Methods

The Ethics Committee on Human Research of Kitasato University approved the present study. Outpatients who regularly visited the cardiovascular center of Kitasato University Hospital from January 2009 to December 2010 because of diabetes mellitus, hypertension, and hyperlipidemia were eligible to participate. Patients with other major illnesses, such as old myocardial infarction, angina pectoris, chronic renal failure, and frequent arrhythmias, were excluded. A total of 244 patients who satisfied the initial criteria were informed about the study, and all patients provided written informed consent.

The patients were classified into 2 groups according to pet ownership: pet owner and nonowner groups. A pet owner was defined as a patient who currently had a pet and had had the pet for >6 months at study enrollment. Those patients who had had pets in the past, but not currently, were excluded from the present study. In the pet owner group, the patients were interviewed about their pet ownership status, including the number and species of pets, pet-keeping duration, where the pet is kept (inside or outside the house), the amount of time spent a day by the owners walking their dogs, and whether the patient was the main caretaker of the pet. Pet ownership status is listed in Table 1 .

We assessed the disease-specific cardiovascular functional status using the specific activity scale. The specific activity scale included 21 items of self-reported data on the performance of well-defined daily activities (e.g., walking, climbing stairs, showering, dressing), which was developed as an alternative to the New York Heart Association functional classification.

After the interview, the patients underwent physical and blood examinations, 24-hour Holter electrocardiogram, echocardiogram, and a measurement of brachial-ankle pulsewave velocity. After the 24-hour Holter electrocardiogram, the patients who exhibited atrial fibrillation, marked sinus arrhythmia, or frequent atrial or ventricular premature beats were excluded. Thus, 191 patients participated in the present study ( Figure 1 ) .

Patient flow charts.

The clinical and sociodemographic variables, including gender, age, systolic blood pressure, diastolic blood pressure, heart rate, diagnosis, current smoking status, and medications, were obtained from the medical records at the beginning of the study. The routine biochemical parameters, such as the serum concentrations of low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglycerides, and hemoglobin A1c, were evaluated. The body mass index was calculated as the body weight in kilograms divided by height in square meters. Echocardiographic analysis was performed by an investigator who was unaware of the clinical and treatment status of patients. Left atrial diameter, left ventricular end-diastolic diameter, left ventricular end-systolic diameter, left ventricular posterior wall thickness, and interventricular septal thickness were measured to calculate the left ventricular ejection fraction and left ventricular muscle mass. The ventricular muscle mass was corrected by body surface area and expressed as the ventricular muscle mass index.

Autonomic nervous activity was evaluated by heart rate variability (HRV) analysis using a 24-hour Holter electrocardiogram (Aria, Del Mar Reynolds Medical, Irvine, California). The HRV based on the beat-to-beat RR intervals of normal sinus beats was assessed for 24 hours, during the day (8.00 a.m. to 5.00 p.m. ) and during the night (0:00 a.m. to 6.00 a.m. ). A Holter recording was considered suitable for time-domain, frequency-domain, or nonlinear-domain HRV analysis if it contained ≥95% analyzable data for each period.

The time-domain indexes included the standard deviation of a series of all-normal RR intervals (SDNN), standard deviation of mean RR intervals of a 5-minute electrocardiogram (SDANN), the square root of the average of squares of differences between consecutive RR intervals (RMSSD), and the percentage of RR intervals that differed from each other by >50 ms. SDNN and SDANN reflect overall HRV. RMSSD and percentage of RR intervals that differed from each other by >50 ms reflect the cardiac parasympathetic nervous activity. The time-domain parameter indexes for the 24-hour, daytime, and nighttime periods are presented as the SDNN _24h , SDNN _day , SDNN _night , RMSSD _24h , RMSSD _day , RMSSD _night , SDANN _24h , SDANN _day , and SDANN _night , respectively.

Using RR intervals obtained from a 24-hour Holter electrocardiogram, a beat-to-beat spectral analysis was performed with a combination of the maximum entropy method for spectral analysis and the nonlinear least squares method for fitting analysis (MemCalc, GMS, Tokyo, Japan) to obtain the HR and 2 frequency bands: a low frequency (LF) component of 0.04 to 0.15 Hz and a high frequency (HF) component of 0.15 to 0.4 Hz. The HF and LF/HF ratio indicate parasympathetic nervous activity and the dominance of sympathetic nervous activity over parasympathetic activity, respectively. The HR and frequency-domain parameters for the different periods are presented as HR _24h , HR _day , HR _night , HF _24h , HF _day , HF _night , LF/HF _24h , LF/HF _day , and LF/HF _night , respectively.

