Journal of Diabetes and Its Complications 31 (2017) 1691–1697

Contents lists available at ScienceDirect

Journal of Diabetes and Its Complications

journal homepage: WWW.JDCJOURNAL.COM

Relation of elevated serum uric acid levels to first-degree heart block and
other cardiac conduction defects in hospitalized patients with type
2 diabetes
Alessandro Mantovani a,⁎, Riccardo Rigolon a, Isabella Pichiri a, Giovanni Morani b, Stefano Bonapace c,
Clementina Dugo c, Giacomo Zoppini a, Enzo Bonora a, Giovanni Targher a
a
Section of Endocrinology, Diabetes and Metabolism, Department of Medicine, University and Azienda Ospedaliera Universitaria Integrata of Verona, Verona, Italy
b
Section of Cardiology, Department of Medicine, University and Azienda Ospedaliera Universitaria Integrata of Verona, Verona, Italy
c
Division of Cardiology, “Sacro Cuore” Hospital, Negrar (VR), Italy

a r t i c l e i n f o a b s t r a c t

Article history: Aims: Several studies have reported that moderately elevated serum uric acid levels are associated with an in-
Received 30 June 2017 creased risk of tachyarrhythmias (mainly atrial fibrillation) in patients with and without type 2 diabetes mellitus
Received in revised form 21 August 2017 (T2DM). It is currently unknown whether an association also exists between elevated serum uric acid levels and
Accepted 17 September 2017 cardiac conduction defects in patients with T2DM.
Available online 20 September 2017
Methods: We retrospectively analyzed a hospital-based sample of 967 patients with T2DM discharged from our
Division of Endocrinology over the years 2007–2014. Standard electrocardiograms were performed on all pa-
Keywords:
Bradyarrhythmias
tients and were interpreted by expert cardiologists.
Conduction defects Results: Overall, 267 (27.6%) patients had some type of conduction defects on electrocardiograms (defined as at
Cardiovascular disease least one block among first-degree atrio-ventricular block, second-degree block, third-degree block, left bundle
Serum uric acid branch block, right bundle branch block, left anterior hemi-block or left posterior hemi-block). Patients in the
Hyperuricemia 3rd serum uric acid tertile had a higher prevalence of any cardiac conduction defects than those belonging to
2nd or 1st tertile, respectively (35.8% vs. 25.0% vs. 22.6%; p b 0.0001). Elevated serum uric acid levels were asso-
ciated with a nearly twofold increased risk of cardiac conduction defects after adjustment for age, sex, hemoglo-
bin A1c, diabetes duration, metabolic syndrome, chronic kidney disease, chronic obstructive pulmonary disease,
ischemic heart disease, valvular heart disease and medication use (adjusted-odds ratio 1.84, 95% confidence in-
tervals 1.2–2.9; p = 0.009).
Conclusions: Moderately elevated serum uric acid levels are associated with an increased prevalence of any car-
diac conduction defects in hospitalized patients with T2DM, independent of multiple risk factors and potential
confounding variables.
© 2017 Elsevier Inc. All rights reserved.

