Impact of diabetes mellitus type two on the incidence of Parkinson’s disease

As is shown in Table 1, in 6 out of 7 publications addressing the topic, T2DM was described as a risk factor for PD (Hu et al. 2007; Xu et al. 2011; Yang et al. 2017; De Pablo-Fernandez et al. 2018; Rhee et al. 2020; Gialluisi et al. 2023).

Table 1 Summary of studies that evaluate the association between T2DM and incidence of PD

A population-based Korean cohort study demonstrated that T2DM as a comorbidity increased the risk of PD (Rhee et al. 2020). This study included 15,168,021 adults aged 40 years and older, divided into groups based on the presence of T2DM, impaired fasting glucose, duration of T2DM < 5 years, and duration of T2DM ≥ 5 years. During the observation period of 49,076,148.74 person-years, 31,577 patients developed PD. Compared to the group without DM or impaired fasting glucose, the hazard ratio (HR) was 1.03 (95% CI 1.009—1.067) in the group with T2DM and a duration < 5 years, and 1.618 (95% CI 1.566—1.672) in the group with T2DM and a duration ≥ 5 years, indicating that T2DM increased the risk of PD. Especially in those patients with longer duration of T2DM, the risk of PD significantly increased with the duration of diabetes.

In a Finnish study by Hu et al., including 51,552 men and women without a history of PD but with or without prior T2DM, 324 men and 309 women developed incident PD during a mean follow-up of 18.0 years. The HR for the occurrence of PD in patients with T2DM compared to those without T2DM was 1.85 (95% CI, 1.23—2.80) (Hu et al. 2007).

Gialluisi et al. examined various influencing factors on PD in a prospective study, including T2DM. Of the 23,901 individuals included, 213 developed incident PD during a median follow-up of 11.18 years. The risk of developing PD with comorbid T2DM in the multivariable Cox proportional hazards analysis yielded a HR of 1.85 (95% CI, 1.24—2.76) (Gialluisi et al. 2023).

Another prospective study from the USA investigated the association between self-reported T2DM and the future risk of PD in 288,662 participants. From 1995 to 2011, 1,565 participants developed PD. The risk of developing PD in patients with T2DM was derived from logistic regression models and showed an odds ratio (OR) of 1.41 (95% CI 1.20—1.66). Additionally, the risk was increased in individuals who had been diagnosed with T2DM for more than 10 years at baseline (OR = 1.75 (1.36—2.25) compared to those with a duration of less than 10 years (OR = 1.11 (0.89—1.38) (Xu et al. 2011).

De Pablo-Fernandez et al. analyzed data from a retrospective cohort study from 1999 to 2011 to estimate the risk of Parkinson's disease in patients with T2DM using Cox regression models. A total of 2,017,115 subjects with T2DM and 6,173,208 subjects without DM were included in the analysis. An increased risk of PD was observed in subjects with T2DM, with a HR of 1.32 (95% CI 1.29—1.35; p < 0.0001), compared to subjects without DM (De Pablo-Fernandez et al. 2018).

Yang et al. included a total of 36,294 patients with newly diagnosed T2DM and 108,882 individuals in the control group without DM between January 1, 2000, and December 31, 2006. Participants were followed until the first manifestation of PD or until December 31, 2011. The HR for the occurrence of PD in patients with pre-existing T2DM was estimated using the Cox proportional hazards regression model. Compared to the cohort without pre-existing DM, the incidence density of PD in the cohort with T2DM was 1.36 times higher (1.53 vs. 2.08 per 1000 person-years), with an HR of 1.19 (95% CI 1.08—1.32) adjusted for age, sex, comorbidities, and medication use (Yang et al. 2017).

A study by Simon et al. (Simon et al. 2007) reported no significant association between T2DM and the risk of developing PD. In this large prospective cohort study, using data from the Nurses' Health Study and the Health Professionals Follow-up Study, the relative risk (RR) was calculated as 1.04 (95% CI: 0.74–1.46). Participants, including 121,046 women and 50,833 men, were followed for a median duration of 22.9 years (women) and 12.6 years (men), with 530 cases of PD identified during the observation period. Exposures were assessed through self-reported doctor-diagnosed disease, rather than by direct clinical examination.

Impact of type two diabetes mellitus on the clinical progression of Parkinson patientsInfluence of type two diabetes mellitus on motor progression

Pagano et al. investigated the influence of T2DM on clinical progression in a case–control study. 25 patients with both T2DM and PD were compared with 25 patients with PD alone and 14 patients with T2DM alone. Additionally, the control group consisted of 14 patients without DM or PD. The clinical diagnosis of T2DM was confirmed by two consecutive fasting serum glucose measurements with values > 126 ml/dL. The endpoint for motor progression was defined as a change of one point on the Hoehn & Yahr scale. The cohorts were followed for 36 months with assessments conducted every 6 months. In the Cox proportional hazards analysis, the presence of T2DM in PD patients was associated with a higher risk of faster motor decline with an HR = 4.521 (95% CI 1.468 – 13.926; p < 0.01) (Pagano et al. 2018).

