Binary reversals: a diagnostic sign in primary progressive aphasia

Background Binary reversals (exemplified by ‘yes’/‘no’ confusions) have been described in patients with primary progressive aphasia (PPA) but their diagnostic value and phenotypic correlates have not been defined. Methods We conducted a retrospective cohort study analysing demographic, clinical, neuropsychological, linguistic and behavioural data from patients representing all major PPA syndromes (non-fluent/agrammatic variant, nfvPPA; logopenic variant, lvPPA; semantic variant, svPPA) and behavioural variant frontotemporal dementia (bvFTD). The prevalence of binary reversals and behavioural abnormalities, illness duration, parkinsonian features and neuropsychological test scores were compared between neurodegenerative syndromes, and the diagnostic predictive value of binary reversals was assessed using logistic regression. Results Data were obtained for 83 patients (21 nfvPPA, 13 lvPPA, 22 svPPA, 27 bvFTD). Binary reversals occurred in all patients with nfvPPA, but significantly less frequently and later in lvPPA (54%), svPPA (9%) and bvFTD (44%). Patients with bvFTD with binary reversals had significantly more severe language (but not general executive or behavioural) deficits than those without reversals. Controlling for potentially confounding variables, binary reversals strongly predicted a diagnosis of nfvPPA over other syndromes. Conclusions Binary reversals are a sensitive (though not specific) neurolinguistic feature of nfvPPA, and should suggest this diagnosis if present as a prominent early symptom.


INTRODUCTION
Diagnosis of primary progressive aphasia (PPA) is challenging even for expert clinicians. 1 Given the current dearth of objective biomarkers in these diseases, clinical phenotyping remains paramount.3][4][5] This symptom may constitute a specifically neurolinguistic feature, rather than reflecting a more generalised deficit of behaviour regulation. 5However, the diagnostic value of binary reversals has not been clarified.
Here we addressed this issue in a large, well characterised patient cohort representing all canonical syndromes of PPA and the behavioural variant of frontotemporal dementia (bvFTD).We assessed the prevalence of binary reversals in relation to syndromic diagnosis and associated clinical, neuropsychological and behavioural features.Based on clinical experience and previously published observations, [2][3][4][5] we hypothesised that binary reversals would be more prevalent in the non-fluent/ agrammatic variant of PPA (nfvPPA) than other syndromes and would be associated with linguistic deficits rather than behavioural abnormalities.

Assessment of patients
We assessed all patients in our active research cohort at the Dementia Research Centre who fulfilled consensus diagnostic criteria for nfvPPA, logopenic variant PPA (lvPPA), semantic variant PPA (svPPA) or bvFTD. 6 7All had syndromes of mild-to-moderate severity and supportive brain MRI with minimal cerebrovascular burden.Patient group characteristics are summarised in table 1.
Using a structured clinical survey, we recorded the presence (or absence) of binary reversals and other potentially relevant behavioural symptoms following illness onset (online supplemental table S1) in online supplemental file 1), consulting with each patient's primary caregiver or equivalent close informant; informants were invited to provide examples of the symptom.Patients underwent neurological examination and a comprehensive neuropsychological assessment (table 1).In addition, we recorded whether binary reversals were associated with parkinsonism and/or a diagnosis of corticobasal syndrome (CBS) or progressive supranuclear palsy (PSP).

Statistical analyses
Statistical analyses were performed using SPSS V.28.0 and R (V.4.
Parkinsonian features were present in 62% of patients with nfvPPA (half with a diagnosis of PSP or CBS) but were less prevalent in other syndromic groups and not consistently associated with binary reversals (table 1).

