Diabetic ketoacidosis and hyperglycaemic hyperosmolar syndrome – clinical guidelines
Karen De Beer, Sindhu Michael, Meera Thacker, Elizabeth Wynne, Caroline Pattni, Mandy Gomm, Carol Ball, Dominic Walsh, Andrew Thomlinson and Kevin Ullah
ABSTRACT Background: The aim of this study was to establish a standardized approach to the initial care of patients with diabetic ketoacidosis (DKA) and hyperglycaemic hyperosmolar syndrome (HHS). DKA and HHS are metabolic emergencies. Effective and efficient management is the responsibility of the multidisciplinary team. The admission of patients to the intensive care unit (ICU) with DKA and HHS is rare, and management of patients’ diverse problems is prone to error because of a lack of familiarity. Aim: The paper’s aim is to set the developmental process of a clinical guideline following a review of the literature. Discussion: This clinical guideline is based on a review of the evidence available within the literature in the early phase of resuscitation. Collaborative working among the multidisciplinary team through clinical practice group was the method adopted. Management of DKA and HHS is divided into three main areas: intravenous fluid replacement, insulin therapy and electrolyte management. The controversy associated with the administration of sodium bicarbonate is discussed. Conclusion: Effective treatment requires a rapid initial assessment of the patient based on current medical history and clinical presentation. To this end, a quick reference algorithm and guide to management were also developed. Key criteria for evaluating the effectiveness of treatment are provided and complications of treatment are addressed. The formation of the practice development group that led to this innovation is outlined, and in conclusion, the success of the group is reflected upon.
Key words: Clinical guidelines • Diabetic ketoacidosis • Hyperglycaemic hyperosmolar syndrome
Admission to hospital for the treatment of diabetic ketoacidotic coma has been estimated at 26–44 per million people living in the UK (Williams, 1989). These are the most recent figures available. Current methods of coding for admission are confusing as similar codes are used to identify admission for hypoglycaemic coma (Home et al., 1999). Therefore, the true incidence
Authors: K De Beer, RN, Staff Nurse, ICU, Royal Free Hampstead NHS Trust, London, UK; S Michael, RN, Staff Nurse, ICU, Royal Free Hampstead NHS Trust, London, UK; M Thacker, BSc (Hons), MRPharmS, DipClinPharm, ICU Pharmacist, ICU, Royal Free Hampstead NHS Trust, London, UK; E Wynne, RN, Sister, ICU, Royal Free Hampstead NHS Trust, London, UK; C Pattni, RN, Staff Nurse, ICU, Royal Free Hampstead NHS Trust London, UK; M Gomm, RN, Sister, ICU, Royal Free Hampstead NHS Trust, London, UK; C Ball, RN, MSc, PhD, Consultant Nurse, ICU, Royal Free Hampstead NHS Trust, Senior Research Fellow, City University, London, UK; D Walsh, RN, Staff Nurse, ICU, Royal Free Hampstead NHS Trust, London, UK; A Thomlinson, RN, Staff Nurse, ICU, Royal Free Hampstead NHS Trust, London, UK; K Ullah, RN, Charge Nurse, Royal Free Hampstead NHS Trust, London, UK Address for correspondence: C Ball, ICU, Royal Free Hampstead NHS Trust, Pond Street, London NW3 2QY, UK E-mail: firstname.lastname@example.org
of diabetic ketoacidosis (DKA) and hyperglycaemic hyperosmolar syndrome (HHS) is unknown in the UK. The admission of patients to the intensive care unit (ICU) for the management of DKA is rare. The total number of patients admitted over the past year to our ICU was two. The incidence of HHS (previously hyperosmolar non-ketotic coma) is even more infrequent. This entails considerable risk because the multidisciplinary team cannot gain in-depth experience, and competence is difficult to maintain. Specific areas of concern identified by members of the ICU team were associated with the safety of commencing an insulin infusion when the potassium level was low, the speed at which the blood glucose should be lowered, when point of care (POC) testing was safe and laboratory analysis was no longer required, and the appropriate concentration of saline in treating dehydration.
