Translation, cross-cultural adaptation and applicability of the Brazilian version of the Frontotemporal Dementia Rating Scale (FTD-FRS)

BACKGROUND Staging scales for dementia have been devised for grading Alzheimer's disease (AD) but do not include the specific symptoms of frontotemporal lobar degeneration (FTLD). OBJECTIVE To translate and adapt the Frontotemporal Dementia Rating Scale (FTD-FRS) to Brazilian Portuguese. METHODS The cross-cultural adaptation process consisted of the following steps: translation, back-translation (prepared by independent translators), discussion with specialists, and development of a final version after minor adjustments. A pilot application was carried out with 12 patients diagnosed with bvFTD and 11 with AD, matched for disease severity (CDR=1.0). The evaluation protocol included: Addenbrooke's Cognitive Examination-Revised (ACE-R), Mini-Mental State Examination (MMSE), Executive Interview (EXIT-25), Neuropsychiatric Inventory (NPI), Frontotemporal Dementia Rating Scale (FTD-FRS) and Clinical Dementia Rating scale (CDR). RESULTS The Brazilian version of the FTD-FRS seemed appropriate for use in this country. Preliminary results revealed greater levels of disability in bvFTD than in AD patients (bvFTD: 25% mild, 50% moderate and 25% severe; AD: 36.36% mild, 63.64% moderate). It appears that the CDR underrates disease severity in bvFTD since a relevant proportion of patients rated as having mild dementia (CDR=1.0) in fact had moderate or severe levels of disability according to the FTD-FRS. CONCLUSION The Brazilian version of the FTD-FRS seems suitable to aid staging and determining disease progression.


INTRODUCTION
T he term Frontotemporal Lobar Degeneration (FTLD) was first introduced in 1998 by a group of Swedish and English researchers, 1 who used it to describe a clinical syndrome characterized by progressive behavioral changes associated with atrophy of the frontal lobes and of the anterior portions of the temporal lobes. The term was introduced in order to replace terminology such as "frontal lobe degeneration of non-Alzheimer type" and "dementia of frontal lobe type". 1 Three main conditions are described in the FTLD group: frontotemporal dementia (FTD) or behavioral variant frontotemporal dementia (bvFTD), 2,3 semantic dementia (SD), 4 and progressive non-fluent aphasia (PNFA). [4][5][6] Recent studies have suggested that FTLD-related diseases have a significant impact on the ability to carry out daily activities. However, studies on disability severity in these conditions are scarce. In addition, disease staging in FTLD remains a challenge as most dementia staging tools have been developed for Alzheimer's disease (AD). For instance, the Clinical Dementia Rating, 7 and other similar instruments may not capture the functional changes that are specific to FTLD. A recently developed scale specifically designed to examine the behavioral and functional changes associated with FTLD, the Frontotemporal Dementia Rating Scale (FTD-FRS), has been found to be helpful for assessing severity and the rate of functional decline. 8 In the validation study of the FTD-FRS, 8 by crosssectional analyses involving a sample with three FTLD variants (bvFTD, n=29; SD, n=20; PNFA, n=28), the authors were able to identify six levels of disease severity (very mild, mild, moderate, severe, very severe and advanced/profound) with the use of the FTD-FRS. There was greater severity of functional impairment in bvFTD than in language variants, and limited correlation with cognitive measures. Follow-up analyses of a sub-sample carried out using the FRS after 12 months revealed that patients with bvFTD advanced more rapidly through the severity stages than the other variants. Therefore, the FTD-FRS was able to distinguish the functional profile of FTLD variants and identify differential rates of decline.
In Brazil, no studies investigating FTLD staging have yet been conducted and validated tools for this purpose are lacking. Therefore, the primary aim of the present study was to translate the FTD-FRS to Brazilian Portuguese and adapt it to the Brazilian cultural context.

