Abstract
Background:
Alzheimer’s disease (AD) is a clinically heterogeneous disease. Multiple atypical syndromes, distinct from the usual amnesic phenotype, have been described. In this context, the existence of a right variant of AD (RAD), characterized by enduring visuospatial impairment associated with right-sided asymmetric brain damage, has been proposed. However, to date, this phenotype remains controversial. In particular, its peculiar characteristics and the independence from more prevalent cases (especially the posterior cortical atrophy syndrome) have to be demonstrated.
Objective:
To explore the existence of focal RAD on the basis of existing literature.
Methods:
We performed a literature search for the description of atypical AD presentations, potentially evoking cases of focal RAD. To be considered as affected by RAD, the described cases had to present: 1) well documented right-sided asymmetry at neuroimaging; 2) predominant cognitive deficits localizable on the right hemisphere; 3) no specific diagnosis of a known variant of AD.
Results:
Twenty-one cases were found in the literature, but some of them were subsequently excluded because some features of a different clinical syndrome were overlapped with the clinical features of RAD. Thirteen positive cases, three of them with pathologically confirmed AD, remained. A common right clinical-radiological syndrome, characterized by memory and visuospatial impairment with temporal and parietal involvement, consistently emerged. However, the heterogeneity among the reports prevented a definitive and univocal description of the syndrome.
Conclusion:
Even if sporadic observations strongly support the existence of a focal RAD, no definitive conclusions can still be drawn about it as an independent condition.
Keywords
INTRODUCTION
Alzheimer’s disease (AD) is a quite heterogeneous disease grouping different clinical-anatomical syndromes. In particular, some atypical phenotypes, which appear different from the usual amnesic syndrome, have been reported [1–3]. The amnesic syndrome is characterized by an initial dominant episodic memory impairment accompanied by impairments in other cognitive functions varying over the disease course (e.g., language, praxis, spatial abilities, executive functions). The atypical presentations of AD (i.e., variants of AD) are often focal syndromes with a predominant impairment of a cognitive function different from memory, especially at the early stages of the disease. In detail, three main atypical presentations of AD are currently recognized and included in the diagnostic criteria for AD together with the typical amnesic syndrome [4]: 1) a left variant showing a definite pattern of fluent aphasia (i.e., the logopenic) [5, 6]; 2) a frontal or behavioral/dysexecutive variant with behavioral features and executive impairment, sometimes clinically indistinguishable from the behavioral variant of frontotemporal dementia (bvFTD [7]; 3) a posterior variant (i.e., the posterior cortical atrophy syndrome, PCA) characterized by prominent visuospatial and visuoperceptive impairments [8, 9]. Furthermore, some authors have recently proposed specific diagnostic criteria for each of these atypical forms of AD (i.e., the posterior, logopenic, and frontal variants) as well as the Down’s syndrome variant of AD [10]. Interestingly, the variants accepted in the current nosology seem not to complete the full spectrum of clinical presentations of AD. In particular, a temporal variant characterized by isolated episodic memory impairment with a slow decline and a left variant showing a pattern of progressive non-fluent aphasia has been also described in some cases [3]. Moreover, also the corticobasal syndrome (CBS) [11, 12] can be considered as unusual presentations of AD. Finally, some authors have reported further uncommon clinical AD phenotypes, which do not entirely fit with the defined syndromes (e.g., a memory/semantic presentation of AD [13]; slightly different profiles from logopenic aphasia [14]; a transcortical cortical sensory aphasia [15]; a confabulating phenotype [16, 17]). In this context, the occurrence of a right variant of AD (RAD) distinct from the PCA syndrome has been hypothesized in a recent review [3]. In particular, this phenotype would be characterized by an enduring pattern of visuospatial dysfunction associated with greater right versus left hemisphere pathology and atrophy. Describing this variant, the authors presented the case of a 64-year-old woman with pathologically confirmed AD, who presented with short-term memory deficits and difficulties with visual guided tasks. She showed temporal and parietal atrophy as well as hypoperfusion, mainly located at the right parietal region.
Unfortunately, to date, the existence of RAD is still debated, in particular, due to 1) poor attention to this possible condition, and 2) probable underestimation of the phenomenon in the clinical and research setting [1, 18]. An accurate characterization of the primary clinical-radiological syndromes of AD is crucial importance to improve clinical decisions (e.g., for early diagnosis, management of the symptoms, counseling to caregivers, planning proper cognitive rehabilitation) and support the development of ad hoc interventions [18]. Moreover, understanding phenotypic heterogeneity in neurodegenerative diseases is an essential aim of current research, especially for theories concerning the scaling of degenerative molecular pathologies to the macroscopic brain [18].
