https://orcid.org/0009-0009-0974-2194
Figures
Abstract
Background
Peripheral arterial disease (PAD) is more common among patients with diabetes mellitus (DM) and can lead to critical limb ischemia and amputation. Data on Doppler-diagnosed PAD are scarce in Tanzania. This study aimed to determine the prevalence of PAD and associated risk factors among DM patients referred for lower limb Doppler assessment at a tertiary hospital. Patients with ≥Grade III stenosis according to Jager’s criteria were classified as having peripheral arterial disease.
Methods
Analytical cross-sectional design was conducted among 80 adult DM patients suspected of PAD at Muhimbili National Hospital between April 3, 2023, and March 30, 2024. Demographic, clinical, and sonographic data were collected. PAD was defined as ≥Grade III stenosis by Jager’s criteria. Associations between risk factors and PAD were analyzed using Firth penalized logistic regression to account for small event numbers.
Results
Mean participant age was 61.2 ± 11.2 years; 55% were male. 18% had PAD, and 63.7% had hypertension. The dorsalis pedis artery was the most commonly affected segment (32.5%). Only DM duration was significantly associated with PAD (AOR = 1.19; 95% CI: 1.00–1.41; p = 0.044).
Conclusion
PAD is present in a notable proportion of DM patients referred for Doppler evaluation. Longer DM duration is an exploratory predictor. Findings are limited by small sample size, few PAD events, and missing HbA1c data. Larger, multicenter studies are needed to confirm risk factors and improve generalizability.
Citation: Sumawe JA, Salingwa L (2026) Doppler sonographic evaluation of peripheral arterial disease and its associated factors among diabetes mellitus patients at Muhimbili National Hospital, Tanzania: A hospital-based cross-sectional study. PLoS One 21(4): e0328852. https://doi.org/10.1371/journal.pone.0328852
Editor: Esha Arora, Asian Institute of Medicine Science and Technology: AIMST University, MALAYSIA
Received: July 20, 2025; Accepted: March 15, 2026; Published: April 2, 2026
Copyright: © 2026 Sumawe, Salingwa. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results has been uploaded as supplementary information in a file named supporting information.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: DM, diabetes mellitus; PAD, peripheral arterial disease; AOR, adjusted odds ratio; COR, crude odds ratio; CI, confidence interval; MNH, Muhimbili National Hospital; Hb, Hemoglobin; HbA1c, glycated hemoglobin; ABI, Ankle-Brachial Index
Introduction
Peripheral arterial disease (PAD) is the partial or complete obstruction of limb arteries, leading to reduced blood flow, tissue ischemia, and, in severe cases, ulceration or amputation [1]. Globally, PAD affects over 200 million people, and in sub-Saharan Africa (SSA) prevalence ranges from 1.7% to 52.5%, with higher rates reported when using Doppler-based diagnostics compared to clinical assessment alone [1,2].
Diabetes mellitus (DM) significantly increases PAD risk due to hyperglycemia, insulin resistance, hypertension, and Older age, smoking, obesity, and family history of cardiovascular disease further elevate risk [3–6]. PAD in DM can lead to serious complications, including non-healing ulcers, gangrene, and amputation [7,8].
While the ankle-brachial index (ABI < 0.9) is widely used internationally, it may underestimate PAD in DM due to arterial calcification [9,10]. Duplex ultrasonography provides non-invasive, accurate assessment of arterial stenosis and hemodynamic significance [7,11–13]. In this study, PAD was defined as ≥Grade III stenosis by Jager’s criteria (≥50% narrowing with hemodynamic compromise), allowing objective detection of clinically meaningful disease.
Despite increasing DM prevalence in Tanzania, local data on Doppler-diagnosed PAD are lacking. This study aimed to determine the prevalence, arterial distribution, and risk factors of PAD among adults with DM referred for lower limb Doppler sonography at a tertiary hospital. The findings will guide targeted screening and interventions to reduce PAD-related complications.
