D50 vs D10 for Severe Hypoglycemia in the Emergency Department

D50 vs D10 for Severe Hypoglycemia in the Emergency Department

d50Think back to your last severely hypoglycemic and lethargic patient presenting to the ED. What was the first treatment modality that came to mind? The initial knee-jerk reaction might be to reach for that big blue box of D50 if the patient has IV access. After all, top priority is to reverse hypoglycemia as fast as possible. But in the midst of stabilizing the patient, how often do we consider the potential aftermath of concentrated glucose?

Drawbacks of D50

1. Rebound hypoglycemia

After administration of D50 there is an excess amount of glucose available, leading to increased uptake and utilization by the tissues, which suppresses both gluconeogenesis and glycogenolysis. Without continued administration of dextrose-containing fluids, this may result in rebound hypoglycemia. Moreover, with the loss of IV access, a busy ED, and a subsequent fingerstick in the 200’s, the decision to initiate such fluids may be overlooked or delayed, thereby facilitating this risk of rebound hypoglycemia.

Unfortunately, the prevalence of rebound hypoglycemia from D50 is not well documented. One study reported subsequent hypoglycemic events occurring among 18% of patients who received both D50 and D10.1 Another study reported that glucose levels after administration of D50 returned to baseline within 30 minutes in a sample of healthy volunteers.2

2. Overshooting glycemic targets after treatment

Several studies have reported post-administration glucose levels between 160-250 mg/dL1,2; While this may be perceived as inconsequential during initial management, a recent retrospective, single-center study in the critically ill suggests that both rapid excursions in glucose and subsequent glycemic variability may be the best predictors of overall morbidity and mortality among these patients.3 It may be reasonable to consider other treatment approaches that may be less likely to complicate further glycemic management upon admission to the hospital, instead of one that can exacerbate glycemic variability.

3. Hypertonic toxicity

It is well known that intravenous administration of hypertonic fluids may result in harm to surrounding tissues. In fact, the osmolarity of D50 is even greater than that of 8.4% sodium bicarbonate (2500 mOsm/L compared to 2000 mOsm/L, respectively). Most would advocate for administration via a central line with osmolarities in excess of 900 mOsm/L.4 While the literature varies, rates of extravasation from such fluids have been cited to be as high as 10-30%.5 This risk can be mitigated somewhat by pushing D50 slowly over 2-5 minutes. Consider however, the osmolarity of 10% dextrose (D10) is 500 mOsm/L, and within range for safer peripheral administration.

Adverse effects of D50 include:

  • Local skin irritation
  • Thrombophlebitis
  • Extravasation with subsequent tissue necrosis

Evidence for use of 10% Dextrose (D10)

One small randomized controlled trial evaluated the effectiveness of D10 vs D50 for the treatment of hypoglycemia by EMS providers in the pre-hospital setting.1 The study included 51 profoundly hypoglycemic patients (median blood glucose of 26 mg/dL evenly distributed among both arms. The investigators pushed small 5 g aliquots of each (50 mL of D10 and 10 mL of D50), while waiting one minute in between doses to reassess the patient. The maximum cumulative dose of dextrose permitted was 25 g.

The primary objective of the study was time to Glasgow Coma Scale (GCS) score of 15. The investigators also took note of post-treatment glucose levels, and the total dose of dextrose administered. Here are the pertinent results:

  • Median pre-treatment GCS was 4 and 6 in the D10 and D50 arms, respectively.
  • Median time to GCS of 15 was 8 minutes in both groups.
  • Median total doses administered were 10 g (100 mL) and 25 g (50 mL) in the D10 and D50 arms, respectively.
  • Median post-treatment glucose levels were also significantly different: 111 mg/dL in the D10 arm and 169 mg/dL in the D50 arm.
  • Rates of rebound hypoglycemia were identical between both groups (18%).