METHODS
The translation and cross-cultural adaptation processes consisted of the following steps: translation, back-translation (prepared by independent translators), evaluation of the back-translated version against the original version, discussion of the Portuguese version of the FTD-FRS with specialists, development of a final version after minor adjustments, and pilot application in patients with diagnoses of bvFTD and AD. The original instrument, translation, back-translation and the final version of the FTD-FRS are given in Table 1 and Appendix A. Table 2 shows percentage scores and logarithmic score conversion for the FTD-FRS correction.
Participants. For this stage of the study it was decided to include in the research sample only patients with bvFTD. Additionally, this variant of FTLD presents features discussed in the scale (disorders of behavior and impact on activities of daily living) that could help in the detection of its applicability in Brazil.
The study sample consisted of 23 individuals aged 45 or older, with at least two years of formal education -12 had been diagnosed with bvFTD and 11 with AD. Patients were matched for disease severity (CDR=1.0). This study was conducted from February 2011 to July in 2013.
Dementia was diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders -DSMIV criteria. 9 For the bvFTD diagnosis, the international consensus criteria were used. 2 AD diagnosis followed the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association -NINCDS-ADRDA criteria for probable AD dementia. 10 The exclusion criteria were as follows: CDR>1, visual, hearing or motor impairments which could hinder comprehension of instructions and execution of cognitive tasks, uncontrolled clinical conditions, severe psychiatric disorders, and significant cerebrovascular disease on neuroimaging.
Evaluation procedures. The evaluation protocol included: sociodemographic and clinical questionnaires; Addenbrooke's Cognitive Examination-Revised (ACE-R) Mini-Mental State Examination (MMSE); Executive Interview (EXIT-25).The protocol for caregivers included the Cornell Scale for Depression in Dementia, Disability Assessment for Dementia (DAD), Neuropsychiatric Inventory (NPI), the Frontotemporal Dementia Rating Scale (FRS) and Clinical Dementia Rating scale (CDR).
The ACE-R and the EXIT-25 were applied to assess cognitive performance. The ACE-R consists of a brief cognitive assessment battery testing five different cog- Tem problemas para tomar suas medicações no horário correto (esquece ou se recusa a tomá-las) Has problems taking their medications at the right time (forgets or refuses to take them) (esquece ou se recusa a tomá-las) Tem problemas para tomar suas medicações no horário correto (esquece ou se recusa a tomá-las) 18 Has difficulties taking his/her medications as prescribed (according to the right dosage) Tem dificuldade para tomar suas medicações como foram prescritas (na dosagem correta) Has difficulties taking their medications in the manner prescribed (at the right dose) Tem dificuldade para tomar suas medicações como foram prescritas (na dosagem correta)

