Sr.

n Variable Powe Confid Assumptio AHR/ Fleiss Possible Refere

o s r ence ns AOR total incomplete and nces
level samp lost medical
le records (10%)
size
1 Comorbi 80% 95% P1=84.2% 5.6 184 202 Jimma(
dities P2=66.4% AOR 84)
2 Recent 80% 95% P1=50.9 % 2.748 134 147 Iraq(69)
acute P2=27.4% AOR
illness
3 Missed 80% 95% P1=48.8 % 1.870 336 369
insulin P2=33.8% AOR
dose
4 Comorbi 80% 95% P1=31.9 % 2.57 268 294 Wollo(
dity P2=17.2% AOR 81)
5 communi 80% 95% P1=24.4% 1.63 306 336 Bahir
ty health P2=39.3% AHR dar(63)
insurance
hildren diabetic ketoacidosis is an acute life threatening and recurrent medical emergency that
requires frequent hospitalizations, treatment and monitoring for multiple metabolic abnormalities
and vigilance for complications. It is the leading cause of morbidity and mortality that continued
to be a major public health concern [1]. It has considerable costs to the health care systems and
adds burden of costs to the patients and families [2]. Despite the improvement in the treatment
and care of type 1 diabetes mellitus and the development of guidelines, diabetic ketoacidosis is
still a major cause of hospitalization and the leading cause of death in children with type 1
diabetes mellitus [1, 3–7]. According to international diabetic federation’s estimation, the
incidence of diabetes mellitus in Ethiopian children (0–14) was 30 per 1000 populations. The
mean diabetes related expenditure per a child was 29 United States Dollars [8]. Evidences
revealed that the common predictors of diabetic ketoacidosis among children with type 1
Diabetes Mellitus were age at diagnosis [9–20], sex [9–12, 14–16, 18–20], lack of healthcare
insurance [16, 19, 20], residence [9, 14, 15, 19], primary care giver [14, 19], insulin dose [9–12,
14, 15, 19–22], missed insulin [9], comorbidities like depression, epilepsy, bronchial asthma and
down syndrome [13, 15, 21, 22], childhoodinfections [9, 15, 20, 23], lower frequency of clinical
visits [11, 16, 18], omission of insulin [11, 18, 20], absence ofdiabetic education [11, 15, 19].
Moreover, protocols currently used for the management of DKA in sub-Saharan Africa were
largely based on those used in developed world, which do not take into consideration co
variables, associated with DKA unique to developing countries [24]. Diabetic ketoacidosis has
an overall mortality between 15 and 31 per 10,000 patients, and cerebral edema (CE), which is a
devastating complication, accounts for between 57% and 87% of DKA related deaths [12]. Even
in developed countries, there is significantly excess mortality from ketoacidosis among children
with type 1 diabetes Page 3/16 mellitus [25]. The incidence of DKA varies considerably between
different countries and studies [26]. The majority of DKA cases occur in patients with previously
diagnosed diabetes [24]. It occurs at a rate of 1–10/100 patientyears in children with established
diabetes globally. The most recent rate of DKA was 4.81/100 patient-years in children with
established type 1 diabetes mellitus in Germany [10]. Up to 90% of children with type 1 DM
reported one or more episodes of DKA over the last 6 months in Sub Saharan Africa. Other
studies show a variation from 25–90% [11, 27]. In developing countries like Ethiopia, the risk of
dying from DKA is greater. Even though many patients with diabetes mellitus in Ethiopia keep
dying from DKA, there is little documentation about DKA. This creates double burden of
communicable and non-communicable diseases [28]. Therefore, this study aimed to determine
the incidence and predictors of diabetic ketoacidosis among children with type 1 diabetes
mellitus at Debremarkos and Felegehiwot referral hospitals, 2013–2017.

The risk of dying from DKA is higher in poor nations like Ethiopia. Despite the fact that DKA
continues to kill a lot of patients with diabetes mellitus in Ethiopia, little is known about the
condition. Due to this, the burden of both communicable and non-communicable diseases is
increased.

The likelihood of dying from DKA is higher in poor nations like Ethiopia. Although DKA
continues to kill many patients with diabetes mellitus in Ethiopia, little is known about the
condition. Due to this, the burden of both communicable and non-communicable diseases is
increased.

Despite all challenges and costs incurred by diabetes and its complications, studies are limited to the
overall determinants in Ethiopia. Therefore, identification of determinants of the disease and the
possible preventive measures that could be instituted to arrest or delay the onset of diabetic
ketoacidosis. Thus, this study was aimed to assess the possible determinants of diabetic ketoacidosis
among diabetic patients in North Wollo and Waghimra Zone Public Hospitals.

For the second objective, sample size is calculated using predictors significantly associated with
the outcome variable. Accordingly, the sample size is determined using a double population
proportion formula by considering the presence of recent acute illness, missed insulin dose,
residence and T1DM duration as predictor variables of incidence of DKA based on a study done
previously by using Epi -info version 7 statistical package. It is calculated by taking a two-sided
significant level (α) of 5 %, 95% confidence level, power 80 %, and 1:1 ratio of non-exposed to
expose. Finally 653 participants will be selected as the final sample size for the study. The
sample size calculation is shown below in table 1.

Table 1: sample size calculation for predictors of diabetic ketoacidosis among type 1 diabetic
mellitus patients diagnosed adults in Amhara Comprehensive Specialized Referral Hospitals,
Northwest Ethiopia, 2023.

Sr.n Variable Powe Confid Assumptio AOR Fleiss Possible incomplete Referen
o s r ence ns total and lost medical ces
level sample records (10%)
size
1 Residenc 80% 95% P1=57.37% 2.762 72 79 Iraq(51)
e P2=25.34% AOR
2 Recent 80% 95% P1=50.9 % 2.748 134 147
acute P2=27.4% AOR
illness
3 Missed 80% 95% P1=48.8 % 1.870 336 369
insulin P2=33.8% AOR
dose
4 T1DM 80% 95% P1=44.5 % 1.106 594 653
duration P2=33.3% AOR
Key: P1=% of exposed with the outcome, and P2=% of non-exposed with the outcome

The largest sample size will be 653.

It is essential to obtain accurate global and regional estimates of the prevalence of GDM for allocation of
health resources and shaping of prevention and management policies to cope with the rising prevalence
of GDM [

[11]
]. In the 10th edition of the IDF Diabetes Atlas, we have made an additional methodological attempt to standardize
the global, regional and country income level specific prevalence of GDM by using the IADPSG's diagnostic criteria
and universal oral glucose tolerance test (OGTT) strategy.

According to the WHO report, in 2014 about 422 million people worldwide have diabetes, the
majority living in low-and middle-income countries, and in 2019, diabetes was the direct cause
of 1.5 million deaths and 48% of all deaths due to diabetes occurred before the age of 70 years.
In 2017 there were 9 million people with type 1 diabetes.

In 2014, 8.5% of adults aged 18 years and older had diabetes. In 2017 there were 9 million
people with type 1 diabetes. In April 2021 WHO launched the Global Diabetes Compact, a
global initiative aiming for sustained improvements in diabetes prevention and care, with a
particular focus on supporting low- and middle-income countries. The Compact is bringing
together all stakeholders to work on a shared vision of reducing the risk of diabetes and ensuring
that all people who are diagnosed with diabetes have access to equitable, comprehensive,
affordable and quality treatment and care.

In May 2021, the World Health Assembly agreed a Resolution on strengthening prevention and
control of diabetes. It recommends action in areas including increasing access to insulin;
promoting convergence and harmonization of regulatory requirements for insulin and other
medicines and health products for the treatment of diabetes. In May 2022 the World Health
Assembly endorsed five global diabetes coverage and treatment targets to be achieved by 2030.

Diabetes
16 September 2022

‫العربية‬

中文

Français

Русский

Español

Key facts
  The number of people with diabetes rose from 108 million in 1980 to 422 million in 2014.
Prevalence has been rising more rapidly in low- and middle-income countries than in high-
income countries.
 Diabetes is a major cause of blindness, kidney failure, heart attacks, stroke and lower limb
amputation.
 Between 2000 and 2019, there was a 3% increase in diabetes mortality rates by age.
 In 2019, diabetes and kidney disease due to diabetes caused an estimated 2 million deaths.
 A healthy diet, regular physical activity, maintaining a normal body weight and avoiding tobacco
use are ways to prevent or delay the onset of type 2 diabetes.
 Diabetes can be treated and its consequences avoided or delayed with diet, physical activity,
medication and regular screening and treatment for complications.