The present article is aimed at exploring further evidence of a specific right clinical- radiological syndrome of AD (i.e., a RAD syndrome), supporting the view [3] of a new variant of AD different from the already known presentations. To this aim, a systematic search in the literature was conducted to report previously described atypical AD presentations consistent with RAD.
MATERIALS AND METHODS
We performed a literature search on PubMed for previous reports of AD patients presenting atypical asymmetric right-sided manifestations of the disease. No time limit for the publication of the reports was applied. Multiple terms were selected to render the search as comprehensive as possible. Each term (e.g., right, variant, atypical, visuospatial, prosopagnosia, unilateral spatial neglect, etc.) and appropriate combinations of different terms (e.g., right variant, visuospatial variant, etc.) were crossed-checked with the term “Alzheimer’s disease” (e.g., Alzheimer’ disease[title] AND right hemisphere[title], Alzheimer’s disease[title] AND right atrophy[title], etc.). Besides, some terms (i.e., visuospatial, visual-spatial, neglect, unilateral spatial neglect, prosopagnosia, topographical disorientation) have also been combined with the terms “primary progressive” and “progressive.” The details of the search are reported in the Supplementary Material. We set a PubMed advanced search exclusively in Titles, because of the high number of terms selected.
Given the exploratory aim of the study, in addition to case-studies reporting accurate clinical descriptions of possible RAD cases, we have included reviews on atypical AD presentations or asymmetries in AD too. Regarding the case-studies, a case was considered admissible if the following three criteria were respected: 1) describe the AD patient(s) presenting a definite right-sided (morphological and/or functional) asymmetry at imaging; 2) have a corresponding right-sided asymmetry in clinical syndrome and cognitive phenotype with preponderant either visuospatial impairment, or deficits of other cognitive abilities localizable at the right hemisphere (e.g., of visual memory, face processing, visual semantic, topographic orientation); 3) have no specific diagnosis of a well-known variant of AD different from RAD (e.g., PCA, CBS, frontal variant) or a well-known variant of a different degenerative dementia (e.g., right temporal variant of FTD or right semantic variant-FTD) made by the original authors. We considered both clinical and pathologically confirmed diagnoses of AD, to conduct a conservative evaluation of the studies. The availability of pathological data confirming AD was recorded when available.
RESULTS
PubMed search results are reported in detail in the Supplementary Material. The flow chart of the search is shown in Fig. 1. We found three different types of studies about atypical AD presentations: 1) case reports and small case series, 2) some early imaging studies on groups of AD patients, and 3) reviews.

Flow chart of the literature search.
1) Considering the case reports and small case series, we found 19 eligible studies reporting in a total of twenty-one AD patients, including one positive case recently studied by our group (i.e., case 1 in Table 1) [19], who presented with a right-sided asymmetric atrophy or hypometabolism/hypoperfusion at the neuroimaging (Table 1). A corresponding right-sided asymmetric clinical syndrome or cognitive phenotype emerged in nineteen cases. The description of the clinical syndrome in the two remaining case-reports (i.e., case 3 and 4 in Table 1) was rather limited. However, some delusions associated with mild memory impairment emerged, probably related to the right-sided brain involvement. None of the cases had received a diagnosis of an AD variant. Instead, several reports were excluded from the present analysis because the right-sided asymmetry of cognitive impairment was not supported by consensual asymmetry at the imaging [20–26]. Other case-reports were excluded because the authors made a diagnosis of PCA syndrome [27] or the right semantic variant of frontotemporal dementia [28, 29].
The Literature analysis. Case-reports of AD found in the literature who presented with a right-sided asymmetric clinico-anatomical syndrome
NR, not reported; FDG, [18F]-fluorodeoxyglucose; PET, positron emission tomography; CSF, cerebrospinal fluid; AD, Alzheimer’s disease; SPECT, single photon emission computed tomography; rCBF, regional cerebral blood flow; FTD, frontotemporal dementia; HMPAO, 99mTc hexamethylpropyleneamine oxime; CBS, corticobasal syndrome; PCA, posterior cortical atrophy; IMP, iodine-123-N-N’, N, -trimethyl-N’-[2-hydroxyl-3-methyl-5-iodo-benzyl]-1, 3 propane diamine; DLB, dementia with Lewy bodies; CMRO, cerebral metabolic rate of oxygen. §clinical syndrome including features of CBS. *clinical syndrome including features of posterior cortical atrophy. °clinical syndrome including features of semantic variant FTD. All patients included in the table were right-handed.
