You are here

  • Home
  • Archive
  • Volume 10, Issue 2
  • Analysis of factors related to diabetic retinopathy in patients with newly diagnosed type 2 diabetes: a cross-sectional study

Article Text

Article menu

Download PDFPDF

XML
Diabetes and endocrinology
Original research
Analysis of factors related to diabetic retinopathy in patients with newly diagnosed type 2 diabetes: a cross-sectional study
  1. Zhaohu Hao1,2,
  2. Xiao Huang1,
  3. Yongzhang Qin1,3,
  4. Huanming Li2,
  5. Fengshi Tian2,
  6. Rong Xu2,
  7. Baocheng Chang1,
  8. Hailin Shao2
  1. 1 NHC Key Laboratory of Hormones and Development (Tianjin Medical University), Tianjin Key Laboratory of Metabolic Diseases, Tianjin Medical University, Chu Hsien-I Memorial Hospital, Tianjin Institute of Endocrinology, Tianjin, China
  2. 2 Metabolic Disease Management Center, Tianjin 4th Central Hospital, The 4th Central Hospital Affiliated to Nankai University, The 4th Center Clinical College of Tianjin Medical University, Tianjin, China
  3. 3 Department of Endocrinology, The First Affiliated Hospital of Gannan Medical University, Ganzhou, China
  1. Correspondence to Dr Baocheng Chang; Changbaocheng2019{at}163.com; Dr Hailin Shao; Shaohailin1988{at}sohu.com

Abstract

Aim To investigate the related factors of diabetic retinopathy (DR) and explore the correlation between smoking and DR in patients with newly diagnosed type 2 diabetes mellitus (T2DM).

Design A single-centre cross-sectional study.

Setting Tianjin 4th Central Hospital.

Participants Patients with newly diagnosed T2DM who visited the outpatient department of the hospital from December 2018 to April 2019.

Methods A total of 947 patients were enrolled in the study. They were divided into two groups according to whether they were diagnosed with DR (diabetic retinopathy group, DR group; non-diabetic retinopathy group, NDR group). The smoking index (SI) was calculated to assess smoking status. Factors such as sex, age, hypertension, T2DM diagnosed age, family history of diabetes, drinking history, haemoglobin A1c (HbA1c), body mass index (BMI) and smoking status were compared between the two groups. Logistic regression was used to analyse the relationship between DR and the above factors.

Results There was no statistically significant difference between the two groups in sex, age, hypertension, DM diagnosed age, family history of diabetes, drinking history and HbA1c. BMI was significantly higher in DR patients (27.7±4.2 vs 26.7±4.4, p=0.004). Smoking status was also different between the two groups (χ2=6.350, p=0.042). BMI was shown to be a related factor for DR in patients with newly diagnosed diabetes (OR=0.592, p=0.004). When BMI was ≥28 kg/m2, heavy smoking was significantly associated with DR (OR=2.219, p=0.049), and there was a negative correlation between DR and the age of diagnosis of diabetes ≥60 years (OR=0.289, p=0.009).

Conclusions Heavy smoking was an important related factor for DR in patients with newly diagnosed diabetes mellitus when BMI was ≥28 kg/m2. Delaying the age of diabetes might prevent the occurrence of DR. To elucidate the correlation, long-term cohort studies with large samples are needed.

  • smoking index
  • diabetic retinopathy
  • type 2 diabetes mellitus
  • newly diagnosed diabetes mellitus
http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjopen-2019-032095

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This study is a representative population-based study.

    • There are few studies on the factors related to newly diagnosed diabetic retinopathy in patients with type 2 diabetes mellitus in China.

    • This study was the first to use the smoking index to assess smoking severity to study diabetic retinopathy.

    • The study is a single-centre cross-sectional study.

