|
|
||||||||
Ann Thorac Surg 1996;61:660-666
© 1996 The Society of Thoracic Surgeons
Departments of Thoracic and Cardiovascular Surgery and Ophthalmology, Lahey Hitchcock Medical Center, Burlington, Massachusetts
Accepted for publication September 26, 1995.
| Abstract |
|---|
|
|
|---|
Methods. A rigorous analysis was conducted of all 602 patients who underwent operation during this period.
Results. No preoperative risk factors were identified. The development of anterior ischemic optic neuropathy was associated with prolonged cardiopulmonary bypass time, low hematocrit levels, excessive perioperative body weight gain, and the use of epinephrine and amrinone. Other hypothetical risk factors include systemic hypothermia, anemia, increased intraocular pressure, and microembolization. Treatment options include the use of corticosteroid medications, reduction of intraocular pressure, and optic nerve fenestration, although recent evidence and our experience indicate that the fenestration procedure is of no benefit.
Conclusions. Because all methods of treatment have had limited success, efforts to prevent this complication are of paramount importance.
| Introduction |
|---|
|
|
|---|
Anterior ischemic optic neuropathy (AION) is found most often in the fifth and sixth decades of life and is manifested clinically as sudden onset of painless unilateral or bilateral visual loss. Giant cell arteritis (arteritic ischemic optic neuropathy) is present in only 20% of these patients. In most patients, the cause is unknown (nonarteritic ischemic optic neuropathy). A small cup-to-disk ratio is an important anatomic risk factor [1]. Numerous conditions, including hypertension, diabetes, chronic migraine, atherosclerosis, anemia, and hemorrhage, have also been implicated to explain the pathophysiologic findings [2, 3], but many cases remain idiopathic.
The development of nonarteritic AION after cardiopulmonary bypass (CPB) was first described in 1982 by Sweeney and associates [4], and a few additional reports have followed [58]. The overall incidence of symptomatic AION after open heart operations at the Lahey Hitchcock Medical Center is less than 0.5%. However, an increased incidence of this complication (1.3% or 8 of 602 patients) was noted during the period from March 1, 1990, to March 1, 1992. Patients who underwent operation during this period were subjected to extensive retrospective analysis to identify predisposing factors.
| Material and Methods |
|---|
|
|
|---|
Numerous clinical, technical, and laboratory variables were analyzed. Preoperative variables included age, sex, history of hypertension, diabetes mellitus, obesity, hyperlipidemia, smoking, renal insufficiency (levels of serum creatinine >2.5 mg/dL), dialysis, and transient ischemic attacks or strokes. Operative variables included the type of procedure, urgency of operation (elective versus urgent or emergent procedure), CPB time, highest and lowest CPB flow rate, highest and lowest perfusion pressure, highest levels of carbon dioxide tension (PCO2), the use of corticosteroids (a bolus of methylprednisolone before initiation of CPB), and the use of inotropic or pressor medications. Postoperative variables included perioperative myocardial infarction (defined as new permanent electrocardiographic changes or the elevation of the MB fraction of creatine kinase to greater than 150 IU in the immediate 24-hour period after operation), the use of intraaortic balloon pump, weight gain in the first 24 hours after operation, development of transient or permanent neurologic deficit other than AION, postoperative atrial fibrillation, and acute renal failure. Laboratory parameters included lowest hematocrit and highest level of PCO2 during CPB.
The clinical course of AION was assessed with respect to type and onset of symptoms and the course of the disease. Complete ophthalmologic examination of each patient was performed, which usually included measurement of visual acuity, perimetry, and intraocular pressure, dilated fundus examination with indirect ophthalmoscopy, and slit-lamp examination of the anterior segment. Retinal photography and fluorescein angiography were performed in most patients, and several patients underwent electroretinography or B-mode ultrasonography of the orbit.
