RESEARCH. Introduction: Re-collection of hemolyzed blood specimens delays. Methods: A prospective, cross-over study of blood collection

Transcription

1 RESEARCH Nursing Blood Specimen Collection Techniques and Hemolysis Rates in an Emergency Department: Analysis of Venipuncture Versus Intravenous Catheter Collection Techniques Authors: Glynnis Lowe, RN, CEN, Rose Stike, MEd, EMT-B, Marc Pollack, MD, PhD, Jenny Bosley, RN, BSN, CEN, Patti O Brien, RN, SANE, Amy Hake, RN, BSN, CEN, PHRN, GretaLandis,RN,BSN,CEN,NatalieBillings,RN,BSN,CEN, Pam Gordon, RN, Steve Manzella, PhD, and Tina Stover, MT(ASCP), York, Pa Glynnis Lowe is Clinical Leader, Department of Emergency Medicine, Rose Stike is Performance Improvement Specialist, Department of Emergency Medicine, Marc Pollack is Research Director, Department of Emergency Medicine, Jenny Bosley is Clinical Nurse Educator, Department of Emergency Medicine, Patti O Brien is Clinical Nurse, Department of Emergency Medicine, Amy Hake is Clinical Nurse, Department of Emergency Medicine, York Hospital, York, Pa. Greta Landis is Clinical Nurse, Department of Emergency Medicine, Natalie Billings is Clinical Nurse, Department of Emergency Medicine, Pam Gordon is Clinical Nurse, Department of Emergency Medicine, Steve Manzella is Core Laboratory Director, Department of Laboratory Services, Tina Stover is Operations Manager of Laboratory Services, Department of Laboratory Services, For correspondence, write: Glynnis Lowe RN, CEN, York Hospital, Emergency Department, 1001 South George St, York, PA 17405; glowe@wellspan.org. J Emerg Nurs 2008;34: Available online 25 September /$34.00 Copyright n 2008 by the Emergency Nurses Association. doi: /j.jen Introduction: Re-collection of hemolyzed blood specimens delays patient care in overcrowded emergency departments. Our emergency department was unable to meet a benchmark of a 2% hemolysis rate for the collection of blood samples. Our hypothesis was that hemolysis rates of blood specimens differ dependent on the blood collection technique by venipuncture or intravenous catheter draw. Methods: A prospective, cross-over study of blood collection techniques in a 64,000 annual visit, community teaching hospital emergency department was conducted. Eleven experienced registered nurses with more than 2 years ED experience completed a standardized phlebotomy retraining session. Registered nurses were randomly assigned to collect samples via intravenous catheters or venipuncture. After nurses collected 70 samples, they then collected samples via the other method. A standardized data collection form was completed. Blood samples were processed and assessed for hemolysis using standard procedures by laboratory technicians who were blinded to the collection method. Results: A total of 853 valid samples were collected; 355 samples (41.6%) were drawn via venipuncture and 498 samples (58.4%) were drawn through an intravenous catheter. Of these, 28 intravenous catheter samples (5.6%) were found to be hemolyzed, whereas only 1 venipuncture sample (0.3%) was hemolyzed. This finding was significant (x 2 b 0.001). Discussion: Experienced ED nurses can reduce the number of hemolyzed specimens by collecting via venipuncture instead of through intravenous catheters. This practice should be considered as standard of care in the ED setting. Limitations: Total samples by nurse were affected by EMS patients arriving with existing intravenous lines, and nurse schedules affected total samples per nurse. 26 JOURNAL OF EMERGENCY NURSING 34:1 February 2008

2 Overcrowding and increased lengths of stay in emergency departments have become nationwide issues. Patient admission rates from the emergency department and hospital elective surgery schedules contribute to hospital gridlock because of a lack of open inpatient beds to meet these demands. This situation can negatively affect ED patient throughput. In addition to external factors, process delays intrinsic to the emergency department impede patient f low. One such factor is the time involved in the re-collection of hemolyzed blood specimens, which can cause delays in patient care decisions that make additional ED treatment beds available or in identifying potentially critical situations. Hemolysis is defined as a rupture of red blood cells with release of hemoglobin and other intracellular contents into the plasma that can alter laboratory test results. Collection technique can affect the quality of blood specimens, with the main factors causing hemolysis being: (1) intravenous catheter size, (2) pressure exerted by vacutainer, (3) vigorously shaking the tube, and (4) overmanipulation of the extremity. 