WASHINGTON, DC — New research shows that that younger, overweight stroke patients have a lower 10-year mortality compared with normal-weight survivors and that compared with nonsmokers, those who smoke have a lower in-hospital mortality rate after an acute ischemic stroke, whether or not they received intravenous tissue plasminogen activator (IV tPA).
Research exploring the obesity paradox and the smoking paradox was presented in separate reports here at the American Academy of Neurology (AAN) 67th Annual Meeting.
In one presentation, Hugo Javier Aparicio, MD, vascular neurology fellow, Department of Neurology, Boston University School of Medicine, and an investigator in the Framingham Heart Study, said the obesity paradox has been observed in myocardial infarction, congestive heart failure and end stage kidney disease, as well as in stroke.
"Obese and overweight patients have been shown to have lower mortality and also suffer lower stroke recurrence," he said.
The reasons for this paradox are unclear, but some experts have questioned whether there's some physiologic advantage to being obese, of having excess fat storage or a "metabolic reserve" at the time of an illness, said Dr Aparicio.
Selection Bias
Other experts have focused on confounding or selection bias possibly being at play, he said. For example, obese patients may present earlier, or get treated at a younger age.
For this new analysis, Dr Aparicio and his colleagues used a sample from the Framingham Study, both the original and the offspring cohorts. They used the following weight categories: underweight: body mass index (BMI) less than 18.5; normal weight, BMI 18.5 to less than 25; overweight, BMI 25 to less than 30 and; obese 30 or greater.
They also subcategorized the groups into low overweight (BMI, 25 to 27.5), high overweight (BMI, 27.5 to 30), mildly obese (BMI, 30 to 32.5), and very obese (BMI, 32.5 or greater).
The analysis included 677 participants with stroke and 2031 controls without stroke matched for age, sex, and BMI. Patients with stroke were more likely to smoke, have other cardiovascular risk factors, and be receiving blood pressure and diabetes treatment before inclusion in the trial.
In both the stroke and controls group, 30% were normal weight, 44% were overweight, and 26% were obese.
No patients with stroke were underweight at the examination before their stroke.
Using a multivariable Cox proportional hazard model with mortality as outcome and adjusting for age, sex, time between examinations, event date, education, marital status, and smoking, researchers found no difference between the weight groups in the 1-year mortality rate among stroke patients.
However, compared with the normal BMI group, there was a significant difference in 10-year mortality in the overweight stroke group (hazard ratio [HR], 0.70; 95% confidence interval [CI], 0.54 - 0.89; P = .004). The difference did not reach significance for those who were obese (HR 0.80, 95% CI0.60-1.05; P = .111).
In controls, 10-year mortality did not significantly differ in the various weight groups.
The effect seen in the overweight group was maintained in both the low and high overweight groups. There was also a significant difference in the mildly obese (HR, 0.60; 95% CI, 0.42 - 0.86; P = .005) but not in the very obese.
In the control population, although none of the results were statistically significant, the very obese showed a trend toward worse mortality (HR, 1.20; P = .179).
Dr Aparicio reported that there was no interaction between weight groups and stroke severity or stroke subtype (eg, lacunar) among the stroke patients. Further, among the cases and controls, there was no interaction between weight groups and sex or smoking affecting 10-year mortality, he said.
Interaction for Age
However, there was a significant interaction for age. In stroke patients younger than age 70 years, a higher BMI (25 or more) on 10-year mortality was protective (HR, 0.52; 95% CI, 0.34 - 0.80; P = .003 compared with normal weight), but this was not the case for controls (HR, 1.53; 95% CI, 0.94 - 2.48; P = .85).
"This was a good exploratory study," commented Dr Aparicio. "I would like to now focus back on the stroke cases and look at other more subtle differences between the stroke cases in the different weight groups."
He stressed that further research is needed to understand how weight affects mortality and whether types of fat and fat distribution play a role in the obesity paradox.
"In the Framingham study, we have a lot of biomarkers and a lot of imaging markers. I would be interested to see how lipid hormones and other circulating biomarkers might be cardioprotective," he said.
Asked about the causes of death in the various weight groups, Dr Aparicio said these data are being collected. At 10 years, causes of death could be "wide ranging both in controls and in patients."
Following his presentation, Dr Aparicio elaborated on what might be causing the obesity paradox. He said he and his colleagues have already found no difference in terms of cancer prevalence and will be further investigating cardiovascular morbidity. "I think that's where some of these answers may lie, particularly in the younger participants," he said.
Today, about 35% of the US population is obese and by the year 2020, it's estimated that 75% of the population will be overweight or obese, he said.
Smoking and Stroke
In a separate presentation, Vishal Jani, MD, interventional neurology fellow, Michigan State University, East Lansing, presented results of a study showing that current or former smokers have a lower adjusted in-hospital mortality from acute ischemic stroke (AIS).
Researchers used information from 2000 to 2011 from a nation-wide inpatient database. They isolated a cohort of patients with AIS using relevant codes, stratified the cohort according to whether they received IV tPA, and then looked at in-hospital mortality among smokers vs nonsmokers.
There were 5,206,102 AIS hospitalizations. Among these, 132,157 (2.6%) patients received tPA.
The in-hospital mortality rate without tPA was 6.36% for nonsmokers compared with 2.83% for smokers (P < .001). The in-hospital mortality rate with tPA was 10.7% for nonsmokers compared with 6.8% for smokers (P < .001).
After adjustment for age, sex, ethnicity, primary payer, median household income, hospital region, teaching status, location, and bed size, among other factors, the odds ratio (OR) for the group with IV tPA was 0.81 (95% CI, 0.71 - 0.91; P < .001). For the non–IV-tPA group, the adjusted OR was 0.62 (95% CI, 0.60 - 0.64; P < .001).
"It's interesting that the finding of lower mortality is the same whether they get IV tPA or not," he commented.
As with obesity, the mechanism of smoking protection is up for debate. Most likely, said Dr Jani, it's due to the "cumulative effect of reconditioning."
The authors have disclosed no relevant financial relationships.
American Academy of Neurology (AAN) 67th Annual Meeting. Abstracts S5.005, S5.006. Presented April 21, 2015.
Medscape Medical News © 2015 WebMD, LLC
Send comments and news tips to news@medscape.net.
Cite this: New Insights Into Obesity and Smoking Stroke Paradoxes - Medscape - Apr 23, 2015.
Comments