Author’s Accepted Manuscript

Hamilton rating scale for depression-24 (HAM-

D24) as a novel predictor for diabetic microvascular
complications in type 2 diabetes mellitus patients

Shuo Pan, Zhong-Wei Liu, Shuang Shi, Xun Ma,

Wen-Qian Song, Gong-Chang Guan, Yong Zhang,
Shun-Ming Zhu, Fu-Qiang Liu, Bo Liu, Zhi-Guo
www.elsevier.com/locate/psychres
Tang, Jun-Kui Wang, Ying Lv

PII: S0165-1781(17)30089-6
DOI: http://dx.doi.org/10.1016/j.psychres.2017.07.050
Reference: PSY10684
To appear in: Psychiatry Research
Received date: 15 January 2017
Accepted date: 25 July 2017
Cite this article as: Shuo Pan, Zhong-Wei Liu, Shuang Shi, Xun Ma, Wen-Qian
Song, Gong-Chang Guan, Yong Zhang, Shun-Ming Zhu, Fu-Qiang Liu, Bo Liu,
Zhi-Guo Tang, Jun-Kui Wang and Ying Lv, Hamilton rating scale for
depression-24 (HAM-D24) as a novel predictor for diabetic microvascular
complications in type 2 diabetes mellitus patients, Psychiatry Research,
http://dx.doi.org/10.1016/j.psychres.2017.07.050
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Hamilton rating scale for depression-24 (HAM-D24) as a novel

predictor for diabetic microvascular complications in type 2

diabetes mellitus patients

Shuo Pana1, Zhong-Wei Liua1, Shuang Shia1, Xun Mab, Wen-Qian Songc, Gong-Chang
Guana, Yong Zhanga, Shun-Ming Zhua, Fu-Qiang Liua, Bo Liua, Zhi-Guo Tanga,
Jun-Kui Wanga*, Ying Lva*

a
First Department of Cardiology, People's Hospital of Shaanxi Province, Xi’an,
Shaanxi, People’s Republic of China

b
Department of Emergency Medicine, People's Hospital of Shaanxi Province, Xi’an,
Shaanxi, People’s Republic of China

c
Department of Internal Medicine, University Hospital of Northwest University,
Xi’an, Shaanxi, People’s Republic of China

*Correspondence to: Jun-Kui Wang M.D, Ph.D and Ying Lv M.D, Ph.D, First
Department of Cardiology, People's Hospital of Shaanxi Province, Xi’an, People’s
Republic of China, 710068, Tel: +86 29 85251331 E-mail: wjk_sx@163.com (J-KW)
and springlvy@163.com (YL)

1
These authors contributed equally to the manuscript.
ABSTRACT

The study was designed to investigate whether the hamilton rating scale for
depression (24-items) (HAM-D24) can be used to predict the diabetic microvascular
complications in type 2 diabetes mellitus (T2DM) patients. 288 hospitalized patients
with T2DM were enrolled. Their diabetic microvascular complications including
diabetic nephropathy, diabetic retinopathy, diabetic peripheral neuropathy and diabetic
foot as well as demographic, clinical data, blood samples and echocardiography were
documented. All the enrolled patients received HAM-D24 evaluation. The HAM-D24
score and incidence of depression in T2DM patients with each diabetic microvascular
complication were significantly higher than those in T2DM patients without each
diabetic microvascular complication. After the adjustment of use of insulin and
hypoglycemic drug, duration of T2DM, mean platelet volume, creatinine, albumin,
fasting glucose, glycosylated hemoglobin type A1C, left ventricular ejection fraction,
respectively, HAM-D24 score was still significantly associated with diabetic
microvascular complications (OR=1.188-1.281, all P﹤0.001). The AUC of HAM-D24
score for the prediction of diabetic microvascular complication was 0.832
(0.761-0.902). 15 points of HAM-D24 score was considered as the optimal cutoff with
the sensitivity of 0.778 and specificity of 0.785. In summary, HAM-D24 score may be
used as a novel predictor of diabetic microvascular complications in T2DM patients.

Keywords: hamilton rating scale for depression; diabetic complications;

microvascular; type 2 diabetes mellitus
1. Introduction

Type 2 Diabetes Mellitus (T2DM) is a wide-spreading disease worldwide,

estimated of 415 million people suffer from T2DM, and the number is expected to
reach approximately 642.6 million by 2040 according to the International Diabetes
Federation (IDF) (IDF Diabetes Atlas Group, 2013). The microcirculation in T2DM
patients has varying degrees of abnormalities, the basement membrane lesions often
interact with the microcirculation, promoting the aggravation of microvascular disease
(Milligan, 2016). The microvascular diseases mainly occur in the kidney, retina, nerve
tissue and toes. Clinically, they were named as diabetic nephropathy, diabetic
retinopathy, diabetic peripheral neuropathy and diabetic foot (Ali et al., 2013).

The prevalence of depression in T2DM patients is higher than that in people

without this disease (Lesperance and Frasure-Smith, 2000; Anderson et al., 2001;
Pouwer et al., 2010), the overall prevalence ranges from 8% to 44.66% (Lloyd et al.,
2000; Sweileh et al., 2014; Mocan et al., 2016). Depression was reported to decrease
the quality of life (Schram et al., 2009; Jacobson et al., 1997; Egede and
Hernandez-Tejada, 2013), to increase mortality rates (van Dooren et al., 2013; Katon
et al., 2006) and economic burden in patients with T2DM (Subramaniam et al., 2009;
Egede etal., 2002). The depression also was reported to be associated with lack of
physical exercise, unhealthy diet, smoking, and poorer physical functioning (Lin et al.,
2004 ), which would in return promote the development and aggravation of T2DM.

