Reviewed By Dr. Sage Wiener
It’s 9 o’clock on a Saturday, the usual crowd shuffles in, Stacey is in the stretcher sitting next to you on this night in November. She’s a 28-year-old woman, presenting to the Emergency Department complaining of three days of headache, dizziness, nausea, and difficulty concentrating. Vitals: HR 102/min, BP 95/70 mm Hg, POx 100% on room air, Temp 37.3ºC, Finger Stick Glucose 85 mg/dL, UCG Negative. You ask your normal headache questions:
Headaches have a fairly broad differential diagnosis, and it’s sometimes hard to ask everyone every question and not to jump to the most likely answer. Some additional questions worth asking are:
This really narrows down your differential and is especially important during colder months and in more industrial areas. Other questions that will help you better interpret their blood levels are whether or not they smoke or use hookah frequently, which can cause toxicity from carbon monoxide (CO) exposure.1–3
Background “Carbon monoxide is formed during the incomplete combustion of virtually any carbon-containing compound,”4 meaning fossil fuels (natural gas, propane, gasoline, coal), plastics, wood, fabrics, etc… CO can be an extremely insidious killer due to the fact that it is colorless and odorless. Around 15,000 people report to EDs every year with unintentional, non-fire-related CO exposures and over 400 people dying.5,6 These numbers don’t even take into account the morbidity related to CO exposure.7 Although CO exposures can happen at any time of the year, there is a distinct seasonal pattern with peaks in the winter, and there have also been recorded cases of peaks associated with generator use after natural disasters like hurricanes.6,8,9 However, these numbers for burden of disease probably grossly underestimate the effects of CO. Three studies present drastically different numbers for exposures with one taking into account suspected exposures only after excluding intentional and fire related exposures. This likely underestimates disease burden even further.10–12 As stated before, numerous exposures can lead to CO toxicity, including methylene chloride and bromide,13 riding in the back of pickup trucks,14 Zamboni (ice skating rink resurfacer) operation indoors,15 and indoor forklift operation,16 just to name a few. So. remember to always think about CO, especially when a patient reports a history involving exposure to fumes from combustion of any materials and especially in enclosed spaces. Oh, and don’t forget to look out for yoga balls.17
Kinetics and Pathophysiology Carbon monoxide is rapidly absorbed and then predominantly carried in the blood tightly bound to hemoglobin with a 200-250 times greater affinity than oxygen.4 Eventually it is absorbed by various tissues, mostly tightly bound to myoglobin and to cytochromes in the electron transport chain. Tracking plasma concentrations of CO could possibly better clarify its potential for disease as it is through this compartment that CO will be distributed to other tissues. The half-life of carboxyhemoglobin can be reduced from approximate 240 minutes on room air at 1 atmosphere of pressure to 90 minutes on 100% oxygen and to 20 minutes on 100% oxygen at 3 atmospheres of pressure (also called hyperbaric oxygen, or HBO).18–20 Carbon monoxide causes toxicity by binding to hemoglobin and decreasing oxygen-carrying capacity and causing a left shift in the oxyhemoglobin dissociation curve. However, this is not the end of the story. To test how much of CO’s toxic effects were due to the reduced oxygen delivery to tissues, researchers tested three groups of dogs: Group 1 was exposed to a lethal dose of CO, Group 2 were bled to an average hematocrit of 25% (representing similar oxygen carrying capacity as group 1), while Group 3 dogs were bled and then transfused with blood from Group 1 to have a similar carboxyhemoglobin level. 100% of group 1 died, whereas 100% of groups 2 and 3 survived (see Figure 1.21) Figure 1 One of CO’s most important molecular mechanisms of toxicity is through binding to cytochrome C oxidase (which contains heme), interfering with cellular respiration, and ultimately leading to oxidative stress with protein oxidation and lipid peroxidation.21 There is further evidence that CO interferes with nitric oxide, leading to endothelial injury, free radical formation, and platelet-neutrophil aggregation. These effects and activation of other signaling molecules likely also lead to programed cell death, apoptosis. Its myoglobin binding also likely leads to myocardial injury. CO is an innocuous appearing, simple molecule and is a product of incomplete combustion and even normal metabolic activity. But it is toxic to our bodies through numerous different pathways.
