| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the Departments of Anesthesiology, Surgery, and Critical Care, University of Pennsylvania, Philadelphia, Pennsylvania.
Address correspondence and reprint requests to Albert T. Cheung, MD, Department of Anesthesia and Critical Care Medicine, University of Pennsylvania, 3400 Spruce St., Dulles 680, Philadelphia, PA 19104-4283. Address e-mail to cheunga{at}uphs.upenn.edu.
Abstract
BACKGROUND: Cyanide toxicity is a complication of sodium nitroprusside administration. Cardiac surgery may increase the risk of cyanide toxicity, because hemolysis during cardiopulmonary bypass (CPB) may catalyze the release of free cyanide from sodium nitroprusside.
METHODS: We obtained serial blood specimens from 25 cardiac surgical patients during CPB. Plasma specimens were analyzed for free hemoglobin concentration and ability to generate free cyanide anion upon exposure to sodium nitroprusside.
RESULTS: Hemolysis based on plasma-free hemoglobin concentration increased over time during CPB at an average rate of 0.27 mg·dL–1·min–1 (P < 0.001). The concentration of free cyanide generated by the addition of sodium nitroprusside to the plasma samples was directly related to the plasma-free hemoglobin concentration (P < 0.001).
CONCLUSION: CPB-associated hemolysis and free hemoglobin release accelerated the immediate release of free cyanide from sodium nitroprusside. These in vitro findings suggest that cardiac surgical patients may be at increased risk of cyanide toxicity in response to the perioperative administration of sodium nitroprusside.
Cyanide toxicity is a complication of sodium nitroprusside (SNP) therapy. Free cyanide is released from SNP in a molar ratio of 5 molecules of cyanide for each molecule of SNP through a nonenzymatic reaction within the body. Under normal conditions, free cyanide anion is detoxicified by enzymatic conversion to thiocyanate through the action of the enzyme thiosulfate sulfur transferase. Thiocyanate is then excreted by the kidney. Cyanide toxicity is believed to occur if the amount of cyanide generated from SNP exceeds the metabolic capacity for detoxicification. Clinical experience suggests that most cases of cyanide toxicity from SNP administration are the consequence of prolonged drug infusion or administration at excessive doses. However, cyanide toxicity from SNP administration has also been reported in patients receiving infusions at the recommended rate, and in one case, after an infusion that lasted for only 50 min (1,2). Furthermore, a study examining the risk of cyanide toxicity from SNP detected an incidence of 2.4% in cardiac surgical patients (3). Conditions that accelerate the nonenzymatic breakdown of sodium SNP in the circulation could provide an explanation for cases of cyanide toxicity observed in cardiac surgical patients or in the cases in which the recommended dose of SNP was not exceeded.
Hemolysis as a consequence of cardiopulmonary bypass (CPB) may explain, in part, the increased risk of cyanide toxicity that has been observed in cardiac surgical patients. Although the precise pathway of SNP breakdown after IV administration has not been completely established, studies have determined that SNP preferentially reacts with sulfhydryl groups on hemoglobin to release free cyanide (4). Normally, the reaction between SNP and hemoglobin may be limited by the erythrocyte membrane that prevents exposure of SNP to hemoglobin within cells. Experiments have demonstrated that although SNP incubation with erythrocytes resulted in free cyanide release, the amount of free cyanide release was increased by a factor of 5- to 8-fold when SNP was incubated with human red blood cell lysates (4,5).
The objectives of this study were to quantify the amount of hemolysis that occurs during CPB in cardiac surgical patients and to determine if the quantity of free hemoglobin in the plasma was related to the capacity of blood to generate free cyanide in response to SNP exposure.
METHODS
In an investigational protocol approved by the IRB, blood samples were obtained from 25 patients undergoing elective cardiac surgical operations requiring CPB. Written informed consent was waived by the IRB because blood samples used for study purposes were acquired through the normal conduct of routine laboratory testing. Blood was sampled after systemic heparinization immediately before CPB, every 30-min during CPB, and immediately after CPB. Samples were immediately centrifuged at 3500 rpm for 15 min in a refrigerated centrifuge to separate plasma from the cellular fraction. Plasma samples were stored at –70°C in sealed containers before analysis. Patients who received SNP before or during operation were excluded from study.