Methods and materials
Study design and setting
A hospital-based analytical cross-sectional study was conducted at the Radiology Department of Muhimbili National Hospital (MNH), Tanzania, between April 03, 2023 and March 30, 2024. Adult DM patients (≥18 years) with clinical suspicion of PAD referred for lower limb Doppler sonography were eligible. Patients with conditions affecting image quality (e.g., filariasis, severe swelling) or who were critically ill were excluded.
Sample size estimation
Sample size was calculated using a finite population formula:
- Population (N) = 100 DM patients referred annually for lower limb Doppler at MNH
- Expected PAD prevalence (P) = 30.7% (Ethiopia)(12)
- 95% confidence level (Z = 1.96), margin of error (d = 0.05)
The minimum required sample size was 80 participants.
Given the small population and convenience sampling, findings are exploratory and may not be generalizable to all Tanzanian DM patients.
Data collection
Demographic and clinical data were collected using a standardized questionnaire. Clinical variables included hypertension (measured), smoking history (lifetime, pack-years not assessed), duration of DM (from medical records), type of DM, and comorbidities. HbA1c was recorded when available but excluded from multivariable analysis due to missing values (54%).
Doppler ultrasound
Lower limb arterial Doppler examinations were performed by trained sonographers and verified by an experienced radiologist. Ultrasound systems used included GE Healthcare and Siemens ACUSON NX3 Elite platforms, linear transducer 7–11 MHz. Arterial stenosis was graded according to Jager’s criteria:
- Grade I: 1–19% stenosis
- Grade II: 20–49% stenosis
- Grade III: 50–99% stenosis (PAD)
- Grade IV: 100% stenosis
Statistical analysis
Data (S1 File) were analyzed using SPSS version 23. Descriptive statistics were used to summarize demographic, clinical, and sonographic characteristics. Binary logistic regression was first performed to estimate crude associations between independent variables and PAD. Variables with p ≤ 0.20 in crude analysis were considered for multivariable modelling. Given the small number of PAD events (n = 14), Firth’s penalized logistic regression was applied to reduce small-sample bias and mitigate separation issues. Statistical significance was set at p ≤ 0.05.
Multicollinearity among candidate predictors was assessed using variance inflation factors (VIF). All VIF values were below 2.0, indicating no evidence of problematic multicollinearity. Due to the use of penalized regression and the limited number of outcome events, conventional goodness-of-fit tests such as Hosmer–Lemeshow were not considered reliable. Therefore, model fit was primarily evaluated using penalized likelihood estimation and inspection of confidence interval width. Results should be interpreted cautiously given the small event-per-variable ratio.
Consent
All research participants provided informed consent, a written one by signing the consent form after reading and understanding the aim of the study. A study included only adults participants (18years and above). A study did not include any minors.
Ethical approval
The original data collection was conducted according to the guidelines of the Declaration of Helsinki and approved by Institutional Review Board of Muhimbili University of Health and Allied Sciences ethical approval numbers MUHAS-REC-03-2023-1606.
We recruited human participants for the study by distributing questionnaires which had demographic and clinic-pathological characteristics. Also data on Doppler sonographic findings was collected after performing lower limbs Doppler arterial sonography.
Start Date: 03/04/2023
End Date: 02/04/2024.
Results
Study participants baseline characteristics
Eighty participants were enrolled. Mean age was 61.2 ± 11.2 years; 55% were male. Most had at least primary education (53.8%) and were married (71.3%) (Table 1).
Clinico-pathological characteristics of the study participants
Claudication and loss of sensation were the most common presenting symptoms, reported in about two-thirds of the study subjects (65% and 62.5% respectively). The majority of the study subjects (63.7%) had co-existing hypertension as a comorbid condition while approximately one-fifth admitted to having ever smoked cigarettes nearly half of the participants (46.3%) had been diagnosed with diabetes for more than 10 years. More participants had type 2 DM(95%) and the most recent value of glycated hemoglobin was abnormal in 95%.The mean value of glycated hemoglobin was 10.2 ± 5.2. More information in Table 2.
