Understanding serum tryptase variability helps clinicians correctly confirm anaphylaxis
Published: August 20, 2021
A substantial increase in blood levels of the mast cell protein tryptase above what is measured when a patient is asymptomatic (in ng/ml), is the most widely accepted laboratory test used to support the clinical diagnosis of anaphylaxis, a severe systemic allergic reaction. However, baseline serum tryptase (BST) levels in blood can already be elevated when a patient carries the diagnosis of hereditary alpha-tryptasemia (HaT) that results from increased copies of the tryptase gene TPSAB1and affects about 6% of the general population. BST levels in people with HT are typically at least three times higher than those with common allergic conditions. Less commonly, rare disorders such as mastocytosis - where there is an increase in mast cell number - may also be associated with elevated BST. The current approach for using tryptase to confirm anaphylaxis is to apply a formula termed the “20+2 rule” where the tryptase level during a symptomatic episode must equal or exceed a 20% increase over a BST measurement plus 2 ng/ml. While this 20+2 rule has been shown to be helpful in identifying patients with anaphylaxis (e.g., has good sensitivity), a concern was raised that it might not be sufficiently specific (e.g., lacks specificity) since it has not been tested among individuals when they are not having reactions. If indeed the 20+2 rule was insufficiently specific, it would incorrectly be positive in some individuals not experiencing anaphylaxis, in part because some patients have an elevated BST (e.g., patients with HT or mastocytosis) that might exaggerate a non-meaningful variation in baseline levels.
Based upon observations that BST levels are likely variable and may affect interpretation of what constitutes a significant rise in serum tryptase, Mateja et al. set out to define how variable this blood test was in two groups - individuals with common allergic problems, and individuals with elevated BST due to HT. BST levels were thus measured over multiple timepoints while patients were not experiencing symptoms that would suggest anaphylaxis. Based upon the variability found in these serial blood samples, Mateja and colleagues then created a model to identify the percent change in two sequential tryptase levels where the ability to both accurately identify patients with anaphylaxis (i.e., sensitivity) and correctly classify subjects as not having meaningful increases in tryptase when they are not experiencing anaphylaxis (i.e., specificity), were jointly optimized (i.e., their sum was effectively maximized). They went on to compare this new model to the 20+2 rule, testing sensitivity among a group of individuals known to have had anaphylaxis, as well as its ability to correctly identify baseline tryptase changes in three additional groups of patients with elevated BST resulting from HT, mastocytosis, or both. Their findings are reported in The Journal of Allergy and Clinical Immunology (JACI).
Variability in BST was found to be substantial - in some cases almost doubling from one serial measurement to the next. When this variability was modeled, it was found that the percent change in tryptase levels from one blood draw to the next was comparable among individuals with HT and those with common allergic conditions. However, among those with elevated BST, 24% (33/139) of serial measurements met the 20+2 rule at baseline. Grounded on their new modeling, the authors proposed a threshold-based approach, where an increase of 68.5% over BST (or 1.685 times BST) was found to be optimal, above which a rise in blood tryptase levels no longer would be consistent with baseline variability. Using this threshold, only 4.3% (6/139) of individuals with elevated BST would have been misidentified as having a clinically meaningful increase that would be suggestive of anaphylaxis. This improvement in specificity and reduction in the number of false positives did not impact the ability of the new approach to correctly identify anaphylaxis in this study. Because anaphylaxis is a clinical diagnosis, they also proposed two additional thresholds to be used at a care provider's discretion based upon the clinical likelihood of anaphylaxis. For a reaction thought unlikely to be anaphylaxis - such as a systemic reaction following passive inhalation of a specific odor - the authors proposed raising the bar higher (i.e., high specificity), requiring an increase of 86.8% over. Likewise, for reactions deemed likely to be anaphylaxis - such as a systemic reaction resulting during an oral food challenge in a food allergic individual - the authors proposed lowering the bar (i.e., high sensitivity), requiring an increase of 37.4% over BST. All three thresholds were incorporated into an online application that can be used to help confirm the clinical diagnosis of anaphylaxis: https://triptase-calculator.niaid.nih.gov/
This is the first study to systematically and clinically characterize baseline variability of serum tryptase levels. Using a threshold approach to determine clinically meaningful changes in serum tryptase levels, an increase of 68.5% over BST significantly improved correct identification of changes resulting from baseline variability, rather than a systemic allergic reaction, compared to the currently accepted 20+2 rule, without reducing the correct identification of anaphylaxis. The advantages of the new threshold over the 20+2 rule were observed among individuals with elevated BST. Individuals with elevated BST due to both HT and mastocytosis have been reported to have more frequent or severe anaphylaxis - but also may have symptoms that could be mistaken for anaphylaxis – so an improvement in specificity is of direct clinical benefit. Because the 68.5% threshold and the 20+2 rule are identical for the average BST level of healthy volunteers without HT or mastocytosis, both would be applicable and no different in this population. However, for individuals with very low BST levels (<4 ng/mL) an increase of 68.5% over BST would be less than the change required to meet the 20+2 rule, and thus potentially more likely to identify very small changes in serum tryptase as being clinically meaningful and consistent with anaphylaxis (i.e., more sensitive), particularly if clinicians use the higher sesntivity threshold of 37.4%. Future studies are needed to see how the two approaches compare in other contexts.
The Journal of Allergy and Clinical Immunology (JACI) is an official scientific journal of the AAAAI, and is the most-cited journal in the field of allergy and clinical immunology.
Full Article