A nonlinear-domain analysis of HRV was also performed with MemCalc to assess entropy as a complexity of the cardiovascular system. Entropy was calculated from a pulse time series of 4 RR intervals and expressed as a scale from 0% (indicating no HRV randomness) to 100% (indicating complete HRV randomness). Entropy for the different periods is presented as entropy _24h , entropy _day , and entropy _night , respectively.

The Mann-Whitney U test, Student’s t test, and chi-square test were used to examine the differences in the clinical characteristics and autonomic nervous activity between pet owner and nonowner groups. The comparisons of the mean values for normally distributed continuous variables were performed using Student’s t test. For continuous variables with non-normal distributions, the Mann-Whitney U test was used. Univariate and multivariate linear regression analyses of the HF and LF/HF measurements were conducted for potential confounding variables: gender, age, HR, body mass index, left ventricular ejection fraction, brachial-ankle pulsewave velocity, diabetes mellitus, hypertension, hyperlipidemia, smoking status, and pet ownership, to evaluate potential predictive factors for autonomic nervous activity in patients with lifestyle-related diseases. p <0.05 was considered statistically significant. All statistical analyses were performed using SPSS, versions 16.0J for Windows (SPSS Japan, Tokyo, Japan).

Results

The patient characteristics are presented in Table 2 . No significant differences were found in the demographic and physiologic characteristics between the pet owner and nonowner groups.

The HR and time-domain analysis of HRV parameters are presented in Table 3 . HR _night was significantly lower in the pet owner group than in the nonowner group (p <0.05). RMSSD _24h , RMSSD _day , RMSSD _night , and percentage of RR intervals that differed from each other by >50 ms were significantly greater in the pet owner group than in the nonowner group (p <0.01 for each). HF _24h , HF _day , and HF _night were significantly greater in the pet owner group than in the nonowner group (p <0.01 for each; Figure 2 ) . The LF/HF _24h and LF/HF _night were significantly lower in the pet owner group than in the nonowner group (p <0.05 for each; Figure 3 ) ; and entropy _24h , entropy _day , and entropy _night were significantly greater in the pet owner group than in the nonowner group (p <0.01 for each; Figure 4 ) .

HF component in pet owner and nonowner groups. Data are presented as box plots displaying median, 25th and 75th percentiles (boxes), and 10th and 90th percentiles (whiskers). *p <0.05 and **p <0.01 (pet owner vs nonowner groups).

LF/HF in pet owner and nonowner groups. Data are presented as box plots displaying median, 25th and 75th percentiles (boxes), and 10th and 90th percentiles (whiskers). *p <0.05 (pet owner vs nonowner groups).

Entropy in pet owner and nonowner groups. Data are presented as box plots displaying median, 25th and 75th percentiles (boxes), and 10th and 90th percentiles (whiskers). *p <0.05 and **p <0.01 (pet owner vs nonowner groups).

The univariate and multivariate regression analysis results of HF are presented in Table 4 . On multivariate analysis, pet ownership was positively associated with HF _24h , HF _day , and HF _night (p <0.01 for each), and HR was inversely associated with HF _24h , HF _day , and HF _night (p <0.01, p <0.05, and p <0.01, respectively). Diabetes mellitus was inversely associated with HF _24h and HF _day (p <0.01 for each), and hyperlipidemia was inversely associated with HF _24h , HF _day , and HF _night (p <0.01 for each).

