitivity in the visual field of this eye).

Nevertheless, Schiefer et al. have described that up to 19% of chiasmal lesions can present with unclassifiable visual field patterns, 9% can present with a monocular visual field defect that does not respect the midline, and 3% can present as pseudoconstriction of the visual field.⁵ Furthermore, due to multiple adenoma resections and chronic chiasmal compression, the normal anatomy could be distorted, generating bizarre visual field defects and contributing to the changes on the OCT.

Sellar masses represent the most common cause of a visual defect known as the junctional scotoma of Traquair (JST), which is characterised by a unilateral temporal visual field defect or, more rarely, nasal hemianopsia due to compression of the ipsilateral optic nerve where it connects to the chiasm.^8–10 Even though the brain MRI showed asymmetrical compression of the left optic nerve in our patient, we could not demonstrate that the right optic nerve was affected on imaging. Furthermore, a retrospective analysis of the thickness of the ganglion cell complex (GCC) on the OCT scan showed predominant thinning of the nasal retina of the right eye, which could explain the temporal visual field deficit in that eye and could be consistent with a JST. In contrast, the thickness of the GCC in the left eye was normal. Miller et al. have shown that GCC involvement correlates with the visual field defect and with the location of the lesions in cases of optic chiasm compression from pituitary tumours.¹¹ However, if we consider the visual findings in our patient’s left eye (superior temporal reduction in sensitivity), the clinical picture would be bilateral, and the concept of JST would not apply. In keeping with this, we hypothesise that an anterior junctional scotoma occurred due to compression of the patient’s right optic nerve (given that the visual field alterations related to the thinnest nasal fibres and the thinning of the temporal and superior fibres with no repercussion on the visual field), of the left optic nerve (more pronounced on the MRI), and of the optic chiasm centrally (possibly the nasal fibres originating from both eyes, since the hemihypokinesia was bitemporal).

Heddaeus¹² and Wilbrand¹³ were the first to describe the phenomenon of pupillary hemihypokinesia, coining the term “hemianopic rigidity.” Two years later, Wernicke¹⁴ used the term ‘hemianopic pupillary reaction’ and added the hypothesis that this phenomenon would only occur in pregeniculate lesions. This idea was later refuted by other researchers in light of modern technology,^3,4,15,16 which has resulted in a considerable decrease in publications on the topic of pupillary hemihypokinesia in the medical literature. The “hemianopic pupillary reaction” became known as “Wernicke’s hemianopic phenomenon.”¹⁶ We now add three new observations: 1) how to proceed in relation to the untested eye during the manoeuvre; 2) the reversibility of the phenomenon; and 3) the presence of the hemianopic phenomenon in chiasmal lesions.¹⁴

Notably, hemihypokinesia and pupillary hemiakinesia have always been described in classical studies with pupillometry with the obligatory presence of a visual field defect proportional to the pupillary motor defect.^3,4 However, several cases have been reported since the 1920s of hemihypokinesia and/or an RAPD without a corresponding visual field defect. These patients presented with structural damage exclusively affecting the brachium of the superior colliculus or its synapses in the pretectal nucleus/pretectal region (pretectal afferent pupillary pathway). Such dissociation is expected, as the afferent pupillomotor fibres are anatomically separate from the visual fibres at these locations.^7,16 It is currently accepted that both an RAPD and hemihypokinesia can also occur with retrogeniculate lesions and without involving the classical pupillary reflex arc, as the pupillary light reaction is controlled by two systems: subcortical (pregeniculate) and cortical (suprageniculate neurons and visual cortex).^7,16,17 Thus, the pupillary response encompasses the subcortical and cortical visual pathways with functions and interactions that are still not fully understood.

Based on our review of the literature obtained from the databases (PubMed/MEDLINE and Embase) and from the Google Scholar, we conclude that the present work highlights two new findings: 1) the presence of bitemporal hemihypokinesia in the absence of hemianopia in a chiasmal lesion; and 2) the reversibility of the referred sign after treatment, pointing to a probable causal relationship with the compressive lesion. It is conventionally accepted that there is no separation between the visual afferent fibres and the pupillomotor fibres in the optic chiasm. Therefore, this dissociation between our patient’s visual field defect and the bitemporal pupillary hemihypokinesia show us, once more, the complexity of the optic chiasm.^{1,2,5–7,15,16} Regarding the reversibility of the hemihypokinesia after treatment, we did not find any work that specifically reported this fact; however, Weinstein et al. studied the pupillary cycle time in 11 patients (nine with chiasmal tumours and bilateral visual field defects), before and after treatment, without isolating the hemiretinas during the tests.¹⁸ They showed that there was a reversal of changes in the pupillary cycle time after surgical decompression and that it could serve as a prognostic marker. Recent works have shown that in patients who underwent surgical chiasmal decompression, postoperative visual field improvement was correlated with the thickness of the preoperative GCC, showing that those with greater preoperative GCC thickness had better postoperative visual fields results.^9,19

There was no correlation of the bitemporal hemihypokinesia with the analysis of the GCC thickness in our case, especially in the left eye. This suggests that the indicated semiological sign occurs earlier than the involvement of the GCC on the OCT, pointing to the potential use of bitemporal hemihypokinesia as a screening test.

Additionally, we adapted Wernicke’s manoeuvre by occluding the untested eye during the examination of our patient, which prevented the light beam from directly stimulating that eye. By adapting Wernicke’s manoeuvre and applying it in this challenging case, we showed that, even in the age of technology, semiology is always a powerful tool in the hands of physicians.

The main limitation of our work is the scattering of light from the tested to the untested hemiretina during the examination with a flashlight (the light beam was not narrow); however, even modern methods of infra-red campimetry show that there is always a scattering of light between the hemiretinas.³ Furthermore, in some cases, the flashlight can test the phenomenon better than computerised pupillometry.¹⁶

Still, Papageorgiou et al. reported a case in which the patient had hemihypokinesia (upon computerised pupillometry) without hemianopia due to a lesion involving the brachium, but the examination of the hemiretinas using the slit lamp was normal.²⁰ On the other hand, examination with a flashlight has proven to be a powerful tool in low-income settings, as it is simple, cheap, reproducible, poses no risk to patients, and can be used in emergency and outpatient settings, especially in places that do not have ophthalmologists and/or slit lamps.

The main question of the present study is as follows: how can the dissociation between the presence of the bitemporal hemihypokinesia with the absence of a hemianopia (sensory-motor dissociation) in a chiasmal lesion be explained? We began with the following hypothesis: there is separation between the afferent visual fibres and the pupillomotor fibres in the chiasm, with the latter being responsible for the referred chiasmal sign (like an ‘alarm signal’) and superficially distributed in this anatomical structure, similarly to the parasympathetic fibres along the third cranial nerve. They would be, therefore, susceptible to extrinsic compressive lesions (which would be reversible after decompression).

Finally, we hope that other researchers will be able to address the main question of our study and assess the accuracy of the indicated chiasmal sign in the diagnosis and screening of patients with chiasmal diseases of different aetiologies by comparing the results of automated methods (e.g., pupillometry, pupillary cycle, GCC analysis) and non-automated methods (flashlight, slit lamp).

Funding Statement

The authors reported there is no funding associated with the work featured in this article

Declaration of interest statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/01658107.2022.2041674