    Introduction

    Diabetes mellitus (DM) is a chronic metabolic disease characterised by hyperglycaemia. It has various aetiologies, and the incidence is increasing yearly worldwide. According to the report of the International Diabetes Federation (IDF) in 2017, about 425 million people were affected by DM, and that number is expected to reach 629 million by 2045.1 China has the highest number of patients with DM, and the prevalence has increased from 0.67% in 1980 to 10.4% in 2013.2 A epidemiological-based meta-analysis conducted in 35 research centres around the world showed that the prevalence of diabetic retinopathy (DR) in the total diabetic population was 35.4%.3 DR is a highly specific microvascular complication caused by diabetes, and its prevalence is significantly correlated with the course of DM and blood glucose levels.4 DR is the most important cause of irreversible blindness in working-age adults.4 Chronic hyperglycaemia causes not only retinal vascular diseases but also damage to retinal neurons, both of which are factors that lead to vision loss.5 Twenty per cent of patients with newly diagnosed type 2 diabetes mellitus (T2DM) are diagnosed with DR.3 The explanation for this phenomenon is that the onset of T2DM is often insidious, and patients may have a long history of untreated hyperglycaemia before diagnosis.3 Are there any other risk factors for DR in these newly diagnosed patients? There is no relevant research report at present.

    Smoking is harmful to health, and it is an important risk factor for cardiovascular, cerebrovascular and respiratory diseases and also for malignant tumours.6 Smoking is one of the most serious public health issues throughout the world, and it is responsible for about 7 million deaths per year worldwide.6 In a 14-year prospective cohort study, Jee et al found that people smoking >20 cigarettes a day had a 1.55-fold higher incidence of diabetes than non-smokers.7 It is well known that smoking may cause an increase in proteinuria.8 However, many studies have shown that smoking has no significant correlation with DR.9 Even some studies in China have shown that smoking is a protective factor for DR.10 Many patients with DM had started smoking before they were diagnosed with diabetes. Does smoking contribute to the complications of DR in newly diagnosed patients? Little research has been done on the relationship between smoking and DR in patients with newly diagnosed diabetes. The present study investigated the factors related to DR and explored the correlation between smoking and DR in patients with newly diagnosed T2DM who visited the outpatient department of the hospital from December 2018 to April 2019.

    Research design and methods

    The present study was a single-centre cross-sectional study of factors associated with DR in patients with newly diagnosed T2DM in Tianjin 4th Central Hospital. The hospital is a regional medical centre in Tianjin. The hospital had a diabetes identification centre. The patients in the region are identified by the department before they can enjoy more preferential health insurance policies in the clinic. The Metabolic Disease Management Center (MMC) was responsible for the screening for complications in patients with newly diagnosed diabetes, such as urinary protein, fundus examination, peripheral vascular and neuropathy. The diabetes nursing team undertook the measurement of height, weight and blood pressure, and gathered information about smoking, drinking and disease history. Electronic medical records were generated at the same time.

    Participants

    Study participants were patients with newly diagnosed T2DM who visited MMC from December 2018 to April 2019. The patient information came from an electronic database of the hospital’s diabetes identification centre. All the patients who underwent fundus examination were enrolled in this study. All patient identifiers were removed prior to analysis. All the steps of this clinical study protocol were conducted in accordance with the principles of the World Medical Association Declaration of Helsinki.11 Written informed consent was obtained from each patient.

    Inclusion criteria

    The study enrolled participants with a baseline of being newly diagnosed with T2DM within the past 6 months; patients who could communicate independently and describe their smoking status; none of the smokers had given up smoking before the study; and all patients underwent DR screening in the study.

    According to the Chinese guidelines for the prevention and treatment of diabetes, all patients with diabetes in our hospital are required to be diagnosed with an oral glucose tolerance test. The diagnostic criteria during the execution were as follows: (1) fasting plasma glucose ≥7.0 mmol/L. Fasting was defined as no caloric intake for at least 8 hours; or (2) 2-hour plasma glucose ≥11.1 mmol/L during an oral glucose tolerance test. The test was performed as described by the WHO, using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water. A haemoglobin A1c (HbA1c) test was not standardised in China, so it could not be used as a diagnostic standard.12

    Exclusion criteria

    Type 1 diabetes and other specific types of diabetes; acute complications of diabetes, such as diabetic ketoacidosis or lactic acidosis; hypovolaemia; subjects with diabetic nephropathy complications, rheumatic diseases, autoimmune diseases, pregnancy, malignant tumours, infection, foot ulcer, severe mental illness, thyroid dysfunction, severe hepatic insufficiency or renal insufficiency, severe anaemia, heart failure or respiratory insufficiency; cognitive impairment and communication disorders.

