Citations

Abstract Background The use of autonomous mobile robots (AMRs) to move materials/supplies is long-standing in manufacturing/warehouse settings and is expanding into hospitals and clinical laboratories. The recently launched Proxie robot (Collaborative Robotics i.e. Cobot) is an AMR that can autonomously move existing carts between defined locations. However, AMRs pose new challenges compared to traditional lab automation. The aims of this study were to: i) perform functional validation to verify Proxie*s manufacturer performance specifications within the laboratory, ii) assess the risk of using Proxie to deliver laboratory specimens, and iii) pilot the use of the Proxie robot to autonomously deliver loaded carts from the Pathology case intake area to the sorting station at Mayo Clinic Laboratories. Methods Functional validation of Proxie consisted of verifying manufacturer’s performance specifications for the following robot parameters (5 test runs): maximum speed, emergency-stop, and ability to maintain localization, travel autonomously, stop and/or avoid moving/fixed objects, and park/unpark. Risk assessment for the robotic use case was performed by a multi-disciplinary team utilizing an AMR risk assessment tool developed by the institutional robotics group. Hazards were categorized as mechanical, electrical, noise, ergonomic, environmental, material/substance or combination risks. Injury severity, exposure(frequency), and avoidance scores were assigned to each hazard to determine negligible-, low-, medium-, high- or very high-risk levels. Proxie’s workstream was evaluated weekly over four weeks by measuring number of cart deliveries, total time/miles traveled, % autonomous deliveries(goal >80%), % on-time deliveries, average±standard deviation(SD) time (min:sec±min:sec) for delivery/return trips, and dwell times at intake/sorting stations. Results Manufacturer’s performance specifications were verified in 100% of test runs. Proxie traveled at human walking speed, stopped within 1 second of emergency stop being enabled, maintained localization, and traveled autonomously for =5 consecutive workflow cycles. Proxie navigated through spaces =4’ wide along fixed pathways and stopped safely when encountering obstacles. Proxie navigated around people and carts in the main 6’ hallway if they were blocking <2’ hallway width. For objects blocking >2’, Proxie stopped and resumed motion after the object moved. After task completion, Proxie side-shifted into the AMR parking space stopping no closer than 7.8” from any object. Twenty-four potential hazards were identified through the risk assessment tool with all risk levels scored as negligible or low. Proxie completed 304 cart deliveries/returns operating 17:35 hours and traveling 11.5 miles during the pilot. The % autonomous deliveries for week 1-4 was 82%,88%,91% and 88%, respectively with =99% on-time deliveries. Delivery from pathology case intake to sorting station averaged 5:43±0:25 min and return trips averaged 3:32±0:34. Proxie dwell times were 13:10±0:33 and 37:34±0:23 min at the pathology intake and sorting stations, respectively. Conclusions The Proxie AMR was capable of successfully and autonomously moving carts with laboratory specimens between the pathology intake and sorting stations of the laboratory during the pilot study. Adding Proxie to this workstream translated to eliminating 5,800 steps and 4 hours 24 min of human cart-pushing per week. Incorporating Proxie into laboratory operations as a cart-moving AMR would increase efficiency and allow laboratory staff to focus on technical, non-repetitive work.

