Written by Clay Smith
Lowering glucose prior to discharge may negatively impact ED length of stay (LOS) without short-term benefit for the patient, but this study wasn’t able to prove it.
Why does this matter?
The same group found in a prior retrospective study that there was no correlation with discharge glucose level and 7-day rate of ED return. They took it further with this prospective randomized trial.
Study on sugar needs grain of salt
This was a RCT of 110 patients with type 2 diabetes who had hyperglycemia (>400 to <600) in the ED. They were assigned to either loose control (discharge glucose <600) or moderate control (discharge glucose <350). For the primary outcome of ED LOS, there was no statistical difference based on intention-to-treat, 211 minutes for moderate control, 216 minutes for loose control. In post hoc analysis, those who actually achieved a glucose of <350 had ED LOS 29 minutes longer (95%CI, 1-59 min). For secondary outcomes, there was no difference at 7 days in repeat ED visits for hyperglycemia, hospitalization for hyperglycemia, or hospitalization for any reason. Keep in mind: type 1 diabetic patients were excluded; a large number of patients were lost to follow-up for the secondary outcomes; and the authors commented, “In retrospect, the choice of LOS as a primary outcome was not ideal.” For me, it’s hard to not treat a glucose in the 400-600 range. What this study adds is weak evidence that this practice probably makes ED LOS longer with no impact on short-term outcomes.
For those who prefer glucose in mmol/L, you can use this calculator.
EM Ottawa Blog had a helpful post on discharge glucose targets.
Comparison of two glycemic discharge goals in ED patients with hyperglycemia, a randomized trial. Am J Emerg Med. 2018 Oct 5. pii: S0735-6757(18)30809-X. doi: 10.1016/j.ajem.2018.09.053. [Epub ahead of print]
Open in Read by QxMD