Portal de Periódicos da CAPES

Postoperative visual loss following prone spinal surgery

2005; Elsevier BV; Volume: 95; Issue: 2 Linguagem: Inglês

10.1093/bja/aei173

1471-6771

Damon Kamming, Stephen Clarke,

Anesthesia and Pain Management

Postoperative visual loss is a rare but disastrous complication that has an estimated incidence of 0.01–1% after non-ocular surgery. We report a patient who underwent a prolonged spinal operation in the prone position and complained of blindness in one eye postoperatively. We consider the potential aetiological factors contributing to this unilateral postoperative visual loss and suggest strategies to reduce the incidence of this complication in spinal surgery. Postoperative visual loss is a rare but disastrous complication that has an estimated incidence of 0.01–1% after non-ocular surgery. We report a patient who underwent a prolonged spinal operation in the prone position and complained of blindness in one eye postoperatively. We consider the potential aetiological factors contributing to this unilateral postoperative visual loss and suggest strategies to reduce the incidence of this complication in spinal surgery. Postoperative visual loss after non-ocular surgery is a rare but devastating complication with an estimated incidence varying from 0.01 to 1% depending on the type of surgery.1Roth S Thisted RA Erickson JP et al.Eye injuries after non-ocular surgery: a study of 60,965 anaesthetics from 1988 to 1992.Anesthesiology. 1996; 85: 1020-1027Crossref PubMed Scopus (211) Google Scholar, 2Warner ME Warner MA Garrity JA et al.The frequency of perioperative vision loss.Anesth Analg. 2001; 93: 1417-1421Crossref PubMed Scopus (118) Google Scholar, 3Williams EL Hart Jr, WM Tempelhoff R Postoperative ischemic optic neuropathy.Anesth Analg. 1995; 80: 1018-1029PubMed Google Scholar, 4Kumar N Jivan S Topping N Morrell AJ Blindness and rectus muscle damage following spinal surgery.Am J Ophthalmol. 2004; 138: 889-891Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar The three recognized causes of postoperative visual loss are ischaemic optic neuropathy, central retinal artery thrombosis and cortical blindness. Ischaemic optic neuropathy is the most frequently cited cause of postoperative visual loss following general anaesthesia1Roth S Thisted RA Erickson JP et al.Eye injuries after non-ocular surgery: a study of 60,965 anaesthetics from 1988 to 1992.Anesthesiology. 1996; 85: 1020-1027Crossref PubMed Scopus (211) Google Scholar. Ischaemic optic neuropathy following spinal surgery in the prone position has been reported previously, although the incidence and risk factors are not fully appreciated by many anaesthetists and surgeons.5Cheng MA Sigurdson W Tempelhoff R Lauryssen C Visual loss after spine surgery: a survey.Neurosurgery. 2000; 46: 625-630Crossref PubMed Scopus (93) Google Scholar, 6Myers MA Hamilton SR Bogosian AJ et al.Visual loss as a complication of spine surgery: a review of 37 cases.Spine. 1997; 22: 1325-1329Crossref PubMed Scopus (233) Google Scholar, 7Ho VT Newman NJ Song S Ksiazek S et al.Ischemic optic neuropathy following spine surgery.J Neurosurg Anesthesiol. 2005; 17: 38-44PubMed Google Scholar, 8Lee LA Lam AM Unilateral blindness after prone lumbar spine surgery.Anesthesiology. 2001; 95: 793-795Crossref PubMed Scopus (61) Google Scholar, 9Abraham M Sakhuja N Sinha S Rastogi S Unilateral visual loss after cervical spine surgery.J Neurosurg Anesthesiol. 2003; 15: 319-322Crossref PubMed Scopus (27) Google Scholar Recognized preoperative risk factors include hypertension, diabetes, polycythaemia, smoking, renal failure, narrow-angle glaucoma, atherosclerotic vascular disease and collagen vascular disorders.10Roth S Barach P Postoperative visual loss: still no answers yet.Anesthesiology. 2001; 95: 575-577Crossref PubMed Scopus (77) Google Scholar Recognized intraoperative risk factors for developing ischaemic optic neuropathy include hypotension and anaemia.11Brown RH Schauble JF Miller NR Anemia and hypotension as contributors to perioperative loss of vision.Anesthesiology. 