The study took place at Department of Clinical Veterinary and Animal Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark
Corresponding author: M. Berendt, Department of Veterinary Clinical and Animal Sciences (DVCAS), Faculty of Health and Medical Sciences, University of Copenhagen (UC), Dyrlægevej 16, Frederiksberg C 1870, Denmark; e-mail: firstname.lastname@example.org.
Canine cognitive dysfunction (CCD) is a neurodegenerative condition affecting geriatric dogs and sharing several characteristics with human Alzheimer's disease (AD). CCD manifests as alterations of behavioral patterns and daily routines. Clinical signs are associated with neurodegenerative changes (eg, cortical atrophy and amyloid-beta deposits).
To investigate clinical characteristics, survival, and risk factors with CCD. Vitamin E was investigated as a potential marker of CCD.
Ninety-four dogs >8 years of age were investigated with a validated CCD questionnaire and allocated to CCD, borderline CCD (b-CCD) and non-CCD groups. The dogs were included in 2008–2009 and followed up in an observational study until follow-up in 2012.
Four key clinical signs dominated in dogs with CCD: sleeping during the day and restless at night, decreased interaction, disorientation at home, and anxiety. A number of borderline CCD cases developed into CCD over time indicating that a prodromal stage of CCD may exist. CCD did not influence survival negatively. Small breeds did not show better survival than large breeds (P = .055) and there was no difference between sexes (P = .99).
Conclusions and Clinical Importance
A few key questions addressing sleep-wake cycle, interaction, and signs of confusion and anxiety can be used as a clinical marker of CCD. Special attention should be paid to anxiety in dogs with CCD because it may be especially stressful to both dog and owner. Dogs with CCD seem to have a good chance of living a full lifespan if supported by the veterinarian and the owner.
Canine cognitive dysfunction (CCD), also called canine cognitive dysfunction syndrome, is a neurodegenerative condition affecting geriatric dogs.[1-3] Prevalence estimates in dogs >8 years of age have ranged from 14% to 60% with increasing age.[3-6]
Canine cognitive dysfunction shares many similarities with human Alzheimer's disease (AD). As in AD, the clinical picture is dominated by a variety of changes in behavioral patterns and daily routines, and clinical signs may worsen over time.[2, 5, 7-9] Dogs naturally accumulate amyloid beta (Aβ) in the brain with an identical amino acid sequence to human Aβ.[10, 11] The extent and location of Aβ deposits in the canine brain have been found to correlate with the severity of cognitive deficits assessed by owner-based questionnaires or laboratory tests, including different visual discrimination tasks.[12-15] Cortical atrophy and an enlarged ventricular system are associated with AD and CCD and have been demonstrated at necropsy and with magnetic resonance imaging (MRI), but can also be found in individuals with normal aging.[14, 16, 17]
Oxidative stress is believed to play a major role in the pathogenesis of CCD and AD, especially because the brain is particularly vulnerable to free radical damage.[18, 19] A decreased concentration of vitamin E has been identified in brain tissue from dogs with CCD and in blood of AD patients.[19, 20]
A number of human geriatric patients experience some of the clinical signs seen with AD, but without qualifying for this diagnosis. This intermediate stage between the expected cognitive decline of normal aging and a more pronounced decline of cognitive function is recognized as mild cognitive impairment (MCI).[21-23] MCI increases the risk of later developing dementia, including AD.[22-24] It is unknown if a similar condition exists in dogs, but so-called predementia or query cases have been reported.[6, 25]
The current knowledge of CCD mainly originates from questionnaire studies investigating signs of CCD by telephone interviews or by internet-based questionnaires, and has consequently rarely included primary investigator-owner contact and clinical evaluation.[4-6, 13, 14, 25] One study has investigated dogs with CCD clinically with an observational period of 3 months. There is a need for long-term clinical studies investigating CCD and from which information about the natural history of this condition (ie, clinical manifestations and prognosis) can be extracted.
The aim of this study was to investigate the clinical characteristics of CCD (including a possible prodromal stage of CCD) and survival and risk factors for CCD. In addition, we investigated vitamin E (alpha-tocopherol) as a possible blood marker for CCD.