We conducted a re-analysis of the clinical data reported in the eligible reports and found that many features possibly pertaining to some already known atypical syndrome of AD, like CBS (i.e., case number 9, 10, 11, 17, 19, and 21 in Table 1) or posterior cortical atrophy syndrome (i.e., case number 11) were present together with some of the described characteristics of possible RAD. In other two studies (missing a definitive pathological confirmation of AD), the patients showed features of the RAD, but together with some signs and symptoms resembling the clinical syndrome of the right semantic variant of FTD (i.e., case number 16 and 20). Thus, our evaluation was restricted to the most credible evidence of RAD, after exclusion of all these patients potentially affected by a different condition. Interestingly, thirteen reports of possible RAD (i.e., case number 1, 2, 3, 4, 5, 6, 7, 8, 12, 13, 14, 15, and 18 in Table 1) remained. Moreover, three cases among them had a pathologically confirmed AD (i.e., the number 1, 2, and 18).
A preliminary evaluation of the cognitive impairments in the thirteen retained patients revealed that all cases had impairment of recent memory. A dissociation between preserved verbal memory and impaired visual memory emerged in one case (i.e., case 1). Moreover, all patients showed at least one impaired function among those lateralized to the right hemisphere (e.g., visuoconstructional functions, topographical orientation, face-processing, visuospatial attention). In detail, six patients showed deficits involving visuoconstructional functions, whereas five patients showed left unilateral spatial neglect. Topographic disorientation emerged in four patients, and three patients presented with prosopagnosia. Two patients showed visual agnosia for objects, and two presented with impaired visual attention. From morphological and functional imaging data, the right temporal and parietal regions were the most frequently involved in the possible cases of RAD.
Data available about other clinical features are quite scant. However, no significant behavioral disturbance was reported. Instead, early mild anxiety and depression emerged in two case-reports (i.e., case 1 and 14) presenting longitudinal data. Besides, psychotic symptoms seem to emerge only in two cases (i.e., case 3 and 4), which were both early-onset AD patients (i.e., 52 and 58 years respectively). Neurological examinations were all negative. All the eligible RAD cases showed an insidious onset. Also, the four case-reports including quite detailed longitudinal assessment (i.e., 1, 6, 12, 14) showed a gradual worsening of cognitive impairment, but with preservation of a quite unvaried RAD phenotype during the first years. No fluctuations of cognition and attention were reported. Finally, data about functional autonomy were lacking.
It is worth noting that considerable heterogeneity emerged among patients in the stage of cognitive decline when they had been studied. In particular, the mean time from symptoms onset to the first visit was 3.6 years, corresponding to a relatively early stage of dementia. Nonetheless, the range was ample, from 1 to 7 years from onset, so including patients from very early to more advanced stages. Moreover, mean Mini-Mental State Examination (MMSE) score was 21.3 at the time of the first assessment, corresponding to a stage of mild dementia. However, the score range varied from 13 to 28 points. This heterogeneity has limited our possibilities to define the core features of RAD more accurately.
2) Considering the imaging studies conducted in groups of AD patients, we found some preliminary studies proposing the concept of RAD [30–36]. In general, these studies reported that a subgroup of AD patients presenting with preponderant and long-lasting visuospatial impairments shows more pronounced hypometabolism on right hemisphere at PET scan. Interestingly, Fischer et al. [33] hypothesized a specific progression of the cognitive impairments in patients with RAD, first involving material-specific memory loss (i.e., visual-spatial), followed by visual-spatial deficits, then global memory impairment, and finally global cortical deficits (i.e., both visual-spatial and semantic, but more severe for the visual-spatial domain). Many of these first RAD cases seem to show a right-sided asymmetry on cerebral hypometabolism, especially on posterior regions (e.g., parietal, parietal-occipital). Thus, a posterior variant of AD (i.e., a PCA syndrome), that sometimes is asymmetric and presents a more extensive involvement on the right side [8], may represent the most likely diagnosis and potentially exclude/complement the RAD one.
3) Considering the reviews retrieved from PubMed, we found only another study [37] including a possible right variant (e.g., AD patients with right-sided hypometabolism associated with disproportionate impairment in visuospatial abilities reported in a section distinct from PCA). However, a detailed case history has not been described in this article. Many other past and current reviews on the same topic did not explicitly talk about RAD [13,15, 38– 48].
