Atherosclerosis and physical functioning in older men, a longitudinal study

Abstract

Objective

Functional decline is a major threat to independency, progressing into functional limitations and eventually leading to disability. Chronic diseases, especially cardiovascular diseases, are important determinants of functional limitations and disability. Vascular damage exits long before it is clinically manifest and can have adverse effects on health, physical and cognitive functioning. The objective was to investigate the association between non-invasive atherosclerosis measures and physical functioning in older men.

Design

Prospective cohort study.

Setting

The study was conducted in the general community.

Participants

195 independently living older men.

Measurements

Atherosclerosis was measured by intima media thickness (CIMT) of the common carotid artery using ultrasonography and assessment for presence of atherosclerotic plaques. Physical functioning was measured by isometric handgrip strength and leg extensor strength using a hand held dynamometer, lower extremity function using the physical performance score and ability to perform activities of daily life using the modified Stanford Health Assessment Questionnaire. Linear regression analysis was performed to estimate the associations between CIMT or plaques and physical functioning.

Results

After adjustment for confounders, higher baseline CIMT was associated with lower isometric handgrip strength at follow up (β^CIMT =−7.21, 95% CI[−13.64;−0.77]). No other associations were found between CIMT and physical functioning. In addition, no associations were found for the presence of plaques and physical functioning either at baseline, or at follow-up.

Conclusion

Atherosclerosis, as measured by higher CIMT, is related to a lower isometric handgrip strength at follow-up, but no further associations with physical functioning were found in this longitudinal study among independently living older men.

Introduction

Functional decline is a major threat to independency, progressing into functional limitations and eventually leading to disability (1, 2). Although functional limitations and disability are the subject of a growing number of studies, the causal pathways remain unclear.

The prevalence of disability in activities of daily life (ADL) in community-living older people is estimated at 20-30%, this percentage increases with age (1). Chronic diseases are important determinants of functional limitations and disability in older people (3, 4). An important chronic condition that predisposes older people to functional limitations and disability is cardiovascular disease (CVD) such as coronary ischemic disease, heart failure and cerebrovascular accidents (1, 2, 5). It has been suggested that over 25% of disability is attributable to CVD (1). In the Netherlands, CVD ranks first among all diseases in loss of disability-adjusted life years (6), and CVD is also a major contributor to reduced disability free life expectancy (5, 7). Facing the expected increase of older people on one hand and people with CVD on the other hand in the upcoming twenty years, disability will become an increasing burden for patients and society.

Although part of end-stage disability is the result of acute clinical events, over 50% results from gradual functional decline (3, 8). Already in younger and middle-aged populations vascular damage exists long before it is clinically manifest (9., 10., 11.). It is also known that subclinical CVD can have adverse effects on health, physical and cognitive functioning (12). This stresses the importance of intervening in an early stage of the disabling process. However, the relation between subclinical CVD and physical functioning is still unknown. Both Carotid Intima Media Thicknes (CIMT) and atherosclerotic plaques are a non-invasive, reproducible measures as markers of atherosclerosis (13, 14). Physical functioning (e.g. muscle strength and lower extremity function) is a predictor of future functional limitations and disability in older people (15, 16). Handgrip strength has often been used as an indicator of overall muscle strength (15, 17). Lower extremity function can be measured with physical performance tests (e.g. tests of balance, walking speed and ability to rise from a chair) (18).

Insight in the aetiology of the pathway from subclinical atherosclerosis to functional decline and disability will aid in the development of effective preventive interventions. This could enable people, even in old age and with (subclinical) CVD, to live independently. The aim of the present study was to investigate the prospective association of carotid intima media thickness and presence of plaques, as measures of atherosclerosis, with functional decline in older men.

Subjects and Methods

Study design and population

In 1996, 403 participants who lived independently and had no severe mobility problems were recruited from male inhabitants of Zoetermeer, a medium-sized town in the Midwestern part of the Netherlands. Participants were judged sufficiently healthy to participate in the study if they were physically and mentally able to visit the study center independently. All participants provided informed consent, and the study was approved by the Institutional Review Board of the University Hospital Rotterdam. The 327 participants who were still alive in 2000 were reinvited for a follow-up examination, of whom 247 (61%) participated. Because of technical problems and inability to visit the study center, 52 of the 247 participants were visited at home, and therefore a second measurement of the carotid CIMT was not possible, leaving 195 participants eligible for the present analyses.