CLINICAL GUIDELINE DEVELOPMENT – THE PROCESS
As a result of these concerns, the practice development group decided to produce clinical guidelines. The
ª 2008 The Authors. Journal Compilation ª 2008 British Association of Critical Care Nurses, Nursing in Critical Care 2008 • Vol 13 No 1
5 – 5. Give 40 mmol potassium per litre of fluid i. fluids and full monitoring. a rapid physical assessment is required using an ABCDE approach familiar to all members of the ICU team.45% saline 4 – 14ml/kg/h
0. They are then revised as appropriate. This self selection results in the formation of the group which will take the topic area forward.
ª 2008 The Authors. Other members of the multidisciplinary team are co-opted as appropriate. Particular attention should be given to any difficulty the patient may have in maintaining their airway and the presence of Kussmaul breathing. Hypotension (<90 mmHg) and tachycardia (>125 b/min).3 mmol/L do not commence insulin. It is important to identify the recent occurrence of prodromal illness. Complete initial assessment: start i.v. On completion of the project. Recheck electrolytes hourly Unless anuric
135 – 155
<135 20–40 mmol/L potassium if K+ 3. the clinical guideline is added to the bedside clinical practice folder and placed on the local ICU intranet for easy reference and support. pathophysiology was reviewed by the pharmacist and a nurse. Fluids Determine hydration status Hypovolaemic shock systolic <90 mmHg heart rate > 100 Colloid/ 0.DKA and hyperglycaemic hyperosmolar syndrome
practice development group is formed from a team of nurses who then choose if they wish to participate in the development of a chosen topic.
The aim of the practice development group in this case was to provide a clinical guideline for the initial management of DKA and HSS at the bedside (Figure 1 and Table 1) and supply an in-depth explanation of the rationale supporting this for general reference within the ICU. Intubation and ventilation may be required. The restoration of these to normal values can aid the evaluation of treatment delivered. Exposure of the patient
i. The ICU pharmacist is a permanent member of the group.
THE CLINICAL GUIDELINE – THE EVIDENCE BASE Assessment
This may require two members of the ICU team. A maximum of 10 h is made available to group members to work on the project. until K+ 3. in this emergency situation. The group then met on a regular basis over a 6-month period to review the work undertaken and guide the development of the guideline.5mmol/L
If blood glucose falls by less than 3 mmol/L in first hour Double insulin dose When blood glucose reaches 12 – 14 mmol/L Halve insulin rate
0. Each member of the group chose a particular dimension to work on. which can then be claimed back from the succeeding on-duty rotas.5 mmol/L do not give potassium.v. For example.9% saline 1 L/h to restore BP and HR Calculate corrected sodium
Insulin Initial rate Start i. However. infection or non-compliance with the insulin regime. Measure blood sugar hourly until within range of 10–12 mmol/L.9% saline 10 – 20ml/kg/h
Serum osmolality should not decrease by > 3 mOsmol/kg H2O/h When blood glucose reaches 12 – 14 mmol/L Change to glucose saline or 5% glucose 100 – 125 ml/h Maintain higher fluid volume until lactate is normal
Titrate insulin to maintain blood glucose of 4–10 mmol/L until DKA resolved and serum osmolality <320 mOsmol/L
See quick reference guide to evaluate effectiveness of treatment
Figure 1 Management of adult diabetic ketoacidosis and hyperglycaemic hyperosmolar syndrome. fluid and electrolyte management by individual members of the group with special responsibility being taken for identification of current algorithms for the management of DKA and HHS. insulin at 0.1 iu/kg/h Check blood glucose hourly
Blood glucose should not fall by more than 4 mmol/L per hour
If serum K+ <3. It also prevents the same people always being responsible for practice development. Once agreed. the clinical guidelines are presented to senior nurses and consultant medical staff for approval. cool peripheries and a reduced level of consciousness all indicate the presence of hypovolaemic shock. Journal Compilation ª 2008 British Association of Critical Care Nurses
.5 mmol/L If serum K+ >5.v.v.