1,2 The patient also can be a difficult stick, especially in the case of elderly or dehydrated patients, which predisposes a collected specimen to hemolysis. In the ED setting, time is an important factor in the assessment, diagnosis, stabilization, and subsequent treatment of ill or injured patients, including trying to save time in collecting laboratory specimens from patients. Nurses often will collect blood specimens from peripheral intravenous catheters when the line is initiated to increase efficiency and decrease patient discomfort by avoiding a second venipuncture. If acceptable blood specimens could consistently be collected from a newly placed peripheral intravenous catheter, as opposed to venipuncture, patients would be spared further discomfort, money could be saved in both staff time and supplies, and specimen collection may be easier for patients with multiple running intravenous lines or difficult peripheral access. 3 Both phlebotomy techniques are routinely used in the emergency department and are acceptable standard nursing practice. However, increased hemolysis rates have been attributed to the practice of obtaining blood specimens through intravenous catheters. This results in the recollection of many specimens, causing the patient to indeed undergo the separate stick that the nurse wanted to avoid and an overall delay in care. The American Society of Clinical Pathology has identified the benchmark of a 2% or less hemolysis rate as best practice. The Emergency Service Line and Laboratory Services in this institution monitor hemolysis rates for specimens collected in the emergency department. In May 2004, a process change occurred where the Laboratory staffed the emergency department with a dedicated phlebotomist for 12 consecutive hours (10 AM-10 PM) each day to assist with blood specimen and culture collection. Phlebotomists were available by pager during off-hours. The phlebotomist assigned to the emergency department performed blood specimen collection using only venipuncture technique. Hemolysis rates decreased to meet benchmark levels during this pilot test period. However, because of inconsistent laboratory phlebotomist coverage and ED staff preference, the phlebotomist coverage eventually was eliminated in June Subsequently, hemolysis rates for ED specimens increased and the benchmark of 98% acceptable specimen rate could not be sustained for specimens collected by ED staff. Studies have confirmed the observation that laboratory phlebotomists are better able to collect acceptable specimens compared with ED staff collectors. 4 Because of laboratory staffing shortages, dedicating a phlebotomist to the emergency department for specimen collection was not possible in our institution. Other studies, some with limited sample sizes, have reported the effect of collection tube size and effect of intravenous site versus venipuncture collection technique on hemolysis rates. 2,5-8 Further investigation into hemolysis rates by collection technique used by ED staff was warranted. The objective of this project was to measure the incidence of hemolysis in blood samples drawn by venipuncture versus through newly placed intravenous catheters in ED patients. The outcome of this project will guide policy and best practice standard development regarding phlebotomy in our emergency department. Methods This research project was conducted as a prospective, crossover study from a nonconsecutive sample of ED patients (aged 18 years and older) in a 450-bed level II trauma February :1 JOURNAL OF EMERGENCY NURSING 27

3 TABLE 1 Hemolysis project data collection form contents Patient identification label Venipuncture data Intravenous catheter data Shift when collection done Butterfly size Unable to start or draw through intravenous line (yes/no) Date of collection Unable to collect (yes/no) Number of attempts Catheter size Assigned study nurse No. No. of attempts Rating for ease of f low: normal, slow, or unable to obtain Rating for ease of flow: normal, slow, or Blood collection site unable to obtain Blood collection site Collection tube size and top color Collection tube size and top color Hemolyzed specimen (yes/no) Hemolyzed specimen (yes/no) Comments Comments center and community teaching hospital. Institutional Review Board approval was obtained. There were no changes in the standard of care for blood specimens obtained from patients included in the study. The nurse collector group consisted of 11 experienced ED nurses (EDRNs) with more than 2 years of ED experience who completed a standardized phlebotomy retraining session. All members of the nurse collector group signed a consent form to confirm voluntary participation in the study and adherence to study protocol. ED registered nurses (RNs) were randomly assigned a number as his or her unique identifier and were randomly assigned to collect samples via intravenous catheters or venipuncture. Collectors were to start with their original collection technique assignment for 70 samples and then cross over to the other technique. A data collection form was completed for each specimen sent to the laboratory. Data collection points are listed in Table 1. The nurse collector determined the size of butterf ly or intravenous catheter used. For the intravenous catheter group, the blood was drawn only through a newly started intravenous line. Specimen collection technique is described in Table 2. After obtaining blood specimens, the nurse collector was instructed to immediately send the specimen to the laboratory. Personnel in the laboratory were