A popular Sherlock Holmes short story focuses on the disappearance of the racehorse, Silver Blaze, on the eve of an important race. The horse was stolen from the stables during the night in Victorian England.
Gregory (Scotland Yard police detective): “Is there any other point to which you would wish to draw my attention?”
Sherlock Holmes: “To the curious incident of the dog in the night-time.”
Gregory: “The dog did nothing in the night-time.”
Sherlock Holmes: “That was the curious incident.”
The “curious incident of the dog in the night-time” to which Holmes alludes is easily explained: The dog made no noise because the thief was no stranger to the dog.
HIL Indexes: Lipemia and More
Laboratorians utilize the HIL index to detect potential analytical interference from hemolysis, icterus, and lipemia (HIL) in serum and plasma samples. Thus, implicit to an HIL index alert is the presence of significant hemolysis, icterus, or lipemia in a sample. Indeed, in our experience, lipemia index alerts are accurate 99.9% of the time. But what does it mean when a lipemia alert flags a specimen with no visible evidence of lipemia or turbidity?
Measuring Wavelengths
Hemoglobin, unconjugated bilirubin, and lipids are detected photometrically at specific wavelengths. While the absorption spectra of hemoglobin and bilirubin overlap significantly, the photometric light scatter induced by lipids can be quite variable. This is due to the heterogeneity of endogenous lipids.
Our instrument at Rocky Mountain Regional VA Medical Center in Aurora, Colorado—the Siemens Dimension Vista 1500—measures lipid interference at the wavelength of approximately 700 nm (determined using intralipid fat emulsion). When HIL interference is detected, the analyzer records the photometric data as an HIL index value. The HIL index, in turn, corresponds to the approximate concentration of hemoglobin, bilirubin, or lipid present in the sample. On our analyzer, lipemia is reported as an index value between 1 and 8. An HIL index alert notifies our staff when a sample has an HIL index value known to produce a 10% or greater change in analyte quantification.
Not all analytes are affected by HIL and those that are can have differing thresholds for interference. We utilize a lipemia index of 3 as our threshold value for flagging samples with HIL index alerts. This translates to a lipid concentration of approximately 100-200 mg/dL.
False lipemia alerts have been reported with certain chemotherapy drugs, circulating immune complexes, and monoclonal proteins. In the case of monoclonal proteins, the sample itself is not turbid, but becomes turbid when mixed with water in the test cuvette during the HIL measurement process. This explains why some samples that trigger a lipemia index alert lack observable lipemia.
Alerts Without Visually Observable Lipemia
There is a tendency to accept as merely outliers alerts for samples that don’t have visible lipemia. However, our lab took the opportunity to investigate these cases further and found our version of “the curious incident of the dog in the night-time.”
Of the 80,000 comprehensive metabolic panels run annually at our center, false lipemia alerts occur 1 to 2 times each month. Over a 2-year period, we had 25 samples flagged as having a high lipemic index (≥3 out of 8) but with no evidence of lipemia upon visual inspection or no previously documented monoclonal gammopathy (J Appl Lab Med 2019;3:1062). In these cases, we notified the ordering physician and obtained permission to evaluate the samples further by serum protein electrophoresis. In so doing we found that 12/25 (48%) of samples harbored monoclonal immunoglobulins. When we characterized the immunoglobulin heavy chains using immunofixation electrophoresis we found IgM monoclonal protein in 11 cases and IgG in one.
Monoclonal gammopathy of unknown significance (MGUS) carries a risk of progressing to multiple myeloma, a plasma cell disorder that often presents with vague, nonspecific symptoms including aches and pain, headaches, fatigue, constipation, and fevers. The paucity of distinct clinical features can delay suspicion for and diagnosis of this and similar disorders. Early diagnosis helps clinicians set up a monitoring program for patients to see if their MGUS is changing and determine when treatment should begin.
We have found that using the lipemia index alert in the absence of visible lipemia is a value-added tool in identifying previously unsuspected, undiagnosed monoclonal gammopathies. We believe the lipemia index alert would work similarly on other analyzers, and we encourage our colleagues to dig a little deeper into any seemingly false alerts. Doing so might set up patients for better health outcomes.
Amy Guimaraes-Young, MD, PhD, is a first-year pathology resident at the University of Colorado Anschutz Medical Campus in Aurora, Colorado.+Email: [email protected]
Yashpal Agrawal, MD, PhD, is section chief of chemistry, toxicology, and point-of-care at the Rocky Mountain Regional VA Medical Center in Aurora, Colorado.+Email: [email protected]