Cognitive impairment is one of the most important non-motor symptoms (NMSs) of Parkinson’s disease (PD; Aarsland et al., 2021), even in the early stages of the disease (Aarsland et al., 2009; Weil et al., 2018). The full spectrum of cognition occurs among individuals with PD, which ranges from normal cognition through mild cognitive impairment (MCI) to PD dementia (PDD; Aarsland et al., 2017). MCI is an intermediate stage that occurs between normal cognition and dementia in PD (Caviness et al., 2007). Approximately 50% of PD patients with normal cognition (PD-NC) develop MCI within 6 years (Pigott et al., 2015). Meanwhile, after 5 years, 27.8% of PD patients with mild cognitive impairment (PD-MCI) patients revert to normal cognition (Pedersen et al., 2017), and almost 40% of PD patients with MCI (PD-MCI) go on to develop PDD (Pedersen et al., 2017). During the past decade, MCI have become increasingly recognized as an important pre-dementia stage of PD.

Subjective cognitive complaints (SCCs) are subjectively identified cognitive decline of people who, may or may not, have detected impairment on neuropsychological tests (Barbosa et al., 2019). A meta-analysis suggests that elderly people with SCCs are at increased risk of MCI and dementia (Mitchell et al., 2014). SCCs are also common among PD patients (Lehrner et al., 2014), with a prevalence that ranges from 30.3 to 60% (Erro et al., 2014; Baschi et al., 2018). Approximately 70% of PD-NC patients who develop cognitive impairment over time have a cognitive complaint at baseline (Purri et al., 2020), and SCCs have been suggested to be an independent predictor of development of MCI in cognitively normal PD patients (Erro et al., 2014; Hong et al., 2014). Furthermore, Hong et al. demonstrated that, in whole PD cohort, SCCs were correlated with neuropsychological tests (Hong et al., 2018). However, among non-demented PD subjects, there was little evidence to suggest that SCCs were related to impaired cognitive testing (AlDakheel et al., 2019), and a longitudinal research showed that depression occured secondary to or independent of cognitive complaints (Purri et al., 2020). The inconsistency of results may be due to the included PD patinets with different cognitive status, for Baschi et al. (2018) suggested SCCs in PD with or without MCI seemed to have a different etiology. Besides, some studies have also discovered that SCCs are correlated with psychiatric symptoms (Lehrner et al., 2014; Chua et al., 2021). Therefore, the relationship of SCCs, MCI and clinical factors (e.g., neuropsychological tests, psychiatric symptoms) is not yet well understood.

In this study, we hypothesized that the prevalence and risk factors of SCCs among early PD patients differ depending on cognitive status. Using a cohort of newly diagnosed, untreated PD patients, this study aims to: (1) investigate the prevalence of SCCs in early PD patients with different cognitive status, and (2) examine the implicated risk factors of SCCs in early PD-MCI and PD-NC patients.

In the present study, we investigated the frequency of SCCs and its relationship to clinical factors, including neuropsychological tests and NMSs, among early PD patients with different cognitive status. The results of present study demonstrate that more than half of PD-MCI patients reported SCCs, while less than a third of PD-NC patients showed SCCs. In addition, although they don’t fit the diagnostic criteria for MCI, cognitively normal PD patients with SCCs performed worse in some cognitive measures evaluating attention/working memory, compared with those without SCCs. Finally, the PD-NC patients who did worse on SCWT were more likely to report experiencing SCCs. However, the PD-MCI patients who had more NMSs were more likely to report SCCs. Our findings suggest that SCCs in early PD patients have different implications that are dependent on the underlying cognitive status.

