Journal of Diabetology

: 2020  |  Volume : 11  |  Issue : 3  |  Page : 183--190

Management of diabetic ketoacidosis after the introduction of local hospital protocol in the secondary care hospital

Salim S Qassabi1, Shiju Ramanunni1, Melba S D’Souza2, Jeffrey Singh3, Padma M J Kurup3,  
1 Department of Internal Medicine, Al Nahda Hospital, Ministry of Health, Muscat Governorate, Sultanate of Oman
2 Nursing Department, Sultan Qaboos University Hospital, Muscat Governorate, Sultanate of Oman
3 Department of Communicable Disease, Surveillance and Control, Directorate General of Health Services, Muscat Governorate, Sultanate of Oman

Correspondence Address:
Dr. Shiju Ramanunni
Department of Internal Medicine, Al Nahda Hospital, Ministry of Health (MOH), Muscat.
Sultanate of Oman


Background: Diabetic ketoacidosis (DKA) is an acute, severe and life-threatening metabolic complication of diabetes. Objective: The objective of this study was to conduct a clinical audit of the management of DKA based on the hospital protocol in the selected secondary care hospital. Design: An observational retrospective longitudinal study design was used to review the data of the patients admitted with DKA. Settings: The study was conducted in a secondary care government hospital, which has 200 beds, including 48 beds in medical ward and 6 beds in the high dependency unit. Materials and Methods: A survey questionnaire was used based on local hospital protocol, and the data was collected from patients admission notes between January 2010 and December 2014, using electronic patients records. Outcome Measures: The study looked at outcomes such as how DKA protocol was followed, complications, adherence, causes, investigations carried out or not, severity, readmissions, and duration of the stay. Sample Size: The audit selected 49 patients from a total of 83 admission notes with DKA who fits the sampling criteria. Results: Of 49 patients, 38 patients were having type 1 diabetes mellitus and 11 patients having type 2 diabetes mellitus. The most common cause in both groups is omission of insulin. Intravenous 0.9% sodium chloride was initiated in the early first hour of diagnosis of DKA for most patients. Readmission rate was 25%. Insulin was commenced in less than 1h for two-thirds of the total patients. Poor adherence to the protocol such as monitoring serum sodium bicarbonate, serum potassium levels and replacement of potassium levels in the early period of management were observed. Conclusion: Continued evidence-based practice and education for medical and paramedical staffs is needed to reduce the complications of DKA and efficiently resolve DKA, improve patient outcomes, and reduce the length of hospital stay. New DKA protocol has been introduced in the hospital as the result of this audit.

How to cite this article:
Qassabi SS, Ramanunni S, D’Souza MS, Singh J, J Kurup PM. Management of diabetic ketoacidosis after the introduction of local hospital protocol in the secondary care hospital.J Diabetol 2020;11:183-190

How to cite this URL:
Qassabi SS, Ramanunni S, D’Souza MS, Singh J, J Kurup PM. Management of diabetic ketoacidosis after the introduction of local hospital protocol in the secondary care hospital. J Diabetol [serial online] 2020 [cited 2020 Oct 31 ];11:183-190
Available from:

Full Text


Diabetic ketoacidosis (DKA) is an acute, severe and life-threatening metabolic complication of diabetes. DKA is defined as uncontrolled catabolism with insulin deficiency. This occurs due to catabolic stress.[1] This study is needed in the Middle East to assess the gaps in the standard guidelines and protocol. Strict adherence to the protocol will reduce morbidity, mortality, and length of hospital stay.[2],[3],[4],[5]

 Materials and Methods

The study has been approved by the institutional research and ethics committee. Hospital permission was obtained, and the unit was informed of the study. Confidentiality and anonymity were ensured in the study. A questionnaire was used based on local hospital protocol, and the data were collected from patient admission notes between January 2010 and December 2014. A preset pro forma sheet of the causes of DKA, age, gender, medical parameters, blood investigations, diagnostics, medical management, fluid resuscitation, insulin, and other medications was used from the existing electronic patient records (EPR). Hospital protocol approved by the hospital committee was used. Data was collected from the EPR by the endocrinologists to avoid error and was entered in the Statistical Package for the Social Sciences (SPSS) software. Epidemiological software EpiData and SPSS, version 22.0, manufactured by the International Business Machines Corporation (IBM) located in Armonk, New York, USA was used for data analysis. Descriptive analysis (frequency, percentage, and percentile) was used for the study [Figure 1].{Figure 1}