If hypovolaemic shock is present (systolic blood pressure <100 mmHg) (English and Williams. hyperglycaemia – blood glucose>35 mmol/L Hyperosmolality – calculated serum osmolality >320 mmol/L Absent ketones pH > 7Á2 HCO3 >15 mmol/L Deteriorating conscious level (seizure)
is necessary to identify dehydration and potential causes of DKA or HSS such as infection. phenytoin. 350–500 mmol of phosphate and 200– 350 mmol of chloride are likely in DKA (English and Williams. 1999) (Appendix). The American Diabetes Association (2003) suggests that the serum osmolality should not decrease by >3 mOsmol/kg H2O/h (Appendix). corticosteroids Peritoneal dialysis/haemodialysis Polyuria
Signs and symptoms Polyuria Polydipsia Nausea/vomiting/abdominal pain Diagnostic criteria Predominant feature – ketoacidosis pH < 7Á3 Serum bicarbonate <15 mmol/L Blood glucose >13Á9 mmol/L Urinary ketones Deteriorating conscious level
Predominant feature – hyperosmolality. 2004). 2000) and 9 L in a patient with HHS (Kitabchi et al... (2001) is 10–20 mL/kg/h and the American Diabetes Association (2003) is 15–20 mL/kg/h. renal and mental status is advised by the American Diabetes Association (2003) to avoid iatrogenic fluid
ª 2008 The Authors.. There would be an estimated fluid loss of 6–10 L in a patient with DKA (Hand. The recommended volume of 0Á9% NaCl by Kitabchi et al. Journal Compilation ª 2008 British Association of Critical Care Nurses
. 2001) with an associated raised serum osmolality of 320– 340 mOsmol/L (English and Williams. Savage et al. The dominant feature of HHS is hyperosmolarity (>320 mmol/L) (Kitabchi et al.. 2004). then intravenous colloid should be considered (Savage et al. 2000). 2004). In mild to moderate depletion..DKA and hyperglycaemic hyperosmolar syndrome
Table 1 Management of adult diabetic ketoacidosis (DKA) and hyperosmolar hyperglycaemic syndrome (HHS) Rapid initial assessment DKA History Type 1 diabetic Non-compliance with diabetes management Prodromal illness (days) Possible infection Weight loss HHS
Type 2 diabetic Inability to take fluid – elderly patients Prodromal illness (weeks) Possible complication of: Parenteral feeding Drug therapy – B-adrenergic blockers. The choice of replacement fluid should consider factors
such as age. potassium and chloride can also be expected in HHS (Kitabchi et al. 2004). 2001). 1999. 2001). Frequent assessment of cardiac. 200–500 mmol of potassium. A reduced rate of intravenous fluid replacement is recommended in the elderly. 2001). 1999).
Management of DKA and HHS Intravenous fluid replacement
Dehydration and sodium depletion develop as a result of the osmotic diuresis.g... Sudden falls in serum sodium can cause cerebral oedema. An estimate of true sodium should be made because of the degree of hyperglycaemia. cardiac disease (Hand. 2003). 2004). Isotonic saline (0Á9% NaCl) is seen as the initial fluid of choice by most authors (Charalambos. The goal is to replace half of the fluid deficit over the first 8 h and the remaining fluid should be replaced over the next 16 h (Brenner. The main objective in the treatment of DKA and HHS is to correct water and electrolyte deficits over the first 24–48 h to replete extracellular fluid volume and restore intravascular volume (Kitabchi et al. Incorrect use of resuscitation fluid can cause an increase in plasma sodium levels and further increase plasma osmolality. Miller. 2004). those with cardiac disease and mild DKA (Savage et al. Electrolyte deficits of 500–700 mmol of sodium. diuretics. which has been associated with pontine myelinolysis (Keays. Larger deficits of especially sodium. e. degree of hydration and patient history. rates of 7 mL/kg/h have also shown the desired effect (Kitabchi and Wall. 1999. Glucose induces water movement from the intracellular to extracellular space and dilutes the sodium concentration (Miller. 2006).