Limited studies have indicated that the depression is also associated with a
variety of diabetic microvascular complications such as diabetic neuropathy,
retinopathy, nephropathy (De Groot et al., 2001; Lin et al., 2010). Considering the
association between the depression and diabetic microvascular complications, we
hypothesized that some indicator of depression may be used as the predictors of
microvascular complications in T2DM patients. In this present study, a commonly
used scale for depression evaluation, the Hamilton Rating Scale for Depression
(24-items) (HAM-D24) (Bech et al., 2015; Carneiro et al., 2015) was applied to predict
the microvascular complications in T2DM patients. In this case, a simple depression
scale would first demonstrate the depression status in T2DM patients in time, then the
psychological treatment and antidepressant medication could be prescribed
immediately to release the depression mood in this population. Second, via the
HAM-D24 depression score, we would be able to predict the microvascular
complications to some extent, which will guide our diagnosis confirming procedures,
meanwhile, to guarantee the early and adequate detection of diabetic microvascular
complications in T2DM patients.

2. Materials and methods

2.1. Subjects

We performed this present cross-sectional and observational study in 1st

Cardiology Department of People's Hospital of Shaanxi Province from January 2015
to October in 2016. We enrolled hospitalized patients with T2DM in our department.
T2DM was defined as fasting plasma glucose ≥ 7.0mmol/L, or random plasma
glucose ≥ 11.1mmol/L, or 2 hour plasma glucose in oral glucose tolerance test (OGTT)
≥ 11.1mmol/L, or use of insulin or oral hypoglycemic agents, or a previous history of
type 2 diabetes mellitus (Grimaldi and Heurtier, 1999). The main exclusion criteria
included significant cognitive impairment, alcohol or drug dependence within the
previous year, psychoses, bipolar disorder, severe personality disorder, active suicidal
ideation, life-threatening comorbidity, and current use of antipsychotic or
antidepressant medications (O'Connor et al., 2010).

288 hospitalized patients with T2DM were enrolled, 168 (58.3%) patients were
male and 120 (41.7%) patients were female. The average age was 63.16±10.51 years.
Among the 288 hospitalized T2DM patients, 51 patients were documented to have
diabetic microvascular complication. 20 patients had diabetic nephropathy, 10 patients
had diabetic retinopathy, 15 patients had diabetic peripheral neuropathy, 6 patients
had diabetic foot. The definitions of the microvascular diabetic complications were as
follows: I. Diabetic nephropathy was defined as urinary albumin excretion rate
(UAER) ≥200 ug/min twice during hospitalization or urine protein﹥0.5g/24h via
routine urinary examination after the T2DM was diagnosed, or a previous history of
diabetic nephropathy. II. Diabetic retinopathy was diagnosed using results of
nonmydriatic digital stereoscopic retinal imaging conducted by the ophthalmologists
in ophthalmology department after the T2DM was diagnosed, or a previous history of
diabetic retinopathy. III. Diabetic peripheral neuropathy was defined as 2 or more
abnormalities of following 5 abnormalities: a. temperature sensory abnormality; b.
diminished or vanished foot sensory; c. vibration sensory abnormality; d. vanished
ankle reflex; e. slow nerve conduction velocity after the T2DM was diagnosed, or a
previous history of diabetic peripheral neuropathy. IV. Diabetic foot was defined as
ulcer, or infection, or osteomyelitis, or Charcot joint disease, or toe deformity, or
amputation after the T2DM was diagnosed, or a previous history of diabetic foot.

2.2. Depression assessments

The heart failure patients were evaluated for depressive symptoms by using
Hamilton Rating Scale for Depression (24-items) (HAM-D24), it contained a total of
24 items (10 item was defined from 0 to 2, and 14 items were defined from 0 to 4).
Questions of 0-2 points were defined as none (0), mild-moderate (1), severe (2).
Questions of 0-4 points were defined as none (0), mild (1), moderate (2), severe (3),
very severe (4). HAM-D24 score <8 points was defined as non-depression and
HAM-D24 score ≥8 points was defined as depression. HAM-D24 score of 8-19 points
was defined as mild depression, HAM-D24 score of 20-34 points was defined as
moderate depression, HAM-D24 score of ≥35 points was defined as severe depression
(Zhu et al., 2015). The measurements of HAM-D24 were conducted by trained
physicians on the first day after admission. Two physicians in our department were
sent to psychology department for a week to learn the talking, observing and
interpreting skill of HAM-D24. The two physicians independently scores for each
patients, the mean scores taken by the physicians were used as the final score. If the
two scores were severely diverted, the repeated test will be performed by a
psychologist in psychology department.

2.3. Demographic and clinical data

Demographic data and cardiovascular risk factors were obtained from the
medical records. Body weight was measured while the subjects were without shoes by
using a double balance placed on a firm surface. Height was measured by using a
Frankfort plane positioned at a 90° angle against a wall-mounted metal tape. The
waist circumference measurements were taken at the end of normal expiration and to
the nearest 0.1 cm, measuring from the narrowest point between the lower borders of
the rib cage and the iliac crest (Koivukangas et al., 2016).

2.4. Blood samples and echocardiography

Peripheral blood was sampled from patients in a fasting state on the morning
following the admission day. Venous blood samples were sent to Clinical Laboratory
Department of People's Hospital of Shaanxi Province for red blood cells (RBC)
counts, hemoglobin, platelet counts, plateletcrit, mean platelet volume (MPV), platelet
distribution width (PDW), white blood cells (WBC) counts, neutrophilic granulocyte
percentage (NEUT%), total cholesterol (TC), triglyceride, high-density
lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C),
apolipoprotein A, apolipoprotein B, urea nitrogen (BUN), creatinine, cystatin-C, total
bilirubin (TBIL), direct bilirubin (DBIL), albumin, brain natriuretic peptide (BNP),
fasting glucose (FG) and glycosylated hemoglobin type A1C (HbA1C) detection
using standard biochemical techniques (Pan et al., 2016). Echocardiographic data (left
ventricular ejection fraction [LVEF]) was obtained using Doppler echocardiography
conducted within 3 days of admission (Korkmaz et al., 2016).

2.5. Definition of risk factors

Coronary artery disease was defined as the presence of at least one significant
coronary artery stenosis of more than 50% luminal diameter in coronary angiography
or coronary computed tomographic angiography (CTA).