    Evaluation of clinical variables

    Information about the patients’ names, sex, phone numbers, body mass index (BMI), age, smoking history, drinking history, family history of DM, hypertension and HbA1c was collected using a uniform information table. The formula for BMI was weight in kilograms divided by the square of height in metres. BMI (≥28 kg/m2) met obesity criteria according to Chinese standards.13 Venous blood samples were collected in EDTA tubes from fasting patients in the morning. The level of HbA1c was determined by affinity chromatography in the hospital standard laboratory (Tosoh Corporation, Japan). Smoking status assessment: according to the WHO (1997) smoking survey method, smoking at least one cigarette per day for >6 months was considered as a smoking history.14 The smoking index (SI) was calculated by multiplying the number of cigarettes smoked per day by the number of years the person had smoked. In our study, according to the SI, 0 meant non-smoking, ≤500 meant light and moderate smoking, and >500 meant heavy smoking.

    Screening for DR was carried out according to EURODIAB guidelines using one 45° field retinograph centred on the fovea. If DR was suspected, the other two 45° retinographs were obtained.15 Ophthalmologists judged whether the patient had DR according to the results of the funduscopic examination and guidelines. In the study, the results were reported as either positive or negative (DR or non-diabetic retinopathy (NDR)), and the patients were divided into two groups (DR group and NDR group). NDR here included a normal fundus and other retinopathy that may be caused by other causes (non-diabetic) such as fundus arteriosclerosis.

    Statistical analysis

    Descriptive analysis was used to illustrate the basic demographic characteristics. For continuous variables, a one-sample Kolmogorov-Smirnov normality test was used to check the normality of the distribution of such variables. Once normal distribution and homogeneity of variance were satisfied, values were represented by the‾x±SD and Student’s t-test was used. For categorical variables, a χ2 test was performed. The occurrence of DR was identified as the dependent variable, and the related factors (sex, hypertension, age at DM diagnosis, family history of diabetes, drinking history, HbA1c, BMI and smoking status) were listed as independent variables. To analyse the factors associated with DR and adjust for potential confounding effects, we examined the factors associated with DR using multivariable logistic regression analysis. All the analyses conducted using SPSS V.16.0 software with p<0.05 (two-sided) indicated statistical significance.

    Patient and public involvement

    Patients were not involved in setting the research questions or planning the study. Investigators did not know the identities of the study participants. In this study, the electronic data were obtained from the health records at the institution. All patient identifiers were removed before the analysis was conducted. There was no direct patient or public involvement.

    Results

    Characteristics of the study population

    In this survey, 952 patients entered the study initially and 5 were excluded because their fundus could not be seen clearly. Finally, a total of 947 patients were enrolled in the study. Their average age was 53.2±11.6 years. The average HbA1c level was 9.53%±2.13%. The average BMI of the patients was 27.5±4.3 kg/m2. A total of 430 patients had a history of smoking. The average smoking age was 27.8±12.2 years (2–60 years). They smoked 21.8±12.3 cigarettes per day. Baseline characteristics are shown in table 1.

    View this table:
    Table 1

    Baseline characteristics of 947 participants with newly diagnosed T2DM

    Comparison of clinical data between the DR and NDR groups

    A total of 755 patients (79.7%) were in the NDR group, and 192 patients were in the DR group. Among the patients in the DR group, there were 12 patients with PDR. The comparison of clinical data between the DR and NDR groups is shown in table 2. There were no statistically significant differences between the two groups in sex, age, hypertension, family history of diabetes, drinking history and HbA1c. The BMI was significantly higher in NDR patients (27.7±4.2 vs 26.7±4.4, p=0.004). Smoking status was different between the two groups (χ2=6.350, p=0.042).

    View this table:
    Table 2

    Comparison of clinical data between DR and NDR groups

    The occurrence of DR was identified as the dependent variable, and the related risk factors were listed as independent variables. Both of the variables were progressed through multivariate logistic regression analysis; age, BMI and smoking status were adjusted, as shown in table 3. Patients with BMI ≥28 kg/m2 were less likely to have DR in patients with newly diagnosed diabetes (OR=0.592, p=0.004). The age of DM diagnosis was also statistically significant in the regression analysis (p=0.047). The incidence of DR in patients over 60 years of age who were diagnosed with diabetes was significantly lower than that in patients under 50 years of age (OR=0.596, p=0.024). The incidence of DR in severe smokers was significantly higher than that in non-smokers (OR=1.664, p=0.043).

