Review article

Impaired vascular relaxation in type 2 diabetes: A systematic review and meta-analysis

Sajad Jeddi1, Zahra Bahadoran2, Parvin Mirmiran2, Khosrow Kashfi3[*], Asghar Ghasemi1

1Endocrine Physiology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran

2Nutrition and Endocrine Research Center, Research Institute for Endocrine Sciences,Shahid Beheshti University of Medical Sciences, Tehran, Iran

3Department of Molecular, Cellular, and Biomedical Sciences, Sophie Davis School of Biomedical Education, City University of New York School of Medicine, NY, USA

EXCLI J 2024;23:Doc937

 

Abstract

Type 2 diabetes (T2D) significantly increases the risk of vascular complications (12-32 %), which are a major cause of death (over 50 %) in T2D patients. In T2D, both endothelial (ET) and vascular smooth muscle (VSM) cells are impaired, which act as independent risk factors for cardiovascular disease. Thus, the question of this systematic review and meta-analysis is: Do ET-dependent and -independent VSM relaxation impair in T2D? We systematically searched PubMed and Scopus databases until March 2024; 44 eligible clinical trial studies (68, 16, 30, and 50 study arms for acetylcholine (ACh), methacholine (MTH), sodium nitroprusside (SNP), and glyceryl trinitrate (GTN)) published were included. ET-dependent VSM relaxation in response to ACh (overall ES = -28.9 %, 95 % CI: -35.2, -22.7; p<0.001) and MTH (overall ES = -55.3 %, 95 % CI: -63.6, -47.1; p<0.001) decreased in T2D patients compared to controls. ET-independent VSM relaxation in response to SNP (overall ES = -17.2 %, 95 % CI: -35.2, -22.7; p<0.001) and GTN (overall ES = -63.2 %, 95 % CI: -81.0, -45.5; p<0.001) decreased in T2D patients compared to controls. Our meta-analysis showed reductions in both ET-dependent (~40 %) and ET-independent (~25 %) VSM relaxation. The decrease was more pronounced for MTH (~55 %) compared to ACh (~30 %) and for GTN (~63 %) compared to SNP (~17 %). These findings suggest that dysfunction of both ET and VSM contributes to impaired VSM relaxation in T2D patients.

See also the graphical abstract(Fig. 1).

Keywords: Type 2 diabetes, vascular complications, vascular smooth muscle, acetylcholine, methacholine, sodium nitroprusside, glyceryl trinitrate

Introduction

Globally, over 536 million adults (10 %) had type 2 diabetes (T2D) in 2021, and it is estimated that this number will rise to 783.2 million (12 %) by 2045 (Sun et al., 2022[72]). Patients with T2D have a 12-32 % risk of vascular complications (Einarson et al., 2018[21]; Aikaeli et al., 2022[2]), which increase cardiovascular morbidity and mortality by 2-4 times compared to those without T2D (Creager et al., 2003[18]; Lüscher et al., 2003[48]) and are responsible for over 50 % of deaths in T2D patients (Einarson et al., 2018[21]). In T2D, both endothelial and vascular smooth muscle (VSM) cells are impaired, which act as independent risk factors for cardiovascular diseases (CVD) in T2D (Ganz and Vita, 2003[24]; Lerman and Zeiher, 2005[46]; Kawano et al., 2012[40]).

Endothelium (ET) produces several vasoactive substances to maintain vascular homeostasis (Park and Park, 2015[58]), of which nitric oxide (NO) is the most important (Kelm, 1999[42]). ET produces about 40-60 % of the whole-body NO (i.e., 1100 μmol/day) (Ghasemi and Jeddi, 2022[27]), with over 60 % of this NO transferred to neighboring VSMs in the blood vessel wall (Malinski et al., 1993[50]). Vascular NO resistance in T2D is characterized by decreased NO production by ET, increased NO inactivation, and impaired responsiveness of VSMs to vasodilatory effects of NO (Bahadoran et al., 2023[5]). Vascular NO resistance, independent of other well-known risk factors (Schächinger et al., 2000[65]; Halcox et al., 2002[31]), is associated with future cardiovascular events such as heart attack and CVD mortality (Bahadoran et al., 2023[5]). Understanding the link between VSM dysfunction and NO is important for developing strategies to reduce the risk of life-threatening events.

We recently reviewed the evidence for and the underlying mechanisms of impaired vascular relaxation in animals and humans with T2D (Bahadoran et al., 2023[5]). Acetylcholine (ACh) and methacholine (MTH)-related VSM relaxation decreased in T2D (Caballero et al., 1999[14]; Brooks et al., 2008[10]) by ~13-94 % (Bahadoran et al., 2023[5]). However, VSM relaxation in responses to sodium nitroprusside (SNP) and glyceryl trinitrate (GTN) in T2D is controversial; some studies report decreases of 6-42 % (Natali et al., 2006[54]; Beer et al., 2008[7]), and others show preserved (Steinberg et al., 1996[70], 2000[71]) or even increased relaxation (Goodfellow et al., 1996[29]; Steinberg et al., 1996[70]). A meta-analysis conducted in 2013 reported that ET-independent VSM relaxation significantly decreased in T2D patients compared with controls [standardized mean difference (SMD) = −0.68; 95 % confidence interval (CI) −0.84, −0.52)] (Montero et al., 2013[52]). However, this analysis did not consider several factors: ET-dependent vasodilation (ACh and MTH-related VSM relaxation), separate main effects and dose-response analyses of each vasodilator (ACh, MTH, SNP, and GTN), and the duration of T2D. ET-dependent (in response to ACh and MTH) and ET-independent (in response to SNP and GTN) VSM relaxation in T2D are influenced by the type of vasodilators used and the duration or severity of the disease (Bahadoran et al., 2023[5]). Thus, the question of this systematic review and meta-analysis is: Do endothelium-dependent and -independent VSM relaxation impair in patients with T2D? We also assess the association between vasodilation and duration of diabetes.

Materials and Methods

Search strategy and study selection

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 updated guideline (Page et al., 2021[57]) to design and perform this study and analyze and report the results. Databases (PubMed and Scopus) were searched until March 2024 using the combination of the following search terms in Title/Abstract: (Diabetes) AND [(acetylcholine) OR (methacholine) OR (nitroglycerin) OR (sodium nitroprusside) OR (glyceryl trinitrate) OR (trinitroglycerin) OR (endothelium-dependent vascular relaxation OR (endothelium-independent vascular relaxation OR (forearm blood flow OR (brachial artery)].

The records were extracted from each database to Endnote version 18 software, and the duplicate reports were removed. The first screening step evaluated titles and abstracts to exclude irrelevant articles. Excluded articles were studies on non-human subjects, studies not in English, review articles (including systematic reviews and meta-analyses), and other report types besides journal articles (e.g., conference abstracts, comments, letters to the editor, and books). Full texts of the relevant studies were then sought and screened cautiously and thoroughly for their eligibility, and studies on type 1 diabetes were excluded. All publications assessed the effect of ET-dependent and -independent VSM relaxation in T2D patients were included. Eventually, eligible studies (44 studies with 68, 16, 30, and 50 study arms for ACh, MTH, SNP, and GTN, respectively) were considered for final inclusion (Figure 2(Fig. 2)). The references of the final included studies were also screened to reduce the risk of missing relevant studies.

Data extraction

Information extracted included the first author's name, year of publication, age, sex (% of men), sample size, duration of T2D, vascular region assessed, and dose of vasodilators used. The mean and standard deviation (SD) of response to vasodilators were also extracted. If a study had measured response to vasodilators following several doses and/or durations of T2D, each measurement was considered a separate experimental study. These data were extracted from the article texts or tables; if a study had represented the results only as a graph, the data was extracted by the Adobe Photoshop software using a previously described method (Gheibi et al., 2019[28]).

Statistical analysis

The Weighted Mean Difference (WMD) was used as the primary outcome for measuring the difference in VSM relaxation in response to ET-dependent (ACh and MTH) and ET-independent (SNP and GTN) vasodilators. WMD was used as the effect size (ES) instead of SMD because all included studies reported the vasodilatory effect of ACh, MTH, SNP, and GTN as a percentage change. For studies lacking the percentage change data, it was calculated from the reported raw data.

The meta-analysis was performed after separating ACh, MTH, SNP, and GTN data. STATA version 17 was used for statistical analysis and making illustrations. The random-effect model was chosen instead of the fixed-effect model due to its suitability when significant heterogeneity (variation) exists between studies. The WMD was analyzed and reported separately with a 95 % CI for each vasodilator (ACh, MTH, SNP, and GTN). Two-sided P-values<0.05 and between 0.05 and 0.10 were considered statistically and marginally significant, respectively.