Measurements

At the baseline as well as the follow-up visit, participants visited the study center (except for scheduled home visits at follow-up). Height and weight of the participants were measured in standing position without shoes. Body mass index was calculated as weight in kilograms divided by the square of height in meters. Fat mass was measured using dual energy X-ray absorptiometry (DXA, Lunar, Madison WI,USA). Participants were asked about smoking status and current use of medications. Educational level was categorized as low, middle, high and university. Cardiovascular drug therapy was defined as using one or more of the following drugs: angiotensin-converting enzyme inhibitors, calcium antagonists, diuretics, B-inhibitors (any of these four drugs were not included if they were used only to lower hypertension), glycosides, nitrates, amiodarone, cholesterol-lowering drugs, and anticoagulants (19, 20). Blood pressure was measured twice with a semi-automated device (Dynamap) after 10 minutes of rest and 5 minutes apart. The average of the two measurements was used for analysis and further calculation.

Generalized non-invasive atherosclerosis measures

Carotid intima media thickness

Carotid intima media thickness (CIMT) is a measure of the thickness of the intima and media layer of the carotid artery and is commonly used as a marker of atherosclerosis (1 3). Ultrasonography of the left and right common carotid artery and the bifurcation was performed with a 7.5-MHz linear array transducer (ATL Ultramark IV; Advanced Technology Laboratories, Inc., Bothell, Washington). A careful search was conducted for all interfaces of the near and far walls of the distal common carotid artery and the far wall of the carotid bifurcation. The actual CIMT measurements were performed off-line, as described previously (19, 20). The baseline and follow-up CIMT images were read by two separate readers. To assess the interobserver variability, the CIMT images of 10 subjects were independently read by both readers. The absolute mean difference between the readers was 0.09 mm, which was subtracted from the CIMT at follow-up (21).

Carotid plaques

At baseline, the common carotid artery, carotid bifurcation and internal carotid artery were evaluated for the presence (yes/no) of atherosclerotic lesions on both the near and far walls of the carotid arteries. Plaques were defined as a focal widening relative to adjacent segments, with protrusions into the lumen composed of only calcified deposits or a combination of calcification and noncalcified material. The size of the lesions was not quantified. The number of plaques present was used as an indicator of the presence of atherosclerosis (range from zero (no plaques present) to maximum of 12 plaques) (20).

Physical functioning

Isometric handgrip strength

Isometric handgrip strength was measured using an adjustable hand held dynamometer (JAMAR). Participants were asked to sit or stand comfortably with their shoulder adducted and neutrally rotated. The elbow was flexed at 90°. The forearm and wrist were in a neutral position. The dynamometer was held freely, without support and the participant was asked to squeeze with maximum strength. The dynamometer automatically recorded the highest force exerted. Each test was repeated three times. The average of the three tests was used in the analyses.

Leg extensor strength

Leg or knee extensor strength was measured using the Hoggan MicroFET hand held dynamometer. Maximum leg strength was defined as the maximum strength for the right or left leg, whichever was largest, in a position of 120-degree extension. Statistical analyses were based on the physical unit momentum (Nm), obtained by multiplying the maximum strength (in Newton) and the distance of the dynamometer to the knee joint (in meters).

Physical Performance Score

According to Guralnik, physical performance, or lower extremity function, can be evaluated by using tests of gait speed, standing balance and ability to rise from a chair (16, 18). Standing balance was assessed using three tests, namely side-by-side, semi-tandem and full-tandem stand. Standing balance was scored 0 to 4, depending on the number of tests a person could complete. For the 8-foot walk and repeated chair stands, those who could not complete the task were assigned a score of 0. Those completing the task were assigned scores of 1 to 4, corresponding to the quartiles of time needed to complete the task, with the fastest times scored as 4. The Physical Performance Score (PPS) was created by summing the category scores of walking, chair stand, and balance tests, which ranged from 0 (worst performance) to 12 (best performance).

