Improving accuracy in calculating estimated blood loss
Perioperative nurses partner with a software engineer to create an EBL calculator program
In the perioperative environment, blood loss is an inevitable aspect of most surgical procedures. An accurate measurement of blood loss is essential for recognizing potential life-threatening hemorrhage and managing blood product replacement. Intraoperative blood loss plays a significant role in predicting surgical outcomes, specifically in regard to patient mortality, which has been shown to increase when blood transfusions are given (Stahl, Groeben, Kroepfl, Gautam, & Eikermann, 2012).
In the perioperative setting in our tertiary care teaching hospital, the practice used to measure estimated blood loss (EBL) is a combination of direct and gravimetric (weight) measurements. In our facility, it is the perioperative nurse’s role to perform EBL calculations, using a formula of adding suction amounts (fluid volume) and weight of the used surgical sponges, then subtracting the amount of irrigation used. This formula takes into account the weight of the sponges before use and the amount of irrigation left on the sterile field. The calculation process is labour-intensive and prone to error. The fast pace of the surgical environment further adds to the potential for making a mistake.
Each calculation, which may be repeated many times during a surgery, and result are captured on an operating room whiteboard for the entire surgical team to see. Results are also verbally reported to the anesthesia team. This information is an integral part of the decision-making about whether to transfuse blood or blood products.
When our team conducted a root-cause analysis stemming from an investigation of two sentinel events in the OR related to blood loss, one of the anesthesia residents noted that the length of time required to calculate EBL was causing delays in the decision-making process.
We decided to submit a proposal for a quality improvement project that would improve the accuracy and efficiency of the calculation. We received funding from a Krembil Foundation nursing research award, which allowed us time away from the operating room to plan and implement the project and design a computerized calculator tool. We were given access to mentors to guide us through the research process and financial support to disseminate our results.
In the first phase of the project, we interviewed novice and experienced perioperative nurses, as well as members of the anesthesia team, for feedback on how EBL was being calculated. Frequently, we heard that there were problems with the process. In addition to the interviews, we observed 30 neurosurgery and spine surgery cases from start to finish, using a checklist we developed to look for inconsistencies in the nurses’ practice when calculating EBL. We also recorded the length of time it took for nurses to do the calculation.
Our observations revealed that 49 per cent of the time, at least one type of error was made. These included making mathematical errors, omitting a value required for the calculation and using containers that are not designed for measuring. The average time to calculate EBL was 146 seconds. The frequency of errors was surprising but confirmed for us the need to come up with a better process.
In phase 2, we worked with the anesthesia resident to create the first version of the tool: an Excel spreadsheet that automatically calculated EBL once all the required values had been entered. However, in testing the tool with nurses, it was obvious they did not find the layout user-friendly. To address this, we enlisted the help of a software engineer, a friend who volunteered his time to create a customized calculator program to guide nurses through the process.
When we demonstrated the program to the perioperative nurses in an in-service session, they gave us much more positive feedback. With our manager’s support, we submitted a special request to our IT department to load the program onto the computers in each nursing station in the ORs. Although we were confident this tool was going to reduce errors, getting buy-in from the IT department took longer than expected because the program had been created externally.
About a week after the tool had been fully implemented, we again asked for feedback from the nurses and, using the same checklist, conducted an observation of a second group of 30 neurosurgery and spine surgery cases.
The positive feedback confirmed that the tool is a more accurate and efficient means of calculating EBL. Calculation errors had decreased to 22 per cent, and the average time taken for the calculation was 109 seconds. Overall, errors decreased by 27 percentage points, and the time taken to calculate blood loss decreased by more than 25 per cent.
Despite this obvious success, we noticed that some of the more experienced nurses were hesitant about continuing to use the new technology. To address their concerns about this change in practice, we provided additional one-on-one support.
Results of a post-implementation survey of perioperative nurses we conducted a few months later indicated that the tool was being used most often in the neurosurgery and spine surgery cases as well as in high-volume blood loss orthopaedic surgery cases. For surgeries in which the expected volume of blood loss was low, the tool was not commonly used.
The entire process from design to implementation took more than two years. We learned early on the importance and value of input from the perioperative nursing staff — our key stake-holders — to help guide the evolution of the tool. With their input, we were able to create a truly user-friendly calculator program, with capability and function the earlier version did not provide.
As our second round of observations showed, use of a computerized calculator program does not eliminate errors made in calculating EBL; it cannot replace the clinical judgment and experience nurses bring to the operating room. However, this finding reinforced what we knew about the importance of ensuring new staff have a good understanding of the formula and all the values required to get an accurate and timely result.
We are fortunate to have been able to disseminate our results through poster and oral presentations, at our facility and at provincial and national conferences dedicated to perioperative nursing practice.
Two years after completing the project, we can report that the tool is being used in additional surgical services at our facility. Our hope is that it will eventually be available in ORs in other hospitals within University Health Network.
The authors thank the Krembil Foundation for its support and funding throughout this quality improvement project
Stahl, D. L., Groeben, H., Kroepfl, D., Gautam, S., & Eikermann, M. (2012). Development and validation of a novel tool to estimate peri-operative blood loss. Anaesthesia, 67(5), 479-486. doi:10.1111/j.1365-2044.2011.06916.x