Overview
Diabetes is a chronic disease that occurs either when the pancreas does not produce enough
insulin or when the body cannot effectively use the insulin it produces. Insulin is a hormone that
regulates blood glucose. Hyperglycaemia, also called raised blood glucose or raised blood sugar,
is a common effect of uncontrolled diabetes and over time leads to serious damage to many of
the body's systems, especially the nerves and blood vessels.

In 2014, 8.5% of adults aged 18 years and older had diabetes. In 2019, diabetes was the direct
cause of 1.5 million deaths and 48% of all deaths due to diabetes occurred before the age of 70
years. Another 460 000 kidney disease deaths were caused by diabetes, and raised blood glucose
causes around 20% of cardiovascular deaths (1).

Between 2000 and 2019, there was a 3% increase in age-standardized mortality rates from
diabetes. In lower-middle-income countries, the mortality rate due to diabetes increased 13%.

By contrast, the probability of dying from any one of the four main noncommunicable diseases
(cardiovascular diseases, cancer, chronic respiratory diseases or diabetes) between the ages of 30
and 70 decreased by 22% globally between 2000 and 2019.

Type 2 diabetes
Type 2 diabetes (formerly called non-insulin-dependent, or adult-onset) results from the body’s
ineffective use of insulin. More than 95% of people with diabetes have type 2 diabetes. This type
of diabetes is largely the result of excess body weight and physical inactivity.

Symptoms may be similar to those of type 1 diabetes but are often less marked. As a result, the
disease may be diagnosed several years after onset, after complications have already arisen.

Until recently, this type of diabetes was seen only in adults but it is now also occurring
increasingly frequently in children.

Type 1 diabetes
Type 1 diabetes (previously known as insulin-dependent, juvenile or childhood-onset) is
characterized by deficient insulin production and requires daily administration of insulin. In 2017
there were 9 million people with type 1 diabetes; the majority of them live in high-income
countries. Neither its cause nor the means to prevent it are known.

Symptoms include excessive excretion of urine (polyuria), thirst (polydipsia), constant hunger,
weight loss, vision changes, and fatigue. These symptoms may occur suddenly.

Gestational diabetes
Gestational diabetes is hyperglycaemia with blood glucose values above normal but below those
diagnostic of diabetes. Gestational diabetes occurs during pregnancy

Women with gestational diabetes are at an increased risk of complications during pregnancy and
at delivery. These women and possibly their children are also at increased risk of type 2 diabetes
in the future.

Gestational diabetes is diagnosed through prenatal screening, rather than through reported
symptoms.

Impaired glucose tolerance and impaired fasting glycaemia

Impaired glucose tolerance (IGT) and impaired fasting glycaemia (IFG) are intermediate
conditions in the transition between normality and diabetes. People with IGT or IFG are at high
risk of progressing to type 2 diabetes, although this is not inevitable.

Health impact
Over time, diabetes can damage the heart, blood vessels, eyes, kidneys, and nerves.

 Adults with diabetes have a two- to three-fold increased risk of heart attacks and strokes (2).
 Combined with reduced blood flow, neuropathy (nerve damage) in the feet increases the
chance of foot ulcers, infection and eventual need for limb amputation.
 Diabetic retinopathy is an important cause of blindness and occurs as a result of long-term
accumulated damage to the small blood vessels in the retina. Close to 1 million people are blind
due to diabetes (3).
 Diabetes is among the leading causes of kidney failure (4).
 People with diabetes are more likely to have poor outcomes for several infectious diseases,
including COVID-19.

Prevention
Lifestyle measures have been shown to be effective in preventing or delaying the onset of type 2
diabetes. To help prevent type 2 diabetes and its complications, people should:
  achieve and maintain a healthy body weight;
 be physically active – doing at least 30 minutes of regular, moderate-intensity activity on most
days. More activity is required for weight control;
 eat a healthy diet, avoiding sugar and saturated fats; and
 avoid tobacco use – smoking increases the risk of diabetes and cardiovascular disease.

Diagnosis and treatment

Early diagnosis can be accomplished through relatively inexpensive testing of blood glucose.

Treatment of diabetes involves diet and physical activity along with lowering of blood glucose
and the levels of other known risk factors that damage blood vessels. Tobacco use cessation is
also important to avoid complications.

Interventions that are both cost-saving and feasible in low- and middle-income countries include:

 blood glucose control, particularly in type 1 diabetes. People with type 1 diabetes require
insulin, people with type 2 diabetes can be treated with oral medication, but may also require
insulin;
 blood pressure control; and
 foot care (patient self-care by maintaining foot hygiene; wearing appropriate footwear; seeking
professional care for ulcer management; and regular examination of feet by health
professionals).

Other cost saving interventions include:

 screening and treatment for retinopathy (which causes blindness);

 blood lipid control (to regulate cholesterol levels);
 screening for early signs of diabetes-related kidney disease and treatment.

WHO response
WHO aims to stimulate and support the adoption of effective measures for the surveillance,
prevention and control of diabetes and its complications, particularly in low- and middle-income
countries. To this end, WHO:

 provides scientific guidelines for the prevention of major noncommunicable diseases including
diabetes;
 develops norms and standards for diabetes diagnosis and care;
 builds awareness on the global epidemic of diabetes, marking World Diabetes Day (14
November); and
 conducts surveillance of diabetes and its risk factors.

The WHO Global report on diabetes provides an overview of the diabetes burden, interventions
available to prevent and manage diabetes, and recommendations for governments, individuals,
the civil society and the private sector.
The WHO module on diagnosis and management of type 2 diabetes brings together guidance on
diagnosis, classification and management of type 2 diabetes in one document.

In April 2021 WHO launched the Global Diabetes Compact, a global initiative aiming for
sustained improvements in diabetes prevention and care, with a particular focus on supporting
low- and middle-income countries. The Compact is bringing together all stakeholders to work on
a shared vision of reducing the risk of diabetes and ensuring that all people who are diagnosed
with diabetes have access to equitable, comprehensive, affordable and quality treatment and care.

In May 2021, the World Health Assembly agreed a Resolution on strengthening prevention and
control of diabetes. It recommends action in areas including increasing access to insulin;
promoting convergence and harmonization of regulatory requirements for insulin and other
medicines and health products for the treatment of diabetes. In May 2022 the World Health
Assembly endorsed five global diabetes coverage and treatment targets to be achieved by 2030.

References

1. Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019.
Results. Institute for Health Metrics and Evaluation. 2020 (https://vizhub.healthdata.org/gbd-
results/).

2. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a
collaborative meta-analysis of 102 prospective studies. Emerging Risk Factors Collaboration.
Sarwar N, Gao P, Seshasai SR, Gobin R, Kaptoge S, Di Angelantonio et al. Lancet. 2010;
26;375:2215-2222.

3. Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence
of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global
Burden of Disease Study GBD 2019 Blindness and Vision Impairment Collaborators* on behalf
of the Vision Loss Expert Group of the Global Burden of Disease Study† Lancet Global Health
2021;9:e141-e160.

4. 2014 USRDS annual data report: Epidemiology of kidney disease in the United States.
United States Renal Data System. National Institutes of Health, National Institute of Diabetes
and Digestive and Kidney Diseases, Bethesda, MD, 2014:188–210.

Dependent Variable: incidence of AKI

Independent Variables:
The independent factors of the study are socio-demographic factors (age, sex, residence, occupation,
marital status, educational status, and family history of diabetes mellitus), behavioral factors (smoking
history, history of alcohol intake, and glycemic control), clinical factors and laboratory factors includes
fasting blood sugar, level of HgA1C, infection, Diabetic Keto Acidosis, Hyperglycemia Hyperosmolar
State,
hypoglycemia, sepsis, comorbid conditions (dyslipidemia, myocardial infarction, chronic kidney disease,
congestive heart failure, diabetic foot ulcer, chronic liver disease, diabetic nephropathy, and
hypertension),
BMI, and treatment-related factors including insulin, oral hypoglycemic agent, non-steroidal anti-
in
Provides chronic care, emergency, antiretroviral therapy services, surgical, dental, medical,
pediatric, gynecologic, obstetric, and other services for more than 4 million clients.

