Exploring the Importance of Repeated Health Screening in Healthy Older Dogs

2.1 Study Cohort

The study was completed between July 2019 and June 2023. All dogs were privately owned. Owners signed an informed consent form, and the study was approved by the local ethical committee (EC2019/39).

Apparently healthy senior or geriatric dogs, as defined based on a previously published human/pet analogy chart [3] were prospectively recruited as previously described [23]. Weight categories were adapted slightly to avoid exclusion of certain weights (≤ 9.1 kg: ≥ 8 years; > 9.1–≤ 22.7 kg: ≥ 7 years; > 22.7–≤ 54.5 kg: ≥ 6 years; > 54.5 kg: ≥ 4 years). Confirmed healthy dogs were included in the follow-up study if baseline examinations did not indicate metabolic or systemic disease (e.g., malignant neoplasia, azotemic chronic kidney disease [CKD], diabetes mellitus, urolithiasis or clinically relevant cardiovascular disease [cardiac disease ≥ stage B2 or hemodynamically relevant arrhythmia]). Chronic kidney disease was diagnosed if serum creatinine concentration was ≥ 1.8 mg/dL (159 μmol/L) or symmetric dimethyl arginine (SDMA) was > 18 μg/dL in combination with urine specific gravity (USG) ≤ 1.030. International Interest Renal Society (IRIS) CKD stage 1 was diagnosed if persistent urinary protein: creatinine ratio [UPC] ≥ 0.5 or persistent USG ≤ 1.020 in the absence of azotemia was present. All dogs were fasted overnight and underwent complete health screening at 12-month intervals [5]. During the 2-year follow-up, medical or surgical treatment was allowed.

2.2 Health Screening Protocol

The study protocol was based on a previously published health study [5]. In brief, at each visit a history questionnaire [5] was completed and a standard physical examination was performed including assessment of body condition score (BCS) and MCS according to World Small Animal Veterinary Association (WSAVA) guidelines [24, 25] and rectal examination [5]. For dental disease, a categorization system previously described in cats was applied to assign each dog a calculus score of 0–3 and a gingivitis score of 0–3. The combination of both scores resulted in a dental disease score of 0–6, implying no (score 0), mild (score 1–2), moderate (score 3–4) or severe (score 5–6) dental disease [26]. Standard orthopedic, neurologic, and brief ophthalmic examinations were performed [5]. Systolic blood pressure (SBP) measurement (Doppler ultrasonography according to the American College of Veterinary Internal Medicine [ACVIM] consensus statement) [27] and direct fundoscopy were performed, and results are described in detail elsewhere [23, 28]. Fine needle aspiration and cytology of cutaneous or SC masses were performed. At each time point, blood was collected from the jugular vein (21 G needle) and urine was taken by ultrasonographic-guided cystocentesis (22 G needle) or voiding. Complete blood count and serum biochemistry profile, including SDMA and electrolytes (Procyte One Hematology, IDEXX SDMA ELISA and Catalyst Dx Chemistry Analyzer, IDEXX laboratories Inc., United States) were performed on each blood sample. Urinalysis consisted of measurement of USG using a manual refractometer (master refractometer, ATAGO CO. LTD), UPC (protein: colorimetric reaction with pyrogallol red molybdate, Beckman Coulter; creatinine: modified Jaffe assay, Beckman Coulter), urinary dipstick (iChem velocity stick, Beckman Coulter) and sediment analysis. Urinary sediment, prepared by centrifuging 5–10 mL of urine at 447 g for 3 min at 2°C (Jouan B4i, Thermo Scientific) was considered active in the presence of bacteriuria or > 3 casts or > 5 red blood cells, white blood cells, or epithelial cells per high power field (40× objective), respectively [29]. Aerobic bacterial culture was performed at baseline only.

Diagnosed diseases were grouped into specific categories (e.g., malignant neoplasia, neurologic disorder, gastrointestinal disease, orthopedic disorder, dermatologic disorder, urogenital disorder, CKD, cardiac disorder, hematologic disorder). Neurologic disorder was defined as neurologic abnormalities beyond hearing or vision deficiency or both, decreased or absent patellar reflexes, or previously known abnormalities (e.g., facial nerve paresis or paralysis). Dogs with unexplained increased liver enzyme activities (ALKP > 600 U/L or ALT > 300 U/L) or confirmed pancreatitis or cholecystitis, or cholelithiasis also were grouped under gastrointestinal disease. Renal disease was defined as the presence of azotemic CKD, persistent renal proteinuria (UPC ≥ 0.5) or USG ≤ 1.020 based on ≥ 2 urine samples. Dogs with orthopedic problems necessitating surgery or medical treatment beyond nutraceuticals were grouped under orthopedic disorder. Time and reason of death during or after the study period were noted. All owners of dogs alive at the end of the study period were contacted in March 2024 for follow-up.