Presentation The presentation of a patient with CO toxicity is broad and non-specific with the leading misdiagnosis being the flu.22 The most common symptoms after mild exposure are: – Headache – Nausea – Dizziness23 It’s not uncommon to misdiagnose children as having gastroenteritis, colic, and food poisoning. The headache is most often described as a non-remitting, dull, frontal headache. Patients with more prolonged exposure can develop decreased EF, elevated cardiac markers, myocardial ischemia, and dysrhythmias, and may have increased cardiac-related mortality 8 years after exposure.7,24–26 Perhaps the organ that most concerns us is the brain. Neurologic symptoms and signs include: – Headaches – Dizziness – Ataxia – Syncope – Seizures – Coma – Stroke mimics including AMS and focal neuro deficits Patients might present as a stroke mimic and get worked up for undifferentiated AMS. Head CT and MR might show decreased density in the central white matter and globus pallidus. An EEG might show diffuse frontal slow-wave activity. Labs aren’t only remarkable for elevated carboxyhemoglobin concentrations. Mild levels of exposure might lead to compensatory respiratory alkalosis secondary to acidosis from impaired cellular respiration, while more substantial exposures lead to lactic acidosis.27 Both initial lactate concentrations and degree of pH have been shown to be useful for prognostication. In the setting of a fire, a lactate > 10 mmol/L should be considered evidence of cyanide toxicity, and the patient should be empirically given hydroxocobalamin.
Neurocognitive effects Perhaps our most feared complications of CO exposure are the neurocognitive effects, which can be irreversible and are not present during the initial evaluation. In patients presenting with acute neuropsychiatric changes during winter months or with other specific risk factors as discussed above, CO poisoning should be on the differential. The possible presenting symptoms and signs are broad and varied and can occur 2-40 days after exposure:28,29 – dementia – amnestic syndromes – psychosis – parkinsonism – paralysis – chorea – cortical blindness – apraxia and agnosia – peripheral neuropathy – incontinence It’s also important to note that there are also neuropsychiatric symptoms, which are often more benign and insidious-appearing. Examples are: – trouble concentrating – memory problems – insomnia – change in behavior – change in patience
Management The most important component in the field is to remove the exposure. As in all emergencies, ABCs are paramount, but in the case of CO poisoning, the airway is actually the route of administration for your antidote: 100% oxygen therapy. And though there is literature arguing against 100% oxygen administration for most patients due to local tissue injury and free radical formation, in this case, it is the antidote. The best way to deliver 100% oxygen is via endotracheal intubation, but this may be too aggressive for most patients. For these patients, non-invasive positive pressure ventilation (NIPPV) is probably the best way to go. For those who can’t tolerate the NIPPV mask, high flow oxygenation with nasal cannula would likely be a good alternative. It is important to note that a non-rebreather mask won’t deliver more than 70%-90% oxygen.4 As stated previously 100% oxygen reduces the half-life of carboxyhemoglobin from approximately 5 hours to 1 hour.4 Since CO can cause cardiac toxicity, maintaining cardiac monitoring and assessing for cardiac function and ischemia with a 12-lead ECG is also important. Pulse oximeters that can read carboxyhemoglobin are not commonly available, but they have been shown to correlate well with blood gas analyzers. It is also important to consider cyanide toxicity or concomitant cyanide and CO toxicity in patients who have been extricated from fires. Many of these patients might have already been given hydroxocobalamin by EMS in the field, which will affect pulse oximetry. And if prehospital personnel have administered a cyanide antidote kit containing nitrites (generally contraindicated in the setting of potential CO), you should also pay attention to the methemoglobin levels on your co-oximetry. And then there is HBO (not the network broadcasting GOT, but Hyperbaric