P088: Dextrose 50% versus dextrose 10% or dextrose titration for the treatment of out-of-hospital hypoglycemia; a systematic review and meta-analysis

Introduction: Paramedics commonly administer intravenous dextrose to severely hypoglycemic patients. Typically, the treatment provided is a 25g ampule of 50% dextrose (D50). This dose of D50 is meant to ensure a return to consciousness. However, this dose may be unnecessary and lead to harm or difficulties regulating blood glucose post treatment. We hypothesize that a lower dose such as dextrose 10% (D10) or titrating the D50 to desired level of consciousness may be optimal and avoid adverse events. Methods: We systematically searched Medline, Embase, CINAHL and Cochrane Central on June 5th 2019. PRISMA guidelines were followed. The GRADE methods and risk of bias assessments were applied to determine the certainty of the evidence. We included primary literature investigating the use of intravenous dextrose in hypoglycemic diabetic patients presenting to paramedics or the emergency department. Outcomes of interest were related to the safe and effective reversal of symptoms and blood glucose levels (BGL). Results: 660 abstracts were screened, 40 full text articles, with eight studies included. Data from three randomized controlled trials and five observational studies were analyzed. A single RCT comparing D10 to D50 was identified. The primary significant finding of the study was an increased post-treatment glycemic profile by 3.2 mmol/L in the D50 group; no other outcomes had significant differences between groups. When comparing pooled data from all the included studies we find higher symptom resolution in the D10 group compared to the D50 group; at 99.8% and 94.9% respectively. However, the mean time to resolution was approximately 4 minutes longer in the D10 group (4.1 minutes (D50) and 8 minutes (D10)). There was more need for subsequent doses in the D10 group at 23.0% versus 16.5% in the D50 group. The post treatment glycemic profile was lower in the D10 group at 5.9 mmol/L versus 8.5 mmol/L in the D50 group. Both treatments had nearly complete resolution of hypoglycemia; 98.7% (D50) and 99.2% (D10). No adverse events were observed in the D10 group (0/871) compared to 12/133 adverse events in the D50 group. Conclusion: D10 may be as effective as D50 at resolving symptoms and correcting hypoglycemia. Although the desired effect can take several minutes longer there appear to be fewer adverse events. The post treatment glycemic profile may facilitate less challenging ongoing glucose management by the patients.