1994; 80: 222-226Crossref PubMed Scopus (145) Google Scholar As the aetiology of postoperative visual loss remains incompletely understood, we discuss the details of a case of ischaemic optic neuropathy resulting in unilateral postoperative visual loss following prolonged spinal surgery in the prone position. A 60-yr-old man (180 cm, 75 kg, BMI 31.25 kg m2) presented for elective decompressive laminectomy with instrumented cervical thoracic fusion of C3–T5 for treatment of an invasive metastatic lesion of C7 with bony collapse and spinal cord impingement. His symptoms consisted of pain over the C7 dermatome and weakness and muscle wasting in his dominant right hand. He had a past medical history of medullary carcinoma of the thyroid for which he had been treated with a total thyroidectomy in 1988 followed by multiple neck operations for local recurrences. He had also received a course of radical external beam radiotherapy to the neck and upper mediastinum. There was further recurrence of the tumour in 2002 for which he underwent resection of the mediastinal lymph nodes and oesophageal reconstruction. He had also been treated with palliative radiotherapy for a supraorbital metastasis and a metastasis at T12. Horner's syndrome affected his right eye following surgery in 2002. Consistent with this, his right pupil was small and unreactive whilst his left pupil was normal and reactive. Coincidentally it was noted that he had early cataracts of both eyes. The patient's regular medications were gabapentin, thyroxine, lansoprazole, alfacalcidol and dothiepin. He had undergone numerous previous general anaesthetics without any complications. He was a non-smoker and did not drink alcohol. On examination his arterial pressure was 128/70 mm Hg and his heart rate was 80 beats min−1. His preoperative 12-lead ECG showed a normal sinus rhythm. His preoperative haemoglobin was 15.8 g dl−1. Anaesthesia was induced with midazolam 2 mg, propofol 150 mg and a bolus of remifentanil 50 µg. Tracheal intubation with a size 8 reinforced oral endotracheal tube was facilitated with atracurium 50 mg. Three 14G peripheral intravenous cannulae were sited and a triple lumen central line was inserted into the femoral vein. An arterial cannula was sited in the right dorsalis pedis artery. The patient's eyes were protected with chloramphenicol ointment before being taped shut and padded. A Mayfield clamp was positioned and a urinary catheter was inserted before the patient was positioned prone on a Montreal mattress. His neck was maintained in the midline position with slight flexion, so that his back was in a neutral/horizontal position, resulting in the head being slightly dependent. The Mayfield clamp supported the head without any pressure on the orbits. Tranexamic acid 1 g i.v. was given at the start of surgery. Anaesthesia was maintained with isoflurane 0.9–1.2% and an infusion of remifentanil, initially at 0.2 µg kg−1 min−1 for ∼1 h and then at 0.1 µg kg−1 min−1 for the next 6 h, to maintain the systolic blood pressure between 90 and 110 mm Hg (within 28% of the preoperative value). His heart rate remained between 65 and 80 beats min−1 throughout the procedure. Pressure-controlled ventilation maintained the PE'CO2 between 4.1 and 4.5 kPa. Arterial blood gases and haemoglobin concentration were monitored periodically (four times) throughout the 7 h procedure. The PaO2 ranged from 19.8 to 29.9 kPa, and the PaCO2 ranged from 4.5 to 4.8 kPa. The lowest intraoperative bicarbonate was 22.8 mmol litre−1 and the lowest base excess was −2.1. Intraoperatively, the haemoglobin concentration varied between 12.2 and 7.6 g dl−1. The rectal temperature varied between 36.4 and 37.2°C. An extensive decompressive laminectomy and spinal fusion of C3 to T5 was performed over a period of 7 h. The blood loss was estimated at 6000 ml and was replaced with 6000 ml of Hartmann's solution, 2000 ml of gelofusine, 2000 ml of eloHAES and 8 units of packed red blood cells. A further 1 g of tranexamic acid was given intravenously towards the end of surgery. Diamorphine 20 mg was given gradually towards the end of surgery. At the end of the operation, the patient was turned into the supine position, disconnected from the Mayfield pins and transported to the intensive care unit (ICU) with the endotracheal tube in place. On arrival in the ICU, his vital signs were as follows: arterial pressure 136/78 mm Hg, heart rate 88 beats min−1 and tympanic temperature 37.4°C. Arterial blood gas analysis showed a PaO2 of 40.8 kPa, a PaCO2 of 5.0, bicarbonate of 21 mmol litre−1 and a base excess of −3.4. The haemoglobin concentration on arrival in the ICU was 10.2 g dl−1. It was noted on admission to the ICU that the patient had significant facial swelling and bilateral conjunctival chemosis. He was cardiovascularly stable overnight, during which time he was electively ventilated before being extubated uneventfully 12 h postoperatively. On awakening after extubation on the first postoperative day the patient complained of reduced vision in his left eye, which he described as a 'snowstorm'. The patient was subsequently examined by an ophthalmologist on the second and twelfth postoperative days. He confirmed that the patient