Hypertension was defined as an average systolic blood pressure ≥ 140 mm Hg,

or an average diastolic blood pressure ≥ 90 mm Hg, or both, or self-reported use of
antihypertensive medication, or a self-reported history of hypertension.

Atrial fibrillation was diagnosed mainly via electrocardiogram (ECG),

characteristic findings included absence of P waves and irregular R-R intervals.

Dilated cardiomyopathy was diagnosed mainly via echocardiogram,

which showed left ventricular dilatation with normal or thinned walls and
reduced ejection fraction. Meanwhile, the history of coronary artery disease,
hypertension, valvular heart disease and other heart diseases should be excluded
before making the present diagnosis.

Smoking index was defined as number of cigarettes smoked per day × years of
smoking.

BMI was calculated as weight in kg divided by height in m2.

2.6. Statistical analysis

The statistical analysis was conducted using SPSS version 16.0 for Windows
(SPSS Inc., Chicago, IL, USA). Continuous variables were expressed as mean ±
standard deviations and the differences between the two groups were analyzed using
the Mann-Whitney U-test. Categorical variables were expressed as proportions and
the differences in categorical variables were analyzed using chi-square or fisher exact
test. The HAM-D24 score in T2DM patients with and without each diabetic
microvascular complication was compared using Mann-Whitney U-test. The
incidence of depression determined by HAM-D24 score were compared in T2DM
patients with and without diabetic microvascular complication using chi-square or
fisher exact test. The Logistic regression analysis was performed to determine
whether the HAM-D24 score was associated with diabetic microvascular complication
after the adjustment of each confounding factor of diabetic microvascular
complication in T2DM patients. The Receiver Operating Characteristic (ROC) curve
and Area Under roc Curve (AUC) of HAM-D24 score for the prediction of diabetic
microvascular complication in T2DM patients was performed. Additionally, the
distance on the ROC curve of each HAM-D24 score was calculated as the square root
of [(1 -sensitivity)2 + (1 - specificity)2]. The HAM-D24 score with the shortest distance
on the ROC curve was considered as appropriate cutoff. Statistical significance was
established at P<0.05.

3. Results

3.1. Clinical characteristics in diabetic patients with or without diabetic

microvascular complication

The baseline characteristics of diabetic patients with or without diabetic

microvascular complication were presented in Table 1. The incidences of use of
insulin, use of hypoglycemic drug in T2DM patients with diabetic microvascular
complication were significantly higher than those in T2DM patients without diabetic
microvascular complication. Duration of T2DM, MPV, creatinine, albumin, FG, LVEF,
HAM-D24 score in T2DM patients with diabetic microvascular complication were
significantly higher when compared with those in T2DM patients without diabetic
microvascular complication. The sex, habitant area, incidence of coronary artery
disease, hypertension, atrial fibrillation, dilated cardiomyopathy, age, BMI, waist
circumference, monthly family income, inhabited family member, smoking index,
RBC counts, hemoglobin, platelet counts, plateletcrit, PDW, WBC counts, NEUT%,
TC, triglyceride, HDL-C, LDL-C, apolipoprotein A, apolipoprotein B, BUN,
cystatin-C, TBIL, DBIL, BNP showed no significant difference between the two
groups.

3.2. Comparison of HAM-D24 score in T2DM patients with and without diabetic
microvascular complication

Comparison of HAM-D24 score in T2DM patients with and without

microvascular diabetic complication was presented in Figure 1. The HAM-D score in
patients with diabetic nephropathy was significantly higher than that in patients
without diabetic nephropathy (19.43±10.13 vs 11.14±8.30, P=0.001). The HAM-D
score in patients with diabetic retinopathy was significantly higher than that in
patients without diabetic retinopathy (27.67±1.32 vs 11.23±8.35, P﹤0.001). The
HAM-D score in patients with diabetic peripheral neuropathy was significantly higher
than that in patients without diabetic peripheral neuropathy (18.40±4.47 vs
11.38±8.73, P﹤0.001). The HAM-D score in patients with diabetic foot was
significantly higher than that in patients without diabetic foot (37.50±1.64 vs
11.20±7.93, P﹤0.001).

3.3. Comparison of incidence of depression in T2DM patients with and without

diabetic microvascular complication
 The incidence of depression determined using HAM-D24 score were compared in
T2DM patients with and without diabetic microvascular complication (Figure 2). The
incidence of depression in patients with diabetic nephropathy was significantly higher
than that in patients without diabetic nephropathy (87.5% vs 51.1%, P=0.002). The
incidence of depression in patients with diabetic retinopathy was significantly higher
than that in patients without diabetic retinopathy (100% vs 52.1%, P=0.004). The
incidence of depression in patients with diabetic peripheral neuropathy was
significantly higher than that in patients without diabetic peripheral neuropathy (100%
vs 51.1%, P﹤0.001). The incidence of depression in patients with diabetic foot was
significantly higher than that in patients without diabetic foot (100% vs 52.1%,
P=0.032).

3.4. Logistic regression analysis of HAM-D24 score after adjustment of

confounding factor of diabetic microvascular complication

The Logistic regression analysis was performed to determine whether the

HAM-D24 score was associated with diabetic microvascular complication after the
adjustment of confounding factor of diabetic microvascular complication in T2DM
patients (Table 2). After the adjustment of use of insulin, use of hypoglycemic drug,
duration of T2DM, MPV, creatinine, albumin, FG, HbA1C, LVEF, respectively,
HAM-D24 score was still significantly associated with diabetic microvascular
complication. The OR values ranged from 1.188 to 1.281 (all P﹤0.001).