Herein, we discovered that 42.3% of newly diagnosed non-demented PD patients report experiencing SCCs, which is higher than previous reports (Lehrner et al., 2014; Chua et al., 2021). The reasons for this discrepancy may be due to a more appropriate subjective measure. The CCI includes 10 questions which explore the cognitive, memory and behavioral domains (Catherine et al., 2006). Thus, it seems to be more appropriate than single yes/no question (i.e., “Do you feel that your memory or thinking have gotten worse?”) to assess SCCs in PD. Indeed, among the few studies that have evaluated the prevalence of SCCs among newly diagnosed PD patients with MCI. Erro et al. (2014) have discovered that 30.3% of 76 newly diagnosed, untreated PD complained of subjective memory impairment (SMI), and approximately 43.4% fulfilled MDS recommend level I MCI criteria. However, this study did not use a standardized instrument to assess cognitive complaint, only relying on a single question in the NMSQ regarding the act of forgetting. Additionally, the author did not evaluate the prevalence of SCC in PD isolation or within the MCI frame. The prevalence of PD-NC + SCC in our cohort was 12.1%, while PD-MCI + SCC accounted for 30.3%. These data are slightly different from Baschi’s research, which demonstrated that 40 (27.2%) PD-NC patients and 48(32.6%) PD-MCI patients reported SCCs (Baschi et al., 2018). Yet, it’s worth noting that the PD patients that the authors concerned about are in mid-moderate stage of the disease, and may be more prone to significant cognitive decline. On the contrary, the prevalence of PD-MCI + SCC in Early Parkinson’s Disease Longitudinal Singapore (PALS) study was 10.7% (Chua et al., 2021), which is lower than our results. This discrepancy is likely due to a simple subjective measure, which may not be the most comprehensive measure for SCC (Kjeldsen and Damholdt, 2019). Besides, the PD-MCI patients in the PALS study are older than those of our research, and older patients are more likely to be aware that their cognitive difficulties are associated with general ageing, instead of PD, and, as a result, do not complain. Hence, differences between present and previous study results might rely on patient clinical characteristics (i.e., age and stage of the disease), diverse assessment measures, and different diagnostic criteria for MCI and SCCs.

Regarding performance on neuropsychological tests, data in our research revealed that the PD-NC + SCC group performed significantly worse with regards to attention/working memory than PD-NC-SCC, while the PD-MCI + SCC group did substantially at the same level of PD-MCI-SCC across all neuropsychological tests. In newly diagnosed PD patients, attention/working memory was the most frequently affected domain (Monastero et al., 2018). In addition, a previous study has that found that cognitive changes in attention/working memory were present in prodromal PD (Darweesh et al., 2017; Weintraub et al., 2017). Thus, our results suggest the presence of cognitive complaints in PD-NC patients may represent a stage of cognitive decline between normal cognition and MCI, as has been reported in the general population (Reid and Maclullich, 2006; Reisberg and Gauthier, 2008; Jessen et al., 2010). Additionally, SCC in preclinical Alzheimer’s disease (AD) is related to increased risk of conversion to dementia (Jessen et al., 2014; Mitchell et al., 2014). The presence of SCCs in PD is an independent risk factor for MCI and, at the same time, can be used as a predictor of future cognitive decline (Hong et al., 2014). Consequently, the report of SCCs, simple to perform, is meaningful in clinical work, which may alert clinicians to focus on early attention/working memory changes. Then, the clinicians know when to start comprehensive neuropsychological assessment evaluation and establish a clear baseline.