Materials and Methods
The study was designed as an observational clinical study with long-term follow-up and was carried out at the Department of Veterinary Clinical and Animal Sciences (DVCAS), University of Copenhagen (UC), Denmark. The study population was recruited from the Small Animal University Hospital Community Practice in a 12-month period in 2008–2009. Owners of dogs >8 years of age visiting the clinic for regular vaccination consults and minor health complaints were asked to volunteer to participate in the investigation. The study was conducted in 2 phases. In phase I, dogs were investigated with a structured CCD questionnaire interview and clinical evaluation. In phase II, dogs were monitored until a final telephone questionnaire follow-up was conducted in July 2012.
Definition of Study Groups
Owners with dogs >8 years of age visiting the community practice for regular vaccination consults and minor health complaints were invited to participate in an oral interview with the purpose of investigating cognitive function. The study aimed at including a total of 100 dogs. A validated CCD questionnaire, previously developed by Rofina et al and named questionnaire C in the publication, was used to interview the owners. This questionnaire addresses alterations of behavior related to disorientation, socioenvironmental interaction, sleep-wake cycle, house soiling, drinking, appetite, and aggressiveness. The lowest total score that can be obtained is 10, which corresponds to a normal cognitive status (non-CCD) and no signs of CCD. The highest total score that can be obtained is 41. The increase in the total score from 10 to 41 indicates the severity of CCD.
We also investigated signs of anxiety and abilities associated with learning and memory, because such signs are common in MCI and Alzheimer's patients.[26, 27] Six additional standardized questions were posed by the investigators to address these categories. The questions from the validated questionnaire and the additional questions are presented in Table 1.
Table 1. Distribution of clinical signs displayed by dogs in the b-CCD group and the CCD group
Separation anxiety started after the age of 8 years
Irrational reaction/fear to well-known objects
Trying to escape specific well-known situations, seems frightened
Reduced ability to learn new tasks
Reduced ability to perform known tasks
Reduced ability to adjust to changes
To secure a uniform interview, the primary investigator (RF) performed the structured interviews. Answers were recorded for each dog in a separate file. Each interview lasted for a minimum of 20 minutes.
Based on the score obtained in the validated questionnaire, dogs were allocated into 3 groups; dogs with a score = 10 (the non-CCD group), dogs with a low CCD score ranging from 11 to 15 (the borderline CCD [b-CCD], and the CCD group with a total score >15). The motivation for investigating a b-CCD group separately was prompted by human geriatric medicine, which employs the term MCI. People identified with MCI have an increased risk of later developing dementia, especially of the AD type, and it is of interest to identify such patients, because they can be targeted for early therapeutic intervention.[21, 24, 28] The cutoff of 15 for the b-CCD group was chosen, because dogs with CCD scores from 11 to 15 were characterized by signs appearing in the questionnaire, which also may appear with natural aging (eg, increased sleeping).
To be included in the CCD group, it was mandatory for owners to allow a clinical and neurological examination, standard hematological and serum biochemistry profiles, and a thyroid panel including T4 and TSH results. A urinalysis was also performed if indicated. In addition, owners were offered an MRI scan of their dog's brain to document possible cortical atrophy that may support a diagnosis of CCD and exclude other intracranial pathology. This procedure was optional attributable to professional and ethical reasons because anesthesia was required. The owners, therefore, were given the possibility to decline MRI. However, MRI was mandatory if the dog showed neurological signs indicating a lesion localized to the brain. Diagnostic imaging was evaluated by a board-certified veterinary radiologist.
To be included in the b-CCD group and the non-CCD group, it was mandatory for owners to allow a clinical examination, whereas a neurological examination and standard hematological and serum biochemistry profiles and a thyroid panel including T4 and TSH could not be performed for professional and ethical reasons without clinical indication and the owner's consent.
Based on the clinical evaluation, dogs were excluded from the study if abnormalities were detected on clinical and on diagnostic tests that suggested concurrent systemic or brain disease that may mimic signs of CCD.