Ability to perform activities of daily living (ADL)

The ability to perform activities of daily life was measured using the modified Stanford Health Assessment Questionnaire (HAQ) (22). The HAQ consists of eight items, i.e. activities, grip, reach, hygiene, walking, eating, arising and dressing/grooming. A total score of the previous items was calculated.

Data analysis

Descriptive analyses included calculation of means and standard deviation for continuous variables and frequencies and percentages for categorical variables. Differences in physical functioning and CIMT between follow-up and baseline were calculated.

Missing data at follow-up occurred because occasionally a test or measurement did not succeed in a participant, and because CIMt-measurement could not be performed in participants who were visited at home. For the following follow-up variables missings occurred: isometric grip strength (n=1), physical performance score (n=3), leg strength (n=2) and ADL-score (n=2). In addition, a total of 80 men were not willing or mentally or physically able to participate in the follow-up visit. In order to investigate whether loss to follow-up was selective, and may have biased the results, we also imputed the values of all study variables for persons who were not willing to participate in the follow-up study (n=80). Missings at follow-up were imputed by multiple imputation (m=10) using the statistical program PASW Statistics (version 17.0). Baseline characteristics were used as imputation variables.

The associations of baseline-CIMT and baseline plaques score with isometric grip strength, physical performance score, leg extensor strength and ADL-score were studied cross-sectionally using baseline data, and prospectively using follow-up data. We used multivariate linear regression analysis, and pooled the results of the ten multiple imputation data sets. The results of the multivariate linear regression analysis are presented as linear regression coefficients (B) and 95% Confidence Interval. In de first model, crude regression coefficients were estimated. In the second model we adjusted for age, and in the third model we additionally adjusted for BMI, fat mass, smoking status (current/former/never), use of ACE-inhibitors (yes/no), diastolic and systolic blood pressure and educational level. Finally, we repeated the regression analyses including the imputed follow-up data for the men who had missing data on physical function or CIMT and who did not participate in the follow-up visit.

The statistical program PASW Statistics (version 17.0) was used to analyze the data. Significance levels were set at α=0.05 for all tests.

Results

General characteristics of the participants at baseline and follow-up are shown in table 1. The mean age at baseline was 77.5 years (SD 3.4, range 73-91) and 81.4 years (SD 3.2, range 76-96). At baseline, the majority of the participants had a low or middle educational level (63.4%). A history of symptomatic cardiovascular disease was present in 99 men (40.0%). The mean CIMT was 0.97 mm (SD 0.15mm) at baseline and increased to 1.00 mm (SD 0.23) at follow-up (^CIMT = 2.61 mm, 95% CI [-11.31; 6.09]) (figure 1). In the majority (69.6%) of the study population more than one atherosclerotic plaque was present at baseline. Mean grip strength at baseline was 35.5 kg (SD 6.9) compared to 34.2 kg (SD 7.0) at follow-up (^{grip strength} = -2.63 kg, 95% CI [-11.27; 6.01]). The mean maximal leg extensor strength at baseline was 107.1 Nm (SD 20.9) compared to 109.9 Nm (SD 21.8) at follow-up (^{leg strength} = -1.20 Nm, 95% CI [-34.46; 32.06]). The mean ADL-score was 10.5 (SD 4.2) at baseline compared to 13.0 (SD 4.5) at follow-up (A ^ADL = 2.54, 95% CI [-7.75; 12.83]). In 26% of the participants the ADL score was similar at follow-up compared to baseline (n=65), 61% (n=150) experienced more difficulty in ADL and 13% (n=32) experienced less difficulty in ADL.