DBCSRH is located in Debre Birhan town of Amhara Regional state, about 130 kilometers
North-East of Addis Ababa. Currently, DBRH is the only public Hospital in Debre Birhan Town,
it serves for 2.8 million catchment population.
Woldiya hospital is found in north Wollo zone in Amhara region, Ethiopia

First 5 comprehensive specialized referral hospital will be selected by simple random sampling
methods from a total of 8 comprehensive specialized referral hospital in Amhara region and
these are UoGCSRH, FHCSRH, TGCSRH, DMCSRH, and DTCSRH. Between September 11,
2017 to September 10, 2022, there are a total of 10,401 adults who have been diagnosed with
T1DM, of which 3571, 3152, 1527, 1249, and 902, are at UoGCSRH, FHCSRH, DMCSRH,
DTCSRH, and TGCSRH respectively. Proportional allocation will be done for each hospital and
the sampling frame will be prepared by collecting the identification numbers of T1DM patients
from the registration book. Finally, after identifying the patients who fulfill inclusion criteria,
study participants will be selected by a simple random sampling technique using computer-
generated random numbers. Schematic presentation of the sampling procedure is described in
figure 2 below.
Different methods are used to diagnose DKA, but they all require the presence of the three
criteria like blood sugar levels (>250 mg/dL), ketones, and high anion gap metabolic acidosis
(serum bicarbonate <18 mEq/L and/or pH <7.30)(15).

A global systematic review show different value of IR from 0-56 and about 263/1000. Therefore,
here we have got a discordant finding among these literatures. This shows us that there are other
plenty of possible factors that might have a chance of explaining, hindering, predisposing or
preventing the occurrence of DKA.
With an average global incidence of 0-56 per 1000 person-years (PYs) and an outlier study
indicating an incidence of 263 per 1000 PYs(17).
Lack of funding for non-communicable diseases, poor public health care, inadequate health
education, and poor self-glycemic control are all major contributors to the expanding DKA
burden in Sub-Saharan nations including Ethiopia (18, 32-34).

In Ethiopia, the cost of inpatient DM management, medicine consumption, and bed occupancy
all significantly depend on complications of DM especially on DKA(17).

Furthermore, the results are crucial for identifying potential intervention areas and the
development of therapies to lessen the incidence of DKA. In consequence, this can result in
fewer hospital beds being occupied and less personnel and material resources being required to
maintain occupancy levels. In the study area, comprehensive specialized referral hospitals will be
forewarned about DKA.
After evaluating several literatures (27, 45, 52, 65), the conceptual framework is modified, and it
now demonstrates a link between the dependent variable (incidence of DKA) and the
independent variables (socio-demographic and behavioral predictors, Clinical predictors and
treatment-related predictors).

Socio-demographic and Clinical predictors

behavioral predictors  Recent acute illness
 Sex  Comorbidities
 Age  Glycemic control
 Residence  DM duration
 Community health insurance  Chronic DM complication
 Smoking  HA1C and BMI
 Family history of T1DM

Incidence of DKA

Treatment predictors

 Type of medication used

 Medication noncompliance/
missed insulin dose
 Follow up frequency
 treatment started duration

Figure 1: Conceptual framework of incidence and predictors of DKA among adults with T1DM
in Amhara regional state Comprehensive Specialized Referral Hospitals, Ethiopia, 2023.
 First 5 Comprehensive Specialized Referral Hospital will be selected by simple random sampling
methods from a total of 8 Comprehensive Specialized Referral Hospital in Amhara region and
these are UoGCSRH, FHCSRH, TGCSRH, DMCSRH, and DTCSRH.
Amhara region Comprehensive
Specialized Referral Hospitals

Simple random sampling

UoGCSRH, FHCSRH, TGCSRH, DMCSRH, and DTCSRH

= 10,401 T1DM

Proportional allocation

TGCSRH DMCSRH DTCSRH

UoGCSRH FHCSRH

N=902 N=1527 N=1249

N= 3571 N=3152

n3=c n4=d n5=e

n1=a n2=b

Simple random sampling

n=387
Figure 2: Schematic presentation of the sampling procedure of adult diabetic patients in Amhara
Comprehensive Specialized Referral Hospitals, Northwest Ethiopia, 2023.

Why you chose this ratio. Make it self-explanatory.

See this for your further understanding

3.5.1 Sample size determination

For the first objective, the sample size was calculated by using the single population proportion

formula with considerations of the following statistical assumptions based on previous study

(34) : a 7.85 proportion of death (p = 0.0785 and q = 0.9215), a 95% confidence level, and a 5%

margin of error.

Sample size = (Z α/2 ) 2 P (1-P)

D2

Where; ni= initial sample size

Z=1.96 the corresponding Z-score for the 95% CI

α=confidence interval (95%)

p = proportions of death, 0.0785.

ni=((1.96)2.(0.0785)(0.9215))/(0.05)2= 111.15 and by considering incomplete patient records, 10

of the initial sample size was added and the final sample size was 123.

For the second objective, it is also calculated by using the second objective to check the
Adequacy of the sample size used for a survival sample size calculation power approach using
STATA 17 software with Cox proportional assumptions. The four predictive variables included
Are follow up frequency, history of medication noncompliance, DM duration, acute recent
infections, and comorbidities from the previous study conducted in Bahir dar(34) . By using
Schoenfeld Formula (Schoenfeld DA, Sample-size formula for the proportional-hazards
regression model,
Biometrics 1983;39 (2):499-503.) (68)
E= , pev = 1- (+, N=
Where E= number of event.
N= number of sample size.
P1 = proportion of event among exposed.
Pev. = probability of an event.
P2 = proportion of event among non-exposed.

22

Table 1 :- Summary of sample size calculation with second objective to access incidence and

predictors of mortality among neonates with perinatal asphyxia admitted at West Oromia

region referral hospitals from January 1, 2018 to December 31, 2021

Variables Assumptions CHR Probability of

events

Total sample size Reference

Cord prolapse Two side 95% CI 13.68

7.85

65

(34)

Pregnancy induced

hypertension

Power= 80% 3.23 324

Maternal Iron

deficiency anemia

Ratio= 1:1 2.28 655

Convulsion 3.27 317

From those predictors, the sample size obtained from maternal iron deficiency anemia was

considered the final sample size of the study because it gives a maximum sample size. Using the

probability of events (7.85) and crude hazard ratio of (2.28), proportional withdrawals (10%),

95% level of confidence &amp; power of (80%); the final sample size become 655.

stpowerlogrank, hratio (1.29) power (0.8) wd prob (0.1)

To date, a total of …. T1DM adult patients are diagnosed at this site in the period between
September 11, 2017 and September 10 2022, of which about…. T1DM adult patients are
receiving active follow-up between September 11, 2017 and September 10 2022.

Diabetes mellitus (DM) describes a group of metabolic disorders characterized by increased blood
glucose concentration; and the main threat to human health in the 21st century.1 In 2017, there were
425 million DM sufferers globally, and 14.2 million of them were in Africa, 77% were in developing
countries, and 5.2% of them were in Ethiopia.2 Diabetic ketoacidosis (DKA) is an acute life-threatening
complication of diabetes mellitus characterized by the triad of hyperglycemia, acidosis, and ketosis that
occurs in the presence of very low levels of effective insulin action.1,3–7 It is one of the common
emergencies among patients with diabetes mellitus and is associated with considerable morbidity and
mortality.8

it is still the leading cause of death; and mortality risk is substantially increased in patients with
chronically poor glycemic control and frequent DKA.3 Frequent DKA is a serious and common health
problem.9 Globally, around 65,000 children aged under 15 years develop type 1 diabetes each year and
up to 80% present with DKA.10 In Africa, the mortality rate due to DKA is unacceptably higher.8 In
developing countries, it ranges from 6% to 24%. In Ethiopia, the risk of dying from DKA is higher due to
poor medical services in the country.4,6,11–13 In addition to the risk of mortality, frequent DKA has a
major impact on the quality of life of the individuals and their families, accelerated micro-vascular
complications, and costs for families and health care.4,9,14 According to studies conducted in different
parts of the world, being female, those diagnosed at the younger age, family history of DM, missing an
insulin dose, infection, and psychiatric comorbidities were the common determinants of frequent
DKA.9,10,15–20, The Federal Ministry of Health (FMOH) of Ethiopia developed a National Strategic
Action plan (NSAP) for four priority Non-Communicable Diseases (NCD), including DM, by focusing on
integration to a three-tier health care system. However, due to limited resources, the country mainly
focuses on infectious diseases and pays little attention to noncommunicable diseases.2 Besides this little
attention, the control of diabetes and the prevention of ketoacidosis were hampered by socioeconomic
factors; particularly the cost and unreliability of insulin supplies.21 As a result, DKA remains a common
public health issue, and a major child killer in subSaharan African countries, including Ethiopia.19
Moreover, there is limited data on the incidence of DKA and its determinants in Amhara Regional State,
Ethiopia. Therefore, this study assessed the frequency of DKA and its determinants among pediatric
diabetes mellitus patients in public hospitals of Amhara Regional State, northwest Ethiopia.
By concentrating on integration to a three-tiered healthcare system, Ethiopia's Federal Ministry
of Health (FMOH) created a National Strategic Action Plan (NSAP) for four priority Non-
Communicable Diseases, including DM. The nation, however, primarily concentrates on
infectious diseases and gives little attention to non-communicable diseases because of inadequate
resources. Together with this lack of attention, socioeconomic considerations, particularly the
high cost and unpredictability of insulin supply negatively impacted the management of DM and
the prevention of ketoacidosis.