Description of Study Groups
Owners of 94 privately owned geriatric dogs of various breeds, >8 years of age, were interviewed in 2008–2009. Based on the results of the validated questionnaire, 23 dogs had a total score of 10 and represented the non-CCD group. Twenty-seven dogs had a total score between 11 and 15 (mean, 12.1; range, 11–15; SD, 1.2) and represented the b-CCD group and 44 dogs had a total score >15 (mean, 20.4; 16–26; SD, 3.6) and represented the CCD group (Fig 1).
All dogs from the CCD group had clinical and neurological examinations and standard blood profiles including a thyroid profile. Fifteen owners allowed their dog to have an MRI scan of the brain. No abnormalities other than suspected cortical atrophy were detected in these dogs.
In the b-CCD group and the non-CCD group, all dogs had clinical examinations. In the b-CCD group, 21 dogs also had neurological examinations and 22 dogs had hematology and blood biochemistry. In the non-CCD group, 15 dogs also had neurological examinations and 19 dogs had hematology and blood biochemistry performed.
Based on the clinical and other diagnostic tests, 7 dogs were excluded from the study because of concurrent medical problems that could possibly mimic signs of cognitive impairment, including severe cardiac, liver, or kidney disease and hypothyroidism.
Finally, 87 dogs were included in the study population, 37 dogs in the CCD group, 27 dogs in the b-CCD group, and 23 dogs in the non-CCD group.
The CCD group was comprised of 24 different breeds and 4 mixed breeds. The age range was 9–17 years (mean, 12.5; SD, 1.9). The b-CCD group was comprised of 16 different breeds and 5 mixed breeds. The age range was 8–14 years (mean, 12.3; SD, 1.8). The non-CCD group was comprised of 16 different breeds and 4 mixed breed dogs. The age range was 9–13 years (mean, 11.6; SD, 1.5).
The antioxidant vitamin E (alpha-tocopherol) was measured in serum for a sample of dogs from each of the 3 groups. Serum vitamin E (alpha-tocopherol) concentration was only measured in dogs if the owner gave her or his consent. A standard CBC including leukocyte differential count was performed on an ADVIA 120 analyzer.1 Serum was used for biochemical analysis on an ADVIA 16502 including ion-selective electrodes and a thyroid profile (free T4, cTSH) was analyzed on an Immulite 2000.1 The hematological, biochemical, and thyroid panels all were analyzed at the Central Laboratory,3 whereas the vitamin E (alpha-tocopherol) assay was performed at the Vet Med Lab, Ludwigsburg, Germany, by ultraperformance liquid chromatography method with diode array detection (UPLC-DAD).
The dogs from the 3 groups were followed from inclusion and clinical evaluation in 2008–2009 until a final telephone follow-up in July 2012. Dog owners could contact the primary investigator for advice during the study, and dogs were monitored by control visits if new signs of CCD or additional clinical signs developed during the study. Owners of dogs from the CCD group were advised regarding supportive care such as medical treatment with selegeline hydrochloride, specific diets intended to promote cognitive health, and cognitive stimulating toys. At follow-up in 2012, the owners of all dogs were contacted by phone by the primary investigator (RF) and a coinvestigator (TS). The status of the dogs, alive or dead and cause of death, was recorded and the questionnaire and additional questions in 2008–2009 also were repeated.
The answers given by the owners were analyzed with respect to clinical signs and CCD score and compared with the information obtained in 2008–2009, and a survival analysis was performed.
The local ethics committee at DVCAS, UC, Denmark, approved the protocol for the study. Animals were cared for according to the principles outlined in the NIH Guide for the care and use of animals.
Descriptive analyses of the results obtained from the questionnaires and the additional questions addressing anxiety and learning-memory in phase I and II were carried out stratified by cognitive status.
Survival functions of the CCD, b-CCD, and non-CCD dogs were calculated by Kaplan–Meier, and differences in survival were tested by the log-rank test.
Univariable analyses of the association between cognitive status, sex, and breed (small or large) were examined by Fischer's exact test.
The potential association between vitamin E (alpha-tocopherol) concentration and cognitive status was evaluated by ANOVA. For all the statistical analyses, a P-value <.05 was considered statistically significant.
