Sage Journals: Discover world-class research

Abstract

Background:

A lower incidence of cardiovascular events has been reported in type 2 diabetes patients treated with insulin analogs (IAs). Corresponding data on people affected by type 1 diabetes are not available yet.

Methodology:

We investigated demographic and clinical data from 509 type 1 diabetics, who were treated in an outpatient clinic from 2006 to 2012. Multiple logistic regression was used to investigate the relationship between the type of insulin treatment and the prevalence of cardiovascular (CV) complications, that is, presence of coronary heart, cerebrovascular and peripheral arterial diseases, adjusting for potential confounders.

Results:

Results from multiple logistic regression revealed that patients with impaired renal function [estimated glomerular filtration rate (eGFR) < 90 ml/min] show lower CV complication rates when treated with IAs (25%) compared with patients treated with human insulin (HI; 28%) and HI/IA (38%, p = 0.06). CV complication rates in the complete patient collective amounted to 17% (IA), 21% (HI) and 21% (HI/IA, p = 0.08). Examination of CV complications according to the type of IA revealed the lowest complication rates in type 1 diabetics treated with insulin lispro (5.9%) and glargine (16%). However, complication rate differences among insulin treatments did not reach statistical significance.

Conclusion:

The present cross-sectional study shows a borderline significantly lower CV morbidity in people with type 1 diabetes and impaired renal function when treated with IA compared with HI treatment after adjustment for multiple potential confounders [odds ratio (OR) = 0.78, which translates into a 22% lower complication rate]. Validation of these preliminary findings in confirmatory, prospective studies may have important clinical implications.

Keywords

Diabetes Type 1 cardiovascular events macroangiopthy insulin insulin analogs glargine lispro detemir aspart

Introduction

The treatment of type 1 diabetes has markedly improved over the last two decades, not only due to novel insulin administration devices (insulin pump, continuous glucose monitoring systems) but also due to new, genetically modified insulin (= insulin analogs; IAs). Short-acting IAs such as insulin lispro, aspart, glulisine can prevent the occurrence of postprandial blood glucose peaks better than human insulin (HI) due to their faster onset of action. Long-acting IAs, that is, insulin glargine and detemir, have a flatter and longer active profile than human basal insulin (NPH insulin) and thus reduce the risk of hypoglycemia, especially at night. Postprandial blood glucose peaks and hypoglycemia have long been identified as a risk factor for the occurrence of cardiovascular (CV) complications in patients with type 1 and type 2 diabetes.^1–4

A number of studies on type 2 diabetics have reported a lower occurrence of CV events when using IA.^5–11 Corresponding studies have not been performed in patients with type 1 diabetes. In a study based on approximately 500 type 1 diabetics, we have recently shown that treatment with IAs, in particular, insulin lispro and glargine, is associated with an improved renal function [higher estimated glomerular filtration rate (eGFR) and lower albumin excretion] and slower hemoglobin decline as renal function deteriorates.¹² In the present study, we investigate whether IA treatment is associated with a lower prevalence of CV complications.

Patients and methodology

Data from 509 patients with type 1 diabetes who attended the outpatient clinic of St. Josefshospital Heidelberg GmbH between 2006 and 2012 and agreed to participate in the study were evaluated. The study design has been described in detail elsewhere.¹² Briefly, main inclusion criteria were age from 20 to 85 years, duration of diabetes at least 3 years, and no change in insulin therapy during the previous 12 months. Exclusion criteria included severe illnesses including malignoma, liver cirrhosis, congestive heart failure, renal insufficiency (eGFR < 15 ml/min), acute infection, any treatment with steroids, pregnancy and organ transplantation.

At the first patient visit, medical history, demographic data and information on concomitant medication were retrieved, and laboratory as well as technical tests [electrocardiogram (ECG), ankle brachial index] were conducted. The presence of following CV complications were assessed: (a) coronary heart disease, defined as history of myocardial infarction, procedure of percutaneous transluminal coronary angioplasty or coronary artery bypass graft, coronary vessel disease confirmed by coronary angiography; (b) cerebrovascular vessel disease, defined as history of ischemic stroke, transient ischemic attack or stenosis of a. carotis >50% (c) peripheral artery disease defined as history of bypass operation, procedure of percutaneous transluminal angioplasty, amputation or ankle brachial index <0.8.