DISCUSSION
The literature analysis supported the occurrence of both early- and late-onset AD cases, who presented with a right clinical-radiological syndrome to some extent similar to the RAD phenotype described by Lam et al. (2013) [3]. The degree of diagnostic certainty of RAD is quite high, considering that some studies documented pathologically confirmed AD patients with such right presentation [3, 50]. Moreover, in some cases, the RAD syndrome remains as focal for a long time [19, 33].
In particular, thirteen RAD cases (three supported by pathological confirmation) emerged from our analysis of the literature. Preliminary scrutiny of their cognitive phenotypes revealed that memory and visuospatial impairments seem to be shared. In particular, mainly an impairment of recent memory, sometimes with a dissociation between impaired visual memory and unimpaired verbal memory, associated with constructional deficits or impairments of other different functions lateralized to the right hemisphere (e.g., left unilateral spatial neglect, topographic disorientation, and prosopagnosia) emerged. Consensually, right-sided asymmetry was more frequently found in temporal and parietal regions at morphological and functional imaging. Data available about other clinical features were quite scant. However, except from early anxiety and depression reported in two cases, RAD syndrome seemed not to be present with significant behavioral disturbances at an early and intermediate stage. It is interesting to note that psychotic symptoms emerged only in the two early-onset RAD cases. Neurological examination was usually negative. Most of the cases seemed to show insidious onset and gradual progression. However, few studies reported a detailed longitudinal assessment. Finally, no fluctuation of cognition or attention was reported.
Unfortunately, it has not been possible to delineate any more detailed clinical or clinical-radiological syndrome of RAD from the available data. The eligible case-studies of RAD found in the literature are still few. Furthermore, considerable heterogeneity emerged among them in the stage of cognitive decline when the patients had been studied. Also, the general cognitive impairment as assessed by MMSE score was quite variable. This heterogeneity is an obstacle to the aim of recognizing a set of core features of a new supposed AD phenotype. If studied too early, patients may not have yet developed all the core symptoms of an atypical syndrome; if too late, they could present deficits in different cognitive domains as well as the core ones. Another limit was that some terms used in the literature search (e.g., visuospatial) were quite general and might not necessarily or exclusively denote right-hemisphere cognitive functions. For instance, several studies showed that regions in both the left and right hemisphere play a role in constructional tasks [69]. A further limitation was that right-sided asymmetry at imaging had been not corroborated by quantitative analyses in all the reviewed studies. At the same time, only nine of thirteen studies presented the results from a detailed neuropsychological assessment supporting the right-sided asymmetric cognitive phenotype by quantitative data (i.e., case 1, 5, 6, 7, 8, 12, 13, 14, and 15). Moreover, the tests used were different among the studies. Finally, the lack of pathological confirmation of AD was another critical limitation. Indeed, only two of the selected patients had pathological data, and one had positive biomarkers for AD. Whereas, all the remaining patients considered as RAD had not pathological information. Consequently, RAD phenotype remains quite approximate. Moreover, the relation between it and the other phenotypes of AD presenting the right involvement is to be clarified. For example, even if atrophy is usually more significant on the left more than right hippocampus in typical AD [51], some data showed more extensive involvement of the right side [52]. Moreover, many studies found that PCA syndrome is sometimes asymmetric, with a preferential right side involvement, although some authors have recently challenged this observation suggesting the possibility of selection bias in the reported literature [7]. Furthermore, CBS (another focal variant of AD) [12, 53] is frequently asymmetric, with a preponderant right-sided brain involvement [12, 54]. According to this last finding, we found some case reports in the literature, which presented many features of CBS overlapped with the clinical features of RAD (Table 1). However, we excluded them from the analysis since potentially affected by a different condition, and cases of RAD remained, which did not present core signs of CBS. The overlap between RAD and PCA was more difficult to disentangle. In particular, many RAD patients selected in the study showed at least one or more posterior cognitive features (e.g., especially constructional dyspraxia and topographical disorientation) which are included in the diagnostic criteria of PCA clinical syndrome [8, 9]. At the same time, parietal regions are targeted in both the RAD and PCA radiological syndromes. However, no RAD patient seemed present with three or more cognitive features of PCA syndrome simultaneously, as requested in diagnostic criteria. Besides, significant impairment of recent memory associated with temporal involvement seemed to be a common feature in many RAD cases. On the contrary, PCA diagnostic criteria include relatively spared anterograde memory function and brain damage centered in the posterior regions (i.e., occipital-parietal and occipital-temporal). Finally, the right-sided asymmetry is the core feature of the RAD clinical-radiological syndrome and was shared by all the patients in the study. Instead, even if PCA syndrome can be asymmetric in some cases, more often just right-sided, this asymmetry is not considered as a core diagnostic feature of the syndrome. On the whole, there seems to be a core difference between RAD and PCA syndrome. Indeed, the hub of the RAD clinical-radiological syndrome seems to be on the right hemisphere, especially temporal-parietal. Vice-versa, the hub of PCA clinical-radiological syndrome is on posterior regions, more often bilaterally, especially occipital-parietal or occipital-temporal.