This study suggests that higher concentrations of dextrose-containing fluids do not expedite reversal of hypoglycemia as compared to lower concentrations, and that lower concentration formulations are more likely to achieve normal glycemic targets. However, because the investigators used aliquots of dextrose to permit intravenous pushing of both formulations, concern may arise over the study’s external validity- -as lower-concentrations of dextrose are typically given as piggybacks or continuous infusions in the Emergency Department. Additionally, no information was provided regarding rates of administration in either arm.

More recently, the California Contra Costa County EMS system revised their current protocol adopting the use of a 100 mL bolus of 10% dextrose for treatment of hypoglycemia in the field.6 An observational cohort of 162 patients treated after protocol implementation demonstrated the feasibility, safety and efficacy of this treatment modality. The median initial blood glucose of patients in the cohort was 38 mg/dL, and median post-treatment glucose levels were 98 mg/dL. The median time to second recheck of blood glucose was 8 minutes. Of those treated, 18% required an additional 100 mL bolus. There were no adverse events or deaths reported. Interestingly, a linear regression analysis of the elapsed time between initial and subsequent glucose values suggested minimal short-term decay in blood glucose after administration – a frequently cited concern with administration of D50.

Infusion Misconceptions

A common over-simplification made during acute situations is that medications need to be given by intravenous push for rapid administration. While this may be true in many cases, it may not be when dealing with vesicant substances, such as dextrose. Although D50 is intended to be given via intravenous push, it still requires slow administration. Another misconception is that administration via piggyback is “too slow.” While flow rates will vary depending upon the cooperation of the patient and the intravenous access obtained, even small, 22 gauge catheters can achieve flow rates of between 35-40 mL/min. Additionally, small amounts of pressure applied to the bag may effectively double the rate of administration.7 Thus, a 200 mL bolus of D10 (20 g) can be administered about just as fast as an amp of D50 should be.

Potential Scenario

“Our hypoglycemic patient is much more alert and oriented after 25 grams of D50, however, we completely blew the IV line in their hand, we can’t get another line in, and the repeat fingerstick is now 250 mg/dL. Is there a way to avoid these complications in the future?”

Suggested approach

If intravenous access has been obtained:

  • Hang a 250 mL bag of 10% dextrose, and administer a bolus of 100-200 mL.
  • If a very small catheter has been used (22-24G), may apply light pressure to bag to facilitate administration.
  • Clamp tubing between boluses, and reassess for improvements in mental status. Re-check the fingerstick glucose.
  • Repeat boluses as needed until the patient becomes alert and oriented, and/or until normoglycemia is achieved.

The aforementioned approach may be most applicable when:

  • A patient has a poor-quality line
  • A patient is a known brittle diabetic
  • Ongoing drug shortages preclude access to D50
1.
Moore C, Woollard M. Dextrose 10% or 50% in the treatment of hypoglycaemia out of hospital? A randomised controlled trial. Emerg Med J. 2005;22(7):512-515. [PubMed]
2.
Balentine J, Gaeta T, Kessler D, Bagiella E, Lee T. Effect of 50 milliliters of 50% dextrose in water administration on the blood sugar of euglycemic volunteers. Acad Emerg Med. 1998;5(7):691-694. [PubMed]
3.
Krinsley J. Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med. 2008;36(11):3008-3013. [PubMed]
4.
Kuwahara T, Asanami S, Kubo S. Experimental infusion phlebitis: tolerance osmolality of peripheral venous endothelial cells. Nutrition. 1998;14(6):496-501. [PubMed]
5.
Wiegand R, Brown J. Hyaluronidase for the management of dextrose extravasation. Am J Emerg Med. 2010;28(2):257.e1-2. [PubMed]
6.
Kiefer M, Gene H, Alter H, Barger J. Dextrose 10% in the treatment of out-of-hospital hypoglycemia. Prehosp Disaster Med. 2014;29(2):190-194. [PubMed]
7.
Reddick A, Ronald J, Morrison W. Intravenous fluid resuscitation: was Poiseuille right? Emerg Med J. 2011;28(3):201-202. [PubMed]

ALiEM Copyedit

Dr. Spaulding:

Thanks for the excellent post! Here are a few changes that I\'ve made and others I recommend:

  1. I\'ve added a preview image to the post which will be displayed as the default image on the ALiEM home page, please let me know if this works for you.
  2. I\'ve shortened the title of the post for better SEO optimization
  3. I\'ve reformatted various sections of the text to be in line with ALiEM\'s F-Style formatting, including adding bulleted lists for various results and new subheadings. Please look these over and make any changes you see fit.
  4. Changed the heading \"Nursing 101\" to \"Infusion Misconceptions\" to be more consistent with section content.