1. Introduction Conversely, acquired and persistent AV conduction defects are infre-

Bradycardia may occur when the conduction of impulse across the
atrio-ventricular (AV) node is altered with the eventuality of concomi-
tant symptoms (e.g., fatigue or syncope) and even death, particularly
when ventricular rates are entirely inefficient.1,2
Transient AV conduction defects are relatively frequent in healthy
young individuals and are caused by high vagal tone in most cases. 1,2
Abbreviations: AV, atrio-ventricular; BMI, body mass index; CKD, chronic kidney disease;
COPD, chronic obstructive pulmonary disease; ECG, electrocardiograms; HbA1c, hemoglobin
A1c; IHD, ischemic heart disease; LAH, left anterior hemi-block; LBBB, left bundle branch
block; LPH, left posterior hemi-block; RBBB, right bundle branch block; SUA, serum uric
acid; T2DM, type 2 diabetes mellitus; VHD, valvular heart disease.
Disclosure statement: The authors have no potential conflicts of interest to disclose.
⁎ Corresponding author at: Section of Endocrinology, Diabetes and Metabolism, University
and Azienda Ospedaliera Universitaria Integrata, Piazzale A, Stefani, 1, 37126 Verona, Italy.
E-mail address: alessandro.mantovani@univr.it (A. Mantovani).
https://doi.org/10.1016/j.jdiacomp.2017.09.011
1056-8727/© 2017 Elsevier Inc. All rights reserved.
quent in healthy individuals, but becomes more frequent in the set-
ting of older age, myocardial ischemia, type 2 diabetes mellitus
(T2DM) or infiltrative diseases. 1,2 Although it is known that high-
degree AV blocks (i.e., second-degree AV block and third-degree AV
block) strongly increase risk of future cardiovascular events in affected
patients, accumulating evidence now suggests that even PR interval
prolongation, first-degree AV block or bundle branch blocks are inde-
pendently associated with a poor cardiac prognosis. 3–8 In a meta-
analysis of 14 observational studies involving N400,000 individuals,
Kwok et al. reported a strong association between PR interval prolonga-
tion or first-degree AV block and increased risks of atrial fibrillation,
heart failure and death.9
For several decades, moderately elevated levels of serum uric acid
have been considered as a simple biochemical abnormality with little
or no clinical significance. However, in the last years, it has become
1692 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697
increasingly clear that moderately elevated serum acid uric levels are
independently associated with increased cardiovascular morbidity
and mortality in the general adult population, in patients with T2DM
and in other high-risk patient groups. 10,11 Convincing evidence has
also emerged of a strong association between elevated serum uric acid
levels and increased risk of atrial fibrillation in patients with and with-
out T2DM. 12–19
To our knowledge, it is currently unknown whether an association
also exists between moderately elevated serum uric acid levels and
risk of cardiac conduction defects in patients with T2DM, a group of in-
dividuals in whom these two pathologic conditions are somewhat re-
current. We believe that this topic merits further in-depth
investigation as it may further explain the increased risk of
cardiovascular mortality and morbidity observed in patients with
chronic hyperuricemia.
Thus, the main aim of this hypothesis-generating study was to ascer-
tain whether elevated serum acid uric levels are associated with an in-
creased prevalence of cardiac conduction defects on standard
electrocardiograms in a large hospital-based sample of patients with
T2DM.
2. Methods
2.1. Patients
We conducted a retrospective, cross-sectional study identifying all
white patients with known T2DM, who were discharged from our Divi-
sion of Endocrinology during 2007–2014. If a patient had multiple dis-
charges from the hospital during this period, the first discharge with
complete data was taken into consideration for statistical analysis.
Most of these patients were admitted to the hospital for chronic decom-
pensated diabetes, diabetic foot ulcers or infections.
A total of 1252 hospitalized patients with T2DM were initially iden-
tified in our database. We subsequently excluded 285 patients (22.8% of
total) with pacemakers or implantable cardioverter defibrillators, pre-
existing atrial fibrillation or flutter, severe valvular heart disease
(VHD) (including those with prosthetic heart valves), end-stage renal
disease, acute electrolyte abnormalities, decompensated cirrhosis, thy-
roid dysfunction as well as those with missing serum uric acid data
and those treated with amiodarone, propafenone, digitalis, non-
dihydropyridine calcium-channel blockers (CCB) or other anti-
arrhythmic agents, except beta-blockers. As a result of this selection,
967 (77.2%) patients were included in the final analysis.
The local ethics committee approved the study protocol. The ethics
committee exempted our research from the informed consent require-
ment because we only accessed retrospectively a de-identified database
for the purpose of data analysis.
2.2. Clinical and laboratory data
Data on age, sex, anthropometric variables (weight, height and body
mass index [BMI]) and blood pressure were recorded during medical visits.
Patients were considered to have arterial hypertension if their blood pres-
sure was ≥140/90 mm Hg or if they were taking any anti-hypertensive
agents (including beta-blockers or diuretics). Detailed information on co-
morbid conditions and current use of medications was collected in all pa-
tients by interviews during medical visits.
Venous blood samples were performed in the morning after an over-
night fast. Complete blood count, serum uric acid, creatinine, electrolytes
and other biochemical blood measurements were determined using stan-
dard laboratory procedures. Hemoglobin A1c (HbA1c) was measured by a
high-performance liquid chromatography analyzer on Tosoh G7 auto-
mated analyzer (Tosoh Bioscience Inc., San Francisco, CA; USA). Albu-
minuria was measured by an immuno-nephelometric method on a
morning spot urine sample and expressed as the albumin/creatinine
ratio on Beckman-Coulter IMMAGE (Beckman-Coulter Instruments,
Fullerton, CA; USA); macroalbuminuria was defined as an urinary albu-
min/creatinine ratio N 300 mg/g creatinine. The presence of chronic kid-