    View this table:
    Table 3

    Analysis of related factors with DR, adjusted for age, BMI and smoking status

    Did obesity have a protective effect on DR?

    Our study compared the age, HbA1c, smoking and drinking history, family history of DM and complications between the obese and non-obese patients, as shown in table 4 (BMI ≥28 vs <28 kg/m2). There was no statistically significant difference between the two groups in sex, family history of DM, drinking history, smoking history and HbA1c. The age of obese patients with diabetes was significantly lower than that of non-obese patients (50.4±12.4 vs 55.2±10.5, t=−6.184, p<0.001). Because the onset of T2DM was insidious, patients might have a long history before diagnosis.3 BMI could be a possible confounding variable.

    View this table:
    Table 4

    Comparison of clinical data between patients (BMI ≥28 vs <28 k g/m2)

    After adjustment for BMI and age, DR-related factors were further analysed. First, the participants were divided into two groups according to whether the BMI was ≥28 kg/m2. The incidence of DR in patients with different smoking statuses was compared (table 5). The incidence of DR in heavy smoking patients was significantly higher in obese patients. There was no significant difference in the incidence of DR with different smoking statuses among patients over 60 years (table 5). Then after adjustment for age, multiple logistic regression analysis was used to analyse the related factors of DR (table 6). When BMI was ≥28 kg/m2, heavy smoking was significantly associated with DR (OR=2.219, p=0.049), and there was a negative correlation between DR and the age of diagnosis of diabetes at ≥60 years (OR=0.289, p=0.009).

    View this table:
    Table 5

    The incidence of DR in patients with different smoking statuses

    View this table:
    Table 6

    Logistic regression analysis of related factors for DR (BMI <28 kg/m2; ≥28 kg/m2)

    Discussion

    Summary of results

    In our study, among the patients newly diagnosed with T2DM, 20.3% presented with DR. After analysing the related factors of DR, the present study found that obese patients may have a lower incidence of DR (OR=0.592, p=0.004). Many researchers have shown that duration of DM, blood glucose, blood pressure and blood lipids are important risk factors for DR.16–18 The duration is the most important factor.19 The subjects of the present study were all patients with newly diagnosed T2DM. Our study showed that DR was related to BMI, smoking status and age of DM diagnosis. To the best of our knowledge, the present study is the first that was designed to investigate the correlation between the smoking level and DR in patients with newly diagnosed T2DM. The results of the study suggested that heavy smoking was associated with DR in obese (BMI ≥28 kg/m2) patients with newly diagnosed T2DM (OR=2.219, p=0.049), and there was a negative correlation between DR and the age of diagnosis of diabetes ≥60 years (OR=0.289, p=0.009). Those correlations did not exist in non-obese (BMI <28 kg/m2) patients.

    The relationship between BMI and DR

    In our study, DR was negatively correlated to BMI. This was not consistent with other research results. Grey et al followed up with 14 657 patients with diabetes for an average of 6.68 years and found that increased risk of DR was associated with higher BMI.20 It has also been reported that BMI was not correlated or even negatively correlated with DR.21 The subjects in our study had newly diagnosed T2DM. The average age of patients with newly diagnosed T2DM with obesity (BMI ≥28 kg/m2) was 50.4±12.4 years, whereas that of patients without obesity was 55.2±10.5 years. The former was significantly lower than the latter (t=−6.184, p<0.001). As the onset of T2DM is insidious, patients may have had a long history before diagnosis.3 Based on the usual clinical experience and their different diagnostic ages, our study suggested that the diagnostic age of patients with T2DM with obesity might be closer to the real age of onset. Perhaps because non-obese patients might be confused with more complex diabetic course factors, we did not find any factors in the analysis of the related factors of DR in these patients. In conclusion, we do not think that a high BMI is a protective related factor for DR. It is very likely that there are confounding factors, such as an unobserved course of T2DM.