3.5. ROC curve and AUC of HAM-D24 score for the prediction of diabetic
microvascular complication

The ROC curve of HAM-D24 score for the prediction of diabetic microvascular
complication in T2DM patients was presented in Figure 3. The AUC of HAM-D24
score for the prediction of diabetic microvascular complication was 0.832
(0.761-0.902). The cutoff points of HAM-D24 score to predict diabetic microvascular
complication were calculated in Table 3. The distances in ROC curve were calculated
using various HAM-D24 scores (1-27). The shortest distance on the ROC curve was
0.309 and 15 points of HAM-D24 score was considered as the optimal cutoff in the
prediction of diabetic microvascular complication in T2DM patients with the
sensitivity of 0.778 and specificity of 0.785.

4. Discussion

In this present study, the significantly different confounding factors in the T2DM
patients with or without diabetic microvascular complication were selected as the
adjustment factors in Logistic regression analysis. We found that after the adjustment
of use of insulin, use of hypoglycemic drug, duration of T2DM, MPV, creatinine,
albumin, FG, HbA1C, LVEF, respectively, HAM-D24 score was still significantly
associated with diabetic microvascular complication. The OR values ranged from
1.188 to 1.281 (all P﹤0.001). Our results were consistent with previous study
including 4,623 patients with T2DM, the study found that after adjustment for prior
complications and demographic, clinical, and diabetes self-care variables, depression
was associated with significantly higher risks of adverse microvascular outcomes
(OR=1.36 [95% CI 1.05–1.75]) (Lin et al., 2010). The OR value in this study was
slightly higher than that in our study since the microvascular complications were
defined as blindness, end-stage renal disease, amputations, and renal failure deaths,
which were even more severe and may elevate OR value of depression when
compared with our study. Several mechanisms may be involved in the association
between depression and diabetic microvascular complication. Depression may
activate the hypothalamic pituitary adrenal axis which could stimulate the sympathetic
nervous system and increase inflammatory and platelet aggregation responses, these
effects may lead to poor diabetes control (Golden, 2007). On the other hand,
depression may have negative effects on behaviors such as diet, exercise, blood
glucose check and hypoglycemic treatment (Almeida et al., 2016; Lustman et al.,
2007; Fisher et al., 2007), these effects would also impair glycemic control. Therefore,
poor glycemic control may finally increase the risk of diabetes microvascular
complications (Stratton et al., 2000).

In this study, we used the ROC (Receiver Operating Characteristic) curve of

HAM-D24 score for the prediction of diabetic microvascular complication in T2DM
patients. We found that the AUC (Area Under roc Curve) of HAM-D24 score for the
prediction of diabetic microvascular complication was 0.832 (0.761-0.902). The AUC
is at a very high level since the higher 95% CI. reached 0.902. Then, we calculated the
optimal cutoff of HAM-D24 score for prediction of diabetic microvascular
complication in T2DM patients, we found that 15 points of HAM-D24 score was
considered as the optimal cutoff with the sensitivity of 0.778 and specificity of 0.785.
Previous study using HAM-D24 scores for the prediction of diabetic microvascular
complication in T2DM patients was not documented by now. 15 points also belonged
to the mild depression (8-19) in HAM-D24 score, the patients with HAM-D24 score
higher than 15 points should be paid special attention to since they had significantly
higher chance of suffering from diabetic microvascular complication.

The HAM-D24 scores in our patients with diabetic nephropathy, diabetic

retinopathy, diabetic peripheral neuropathy and diabetic foot were all significantly
higher than those in patients without these diabetic microvascular complications.
Interestingly, the HAM-D24 score in diabetic foot were presented as the highest score
among all four microvascular complications with the mean±SD of 37.50±1.64 points,
which was higher than the definition of severe depression score of 35 points. Our
results were consistent with previous researches focusing on the depression level of
diabetic foot patients (Vedhara et al., 2016; Yildiz and Aşti, 2015; Chapman et al.,
2014). The reason may be that the diabetic foot syndrome usually occurs under the
combined effects of neuropathy and vascular insufficiency, the symptoms such as
severe foot pain and movement difficulties could be unbearable, the combined effects
of poorer self-care and poorer outcomes would also lead to severe depression in these
patients (Hjelm and Apelqvist, 2016). Second, psychological comorbidity such as
depression would confer additional risks in patients with T2DM (Lustman et al.,
2000). Depression in T2DM was reported confer twice the rate of a first diabetic foot
ulcer over a 4 year follow-up period (Williams et al., 2010) and higher rates of
amputation (Williams et al., 2011), which would in return cause more severe
depression in this population.

Meanwhile, the incidence of depression for diabetic nephropathy, diabetic

retinopathy, diabetic peripheral neuropathy and diabetic foot were 87.5%, 100%, 100%
and 100%, respectively. We noticed that the all patients with diabetic retinopathy,
diabetic peripheral neuropathy and diabetic foot were depressed. The results were
consistent with the previous literatures (Prinz et al., 2016; Mushtaque et al., 2016).
Since we enrolled the T2DM patients in cardiology department, many patients had
macrovascular complications such as coronary artery disease or even heart failure.
Relevant research has also shown a higher prevalence of depression in patients with
more severe coronary artery disease and heart failure (Carels, 2004; Faller et al.,
2009). That may be the reason why the vast majority of our T2DM patients with
microvascular complications were more depressed.

The incidence of depression determined using HAM-D24 score was 54.2% in this
present study. According to other literatures, the prevalence of depression was found
to range from 8% (Lloyd et al., 2000) to 44.66% (Sweileh et al., 2014; Mocan et al.,
2016). The incidence of depression in our study was much higher than that in other
studies, the possible reason may be as follows: Our patients were enrolled in
cardiology department, most patients had cardiovascular disease such as coronary
artery disease, hypertension, atrial fibrillation and dilated cardiomyopathy besides
T2DM, the combined effects of cardiovascular disease and T2DM may significantly
elevate the levels of depression. Meanwhile, 33.3%, 17.7% and 3.1% of the T2DM
patients had mild depression, moderate depression, and severe depression,
respectively. Especially, 3.1% of the depressed T2DM patients were experiencing
severe depressive symptoms compared with 1.4% in study conducted in Romania
(Mocan et al., 2016), whose patients were recruited from center for diabetes, nutrition
and metabolic Diseases. The different enrollment methods not only caused the
increased incidence of depression, but also caused the increased severity of the
depression in T2DM patients. Another possible explanation for these diverted findings
might be that when depression is screened, different measurement instruments with
different cutoff points are used (Asuzu et al., 2016; Vijayalakshmi et al., 2016). Lloyd
et al (Lloyd et al., 2012) reported that prevalence differences of depression in diabetes
patients might be associated with that people might experience differently according
to their present stressors, their culture and social dynamics.

