Implementation and Assessment of an Electronic Health Record Alert Preventing Misleading Serology Test Results in Patients Recently Treated with Intravenous Immunoglobulin

To the Editor: Intravenous immunoglobulin (IVIG) is concentrated immunoglobulin from human donors. It is FDA-approved for a large number of conditions, including primary immunodeficiency disease, immune thrombocytopenic purpura, dermatomyositis, and chronic lymphocytic leukemia (1). The literature has shown that patients with exposure to IVIG can demonstrate false positivity in various serology tests (2–5). We have previously published findings supporting a false positivity rate of 1.4% in serum protein electrophoresis tests with recent IVIG usage (5). As a result, at The Johns Hopkins Hospital, we implemented an OurPractice Advisory (OPA) in the Epic electronic health record to alert clinicians when ordering selected serum tests where evidence has supported that results could be compromised by recent IVIG exposure (2–5). The alert triggers when a test is ordered in a patient who was treated with IVIG within 30 days (in concordance with the average half-life of IVIG preparations), and the alert provides background information and offers options for test acknowledgment/continuation or for cancellation. The tests included in the alert, based on a literature review, are direct antiglobulin tests, Treponema palladium antibodies, hepatitis B surface and core antibodies, and serum protein electrophoresis tests, although it is acknowledged that this list is not exhaustive. Approximately 1 year postimplementation, we analyzed the number of alert triggers and test cancellations (aka “remove orders”). Twenty-nine of 332 tests were canceled, giving a cancellation rate at the time of alert of 8.7%. We further investigated the 50 most recent alert triggers through a manual review, which showed a cancellation rate of 26% (13/50 tests) (Fig. 1). Investigation into the differences in these results revealed that if clinicians canceled the test order through alternative means during the alert process (such as terminating the entire prompt, navigating out of Epic, etc.) without hitting the “remove order” button on the alert prompt, the order would be removed and traceable during manual review; however, the electronic health record system would not identify this as a test cancellation. Further investigation of the 37/50 ordered tests found that 8 tests were subsequently canceled post-OPA but prior to laboratory analysis, leading to a cancellation rate of 21/50 (42%). IVIG alert analysis. Bar graph depicts tests canceled at the alert (orange) vs ordered (gray) after longitudinal review of approximately 1 year in duration (n = 332). An in-depth review of the last 50 alerts assessed tests canceled after the alert was triggered (post-alert). A breakdown of the ordered tests (29 total) revealed a false positivity rate of 10% (3/29). Of the 29 tests that were completed, 8 tests produced negative results, and 21 produced positive findings (Fig. 1). Due to the interfering nature of recent IVIG exposure, the validity of these positive results was investigated by reviewing the medical history, clinical context, patient workup, and other laboratory findings and conclusions from the managing clinical team. Any completed tests not meeting these criteria were labeled as clinically discordant, potentially due to IVIG-related serologic interference, with 3/29 (10%) completed tests identified. The specifics are as follows: Direct antiglobulin testing to clarify the etiology of hemolytic anemia. The direct antiglobulin test was positive for IgG and negative for C3, and IVIG-induced hemolysis was suspected given the clinical context/recent exposure and subsequent eluate showing anti-A antibodies in a patient with blood type A positive. A suspected false-positive hepatitis B core antibody test (IgG and IgM) in a patient who was negative for hepatitis B surface antigen and positive for surface antibody. The patient was previously vaccinated and had prior hepatitis B core IgM testing 15 days prior (and before IVIG), which was negative. A suspected false-positive hepatitis B core antibody test (IgG and IgM) in a patient who was previously vaccinated and had positive hepatitis B surface antibody and negative antigen. There was no follow-up with the primary team for interpretation of these findings. Together, these data support the findings that recent IVIG usage affects certain serum tests, leading to clinically discordant results and a finding that an electronic health record alert is an effective method to reduce the number of orders for which results could be compromised (a test cancellation rate of up to 41% in the most recent 50 tests). Additionally, our analysis also highlighted another benefit of the OPA, which raises awareness of the potential interference leading to a substantial proportion of late test cancellations (16%) after the initial alert. We suspect that after awareness of the issue and discussion by the clinical team, the risks of clinical decisions being made from potentially compromised serology results prompted test cancellation. This analysis supports the finding that an alert influences decisions at the time of the alert, affects downstream clinical decisions, and raises awareness within the clinical team. This underscores the importance of evaluating the impact of an alert both at the initial alert stage and afterward to accurately determine its efficacy. Nonstandard Abbreviations: IVIG, intravenous immunoglobulin; OPA, OurPractice Advisory. Author Contributions: The corresponding author takes full responsibility that all authors on this publication have met the following required criteria of eligibility for authorship: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Nobody who qualifies for authorship has been omitted from the list. Andrew Sulaiman (Conceptualization [equal], Data curation, Formal analysis, Investigation, Writing—original draft [lead]), Kristi Beers (Data curation, Writing—review & editing [supporting], Software [lead]), and Lori J. Sokoll (Conceptualization [equal], Supervision, Writing—review & editing [lead]). Authors’ Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.