blinded to study specimens and phlebotomy technique. Variables related to tube handling or time between draw and specimen transport were not measured. Data collection forms were collected daily by the ED Clinical Leader, and data were entered into the Remark Office OMR scanning software program. For the purposes of this study, hemolysis was defined as any level of visually detectable hemoglobin as determined by an experienced laboratory technician. Prior to beginning this study, efforts were made to standardize the visual assessment of hemolysis levels between laboratory personnel. Digital photographs were posted in the lab showing the 5 levels of hemolysis (Figure 1). The pictures were displayed in the laboratory along with the list of tests most affected by hemolysis and to what degree. Laboratory personnel were trained to use the guidelines listed in Table 3 and standardized digital pictures to assess hemolysis. A Beckman Coulter LX-20 serum indexes monitor was used to assign hemoglobin quantities for each hemolyzed specimen. The Core Lab Director provided weekly hemolysis reports for blood specimens collected in the emergency department. Study nurse collectors were identified from this list, and hemolyzed specimen information was entered into the Remark database. A 100% verification of the data was completed prior to the analysis of the data using SPSS software (SPSS for 28 JOURNAL OF EMERGENCY NURSING 34:1 February 2008

4 TABLE 2 Specimen collection technique* Venipuncture 1. Perform phlebotomy stick, with gloves on, using a 21-gauge or 23-gauge BD butterfly device with vacutainer attachment at end of tubing Intravenous catheter 1. Perform intravenous stick with gloves on 2. If flash, thread catheter and withdraw stylet 3. If no flash, reposition intravenous needle until flash 2. If flash, collect specimens using correct order of draw obtained or abort attempt 3. If no flash, reposition needle until flash obtained, or abort attempt 4. Attach vacutainer to hub of intravenous catheter and collect specimens, using correct order of draw 4. Remove tourniquet upon obtaining specimens 5. Remove tourniquet upon obtaining specimens 5. Withdraw needle while still engaged in vein; apply cotton ball or gauze to site prior to disengaging needle from vein/skin 6. Remove vacutainer from hub of intravenous catheter and attach saline lock; flush intravenous catheter and dress intravenous site 6. Apply pressure to site 7. Rotate tubes 7. Rotate tubes 8. Label tubes with patient labels 8. Label tubes with patient labels 9. Place tubes in bag and label outside of bag with departmental collector identification and time of collection 9. Place tubes in bag and label outside of bag with departmental collector identification and time of collection 10. Place specimens in pneumatic tube for transport to laboratory 10. Place specimens in pneumatic tube for transport to laboratory *Hospital phlebotomy policy was applied to the standardized training curriculum for specimen collection, including the general procedures of verification of laboratory tests ordered, gathering supplies and taking to bedside, introduction of self and explanation of procedure to patient, hand washing, check of patient identification, application of tourniquet and collection site selection, removal of tourniquet while prepping site and assembling equipment, and donning gloves. The table shows the steps that were utilized for each collection technique. Windows, Rel. 14, 2006, Chicago: SPSS Inc). Data analysis was obtained using SPSS software. Chi-square and cross tab analysis were completed. Tests were considered significant if P values were less than.05. Results A total of 857 data collection forms were completed by the nurse collector group, with 4 excluded because of missing information regarding collection method. A total of 853 blood specimens were included in data analysis for this study during the period of April 5 to May 30, Sixty-two percent of the specimens were drawn during day shift (7 AM-3 PM), 29% were collected during evening shift (3 PM-11 PM), and 9% were collected during night shift (11 PM-7 AM). There was no selection bias for nurse collectors and shift worked. Cross tab analysis of specimen types and hemolysis rating is listed in Table 4. Hemolysis rate was higher in the intravenous catheter collection technique group compared with the venipuncture group ( P b.001). The total number of specimens for each individual nurse varied because of EMS patients arriving with intravenous access already obtained, collector staffing schedules, and assignment at triage area resulting in less phlebotomy being performed. Two ED RNs had higher hemolysis rates than other collectors. Separate analysis was completed to remove those 2 ED RNs from analysis, and intravenous catheter collection technique still resulted in a higher hemolysis rate than venipuncture technique ( P =.006). The single hemolyzed specimen from the venipuncture group was drawn from the antecubital site. The intravenous catheter collection group had hemolyzed specimens as shown in Table 5. Hemolysis rates by collection site was not statistically significant ( P =.088). Further analysis into hemolysis rate by blood collection site for intravenous catheters revealed a 3.1% hemolysis rate for 20-gauge needles at the antecubital site. While this was the lowest hemolysis rate for the intravenous catheter group, this still exceeds the institutional benchmark of 2% set by the February :1 JOURNAL OF EMERGENCY NURSING 29

