Visual evoked potentials monitoring in a case of transient post-operative visual loss.

Post-operative visual loss (POVL) is a rare, albeit potentially serious complication of general anaesthesia. This report describes the case of a 54-year-old woman who developed transient POVL after general anaesthesia following a left posterior parietal meningioma surgery in the prone position and discusses the usefulness of visual evoked potentials monitoring in such situations.

INTRODUCTION

Post-operative visual loss (POVL) represents a puzzling event for both anaesthesiologist and surgeon and may have a serious impact on the patient's visual prognosis after non-ophthalmic surgery.[1] The identification of individual risk factors remains difficult as no single factor has been shown to predispose to POVL.[2] Some factors (intraoperative blood loss, duration of the procedure) cannot easily be predicted at the beginning of surgery. The role of other factors such as prone position or history of chronic hypertension is still debatable.[2] We describe a patient who developed transient POVL after meningioma surgery in the prone position. This case report emphasises the usefulness of visual evoked potentials (VEPs) monitoring to evaluate POVL and to precisely locate the site of the injury.

CASE REPORT

A 54-year-old woman (height, 169 cm and weight, 74 kg) underwent elective surgery for a left posterior parietal meningioma. Her medical history consisted of arterial hypertension well controlled by nebivolol therapy (10 mg/day) that was not interrupted before surgery. The patient was free of visual complaints before surgery. The pre-operative anaesthesia evaluation found her to be in excellent condition. The patient was accepted in the American Society of Anesthesiologists’ physical status 1 with a Lee's cardiac risk index of 1. Pre-operative laboratory investigations were within normal limits. Pre-operative arterial blood pressure was 160/80 mm Hg. Haemoglobin concentration at the beginning of surgery, however, was 11.2 g/dL and haematocrit 35%. She was induced with controlled administration of propofol and sevoflurane. Cisatracurium was used to achieve orotracheal intubation with the help of a No. 8 endotracheal tube. Anaesthesia was maintained with propofol target-controlled infusion (target: 4-6 μg/ml) and continuous infusion of remifentanil (0.5-1 μg/kg/min). Central venous pressure (CVP) and arterial pressure were monitored after cannulation. She was then positioned for surgery in the prone position, with a mild inclination of the table towards a reverse Trendelenburg position. Her head was maintained in a Mayfield frame, with a mild flexion. There was no external compression of the ocular globes. During anaesthesia, end tidal CO2 was continuously kept in the range of 30 mm Hg and oxygen saturation always above 95%. The patient received 200 ml mannitol 15% for brain relaxation 1 h after the beginning of surgery. Her systolic and mean blood pressure was maintained in the range of 80-100 mm Hg and around 60 mm Hg, respectively, throughout the surgery [Figure 1]. Total duration of anaesthesia was 7 h and the patient remained in prone position for 6 h. Total blood loss was around 50 ml. Total fluid infused was 2000 ml crystalloids and intraoperative urine output was 700 ml. There was no use of vasopressors at any point. Extubation was performed 49 min after the surgery. The patient did not receive neuromuscular reversal agents. At the end of surgery, the haemoglobin concentration was 11.7 g/dL and the haematocrit was 36%. The patient was then transferred to Intensive Care Unit (ICU) for further management. On admission, her CVP was 8 mm Hg, systolic arterial pressure 135 mm Hg and mean pressure 85 mm Hg, and these values were maintained during the whole ICU stay. Fluid balance over the first 24 h was slightly negative (−300 mL). After recovery from anaesthesia in ICU, the patient began complaining of a significant decrease in visual acuity without any ocular pain. She was unable to count fingers at a distance of 2 m. The ophthalmological examination was otherwise unremarkable, except for a sluggish pupillary light reflex. There was no evidence of papilloedema on fundoscopy. Intraocular pressure was within normal range. The post-operative brain computed tomography ruled out ischaemic or haemorrhagic complications. The patient was assessed with VEPs on the 1st post-operative day. The examination consisted of flash VEPs (fVEPs) (day 1) and monocular pattern reversal VEPs (pVEPs) (day 1 and 5). The fVEPs were elicited by light-emitting diode goggles (monocular stimulation; red light; flash duration: 1000 μsec; rate: 1.7 Hz; analysis time 500 ms) and recorded at left and right occipital location (O1 and O2). The pVEPS were obtained on day 1 and day 5 by reversal of a black-and-white checkerboard (monocular stimulation, using 3 check sizes, respectively, of 1΀, 30’ and 15’ of arc, at a rate of 1.7 Hz). The pVEPs were recorded at the medial occipital location (Oz) referred to the central medial location (Cz). The fVEPs remained normal at day 1 and were not retested (data not shown). In contrast, the latency of the pVEPs showed significant slowing in both eyes, irrespective of the check size, with a typical paramacular profile (positivity-negativity-positivity) [Figure 2 above panel]. This alteration in the structure of the VEPs suggested that all nervous fibres in the optic nerve were not affected to the same extent. The latency and structure of pVEPs were normalised by day 5 [Figure 2, lower panel].