had 6/9 normal visual acuity in his right eye but could only detect hand motions with the left eye. Intraocular pressures were normal in both eyes. A dense afferent pupillary defect was demonstrated in the left eye. Anterior segments were normal in both eyes. Fundal appearances showed normal discs and maculae. There was no disc swelling or disc pallor in the left eye. No retinal haemorrhages were seen. In view of the above findings a diagnosis of left posterior ischaemic optic neuropathy was made. This diagnosis and the lack of possible treatment were discussed with the patient. However, he was assured that, given the good level of visual acuity in his right eye, he would still be allowed to drive. The patient made an otherwise unremarkable recovery from surgery and was discharged home. Ischaemic optic neuropathy is the most common cause of postoperative visual loss. The American Society of Anesthesiologists (ASA) closed claims database (www.asaclosedclaims.org) reports that 81% of postoperative visual loss is diagnosed as ischaemic optic neuropathy and the remainder as retinal artery thrombosis. Sixty-seven per cent of all the reported cases of postoperative visual loss occurred after spinal surgery in the prone position. Ischaemic optic neuropathy is caused by an ischaemic insult to the optic nerve and can be divided into anterior and posterior ischaemic optic neuropathy. The diagnosis depends on which part of the nerve is affected, since the anterior and posterior sections have different blood supplies. The posterior part of the optic nerve is furthest from an arterial supply and is most commonly implicated in postoperative visual loss associated with haemorrhagic hypotension. The posterior portion of the optic nerve has its main vascular supply from pial vessels derived from branches of the ophthalmic artery. These vessels are incapable of autoregulatory control and therefore this posterior portion of the optic nerve is particularly vulnerable to a fall in perfusion pressure or anaemia. Posterior ischaemic optic neuropathy can be a diagnosis of exclusion as there are often no abnormal ophthalmoscopic findings.12Rizzo III, JF Lessell S Posterior ischemic optic neuropathy during general surgery.Am J Ophthalmol. 1987; 103: 808-811Abstract Full Text PDF PubMed Google Scholar A diagnosis of ischaemic optic neuropathy following anaesthesia does not suggest that there was intraoperative extrinsic pressure on the eye. Postoperative visual loss secondary to intraoperative extrinsic pressure on the eye usually has a different opthalmological examination finding. In reported cases the diagnosis is commonly retinal artery thrombosis, with the fundoscopic appearance of retinal pallor with a characteristic 'cherry red spot'.6Myers MA Hamilton SR Bogosian AJ et al.Visual loss as a complication of spine surgery: a review of 37 cases.Spine. 1997; 22: 1325-1329Crossref PubMed Scopus (233) Google Scholar Several factors could have contributed to this man's postoperative visual loss. Two of the most commonly blamed risk factors are intraoperative hypotension and anaemia. In this case, blood pressure was tightly controlled with a remifentanil infusion, aiming for a systolic blood pressure of 90–110 mm Hg in an attempt to limit blood loss and the need for blood transfusion during surgery. Despite this, the patient lost ∼6000 ml of blood and required a significant blood transfusion (8 units). During the procedure, the lowest recorded haemoglobin was 7.6 g dl−1. Throughout the operation there was no evidence of reduced end-organ perfusion, as manifested by a metabolic acidosis, decreased urine output or ECG changes, suggesting that a moderate degree of hypotension and anaemia may not fully explain the postoperative ischaemic optic neuropathy. It is also recognized that a significant risk factor for postoperative visual loss is increased intraocular pressure. This can be attributed to a number of factors including direct pressure on the globe and increased central venous pressure (CVP). CVP is increased by reduced venous return in the head-down position and obstruction to venous outflow if the head is turned to one side. CVP is also increased if there is direct pressure on the abdomen due to malpositioning on an operating table compressing the abdomen and obstructing venous return to the heart. This patient's head was secured after induction of anaesthesia with Mayfield pins without any pressure on the globe and was positioned correctly on a Montreal mattress in a neutral position. The CVP varied between 6 and 13 mm Hg throughout and therefore might not be viewed as a single significant contributing factor to the visual loss. However, CVP readings may not necessarily accurately reflect venous pressure within the globe, and doubtless there could still be marked venous congestion of the head and neck with a normal CVP reading. It has been demonstrated that, because of the absence of venous valves, changes in CVP do translate into concomitant changes of ocular venous pressures and therefore can affect intraocular pressure.13Lee LA Vavilala MS Sires BS Intraocular pressure is partially dependent on central venous pressure during prone spine surgery.Anesthesiology. 