Heterogeneity was assessed using I² (I-squared), which quantifies the proportion of variation in effect sizes attributable to factors beyond sampling error (percentage of heterogeneity indicates the extent of the studies that the CI of their outcome does not overlap with the overall ES), and Cochran's Q test (test of homogeneity), which statistically evaluates the likelihood that the observed differences in study outcomes are due to chance and representing the probability that the outcomes of the studies are homogenous (Higgins et al., 2003[35]). The results of I² are reported and interpreted according to the established categories: 0-25 % = no significant heterogeneity, 25-50 % = low heterogeneity, 50-75 % = moderate heterogeneity, and >75 % = high heterogeneity (Higgins et al., 2003[35]). The P-value is reported for Cochran's Q test, and a value less than 0.1 typically suggests significant heterogeneity. Publication bias was evaluated using Egger's regression test and funnel plots (Egger et al., 1997[20]). A funnel plot is a scatter plot of the intervention effect estimates from individual studies (X-axis) against the standard error of the estimated effect (Y-axis). Egger's test examines potential publication bias in a meta-analysis by examining funnel plot asymmetry. Specifically, Egger's test involves a linear regression of the intervention effect estimates on their standard errors, weighted by their inverse variance (Egger et al., 1997[20]).

Random effects meta-regression was performed to explore potential sources of heterogeneity among the included studies. This analysis examined the dose of each vasodilator (ACh, MTH, SNP, and GTN) and the duration of T2D as possible influencing factors. For meta-regression, the R2 was reported, indicating the amount of heterogeneity explained by the moderator; if the dose of vasodilators and duration of T2D were not available for a study, the study was excluded from the meta-regression analysis. Additionally, subgroup analysis was conducted based on the type of vessels studied to determine if the impairment varied depending on the specific vessel type (Yugar-Toledo, 2004[85]).

Variability in vasodilator responses across studies was observed due to differences in the administered doses and the duration of the study. To address this variation and provide a more precise evaluation of the studied vasodilators' effects, separate dose-response analyses were conducted for each vasodilator, including ACh, MTH, SNP, and GTN.

Results

Characteristic of studies: ET-dependent VSM relaxation

Twenty-two studies assessed the effect of ACh on VSM relaxation in T2D patients (McVeigh et al., 1992[51]; Watts et al., 1996[80]; Avogaro et al., 1997[3]; Hogikyan et al., 1998[36]; Caballero et al., 1999[14]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Heitzer et al., 2000[33]; Preik et al., 2000[62]; Kingwell et al., 2003[44]; Pistrosch et al., 2004[60]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Yeh et al., 2009[84]; Chousos et al., 2010[16]; Bock et al., 2021[8]). These studies involved 664 patients with T2D and 415 control subjects. The mean age of T2D patients in these studies was 56.2±4.9 years with a range of 42 (Avogaro et al., 1997[3]) to 65 (Butler et al., 2000[13]) years, and the mean age of control subjects was 53.8±5.8 years with a range of 38 (Avogaro et al., 1997[3]) to 58 (Schmiedel et al., 2007[66]) years. Nineteen studies included both sexes (McVeigh et al., 1992[51]; Hogikyan et al., 1998[36]; Caballero et al., 1999[14]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Heitzer et al., 2000[33]; Preik et al., 2000[62]; Pistrosch et al., 2004[60]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Beer et al., 2008;[7] Brooks et al., 2008[10]; Yeh et al., 2009[84]; Chousos et al., 2010[16]; Bock et al., 2021[8]), and three studies were conducted in men (Watts et al., 1996[80]; Avogaro et al., 1997[3]; Kingwell et al., 2003[44]). The mean duration of T2D was 6.5±2.4 years, with a range of 3.6 (Watts et al., 1996[80]) to 10.8 (Beer et al., 2008[7]) years. The ACh was administered in the brachial artery (McVeigh et al., 1992[51]; Watts et al., 1996[80]; Avogaro et al., 1997[3]; Hogikyan et al., 1998[36]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Heitzer et al., 2000[33]; Preik et al., 2000[62]; Pistrosch et al., 2004[60]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Yeh et al., 2009[84]; Bock et al., 2021[8]), femoral artery (Kingwell et al., 2003[44]), and forearm skin microcirculation (FMS) (Caballero et al., 1999[14]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Chousos et al., 2010[16]). The ACh dose range used was 0.45 (Natali et al., 2006[54]) and 35 (McVeigh et al., 1992[51]) µg/min. Characteristics of studies with more details are provided in Table 1(Tab. 1) (References in Table 1: Avogaro et al., 1997[3]; Beckman et al., 2010[6]; Beer et al., 2008[7]; Bock et al., 2021[8]; Brooks et al., 2008[10]; Butler et al., 2000[13]; Caballero et al., 1999[14]; Chousos et al., 2010[16]; Gazis et al., 1999[26]; Heitzer et al., 2000[33]; Hogikyan et al., 1998[36]; Kingwell et al., 2003[44]; McVeigh et al., 1992[51]; Natali et al., 2006[54]; Pistrosch et al., 2004[60]; Preik et al., 2000[62]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Steinberg et al., 1996[70], 2000[71]; Ting et al., 1996[74]; Vavuranakis et al., 1999[77]; Vehkavaara and Yki-Järvinen, 2004[78]; Watts et al., 1996[80]; Williams et al., 1996[81]; Woodman et al., 2006[83]; Yeh et al., 2009[84]).

The effect of MTH on VSM relaxation in T2D patients has been reported in five studies (Steinberg et al., 1996[70]; Ting et al., 1996[74]; Williams et al., 1996[81]; Steinberg et al., 2000[71]; Beckman et al., 2010[6]). These studies involved 60 patients with T2D and 105 control subjects. The mean age of T2D patients in these studies was 42.1±5.5 years with a range of 36 (Steinberg et al., 2000[71]) to 53 (Beckman et al., 2010[6]) years, and the mean age of control subjects was 38.5±4.9 years with a range of 34 (Schächinger et al., 2000[65]) to 44 (Beckman et al., 2010[6]) years. Three studies included both sexes (Ting et al., 1996[74]; Williams et al., 1996[81]; Beckman et al., 2010[6]), one study was conducted in men (Steinberg et al., 2000[71]), and one study did not specify sex (Steinberg et al., 1996[70]). The duration of T2D was reported only in 2 studies (3.7 (Ting et al., 1996[74]) and 4 (Williams et al., 1996[81]) years). The MTH was administered in brachial artery (Ting et al., 1996[74]; Williams et al., 1996[81]; Beckman et al., 2010[6]), femoral artery (Steinberg et al., 2000[71]), and FMS (Steinberg et al., 1996[70]). The MTH dose range used was 0.3 (Beckman et al., 2010[6]) and 15 (Steinberg et al., 2000[71]) µg/min.

Meta-analyses: ET-dependent VSM relaxation

As shown in Figures 3(Fig. 3) and 4(Fig. 4), ET-dependent VSM relaxation in response to ACh (overall ES = -28.9 %, 95 % CI: -35.2, -22.7; p<0.001) (Figure 3(Fig. 3); References in Figure 3: Avogaro et al., 1997[3]; Beer et al., 2008[7]; Bock et al., 2021[8]; Brooks et al., 2008[10]; Butler et al., 2000[13]; Caballero et al., 1999[14]; Chousos et al., 2010[16]; Gazis et al., 1999[26]; Heitzer et al., 2000[33]; Hogikyan et al., 1998[36]; Kingwell et al., 2003[44]; McVeigh et al., 1992[51]; Natali et al., 2006[54]; Pistrosch et al., 2004[60]; Preik et al., 2000[62]; Schmiedel et al., 2007[66]; Sivitz et al., 2007;[68] Vavuranakis et al., 1999[77]; Vehkavaara and Yki-Järvinen, 2004[78]; Watts et al., 1996[80]; Woodman et al., 2006[83]; Yeh et al., 2009[84]) and MTH (overall ES = -55.3 %, 95 % CI: -63.6, -47.1; p<0.001) (Figure 4(Fig. 4); References in Figure 4: Beckman et al., 2010[6]; Steinberg et al., 1996[70], 2000[71]; Ting et al., 1996[74]; Williams et al., 1996[81]) decreased in T2D patients compared to controls. In addition, high heterogeneity was observed in this analysis, indicated by I² values of 93.8 % (p<0.001) and 99.9 % (p<0.001) for ACh and MTH, respectively. Overall, ET-dependent VSM relaxation in response to ACh/MTH in T2D patients decreased by 39.2 % (95 % CI: -46.0, -33.5; p<0.001).