. Some patients with DKA initially present with hypokalaemia and insulin administration may decrease potassium levels even further. 1999).DKA and hyperglycaemic hyperosmolar syndrome
overload. although subcutaneous methods are now cited in the management of uncomplicated DKA (Umpierrez et al. Insulin administration seeks to restore normal glucose uptake by cells. Savage et al. 2004).
Management of electrolytes Potassium
Hyperosmolality causes a shift of potassium from within the cells to the extracellular space. The American Diabetes Association (2003) recommends a rate of 4–14 mL/kg/h of 0Á45% NaCl if the corrected sodium is normal or elevated. If the potassium level is below 3Á3 mmol/L. (2003) suggest that the aim should be to reduce blood glucose by 2–3 mmol/L/h. A similar suggestion was made by English and Williams (2004) and 5% glucose can be started at 100–125 mL/h. If this is not achieved. thus avoiding complications such as cerebral oedema (Brenner. The serum potassium should be corrected to 3Á5 mmol/L before insulin is commenced (Keays. 2006). Potassium levels may decline as
It is essential to obtain a serum potassium level before insulin is started. As the cells are rehydrated and insulin therapy is commenced. This is then lost because of the osmotic diuresis.. Literature strongly suggests that intravenous infusion is the best method of delivery in the acute management of DKA (Miller. there are no data to support any advantage in giving an initial bolus. 1999). (2003) claim that continuous intravenous infusion is the most effective as all other routes are unreliable in the volume-depleted patient where erratic absorption can occur with intramuscular or subcutaneous administration. but this is not generally exceeded (English and Williams. Patients with an initially low serum potassium (<3Á3 mmol/L) are at risk of cardiac arrhythmias and muscle weakness. NaCl 0Á9% can be continued at a slower rate to complete rehydration and electrolyte replacement. Currently. 2003).. However. 2006). Further decisions should then be based on the patient’s potassium levels and urine output. 2004b). Savage et al. Ten percent glucose can be given if less fluid is required (English. insulin should not be commenced. If levels are between 3Á5 and 5Á5 mmol/L. When the blood glucose reaches 12– 14 mmol/L. excessive insulin administration and associated hypokalaemia and hypophosphataemia must be avoided. the insulin rate may be reduced by half and 5% glucose or glucose saline replaces 0Á9% or 0Á45% NaCl.g. but obese patients with type 2 diabetes might require larger doses to reduce hyperglycaemia. resulting in muscular weakness. 0Á9% NaCl should be replaced with 0Á45% NaCl (Savage et al. 20– 40 mmol of potassium can be given initially in the first hour of treatment. However. English and Williams (2003) recommend 40 mmol potassium/L of saline until the serum potassium reaches 3Á5 mmol/L. Chiasson et al. It is important that the serum glucose is not lowered too rapidly. Journal Compilation ª 2008 British Association of Critical Care Nurses
. Patients with HHS are very sensitive to insulin. 2004) (Appendix). The initial recommended starting dose is 0Á1 unit insulin/kg/h (Kitabchi et al. 2004). 2003) as the use of insulin and saline will further decrease extracellular potassium levels (Morton et al. If the corrected serum sodium rises above 155 mmol/L. (2004) recommended 1 L of 5% glucose over 8 h. Administration of 5% glucose can be continued until ketonaemia is controlled and iatrogenic hypoglycaemia is avoided (Kitabchi and Wall. Serum potassium levels may also be high and potassium therapy should be omitted if blood levels exceed 5Á5 mmol/L (Keays.