Oxygen) therapy. This is the area of the literature for CO poisoning management that is most controversial. There are two major ways in which HBO therapy is supposed to treat CO poisoning. First, it increases the amount of dissolved oxygen in the plasma to levels high enough for cellular respiration and by increasing the amount of oxygen, it shifts the equilibrium of hemoglobin bound to CO and oxygen, thus leading to a reduced amount of carboxyhemoglobin. The half-life of CO decreases from on average about 1 hour on 100% oxygen to about 20 minutes on average with 100% oxygen and 2.5 ATMs (lower end of HBO used for therapy).19 A multitude of animal studies support the use of HBO through multiple other pathways leading to decreased reperfusion injury, and mitochondrial function. More importantly, HBO has been shown in a randomized, placebo-controlled trial to significantly reduce delayed neurocognitive sequelae. At present, there are a limited number of studies that support the routine use of HBO, and perhaps the most well-known is a Cochrane systematic review30 that concluded that there is inadequate evidence to support a specific role for HBO in the treatment of CO therapy. The general problem appears to be related to poor quality of study protocols, examples of negative findings with potential flawed protocols and delays in treatment, sub therapeutic levels of pressure, lack of including the sickest patients, and lack of patients with concerning presentations including episodes of syncope. This is in contrast to positive studies, including the only randomized placebo-controlled trial,19 which enrolled mostly ill patients on average within 3 hours, and for its initial dive, used 3 ATMs of pressure. There is much confusion, so here is my distillation along with expert advice from toxicologists, and most especially from Goldfrank’s Toxicologic Emergencies.4 HBO is a relatively benign treatment, that offers your sickest, highest-risk patients the best chance of an improved outcome. The basic science research supports its use, and though the literature doesn’t definitively support its use for all patients, the vast majority of negative studies exclude these very patients. The most well-designed study of the therapy showed a clear benefit. And important points of HBO therapy: Carbon monoxide poisoning is a devastating illness, and for a subset of patients who are very ill or have high-risk presentations including loss of consciousness, you should do everything you can to reduce morbidity and long term neurocognitive sequelae. There are no validated indications for HBO, however the following is a generally well agreed upon list:4 – Syncope (loss of consciousness) – Coma – Seizure – Altered mental status (GCS<15) or confusion – Abnormal cerebellar function – Prolonged CO exposure (≥24 hours) – Fetal distress in pregnancy – Myocardial Ischemia – Carboxyhemoglobin >25% (not evidenced-based, but is often what you will be tested on) – Pregnant with carboxyhemoglobin >15-20% (no specific evidence) However, one study shows that even patients who have none of these risk factors but who are young still have high rates of neurocognitive sequelae.31
Your Patient’s Course You put your patient on NIPPV with 100% oxygen, and the lab results come back. Your patient isn’t pregnant and has a carboxyhemoglobin of 14%, normal lactate, and normal pH. Based on the elevated carboxyhemoglobin and lack of risk factors for a baseline elevated carboxyhemoglobin, you suspect CO poisoning and alert your local poison control center. Your neurologic exam is normal with no cerebellar abnormalities. You contact the fire department and other regulatory bodies to inform them, remembering there are other people living in the buildings. The poison control center consultant states that the patient doesn’t meet any indication for HBO. You place the patient in observation on 100% oxygen. You get a repeat carboxyhemoglobin 8 hours later which comes back at 2%. The patient is discharged with neurology follow up.
Summary/Take Home Points
References
Noah Berland
Latest posts by Noah Berland (see all)
- Teaching Shifts: “Sore Throat” - November 12, 2020
- Teaching Shifts - October 20, 2020
- Social EM Journal Club: Food Insecurity - October 16, 2020
0 Comments