Interest in blood conservation techniques is growing in the United States. This interest stems from a shortage of blood components available for transfusion (1). Likewise, a growing awareness of errors in medicine has attracted attention because of a recent Institute of Medicine (IOM) report on errors (2). This IOM report documents that from 44,000 to 98,000 deaths a year result from a broad range of medical errors. The conclusion of the IOM report was that most medical errors result from the basic organization of the health care system. In this case report, a description of two errors that were made during the course of cell salvage is reported. In the first case, absence of a quality assurance system and lack of an organized cell salvage program contributed to a patient receiving a cell salvaged unit of blood that was contaminated with glycine. In the second case report, changes instituted as a result of a quality improvement process prevented a similar occurrence.
A 73-yr-old male was undergoing the repair of an 11-cm Type IV thoracoabdominal aneurysm. Cell salvage was requested for the surgical procedure. Cell salvage was initiated with a Hemonetics Cell Saver 4(Hemonetics Corp., Braintree, MA) managed by the operating room (OR) circulating nurse. Shed blood was heparinized and scavenged into a standard 120-μm cardiotomy reservoir. Blood processing used a 225-mL Latham bowl with 0.9% saline solution as the wash solution. Immediately after release of the aortic cross-clamp, vigorous blood loss required 6 U of blood to be processed rapidly. Besides managing the cell salvage equipment, the circulating nurse was performing other operating room tasks. These duties included obtaining supplies for the operating room technician, computer entry for supply charges, answering pages for the surgeons, maintaining OR paperwork, and assisting the anesthesia provider as needed. During this period of heavy blood loss, the bag of wash solution was changed. Instead of 0.9% saline solution, the nurse connected 1.5% glycine solution, which is a solution normally reserved for bladder irrigation during the performance of transurethral resection of the prostate (TURP). One unit of blood was processed and administered to the patient. On processing of the second unit, the anesthesiologist managing the patient noticed the solution error. Cell salvage was discontinued and the surgery was completed without other complications. Glycine and ammonia levels were sent immediately after discovery of the error but were found to be within normal limits. After the surgery, the patient was transported to the intensive care unit, where he had an uneventful recovery.
A 74-yr-old patient was undergoing an 8-cm type IV thoracoabdominal aneurysm repair. Cell salvage was requested for the surgical procedure. Cell salvage was initiated using a Medtronic Sequestra (Medtronic, Inc., Minneapolis, MN) which was managed by a technician dedicated to the cell salvage equipment. Shed blood was heparinized and scavenged into a standard 120-μm cardiotomy reservoir. Blood processing used a 225-mL Latham bowl with 0.9% saline solution as the wash solution. After the processing of five bowls of blood, the technician noticed that the effluent supernatant line did not clear of its red coloration and that the color of the blood did not appear to have a normal color (Fig. 1). A routine hemoglobin check of the processed blood was found to be 9 gm/dL, well below expected values. On noticing these three events, the technician noticed that instead of using 0.9% saline as a wash solution, he had used sterile water. The processed blood was discarded, the system was changed, and the case proceeded uneventfully, with the patient ultimately losing 21,000 mL of blood with 5045 mL being processed and returned from cell salvage.
As anesthesiologists become more interested in blood conservation, an understanding of the complications related to cell salvage and how to prevent complications related to cell salvage is important. These two cases illustrate how similar labeling of three irrigation solutions often used in an OR can lead to errors during cell salvage. It also demonstrates the need for a quality assurance system to help prevent errors during cell salvage.
In the first case, the cell salvage equipment was managed by a circulating nurse who was responsible for multiple OR tasks in addition to managing the cell salvage equipment. In this system, cell salvage processing generally occurred at the convenience of the nurse because of the multiple tasks that were assigned to her. As a result, shortcuts have been taken that included processing of incomplete bowls, which can lead to inadequate wash quality (3), a failure to adequately anticoagulate the salvaged blood, which can lead to poor red cell recovery, and occasional errors in processing as described here. The error described in the first case occurred at a time when the nurse was trying to respond to multiple demands during a critical period of the surgery when acute blood loss was occurring.