ney disease (CKD) was defined as presence of eGFRMDRD b 60 ml/min/
1.73 m 2 (estimated by using the Modification of Diet in Renal Diseases
[MDRD] study equation) or macroalbuminuria9. 20 Metabolic
syndrome was diagnosed by a modified Adult Treatment Panel (ATP)-
III definition, 21,22 as waist circumference was not available for all pa-
tients. In accordance with this modified ATP-III definition, a T2DM pa-
tient was classified as having the metabolic syndrome if he/she had at
least two of the following four risk abnormalities: (i) BMI N 28 kg/m 2
in men or N27 kg/m 2 in women; (ii) triglycerides ≥ 150 mg/dl; (iii)
HDL-cholesterol b 40 mg/dl in men and b50 mg/dl in women or receiv-
ing any lipid-lowering drugs; and (iv) blood pressure ≥ 130/85 mm Hg
or receiving any anti-hypertensive drugs.21,22
Ischemic heart disease (IHD) was defined as a documented history
of myocardial infarction, angina, coronary revascularization procedures
or typical electrocardiographic abnormalities according to the Minneso-
ta code.23 Pre-existing history of mild-to-moderate VHD was confirmed
by reviewing medical records of the hospital, including diagnostic
symptoms patterns and echocardiograms (when available). As previ-
ously reported, those with severe VHD or prosthetic valves were exclud-
ed from the study. The pre-existing history of peripheral artery disease
was based on medical history and examination (e.g., intermittent clau-
dication, rest pain or lower-extremity revascularizations) and was con-
firmed by reviewing medical records of the hospital of patients,
including radiologic imaging results. The pre-existing history of chronic
obstructive pulmonary disease (COPD) was confirmed by reviewing med-
ical records of the hospital, including diagnostic symptoms patterns, and
results of lung function tests. In most patients, the presence of microvas-
cular diabetic complications such as lower-extremity sensory
polyneuropathy (by biothesiometer or 5.07/10-g monofilament) and dia-
betic retinopathy (by fundoscopy) were also recorded.
2.3. Resting electrocardiograms
A standard 12-lead electrocardiogram (ECG) was performed in all
patients during the first 1–2 days of the hospital stay (and then repeated
when necessary). A 24-hour ECG Holter monitoring was not regularly
carried out. The diagnosis of cardiac conduction defects was made on
the basis of ECGs and confirmed by expert cardiologists, who were
blinded to patient's clinical data. In particular, the first-degree AV
block was defined as a PR interval duration of 200 ms or more with no
variation. 23 The diagnosis of second-degree AV block was made in ac-
cordance with the progressive prolongation of PR interval, culminating
in a non-conducted P wave.23 Afterwards second-degree AV block was
classified in Mobitz type I or Mobitz type II. A third-degree AV block
was identified, when P waves were not followed by QRS complexes.23
Analogously, the presence of complete right bundle branch block
(RBBB), left bundle branch block (LBBB), left anterior hemi-block (LAH
or left anterior fascicular block) or left posterior hemi-block (LPH or
left posterior fascicular block) were diagnosed in accordance with stan-
dard ECG criteria.23 In this study, a conventional echocardiography was
not available for all patients.
2.4. Statistical analysis
Data are reported as means ± SD, medians and inter-quartile ranges
(IQR) or percentages. Differences in main clinical and biochemical charac-
teristics of patients grouped according to tertiles of serum uric acid levels
(i.e., 1st tertile: ≤4.2 mg/dl, 2nd tertile: 4.3–5.7 mg/dl, and 3rd tertile:
≥5.8 mg/dl) were assessed using the one-way analysis of variance for nor-
mally distributed variables and the Kruskal-Wallis test for non-normally
distributed variables (i.e., diabetes duration, serum triglycerides and liver
enzymes). The chi-squared test was used to test for between-group differ-
ences among the categorical variables. Logistic regression analysis was used
to examine the association between serum uric acid tertiles and risk of
 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697
cardiac conduction defects (defined as presence of at least one electrocar-
diographic conduction defect among first-degree AV block, second-degree
AV block, third-degree AV block, LBBB, RBBB, LAH or LPH) after adjustment
for potential confounding variables. We also repeated the same logistic re-
gression analyses including serum uric acid levels as a continuous measure
(for 1-SD increment). We performed four forced-entry logistic regression
models: the first model was unadjusted; the second model was adjusted
for age and sex (model 1); the third model was adjusted for age, sex, dura-
tion of diabetes, HbA1c, metabolic syndrome, CKD, COPD, IHD and VHD
(model 2). Finally, the fourth regression model was further adjusted for
the use of medications, such as statins, anti-platelet drugs, anticoagulants
or allopurinol (model 3). Covariates included in multivariate logistic regres-
sion models were selected as potential confounding factors based on their
significance in univariate analyses or based on their biological plausibility.
A p-value b 0.05 was considered to be statistically significant. Statistical
analyses were performed using the SPSS software, version 21.0 (IBM
Corp, Armonk, NY).
3. Results
A total of 967 patients with T2DM were included in the study (56%
men, mean age of 65 years, median duration of diabetes of 12 years).
No patients had a history of gout. There were 318 total cardiac conduc-
tion defects among 267 (27.6%) patients, who had some type of persis-
tent conduction defects demonstrated on standard ECGs. In particular,
118 patients had a first-degree AV block, 4 had a second-degree AV
block (3 of whom had Mobitz type I and 1 had Mobitz type II), 1 had a
third-degree AV block, 95 had RBBB, 25 had LBBB, 73 had LAH, 2 had
LPH; 43 patients had the combination of first-degree AV block and
intra-ventricular defects (LBBB or RBBB).
The clinical, biochemical and electrocardiographic characteristics of
patients grouped according to serum uric acid tertiles are summarized
in Table 1. Compared with patients in the 1st and 2nd tertile, those in
the 3rd tertile of serum uric acid levels were more likely to be over-