2003; 99: A289Crossref PubMed Scopus (22) Google Scholar This patient had undergone previous neck surgeries and radiotherapy, which could also have resulted in abnormal venous drainage. He had previously had radiotherapy for a supraorbital metastasis, and this might also have had some effect on local venous drainage and therefore intraocular venous pressure. With regard to other patient risk factors that might have contributed, malignant disease in itself can sometimes produce a hypercoaguable state and affect platelet function. This may also have been compounded by the antifibrinolytic effect of tranexamic acid which could in theory alter local ocular haemodynamics. However, there is currently no clear-cut evidence that tranexamic acid is associated with an increased risk of thromboembolic complications.14Mahdy AM Webster NR Perioperative systemic haemostatic agents.Br J Anaesth. 2004; 93: 842-858Abstract Full Text Full Text PDF PubMed Scopus (125) Google Scholar Recently it has been demonstrated that the prone position itself significantly increases intraocular pressure.15Hunt K Bajekal R Calder I et al.Changes in intraocular pressure in anaesthetised prone patients.J Neurosurg Anesthesiol. 2004; 16: 287-290Crossref PubMed Scopus (80) Google Scholar A recent study in awake volunteers noted that there was a significant increase in intraocular pressure in all subjects when positioned prone. Careful attention was made to ensure no direct pressure on the globe or dependent positioning. It was found that a 10° reverse Trendelenberg position normalized the intraocular pressure whilst prone.16Ozcan MS Praetel C Bhatti MT et al.The effect of body inclination during prone positioning on intraocular pressure in awake volunteers: a comparison of two operating tables.Anesth Analg. 2004; 99: 1152-1158Crossref PubMed Scopus (92) Google Scholar In our case the head was positioned in a slightly dependent position to ensure a neutral/horizontal back position. In view of the above findings, this may be a significant contributory factor in increasing intraocular pressure, thereby reducing perfusion pressure to the eye. It is likely that a combination of mild to moderate hypotension, anaemia and increased intraocular pressure secondary to being prone with head slightly dependent all contributed to a reduction in perfusion pressure (the difference between mean arterial pressure and intraocular pressure) to the optic nerve intraoperatively and resulted in the postoperative visual loss. Even modest hypotension in combination with raised intraocular pressure can result in decreased ocular perfusion and optic ischaemia. In conclusion, anaesthetists need to be aware that reductions in ocular blood flow are caused by a number of factors which, when combined, can cumulatively increase the risk of postoperative visual loss. In future cases where the patient is in the prone position for any significant length of time, it may be prudent to consider a 10° reverse Trendelenburg position to reduce the increased intraocular pressure during prolonged surgery. Patients with preoperative risk factors (glaucoma, hypertension, diabetes, smoking and atherosclerosis) having prone spinal surgery at risk of severe blood loss are now known to be at significantly increased risk of postoperative visual loss secondary to ischaemic optic neuropathy, and consideration should be given to warning them of such a risk. Consideration should also be given to a higher transfusion trigger in prone spinal patients at risk of postoperative visual loss. With regard to hypotension, a safe lower limit for the perfusion pressure of the optic nerve remains unknown, and the risk–benefit ratio should be assessed for each patient based on his or her preoperative blood pressure. In addition, every effort must continue to be made to ensure there is never any extrinsic pressure on the eye in patients having surgery in the prone position to avoid the risk of postoperative visual loss secondary to retinal artery thrombosis Download .zip (.0 MB) Help with zip files Download .zip (.0 MB) Help with zip files