Brought in by Ambulance, #4: Pour Some Sugar on Me

Case Scenario: 

Your unit is responding on the quiet to a 36 year-old F with chief complaint of “sick.”  You arrive to find a cachectic woman (BMI 16) who suffers from lifelong anorexia. She complains of shortness of breath, abdominal pain and general malaise. She walks herself to the ambulance and after her vitals are taken, a FSBS is 56.  According to your protocol, this should be treated in the following manner: the EMT-B administers 15 gm oral glucose solution and repeat as needed, while the EMT-P administers 25 gm D50W IV/IO or D10W IV/IO, or glucagon 1 mg IM if no IV/IO access available.  The medic asks your opinion and as the patient was symptomatic, you suggested starting with oral glucose solution. However, this made her feel nauseated. Once again, the medic asks you whether to give D10W or D50W. You elected for D10W, but you admittedly have no evidence for doing so. After an uneventful ride to the hospital, the patient’s repeat blood glucose was 254 mg/dL.

Clinical Question:
What is the best way to treat hypoglycemia in the prehospital environment?

Literature Review:  

Most common sources define hypoglycemia as less than 60-70 mg/dL. According to the Endocrine Society’s Clinical Practice Guidelines, hypoglycemia should only be treated and investigated in patients showing signs of "Whipple’s triad" -- signs and symptoms of hypoglycemia, a low plasma glucose reading, and resolution of symptoms with elevation of plasma glucose concentration [1]. The signs and symptoms of hypoglycemia are quite vague and nonspecific; they can include shakiness, anxiety, headaches, and weakness, progressing to seizures and unconsciousness [2]. Severe symptoms typically occur with serum glucose values less than 40 mg/dL [3]. It seems obvious that seizures and unconsciousness attributed to hypoglycemia should be treated, but any of the minor signs and symptoms could easily be appreciated on almost any primary scene response. Therefore the vast majority of hypoglycemic patients encountered by EMS providers are most likely going to be treated.

Once we have decided to treat prehospital hypoglycemia, we must determine the optimum modality for doing so. We could find no published work doing a heads-up comparison of oral versus intravenous carbohydrate administration, but there has been work related to D10W versus D50W. An EMS system in California published data from their experience replacing 50mL D50W with 100mL D10W as the standard treatment for hypoglycemia [7]. In 164 treated patients, the median pretreatment glucose was 38 mg/dL, and at 8 minutes after treatment was 98 mg/dL.Twenty-nine patients required an additional dose, and one patient required a third. There were no adverse events reported. The authors conclude these results demonstrate the feasibility, safety, and efficacy of D10 as an alternative to D50.

One unblinded randomized controlled trial compared D10W to D50W in prehospital hypoglycemic patients with GCS < 15 (approx. 25 pts in each group) with regards to the ability of the solutions to raise glucose to normal values and the time to reach a GCS of 15 [4]. The mean repeat glucose value in these groups was 112 and 169 in the D10W and D50W groups, respectively (p = 0.003). The authors note that the average D10 given was 10g (100 mL) and D50 given was 25g (50 mL) (p < 0.001). There were no significant differences between the groups in median time to recovery, median post-treatment GCS, or number of patients experiencing another hypoglycemic episode within 24 hours. The authors conclude that due to the efficacy of treatment with lower risk of hyperglycemia, D10 should be the preferred agent.

   

Not only is D50 more likely to cause hyperglycemia after administration, but it also carries with it the risk of hypertonicity and tissue necrosis. D50 has an osmolarity of 2,525 mOsm/L and a pH between 3.5 and 6.5. This is in contrast to D10 which has an osmolarity of 506 mOsm/L and is pH neutral. Usual IV therapy recommendations state that solutions with osmolarity greater than 900 mOsm/L should be administered through central access. There have also been several case reports of extremity amputation after dextrose extravasation [5,8].

To illustrate the difficulty in administering D50, one of our toxicology faculty compared the injection of D50 to injecting maple syrup. This is not a facetious analogy, as a simple look at the nutrition label of a bottle of syrup reveals:

Note that D50 = 50g of sugar per 100mL of soluion. Each "amp" of D50 contains 25g of sugar in 50mL. Compare that to Aunt Jemima's, which contains 32g of sugar in 60mL solution.

The primary EMS agency bringing patients to our facility recently updated their protocols to make D10 the preferred agent for treating hypoglycemia prehospital. The Medical Director of this agency reiterated the preferable safety profile of D10 compared with D50, and remarked on another important advantage of D10 -- it does not need to be diluted for use in pediatric patients.

