Background: Current practice standards regarding telemetry utilization do not address the usage of telemetry for non-cardiac indications in the acute care, non-ICU, setting. The American Heart Association (AHA) practice standards make no recommendation on the use of cardiac monitoring for non-cardiac conditions (non-AHA) with the potential for hemodynamic decompensation, such as sepsis, gastrointestinal bleeding, alcohol withdrawal, and respiratory infections. Telemetry utilization in these settings is often labeled as overuse with ordering patterns subject to personal variation in practice. There are significant gaps in the literature regarding its use and utility.
Methods: A clinical decision support tool was developed to increase concordance with AHA practice standards for continuous cardiac monitoring within a 10-hospital academic health system. The telemetry orders were tied to AHA practice standards, which included a write-in option for non AHA-indications. We explored overall use of telemetry for AHA and non-AHA indications, as well as in-hospital mortality, all cause 30-day mortality, length of stay, ICU transfer rate, episodes of bradycardia (HR < 50), tachycardia (HR > 160), and hypotension (mean arterial pressure < 65) on and off telemetry, comparing against the non-telemetry population. We confirmed the stability of our findings through manual retrospective chart review of telemetry orders placed 6 months prior to the intervention. For the purpose of this study, we focused on the first telemetry order of a hospitalization.
Results: Following implementation of a clinical decision support tool, 11,025 orders for telemetry were placed and of which 8,037 had a documented indication. Of those 1,525 admissions (19.0%) had a non-AHA indication reported. Patients with a non-AHA indication had lower ICU transfer rates (14.2% vs 30.7% p < 0.001), similar length of stay (6.3 days vs 6.3 days, p = 0.98), and lower in-hospital mortality (3.5% vs 4.5%, p = 0.009), when compared with patients with guideline concordant telemetry. However, patients on telemetry outside of AHA indications had markedly higher ICU transfer rates (14.2% vs 3.8%), and mortality rates (3.5% vs 0.1%) than patients with no telemetry order. The most common non-cardiac indications provided were: respiratory failure (180 orders), sepsis (179), gastrointestinal bleeding (123), diabetic ketoacidosis (98), pulmonary embolism (94), and acute alcohol withdrawal (71). Hypotensive episodes were observed both on telemetry (7.9%) and after discontinuation (8.3%) for this population. Discontinuation of telemetry prior to hypotensive episodes was not associated with worse outcomes.
Conclusions: Standardization of telemetry ordering to adhere to AHA guidelines assists with exploring gaps in evidence. We find that approximately 20% of telemetry orders fall outside of AHA indications. These patients have higher overall risk than the general hospital population, but lower when compared to patients on telemetry with cardiac indications. Exploration of the impact of telemetry on the outcomes of patients who are at risk for hemodynamic instability is important and now feasible.