Analysis of Questionnaires and Additional Questions
The most frequently reported sign for the CCD group was “sleep during the day, restless at night” (57%). The 2nd most frequent sign was decreased interaction (51%) followed by disoriented at home (49%) and anxiety (46%).
The dogs included in 2008–2009 in the CCD group had a moderate CCD score (mean score, 20.4; range, 16–26; SD, 3.6).
For the b-CCD group, the predominant sign was increased daytime sleeping, reported for 70% of the dogs. Signs related to anxiety were reported for 11%.
In the non-CCD group, signs of anxiety were only reported in 4% of the dogs.
The distribution of all signs recorded for dogs in the b-CCD and CCD group is presented in Table 1.
Analysis of Risk Factors
The distribution of CCD, b-CCD, and non-CCD dogs with regard to sex (male-female, neutered-spayed) or size of breed is given in Table 2 with the P-value of the Fischer's exact test. In the univariable analyses, no variable was found to be significant.
Table 2. Association between sex (intact, neutered/spayed), size of breed, and cognitive status. P-value calculated by Fischer's exact test
The vitamin E (alpha-tocopherol) concentration was measured in serum for all dogs in the CCD group (mean, 35.6 mg/L; range, 7.4–97.3; SD, 20.6), in 17 dogs in the b-CCD group (mean, 38.4 mg/L; range, 18.5–66.5; SD, 16.4), and in 12 dogs in the non-CCD group (mean, 32.6 mg/L; range, 3.2–48.1; SD, 14.4). No significant difference was found in the vitamin E (alpha-tocopherol) concentration between the groups (P = .72).
Eighty-two dogs participated in the follow-up investigation, whereas 5 dogs were lost to follow-up. At follow-up, 74 dogs were dead and 8 dogs were alive.
Only 6 owners reported CCD to be the primary cause for euthanasia. Median survival time for the non-CCD group was 12.6 years (range, 9.7–15.0), for the b-CCD group 13.5 years (range, 9.3–17.0), and for the CCD group 13.9 years (range, 10.0–17.0).
When analyzing the development of CCD score for the individual dogs, we found that 11 dogs (58%) from the non-CCD group moved to the b-CCD group and 3 dogs (14%) moved from the b-CCD group to the CCD group from inclusion to follow-up.
None of the dogs from the non-CCD group moved into the CCD group during the study.
In the CCD group, 8 dogs (27%) developed higher scores with time. Nineteen dogs (63%) had a lower CCD score at follow-up than at inclusion, but remained in the CCD group.
The log-rank test gave a P = .0155, showing a significant difference in survival between the groups (Fig 2). From the graph, it is apparent that the difference lies in the non-CCD (control) group having generally shorter survival times, whereas the survival functions for CCD and b-CCD dogs overlapped.
Similar analyses for breed and sex (divided into male/female regardless of neuter status) showed that there was no difference in survival between small and large breed dogs (P = .055) and no difference between sexes (P = .99).
The overall aim of this study was to investigate the characteristics of CCD, including survival and risk factors. We followed a cohort of geriatric dogs in an observational clinical study with long-term follow-up. We found that 4 key clinical signs characterize dogs with CCD and that a prodromal stage of CCD possibly precedes CCD. In this study, a diagnosis of CCD did not affect survival, and our results suggest that dogs with CCD may benefit from veterinary support and owner counseling.
Owner interviews with questionnaires are an important tool when assessing patients suspected to have CCD.[13, 14, 25, 29, 30] However, it is equally important to exclude other systemic or brain diseases that could account for the signs displayed by the patient and mimic CCD before reaching a diagnostic conclusion. Most studies that have investigated CCD have not included a clinical evaluation and thus may suffer from difficulty associated with case definition.[4, 8, 13, 25, 31] In this study, all dogs in the CCD group were investigated with questionnaires and a full diagnostic evaluation was performed to establish the diagnosis and strengthen case ascertainment. For professional and ethical reasons, it was, however, only possible to perform MRI in 40% of the dogs included in the CCD group. Therefore, we cannot exclude that intracranial disease (other than CCD) has been present in some of the cases included in this group. Clinical, as opposed to experimental, studies are associated with certain limitations. One is that investigations including risks (such as anesthesia) cannot be performed unless justified diagnostically, and even when justified, the owner must still agree to such a procedure. All of the dogs included in the CCD group had a full diagnostic evaluation with extensive questionnaire owner interviews, clinical and neurological examinations, hematology and blood biochemistry, and thyroid profile. None of the dogs showed signs of disease other than the signs indicating CCD at inclusion and neither did they develop such signs until death or follow-up after 3 years. We, therefore, believe that other intracranial or severe systemic disease is highly unlikely in the dogs included in the CCD group.