CPB was performed with a centrifugal pump (Revolution, Cobe Cardiovascular, Arvada, CO), membrane oxygenator (Optima, Cobe Cardiovascular), heat exchanger (Terumo Sarns, Terumo Cardiovascular Systems, Ann Arbor, MI), 20-µm arterial line filter, and closed venous reservoir bag. All tubing and surfaces of the extracorporeal circuit consisted of a polyvinylchloride base resin with a polysiloxane-containing copolymer surface modification (SMARxT, Cobe Cardiovascular). A cardiotomy suction system was used for all cases and scavenged blood from the mediastinum and CPB circuit was processed by filtration, centrifugation, and washing (Hemonetics Cell Saver, Hemonetics, Braintree, MA).
Measurement of Plasma-Free Hemoglobin
Plasma-free hemoglobin concentration [free Hb] was measured by spectrophotometry, using the HemoCue plasma/low hemoglobin system (HemoCue, Lake Forest, CA) from a 20-µL aliquot of plasma added to a chemically pretreated cuvette. The HemoCue plasma/low hemoglobin instrument had a sensitivity to detect and quantify plasma [free Hb] in the range of 0.03–3.00 g/dL after conversion of hemoglobin to azide methemoglobin with absorption peaks at 565 and 880 nm. The calibration of the HemoCue plasma/ low hemoglobin instrument was verified with standard solutions with hemoglobin concentrations of 70 mg/dL, 490 mg/dL, and 2020 mg/dL to ensure accuracy for each series of measurements.
Measurement of Free Cyanide Anion Release From SNP Added to Plasma Samples
To determine cyanide anion generation in individual plasma samples (2.0 mL), each sample (n = 172) was incubated in vitro with SNP at a concentration of 250 µg/mL for 2 h at room temperature (25°C) in a sealed aeration apparatus to capture dissolved and volatile cyanide. A spectrophotometric assay was used to measure free cyanide anion concentrations based on absorbance at 532 nm after conversion to cyanogen bromide using a standard calibration curve (6). The assay quantified the generation of free cyanide in concentrations ranging from 0.0 to 5.0 µg/mL with a sensitivity of 0.03 µg/mL.
The capacity of plasma to generate cyanide upon exposure to different concentrations of SNP was determined by adding SNP to achieve concentrations ranging from 50 to 600 µg/mL in 2 mL pooled plasma samples obtained during CPB with a final plasma [free Hb] of 120 mg/dL. The time course of cyanide generation was further determined by measuring the amount of cyanide generated from 2 mL pooled plasma samples with a plasma [free Hb] of 120 mg/dL exposed to SNP at a concentration of 250 µg/mL after incubation periods ranging from 10 min to 4 h.
Statistics
Plasma-free Hb concentrations and SNP-induced release of free cyanide concentrations were treated as continuous variables. Regression analysis with a generalized estimating equation (STATA8, STATA Corp., TX) was used to compensate for multiple measurements in each patient. Multiple regression analysis tested for relationships between the plasma [free Hb], sampling time in relation to the start of CPB, SNP-induced cyanide release, and confounders.
RESULTS
Patients (8 females and 17 males) ranged in age from 37 to 86 yr with a mean age of 65 ± 3 yr (Table 1). Duration of CPB ranged from 75 to 322 min with a mean of 186 ± 15 min. Deep hypothermic circulatory arrest was used in six patients for ascending aorta or aortic arch reconstruction.
|
Plasma-free Hb concentrations increased over time during CPB (Fig. 1, P < 0.001). The concentration of cyanide anion generated by the addition of SNP to plasma was directly related to the plasma-free Hb levels (Fig. 2, P < 0.001). Based on a linear model, plasma-free Hb increased at an average rate of 0.27 mg/dL/min during CPB (P < 0.001). Linear modeling predicted that at a plasma SNP concentration of 250 µg/mL, each 100 mg/dL of free Hb increased the free cyanide anion concentration by 0.58 µg/mL (95% CI 0.53–0.63 µg/mL). The amount of free cyanide released varied directly with the concentration of SNP. At a plasma-free Hb concentration of 120 mg/dL, cyanide production increased in response to increasing SNP concentrations over a range of 71–536 µg/mL (Fig. 3, P < 0.001). Free cyanide generation in plasma samples upon exposure to SNP occurred largely independent of the incubation time. The maximum free cyanide concentration was detected within 30 min of incubation and did not increase further in response to longer incubation periods up to 4 h.