Table 1.

characteristics of the participants

	Baseline (n= 327)			Follow-up complete cases (n= 195)			Follow-up imputation (n= 327)
	n	(%)	Mean ± sd	n	(%)	Mean ± sd	n (%) Mean ± sd
Age (years)			77.5 ± 3.4			81.4 ± 3.2			81.7 ± 3.3
Body mass index (kg/m2)			25.6 ± 3.0			26.2 ± 3.2			26.1 ± 3.3
Waist-Hip ratio			0.98 ± 0.05			0.98 ± 0.05			0.99 ± 0.05
Fat mass (kg)			21.2 ± 5.6			20.1 ± 6.4			20.1 ± 6.3
Lean mass (kg)			51.9 ± 5.7			57.3 ± 6.8			56.7 ± 7.0
Mean intima media thickness (mm) Isometric grip strength (kg)			0.97 ± 0.15 35.5 ± 6.9			1.00 ± 0.23 34.2 ± 7.0			1.02 ± 0.25 31.4 ± 8.7
Maximum leg extensor strength (Nm)			107.1 ± 20.9			109.9 ± 21.8			100.4 ± 30.3
Total HAQ score			10.5 ± 4.2			13.0 ± 4.5			13.8 ± 8.8
Carotid atherosclerotic plaques			3.2 ± 2.6
Blood pressure
Systolic (mm Hg)			142 ± 20			153 ± 24			153 ± 24
Diastolic (mm Hg)			79 ± 11			83 ± 11			83 ± 11
Smoking
Current	57	(17.4)
Former Never	224 46	(68.5) (14.1)
Muscle joint complaints
Kneepain	60	(18.3)
Hippain	25	(7.6)
Backpain	82	(25.3)
Cardiovascular disease	99	(40.0)
Myocard infarct	47	(14.4)
Angina pectoris	41	(12.5)
ACE-inhibitors use	31	(9.5)
Claudicatio intermittens	11	(3.4)
Diabetes Mellitus	19	(5.8)
Physical performance score
0-3	8	(2.4)		11	(5.6)		42	(12.8)
4-6	56	(17.1)		33	(16.9)		72	(22.0)
7-9	120	(37.0)		51	(26.2)		73	(22.3)
10-12	143	(44.4)		100	(51.3)		140	(42.8)

Figure 1.

Distribution of Carotid Intima Media Thickness at baseline and 4-year follow-up

Table 2 shows the associations of baseline-CIMT and isometric grip strength, physical performance score, leg strength and the ability to perform ADL at baseline and follow-up. After adjustment for confounders, higher baseline CIMT was associated with lower isometric grip strength at follow up (^CIMT =-7.21, 95% CI [-13.64; -0.77]) (figures 2 and 3). No associations were found for baseline CIMT and physical performance score, leg strength and ability to perform ADL at baseline and follow-up. Table 3 shows the associations between the presence of plaques with physical functioning. No associations were found for the presence of plaques and physical functioning measures.

Table 2.

Linear regression coefficients for intima media thickness (t0) with physical performance measures at baseline and at follow-up (n=195)

Isometric grip	Isometric grip	Leg strength (t0)	Leg strength (t1)	Physical	Physical	ADL - score # (t0)	ADL - score
	strength (t0)	strength (t1)			Performance Score (t0)	Performance Score (t1)		(t1)
Crude model
B 95% CI	−2.19 -6.62; 2.24	−8.81 -15.24; -2.39	2.73 -10.74; 16.19	−6.99 -26.78; 12.79	−1.06 -2.62; 0.51	−2.33 -5.04; 0.38	1.19 -1.61; 3.99	0.21 -0.15; 0.57
Age-adjusted model *
B 95% CI	−1.63 -5.86; 2.61	−6.80 -13.07; -0.53	4.30 -8.71; 17.30	−1.34 -20.57; 17.89	−0.84 -2.33; 0.65	−1.23 -3.75; 1.30	1.10 -1.69; 3.88	0.13 -0.23; 0.48
Fully adjusted model §
B 95% CI	−3.06 -8.89; 2.76	−7.21 -13.64; -0.77	−7.44 -24.46; 9.58	−7.05 -26.74; 12.64	−1.03 -2.91; 0.85	−1.76 -4.26; 0.74	−0.37 -4.07; 3.32	3.16 -2.45; 8.77

Note: # ADL= activities of daily living;

^*Adjusted for age; § additionally adjusted for BMI, smoking, use of ACE-inhibitors, educational level, fat mass, diastolic blood pressure and systolic blood pressure.