often involve small sample sizes (e. 2001). 2004a.. Authors disagree on the absolute blood glucose value (Table 2).. 1999).. 2006). 2004). Electrocardiograph monitoring is essential at this time. n = 40) and intermediate care units are not always available in the UK. Caution should be taken if blood pressure stability is not achieved after 2 h of intravenous fluid administration (Kitabchi et al. the insulin dose should be doubled every hour until a decrease of 3–4 mmol/L/h is achieved. This depletion can be more severe in HHS than in DKA. Starting potassium with an unknown level and compromised renal function can be fatal (Morton et al. 2001).. (2004) Chiasson et al. Initial potassium levels may vary according to the individual. The initial fluid choice of 0Á9% or 0Á45% saline is followed by 5% glucose or glucose saline once the blood glucose levels decrease below 13 mmol/L (Kitabchi et al. The administration of hypotonic saline (0Á45% NaCl) is similar to the fluid loss during osmotic diuresis and would gradually replenish both intracellular and extracellular compartments (Kitabchi and Wall.. Chiasson et al. these studies
Table 2 Suggested glucose values Blood glucose 13Á9 mmol <14 mmol/L <15 mmol/L 12–14 mmol/L Author Kitabchi and Wall (1999) English (2004) Savage et al. this shift is reversed. (2003)
ª 2008 The Authors. 2001.
On an hourly basis. This may differ according to POC testing in different ICUs. The clinical guideline developed for use at the bedside. The final aim of management was to attain a blood glucose level of 10 mmol/L. 2003). Below a pH of 6Á9. Journal Compilation ª 2008 British Association of Critical Care Nurses
Phosphate levels are normal or elevated on initial presentation of DKA (American Diabetes Association. These include increased CO2 production. particularly potassium. fluid replacement may change from saline to 5% glucose or glucose/saline. 2003). most authorities would recommend the use of bicarbonate to correct the pH to a threshold of 6Á9–7Á15. To precipitate a fall in blood glucose may result in fluid shifts between the intravascular and extravascular space and further electrolyte disturbance.
Implementation and evaluation Reflection on the process of practice development for DKA and HHS
This method of practice development worked well. Renal function should be assessed on a daily basis to ensure renal function returns to normal (Table 3). The aim was to maintain serum potassium >4 mmol/L. 20– 30 mmol of potassium could be given in each litre of intravenous fluid (English and Williams. assessment should be made of the corrected sodium level in order to ascertain the need for hypotonic saline. The enthusiasm of the group was maintained because
Table 3 Evaluating the effectiveness of treatment Blood glucose Monitor blood glucose hourly via biochemistry if over 27 mmol/L Reduce no faster than 3 mmol/L/h Initial aim blood glucose – 12–14 mmol/L When initial aim achieved change fluid replacement to glucose 5% or glucose saline Halve insulin rate Final aim blood glucose <10 mmol/L Corrected Na+ 135–145 mmol/L Corrected anion gap = 10–17 Potassium 3Á5–4Á5 mmol/L Serum osmolality – do not reduce by more than 3 mOsmol/kg H2O/h: aim – serum osmolality 285–295 mOsmol/L Phosphate. blood glucose should be monitored hourly via biochemistry until levels can be accurately analysed by the blood gas analyser. Excessive replacement of phosphate can lead to hypocalcaemia. The administration of sodium bicarbonate is associated with side effects that may preclude any potential benefits. Hypophosphataemia should be carefully corrected as it can result in muscular weakness and neurological dysfunction. although this threshold has been disputed. 2004). this should be clarified at an individual hospital level. Phosphate. thus if serum potassium is >3Á5–5Á0 mmol/L. Insulin therapy lowers serum phosphate levels. However.
manner. 2003). acidosis of cerebrospinal fluid. Once the blood glucose reaches 12–14 mmol/L. The same is also necessary for the measurement of potassium levels.
The use of sodium bicarbonate has been the subject of controversy. therefore. serum calcium levels should be also monitored (Keays.