It is essential to study the incidence and predictors of the DKA to create strategies to reduce
DKA occurrence, along with accompanying morbidity, mortality, and medical expense.

data were collected by nurses trained and actively working in the given hospital wardrs by reviewing
charts using kobo toolbox software and exported to stata version 17 for analysis. cox proportional
hazard assumptions Will be checked. the appropriate model for the data will be selected by akaike
information criteria. cox snell residuals will used to check the goodness of fit. finally, an adjusted hazard
ratio with 95% ci was computed, and variables with a p-value & lt;0.05 in the multivariable analysis wil
taken as significant predictors for the incidence of death from perinatal asphyxia.

A five year institution-based retrospective follow up study will be conducted from September
11, 2017 to September 10, 2022 at selected comprehensive specialized hospitals in Amhara
regional state and the data extraction period will be from April 12, 2023 to May 28, 2023. A
Simple random sampling will be used to select the study subjects. Data will be collected by
trained nurses and data entry will be performed by using EPI info version 7.2.5.0 and will be
exported to STATA version 17.0 for statistical analysis. The frequencies and summary statistics
will be computed using descriptive and summary statistics by using median with interquartile
range (IQR) after checking the distribution of the data with histogram and Shapiro-Wilk test
whereas categorical variables will be displayed by using frequency with percentage. The Cox-
proportional hazard model will be used to evaluate the relationship between the independent
factors and the outcome variable. Bivariable variables with a p-value of less than 0.25 will be
candidates for multivariable Cox proportional hazard analysis, and 0.05 in the multivariable
analysis will be taken as significant predictors for the incidence of diabetes ketoacidosis.
The incidence and predictors of DKA among adult T1DM patients are not well studied in
developing countries. DKA is one of the major public health challenges which need
appropriately designed studies to establish a strong evidence for the healthcare system in order to
provide information for the planning and provision of health services. It is essential to study the
incidence and predictors of the DKA to create strategies to reduce DKA occurrence, along with
accompanying morbidity, mortality, and medical expense. However there is a limited data about
the incidence of DKA in Ethiopia. And as to the investigators knowledge, there is no specific
study to assess the incidence and predictors of DKA among adult T1DM patients in the country.

The Federal Ministry of Health (FMOH) of Ethiopia developed a National Strategic Action plan
(NSAP) for 4 priorities by concentrating on integration to a three-tiered healthcare system, Non-
Communicable Diseases (NCD), includes DM. Unfortunately, due to a lack of resources, the
country primarily prioritizes infectious diseases and gives non-communicable diseases little
attention. So the result of this study will provide key information to monitor and evaluate this
strategic action plan.

Determining the incidence and identifying the possible predictors of DKA among adult T1DM
patients can give important information to improve the quality of care and to reduce the
occurrence of DKA. The development of preventive measures and treatment regimens may be
aided by identifying the DKA predictors. The riskiest time following T1DM diagnosis to develop
DKA will be identified with the aid of the time to develop DKA assessment. In addition,
identifying common predictors has a great clinical significance that will support the health
professionals to set priorities for interventions and to adhere with evidence based practice, and
also institutional managers to monitor and evaluate health service programs and allocate
resources appropriately.

The results are anticipated to be crucial for identifying adults with T1DM who are at a high risk
of developing DKA, as well as crucial for identifying potential intervention areas and the
development of therapies to lessen the incidence of DKA. In consequence, this can result in
fewer hospital beds being occupied and less personnel and material resources being required to
maintain occupancy levels. For patients, it will provide knowledge about DKA that helps them to
avoid it, enhance their quality of life, and lengthen the time they spend living without DKA. In
the study area, comprehensive specialized referral hospitals will be forewarned about DKA.

The Federal Ministry of Health (FMOH) of Ethiopia developed a National Strategic Action plan (NSAP) for
four priority Non-Communicable Diseases (NCD), including DM, by focusing on integration to a three-tier
health care system. However, due to limited resources, the country mainly focuses on infectious
diseases and pays little attention to non-communicable diseases.2 Besides this little attention, the
control of diabetes and the prevention of ketoacidosis were hampered by socioeconomic factors;
particularly the cost and unreliability of insulin supplies.21 As a result, DKA remains a common public
health issue, and a major child killer in sub-Saharan African countries, including Ethiopia
It is unclear why some patients with T1DM present in diabetic ketoacidosis whereas others do not and
whether the development of diabetic ketoacidosis is a consequence of delayed diagnosis and treatment
or whether it reflects a particularly aggressive form of diabetes.10

Understanding which factors are associated with diabetic ketoacidosis and the relative importance of
delayed diagnosis and treatment is, therefore, important. This potentially informs both our
understanding of the disease as well as the development of patient, professional, and population based
interventions to reduce the proportion of adults presenting in diabetic ketoacidosis. Several individual
factors from individual studies have been quoted in guidelines and consensus statements.11 12 13 To
our knowledge, this is the first systematic review of all factors associated with diabetic ketoacidosis at
the diagnosis of type 1 diabetes in children and young adults

It is crucial to comprehend what causes diabetic ketoacidosis and the relative significance of delayed
diagnosis and therapy.

It is crucial to comprehend the predictors of diabetic ketoacidosis as well as the relative significance of
early detection and therapy. This may contribute to our understanding of the condition as well as the
development treatments to lower the number of people presenting with diabetic ketoacidosis.

These observed variations may be adduced to racial, geographic, and socio-demographic differences.
The predominance of urban-dwellers may be indicative of the enhanced exposition to various
sociocultural (higher current alcohol consumption rate observed) and environmental factors linked to
TIDM incidence with subsequent DKA evolution.[27] The urban residents were mostly females,
presented at a younger age, and were mostly tertiary-level educated at the time of TIDM with
concurrent DKA diagnosis. These three demographic characteristics have been reported in association
with DKA and could probably explain the preponderance of the urban residents in the current study.
[10,18,28] First, besides their greater predisposition to autoimmune disorders, the female
preponderance may also reflect the global gender-based pattern of TIDM wherein regions of low
incidence (Africa) exhibit female predominance while regions of high incidence (Europe) exhibit male
predominance. [28-30] Secondly, the diagnosis of DKA at a younger age may indicate poor recognition of
symptoms and poor health- seeking behaviors which could trigger DKA episodes. [18,28] Thirdly,
tertiary-level educated individuals are likely to increase once awareness of health status and enhance
recognition of clinical features of TIDM and possibly DKA, thereby prompting early presentation for
emergency medical care.

The majority of the urban population at the time of TIDM with concurrent DKA diagnosis were female,
younger in age, and primarily had tertiary education. These three demographic traits have been linked
to DKA and may help to explain why there are a disproportionate number of urban dwellers in the
current studies. First, in addition to having a higher propensity for autoimmune diseases, the female
preponderance may also reflect the global gender-based pattern of TIDM, which shows a female
preponderance in regions with low incidence (Africa), while a male preponderance is seen in regions
with high incidence (Europe). Second, a diagnosis of DKA at a younger age may reflect a lack of symptom
identification and unhealthy health-seeking practises, both of which have been linked to bouts of DKA.
Lastly, people with tertiary education are more likely to be aware of their health state and better
recognise the clinical signs of TIDM and perhaps DKA, leading to earlier presentation for emergency
medical care.