2.3 Statistical Analysis

Statistical analysis was performed using the statistical software package R (R module of SAS version 9.3, SAS Institute, NC, United States). Baseline comparison between the senior and geriatric age groups was performed, based on the Welch two-sample t-test, for body weight, dental score, presence of systolic murmur, BCS, MCS, SBP, Schirmer tear testing results, USG, UPC, CBC variables, biochemistry variables, SDMA, and electrolytes. The predictive value of the following variables at baseline to onset of death was assessed for the confirmed healthy older dogs at baseline, using the Cox proportional hazards model to manage right censoring: age group, body weight, body and MCS, dental score, hematocrit, albumin, ALT, and ALKP. The level of significance was set at 5%.

3.1 Study Cohort

One-hundred twenty-two apparently healthy older dogs were included at baseline, with 67 senior and 55 geriatric dogs and 41, 33, 46, and 2 dogs in each weight category [3]. Fifty-one (42%) dogs were male (30 neutered) and 71 (58%) were female (43 spayed). The study cohort consisted of 14 mixed breed dogs, 10 Border Collies, 9 Belgian Shepherds, 8 Golden Retrievers, 6 Chihuahuas, 6 Labrador Retrievers, 6 Dachshunds, 5 Shetland Sheepdogs, 5 Jack Russell Terriers, 4 Cavalier King Charles Spaniels, 4 Shih Tzus, 4 English Cocker Spaniels, 4 Rottweilers, and ≤ 3 dogs of 28 other breeds. Follow-up data were available for 88 (time point T12) and 77 (T24) of the 98 dogs confirmed healthy at baseline (Table 1).

3.2 Health Screening: Findings in History and Physical Examination

Results of the completed questionnaires, at baseline and during follow-up, can be found in Table S1. The majority of owners reported good to excellent quality of life (QOL) for their dogs at baseline. However, > 25% of owners of confirmed healthy dogs reported decreased QOL by the end of the study, most commonly related to orthopedic (13/77, 17%) or neurologic (10/77, 13%) issues. In 10 confirmed healthy dogs, owners reported weight loss at baseline, but the majority (8/10) of these had an ideal BCS (4–5/9). Weight gain also was reported at baseline in 10 confirmed healthy dogs, of which 6 were overweight (BCS 6–7/9). The majority (80%, 85/107) of owners went to the veterinarian with their dog annually, with 10% (11/107) reporting a biannual visit. Vaccination in accordance with recent guidelines [31] and quarterly deworming was not up to date in 11% (12/110) and 24% (26/107) of dogs at baseline, respectively.

Findings on physical examination, at baseline and during follow-up, are presented in Tables 1 and 2. At baseline, only body weight was significantly (p < 0.05) higher in the geriatric compared to the senior age group. The majority of confirmed healthy dogs (90%) had an optimal BCS throughout the study, whereas a worse MCS was more commonly seen at the end (22%) of the study compared to T0 (5%). Overall, BCS and MCS were similar between apparently healthy and confirmed healthy dogs at baseline. None of the 13 of 122 (11%) dogs with obesity (BCS ≥ 7/9) at baseline showed signs of muscle wasting. Eight confirmed healthy dogs developed a new systolic murmur within 2 years of follow-up (grade 1–2, n = 6; grade 3–4, n = 2), and in all cases the murmur was asymptomatic. Echocardiography was performed in 2/8 dogs and identified myxomatous mitral valve disease ACVIM stage B1. None of the confirmed healthy dogs had an irregular heart rhythm at any time point, other than respiratory sinus arrhythmia. Throughout the study period, rectal examination identified 5 palpable anal sac masses (right-sided, n = 3; left-sided, n = 2). The majority (4/5) were surgically removed; 1 dog was euthanized because of concurrent hypercalcemia and renal azotemia. In 3/4 surgical cases, histology of the mass confirmed an anal sac adenocarcinoma; in the remaining dog, no histopathology was performed. Six female dogs had undergone partial or complete unilateral mastectomy before inclusion. In three intact female dogs where a mammary gland nodule was observed at baseline, histopathology identified a benign process. These dogs therefore still met the criteria for “confirmed healthy” and were included for follow-up. One of these dogs developed a new mammary gland nodule at a different location by T12. Subcutaneous masses were commonly observed at baseline and during follow-up. Of nodules on which cytology was performed, the majority (> 90%) were classified as lipomas or sebaceous gland cysts.

Schirmer tear testing (STT) was performed in 50% of dogs at baseline. In 50 of 63 (79%) apparently healthy older dogs, STT was bilaterally normal, in 11 of 63 dogs (17%) unilaterally decreased (< 15 mm/min) and in 2 dogs bilaterally decreased, of which one was previously diagnosed with keratoconjunctivitis sicca (KCS). In three confirmed healthy dogs with normal STT at baseline, STT decreased to < 15 mm/min by T12. Lens opacification was the most commonly observed abnormality on ophthalmic examination, with an increasing frequency by T24. Of 52 confirmed healthy dogs without lens opacification at baseline, 24 (46%) showed mild opacification by T24.

