2025 Southeastern Residency Conference

Implementation and Analysis of a Pharmacist-Managed Heparin Infusion Service at an Academic Medical Center 
Abigail Mason, Brittany White, Emily Goodwin, Ashley Williams, Jesse Briscoe, Kyle Knapp

Background
Unfractionated heparin is widely used as a first-line anticoagulant for hospitalized patients due to its rapid onset of effect and short duration. Given the inherent risk of bleeding with heparin administration, frequent lab monitoring is required to maintain target lab levels and to minimize risk of adverse effects. An internal analysis of a historic nurse-led heparin infusion protocol revealed frequent protocol non-compliance and deviations resulting in a facility change to a pharmacist-led protocol in December 2023. This project compares historic lab outcomes and safety events between nurse-driven and pharmacy-driven heparin protocols.

Methods
This IRB-approved, retrospective cohort analysis compared laboratory and safety outcomes between historic nurse-led protocol and pharmacist-led protocol. Patients meeting the following criteria were identified for inclusion in the pharmacist-led cohort: age greater than 18 years, admission to Erlanger Baroness Hospital, and receipt of the standard, reduced-dose, or low-dose Heparin Infusion Protocols between March 1, 2024 and March 30, 2025. Patients were excluded if they received heparin therapy for less than 24 hours or if the baseline activated partial thromboplastin time (aPTT) exceeded 40 seconds. The primary outcome of this study was difference between cohorts in mean time, in hours, to first aPTT result at or above the protocol-specified target range. Results were stratified by heparin infusion protocol. Secondary outcomes included pharmacist-led heparin protocol adherence, mean number of aPTT checks in therapeutic range, and documented bleeding events during the hospitalization. Adherence to heparin protocols in the pharmacist-led group was assessed in three domains. This included selection of correct initial heparin bolus dose, correct initial infusion rate, and correct subsequent rate adjustments as specified by the ordered heparin protocol. Additionally, the mean time from aPTT result to pharmacist order change and time from pharmacist order change to nurse medication administration was analyzed.

Results: The implementation of a pharmacist-managed heparin service at Erlanger resulted in a statistically significant reduction in the median time to achieve the primary outcome of first therapeutic aPTT or higher, decreasing from 10.1 hours in the nurse-managed group to 6.3 hours (p<0.0001). This significant improvement was observed across standard-dose (10.4 to 6.3 hours, p<0.0001), reduced-dose (10.0 to 6.4 hours, p=0.0207), and low-dose protocols (9.0 to 6.2 hours, p=0.0099). Regarding secondary outcomes, the pharmacist-managed group demonstrated a statistically significant decrease in the percentage of sub-therapeutic aPTT checks per patient (40% to 31%, p=0.0003) and a significant increase in therapeutic checks (p=0.0180) compared to the nurse-managed group. Notably, there were no significant differences between groups in the rates of supratherapeutic aPTT checks or aPTTs exceeding 150 seconds. Pharmacist adherence to the protocol was high at 95.4%, with a median time of 4.5 minutes from lab result to order entry and 18 minutes to rate change documentation.

Conclusions: In conclusion, the pharmacist-managed heparin service at Erlanger effectively improved the time to therapeutic anticoagulation and improved the proportion of therapeutic aPTTs without increasing the risk of excessive anticoagulation.

Title: Does corrected calcium adequately reflect calcium levels in critically ill patients?


Authors: Allison Krueger, Caitlin Thomas


Objective: Review methods to assess patients’ calcium status and whether corrected calcium accurately represents the status of critical care patients.


Self Assessment Question: Which of the following is TRUE regarding the original Payne corrected calcium equation?


Background: Abnormalities in calcium status are common among critically ill patients, and disturbances in calcium status have been linked with increased mortality and morbidity. Accurate representation of calcium status is key in managing patients in the intensive care unit (ICU). There are two main ways of measuring calcium in a blood sample: total serum calcium (totCa) and ionized calcium (measurement of unbound calcium). Since only approximately half of the serum calcium is biologically active under normal conditions, equations were developed to estimate that value using serum calcium levels before labs were capable of directly testing ionized calcium. The most well know adjustment is the modified Payne equation that “corrects” totCa from reduced albumin levels. A formula by Pftizenmeyer et al. in 2007 was designed to “correct” totCa for very elderly patients in a facility that does not utilize iCa. The purpose of this study is to evaluate the accuracy of albumin corrected calcium and total serum calcium compared to ionized calcium at discerning calcium homeostasis in patients requiring critical care.


Methods: This is a single-center, retrospective cohort study that was deemed exempt from Institutional Review Board approval. It was conducted in a large, tertiary level, community teaching hospital with patients across seven adult ICUs. The electronic medical record was reviewed for inclusion in the study. Patients were included if the following labs were collected with 10 minutes of each other: ionized calcium, serum calcium, and serum albumin. Patients were excluded if they received albumin within the 24 hours prior to the lab collection and/or received intravenous calcium within the 12 hours prior to lab collection. The primary outcome is to assess total calcium and modified Payne corrected calcium for noninferiority to ionized calcium.  


Results: A total of 25 patients were included in this study. The mean levels for iCa, totCa, and corCa were 1.12 mmol/L, 8.4 mg/dL, and 9.5 mg/dL respectively. There was a statistical difference between the three mean levels (F = 11.35, p < 0.001). There was a statistical difference between the three methods at categorizing calcium status (Q = 7.98, p = 0.0185). Total calcium was found to be the outlier. Bland-Altman analysis of totCa shows a mean difference of 0.13 mmol/L (95% CI -0.06 – 0.32). Bland-Altman analysis of corCa shows a mean difference of 0.01 mmol/L (95% CI -0.15 – 0.17).