Other recommended changes:

  1. Under \"Suggested Approach\" you might want to specify the gauge of a \"very small\" catheter.
  2. Consider condensing bullets 1 and 2 under \"approach may be most applicable\" to simply: \"When patient has a poor-quality line\" or \"When suspicion of extravasation or line occlusion is high\"

Thank you, feel free to contact me with any questions or concerns.

Scott Kobner
Medical Student @ New York University
ALiEM-EMRA Social Media and Digital Scholarship Fellow
Founder, EdintheED.com

Thank you for the picture and re-formatting. All looks fantastic! I agree with your recommended changes, and these have been reflected in the post. Thanks!

Adam Spaulding, PharmD BCPS
Emergency Medicine Pharmacist
Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog

Expert Peer Review

Dr. Spaulding,
Great article on a drug that is given on a daily basis but it\'s evidence or rational is often not given much though. The previous ALiEM copyeditors have done an excellent job with you and this article and I don\'t have much more to add.

I agree with Dr. Coralic that there may be obstacles from an operational perspective in administering D10. At the institutions I\'ve worked at, D10 comes as 250 mL, 500 mL, and 1000 mL bags, so administering volumes other than the complete bag would likely require a pump. This is probably a better/safer solution overall but it would be something to think of.

Separately, it\'s worth adding something on D50 in patients on insulin or sulfonylureas, although it could be an entirely separate article. But for these types of patients who are otherwise able to tolerate food, enteral glucose administration (ie, a sandwich) would lead to less fluctuation in glucose.

Craig Cocchio, PharmD, BCPS
Clinical Assistant Professor
Residency Program Director
Emergency Medicine
Rutgers University and Robert Wood Johnson University Hospital
Editor-in-Chief of Emergency Medicine PharmD Blog

Dr Cocchio: Thank you – much appreciated.In terms of implementing the use of D10 from an operational perspective, please refer to my rationale above in response to Dr. Coralic. Briefly, if protocols for the management of severe hypoglycemia include only 250mL bags of D10 (small volumes), I think it is safe to administer off of a pump. Think about how variable rates of administration are with D50 (rate is dependent upon the user). The infusion of 250 mL by gravity (which typically takes about 5 minutes, all else constant) likely results in little consequence.In terms of implementing the use of D10 from an operational perspective, please refer to my rationale above in response to Dr. Coralic. Briefly, if protocols for the management of severe hypoglycemia include only 250mL bags of D10 (small volumes), I think it is safe to administer off of a pump. Think about how variable rates of administration are with D50 (rate is dependent upon the user). The infusion of 250 mL by gravity (which typically takes about 5 minutes, all else constant) likely result in little consequence. Moreover, I would advocate for even smaller volumes. Agreed. Hypoglycemia from sulfonylureas, in particular, is worthy of an entirely separate post. Oral glucose or food is certainly appropriate in certain situations, but likely not in severe hypoglycemia which is the focus of my post.

Agreed. Hypoglycemia from sulfonylureas, in particular, is worthy of an entirely separate post. Oral glucose or food is certainly appropriate in certain situations, but likely not in severe hypoglycemia which is the focus of my post.

Adam Spaulding, PharmD BCPS
Emergency Medicine Pharmacist
Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog

Expert Peer Review

Thank you for your time and effort in writing this post Dr. Spaulding. Overall, this is a great topic and very much applicable to pre-hospital and emergency medicine.

