Presented by: Dr. Cassie Thomassin
Summarized by: Dr. Yonatan Yohannes
Reviewed by: Dr. Ian deSouza, Dr. Wendy Chan
The Fictitious Case
A 22 y/o F with PMH insulin-dependent DM and hypothyroidism is brought in by EMS for altered mental status. Family members on the scene noted she has been very sleepy, difficult to arouse, and confused once she wakes up. The patient is on levothyroxine and has an insulin-pump which has not functioned properly over the last 2 days. She complains of shortness of breath, abdominal pain, and nausea. Vitals on arrival show afebrile, HR 155, BP 134/104, RR 27, O2sat 100% on room air, and fingerstick > 451. Exam significant for lethargy, oriented x 3, dry mucous membranes, lungs CTA, soft and nontender abdomen, RLQ insulin pump.
Initial ECG shows tachycardia with diffuse peaked T-waves. Labs were remarkable for pH 6.87, potassium of 9, bicarbonate of 2, glucose of 1920, anion gap of 39, moderate serum acetone, and serum osmolality of 373. The patient was given calcium gluconate, nebulized albuterol (high dose), and started on IV normal saline with an insulin bolus & drip.
Shortly after the initial assessment, the patient became progressively more bradycardic and hypotensive with HR dropping into the 40s-50s and SBP 50s-80s with decreased mentation. 1 ampule of sodium bicarbonate was given for severe academia with cardiovascular collapse. She was subsequently intubated for airway protection and a central line was placed to start a norepinephrine infusion. The patient was admitted to MICU for diabetic ketoacidosis (DKA).
The patient’s anion gap closed on hospital day 2. Subcutaneous insulin started, insulin drip discontinued, troponin trended down, norepinephrine drip tapered, and patient was extubated. The patient was transferred out of MICU on hospital day 3 and discharged home on hospital day 5.
Pathophysiology of DKA
The onset of DKA typically has a precipitating event which stimulates a stress and/or inflammatory response, such as infection (commonly urinary tract infection or pneumonia), acute organ ischemia (myocardial infarction or stroke), GI bleeding, pancreatitis, trauma/surgery, or as an adverse effect of a medication.
A common misperception is that DKA is caused by hyperglycemia, when the true driving mechanism is lack of insulin. They may sound as if they are one in the same, however hyperglycemia is a result of DKA, not the cause of it. Insulin deficiency while the body is in an inflammatory state causes an increase in counterregulatory hormones (catecholamines, cortisol, glucagon, growth hormone) contributing to hyperglycemia and ketogenesis. The mechanism of hyperglycemia is a combination of impaired glucose uptake, gluconeogenesis, and accelerated glycogenolysis. Inability to utilize glucose for energy production leads to increased lipolysis and the release of free fatty acids. Free fatty acids are then metabolized to ketone bodies (acetoacetate and beta-hydroxybutyrate) for energy production, thus causing a metabolic acidosis. Hyperglycemia, if unchecked, leads to osmotic diuresis, electrolyte losses, dehydration, hyperosmolarity, and a drop in glomerular filtration rate.
Serum potassium may appear normal or elevated primarily due to decreased intracellular uptake caused by insulin deficiency and hyperosmolarity, in addition to extracellular potassium shifting via the H+/K+ pump to compensate for acidosis. These potassium shifts result in large amounts of potassium loss due to osmotic diuresis. Therefore, total body potassium is usually depleted.
Difference between DKA and HHS (Hyperosmolar Hyperglycemic State)
Patients with HHS have a combination of insulin resistance with some component of insulin deficiency. However, most patients still secrete enough insulin to prevent lipolysis and ketogenesis even if it is not enough to prevent hyperglycemia. Therefore the hyperglycemia in HHS is usually more profound with a relatively normal pH and anion gap. Due to the significant elevation in plasma osmolality in HHS, it is usually associated with a greater depression of mental status, although in severe DKA a similar change in mental status may also be seen.
Diagnostic Criteria
| DKA | ||||
| Mild | Moderate | Severe | HHS | |
| Plasma glucose | >250mg/dl | >250mg/dl | >250mg/dl | >600mg/dl |
| pH | 7.25-7.30 | 7.00-7.25 | <7.00 | >7.30 |
| Serum bicarb | 15-18 | 10-15 | <10 | >15 |
| Serum ketones | Positive | Positive | Positive | Small |
| Urine ketones | Positive | Positive | Positive | Small |
| Anion gap | >10 | >12 | >12 | >320mOsm/kg |
| Osmolality | Variable | Variable | Variable | <12 |
| Mental status | Alert | Alert/drowsy | Stupor/coma | Stupor/coma |
*chart edited from Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crisis in adult patiens with diabetes. Diabetes Care. 2009. 32; 7:1335-1343.