    The relationship between smoking and DR

    Cigarette smoking is a well-known risk factor for many chronic diseases, including cardiovascular diseases and various malignancies.6 Many researchers have reported the negative effects of smoking on DM.22 23 A 14-year prospective cohort study in South Korea showed that the risk of DM among men and women who smoked 20 or more cigarettes a day was 1.55 times higher than that in non-smokers.7 Cigarette smoking is independently associated with the incidence of T2DM.24 Although the mechanisms involved are not clear yet. Smoking is associated with insulin resistance, inflammation and dyslipidaemia.25 26 When BMI was ≥28 kg/m2, we found that the incidence of DR was 12.4% in non-smokers, 12.1% in light and moderate smokers and 28.2% in heavy smokers (χ2=11.163, p=0.004). Some studies have reported that no association was found between smoking and DR in patients with T2DM.27 28 Smoking cessation is one of the important targets for diabetes control and the prevention of chronic diabetes complications.29 30

    Studies have shown that cigarette smoking impairs nitric oxide-mediated endothelial function via increased generation of superoxide anions, which may increase the risk of DR in patients with diabetes and could cause eye tissue ischaemia, eye tissue hypoxia, retinal arteriosclerosis and a decrease in choroidal blood flow, eventually leading to retinal ischaemia.30 31 When BMI was ≥28 kg/m2, the present study showed that the incidence of DR was 28.2% in heavy smokers. Heavy smoking was significantly associated with DR. Smoking may aggravate retinopathy damage by increasing arteriosclerosis.32 DR remains the leading cause of acquired blindness in working-age adults.4 Although cutting-edge research in the field has identified many molecular, functional and structural abnormalities, the exact molecular mechanism of this devastating disease remains obscure.33 A diabetic environment drives the dysfunction of the power generator of the cell and disturbs the homeostasis of the mitochondrial dynamic. Mitochondria seem to have a significant role in the development of DR, and unravelling the mechanism responsible for their damage as well as the role of epigenetic modifications in mitochondrial homeostasis should identify novel therapeutic targets.33 Further research is needed to determine whether smoking affects mitochondrial DNA function.

    In the present study, DR was found in 20.3% of patients with newly diagnosed T2DM. This was consistent with other findings.3 The study suggested that smoking may be an important related factor for the occurrence of DR in patients with newly diagnosed diabetes. Tyrberg et al investigated 794 patients diagnosed with diabetes in 1987–1988 and found that smoking history could increase the risk of DR during 9–17 years after diagnosis.34 Because smoking behaviour occurred before the diagnosis of T2DM in this study, heavy smoking is very likely to be a risk factor for DR.

    When BMI was ≥28 kg/m2, there was a negative correlation between DR and the age of diagnosis of diabetes ≥60 years (OR=0.289, p=0.009). Wong et al investigated >1400 patients with T2DM and found that the risk of DR in young patients was significantly higher than that in old patients.35 After adjusting for factors such as the course of diabetes, the risk of DR was increased two times in those with T2DM who were <45 years.35 Our report agrees with other studies. Age at onset of diabetes was strongly associated with an increasing number of complications.36 In other regions, studies that have looked at the age of onset of diabetes have found that early onset diabetes increases disease severity. Genetic determinants of diabetes are likely to be more common in patients with early onset of diabetes.37 This observation suggests that patients with early onset of diabetes have a more aggressive disease and are prone to develop more complications at an earlier age.38 Delaying the age of diabetes might prevent the occurrence of DR.

    Limitations of the study

    The study was a single-centre cross-sectional study. The results only showed a correlation between smoking and DR. Prospective cohort studies are needed to confirm whether there is a causal relationship between smoking and DR in patients with newly diagnosed T2DM. Because of the large number of patients with newly diagnosed T2DM, we have set relatively wide exclusion criteria for the population, which may affect external validity. Although our department is responsible for the first diagnosis and identification of diabetes in Tianjin, it is still unable to achieve a timely diagnosis for all those patients who newly develop diabetes. Because the patients are all from Tianjin, China, the conclusion has regional limitations.

    Conclusions

    The study found that heavy smoking and the age of T2DM diagnosis were related factors for the occurrence of DR in newly diagnosed patients when BMI was ≥28 kg/m2. Although the subjects were all newly diagnosed patients, it is known that the onset of T2DM is relatively insidious. It is of great importance for people in populations at high risk of diabetes mellitus to quit smoking. Delaying the age of diabetes may prevent the occurrence of DR. Therefore, long-term cohort studies are needed to elucidate the correlation.