weight/obese and hypertensive and had higher levels of serum triglyc-
erides and creatinine, longer electrocardiographic PR intervals and
tended to have longer QRS intervals. They also had a higher prevalence
of metabolic syndrome, CKD and COPD and lower circulating levels of
HbA1c, HDL-cholesterol and eGFRMDRD. The proportion of patients
treated with diuretics, dihydropyridine CCBs (it is important to remem-
ber that patients treated with non-dihydropyridine CCBs were excluded
from the study), beta-blockers or allopurinol was significantly greater
among patients in the 3rd tertile of serum uric acid levels than among
those belonging to 1st and 2nd tertile, whereas the proportion of pa-
tients treated with metformin was lower. Conversely, all other clinical
and biochemical (including serum electrolyte levels) variables did not
differ significantly among the three subgroups of patients; similarly,
the prevalence of the first-degree AV block and intra-ventricular defects
(combined) did not differ among the three subgroups.
Supplemental Table 1 shows the clinical, biochemical and electro-
cardiographic characteristics of patients stratified by presence or ab-
sence of cardiac conduction defects. Compared to those without any
cardiac conduction defects, patients with conduction defects were
more likely to be male, older and had longer duration of diabetes and
higher values of serum creatinine and uric acid as well as longer electro-
cardiographic PR, QRS and QTc intervals. They also had lower levels of
eGFRMDRD, total/LDL cholesterol, HbA1c, ALT and hemoglobin. Finally,
the prevalence of prior IHD, VHD, peripheral artery disease, CKD,
lower-extremity sensory neuropathy as well as the use of diuretics,
dihydropyridine CCBs, anti-platelet drugs, anticoagulants,
nitroderivates and statins were significantly higher in patients with
conduction defects than in those without. Notably, the percentage of
using beta-blockers was comparable between those with and those
without conduction defects.
As shown in Fig. 1, patients in the 3rd tertile of serum uric acid levels
had a remarkably higher prevalence of cardiac conduction defects
1693
(combined endpoint) compared with those in the 2nd and 1st tertile,
respectively.
Investigating each type of cardiac conduction defect separately
(Supplemental Fig. 1), the prevalence of first-degree AV block as well
as the prevalence of any AV blocks was higher in patients in the 3rd
tertile of serum uric acid levels compared with the other subgroups of
patients. Conversely, no significant differences were found in the prev-
alence of second-degree AV block, third-degree AV block, LBBB, RBBB,
LAH, LPH or any bundle branch block (combined) among the three sub-
groups of patients.
Table 2 shows the association between elevated serum uric acid
levels and the risk of cardiac conduction defects (combined endpoint).
In univariate logistic analysis (unadjusted model), patients in the 3rd
tertile of serum uric acid levels had an approximately twofold higher
risk of prevalent cardiac conduction defects. This association remained
statistically significant even after adjustment for age, sex, diabetes dura-
tion, HbA1c, metabolic syndrome (that also included hypertension sta-
tus and use of any anti-hypertensive agents), COPD, CKD, IHD and VHD
(adjusted models 1 and 2). Further adjustment for the use of certain
medications (including also allopurinol) did not weaken this association
(adjusted model 3). Similar findings were also observed when the same
analyses were repeated using serum uric acid levels as a continuous
measure (Table 2). Also in such case, a 1-SD increment (i.e., 1.81 mg/
dl) in serum uric acid levels was associated with an approximately
20% higher risk of cardiac conduction defects after adjusting for all po-
tential confounding factors included in the above-mentioned regression
model 3.
Almost identical results were also observed when patients, who took
beta-blockers (n = 292), were excluded from analysis (adjusted-odds
ratio for 3rd serum uric acid tertile: 1.59, 95% CI 1.05–2.5, p = 0.037;
adjusted-odds ratio for a 1-SD increment in serum uric acid levels:
1.20, 95% CI 1.01–1.5, p = 0.039).
As reported in Supplemental Table 2, when we performed further
separate analyses for evaluating the association of serum uric acid
tertiles with the risk of any AV block (panel A) or any bundle branch
block (panel B) or the presence of at least a conduction defect among
first-degree AV block, LBBB or RBBB (panel C), we found that moderate-
ly elevated serum uric acid levels were independently associated with
both the presence of any AV blocks and the presence of at least a con-
duction defect among first-degree AV block, LBBB or RBBB, but not
with the presence of any bundle branch blocks.
4. Discussion
The novel findings of our hypothesis-generating study are as fol-
lows: (1) patients with T2DM and moderately elevated levels of
serum uric acid had a markedly higher prevalence of cardiac conduction
defects on standard electrocardiograms (primarily first-degree AV
block) than those with normal serum uric acid levels; (2) moderately el-
evated serum uric acid levels were associated with an approximately
twofold increased risk of cardiac conduction defects; and (3) this asso-
ciation was only slightly attenuated after adjustment for established
cardiovascular risk factors, diabetes-related variables and other
coexisting comorbid conditions.
To our knowledge, this is the first cross-sectional study aimed at ex-
amining whether moderately elevated serum uric acid levels are associ-
ated with cardiac conduction defects in a large sample of hospitalized
patients with T2DM.
In clinical practice, it is frequent to observe acquired and persistent
conduction defects among patients with atherosclerosis, IHD or T2DM.
High-degree AV blocks (i.e., second-degree AV block and third-degree
AV block) are known risk factors for adverse cardiovascular outcomes
in affected patients.5,6,8 In the last years, however, a number of studies
have also suggested that prolonged PR interval, first-degree AV block
or RBBB were independently associated with an increased risk of all-
cause and cardiovascular mortality. 3,4,7,9 It is important to underline
1694 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697

