Prevalence and associated factors of uncontrolled hyperlipidemia among Thai patients with diabetes and clinical atherosclerotic cardiovascular diseases: a cross-sectional study
BMC Research Notes volume 14, Article number: 118 (2021)
The research aimed to determine the prevalence and associated factors of uncontrolled hyperlipidemia among Thai patients with the disease and Clinical ASCVD.
A total of 1,527 Thai diabetic patients with a history of ASCVD were included in the study. Uncontrolled hyperlipidemia was detected among 1,216 patients (79.6%; 95% CI 77.6–81.7). The independent factors associated with uncontrolled hyperlipidemia included being female (adjusted odds ratio (AORs); 1.5, 95% CI 1.2–2.0), using thiazolidinedione (AORs; 1.7, 95% CI 1.1–2.7), community hospital (AORs; 4.3, 95% CI 1.0–18.0) and BMI level at 18.5–22.9 kg/m2 (AORs; 2.2, 95% CI 1.2–4.0), 23.0–24.9 kg/m2 (AORs; 1.8 95% CI 0.9–3.3), 25.0–29.9 kg/m2 (AORs; 2.3 95% CI 1.3–4.3) and ≥ 30 kg/m2 (AORs; 2.5 95% CI 1.3–4.9).
Dyslipidemia (DLP) is one of the major risk factors of Clinical Atherosclerotic Cardiovascular Diseases (ASCVD) which is a group of diseases with one of the major mortality incidences in the world . The World Health Organization (WHO) reported that cardiovascular diseases were the most common cause of death globally; however, these diseases could be attenuated by lifestyle modification and medication used [2, 3]. DLP is defined by the elevated levels of plasma cholesterol, triglycerides, or both, or reduced levels of high density lipoprotein cholesterol (HDL-C) . Therefore, DLP can contribute to atherosclerosis [5, 6]. Then these vascular consequences proceed to Clinical ASCVD comprising ischemic heart disease (IHD), ischemic stroke, transient ischemic attack (TIA) and peripheral artery disease (PAD) [5, 6]. Diabetes mellitus (DM) serves an important role in DLP by decreasing catabolism of triglyceride rich lipoproteins of intestinal and hepatic origin . Therefore, DM affects lipid metabolism increasing cholesterol levels especially among patients with DM and DLP . However, limited information is available regarding the prevalence of uncontrolled DLP among patients with DM with ASCVD, especially in Thailand. We aimed to determine the prevalence and associated factors of uncontrolled hyperlipidemia among Thai diabetic patients with Clinical ASCVD.
The data were retrieved from the database: an Assessment on Quality of Care among Patients Diagnosed with Type 2 Diabetes (T2D) and Hypertension Visiting Hospitals under the Authority of the Ministry of Public Health (MoPH) and Hospitals in Bangkok, Thailand, (the Thailand DM/HT study) after obtaining permission from the National Health Security Office (NHSO) and the Medical Research Network of the Consortium of Thai Medical Schools (MedResNet). A standardized case report form was used to obtain the required information from medical records of DM treatment between February and May in 2018 and was sent to the Thailand DM/HT study of the NHSO Central Data Management Unit in Nonthaburi. This study was reviewed and approved by the Royal Thai Army Medical Department Institutional Review Board, reference number R045h/63_Exp.
The inclusion criteria comprised patients with T2D aged at least 18 years with history of Clinical ASCVD including IHD, ischemic stroke, TIA and PAD. Any patient who had participated in a clinical trial was excluded. Those patients may have received trial medication or placebo, influencing the outcome of the study. The participant population totaled 1,527 patients. Data collected included demographics (sex, age, occupation, religion and health scheme), weight, height, body mass index (BMI), smoking behavior, alcohol consumption behavior, hospital level, diabetic complications such as diabetic retinopathy, diabetic nephropathy, diabetic neuropathy, blood chemistry data including fasting plasma blood glucose, hemoglobin A1c (HbA1c), lipid profile including total cholesterol, triglyceride (TG), HDL-C and low density lipoprotein cholesterol (LDL), history of anti-hyperglycemic, anti-hypertensive and lipid-lowering drug use and glomerular filtration rate calculated using the epidemiology collaboration formula.