    References

    1. ↵
      1. American Diabetes Association
      . 2. classification and diagnosis of diabetes. Diabetes Care 2017;40:S11–24.doi:10.2337/dc17-S005
      OpenUrlFREE Full Text
    2. ↵
      2. Wang L ,
      3. Gao P ,
      4. Zhang M , et al
      . Prevalence and ethnic pattern of diabetes and prediabetes in China in 2013. JAMA 2017;317:2515–23.doi:10.1001/jama.2017.7596
      OpenUrlCrossRefPubMed
    3. ↵
      2. Yau JWY ,
      3. Rogers SL ,
      4. Kawasaki R , et al
      . Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35:556–64.doi:10.2337/dc11-1909
      OpenUrlAbstract/FREE Full Text
    4. ↵
      2. Pang C ,
      3. Jia L ,
      4. Jiang S , et al
      . Determination of diabetic retinopathy prevalence and associated risk factors in Chinese diabetic and pre-diabetic subjects: Shanghai diabetic complications study. Diabetes Metab Res Rev 2012;28:276–83.doi:10.1002/dmrr.1307
      OpenUrlCrossRefPubMed
    5. ↵
      1. Expert Committee on the Diagnosis and Clasification of Diabetes Mellitus
      . American diabetes association: clinical practice recommendations 2002. Diabetes Care 2002;25 Suppl 1:S1–147.doi:10.2337/diacare.25.2007.s1
      OpenUrl
    6. ↵
      1. GBD 2015 Risk Factors Collaborators
      . Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990-2015: a systematic analysis for the global burden of disease study 2015. Lancet 2016;388:1659–724.doi:10.1016/S0140-6736(16)31679-8
      OpenUrlCrossRefPubMed
    7. ↵
      2. Jee SH ,
      3. Foong AW ,
      4. Hur NW , et al
      . Smoking and risk for diabetes incidence and mortality in Korean men and women. Diabetes Care 2010;33:2567–72.doi:10.2337/dc10-0261
      OpenUrlAbstract/FREE Full Text
    8. ↵
      2. Ohkuma T ,
      3. Nakamura U ,
      4. Iwase M , et al
      . Effects of smoking and its cessation on creatinine- and cystatin C-based estimated glomerular filtration rates and albuminuria in male patients with type 2 diabetes mellitus: the Fukuoka diabetes registry. Hypertens Res 2016;39:744–51.doi:10.1038/hr.2016.51
      OpenUrl
    9. ↵
      2. Moss SE ,
      3. Klein R ,
      4. Klein BEK
      . Cigarette smoking and ten-year progression of diabetic retinopathy. Ophthalmology 1996;103:1438–42.doi:10.1016/S0161-6420(96)30486-7
      OpenUrlPubMedWeb of Science
    10. ↵
      2. Li Y ,
      3. Wang J ,
      4. Qu B , et al
      . Prevalence and risk factors of diabetic retinopathy in hospital patients (in Chinese). Natl Med J China 2018;98:440–4.
      OpenUrl
    11. ↵
      1. General Assembly of the World Medical Association
      . World Medical association Declaration of Helsinki: ethical principles for medical research involving human subjects. J Am Coll Dent 2014;81:14–18.
      OpenUrlPubMed
    12. ↵
      2. Xu Y ,
      3. Wang L ,
      4. He J , et al
      . Prevalence and control of diabetes in Chinese adults. JAMA 2013;310:948–59.doi:10.1001/jama.2013.168118
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      2. Chen CM ,
      3. Kong LZ
      . The guideline for prevention and control of overweight and obesity in Chinese adults[M]. Beijing: People’s Medical Publishing House, 2006.
    14. ↵
      1. World Health Organisation
      . Guidelines for controlling and monitoring the tobacco epidemic. Genva: Tobacco or Health Programme,WHO, 1997.
    15. ↵
      2. Romero P ,
      3. Sagarra R ,
      4. Ferrer J , et al
      . The incorporation of family physicians in the assessment of diabetic retinopathy by non-mydriatic fundus camera. Diabetes Res Clin Pract 2010;88:184–8.doi:10.1016/j.diabres.2010.02.001
      OpenUrlCrossRefPubMed
    16. ↵