Data were analyzed using BMDP Statistical Software and were expressed as means ± standard deviation. Continuous variables were analyzed using an unpaired Student's t test. Categoric variables were analyzed using the Yate's corrected
2 test or the Mietinnen modification of the Fisher exact test. Probability values are two-tailed, with p less than 0.05 considered significant. Multivariate analysis of risk factors could not be performed because of insufficient adverse outcomes.
| Results |
|---|
|
|
|---|
A comparison of the clinical characteristics of the patients with AION and the control group is depicted in Table 1
. The groups were similar in age, sex distribution, and the frequency of preoperative risk factors.
|
|
|
Table 4
summarizes the ophthalmologic findings in these 8 patients. Intraocular pressures, measured at variable times after symptoms were noted, were not elevated. Examination of the optic disc usually revealed pallid edema in areas corresponding to the visual field deficit, occasionally associated with scattered nerve fiber layer hemorrhage. Three patients had some associated posterior pole retinal ischemia and splinter hemorrhages. An afferent pupillary defect was noted in all but 1 patient. Figure 1
is a representative photograph of the optic disks in patient 3, comparing the normal eye with the affected eye.
|
|
| Comment |
|---|
|
|
|---|
In the Shaw study [10] and others [8, 9, 11], the reported postoperative visual complications included retinal emboli, retinal infarcts, visual field deficits of cerebral origin (usually occipital infarcts), Horner's syndrome, and ischemic optic neuropathy. Of these complications, ischemic optic neuropathy is the least well understood and has probably been underreported in the past, with some cases incorrectly attributed to stroke or retinal embolism. Although it is infrequent, AION is now the most commonly reported cause of postoperative loss of vision for all surgical procedures [11]. In the first reported series of AION after coronary artery bypass grafting, Sweeney and associates [4] reported an incidence of 0.1% (7 of 7,685 patients).
Etiology and Pathophysiology
Although numerous etiologic theories have been proposed, it is now believed that AION develops secondary to interruption of the oxygen supply to the optic nerve head anterior to the lamina cribrosa and that it may follow a critical reduction in blood supply or oxygen-carrying capacity [1113]. The anterior optic nerve is supplied by multiple branches of the posterior ciliary arteries, forming the peripapillary choroidal plexus [14]. Several factors have been implicated in the reduction of blood flow in these arteries, including intrinsic small vessel disease secondary to diabetes mellitus and hypertension, prolonged periods of hypotension, and vasospasm induced by high levels of circulating catecholamines or other vasoconstrictors [24, 11]. A decrease in the red cell mass may result in a reduction of oxygen-carrying capacity, and the association of AION with hemorrhage is well known [3]. It has also been demonstrated that the cup-to-disk ratio in patients with AION is significantly lower than normal [1]. This ratio correlates anatomically with a narrow scleral canal, which may be more susceptible to a vicious cycle of ischemia, edema, and further ischemia. Katz and colleagues [15] provided further support to this anatomic hypothesis by demonstrating that myopic individuals are less likely than hyperopic individuals to experience AION.
In our series, the preoperative incidence of diabetes mellitus, hypertension, smoking, and hypercholesterolemia was not higher in patients who experienced AION. However, significant differences in certain operative and postoperative variables were identified. The duration of cardiopulmonary bypass was significantly longer in patients with AION compared with patients who were unaffected. It has been shown that cardiopulmonary bypass is associated with high levels of circulating endogenous vasoactive amines [16, 17]. Epinephrine and amrionine were also used more frequently in patients with AION. In our practice, these agents are only used to treat patients with severe postoperative low cardiac output syndrome. Thus, a combination of exogenous vasoactive drugs and prolonged exposure to endogenous amines stimulated by CPB or postoperative low cardiac output syndrome may have acted synergistically to produce vasoconstriction and ischemia in the posterior ciliary circulation. The use of a membrane oxygenator and pulsatile pump flow in all our patients did not appear to have a protective effect despite some evidence that these modes may be associated with decreased production of C3a and other endogenous vasoconstrictors [18, 19].