In the remaining 2 groups, 6 dogs from the b-CCD group and 8 dogs from the non-CCD group did not have a neurological examination at inclusion, and not all dogs had standard blood profiles. A full diagnostic evaluation was, in some cases, not considered appropriate because of ethical or professional reasons and in some cases, the owners declined. This is a limitation of the study and may have influenced our case definition for the b-CCD group and the non-CCD group as it is possible that some of these dogs may have suffered from systemic disease. However, the fact that none of the dogs developed any signs of neurological or other disease known to mimic CCD from inclusion to death or follow-up supports the case definition in the non-CCD and b-CCD groups.
A study of Salvin and coworkers (2010) estimated the prevalence of CCD to 14.2% in dogs >8 years of age, but also found that only 1.9% of the dogs had a previous CCD diagnosis from a veterinarian. The study highlighted the importance of identifying dogs with CCD, especially if they can benefit from supportive initiatives.[3, 32-34]
We wished to investigate if specific signs are more prevalent in CCD than others, because signs may serve as clinical markers and alert the clinician to investigate the patient further. Previous studies have reported signs of CCD, but results were reported according to behavioral category rather than specific signs.[3, 4, 8, 31] Our study identified 4 key signs of CCD—sleeping during the day and restless at night, altered interaction, signs of disorientation at home, and anxiety. The only prominent sign characterizing the b-CCD group was increased daytime sleeping.
A disturbed sleep-wake cycle has previously been reported in geriatric dogs and dogs with CCD.[3, 32, 35] Sleep-wake cycle disturbances also are a well-known feature of AD and believed to reflect a disruption of the circadian rhythm. Aimless behavior characterized by wandering, delusions, or hallucinations also are common in AD patients and increase with cognitive decline. In dogs, sleep-wake cycle disturbances, disorientation, social interaction, and house soiling have been linked to learning and memory, and Aβ deposits in the cortex and hippocampus have been related to the loss of these capacities.[13, 14]
Almost half of the CCD group (46%) expressed signs of anxiety, whereas this was not a dominant sign in the b-CCD group and the non-CCD group. Anxiety in CCD previously has only been sparsely documented. Our results stress the importance of addressing anxiety, related behavior in dogs with CCD, and considering if anxiolytics could be beneficial. Owners should be informed that even small changes in daily routines can have a great impact on the dog affected with CCD, as increased sensitivity to changes, altered response to stimuli, and a decreased capacity to adapt also will influence anxiety. Anxiety is a well-known problem in AD, where 20–60% of patients experience such problems, and anxiety has been reported to be a possible marker of early stage AD.[26, 39, 40] A relationship between anxiety and nighttime behavioral disturbances, with AD patients awake at night having higher levels of anxiety, also has been reported.
Decreased interaction and signs of disorientation were other predominant signs in the CCD dogs and also are recognized in patients with AD, even in those with MCI.