PAD prevalence and arterial involvement
PAD (≥Grade III stenosis) was detected in 18% (n = 14). Dorsalis pedis artery was most commonly affected (32.5%), left deep femoral artery least affected (16.3%) (Fig 1). Right and left limb involvement were equal (35% each); bilateral involvement occurred in 31%. Grade I stenosis was most frequent (42.5%); Grade IV stenosis was rare (2.5%) (Figs 2 and 3).
Firth penalized logistic regression
In crude analysis, leg ulcer, smoking, and DM duration ≥10 years were associated with PAD. After adjustment using Firth penalized logistic regression, only DM duration remained statistically significant (AOR = 1.19, 95% CI: 1.00–1.41, p = 0.044). Other associations were not statistically significant. Confidence intervals were wide for several predictors, reflecting the limited number of PAD events and reduced statistical power. Although Firth’s method reduces small-sample bias, the low event-per-variable ratio may still contribute to instability in regression estimates. Therefore, effect sizes should be interpreted as exploratory. More information in Table 3.
Discussion
This study evaluated Doppler sonographic patterns of PAD and associated factors among 80 DM patients referred for lower limb arterial assessment at a tertiary hospital in Tanzania. We found that 18% of participants met the operational definition of PAD (≥Grade III stenosis by Jager’s criteria). The dorsalis pedis artery was the most commonly affected segment, and right and left limbs were equally involved.
Clinical presentation and comorbidities
The most frequent presenting symptoms were claudication (65%), loss of sensation (62.5%), and leg ulcers (62.5%), consistent with previous studies in Pakistan and Ethiopia [12,14]. Less than one-third of participants presented with frank gangrene or leg swelling, similar to the Pakistani study [14] but lower than some reports from Egypt [4]. Hypertension was the most common comorbidity (63.7%), higher than proportions reported in Ethiopia and Pakistan [12,14], potentially reflecting differences in population characteristics or healthcare access. Ischemic heart disease was present in 11.2% of participants, consistent with the lower prevalence reported in sub-Saharan Africa [15]. Smoking prevalence (17.5%) was comparable to Ethiopian data [12] but lower than in Dominican Republic and Egypt [4].
Prevalence of PAD
The prevalence of PAD in our study (18%) falls within the lower range of reported prevalence in Sub-Saharan Africa. Recent meta-analyses done in 2025 [16] indicate PAD prevalence among diabetic patients in the region ranges widely from 12% to 28% [16], depending on diagnostic modality and population characteristics. Our use of duplex Doppler sonography aligns with contemporary recommendations, as ABI measurements can underestimate PAD in diabetic populations due to arterial calcification. By integrating these recent regional analyses, our findings suggest that while PAD remains an important complication of diabetes, estimates vary substantially depending on the diagnostic method and patient selection. This highlights the need for standardized, regionally adapted screening strategies.
Factors associated with PAD
In crude analyses, leg ulcer, smoking, and DM duration ≥10 years were associated with PAD. After adjustment using Firth penalized logistic regression, only DM duration remained statistically significant (AOR = 1.19, 95% CI 1.00–1.41). This is consistent with previous studies from Ghana, Ethiopia, and Brazil [12,17,18]. Longer DM duration likely reflects cumulative vascular injury from chronic hyperglycemia, dyslipidemia, inflammation, and endothelial dysfunction [19].
Other classical well established cardiovascular risk factors such as hypertension and smoking were not significant in multivariable analysis. This may reflect limited statistical power due to small sample size (n = 80) and few PAD events (n ≈ 14), as well as missing data for HbA1c (54%). The small event-per-variable ratio also raises the possibility of overfitting and wide confidence intervals, limiting confidence in effect estimates. Our use of Firth penalized regression partially mitigates this risk, but findings should be interpreted cautiously.