The study included those patients with a diagnosis of T2D and history of Clinical ASCVD. All patients needed to receive ongoing medical treatment in a registered hospital. Thai hospitals use the American Diabetes Association standard to diagnose and treat patients with T2D . Criteria to diagnose diabetes were defined as one of the following indicators. The first was fasting plasma glucose ≥ 126 mg/dL (7.0 mmol/L). Fasting is defined as no caloric intake for at least 8 h. The second was 2-h postprandial glucose (2-h PG) ≥ 200 mg/dL (11.1 mmol/L) during the oral glucose tolerance test which uses a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water. In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose comprises ≥ 200 mg/dL (11.1 mmol/L). The first three tests are confirmed by repeat testing at a second visit. Clinical ASCVD is defined by one of the following: (1) IHD including ST-elevation myocardia infraction, nonST-elevation myocardial infraction and unstable angina, (2) ischemic stroke and TIA and (3) PAD .
Extrapolating from the available data, an absolute reduction to an LDL-C level, 1.8 mmol/L (less than 70 mg/dL) or at least a 50% relative reduction in LDL-C provides the best benefit in terms of CVD reduction . Therefore, the target LDL-C of patients with very high cardiovascular risk as participants in the study had less than 70 mg/dL or a ≥ 50% reduction from baseline LDL-C. Thus, uncontrolled hyperlipidemia among patients with T2D and Clinical ASCVD was defined by LDL-C level ≥ 70 mg/dL (1.8 mmol/L) .
Data were analyzed using IBM SPSS Statistics for Windows, Version 22.0. Categorical data were presented as number and percentage. Prevalence was analyzed using descriptive statistics and reported as percentage and 95% confidence interval (CI). Multivariate analysis was performed using logistic regression analysis, and a p-value less than 0.05 was considered statistically significant.
A total of 38,568 patients with T2D were included in the Thailand DM/HT study in 2018. Of these, 1,527 patients with T2D with a history of Clinical ASCVD were enrolled in the present study. The patients’ baseline characteristics are presented in Table 1. Among enrolled patients, average age was 66.6 ± 10.3 years, and 842 (55.1%) patients were female. Most participants with hyperlipidemia were treated by statin and fibrates accounting for 80%. Of the 1,527 eligible participants, 1,216 patients did not achieve the LDL-C level goal according to 2018 AHA/ACC guidelines. Overall prevalence of uncontrolled hyperlipidemia (LDL-C ≥ 70 mg/dL) in the study was 79.6% (95% CI 77.6–81.7). This prevalence increased with older age and was more common among females. After adjusting for potential confounders, risk factors associated with uncontrolled hyperlipidemia were being female (adjusted odds ratio (AORs); 1.5, 95% CI 1.2–2.0), using thiazolidinedione (AORs; 1.7, 95% CI 1.1–2.7), community hospital (AORs; 4.3, 95% CI 1.1–18.0) and higher BMI levels (Tables 2 and 3).
These results demonstrated important implications to the Thai public health system because uncontrolled hyperlipidemia is a major risk factor for recurrent ASCVD and its complications. Our study revealed that the essential evidence of a high prevalence of uncontrolled hyperlipidemia among Thai patients with T2D and Clinical ASCVD was approximately 80%. Additionally, we found that being female, using TZD, hospital level and higher BMI level were associated with uncontrolled hyperlipidemia.