3.3 Health Screening: Laboratory Findings

Results of routine blood tests and urinalysis at baseline and during follow-up are presented in Tables 3 and 4. The most common deviations on CBC in dogs at baseline were mild anemia (12%; of which 17% [2/12] showed signs of regeneration [reticulocyte count > 60 K/μL]), mild lymphocytosis (18%), mild leukopenia (8%) and mild neutropenia (6%); these frequencies remained stable over time (Table 3). Increased ALKP or ALT activity was observed in > 10% of dogs at baseline and during follow-up. In ≥ 60% of cases, these increases were mild (≤ 2 times the upper limit of the reference interval [RI]). Of the six dogs with moderate to severely (> 2 times the upper limit of the RI) increased liver enzyme activities, four underwent further evaluation; in two dogs, vacuolar hepatopathy was found, and in two dogs, abdominal ultrasonography did not identify an obvious cause. Deviations in electrolytes were rare and usually minor, with the exception of hypophosphatemia. At baseline, 11% of dogs had a decreased serum phosphorus concentration, with a lowest concentration of 0.42 mmol/L. Because of the high number of apparently healthy dogs with hypophosphatemia, an age-specific RI (0.72 [0.42–0.75]–1.74 [1.38–2.6] mmol/L) was calculated using a non-parametric percentile method according to the American Society of Veterinary Clinical Pathology (ASVCP) guidelines [33, 34]. A histogram is provided in Figure 1 to illustrate the distribution of serum phosphorus concentration in our study cohort.

At baseline, only USG was significantly (p < 0.05) lower in the geriatric compared with the senior age group. Bacteriuria was observed at baseline in 11% of confirmed healthy dogs; in 7% it was confirmed by positive urine culture (three voided and four cystocentesis samples). These dogs did not show lower urinary tract signs, and this subclinical bacteriuria was not treated, in accordance with recent guidelines [35]. The incidence of subclinical microscopically detected bacteriuria remained similar during follow-up (10%–14%; Table 4).

3.4 New or Unknown Diseases Detected at Baseline

Based on the complete health evaluation at baseline, 24/122 (20%) apparently healthy older dogs were diagnosed with ≥ 1 disease, excluding them from further inclusion in the longitudinal health screening study (Table 5). In total, five dogs were diagnosed with a malignant neoplasia at baseline (n = 3, anal sac tumors; n = 1, mammary gland tumor and pulmonary mass; n = 1, cutaneous mastocytoma). Azotemic CKD [31] was not diagnosed at baseline, but 16% (20/122) showed CKD IRIS stage 1 [31] (n = 15, persistent renal proteinuria; n = 5, persistent USG ≤ 1.020) and one other dog was diagnosed with non-obstructive sterile struvite urolithiasis and excluded from follow-up.

3.5 New or Unknown Diseases Detected During Follow-Up

During the first year of follow-up, 1 dog dropped out for owner-related reasons and 7 dogs were euthanized. During the second year of follow-up, 15 dogs died (euthanasia, n = 14; hit-by-car, n = 1). In 43% (9/21) euthanasia was elected because of malignant neoplasia (lymphoma, n = 2; hemangiosarcoma, n = 2; mammary and pulmonary mass, n = 1; intramedullary sarcoma, n = 1; pancreatic adenocarcinoma, n = 1; gastric adenocarcinoma, n = 1; heart base tumor, n = 1). Remaining dogs were mainly euthanized because of a (progressive) neurologic disorder (n = 9, 41%) including cognitive dysfunction, and less frequently because of progressive CKD (n = 1), hematologic disorder (n = 1) or gastrointestinal disorder (n = 1). In 4 confirmed healthy dogs, health screening was repeated at T24 but not at T12.

Diseases that confirmed healthy dogs developed within 2 years are listed in Table 6. In 3/12 dogs where malignant neoplasia developed, it was detected on health screening (n = 2, anal sac tumor; n = 1, indolent splenic lymphoma). In the nine other dogs, a diagnosis was made independent of health screening time points. Including the 5 dogs diagnosed with neoplasia at baseline, these 8/17 (47%) malignant neoplasms were detected during health screening.

3.6 Prediction of Mortality

The normal distribution assumption for the biomarkers was not rejected based on the Shapiro-Wilks test. Age group had a significant predictive value for onset of death (p < 0.0001) and was implemented as a stratification factor for the remaining analysis. None of the remaining assessed variables showed a predictive value for onset of death in our cohort of 98 confirmed healthy older dogs at baseline (Table S2).