Our study has several strengths. First, we have shown for the first time that the
HAM-D24 scores could be used to predict the diabetic microvascular complication in
T2DM patients. Second, we provide clinical and laboratory data in T2DM patients
with or without diabetic microvascular complications, which conveyed valuable clue
for other studies searching new risk factors for diabetic microvascular complications.

Third, the HAM-D24 scores is a simple, cheap and commonly used scale for
depression, to perform the test when the T2DM patients were admitted would not only
provide the information on depression status, it could also alert us about the diabetic
microvascular complications when the HAM-D24 scores reached the optimal cutoff of
15 points.

Our study has several limitations. This study was a cross-sectional and
observational study, the causal relationship between depression and diabetic
microvascular complications needs to be clarified. Though, the relationship between
depression and T2DM is proved to be bidirectional (Golden et al., 2008). Further
cohort studies may focus on which comes first, depression or diabetic microvascular
complications. Second, we chose the diabetic nephropathy, diabetic retinopathy,
diabetic peripheral neuropathy and diabetic foot as the definition of diabetic
microvascular complications, which may be insufficient to cover all the diabetic
microvascular complications. So the incidence of diabetic microvascular
complications may be underestimated. Third, we used nonmydriatic digital
stereoscopic retinal imaging instead of fluorescein fundus angiography for the
screening of diabetic retinopathy, it may underestimate the incidence of diabetic
retinopathy. Meanwhile, we performed urinary albumin excretion rate (UAER) ≥200
ug/min twice during hospitalization, however, the previous UAER elevation could not
be documented, so the incidence of diabetic retinopathy may also be underestimated.

5. Conclusions

In conclusion, the HAM-D24 scores and incidence of depression in T2DM

patients with diabetic microvascular complications were all significantly higher than
those in patients without diabetic microvascular complications. After the adjustment
of confounding factors such as use of insulin, use of hypoglycemic drug, duration of
DM, MPV, creatinine, albumin, FG, HbA1C, LVEF, respectively, HAM-D24 score was
still independently associated with diabetic microvascular complication. The AUC of
HAM-D24 score for the prediction of diabetic microvascular complication was 0.832
(0.761-0.902). 15 points of HAM-D24 score was considered as the optimal cutoff for
prediction of diabetic microvascular complication in T2DM patients with the
sensitivity of 0.778 and specificity of 0.785. We believe that the HAM-D24 score
should be applied when T2DM patients were admitted, the patients with HAM-D24
score higher than 15 points should be paid special attention to since they had
significantly higher chance of suffering from diabetic microvascular complication.

Funding sources

This paper was funded by National Natural Science Funds of China (Grant No.
81500308).
Conflicts of interest

None.

Authors’ contributions

Conceived and designed the experiments: SP, Z-WL, YL, J-KW. Performed the
experiments: SS, W-QS, XM, G-CG, YZ, S-MZ, F-QL, BL, Z-GT. Analyzed the data:
SP Z-WL. Contributed reagents/materials/analysis tools: SP, Z-WL. Wrote the
manuscript: SP, YL , J-KW.

Ethics approval and consent to participate

This study was approved by the Ethics Committee of the People's Hospital of Shaanxi
Province (Xi’an, China). Written informed consent was obtained from all participants.
We have obtained consent to publish from the participants to report individual patient
data. It was conducted according to the standards of the Declaration of Helsinki.

Consent to publish

The consents to publish the paper were obtained from all authors.

References
Ali, M.K., Bullard, K.M., Saaddine, J.B., Cowie, C.C., Imperatore, G., Gregg, E.W.,
2013. Achievement of goals in U.S. diabetes care, 1999-2010. N. Engl. J.
Med.368,1613-1624.

Almeida, O.P., McCaul, K., Hankey, G.J., Yeap, B.B., Golledge, J., Norman, P.E.,
Flicker, L., 2016. Duration of diabetes and its association with depression in later
life: The Health In Men Study (HIMS). Maturitas.86,3-9.

Anderson, R.J., Freedland, K.E., Clouse, R.E., Lustman, P.J., 2001. The prevalence of
comorbid depression in adults with diabetes: a meta-analysis. Diabetes
Care.24,1069-1078.

Asuzu, C.C., Walker, R.J., Williams, J.S., Egede, L.E., 2016. Pathways for the
relationship between diabetes distress, depression, fatalism and glycemic control
in adults with type 2 diabetes. J. Diabetes Complications,S1056-8727,30635-3.

Bech, P., Paykel, E., Sireling, L., Yiend, J., 2015. Rating scales in general practice
depression: psychometric analyses of the clinical interview for depression and
the Hamilton rating scale. J. Affect Disord.171,68-73.

Carels, R.A., 2004. The association between disease severity, functional status,
depression and daily quality of life in congestive heart failure patients. Qual. Life
Res.13,63-72.

Carneiro, A.M., Fernandes, F., Moreno, R.A., 2015. Hamilton depression rating scale
and montgomery-asberg depression rating scale in depressed and bipolar I
patients: psychometric properties in a Brazilian sample. Health Qual. Life
Outcomes.13,42.

Chapman, Z., Shuttleworth, C.M., Huber, J.W., 2014. High levels of anxiety and
depression in diabetic patients with Charcot foot. J. Foot Ankle Res.7,22.
De Groot, M., Andreson, M., Freedland, K.E., Clouse, R.E., Lustman, P.J., 2001.
Association of depression and diabetes complication: a meta-analysis.
Psychosom. Med.63,619-630.