An observational retrospective longitudinal study design was used to review the data of the patients admitted with DKA in a secondary care government hospital that has 200 beds, with 48-beds in the medical wards including 6-beds in the high dependency unit. Patients admitted in this selected secondary care hospital were diagnosed as DKA if they fulfill all the three following diagnostic criteria: hyperglycemias, ketonuria, and pH less than 7.3 or bicarbonate less than 18 or anion gap more than 10.[6] Patients admitted with DKA in the consecutive 5 years, from January 2010 to December 2014, were the population in the selected secondary care hospital. A modified DKA protocol adopted from the American Diabetes Association Consensus Statement 2009 has been in use in the hospital during the study period. The protocol is similar to two major tertiary care hospitals in the region, so that the results of the study can be extrapolated to general health care practice and national policy in the region. This is the first DKA study in adults in the country.

A convenient sample of 83 admission notes from 49 patients who fit the sampling criteria was appropriate for the study aim. Inclusion criteria consisted patients with diabetes mellitus, aged more than 12 years, admitted in a selected secondary care hospital with the confirmed primary diagnosis of DKA across 5 years with accessible EPR.

 Ethical Clearance

Ethical approval for this study (Ethical Committee ALN/RC/2018) was provided by the Research and Ethical Committee Al Nahdha Hospital, Ministry of Health, Sultanate of Oman, on 04th March, 2018.


From the admission notes of hospital records (Al-Shifa system), we were able to analyze 83 patients records from 49 patients. A total of 30 patients (61%) were between the age-group of 13–30 years, 18 patients (37%) between the age-group of 31–65 years, and 1 patient (2%) more than 65 years with median age of 27.5 years and mode of 15 years [Figure 2]. Of the 49 patients, 32 (65%) were males [Table 1].{Figure 2} {Table 1}

Of the 83 patients’ admissions notes, we found that 46 patients’ admissions notes were repeat admissions from just 12 patients (24%). Of this, one patient had 10 repeat admissions, one had six repeat admissions, one had five repeat admissions, two patients had four repeat admissions, three patient had three repeat admissions, and four patients had two repeat admissions [Table 1].

There were no deaths reported; 55% of the patients (27 patients) were discharged in 1–3 days of hospitalization, 31% of the patients (15 patients) stayed for 4–7 days, 12% patients (6 patients) stayed for 8–14 days, and 2% patients (1 patient) stayed for 15–42 days. In total, the length of hospital stay varied from 1–42 days with median stay of 3 days and mode of 2 days [Table 1].

Of the 49 patients, only for the 47 patients (96%), causes were documented. For the two patients (4%), causes were not documented. Of which, 38 patients (78%) were diagnosed as type 1 diabetes mellitus and 11 patients (22%) were diagnosed as type 2 diabetes mellitus. The most common cause was omission of insulin in patients with both type 1 and type 2 diabetes mellitus. There were 18 patients with type 1 diabetes mellitus and 5 with type 2 diabetes mellitus with a total of 23 patients (47%). Eleven patients (22%) were newly diagnosed as DKA, of which, eight and three patients belonged to type 1 diabetes mellitus and type 2 diabetes mellitus group, respectively.

The third common cause was infection, with a total of 10 patients (20%), of which 7 patients had type 1 diabetes mellitus and 3 had type 2 diabetes mellitus. They were diagnosed to have pneumonia, otitis media, gastroenteritis, Urinary tract infection, herpes zoster, infected bed sores, and boils. one patient (2%)with type 2 diabetes mellitus had myocardial infarction (2%), others were two patients (4%) with type 1 diabetes mellitus admitted because of nausea, vomiting, and poor oral intake (dehydration), and pancreatitis, respectively [Figure 3].{Figure 3}

The majority of the patients with type 1 diabetes mellitus belongs to adolescent and young age-group between 20 and 30 years of age except three patients around 40 years of age. However, the majority of patients with type 2 diabetes mellitus were older than 40 years of age, with two patients around 30 years of age.