Figure 2.

Correlation between Carotid Intima Media Thickness at baseline and isometric handgrip strength at follow-up

Figure 3.

Correlation between Carotid Intima Media Thickness at follow-up and isometric handgrip strength at follow-up

Table 3.

Linear regression coefficients for carotid atherosclerotic plaques (t0) with physical performance measures at baseline and at follow-up (n=195)

	Isometric grip strength (t0)	Isometric grip strength (t1)	Leg strength (t0)	Leg strength (t1)	Physical Performance Score (t0)	Physical Performance Score (t1)	ADL - score# (t0)	ADL - score # (t1)
Crude model
B 95% CI	−0.23 -0.57; 0.12	−0.40 -0.78; -0.04	−0.71 -1.72; 0.30	−0.50 -1.62; 0.63	−0.02 -0.14; 0.09	−0.06 -0.21; 0.10	−0.01 -0.21; 0.22	0.03 -0.19; 0.24
Age-adjusted model *
B 95% CI	−0.18 -0.51; 0.15	−0.35 -0.71; -0.01	−0.58 -1.56; 0.39	−0.37 -1.45; 0.72	−0.01 -0.12; 0.10	−0.03 -0.17; 0.11	−0.01 -0.22; 0.20	0.02 -0.19; 0.23
Fully adjusted model §
B 95% CI	−0.19 -0.53; 0.16	−0.33 -0.71; 0.06	−0.48 -1.52; 0.55	−0.24 -1.45; 0.97	−0.03 -0.14; 0.09	−0.06 -0.21; 0.10	−0.04 -0.25; 0.18	0.08 -0.26; 0.42

Note: # ADL= activities of daily living;

^*adjusted for age; § additionally adjusted for BMI, smoking, use of ACE-inhibitors, educational level, fat mass, diastolic blood pressure and systolic blood pressure.

Additionally adding the imputed data for persons who were not willing to participate in the follow-up study yielded similar results (tables 4 and 5).

Table 4.

Linear regression coefficients for intima media thickness (t0) with physical performance measures at baseline and at follow-up (n=327)

	Isometric grip strength (t0)	Isometric grip strength (t1)	Leg strength (t0)	Leg strength (t1)	Physical Performance Score (t0)	Physical Performance) Score (t1)	ADL - score (t0)	ADL - score (t1)
Crude model
B	−2.50	−7.47	−0.74	−4.73	−1.35	−1.99	0.90	2.59
95% CI	−7.44; 2.44	−13.15; -1.79	−15.65; 14.17	−23.99; 14.54	−3.06; 0.35	−5.04; 1.07	−2.08; 3.88	−3.34; 8.52
Age-adjusted model *
B	−1.51	−6.42	1.97	−1.82	−0.95	−1.45	0.58	2.03
95% CI	−6.32; 3.31	−11.98; -0.85	−12.63; 16.56	−20.98; 17.35	−2.58; 0.69	−4.47; 1.57	−2.39; 3.55	−3.89; 7.95
Fully adjusted model §
B	−3.13	−7.21	−7.44	−6.49	−1.03	−1.76	−0.37	1.67
95% CI	−8.90; 2.65	−13.64; -0.77	−24.46; 9.58	−26.27; 13.30	−2.91; 0.85	−4.26; 0.74	−4.07; 3.32	−2.26; 5.59

Note: The values of CIMT were imputed for the participants in whom CIMT was not measured at follow-up (n=52) and all variables in persons who were not willing to participate in the follow-up measurements (n=80).

^*Adjusted for age; § additionally adjusted for BMI, smoking, use of ACE-inhibitors, educational level, fat mass, diastolic blood pressure and systolic blood pressure.

Table 5.