Please consider some of the following suggestions:

  1. I hesitate calling D50 a “Myth,” as this may be misinterpreted. Most resuscitation areas are likely to have D50 more readily available than bags of D10. My concern is having a bedside clinician delaying prompt treatment of hypoglycemia. Would you consider “D50 vs D10 for Severe Hypoglycemia in the Emergency Department” as a title?
  2. Please clarify the following statement, “Unfortunately, the prevalence of rebound hypoglycemia from D50 [or D10] is not well documented. One study reported subsequent hypoglycemic events from [both dextrose concentrations] occurring among 18% of studied patients [1].” Since you are comparing these two agents, it is important to state that at 24 hours, in this small study, there was no difference in incidence of hypoglycemia. As originally written in this paragraph, it may seem that only D50 is the culprit. It may be useful bolding this information in the later paragraph as well.
  3. To keep with the tone of the article, would recommend deleting “undoubtedly overkill for treating any degree of hypoglycemia.”
  4. Please consider revising, “recent literature among the critically ill,” to “a recent retrospective single center study in the critically ill…”
  5. Please consider revising, “It is well known that administration of hypertonic fluids is not without the additional risk of causing harm to surrounding tissues, particularly after overzealous administration,” to “Intravenous administration of hypertonic fluids may result in harm to surrounding tissues.”
  6. It would be interesting to see you elaborate more on feasibility of administering D10 in the emergency department. For example, you state 200 ml can be administered about just as fast as D50 would; however, this does not take into account the set-up time. Most hospitals will have 500-1000 ml bags of D10, and the time it takes to retrieve the bags, measure out the correct amount, prime the line, and program the pump must be taken into account. This time may be negligible for a seasoned nurse when they only have this one task at hand, but it may be challenging for a less-experienced nurse or a clinically demanding patient, especially if the patient is combative due to altered mental status where the hand must be physically restrained to ensure adequate flow through the cannula.
  7. Another thing that would be very useful is to review the evidence of using “Infant” dextrose or D25 in adult patients. D25 is half as concentrated, and it may be useful where D10 tends to be an issue (setting up an infusion). D25 comes in 10 mL sticks, and maybe one or two sticks of D25 would be a better approach than a large bolus of D50? It would certainly give the clinician better control of the amount of dextrose infused and glucose variability. If there is no evidence for D25, it would be beneficial in stating the lack of research on this topic and as a possible area of research.
  8. Also, please consider elaborating on subsequent infusions of D10, regardless of the dextrose concentration used for rescue of the initial hypoglycemic episode. From the small study [1] we know that many patients will have rebound hypoglycemia. Are there any official recommendations for D10 infusions (i.e., start infusion at 100 mL/hr with Q1H blood glucose checks) to prevent rebound hypoglycemia? Should we start these infusions on all rescued hypoglycemic cases?
  9. For Potential Scenario patient, it may be useful to briefly discuss how troubling the extravasation of D50 is now that it has occurred. Under suggested approach consider including a couple of evidence-based bullet points: arm elevation, compresses, or the use of hyaluronidase. This always causes confusion: hot or cold compress, hyaluronidase or not).

Thank you again for this very pragmatic post. Please feel free to contact me for any clarifications.

Zlatan Coralic, PharmD
Emergency Medicine Clinical Pharmacist
University of California, San Fransisco

Hi Dr. Coralic,
Thank you for your extensive feedback; this is much appreciated and will further enhance the quality of my post.