Egede, L.E., Zheng, D., Simpson, K., 2002. Comorbid depression is associated with
increased health care use and ecpenditures in individuals with diabetes. Diabetes
Care.25,464-470.

Egede, L.E., Hernandez-Tejada, M.A., 2013. Effect of comorbid depression on

quality of life in adults with Type 2 diabetes. Expert Rev. Pharmacoecon.
Outcomes Res.13,83-91.

Faller, H., Störk, S., Schuler, M., Schowalter, M., Steinbüchel, T., Ertl, G.,
Angermann, C.E., 2009. Depression and disease severity as predictors of
health-related quality of life in patients with chronic heart failure-a structural
equation modeling approach. J. Card. Fail.15,286-292.e2

Fisher, E.B., Thorpe, C.T., Devellis, B.M., Devellis, R.F., 2007. Healthy coping,
negative emotions, and diabetes management: a systematic review and appraisal.
Diabetes. Educ.33,1080-103; discussion 1104-1106.

Golden, H.S., 2007. A review of the evidence for a neuroendocrine link between
stress, depression and diabetes mellitus. Curr. Diab. Rev.3,252–259

Golden, S.H., Lazo, M., Carnethon, M., Bertoni, A.G., Schreiner, P.J., Diez, Roux,
A.V., Lee, H.B., Lyketsos, C., 2008. Examining a bidirectional association
between depressive symptoms and diabetes. JAMA.299,2751–2759

Grimaldi, A., Heurtier, A., 1999. Diagnostic criteria for type 2 diabetes. Rev.
Prat.49,16-21.
Hjelm, K., Apelqvist, J., 2016. Influence of beliefs about health and illness on
self-care and care-seeking in foreign-born people with diabetic foot ulcers:
dissimilarities related to origin. J. Wound Care.25,602-616.

IDF Diabetes Atlas Group. (2015). Update of mortality attributable to diabetes for the
IDF Diabetes Atlas: Estimates for the year 2013. Diabetes Res. Clin.
Pract.109,461-465.

Jacobson, A.M., de Groot, M., Samson, J.A., 1997. The effects of psychiatric
disorders and symptoms on quality of life in patients with type I and type II
diabetes mellitus. Qual. Life. Res.6,11-20.

Katon, W.J., Rutter, C., Simon, G., Lin, E.H., Ludman, E., Ciechanowski, P., Kinder,
L., Young, B., Von, Korff, M., 2005. The association of comorbid depression
with mortality in patients with type 2 diabetes. Diabetes Care.28,2668-2672.

Koivukangas, J., Björnholm, L., Tervonen, O., Miettunen, J., Nordström, T.,
Kiviniemi, V., Mäki, P., Mukkala, S., Moilanen, I., Barnett, J.H., Jones, P.B.,
Nikkinen, J., Veijola, J., 2016. Body mass index and brain white matter structure
in young adults at risk for psychosis - The Oulu Brain and Mind Study.
Psychiatry Res.254:169-76.

Korkmaz, S., Korkmaz, H., Özer, Ö., Atmaca, M., 2016. Assessment of left ventricle
systolic and diastolic functions in schizophrenia patients. Psychiatry Res.
240,348-51.

Lesperance, F., Frasure-Smith, N., 2000. Depression in patients with cardiac disease:
a practical review. J. Psychosom. Res.48,379-391.

Lin, E.H., Katon, W., Von, Korff, M., Rutter, C., Simon, G.E., Oliver, M.,
Ciechanowski, P., Ludman, E.J., Bush, T., Young, B., 2004. Relationship of
depression and diabetes slef-care, medication adherence, and preventive care.
Diabetes Care.27,2154-2160.
Lin, E.H., Rutter, C.M., Katon, W., Heckbert, S.R., Ciechanowski, P., Oliver, M.M.,
Ludman, E.J., Young, B.A., Williams, L.H., McCulloch, D.K., Von, Korff, M.,
2010. Depression and advanced complications of diabetes: a prospective cohort
study. Diabetes Care.33,264-269.

Lloyd, C.E., Dyer, P.H., Barnett, A.H., 2000. Prevalence of symptoms of depression
and anxiety in diabetic clinic population. Diabet Med,17,198-202.

Lloyd, C.E., Roy, T., Nouwen, A., Chauhan, A.M., 2012. Epidemiology of depression
in diabetes: International and cross-cultural issues. J. Affect. Disord.142
Suppl,S22-S29.

Lustman, P.J., Anderson, R.J., Freedland, K.E., De Groot, M., Carney, R.M., Clouse,
R.E., 2000. Depression and poor glycemic control: a meta-analytic review of the
literature. Diabetes Care.23,934-942.

Lustman, P.J., Penckofer, S.M., Clouse, R.E., 2007. Recent advances in

understanding depression in adults with diabetes. Curr. Diab. Rep.7,114-122

Milligan, S., 2016. Combination therapy for the improvement of long-term

macrovascular and microvascular outcomes in type 2 diabetes: Rationale and
evidence for early initiation. J. Diabetes Complications.30,1177-1185.

Mocan, A.S., Iancu, S.S., Duma, L., Mureseanu, C., Baban, A.S., 2016. Depression in
romanian patients with type 2 diabetes: prevalence and risk factors. Clujul.
Med.89,371-377.

Mushtaque, A., Gulati, R., Hossain, M.M., Azmi, S.A., 2016. Prevalence of
depression in patients of type 2 diabetes mellitus: A cross sectional study in a
tertiary care centre. Diabetes Metab. Syndr.10,238-241.

O'Connor, C.M., Jiang, W., Kuchibhatla, M., Silva, S.G., Cuffe, M.S., Callwood, D.D.,
Zakhary, B., Stough, W.G., Arias, R.M., Rivelli, S.K., Krishnan, R.,
 SADHART-CHF Investigators, 2010. Safety and efficacy of sertraline for
depression in patients with heart failure: results of the SADHART-CHF
(Sertraline Against Depression and Heart Disease in Chronic Heart Failure) trial.
J. Am. Coll. Cardiol.56,692-629.