After this case, a cell salvage program was established that complied with the recommendations of the American Association of Blood Banks (AABB) which proposes “Standards for perioperative collection and Transfusion” (Table 1) (4). In these proposed standards is guidance for product testing. For cell salvage, the recommendations are for inspection of the effluent supernatant color, a measure of the product hematocrit, measurement of the weight of the product, and visual inspection of the final blood product. In addition, these guidelines recommend that a dedicated technician who has undergone training and periodic continuing education, rather than an individual who has multiple OR job responsibilities, be in charge of the cell salvage process. In our program, technicians with a minimum educational background of an associate’s degree were hired and trained via the cell salvage equipment manufacturer’s training program. Additionally, weekly educational meetings are held to discuss cell salvage management and program problems. Because the educational background of these individuals is not extensive, minimal increases in the cost of producing a cell salvage product have occurred from implementation of the AABB standards. The value of these quality checks and a dedicated technician is illustrated in the second case. Here, a similar wash solution identification error was made but, unlike the first case, the error was recognized before administration of the blood.
The similar labeling of the 0.9% saline, glycine, and sterile water infusion bags led to these errors. (Fig. 2) When we inspected how this could have happened, we noticed that in our central core storage area all types of irrigation fluid (0.9% saline, sterile water, and 1.5% glycine) are kept in close proximity. Furthermore, we found several bags of glycine stored with 0.9% saline. In 1981, Wang and Turndorf (5) noted that up to 12% of drugs administered in hospitals were in error. This lead to an effort to standardize and enhance the labeling of drugs and prefilled syringes (6). This case would suggest that the same effort should be applied to the labeling of these irrigation fluids.
No injury resulted from either of these mistakes; however, in the second case, the consequences may have been significant. Administration of blood washed with sterile water has not been previously reported but a similar problem has been described with albumin dilution. From 1994 to present there have been ten reports of 25% albumin being diluted with sterile water and being administered to patients (7,8). In these albumin cases, the red cell hemolysis which resulted from the hypotonic solution caused renal insufficiency and failure, decreases in hematocrit, increases in serum lactate dehydrogenase, increases in total serum bilirubin concentration, disseminated intravascular coagulation, and, in two patients, death. Thus, it is extremely important to recognize this error to prevent serious patient injury.
The consequences of the glycine infusion may be less significant to patient outcome because of the isoosmotic character of the fluid, though the glycine administration may have caused transient blindness as may occur during the TURP syndrome. Glycine has a distribution similar to that of γ-aminobutyric acid, an inhibitory transmitter in the brain, and is believed to act as an inhibitory transmitter also at the spinal cord level, brainstem, and retina. In addition, absorption of glycine may result in central nervous system toxicity based on its oxidative biotransformation to ammonia. Subsequent analysis of the patient’s blood glycine and ammonia levels, and a finding that these levels were within normal range, suggested that the patient received small amounts of glycine with the unit of glycine-washed blood. This conclusion would be supported by the volume of glycine that was in the administered unit of washed cells. In general the average unit of cell salvage blood has a hematocrit of 60% with a standard volume of 225 mL. This means that approximately 90 mL of the unit was glycine solution.
In conclusion, these cases highlight the need to be vigilant when performing cell salvage. This report is also illustrative of the importance of adopting the AABB standards for perioperative cell salvage. In addition, it demonstrates that product labeling can be important in areas other than drug labels.
1. Wallace EL, Churchill WH, Surgenor DM, et al. Collection and transfusion of blood and blood components in the United States. Transfusion 1998;1994:38:625–36.
2. Kohn LT, Corrigan JM, Donaldson MS, eds. To err is human: building a safer health system. Washington, DC: National Academy Press, 2000.
3. Szpisjak DF. Debris elimination from partially filled cell salvage bowls. Anesth Analg 2001; 92: 1137–8.
5. Wang BC, Turndorf H. Prevention of medication error. N Y State J Med 1981; 81: 395–402.
6. Rendell-Baker L. Better labels will cut drug errors. Anesthesia Patient Safety Foundation Newsletter 1987; 2: 29–40.
7. Pierce LR, Gaines A, Varricchio F, Epstein J. Hemolysis and renal failure associated with use of sterile water for injection to dilute 25% human albumin solution. Am J Health Syst Pharm 1998; 55: 1057–70.
8. Centers for Disease Control and Prevention. Hemolysis associated with 25% human albumin diluted with sterile water–United States, 1994–1998. JAMA 1999:281:1076–7.