Diagnostic accuracy of ankle-brachial index versus pulse wave index measurements in the diagnosis of peripheral arterial disease in diabetic patients: A retrospective record-based study

Technical Summary: Diagnostic Accuracy of ABI vs. PWI in Diabetic PAD

Problem Statement
Peripheral Arterial Disease (PAD) is a critical macrovascular complication in diabetic patients, significantly increasing the risks of lower-extremity amputation and mortality. The Ankle-Brachial Index (ABI) is the standard non-invasive tool for diagnosing PAD; however, its accuracy is compromised in diabetic populations due to medial arterial calcification (media sclerosis). This calcification renders vessels non-compressible, often leading to falsely elevated or normal ABI readings (false negatives), thereby delaying diagnosis and treatment. While the Toe-Brachial Index (TBI) is recommended for such cases, there is a need to evaluate alternative, accessible metrics. This study addresses the limitation of ABI in diabetic patients by investigating whether the Pulse Wave Index (PWI), derived from arterial pulse waveform analysis, offers superior diagnostic accuracy for early PAD detection.

Methodology
The study employed a retrospective, record-based design utilizing electronic file system data from the vascular surgery outpatient clinic at As-Salam Hospital in Port Said, Egypt.

• Population: The cohort consisted of 184 diabetic patients (males and females) aged 40 years or older. Patients with upper limb vascular procedures or arteriovenous fistulas were excluded.
• Data Collection: Demographics, glycemic control (HbA1c), comorbidities (hypertension, dyslipidemia, CKD), and clinical history were extracted.
• Measurements: A semi-automatic four-point oscillometric instrument (AngE, SOT Medical Systems) was used to measure:
  • ABI: Calculated as the ratio of maximal amplitude between upper and lower extremities. PAD was defined as ABI < 0.9.
  • PWI: Calculated by dividing the maximal pulse amplitude of the lower leg by that of the upper extremity, multiplied by the pulse wave's peak time. PAD was defined as PWI > 300.
• Statistical Analysis: Diagnostic performance was evaluated using Receiver Operating Characteristic (ROC) curve analysis to determine the Area Under the Curve (AUC). The Youden index was used to identify optimal cutoff values. Metrics included sensitivity, specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV), and total diagnostic accuracy.

Key Results

• Prevalence: The study found a PAD prevalence of 25% based on PWI criteria, compared to 20% based on ABI criteria.
• Diagnostic Performance (PWI vs. ABI):
  • PWI: Demonstrated superior discriminative ability with an AUC of 0.805 (95% CI: 0.712–0.898). At an optimal cutoff of > 156, PWI achieved high sensitivity (88.9%) and a strong Negative Predictive Value (93%), making it effective for ruling out disease. However, it showed lower specificity (35.8%).
  • ABI: Showed moderate discriminatory capacity with an AUC of 0.663 (95% CI: 0.543–0.783). While it maintained high specificity (95.9%), its sensitivity was notably low (41.7%) at a cutoff of < 0.895, indicating a high rate of missed diagnoses in this population.
• Combined Model: Combining ABI and PWI increased specificity to 97.9% and PPV to 82.4%, though the overall AUC (0.684) did not surpass PWI alone.
• Predictive Value: Multivariable analysis indicated that PWI was a stronger independent predictor of PAD than ABI.

Key Contributions

1. Validation of PWI in Diabetics: The study provides empirical evidence that PWI outperforms ABI in sensitivity and overall diagnostic efficacy for detecting PAD in high-risk diabetic patients, specifically addressing the issue of non-compressible vessels.
2. Quantitative Comparison: It offers a direct head-to-head comparison of diagnostic metrics (AUC, sensitivity, specificity) for ABI and PWI within a specific demographic (Egyptian diabetic cohort), highlighting the limitations of relying solely on ABI.
3. Screening Strategy: The findings suggest a potential shift in screening protocols where PWI serves as a primary sensitive screening tool to rule out PAD, while ABI or other confirmatory tests are used for rule-in scenarios due to ABI's high specificity.

Significance and Claims
The paper claims that the Pulse Wave Index is a more effective tool than the Ankle-Brachial Index for the early detection of Peripheral Arterial Disease in diabetic patients, primarily due to its ability to overcome the false-negative results caused by arterial calcification. The authors conclude that incorporating arterial stiffness assessment (via PWI) into screening strategies is warranted for diabetic populations where ABI may underestimate disease burden.

The study modestly positions its findings as a step toward reducing PAD-related complications through improved detection. It acknowledges limitations, including the cross-sectional design, the lack of a universal angiographic gold standard, and the single-center setting. Consequently, the authors do not claim PWI as a definitive replacement for all other methods but advocate for its evaluation as a primary screening method and call for future prospective, multicenter research to validate PWI cutoffs and assess its long-term prognostic value for limb and cardiovascular outcomes.