Evaluating the effectiveness of treatment
The aims of treatment are to attain normal biochemical values as soon as possible in a safe and effective
ª 2008 The Authors. rebound alkalosis. 2003).
Chronic magnesium depletion can be present in uncontrolled diabetes and can contribute to insulin resistance. pH and serum osmolality. Research has found no benefit in the administration of sodium bicarbonate in DKA (Gehlbach and Schmidt. hypokalaemia. the blood gas analyser should not be used to monitor blood glucose levels until the level has fallen below this figure. is given in Figure 1 and Table 1. Therefore. calcium and magnesium four hourly – maintained within normal levels Test for urinary ketones four hourly (ketones may persist for up to 2 days after correction of acidosis) Renal function – creatinine 55–120 mmol/L
Obtaining glucose levels – biochemistry procedures
POC testing is inaccurate above 27 mmol/L. 2003). life-threatening hyperkalaemia is an undisputed indication for bicarbonate therapy (Keays. In the initial resuscitation period. magnesium and calcium levels should also be assessed four hourly together with urinary ketones. therefore. The insulin rate should be halved at this point to avoid potential hypoglycaemia. based on the evidence above.DKA and hyperglycaemic hyperosmolar syndrome
potassium returns to the intracellular space when insulin is administered and hydration is restored. research has yet to show any benefits in the administration of magnesium in diabetic emergencies (Keays. However. Urinary ketones may be raised for up to 2 days after correction of the acidosis. volume overload and altered tissue oxygenation (Keays.
Havrankova J. Belanger R. 8: 259–265.
WHAT IS KNOW ABOUT THIS TOPIC
• Most recent figures available indicate 26–44 per million people in the UK will be admitted to hospital for the treatment of DKA.g. The evaluation indicated that the ICU team responsible for managing the patient utilized the clinical guideline immediately to prescribe treatment and monitor its effectiveness. (2003). (2003). (2006). although there was some delay in achieving the goals set for completion during an exceptionally busy period on the ICU. 168: 859–866.
WHAT THIS PAPER ADDS
• DKA and HHS rarely require admission to the ICU. e. Since that time it has been used twice. Journal Compilation ª 2008 British Association of Critical Care Nurses
. Charalambos C. • The incidence of admission to ICU is unknown. (2004). This led to the amendment of the clinical guideline to include fluid replacement to titrate against serum lactate level (Figure 1). The aim of this
study was to both inform staff of the reasons why management was so multi-faceted. Fournier H. Acute diabetic emergencies and their management. insulin administration prior
The authors would like to thank the members of the practice development team who also contributed to the development of these clinical guidelines: Dr Julian Howard Consultant Intensivist. Postgraduate Medical Journal. 15: 132–134. Aris-Jilwan N. The utilization of consultant nurse and matronled ward rounds also aids the implementation and maintenance of the clinical guidelines.
• To reduce this risk.DKA and hyperglycaemic hyperosmolar syndrome
members had self selected and the guidelines were completed on time. Blood glucose levels and normal pH were established in a timely manner with the complete absence of adverse events. Management of hyperglycemic emergencies.
This method of practice development has thus far proved successful in the development and acceptance of clinical guidelines and innovations to practice on the ICU. The clinical resource file is particularly useful to junior medical staff and agency nurses. Areas that were the cause of most discussion were the rate at which the blood glucose level should fall and the calculation of corrected sodium. we developed a set of clinical guidelines based on the pathophysiology of these abnormalities. Of most note was the ability of nursing staff to maintain treatment goals when the ICU registrar had to attend a prolonged trauma call. Dominic Cox – Clinical Scientist and Peta Collins – Clinical Audit Manager. In this lies considerable clinical risk as the multidisciplinary team may have little
experience of the complex management required by this patient population. (1999). One deficit was identified in one patient.