The present study showed that four fifths of the study participants could not control their LDL-C. This situation could be explained by several points. Firstly, this study was conducted among patients receiving a diagnosis of diabetes. In the setting of insulin resistance and DM, lipoprotein lipase activity, a rate-limiting enzyme in the hydrolysis of triglyceride, is reduced . Very low-density lipids (VLDL) can be incompletely lipolyzed, yielding increased serum levels of VLDL remnants, after catabolism and triglyceride exchange were completed. The LDL particles become progressively more enriched with triglycerides, which are lipolyzed to smaller, denser and more numerous particles. These smaller particles are more atherogenic than those of larger size and less dense LDL particles . Related studies in five Asian countries reported that the prevalence of uncontrolled LDL-C among patients with a history of coronary heart diseases was 92.0%; however, after 12 months follow-up, 33.8% of patients reached LDL-C goals (< 100 mg/dL) . Compared with a pool analysis of observational studies, the present study reported a lower prevalence of uncontrolled hyperlipidemia . One related study in China found a lower prevalence of uncontrolled hyperlipidemia (63.8%) among those experiencing acute coronary syndrome undergoing statin treatment (> 2 weeks) . Additionally, compared with one related study in the US, the prevalence of uncontrolled hyperlipidemia in the study was comparable with those in the US study accounting for 74.7%. However, uncontrolled hyperlipidemia in the US study was defined by an LDL-C level > 100 mg/dL because the participants were only patients with DM without history of ASCVD .
The present study showed that patients with uncontrolled hyperlipidemia tended to be higher among females. Similar to the US study, sex differences in the prevalence of, and trends in cardiovascular risk factors, treatment, and control in the US, 2001 to 2016, indicated men were more likely to be treated and to have controlled DLP, especially at older age . The phenomenon can be explained by several ways. Firstly, several epidemiologic studies found that postmenopausal women had different lipid profiles when compared with premenopausal women [17,18,19]. In the study, the average age of women with DM and history of ASCVD was 67.3 ± 10.6 years and menopausal. The reason for uncontrolled hyperlipidemia among females was the menopausal transition [20, 21]. Additionally, the causes of uncontrolled hyperlipidemia are mostly found among females experienced with Clinical ASCVD because females with DM have a cardiovascular mortality risk greater than males with DM: 20.9 versus 14.9% [22,23,24]. Moreover, postmenopausal women lack estrogen having two effects on lipid metabolism. Firstly, estrogen depends on regulation of LDL receptors resulting in increased LDL particle clearance by hepatocytes; thus, decreasing plasma LDL-C . Another study revealed that estrogen receptors are present in adipocytes which have a 17-beta-estradiol . The high level of this hormone directly affects marked inhibition of adipose tissue lipoprotein lipase (LPL) activity and hormone-sensitive lipase (HSL) . LPL initiates chylomicron metabolism and conversion of VLDL to LDL . HSL was studied in mice showing that HSL had compensatory mechanisms to increase LDL receptor expression . Thus, elderly females tended to have high LDL levels compared with males.
Another factor associated with uncontrolled hyperlipidemia among patients with DM and Clinical ASCVD was thiazolidinediones (TZDs) use. TZDs have the potential to benefit the full “insulin resistance syndrome” associated with DM . According to Thai rational drugs use in 2018, pioglitazone hydrochloride was the only drug in the TZD group. Some studies have revealed that treatment with pioglitazone modestly increased LDL-cholesterol levels by ~ 10 to 15% and increased LDL particle size [31, 32]. Another study in Germany, reporting that pioglitazone reduced atherogenic dense LDL particles among patients with T2D, showed that pioglitazone decreased the amount of LDL-6 (the densest LDL subfraction), resulting in larger and less dense LDL particles . However, the present study found that the TZDs were administered among patients with DM and history of ASCVD, who had uncontrolled hyperlipidemia. The reason may be from potential benefits on the secondary complications of DM. Another study found that among patients with T2D, pioglitazone plus sulfonylurea significantly improved HbA1C and fasting plasma glucose levels with beneficial effects on serum triglyceride and HDL-C levels .
Hospital levels in Thailand are classified in two categories: standard/advanced level and community level. We found that community level hospitals were significantly associated with uncontrolled hyperlipidemia among patients with DM and history of Clinical ASCVD. This finding may reflect the fact that community level hospitals are located in district areas and have limited resources including specialists, various types of medication and laboratory testing . Patients recovering from diseases were referred to and received regular medication in community hospitals . At the community level, patients’ cholesterol levels may be under monitored. In addition, a related survey conducted in a primary care setting in Thailand showed that necessary and routine aspects of diabetic care were not performed by the healthcare systems regularly . Thus, a higher prevalence of uncontrolled hyperlipidemia was observed among community hospitals when compared with the standard/advanced hospitals.