      2. van Leiden HA ,
      3. Dekker JM ,
      4. Moll AC , et al
      . Blood pressure, lipids, and obesity are associated with retinopathy: the Hoorn study. Diabetes Care 2002;25:1320–5.doi:10.2337/diacare.25.8.1320
      OpenUrlAbstract/FREE Full Text
    17. ↵
      2. Kumari N ,
      3. Bhargava M ,
      4. Nguyen DQ , et al
      . Six-year incidence and progression of diabetic retinopathy in Indian adults: the Singapore Indian Eye study. Br J Ophthalmol 2019;103:1732–9.doi:10.1136/bjophthalmol-2018-313282
      OpenUrlAbstract/FREE Full Text
    18. ↵
      2. Kawasaki R ,
      3. Kitano S ,
      4. Sato Y , et al
      . Factors associated with non-proliferative diabetic retinopathy in patients with type 1 and type 2 diabetes: the Japan diabetes complication and its prevention prospective study (JDCP study 4). Diabetol Int 2019;10:3–11.doi:10.1007/s13340-018-0357-z
      OpenUrl
    19. ↵
      2. Cohen O ,
      3. Norymberg K ,
      4. Neumann E , et al
      . Complication-Free duration and the risk of development of retinopathy in elderly diabetic patients. Arch Intern Med 1998;158:641–4.doi:10.1001/archinte.158.6.641
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      2. Gray N ,
      3. Picone G ,
      4. Sloan F , et al
      . Relation between BMI and diabetes mellitus and its complications among US older adults. South Med J 2015;108:29–36.doi:10.14423/SMJ.0000000000000214
      OpenUrlCrossRefPubMed
    21. ↵
      2. Rooney D ,
      3. Lye WK ,
      4. Tan G , et al
      . Body mass index and retinopathy in Asian populations with diabetes mellitus. Acta Diabetol 2015;52:73–80.doi:10.1007/s00592-014-0602-2
      OpenUrl
    22. ↵
      2. Kim SJ ,
      3. Jee SH ,
      4. Nam JM , et al
      . Do early onset and pack-years of smoking increase risk of type II diabetes? BMC Public Health 2014;14:178.doi:10.1186/1471-2458-14-178
    23. ↵
      2. Pan A ,
      3. Wang Y ,
      4. Talaei M , et al
      . Relation of active, passive, and quitting smoking with incident type 2 diabetes: a systematic review and meta-analysis. Lancet Diabetes Endocrinol 2015;3:958–67.doi:10.1016/S2213-8587(15)00316-2
      OpenUrl
    24. ↵
      2. Sung Y-T ,
      3. Hsiao C-T ,
      4. Chang I-J , et al
      . Smoking cessation carries a short-term rising risk for newly diagnosed diabetes mellitus independently of weight gain: a 6-year retrospective cohort study. J Diabetes Res 2016;2016:1–7.doi:10.1155/2016/3961756
      OpenUrl
    25. ↵
      2. Chang SA
      . Smoking and type 2 diabetes mellitus. Diabetes Metab J 2012;36:399–403.doi:10.4093/dmj.2012.36.6.399
      OpenUrlPubMed
    26. ↵
      2. Hilawe EH ,
      3. Yatsuya H ,
      4. Li Y , et al
      . Smoking and diabetes: is the association mediated by adiponectin, leptin, or C-reactive protein? J Epidemiol 2015;25:99–109.doi:10.2188/jea.JE20140055
      OpenUrl
    27. ↵
      2. Moss SE ,
      3. Klein R ,
      4. Klein BE , et al
      . Cigarette smoking and ten-year progression of diabetic retinopathy. Ophthalmology 1996;103:1438–42.doi:10.1016/S0161-6420(96)30486-7
      OpenUrlPubMedWeb of Science
    28. ↵
      2. Stratton IM ,
      3. Kohner EM ,
      4. Aldington SJ , et al
      . UKPDS 50: risk factors for incidence and progression of retinopathy in type II diabetes over 6 years from diagnosis. Diabetologia 2001;44:156–63.doi:10.1007/s001250051594
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      2. Omae T ,
      3. Nagaoka T ,
      4. Yoshida A
      . Effects of habitual cigarette smoking on retinal circulation in patients with type 2 diabetes. Invest Ophthalmol Vis Sci 2016;57:1345–51.doi:10.1167/iovs.15-18813