House soiling was not a dominant sign in dogs with CCD in this study. Urinating and defecating in the house was recorded in 30% of the dogs in the CCD group, without any medical explanation. However, the dogs in the CCD group in our study, in general, had a moderate CCD score, which may explain this result. Urinary or fecal incontinence is a common problem in human patients with dementia. Many studies of geriatric dogs have found decreased interaction and house soiling to be associated with CCD.[3-5, 8, 31, 32, 38]
Dogs were distributed into a non-CCD, a b-CCD, and a CCD group. Thereby, we could both explore the hypothesis that a potential prodrome of CCD exists, and investigate progression of CCD signs separately for each group. We decided to set the upper cutoff at 15 for the b-CCD group because dogs with scores of 11–15 were dominated by increased sleeping, which may be associated with natural aging, whereas dogs with scores >15 displayed signs that have been described to characterize CCD. Our study documented that some of the dogs belonging to the b-CCD group developed CCD with time. A prodromal stage of CCD seems to be present, but may be difficult to detect because of the vagueness of the clinical signs expressed. We acknowledge that our results are influenced by our choice of cutoff. Our results, however, are supported by what has been shown in human geriatrics, where some people who experience MCI will develop AD, whereas others never progress and some even get better.[43, 44] Our results also are supported by previous questionnaire surveys that have identified equivocal CCD cases and named them predementia or query cases.[2, 6, 25] As in humans with MCI, it may be important to detect early CCD because subjects targeted for early therapeutic intervention may have a better prognosis.[33, 45]
Survival with CCD has not been investigated previously. For the dogs investigated in this study, it was found that a diagnosis of CCD did not affect survival negatively. Dogs in the CCD group surprisingly had a better survival than dogs in the non-CCD and b-CCD group and only 6 dogs of 37 were euthanized because of CCD. This result may partly reflect the fact that the owners of dogs with CCD had access to the primary investigator for advice during the study and were educated about CCD and advised regarding home activities, medical treatment, nutritional support, and cognitive stimulation, including medical treatment such as selegeline hydrochloride, specific diets proven to promote cognitive health, and cognitive stimulating toys.[3, 32, 33] The treatment regimen was variable and influenced by individual patient needs and of the owners opinions, and thereby reflects the fact that a clinical therapeutic approach that fits all is not practical in CCD. It is possible that the investigator-owner contact, supportive initiatives, and owner education resulted in an intensified human-animal bond and thus motivated the owners to keep the dog even longer than expected for dogs in general. The reduction in the questionnaire score that occurred for some of the dogs in the CCD group from inclusion to follow-up also suggests that supportive initiatives either helped the dogs or resulted in a situation in which the owners were not as bothered by the signs as previously. Another explanation, however, may be that the dogs in the CCD group were characterized by a moderate CCD score when included in the study. If more dogs had experienced severe CCD at inclusion, our results might have been different and may have caused more owners to elect euthanasia. Some recall bias or glorification of the lost dog also may have occurred.
Our study found that breed size did not influence survival and no association was found between CCD and size of breed. Previous studies have estimated a median age at death for dogs in general of 10 years and have shown that purebred dogs live less long than mongrels and that large-sized breeds live less long than small-sized breeds.[46-48] Small size dogs (≤15 kg) previously have been shown to have a greater risk of developing signs of CCD.
We investigated vitamin E (alpha-tocopherol) as a potential marker for CCD, because oxidative stress has been discussed as an important contributor to CCD and AD[18, 19] The antioxidant vitamin E (alpha-tocopherol) plays an important role through protective action, while being consumed, and thereby hypothetically may be decreased, when exposed to major radical production. A decreased concentration of vitamin E (alpha-tocopherol) has been found in brain tissue of dogs with CCD and in plasma from humans with AD.[19, 20] No correlation between cognitive status and Vitamin E (alpha-tocopherol) was found in this study, which may have been caused by the small sample size and the degree of CCD. Had more dogs in this study suffered from severe CCD, a difference might have been detected, when comparing with a control group.
Many geriatric dogs apparently are not recognized as suffering from CCD and consequently do not receive treatment. We propose that a few key questions addressing sleep-wake cycle, interaction, signs of confusion, and anxiety be posed to any owner of geriatric dogs and used as a clinical marker to decide if further investigation with an extended CCD questionnaire and diagnostic evaluation could benefit the dog. Special attention should be paid to anxiety in dogs with CCD, because anxiety is especially stressful to both dog and owner.
Based on our results, dogs with CCD seem to have a good chance of living a full lifespan if supported. It would be useful to investigate a larger cohort of dogs from the age of 8 years until death in a longitudinal study design with regular status visits to further explore the results of this study.
The authors thank veterinarian Line Rühmecke Jepsen for assistance in the clinic in the initial phase of the study.
The study was supported by a grant from SHARE-Synergy in Human and Animal Research, University of Copenhagen.
Conflict of Interest Declaration: Authors disclose no conflict of interest.