|
|
|
In the generalized estimating equation analysis, only the free Hb concentration of the plasma sample (P < 0.001) and the time of plasma sampling after the start of CPB (P < 0.001) were significant determinants of the concentration of cyanide anion generated by the addition of SNP. The concentration of cyanide generated by the addition of SNP to plasma samples was independent of patient age, patient weight, gender, baseline total hemoglobin concentration, type of operation performed, lowest nasopharyngeal temperature during CPB, use of deep hypothermic circulatory arrest, or volume of cell saver blood harvested during the operation.
DISCUSSION
Sodium nitroprusside contains 44% cyanide by weight and, thus, an important safety concern of SNP administration to humans is the potential for cyanide toxicity. The exact pathways of SNP breakdown within the body after IV administration have not been completely established, but SNP may undergo rapid degradation by reacting with white blood cells, hepatocytes, or erythrocytes to release cyanide anion (4,5,7,8). Among reactive species, hemoglobin was found to be the most biologically active substance capable of generating cyanide anion from SNP (4). Hemolysis, as a consequence of injury to cells during CPB, may expose a greater quantity of hemoglobin available to react with exogenously administered SNP, posing an increased risk of cyanide toxicity.
The observation that plasma from patients undergoing CPB generated cyanide anion upon SNP exposure in amounts that correlated with the plasma-free Hb concentrations was consistent with the ability of SNP to react with hemoglobin. The finding that very little cyanide anion was released in response to SNP from the plasma of patients when plasma-free Hb concentrations was low or undetectable was also consistent with previous experimental studies (4,5). In those experiments, cyanide was released when SNP was incubated with erythrocytes, but the amount of cyanide produced was increased 5- to 8-fold upon lysis of the erythrocytes (4,5). The reaction between SNP and hemoglobin is believed to occur as a nonenzymatic reaction with oxy- or deoxyhemoglobin, or by a nonspecific reaction with free sulfhydryl groups on the protein. The stoichiometric reaction between SNP and hemoglobin has the potential to generate the equivalent of one cyanomethemoglobin molecule and four free cyanide anions (9). The intact erythrocyte cell membrane likely serves as a barrier to the nitroferricyanide anion protecting hemoglobin within the cell from direct exposure to SNP in the plasma, and thereby limiting the amount of free cyanide generated in response to SNP administration.
The reaction between SNP and free hemoglobin in the plasma samples was rapid with maximum cyanide production attained within 30 min. This finding is also consistent with a nonenzymatic reaction between hemoglobin and SNP. The amount of cyanide released from a given plasma sample with a fixed plasma-free Hb, (i.e., 120 mg/dL) was dependent on the concentration of SNP added. Cyanide production at a fixed concentration of SNP also correlated significantly with plasma-free Hb concentrations, even though the measured values for plasma-free Hb encountered during the course of CPB were in the lower range of sensitivity for the spectrophotometric assay (Fig. 2). This plasma-free Hb-dependent and SNP dose-dependent release of cyanide indicated that even a small amount of hemolysis as a consequence of CPB had the capacity to generate a significant amount of cyanide anion.