Finally, we found a dose response relationship between higher BMI level and uncontrolled hyperlipidemia. This result was similar to that of related studies in Spain and Indonesia showing that BMI was associated with a high LDL level [37, 38]. Another study in Thailand indicated the prevalence of small dense LDL increased obesity status in a Thai population which may be assumed to be an increased BMI . The phenomenon can be explained by metabolic processes. Hyperlipidemia enhancing in obesity involves elevated fasting and postprandial TG combined with the essential small dense LDL and low HDL-C. An increase in TG level may be the principal cause of other lipid abnormalities because it leads to delayed clearance of the TG-rich lipoproteins and formation of small dense LDL [40,41,42]. However, some studies have reported that LDL-C showed no significant correlation with BMI level [43, 44].
The strength of this study included the scope for uncontrolled hyperlipidemia among patients with DM and history of ASCVD. One implication of the study is to reduce the prevalence of uncontrolled hyperlipidemia, resulting from improving DM management. Moreover, the health literacy of our patients should be improved to increase awareness of their behaviors especially body weight control. Eventually, healthcare services access by patients with DM especially at community levels should be adjusted and improved to alleviate their cardiovascular complications.
This study employed a cross-sectional design, so the results could show only factors associated with uncontrolled hyperlipidemia. We were aware of missing data from this observational study. However, some data were missing as from the nationwide observational study, so few associations between factors and outcomes could be presented.
Availability of data and materials
The datasets generated and/or analyzed during the current study are available at http://www.damus.in.th after the permission of the Thailand DM/HT study of the MedResNet.
Atherosclerotic cardiovascular diseases
Systolic blood pressure
Diastolic blood pressure
Body mass index
Ministry of Public Health
National Health Security Office
Low density lipoprotein cholesterol
Roy S. Atherosclerotic cardiovascular disease risk and evidence-based management of cholesterol. N Am J Med Sci. 2014;6:191–8.
Rosenzweig JL, Bakris GL, Berglund LF, Hivert M-F, Horton ES, Kalyani RR, et al. Primary prevention of ASCVD and T2DM in patients at metabolic risk: an endocrine society* clinical practice guideline. J Clin Endocrinol Metab. 2019. https://doi.org/10.1210/jc.2019-01338.
World Health Organization. Cardiovascular diseases (CVDs) fact sheet. 2017. https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds).
Narindrarangkura P, Bosl W, Rangsin R, Hatthachote P. Prevalence of dyslipidemia associated with complications in diabetic patients: a nationwide study in Thailand. Lipids Health Dis. 2019;18:90.
Houston M. Chapter 27 - Dyslipidemia. In: Fourth E, editor. Rakel DBT-IM. New York: Elsevier; 2018. p. 264–75.
Association AD. Cardiovascular disease and risk management. Diabetes Care. 2016;39:S60-71.
Wu L, Parhofer KG. Diabetic dyslipidemia. Metabolism. 2014;63:1469–79.
Association AD. Classification and diagnosis of diabetes: standards of medical care in diabetes—2018. Diabetes Care. 2018;41:S13-27.
Stone NJ, Robinson JG, Lichtenstein AH, Goff DC Jr, Lloyd-Jones DM, Smith SC Jr, et al. Treatment of blood cholesterol to reduce atherosclerotic cardiovascular disease risk in adults: synopsis of the 2013 American College of Cardiology/American Heart Association cholesterol guideline. Ann Intern Med. 2014;160:339–43.
Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2019;73:e285-350.
Phan BAP, Toth PP. Dyslipidemia in women: etiology and management. Int J Womens Health. 2014;6:185–94.
Taskinen MR. Lipoprotein lipase in diabetes. Diabetes Metab Rev. 1987;3:551–70.
Unniachan S, Bash LD, Khovidhunkit W, Sri RZT, Vicaldo E, Recto C, et al. Prevalence of lipid abnormalities and attainment of normal lipid levels among patients with dyslipidaemia: a pooled analysis of observational studies from five Asian countries. Int J Clin Pract. 2014;68:1010–9.