Meta-regression, subgroup analysis, and publication bias: ET-dependent VSM relaxation

ET-dependent VSM relaxation in response to ACh has not been affected by the dose of ACh (p=0.875, B = -0.064; Supplementary Figure 1Aexcli2024-7330_supplementary_information.pdf) and the duration of T2D (p=0.198, B = 1.96; Supplementary Figure 1Bexcli2024-7330_supplementary_information.pdf). In contrast, ET-dependent VSM relaxation in response to MTH affected by dose of MTH (p=0.007, B = 2.25; Supplementary Figure 1Cexcli2024-7330_supplementary_information.pdf) and duration of T2D (p= 0.058, B = 29.84; Supplementary Figure 1Dexcli2024-7330_supplementary_information.pdf); higher doses of MTH and longer duration of T2D were associated with a lower WMD (lower impairment in response to MTH) of VSM relaxation between T2D patients and controls.

The test of group differences was statistically significant between the subgroups for the type of vessels in studies that used ACh (p=0.004) but not for MTH (p=0.266) (Table 2(Tab. 2)).

The funnel plot for the mean difference in the VSM function of studies included in the meta-analysis was asymmetrical (Supplementary Figure 2Aexcli2024-7330_supplementary_information.pdf for ACh and 2B for MTH). However, Egger's test detected that the observed asymmetry was not due to publication bias (p=0.529, p=0.504 for ACh and MTH, respectively).

Characteristic of studies: ET-independent VSM relaxation

Twenty-eight studies assessed the effect of SNP on VSM relaxation in T2D patients (Steinberg et al., 1996[70]; Ting et al., 1996[74]; Watts et al., 1996[80]; Williams et al., 1996[81]; Avogaro et al., 1997[3]; O'Brien et al., 1997[55]; Hogikyan et al., 1998[36]; Caballero et al., 1999[14]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Preik et al., 2000[62]; Steinberg et al., 2000[71]; Van de Ree et al., 2001[75]; Van Etten et al., 2002[76]; Kingwell et al., 2003[44]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Sokolnicki et al., 2007[69]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Yeh et al., 2009[84]; Chousos et al., 2010[16]; Shemyakin et al., 2012[67]; Mahdi et al., 2018[49]). These studies involved 724 patients with T2D and 519 control subjects. The mean age of T2D patients in these studies was 54.6±7.4 years with a range of 36 (Steinberg et al., 2000[71]) to 53 (Beckman et al., 2010[6]) years, and the mean age of control subjects was 51.7±8.2 years with a range of 28 (Steinberg et al., 2000[71]) to 66 (Schmiedel et al., 2007[66]) years. Nineteen studies included both sexes (Ting et al., 1996[74]; Williams et al., 1996[81]; Hogikyan et al., 1998[36]; Caballero et al., 1999[14]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Preik et al., 2000[62]; Van Etten et al., 2002[76]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sokolnicki et al., 2007[69]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Yeh et al., 2009[84]; Chousos et al., 2010[16]; Mahdi et al., 2018[49]); seven studies were conducted only in men (Watts et al., 1996[80]; Avogaro et al., 1997[3]; O'Brien et al., 1997[55]; Steinberg et al., 2000[71]; Van de Ree et al., 2001[75]; Kingwell et al., 2003[44]; Shemyakin et al., 2012[67]); one study was conducted only in women (Sivitz et al., 2007[68]), and one study did not specify sex (Steinberg et al., 1996[70]). The mean duration of T2D was 6.6±2.5 years with a range of 3.6 (Watts et al., 1996[80]) to 11 (Mahdi et al., 2018[49]) years. The SNP were administered in brachial artery (Ting et al., 1996[74]; Watts et al., 1996[80]; Williams et al., 1996[81]; Avogaro et al., 1997[3]; O'Brien et al., 1997[55]; Hogikyan et al., 1998[36]; Gazis et al., 1999[26]; Vavuranakis et al., 1999[77]; Butler et al., 2000[13]; Preik et al., 2000[62]; Van de Ree et al., 2001[75]; Van Etten et al., 2002[76]; Vehkavaara and Yki-Järvinen, 2004[78]; Natali et al., 2006[54]; Woodman et al., 2006[83]; Schmiedel et al., 2007[66]; Sivitz et al., 2007[68]; Yeh et al., 2009[84]; Shemyakin et al., 2012[67]; Mahdi et al., 2018[49]), femoral artery (Steinberg et al., 2000[71]), femoral vein (Kingwell et al., 2003[44]), and FMS (Steinberg et al., 1996[70]; Caballero et al., 1999[14]; Sokolnicki et al., 2007[69]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Chousos et al., 2010[16]). The SNP dose range was 0.01 (Caballero et al., 1999[14]) and 10 (Avogaro et al., 1997[3]) µg/min. Characteristics of studies with more details are provided in Table 3(Tab. 3) (References in Table 3: Avogaro et al., 1997[3]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Butler et al., 2000[13]; Caballero et al., 1999[14]; Chousos et al., 2010[16]; Enderle et al., 1998[22]; Furuta et al., 2013[23]; Gazis et al., 1999[26]; Goodfellow et al., 1996[29]; Henry et al., 2004[34]; Hogikyan et al., 1998[36]; Huvers et al., 1997[38]; Kimura et al., 2001[43]; Kingwell et al., 2003[44]; Liu et al., 2018[47]; Mahdi et al., 2018[49]; McVeigh et al., 1992[51]; Natali et al., 2006[54]; O'Brien et al., 1997[55]; Preik et al., 2000[62]; Schmiedel et al., 2007[66]; Shemyakin et al., 2012[67]; Sivitz et al., 2007[68]; Sokolnicki et al., 2007[69]; Steinberg et al., 1996[70], 2000[71]; Tan et al., 2004[73]; Ting et al., 1996[74]; Van de Ree et al., 2001[75]; Van Etten et al., 2002[76]; Vavuranakis et al., 1999[77]; Vehkavaara and Yki-Järvinen, 2004[78]; Watts et al., 1996[80]; Williams et al., 1996[81]; Woodman et al., 2006[83]; Yeh et al., 2009[84]; Yugar-Toledo, 2004[85]).

The effect of GTN on VSM relaxation has been reported in thirteen studies (McVeigh et al., 1992[51]; Goodfellow et al., 1996[29]; Huvers et al., 1997[38]; Enderle et al., 1998[22]; Vavuranakis et al., 1999[77]; Kimura et al., 2001[43]; Henry et al., 2004[34]; Tan et al., 2004[73]; Woodman et al., 2006[83]; Yeh et al., 2009[84]; Furuta et al., 2013[23]; Yugar-Toledo, 2004[85]; Liu et al., 2018[47]). These studies involved 859 patients with T2D and 583 control subjects. The mean age of T2D patients in these studies was 56.4±7.4 years with a range of 42.7 (Liu et al., 2018[47]) to 70 (Kimura et al., 2001[43]) years, and the mean age of control subjects was 52.8±8.6 years with a range of 42 (Furuta et al., 2013[23]) to 68 (Henry et al., 2004[34]) years. Eleven studies included both sexes (McVeigh et al., 1992[51]; Goodfellow et al., 1996[29]; Huvers et al., 1997[38]; Enderle et al., 1998[22]; Vavuranakis et al., 1999[77]; Kimura et al., 2001[43]; Henry et al., 2004[34]; Tan et al., 2004[73]; Woodman et al., 2006[83]; Yeh et al., 2009[84]; Furuta et al., 2013[23]); one study was conducted only in women (Liu et al., 2018[47]), and one study did not specify sex ( Yugar-Toledo, 2004[85]). The mean duration of T2D was 7.7±5.2 years with a range of 3.8 (Goodfellow et al., 1996[29]) to 19.7 (Furuta et al., 2013[23]) years. The GTN was administered in the brachial artery (McVeigh et al., 1992[51]; Goodfellow et al., 1996[29]; Huvers et al., 1997[38]; Enderle et al., 1998[22]; Kimura et al., 2001[43]; Henry et al., 2004[34]; Tan et al., 2004[73]; Woodman et al., 2006[83]; Yeh et al., 2009[84]; Furuta et al., 2013[23]; Yugar-Toledo, 2004[85]; Liu et al., 2018[47]) and coronary artery (Vavuranakis et al., 1999[77]). The range of doses of GTN used were 68 (McVeigh et al., 1992[51]) and 400 (Goodfellow et al., 1996[29]) µg/min. Characteristics of studies with more details are provided in Table 3(Tab. 3).