ª 2008 The Authors. Diagnosis and treatment of diabetic ketoacidosis and the hyperglycemic hyperosmolar state. on both occasions at the weekend when the number of senior staff available for consultation is reduced. 17: 56–65. Gehlbach BK. Critical Care. This related to the reduction in fluid replacement as near normal blood glucose levels were achieved and as a result the serum lactate level failed to resolve. precipitate reduction in blood sugar. 26: S109–S117. Beauregard H. 80: 253–261. and produce a set of rapid review bedside guidelines to promote safe and effective management. Care of the Critically Ill. Diabetes Care. Schmidt GA.
to correction of potassium levels. Chiasson JL. Williams G. and potentially hazardous.
The clinical guideline
The clinical guideline was introduced in November 2006. The specific benefit to patients is the reduction of risk in relation to the management of forms of critical illness seen infrequently in our ICU and the promotion of best practice to which all members of the multidisciplinary team adhere. Hyperglycaemic crises and lactic acidosis in diabetes mellitus. English P. The process was evaluated by the consultant intensive care doctor. Bench-to-bedside review: treating acid-base abnormalities in the intensive care unit – the role of buffers. Each bedside has a clinical resource file to aid the multidisciplinary team in the management of critically ill patients and the DKA-HHS clinical guidelines have been added to this and placed on the local ICU intranet. Hyperglycemic crises in patients with diabetes mellitus. AACN Clinical Issues.
American Diabetes Association. which have been developed for the ICU. Canadian Medical Association Journal. Brenner ZR. (2004). Ekoe J-M. Bertrand S.
Stoever J. Management of diabetic ketoacidosis. Oxford: Butterworth-Heinemann. Williams DRR. Efficacy of subcutaneous insulin lispro versus continuous intravenous regular insulin for the treatment of patients with diabetic ketoacidosis. Journal Compilation ª 2008 British Association of Critical Care Nurses
. Barrett EJ. Oh’s Intensive Care Manual. 15: 47–52. Outcome indicators for diabetes services: what do we have and what do we need? Community Medicine. Mason A. (2006). Goldacre M. (2000). Management of hyperglycemic crisis in patients with diabetes. Malone JI. Wilkinson E (eds). Newell-Price J. Philadelphia: Lippincott. Weetman AP. 80: 506–515. 8th edn. Karabell A. (2004b).
APPENDIX – CALCULATIONS
Calculation 1: Corrected sodium calculation Calculate the true sodium by dividing the blood glucose by three and add this to the measured sodium Calculation 2 2 (Na+ + K+) + glucose + urea = serum osmolality
ª 2008 The Authors. Postgraduate Medical Journal. (1999). 11: 57–64. Latif K. 24: 131–153. (2003). (1989). 60: 455–464.DKA and hyperglycaemic hyperosmolar syndrome
Hand H. Freire AX. (eds). Keays R. Mah PM. American Journal of Medicine. Management of diabetic ketoacidosis. Diabetes Care. Umpierrez GE. Murphy MB. Kitabchi AE. (2001). (2004a). Gallo BM. Williams and Wilkins. Nursing Standard. Umpierrez GE. Treatment of diabetic ketoacidosis with subcutaneous insulin aspart. Critical Care Nursing: A Holistic Approach. Miller J. Park L. In: Bernstein A.
Morton PG. Coles J. Kitabchi AE. Wall B. 5th edn. (2004). Kreisberg RA. Diabetes Care. Endocrine emergencies. (1999) Health Outcome Indicators: Diabetes.25: 514–519. Cuervo R. American Family Physician. Fontaine DK. Diabetic emergencies. Journal of Emergency Nursing. The development of diabetic ketoacidosis. Soni N. 27: 1873–1878. Kitabchi AE. Report of a Working Group to the Department of Health. Umpierrez GE. Oxford: National Centre for Health Outcomes Development. Freire AX. Savage MW. (1999). Home P. Kitabchi AE. 117: 291–296. Cuervo R. Hudak CM. Wall BM. Latif K. Oh TE.