Abstract Background Hemolysis, a common pre-analytical interferent in clinical laboratory testing, can significantly affect accuracy and reliability of laboratory results. One test particularly susceptible to interference is direct bilirubin (D-bil). For example, during December 2024, 40% (500/1239) of specimens rejected due to hemolysis were orders for D-bil. The recently released Roche (Roche Diagnostics) direct bilirubin reagent (BILD2) has a lower threshold for hemolysis (H-index [HI]=25)) than the previous generation (HI=30). The aims of our study were to (1) establish concentration-specific hemolysis thresholds for D-bil and (2) investigate the instrument’s reduced volume function (dilution) as a possible mitigation strategy for samples with hemolysis indices exceeding the established thresholds. Methods Residual serum specimens from physician-ordered D-bil testing were used and analyzed on Roche cobas c701 and c501 analyzers. The instrument’s decreased volume function, intended for results outside the measuring range, was used as a 1:2 dilution and validated using 9 non-hemolyzed negative control specimens with D-bil results from 9.8-13.4 mg/dL. Recovery within 10% neat value was considered acceptable. EDTA whole blood was used to prepare hemolysate pools by washing red cells and lysing by freezing (-80 °C). Twenty-six serum pools with varying D-bil concentrations (0.2-3.4 mg/dL) were prepared. Each pool was split into 10 aliquots and spiked with increasing amounts of hemolysate, not exceeding 10% serum to interferent ratio. The decreased volume function (dilution) was utilized on a subset of the hemolysate-spiked aliquots (N=55) with D-bil results from 0.2-2.5 mg/dL. HI values were estimated to be 50% of neat HI due to the instrument’s inability to measure indices on specimens tested with the decreased volume function. Acceptance criteria for experiments utilizing hemolyzed specimens were ±0.1 mg/dL or 20% of neat sample. Results All non-hemolyzed specimens tested with the decreased volume function passed within acceptance criteria with average recovery of 109% (range: 106-110%) from undiluted specimens. Serum pools with D-bil <0.3 mg/dL (N=7) spiked with hemolysate (HI range: 7-191) had an average difference of -0.1 mg/dL (range: -0.2-0.0 mg/dL) from non-hemolyzed pools, D-bil pools 0.4-0.9 mg/dL (N=10; HI range: 6-173) had an average difference of -0.2 mg/dL (range: -0.5-0.0 mg/dL), and D-bil pools >1.0 mg/dL (N=9; HI range: 5-182) had an average difference of -20.7% (range –45.5-0.0%). Serum pools with D-bil <0.3 mg/dL (N=23; HI range: 43-191) tested using the decreased volume function had an average difference –0.1 mg/dL (range: -0.2-0.0 mg/dL; HI range: 22-96) from the non-hemolyzed specimens, D-bil pools 0.4-0.9 mg/dL (N=20; HI range: 39-173) had an average difference -0.1 mg/dL (-0.3-0.0 mg/dL; HI range: 19-86), and D-bil pools >1.0 mg/dL (N=12; HI range: 37-182 ) had an average difference of -0.2 mg/dL (-0.5-0.2 mg/dL; HI range: 19-91). Conclusion We established a new hemolysis interference threshold of 75 for samples with D-bil <0.3 mg/dL. Using the decreased volume function, we expanded the hemolysis interference threshold to HI=100 for specimens =0.3 mg/dL. All specimens with HI>100 will result in a recollection. With these changes, we estimate that we can reduce our D-bil recollections by up to 6% over six months.