Diabetic patients were grouped according to the type of insulin treatment: long-acting insulin (HI, insulin detemir, insulin glargine) and short-acting insulin (HI, insulin aspart, insulin glulisine, insulin lispro). In general, treatment with IA was initiated by the patient’s GP/diabetologist. Reasons for change from HI to IA were mostly hypoglycemic events during night, high postprandial glucose levels or insufficient glycemic control. The time of treatment with the specific type of IA was reported by the patient or provided by the GP/diabetologist practices.

Relative frequencies were used to describe ordinal-scaled factors, for example, to examine complication rates according to the type of insulin treatment. The location and spread of ratio scaled variables were represented by medians with 5th and 95th percentiles. Differences between groups of patients treated with different insulin types were assessed by uncorrected Chi-square or Wilcoxon tests as appropriate. Univariate logistic regression was used to identify patient characteristics associated with rate of CV complications, and to quantify rate difference among insulin treatments. Subsequently, associated patient characteristics were taken into account in multiple logistic regression models, which permitted to estimate complication odd ratios with 95% confidence intervals adjusted for potential confounders. Statistical analyses were conducted using SAS version 9.3 (SAS Institute Inc, Cary, NC, USA).

The study was approved by the local ethics committee and written consent was obtained from all patients.

Results

Most investigated type 1 diabetes patients received an intensive insulin therapy (91%), 7.3% of them used an insulin pump. About one third (31.4%) was exclusively treated with HI; almost half of the patients (46.8%) were treated with IA, the remaining (21.8%) received a combination of HI and IA. Reported average treatment times with IA were 4.9 years for lispro, 4.6 years for aspart, 3.8 years for glulisine, 5.1 years for glargine and 4.5 years for detemir. In 78% of the patients, the treatment did not change for at least 3 years before examination date.

The most important characteristics of the investigated patients are summarized in Table 1. Patients treated with HI and IA showed no differences (p > 0.05) in sex proportions, age, diabetes duration, body mass index (BMI), hemoglobin (Hb)A1c, high-sensitivity C-reactive protein (hs-CRP), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) cholesterol. Overall cholesterol and triglyceride levels were slightly higher in the HI group. Approximately 19–24% of patients received a lipid-lowering regimen with statins, without statistically significant proportion differences between treatment groups. The prevalence of hypertension, high-pressure therapy with angiotensin-converting-enzyme (ACE)/AT1 blockers or beta blockers and the quality of the blood pressure adjustment were also similar in the HI and IA groups.

Table 1.

Main patient characteristics (median values with 5th and 95th percentiles, and percentages) and probability values from Wilcoxon and Chi-square tests using the HI group as a reference.

Characteristic	Unit	Type of insulin
Characteristic		HI (n = 160)		IA (n = 238)			HI/IA (n = 111)
		Med	5th to 95th	Med	5th to 95th	p	Med	5th to 95th	p
Age	Years	51	28–71	51	25–72	0.63	48	24–71	0.01
Sex	%	53% women		46% women		0.18	41% women		0.04
Diabetes duration	Years	19	3–44	20	5–45	0.25	18	2–44	0.44
BMI	kg/m²	24	20–30	25	20–33	0.16	25	20–31	0.53
HbA1c	%	8	6–10	8	6–10	0.42	8	6–11	0.42
hs-CRP	mg/l	1.7	0.0–8.0	1.4	0.2–9.9	0.71	1.0	0.2–8.3	0.26
Triglycerides	mg/dl	87	43–308	80	40–244	0.01	78	37–182	0.05
Cholesterol	mg/dl	198	136–263	192	139–262	0.04	188	127–241	0.05
HDL	mg/dl	58	34–110	63	39–95	0.33	62	38–95	0.47
LDL	mg/dl	115	65–169	107	63–170	0.18	103	64–160	0.14
eGFR	ml/min	83	34–128	94	45–134	0.01	101	40–137	0.01
Albumin	mg/g	7	2–205	6	1–383	0.01	6	1–233	0.10
Hemoglobin	g/dl	13.9	11.6–16.2	13.7	11.3–15.8	0.02	14.1	11.8–16.3	0.24
Hypertension	%	58		61		0.64	50		0.16
Systolic BP	mmHg	134	104–180	132	110–169	0.95	130	108–163	0.12
Diastolic BP	mmHg	78	60–93	78	61–95	0.47	76	63–92	0.63
ACE inhibitors/AT1 blockers	%	43		47		0.37	39		0.56
Beta blocker	%	17		21		0.27	18		0.31
Statins	%	20		24		0.34	19		0.78

Bold type denotes statistical significance at the 5% confidence level.