  • I did change the title. While perhaps controversial, I think the term mythbusting would still work, as I have built the argument for why D50 may not be preferred in all scenarios of severe hypoglycemia. Also, as explained below, I believe administration of D10 is just as feasible as appropriate administration of D50, given certain process changes are established. Nevertheless, I do like your suggestion, and agree with its appropriately neutral stance for a title.
  • I agree with your statement about rates of rebound hypoglycemia between D50 and D10 in the small study, and have added a short statement clarifying this. I would add, however, that there is some evidence to suggest that D10 may not result in as much or as pronounced rebound hypoglycemia (see reference 6) despite the same rates seen in this small study.
  • I did remove the statement, “undoubtedly overkill for treating any degree of hypoglycemia\", since the sentence doesn\'t add nor detract from the overall message.
  • The statement, “recent literature among the critically ill\", was revised as suggested, since this is more appropriate disclosure. I would add, however, that this is not the only literature suggesting increased glycemic variability is a more important predictor of diabetic morbidity/mortality than the magnitude of glycemic excursions.
  • The statement, “It is well known that administration of hypertonic fluids is not without the additional risk of causing harm to surrounding tissues, particularly after overzealous administration”, was revised as suggested.
  • With regard to the feasibility of administering D10 relative to D50, this was actually a relatively common practice in the Emergency Department where I completed my residency training. First off, while I do acknowledge that departmental education, and perhaps, various logistical and inventory considerations/changes may be required on a system level, expedient administration of D10 is absolutely feasible. Among the institutions in which I have practiced, they have carried at least the 250 and 500 mL bags of D10. Specifically, bags of 250 mL D10 were available in our stock areas of the ED, as well as within IV line carts right outside the patient rooms. Also, in my experience, infusion pumps have not been required, only primary tubing, which is also in the IV line cart. Nurses simply spike the bag and hang like regular fluids, but utilize the graduation marks present on the bags when infusing less than the entire volume. They would then clamp the tubing and check the fingerstick. If more dextrose is needed, they unclamp and finish infusing the rest of the bag. Clearly, for this method I would not condone the use of larger volume bags of D10. With regard to safety and rate of administration, again the time it takes to infuse 250mL of D10 is about the same as administering D50 appropriately, yet volumes less than 250 mL are likely more than sufficient, given the literature. All together, infusing such small volumes of D10 by gravity is likely inconsequential.
  • Good point. I have not seen any literature regarding the use of D25 (infant dextrose). I certainly think that this approach would be just as convenient as pushing D50, while theoretically attenuating the risks inherent with D50. Because I would like to keep the post focused and concise (I\'m already approaching the suggested word limit), I have chosen not to elaborate on this approach, but again certainly an area I would advocate for further study.
  • With regard to the need for subsequent dextrose infusions, this should be considered on a case by case basis, given the patient\'s previous history, amount of dextrose required for initial rescue, etc. As mentioned above, while the small study [1] did show equal rates of subsequent hypoglycemia between D10 and D50, there is other evidence that supports lower rates of rebound hypoglycemia with D10 [6]. Given this, patients may not necessarily require subsequent dextrose-containing fluids after receiving D10 for the initial rescue. With that said, there are no official recommendations I am aware of that advocate for subsequent D10 infusions.
  • I agree that the complications of extravasation (particularly with hyperosmolar substances) are a relatively infrequent, complicated and confusing situation to manage for many practitioners. Because of this, I feel that this topic would be deserving of an entirely separate post, and I’m not sure I could do it justice in just a few lines.]
Adam Spaulding, PharmD BCPS
Emergency Medicine Pharmacist
Pharmacy Residency Program Director
Waterbury Hospital Health Center
Adjunct Assistant Professor - UCONN School of Pharmacy
Contributor to Emergency Medicine PharmD Blog
Adam Spaulding, PharmD BCPS

Adam Spaulding, PharmD BCPS

Emergency Medicine Pharmacist,
Pharmacy Residency Program Director,
Waterbury Hospital Health Center,
Adjunct Assistant Professor - UCONN School of Pharmacy,
Contributor to Emergency Medicine PharmD Blog
Adam Spaulding, PharmD BCPS

Latest posts by Adam Spaulding, PharmD BCPS (see all)

  • Meghan Groth

    Thanks for the post Adam! I’ve had a number of providers ask me about this topic recently and the information you’ve presented will be very helpful in determining safe and effective care, especially for the pre-hospital setting (and in an era of ever-present drug shortages).