Table 1
Clinical, biochemical and electrocardiographic characteristics of patients with type 2 diabetes, who were stratified by serum uric acid tertiles.

1st tertile (n = 332) 2nd tertile (n = 328) 3rd tertile (n = 307) P value

Age (years) 64.4 ± 13 65.3 ± 13 66.0 ± 12 0.284

Sex (male) (%) 53.0 55.2 60.0 0.169
Weight (kg) 75.8 ± 18.2 83.4 ± 21.8 90.3 ± 22.5 b0.001
BMI (kg/m2) 27.9 ± 6 30.4 ± 7 32.4 ± 7 b0.001
BMI N 30 kg/m2 (%) 30.7 40.4 58.4 b0.001
Current smokers (%), n = 219 49.5 53.5 46.2 0.595
Duration of diabetes (years) 11 (5–20) 13 (7–20) 12 (6–20) 0.220
Heart rate (bpm) 76 ± 13 75 ± 13 76 ± 13 0.289
Systolic blood pressure (mm Hg) 138 ± 20 142 ± 20 141 ± 20 0.034
Diastolic blood pressure (mm Hg) 79 ± 11 80 ± 11 81 ± 11 0.171
Hypertension (%) 72.3 82.0 85.3 b0.001
Metabolic syndrome (%) 73.8 85.4 93.2 b0.001
Fasting glucose (mg/dl) 204.7 ± 120 194.4 ± 117 190.5 ± 135 0.324
HbA1c (%) 10.1 ± 2.7 9.6 ± 2.5 9.4 ± 2.5 0.001
Total cholesterol (mg/dl) 171 ± 53 172 ± 49 175 ± 52 0.708
LDL-cholesterol (mg/dl) 98 ± 44 96 ± 38 95 ± 40 0.602
HDL-cholesterol (mg/dl) 43 ± 16 42 ± 14 38 ± 14 b0.001
Triglycerides (mg/dl) 123 (90–170) 142 (98–195) 170 (120–260) b0.001
ALT (U/l) 22 (16–34) 22 (15–33) 23 (16–37) 0.496
Creatinine (mg/dl) 0.92 ± 0.4 1.21 ± 0.7 1.40 ± 0.9 b0.001
eGFRMDRD (ml/min/1.72 m2) 82.9 ± 34 68.2 ± 26 55.9 ± 26 b0.001
Na (mmol/l) 140 ± 1.8 139 ± 2.1 139 ± 1.9 0.373
K (mmol/l) 4.1 ± 0.3 3.9 ± 0.4 3.9 ± 0.4 0.334
Hemoglobin (g/dl) 12.9 ± 1.7 12.9 ± 1.9 12.9 ± 1.8 0.862
Platelets (×109/l) 231 ± 79 239 ± 84 232 ± 68 0.297
IHD (%) 17.8 19.8 23.5 0.197
VHD (%) 6.0 7.9 10.4 0.124
COPD (%) 3.3 4.9 8.1 0.023
CKD (%) 21.6 42.1 65.7 b0.001
Macroalbuminuria (%) 6.8 10.9 18.2 b0.001
Diabetic retinopathy (%), any degree 37.0 44.0 45.4 0.090
Peripheral sensory neuropathy (%) 24.1 30.8 30.2 0.113
Peripheral artery disease (%) 47.2 53.6 57.1 0.096
Neoplasms (%) 5.1 9.1 5.9 0.091
Insulin users (%) 73.8 69.5 70.4 0.438
Metformin users (%) 36.1 44.5 32.9 0.008
Sulfonylurea users (%) 19.0 17.7 22.8 0.245
Incretin users (%) 9.7 11.6 9.1 0.288
ACE-inhibitor users (%) 46.1 54.3 52.4 0.088
ARB users (%) 19.0 22.6 21.8 0.492
Beta-blocker users (%) 23.8 30.2 37.1 0.001
Dihydropyridine CCB users (%) 26.8 34.5 35.2 0.040
Diuretic users (%) 33.4 46.0 64.8 b0.001
Antiplatelet drug users (%) 56.6 63.4 61.6 0.183
Anticoagulant drug users (%) 6.3 5.2 6.2 0.795
Nitro-derivate drug users (%) 9.0 11.9 14.0 0.143
Statin users (%) 52.4 61.0 55.4 0.080
Allopurinol users (%) 5.6 6.6 19.2 b0.001
Serum uric acid (mg/dl) 3.4 ± 0.6 4.9 ± 0.4 7.2 ± 1.5 ND
Any cardiac conduction defect (%) 22.6 25.0 35.8 b0.0001
Combination of 1st AV block and intra-ventricular defects (LBBB or RBBB) (%) 3.6 4.6 5.2 0.602

Electrocardiographic characteristics
Heart rate (bpm) 76.3 ± 13.1 74.7 ± 13.1 75.7 ± 13.3 0.289
PR interval (ms) 163 ± 32 166 ± 33 177 ± 39 0.039
QRS interval (ms) 95 ± 20 97 ± 21 99 ± 20 0.057
QTc interval (ms) 434 ± 36 433 ± 43 440 ± 31 0.293

Sample size, n = 967 except where indicated. Data are expressed as means ± SD, medians and IQR (in parenthesis) or percentages.
Differences between the two groups were tested by the chi-squared test for categorical variables, the one-way ANOVA for normally distributed continuous variables and the Kruskal-Wallis
test for non-normally distributed continuous variables (i.e., diabetes duration, serum triglycerides and ALT).
Abbreviations: ACE, angiotensin-converting-enzyme; ALT, alanine aminotransferase; ARB, angiotensin II receptor blockers; BMI, body mass index; CCB, calcium-channel blocker; CKD,
chronic kidney disease; COPD, chronic obstructive pulmonary disease; eGFR, glomerular filtration rate estimated by using the Modification of Diet in Renal Diseases (MDRD) study equa-
tion; IHD, ischemic heart disease; K, potassium; Na, sodium; ND, not determined; VHD, valvular heart disease.
Note: CKD was defined as eGFRMDRD b 60 ml/min/1.73 m2 or macro-albuminuria; the metabolic syndrome was defined by a modified ATP III definition; hypertension was defined as
blood pressure ≥ 140/90 mm Hg or use of any anti-hypertensive drugs; presence of any cardiac conduction defect was defined as the presence of at least one electrocardiographic con-
duction defect among first-degree AV block, second-degree AV block, third-degree AV block, LBBB, RBBB, LAH or LPH.

that, in our study, the prevalence of any cardiac conduction defects was
approximately 36% in patients in the 3rd serum uric acid tertile and that
the prevalence of any AV block was approximately 20% in these pa-
tients. These two prevalences are much higher than those observed
both in the general adult population 24 and in other high-risk patient
groups. 25 A possible explanation for these findings is that our sample
was composed of hospitalized patients with T2DM and that those in
the upper serum uric acid tertile also had more comorbidities than
those belonging to the 2nd and 1st tertile.
Several studies have suggested that moderately elevated serum uric
acid levels are an emerging, non-traditional risk factor for cardiovascu-
lar disease both in general adult population and in some high-risk pa-
tient groups. 10,11 Large observational studies also reported a
significant association between moderately elevated serum uric acid
 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697 1695