To reliably capture and quantify precisely all cyanide anion generated from plasma samples upon exposure to SNP, the experiment was performed in the laboratory within a closed aeration apparatus. This approach was chosen to avoid spontaneous loss or metabolism of cyanide that limits the accuracy of analytic assays for detecting cyanide in stored clinical specimens. A limitation of this approach was that the addition of nitroprusside to plasma samples in the laboratory may not have duplicated actual conditions within the body. For example, the measured cyanide concentrations in plasma may have underestimated actual cyanide anion generation in whole blood in response to SNP, because cyanide anion concentration within the cellular fraction of blood was not analyzed. Alternatively, the measured cyanide concentrations may overestimate the actual risk of toxicity because the metabolic capacity to detoxify free cyanide was not considered. The concentrations of SNP studied may also not precisely reflect the actual plasma SNP concentrations achieved during IV administration. Therapeutic plasma concentrations of SNP are not precisely known because there are no analytic assays to determine the SNP concentration in clinical specimens. However, two previous clinical studies have reported increased cyanide concentrations as a consequence of SNP administration during CPB. In one of the studies, SNP infusion at 7.3 µg·kg–1·min–1, a dose in the high end of the recommended range, for a period of 20 min during hypothermic CPB in six patients caused a significant increase in erythrocyte cyanide concentrations (10). In the other report, SNP administered at the seemingly low dose of 1 µg· kg–1·min–1 during hypothermic CPB was associated with increased erythrocyte cyanide concentrations (11). The authors speculated that deliberate hypothermia during CPB increased cyanide concentrations in response to SNP administration by impairing the enzymatic conversion of cyanide to thiocyanate through limiting the availability of thiosulfate substrate or decreasing the activity of thiosulfate sulfate transferase (rodanase). Based on the present study, the accelerated breakdown of SNP due to hemolysis as a consequence of CPB provided an additional, or alternative, explanation for the observation of elevated cyanide levels in response to SNP administration during cardiac operations as observed in those earlier studies.
Among verified cases of cyanide toxicity caused by SNP, CPB has been identified as a risk factor (1). In one reported clinical series, 7 of 292 patients undergoing coronary artery bypass grafting were diagnosed with cyanide toxicity as a consequence of perioperative SNP administration (3). In that report, cyanide toxicity occurred even at doses of SNP that were considered to be safe in humans. In another clinical investigation, elevated thiocyanate levels were detected in 44% of patients who received SNP after cardiac surgery, but no overt cases of cyanide toxicity were identified (12). Critical illness causing increased sensitivity to the toxic effects of cyanide, impaired ability to detoxify cyanide by metabolism to sodium thiocyanate, or defects in the hepatic or renal clearance of cyanide were proposed explanations of the increased risk of cyanide toxicity in cardiac surgical patients treated with SNP. The demonstration in this study that hemolysis from CPB has the potential to increase cyanide release in response to SNP provides an additional explanation for the occurrence of SNP-induced cyanide toxicity in cardiac surgical patients. It is possible that clinically significant cyanide toxicity may be under-diagnosed among cardiac surgical patients, because laboratory testing for cyanide toxicity is rarely performed and is not always reliable. Furthermore, the signs and symptoms of cyanide toxicity, such as tachyphylaxis to SNP, hypotension, encephalopathy, or metabolic acidosis, are nonspecific and common among cardiac surgical patients immediately after operation (2,3).
In conclusion, CPB was associated with hemolysis that increased over time during CPB. Hemolyzed plasma from cardiac surgical patients increased the release of free cyanide anion from SNP. The release of cyanide anion from SNP in the presence of hemoglobin may provide an explanation for reported cases of cyanide toxicity after SNP administration in patients undergoing operations requiring CPB. Although a heterogeneous cardiac surgical patient population with a wide range of CPB duration was studied to examine a broad range of conditions that may lead to cyanide generation from SNP, it is possible that the potential for cyanide toxicity from SNP may vary depending on the type of operation performed, the specific equipment used to conduct CPB, and other factors unique to an individual practice setting. It was also possible that other effects of CPB unrelated to hemolysis, such as the activation of leukocytes, release of mitochondrial enzymes into the circulation, or the generation of reactive oxygen species, that were not measured, also contributed to the observed release of free cyanide anion from SNP. Even if these other effects of CPB accelerated the release of cyanide from SNP, plasma-free Hb may still serve as a marker indicating the potential for SNP-induced cyanide toxicity in cardiac surgical patients.
ACKNOWLEDGMENTS
The authors gratefully acknowledge Doreen C. Cowie, John C. Haddle, Lin Tian, and other members of the Perfusion Department of the Hospital of the University of Pennsylvania for their invaluable assistance with the conduct of the study.
Footnotes
Accepted for publication March 7, 2007.
The study was funded in part by an unrestricted research grant from ESP Pharma under a sponsored-research agreement with the University of Pennsylvania. ESP Pharma marketed nicardipine.
Albert T. Cheung, MD, has received speaking honoraria from ESP Pharma.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