    • Adam M. Spaulding

      Thank you for the comment, Meghan. Although often overlooked, the administration of D10 for severe hypoglycemia in the ER is feasible and is something I have had experience with. As you mentioned, my initial experience came as the result of a D50 shortage, where we were forced to utilize D10. I would reiterate, however, that the feasibility of utilizing D10, particularly in these emergent situations, is contingent upon both systematic process changes and effective department-wide education (i.e. it needs to be readily available, a standardized process for administration needs to be developed, and all staff need to be educated). In addition to the post, additional administration logistics/details were also addressed in the Expert Peer Review critiques at the bottom.

  • PreHospital FOAM

    Fantastic article! I’m curious what you would recommend as an approach in an EMS system where we have out-of-box D50 with 500mL and 1000mL bags of normal saline. D10 is in our drug box, but many won’t break into it if it can be avoided. Most providers tend to push D50 intermittently with NS running–would this produce the same benefit of using a bag of D10, or would mixing the D50 in a 500 bag be a better option?

    • Adam M. Spaulding

      Hello. Thank you for the great question! However, I’m not sure if I could offer a blanket recommendation for one method over the other. I would take patient-specific factors into consideration, such as difficulty of obtaining access and quality of the line placed. For those difficult patients, where low quality access is obtained, the higher osmolar D50 would be a riskier option. Pushing D50 concurrently through a line with NS may indeed mitigate some of the risk of infiltration, though it does not address the fact that 25g of dextrose seems to repeatedly demonstrate in the literature to over-shoot glycemic targets (albeit of little concern during acute resuscitation). Just curious, why do most of your colleagues actively avoid D10? A few EMS services, one of which I have referenced, have actually resorted to D10 as their preferred option for treating hypoglycemia in the field. Small volumes (100-200mL) of D10 have been shown to be adequate to reverse severe hypoglycemia without over-shooting, is feasible and can be administered quickly, and confers a lower risk of infiltration via peripheral access.

      • PreHospital FOAM

        The unfortunate answer is that using D10 requires opening a drug-box, which then needs to be documented and replaced upon arrival at the hospital. D50 on the other hand is readily available to restock in our own supply room. It isn’t an excuse, but is the root cause. Also, there isn’t sufficient education in EMS (at least in northern New England) surrounding the issue–most providers don’t bat an eye at pushing a whole amp of D50 for the altered hypoglycemic patient.

        Thank you again for your time, and this is a resource I’ll use to pursue this issue within my own agency!

        • Emergentt

          As one of the authors on the EMS paper in Contra Costa County in California, it was initially met with confusion but we started it in part due to the D50 shortage. We now have 2 years of data with almost 2000 patients (analyzing now) and more and more EMS systems have started to use our protocol. (The neighboring Alameda County started Jan 1st.) There will be a session on D10 vs D50 at the NAEMSP meetings next week. Gene Hern

          • Adam M. Spaulding

            Thank you for your input. Your article was well done!

          • Emergentt

            I think we should eliminate d50 from all Code Carts in the hospital as well. Email me at emergentt@gmail.com. I have an idea about the next paper.

  • DP Smith

    Another EMS question for you. My recent case was an unresponsive, known diabetic pregnant (2nd trimester) woman with a glucose reading of 7. My protocols state administering an amp of D50, but in this case, I opted to mix D10 quickly using a 250mL bag of saline. Are you aware of any evidence to support best practices in regards to pregnant hypoglycmeic patients?

    • Adam M. Spaulding

      Hi DP Smith,
      Thank you for the comment. I do realize that most protocols continue to recommend D50 in the prehospital setting. I imagine that this determination is made, at least partially, with convenience in mind, along with an under-appreciation of the risks of D50. In your case, the fact that you alluded to mixing makes me wonder why you do not have 250mL bags of D10 available; this is something I would be an advocate for. Finally, I am not aware of any evidence to suggest that the management of the hypoglycemic pregnant patient should be any different than standard care. Thanks.

      • DP Smith

        Thanks for the reply. We don’t carry D10, some of us have advocated for it but have met with resistance from our management. In the meantime I’ll continue to mix my own, the science obviously supports it.