TV viewing may not be risk factor for venous thromboembolism, study shows
Click Here to Manage Email Alerts
TV or not TV: The relationship between viewing it and venous thromboembolism risk, when accounting for physical activity and other risk factors, is the question.
In a study published in Journal of Thrombosis and Haemostasis, researchers hypothesized that more TV viewing would correlate with higher VTE risk, with physical activity attenuating the association.
However, the REGARDS (Reasons for Geographic and Racial Differences in Stroke) study, which included a contemporary cohort of people in the United States, showed no evidence of interaction among TV viewing, physical activity and risk for VTE.
The findings appear to be an outlier in much of the research within the space examining VTE and physical activity in the last 5 years.
“It is important to read the medical literature as a whole, rather than focusing on one particular study,” Neil A. Zakai, MD, MSc, hematologist at University of Vermont Medical Center and associate professor at Larner College of Medicine at University of Vermont, said in an interview with Healio. “Most of the time, it is possible to find a study that may support a particular point. However, it is key to look at everything available and develop an opinion based on the entirety of the literature and not just one study.”
REGARDS study
Zakai characterized prior literature on TV viewing and VTE risk as “sparse” but said previous studies had raised interesting points, including the finding that physical activity does not mitigate the effect seen in the Atherosclerosis Risk in Communities cohort. That motivated Zakai and colleagues to focus on determining the association of TV viewing with incident VTE risk and whether the relationship is modified or confounded by physical activity.
Their analysis included 30,239 participants aged 45 years or older (55% women, 43% Black) who participated in the REGARDS study from 2003 to 2007. Those with self-reported VTE prior to baseline or lack of baseline data or follow-up were excluded.
Researchers surveyed participants for baseline TV viewing and followed them for VTE events. They characterized TV viewing as light (fewer than 2 hours per day), moderate (2-4 hours per day) and heavy (more than 4 hours per day), and classified physical activity as poor, intermediate or ideal based on weekly reported activity. They calculated hazard ratios for TV viewing and physical activity with adjustment for VTE risk factors.
Median follow-up was 5.06 years.
Results showed 214 VTE events during 96,813 person-years of follow-up. Researchers found no association of heavy TV viewing with VTE risk in the unadjusted and fully adjusted model (adjusted HR = 0.92; 95% CI, 0.62-1.36). They noted a trend of ideal vs. poor physical activity toward reduced VTE risk (HR = 0.71; 95% CI, 0.51-1.01) but observed no evidence of an interaction between TV viewing, physical activity and risk of VTE.
“We hypothesized that TV viewing would be associated with greater VTE risk,” Zakai said. “We did not think it was causal, but that it was a marker of behaviors that may be associated with VTE risk. Based on the prior literature, we were a little surprised [by our findings]. However, the findings are valid within the context of the study, and it is just a reminder that we need to confirm results in multiple studies and populations before we make definitive conclusions.”
Other studies
The results from Zakai and colleagues differ from those of other studies, including two within the past 5 years that also looked at TV viewing specifically.
In a study published in 2018 in Journal of Thrombosis and Thrombolysis, Kubota and colleagues obtained information on the frequency of TV viewing among 15,158 participants aged 45 to 64 years in the Atherosclerosis Risk in Communities Study who did not have VTE prior to baseline. Results showed 691 VTE events during 299,767 person-years of follow-up.
Researchers reported “a positive dose-response relation” of TV viewing frequency with VTE incidence (P for trend = .036), with viewing TV “very often” carrying a 1.71 (95% CI, 1.26–2.32) times higher risk for VTE than “never or seldom” viewing TV. Additionally, they found even individuals who achieved recommended levels of physical activity who watched TV “very often” had 1.8 (95% CI, 1.04-3.09) times the risk for VTE than those who “never or seldom” watched.