Meal Preparation and Eating
Preparo de refeições e alimentação  (14); language (28); and visuo-spatial abilities (5). Higher scores indicate better performance. The scores regarding each of the six domains can be computed separately and their sum generates the total ACE-R score of which 30 points corresponds to the MMSE. 11,12 The EXIT-25 assesses different aspects of executive function. It consists of 25 sub-items with scores ranging from 0 to 2, with total score ranging from 0 to 50, and lower scores indicating better performance. It assesses verbal fluency, design fluency, anomalous sentence repetition, and interference, among others. Studies have suggested that a score higher than 15 is consistent with dementia. 13,14 For dementia staging, the CDR was completed. It evaluates six domains related to cognitive and functional performance: memory, orientation, judgment and problem solving, community affairs, home and hobbies, First. make sure that all not applicable (N/A) questions are excluded from the final score. E.g.
if the patient does not take any medication then maximum score is 28 (not 30). Divide the number of "never" questions by the number of maximum applicabe questions. This percentage score should be checked against this and personal care. 7,15 A pre-defined algorithm allows the calculation of a total score, with 0 indicating preserved performance and higher scores indicating increased impairment. 7 The Neuropsychiatric Inventory (NPI) in its short version is a 10-item questionnaire that makes it possible to determine the presence of neuropsychiatric and behavioral symptoms, their frequency and severity. Scores range from 0 to 144. Each behavior has a maximum score of 12 points, calculated by multiplying symptom frequency by its severity. The assessed behaviors are: delusions, hallucinations, agitation and aggression, dysphoria, anxiety, euphoria, apathy, disinhibition, irritability/lability, aberrant motor activity, nighttime behaviors, and changes in appetite. The higher the score, the greater the severity and frequency of these behaviors. 18,19 The FTD-FRS was developed based on questions from the Cambridge Behavioral Inventory (CBI) 20 and the Disability Assessment for Dementia (DAD). 21 It is a 30-item questionnaire that assesses: Behavior, Outing and Shopping, Household Chores, Telephone, Finances and Correspondence, Medications, Meal Preparation, Eating, Self-care and Mobility. It was developed with the purpose of assessing disease severity and progression in FTLD. 8 The response options for each question are: all the time=0; sometimes=0 and never =1. The examiner must add the number of alternatives marked as "never" and then divide by the number of questions answered. This will generate a percentage (an index of functional preservation) which takes into account the pre-morbid state of the patient (as the tasks which were never per- formed are not considered in the score). After calculating the percentage of preservation the score should be converted to a logarithm (  Table 3 shows the sociodemographic characteristics of participants. It can be noted that the groups were homogeneous with regards to gender, age and education. On the MMSE and the EXIT-25 there was a significant difference among the three groups, with the AD group exhibiting worst performance. Preliminary results for the FTD-FRS revealed greater levels of disability in bvFTD than in AD patients (bvFTD: 25% mild, 50% moderate and 25% severe; AD: 36.36% mild, 63.64% moderate), in spite of having similar CDR ratings (see Table 3 and Figure 1).

DISCUSSION
In this report, we present a culturally adapted, translated version of the FTD-FRS in Brazilian Portuguese. Confrontation between original and back-translated scales, and the preliminary staging results achieved in bvFTD patients suggest that our version is suitable for clinical purposes.
Results from the scale's pilot application are in line with those from the validation study, 8 as FTD-FRS seemed to be capable of capturing functional and behavioral change not identified by the CDR. All participants had a score on the CDR=1, and yet, according to the FTD-FRS, 25% of bvFTD patients were severely impaired. Also, in agreement with previous studies, 20,21 our findings suggest that bvFTD is associated with greater functional loss and behavioral change compared to AD.
Determining disease severity in dementia, and especially in less prevalent sub-types, remains a controversial issue. There is currently a lack of consensus regarding the definition of severity in dementia and its ideal staging tools. 8,15,22 Our study suggested that severity in bvFTD needs to be measured with a tool specifically designed to detect its early symptoms. Cognitive-based staging strategies are limited, since they are heavily dependent on language skills, which might overestimate disease severity, as observed in primary progressive aphasias. 23 Additionally, in developing countries, cutoff scores in cognitive tests are unsuitable for dementia staging because of great variability in educational background. The FTD-FRS may provide a better understanding of disease progression in FTD, by showing which abilities are lost early and late in the disease, as it relies on collateral information. Also, in patients with AD, the  scale showed sensitivity in detecting severity of dementia, where a great proportion of patients with a low CDR 1 had in fact moderate severity on the FTD-FRS (64%). The Brazilian version of the FDT-FRS seems suitable to aid staging and determining disease progression. This study had some potential limitations. The dementia groups consisted of patients currently attending our clinics, which excludes more impaired patients living in nursing homes. We were unable to include neuropathology, which is ideally needed to confirm a definitive diagnosis. Additionally, the analyses were cross-sectional, restricting some of our interpretations. As to the strengths of the study, we may cite the fact that the sample was homogeneous as only early dementia cases were included (CDR=1).
Our preliminary results suggest that the Brazilian version of the FTD-FRS is appropriate for clinical use, as it was easily understood by caregivers and family members. In addition, results are in line with previous studies using the scale, as they suggested greater functional and behavioral changes among bvFTD patients. Future studies should continue to examine the psychometric characteristics of this instrument as it may play an important role in the early diagnosis of FTLD.