Most T1DM patients with acute infections were presented with nausea, vomiting and abdominal
pain, which was interpreted as an indication to reduce or stop their insulin. This is a deep-rooted
belief shared by some doctors and diabetic educators. The situation is made worse when such
patients consume large amounts of sugar-rich fluid to counteract presumed hypoglycemia. And
also higher glycemic status depresses immune responses with attendant higher infection rates among
diabetics

Most T1DM patients with acute infections complained of nausea, vomiting, and abdominal pain,
which was taken as a sign that they needed their insulin reduced or stopped. Some medical
professionals and diabetes educators have also this deep-rooted belief. When such individuals
take a lot of sugar-rich fluid to combat perceived hypoglycemia, the issue is made worse. And
also higher glycemic status depresses immune responses with attendant higher infection rates among
T1DM.

This might be because the presence of comorbidities can directly or indirectly disturb the normal
functions of insulin and other counter regulatory hormones. In addition, diabetic patients with
chronic medical comorbidities had poor glycemic control and psychiatric comorbidities can
compromise adherence to treatment. And documented reasons for poor drug compliance in patients
ranged from poor accessibility to health facilities, high costs of drugs often resulting from polypharmacy
because of co-morbidities and also ignorance on self-care habits of T1DM.

Based on their assumptions it may be the case because comorbidities can either directly or
indirectly interfere with insulin's and other counter-regulatory hormones' regular functions. T1DM
individuals with chronic medical comorbidities had poor glucose control and psychological
comorbidities can impact adherence to therapy. In addition, the reported causes of patients'
poor drug compliance included frequently brought on by polypharmacy due to co-morbidities.
This discrepancy could be due to methodology and clinical characteristics differences. In
addition, T1DM patients during the early years of diagnosis had stress and anxiety about the
disease and low knowledge and practice about the prevention of DKA

According to their views, the variation in methodology and clinical traits could be the cause of
this mismatch. Moreover, T1DM patients had limited awareness and practice about the
prevention of DKA in the early years following diagnosis, which added to their stress and worry
about the condition.

The reason might be that both lower and higher blood sugar level from the normal range can
lead to a point spectrum of DKA by increasing counter-regulatory hormones, altered glucose
production, and lipolysis.

The reason may be due to the fact that both lower and higher blood sugar levels than the usual
range can cause DKA at specific points in the spectrum by raising counter-regulatory hormones,
changing glucose synthesis, and enhancing lipolysis.

Enter search term Search

Journal Article

The Long-Term Incidence of Hospitalization

for Ketoacidosis in Adults with Established
T1D—A Prospective Cohort Study
Merlin Thomas, Valma Harjutsalo, Maija Feodoroff, Carol Forsblom, Daniel Gordin, Per-Henrik
Groop FinnDiane Study Group
Author Notes
The Journal of Clinical Endocrinology & Metabolism, Volume 105, Issue 1, January 2020,
Pages 231–241, https://doi.org/10.1210/clinem/dgz003
Published:
14 September 2019
Article history

 PDF
 Split View
 Cite
 Permissions Icon Permissions


Share
Abstract
Context

The long-term natural history of diabetic ketoacidosis (DKA) and its risk factors are poorly
understood.

Objective

To determine the long-term incidence and predictors of DKA in adults with longstanding type 1
diabetes (T1D).

Design

All hospitalizations and deaths due to DKA between 1996 and 2016 were identified in 4758
adults with T1D from the Finnish Diabetic Nephropathy Study (FinnDiane), and a cohort of 16
224 adults with T1D from the Finnish general population.

Results

Between 1996 and 2015, there were 1228 DKA events in the FinnDiane participants (1.4/100
person-years) and 4914 DKA events (1.8/100 person-years) in adults with T1D from the general
population. The majority were hospitalized only once. There was a modest increase in the
frequency of DKA in the FinnDiane over the follow-up (~2.4%/year [95% CI, 0.3–4.5%]; P =
0.03). Predictors of DKA were glucose control, CSII, smoking and alcohol consumption, and
raised high-density lipoprotein cholesterol and triacylglycerides. Diabetic nephropathy and renal
impairment were associated with DKA; patients with end-stage renal disease, macroalbuminuria,
and microalbuminuria had 2.09-fol (95% CI, 1.40–3.12), 1.65-fold (95% CI, 1.23–2.19), and
0.87-fold (95% CI, 0.61–1.24) risk of DKA compared with patients with normal albumin
excretion rate, respectively. Patients with an estimated glomerular filtration rate <60
mL/min/1.73 m2 were also more likely to be hospitalized for DKA (HR 1.71 [95% CI, 1.26–
2.67]).

Conclusions

DKA remains a common cause of hospitalization in individuals with longstanding T1D. These
data suggest that the goal to use SGLT2 inhibitors for their vasculo- and renoprotective actions
may be problematic, as those most likely to benefit may also have the highest risk for DKA.

Issue Section:
Clinical Research Articles

Diabetic ketoacidosis (DKA) is one of the most important emergencies in type 1 diabetes (T1D)
and a common reason for hospitalization (1). Despite improving diabetes care and new
technologies for insulin delivery and glucose monitoring, at least three in every 100 persons with
T1D may be hospitalized annually in the United States and this incidence appears to be
increasing both in the United States (2, 3) and in other countries (4). The total healthcare cost
associated with DKA in the United States has been estimated to be over 5 billion dollars annually
(1). Although mortality rates from DKA are declining, it remains one of the biggest contributors
to loss of life years associated with T1D (5, 6).

Many DKA events are observed in and around the time of the diagnosis of T1D, or during the
first few years of management when titration, accidental omission, or poor adherence may
sometimes result in inadequate insulin therapy, especially in children and adolescents. Cross-
sectional studies exploring the prevalence of DKA are therefore potentially dominated by newly
diagnosed cases of T1D, especially in young individuals and their associated risk factors. These
risks may not be the same as older adults with well-established diabetes, in whom the annual
incidence of acute DKA is lower (2) but likely remains significant over time (6).

The long-term natural history of DKA and its risks are poorly understood. Determining the
ongoing risks for DKA in adults with established T1D is also now particularly important, as the
future relevance of vasculo-protective outcomes of Sodium GLucose coTransporter (SGLT) 2
inhibitors may be offset by an increased risk of DKA following treatment with these agents in
this setting (7). In this study, we examine the long-term incidence and predictors of
hospitalization for DKA in a prospective cohort study of 4758 Finnish adults with T1D from the
FinnDiane study, and a retrospective cohort of 16 224 adults with T1D from the Finnish general
population followed over the same time period.

Methods
The Finnish Diabetic Nephropathy Study (FinnDiane) is an ongoing, nationwide, multicenter
study, initiated in order to identify genetic and environmental risk factors for diabetic
complications, with special emphasis on diabetic kidney disease in patients with type 1 diabetes.
A detailed description of the FinnDiane recruitment protocol has previously been presented (8).
Briefly, adult patients (≥18 years old) with type 1 diabetes across Finland were asked to
participate, and the participation rate was 78%. Type 1 diabetes was defined as insulin
dependence and C-peptide <0.30 nmol/L, age at onset of diabetes <40 years and insulin
treatment initiated within 1 year of diagnosis. The study protocol was in accordance with the
principles of the Declaration of Helsinki as revised in 2000 and was approved by the Ethics
Committee of Helsinki and Uusimaa Hospital District. Written informed consents were obtained
from each FinnDiane patient.

The baseline visit for participants occurred between the years 1994 and 2015 (n = 4758), at
which time participants underwent a thorough clinical examination and blood and urine samples
were collected. Information on the presence of diabetic complications including renal and
cardiovascular disease were obtained from the medical files by the attending physician using a
standardized questionnaire. In addition, albuminuria status was determined based on
measurements of urinary albumin excretion rate (UAER) from at least two timed overnight
(g/min) or 24-hour urine (mg/24 hour) collections. Normoalbuminuria was defined when in at
least two out of three urinary samples UAER was <20 μg/min or <30 mg/24 hour;
microalbuminuria, when UAER ≥20 and <200 μg/min or ≥30 and <300 mg/24 hour; or
macroalbuminuria, when UAER ≥200 μg/min or ≥300 mg/24 hour; end-stage renal disease
(ESRD) as dialysis treatment or having received a kidney transplant. Serum creatinine was
determined at a central laboratory by a kinetic Jaffe method using a Hitachi 911 autoanalyzer
until January 2002. Thereafter it was determined by a photometric enzymatic method using a
Hitachi 917 or Modular analyzer. We have taken account of this change in the measurement of
creatinine by using the modifying coefficient (1/1.0465). The estimated glomerular filtration rate
(eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-
EPI) formula.