Of the 49 patients, 24 patients (49%) developed severe DKA, 14 patients (29%) developed moderate DKA, and 11 patients (22%) developed mild DKA [Table 1]. Severity classification was based on American diabetes association criteria.[7] Similar findings were noted in other studies.[5]

Regular insulin/medications were documented only in 35 patients (71%) of 49 patients [Table 1]. Routine blood and urine investigations were carried out for all patients, but electrocardiography, chest X-ray, and blood culture were performed for 31 (63%), 36 (74%), and 34 patients (69%), respectively [Figure 4].{Figure 4}

The main treatment of DKA is early administration of intravenous (IV) fluids. Of the 49 patients, only for the 4 patients (8%), there was a delay of around 1–4h after admission in starting IV fluids, for remaining 45 patients (92%), IV saline was started in stipulated time of less than 1h [Table 2].{Table 2}

Of the 49 patients, IV saline was not followed as per protocol in the 37 patients (76%), and it was followed only for the 12 patients (24%), and in none of these patients’ file, reason for not following the protocol was documented. Potassium was replaced early only in 9 (18%) of the 49 patients. For the remaining 40 patients (82%), potassium was not started at early hours [Table 2].

Variable rate IV insulin should be commenced within 1h after obtaining urea and electrolytes results, but it was administered only for the 35 patients (71%) of the 49 patients, for the 12 patients (25%), it was started within 1–4h, and for the 2 patients (4%), it was delayed for more than 4h [Table 2].

Variable rate IV insulin should be continued until the resolution of DKA, and it was continued for 48 patients (98%); only for 1 patient, it was not continued and that patient developed recurrence of DKA during same the admission [Table 2].

Venous pH or bicarbonate should be checked every 2h initially until the resolution of DKA, it was carried out only for the 21 patients (43%) and not for the 28 patients (57%). As per our protocol, sodium bicarbonate can be administered in very severe DKA if pH less than or equal to 6.9, after consulting consultant endocrinologist; only one patient of the 49 patients fit this criteria, and sodium bicarbonate infusion was started [Table 2].

Potassium should be checked initially within 4h but it was checked only for the 19 patients (39%) and not for the 30 patients (61%). Dextrose 5% should be initiated if blood glucose level was equal or less than 12 mmol/L, it was started for all 49 patients (100%). As per our protocol, blood glucose level should be checked every hour until the resolution of DKA, it was checked for all the patients (100%). Seventeen of the 49 (35%) patients developed hypoglycemia (less than 4mol/L) during insulin infusion, which was mild to moderate and was corrected as per protocol. None of them had severe hypoglycemia [Table 2].

The median length of hospital stay was 3 days. As per hospital protocol, DKA should be managed in high dependency unit. A total of the 48 patients (98%) managed in high dependency unit, only 1 patient (2%) managed in medical ward for unknown reason [Table 1].

DKA-related complications during treatment included 17 patients with mild to moderate hypoglycemia and 18 patients with mild hypokalemia. Eight patients each with moderate and severe hypokalemia respectively. No major complications such as pulmonary and cerebral edema were observed [Figure 5].{Figure 5}

The median time taken for resolution of DKA was 22h (0–82h) with 25% of the patients resolved in 17h, 50% of patients in 22h, and 25% in 27.5h, and the mode time being 26h [Figure 6].{Figure 6}

The median and mode time taken for conversion of IV insulin to subcutaneous insulin after resolution of DKA was 2h with 25% of patient’s conversion carried out in 1h (which was the time frame in our protocol), for the 50% of patients, it was performed in 2h, and for the 25% of patients, it was carried out in 2h 30min [Figure 7].{Figure 7}

Considering all the parameters of the overall protocol, only five patients (10%) were strictly adherent to the protocol, whose median and mode of stay was 2 days and median resolution of DKA occurred in 22h [Table 2].


We aimed to audit our patients admitted with DKA between January 2010 and December 2014 after the introduction of our DKA local protocol. We found most of them were young patients. One-fourth had repeat admissions because of poor compliance to medications. As a result of this audit, we found poor adherence in important areas of the protocols such as early introduction of insulin and potassium, monitoring serum bicarbonate and potassium, timely administration of IV sodium chloride, and documentation. Similar findings were highlighted in previous studies.[5],[7],[8],[9],[10],[11]

The reason for not following adequate IV fluids in the stipulated time was due to delay in the administration of initial bolus fluids in emergency department, which could be due to the lack of DKA protocol in emergency department and delay in transporting patients to medical admission units due to the lack of adequate staff and availability of bed. Also the flow rate for IV fluids per hour in DKA protocol in medical unit was not clearly expressed.

