Virtual Critical Care at Carolinas HealthCare System

Sonia Everhart, Pharm.D., BCPS, BCCCP; Desiree Kosmisky, Pharm.D., BCCCP; Colleen Karvetski, Ph.D.; Nehal Thakkar, M.D., FCCP; Michael Reif, M.D., FCCP; Kimberly Purtill, R.N., M.S., CCRN

Virtual Critical Care at Carolinas HealthCare System, Charlotte, North Carolina

Hyperglycemia, a commonly observed metabolic abnormality occurring in up to 77.3% of critically ill patients with and without diabetes, is linked to mortality. Recommendations by several national organizations led Carolinas HealthCare System (CHS) to target blood glucose (BG) values < 180 mg/dL and > 70 mg/dL, with tele-intensive care unit (ICU) pharmacy services hypothesized to improve glycemic control.

The tele-ICU pharmacy program began in September 2015 with several anticipated benefits including extension of critical care pharmacy services to 137 ICU beds at eight hospitals without dedicated personnel, promotion of medication safety, and provision of drug information. Glycemic control represented the most frequently performed tele-intensivist intervention that could be delegated to a pharmacist. Baseline glycemic control in the facilities selected for intervention was poor, with CHS ranking 22^nd of 39 on percent of patient days with average BG ≤ 150 mg/dL among health-care systems supported by the Philips VISICU^TM platform. Two tele-ICU pharmacists developed a standard workflow utilizing custom TheraDoc^® alerts for abnormal glucose (< 70 mg/dL or >180 mg/dL), electrolyte, and lactate levels.

Since inception, 10,265 TheraDoc^® alerts were reviewed and 2,843 interventions performed in 1,556 patients. Of these interventions, 1,277 related to glycemic control in 788 patients, accounting for 44.9% of tele-pharmacist activity. To evaluate the impact of tele-pharmacy on glycemic control, a non-randomized, retrospective study was designed with the primary outcome of percent of patient days with an average BG > 180 mg/dL divided by total ICU patient days with at least one reportable glucose value. The secondary outcome was number of hypoglycemic events per patient days with at least one reportable glucose value. During the control timeframe (September 1, 2014 to August 31, 2015), 18.8% of patient days had an average BG > 180 mg/dL (5669/30,163 patient days with at least one reportable BG). In the tele-pharmacy intervention period (September 1, 2015 - June 30, 2016), 16.1% of patient days had an average BG > 180 mg/dL (3861/24,054). Overall, the percentage of patients with average BG ≥ 180 mg/dL in the tele-ICU intervention period significantly decreased by 2.7% (P < 0.001). During the control period, severe hypoglycemic events (BG < 50 mg/dL) per patient days with at least one reportable glucose value totaled 1,119 of 30,163 patient days (3.7%) compared to 923 of 24,054 patient days (3.8%) in the intervention period (P = 0.4390), suggesting no predisposition to harm. Additionally, in Quarter 1 of 2016, CHS ranked 10^th of 42 health-systems supported by the Philips VISICU^TM platform, with average BG < 150 mg/dL on 70% of patient days.

This innovative tele-ICU pharmacy model resulted in significant glycemic control improvements and expanded critical care pharmacist coverage across CHS facilities.

References

1. Plummer MP, Bellomo R, Cousins CE et al. Dysglycemia in the critically ill and the interaction of chronic and acute glycaemia with mortality. Intensive Care Med. 2014; 40:973-80.
2. Finney SJ, Zekveld C, Elia A et al. Glucose Control and Mortality in Critically Ill Patients. JAMA. 2003; 290:2041-7.
3. Krinsley JS. Association between hyperglycemia and increased hospital mortality in a heterogeneous population of critically ill patients. Mayo Clin Proc. 2003; 78:1471-8.