Light and Shade in Patrick Lewis et al's Paper on the First Photographs of Parkinson's Disease

Patrick Lewis et al’s article, “Pierre D. and the First Photographs of Parkinson’s Disease,” discusses the case of a patient first reported by Albert Boucher (Observation IX) in his MD thesis of 1877, which was later extended by Paul de Saint-Léger (Observation VII) in his thesis of 1879. Pierre D.’s severe bilateral tremor began days after he was lined up against a wall with others to be executed during the Paris Commune; he was spared only because an alert caused the insurgents to flee “leaving behind more dead than alive” (p 54). A severe, acute-onset postural and action tremor of the hands and head prevented Pierre D. from working as a master mason and made it impossible for him to drink a glass of wine without spilling the contents. This symptom remained prominent for 6 years until 1877, when after a hospital admission and treatment with sulfurous baths, the tremor improved markedly, becoming limited to a mild quiver of the hands at rest. Considerable improvement in Pierre’s stiffness, gait, and speed of movement was also observed, which led Boucher to comment:

nature of her condition, which gives precedence to Regnard's earlier photograph of her over those of Pierre D.
Lewis et al further claim that the clinical descriptions of Pierre D., which spans an 8-year period, provide "a detailed longitudinal aspect to the case that was lacking from James Parkinson's original description of the shaking palsy" (p 1). 1 The authors have underplayed Parkinson's contribution in an attempt to over-egg the significance of their own report. In the Preface to his essay on the shaking palsy, Parkinson wrote: "The disease is of long duration: to connect, therefore, the symptoms which occur in its later stages with those which mark its commencement, requires a continuance of observation of the same case, or at least a correct history of its symptoms, even for several years" (pii). 8 His sixth clinical case concerned a 72-year-old man whom Parkinson noted had suffered from the condition for "[a]bout eleven or twelve, or perhaps more, years" (p 15). 8 In recounting the case Parkinson first summarized its trajectory and divided it into 3 triennial periods and reported that in year 11 the patient suffered a stroke from which he "nearly lost the use of the right side": "During the time of their having remained in this state, neither the arm nor the leg of the paralytic side was in the least affected with the tremulous agitation; but as their paralysed state was removed, the shaking returned" (p 16). 8 Parkinson's sixth case is a detailed and closely observed account whose "longitudinal aspect" is a prominent aspect of the description and confutes Lewis et al's assertion that Parkinson's exemplars were described only "in passing" (p 2). 1 Parkinson's appreciation of the course of the malady is exemplified in relation to a man whose shaking he noticed had completely ceased after 10 years, whom he concluded did not suffer from the shaking palsy. Its "longitudinal aspect" is also apparent in his generic account of the condition, set out in chapter 1 of the essay, which developed a detailed chronological clinical picture of the condition, from onset to death (p 498). 9 Increasing disability over the years was a constitutive aspect of Parkinson's belief that he had recognized a new syndrome.
The photographs of Pierre D. are striking visual illustrations of the postural abnormalities characteristic of the later stages of Parkinson's disease, but a still photograph can only hint at the changes of facial expression described by Charcot and cannot reliably demonstrate the cardinal features of rigidity, bradykinesia, or tremor found by inspection and neurological examination. For Lewis  We extend our thanks to Profs. Hurwitz, Lees, and Walusinski for their detailed and insightful commentary on our article relating to Pierre D., in particular for bringing to our attention to the photograph of Anne-Marie Gavr, of which we were unaware.

Spinal Cord Stimulation for
Parkinson's Disease: Dynamic Habituation as a Mechanism of Failure?
We read with great interesting the study from Prasad and colleagues 1 on the effects of spinal cord stimulation on gait in patients with advanced Parkinson's disease (PD), in which no clinically relevant influence was found over the 12 months of follow-up. We share the authors' view that the variety of stimulation parameters and heterogeneity of the studied population throughout spinal cord stimulation (SCS) trials makes a robust consensus unattainable at present.