Patients filled in questionnaires regarding their lifestyle, including their smoking habits, alcohol
consumption, and socioeconomic status. Serial hemoglobin A1 (HbA1c) values from the medical
files were obtained and intra-individual mean and coefficient of variation (CV) were calculated.
The CV was calculated as a ratio of intra-individual standard deviation and mean.

Ascertainment of outcomes

For each participant in the FinnDiane cohort, the composite outcome of admission to hospital,
attendance at an emergency department or specialized outpatient care with a diagnosis of DKA
or death from DKA (hitherto denoted as DKA hospitalization) was determined. Hospitalization
for DKA was ascertained from the Finnish Care Register for Health Care (HILMO) database that
collects all discharges from Finnish hospitals (including specialized outpatient care) and
associated diagnosis and treatment codes. All events in which the International Classification of
Disease (ICD) 9th Revision code of 2501B (1987–1995 for prevalent cases before visit) and the
10th Revision code of E10.1 (1996–2015) were used were designated as DKA. However, if there
were several successive admissions within less than 5 days they were calculated as one event.
Fatal DKA events were drawn from the Finnish Cause of Death Register.

For determining the frequency of DKA hospitalization in FinnDiane participants, all DKA events
between 1996 and 2015 were determined, including events that occurred prior to enrollment and
baseline visits. All DKA hospitalization events that occurred before the age of 18 or within the
first 2 years of their diabetes were excluded, as a recent diagnosis, honeymoon phases, and
subsequent titration of insulin may confound an analysis of DKA incidence in these cases.
Follow-up of outcomes commenced from the start of their third year of diabetes, from 18 years
of age, or the year 1996, whichever was most recent. For analysis of the predictors of incident
DKA, only DKA events that occurred after the baseline visit were considered as the outcome. A
history of prior DKA was determined as a coded diagnosis for hospitalization between 1987 and
baseline visit in the HILMO dataset and used as a predictive variable.

Retrospective cohort validation

The frequency of hospitalization for DKA over the same time period was also determined in a
larger retrospective cohort of adults with type 1 diabetes in Finland. To do this, all adults (age
>18) with T1D (defined by a diagnosis of diabetes under the age of 30 years and ongoing insulin
requirements) present on January 1 1994 were identified using the National Institute for Health
and Welfare database, as previously described (9–11). Admission to hospital for DKA between
January 1996 and December 2015 was then retrospectively identified using the HILMO
hospitalization and ICD-10 codes as detailed above. This inclusion strategy effectively excluded
events that occurred in patients with potentially less than two years duration of diabetes. The
study has permissions from the Finnish National Institute for Health and Welfare
(THL/786/6.02.00/2016) and Statistics Finland (TK53-26-16).

Statistical methods

Crude DKA rates and DKA mortality rates were calculated per 100 person-years with 95%
Poisson confidence intervals. We used linear regression to model the age-standardized
prevalence curves. In order to model the prognostic variables related to DKA, Cox regression
analyses were conducted on prospective data obtained from the time of baseline visit. Time-to
event was estimated by Kaplan–Meier survival analyses and the log rank test was used to test for
between-group differences. Poisson regression was used to model predictors of all DKA events
including recurrent events. Statistical analyses were performed using SAS 9.4 version (SAS
Institute Inc, Cary, NC, USA).

Results
Baseline characteristics of the FinnDiane cohort

The baseline characteristics of individuals experiencing hospitalization for DKA during the
prospective follow-up phase are shown in Table 1, compared with those not experiencing DKA
during follow-up. Those with at least one DKA hospitalization had more frequently prior DKA,
insulin pump therapy, higher HbA1c, mean and variability of serial HbA1c, dose of weekly
alcohol consumption, insulin dose, total cholesterol, and triglycerides, but lower age at onset of
diabetes, body mass index, and estimated glucose disposal rate than patients without DKA. In
addition, they were more likely to have smoking history, hypertension, kidney disease, and
severe diabetic retinopathy. Patients with two or more DKAs had further more likely history of
excess alcohol usage.

Table 1.

Baseline characteristics of adults with T1D from the Finn Diane study cohort, stratified
according to incident DKA during follow-up after enrolment

No. of DKA At least 1 DKA 1 DKA event ≥2 DKA events

events during the event during the during the during the
baseline visit baseline visit baseline visit baseline visit
until 2015 (n = until 2015 (n = until 2015 (n = until 2015 (n =
4297) 461) 273) 188)
Prior DKA (%) (1987-
8.1 20.0a 19.2a 22.7a
baseline)d
Insulin pump therapy
4.1 8.2a 9.0a 7.1c
(%)
Sex (female, %) 47.4 48.4 51.4 49.2
Age (years) 37.9 (29.0–47.4) 38.2 (28.5–46.4) 38.7 (28.5–45.9) 36.3 (27.2–45.0)
 No. of DKA At least 1 DKA 1 DKA event ≥2 DKA events
events during the event during the during the during the
baseline visit baseline visit baseline visit baseline visit
until 2015 (n = until 2015 (n = until 2015 (n = until 2015 (n =
4297) 461) 273) 188)
Age at diabetes
14.3 (9.3–23.0) 13.5 (8.9–20.9)c 13.4 (9.0–20.9) 13.2 (7.8–20.8)c
diagnosis (years)
Age at diagnosis of
diabetes
0–4 9.5 11.3 9.4 15.5
5–9 19.0 19.7 20.0 21.0
10–14 24.1 26.3 29.0 21.0
≥15 47.4 42.7 41.6 42.5c
Duration of diabetes 21.4 (12.2–31.0) 23.1 (13.0–31.4) 22.9 (11.5–31.3) 22.8 (13.3–31.0)
HbA1c (%) 8.3 ± 1.4 9.1 ± 1.7a 8.9 ± 1.6a 9.5 ± 1.9a
Intra-individual mean
8.3 (7.6–9.1) 8.9 (8.1–9.9)a 8.8 (8.0–9.7)a 9.1 (8.4–10.3)a
of HbA1c (%)
Coefficient of
0.09 (0.07–0.12) 0.11 (0.08–0.14)a 0.11 (0.08–0.14)a 0.11 (0.08–0.15)a
variability in HbA1c
Smoking history (%)e 45.8 58.8a 61.2a 53.2
Alcohol consumption
72.1 72.5 75.8 66.4
(yes, %)
Alcohol excess (yes,
16.4 20.2 16.9 24.0c
%)
Alcohol consumption
48 (24–90) 60 (24–120)a 48 (24–120) 72 (36–144)a
(g/week)f
Blue collar workers 50.0 55.1c 54.7 53.6
Without education 16.5 21.3 21.6 18.8
With education 33.5 33.8 33.1 34.8
White collar workers 24.5 21.3 21.2 23.8
Lower 20.7 15.7 14.3 18.8
Upper 8.3 5.6 6.9 5.0
Other or not known 25.5 23.6 24.1 22.6
Body Mass Index
25.1 ± 3.7 24.4 ± 3.6a 24.5 ± 3.5c 24.0 ± 3.4a
(kg/m2)
Waist-to-hip ratio 0.87 ± 0.09 0.87 ± 0.08 0.87 ± 0.08 0.87 ± 0.08
Systolic blood pressure
134 ± 19 136. ± 21 136 ± 22 136 ± 20
(mmHg)
Diastolic blood
79 ± 10 80 ± 11 79 ± 10 82 ± 11b
pressure (mmHg)
Blood pressure
42.4 52.7a 50.2c 53.6b
lowering therapy (%)
Hypertension (%)g 56.0 62.5b 59.6 64.1c
 No. of DKA At least 1 DKA 1 DKA event ≥2 DKA events
events during the event during the during the during the
baseline visit baseline visit baseline visit baseline visit
until 2015 (n = until 2015 (n = until 2015 (n = until 2015 (n =
4297) 461) 273) 188)
Insulin dose/kg body
0.69 ± 0.27 0.72 ± 0.27c 0.70 ± 0.28 0.75 ± 0.27b
weight (IU)
Estimated glucose
6.3 ± 2.5 5.6 ± 2.4a 5.8 ± 2.4b 5.3 ± 2.3a
disposal rate
Total cholesterol
4.89 ± 0.98 5.18 ± 1.07a 5.15 ± 1.10a 5.21 ± 1.04a
(mmol/L)
HDL cholesterol
1.34 ± 0.39 1.36 ± 0.44 1.36 ± 0.44 1.36 ± 0.46
(mmol/L)
Triglycerides
1.02 (0.77–1.46) 1.21 (0.85–1.80)a 1.21 (0.84–1.84)a 1.25 (0.82–1.79)a
(mmol/L)
Detectable C-peptide
21.9 18.3 22.2 13.4c
level (%)
Lipid lowering therapy
15.8 16.5 16.3 16.0
(%) at baseline
Normoalbuminuria 61.9 48.4a 51.2a 45.9a
Microalbuminuria 12.1 12.2 11.4 13.3
Macroalbuminuria 13.2 21.0 20.4 21.6
Albuminuria not
5.1 7.1 6.4 7.6
known
ESRD 7.7 11.3 10.6 11.6
eGFR
95 (78–110) 94 (73–112) 95 (74–112) 93 (66–112)
(mL/min/1.73m2)h
eGFR <60
10.8 17.9a 18.7 18.1b
mL/min/1.73m2 (%)h
SDR, laser treatment
33.7 43.5a 39.5 47.2a
(%)
CVD (%)i 6.3 7.6 6.6 9.0
hs-CRP (mg/L) 4.42 ± 8.98 5.40 ± 11.01 5.55 ± 12.43 5.42 ± 9.66

