Journal of Neuro-Ophthalmology, March 1983, Volume 3, Issue 1

Occipital lobe infarction after open heart surgery.

The most common permanent neuro-ophthalmologic complication of cardiopulmonary bypass surgery is visual loss. Bilateral lower altitudinal visual field defects were documented in a patient who noted blurred vision following open heart surgery. A difference of opinion existed as to whether the field defects were due to retina-optic nerve or occipital lobe lesions. Two points are emphasized in this report: 1) the field defects were much easier to define on the tangent screen than on Goldman perimetry, and 2) occipital coronal high resolution CT scan confirmed bilateral upper bank calcarine cortex infarctions in this patient. Occipital coronal, thin-section, high-resolution computed tomographic scans are helpful in studying patients with occipital lobe visual field defects.

]. Clin. Neuro-ophthalmoJ 3: 2.3-30, 198.3. Occipital Lobe Infarction after Open Heart Surgery ]. LAWTON SMITH, M.D. SHELLEY A. CROSS, MD. Abstract The most common permanent neuro-ophthalmologic complication of cardiopulmonary bypass surgery is visual loss. Bilateral lower altitudinal visual field de­fects were documented in a patient who noted blurred vision following open heart surgery. A difference of opinion existed as to whether the field defects were due to retina-optic nerve or occipital lobe lesions. Two points are emphasized in this report: 1) the field defects were much easier to define on the tangent screen than on Goldman perimetry, and 2) occipital coronal high resolution CT scan confirmed bilateral upper bank calcarine cortex infarctions in this patient. Occipital coronal, thin-section, high-resolution com­puted tomographic scans are helpful in studying pa­tients with occipital lobe visual field defects. Visual loss following coronary artery bypass surgery has received much attention in the litera­ture as the number of such cases has increased. I-I Various mechanisms have been proposed to ex­plain this complication These have included em­bolism. cerebral hypoperfusion, ischemic optic neuropathy, and combinations of such lesions. The patient described below noted blurred vision im­mediately after open heart surgery and presented with unusual visual field defects. Opinions varied as to whether the field defect was due to retinal­optic nerve ischemia or was secondary to occipital lobe involvement. Coronal computed tomography of the occipital lobes W.:JS helpful in confirming bilateral upper bank calc.:Jrine infarctions in this patient. The use of vertical prism glasses in helping patients with this type of altitudinal field defect is also documented. Case Report A 56-year-old, right-handed white male W.:JS first seen on June 3, 1982, through the courtesy of Dr. T. C. Spoor. The chief complaint was blurred vision since October 1981. He enjoyed good health From the Bo"com Palmer Eye In,tilllte. DepoHtmenl l)f Uph­lh. llmok)gy. Universily of Miami School of Medicine. Mi.lIni. FkHidol (jLS); Jnd the DepoHlnwnt of !\:(,IIrc>logy. The Molvl' Clinic. RllChl'>ll'r. MlIlneSOlJ (SAC). March 1983 until March 1981, when he suffered a myocardial infarction I day following surgical repair of an anal fistula. Seven months later, he was hospitalized to reevaluate his cardiac status. Cardiac catheteriza­tion revealed significant coronary artery disease, and surgical bypass was advised. The patient was operated on October 26, 1981, and had a four­vessel coronary artery bypass and excision of a myocardial aneurysm. He tolerdted the surgery uneventfully to his knowledge; postoperatively, however, he noted immedidtely that something was wrong with his vision. The patient was not particul.lrly alert dfter sur­gery and was sedated for a time; however, while in the intensive care unit he recdlled that when he would try to pick up a newspdper and look at it, the entire print would suddenly disappear. He called this a "whiteout." Within a few seconds vision would return, and he described this vividly as "like the tide rolling back in." This pattern recurred for .1 day or two, and when he mentioned it to his physici~ns, they became concerned. He had several tests, including an electroencephalo­gram and a CT scan. He was seen by a neurologist and an ophthalmologist, and apparently there was some difference of opinion as to whether the prob­lem was in his eyes or in his brain. Within a short time, the blurred vision began to slowly