ACE, angiotensin-converting-enzyme; BP, blood pressure; BMI, body mass index; eGFR, estimated glomerular filtration rate; HbA1c, hemoglobin A1c; HDL, high-density lipoprotein; HI, human insulin; hs-CRP, high-sensitivity C-reactive protein; IA, insulin analog; LDL, low-density lipoprotein.

The hemoglobin concentration in the IA group was 13.7 g/dl, slightly lower than that in the HI group at 13.9 g/dl (p = 0.02). The renal function (eGFR in ml/min) was 13.3% higher, and the albumin excretion was 19% lower in IA-treated patients (both p = 0.01). Patients treated with a combination of HI/IA also showed a higher average eGFR level (101 ml/min) than the HI group (83 ml/min), but albumin in urine and hemoglobin concentrations were not different (Table 1).

Table 2 shows the results from univariate logistic regression analyses on possible associations with the occurrence of CV complications. According to these results, age, diabetes duration, BMI, lipids, systolic blood pressure, renal function, hemoglobin, hypertension as well as administration of ACE inhibitors or AT1 blockers were taken into account as potential confounders in the subsequent multiple logistic regression analyses.

Table 2.

Univariate analysis of possible associations with CV complication rates in the investigated patient collective.

Explanatory variable	OR	95%	CI	p-value
Age	1.09	1.07	1.11	<.0001
Diabetes duration	1.07	1.05	1.09	<.0001
BMI	1.09	1.02	1.16	0.01
Systolic blood pressure	1.03	1.02	1.04	<.0001
Triglyceride	1.00	1.00	1.01	0.01
HDL cholesterol	0.98	0.97	1.00	0.02
LDL cholesterol	0.99	0.99	1.00	0.05
eGFR	0.97	0.97	0.98	<.0001
Hemoglobin	0.62	0.52	0.74	<.0001
Hypertension	11.4	5.41	24.2	<.0001
ACE inhibitor/AT1 block.	5.29	3.20	8.76	<.0001

Bold type denotes statistical significance at the 5% confidence level.

ACE, angiotensin-converting-enzyme; BMI, body mass index; CI, confidence interval;

eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; OR, odds ratio.

CV complications were observed in 96 patients. The most common complication was a peripheral arterial occlusive disease (43%), followed by cardiac (34%) and cerebrovascular manifestations (23%). The prevalence of CV events in the overall patient collective in the treatment groups HI, IA and HI/IA is represented in Figure 1(a). Treatment with IAs showed a lower CV morbidity (17%) than undergoing therapy with HI alone or HI/IA (21% for both types of treatment). However, treatment differences in CV complication rates considering the above listed potential confounders did not reach statistical significance (p = 0.08, Table 3).

Figure 1.

1) Prevalence of cardiovascular complications in the treatment groups “human insulin (HI)“, “analog-insulin (IA)“ and combination of “HI & IA“. 1 a) = all patients; 1 b = patients with impaired renal function (eGFR < 90 ml/min).

Table 3.

Association between complication rates and the type of insulin treatment.

Insulin treatment	Patients	Events	Univariate logistic regression				Multiple* logistic regression
Insulin treatment	Patients	Events	p	OR	95%	CI	p	OR	95%	CI
All patients
Human	159	33	0.56	Ref.			0.08	Ref.
Analog	236	40		0.78	0.47	1.30		0.90	0.43	1.91
Human and analog	111	23		1.00	0.55	1.81		2.12	0.90	4.98
Human	159	33	0.42	Ref.			0.52	Ref.
Lispro	34	2		0.24	0.05	1.05		0.14	0.02	1.31
Lispro glargine	75	12		0.73	0.35	1.5		1.20	0.44	3.24
Lispro detemir	23	6		1.35	0.49	3.69		1.64	0.38	7.06
Aspart glargine	47	9		0.90	0.40	2.06		0.88	0.26	3.00
Aspart detemir	41	9		1.07	0.47	2.47		1.16	0.38	3.55
Human	159	33	0.57	Ref.			0.87	Ref.
Glargine-based	122	21		0.79	0.43	1.46		1.03	0.42	2.51
Detemir-based	64	15		1.17	0.58	2.34		1.28	0.48	3.39
eGFR < 90 ml/min
Human	103	29	0.29	Ref.			0.06	Ref.
Analog	100	25		0.85	0.46	1.59		0.78	0.31	1.94
Human and analog	45	17		1.55	0.74	3.25		2.65	0.91	7.70
Human	103	29	0.90	Ref.			0.97	Ref.
Lispro	12	2		0.51	0.11	2.47		0.36	0.03	4.98
Lispro glargine	33	7		0.69	0.27	1.76		0.81	0.22	2.98
Lispro detemir	9	2		0.73	0.14	3.72		0.56	0.07	4.53
Aspart glargine	21	6		1.02	0.36	2.89		1.09	0.25	4.83
Aspart detemir	19	6		1.18	0.41	3.39		0.94	0.22	4.14
Human	103	29	0.84	Ref.			0.94	Ref.
Glargine-based	54	13		0.81	0.38	1.73		0.91	0.30	2.75
Detemir-based	28	8		1.02	0.40	2.58		0.79	0.22	2.88