      OpenUrl
    30. ↵
      2. Aggarwal S ,
      3. Khandelwal D ,
      4. Dutta D
      . Diabetes and smoking: the burden of evidence. In: Rodriguez-Saldana J , ed. The diabetes textbook. Cham: Springer, 2019.
    31. ↵
      2. Jaimes EA ,
      3. DeMaster EG ,
      4. Tian R-X , et al
      . Stable compounds of cigarette smoke induce endothelial superoxide anion production via NADPH oxidase activation. Arterioscler Thromb Vasc Biol 2004;24:1031–6.doi:10.1161/01.ATV.0000127083.88549.58
      OpenUrlAbstract/FREE Full Text
    32. ↵
      2. Zhang X ,
      3. Lim SC ,
      4. Tavintharan S , et al
      . Association of central arterial stiffness with the presence and severity of diabetic retinopathy in Asians with type 2 diabetes. Diab Vasc Dis Res 2019;16:498–505.doi:10.1177/1479164119845904
      OpenUrl
    33. ↵
      2. Kowluru RA
      . Mitochondrial stability in diabetic retinopathy: lessons learned from epigenetics. Diabetes 2019;68:241–7.doi:10.2337/dbi18-0016
      OpenUrlAbstract/FREE Full Text
    34. ↵
      2. Tyrberg M ,
      3. Nyström L ,
      4. Arnqvist HJ , et al
      . Overweight, hyperglycemia and tobacco use are modifiable risk factors for onset of retinopathy 9 and 17 years after the diagnosis of diabetes – a retrospective observational nation-wide cohort study. Diabetes Res Clin Pract 2017;133:21–9.doi:10.1016/j.diabres.2017.08.009
      OpenUrl
    35. ↵
      2. Wong J ,
      3. Molyneaux L ,
      4. Constantino M , et al
      . Timing is everything: age of onset influences long-term retinopathy risk in type 2 diabetes, independent of traditional risk factors. Diabetes Care 2008;31:1985–90.doi:10.2337/dc08-0580
      OpenUrlAbstract/FREE Full Text
    36. ↵
      2. Cheema S ,
      3. Maisonneuve P ,
      4. Zirie M , et al
      . Risk factors for microvascular complications of diabetes in a high-risk middle East population. J Diabetes Res 2018;2018:8964027.doi:10.1155/2018/8964027
    37. ↵
      2. Song SH ,
      3. Hardisty CA
      . Early onset type 2 diabetes mellitus: a harbinger for complications in later years--clinical observation from a secondary care cohort. QJM 2009;102:799–806.doi:10.1093/qjmed/hcp121
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      2. Zou W ,
      3. Ni L ,
      4. Lu Q , et al
      . Diabetes onset at 31–45 years of age is associated with an increased risk of diabetic retinopathy in type 2 diabetes. Sci Rep 2016;6:38113.doi:10.1038/srep38113

    Footnotes

    • ZH and XH are joint first authors.

    • ZH and XH contributed equally.

    • Contributors ZH: wrote the manuscript. XH: in-charge of fundus examinations. YQ: responsible for basic information collection. FT: responsible for the statistical analysis of the data. HL: responsible for auditing the fundus images and reports. RX: responsible for maintaining patient activities. BC: proposed the necessity and design of the study. HS: director of the department and also provided project support for the research.

    • Funding This work was supported by the National Key R&D Program of China (2018YFC1314000),Tianjin Major Science and Technology Projects（17ZXMFSY00200) and Tianjin Science and Technology Development Strategy Research Projects (18ZLZXZF00740).

    • Competing interests None declared.

    • Patient consent for publication Obtained.

    • Ethics approval This study was approved by the IRB of Tianjin 4th Central Hospital (2017-SZXL017) and the Chinese Clinical Trial Registry (No. ChiCTR1900020728).

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement Data are available upon reasonable request. Data are available upon reasonable request.