The Case

A 72-year-old man with congestive heart failure due to nonischemic cardiomyopathy, stage 3 chronic kidney disease, atrial fibrillation, and type 2 diabetes mellitus presented to the emergency department (ED) with new onset generalized weakness. The ED physician ordered a chemistry panel, which came back as showing evidence of acute kidney injury, as well as a serum potassium level of 7.4 mEq/L (normal range 3.5 to 5.0 mEq/L). However, the laboratory noted that the patient's blood sample was hemolyzed, which can spuriously increase the measured serum potassium.

The ED physician assumed that the first result must have been incorrect due to hemolysis, but felt that the patient should be admitted to the hospital for further workup and treatment of his acute kidney injury. She ordered a repeat potassium level and called the hospitalist for admission. The hospitalist noted mild electrocardiographic abnormalities that could have been consistent with hyperkalemia (high potassium levels), but she did not institute immediate treatment for hyperkalemia as the repeat level was still pending. The patient was admitted to the medical ward.

When the patient arrived at the floor, he started to feel more confused and lethargic and he became progressively more hypotensive and bradycardic. The laboratory then called the hospitalist with a panic result—the repeat potassium level had come back at 8.4 mEq/L, a level associated with a high risk of cardiac complications. Almost immediately thereafter, the patient went into cardiac arrest. The patient was resuscitated and emergently administered calcium gluconate, sodium bicarbonate, and insulin to treat the hyperkalemia. He was able to be resuscitated and was emergently transferred to the intensive care unit for initiation of hemodialysis. The patient had a lengthy hospital course, but he ultimately survived and was discharged home off dialysis and with improved renal function.

The Commentary

by Christopher M. Lehman, MD

Hemolysis is the most common reason for laboratory rejection of blood specimens, accounting for roughly 25% of rejected specimens.(1) Published rates of hemolysis range from under 1% to 30% or greater, and reportedly vary by location of collection (highest in emergency departments), patient population (higher in newborns), anatomic site of phlebotomy (lower from antecubital fossa), collection method (phlebotomy lower than collection through an intravenous line), specimen transport method (potentially higher with pneumatic tube transport), and the proficiency of collecting personnel (lower with trained, experienced phlebotomists).

Accurate detection of hemolyzed specimens by laboratory personnel is known to be inconsistent (2), complicated by the degree of hemolysis in the specimen, the capability and focus of the individual assessing the specimen, and the ability to visualize the supernatant of a centrifuged tube of blood that is often partially, or completely, covered by labels. Consequently, manufacturers have incorporated automated methods into chemistry analyzers to detect even subtle amounts of hemolysis (known as the hemolysis index). Assessment of the presence and degree of hemolysis in blood specimens submitted for analysis on nonchemistry analyzers must still be done visually by the laboratorian.

When a hemolysis index is reported by a chemistry analyzer, a semi-quantitative or quantitative estimate of the degree of hemolysis is provided. Hemolysis can interfere with assays through the following mechanisms: a) release of red blood cell (RBC) contents (e.g., potassium, lactate dehydrogenase [LDH]); b) interference with spectrophotometric measurements due to the broad absorption spectrum of hemoglobin (e.g., bilirubin); and c) nonspecific RBC membrane binding to components of immunoassays resulting in aberrant results (e.g., troponin). The hemolysis index can be used to assess the degree of interference with an assay through mechanism b cited above.

In many cases, mild to moderate hemolysis does not substantively interfere with assays, and so results can be released. However, potassium and LDH concentrations are exquisitely sensitive to hemolysis, even at the lowest hemolysis index, and the index cannot be used to estimate the increase in their concentrations due to hemolysis.(3) Consequently, identification of hemolyzed specimens and immediate communication with the clinician to determine if a second, nonhemolyzed specimen can be collected is critical. In occasional cases, acquiring a nonhemolyzed specimen may be impossible (e.g., patient with hemolytic anemia). In those instances, the laboratory should confer with the clinician about the need to release affected results and note in the report the potential effect of hemolysis. Clinicians should be aware that analysis of whole blood specimens on point-of-care analyzers does not allow for detection of hemolyzed specimens, unless the specimen is centrifuged after analysis and the supernatant examined. Therefore, clinicians should not assume that hemolysis does not affect point-of-care results.