Meta-analyses: ET-independent VSM relaxation

ET-independent VSM relaxation in response to SNP (overall ES = -17.2 %, 95 % CI: -35.2, -22.7; p<0.001) (Figure 5(Fig. 5); References in Figure 5: Avogaro et al., 1997[3]; Beer et al., 2008[7]; Brooks et al., 2008[10]; Butler et al., 2000[13]; Caballero et al., 1999[14]; Chousos et al., 2010[16]; Gazis et al., 1999[26]; Hogikyan et al., 1998[36]; Kingwell et al., 2003[44]; Mahdi et al., 2018[49]; Natali et al., 2006[54]; O'Brien et al., 1997[55]; Preik et al., 2000[62]; Schmiedel et al., 2007[66]; Shemyakin et al., 2012[67]; Sivitz et al., 2007[68]; Sokolnicki et al., 2007[69]; Steinberg et al., 1996[70], 2000[71]; Ting et al., 1996[74]; Van de Ree et al., 2001[75]; Van Etten et al., 2002[76]; Vavuranakis et al., 1999[77]; Vehkavaara and Yki-Järvinen, 2004[78]; Watts et al., 1996[80]; Williams et al., 1996[81]; Woodman et al., 2006[83]; Yeh et al., 2009[84]) and GTN (overall ES = -63.2 %, 95 % CI: -81.0, -45.5; p<0.001) (Figure 6(Fig. 6); References in Figure 6: Enderle et al., 1998[22]; Furuta et al., 2013[23]; Goodfellow et al., 1996[29]; Henry et al., 2004[34]; Huvers et al., 1997[38]; Kimura et al., 2001[43]; Liu et al., 2018[47]; McVeigh et al., 1992[51]; Tan et al., 2004[73]; Vavuranakis et al., 1999[77]; Woodman et al., 2006[83]; Yeh et al., 2009[84]; Yugar-Toledo, 2004[85]) decreased in T2D patients compared to controls. In addition, high heterogeneity was observed in these analyses, indicated by I² values of 97.7 % (p<0.001) and 98.6 % (p<0.001) for SNP and GTN, respectively. Overall, ET-independent VSM relaxation in response to SNP/GTN in T2D patients decreased by 25.8 % (95 % CI: -32.7, -18.8; p<0.001).

Meta-regressions, subgroup analysis, and publication bias: ET-independent VSMrelaxation

ET-independent VSM relaxation in response to SNP was affected by the doses of SNP (p= 0.084, B=-1.36; Supplementary Figure 1Eexcli2024-7330_supplementary_information.pdf) and the duration of T2D (p= 0.047, B=-3.36; Supplementary Figure 1Fexcli2024-7330_supplementary_information.pdf); higher doses of SNP and longer duration of T2D were associated with a higher WMD (higher impairment in response to SNP) of VSM relaxation between T2D patients and controls. ET-independent VSM relaxation in response to GTN has not been affected by GTN doses (p=0.942, B = -0.006; Supplementary Figure 1Gexcli2024-7330_supplementary_information.pdf) but affected by the duration of T2D (p=0.071, B = -0.2631.96; Supplementary Figure 1Hexcli2024-7330_supplementary_information.pdf); longer durations of T2D were associated with a higher WMD (higher impairment in response to GTN) of VSM relaxation between T2D patients and controls.

The test of group differences was statistically significant between the subgroups for the type of vessels in studies that used SNP (p=0.045) but not for GTN (p=0.809) (Table 4(Tab. 4)).

The funnel plot for the mean difference in the VSM function of studies included in the meta-analysis was asymmetrical (Supplementary Figure 2Cexcli2024-7330_supplementary_information.pdf for SNP and 2D for GTN). However, Egger's test detected that the observed asymmetry was not due to publication bias (p=0.315, p=0.879 for SNP and GTN, respectively).

Discussion

In this study, following our previous narrative review study (Bahadoran et al., 2023[5]), we systematically reviewed and meta-analyzed data from 44 non-randomized clinical trials (with 68, 16, 30, and 50 study arms for ACh, MTH, SNP, and GTN, respectively). These trials compared ET-dependent and -independent VSM relaxation in T2D patients and controls. Our analysis revealed a decrease in ET-dependent VSM relaxation in T2D patients by ~40 %, with a greater decrease for MTH (~55 %) compared to ACh (~30 %). Similarly, ET-independent VSM relaxation in T2D patients decreased by ~25 %. We observed a greater decrease for GTN (~63 %) compared to SNP (~17). Therefore, the current, up-to-date meta-analysis confirmed the results of Montero et al. (2013[52]) and, for the first time, showed that impairment of ET-dependent and -independent VSM relaxation in T2D depends on the type and dose of vasodilators used. In addition, ET-dependent (MTH) and ET-independent (SNP/GTN) VSM relaxation in T2D patients are affected by the duration of T2D and the type of vessels studied.

Our results showed that ET-dependent (ACh/MTH) VSM relaxation in T2D patients decreased by ~40 %, and the decrease was higher in response to MTH (~55 %) compared to ACh (~30 %), a difference of about two-fold. Although both ACh and MTH are non-selective cholinergic agonists with similar affinity for muscarinic receptors, ACh exhibits significantly lower potency in inducing vasodilatation (-log EC50 = 6.43 vs. 7.24 mol/L) (Bruning et al., 1995[11], 1996[12]). This difference can be attributed to the rapid degradation of ACh by cholinesterases. Consequently, MTH has a longer-lasting effect, leading to a more pronounced (10-fold) vasodilator response (Bruning et al., 1995[11], 1996[12]).

Our results showed that ET-independent (SNP/GTN) VSM relaxation in T2D patients decreased by ~25 %, and this decrease was higher in response to GTN (~63 %) compared to SNP (~17 %) in T2D patients. It has been reported that GTN can be more potent than SNP under specific conditions, exhibiting ~1,000-fold higher vasodilatory activity in certain conditions in rat chest vessels (Badejo et al., 2010[4]) and in human coronary arteries (He and Yang, 1997[32]). Unlike the consistent findings on impaired ET-dependent VSM relaxation, studies investigating ET-independent VSM relaxation in T2D patients in response to GTN and SNP have yielded conflicting results. Some studies reported a decrease of 6-42 % in SNP (Morris et al., 1995[53]; Watts et al., 1996[80]; Preik et al., 2000[62]) and GTN (McVeigh et al., 1992[51]; Vavuranakis et al., 1999[77]) induced VSM relaxation, while others show preserved VSM relaxation (Karasu et al., 1995[39]; Steinberg et al., 1996[71]; Heitzer et al., 2000[33]) in T2D patients. These discrepancies likely arise from several factors, including the presence of diabetes complications (Avogaro et al., 1997[3]), disease severity and duration (Natali et al., 2006[54]), and T2D management (Bock et al., 2021[8]). These factors are estimated to explain 32-37 % of the variation observed in ET-independent VSM relaxation (Caballero et al., 1999[14]; Bahadoran et al., 2023[5]). Interestingly, ET-independent VSM relaxation remains preserved in the brachial artery of T2D patients without complications (Avogaro et al., 1997[3]). Additionally, reduced SNP-induced vasodilation was only observed in neuropathic T2D patients (Pitei et al., 1997[61]). Studies also report a progressive decrease in ET-independent VSM relaxation from 3 % in insulin-sensitive patients to 27 % in insulin-resistant patients (Natali et al., 2006[54]). Furthermore, insulin therapy for 6 months and a 3.5-year follow-up can improve SNP-induced vasodilation in T2D patients (Vehkavaara and Yki-Järvinen, 2004[78]).

Results of meta-regression analysis showed that the duration of T2D affects MTH and SNP/GTN-induced VSM relaxation in T2D patients. For MTH (with a maximum T2D duration of 4 years), longer T2D was associated with a smaller difference in VSM relaxation in T2D patients compared to controls. In contrast, SNP/GTN studies (with a maximum T2D duration of 20 years) showed a greater difference in VSM relaxation in T2D patients than in controls. Importantly, this difference in study duration between MTH and SNP/GTN studies may explain the opposing findings.

Decreased NO production by the ET and impaired responsiveness of VSMs to NO have been reported in the late stages of T2D (Bahadoran et al., 2023[5]). Early T2D stages, however, may exhibit increased NO production (Adela et al., 2015[1]), potentially as a compensatory mechanism. Over time, this ability diminishes, leading to a state of vascular NO deficiency in established T2D (Pieper, 1998[59]; Salvatore et al., 2022[64]). In support, ACh-induced NO release in the thoracic aorta of rats enhanced at 12 weeks but decreased at 36 weeks (Kobayashi et al., 2004[45]). A progressive decreased endothelial NO synthase sensitivity to ACh was also shown to be worsened with the diabetes duration in rats; EC50 for ACh in diabetic arteries was increased from 13.5 nM after 6 weeks to 100 nM after 24 weeks of diabetes (Diederich et al., 1994[19]).