A striking feature common to all our patients with AION was a 10.0% to 26.0% gain in body weight that substantially exceeded that of the unaffected patients. This uptake of fluid resulted from long bypass times and the need to administer large volumes of fluid, both colloid and crystalloid, in the perioperative period. The acute development of tissue edema, particularly in patients with a narrow scleral canal [1], could result in progressive capillary compression and ischemia of the optic nerve. This condition may induce axoplasmic flow stasis [20] and lead to a vicious cycle of edema, ischemia, and nerve injury.
Some relationship may exist among perioperative edema, elevated intraocular pressures, and the development of AION after CPB [2126] Larkin and colleagues [23] demonstrated an increase in intraocular pressure during cardiopulmonary bypass that correlated with the degree of intraoperative hemodilution. Deutch and Lewis [24] reported that this effect may persist as long as 3 days after the operation and may correlate with weight gain. This effect may be noted particularly when crystalloid prime is employed, which results in decreased plasma oncotic pressure [25]. Furthermore, James and Smith [26] have demonstrated that assumption of the supine position, as would occur during CPB, results in a higher increase in intraocular pressure in patients with AION. Because the optic disk and peripapillary choroidal vessels are in a ``watershed'' area, they may be adversely affected by increased intraocular pressure, especially in the presence of small cup-to-disk ratios [21]. Katz and colleagues [21] have demonstrated higher levels of intraocular pressure in patients with AION, although this finding was not confirmed in subsequent studies by Kalenak and associates [22].
Patients with preexisting glaucoma and elevated intraocular pressures may be at higher risk for postoperative AION. One such patient has been reported to have undergone aggressive intraoperative monitoring and reduction of intraocular pressure by withdrawal of aqueous fluid [27].
Perioperative anemia is another possible etiologic factor contributing to AION. Low hematocrit levels were documented in all our affected patients and were significantly lower than in the unaffected patients. Low hematocrit levels are universal findings in modern cardiac operations using hemodilution prime. This condition may be tolerated in most patients, but profound anemia may enhance the adverse effects on oxygen delivery of other risk factors, such as excessive weight gain and tissue edema, use of vasoactive drugs, and prolonged CPB time. In critically ill patients with other risk factors for AION, it may be advisable to keep the hematocrit at levels of at least 25% to 30% [5, 28]. Hemoconcentration at the conclusion of CPB should also be considered to decrease postoperative weight gain (thereby resulting in less increase in intraocular pressure) and increase hematocrit levels without homologous transfusions.
Hypothermia, a technique used in most open heart operations, is generally thought to be protective of the brain but, in some instances, may contribute to ocular ischemia. Each centigrade degree decrease in body temperature is followed by a 6% to 7% decrease in cerebral blood flow [29]. Johnston and colleagues [30] demonstrated a 48% to 62% decrease in cerebral blood flow during prolonged hypothermic bypass procedures in a canine model, and this decrease persisted after the animal's temperature was raised to normal. They documented ischemic degeneration in the biopolar cells of the retinal inner nuclear layer in 67% of animals undergoing hypothermic CPB [30]. This cell layer is located in the vascular watershed boundary between the retinal and choroidal circulation and may be most susceptible to hypoperfusion. Plasma viscosity may also be increased during hypothermia [11]. In our study, the lowest bypass temperature was constant in all patients at 25°C, but more recently we have kept systemic temperature at 30°C.
Although no differences in levels of PCO2 between the AION and normal patients were documented in our series, alterations in cerebral autoregulation during CPB may be induced by inappropriate manipulation of the arterial PCO2. This manipulation may have adverse effects on cerebral perfusion, intracranial pressure, intraocular pressure, uncoupling of flow/metabolism, and microembolism [11, 30, 31].
Although rarely implicated in the development of AION, embolization to the optic nerve and retinal vessels may be an eitiologic factor in some patients [6, 7, 11]. Evidence of retinal embolization during CPB has been well documented in human and animal models using fluorescein angiography and retinal histopathology [30, 32]. Several of our patients had evidence of concomitant retinal ischemia.