Management
Hydrate, check potassium, give insulin, re-check potassium every 2 hours. Historically insulin is given as both a bolus and drip, however literature suggests there is no significant difference in the drop in serum glucose and correction of pH when insulin drip is started without a bolus in DKA.
IV fluids
The IV fluid historically administered and currently recommended in most guidelines is normal saline (NS). However, after large amounts of normal saline there can be a dramatic increase in the serum chloride (Cl) level, with an initial decline in serum bicarbonate (HCO3) due to a hyperchloremic metabolic acidosis. Resuscitation with more balanced solutions, such as Ringer’s lactate (RL) or Plasmalyte, are thought to avoid this adverse effect of NS.
Van Zyl, et al compared RL vs NS in DKA patients and found no difference in time to normalization of pH, however the RL group took significantly longer for resolution of hyperglycemia.
On the other hand, Chua et al compared Plasmalyte to NS in resuscitating DKA patients and found a significant increase in bicarbonate level at 12 hours in the Plasmalyte group compared to the NS group, as well as avoidance of the marked increase in chloride levels. However, even with the increased bicarbonate levels at 12hours the rate of change of pH between the two groups was not found to be statistically significant.
While NS may not be the ideal fluid solution for resuscitating DKA patients, there does not seem to be a difference in overall outcomes when compared to more balanced solutions. Resuscitation of DKA patients may begin with 1-2 liters NS then transition to 0.45% saline (to replace calculated free water deficit) or one of the more balanced solutions.
Insulin administration
Continuous infusion of IV insulin is a cornerstone of DKA management. Over the past decade, a second option for insulin delivery has emerged consisting of subcutaneous rapid-acting insulin. Subcutaneous lispro has been shown to be equal in safety and efficacy to IV regular insulin in uncomplicated DKA. No difference was shown in rate of correction of hyperglycemia, amount of insulin delivered, events of hypoglycemia, or length of stay in hospital. Without the need for continuous IV insulin, patients may theoretically be managed on a well-staffed medicine floor, avoiding a costly ICU admission.
When comparing patients who have received IV insulin bolus followed by drip versus IV drip alone, the insulin bolus has not been associated with any significant benefit to adult DKA patients (it is not recommended in pediatrics for concern of increasing risk of cerebral edema). An insulin bolus was found to have a larger decline in serum glucose within the first 1-2 hours of treatment that was not sustained, showing no difference in time to achieve normoglycemia, time spent of insulin drip, or rate of change in pH, when compared to IV insulin alone.
Mechanical ventilation
The goal here is to avoid intubation as long as possible. It is difficult to match the physiologic minute ventilation of patients in DKA or any severe metabolic acidosis with mechanical ventilation. If intubation is indicated, usually for profound altered mental status and airway protection, the goal is to set your initial vent settings with a respiratory rate that matches the patient’s pre-intubation respiratory rate (usually 25-30) and adjust your respiratory rate based on subsequent blood gas pH to resolve the acidemia. Deep sedation and neuromuscular blockade with a non-depolarizing agent may be needed to prevent patient-ventilator dyssynchrony at such high respiratory rates.
Sodium Bicarbonate
The primary concern with severe acidemia is the deleterious hemodynamic effect, including decreased cardiac contractility and catecholamine resistance, resulting in decreased cardiac output, hypotension, and predisposing to dysrhythmia. The American Diabetes Association recommends giving bicarbonate for pH < 6.9 to prevent cardiovascular collapse. However, sodium bicarbonate therapy is associated with worsening academia, decreasing cerebral pH, hypokalemia, and in pediatrics, an increased risk of cerebral edema. Remember this equilibrium H2O + CO2 <-> H2CO3 <-> H+ + HCO3? Therefore giving bicarbonate can shift this equilibrium to the left, producing more CO2 and worsening intracellular acidosis (since CO2 is lipid soluble and moves freely across cell membranes). If someone is breathing normally, they can increase their physiologic minute ventilation to expel the additional CO2 over time. However, patients in severe metabolic acidosis are likely already breathing at their maximum physiologic respiratory rate and are unable to increase their minute ventilation much further on their own. There are numerous studies that have shown sodium bicarbonate has no benefit in DKA, regardless of pH.
Complications
Cerebral edema
Mostly seen in children and adolescents. Symptoms usually develop 12-24 hours after onset of treatment. The first sign is headache, followed by lethargy and decreased alertness. Symptoms may progress rapidly, before papilledema can develop. Best way to avoid this complication is by replacing sodium and free water deficit gradually, especially in hyperosmolar patients. Monitor mental status throughout course of treatment.