Abbreviations: DKA, diabetic ketoacidosis; ESRD, end-stage renal disease; eGFR, estimated
glomerular filtration rate using CKD-EPI formula; SDR, severe diabetic retinopathy; hs-CRP,
high sensitive CRP.aP < 0.001,bP < 0.01,cP < 0.05 for comparison with no DKA;dPrior
hospitalization for DKA as an adult (aged > 18 years) and after the first two years of
diabetes;eSmoking history was defined as smoking at least 1 cigarette per day for at least 1
year;fAmong those who consume any alcohol;gHypertension defined as either systolic blood
pressure ≥140 mmHg or diastolic blood pressure ≥90, based on the average of 2 measurements,
or any use of antihypertensive medication; hPatients with ESRD excluded;

iIncludes coronary artery disease and stroke.

Open in new tab

The incidence of hospitalization for DKA for participants in the FinnDiane

cohort

Between 1996 and 2015, there were 1206 hospitalizations for DKA and 22 fatal DKA events in
the FinnDiane participants resulted in an overall rate of 1.4 per 100 person-years. The proportion
individuals managed as inpatients was similar in those with eGFR <60 (62.4%) and with eGFR
≥60 (63.0%) (P = 0.81). The annualized frequency of DKA events is shown in Fig. 1A. There
was a modest but significant increase in the frequency of DKAs over this time period (~2.4% per
year (95% CI 0.3–4.5%, P < 0.03). The modelled mean incidence was 1.1 per 100 person-years
in 1996 while in 2015 it had risen to 1.6 per 100 person-years. These DKA events occurred in
617 participants (13.0%), meaning the majority of individuals experiencing hospitalization for
DKA, only experienced DKA on a single occasion in this time period (65%). Notably the
frequency of these “one-off” DKAs remained constant in this cohort (Fig. 1A).

Figure 1.

Open in new tabDownload slide

The increasing frequency in incidence of hospitalization for diabetic ketoacidosis (DKA) or

death from DKA between 1996 and 2015 in adults with type 1 diabetes (T1D) (A) participating
in the FinnDiane study (n = 4758) and (B) in adults with T1D from the Finnish general
population (n = 16 219). Black circles show annualized rates for all DKA hospitalizations (upper
line) white circles (middle line) show annualized rates for a DKA hospitalization that occurred in
an individual who subsequently experienced no further admissions for DKA (ie, one-off
admissions) and triangles (lower line) shows annualized DKA mortality rates. Dashed line shows
mean with 95% upper and lower confidence intervals.

Predictors of the incidence of hospitalization for DKA in the FinnDiane cohort

From the time of their enrollment in the FinnDiane cohort, there were 969 non-fatal or fatal
DKAs that occurred in 461 individuals over a median follow-up of 14.4 (10.6–16.6) years from
baseline visit, equating to an overall event rate of 1.5 per 100 person-years only over the
prospective follow-up period. This number is slightly higher than the total hospitalizations
detailed above, in the same individuals that occurred between 1996 and 2015.

The majority of these DKAs (n = 709, 73%) occurred in individuals that did not have a prior
recorded DKA hospitalizations as an adult (aged > 18) in the HILMO database, beyond the first
2 years of their diabetes. Only 92 individuals (20%) experiencing a DKA hospitalization after
enrollment had been hospitalized for DKA previously as an adult. Total of 369 participants that
had a DKA hospitalization event prior to enrollment, did not have a subsequent event after
baseline during prospective follow-up.

After adjusting for a prior history of hospitalization for DKA, incident DKA in this population
was independently associated with smoking and alcohol consumption at baseline, raised high-
density lipoprotein (HDL) cholesterol and plasma triglycerides concentrations, insulin pump
therapy, and poor glucose control (HbA1c) at baseline (Table 2). In particular, HbA1c levels
over the course of follow-up and HbA1c variability trait were also independently associated with
the incidence of hospitalization for DKA in adults with type 1 diabetes (Fig. 2, Table 2)—such
that those individuals with the best glycemic control and the least variability in their HbA1c also
had the lowest incidence of DKA, while those individuals with the highest rates of DKA also had
worse glucose control and more variability in their control. The predictors of time to first event
on Cox regression analysis were the same as those for all DKA events determined using Poisson
regression (data not shown).

Table 2.

Independent predictors of incident DKA in adults with T1D during the FinnDiane study cohort
prospective follow-up period identified using Cox regression analysis. Patients with end-stage
renal disease at baseline were excluded for this analysis

Hazard ratio 95% CI P value

Prior history of DKA (yes) 2.05 1.58–2.67 <0.0001
Insulin pump therapy (yes) 2.44 1.60–3.72 <0.0001
Smoking history (yes) 1.31 1.04–1.66 0.02
Weekly alcohol consumption (per 10g) 1.02 1.01–1.03 0.002
Serial HbA1c (%) 1.48 1.35–1.62 <0.0001
HbA1c variability (per 10 percentage points) 1.31 1.07–1.61 0.01
HDL cholesterol (mmol/L) 1.82 1.36–2.45 <0.0001
Triglycerides (mmol/L) 1.15 1.03–1.29 0.01
2
eGFR <60 mL/min/1.73 m (yes) 1.71 1.26–2.67 <0.0001

Abbreviations: DKA, diabetic ketoacidosis; eGFR, estimated glomerular filtration rate; HDL,
high-density lipoprotein.

Open in new tab

Figure 2.

Open in new tabDownload slide

The cumulative incidence of first hospitalization for diabetic ketoacidosis (DKA) following
enrolment of adults with type 1 diabetes (T1D) into the FinnDiane study, stratified according to
mean and variability (coefficient of variation, CV) of serial HbA1c above and below the median.

Most importantly, diabetic nephropathy was associated with increased risk of DKA. After
adjustment for these other risk factors patients with ESRD, and macro- and microalbuminuria the
risk was 2.09-fold (95% CI 1.40–3.12; P < 0.0001), 1.65 (1.23–2.19; P = 0.0007), and 0.87-fold
(0.61–1.24; P = 0.43) compared with patients with normal albumin excretion rate (Fig. 3A).
After excluding participants with ESRD, and adjusting for other risk factors, the presence of
moderate to severe impairment in kidney function (eGFR < 60 mL/min/1.73 m2) remained
associated with the incidence of hospitalization for DKA in adults with T1D, such that those
individuals with eGFR <60 mL/min/1.73 m2 were twice as likely to be hospitalized for DKA
(HR 1.71, [1.26–2.67]; P < 0.0001) when compared with those with an eGFR >60 mL/min/1.73
m2 (Table 2, Fig. 3B). Hyperfiltration at baseline (eGFR >130 mL/min/1.73 m2) was also
associated with an increased risk of incident DKA (HR 2.35; 95% CI 1.50–3.66; P = 0.0002).
However, after adjusting for glycemic control and glycemic variability during the study this
association also became non-significant (P = 0.07). Notably, the duration of T1D or the age at
diagnosis were not associated with incidence of DKA in adults with type 1 diabetes. In addition,
there was no association between incident DKA and sex, socioeconomic status, the control of
LDL cholesterol or blood pressure, or the presence of retinal complications at baseline (defined
by the need for laser therapy).