In the absence of IV/IO access, evidence exists for the efficacy of IM glucagon in treating prehospital hypoglycemia [9]. The response to glucagon may take longer than when oral or IV glucose solutions are used [10, 11], though may cause a greater increase in blood glucose levels than 10g of glucose solution [12].
   

Take-home:

- If mental status allows, hypoglycemic patients should first be offered oral glucose -- preferably 15gm of a standard glucose formula, but syrup, juice, or honey can also be used. (Note that honey has a lower glycemic index than most juices or glucose solution and thus theoretically may not be as effective.) 
- The next line of therapy should probably be IV D10, as there is good evidence to suggest efficacy on par with that of D50 but with a more favorable safety profile.
- If no IV/IO access is available and oral glucose is unable to be administered, it is reasonable to administer 1mg IM glucagon.

References:
[1] Evaluation and management of adult hypoglycemic disorders.  J Clin Endocrinol Metab, 2009, 94(3), 709.
[2] American Diabetes Association.  Diabetes.org
[3] National Diabetes Information Clearinghouse.  Diabetes.niddk.nih.gov
[4] Moore and Woolard. Emerg Med J, 2005, 22, 512-515.
[5] Kumar et al. ANZ Journal of Surgery, 2001, 71, 285-289.
[6] Is D50 Too Much of a Good Thing?  Stephen Wood, 2007, m.jems.com/article/patient-care
[7] Kiefer et al. Prehosp Disaster Med, 2014, 29(2), 190-194.
[8] Lawson et al. Am J Emerg Med, 2013, 31(5), 886:e3-5.
[9] Vukmir et al. Ann Emerg Med, 1991, 20(4), 375-9.
[10] Howell et al. J Accid Emerg Med, 1997, 14(1), 30-2.
[11] Carstens and Sprehn. Prehosp Disaster Med, 1998, 13(2-4), 44-50.
[12] Vermeulen et al. Diabetes Care, 2003, 26(8), 2472-3.

Submitted by Chris Miller, PGY-2.
Edited by C. Sam Smith (@CSamSmithMD), PGY-3.
Faculty reviewed by Hawnwan P. Moy.

#FOAMed Digest No. 4: Butter My Biscuit, Baby

Welcome back, to the brand new edition of the WUEMR FOAMed Digest. Get out your Tintinalli’s and strap in, because we’re going back to basics today. It’s all about the bread and butter. The things any PGY-2 setting off to an overnight Saturday shift in the Deuce should have down cold…yet us seniors still screw up on the daily.

FOAMed…ENGAGE!

Three Stars:

1. If my last shift at Children’s is any indication, the season is upon us – pharyngitis in every exam room. Casey Parker over at Broome Docs (a blog authored by EPs & GPs practicing in rural Australia), presents a magnificent summary of the data surrounding rapid strep swabs, antibiotic use for symptom relief, and antibiotic use for preventing secondary complications of strep. As always, be sure to check out the original literature for yourself. And don’t miss Minh Le Cong’s excellent counterpoint in the comments, which is also well-referenced.

2. What’s your record for most C-collars cleared in one shift? (When you hit double-digits, then we can talk.) The best tools in our arsenal for clearing C-spine in low-risk patients remain the Canadian C-spine and NEXUS instruments. But which one should you use? Do you even remember which criteria belong in each rule, or do you find yourself trying to apply the “Canadi-EXUS” criteria, like I do? Luckily for us, Alayna Hawling at BoringEM authored an excellent rundown and comparison – with a pretty flowchart!

3. As much as you want to start the fist-pumping and beer-chugging as soon as you drop that tube past the cords, your work with the intubated patient is not done, my friend! We’ve already touched on our persistently poor rates of achieving adequate analgesia & sedation in the intubated patient. Another part of quality post-intubation care is knowing what to do if your ventilated patient acutely decompensates. Check out Chris Cresswell’s summary of the DOTTS mnemonic over at EM Tutorials.

(EXTRA CREDIT: He also included a link to Scott Weingart’s notes regarding care of the crashing ventilated patient, which are well worth a look.)

Oldie But Goodie:

There’s been some e-mail discussion lately among our attendings regarding the best way to clean lacs prior to closure. Back in February, Ken Milne at the Skeptic’s Guide (along with Eve Purdy, a rockstar med student and creator of the excellent Manu et Corde blog) published a piece dedicated to breaking down the dogma of management of simple lacerations. Tap water vs sterile water, sterile gloves vs clean gloves, to sew or not to sew…it’s all covered here. Plus there’s links to other excellent FOAMed resources regarding wound care dogma.