In this case study, a new specimen was appropriately collected and analyzed to accurately assess the patient's blood potassium level, which was actually greater than the level in the hemolyzed specimen. This discrepancy could have been due to misidentification of hemolysis in the original specimen, or to an increase in the patient's actual potassium concentration during the interval between the collection and analysis of the two specimens. However, it was unlikely to have been due to an analytical error.(4,5)

Since it was not possible for the laboratory to estimate the potential impact of hemolysis on the potassium level of the patient, the laboratory's only recourse was to request and analyze a second specimen. Unfortunately, that did not help the clinicians determine if the signs and symptoms of the patient were due to an elevated potassium level, or some other cause, until the result from the analysis of the second specimen was reported. Clinicians must rely on their assessment of the patient's condition and other data to care for the patient while they await results from a nonhemolyzed specimen. In this case, the hospitalist probably should have instituted treatment for hyperkalemia before the results of the second specimen were available, since the patient had electrocardiographic changes consistent with hyperkalemia and the risk of instituting treatment was relatively low.

Decreasing hemolysis rates is challenging. Since hemolyzed specimens can put both clinicians and laboratory personnel in a difficult position, laboratories should proactively monitor and report hemolysis rates by clinical service and/or facility location to appropriate hospital and/or clinic personnel, identify locations with elevated rates, and work with nurses, phlebotomists, and clinicians to improve collection practices. Although various factors have been correlated with increased hemolysis rates (as described earlier), the published literature contains few well-designed studies that evaluate specific interventions to decrease hemolysis. The Laboratory Medicine Best Practices Workgroup has made recommendations for limiting hemolysis based on a systematic review and meta-analysis of the literature (Table).(6,7) The strongest evidence-based recommendation supports specimen collection in the emergency department using straight needle phlebotomy, rather than collection from an intravenous catheter. Targeting hemolysis rates in the emergency department is logical, as hemolyzed specimens are more common there and are more likely to directly affect patient care. However, the extent to which these best practices have been adopted is unclear.

Take-Home Points

• Hemolysis is the most common reason for blood specimen rejection by clinical laboratories.
• Hemolyzed specimens are most common when drawn in the emergency department.
• The laboratory cannot correct for the increase in analyte concentration resulting from release of red blood cell contents secondary to hemolysis (e.g., potassium).
• There are evidence-based recommendations for decreasing the rate of hemolyzed specimens, but few quality improvement interventions to reduce hemolysis have been published.

Christopher M. Lehman, MD Clinical Professor of Pathology University of Utah, School of Medicine Medical Director, University Hospital, University Neuropsychiatric Institute South Jordan Health Center Clinical Laboratories Salt Lake City, UT

References

1. Nakhleh RE, Souers RJ, Bashleben CP, et al. Fifteen years' experience of a College of American Pathologists program for continuous monitoring and improvement. Arch Pathol Lab Med. 2014;138:1150-1155. [go to PubMed]

2. Simundic A, Nikolac N, Ivankovic V, et al. Comparison of visual vs. automated detection of lipemic, icteric and hemolyzed specimens: can we rely on a human eye? Clin Chem Lab Med. 2009;47:1361-1365. [go to PubMed]

3. Mansour MM, Azzazy HM, Kazmierczak SC. Correction factors for estimating potassium concentrations in samples with in vitro hemolysis: a detriment to patient safety. Arch Pathol Lab Med. 2009;133:960-966. [go to PubMed]

4. Deetz CO, Nolan DK, Scott MG. An examination of the usefulness of repeat testing practices in a large hospital clinical chemistry laboratory. Am J Clin Pathol. 2012;137:20-25. [go to PubMed]

5. Lehman CM, Howanitz PJ, Souers R, Karcher DS. Utility of repeat testing of critical values: a Q-probes analysis of 86 clinical laboratories. Arch Pathol Lab Med. 2014;138:788-793. [go to PubMed]

6. Laboratory Medicine Best Practices. Effective practices to reduce blood sample hemolysis in emergency departments. [Available at]

7. Heyer NJ, Derzon JH, Winges L, et al. Effectiveness of practices to reduce blood sample hemolysis in EDs: a laboratory medicine best practices systematic review and meta-analysis. Clin Biochem. 2012;45:1012-1032. [go to PubMed]

Table