5 FIGURE 1 Standardized Visual Laboratory Hemolysis Assessment. Extent of hemolysis for patient specimens was quantitated, in mg/dl hemoglobin, by use of a Beckman LX-20 specimen integrity monitor. This information was displayed with the picture above in the laboratory testing area along with a corresponding list of tests affected by these levels of hemolysis (Table 3). laboratory. Analysis of the data regarding catheter size and collection site was not significant. The number of specimens in each group was too small to draw any conclusions. Discussion As emergency departments across the United States face the same overcrowding dilemma, different ways to improve patient throughput should be explored. One potential way to decrease laboratory turnaround and patient disposition times is by sending nonhemolyzed specimens to the laboratory and avoiding the need for re-collection. 9 It has already been established that trained phlebotomists who have the primary responsibility for and certification in blood specimen collection can attain acceptable specimen rate goals. 4 The issue in our institution was the lack of available phlebotomist coverage for our emergency department to institute this approach. The logical next step was for a clinical research project to be dedicated to further investigating best practice guidelines regarding phlebotomy. Regulatory agencies often mandate patient care standards. A common approach with management teams is to TABLE 3 Laboratory tests most affected by hemolysis Analyte Ammonia Amylase Aspartate aminotransferase Iron Lactate dehydrogenase Potassium Folate Vitamin B12 Total bilirubin Magnesium Lipase Lithium Ferritin Uric acid Acceptable limit of hemoglobin 1 (50 mg/dl) 3 (150 mg/dl) 4 (200 mg/dl) 5 (250 mg/dl) 6 (300 mg/dl) create policy and standard operating procedure documents based on these regulatory or core measure patient care indicators. Performance standards oftentimes are not met because of nursing staff current practices, ED patient volumes placing overwhelming workload volumes on staff, and skepticism about the reasons for policy and practice changes. Evidence-based practice provides a way to frame and address questions about how to provide the best patient care. 10 When nurses are actively involved in answering best practice questions through clinical research, management is more likely to realize sustained success in implementing process changes. Our study showed that acceptable blood specimen rates can be achieved by nurses using venipuncture as the preferred collection technique over intravenous catheter collection. The sample size in this study was larger than others exploring this issue. This study was effective in demonstrating that shared decision making is an effective model for involving staff in performance improvement activities. In the month since completing this study and instituting a best-practice blood specimen collection policy in our emergency department, we have been able to keep the hemolysis rate for specimens collected by ED staff to less than 2%, thus meeting the standard for care. 30 JOURNAL OF EMERGENCY NURSING 34:1 February 2008

6 TABLE 4 Specimen hemolysis by group cross tabulation Implications For ED Nurses Most of the literature dedicated to the topic of hemolyzed laboratory specimens recommends a dedicated phlebotomist in the emergency department. The study has proven that by using the most effective collection technique, ED nurses can obtain an increased rate of acceptable blood specimens similar to that of laboratory phlebotomists. Reduced recollection of specimens can potentially improve physician decision-making time and decrease overall lengths of stay for ED patients, especially those who are discharged from the emergency department when laboratory results are received and can be acted upon, making scarce ED beds available for other patients. Applying evidence-based practice theory to ED nursing is vitally important not only to promote patient care that is safe, timely, efficient, effective, equitable and patient-centered but also that will empower ED nurses to become involved in clinical activities that will make their work more rewarding. This research model could be applied to other patient care issues that involve nursing, such as blood culture contamination rates. Limitations Specimen hemolyzed Group No Yes % hemolysis Total Group Venipuncture b1 355 Intravenous catheter Total The study was designed for each nurse collector to submit the same number of specimens and data collection forms. One nurse collector dropped out of the study as a result of bereavement leave. Specimen collection for the intravenous catheter group was limited by patients arriving by EMS with intravenous