Two years earlier, in a prospective, population-based study published in Circulation and involving 110,585 participants aged 40 to 79 years in Japan, Shirakawa and colleagues concluded that prolonged TV watching was a substantial risk factor for mortality due to pulmonary embolism. They calculated multivariable HRs of 1.7 (95% CI, 0.9-3) for participants who viewed 2.5 to 4.9 hours of TV per day and 2.5 (95% CI, 1.2-5.3) for those who viewed more than 5 hours a day vs. less than 2.5 hours a day.
“The results of this current study (by Zakai and colleagues) were surprising because the emerging evidence suggests that leisure sedentary activities such as prolonged TV viewing could increase the risk [for] VTE. But also, the results are not surprising, because of reasons that are specific to this study,” Setor K. Kunutsor, BSc, MBChB, MPhil, PhD, research fellow in evidence synthesis at Bristol Medical School of University of Bristol in the U.K., told Healio.
Kunutsor also has done extensive research into VTE and physical activity, including a meta-analysis of 14 prospective cohort studies comprising 1,286,295 participants and 23,753 VTE events published in European Journal of Epidemiology in 2019. In that analysis, Kunutsor and colleagues reported a pooled, fully adjusted RR of 0.87 (95% CI, 0.79-0.95) for VTE comparing groups with the most vs. least physical activity.
‘We all need to be very active’
Regarding the work of Zakai and colleagues, Kunutsor said the null findings could be attributed to factors such as the high burden of VTE risk factors in the REGARDS study, the low VTE events that could contribute to “low power” of the study, and cutoffs used for TV viewing “that may not have been very appropriate.”
“They used cutoffs of less than 2 hours a day, 2 to 4 hours a day, and 4 or more hours a day,” Kunutsor said. “It’s possible to watch TV for 4 hours a day and still be very physically active.”
Evidence in favor of “low power” was the inability of the study to demonstrate an association between physical activity and VTE risk.
Kunutsor added that the vast majority of research suggests VTE decreases with regular physical activity and, thus, physicians should encourage patients to achieve their daily or weekly exercise goals.
“The message is pretty straightforward: immobility or sedentary behavior increases the risk [for] cardiovascular diseases, including VTE. So, we all need to be very active,” he said. “Physical activity has been shown over and over again to have a lot of health benefits. It prevents cardiovascular disease, diabetes and a lot of other diseases. And now, we’ve also shown that physical activity reduces the risk [for] VTE.
“People tend to snack in front of the TV, sit for long periods,” Kunutsor added. “So when you’re doing it, try to get some breaks, stand up and just stretch your legs a bit, that sort of thing. That’s what doctors should be telling their patients.”
In comparing his own work to that of others who investigated VTE and physical activity, Zakai said it is important to recognize that physical activity and physical inactivity are not the same risk factor. He concurred with Kunutsor that just because a person watches a lot of TV doesn’t necessarily mean they are also inactive or not exercising enough.
“Television watching in and of itself is not a cause for many adverse health outcomes,” Zakai said. “Someone who gets on a stationary bike for an hour or 2 every day and watches television vs. someone who sits immobile and (eats) snacks are two disparate examples. Our study suggests that TV viewing is a poor surrogate for asking about physical activity and inactivity.”
References:
Munger JA, et al. J Thromb Haemost. 2021;doi:10.1111/jth.15408.
Kubota Y, et al. J Thromb Thrombolysis. 2018;doi:10.1007/s11239-018-1620-7.
Kunutsor SK, et al. Eur J Epidemiol. 2020;doi:10.1007/s10654-019-00579-2.
Shirakawa T, et al. Circulation. 2016;doi:10.1161/CIRCULATIONAHA.116.023671.
For more information:
Setor K. Kunutsor, BSc, MBChB, MPhil, PhD, can be reached at Bristol Medical School, University of Bristol, Learning & Research Building (Level 1), Southmead Hospital, Bristol, UK BS10 5NB; email: skk31@cantab.net.
Neil A. Zakai, MD, MSc, can be reached at Laboratory for Clinical Biochemistry Research, University of Vermont, 360 South Park Drive, Colchester, VT 05446; email: neil.zakai@med.uvm.edu.
Click Here to Manage Email Alerts