Adjusted for age, diabetes duration, BMI, systolic blood pressure, triglyceride, HDL, LDL, eGFR, hemoglobin, hypertension and ACEAT.

ACEAT, ACE inhibitors/AT1 blockers; BMI, body mass index; CI, confidence interval; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

Figure 1(b) shows the corresponding prevalence in patients with renal impairment (eGFR < 90 ml/min). As expected, CV morbidity was higher in this group of patients than in the whole patient collective [Figure 1(a)]. Patients under IA therapy showed the lowest rate of CV complications (IA 25%; HI 28.2% and HI/IA 37.8%). Differences among treatment groups in CV complication rates were of borderline significance when multiple confounders were taken into account, p = 0.06; Table 3. The odds ratio (OR) of 0.78 for AI-treated patients translates into a 22% decreased risk, and the OR of 2.65 for HI/AI-treated patients translates into a 165% increased risk of CV complications compared with HI-treated patients. No differences were found in patients with normal renal function (eGFR > 90 ml/min; data not shown).

To investigate possible differences in CV morbidity among the types of IA treatment, patients were accordingly stratified. The most common combinations of IAs were lispro/glargine (32%) and aspart/glargine (20%). A total of 11 patients treated with insulin glulisine were not separately considered because of the low number.

Figure 2 shows the prevalence of CV complications according to the type of IA therapy. Patients treated with lispro alone or in combination with glargine presented the lowest CV morbidity compared with other IA combinations, both in the whole patient collective [Figure 2(a), 5.9% and 16% respectively] and in patients with renal impairment [Figure 2(b), 16.7% and 21.2% respectively]. However, differences among CV complication rates adjusted for potential confounders did not reach statistical significance, as shown in Table 3.

Figure 2.

1) Prevalence of cardiovascular complications in the treatment groups “human insulin (HI)“, “analog-insulin (IA)“ and combination of “HI und IA“. 2 a) = all patients; 2 b = patients with impaired renal function (eGFR < 90 ml/min).

When patients were stratified according to basal insulin treatment, observed CV complication rates were 17% in the glargine-based group and 23% in detemir-based group, compared with 21% for HI-treated patients. The most common combination partner for bolus insulin was lispro (62%) in the glargine group and aspart (63%) in the detemir group. Similar ratios of CV complication rates were found among patients with impaired renal function (Table 3).

Discussion

This study shows for the first time that treatment of people with type 1 diabetes with IAs seems to be associated with a lower prevalence of CV events (17%) compared with therapy with HI or a combination of HI/IA (both 21%, p = 0.08). In the stratified analysis of patients with renal impairment, the difference showed borderline significance (p = 0.06) after adjustment for multiple potential confounders.

In the examination of the different IAs, CV morbidity was the lowest for lispro (5.9%) and lispro/glargine (16%). Similar findings were also found in the examination of the subgroup with impaired renal function. However, the sample size of the study was relatively low, particularly considering the large number of associated confounders.

Although IAs have been used in diabetic therapy for two decades, no studies are available for type 1 diabetes on the influence of different IAs on the CV system. For patients with type 2 diabetes, various retrospective observation studies were published in which databases from health insurance funds and general practitioners were evaluated. A reduced incidence of CV complications could be observed for glargine compared with long-acting human insulin.^5–7 The use of short-acting IAs also showed favorable effects on CV morbidity compared with HI.^8–11 The advantage of these investigations is the availability of large amounts of data. A major drawback, however, is the fact that the data is only based on International Classification of Diseases encodings and therefore does not contain any further information for example, diabetes status, blood pressure adjustment, lipids, and other CV risk factors. In the Origin study, no difference was seen in the occurrence of CV complications with early insulin treatment with glargine.¹³ However, the patient group included in this study differed fundamentally from that of the studies cited above.