improve. When he alternately covered his eyes with his hands (which he did frequently). the problem appeared to him to be ex.Ktly the S.mle in each eye. He noted that as vision returned, he could recognize the first two letters of .1 six-letter word, and bv knowing these letters could then deduce what' the word was. The patient W.1S dis­charged from the hospital in 3 weeks. In J.muary 1982, he was rehospitalized after some p.llpit.1tions. Thereafter, he considered his visu.11 problem st.l­ble. Because of a visually demanding occupation, his difficulty in reading prompted him to seek another opinion. He also described .1 slight problem with gait. When walking, at times his right leg might "tingle" and tend to deviate out. The veins grafts h.ld been removed from his left leg, not the right. Also, at times the right arm fE'lt like it was going to sleep, but this too was improving. He had absolutely no other complaints. He W.:JS taking Isordil, Inderal, Lanoxin, Valium, and Ascriptine. 23 Occipital Lobe Infarction Figure 1. Goldmann visual field of right eve. Februarv 23. [°82. The patient had been examined elsewhere, and Goldmann visual fields were performed (Figs. 1 and 2). These were interpreted as due to a retinal­optic nerve lesion because no categoric vertical hemianopic shift was detected. He was then re­ferred to the Bascom Palmer Eye Institute for an­other neuro-ophthalmologic opinion. Examination on June 3, 1982, revealed a cor­rected vision of 20/25+3 in the right eye, readin~ slowly letter by letter. The left eye was 20/30-1, agains slowly letter by letter or number by number. With +5.50 spheres he could read J-I print with each eye on a Titmus card at 8~ inches. Although he read slowly, and generally only word by word, with sufficient plus correction he read fairly well. Therefore, there was no apparent organic dyslexia, and his reading difficulty was attributed to the visual field defects. External eye examination was not remarkable. The pupils and ocular motility were intact. Slit lamp and dilated indirect ophthal­moscopic examinations were normal. The applan­ation tensions were 14 in both eyes, and his blood pressure was 130/90. The primary point of interest was related to the visual fields. On Amsler grid testing, the patient drew a line across the lower portion of the graph 24 paper and stated he simply did not see below this line with either eye. This was symmetrical in the two eyes. Peripheral fields were normal to 3/330 white and 1/330 white on the Aimark perimeter in both eyes (Fig. 3). The peripheral fields, therefore, to larger isopters were exactly the same as had been plotted on the Goldmann perimeter. On the tangent screen, however, interesting central field defects were seen. There was an absolute scotoma in both eyes-the same to a 5/1000 white projec­tion light pointer as to an 18/1000 white formed Bausch and Lomb target (Fig. 4). The patient fixed very well. In the right eye, the field defect was seen to come out of the blind spot, definitely crossing the midline. At first blush, it looked like an optic nerve type arcuate scotoma. On the tangent screen, however, there was a nasal-temporal step compa­rable to the midline step previously emphasized by Dr. Max Chamlin. Therefore, the lower altitudinal arcuate scotoma came down about 12° lower on the right half of the field than on the left half, and was congruous. This was not only an absolute scotoma (the same in size to large and small tar­gets). but it was completely congruous. It could be studied more easily by testing him not only at 2 meters. but also at the end of the examination Journal of Clinical Neuro-ophthalmology Smith, Cross Figure 2. Goldmann visual field of left eye, February 23, 1982. VISUAL FIELDS (peripheral) Figure 3. Aimark peripheral visual fields performed June 3, 1982. March 1983 25 OCCIJ:>ltc1! Lobe InLHclion (CeD t ra 1) Wt 5/1000 W & 18/1000 W Right. o Figure 4. Tangent screen visual fields performed June 3. 