Diabetic neuropathy: a single entity with many faces?

Aims: What is the symptom-based temporal relationship between diabetic peripheral neuropathy (DPN), autonomic neuropathy (DAN) and painful DPN (pDPN)?

Methods: An electronic survey was distributed to n=29,155 eligible patients with diabetes in the North Denmark Region, of which n=7,321 patients were finally included in the analysis. Three validated questionnaires were used to assess for DPN, DAN, and pDPN namely the Michigan Neuropathy Screening Instrument questionnaire (MNSIq), the Composite Autonomic Symptom Score-31 (COMPASS-31), and Douleur Neuropathique en 4 Questions Interview (DN4). A MNSIq score ≧4 was used as the threshold for possible DPN, and a DN4 ≧3 independent of MNSIq for possible pDPN. A total COMPASS-31 score ≧16 was used to define DAN, while based on the severity of autonomic dysfunction (SAD) patients were further subcategorized into SAD- (<16), SAD+ (16-31), and SAD++ (≧32). Demographic data such as diabetes duration, treatment, income and others were also collected.

Results: The vast majority of participants had type 2 diabetes (86.3%), with a median age of 66 years (IQR 58-73), mean BMI 29.8 ± 6.1 kg/m2, and median diabetes duration of 10 years (IQR 5-18). The total prevalence of DAN was 40.9% (T1D=35.9%; T2D=41.9%), DPN=23.2% (T1D=23%; T2D=23.3%), and pDPN=8% (T1D=6.1%; T2D=8.2%). Notably, amongst participants with DAN, the total prevalence of DPN rose to  40.3% (T1D=44.6%; T2D=39.6%) and of pDPN to 33% (T1D=35.1%; T2D=31.8%). Furthermore, a relationship between DPN and DAN severity was observed as in the SAD- group, DPN prevlanece was 11.1% (95% CI: 10.2-12.1), rising to 26.4% (24.4-28.6) in SAD+, and 58.6% (55.9-61.3) in SAD++ regardless of diabetes subtype (all P<0.01). Similarly, there was a stepwise increase in pDPN prevalence from 8% (7.2-8.8) in SAD- to 20.2% (18.3-22.1) in SAD+, reaching as high as 50.2 % (47.4-52.9) in SAD++ (all P<0.01). There were moderate associations between COMPASS-31 and MNSIq (r=0.53, P<0.01), COMPASS-31 and DN4 (r=0.48, P<0.01). There were weak associations between the co-existence of DPN/DAN and BMI, diabetes duration, income, and insulin use (all P<0.01). There was no association with age, sex, diabetes type, educational level, or oral antidiabetic medications (all P>0.05). 

Conclusions: There is a strong association between DAN severity and painless and painful DPN suggesting shared underlying pathobiology.

Comments: This study elegantly harnesses the power of patient reported outcomes (PRO) and population registries to investigative at scale the temporal link between DAN, DPN, and pDPN. PROs are scalable and relevant to how patients feel albeit subjective, especially when not administered by trained personnel. This coupled with potential selection bias can limit precise estimation of true disease prevalence in the absence of objective measures. Nevertheless, patients with DAN appear to be at an excessive risk of DPN, which increases markedly with autonomic severity. These results confirm previous findings from smaller cohort studies and indicate potentially shared mechanistic pathways. Longitudinal studies are needed to establish the temporal evolution of DAN in relation to painless and painful DPN based on functional and structural measures. Corneal confocal microscopy is a non-invasive biomarker of small fiber neuropathy, which has also shown to detect autonomic and painful neuropathy. Given that DAN and DPN are underpinned by small fiber neuropathy, corneal confocal microscopy offers an objective and sensitive means to better understand the link between these debilitating complications of diabetes.

Ioannis N. Petropoulos

Reference. Røikjer J, Borbjerg MK, Poulsen MB, Nikontovic A, Ejskjaer N, Vestergaard P, Brock C. Symptom-driven co-existence of diabetic autonomic neuropathy, peripheral neuropathy, and painful neuropathy. Acta Diabetol. 2025 Nov 7. doi: 10.1007/s00592-025-02612-1. Epub ahead of print. PMID: 41201618.

🔗 https://link.springer.com/article/10.1007/s00592-025-02612-1