In the setting of sub Saharan African countries, some studies suggested that there might be undetermined risk or epigenetic factors that may play a role in pathogenesis of PAD among patients with DM. For instance, one meta-analysis raises the suggestions than human immunodeficiency vasculopathy which is highly prevalent in sub Saharan Africa might have contributed to PAD among patients with DM [20]. This could serve as one of the population specific factors.
Importantly, the low number of PAD events (n = 14) may have resulted in unstable regression estimates despite the use of Firth penalization. Wide confidence intervals observed for several predictors suggest imprecision, and small changes in the dataset could potentially alter effect estimates. Therefore, these findings should be interpreted cautiously and viewed as hypothesis-generating rather than confirmatory.
Strengths and Context
This study provides one of the first Doppler-based assessments of PAD in Tanzanian DM patients and highlights the distribution of arterial involvement. The study used standardized imaging protocols and experienced sonographers, ensuring reliable detection of stenosis. The findings contribute to understanding PAD burden and may inform targeted screening strategies.
Limitations and critical reflection
Several limitations affect interpretation and generalizability. First, convenience sampling from a single tertiary hospital introduces potential selection bias, and participants may not represent the broader diabetic population in Tanzania. Second, the relatively small sample size and low number of PAD events reduced statistical power and increased the likelihood of imprecise estimates.
Third, although Firth penalized regression was used to minimize small-sample bias, the limited event-per-variable ratio may still result in unstable regression coefficients and wide confidence intervals. Fourth, missing HbA1c data (54%) prevented inclusion of glycemic control in multivariable modelling and may have obscured relevant associations. Fifth, residual confounding cannot be excluded, as socioeconomic status, medication use, diet, and physical activity were not assessed.
Finally, differences in PAD diagnostic criteria limit direct comparison with studies using ankle–brachial index (ABI < 0.9). Our Doppler-based definition (≥Grade III stenosis by Jager’s criteria) identifies hemodynamically significant disease but may yield prevalence estimates that are not directly comparable to ABI-based studies.
Given these limitations, the results should be interpreted as exploratory. Associations observed, particularly for DM duration, warrant confirmation in larger, multicenter studies.
Conclusion
PAD affects a notable proportion of DM patients referred for lower limb Doppler evaluation, with longer DM duration being the only significant predictor in this study. However, due to small sample size, low number of events, and potential biases, these findings are exploratory. Clinicians should interpret the results cautiously and consider targeted Doppler screening for high-risk patients while acknowledging the limited generalizability.
Recommendation
Clinical: Duplex ultrasonography remains a useful tool for assessing PAD in DM patients, particularly those with long-standing diabetes. Clinicians should consider targeted screening to identify patients at risk of complications.
Research: Larger, multicenter prospective studies are needed to confirm associations, improve generalizability, and explore additional risk factors, including glycemic control, medications, and lifestyle factors.
Methodological: Future studies should include strategies for handling missing data, ensure adequate sample size for multivariable modeling, and assess model assumptions, such as multicollinearity and fit.
Acknowledgments
We extend our gratitude to the management of Muhimbili National Hosipital for granting permission to conduct this study at the facility. A special thanks is extended to Mr. Hussein Lwinde who is a sonographer at MNH for assisting in data collection.
References
- 1. Nativel M, Potier L, Alexandre L, Baillet-Blanco L, Ducasse E, Velho G, et al. Lower extremity arterial disease in patients with diabetes: A contemporary narrative review. Cardiovasc Diabetol. 2018;17(1):138. pmid:30352589
- 2. Okello S, Millard A, Owori R, Asiimwe SB, Siedner MJ, Rwebembera J, et al. Prevalence of lower extremity peripheral artery disease among adult diabetes patients in southwestern Uganda. BMC Cardiovasc Disord. 2014;14:75. pmid:24913468
- 3. Soyoye DO, Abiodun OO, Ikem RT, Kolawole BA, Akintomide AO. Diabetes and peripheral artery disease: A review. World J Diabetes. 2021;12(6):827–38. pmid:34168731
- 4. El-sayed YA, Hak MTA, Mostafa AM. Role of arterial duplex of both lower limbs in the evaluation of diabetic foot. Role of arterial duplex of both lower limbs in the evaluation of diabetic foot. 2021;1(April).