In the case–control study by Kotagal et al., 39 subjects were divided into two groups: 13 with PD and T2DM, and 26 with PD alone, matched for age, sex, disease duration, and comparable controls. All subjects underwent positron emission tomography to assess the volume ratio of [11C]dihydrotetrabenazine binding, a marker for nigrostriatal dopaminergic innervation, along with assessment of motor status using the Unified Parkinson's Disease Rating Scale (UPDRS). PD patients with comorbid T2DM showed increased postural instability compared to those without diabetes, and also exhibited more gait difficulties (p < 0.0005), although there was no significant difference of the mean striatal [11C]dihydrotetrabenazine distribution volume ratio between PD with T2DM and PD without DM (Kotagal et al. 2013).

In another case–control study, 89 patients with PD and comorbid T2DM were compared with 89 patients with PD alone, matched for gender, body mass index, and disease duration. Cereda et al. found that PD patients with comorbid T2DM had higher scores on the Unified Parkinson's Disease Rating Scale (UPDRS) for Part III (Motor Examination) (22.3 ± 9.0 vs. 19.3 ± 7.9; p < 0.019) and Part II (Motor Aspects of Experiences of Daily Living) (9.7 ± 5.1 vs. 8.3 ± 4.3; p < 0.049) compared to PD patients without DM (Cereda et al. 2012).

Ou et al. investigated in a prospective cohort study from March 2009 to August 2020 whether the control status of T2DM affects the progression of PD. 379 subjects were divided into three cohorts. 49 (12.9%) had both T2DM and PD. 22 (44.9%) of them had poorly controlled T2DM based on the HbA1c levels. Multivariate Cox proportional hazard regression models were used to assess predictors of PD progression. Rapid motor progression was defined as an increase of at least 14 points in the UPDRS-III score at a follow-up of 4.0 ± 2.4 years. In the cohort with poorly controlled T2DM, an increase in UPDRS-III scores ≥ 14 points was associated with an HR 2.060 (95% CI 1.165—3.641) compared to the control group without T2DM. In the cohort with well-controlled T2DM, the HR = 1.066 (95% CI 0.572—1.986) adjusted for age, sex, age at disease onset, body mass index, UPDRS-III scores, and Montreal Cognitive Assessment (MoCA) scores. Another endpoint was reaching Hoehn & Yahr stage ≥ 3. In the cohort with poorly controlled T2DM, the hazard ratio was 2.079 (95% CI 1.212 – 3.566) compared to an HR = 0.879 (95% CI 0.413 – 1.871) in the cohort with well-controlled T2DM (Ou et al. 2021).

In the study by Ogaki et al., the influence of diabetes mellitus on motor and non-motor symptoms in PD was investigated. In the cross-sectional study of 140 PD patients with available medical history and HbA1c serum levels, the correlation between HbA1c levels and clinical variables was analyzed. For this purpose, 23 PD patients with T2DM were grouped by severity of diabetes into complicated or uncomplicated DM based on HbA1c levels. The control group consisted of PD patients without DM. PD patients with T2DM had higher the Movement Disorder Society- Unified Parkinson's Disease Rating Scale Part III (MDS-UPDRS-III) scores of 36.3 ± 19.5 compared to MDS-UPDRS-III scores of 29.2 ± 15.3 in PD patients without DM. Additionally, PD patients with HbA1c levels > 5.7 exhibited elevated MDS-UPDRS-III scores of 33.3 ± 16.0 compared to PD patients with HbA1c levels ≤ 5.7 with MDS-UPDRS-III scores of 27.2 ± 15.6, indicating worse motor impairment in PD patients with higher HbA1c levels (Ogaki et al. 2023).

Influence of Diabetes mellitus type two on cognition

Pagano et al. also examined in the aforementioned study the influence of T2DM on cognition in the same cohorts. Subjectively reported memory problems of PD patients were assessed, along with six different neuropsychiatric tests (Letter-Number Sequencing Test, Semantic Fluency Test, Hopkins Verbal Learning Test-Revised Recall, Hopkins Verbal Learning Test-Revised Recognition Discrimination, Symbol Digit Modalities Test, and Benton Judgement of Line Orientation). Endpoints were defined as 2 out of 6 pathological neuropsychiatric tests and affirmation of memory problems. In the Cox proportional hazards analysis, the presence of T2DM in PD patients was associated with a risk of faster cognitive decline with an HR of 9.314 (95% CI 1.164—74.519; p < 0.05) (Pagano et al. 2018).