We would like to contribute to the discussion with another current delivery strategy. Some PD patients from our practice have had a good initial response to SCS that evolves with a disappointing loss of benefit days to weeks following parameters adjustment. Considering the possibility of habituation, we implemented cycling stimulation as part of our SCS protocol. Surprisingly, this led to some benefit in patients unresponsive to tonic stimulation. To illustrate, we present the case of a 55-year-old male patient with PD and 5 years of STN DBS, with advanced gait problems. Thoracic SCS was performed, and no clear sustained effect was observed despite the various parameter tests at tonic stimulation. We compared cycling stimulation (alternating 15 minutes on-SCS, 15 minutes off-SCSarbitrary paradigm) to best continuous on-SCS and off stimulation (off-stim); evaluations were made 4 weeks apart from each scenario during on-medication. The patient improved from 17 seconds off-stim to 15 seconds continuous onstim and 10 seconds cycling on-stim during the 10-m walk test. In addition, he scored 12 on New Freezing of Gait Questionnaire (NFOG-Q) during off and continuous on-stim situations and 9 under the cycling protocol. Although the results were modest and from this single report, we ask whether some of the lack of sustained effect seen could derive from a "short habituation." Assuming that motor improvement after SCS in monkeys was strongly associated with desynchronization of aberrant lowfrequency corticostriatal oscillations, 2 SCS could work by replacing pathologically rhythmic in the gait network. The question is: are we efficiently delivering the current to do that? We hypothesize that delivering intermittent trains (ie, varying the interpulse interval [IPI]) could more powerfully break the pathologically disordered neuronal activity into the gait network 3 and prevent a possibly "quick habituation" of the stimulated circuitry. A more "chaotic" IPI could minimize the tolerance by avoiding phase-coupling ("adaptation") of the pathological oscillation with the SCS current. 3 Phase resetting is fundamental for the synchronization of different neurons, which might be better generated by reshaping IPI. Notably, chronic SCS in a parkinsonian model of rats applied the current only twice per week (30 minutes per session) and led to significant improvement in symptoms. 4 Last, despite the negative results reported by Prasad et al, the individual analysis showed different responsesone patient improved freezing by around 50% 1 month after SCS. We wonder whether predictive response for epidural SCS through transspinal magnetic stimulation applied before the surgery would be useful. This is an emerging method that activates similar target neural structures noninvasively and has recently been explored in the treatment of spasticity after spinal cord injury. 5 A pilot trial on PD is ongoing (clinicaltrial: NCT04171076). We kindly thank Cury and colleagues for their comments on our study and for sharing some of our concerns on spinal cord stimulation (SCS) for Parkinson's disease (PD). It is established that the current literature on this topic is extremely fragmented and heterogenous to make conclusions. Cury and colleagues are now proposing "habituation" as a mechanism explaining the decay of benefit seen in some of our and their patients. They illustrate the example of a single patient with PD who was evaluated for 4 weeks under an arbitrary cycling paradigm (SCS on and off for 15 minutes each epoch) and hypothesize that the delivery of a variable stimulation paradigm could minimize habituation by avoiding the phase coupling of pathological oscillation and SCS.
Although the concept of cycling stimulation is interesting, to our knowledge there has been no published experience with this type of stimulation in SCS patients with PD, or any other type of condition susceptible to SCS. A slightly different type of stimulation is intermittent dosing SCS (using several seconds SCS ON/OFF intervals with unchanged frequency and pulse width) in pain patients, overall showing that this approach is superior to continuous tonic stimulation only in terms of battery life. 1 Furthermore, in Feng et al.'s 2 paper, cited by Cury and colleagues, the hippocampal cells of healthy rats were stimulated with a variable interpulse interval, causing synchronization of neuronal activity at a local level. How this can be extrapolated to patients with PD is difficult to say. Interestingly, there are reports in patients with PD and essential tremor (the latter also susceptible to habituation), as well as computer simulations and animal studies, showing that random deep brain stimulation is less effective than continuous stimulation. [3][4][5] Importantly, one component highly susceptible of habituation is placebo effect, which is how we interpreted the transient 50% improvement of freezing seen in one of our patients 1 month after SCS. In contrast with most SCS studies published so far, we used subthreshold stimulation to lessen the placebo effect, although we changed to suprathreshold stimulation (ie, causing lower limbs paresthesias) during the last month of trial.