DISCUSSION
We have shown that binary reversals strongly predict a diagnosis of nfvPPA, developing in a high proportion (here 100%) of patients with this syndrome and significantly more frequently than in other PPA syndromes or bvFTD.Binary reversals were uncommon in svPPA and though encountered in around half of patients with lvPPA and bvFTD, developed later and/or in the context of more severe cognitive impairment in these syndromes than in nfvPPA.This feature may therefore have higher diagnostic specificity earlier in the course of the illness.
While this study does not elucidate the pathophysiological mechanism, it is noteworthy that, within the bvFTD group, patients with binary reversals performed significantly less well on neurolinguistic measures (phonological fluency, phonological working memory, word comprehension and reading) than patients who did not make reversals, whereas the two subgroups had otherwise comparable executive, general cognitive and behavioural profiles.This suggests that the development of binary reversals can form part of the neurolinguistic phenotype of bvFTD. 10Although it was not possible to analyse the specific associations of binary reversals in the nfvPPA group (since reversals were universal in this group), no single behavioural feature nor the presence of clinical parkinsonism, PSP or CBS was required for binary reversals to manifest.Taken together, our findings suggest that binary reversals are a neurolinguistic phenomenon, rather than a non-specific consequence of executive dysregulation, in line with previous reports of similar reversals in aphasic stroke. 11On the other hand, impaired response inhibition (as indexed by impaired Stroop task performance) was present in all patient groups exhibiting binary reversals, which may signify a complex interplay of causative and permissive factors. 2 3 Further work is needed to characterise the semiology of binary reversals and the circumstances that provoke them.Here we simply recorded the occurrence of the symptom; quantifying the frequency and severity of binary reversals and tracking their longitudinal development in the individual patient would give a more nuanced picture and establish how this symptom relates to other features of the illness.Anecdotally, a similar phenomenon occurs in nfvPPA patients speaking languages other than English: this requires substantiation.The neural basis of the symptom also remains to be defined.Our nfvPPA cohort is fairly typical neuropsychologically and neurologically of other published series, 12 13 and (considered alongside previous observations 2 3 ) the propensity of this syndrome to manifest binary reversals may reflect the targeting of fronto-subcortical circuitry by causative tauopathies.However, unless binary reversals have led to an important communication failure, this symptom may not be volunteered. 4We propose that clinicians suspecting PPA should seek a history of binary reversals and-particularly where early and prominent-this curious phenomenon may constitute a useful diagnostic clue.Recognition will enable investigation and management, including speech and language therapy for patients in whom binary reversals present a significant issue for communication in daily life.
X Charles R Marshall @charl_marshall and Chris JD Hardy @cjdhardy

Figure 1
Figure 1Significant phenotypic associations of binary reversals in the behavioural variant frontotemporal dementia group.The figure shows boxand-whisker plots of clinical and neurolinguistic characteristics significantly associated with the presence of binary reversals, across individual patients with behavioural variant frontotemporal dementia (the patient group in which associations could be most reliably assessed; see text) at the prescribed threshold (p<0.05)(non-significant associations are plotted in online supplemental figure S1; see also text and online supplemental tableS2).Boxes represent the IQR, and whiskers indicate the overall range of values in each group; the horizontal line in each box represents the median; in each panel, data for patients who exhibited binary reversals at the time of assessment (b.r.+) are presented on the right (in blue) and data for patients who did not exhibit reversals (b.r.−) on the left (in green).Binary reversals were significantly associated with younger age at assessment, longer symptom duration and more severe deficits of phonological fluency, phonological working memory, single word comprehension and reading.BPVS, British Picture Vocabulary Scale (a measure of single word comprehensions); DS reverse, reverse digit span (a measure of phonological working memory); NART, National Adult Reading Test; Phon fluency, phonological fluency (number of words generated to a target initial letter in 1 minute).

Table 1
3.1).Kolmogorov-Smirnov tests and Levene's tests were first conducted to check for normality and homogeneity of variance, respectively.A one-way Analysis of Variance (ANOVA) assessed for age differences and a Kruskal-Wallis test for differences in Mini Mental State Examination (MMSE) scores between patient groups (irrespective of binary reversal status).The Kruskal-Wallis test was used to compare illness duration in the nfvPPA cohort versus patients with and without binary reversals in other syndromic categories.For Clinical, cognitive and behavioural characteristics of patient subgroups with/without binary reversalsThe table summarises demographic, clinical, behavioural and neuropsychological data for all participant groups, subdivided on the basis of whether or not they made binary reversals.Mean (SD) data are shown unless otherwise indicated; maximum scores on neuropsychological tests are shown in parentheses.Neuropsychological scores in bold indicate performance below the 10th percentile according to published norms or local normative data from the Dementia Research Centre research cohort of older healthy controls (n=40, 21 males, 19 females, mean age 68.0 (6.1)).bvFTD subgroups with and without binary reversals have been compared statistically, as case numbers in this diagnostic group made the comparison meaningful; significant differences between subgroups (p<0.05) are coded in italics.Not all neuropsychological tests were completed by all patients; numbers in the bvFTD group missing data for each test are presented in online supplemental table S2 in online supplemental material.*All patients in the nfvPPA group exhibited binary reversals; two patients in this group fulfilled criteria for primary progressive apraxia of speech (ie, presentation with 'pure' speech apraxia and normal performance on key language tests: GNT, BPVS, PALPA55 and sentence construction).†Estimated duration of symptoms.‡Missing data left only one patient so SD could not be calculated.