In our study of patients with type 1 diabetes, a large number of individual findings and possible covariates could be used due to a structured recording of the patient-related data and the laboratory values during the initial examination. Our results tend to be consistent with the reported findings in type 2 diabetes, that is, lower CV morbidity when treated with IA compared with HI. The borderline significance is partially due to a low number of CV events in the younger type 1 diabetic patients. Another reason could be that the IA exposure time of an average of 3 years is not sufficient to detect the influence of these insulins on the development of macroangiopathic complications in these patients. In a previous study of the same patient group on the influence of IAs on the prevalence of nephropathy, patients under IA treatment showed a significantly better renal status (eGFR, albumin excretion) than the patients treated with HI.¹² The temporally different response of a treatment to micro- and macroangiopathic complications corresponds to the findings of the DCCT/EDIC study in which the influence of an intensified versus conventional insulin therapy was noticeable much earlier in the occurrence of a retinopathy than a CV complication.^14,15

After stratification of the patients per the IA used, the CV morbidity was lowest when using lispro alone or in combination with glargine, but the difference was not statistically significant for the reasons already indicated. In the aforementioned study on nephropathy, a similar pattern of results has been described: patients treated with lispro or lispro/glargine showed significantly better renal function parameters and higher Hb concentration compared with HI treatment. As it is known that renal status is a good predictor for the development of CV complications it can be discussed whether the lower CV morbidity observed here under treatment with lispro/glargine is the result of better renal status of these patients.^16,17

The reasons for a possibly protective influence of IA on CV morbidity in patients with diabetes are not fully understood; a reduction in glycemic variability under treatment with short-acting IAs is at the forefront of discussions. This is accompanied by a decrease in oxidative stress for the vessels and an improvement in endothelial function. Thus, a reduction of the postprandial glucose increase by the short-acting IA aspart was shown to be associated with a significant improvement in myocardial perfusion and diastolic heart function.^18,19 The microvascular circulation of the skin showed that the IAs lispro and glulisine also improve the microvascular circulation by reducing the postprandial increase in blood glucose.^20,21 The reduction of the risk of hypoglycemia under the treatment with long-acting IAs compared with human NPH insulin is also a factor in reducing CV risk in insulin-dependent patients.^1,2 The extent to which further factors play a role here will require further research.

This study has several limitations. (1) the investigated patients were assigned to the diabetes outpatient clinic or diabetic day clinic for various reasons (diabetes training, metabolic problems, insulin conversion, etc.) from general medical or diabetes centers. Therefore, a ‘selection bias’ cannot be ruled out, even if the longer exposure period and the greater number of examined patients can reduce the influence of this factor. (2) The treatment period with the respective IA could not be exactly defined as discussed above. However, the IA exposure time in the majority of patients could be limited through intensive questioning of the patients and, if necessary, queries in the practices in which the patients were adjusted to IA. (3) Information on some factors important to the development of macroangiopathic complications (e.g. frequency of severe hypoglycemia, cigarette and alcohol consumption, physical and occupational activity, antiplatelet treatment or other concomitant complications such as neuropathy as well as social status) were not systematically recorded. (4) The cross-sectional design was another limitation of the present study. Validation of reported findings in future large-scale, rich-data, prospective studies is needed in advance to clinical translation.

In summary, we show for the first time that people with type 1 diabetes, especially those with impaired renal function, seem to present with a lower CV morbidity when treated with IA compared with HI or HI/IA administration. Validation of these preliminary findings in confirmatory, prospective studies may have important clinical implications.

Footnotes

Funding

The study was kindly supported by an unrestricted grant of Manfred Lautenschläger Stiftung Diabetes, Wiesloch, Gemany and by Sanofi Deutschland, Berlin, Germany.

Conflict of interest statement

The authors declare that there is no conflict of interest.

References

Zoungas

Patel

Chalmers

et al . Severe hypoglycemia and risks of vascular events and death. N Engl J Med 2010; 363: 1410–1418.

Giménez

Gilabert

Monteagudo

et al . Repeated episodes of hypoglycemia as a potential aggravating factor for preclinical atherosclerosis in subjects with type 1 diabetes. Diabetes Care 2011; 34: 198–203.