Finally, we agree with Cury and colleagues 1 that there might be a role for noninvasive stimulation to predict SCS response in patients with PD, although the issue of placebo will still need to be addressed.
In conclusion, a decade after the first cases of SCS for PD were published, the role of this therapy is still unclear beyond what has been claimed by enthusiastic review papers. Doubleblinded prospective studies in humans and using paresthesiafree suprathreshold stimulation (eg, burst stimulation) or, in keeping with what is proposed here, cycling stimulation might help elucidate this issue.

Brief Clinical Rating Scales Should Not Be Overlooked
Piot and colleagues 1 are to be highly commended for developing a brief rating scale for progressive supranuclear palsy (PSP). The PSP Clinical Deficits Scale (PSP-CDS) comprises seven domains, each scored 0-3. Clinimetrically sound and taking ≈5 minutes to complete, the scale is envisaged to be useful in clinical care and research, including clinical trials. By contrast, the current gold standard PSP Rating Scale, introduced in 2007, consists of 28 items, and takes ≈3× longer to administer. 1 In Parkinson's disease (PD), the Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS) is widely considered the gold standard. Encompassing 65 items, it has undergone extensive clinimetric testing and demonstrated strong outcomes. It also includes a broader range of nonmotor symptoms compared with its predecessor (the UPDRS); administration takes ≈30 minutes. The time-consuming nature of this and other comprehensive assessments, however, raises several issues. 2 The first concerns sampling bias in research. Severely disabled individuals are often excluded because of inability to complete more laborious evaluations. Such assessments also increase the likelihood of patient unwillingness to participate, in some cases influenced by low mood, anxiety, or apathy (all common in PD). Data on nonparticipation are infrequently reported, so it is hard to know the true extent of this problem. Recent statistics offer a glimpse; for example, a survey of 103 centers conducting research in PD genetics found that only two-thirds reported having UPDRS/MDS-UPDRS data for "at least a subset" of patients. 3 Unavailability of personnel and constraints on clinician time are critical factors further limiting participation, especially in under-resourced settings. 4 A lack of ethno-geographical diversity in studies has been increasingly highlighted as a major gap in medical research. PD is not exempt; a global listing of cohort studies showed that <10% were in Asia Pacific countries (representing >50% of the global patient population!). 4 This should raise alarm regarding the (un)representativeness of our current research enterprise-and the inequities this could entail.
In contrast, in well-resourced settings, it could be anticipated that the task of administering longer scales will often fall onto juniors, potentially affecting interrater reliability of scoring.
We recently captured "real-world" data on global PD severity in 828 patients within a busy clinic practice, using the Clinical Impression of Severity Index for PD. 5 Akin to the PSP-CDS, several core disease domains (motor signs, disability, motor complications, cognitive status) are rated on a 0-6 Likert scale, emphasizing functional impact. Ratings, typically taking ≤10 minutes, were incorporated into routine consultations, facilitating participation of patients spanning a wide demographic and disease spectrum (Fig. 1A). We reproduced the expected (mostly high) correlations with established measures 5 (Fig. 1B-D). The data also enabled broad comparisons of the clinical severity of selected monogenic PD cases 6 ( Fig. 1E-H).
Undoubtedly, comprehensive rating scales will continue to play an important role given the multifaceted nature of PD, because they capture more information, including nonmotor features that frequently cause significant impairment to the quality of life of people living with PD. At the same time, we should not overlook the usefulness of simple, but still valid and reliable, instruments to rate disease severity, which can be used in appropriate circumstances to improve care and bolster research.
The study was approved by the University of Malaya Medical Centre Medical Research Ethics Committee (ID numbers 732.5 and 20191010-7917) and conducted ethically in accordance with the World Medical Association Declaration of Helsinki. Written informed consent was obtained from all subjects. We thank Professor Lim and colleagues for commending the Progressive Supranuclear Palsy-Clinical Deficits Scale (PSP-CDS) 1 in their letter, "Brief clinical rating scales should not be overlooked." 2 The PSP-CDS is a physician-rated scale that was recently developed by members of the Movement Disorder Society (MDS)-endorsed PSP study group to reliably assess disease severity and progression in patients with PSP, regardless of the clinical phenotype. 1 The full version of the scale (PSP-CDS 7x3 ) comprises seven items with three response categories each (0 = no deficit; 1 = mild deficits not affecting activities of daily living (ADL); 2 = moderate deficits requiring partial external support; 3 = severe deficits necessitating permanent external support). The abbreviated PSP-CDS 7x2 scale omits the response category for mild clinical deficits, to be used in research settings, where such items without ADL-impact are of minor importance (eg, pivotal clinical trials). Finally, the short PSP-CDS 6x2 scale omits the item "Eye movements" from PSP-CDS 7x2 because their importance for ADL is controversial.