Our subgroup analysis showed that ET-dependent (ACh-induced) VSM relaxation in T2D patients was significantly lower (higher impairment) than controls when using brachial and femoral arteries (macrocirculation) compared to FMS. Conversely, ET-independent (SNP-induced) VSM relaxation was significantly lower in FMS compared to macrocirculation. This finding aligns with a meta-analysis reporting higher impairment of ET-independent VSM relaxation in microcirculation versus macrocirculation in T2D patients compared to controls (Montero et al., 2013[52]). Evidence suggests that small coronary arteries might be more susceptible to T2D-induced vascular NO resistance compared to large arteries (Oltman et al., 2006[56]). This could be due to variations in NO production capacity and responsiveness to NO's vasodilatory effects between different blood vessel types (Bahadoran et al., 2023[5]). These observations may imply that different vessels probably exhibit diverse phenotypes of NO resistance in T2D.

In summary, both ET-dependent and ET-independent VSM relaxation are impaired in T2D. Various studies, including a recent review by our group, have investigated the mechanisms underlying this impairment (Bahadoran et al., 2023[5]). Although, the exact mechanisms remain unclear; hyperglycemia is thought to play a significant role. Hyperglycemia increases the production of reactive oxygen species (ROS) in the vessel wall, triggering a series of events: activation of the polyol and hexosamine pathways, formation of advanced glycation end products (AGEs), activation of protein kinase C (PKC) pathways, and increased nicotinamide adenine dinucleotide phosphate (NADPH) oxidase activity. These events collectively impair NO-induced VSM relaxation, resulting in vascular hypo-responsiveness to NO (Figure 7(Fig. 7)). In ET cells, T2D primary blocks nitric oxide synthase (NOS) activity and expression, leading to lower NO bioavailability in VSM cells. In VSM cells, one key mechanism is the impairment of the NO-cyclic guanosine monophosphate (cGMP)-protein kinase G (PKG) pathway at both the receptor and post-receptor levels. This may occur through quenching of NO and desensitization of soluble guanylate cyclase (sGC) at the receptor level (Witte et al., 2002[82]; Goulopoulou et al., 2015[30]), and through decreased cGMP degradation and altered activity and expression of PKG at the post-receptor level (Chirkov and Horowitz, 2007[15]; Russo et al., 2008[63]). More details on the mechanisms of vascular relaxation impairment in T2D can be found elsewhere (Bahadoran et al., 2023[5]).

Our study has some limitations; first, we identified significant heterogeneity, meaning that the results of the included studies varied considerably. Egger's test detected that the observed heterogeneity was not due to publication bias. While subgroup analyses suggested vessel type as the potential source of heterogeneity, the uneven distribution of studies across subgroups limits definitive conclusions. Additionally, the studies themselves varied in design, patient numbers, doses of vasodilators, and T2D duration. These design variations might contribute to the observed heterogeneity. Second, another limitation is that our study did not account for the potential confounding effects of diabetic complications or medications taken by diabetic patients. One strength of our study is that we considered baseline values of VSM in brachial and femoral arteries, as well as FMS of diabetic patients. This analysis revealed a 13 % lower baseline VSM (indicating lower baseline blood flow) in T2D patients compared to controls within the studies included in our meta-analysis. This finding highlights a potentially important factor that was not considered by Montero et al. (2013[52]). This allows for a clearer interpretation of results as we assess the change in relaxation due to T2D rather than simply comparing the values of diabetic and control groups. In addition, the weight of the included studies was balanced across different analyses. This suggests the overall ES is not driven by specific studies and is unlikely to be a false-positive result.

Conclusion

Our meta-analysis of clinical trials investigated the effects of T2D on VSM relaxation. The results showed significant reductions in both ET-dependent (~40 %) and ET-independent (~25 %) VSM relaxation. The decrease was more pronounced for MTH (~55 %) compared to ACh (~30 %) and for GTN (~63 %) compared to SNP (~17 %). These findings suggest that dysfunction of both ET and VSM contributes to impaired VSM relaxation in T2D patients.

Results obtained from ET-dependent (MTH) and ET-independent (SNP/GTN) VSM relaxation in T2D patients may be affected by both the type of studied vessels and the duration of T2D. This knowledge may be relevant for developing treatments specifically targeting different blood vessels in T2D. Considering the diverse phenotypes of NO resistance in different vessels may help develop specific vessel-targeted drug delivery platforms to overcome vascular NO resistance in T2D. Pharmacological approaches such as angiotensin-converting enzyme inhibitors, the anti-anginal agent perhexiline, insulin, statins, nitrite, and sGC activators have been proposed to attenuate vascular NO resistance by decreasing ROS production and increasing eNOS expression and activity (Chirkov and Horowitz, 2007[15]; Velagic et al., 2020[79]). NO-releasing drugs may not affect microcirculation, which has a higher impairment of ET-independent VSM relaxation and is relatively nonresponsive to NO compared to macrocirculation. In these cases, the use of nitroxyl donors, such as Angeli's salt, has been suggested as an effective treatment for overcoming NO resistance in T2D, where NO-releasing drugs do not act effectively (Velagic et al., 2020[79]). Angeli's salt can effectively bypass NO resistance in vessels that become relatively nonresponsive to NO in T2D by affecting on distinct molecules (e.g., thiol residues) and pathways [cyclic AMP (cAMP)-protein kinase A (PKA)]. Moreover, the duration of T2D should be considered when designing treatment strategies for vascular complications. Because various vessels have different capacities of NO production (Bohlen, 2015[9]) and respond heterogeneously to the vasodilatory action of NO, likely due to their different morphologies, functions, and diverse receptor and ion channel populations (Clark and Fuchs, 1997[17]; Gao et al., 1998[25]; Houghton et al., 1998[37]). Vascular NO resistance seems to progress over time, with NO production increasing in the initial stages of T2D (within 5 years of onset) and decreasing in established T2D (Pieper, 1998[59]; Salvatore et al., 2022[64]). Upregulated NOS expression/activity may explain the elevated NO production in the initial stages of T2D (Kazuyama et al., 2009[41]). Although it needs to be proved, small coronary arteries appear to be affected by T2D-induced vascular NO resistance earlier and to a greater extent compared to mesenteric resistance arteries and large elastic vessels like the aorta (Oltman et al., 2006[56]). These observations suggest that some treatments, such as NO-releasing drugs and activators of NOS expression, may not affect impairment in VSM relaxation in the early stage T2D. Management of hyperglycemia, insulin resistance and oxidative stress are essential treatments during this stage (Bahadoran et al., 2023[5]).

Notes

Khosrow Kashfi and Asghar Ghasemi (Endocrine Physiology Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran; E-mail: Ghasemi@sbmu.ac.ir) contributed equally as corresponding author.

Declaration

Acknowledgments and funding information

This study was supported by a grant from Shahid Beheshti University of Medical Sciences.

Declaration of competing interest

The authors declare that they have no competing interests.

Authorships

Sajad Jeddi, Zahra Bahadoran, and Parvin Mirmiran contributed to the screening and quality assessment of studies and data extraction. Sajad Jeddi, Zahra Bahadoran, Parvin Mirmiran, Khosrow Kashfi, and Asghar Ghasemi contributed to the literature review and wrote the article. Sajad Jeddi, Khosrow Kashfi, and Asghar Ghasemi provided critical revision and final approval of the finalized manuscript. All authors have read and approved the final manuscript.

 