1082. Note the nasc11-temporal vertical step at the lower margin of the field defects room (about 10 ft. away) with both formed targets, colored Mydriacyl bottle tops, and a projector light. It was then evident that this was a bilateral occipital lobe visual field defect. The neuroradiologist was called and a specific request made for occipital coronal pictures, as clin­ically the field defect indicted the upper bank of the calcarine cortex on both sides. A previous CT scan had been reported as negative. Therefore, a careful occipital coronal study was performed, and the clinical impression was confirmed (Figs. 5-9) Bilateral upper bank calcarine hypodensities were seen which spared the lower bank of the calcarine cortex. These softenings were near the tip of the occipital lobes, and, therefore, were consistent with the field defects being near fixation (i.e., towards the macula) and sparing the more anterior occipital cortex (i.e., the peripheral fields). The CT scan also showed that there was some slight tongue of exten­sion into the lower bank, particularly on the left side, which might have indicated an extension of the field larger than the defect plotted. However, the occipital coronal studies established the clinical diagnosis in this patient. A final interesting point is that the patient had difficulty ambulating because of the bilateral lower altitudinal occipital lobe field defects. He was given a pair of -0.75 spheres (his distance manifest refraction) with 7-diopters base down ground in before both eyes. In addition, he was given a 26 Figure 5. A'ic11 computed tomographic scan, June 4, 1982, showing bilateral hvpodensities at occipital poles. prescription for +5,50 spheres for reading only. The patient later reported that his reading was better with the +5.50 spheres, but he had difficulty Journal of Clinical Neuro-ophthalmo!ogy Smith, Crus;, Figure 6, Higher cut of axial computed tomographic scan showing occipital lobe involvement. Figure 7. Coronal occipital computed tomography, June 4, 1982. The large white arrows show bilateral hypodense areas in upper banks of the calcarine cortex. The black arrow shows that the lower bank is intact. "t" = the torcula. with the distance pair. The use of vertical prism glasses for bilateral altitudinal field defects is of interest and must be carefully adjusted and ground for each patient. This point is mentioned to help the clinician keep vertical prism glasses in mind for a therapeutic trial in individuals with bilateral altitudinal field defects. March 1983 Discussion As open heart surgery or cardiopulmonary by­pass surgery has become more widely practiced in the past decade, the interest in neuro-ophthalmo­logic complications following such procedures has steadily increased. The pathogenesis of ocular and cerebral complications of open heart surgery has come under increasing scrutiny because of the desire to try to prevent such formidable postoper­ative sequelae. It appears the visual loss is the most common permament extracardiac complication fol­lowing cardiopulmonary bypass surgery. Gilman1 examined 35 patients before and after open heart surgery. Twelve patients (34%) had neurological deficits, and five of these died. An­other six died either without deficits or before they could be examined. Deficits included gnostic dis­orders, hemiplegia, visual field defects (right infe­rior hemianopia with Riddoch phenomenon in the blind quadrants, small bilateral central scotomas resolving into homonymous paracentral defects), and seizures. Etiologies were insufficient cerebral blood flow and embolism Tufo et al. 2 did a prospective study of 100 pa­tients. Neurological, psychometric, and behavioral testing was performed before and after open heart surgery. Fifty percent of the patients had neurolog­ical damage following recovery from anesthesia, 43% had behavioral abnormalities preceded by focal neurologic damage and depressed intellectual function, and 15% of the survivors had neurological damage at the time of discharge from the hospital. 27 Occipital Lobe Infarction 28 Figure 8. OCCipital coronal CT scan only slightly more rostral to Fig. 7, showing that the occipital infarctions were larger near the pole and decreased in size as one went more rostral in occipital lobe. The black arrow shows the upper bank infarction; the small black arrowhead shows intact lower calcarine bank; and the small black arrowheads show the cerebellum. Figure 9. Enlarged occipital coronal CT scan made June 4, 1982. Small white arrowheads are in the upper bank softenings; T = torcula Hereophili; t = tentorium cerebelli; c = cerebellum. Journal of Clinical Neuro-ophthalmology Cerebral damage was related to increasing age and depression of arterial pressure. Neuropathological changes suggested ischemia. Williams:) examined seven patients before and after surgery. The intriguing part of her study was that she noted intravascular retinal changes occur­ring during open heart