Table 4

Clinical predictors for high frequency (HF) in heart rate variability (HRV) using uni- and multivariate linear regression analyses

High Frequency	Univariate			Multivariate
	B	95% CI	p Value	B	95% CI	p Value
24 hours
Gender	7.4	−27.1, 41.9	0.67	8.2	−24.9, 41.3	0.63
Age	−0.79	−2.6, 1.0	0.38	−0.56	−2.4, 1.3	0.55
Heart rate	−2.9	−4.6, −1.2	0.001 ⁎	−2.55	−4.2, −0.9	0.003 ⁎
Body mass index	1.3	−3.6, 6.3	0.60	2.2	−2.5, 6.9	0.35
Left ventricular ejection fraction	−0.04	−2.0, 1.9	0.97	−0.4	−2.24, 1.5	0.70
Brachial ankle pulse wave velosity	−0.04	−0.09, 0.006	0.09	−0.03	−0.08, 0.02	0.27
Diabetes mellitus	−30.53	−60.3, −0.8	0.044 †	−38.3	−67, −9.8	0.009 ⁎
Hypertension	12.5	−17.4, 42.4	0.41	18.6	−11.6, 48.8	0.24
Hyperlipidemia	−44.2	−74.3, −14.1	0.004 ⁎	−40.5	−69.2, −11.9	0.006 ⁎
Smoking	5.2	−42.6, 53.0	0.83	7.9	−37.9, 53.6	0.74
Pet ownership	66	37.3, 94.8	0.00 ⁎	54.0	25.8, 82.1	0.000 ⁎
Day
Gender	4.0	−31.0, 39.1	0.82	0.9	−34.0, 35.8	0.96
Age	−0.3	−2.1, 1.5	0.73	−0.24	−2.2, 1.7	0.81
Heart rate	−2.5	−4.3, −0.8	0.005 ⁎	−1.9	−3.7, 0.09	0.04 †
Body mass index	1.1	−4.0, 6.2	0.66	2.2	−2.8, 7.1	0.39
Left ventricular ejection fraction	0.52	−1.5, 2.5	0.61	0.28	−1.7, 2.2	0.78
Brachial ankle pulse wave velosity	−0.03	−0.08, 0.02	0.26	−0.02	−0.08, 0.03	0.38
Diabetes mellitus	−29	−59.3, 1.2	0.06	−38.1	−68, −8.1	0.01 ⁎
Hypertension	17.5	−12.9, 47.8	0.26	22.5	−9.3, 54.3	0.16
Hyperlipidemia	−46.2	−76.6, −15.7	0.003 ⁎	−42.7	−72.9, −12.6	0.006 ⁎
Smoking	1.2	−47.4, 49.7	1.0	3.3	−44.9, 51.5	0.89
Pet ownership	52.1	22.3, 81.9	0.001 ⁎	42.56	12.93, 72.2	0.005 ⁎
Night
Gender	32.1	−14.5, 78.6	0.18	35.08	−10.2, 80.4	0.13
Age	−2.1	−4.5, 0.31	0.09	−1.74	−4.3, 0.81	0.18
Heart rate	−4.1	−6.4, −1.7	0.001 ⁎	−3.55	−5.9, −1.2	0.003 ⁎
Body mass index	2.6	−4.1, 9.4	0.45	3.05	−3.3, 9.4	0.35
Left ventricular ejection fraction	0.1	−2.5, 2.8	0.91	−0.72	−3.3, 1.8	0.58
Brachial ankle pulse wave velosity	−0.06	−0.13, 0.003	0.06	−0.04	−0.1, 0.03	0.30
Diabetes mellitus	−19.1	−59.7, 21.6	0.36	−28.5	−67.5, 10.5	0.15
Hypertension	18.0	−22.6, 58.5	0.38	25.64	−15.7, 67.0	0.22
Hyperlipidemia	−57.7	−98.6, −16.9	0.006 ⁎	−51.57	−90.8, −12.4	0.01 ⁎
Smoking	18.5	−46.4, 83.3	0.58	20.17	−42.5, 82.8	0.53
Pet ownership	84.6	45.5, 123.8	0.00 ⁎	65.3	26.8, 103.8	0.001 ⁎

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Diabetes Mellitus, Myocardial Reperfusion, and Outcome in Patients With Acute ST-Elevation Myocardial Infarction Treated With Primary Angioplasty (from HORIZONS AMI)
2. Impact of Anemia on Platelet Response to Clopidogrel in Patients Undergoing Percutaneous Coronary Stenting
3. Cardiac Transplantation in Adults With Aortic Valve Disease With Focus on the Bicuspid Aortic Valve
4. Good Books in Cardiovascular Disease Appearing in 2011

Stay updated, free articles. Join our Telegram channel

Join