Pan, S., Liu, Z.W., Lv, Y., Song, W.Q., Ma, X., Guan, G.C., Zhang, Y., Zhu, S.M., Liu,
F.Q., Liu, B., Tang, Z.G., Wang, J.K., 2016. Association between neutrophilic
granulocyte percentage and depression in hospitalized patients with heart failure.
BMC. Psychiatry.16,446.

Pouwer, F., Geelhoed-Duijvestijn, P.H., Tack, C.J., Bazelmans, E., Beekman, A.J.,
Heine, R.J., Snoek, F.J., 2010. Prevalence of comorbid depression is high in
out-patients with Type 1 and Type 2 diabetes mellitus. Results from three
out-patient clinics in the Netherlands. Diabet. Med.27,217-224.

Prinz, N., Ebner, S., Grünerbel, A., Henkelüdecke, U., Hermanns, N., Hummel, M.,
Schäfer, C., Wagner, C., Weiland, J., Welp, R., Holl, R.W., 2016. Female sex,
young age, northern German residence, hypoglycemia and disabling diabetes
complications are associated with depressed mood in the WHO-5 questionnaire -
A multicenter DPV study among 17,563 adult patients with type 2 diabetes. J.
Affect. Disord.208,384-391.

Schram, M.T., Baan, C.A., Pouwer, F., 2009. Depression and quality of life in
patients with diabetes: a systematic review from the European depression in
diabetes (EDID) research consortium. Curr, Diabetes Rev.5,112-119.

Stratton, I.M., Adler, A.I., Neil, H.A., Matthews, D.R., Manley, S.E., Cull, C.A.,
Hadden, D., Turner, R.C., Holman, R.R., 2000. Association of glycaemia with
macrovascular and microvascular complications of type 2 diabetes (UKPDS 35):
prospective observational study. BMJ.321,405-412.
Subramaniam, M., Sum, C.F., Pek, E., Stahl, D., Verma, S., Liow, P.H., Chua, H.C.,
Abdin, E., Chong, S.A., 2009. Comorbid depression and increased health care
utilisation in individuals with diabetes. Gen. Hosp. Psychiatry.31,220-224.

Sweileh, W.M., Abu-Hadeed, H.M., Al-Jabi, S.W., Zyoud, S.H., 2014. Prevalence of
depression amoung people with type 2 diabetes mellitus: a cross sectional study
in Palestine. BMC. Public Health.14,163.

van Dooren, F.E., Nefs, G., Schram, M.T., Verhey, F.R., Denollet, J., Pouwer, F.,
2013. Depression and risk of mortality in people with diabetes mellitus: a
systematic review and meta-analysis. PLoS. One.8,e57058.

Vedhara, K., Dawe, K., Miles, J.N., Wetherell, M.A., Cullum, N., Dayan, C., 2016.
Illness Beliefs Predict Mortality in Patients with Diabetic Foot Ulcers. PLoS.
One.11,e0153315.

Vijayalakshmi, U.B., Bodi, A.V., Sudagani, J., 2016. Biochemical and Clinical Profile
in Type 2 Diabetics with Depression. J. Clin. Diagn. Res.10,BC19-23.

Williams, L.H., Rutter, C.M., Katon, W.J., Reiber, G.E., Ciechanowski, P., Heckbert,
S.R., Lin, E.H., Ludman, E.J., Oliver, M.M., Young, B.A., Von, Korff, M., 2010.
Depression and incident diabetic foot ulcers: a prospective cohort study. Am. J.
Med.123,748-754.

Williams, L.H., Miller, D.R., Fincke, G., Lafrance, J.P., Etzioni, R., Maynard, C.,
Raugi, G.J., Reiber, G.E., 2011. Depression and incident lower limb amputations
in veterans with diabetes. J. Diabetes Complications.25,175-182.

Yildiz, E., Aşti, T., 2015. Determine the relationship between perceived social support
and depression level of patients with diabetic foot. J. Diabetes Metab.
Disord.16,14:59
Zhu, G., Yin, Y., Xiao, C.L., Mao, R.J., Shi, B.H., Jie, Y., Wang, Z.W., 2015. Serum
DHEAS levels are associated with the development of depression. Psychiatry
Res.229,447-53.

Figure 1. Comparison of HAM-D24 score in T2DM patients with and without each
microvascular diabetic complication

Figure 2. Comparison of depression incidence determined using HAM-D24 score in

T2DM patients with and without each diabetic microvascular complication
Figure 3. Receiver Operating Characteristic (ROC) curve and Area Under roc Curve
(AUC) of HAM-D24 score for the prediction of diabetic microvascular complications in
T2DM patients
 Table 1. Baseline characteristics of T2DM patients with and without diabetic
microvascular complications

Without diabetic With diabetic

microvascular microvascular
P value
complication (n=237) complication (n=51)

Sex 0.389

Men 141 (59.5%) 27 (52.9%)

Women 96 (40.5%) 24 (47.1%)

Habitant area 0.190

Urban area 162(68.4%) 30 (58.8%)

Rural area 75 (31.6%) 21 (41.2%)

Coronary artery disease 180 (75.9%) 45 (88.2%) 0.054

Hypertension 138 (58.2%) 36 (70.6%) 0.102

Atrial fibrillation 66 (27.8%) 18 (35.3%) 0.289

Dilated cardiomyopathy 27 (11.7%) 3 (5.9%) 0.316

Age (years) 62.92±10.29 64.22±11.46 0.349

BMI (kg/m2) 23.82±2.88 23.88±3.41 0.681

Waist circumference (cm) 85.09±8.59 85.23±9.05 0.948

Monthly family income

3641.77±3125.70 3583.33±1650.47 0.324
(yuan)

Inhabited family member

4.06±1.32 4.00±1.47 0.576
(person)

