By Lewis Nelson MD
New York City Poison Center
A 34 year old ex-marine arrives to the emergency department via EMS in coma, several hours after ingesting a bottle of muscle relaxants and ethanol. Although he is unresponsive, his vital signs are normal except for a heart rate of 110. His physical examination is notable for a supple neck, and a nonfocal neurologic exam. The patient appears to be becoming more responsive over the short evaluation period. His pulse oximeter is 99% and his finger stick glucose is 110 mg/dL. Pertinent laboratory values: serum bicarbonate 22 mg/dL, anion gap 12, BUN 6, creatinine 1.0, arterial pH 7.39, pCO2 36, pO2 252, osmolar gap 10 and ethanol 120 mg/dL. His electrocardiogram revealed sinus tachycardia with a QRS duration of 110 msec.
After three hours the patient fully awakens and states that he cannot see, although he seems only minimally agitated by this situation. He is refusing to provide further information on his ingestion. On examination, the patient has no light perception, and his pupils are noted to be 6-7 mm and minimally reactive. His extraocular movements and funduscopic examination are normal. There is no sign of extraocular trauma, no chemosis, and his intraocular pressures are normal by tonometry. Opticokinetic nystagmus is absent.
Assuming that his blindness is toxin related, what could be the causes?
Many drugs and toxins are known to cause blindness infrequently. For example, vasodilators can reduce blood pressure to such an extent that an occipital cortical infarct occurs. However, hypotension induced infarction is unlikely in a young patient with presumed normal vasculature, but may be encountered in a more elderly patient. In addition, carbon monoxide and hydrogen sulfide may produce cortical blindness. In these cases, the pupillary reflex should be normal despite complete loss of vision.
Methanol, a one carbon alcohol, is perhaps the most common cause of toxic blindness. Methanol induced blindness is unlikely in this patient for several reasons. Firstly, and most importantly, the methanol must undergo metabolism to formate, which is directly toxic to the retina. Alcohol dehydrogenase is responsible for the metabolism, the enzyme prefers ethanol. This patient had enough ethanol present to prevent the metabolism of methanol (approximately 100 mg/dL is therapeutic). Had the patient ingested methanol a day prior to drinking the ethanol, it is possible that he may be blind from formate with an elevated blood ethanol level. However, as methanol is metabolized to formic acid, an anion gap metabolic acidosis should develop. The absence of such an acidosis, effectively rules out methanol induced blindness. Note that the absence of an osmol gap greater than 10 should not be used to rule out methanol ingestion. It is actually the change in the osmol gap that is diagnostic, not the absolute value. Since any individuals baseline osmol gap is unknown a change in the gap cannot be determined (Hoffman).
Any agent capable of causing diffuse vasospasm could induce blindness via retinal ischemia. Examples include cocaine and ergots. In both cases, retinal changes, such as pallor or vasoconstriction, should be observable. Phencyclidine is associated with "sun-gazers retinopathy" due to prolonged staring at the sun with ultraviolet damage to the retina. The "blind as a bat" associated with the anticholinergic toxidrome refers not to visual loss, but to loss of accomodation with an inability to focus near. It also serves to remind caregivers to look at the patients eyes for mydriasis.
Another agent associated with visual loss is quinine. Upon further questioning, the patient admitted to ingestion of thirty quinine tablets (300 mg each) in a suicide attempt three to four hours prior to arriving in the ED.
Where would a patient in the United States find quinine?
Quinine is derived from the bark of the Cinchona tree, the same tree from which aspirin is derived. It has a long history of use in herbal and homeopathic remedies and is still found in tonic water (about 2mg/ounce). It has been used to adulterate heroin, as the bitter taste resembles heroin. It has only been removed this year from the market as a nonprescription remedy for leg cramps. It has also been used in the past as an abortifacient, often with disastrous results (Dannenberg). Quinine is widely used worldwide as an antimalarial agent, especially in regions where chloroquine resistance has developed. As a recently discharged Marine, the patient had accumulated pills during his service.
How does quinine produce toxicity?
Quinine shares many properties with the two other agents derived from the Cinchona tree. Like salicylates, quinine induces cinchonism, consisting of nausea, vomiting, tinnitus, dizziness, and headache. Like it's optical isomer quinidine, quinine has cardiac effects analagous to the type Ia antidysrhythmics. This consists of QRS and QT prolongation. Unique to quinine is the ability to induce blindness. Visual changes were initially thought to be due to retinal vasospasm with subsequent ischemia. It was even felt to respond favorably to stellate ganglion blocks to relieve the vasospasm, but this may be largely a reporting bias, since several studies have not validated its benefit. Quinine, or possibly a metabolite, is probable a direct retinal toxin (Bacon).
In a large retrospective study of acute quinine poisonings, Boland found that 42% had visual symptoms, 38% had tinnitus, and 14% had altered mental status, with deep coma in 4%. Of those with visual symptoms just over half had total blindess, which was permanent in about half of these. The risk of blindness was roughly corellated with quinine levels: patients with ocular effects are likely to have a level of 10g/mL at 10 hours after ingestion. Currently, there is little clinical utility to a serum level, as the turnaroud time is long, and therapeutic alterations are limited.
Is therapy available?
Although case reports have touted stellate ganglion blocks as curative of the vasospasm and blindness, Boland noted that in 34 cases in which it was performed, it may have helped in only one. Fortunately, the visual changes generally resolve with supportive care.
The use of oral activated charcoal has been shown to reduce the half-life of quinine in human volunteers with non-toxic ingestions.. Sabto et.al. compared methods to enhance elimination of the drug once absorbed. They noted that forced diuresis was better than hemodialysis, plasma exchange, and peritoneal dialysis for increasing clearance of drug. However, neither they nor any others have ever shown that increasing clearance speeds recovery. Indeed, forced diuresis is a procedure almost never indicated in clinical medicine. Although not adequately studied, the clinical valueof charcoal hermoperfusion may surpass the above methods of enhancing elimination (Morgan).
The patient received activated charcaol for several doses. He was given intravenous saline at three times maintenance and his urine pH remained below 7.5. He did not receive stellate ganglion blocks nor was another method of enhanced elimination instituted. His vision improved over the subsequent two days and his electrocardiogram normalized over 12 hours without specific therapy. He was discharged to the psychiatry service.
Bacon P, Spalton DJ, Smith SE. Blindness from quinine toxicity. Br J Ophthal 1988;72:219-224.
Boland ME, Roper SMB, Henry JA. Complications of quinine poisoning. Lancet 1985;I:384-385.
Dannenberg A, Dorman SF, Johnson J. Use of Quinine for self induced abortion. South Med J 1983;76:846-849.
Morgan MDL, Pusey CD, Rainford DJ, Robins-Cherry AM. The treatment of quinine poisoning with charcoal haemoperfusion. Postgrad Med J 1983;59:365-367.
Sabto J, Pierce RM, West RH, Gurr FW. Hemodialysis, peritoneal dialysis, plamsapheresis, and forced diuresis for the treatment of quinine overdose. Clin Nephrol 1981;16:264-268.