Abstract Background Lymphocyte function and immune competence can be measured by proliferative response to stimuli and is useful in several clinical contexts including primary (or genetic) and secondary immunodeficiencies (e.g., immunosuppression following transplantation). Proliferation assays have traditionally been prepared manually due to the difficulty in maintaining the sterile environment required to prevent bacterial and fungal impact on cell growth. On most automated liquid handlers, the sample and reagents are placed inside a cabinet whose inner surfaces include rails, pullies, pumps, and circuitry that face the sample and are not amenable to sterilization. The Andrew+ (Waters, MA) is a small surface cleanable liquid handler with all machinery enclosed. We demonstrate here the use of an Andrew+ placed in a biological safety cabinet (BSC) to perform sterile assay setup. Methods Peripheral blood monocytes (PBMCs) from five individuals were manually generated via buffy coat isolation, then brought to standardized concentration. An Andrew+ was placed in a 4’ BSC and all surfaces disinfected with Oxivir Tb (Diversey, SC). The Andrew+ prepared three serial dilutions of two stimulant reagents (phytohemagglutinin (PHA) and pokeweed (PWM)), then dispensed stimulant and control reagents horizontally and PBMCs vertically to a 48 well plate. This process was duplicated manually from the same specimens to create a reference plate. The plates were incubated at 37°C and 5% CO2 for three days, then pulsed with EdU (Thermo Fisher, MA) and incubated for one additional day. Cells were harvested and counted via flow cytometry. Overall time and hands-on time were measured. Results A total of five samples were tested inside a using both manual and automation setup under eight conditions including with isotype control, media without stimulant, three concentrations of PHA, and three concentrations of PWM. The percent proliferation for various cell types as induced by each stimulant was calculated from the difference of the highest stimulant signal from the background (media-only) signal. Comparing the automated and manual method results showed a slope of 1.0217 and R2 of 0.9966. For a simulated run of 6 samples, technologist time for automated steps was under 8 minutes with a total time of 56 minutes. When these steps were performed manually, the time was 30 minutes. Automation increased total time by 26 minutes but saved 22 minutes of technologist hands-on time. Conclusion The Andrew+ was able to perform sterile setup of a cell culture assay. While slower overall than manual processing, technologist time was reduced. Ergonomic, inter-tech variability, and sample transcription error issues were subjectively observed to be less likely. The Andrew+ can work with up to four “dominos” (reagent or tip blocks) in width and up to three dominos in depth. Non-custom, standard depth BSCs from a variety of manufacturers (Labconco, Nuaire, ThermoFisher) have room for one domino in depth so, except for timings performed in this evaluation, all steps were broken down to be performed with a max of four dominos. Efficient sterile use of the system would require an extended depth BSC.

Abstract Background Community Paramedic (CP) programs use specially trained paramedics to provide healthcare in the community to increase access to care, reduce unnecessary emergency department visits, and improve timeliness of care. Mayo Clinic in Rochester, MN recently launched a CP Program to provide in-home care for patients who require regular intravenous medications, wound care, diuresis for heart failure, medication reconciliation, care plan compliance/education. CP utilizes i-STAT1 CHEM8+ and Lactate (CG4+) testing (Abbott Point of Care, Abbott Park, IL) to make therapy decisions primarily for heart failure patients based on rapid laboratory results. Having results available at the point of care (POC) eliminates delays between testing and treatments, expediting care for the patient. Prior to using i-STAT, CP traveled to the patient’s residence, drew blood tubes, delivered the tubes to Mayo Clinic, waited for test results, returned to the patient’s home, and provided treatment. In this observational study, we assessed the benefit of in-home POC i-STAT1 testing performed by our CP Program to improve patient care and increase efficiency. Methods Patients within a 40-mile radius of Mayo Clinic were eligible to receive CP service. Whole blood venous specimens were collected and tested using the i-STAT1 CHEM8+ metabolic panel cartridge according to manufacturer instructions to measure: ionized calcium, chloride, glucose, hematocrit, total carbon dioxide, blood urea nitrogen, creatinine, potassium, sodium, anion gap while the i-STAT1 CG4+ cartridge measures lactate. The laboratory information system was used to extract laboratory results, and Excel (Microsoft, Redmond, WA) was used to filter data and perform calculations. Results November 2022 through December 2024, CP performed testing with 353 i-STAT cartridges (3862 tests) on 102 unique patients who each had 1-12 separate orders. Patient ages were 25 to 97 years. 41.2% (1590/3862) of the individual results obtained were outside of our reference intervals (23.0% high, 18.0% low), with 10.8% being ionized calcium, 10.2% chloride, 10.3% glucose, 13.0% hematocrit, 7.6% total carbon dioxide, 18.2% blood urea nitrogen, 16.4% creatinine, 6.7% potassium and 6.9% sodium. Additionally, 0.13% of results were critically high (ionized calcium, potassium, and sodium) and 0.10% were critically low (potassium and ionized calcium). Hypokalemic patients were sent to the infusion therapy center for IV potassium. Patients experiencing hyperkalemia were retested and sent to the emergency department (ED). Hyponatremic and hypercalcemic patients were sent to the ED for further treatment. Abnormal (non-critical) results were reviewed by CP personnel to make treatment recommendations or medication adjustments. Conclusion i-STAT1 home testing performed by CP facilitated rapid therapeutic interventions and expedited decision-making. POC testing has increased the group’s efficiency—alleviating the need to transport specimens to the central laboratory and wait on test results before returning to the patient for treatment. Paramedicine emphasizes a proactive and preventative approach to medicine. We anticipate that this model of care and testing will expand to cover other chronic health conditions due to the program’s early success.