Pulmonary edema
Noncardiogenic pulmonary edema is a rare complication thought to be due to a combination of increased permeability of pulmonary capillary membranes and alteration of hydrostatic & oncotic forces due to aggressive fluid resuscitation.
Hypokalemia
If K+ is < 3.3mEq/L, replete with IV KCl prior to starting insulin. IV insulin can be started once K+ is > 3.3mEq/L. If K+ is between 3.3-5.3mEq/L, then supplement each liter of IV fluid with 20-30mEq of KCl. If K+ is > 5.3mEq/L, then no supplementation necessary. Monitor K+ levels every 2-4 hours until stable and supplement as needed.
Hypoglycemia
Completely avoidable. Serial fingersticks and titrate insulin drip accordingly. Once serum glucose is 200-250, change IV fluid to D5 with 0.45%NS. Administration of subcutaneous insulin should start 2 hours before stopping IV insulin drip. Many ICUs have their own guidelines for titration of an insulin drip.
Pitfalls
-Overaggressive fluid resuscitation resulting in pulmonary edema or cerebral edema
-Hyperchloremic metabolic acidosis from administering large amounts of normal saline
-Starting insulin before checking potassium
Conclusion
-Always check potassium before giving insulin IV
-Manage DKA with fluids, insulin, and potassium supplementation as needed, but don’t forget to treat the underlying cause of DKA!
-If mechanical ventilation is needed, initial vent settings should match the patient’s pre-intubation respiratory rate
-Uncomplicated DKA may be an avoidable ICU admission if your medicine floor is well-staffed and nursing staff can deliver subcutaneous insulin every 1-2 hours, in addition to monitoring electrolytes and fingersticks
References
Gosmanov AR, Gosmanova EO, Dillard-Cannon E. Management of adult diabetic ketoacidosis. Diabetes Metab Syndr Obes. 2014. 7:255-264.
Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN. Hyperglycemic crisis in adult patiens with diabetes. Diabetes Care. 2009. 32; 7:1335-1343.
Van Zyl DG, Rheeder P, Delport E. Fluid management in diabetic-acidosis—Ringer’s lactate versus normal saline: a randomized controlled trial. QJM. 2012. 105(4):337-343.
Chua HR, Venkatesh B, Stachowski E, Schneider AG, Perkins K, Ladanyi S, Kruger P, Bellomo. Plasma-Lyte 148 vs 0.9% saline for fluid resuscitation in diabetic ketoacidosis. J Crit Care. 2012. 27(2):138-145.
Umpierrez GE, Latif K, Stoever J, Cuervo R, Park L, Freire A, Kitabchi AE. Efficacy of subcutaneous insulin lispro versus continuous intravenous insulin for the treatment of patients with diabetic ketoacidosis. Am J Med. 2004. 117(5):291-296.
Goyal N, Miller JB, Sankey SS, Mossallam U. Utility of initial bolus insulin in the treatment of diabetic ketoacidosis. J Emerg Med. 2010. 38(4):422-427.
Sprung CL, Rackow EC, Fein IA. Pulmonary edema; a complication of diabetic ketoacidosis. Chest. 1980. 77(5):687-688.
Manthous CA. Avoiding circulatory complications during endotracheal intubation and initiation of positive pressure ventilation. J Emerg Med. 2010;38(5):622-631.
Chua HR, Schneider A, Bellomo R. Bicarbonate in diabetic ketoacidosis – a systematic review. Ann Intensive Care. 2011.1(1):23.
Glaser N, Barnett P, McCaslin I, et al. Risk factors for cerebral edema in children with diabetic ketoacidosis. N Engl J Med. 2001.344:264-269.
Viallon A, Zeni F, Lafond P, Venet C, Tardy B, Page Y, Bertrand JC. Does bicarbonate therapy improve the management of severe diabetic ketoacidosis? Crit Care Med. 1999.27(12):2690-2693.
Green SM, Rothrock SG, Ho JD, Gallant RD, Borger R, Thomas TL, Zimmerman GJ. Failure of adjunctive bicarbonate to improve outcome in severe pediatric diabetic ketoacidosis. Ann Emerg Med. 1998.(1):41-48.
yon.yohannes
Latest posts by yon.yohannes (see all)
- The Critically Ill “Tox Patient” - June 2, 2016
- Shoulder Dislocations in Austere Environments – A Novel Approach - May 12, 2016
- Management of Atrial Flutter with Rapid AV Conduction - March 16, 2016
0 Comments