Linear regression coefficients for intima media thickness (t0) with physical performance measures at baseline and at follow-up (n=327)

	Isometric grip strength (t0)	isometric grip strength (t1)	Leg strength (t0)	Leg strength (t1)	Physical Performance Score (t0)	Physical Performance Score (t1)	ADL - score (t0)	ADL - score (t1)
Crude model
B 95% CI	0.25 -0.51; 0.01	−0.40 -0.78; -0.02	−0.71 -1.48; 0.07	−0.39 -1.67; 0.89	−0.09 -0.19; -0.01	−0.11 -0.29; 0.08	0.11 -0.05; 0.27	0.07 -0.24; 0.39
Age-adjusted model *
B 95% CI	−0.19 -0.44; 0.06	−0.34 -0.71; 0.03	−0.57 -1.32; 0.19	−0.24 -1.51; 1.03	−0.07 -0.16; 0.02	−0.08 -0.26; 0.10	0.10 -0.06; 0.26	0.04 -0.27; 0.35
Fully adjusted model §
B 95% CI	−0.21 -0.55; 0.13	−0.34 -0.71; 0.04	−0.79 -1.78; 0.19	−0.65 -1.80; 0.54	−0.05 -0.16; 0.06	−0.10 -0.24; 0.05	0.02 -0.19; 0.24	0.11 -0.12; 0.34

Note: The values of CIMT were imputed for the participants in whom CIMT was not measured at follow-up (n=52) and all variables in persons who were not willing to participate in the follow-up measurements (n=80).

^*Adjusted for age; § additionally adjusted for BMI, smoking, use of ACE-inhibitors, educational level, fat mass, diastolic blood pressure and systolic blood pressure.

Discussion

In the present longitudinal study the relation between non-invasive atherosclerosis measures and physical functioning was investigated. After adjustment for confounders, higher baseline-CIMT was associated with lower isometric grip strength at follow-up. No associations were found between baseline-CIMT and the physical performance score, leg strength, or ADL-score at baseline and follow-up. In addition, no associations were found for the presence of plaques and physical functioning.

Several limitations of this study need to be acknowledged. Selective loss to follow-up of men with a worse physical or mental condition at baseline could have affected the associations. Participants had higher leg extensor strength (107.1 vs. 97.8 Nm) and had a higher score on the mini mental state examination (27.4 vs. 26.9) compared to non-participants. However, participants and non-participants had comparable age, body mass index, smoking status, isometric grip strength and physical performance score at baseline. The imputation of the missing values of CIMT and physical functioning measures and missing data for variables of the persons who were not willing to participate in the follow-up measurements did not change the findings, suggesting that selective loss-to-follow-up did not play a major role.

Strength of the present study is the inclusion of different measures of physical functioning (i.e. grip strength, leg strength and the physical performance score). These measures also affect the ability to perform ADL, therefore the ADL-score was also studied (23). Atherosclerosis was measured both by intima media thickness and the presence of atherosclerotic plaques.

CIMT is a non-invasive, reproducible measure as a marker of atherosclerosis. In the present study, an increase in CIMT from 0.97 mm at baseline (mean age: 77.5 years) to 1.02 mm at follow-up (mean age: 81.7 years) was observed. In the Carotid Atherosclerosis Progression Study the mean CIMT for men aged 75 years was 0.914mm (25th percentile: 0.814mm to 75th percentile: 1.028mm) and for men of 85 years was 0.937mm (25th percentile: 0.83mm to 75th percentile: 1.208mm) (24, 25). The CIMT levels in our study were in line with this study, suggesting a representative sample of the general population with respect to non-invasive atherosclerosis measures. However, change in CIMT between follow-up and baseline was larger than expected from the literature (2.61mm (95% CI [-11.31; 6.09] per four year) vs 0.0147 mm (95% CI [0.0122; 0.0173] per year) (26).

In the present study, higher baseline-CIMT was associated with lower isometric grip strength at follow-up. Previous studies have shown that lower isometric grip strength is associated with longer length of stay in the hospital, higher probability of disability and death (27, 28). Hence in older persons higher grip strength is important for maintaining independent in ADL. No associations were found for the presence of atherosclerotic plaques and isometric grip strength. We hypothesized a stronger association between the presences of plaques and physical functioning, because atherosclerotic lesions represent a more advanced stage of non-invasive atherosclerosis measures. In approximately 70% of the study population more then 1 plaque was present in the common carotid artery, 11.4% had one plaque and 19% had no plaques. Compared to a previous study the percentage of persons with more than one plaque is much higher (69.6% vs. 28%) (29). The small number of persons in the lowest categories of plaques might have affected the power to detect associations.

