Imagine. You're on a busy red team shift. A clinic calls to send a patient with asymptomatic hyperglycemia to the ED. You roll your eyes, transfer it to the attending. The patient of course lands in the resus bay and you go to evaluate them. Their sugar is 450 mg/dL, they have no symptoms and their vitals are normal. You mindlessly order a VBG, basic labs and a liter of fluid and move on with your day. However, is this all necessary? What should our end goal be? What should I choose as my intervention if necessary? Do I even need labs?
Today, we will discuss the topic of management of hyperglycemia without DKA/HHS in the Emergency Department. While researching, I was surprised to find very few clinical guidelines regarding these scenarios. DKA and HHS protocols are now well built and evidence based. Most hospitals have extensive and thorough guidelines for these critical patients. But what do you do with patients who don't meet these criteria? How do we best care for them while also maximizing resources? Today we’ll explore three specific clinical questions for these scenarios and some studies to support the answers. (It's important to emphasize this conversation is not meant to apply to type 1 diabetics due to a higher propensity for DKA at lower blood glucose levels)
What is my end goal glucose?
We’ve all asked the question to our attendings, “at what glucose do you feel comfortable for discharge?” Ultimately, we must consider this question first or we can’t effectively select our interventions.
Unfortunately, there are not a ton of studies on this topic. The most cited studies on this topic come from Hennepin and the authors Driver et al.
Their initial study in 2016 was a retrospective cohort study evaluating whether discharge glucose in patients with glucose levels >400 mg/dL at any point in the ED affected 7 day outcomes.3 Patients were notably excluded if they were type I diabetics or hypoglycemic on arrival. Outcomes included the occurrence of DKA/HHS, repeat ED visit for hyperglycemia, or hospitalization. Mean discharge glucose was 334 mg/dL. Most patients received glucose lowering therapy, with 60% receiving both IV fluid and subcutaneous insulin. Mean glucose reduction was 150 mg/dL. Only 2 patients returned with DKA on follow-up. No deaths were observed. And no statistically significant increase in hospitalizations, repeat visits, DKA or HHS were observed.
Driver et al then completed an additional study in 2019.4 It was an RCT evaluating whether moderate control vs “loose” glycemic control in the ED effected length of stay, as well considered largely the same 7 day outcomes from the initial study. Patients were included with glucose levels ranging from 400-600 mg/dL, and randomized into discharge glucose targets of less than <350 vs <600 mg/dL. Again, most patients received IV fluid. Very few received insulin in the “loose” control group (only 4 of 56). No significant differences in outcomes were noted, and ED length of stay was reduced on average by 30 minutes in the loose control group.
While there are many limitations with these studies, they do suggest that we may not be reducing harm by dropping our patients’ blood sugars, and may be unnecessarily using resources and increasing visit length.
As a side note, many authors mention many glucometers do not give a value above 600 mg/dL, they simply read “high”. This may be a logical, firm upper limit to allow patients to track their actual glucose on discharge.
Should we give IV fluids or Insulin?
On this topic there is a lot of conflicting opinion and not a lot of high-powered data.
In patients on insulin therapy, it would be reasonable to administer their home dose, especially if non-adherent. Patient’s naïve to insulin may require lower doses, or simply hydration. Notably 2% of patients in the 2016 Driver study did experience iatrogenic hypoglycemia, which is a large risk to undertake with potentially minimal benefit.
For severe hyperglycemia (>600 mg/dL), most experts recommend IV fluids for dilution of the blood glucose and further diuresis for elimination of the glucose from the body.2 By nature, chronically severely hyperglycemic patients may be intravascularly depleted due to the diuretic effect of glucose and benefit from hydration.
However, one study compared IV vs oral rehydration therapy in efficacy of dropping blood glucose and found them to be equivalent.1 So, in moderate hyperglycemia and well appearing patients, it may be reasonable to opt for a pitcher of water in lieu of IVF.
Obviously, always use clinical judgement. If your patient appears dry on exam, is tachycardic, etc., fluid is a largely safe intervention if there are no other contraindications.
Do I need labs?
There is really no data to provide evidence on the benefit of lab work in these patients. Most mild-moderate hyperglycemia patients without symptoms will likely not benefit from lab work. Usually, DKA and HHS are not subtle.
However, those who appear clinically dehydrated, have complex co-morbidities, or infrequent health care exposure may benefit from screening labs to evaluate for renal function, electrolyte abnormalities etc. Additionally, always ask a thorough history to consider if there may be an underlying medical etiology for their hyperglycemia (ischemia, sepsis, medications).
Ultimately, use your clinical judgement. But for most asymptomatic patients with mild hyperglycemia, blood work may not be necessary.
Final Thoughts
Your brief management in the ED of these patients will likely not affect their long term outcomes. Most importantly, educate at bedside on the critical importance of PCP follow-up for optimization of their long-term medications and counselling on dietary changes. Make a GMC Diabetes follow-up clinic when appropriate, this may be the most critical step in your management.
In Summary
Aggressive management of our asymptomatic hyperglycemic patients may not be necessary, and may lead to longer stays, more resource utilization, and potential adverse effects.
However, again I will emphasize the paucity of data and the importance of clinical judgement. Many of us will differ in our risk aversion and opinions on management, but hopefully this will allow you to consider your care for these patients in a more thoughtful way in the future!
References
Arora S, Probst MA, Andrews L, Camilion M, Grock A, Hayward G, Menchine M. A randomized, controlled trial of oral versus intravenous fluids for lowering blood glucose in emergency department patients with hyperglycemia. CJEM. 2014 May;16(3):214-9. doi: 10.2310/8000.2013.131082. PMID: 24852584.
Banh K, Acosta J, Hyperglycemia, SAEM M4 Curriculum. https://www.saem.org/about-saem/academies-interest-groups-affiliates2/cdem/for-students/online-education/m4-curriculum/group-m4-endocrine-electrolytes/hyperglycemia
Driver BE, Olives TD, Bischof JE, Salmen MR, Miner JR. Discharge Glucose Is Not Associated With Short-Term Adverse Outcomes in Emergency Department Patients With Moderate to Severe Hyperglycemia. Ann Emerg Med. 2016 Dec;68(6):697-705.e3. doi: 10.1016/j.annemergmed.2016.04.057. Epub 2016 Jun 25. PMID: 27353284.
Driver BE, Klein LR, Cole JB, Prekker ME, Fagerstrom ET, Miner JR. Comparison of two glycemic discharge goals in ED patients with hyperglycemia, a randomized trial. Am J Emerg Med. 2019 Jul;37(7):1295-1300. doi: 10.1016/j.ajem.2018.09.053. Epub 2018 Oct 5. PMID: 30316635.
Shaw J, Hill J. Hyperglycemia in the ED. Taming the SRU. https://www.tamingthesru.com/blog/diagnostics/hyperglycemia-in-the-ed
Authored by Kathryn McGregor MD, Joanne Routsolias PharmD, and Eric Leser MD