References

1. Adela R, Nethi SK, Bagul PK, Barui AK, Mattapally S, Kuncha M, et al. Hyperglycaemia enhances nitric oxide production in diabetes: a study from South Indian patients. PLoS One. 2015;10(4):e0125270. doi: 10.1371/journal.pone.0125270
2. Aikaeli F, Njim T, Gissing S, Moyo F, Alam U, Mfinanga SG, et al. Prevalence of microvascular and macrovascular complications of diabetes in newly diagnosed type 2 diabetes in low-and-middle-income countries: A systematic review and meta-analysis. PLOS Glob Public Health. 2022;2(6):e0000599. doi: 10.1371/journal.pgph.0000599
3. Avogaro A, Piarulli F, Valerio A, Miola M, Calveri M, Pavan P, et al. Forearm nitric oxide balance, vascular relaxation, and glucose metabolism in NIDDM patients. Diabetes. 1997;46:1040-6. doi: 10.2337/diab.46.6.1040
4. Badejo AM Jr, Hodnette C, Dhaliwal JS, Casey DB, Pankey E, Murthy SN, et al. Mitochondrial aldehyde dehydrogenase mediates vasodilator responses of glyceryl trinitrate and sodium nitrite in the pulmonary vascular bed of the rat. Am J Physiol Heart Circ Physiol. 2010;299:H819-26. doi: 10.1152/ajpheart.00959.2009
5. Bahadoran Z, Mirmiran P, Kashfi K, Ghasemi A. Vascular nitric oxide resistance in type 2 diabetes. Cell Death Dis. 2023;14(7):410. doi: 10.1038/s41419-023-05935-5
6. Beckman JA, Goldfine AB, Goldin A, Prsic A, Kim S, Creager MA. Inhibition of protein kinase Cbeta does not improve endothelial function in type 2 diabetes. J Clin Endocrinol Metab. 2010;95:3783-7. doi: 10.1210/jc.2010-0286
7. Beer S, Feihl F, Ruiz J, Juhan-Vague I, Aillaud MF, Wetzel SG, et al. Comparison of skin microvascular reactivity with hemostatic markers of endothelial dysfunction and damage in type 2 diabetes. Vasc Health Risk Manag. 2008;4:1449-58. doi: 10.2147/vhrm.s4175
8. Bock JM, Hughes WE, Ueda K, Feider AJ, Hanada S, Casey DP. Glycemic management is inversely related to skeletal muscle microvascular endothelial function in patients with type 2 diabetes. Physiol Rep. 2021;9(5):e14764. doi: 10.14814/phy2.14764
9. Bohlen HG. Nitric oxide and the cardiovascular system. Compr Physiol. 2015;5:808-23. doi: 10.1002/cphy.c140052
10. Brooks BA, Franjic B, Ban CR, Swaraj K, Yue DK, Celermajer DS, et al. Diastolic dysfunction and abnormalities of the microcirculation in type 2 diabetes. Diabetes Obes Metab. 2008;10:739-46. doi: 10.1111/j.1463-1326.2007.00803.x
11. Bruning TA, Chang PC, Hendriks MG, Vermeij P, Pfaffendorf M, van Zwieten PA. In vivo characterization of muscarinic receptor subtypes that mediate vasodilatation in patients with essential hypertension. Hypertension. 1995;26(1):70-7. doi: 10.1161/01.hyp.26.1.70
12. Bruning TA, Chang PC, Kemme MJ, Vermeij P, Pfaffendorf M, van Zwieten PA. Comparison of cholinergic vasodilator responses to acetylcholine and methacholine in the human forearm. Blood Press. 1996;5:333-41. doi: 10.3109/08037059609078071
13. Butler R, Morris AD, Belch JJ, Hill A, Struthers AD. Allopurinol normalizes endothelial dysfunction in type 2 diabetics with mild hypertension. Hypertension. 2000;35:746-51. doi: 10.1161/01.hyp.35.3.746
14. Caballero AE, Arora S, Saouaf R, Lim SC, Smakowski P, Park JY, et al. Microvascular and macrovascular reactivity is reduced in subjects at risk for type 2 diabetes. Diabetes. 1999;48:1856-62. doi: 10.2337/diabetes.48.9.1856
15. Chirkov YY, Horowitz JD. Impaired tissue responsiveness to organic nitrates and nitric oxide: a new therapeutic frontier? Pharmacol Ther. 2007;116:287-305. doi: 10.1016/j.pharmthera.2007.06.012
16. Chousos I, Perrea D, Kyriaki D, Liatis S, Katsilambros N, Makrilakis K. Acute hyperhomocysteinaemia blunts endothelial dependent and endothelial independent vasodilatation in diabetic patients. Diab Vasc Dis Res. 2010;7:186-94. doi: 10.1177/1479164110369687
17. Clark SG, Fuchs LC. Role of nitric oxide and Ca++-dependent K+ channels in mediating heterogeneous microvascular responses to acetylcholine in different vascular beds. J Pharmacol Exp Ther. 1997;282:1473-9
18. Creager MA, Lüscher TF, Cosentino F, Beckman JA. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: Part I. Circulation. 2003;108:1527-32. doi: 10.1161/01.cir.0000091257.27563.32
19. Diederich D, Skopec J, Diederich A, Dai FX. Endothelial dysfunction in mesenteric resistance arteries of diabetic rats: role of free radicals. Am J Physiol. 1994;266:H1153-61. doi: 10.1152/ajpheart.1994.266.3.H1153
20. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-34. doi: 10.1136/bmj.315.7109.629
21. Einarson TR, Acs A, Ludwig C, Panton UH. Prevalence of cardiovascular disease in type 2 diabetes: a systematic literature review of scientific evidence from across the world in 2007–2017. Cardiovasc Diabetol. 2018;17(1):83. doi: 10.1186/s12933-018-0728-6
22. Enderle MD, Benda N, Schmuelling RM, Haering HU, Pfohl M. Preserved endothelial function in IDDM patients, but not in NIDDM patients, compared with healthy subjects. Diabetes Care. 1998;21:271-7. doi: 10.2337/diacare.21.2.271
23. Furuta M, Ueyama M, Morita S, Yamana A, Sanke T. Combined examination of glyceryl trinitrate-mediated vascular dilation with flow-mediated vascular dilation is essential for assessment of vascular function in type 2 diabetes. J Diabetes Investig. 2013;4:304-9. doi: 10.1111/jdi.12021
24. Ganz P, Vita JA. Testing endothelial vasomotor function: nitric oxide, a multipotent molecule. Circulation. 2003;108:2049-53. doi: 10.1161/01.cir.0000089507.19675.f9
25. Gao Y, Tolsa J-F, Raj JU. Heterogeneity in endothelium-derived nitric oxide-mediated relaxation of different sized pulmonary arteries of newborn lambs. Pediatric Res. 1998;44:723-9. doi: 10.1203/00006450-199811000-00015
26. Gazis A, White DJ, Page SR, Cockcroft JR. Effect of oral vitamin E (alpha-tocopherol) supplementation on vascular endothelial function in Type 2 diabetes mellitus. Diabet Med. 1999;16:304-11. doi: 10.1046/j.1464-5491.1999.00049.x
27. Ghasemi A, Jeddi S. Quantitative aspects of nitric oxide production in the heart. Mol Biol Rep. 2022;49:11113-22. doi: 10.1007/s11033-022-07889-x
28. Gheibi S, Mahmoodzadeh A, Kashfi K, Jeddi S, Ghasemi A. Data extraction from graphs using adobe photoshop: applications for meta-analyses. Int J Endocrinol Metab. 2019;17(4):e95216. doi: 10.5812/ijem.95216
29. Goodfellow J, Ramsey MW, Luddington LA, Jones CJ, Coates PA, Dunstan F, et al. Endothelium and inelastic arteries: an early marker of vascular dysfunction in non-insulin dependent diabetes. BMJ. 1996;312(7033):744-5. doi: 10.1136/bmj.312.7033.744
30. Goulopoulou S, Hannan JL, Matsumoto T, Ogbi S, Ergul A, Webb RC. Reduced vascular responses to soluble guanylyl cyclase but increased sensitivity to sildenafil in female rats with type 2 diabetes. Am J Physiol Heart Circ Physiol. 2015;309:H297-304. doi: 10.1152/ajpheart.00079.2015
31. Halcox JP, Schenke WH, Zalos G, Mincemoyer R, Prasad A, Waclawiw MA, et al. Prognostic value of coronary vascular endothelial dysfunction. Circulation. 2002;106:653-8. doi: 10.1161/01.cir.0000025404.78001.d8
32. He GW, Yang CQ. Comparison of nitroprusside and nitroglycerin in inhibition of angiotensin II and other vasoconstrictor-mediated contraction in human coronary bypass conduits. Br J Clin Pharmacol. 1997;44:361-7. doi: 10.1046/j.1365-2125.1997.t01-2-00589.x
33. Heitzer T, Krohn K, Albers S, Meinertz T. Tetrahydrobiopterin improves endothelium-dependent vasodilation by increasing nitric oxide activity in patients with Type II diabetes mellitus. Diabetologia. 2000;43:1435-8. doi: 10.1007/s001250051551
34. Henry RM, Ferreira I, Kostense PJ, Dekker JM, Nijpels G, Heine RJ, et al. Type 2 diabetes is associated with impaired endothelium-dependent, flow-mediated dilation, but impaired glucose metabolism is not;The Hoorn Study. Atherosclerosis. 2004;174:49-56. doi: 10.1016/j.atherosclerosis.2004.01.002
35. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557-60. doi: 10.1136/bmj.327.7414.557
36. Hogikyan RV, Galecki AT, Pitt B, Halter JB, Greene DA, Supiano MA. Specific impairment of endothelium-dependent vasodilation in subjects with type 2 diabetes independent of obesity. J Clin Endocrinol Metab. 1998;83:1946-52. doi: 10.1210/jcem.83.6.4907
37. Houghton JL, Davison CA, Kuhner PA, Torossov MT, Strogatz DS, Carr AA. Heterogeneous vasomotor responses of coronary conduit and resistance vessels in hypertension. J Am Coll Cardiol. 1998;31:374-82. doi: 10.1016/S0735-1097(97)00505-6
38. Huvers FC, Schaper NC, Houben AJ, Hofstra L, Kitslaar PJ, De Leeuw PW, et al. Impaired arterial but not venous responsiveness to nitroglycerin in non-insulin-dependent diabetes mellitus. Eur J Clin Invest. 1997;27:360-5. doi: 10.1046/j.1365-2362.1997.1210673.x
39. Karasu C, Soncul H, Altan VM. Effects of non-insulin dependent diabetes mellitus on the reactivity of human internal mammary artery and human saphenous vein. Life Sci. 1995;57:103-12. doi: 10.1016/0024-3205(95)00251-z
40. Kawano N, Emoto M, Mori K, Yamazaki Y, Urata H, Tsuchikura S, et al. Association of endothelial and vascular smooth muscle dysfunction with cardiovascular risk factors, vascular complications, and subclinical carotid atherosclerosis in type 2 diabetic patients. J Atheroscler Thromb. 2012;19:276-84. doi: 10.5551/jat.10629
41. Kazuyama E, Saito M, Kinoshita Y, Satoh I, Dimitriadis F, Satoh K. Endothelial dysfunction in the early- and late-stage type-2 diabetic Goto-Kakizaki rat aorta. Mol Cell Biochem. 2009;332:95-102. doi: 10.1007/s11010-009-0178-2
42. Kelm M. Nitric oxide metabolism and breakdown. Biochim Biophys Acta. 1999;1411:273-89. doi: S0005-2728(99)00020-1 [pii]
43. Kimura Y, Matsumoto M, Miyauchi E, Deng YB, Iwai K, Hattori H. Noninvasive detection of endothelial dysfunction in elderly with NIDDM by ultrasonography. Echocardiography. 2001;18:559-64. doi: 10.1046/j.1540-8175.2001.00559.x
44. Kingwell BA, Formosa M, Muhlmann M, Bradley SJ, McConell GK. Type 2 diabetic individuals have impaired leg blood flow responses to exercise: role of endothelium-dependent vasodilation. Diabetes Care. 2003;26:899-904. doi: 10.2337/diacare.26.3.899