Abstract Background Diagnosis of VZV CNS vasculitis (CNS-VZV) is aided by detecting intrathecally (i.t.) synthesized VZV-specific antibodies (Abs) through testing paired serum (S) and CSF specimens and determining an antibody index (AI). Currently, laboratories in the United States do not provide AI testing for VZV. The currently used methods (e.g., EIAs, IFAs) lack clinical utility as they cannot discriminate if VZV Abs in the CSF are i.t. produced due to CNS infection or whether they are present due to passive diffusion across the blood-CSF-barrier. To optimize diagnostic options for CNS-VZV, we evaluated a quantitative VZV IgG ELISA (Euroimmun AG, Lübeck, Germany) to calculate AI values using Reiber’s formula on paired S/CSF specimens to assess the diagnostic accuracy of this method. Methods The clinical sensitivity cohort consisted of four patients diagnosed with CNS-VZV (meningitis or vasculitis) as determined by symptoms, imaging, response to antiviral therapy, and/or supporting test results (e.g., lymphocytic pleocytosis, elevated protein in CSF). The patient group for clinical specificity included 11 patients with suspected CNS-VZV who ultimately had an unclear (n=4) or non-CNS-VZV (n=7) diagnosis. The reference range cohort included paired clinical S/CSF samples from 25 patients without infectious diseases testing. Paired S/CSF samples were processed on the same batch using a fully automated ELISA analyzer capable of onboard, ad hoc runtime dilution preparation (AP22 ELITE, DAS ITALY). The AI values were calculated using the Reiber formula, and sensitivity (Sn), specificity (Sp), and positive (PPV) and negative (NPV) predictive values were calculated. The AI was calculated using the following equations: If Total IgG CSF (mg/dL):Total IgG serum (mg/dL) < Quotient limit Total IgG, then AI = (VZV IgG Ab CSF [units × dilution]:VZV IgG serum [units × dilution])/(Total IgG CSF [mg/dL]:Total IgG serum [mg/dL]). If Total IgG CSF (mg/dL):Total IgG serum (mg/dL) > Quotient limit Total, then AI = (VZV IgG Ab CSF [units × dilution]:VZV IgG serum [units × dilution])/Quotient limit Total IgG. Quotient limit Total IgG = 0.93 × [(Qalbumin2) + (6 × 10-6)]1/2 - 1.7 ×10-3. Qalbumin = Albumin CSF (mg/dL):Albumin serum (mg/dL). Results The VZV IgG ELISA with Reiber AI calculation method showed Sn, Sp, PPV, and NPV values of 100% (4/4), 91% (10/11), 80% (4/5), and ?0 (10/10), respectively. One of the Sp patients had an equivocal AI result and was classified as negative, whereas the Sp patient with a positive AI result had an unclear diagnosis. VZV AI results for 96% (24/25) of the reference range patients were negative, with one equivocal. The mean AI of the 25 patients was 1.05, and the lower and upper 3 standard deviation limits were 0.6 and 1.5, respectively. These values match the AI qualitative interpretive criteria recommended by Reiber (0.6 to <1.3 = negative, 1.3 to 1.5 = equivocal, >1.5 AI = positive). Conclusion We show that the VZV IgG ELISA, alongside the Reiber formula for AI calculation, is effective for identifying patients with i.t. synthesized VZV IgG Abs. This method can assist clinicians in evaluating patients with suspected CNS-VZV infections who are negative by alternative methods.