Smoking index 94.43±217.90 88.89±177.69 0.802

﹤
Use of insulin 54(22.8%) 45(83.3%)
0.001*

﹤
Use hypoglycemic drug 237(100%) 51(94.4%)
0.001*

Duration of T2DM 7.63±5.26 9.78±5.91 0.013*

RBC counts (×1012/L) 4.33±0.49 4.21±0.81 0.949

Hemoglobin (g/L) 134.99±15.66 131.33±26.23 0.741

Platelet counts (×109/L) 183.48±52.06 183.07±51.71 0.783

Plateletcrit (%) 0.19±0.05 0.17±0.04 0.166

MPV (fL) 10.42±1.53 9.69±1.31 0.007*

PDW (fL) 16.27±2.37 16.95±0.95 0.132

WBC counts (×109/L) 6.40±2.16 6.45±1.63 0.373

NEUT% (%) 63.79±11.73 63.47±8.20 0.970

TC (mmol/L) 4.26±1.13 4.38±0.93 0.249

Triglyceride (mmol/L) 1.52±0.89 1.45±0.54 0.377

HDL-C (mmol/L) 1.16±0.34 1.18±0.35 0.749

LDL-C (mmol/L) 2.10±0.75 2.24±0.60 0.080

Apolipoprotein A (g/L) 1.11±0.26 1.13±0.30 0.964

Apolipoprotein B (g/L) 0.78±0.22 0.80±0.18 0.347

BUN (mmol/L) 6.86±3.97 7.06±2.97 0.228

Creatinine (umol/L) 86.19±45.38 73.71±42.43 0.001*

Cystatin-C (mg/L) 1.22±0.50 1.20±0.49 0.561

TBIL (umol/L) 18.93±11.02 19.73±11.80 0.855

DBIL (umol/L) 6.53±4.15 6.49±3.85 0.787

﹤
Albumin (g/L) 38.48±4.22 34.77±4.34
0.001*

BNP (pg/mL) 502.06±998.22 949.09±1016.44 0.121

FG (mmol/L) 6.63±1.54 7.66±2.78 0.046*

﹤
HbA1C (%) 6.34±1.08 7.34±1.60
0.001*

LVEF (%) 55.97±11.91 43.33±15.30 0.001*

 ﹤
HAM-D24 score 9.47±6.53 21.72±9.99
0.001*

Note: T2DM, Type 2 Diabetes Mellitus; BMI, Body Mass Index; RBC, Red Blood Cells;
MPV, Mean Platelet Volume; PDW, Platelet Distribution Width; WBC, White Blood Cells;
NEUT%, neutrophilic granulocyte percentage; TC, Total Cholesterol; HDL-C, High-Density
Lipoprotein-Cholesterol; LDL-C, Low-Density Lipoprotein-Cholesterol; BUN, Urea Nitrogen;
TBIL, Total Bilirubin; DBIL, Direct Bilirubin; BNP, Brain Natriuretic Peptide; FG, Fasting
Glucose; HbA1C, Glycosylated Hemoglobin Type A1C; LVEF, left ventricular ejection
fraction; HAM-D24 score, Hamilton Rating Scale for Depression (24-items); *P﹤0.05.

Table 2. OR and 95% CI. of HAM-D24 score after the adjustment of each confounding
factors using Logistic regression analysis

OR 95% CI. P value

Adjusted for Coronary artery disease 1.191 1.135-1.248 ﹤0.001

Adjusted for Use of insulin 1.139 1.082-1.199 ﹤0.001

Adjusted for Use hypoglycemic drug 1.188 1.134-1.246 ﹤0.001

Adjusted for Duration of T2DM 1.207 1.143-1.275 ﹤0.001

Adjusted for MPV 1.203 1.139-1.270 ﹤0.001

Adjusted for Creatinine 1.206 1.146-1.269 ﹤0.001

Adjusted for Albumin 1.281 1.190-1.378 ﹤0.001

Adjusted for FG 1.190 1.135-1.248 ﹤0.001

Adjusted for HbA1C 1.214 1.150-1.281 ﹤0.001

Adjusted for LVEF 1.188 1.099-1.283 ﹤0.001

Note: OR, Odds Ratio; CI., Confidence Interval; HAM-D24 score, Hamilton Rating Scale for
Depression (24-items); T2DM, Type 2 Diabetes Mellitus; MPV, Mean Platelet Volume; FG,
Fasting Glucose; HbA1C, Glycosylated Hemoglobin Type A1C; LVEF, left ventricular
ejection fraction.

Table 3. Sensitivity, specificity, and distance in the ROC curve of each HAM-D24 score
for the prediction of diabetic microvascular complications

HADM-D24 score cutoffs (points) Sensitivity Specificity Distance in ROC curve

1 1.000 0.000 1.000

3 0.944 0.133 0.869

5 0.917 0.323 0.682

7 0.889 0.513 0.499

9 0.889 0.430 0.581

11 0.833 0.640 0.397

13 0.806 0.728 0.334

14 0.778 0.754 0.331

15 0.778 0.785 0.309

16 0.723 0.817 0.332

 17 0.667 0.835 0.372

19 0.667 0.880 0.354

21 0.612 0.912 0.398

23 0.389 0.962 0.612

25 0.334 0.999 0.666

27 0.278 1.000 0.722

Note: ROC, Receiver Operating Characteristic; HAM-D24 score, Hamilton Rating Scale
for Depression (24-items).

Highlights

1. HAM-D24 scores could be used to predict the diabetic microvascular

complication in T2DM patients (AUC of 0.832 (0.761-0.902).

2. 15 points of HAM-D24 score was the optimal cutoff for the prediction

of diabetic microvascular complications in T2DM patients (sensitivity of

0.778 and specificity of 0.785).

3. HAM-D24 score was significantly associated with diabetic

microvascular complications after adjustments (OR=1.188-1.281, all P﹤

0.001)

4. HAM-D24 score and incidence of depression in T2DM patients with or

without diabetic microvascular complications were compared.