Clinical Presentation and Diagnosis
Our patients presented with the classic feature of AION, a visual field deficit. The typical deficit is an inferior altitudinal hemianopia [24, 11] resulting from loss of nerve fibers in the superior half of the nerve, and many patients have bilateral symptoms. This painless visual loss is accompanied by ophthalmoscopic findings of pallid edema of the involved optic nerve. Intraocular pressures, perimetry, fluorescein angiography, B-mode orbital ultrasound electroretinography, retinal photography, slit-lamp examination of the anterior segment, and visual evoked potentials are also useful in evaluating patients with suspected AION. Visual acuity may vary from almost normal to no perception of light. An interval of a few hours to several days between the inciting event and the onset of symptoms is typical, recognition of the deficit may be delayed in patients who are critically ill, intubated, or sedated, and the presence of an abnormal visual field may not even be initially appreciated by patient or physician.
None of our patients had progression of symptoms, and except for 2 patients who reported minimal improvement, no appreciable recovery occurred. This finding correlated with the development of chronic atrophy of the optic nerve.
Because the erythrocyte sedimentation rate is elevated in most patients after bypass procedures, it may not be used to distinguish arteritic from nonarteritic AION [7]. However, we agree with Alpert and colleagues [7] that the acute postoperative development of the arteritic form would be highly unlikely.
Treatment
No specific treatment is available for patients with AION, and in most patients it does not improve. Corticosteroid agents, carbonic anhydrase inhibitors, and osmotic diuretics have been employed, but without consistent success [11]. Rarely, immediate recognition of the developing syndrome with swift correction of hypotension or anemia has reversed the process [33]. In 1989, Sergott and colleagues [34] were the first investigators to report on successful application of fenestration of the optic nerve sheath in patients with the progressive form of AION. This procedure presumably decreases the perineural pressure and improves perfusion and axoplasmic transport. However, this treatment has been shown in a randomized trial to be ineffective [35]. This procedure was undertaken in 2 of our patients in whom bilateral profound visual deficit developed, but no appreciable improvement occurred.
Conclusion
Because no treatment has consistently been effective, prevention is the key to avoiding the consequences of AION after CPB, particularly in patients with potential predisposing factors, such as increased intraocular pressure, marked hyperopia, or small cup-to-disk ratio. Reasonable goals, although not always attainable because of the severity of the underlying cardiac disease, include (1) brief duration of CPB, (2) use of colloid prime in patients at high risk for AION, (3) reduction of perioperative edema by judicious use of fluids and hemoconcentration, (4) appropriate PCO2 management to maintain cerebral autoregulation and decrease microembolization, (5) avoidance of prolonged low perfusion states or hypertension, (6) avoidance of profound systemic hypothermia except when specifically indicated, (7) judicious use of exogenous inotropic and vasoconstricting agents, (8) maintenance of adequate levels of hematocrit and hemoglobin in high-risk patients, and (9) intraoperative monitoring and adjustment of intraocular pressure in high-risk patients (for example, patients with severe glaucoma). Prophylactic perioperative reduction of intraocular pressure by installation of agents into the eye might be of theoretic benefit, but a rigorous study of such an intervention would be difficult to justify and perform.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
Y. Shen, M. Drum, and S. Roth The Prevalence of Perioperative Visual Loss in the United States: A 10-Year Study from 1996 to 2005 of Spinal, Orthopedic, Cardiac, and General Surgery Anesth. Analg., November 1, 2009; 109(5): 1534 - 1545. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. A. Lee, A. B. Nathens, B. S. Sires, M. K. McMurray, and A. M. Lam Blindness in the Intensive Care Unit: Possible Role for Vasopressors? Anesth. Analg., January 1, 2005; 100(1): 192 - 195. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. T. Bhatti and F. K. Enneking Visual Loss and Ophthalmoplegia After Shoulder Surgery Anesth. Analg., March 1, 2003; 96(3): 899 - 902. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. A. Nuttall, J. A. Garrity, J. A. Dearani, M. D. Abel, D. R. Schroeder, and C. J. Mullany Risk Factors for Ischemic Optic Neuropathy After Cardiopulmonary Bypass: A Matched Case/Control Study Anesth. Analg., December 1, 2001; 93(6): 1410 - 1416. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |