Antifibrinolytic Therapy Use to Mitigate Blood Loss during Staged Complex Major Spine Surgery: Postoperative Visual Color Changes after Tranexamic Acid Administration

A 58-yr-old, 80-kg man presented in 2003 with new-onset back pain and was found to have a thoracic chordoma for which he underwent surgical resection. The patient underwent repeat surgery in 2004 for tumor recurrence. In 2006, leg weakness developed, and neuraxial imaging confirmed recurrence of his thoracic chordoma. Because of the progressive nature of his symptoms, he was scheduled for en bloc vertebral resection to be staged on two separate days.

Aside from his neurologic disease, the patient had no other significant medical history. After standard monitoring, general anesthesia was induced with intravenous midazolam and fentanyl, and muscle relaxation was achieved with vecuronium. After the patient’s airway was secured, general anesthesia was maintained with an infusion of fentanyl, midazolam, and atracurium. Blood glucose was strictly controlled (goal 80–110 mg/dl) with an insulin infusion. Because of anticipated dissection through multiple previously operated tissue planes, tranexamic acid (10-mg/kg loading dose, 1-mg · kg⁻¹· h⁻¹infusion) was used to minimize blood loss and transfusion requirements. The patient was placed in pinions, was positioned prone, and underwent T4–T8 laminectomies, mobilization of the epidural tumor from T6 to T8, and posterior spinal instrumentation from T2 to T11. The total time in the prone position was 9 h. The total tranexamic acid dose was 1,360 mg, and the estimated blood loss was 550 ml. The patient was awakened at the conclusion of surgery, moved all his extremities to command, and was transferred to the intensive care unit. He was extubated the following morning and had an uneventful recovery.

The second stage of the patient’s surgery was performed 38 h thereafter. Anesthesia was induced with midazolam and fentanyl, and muscle relaxation was achieved with vecuronium. General anesthesia was maintained with an infusion of midazolam, fentanyl, and atracurium. Tranexamic acid was again used (same dose as above). The patient was placed in the left lateral decubitus position (no pressure on eyes noted throughout case) and underwent en bloc tumor resection at T5–T7, spine stabilization with a titanium cage, and vascularized rib graft through a combined anterior (right thoracotomy) and posterior approach. In addition, a latissimus dorsi flap was used for hardware coverage. The total anesthesia time was 13 h. The total tranexamic dose was 1,840 mg, and the estimated blood loss was 1,300 ml. The patient was awakened, moved his upper and lower extremities to command, and was transferred to the intensive care unit sedated and mechanically ventilated. He remained hemodynamically stable, and the trachea was extubated the next day. After extubation, he was alert and oriented to person, place, and time; however, he reported color vision disturbances; specifically, all objects appeared green. As his color vision improved, his perception changed from objects being all green, to objects having a slight green tinge, to a green-tinged halo effect around objects that were now normal in color, to return of normal color vision. A formal ophthalmologic examination, including Ishihara color vision testing for both eyes, was normal. Unfortunately, this examination was performed after significant subjective improvement had already occurred in his color vision. By the second postoperative day, his color visual changes had dissipated. Throughout the remainder of his hospital stay, there was no recrudescence of his ophthalmologic symptoms. He was discharged to a rehabilitation facility on postoperative day 8.

We present an unusual case of transient color vision change after TEA administration during major spine surgery. To our knowledge, such an occurrence has not previously been reported in the medical literature.

It is known that TEA can cause ophthalmologic complications, including blindness. For example, two women receiving TEA for menorrhagia developed central venous stasis retinopathy-induced blindness after 1 week of therapy.1In both cases, visual acuity and funduscopic examinations returned to normal after concomitant TEA withdrawal and treatment with systemic corticosteroid and dipyridamole. In another case, a 56-yr-old dialysis patient experienced blindness on two separate occasions during prolonged TEA therapy for gastrointestinal hemorrhage management.2On both occasions, visual acuity returned after withdrawal of TEA, although some defect in night vision persisted. Because TEA is renally excreted, blindness was attributed to a TEA overdose in this particular case.2Whether the “TEA overdose” resulted in thrombosis of retinal circulation or a direct drug or metabolite effect on retinal cone cells remains uncertain. In yet another case, a patient experienced blindness while receiving TEA for hemorrhagic cystitis.3In all of the above cases, the etiology has yet to be fully elucidated. Although a thromboischemic mechanism would seem to be the most plausible explanation, there is no definite link between TEA and blindness.

In regard to color vision change, we were unable to find any case reports or clinical trials citing such observation. Interestingly, Micromedex (Thomson Micromedex, Greenwood Village, CO) cautions that “acquired defective color vision” is a contraindication to TEA use. However, Micromedex provides no substantiating laboratory or clinical evidence as to why this claim is made.

Color vision requires the intact function of several structures. Cone cells within the retina contain pigments that allow for differential responses to various light wavelengths. The optic tracts then transmit this data to the visual cortex of the occipital lobe. In addition, color vision is processed in area V4 of the temporal lobe. Disruption of function in any of these areas could presumably affect color vision.

In our case, ischemia as the cause of color vision change seems unlikely. An ischemic injury to any of the structures or pathways involved in color vision would also be expected to result in additional findings such as changes in visual acuity or visual field defects. Rather, the nature of our patient’s morbidity was not total loss of color vision, but a green tinge and green halo effect. Visual changes (e.g. , transient or permanent blindness) have been reported in patients after major spine surgery.4More specifically, the American Society of Anesthesiologists Practice Advisory Board has reported prolonged surgical duration (i.e. , exceeding an average of 6.5 h) and substantial blood loss (i.e. , when loss reaches an average of 44.7% of estimated blood volume) as high-risk contributing factors to developing postoperative visual acuity changes after spine surgery.5Regardless of whether these criteria were fulfilled, our patient experienced a completely different entity in that visual acuity was spared. Although it is impossible to definitively identify a pathogenic link between TEA and color vision disturbance, we speculate that our observation could represent a pharmacodynamic effect on one or more of the pigments involved in color differentiation by retinal cone cells. The likelihood of a pharmacodynamic effect of some kind is supported by the fact that the manufacturer lists acquired color vision defects as a contraindication to TEA use.

Last, this patient received several other drugs during his anesthetic, none of which seem likely causes of his impairment. Midazolam is not known to cause color vision problems per se . It does have hallucinogenic potential; however, our patient did not demonstrate any psychotic behavior, altered consciousness, or confusion associated with his color vision impairment that would indicate hallucination. Hallucinations due to fentanyl would be an unlikely cause for similar reasons. Other side effects of fentanyl do include blurred vision, but not color vision disturbance. Vecuronium and atracurium are not known to have visual side effects.

In summary, we report an unusual case of transient color vision disturbance after TEA administration in a patient undergoing staged complex major spine surgery. Although the etiology has yet to be fully elucidated, we speculate that our observations resulted from a pathogenic link between TEA and retinal cones.