In addition, no association was found between CIMT and the physical performance score. The physical performance score is a summary score of three subtests and therefore might be a more robust measurement of physical performance (18). In additional analysis, we investigated the associations between atherosclerosis and the subtests ability to maintain balance, walking speed and the chair stand test (data not shown), but atherosclerosis was not associated to any of the subtests either.

We did not observe an association between atherosclerosis and the ability to perform ADL. A possible explanation is that older persons who start to decline in their physical function may not report that they are having difficulty with ADL (yet) (30, 31). Before experiencing difficulty of performing a task, older persons may modify the methods they use or change the frequency of task performance which can also lead to underestimation of (physical) disability (3). We did not measure whether persons modified the methods they use to perform a task or frequency of performing a task, therefore we might have underestimated the level of disability.

Generalized atherosclerosis was measured in different segments of the carotid artery (i.e. internal and common carotid artery and bifurcation) (12, 29, 32, 33). In addition, both mean and maximum values are used to define wall thickness (12, 29, 32, 33). Maximum CIMT expresses a more advanced stage of non-invasive atherosclerosis measures and is associated with atherosclerotic plaques (34). Compared to maximum CIMT, mean CIMT is less sensitive to outliers (34). Data on maximum CIMT was not available. Although the use of both different measurements and different segments of wall thickness hamper valid comparisons between studies, in general associations between generalized atherosclerosis and physical functioning were reported (12, 29, 32, 33). A strong point of the previous studies are the large sample sizes (400 (33), 2572 (29), 2932 (12) and 4735 (32) persons, respectively) and thus more power to detect small associations. The relatively short follow-up time of four years and the inclusion of more independently living older men with cardiovascular disease (40%) might explain the contradicting results to previous studies.

A cross-sectional study suggested that associations of CIMT as well as plaques with maximum walking speed were only present in the participants in the lowest quartile of maximum walking speed (29). In additional analysis, only including participants in the lowest quartile of usual walking speed, we did not find an association between non-invasive atherosclerosis measures and physical functioning. Including vascular risk factors (i.e. hypertension, history of diabetes and hypercholesterolemia, smoking, bloodpressure and BMI) as potential confounders to the analysis did not change the results either.

The Cardiovascular Health Study (CHS) reported an association between higher maximum common carotid wall thickness and the probability of being frail (32). For that analysis, persons with stroke, Parkinsons disease, severe cognitive impairment (MMSE<18) and persons using Sinemet Aricept and anti-depressants were excluded. An additional analysis of our own data, excluding persons with these diseases or medications (n=22) did not change the results, suggesting that the inclusion of these persons does not explain the contradicting results. In addition, almost half of the participants (48.0%) in the CHS remained free of incident cancer, cardiovascular disease, chronic obstructive pulmonary disease, new and persisted physical disability or cognitive decline (32). This might indicate that more healthy older persons were included in previous studies and that the stage of non-invasive atherosclerosis measures is not comparable between the studies. The mean CIMT of the present study is higher compared to two other studies (0.97 ± 0.15 vs 0.69 ± 0.11 and 0.82 ± 0.2 mm) (29, 33). This difference can be explained by the lower mean age of the study population and the inclusion of more never-smokers and less people with cardiovascular disease included in previous studies (29, 33).

In conclusion atherosclerosis, as measured by higher carotid artery CIMT, is related to a lower isometric handgrip strength at follow-up, but no further associations with physical functioning were found in this longitudinal study among independently living older men. The present study contributes to insight in the aetiology of the pathway from subclinical atherosclerosis to functional decline. Screening of preclinical atherosclerosis can be important to maintain grip strength at older age.

Funding and conflict of interest

M.E.M. den Ouden and E.M.A. Arts are supported by grant: 60-61900-98-146 from the Netherlands organization for health research and development. The funding organizations played no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.