surgery by performing ophthalmoscopy during the procedure. White plugs, some forming in situ, were seen to traverse the retinal circulation. Microinfarcts were docu­mented by fluorescein angiography. Refractile specks were also seen. Necropsy material sup­ported the interpretation that these plugs were formed from blood constituents. Brennan et al. 4 evaluated five dogs undergoing partial bypass. They observed that bubble oxygen­ators generate microemboli which injure the brain during bypass. Significant depression of cerebral blood flow and metabolism accompanied even rel­atively short pump time and was not immediately reversible. The depression was microembolic in origin and could be largely avoided by effective filtration. Aguilar et al. 5 reported on neuropathologic com­plications of cardiac surgery, studying 206 patients who died after open heart surgery and eight who died after cardiac bypass procedures. The bypass procedures had twice the morbidity of other open heart procedures. Widely scattered fresh subarach­noid hemorrhage and intracerebral hemorrhages, acute ischemic neuronal necrosis and focal infarcts, and acute neuronal degeneration were noted. The cause was reduced cerebral blood flow and fibrin platelet emboli, fat emboli (reduced by better fil­ters), and polarizable matter (not seen with good filters). The triad of emboli, acute hemorrhage, and focal or diffuse neuronal necrosis represents the postoperative syndrome of neuropathologic com­plications. The lesions were caused by anoxia, air embolism, hypotension, fibrin, and platelet or an­tifoam microemboli. Changes were seen in 85% of 206 patients undergoing open heart surgery and eight patients undergoing cardiac bypass. Patterson et al.6 studied dogs submitted to car­diopulmonary bypass under conditions generating bubbles and microemboli. Postoperatively, im­paired consciousness and ataxia were seen. These deficits cleared within 1 week. Multiple filling defects were seen in the microcirculation of the brain, demonstrated by ink injection before sacri­fice. The vascular pattern was normal after 1 week. Neuropathologic findings were restricted to the cerebellum in those studied later. Muraoka et al. 7 studied 57 children operated on with mild hypothermia and high-flow bypass, and performed CT scanning before and after surgery. There were 27 bubble oxygenator and 18 mem­brane oxygenator cases. In the bubble group, all 14 with 20 micron filters had normal CT scans. Four March 1983 Smith, Cross of 13 with 40 micron filters or no filters had decreased brain mass on postoperative CT scans. In the membrane group, all postoperative CT scans were normal. Sweeney et al. B described ischemic optic neurop­athy in seven of 7685 bypass patients. Blood loss with hypotension was the proximate cause in four of the seven cases. These authors stated that the optic nerve head behaves like an end organ. Hy­pothermia with decrease in oxygen consumption, also with a decrease of cerebral blood flow of 6 to 7% per degree and increased viscosity and sludging of blood were implicated as well. They proposed that nylon mesh bubble oxygenators might pro­mote activation of complement. It is evident from these reports that neuro­ophthalmologic complications 1) frequently follow cardiopulmonary bypass procedures, 2) involve the microcirculation, and 3) fortunately, the postoper­ative deficits, when documented, usually clear. However, loss of vision or visual field appears to be the most prevalent permanent neuro-ophthal­mologic complication following such procedures. It is evident that there are many contributing fac­tors to the development of retinal optic nerve and brain infarcts in such patients. The calcarine cortex is said to be the most sensitive area of the brain to hypoxia, and this may explain the frequency of permanent visual field defects in patients following open heart surgery. It is obvious that retinal and optic nerve involvement also are not infrequently seen, and in our experience a combination of an­terior and posterior visual loss mechanisms have been encountered. The most common cause of a monocular altitu­dinal visual field defect is optic nerve disease (e.g., ischemic optic neuropathy). The most common cause of binocular altitudinal visual field defects, when not occurring as two different bouts of optic nerve disease separated by a definite time, is either chiasmal or occipital lobe disease. These are usu­ally easy to differentiate in that