The research paper by Bohnen et al. also examined the effects of comorbid T2DM on cognition in patients with PD. In the cross-sectional study, 148 patients with PD, adjusted for age and Hoehn & Yahr stage, were divided into two cohorts. Fifteen PD patients had comorbid T2DM, while 133 patients did not. All subjects underwent [11 C] Methyl-4-Piperidinylpropionate (PMP)-Acetylcholinesterase (AChE)-PET imaging to evaluate cortical cholinergic denervation, [11C]dihydrotetrabenazine (DTBZ)-PET imaging to determine nigrostriatal denervation, and neuropsychological evaluations. A global cognitive Z-score was calculated based on normative data. Cognitive differences between patients with and without comorbid T2DM were assessed using analysis of covariance. The Z-score indicates whether the collected data are above or below the previously obtained average value. In the analyzed cohorts, PD patients with T2DM had a significantly lower mean global Z-score (−0.98 ± 1.01) compared to PD patients without DM (−0.36 ± 0.91; F = 7.76; p = 0.0061) p < 0.0001 (Bohnen et al. 2014).

The aforementioned study by Ogaki et al. also investigated the influences of T2DM on cognitive performance in PD patients. Binary logistic regression analysis, including age, sex, disease duration, MMSE scores (Mini-Mental State Examination), and MDS-UPDRS-III score as independent variables, revealed that a low MMSE score was an independent factor for PD patients with T2DM with an OR 0.823 (95% CI 0.714–0.930; p = 0.032) and PD patients with high HbA1c levels OR of 0.879 (95% CI 0.781–0.989; p = 0.032), meaning that a low MMSE score is associated with a higher risk of PD-DM (Ogaki et al. 2023).

Uyar et al. examined the association between HbA1c levels and cognitive impairment in 195 patients with PD and comorbid T2DM. The HbA1c levels of PD patients correlated significantly with the MoCA score. PD patients with comorbid T2DM had lower MoCA scores, indicating cognitive impairment (Uyar et al. 2022).

Influence of diabetes mellitus type two on treatment response

Cereda et al. investigated whether there were differences in levodopa treatment doses in PD patients with comorbid T2DM compared to PD patients without. 89 patients with newly diagnosed PD and pre-existing T2DM were included from 2007 to 2010. The control group consisted of 89 PD patients without T2DM, matched by gender, body mass index (± 1 kg/m2), and disease duration (± 1 year). PD patients with comorbid T2DM had higher levodopa treatment doses than PD patients without DM (mg/day, 448 ± 265 vs. 300 ± 213; p < 0.0001; mg/kg/day, 5.8 ± 4.0 vs. 3.8 ± 2.9; p < 0.0001) (Cereda et al. 2012).

Influence of diabetes mellitus type two on overall mortality

Pezzoli et al. examined the influence of antidiabetic therapy on the age at onset of PD and its impact on overall mortality. Data from 8,380 patients were included. There was a delayed onset of PD by 6.2 years (p < 0.001) in patients receiving antidiabetic therapy for T2DM before the onset of PD compared with patients in whom T2DM occurred after the onset of PD and those without DM. Pre-existing T2DM also showed a negative effect on prognosis with an adjusted hazard ratio of 1.64 (95% CI 1.33—2.02; p < 0.0019). In patients who developed DM after the onset of PD, there was no negative impact on survival HR = 0.86 (95% CI 0.53 – 1.39; p = 0.54) (Pezzoli et al. 2023).

Impact of diabetes mellitus type two on the biomarker profile of Parkinson patients

In the case–control study by Pagano et al., the influence of T2DM on neurological progression also included the impact on tau protein levels in cerebrospinal fluid in 25 patients with T2DM and PD, 25 patients with PD without DM, and 14 patients with T2DM alone. The control group consisted of 14 subjects without PD or DM. The samples from PD patients with T2DM showed higher pTau181 and total Tau level in cerebrospinal fluid (p < 0.05) compared to the other cohorts (Pagano et al. 2018). Additionally, Pagano et al.'s study showed that PD patients with T2DM exhibited lower striatal dopamine transporter binding on [123I]FP-CIT SPECT imaging compared to Parkinson's patients without DM (Pagano et al. 2018).

In the study by Uyar et al. with 195 subjects, T2DM in PD patients was associated with higher serum neurofilament light chain levels in the linear regression model after adjustment for age and BMI, serving as a marker for more profound neuronal damage in these patients (Uyar et al. 2022).