As a measure of clinical meaningful deficits that generates scores with appropriate annual sensitivity to change and a completion time of~4 minutes, 1 we suggest the PSP-CDS be applied in the context of clinical trials and in clinical routine care to monitor disease progression and to identify clinical deficits requiring medical attention.
We would like to point out, however, that we do not consider the PSP-CDS as a substitute for the Progressive Supranuclear Palsy Rating Scale (PSPRS) 3 in every instance, but rather a complementary instrument. The PSPRS is a 28-item scale to assess disease severity and progression in PSP. The PSPRS proved to be a reliable measure in many PSP randomized controlled trials and observation studies, as well as in routine clinical settings. [4][5][6] Compared to the PSP-CDS, the PSPRS assesses a larger number of items and has slightly better sensitivity to change but is less established for PSP with variant phenotypes (other than Richardson's syndrome) and takes more than three times as long to be completed. 3 This tradeoff needs to be considered based on the situation.
Prof. Lim and colleagues point out that the inability of severely disabled patients to complete comprehensive examinations, unavailability of personnel, and time constraints are potentially creating barriers for the use of lengthy scales in clinical routine and a bias in research applications. 2 Hence, in such situations, shorter scales, such as the PSP-CDS or the Clinical Impression of Severity Index for Parkinson's Disease (CISI-PD) 7 as mentioned by Lim and colleagues, 2 bear significant advantages. Further validation of the PSP-CDS will facilitate the decision on when to use the PSPRS or the PSP-CDS. Translation of the PSP-CDS into other languages is underway and will allow application in diverse geographical locations.
subclinical cardiac markers and motor impairment (Unified Parkinson's Disease Rating Scale Part III and Hoehn and Yahr stage) and cognition (Montreal Cognitive Assessment [MoCA]) concluding that (1) patients with PD presented higher levels of N-terminal pro-B-type natriuretic peptide (NT-proBNP) compared with healthy controls matched for age, sex, and cardiovascular risk factors; (2) increased levels of NT-proBNP and high-sensitivity troponin I were associated with worse motor impairment (Unified Parkinson's Disease Rating Scale Part III and Hoehn and Yahr stage); and (3) MoCA was inversely associated to high-sensitivity troponin I and NT-proBNP levels only in unadjusted models among patients with PD. 1 The population included were patients with advanced-stage PD (12 years of disease duration in average). 1 We have replicated these analyses in an independent population of patients with earlier stage PD (age 66 ± 9 years old, 56% male, disease duration 4.2 ± 2.62 years, Unified Parkinson's Disease Rating Scale Part III 24.5 ± 9.31) being evaluated for cardiovascular risk factors. We conducted a case-control study comparing patients with PD with non-PD controls. Patients with PD were recruited sequentially from the Movement Disorders Unit of the University Hospital of Santa Maria-Lisbon, and controls were recruited from 1 general practitioner list from a primary healthcare center in the same geographic region. All patients were diagnosed with PD by a movement disorders specialist according to the UK PD Brain Bank criteria 2 and had <11 years of disease duration. Controls were randomly selected from a general practitioner's list of patients and matched to the patients with PD according to age and gender. All participants were evaluated by an internal medicine physician and movement disorders expert. Blood samples were taken and the MoCA test was performed by a trained nurse. Dysautonomy/orthostatic hypotension was defined as the decline of at least 20 mmHg in systolic blood pressure and/or 10 mmHg in diastolic blood pressure within 5 minutes of going from supine to standing. Participants with heart failure were excluded.