Conclusion: Total calcium was inferior to iCa when it comes to assessing patient calcium status. This study found that corCa compared to iCa was noninferior at predicting normocalcemia. Furthermore, the Bland-Altman analysis shows that, when comparing methods for getting an accurate value, both total calcium and corrected calcium have too wide of variations to accurately rely on them.

Title: Impact of Required Stop Times for Continuous Intravenous Fluid on Duration of Fluid Therapy

 Authors: 
Lauren Hudson
Jessica Briscoe
Christopher Wilson
 
Background: 
While continuous intravenous (IV) fluids are widely administered, inappropriate use is associated with significant adverse effects, including increased risk of fluid overload, electrolyte disturbances, and mortality. Despite these complications, recommendations regarding continuous IV fluids are lacking. At the study institution, a required IV fluid duration was implemented in the electronic medical record, which mandates a provider to schedule a stop time on continuous IV fluids when placing the initial order. This study aimed to determine the effect of preemptive stop times for continuous IV fluids on duration of fluid therapy and patient outcomes.
 
Methods: 
This single center, retrospective, observational study was approved by the Institutional Review Board. Adult patients admitted to the general ward on a hospitalist service for at least 24 hours with orders for continuous IV fluids for at least 12 hours were included. Exclusion criteria included requirement for renal replacement therapy prior to admission or receipt of continuous fluids for cancer-related complications (i.e. tumor lysis syndrome), dysnatremias, pancreatitis, rhabdomyolysis, diabetic ketoacidosis, high output fistulas, total parenteral nutrition, or sepsis. The primary outcome of this study was to compare the duration of fluid therapy pre-and post-implementation of required stop times on continuous IV fluid orders. Duration of fluid therapy was assessed until hospital discharge or through 30 days after initiation. Secondary outcomes of this study included comparison of total volume of continuous fluids administered through day 5, hospital length of stay (LOS), incidence of intensive care unit (ICU) admission related to fluid overload, and any incidence of electrolyte disturbances throughout fluid administration. Nominal data was analyzed using a Chi-square or Fischer’s exact test. Continuous data was analyzed via Student’s t-test or Mann-Whitney U.
 
Results: 
A total of three hundred and fifty patients were included. Median duration of fluid therapy at 30 days was shorter in the post-protocol group compared to the pre-protocol group (1 day vs 2 days, p-value < 0.001). Median maintenance fluid volume at day 5 was lower in the post-protocol group compared to the pre-protocol group (1875 mL vs 3100 mL, p-value < 0.001). There were statistically significant reductions in the incidence of electrolyte disturbances, fluid overload, diuretic requirement, and increased oxygen requirements in the post-protocol group.
 
Conclusions: 
Implementation of a required stop time on continuous IV fluids orders reduced the duration of fluid therapy and volume of fluids administered. Further evaluations should be performed to assess the role that required stop times play in reducing hospital costs.

Title: A Retrospective Evaluation of an Electronic Medical Record Alert to Pharmacists on the Incidence of Inappropriate Medication Administration in Patients with Feeding Tubes 

Authors: Devin O'Brien, Rosemary Garbowski, Matthew Lane, Saumil Vaghela

Background: Medication administration is an important part of the foundation for medication safety and efficacy. Various studies have demonstrated the frequency of inappropriate administration of medications through a feeding tube. Many of these studies have shown that pharmacist interventions can positively impact the percentage of medications administered appropriately in patients with feeding tubes. Few studies have evaluated the effectiveness of a pharmacist-directed alert on the appropriate administration of medications. The purpose of this study was to evaluate the effectiveness of an electronic medical record alert to pharmacists on the incidence of inappropriate medication administration in patients with feeding tubes.  

Methods: A retrospective chart review was conducted using the electronic medical record (EMR) at an acute care community hospital. Chart reviews were conducted on patients with tube feeding (TF) orders who were admitted prior to and following implementation of the EMR alert. The pre-implementation period was between June 1, 2024 and August 31, 2024. The post-implementation period was between October 1, 2024 and December 31, 2024. A wash-out period was designated between September 1, 2024 and September 30, 2024 to ensure that all patients with a TF order triggered the alert to pharmacists in the EMR. Patients were included if they had a TF order placed and had at least one scheduled medication ordered to be administered enterally. Patients were excluded if they were covered by a service that participated in daily multidisciplinary team rounding, as the medications for these patients were individually evaluated regardless of the pharmacist-directed alert. The primary outcome for this study was the composite incidence of inappropriate medication administration and medication administration omissions in patients with feeding tubes (for example: crushed medications that should not be crushed per the package insert or a liquid administered through the feeding tube that has potential for binding to the feeding tube). The secondary outcome for this study was inappropriate administration of high risk medications. All data points collected for each patient were compiled in an electronic spreadsheet. 

Results: A total of 52 patients were included in the study, 29 in the pre-implementation group and 23 in the post-implementation group. For the primary endpoint of composite incidence of inappropriate medication administration and medication administration omissions,122 errors were found in the pre-implementation group and 71 in the post-implementation group. For the secondary outcome of inappropriate administration of high risk medications, there were zero patients in both groups. 

Conclusion: An EMR alert to pharmacists to evaluate medications in patients with feeding tubes may help to decrease the number of inappropriate medication administrations