40
p<0.0001

Percentage of conduction defects (%)

35

30

25

20

15

10

0
1st Tertile (n=332) 2nd Tertile (n=328) 3rd Tertiles (n=307)

Serum uric acid tertiles

Fig. 1. Prevalence of any cardiac conduction defects (combined endpoint) on standard 12-lead electrocardiograms in a hospital-based sample of 967 patients with type 2 diabetes stratified
by serum uric acid tertiles (i.e., 1st tertile: ≤4.2 mg/dl, 2nd tertile: 4.3–5.7 mg/dl, and 3rd tertile: ≥5.8 mg/dl).

levels and the risk of developing arterial hypertension, vascular demen-
tia and CKD.10,11 In addition, a number of studies also documented a link
between elevated serum uric acid levels and risk of atrial
fibrillation.12–19
In our opinion, the results of this study expand the above-mentioned
observations and may have important clinical implications. Firstly, our
data provide new and additional information regarding a potential ar-
rhythmogenic role of moderately elevated serum uric acid levels. Sec-
ondly, the strong and positive association we observed between
serum uric acid levels and the risk of cardiac conduction defects (princi-
pally first-degree AV block) may partly contribute to explain the in-
creased risk of cardiovascular morbidity and mortality observed
among patients with chronic hyperuricemia. However, the prevalence
of patients with coexisting first-degree AV block and intra-ventricular
defects (LBBB or RBBB), which identifies a subgroup of individuals at
higher risk of developing cardiovascular complications, 26,27 was not sta-
tistically different across serum uric acid tertiles. This finding might also
be partly due to the relatively low number of patients (4.4% of total)
with this more severe cardiac conduction defect. However, larger stud-
ies are needed to better elucidate this issue.
Presently, the underlying pathophysiological mechanisms responsible
for the association between moderately elevated serum uric acid levels
and cardiac conduction defects are not entirely understood. An obvious
explanation for our results is that the significant association between ele-
vated serum uric acid levels and cardiac conduction defects may be an
epiphenomenon of coexisting established risk factors and comorbidities,
including older age, poor glycemic control, metabolic syndrome, IHD,
VHD, COPD, CKD and use of certain medications (e.g., beta-blockers,
dihydropyridine CCBs and diuretics). However, it is important to under-
line that, in our study, the significant association between moderately el-
evated serum uric acid levels and cardiac conduction defects persisted
even after adjustment for the above-mentioned risk factors and comorbid
conditions. Hence, it is reasonable to assume that elevated levels of serum
uric acid might partly contribute to the development and persistence of
cardiac conduction defects (or, perhaps, of some specific conduction de-
fects because in our study the association between serum uric acid levels

Table 2
Logistic regression analyses. Association between serum uric acid levels (considered as either categorical or continuous measure) and the risk of cardiac conduction defects on standard
electrocardiograms in patients with type 2 diabetes.

Logistic regression models Odds ratio ± 95% CI P valuea Odds ratio ± 95% CI P valueb P for trend

Serum uric acid tertiles 1st tertile (≤4.2 mg/dl) 2nd tertile (4.3–5.7 mg/dl) 3rd tertile (≥5.8 mg/dl)

Unadjusted model Ref. 1.14 (0.8–1.6) 0.467 1.91 (1.3–2.7) b0.001 b0.001
Adjusted model 1 Ref. 1.08 (0.7–1.6) 0.663 1.81 (1.2–2.6) 0.001 0.002
Adjusted model 2 Ref. 1.15 (0.8–1.8) 0.504 1.76 (1.1–2.8) 0.012 0.027
Adjusted model 3 Ref. 1.17 (0.8–1.8) 0.472 1.84 (1.2–2.9) 0.009 0.020

Logistic regression models Odds ratio ± 95% CI P value

1-SD increment in serum uric acid (mg/dl)

Unadjusted model 1.26 (1.09–1.4) 0.001
Adjusted model 1 1.25 (1.09–1.5) 0.002
Adjusted model 2 1.19 (1.01–1.4) 0.042
Adjusted model 3 1.22 (1.03–1.5) 0.025