Coutinho

Gerstein

Wang

et al . The relationship between glucose and incident cardiovascular events. A metaregression analysis of published data from 20 studies of 95,783 individuals followed for 12.4 years. Diabetes Care 1999; 22: 233–240.

Hanefeld

Fischer

Julius

et al . Risk factors for myocardial infarction and death in newly detected NIDDM: the Diabetes Intervention Study, 11-year follow-up. Diabetologia 1996; 39: 1577–1583.

Rhoads

Kosiborod

Nesto

et al . Comparison of incidence of acute myocardial infarction in patients with type 2 diabetes mellitus following initiation of neutral protamine Hagedorn insulin versus insulin glargine. Am J Cardiol 2009; 104: 910–916.

Juhaeri

Gao

Dai

. Incidence rates of heart failure, stroke, and acute myocardial infarction among type 2 diabetic patients using insulin glargine and other insulin. Pharmacoepidemiol Drug Saf 2009; 18: 497–503.

Cammarota

Bruzzese

Catapano

et al . Lower incidence of macrovascular complications in patients on insulin glargine versus those on basal human insulins: a population-based cohort study in Italy. Nutr Metab Cardiovasc Dis 2014; 24: 10–17.

Cammarota

Falconio

Bruzzese

et al . Lower rate of cardiovascular complications in patients on bolus insulin analogues: a retrospective population-based cohort study. PLoS One 2013; 8: e79762.

Rathmann

Kostev

. Lower incidence of recorded cardiovascular outcomes in patients with type 2 diabetes using insulin aspart vs. those on human regular insulin: observational evidence from general practices. Diabetes Obes Metab 2013; 15: 358–363.

10.

Scholz

Dippel

. Incidence of cardiovascular outcomes in type 2 diabetic patients treated with insulin glulisine or other rapid acting insulin analogs under real-life conditions: a retrospective database analysis. Diabetol Stoffwechsel 2012; 7: 302–307.

11.

Raz

Ceriello

Wilson

et al . Post hoc subgroup analysis of the HEART2D trial demonstrates lower cardiovascular risk in older patients targeting postprandial versus fasting/premeal glycemia. Diabetes Care 2011; 34: 1511–1513.

12.

Hasslacher

Kulozik

Lorenzo Bermejo

. Treatment with insulin analogs, especially Glargine and Lispro, associates with better renal function and higher hemoglobin levels in Type 1 diabetic patients with impaired kidney function. Ther Adv Endocrinol Metab 2016; 7: 166–177.

13.

Gilbert

Mann

JFE

Hanefeld

et al .; ORIGIN Trial Investigators. Basal insulin glargine and microvascular outcomes in dysglycaemic individuals: results of the Outcome Reduction with an Initial Glargine Intervention (ORIGIN) trial. Diabetologia 2014; 57: 1325–1331.

14.

DCCT Group. The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the diabetes control and complications trial. Diabetes 1995; 44: 968–983.

15.

Lachin

Orchard

Nathan

.; DCCT/EDIC Research Group. Update on cardiovascular outcomes at 30 years of the diabetes control and complications trial/epidemiology of diabetes interventions and complications study. Diabetes Care 2014; 37: 39–43.

16.

Sasso

Chiodini

Carbonara

et al . High cardiovascular risk in patients with type 2 diabetic nephropathy: the predictive role of albuminuria and glomerular filtration rate. The NID-2 Prospective Cohort Study. Nephrol Dial Transplant 2012; 27: 2269–2274.

17.

Svensson

Cederholm

Eliasson

et al .; Swedish National Diabetes Register. Albuminuria and renal function as predictors of cardiovascular events and mortality in a general population of patients with type 2 diabetes: a nationwide observational study from the Swedish National Diabetes Register. Diab Vasc Dis Res 2013; 10: 520–529.

18.

Scognamiglio

Negut

de Kreutzenberg

et al . Effects of different insulin regimes on postprandial myocardial perfusion defects in type 2 diabetic patients. Diabetes Care 2006; 29: 95–100.

19.

von Bibra

Siegmund

Kingreen

et al . Effects of analogue insulin in multiple daily injection therapy of type 2 diabetes on postprandial glucose control and cardiac function compared to human insulin: a randomized controlled long-term study. Cardiovasc Diabetol 2016; 15: 7.

20.

Forst

Strunk

et al . Impact of insulin on microvascular blood flow and endothelial cell function in the postprandial state in patients with type 1 diabetes. J Diabetes Complications 2005; 19: 128–132.

21.

Hohberg