Interestingly, further studies have demonstrated that SCCs in PD-MCI and PD-NC groups are associated with different risk factors. We discovered that the former is independently associated with a score of NMSQ, whereas the latter relates to SCWT. Previous studies on the relationships between SCCs and clinical features (i.e., objective cognition test), behavioral symptoms (i.e., depression, apathy, and anxiety) in PD led to conflicting results (Erro et al., 2014; Lehrner et al., 2014; Hong et al., 2018; AlDakheel et al., 2019). A previous report has proposed that subthreshold depression is associated with SCCs in non-demented PD (Santangelo et al., 2014). Lehrner et al. (2014) revealed that depression and cognitive state is significantly associated with SCCs in PD, but Chua et al. (2021) found that in PD-MCI patients, their mood symptoms were not associated with SCCs. Notably, a recent research from the Parkinson’s Progression Marker Initiative (PPMI) discovered that minor hallucinations were associated with subjective cognitive decline (Bejr-Kasem et al., 2021). In addition, there is literature to suggest that, with the exception of various psychiatric disorders, personality traits, such as anxiety and sensitivity, are related to subjective cognitive decline (Jessen et al., 2014). Therefore, SCCs cannot be explained by separate NMS, which is a significant feature of early PD stage (O’Sullivan et al., 2008). NMSQ is developed to further understand the basis and clinical burden of NMS in PD (Breen and Drutyte, 2012). In addition, it can be divided into nine domains that range from sleep dysfunction, urinary, mood, and sexual to cognitive problems (Chaudhuri et al., 2007). Higher NMSQ scores are known to strongly influence quality of life (QoL; Zipprich et al., 2021), which may remind patients of their cognitive impairment. Consequently, our result that the number of NMS is related to SCCs in PD-MCI may be more robust. On the other hand, in line with previous studies (Yoo et al., 2020), the PD-NC + SCC group did worse in cognition than in the PD-NC-SCC group, especially in SCWT, which reflects attention, working memory, executive function, processing speed and cognitive flexibility (Scarpina and Tagini, 2017), despite the fact that both of their cognition performance are above appropriate norms. Furthermore, a neuroimaging study demonstrated that PD patients with SCCs had poorer performance on tasks related to attention, and more severe cortical atrophy in the anterior cingulate gyrus and right inferior parietal lobule, compared to those without (Hong et al., 2012). Ultimately, These results illustrate that SCCs in PD-NC participants may represent an early sign of the potential pathology of PD.

There exist a few limitations in our study. Firstly, CCI has not been validated in PD, and has been answered only by the patients. But Hong et al. (2018) found that CCI was strongly correlated with objective cognitive function in PD, and was used in many researches of subjective cognitive function (Yoo et al., 2020). What’s more, at present, there are neither authoritative recommendations for suitable assessment tool of SCCs, nor any available guidelines recommended someone that should be regarded as the most reliable when obtaining information from PD patients, caregivers and clinicians (Kjeldsen and Damholdt, 2019). Secondly, all two-by-two comparisons in our research were made without correction for multiple comparison. For our study was exploratory, the purpose of two-by-two comparison was to explore the association between SCCs presence and clinical factors (e.g., neuropsychological tests, and psychiatric symptoms) in PD-MCI and PD-NC groups. Therefore, we did not perform multiple comparison analysis test so as to minimize possible Type II errors resulting from stringent correction. Thirdly, it was a cross-sectional study that merely demonstrates association and the sample was relatively small. Yet, this study is a part of longitudinal study. We’ve been recruiting volunteers to participate, and the follow-up of patients has been proceeding in order. Further studies with larger samples, and follow-up data would be able to confirm these findings. At last, the paticipants of study were recruited in hospital, and a possible selection bias cannot be excluded.

In conclusion, we have demonstrated that SCC is frequently seen in early PD subjects with or without MCI. Furthermore, we found that the attention/working memory of cognitively normal PD patients with SCCs declined. Our findings can help clinicians identify patients who are at risk for cognitive decline in the future, and provide a new understanding of possible neuropathology in PD cognitive dysfunction.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Medical Ethics Committee of The Affiliated Brain Hospital of Nanjing Medical University. The patients/participants provided their written informed consent to participate in this study.

WL, JC, and CP made substantial contributions to the conception or design of the work. CP, PH, LY, YW, GZ, and RZ made substantial contributions to the acquisition of the data. CP, JR, and MY made substantial contributions to the analysis of the data. CP, JC, and WL made substantial contributions to the interpretation of the data. CP drafted the work. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.