Abstract Background Insulin resistance leads to changes in lipid and lipoprotein metabolism prior to developing prediabetes, including an increase in the concentration of large very-low-density lipoprotein particles (LVLDL-p), the mean size of VLDL (VLDL-s), and the concentration of small low-density lipoprotein particles (sLDL-p). Conversely, there is a decrease in the concentration of large high-density lipoprotein particles (LHDL-p), mean size of HDL (HDL-s), and mean size of LDL (LDL-s). Clinically validated studies show that proton nuclear magnetic resonance (NMR) can measure these six parameters and derive a lipoprotein insulin resistance (LP-IR) score, ranging from 0 to 100, with higher scores indicating greater risk for insulin resistance. The reference method for calculating LP-IR relies on a proprietary NMR spectrometer, spectral deconvolution method, and algorithm. This study seeks to determine if lipoprotein subfraction parameters measured from a commercially available NMR system can produce LP-IR scores comparable to the reference method. Methods Serum lipoprotein NMR spectra were collected using a Bruker Ascend 600 MHz magnet, Bruker Avance NEO 500 spectrometer, and the Numares AXINON lipoFIT System to report the six LP-IR score parameters between January 2023 and January 2025 (Cohort A, n = 2,474). During the same period, LP-IR score and parameters were collected using the Vantera Clinical Analyzer (400 MHz) with the LipoProfile LP4 method (Cohort B, n = 2,959). Linear regression of decile-matched data for each parameter yielded adjustment equations to align lipoFIT values with the LipoProfile method. LP-IR scores were then calculated for both adjusted and unadjusted values using a previously published LP-IR scoring algorithm. Finally, LP-IR scores were compared for sera with measures from both instruments was (Cohort C, n = 108). Results The median concentrations of sLDL-p, LDL-s, VLDL-s, HDL-S matched between Cohort A and B (<1% difference). Significant differences were noted for LVLDL-p and LHDL-p, with percent differences of 21.5% and 23.5%, respectively. Despite closely matching medians, the distributions varied greatly, with a mean interquartile range (IQR) percent difference across measures of 31.2% (Table 1). After distribution matching, all parameter median and mean IQR percent differences fell to <5% and <10%, respectively, between Cohort A and Cohort B. In Cohort C the LP-IR score matched well (y=1.1x-15, r=0.92; 95%CI 0.88, 0.94). Following distribution matching adjustment of lipoFIT values for Cohort C, new calculated LP-IR scores matched with reduced bias (y=0.94x-1.2) while correlation was slightly reduced (r=0.88 [0.83, 0.92]). By comparison, adjustments made via scaling lipoFIT values proportionally by the ratio of the medians between Cohort A and Cohort B yielded calculated LP-IR scores in Cohort C that maintained a similar bias prior to adjustments (y=1.2x-21), highlighting the ability of distribution matching to reduce bias. Conclusion Adjustments to measured lipoprotein parameters from a commercially available NMR spectrometer can be achieved via distribution matching. These adjustments greatly reduce the bias between a commercially available method and algorithm compared to the reference method, allowing for wider adoption of LP-IR score reporting.