Another limitation of this study was that our search of the literature has probably missed some RAD cases.
Firstly, we expected to find more studies reporting eligible RAD patients in AD literature by setting the right-sided asymmetry at imaging, rather than asymmetry in cognitive phenotype, as a primary inclusion criterion. It is necessary to perform detailed neuropsychological evaluations, including tests for uncommon impairments as prosopagnosia, unilateral spatial neglect as well as topographical disorientation, to delineate a right-sided cognitive phenotype effectively. However, these tests are often not included in the standard test batteries performed routinely in patients with mild cognitive impairment and/or dementia. On the contrary, a simple CT or MRI scan, as performed routinely in clinical settings, can be enough to highlight, or at least reliably suggest a right-sided radiological asymmetry. However, the choice of prioritizing the right-sided radiological syndrome against the clinical one has probably limited the number of RAD eligible cases retrieved. In particular, we had to exclude some case studies reporting probable RAD patients with a well-documented right-sided cognitive phenotype, but not supported by a consensual right-sided radiological asymmetry at imaging [23, 26].
Secondly, to restrict our evaluation to the most persuasive evidence of RAD, we excluded all the patients potentially affected by a different condition. So, we excluded some cases when many signs of a well-known right-sided syndrome different from RAD (e.g., right PCA, right CBS and the right semantic variant of FTD) overlapped with signs of RAD, or the authors of the original studies diagnosed these such syndromes. In doing so, we have probably missed further possible RAD cases. In any case, it is challenging to change a diagnosis made by another clinician by only examining data presented in sometimes very scant case reports, and without performing a further evaluation on the same patient. Equally, it is challenging to bring into question a diagnosis of a well-established clinical entity as PCA, when the alternative diagnosis proposed is not a well-established and recognized clinical entity yet, just like RAD is.
In sum, the failure to provide a definitive and detailed description of focal RAD in our study was not due to negative findings, but to the limited details and relative heterogeneity of data available. Moreover, although systematic, our search may not be exhaustive and have potentially missed some reports of RAD in the literature.
Considering the limitations of current research on RAD, it will be essential to plan longitudinal studies with detailed descriptions of the patient’s clinical and cognitive phenotype, starting from the early stages of dementia (e.g., mild cognitive impairment and mild dementia stage) in the future. Besides, the clinical descriptions are to be supported by extensive neuropsychological evaluations including tests of cognitive functions lateralized to the right hemisphere (e.g., face-processing, topographical orientation, visuospatial attention) as well as tests of both verbal and visual-spatial memory. Also, imaging studies with quantitative analysis of atrophy will be needed to support the right-sided asymmetry of the clinical-radiological syndrome. Moreover, considering the possible overlap between RAD and other right syndromes of different neurodegenerative diseases (e.g., right semantic variant-FTD, right CBS-CBD), it will be crucial to have confirmation of AD by biomarkers in cerebrospinal fluid, amyloid PET-scan or autopsy. Finally, when the distinctive features of RAD will be more clearly recognized, it will be essential to re-analyze all those cases reported in the literature which showed a progressive neurodegenerative right-sided syndrome, included those cases already diagnosed with a different syndrome (e.g., right PCA, right CBS), to verify if RAD could be a more appropriate diagnosis.
CONCLUSION
We believe that, even if clear evidence of a RAD is still limited, some sporadic observations seem to support this concept preliminarily. Research in this area merits to be continued. The finding of a new unrecognized variant of AD would be, in fact, significant not only for clinical diagnosis but also for a better knowledge of phenotypic diversity in dementia. The comprehension of this last phenomenon may be essential for improving our understanding of the complex AD syndrome [18].
DISCLOSURE STATEMENT
Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/19-0338r2).