45. Kobayashi T, Matsumoto T, Ooishi K, Kamata K. Differential expression of alpha2D-adrenoceptor and eNOS in aortas from early and later stages of diabetes in Goto-Kakizaki rats. Am J Physiol Heart Circ Physiol. 2004;287:H135-43. doi: 10.1152/ajpheart.01074.2003
46. Lerman A, Zeiher AM. Endothelial function: cardiac events. Circulation. 2005;111:363-8. doi: 10.1161/01.cir.0000153339.27064.14
47. Liu J, Hu DJ, Yan H, Liu J, Ai X, Ren Z, et al. Attenuated endothelial function is associated with decreased endothelial progenitor cells and nitric oxide in premenopausal diabetic women. Mol Med Rep. 2018;18:4666-74. doi: 10.3892/mmr.2018.9451
48. Lüscher TF, Creager MA, Beckman JA, Cosentino F. Diabetes and vascular disease: pathophysiology, clinical consequences, and medical therapy: Part II. Circulation. 2003;108:1655-61. doi: 10.1161/01.cir.0000089189.70578.e2
49. Mahdi A, Kövamees O, Checa A, Wheelock CE, von Heijne M, Alvarsson M, et al. Arginase inhibition improves endothelial function in patients with type 2 diabetes mellitus despite intensive glucose-lowering therapy. J Intern Med. 2018;284:388-98. doi: 10.1111/joim.12785
50. Malinski T, Taha Z, Grunfeld S, Patton S, Kapturczak M, Tomboulian P. Diffusion of nitric oxide in the aorta wall monitored in situ by porphyrinic microsensors. Biochem Biophys Res Commun. 1993;193:1076-82. doi: 10.1006/bbrc.1993.1735
51. McVeigh GE, Brennan GM, Johnston GD, McDermott BJ, McGrath LT, Henry WR, et al. Impaired endothelium-dependent and independent vasodilation in patients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia. 1992;35:771-6. doi: 10.1007/bf00429099
52. Montero D, Walther G, Pérez-Martin A, Vicente-Salar N, Roche E, Vinet A. Vascular smooth muscle function in type 2 diabetes mellitus: a systematic review and meta-analysis. Diabetologia. 2013;56:2122-33. doi: 10.1007/s00125-013-2974-1
53. Morris SJ, Shore AC, Tooke JE. Responses of the skin microcirculation to acetylcholine and sodium nitroprusside in patients with NIDDM. Diabetologia. 1995;38:1337-44. doi: 10.1007/bf00401767
54. Natali A, Toschi E, Baldeweg S, Ciociaro D, Favilla S, Saccà L, et al. Clustering of insulin resistance with vascular dysfunction and low-grade inflammation in type 2 diabetes. Diabetes. 2006;55:1133-40. doi: 10.2337/diabetes.55.04.06.db05-1076
55. O'Brien SF, Watts GF, Playford DA, Burke V, O'Neal DN, Best JD. Low-density lipoprotein size, high-density lipoprotein concentration, and endothelial dysfunction in non-insulin-dependent diabetes. Diabet Med. 1997;14:974-8. doi: 10.1002/(sici)1096-9136(199711)14:11<974::aid-dia495>3.0.co;2-i
56. Oltman CL, Richou LL, Davidson EP, Coppey LJ, Lund DD, Yorek MA. Progression of coronary and mesenteric vascular dysfunction in Zucker obese and Zucker diabetic fatty rats. Am J Physiol Heart Circ Physiol. 2006;291:H1780-7. doi: 10.1152/ajpheart.01297.2005
57. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71
58. Park KH, Park WJ. Endothelial dysfunction: clinical implications in cardiovascular disease and therapeutic approaches. J Korean Med Sci. 2015;30:1213-25. doi: 10.3346/jkms.2015.30.9.1213
59. Pieper GM. Review of alterations in endothelial nitric oxide production in diabetes: protective role of arginine on endothelial dysfunction. Hypertension. 1998;31:1047-60. doi: 10.1161/01.hyp.31.5.1047
60. Pistrosch F, Passauer J, Fischer S, Fuecker K, Hanefeld M, Gross P. In type 2 diabetes, rosiglitazone therapy for insulin resistance ameliorates endothelial dysfunction independent of glucose control. Diabetes Care. 2004;27:484-90. doi: 10.2337/diacare.27.2.484
61. Pitei DL, Watkins PJ, Edmonds ME. NO-dependent smooth muscle vasodilatation is reduced in NIDDM patients with peripheral sensory neuropathy. Diabet Med. 1997;14:284-90. doi: 10.1002/(sici)1096-9136(199704)14:4<284::aid-dia348>3.0.co;2-0
62. Preik M, Kelm M, Rösen P, Tschöpe D, Strauer BE. Additive effect of coexistent type 2 diabetes and arterial hypertension on endothelial dysfunction in resistance arteries of human forearm vasculature. Angiology. 2000;51:545-54. doi: 10.1177/000331970005100703
63. Russo I, Del Mese P, Doronzo G, Mattiello L, Viretto M, Bosia A, et al. Resistance to the nitric oxide/cyclic guanosine 5'-monophosphate/protein kinase G pathway in vascular smooth muscle cells from the obese Zucker rat, a classical animal model of insulin resistance: role of oxidative stress. Endocrinology. 2008;149:1480-9. doi: 10.1210/en.2007-0920
64. Salvatore T, Galiero R, Caturano A, Vetrano E, Loffredo G, Rinaldi L, et al. Coronary microvascular dysfunction in diabetes mellitus: pathogenetic mechanisms and potential therapeutic options. Biomedicines. 2022;10(9):2274. doi: 10.3390/biomedicines10092274
65. Schächinger V, Britten MB, Zeiher AM. Prognostic impact of coronary vasodilator dysfunction on adverse long-term outcome of coronary heart disease. Circulation. 2000;101:1899-906. doi: 10.1161/01.cir.101.16.1899
66. Schmiedel O, Schroeter ML, Harvey JN. Microalbuminuria in Type 2 diabetes indicates impaired microvascular vasomotion and perfusion. Am J Physiol Heart Circ Physiol. 2007;293:H3424-31. doi: 10.1152/ajpheart.00558.2007
67. Shemyakin A, Kövamees O, Rafnsson A, Böhm F, Svenarud P, Settergren M, et al. Arginase inhibition improves endothelial function in patients with coronary artery disease and type 2 diabetes mellitus. Circulation. 2012;126:2943-50. doi: 10.1161/circulationaha.112.140335
68. Sivitz WI, Wayson SM, Bayless ML, Sinkey CA, Haynes WG. Obesity impairs vascular relaxation in human subjects: hyperglycemia exaggerates adrenergic vasoconstriction arterial dysfunction in obesity and diabetes. J Diabetes Complications. 2007;21:149-57. doi: 10.1016/j.jdiacomp.2005.12.003
69. Sokolnicki LA, Roberts SK, Wilkins BW, Basu A, Charkoudian N. Contribution of nitric oxide to cutaneous microvascular dilation in individuals with type 2 diabetes mellitus. Am J Physiol Endocrinol Metab. 2007;292:E314-8. doi: 10.1152/ajpendo.00365.2006
70. Steinberg HO, Chaker H, Leaming R, Johnson A, Brechtel G, Baron AD. Obesity/insulin resistance is associated with endothelial dysfunction. Implications for the syndrome of insulin resistance. J Clin Invest. 1996;97:2601-10. doi: 10.1172/jci118709
71. Steinberg HO, Paradisi G, Cronin J, Crowde K, Hempfling A, Hook G, et al. Type II diabetes abrogates sex differences in endothelial function in premenopausal women. Circulation. 2000;101:2040-6. doi: 10.1161/01.cir.101.17.2040
72. Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183:109119. doi: 10.1016/j.diabres.2021.109119
73. Tan KC, Xu A, Chow WS, Lam MC, Ai VH, Tam SC, et al. Hypoadiponectinemia is associated with impaired endothelium-dependent vasodilation. J Clin Endocrinol Metab. 2004;89:765-9. doi: 10.1210/jc.2003-031012
74. Ting HH, Timimi FK, Boles KS, Creager SJ, Ganz P, Creager MA. Vitamin C improves endothelium-dependent vasodilation in patients with non-insulin-dependent diabetes mellitus. J Clin Invest. 1996;97:22-8. doi: 10.1172/jci118394
75. Van de Ree MA, Huisman MV, de Man FH, van der Vijver JC, Meinders AE, Blauw GJ. Impaired endothelium-dependent vasodilation in type 2 diabetes mellitus and the lack of effect of simvastatin. Cardiovasc Res. 2001;52:299-305. doi: 10.1016/s0008-6363(01)00379-0
76. Van Etten RW, de Koning EJ, Verhaar MC, Gaillard CA, Rabelink TJ. Impaired NO-dependent vasodilation in patients with Type II (non-insulin-dependent) diabetes mellitus is restored by acute administration of folate. Diabetologia. 2002;45:1004-10. doi: 10.1007/s00125-002-0862-1
77. Vavuranakis M, Stefanadis C, Triandaphyllidi E, Toutouzas K, Toutouzas P. Coronary artery distensibility in diabetic patients with simultaneous measurements of luminal area and intracoronary pressure: evidence of impaired reactivity to nitroglycerin. J Am Coll Cardiol. 1999;34:1075-81. doi: 10.1016/s0735-1097(99)00331-9
78. Vehkavaara S, Yki-Järvinen H. 3.5 years of insulin therapy with insulin glargine improves in vivo endothelial function in type 2 diabetes. Arterioscler Thromb Vasc Biol. 2004;24:325-30. doi: 10.1161/01.ATV.0000113817.48983.c5
79. Velagic A, Qin C, Woodman OL, Horowitz JD, Ritchie RH, Kemp-Harper BK. Nitroxyl: A novel strategy to circumvent diabetes associated impairments in nitric oxide signaling. Front Pharmacol. 2020;11:727. doi: 10.3389/fphar.2020.00727
80. Watts GF, O'Brien SF, Silvester W, Millar JA. Impaired endothelium-dependent and independent dilatation of forearm resistance arteries in men with diet-treated non-insulin-dependent diabetes: role of dyslipidaemia. Clin Sci (Lond). 1996;91:567-73. doi: 10.1042/cs0910567
81. Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager MA. Impaired nitric oxide-mediated vasodilation in patients with non-insulin-dependent diabetes mellitus. J Am Coll Cardiol. 1996;27:567-74. doi: 10.1016/0735-1097(95)00522-6
82. Witte K, Jacke K, Stahrenberg R, Arlt G, Reitenbach I, Schilling L, et al. Dysfunction of soluble guanylyl cyclase in aorta and kidney of Goto-Kakizaki rats: influence of age and diabetic state. Nitric Oxide. 2002;6:85-95. doi: 10.1006/niox.2001.0363
83. Woodman RJ, Playford DA, Watts GF. Basal production of nitric oxide (NO) and non-NO vasodilators in the forearm microcirculation in Type 2 diabetes: associations with blood pressure and HDL cholesterol. Diabetes Res Clin Pract. 2006;71:59-67. doi: 10.1016/j.diabres.2005.05.008
84. Yeh SY, Doupis J, Rahangdale S, Horr S, Malhotra A, Veves A. Total serum bilirubin does not affect vascular reactivity in patients with diabetes. Vasc Med. 2009;14:129-36. doi: 10.1177/1358863x08098273
85. Yugar-Toledo JC, Tanus-Santos JE, Sabha M, Sousa MG, Cittadino M, Tácito LH, et al. Uncontrolled hypertension, uncompensated type II diabetes, and smoking have different patterns of vascular dysfunction. Chest. 2004;125:823-30. doi: 10.1378/chest.125.3.823
 