F(FN)OAMed:

The good folks over at EB Medicine recently published a stem-to-stern guide to UTI diagnosis and management in the ED, all based on best available evidence. A bit lengthier than your average blog post, but incredibly high-yield and well worth your time. It’s a bit difficult for me to place a direct link here, but you can find it simply by logging into your account at EBMedicine, following the link to browse issues of Emergency Medicine Practice, and opening the July 2014 issue on UTI.
(As always, contact your friendly neighborhood Social Media Committee member if you need help obtaining access to EB Medicine resources.)

The Gunner Files:

1. Hard to get through a Deuce shift without breaking out the prochlorperazine at least once. We’ve all seen patients get jittery, agitated, or downright whacky following its use. Does Benadryl help? A PharmD expert at ALiEM has a good lit review of the topic.

2. Short and sweet: some diabetic medications are more likely to cause harmful hypoglycemia after overdose than others. Quick table-based rundown over at ALiEM.

3. It is asthma season, and you may find yourself in the worst-case-asthma-scenario of impending need for intubation. Check out this post from The Kings of County regarding care for the sick asthmatic, including intubation and mechanical ventilation issues.

4. FOAMed is taking the world by storm! Does the UK College of Emergency Medicine launching a dedicated FOAMed site mean it’s officially gone mainstream? Don’t worry – we were all into FOAMed before it was cool. But seriously, check out this vodcast on diagnostics in EM, and not feel quite so much increase in sphincter tone when Carpenter or Cohn pimp you on likelihood ratios or Bayesian analysis.

5. Another classic from the Skeptic’s Guide, this time addressing another oh-so-common ED complaint: renal colic. Fluids? Flomax? Any good evidence for either? In news that will surprise no one, Ken Milne is skeptical.

Never stop learning,

Sam Smith, PGY-3

Insulin bolus in DKA?

A middle aged male with a history of IDDM presents with 2 days of nausea and vomiting. He reports running out of his insulin. In triage, FSBS is critical high, FS ketones are 5.0. You start to treat for DKA with bolus of IVF and draw basic labs. You order your insulin infusion and give a 0.10U/kg bolus prior to starting the drip. Your attending questions your actions. 

Clinical Question:

Is there any evidence to support giving an insulin bolus prior to starting a drip in DKA patients?

Literature:

The latest ADA recommendations last updated in 2009 still recommend giving a 0.1U/kg bolus prior to initiating an insulin infusion at 0.1U/kg/hr. This recommendation is based on the theory that DKA represents a significant "insulin resistant" state, and a bolus of insulin is needed to overcome this resistance and effectively reduce serum glucose and suppress gluconeogensis. Much of these recommendations are based on limited studies from the 1980's, measuring active levels of insulin in serum, and none were prospective comparing continuous insulin infusion with and without the bolus. (1) In 2008, Kitbachi Et al performed a randomized prospective study with three arms, those on an insulin infusion at 0.07U/kg/hr with and without a bolus, and those who were only on a infusion at 0.14U/kg/hr. Kitbachi effectively demonstrated that an insulin bolus is unnecessary in those who were started on a infusion at 0.14U/kg/hr. There was no difference in time to optimal glucose, pH, Anion Gap, or bicarbonate. (2) In 2010, Goyal Et al performed a similar study but had two arms, insulin infusion at 0.1U/kg/hr and those with an infusion and a bolus. He showed no difference in length of ED stay or hospital stay, and no difference in time to goal glucose or anion gap closure. (3)

Take Home:

In summary, there appears to be a paucity of evidence supporting the ADA recommendation for an insulin bolus prior to infusion, however there seems to be robust evidence showing at least a noninferiority with insulin infusion alone. Therefore I find there is no reason to continue using an insulin bolus during my standard treatment of DKA.

References:

1) Diabetes in the Emergency Department: Acute Care of Diabetes Patients. Clinical Diabetes April 1, 2011 29:51-59 http://care.diabetesjournals.org/content/32/7/1335.short
2) Kitabchi Et al, Is a priming dose of insulin necessary in a low-dose insulin protocol for the treatment of diabetic ketoacidosis? Diabetes Care. 2008 Nov;31(11):2081-5
3) Goyal Et al, Utility of initial bolus insulin in the treatment of diabetic ketoacidosis. J Emerg Med. 2010 May;38(4):422-7. doi: 10.1016/j.jemermed.2007.11.033. Epub 2008 Jun 2.

Submitted by Louis Jamtgaard, PGY-3