All comparisons between patients with PD and matched controls did not reach significance (Table 1). In our analysis of the NT-proBNP marker among patients with PD, we only found higher significant results when comparing patients with PD with and without orthostatic hypotension (254 ng/L vs. 135 ng/L; P = 0.028). We also did not find a significant association of NT-proBNP with motor impairment or MoCA test in patients with PD.
Although our results are not conclusive, they go in the same direction of the conclusions of Choe and colleagues. The purpose of this letter is to add to the discussion the possibility that this association between subclinical cardiac markers and PD should be explored even at an earlier stage of the disease.   We are pleased that Alves and colleagues 1 underline the importance of the cardiac-brain axis in Parkinson's disease (PD). Alves and colleagues observed associations of the cardiac biomarker N-terminal pro-B-type natriuretic peptide (NT-proBNP) with cardiac dysautonomia in patients with early PD in their clinical cohort. Although nonsignificant, NT-proBNP levels were numerically higher and more often above a clinical cut-off of 125 ng/L in patients with PD compared with agematched and sex-matched controls. 1 As pointed out by Alves and colleagues, patients with PD in our Mark-PD cohort had a longer disease duration and therefore more advanced disease. 1,2 Patients with PD from Alves and colleagues were age-matched and sex-matched with our controls and patients with PD (P = 0.34 and P = 0.35, respectively) ( Table 1), but disease duration was much longer in our patients with PD (P < 0.001) ( Table 1). In patients with earlier PD, Alves and colleagues reported NT-proBNP levels of 158 AE 214 ng/L, which were significantly higher compared with age-matched and sexmatched controls from our cohort (NT-proBNP, 69 AE 32 ng/L; P < 0.001, 1-way analysis of variance with Bonferroni's post hoc test) and significantly lower compared with age-matched and sexmatched patients with advanced PD (NT-proBNP, 236 AE 348 ng/ L; P < 0.01, 1-way analysis of variance with Bonferroni's post hoc test) ( Table 1). Concerning the number of individuals with high NT-proBNP levels (>125 ng/L), we observed a similar pattern comparing controls and patients with early and advanced PD (Table 1).
Although data were matched only for age and sex, our combined data might indicate that subclinical cardiac biomarkers are indeed associated with disease duration and stage. Compared with NT-proBNP, we have observed stronger and higher significance levels for high-sensitivity troponin I compared with NT-proBNP concentrations. 2 Therefore, it might be even more interesting to evaluate high-sensitivity troponin I levels in earlier stages of PD in larger clinical trials. Furthermore, Alves and colleagues reported significantly higher NT-proBNP levels in patients with PD with compared without cardiac dysautonomia, whereas we did not observe such a difference in our cohort. At the moment it remains unclear to what extent cardiac dysautonomia influences the association of cardiac biomarkers and disease severity in patients with PD.
Finally, we agree with Alves and colleagues that the clinical relevance of subclinical cardiac biomarkers reflecting microdamage should be further evaluated in patients with PD at different disease stages.

Dopa-Responsive Parkinsonism in a Patient With Homozygous RFC1 Expansions
We read the letter titled "RFC1 Intronic Repeat Expansions Absent in Pathologically Confirmed Multiple Systems Atrophy" published online in April 2020 1 with great interest. In light of that letter, we would like to expand the phenotypic spectrum of RFC1 expansion-related disorders by reporting dopa-responsive parkinsonism in a 63-year-old woman. She developed parkinsonian symptoms in her early 50s, characterized by bradykinesia, resting tremor, and stiffness. The patient was started on levodopa as a symptomatic therapy with overt gait improvement (Video S1). Approximately 1 year later, she noticed oscillopsia and sensory complaints described as asymmetrical limb paresthesia that became confluent and associated with decreased vibration as well as proprioceptive sensation leading to gait unsteadiness. Head impulse test demonstrated absent vestibulo-ocular reflex bilaterally. A 20-year dry cough history was also reported. Brain magnetic resonance imaging and laboratory workup were unremarkable. Nerve conduction studies showed diffuse abnormalities restricted to sensory nerves, quantitative sudomotor axonal reflex was normal, and heart rate variability revealed incipient cardiac dysautonomia. Dopamine transporter scan highlighted a marked reduction of dopaminergic transporters in the bilateral striatum (Fig. 1A). Whole-exome sequencing failed to identify relevant variants, but triplet-repeat primed polymerase chain reaction detected biallelic (AAGGGn) intronic RFC1 expansions (Fig. 1C).
