Circulation 36,825 • Volume 18, No. 2 • Summer 2003

POVL Registry Reports Preliminary DATA

Lorri A. Lee, MD

Postoperative visual loss is one of the most catastrophic and unexpected complications that patients can experience. It can occur after almost any type of surgery (e.g., spine surgery, prostatectomy, peripheral vascular surgery), many of which are unrelated to the head and neck. Visual deficits may range from transient blurred vision to complete blindness. One patient, describing his visual loss after spine surgery, said that it was “like looking through two paper towel holders with cellophane taped over the ends.” He was left with severely constricted visual fields bilaterally and with decreased visual acuity in the remaining field. This degree of vision loss can severely handicap patients in everyday life, such that the simple act of walking down a flight of stairs becomes a challenge.

Because of the devastating impact of this perioperative complication on patients, the American Society of Anesthesiologists (ASA) developed the Postoperative Visual Loss Registry in June 1999. Cases of postoperative visual loss after non-ocular surgery are submitted anonymously on detailed case report forms. The goal of the ASA Postoperative Visual Loss Registry is to obtain a sufficiently large number of cases (n=100) so that associations between patients with postoperative visual loss can be determined. When the goal of 100 patients has been attained, this Registry will be the single largest database on postoperative visual loss, and will contain the most detailed information on perioperative events. These findings can then be used to direct research into the etiology and prevention of this perioperative complication.

Preliminary results from the ASA Postoperative Visual Loss Registry (n=79) demonstrate that ischemic optic neuropathy (ION) was the most common injury associated with postoperative visual loss. Other injuries associated with postoperative visual loss included central retinal artery occlusion (CRAO) and cortical blindness, although these were much less common than ION. The most common procedures associated with postoperative visual loss were spine surgery (n=53, 67%) and cardiac surgery (n=8, 10%). Sixteen cases were associated with a variety of operations including liver transplantation, thoraco- and abdominal aneurysm resection, head and neck surgery, thoracotomy, and others.

Because spine surgery patients who developed ION comprised the largest group of patients (n = 43) in the Registry, these cases will be discussed in greater detail. The data in Table 1 demonstrate the median values with ranges, for age, duration of prone position, estimated blood loss, and lowest hematocrit for patients who were diagnosed with ION. Although most cases occur in middle-aged patients (median age = 49 yrs) with a long duration in the prone position (median time = 8 hrs) and/or with large blood loss (median blood loss = 2.3 L), the lower ends of these ranges demonstrate that almost any spine surgery patient is susceptible to developing this complication. For example, postoperative ION has occurred in patients as young as 19 years old, in procedures as brief as 3 hrs, and with a blood loss as low as 200 ml.

Consistent with these findings, the Registry contains cases in which the intraoperative course is entirely unremarkable, but the patient wakes up with visual deficits. Moreover, patients with known coronary artery disease may develop vision loss, but no cardiac sequelae. These observations suggest that the blood supply to the optic nerve is uniquely vulnerable to circulatory perturbations. Because of the known variability in blood supply to the optic nerve, patients may have unforeseen, and currently unpredictable, and clinically undetectable, anomalies including an atypical anatomical pattern in the optic nerve blood supply, poor watershed perfusion zones, and abnormal autoregulation.

Other associated factors in spine surgery patients who developed ION include the presence of anemia in many cases with a median hematocrit of 25%. Additionally, the largest percentage decrease in blood pressure from baseline was 37% (median), and deliberate hypotension was utilized in 40% of cases. The optimal ranges of hematocrit and blood pressure for adequate oxygen delivery to the optic nerve have not been identified, particularly in the presence of venous congestion in the prone position. Therefore, it is difficult to determine the contribution of anemia and hypotension in these cases to the development of ION, although some concern is warranted.

Table 1 also demonstrates that ION occurred in 8 patients whose heads were placed in Mayfield tongs (18%). Many patients and physicians have previously hypothesized that ION was caused by external pressure on the globe as a result of inadequate positioning and padding. These data provide strong evidence that ION, the most common cause of postoperative visual loss, can occur when the face is free from pressure. This knowledge can now be used to direct research in more productive areas such as the basic vascular physiology of the optic nerve and the relationships between perioperative care and vascular supply and demand by the optic nerve.

Table 1. Preliminary Data from the ASA Postoperative Visual Loss Registry: As sociated Factors from Spine Cases Diagnosed with ION (N=43)
Median Range
Age (yrs) 49 19 – 73
Prone Time (hrs) 8 3 – 24
Estimated Blood Loss (liter) 2.3 0.2 – 20.0
Lowest Hematocrit (%) 25.5% 19% – 40%
Largest % Decrease in BP 37% 19% – 61%
Deliberate Hypotension 17 (40%)
Headrest:
Mayfield Tongs 8 (18%)
Horseshoe 0
Foam 33 (77%)
Unknown 2 (5%)

Additional case submissions are needed to obtain the goal of 100 patients for the ASA Postoperative Visual Loss Registry. A larger number of cases will provide more definitive information regarding associated factors for this devastating perioperative complication. Anonymous case submission forms are available from our website at www.closedclaims.org.

In summary, ischemic optic neuropathy (ION) is the most common injury associated with postoperative visual loss in the ASA Postoperative Visual Loss Registry. ION after spine surgery can occur in young, healthy patients, after operations as brief as 3 hrs, and with an estimated blood loss as low as 200 ml. ION occurs in the absence of external pressure on the globe. Further research will need to be conducted to determine the contribution of anemia and hypotension to the development of ION, particularly in the presence of venous congestion in the prone position.

Dr. Lee is an Assistant Professor in the Department of Anesthesiology at the University of Washington, Seattle. She is also the Director of the ASA Postoperative Visual Loss Registry.

Editor’s Note: The reader is also directed to two previous articles pertaining to postoperative visual loss which appeared in the Winter 2001-02 and Summer 1998 issues of this Newsletter, and two articles in Anesthesia & Analgesia (2001;93:1410-6 and 2001;93:1417-21).