Carboxyhemoglobinemia is usually associated with exposure to exogenous carbon monoxide (CO) and is the leading cause of injury and death due to poisoning worldwide (1). CO, a colorless, odorless, tasteless gas, is produced by the incomplete combustion of any carbon-containing fuel (1). In humans, the sole endogenous source of CO is heme degradation (2). We report the development of significant carboxyhemoglobinemia in a critically ill, mechanically ventilated patient who required massive transfusion because of retroperitoneal hemorrhage.
A 41-yr-old man was transferred to the University of Illinois at Chicago Medical Center (UIC) from a referring hospital for further management of large bilateral spontaneous adrenal hemorrhage. Before transfer, the patient had received transfusion of 10 U of packed red blood cells (PRBC) and 10 U of fresh frozen plasma (FFP) over 3 days in the intensive care unit to maintain hemodynamic stability and to correct coagulopathy. The patient did not smoke before transfer, nor was he exposed to anesthetic gases.
At UIC, the patient required tracheal intubation and mechanical ventilation with an inspired oxygen concentration of 50% on hospital Day 1 for transfusion-related acute lung injury. The patient remained tracheally intubated throughout his hospital stay and received further transfusion of blood products for continuous hemorrhage. On hospital Day 2, the patient underwent bilateral adrenal artery angiography, and embolization was attempted to control hemorrhage. On hospital Day 6, because of the presence of a large amount of retroperitoneal blood, the need for continuing administration of red blood cell transfusions, and an increased serum bilirubin level with normal serum concentrations of hepatic enzymes (aspartate aminotransferase [AST], alanine aminotransferase [ALT], and alkaline phosphatase [AP]), we obtained an arterial blood gas (by using cooximetry) to determine the concentration of carboxyhemoglobin (COHb), which was 3.9%. Three days later, after the transfusion of 27 U of PRBC, the maximum COHb level was 6.4% (hemoglobin, 8.4 g/dL; total bilirubin, 14.7 mg/dL) (Fig. 1).
The patient underwent exploratory laparotomy on hospital Day 14 for continuing retroperitoneal bleeding and the development of abdominal compartment syndrome. Approximately 4000 mL of hematoma was evacuated. The abdomen was packed and left open, and external drainage of blood was therefore possible. In the perioperative period, he received another 12 U of PRBCs, 18 U of FFP, and 2 U of pooled platelets (total: 39 U of PRBC [27 before surgery and 12 during surgery] and 30 U of FFP [12 before surgery and 18 during surgery]). The pathologic examination of the specimen revealed pheochromocytoma.
The measured total bilirubin and hepatic enzymes (AST, ALT, and AP) on the day of admission to UIC were normal. Before surgery, the maximal total/direct bilirubin and AST serum concentrations were 15.1/10.2 mg/dL and 84 U/L, respectively. The ALT and AP remained within normal limits before surgery (Fig. 1).
Over the subsequent weeks, the patient underwent bilateral adrenalectomy, partial pancreatectomy, splenectomy, multiple visceral angiograms, and embolizations. He developed large bilateral pleural effusions requiring chest tube placement, nonoliguric acute renal failure, disseminated intravascular coagulation, and shock liver. Support was withdrawn at the request of the family, and the patient died on hospital Day 58.
This case report highlights that, in patients without external blood loss, the transfusion of large volumes of red cells can lead to the development of prolonged increases in COHb levels from the physiologic metabolism of heme.
The sole physiologic (endogenous) source of CO is heme degradation in the body. Heme oxygenase is a microsomal enzyme that regulates the cellular content of heme. It catalyzes the rate-limiting step in the oxidative catabolism of heme to yield equimolar quantities of biliverdin, CO, and iron. Biliverdin is then reduced by the biliverdin reductase to bilirubin (2).
The affinity of hemoglobin for CO is more than 200 times as great as its affinity for oxygen and shifts the oxyhemoglobin dissociation curve to the left (1). CO binds to cell heme proteins, such as myoglobin or cytochrome oxidase, and interferes with their function (4). CO elimination is related to minute ventilation, duration of exposure, and the fraction of inspired oxygen (Fio2). The half-life of COHb is 4–6 hours when the patient is breathing room air, 40–80 minutes when the patient is breathing 100% oxygen, and only 15–30 minutes when the patient is breathing hyperbaric oxygen (6). Overt signs of toxic effects usually appear at COHb levels of 15%–20%, and a level of 25% is an indicator of severe poisoning (4). In a sedated patient, signs and symptoms of CO poisoning, such as altered levels of consciousness, are difficult to assess (7–9).1
Red blood cell destruction and heme catabolism lead to increased COHb levels in patients with hemolytic disease. The measurement of CO in breath or blood can be used as an index of heme degradation in vivo (3), and COHb levels correlate with unconjugated bilirubin levels and the reticulocyte count (5). Wohlfeil et al. (10) reported a patient with severe intravascular hemolysis secondary to antibiotic-dependent antibodies. This led to a COHb concentration of 7.3% within 48 hours. Mild hyperventilation with an Fio2 of 0.4 decreased the COHb level to 4.3% within eight hours.