 

File-Attachments

  1. excli2024-7330_supplementary_information.pdf (598,59 KB)
    Supplementary information
 
 

Figure 1: Graphical abstract. Type 2 diabetes decreased both ET-dependent (~40 %) and ET-independent (~25 %) vascular smooth muscle (VSM) relaxation that was more pronounced for MTH (~55 %) compared to ACh (~30 %) and for GTN (~63 %) compared to SNP (~17 %). ACh, acetylcholine; GTN, glyceryl trinitrate; MTH, methacholine; NO, nitric oxide; ROS, reactive oxygen species; SNP, sodium nitroprusside

Figure 2: PRISMA flow diagram of the included studies. Created with BioRender.com

Figure 3: Overall effect size (ES) of endothelium (ET)-dependent vascular smooth muscle (VSM) relaxation in response to acetylcholine (ACh) in patients with type 2 diabetes (T2D) compared to controls. CI, confidence interval

Figure 4: Overall effect size (ES) of endothelium (ET)-dependent vascular smooth muscle (VSM) relaxation in response to methacholine (MTH) in patients with type 2 diabetes (T2D) compared to controls. CI, confidence interval

Figure 5: Overall effect size (ES) of endothelium (ET)-independent vascular relaxation in response to sodium nitroprusside (SNP) in type 2 diabetic patients compared to controls. CI, confidence interval.

Figure 6: Overall effect size (ES) of endothelium (ET)-independent vascular relaxation in response to glyceryl trinitrate (GTN) in type 2 diabetic patients compared to controls. CI, confidence interval

Figure 7: Type 2 diabetes (T2D) impairs both ET-dependent and -independent vascular smooth muscle (VSM) relaxation by affecting nitric oxide (NO) action on VSM relaxation. The vascular hypo-responsiveness to NO is evident at the receptor and post-receptor level. AEGs, advanced glycation end products; cGMP, cyclic guanosine monophosphate; L-Arg, l-arginine; NADPH, nicotinamide adenine dinucleotide phosphate; NOS, Nitric oxide synthase; PDE, phosphodiesterase; PKG, protein kinase G; PKC, protein kinase C; ROS, reactive oxygen species; sGC, soluble guanylate cyclase

 

Table 1: Characteristic of studies that reported endothelium-dependent vascular smooth muscle (VSM) relaxation in patients with type 2 diabetes (T2D) in response to acetylcholine (ACh) and methacholine (MTH)

Table 2: Subgroup analysis of endothelium (ET)-dependent vascular smooth muscle (VSM) relaxation in response to acetylcholine (ACh) and methacholine (MTH) in patients with type 2 diabetes (T2D) compared to controls

Table 3: Characteristic of studies that reported endothelium (ET)-independent vascular smooth muscle (VSM) relaxation in patients with type 2 diabetes (T2D) to sodium nitroprusside (SNP) and glyceryl trinitrate (GTN)

Table 4: Subgroup analysis of endothelium (ET)-independent vascular smooth muscle (VSM) relaxation in response to sodium nitroprusside (SNP) and glyceryl trinitrate (GTN) in patients with type 2 diabetes (T2D) compared to controls

[*] Corresponding Author:

PhD FRSC FRSB Khosrow Kashfi, Department of Molecular, Cellular, and Biomedical Sciences, Sophie Davis School of Biomedical Education, City University of New York School of Medicine, NY, USA NY, 10031, USA, eMail: Kashfi@med.cuny.edu