Jiang J, Zhou Y-J, Li J-J, Ge J-B, Feng Y-Q, Huo Y. Uncontrolled hyperlipidemia in Chinese patients who experienced acute coronary syndrome: an observational study. Ther Clin Risk Manag. 2018;14:2255–64.
Jacobs MJ, Kleisli T, Pio JR, Malik S, L’Italien GJ, Chen RS, et al. Prevalence and control of dyslipidemia among persons with diabetes in the United States. Diabetes Res Clin Pract. 2005;70:263–9.
Peters SAE, Muntner P, Woodward M. Sex differences in the prevalence of, and trends in, cardiovascular risk factors, treatment, and control in the United States, 2001 to 2016. Circulation. 2019;139:1025–35.
Kumari P, Sahay G, Bano M, Niranjan R. A comparative study of serum lipid profile and premenopausal, perimenopausal and postmenopausal healthy women: a hospital-based study in Jharkhand. India Int J Contemp Med Res. 2018;5:H7-11.
Matthews KA, Meilahn E, Kuller LH, Kelsey SF, Caggiula AW, Wing RR. Menopause and risk factors for coronary heart disease. N Engl J Med. 1989;321:641–6.
Wu ZY, Wu XK, Zhang YW. Relationship of menopausal status and sex hormones to serum lipids and blood pressure. Int J Epidemiol. 1990;19:297–302.
Ko S-H, Kim H-S. Menopause-associated lipid metabolic disorders and foods beneficial for postmenopausal women. Nutrients. 2020;12:78.
Carr MC, Kim KH, Zambon A, Mitchell ES, Woods NF, Casazza CP, et al. Changes in LDL density across the menopausal transition. J Investig Med Off Publ Am Fed Clin Res. 2000;48:245–50.
Brown HL, Warner JJ, Gianos E, Gulati M, Hill AJ, Hollier LM, et al. Promoting risk identification and reduction of cardiovascular disease in women through collaboration with obstetricians and gynecologists: a presidential advisory from the American Heart Association and the American College of Obstetricians and Gynecolog. Circulation. 2018;137:e843–52.
Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet (London, England). 2004;364:937–52.
Peters SAE, Huxley RR, Woodward M. Diabetes as risk factor for incident coronary heart disease in women compared with men: a systematic review and meta-analysis of 64 cohorts including 858,507 individuals and 28,203 coronary events. Diabetologia. 2014;57:1542–51.
Bade G, Shah S, Nahar P, Vaidya S. Effect of menopause on lipid profile in relation to body mass index. Chronicles Young Sci. 2014;5:87.
Szafran H. Smielak-Korombel W [The role of estrogens in hormonal regulation of lipid metabolism in women]. Przegl Lek. 1998;55:266–70.
Hamosh M, Hamosh P. The effect of estrogen on the lipoprotein lipase activity of rat adipose tissue. J Clin Invest. 1975;55:1132–5.
Saxena U, Klein MG, Vanni TM, Goldberg IJ. Lipoprotein lipase increases low density lipoprotein retention by subendothelial cell matrix. J Clin Invest. 1992;89:373–80.
Kraemer FB, Shen W-J, Harada K, Patel S, Osuga J, Ishibashi S, et al. Hormone-sensitive lipase is required for high-density lipoprotein cholesteryl ester-supported adrenal steroidogenesis. Mol Endocrinol. 2004;18:549–57.
Hauner H. The mode of action of thiazolidinediones. Diabetes Metab Res Rev. 2002;18(Suppl 2):S10–5.
Goldberg RB, Kendall DM, Deeg MA, Buse JB, Zagar AJ, Pinaire JA, et al. A comparison of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes Care. 2005;28:1547–54.
Gauri Singh & Ricardo Correa. Pioglitazone. Treasure Island (FL): StatPearls Publishing; 2020. https://www.ncbi.nlm.nih.gov/books/NBK544287/.
Winkler K, Friedrich I, Baumstark MW, Wieland H, März W. Pioglitazone reduces atherogenic dense low density lipoprotein (LDL) particles in patients with type 2 diabetes mellitus. Br J Diabetes Vasc Dis. 2002;2:143–8. https://doi.org/10.1177/14746514020020021301.