In our patient, transfusion of 27 U of PRBCs over three weeks led to a maximum COHB level of 6.4%. Because he was tracheally intubated and ventilated with an Fio2 of 0.35–0.5 at the time the COHb level was measured, the half-life of COHb was decreased and yet remained increased for days, ranging from 1.7%–5.6% (Fig. 1). This may be because the patient bled into the retroperitoneal space, where the hemoglobin was degraded continuously by cellular and nonhumoral mechanisms. The level of COHb is dependent on the rate of hemoglobin degradation and elimination of COHb and on the degree of endogenous production.
As shown in Figure 1, the COHb level peaks, and this follows the increase in bilirubin levels before surgery. After evacuation of the hematoma with an open abdominal wound and drainage of blood, the COHb level did not increase substantially even with massive transfusion and increases in bilirubin. Shock liver after cardiac arrest led to the second increase of total bilirubin serum concentration with increases in liver enzymes.
Another source of CO may be transfused blood. Blood transfusion may be contaminated with CO and may have COHb levels of up to 12% (11). In our patient (188 cm; 79 kg), the estimated blood volume was 5530 mL (79 kg × 70 mL/kg). At an average hematocrit (Hct) of 27%, the red blood cell volume (RCV) was 1493 mL. Transfusion of 1 U of PRBCs with a COHb level of 10%, an Hct of 70%, and a volume of 300 mL (RCV 210 mL/RCV[COHb] 21 mL) (12) will increase the COHb level by 0.81% (RCV 1703 mL/10.5 mL or 21 mL). Calculation of the endogenous production is difficult and likely to be inaccurate because of the slow rate of hemolysis compared with acute intravascular hemolysis (10) and factors influencing the half-life of COHb (Fio2, minute ventilation, and length of exposure).
Although the maximum level of COHb reached in our patient was not considered to be overtly toxic, we believe that, in critically ill patients requiring similar or larger amounts of red blood cell transfusion without external blood loss or with a high rate of hemolysis, the physician should consider the possibility of the development of carboxyhemoglobinemia and should obtain a blood gas analysis by using cooximetry. If COHb is significantly increased, therapy to decrease the concentration of COHb (such as increasing Fio2 and/or minute ventilation) or removal of the hemoglobin source should be considered.
1. Thom SR. Hyperbaric-oxygen therapy for acute carbon monoxide poisoning. N Engl J Med 2002;347:1105–6.
2. Brittenham GM. Disorders of iron metabolism: iron deficiency and overload. In: Hoffman R, Benz EJ Jr, Shattil SJ et al., eds. Hematology: basic principles and practice. 3rd ed. Philadelphia: Churchill Livingstone, 2000:397–428.
3. Stevenson DK, Vreman HJ. Carbon monoxide and bilirubin production in neonates. Pediatrics 1997;100:252–4.
4. Piantadosi CA. Carbon monoxide poisoning. N Engl J Med 2002;347:1054–5.
5. Sears DA, Udden M, Thomas LJ. Carboxyhemoglobin levels in patients with sickle-cell anemia: relationship to hemolytic and vasoocclusive severity. Am J Med Sci 2001;322:345–8.
6. Ernst A, Zibrak J. Current concepts: carbon monoxide poisoning. N Engl J Med 1998;339:1603–8.
7. Shi Y, Li H, Pan J, et al. Evidence of increased endogenous carbon monoxide production in newborn rat endotoxicosis. Chin Med Sci J 1997;12:212–5.
8. Johnson RA, Colombari E, Colombari DSA, et al. Role of endogenous carbon monoxide in central regulation of arterial pressure. Hypertension 1997;30:962–7.
9. Ping L, Houli J, Chang HY, et al. Angiotensin II induces carbon monoxide production in isolated rat kidney. Hypertension 2000;36:712.
10. Wohlfeil ER, Woehlck H, Gottschall J, Poole W. Increased carboxyhemoglobin from hemolysis mistaken as intraoperative desflurane breakdown. Anesth Analg 2001;92:1609–10.
11. Ehlers M, McCloskey D, Devejian N. Alarming levels of carboxyhemoglobin in a unit of banked blood. Anesth Analg 2003;97:289–90.
12. Miller RD. Transfusion therapy. In: Miller RD, ed. Anesthesia. 5th ed. Philadelphia: Churchill Livingstone, 2000:1613–44.
1 Weber TP, Kirchhefer U, Meiner A, et al. An experimental model detecting a hyperoxic lung injury with pulmonary carbon monoxide production. Crit Care Med 1999;27:A130.