The delay in IV insulin administration in 14 patients (29%) was due to the late result of potassium level owing to delay in the transport of the blood samples to laboratory due to inadequate staff, hemolyzed samples, waiting for medical officers (usually very busy) to collect samples, and delayed results.

Potassium supplements were given at a later stage than requested by the protocol, leading to a low serum potassium level. The delay in giving potassium supplements can be attributed to various reasons, mainly lack of clear instructions in the protocol regarding the maximum dose of potassium per hour (flow rate) that can be administered and the dose of potassium to be administered according to the exact serum potassium level. The target range of serum potassium level for the patient with DKA was also not adequately highlighted in the protocol. Other possible explanations for this problem include blood investigations carried out later than the time specified by the protocol, laboratory delay in issuing results, hemolyzed samples, waiting for doctors to collect blood samples, and low potassium replacement dose in current protocol.

The cause for hypoglycemia is most likely due to change of IV fluids from 0.9% sodium chloride to 5% dextrose at 12 mmol/L rather than higher cutoff values.

A total of the 12 patients (24%) contributed to 46 readmissions. These patients were noncompliant to insulin, had poor attendance at follow-up clinics, and had psychosocial issues, which are comparable to the result published in the large teaching hospital where they found one-third of the patients had repeat admissions.[12]

In spite of poor adherence to protocol, our median hospital stay was 3 days, and no mortality was observed. But we must take every effort for increasing adherence to protocol, especially patient should have received insulin in 1h, initial investigations should be carried out within 1h and repeated timely according to the protocol to reduce the length of hospital stay and to avoid complications. Also, the timely documentation is mandatory, and strict adherence to protocol is possible by integrated and continuous training of the staff in dedicated areas.[13],[14],[15],[16],[17]

Limitations: We selected patients with a diagnosis such as type 1 diabetes, type 2 diabetes, hyperglycemia, and DKA during 2010–2014. This is a small study population. It is possible that we may have missed other patients diagnosed as DKA. The admitting medical and confirmatory diagnosis could have been entered as a differential diagnosis based on the International Classification of Disease in the Al-Shifa EHR system as recorded by the admitting doctor or the physician. There were no standard measures of interrupting the timing of observations and fluids therapy due to some unclear documentation and use of the protocol. Currently a multicenter DKA audit in the tertiary care hospital is under process in the region.

As a result of this audit, the endocrinology team modified the current DKA protocol according to evidence-based medicine and with the help of Joint British Diabetes Society Inpatient care group 2013 DKA[18] guidelines, and carried out the following changes: (1) Introduction of basal insulin analog, glargine, for patients already taking before which will reduce rebound hyperglycemia and reduce time for DKA resolution, thereby reducing length of stay; (2) potassium chloride replacement was increased from 10 to 20 mmol per 500mL with emphasis on flow rate per hour according to the potassium level so that it will reduce hypokalemia event, which is a very significant complication in this study; (3) steps were taken to minimize the delay in investigations by encouraging staff nurse to collect blood and sample to be taken personally to lab if there is delay in transport and alerting labs to release in 1h and to increase staff strength if possible; (4) threshold to start 5% dextrose was increased from blood glucose 12 to 14 mmol to avoid hypoglycemia; and (5) the new protocol highlighted the IV fluid section with flow rate per hour, which is very clearly documented and which will increase the adherence to the protocol.

The importance of increasing nursing staff and regular review of patients by medical officers has been emphasized. All patients have been given structured education by multidisciplinary team comprising endocrinologist, clinical pharmacist, dietician, and nursing staff with sick day rules and in particular not to omit insulin even if very sick before discharge with regular outpatient diabetic clinic follow-up.


The study showed poor adherence to the protocol in some areas such as blood sampling and replacement of potassium in early hours. As a result of this audit, our endocrine department was able to modify the DKA protocol and to implement new protocol. Ongoing work is needed to provide continuous patient and staff education to reduce the risk of morbidity, repeat admission rates, and length of hospital stay for patients admitted with DKA. We shall reaudit the new DKA protocol in 2 years to see the difference in outcomes.