- 5.
Alnima T, Meijer RI, Spronk HMH, Warlé M, Cate H. Diabetes- versus smoking-related thrombo-inflammation in peripheral artery disease. 2023;:1–13.
- 6.
Ogunmuyiwa O. Hypertension and diabetes comorbidity: Factors that are associated with their joint occurrence. 2023;:5–12. https://doi.org/10.57709/35480760
- 7.
Sohail S, Shaheen R, Sohail S. A doppler-based evaluation of peripheral lower limb arterial insufficiency in diabetes mellitus. 2010.
- 8.
Epidemiology C, Gaps M, Criqui MH, Matsushita K, Aboyans V. Lower extremity peripheral artery disease. 2021.
- 9. Hur KY, Jun JE, Choi YJ, Lee YH, Kim DJ, Park SW, et al. Color doppler ultrasonography is a useful tool for diagnosis of peripheral artery disease in type 2 diabetes mellitus patients with ankle-brachial index 0.91 to 1.40. Diabetes Metab J. 2018;42(1):63–73. pmid:29504306
- 10. Suwannasrisuk P, Sattanon S, Taburee W, Singkheaw P, Sowanna N, Boonprasert P, et al. Prevalence and predictors of peripheral arterial disease determined by ankle brachial index in diabetes population treated within primary care services in a non-urban area of lower northern Thailand. Diab Vasc Dis Res. 2020;17(6):1479164120966997. pmid:33158366
- 11. Hwang JY. Doppler ultrasonography of the lower extremity arteries: Anatomy and scanning guidelines. Ultrasonography. 2017;36(2):111–9. pmid:28219004
- 12. Akalu Y, Birhan A. Peripheral Arterial Disease and Its Associated Factors among Type 2 Diabetes Mellitus Patients at Debre Tabor General Hospital, Northwest Ethiopia. J Diabetes Res. 2020;2020:9419413. pmid:32090126
- 13.
Tityiwe JS. Duplex ultrasound assessment of early-stage peripheral artery disease in the lower limbs of Zimbabwean diabetic patients. 2020.
- 14. Shou Z, Zhao Y, Zhang Y, Li S. Risk factors for peripheral arterial disease in elderly patients with Type-2 diabetes mellitus: A clinical study. Pak J Med Sci. 2020;36(6):1344–8. pmid:32968406
- 15. Bosevski M. Peripheral arterial disease and diabetes polyvascular disease. Contrib Sec Biol Med Sci. 2012;1:65–78.
- 16. Haile KE, Asgedom YS, Azeze GA, Amsalu AA, Kassie GA, Gebrekidan AY. Burden of peripheral artery disease and risk factors among patients with diabetes mellitus in sub-Saharan Africa: A systematic review and meta-analysis. BMC Endocr Disord. 2025;25(1):42. pmid:39962460
- 17. Amissah I, Kissiedu E, Mbchb A. The prevalence of lower extremity peripheral artery disease among adults with Type 2 Diabetes Mellitus Attending a Teaching Hospital in Ghana. Int J Sci Res. 2016;5(6):2034–8.
- 18. Cardoso CRL, Melo JV, Santos TRM, Leite NC, Salles GF. Traditional and non-traditional risk factors for peripheral artery disease development/progression in patients with type 2 diabetes: the Rio de Janeiro type 2 diabetes cohort study. Cardiovasc Diabetol. 2021;20(1):54. pmid:33639945
- 19. Zou J, Zhang W, Chen X, Su W, Yu D. Data mining reveal the association between diabetic foot ulcer and peripheral artery disease. Front Public Health. 2022;10:963426. pmid:36062083
- 20. Johnston LE, Stewart BT, Yangni-Angate H, Veller M, Upchurch GR, Gyedu A, et al. Peripheral arterial disease in Sub-Saharan Africa: A review. JAMA Surg. 2016;151(6):564–72. pmid:27050676