access already obtained, which excluded the patient from the intravenous catheter draw group. Given these unanticipated occurrences, a mid-study analysis was completed and showed statistical significance that resulted in a decision to stop the study. Waiting until each nurse collector reached goal specimen numbers would have greatly delayed the study time period. Interim analysis revealed such a major difference that the study was stopped and policy was changed. The nurse collector group was limited to RNs with more than 2 years ED experience. Including all ED RNs could have increased the nurse collector group and the final sample size; however, the majority of the nursing staff in our emergency department has more than 2 years ED experience. Excluding new RNs most likely did not affect study results because all nurse collectors received standardized training prior to specimen collection. Certain components of the phlebotomy process such as tourniquet time, time from specimen collection to analysis, and mode of transport to the lab (pneumatic tube vs manual) were not collected. These data could have further refined analysis and should be considered for future research in this area. Conclusions The outcomes of this study have guided a policy development specifying practice guidelines that venipuncture should be the standard of care for drawing blood samples with the exception of high-acuity patients and patients who have difficult venous access. Initial results since the conclusion of this study have shown that ED nurses can consistently obtain the 98% benchmark set by our laboratory if the standard is venipuncture first. Because change can be a difficult process and our emergency department is staffed with assistance from traveling nurses and agency nurses, a future challenge will be to mandate and monitor compliance with this policy for temporary employees. Another challenge will be educating the newer staff on the policy and enforcing it. By collaborating with laboratory personnel, the emergency department continues to track the percentage of nonacceptable specimen rates and follows up with nurses who are noncompliant. Future research in this area could include intravenous catheter size and hemolysis rates for a larger sample size similar to that obtained in this study. ED nurses of all experience levels February :1 JOURNAL OF EMERGENCY NURSING 31

7 TABLE 5 Hemolysis by blood collection site cross tabulation Blood collection site Specimen hemolyzed Antecubital Blank Forearm Hand Mult Wrist Total No Venipuncture Intravenous catheter Total Yes Venipuncture Intravenous catheter Total could be included in the collector groups and hemolysis rates could be compared by site of venipuncture. This research demonstrates that acceptable phlebotomy technique can be performed by ED nursing staff to provide safe and efficient patient care and facilitate clinical decision making with decreased specimen recollection rates. Improving specimen acceptability results in quicker laboratory results to physicians who can then disposition ED patients sooner and positively affect ED throughput. Acknowledgments The following individuals are gratefully acknowledged for their participation in this project: ED RN nurse collectors: Blythe Stover- Baker, RN, CEN, Theresa Thomas, RN, CEN, Michelle Mahan, RN, and Francis McKniff, RN; study design input and manuscript review: Ron Benenson, MD, Department of Emergency Medicine; initial study design input and data analysis: Melissa Schlenker and Ted Bell, Emig Research Center; phlebotomy training: June Wineholt, RN, CEN, ED Nursing Clinical Leader; and manuscript review and final submission: Amy Daugherty, MS, ED Clinical Research Associate. 5. Cox S, Dages J, Jarjoura D, Hazelett S. Blood samples drawn from IV catheters have less hemolysis when 5-mL (vs 10-mL) collection tubes are used. J Emerg Nurs 2004;30: Dugan L, Leech L, Speroni K, Corriher J. Factors affecting hemolysis rates in blood samples drawn from newly placed IV sites in the emergency department. J Emerg Nurs 2005; 31: Kennedy C, Angermuller S, King R, Noviello S, Walker J, Warden J, Vang S. A comparison of hemolysis rates using intravenous catheters versus venipuncture tubes for obtaining blood samples. J Emerg Nurs 1996;22: Seeman S, Reinhardt A. Blood sample collection from a peripheral catheter system compared with phlebotomy. J Intravenous Nurs 2000;23: Becan-McBride K. Laboratory sampling. J Intravenous Nurs 1999;22: Pipe T, Wellik K, Buchda V, Hansen C, Martyn D. Implementing evidence-based nursing practice. MedSurg Nurs 2005;14: REFERENCES 1. Bush V. The hemolyzed specimen: causes, effects, and reduction. ED Lab Notes 2003;1: Grant M. The effect of blood drawing techniques and equipment on the hemolysis of ED laboratory blood samples. J Emerg Nurs 2003;29: Himberger JR, Himberger LC. Accuracy of drawing blood through infusing peripheral intravenous lines. Heart Lung 2001;30: Yoshikawa N. Hemolysis in serum samples drawn by emergency department personnel versus laboratory phlebotomists. Lab Med 2002;33: JOURNAL OF EMERGENCY NURSING 34:1 February 2008