Financial support and sponsorship

This study was supported by the Ministry of Health, Sultanate of Oman.

Conflicts of interest

There are no conflicts of interest.


1Lee MH, Calder GL, Santamaria JD, MacIsaac RJ Diabetic ketoacidosis in adult patients: An audit of factors influencing time to normalization of metabolic parameters. Internal Med J 2018;48:529-34.
2Devlia D Adherence to protocol during the acute management of diabetic ketoacidosis: Would specialist involvement lead to better outcomes. Int J Clin Pract 2010;64:1580-2.
3Hassan IS, Al-otaibi AD, Al-Bugami MM, Salih SB, Al Saleh Y, Abdulaziz S The impact of a structured clinical pathway on the application of management standards in patients with diabetic ketoacidosis and its acceptability by medical residents. J Diabetes Mellitus 2014;4:264-72.
4Rotter T, Kinsman L, James E, Machotta A, Gothe H, Willis J, et al. Clinical pathways: Effects on professional practice, patient outcomes, length of stay and hospital costs. Cochrane Database Syst Rev 2010;3:CD006632.
5Chakera AJ, Brooks AM, Jones AG, Foster JAH, Carpenter AB Can we use new national guidelines regarding the management of diabetic ketoacidosis to improve care in a teaching hospital? Diabet Med 2011;28:32-203.
6Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN Hyperglycemic crises in adult patients with diabetes. Diabetes Care 2009;32:1335-43.
7Kurdi H, Pinto LP, Smeeton FJ An audit of the management of diabetic ketoacidosis in a district general hospital after introduction of the Joint British Diabetes Societies guidelines. Diabet Med 2014;31:28-183.
8Bondugulapati LNR, Phyu CHO, Irrinki L, Dixon A, Griffiths C Audit on the management of diabetic ketoacidosis in a district general hospital. Diabet Med 2013;30:30-201.
9Manikandan RM, Abel J, Verdaguer R, Rajendran R, Masding MG Management of diabetic ketoacidosis improves with implementation of care pathway. Diabet Med 2012;29:30-177.
10Parthasarathy S, Ellis K, Smith A, Crown A, Vaughan N Management of diabetic ketoacidosis: Comparison against standards using an online audit tool. Diabet Med 2014;31:28-183.
11van Zeller C, Milner T, Herring R, Russell-Jones D A paradigm shift in the management of diabetic ketoacidosis: The introduction of the Joint British Diabetes Societies 2010 guideline into the Royal Surrey County Hospital. Diabet Med 2014; 31:28-183.
12Crasto W, Htike ZZ, Turner L, Higgins K Management of diabetic ketoacidosis following implementation of the JBDS guidelines: Where are we and where should we go? Br J Diabetes Vascular Dis 2015;15:11-6.
13Jervis A, Champion S, Figg G, Langley J, Adams G Prevalence of diabetes ketoacidosis rises and still no strict treatment adherence. Curr Diabetes Rev 2013;9:54-61.
14Dawkins CL, Oyibo SO The management of diabetic ketoacidosis in a busy district hospital: An audit report. Online J Clin Audits2012;4:261-269.
15Thomson-Moore A, Low A, Williams J, Bradley K, Thorogood N Impact of the 2010 diabetic ketoacidosis (DKA) guidelines (based on the 2010 Joint British Diabetes Societies Inpatient Care Group Standards of Care) and IV insulin and fluid prescription chart on DKA management at University Hospitals Bristol NHS Foundation Trust. Diabet Med 2012; 29:30-177.
16Brown CW, Watson WA, Morrison IG, MacLennan KY Novel methods of teaching diabetic ketoacidosis (DKA) diagnosis and management in a clinical curriculum: Sim-DKA. Practical Diabetes 2014;31:207-11.
17Dhatariya KK, Nunney I, Higgins K, Sampson MJ, Higgins K, Iceton G National survey of the management of Diabetic Ketoacidosis (DKA) in the UK in 2014. Diabet Med. 2016;33:252-60.
18Joint British Diabetes Societies Inpatient Care Group. The Management of Diabetic Ketoacidosis in Adults. Second edition updated September 2013 [Last accessed on 2019 Dec 12].