Figure 3.

Open in new tabDownload slide

The cumulative incidence of first hospitalization for diabetic ketoacidosis (DKA) following
enrolment of adults with type 1 diabetes into the FinnDiane study, stratified according to diabetic
nephropathy status at baseline (A) and to the presence of an eGFR <60 mL/min/1.73 m2 at
baseline (B). ESRD, end-stage renal disease.

The frequency of hospitalization for DKA in the Finnish retrospective cohort

In 15 149 adults with type 1 diabetes from the general population, over the same time period of
20 years (1996–2015), there were 4771 hospitalizations for DKA and 143 fatal DKA events
affecting 2376 (15.7%) individuals, equating to an event rate of 1.8 per 100 person-years over
this time period. This retrospective cohort did not show significant increase in annual incidence
(P = 0.09, Fig. 1B). The modelled incidence of hospitalization for DKA was 1.5 in 1996 per 100
person-years and 2.1 in 2015. Again, as in the FinnDiane cohort, most of the individuals
experiencing DKA events in the retrospective cohort only had one event during this time period
and the frequency of “one-off” DKA events remained constant in this cohort (Fig. 1B). Mortality
from DKA though rare, did not appear to decline with time (Fig. 1B), though more patients were
hospitalized with DKA over the same time period, meaning that DKA deaths as percentage of
DKA events declined slightly, consistent with recent registry data (5, 6).

Discussion
DKA is an important and surprisingly common acute complication of T1D that results in
increased healthcare costs (1–3), hospitalization (3, 4), and premature mortality (6, 12). In this
contemporary prospective cohort of Finnish adults with T1D, one in ten were hospitalized or
died from DKA over a 14-year period, 36% of whom were hospitalized on more than one
occasion. The rate of DKA was similar in adults with T1D from the Finnish general population
than in participants in the FinnDiane study. Importantly, we report in both cohorts, the annual
incidence rate in the same individuals did not decline over time and actually increased over the
follow-up time period. Tragically, between 1.8% and 3.0% of all DKA events leading to
hospitalization were associated with patient death.
It is widely regarded that DKA is only a problem of early diabetes management, and becomes
less likely in patients with established and ongoing insulin use and dedicated diabetes care. This
assumption is supported by outcomes in the Diabetes Control and Complications Trial and the
Epidemiology of Diabetes Interventions and Complications cohort that showed a reduction in
DKAs over the follow-up, with an incidence rate of approximately two cases per 100 person-
years at baseline falling to no events at the 12-year follow-up, even in the standard treatment arm
(13). Similarly, in a prospective follow-up of the Pittsburgh Epidemiology of Diabetes
Complications (EDC) study cohort, the incidence of DKA fell to less than one case per 100
person-years at the 18-year follow-up (13). However, it is possible that the trials settings and
close attention to diabetes control both during and after these studies may have progressively
reduced the risk of DKA in these studies. In real-world data from adults with T1D in the
FinnDiane study and adults with T1D in the Finnish general T1D population, DKA clearly did
not decline, even as “one-off” events. These data suggest that regardless of the duration of
diabetes, the risk of DKA remains a constant companion.

Previous cross-sectional registry-based studies exploring the predictors of DKA in populations

have been inevitably dominated by new cases of T1D and the inclusion of children and young
adults that have the highest risks of DKA. The clinical characteristics of these individuals
therefore largely determine the reported risk factors for DKA in these reports. Such risks may be
less relevant to the adult with established T1D, in whom longitudinal risks for DKA have not
been well studied. The FinnDiane study is a large prospective cohort study designed specifically
to look at complications in adults. Taking this prospective approach, and excluding all events in
the first 2 years after diagnosis, we have been able to follow the incidence for hospitalization for
DKA over 14 years within individual adults with T1D. We anticipate that this approach may be
more useful in determining individuals who may be ketosis prone, or risk factors at baseline that
may be usefully associated with subsequent DKA in the long term.

Consistent with previous studies, in participants within FinnDiane cohort, poor glucose control
was associated with an increased risk of hospitalization for DKA, which was clearly associated
with both HbA1c at baseline, mean HbA1c during the follow-up, and intra-individual HbA1c
variability. It is likely that each of these markers captures some of the risk for DKA associated
with recurrent hyperglycemia, inadequate insulin therapy, and/or non-adherence. The presence of
elevated serum triglycerides may also capture some of this risk. However, even in those with
good ongoing glucose control during the study, and the least variability in HbA1c,
hospitalization for DKA still occurred.

In a recent metanalysis of trials, use of continuous insulin subcutaneous infusions (CSII) has also
been associated with a higher incidence of reported DKA than conventional insulin therapy (14).
The small percentage of participants in the FinnDiane cohort using insulin pumps at baseline
(6.6%) also had a higher incidence of hospitalization for DKA (HR = 2.44, 95% CI 1.60–3.72, P
< 0.001) than those receiving standard insulin therapy. Beyond malfunction of insulin pump
and/or catheter occlusion, the reasons behind this association are unclear but may include a
failure to up-titrate insulin during acute illness. Notably, the association between DKA
hospitalization and CSII was independent to poor baseline or achieved glucose control, which is
often an indicator for pump initiation.
Smoking and alcohol consumption were also independently associated with DKA in the
FinnDiane cohort. While these may both be markers of risk-taking behaviors on non-adherent
adults with T1D, the association remained significant after adjusting for ongoing glucose control
and HbA1c variability. Ethanol is metabolized by the liver to generate acetic acid, and
subsequently acetyl-coenzyme A, which is both a substrate for ketogenesis and inhibitor of
peripheral lipolysis. Oxidation of alcohol also increases generation of nicotinamide adenine
dinucleotide, which inhibits gluconeogenesis and further drives the generation of beta-
hydroxybutyrate. These pathways are augmented in the absence of insulin, making excess
alcohol intake and T1D a potentially dangerous combination. It is possible that increasing
alcohol consumption in Finland following a reduction of 33% in alcohol taxes in 2005–2007,
partly contributed to the increasing prevalence of DKA observed in this study, as well as
increased alcohol-related mortality in the general population (15).

Previous studies have identified socioeconomic status and educational attainment as key risk
factors for DKA (16). The nature of society and the cost-of-care support for individuals with
T1D (without requiring medical insurance) may substantially attenuate the impact of their impact
in Finland and no association was observed between socioeconomic status and DKA in our
cohort, after adjusting for alcohol use and glycemic attainment. Some cross-sectional studies
have also suggested that female gender is associated with an increased risk of hospitalization for
DKA, despite similar overall indices of glucose control. By contrast, recent US data reported
higher incidence of DKA hospitalization in males, possibly because male patients with T1D
present at a younger age. By comparison, in our prospective and retrospective cohorts we saw no
association between gender and DKA in Finnish adults with T1D. We have previously
documented higher risk-taking behavior (eg, smoking and drinking excessively) in males with
T1D and a higher risk of chronic complications of their diabetes (6, 17). However, after adjusting
for confounding effects, gender appeared to play no role in incident DKA in our cohort.

In this study, we report for the first time a strong independent association between diabetic
nephropathy, kidney function, and DKA hospitalization. In particular, a reduced eGFR (<60
mL/min/1.73 m2) was associated with a higher risk of hospitalization for DKA. This association
was independent of markers of glycemic control which may also be disturbed in individuals with
complications. It is also possible to speculate that the increasing incidence of DKA in Finnish
adults with T1D over this time period was partly determined by the progressive decline in kidney
function that occurred over the same time period.

It is known that the kidney plays a key role in mitigating the effects of hyperglycemia and
ketonemia, such that in individuals with renal impairment reduced renal excretion of excess
ketones and glucose alongside impaired anion exchange may facilitate progression to DKA.

It is well established that the kidney plays a critical role in reducing the consequences of
hyperglycemia and ketonemia, therefore diminished renal excretion of extra ketones and
glucose together with impaired anion exchange may increase progression to DKA in those with
renal impairment.