Kipnes MS, Krosnick A, Rendell MS, Egan JW, Mathisen AL, Schneider RL. Pioglitazone hydrochloride in combination with sulfonylurea therapy improves glycemic control in patients with type 2 diabetes mellitus: a randomized, placebo-controlled study. Am J Med. 2001;111:10–7. https://doi.org/10.1016/s0002-9343(01)00713-6.
Euswas N, Phonnopparat N, Morasert K, Thakhampaeng P, Kaewsanit A, Mungthin M, et al. National trends in the prevalence of diabetic retinopathy among Thai patients with type 2 diabetes and its associated factors from 2014 to 2018. PLoS ONE. 2021;16:1.
Nitiyanant W, Chetthakul T, Sang-A-kad P, Therakiatkumjorn C, Kunsuikmengrai K, Yeo JP. A survey study on diabetes management and complication status in primary care setting in Thailand. J Med Assoc Thai. 2007;90:65–71.
Schröder H, Marrugat J, Elosua R, Covas MI. Relationship between body mass index, serum cholesterol, leisure-time physical activity, and diet in a Mediterranean Southern-Europe population. Br J Nutr. 2003;90:431–9.
Humaera Z, Sukandar H, Rachmayati S, Sofiatin Y, Roesli RMA. 64 Body mass index correlates with lipid profile in jatinangor population. J Hypertens. 2017;35. https://journals.lww.com/jhypertension/Fulltext/2017/11003/64_Body_Mass_Index_correlates_with_Lipid_Profile.40.aspx.
Kulanuwat S, Tungtrongchitr R, Billington D, Davies IG. Prevalence of plasma small dense LDL is increased in obesity in a Thai population. Lipids Health Dis. 2015;14:30. https://doi.org/10.1186/s12944-015-0034-1.
Klop B, Elte JWF, Cabezas MC. Dyslipidemia in obesity: mechanisms and potential targets. Nutrients. 2013;5:1218–40.
Capell WH, Zambon A, Austin MA, Brunzell JD, Hokanson JE. Compositional differences of LDL particles in normal subjects with LDL subclass phenotype A and LDL subclass phenotype B. Arterioscler Thromb Vasc Biol. 1996;16:1040–6.
Hokanson JE, Krauss RM, Albers JJ, Austin MA, Brunzell JD. LDL physical and chemical properties in familial combined hyperlipidemia. Arterioscler Thromb Vasc Biol. 1995;15:452–9.
Hussain A, Ali I, Kaleem WA, Yasmeen F. Correlation between Body Mass Index and Lipid Profile in patients with Type 2 Diabetes attending a tertiary care hospital in Peshawar. Pakistan J Med Sci. 2019;35:591–7.
Shamai L, Lurix E, Shen M, Novaro GM, Szomstein S, Rosenthal R, et al. Association of body mass index and lipid profiles: evaluation of a broad spectrum of body mass index patients including the morbidly obese. Obes Surg. 2011;21:42–7.
The authors wish to thank the entire staff members of the Department of Military and Community Medicine, Phramongkutklao College of Medicine, for their support in completing this study. The Thai DM/HT study and the MedResNet were supported by The Thailand National Health Security Office.
No funding was received.
Ethics Approval and consent to participate
The data of this study were retrieved from databases: Thailand DM/HT) after receiving permission from the MedResNet. The Thailand DM/HT was approved by local institutional review boards of the participating hospitals. The participants provided written consent in agreement with the WMA Declaration of Helsinki—ethics principles for medical research involving human subjects. This study was reviewed and approved by the Royal Thai Army Medical Department Institutional Review Board, Reference number R045h/63_Exp.
Consent for publication
This manuscript includes details and images not relating to any individual.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Lertwanichwattana, T., Rangsin, R. & Sakboonyarat, B. Prevalence and associated factors of uncontrolled hyperlipidemia among Thai patients with diabetes and clinical atherosclerotic cardiovascular diseases: a cross-sectional study. BMC Res Notes 14, 118 (2021). https://doi.org/10.1186/s13104-021-05535-6