chiasmaI lesions commonly produce asymmetric drops in visual acuity, will lead to optic atrophy in one or both eyes, and usually show diagnostic chiasmal visual field syndromes. Unilateral occipital lobe lesions do not produce a drop in visual acuity, whereas bilateral occipital lobe lesions produce strikingly symmetrical reductions in visual acuity, and the optic discs remain clinically uninvolved. We are excluding the late instances of transsynaptic degen­eration. In any patient complaining of blurred vision following open heart surgery, a careful neuro­ophthalmologic examination is indicated. The measurement of visual acuity in each eye, a careful examination of the pupils, a confrontation field, and a careful look at the optic disc and retina for cotton-wool spots, bright plaques, or other retinal 29 Occipital Lobe Infarction emboli is important. An Amsler grid test may be done at bedside. A "history" field is important to see if the defect is notably congruous (i.e., exactly the same in the two eyes). When feasible postop­eratively, a high-resolution CT scan of the brain with emphasis on the occipital lobes and specifi­cally occipital coronal views may be invaluable. The near acuity should be measured not only with isolated letters and numbers (as the Rosenbaum near card), but also with text (as the Titmus or Sloan cards) in order to help make a differentiation between acuity drop, visual field defects, and or­ganic dyslexia syndromes. A neurological exami­nation is also obviously indicated. Finally, the use of hemianopic Fresnel prisms may be considered. A previous report9 documented the use of these prisms in hemianopias. * It should be emphasized that not all patients with hemiano­pias like hemianopic Fresnel prisms. Those with notable neurologic deficits (e.g., left hemisphere infarcts) usually are not happy with them (as they do not restore their ability to read). Patients with isolated hemianopias (e.g., right occipital lesions) often definitely like them, and they are worth a trial in such cases. The incidence of permanent loss of vision or visual field defects following open heart surgery must be more precisely defined. Such reports should include meticulous detail as to intraopera­tive phenomena (cerebral blood flow, hypotensive * A recent communication from Optical Sciences Group. 1331 Commerce Street, Petaluma, California 94952; 1-800-227-2254 established that they will be making some 40-diopter Fresnel prisms. The largest ones now available arc 30 diopters. The price quoted for these by Mrs. Astrid Harper was $5,75 each (in lots of 1-23) or $4.31 each (in lots of 24 or more), The price quote was effective August 27. 1982, 30 episodes, size and types of filters employed, serum complement studies, bubble oxygenators, etc.). Further intraoperative careful fundus examinations (as in the study by Dr. Isla Wiliams) also may be helpful in learning how to decrease the incidence of this problem, References 1. Gilman, S.: Cerebral disorders after open heart op­erations. N. Eng/. /. Med. 272(10): 489-498, 1965. 2. Tufo, H.M., Ostfeld, AM., and Shekelle, R.; Central nervous system dysfunction following open-heart surgery. /. Am. Med. Assoc. 212(8): 1333-13, 1970. 3. Williams, I.M,: Intravascular changes in the retina during open heart surgery, Lancet 688-691, 1971. 4. Aguilar, M.J., Gerbode, F., and Hill, J.D.; Neuro­pathologic complications of cardiac surgery. /. Tharac. Cardiovasc. Surg. 61(5): 676-685, 1971. 5. Patterson, R.H., Rosenfeld, L., and Porro, R.S.: Tran­sitory cerebral microvascular blockade after cardio­pulmonary bypass. Thorax 31: 736-741, 1976. 6. Muraoka, R., and Yokota, M.: Subclinical changes in brain morphology following cardiac operations as reflected by computed tomographic scans of the brain, /. Tharac. Cardiavasc. Surg. 81: 364-369, 1981. 7. Sweeney, P.I., Breuer, AC, et al.: Ischemic optic neuropathy: A complication of cardiopulmonary by­pass surgery. Neurology 32: 560-562, 1982. 8. Smith, J.L., Weiner, I.G., and Lucero, AJ.: Hemi­anopic fresnel prisms. /. Clin. Neuro-ophtha/mal. 2: 19-22, 1982. Acknowledgments The authors thank Dr. T. C Spoor for referring this patient and for providing the Goldmann visual fields of February 23, 1982; and Dr. Robert Quencer and staff for the occipital coronal computed tomography studies. Write for reprints to: J, Lawton Smith, M, D., Bascom Palmer Eye Institute, P.O. Box 016880, Miami, Florida 33101. Journal of Clinical Neuro-ophthalmology