Abstract Background The Oncology Department at our institution requested faster turn-around times (TAT) for chemistry testing. Prior to their request, serum separator tubes (SST) were collected and tested in our core laboratory. To accommodate the request, we added plasma total bilirubin and aspartate aminotransferase (AST) to our stat laboratory test menu with collection in plasma separator tubes (PST). Upon implementation of this testing change, we observed an increase in specimen recollection rates due to hemolysis. We systematically evaluated laboratory testing variables that could have contributed to the increase: pneumatic tube transport path, differences in hemolysis measurements between chemistry analyzer models, and specimen types. Previous studies have demonstrated that lithium heparin plasma specimens are more susceptible to hemolysis than serum. Likewise, patients undergoing chemotherapy or radiation therapy may have fragile red blood cell (RBC) membranes, further complicating the issue. Our study*s objective was to assess the impact of changing from SST to PST collections. We conducted a 3-month pilot study to determine if we could reduce our recollection rates by collecting/testing Rapid Serum Tube (RST), while maintaining plasma TAT. Methods We compared recollection rates for specimens collected from the oncology patients and tested in our core laboratory (SST, May-June 2021), stat laboratory (PST, September-November 2022), pilot period in stat lab (RST, January-March 2023), and post-pilot period in stat lab (PST, May-July 2023). Data were extracted from our laboratory information system and recollection rates were calculated using Tableau (Salesforce, Inc, San Francisco, CA). Analysis was performed in Excel (Microsoft, Redmond, WA) using Student’s t-test, with p < 0.05 defined as statistically significant. Results SST specimens collected from oncology patients and tested in the core lab (May-June 2021) resulted in a recollection rate of 0.17% (4/2341). PST specimens sent to the stat lab (Sept-Nov 2022) had an increased recollection rate of 0.90% (55/6129, p-value < 0.001). During the pilot period when RST tubes were used (Jan-Mar 2023), our recollection rate decreased to 0.19% (11/5801, p-value 0.014). After the pilot period ended and PST tubes were again utilized for testing (May-July 2023), the recollection rate again increased to 0.34% (22/6500, p-value 0.019). Conclusion Rapid TAT of laboratory results within specialized-care settings, such as a cancer treatment clinic, is necessary to provide patients with the most effective and personalized treatment regimens. Hemolysis presents a challenge as it generates an analytical issue with the release of contents from RBCs that may spuriously affect the concentration of the analyte being measured, delaying result reporting due to the need for specimen recollection. The data from our study highlight the utility of rapid serum tubes. Not only did they achieve stat TAT that was observed with the use of PST, but they also decreased specimen recollection rates that mirror the low rates observed with the use of SST. Thus, this technology may be an excellent replacement for standard SST when specimen integrity is critical to rapidly reporting accurate test results.

Abstract Background Cryoglobulins are immunoglobulins that precipitate at cold temperatures. They are associated with plasma cell disorders, infectious and autoimmune diseases. Laboratory testing is manual and requires strict temperature control, from keeping the blood sample warm from the moment of draw until processing in the laboratory. Serum samples are transferred to Kolmer glass graduated tubes upon arrival, and kept on ice for visualization of a precipitate for up to 7 days. Because of the temperature variation samples are subjected to, positive sample identification is important. The Kolmer glass tubes are not disposable, and undergo several steps of deidentification and cleaning for reuse. The goal of this project was to improve patient safety and label adherence in cryoglobulin testing by using the Managing for Daily Improvement (MDI) method to monitor and implement changes effectively. Methods During a Plan Do Study Act (PDSA) initiative, it was discovered that labels were not being secured on both ends with tape due to safety concerns when using a box cutter to remove the label prior to cleaning. This practice led to an increase in labels detaching from the glass tubes. To address this issue, we employed the MDI method and established a Key Performance Indicator (KPI) to monitor and evaluate potential changes aimed at improving label adherence. The MDI method facilitates problem-solving through daily visual indicators and discussions by empowering technologists to identify and address issues affecting daily operations. This approach encourages open communication and collaborative problem-solving, allowing the entire team to share insights and solutions in real time. Results A PDSA initiative via MDI uncovered the label adherence challenges. A KPI was developed to monitor interventions to improve the process. First, removable tape was used to secure both sides of the label. Initial results were worse than the original method but slight improvement occurred at about 3 weeks. A Just Do It (JDI) approach was initiated to leave additional space between samples. No improvements were observed and the practice discontinued within < 2 weeks. A second JDI was trialed with a different tool to remove labels post-use. The entire sample identification process was reviewed and the original identification process resumed, by adding scotch tape securing both sides of the labels and using a ceramic box cutter to remove the labels. Significant improvements to labels adherence was observed, from about 60% of labels coming off to zero, while the new tool was also safer for the team. Another PDSA was initiated to investigate if plastic, disposable tubes were a viable option for testing, and a full method validation of the new plastic tubes followed, finally eliminating the challenges with the larger process or reusing glass tubes. Conclusion MDI helped recognize a problem from the front-line and find real time solutions that benefited end-to-end operations. Visual daily indicators clearly showed effectiveness of interventions. Identifying real issues, even if they appear small, can lead to improved patient and staff safety, and empowerment to initiate meaningful change.