Figure 1

Haemodynamic monitoring during the procedure (APS: Arterial pressure systolic; APD: Arterial pressure diastolic; APm: Arterial pressure mean)

Figure 2

Pattern reversal visual evoked potentials on day 1 (upper panel) and day 5 (lower panel). Monocular recording of medial occipital activity was performed at different check sizes, respectively, of 1°, 30’ and 15’ of arc. LO = left eye; RO = right eye. Latency and structure of pattern reversal visual evoked potentials are altered on day 1 and returned to normal by day 5

Starting from post-operative day 2, the patient made a complete recovery of visual acuity. This was confirmed by ophthalmological examination at discharge.

DISCUSSION

There is an increasing awareness of the risk of POVL following a large variety of surgical procedures.[123456] Cardiac and spinal surgeries are mentioned as possible predisposing factors; however, the true incidence of serious events remains well below 0.1%.[16] Visual prognosis following POVL is usually poor. Therefore, it is prudent to find out the aetiopathogenesis of POVL so that appropriate treatment is accorded.[3]

In the present case, the probable cause of POVL seems to be an ischaemic optic neuropathy (ION). The specific mechanism and location of the vascular insult remain largely unknown. Visual loss occurs typically 24-48 h after surgery. Among the factors that were identified as increasing the risk of POVL, prone position and duration of surgery are the most commonly cited.[378] Both in normal volunteers and in patients, the increase of intraocular pressure in the prone position is time-dependent.[8] Among the additional risk factors, hypotension, haemodilution and vasopressor use are also of great importance.[13] In patients with chronic arterial hypertension, acute changes in arterial blood pressure may affect the perfusion pressure-dependant autoregulation. While deliberate hypotension is not recommended as a routine procedure in neurosurgery, the use of large doses of propofol may contribute to a decrease in arterial blood pressure during surgery. In addition, uninterrupted administration of β blockers until anaesthesia induction is now largely encouraged. It is difficult to define the blood pressure threshold below which hypotension may be deleterious. Moreover, hypotension itself is probably not sufficient to cause ION. In the present observation, although the mean arterial blood pressure was maintained around 60 mm Hg, this was probably the main triggering factor in a patient with a history of hypertension. Blood loss was limited, with neither haemodilution nor use of vasopressors.

Among advisable investigations, fundoscopy and VEPs appear particularly helpful to distinguish between anterior or posterior ION. This distinction is important for the prognosis that appears to be better in case of anterior ION. In both situations, pupillary light reflex is usually abolished. Papilloedema is noted in anterior ION but not in posterior. In severe forms of ION, anterior or posterior, fVEPs may reveal the absence of any retinal, primary or late occipital activity. fVEPs with recording of both occipital regions can also help in differentiating pre- from post-chiasmal visual dysfunction. In case of pre-chiasmal visual loss, the latency of occipital activities is symmetrically increased. In case of post-chiasmal visual dysfunction, the morphology and latency of occipital activities are asymmetrical. It is provided that ophthalmological examination can exclude diseases affecting the transparent optic media, such as severe cataract or uveitis. The use of full-field pVEPs recording is a sensitive technique to detect optic neuropathy and differentiate between peripheral and macular injury, when fVEPs remain unaltered, as in the case reported here.

Two other mechanisms of POVL should be excluded.[37] The first one is related to central retinal artery occlusion, which mainly results from external compression during surgery; visual loss is then related to ischaemia-reperfusion injury of retinal cells. Rarely, internal compression of the central retinal artery may be caused by the use of nitrous oxide for anaesthesia. The second mechanism is linked to thrombotic or ischaemic events within the occipital lobes during surgery, resulting in cortical blindness. The prognosis of these two types of injury is usually very poor as no specific therapy can be offered.

CONCLUSION

This case highlights the risks of visual impairment during prolonged neurosurgical procedure in prone position. Probably, low blood pressure during anaesthesia in this hypertensive patient could have initiated the ischaemic event resulting in transient POVL which was diagnosed by VEPs monitoring.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.