Since the description of biallelic intronic RFC1 expansions as the underlying cause for cerebellar ataxia with neuropathy and vestibular areflexia syndrome in 2019, the full phenotypic spectrum related to this genetic abnormality remains to be determined. 1-3 Herein we report a patient with RFC1 expansions leading to cerebellar ataxia with neuropathy and vestibular areflexia syndrome and dopa-responsive parkinsonism as part of her clinical picture. Although different cohorts involving patients with parkinsonism in the context of multiple system atrophy failed to demonstrate intronic RFC1 expansions as a causative factor, 1,3 we hypothesize that parkinsonism could be related to RFC1 in our patient. Indeed, the absence of autonomic features (with normal quantitative sudomotor axonal reflex) and hypo/anosmia combined with the lack of rapid eye movement behavior disorder all argue against the diagnosis of classical Parkinson's disease. Her benign clinical course without motor fluctuations more than a decade after levodopa start is also atypical. Lastly, the simultaneous occurrence of 2 rare conditions-homozygous RFC1 expansions and young-onset parkinsonism-just by chance would be possible but, rather, improbable. This description, although anecdotic, raises the possibility that dopa-responsive parkinsonism is part of the RFC1 clinical spectrum. Further studies in young-onset parkinsonian patients should be done to validate this assumption. disorders for the RFC1 expansion, to confirm its prevalence and further define the RFC1 clinical spectrum.

Debugging Adaptive Deep Brain Stimulation for Parkinson's Disease
The review by Little and Brown 1 on adaptive deep brain stimulation (aDBS) was very informative. However, a fundamental conundrum that challenges medicine and is likely a not infrequent cause of failure is the difficulty in caring for the individual patient through extrapolation from aggregate statistical descriptors of research studies. 2 Often, the problem is intractable. However, in the case of adaptive aDBS, we have an opportunity. We can focus on the detection algorithms and ask what is the specificity and sensitivity. Because importantly, we need to know the positive and negative predictive values, which requires knowledge of the prior probabilities of the signal to be detected. For example, if 15% of patients with Parkinson's disease do not have meaningfully increased beta power, the detection algorithm will have, at the minimum, a 15% false-negative rate. It may be that the other 85% of patients will do well with aDBS, and the aggregate clinical benefit in research studies will be positive. But the physician addressing the individual patient does not know a priori whether the patient is among the 15% who will be false negatives. Unfortunately, prior probabilities are difficult to determine because most published studies of beta oscillations pool individuals' data.
Reporting of other analyses such as logistic regression and the receiver operator curve characteristics of the detection algorithm may be helpful.
The second half of the question is the consequent therapeutic stimulation. It is not clear that the dynamics of the therapeutic response have been adequately considered relative to the timing and duration of stimulation. We can take, as a metaphor, drug pharmacokinetics where the dosing intervals are greatly affected by the drug half-life, which affects wash-in and washout periods. With drugs, any dosing interval less than the half-life likely is therapeutically equivalent. Dosing intervals greater than the half-life likely will be less therapeutic. Further, it may take multiple doses for the pharmacological effect to reach steady-state benefit unless one loads the patient. It is not likely that a loading dose of aDBS is feasible; thus, the time it takes DBS to reach steady-state effect is critical. For example, DBS in cycling mode at 500 ms on and 500 ms off was nearly as effective as continuous DBS, whereas cycling at 100 ms on and off was not as effective, even though the same number of stimulation pulses at the same parameters were given. 3 However, a caution, the study was not designed to address the therapeutic kinetics of DBS. The issues of DBS therapeutic kinetics analogous to pharmacokinetics are important. We know some disabilities have significant latencies to gain and loss of benefit, and hence differences in wash-in and washout effects. For many disabilities, we do not know the functional half-life of DBS. If the wash-in period for a DBS effect is on the order of many tens of seconds or more, what does that mean for aDBS? If the washout is on the order of many tens of seconds or more, what are the implications for aDBS?