Abstract Background Platelet function tests (PFT) help assess bleeding risk and guide appropriate hemostatic measures in patients on antiplatelet therapy undergoing emergent surgical interventions. Light transmission aggregometry (LTA, BioData Corporation), the gold standard for assessing platelet function, cannot be performed rapidly enough for emergent interventions. In this study, we compared two PFT, Plateletworks (Helena Laboratories) and TEG 6S (TEG, Haemonetics) adenosine diphosphate-(ADP) and arachidonic acid (AA)-induced platelet function, evaluating analytical imprecision and their ability to differentiate healthy controls from patients with acute coronary syndrome (ACS) on dual antiplatelet therapy. Methods Fresh specimens were collected from healthy donors (HD, N=19) and a patient group (PG, N=17) on both aspirin and clopidogrel taken within 24 hours of procedure. Percent aggregation was measured according to manufacturers’ instructions using Plateletworks and TEG, utilizing both AA and ADP and compared to LTA performed on a PAP-8E Platelet Aggregometer. Two separate venipuncture collections from HD were used to obtain repeat measures of Plateletworks and TEG to assess variability in platelet function due to pre-analytical variables. Using a two-way nested ANOVA model on data from HD, we estimated variability between replicates of a single sample (intra-assay imprecision) and between blood draws within a single patient (inter-assay imprecision), reported as coefficient of variation (CV%). Platelet function results were summarized as median (interquartile range [IQR; 25th-75th percentiles]) and compared between groups using nonparametric Wilcoxon rank-sum test. Receiver operating characteristic (ROC) analysis (using duplicate means from first blood draw) was performed. Results Plateletworks AA % aggregation, medians (IQR) were 96.3 (95.3-97.1) for HD and 48.9 (46.8-54.8) for PG (p<0.001); ADP % aggregation were 97.0 (94.5-98.2) and 37.1 (28.0-53.0) for HD and PG, respectively (p<0.001). TEG AA % aggregation was 97.8 (94.4-99.8) for HD and 13.5 (7.9-26.8) for PG (p<0.001); ADP % aggregation was 98.3 (95.3-99.9) for HD and 83.3 (77.8-87.7) for TG (p<0.001). LTA AA % aggregation was 55.0 (44.0-64.8) for HD and 0.0 (0.0-1.5) for PG (p<0.001); ADP % aggregation of 59.0 (40.8-70.8) and 14.5 (4.5-19.5) for HD and PG, respectively (p<0.001). Intra- and inter-assay imprecision estimates (CV%), respectively, in HD were 1.4 and 5.1 for Plateletworks AA, 5.5 and 15.9 for Plateletworks ADP, 3.0 and 0.0 for TEG AA, and 2.6 and 0.8 for TEG ADP. Intra-assay precision estimates for LTA AA and LTA ADP were 19.3 and 8.2. In PG, intra-assay precision was estimated as 7.0 for Plateletworks AA, 9.6 for Plateletworks ADP, 61.3 for LTA AA, 11.9 for LTA ADP, 37.6 for TEG AA, and 4.1 for TEG ADP. In our limited data set, ROC analysis showed optimal cutpoints (sensitivity/specificity) for Plateletworks AA 65.5 (1.0/1.0) and ADP 84.5 (1.0/0.90), TEG AA 87.1 (1.0/1.0) and ADP 92.7 (1.0/1.0), and LTA AA 8.75 (0.94/1.0) and ADP 25.0 (0.82/0.90) Conclusion Plateletworks most effectively distinguished % aggregation between healthy donors and patients on dual antiplatelet therapy. TEG also differentiated the groups, but differences in median ADP-induced aggregation were smaller than with Plateletworks. Both AA and ADP assays on LTA showed more variability than either Plateletworks or TEG.