Sample size, n = 967. Data are expressed as odds ratios ± 95% confidence intervals (CI) as assessed by either univariate (unadjusted) or multivariate logistic regression analyses. The
presence of cardiac conduction defects (n = 267; defined as presence of at least one of the following electrocardiographic heart blocks: first-degree AV block, second-degree AV block,
third-degree AV block, LBBB, RBBB, LAH or LPH) was included as the dependent variable.
Other covariates included in these multivariate regression models (along with serum uric acid levels) were as follows: model 1: adjusted for age, sex; model 2: adjusted for age, sex, HbA1c,
duration of diabetes, metabolic syndrome (by a modified ATP III definition), CKD, COPD, IHD and VHD; model 3: the same covariates included in model 2 plus the use of statins, anti-plate-
lets drugs, anti-coagulants or allopurinol.
a
2nd tertile group vs. 1st tertile group.
b
3rd tertile group vs. 1st tertile group.
1696 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697
and risk of cardiac conduction defects appeared to be mainly driven by
the presence of the first-degree AV block). Indeed, experimental studies
have shown that chronic, asymptomatic hyperuricemia may promote cir-
culatory endothelial dysfunction, low-grade inflammatory state, in-
creased oxidative stress and activation of the renin-angiotensin-
aldosterone system.18,28–30 Experimentally, it is also known that several
pro-inflammatory factors and multiple reactive oxygen species may pro-
mote important derangements in the structural and electrophysiological
substrates of the myocardium, inducing an increased vulnerability to car-
diac arrhythmias.31,32 Hence, it is also plausible to speculate that all these
hyperuricemia-induced changes may partly contribute to the pro-
gressive fibrosis of the His-Purkinje system and deep derangements
in its distribution and transmission velocities, producing a delay or
even a stop of the impulse conduction across the AV node, His bundle
and bundle branches. However, further research is certainly needed
to clarify these pathophysiological issues.
Our study has some important limitations that should be mentioned.
Firstly, given its retrospective and cross-sectional design, we cannot
draw any conclusion about the temporality or causality of the findings.
Secondly, a potential selection bias of including a hospital-based sample
of patients with T2DM cannot be definitely excluded. So, our results
may not be necessarily generalizable to a more general population of
nondiabetic individuals. Thirdly, a conventional echocardiography was
not available for all patients. Fourthly, although in our study the associ-
ation between moderately elevated serum uric acid levels and risk of
cardiac conduction defects appears to be mainly driven by the presence
of first-degree AV block, we cannot be certain if serum uric acid levels
are a strong predictor of mild conduction defects only (e.g., first degree
AV block) or if, by expanding the sample size of the study, they are also a
predictor of more severe cardiac conduction defects (e.g., high-degree
AV blocks or LBBB) in patients with T2DM. Future larger studies are
timely needed to better elucidate this issue. Finally, although our multi-
variate logistic regression models were extensive, we cannot exclude
the possibility of residual confounding by some unmeasured factors
that might partly explain the results.
Despite these limitations, our study has also important strengths, in-
cluding the relatively large sample size, the ability to adjust for multiple
potential confounding factors and the exclusion of patients with pace-
makers or implantable cardioverter defibrillators, those who took
anti-arrhythmic agents and those with serious comorbidities, such as
atrial fibrillation/flutter, severe VHD or prosthetic valves, decompensat-
ed cirrhosis, malignancies and kidney failure. We believe that the inclu-
sion of these patients might have confounded the interpretation of data.
In conclusion, this pilot, cross-sectional study showed for the first
time that moderately elevated serum uric acid levels are associated
with an increased prevalence of cardiac conduction defects (mainly
first-degree AV block) in a large hospital-based sample of patients
with T2DM, independent of multiple established risk factors and poten-
tial confounding factors. Further studies are required to elucidate the bi-
ological mechanisms responsible for this association, and to establish
whether elevated serum uric acid levels may predict the development
of cardiac conduction defects in patients with T2DM.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jdiacomp.2017.09.011.
Funding sources
GT is supported in part by grants from the School of Medicine,
University of Verona, Verona, Italy.
Authors' contributions
AM and GT conceived and designed the study. RR, IP, SB and CD
researched data and reviewed/edited the manuscript. GM, GZ and EB
contributed to discussion and reviewed/edited the manuscript. AM
and GT analyzed the data and wrote the manuscript draft. AM is the
guarantor of this work and, as such, had full access to all the data in
the study and takes responsibility for the integrity of the data and the
accuracy of the data.
References
1. Braunwald's Heart Disease. A Textbook of Cardiovascular Medicine9th edition. .
2012:818-24.
2. Lee S, Wellens HJ, Josephson ME. Paroxysmal atrioventricular block. Heart Rhythm.
2009;6:1229-34.
3. Cheng S, Keyes MJ, Larson MG, McCabe EL, Newton-Cheh C, Levy D, et al. Long-term
outcomes in individuals with prolonged PR interval or first-degree atrioventricular
block. JAMA. 2009;301:2571-7.
4. Bussink BE, Holst AG, Jespersen L, Deckers JW, Jensen GB, Prescott E. Right bundle
branch block: prevalence, risk factors, and outcome in the general population: results
from the Copenhagen City Heart Study. Eur Heart J. 2013;34:138-46.
5. Coumbe AG, Naksuk N, Newell MC, Somasundaram PE, Benditt DG, Adabag S.
Long-term follow-up of older patients with Mobitz type I second degree atrioventric-
ular block. Heart. 2013;99:334-8.
6. Pokorney SD, Radder C, Schulte PJ, Al-Khatib SM, Tricocci P, Van de Werf F, et al.
High-degree atrioventricular block, asystole, and electro-mechanical dissociation
complicating non-ST-segment elevation myocardial infarction. Am Heart J.
2016;171:25-32.
7. Zhang ZM, Rautaharju PM, Prineas RJ, Loehr L, Rosamond W, Soliman EZ. Ventricular
conduction defects and the risk of incident heart failure in the Atherosclerosis Risk in
Communities (ARIC) Study. J Card Fail. 2015;21:307-12.
8. Auffret V, Loirat A, Leurent G, Martins RP, Filippi E, Coudert I, et al. High-degree atrio-
ventricular block complicating ST segment elevation myocardial infarction in the
contemporary era. Heart. 2016;102:40-9.
9. Kwok CS, Rashid M, Beynon R, Barker D, Patwala A, Morley-Davies A, et al. Prolonged
PR interval, first-degree heart block and adverse cardiovascular outcomes: a system-
atic review and meta-analysis. Heart. 2016;102:672-80.
10. Feig DI, Kang D-H, Johnson RJ. Uric acid and cardiovascular risk. N Engl J Med.
2008;359:1811-21.
11. Zoppini G, Targher G, Bonora E. The role of serum uric acid in cardiovascular disease
in type 2 diabetic and non-diabetic subjects. A narrative review. J Endocrinol Investig.
2011;34:881-6.
12. Tamariz L, Hernandez F, Bush A, Palacio A, Hare JM. Association between serum uric
acid and atrial fibrillation: a systematic review and meta-analysis. Heart Rhythm.
2014;11:1102-8.
13. Kuwabara M, Niwa K, Nishihara S, Nishi Y, Takahashi O, Kario K, et al. Hyperuricemia
is an independent competing risk factor for atrial fibrillation. Int J Cardiol. 2017;231:
137-42.
14. Chao TF, Hung CL, Chen SJ, Wang KL, Chen TJ, Lin YJ, et al. The association between
hyperuricemia, left atrial size and new-onset atrial fibrillation. Int J Cardiol.
2013;168:4027-32.
15. Xu X, Du N, Wang R, Wang Y, Cai S. Hyperuricemia is independently associated with
increased risk of atrial fibrillation: a meta-analysis of cohort studies. Int J Cardiol.
2015;184:699-702.
16. Chuang SY, Wu CC, Hsu PF, Chia-Yu Chen R, Liu WL, Hsu YY, et al. Hyperuricemia and
incident atrial fibrillation in a normotensive elderly population in Taiwan. Nutr Metab
Cardiovasc Dis. 2014;24:1020-6.
17. Tamariz L, Agarwal S, Soliman EZ, Chamberlain AM, Prineas R, Folsom AR, et al. Asso-
ciation of serum uric acid with incident atrial fibrillation (from the Atherosclerosis
Risk in Communities [ARIC] study). Am J Cardiol. 2011;108:1272-6.
18. Mantovani A, Rigolon R, Pichiri I, Pernigo M, Bergamini C, Zoppini G, et al. Hyperuri-
cemia is associated with an increased prevalence of atrial fibrillation in hospitalized
patients with type 2 diabetes. J Endocrinol Investig. 2016;39:159-67.
19. Valbusa F, Bertolini L, Bonapace S, Zenari L, Zoppini G, Arcaro G, et al. Relation of el-
evated serum uric acid levels to incidence of atrial fibrillation in patients with type 2
diabetes mellitus. Am J Cardiol. 2013;112:499-504.
20. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. Modification of diet in Renal
Disease Study Group. A more accurate method to estimate glomerular filtration rate
from serum creatinine: a new prediction equation. Ann Intern Med. 1999;130:461-70.
21. Sattar N, Gaw A, Scherbakova O, Ford I, O'Reilly DS, Haffner SM, et al. Metabolic syn-
drome with and without C-reactive protein as a predictor of coronary heart disease
and diabetes in the West of Scotland Coronary Prevention Study. Circulation.
2003;108:414-9.
22. Girman CJ, Rhodes T, Mercuri M, Pyörälä K, Kjekshus J, Pedersen TR, et al. The meta-
bolic syndrome and risk of major coronary events in the Scandinavian Simvastatin
Survival Study (4S) and the Air Force/Texas Coronary Atherosclerosis Prevention
Study (AFCAPS/TexCAPS). Am J Cardiol. 2004;93:136-41.
23. Prineas RJ, Crow RS, Zhang ZM, eds. The Minnesota code manual of electrocardio-
graphic findings: standards and procedures for ECG measurement in epidemiologic
and clinical trials. 2nd edition. London: Springer-Verlag; 2010.
24. Aro AL, Anttonen O, Kerola T, Junttila MJ, Tikkanen JT, Rissanen HA, et al. Prognostic
significance of prolonged PR interval in the general population. Eur Heart J. 2014;35:
123-9.
25. Uhm JS, Shim J, Wi J, Mun HS, Park J, Park SH, et al. First-degree atrioventricular block
is associated with advanced atrioventricular block, atrial fibrillation and left ventric-
ular dysfunction in patients with hypertension. J Hypertens. 2014;32:1115-20.
26. Miller WL, Hodge DO, Hammill SC. Association of uncomplicated electrocardiographic
conduction blocks with subsequent cardiac morbidity in a community-based population
(Olmsted County, Minnesota). Am J Cardiol. 2008;101:102-6.
 A. Mantovani et al. / Journal of Diabetes and Its Complications 31 (2017) 1691–1697 1697