Abstract Aims Blood biomarkers (BBMs) are promising minimally invasive tools for the evaluation of Alzheimer’s disease (AD). However, recent studies have demonstrated that comorbidities, such as severe chronic kidney disease (CKD), may elevate circulating concentrations of various BBMs including amyloid-ß (Aß) and phosphorylated tau proteins (p-tau). Previous work has found that the effects of CKD on BBM plasma concentrations could be partially mitigated through use of biomarker ratios (e.g., the proportion of p-tau217 to non-phosphorylated tau217 (np-tau217)). In this work, we quantitatively evaluate how different combinations of BBM ratios and single biomarkers are affected in the context of decreased kidney function. Methods Participants (n=83, >57 years of age) with CKD stages 1-2 (eGFR 60-98, n = 50), and stages 3-4 (eGFR 25-59, n = 33) and without amyloid pathology (A-) were included. eGFR (mL/min/1.73 m2) was calculated using the 2021 CKD-EPI equation based on serum-paired creatinine measurements. Plasma p-tau217, np-tau217, p-tau181, np-tau181, Aß42, and Aß40 were measured by mass spectrometry (MS) (C2N Diagnostics). %p-tau217 [(p-tau217/np-tau217) x 100], %p-tau181 [(p-tau181/np-tau181) x 100], Aß42/Aß40 ratio, and p-tau217/Aß42 ratio were calculated. Log-Log (power function) best-fit regression analysis was used to quantitatively predict the relationship of BBM changes as a function of eGFR. Results Median concentrations of p-tau217, np-tau217, p-tau181, np-tau181, Aß42, and Aß40 were all significantly elevated in individuals with eGFR <60 mL/min/1.73 m2 compared to those with eGFR > or =60 (all p < 0.05). There were no differences in the %p-tau217 and Aß42/Aß40 median values between these two groups (eGFR <60 versus > or =60) (p > 0.05); whereas the proportions %p-tau181 and p-tau217/Aß42 median values were higher in the eGFR > or =60 group versus the <60 subset. Observed best-fit analyte regression equations of eGFR versus BBM concentrations predict that an eGFR decrease from 60 to 45 results in p-tau217 and p-tau181 increases of 52% (95% confidence interval (CI): 30–78%) and 40% (95% CI: 29–53%), respectively. A 16% (95% CI: 12–21%), and a 19% (95% CI: 16–23%) increase was predicted in response to the same eGFR decline for Aß42 and Aß40. Furthermore, increases of 19% (95% CI: 2–38%) for %p-tau217, 6% (95% CI: -0.02–13%) for %p-tau181 and 31% (95% CI: 12–53%) for p-tau217/Aß42 were predicted as eGFR declined from 60 to 45, while a minor decrease of 2.3% (95% CI: -5.2–0.6%) was predicted for Aß42/Aß40. Conclusions: Reduction in eGFR is associated with significant elevations in plasma concentrations of p-tau217, p-tau181, Aß42, and Aß40, which may affect the interpretations of these biomarkers. However, the use of %p-tau217, %p-tau181, or p-tau217/Aß42 could partially attenuate this effect. Although the Aß42/Aß40 ratio is not considered an effective diagnostic BBM for AD, this ratio was relatively unaffected by CKD. For example, in comparison to p-tau217 alone, using %p-tau217 or p-tau217/Aß42, the relative increase in analyte values associated with decreased eGFR was reduced by 63% and 40%, respectively. These findings provide a better understanding of the effect of reduced eGFR on AD BBM interpretation in the presence of CKD.