27. Fahy GJ, Pinski SL, Miller DP, McCabe N, Pye C, Walsh MJ, et al. Natural history of iso-
lated bundle branch block. Am J Cardiol. 1996;77:1185-90.
28. Khosla U, Zharikov S, Finch J, Nakagawa T, Roncal C, Mu W, et al. Hyperuricemia in-
duces endothelial dysfunction. Kidney Int. 2005;67:1739-42.
29. Kato M, Hisatome I, Tomikura Y, Kotani K, Kinugawa T, Ogino K, et al. Status of endo-
thelial dependent vasodilatation in patients with hyperuricemia. Am J Cardiol.
2005;96:1576-8.
30. Perez-Ruiz F, Becker MA. Inflammation: a possible mechanism for a causative role of
hyperuricemia/gout in cardiovascular disease. Curr Med Res Opin. 2015:9-14.
31. Hu YF, Chen YJ, Lin YJ, Chen SA. Inflammation and the pathogenesis of atrial fibrilla-
tion. Nat Rev Cardiol. 2015;12:230-43.
32. Wu N, Xu B, Xiang Y, Wu L, Zhang Y, Ma X, et al. Association of inflammatory factors
with occurrence and recurrence of atrial fibrillation: a meta-analysis. Int J Cardiol.
2013;169:62-72.
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.