Acute Coronary Syndrome
Obesity Paradox in Patients With Acute Myocardial Infarction
2022 Volume 86 Issue 4 Pages 640-641
Details
2022 Volume 86 Issue 4 Pages 640-641
Obesity is a major global public health concern, with over 60% of populations in Western countries classified as overweight or obese based on body mass index (BMI) criteria. In general, obesity is a major exacerbating factor for several types of disease, with a higher incidence of comorbidities such as hypertension, diabetes, and dyslipidemia. Therefore, the prevalence of several types of diseases, not surprisingly, increases markedly in obese patients. Nevertheless, numerous studies have documented the obesity paradox, in which overweight and obese patients have a better prognosis than lean or underweight patients. Interestingly, the obesity paradox is observed in various diseases, not only cardiovascular diseases (CVD) but also non-CVDs such as endstage renal disease, advanced cancers, chronic obstructive pulmonary disease, rheumatoid arthritis, and human immunodeficiency virus infection.1
Article p 632
Although Japan has one of the lowest rates of obesity globally, the size of the population with lower BMI values is relatively high, with the incidence of low BMI (<20 kg/m2) consistently reported to be 16–18% in people aged ≥65 years.2 In general, there is a close relationship between low BMI and advanced age. Moreover, lower BMI values are related to low lean body mass, a condition also known as sarcopenia.3 Patients with sarcopenia have limited exercise capacity and reduced mobility, both conditions associated with increased total mortality,4 cognitive decline, and requirement for caregiver support. Considering that Japan is the aging society without parallel in the world, it is crucial to evaluate the prognostic impact of lower BMI in patients with CVD, which is the second most common cause of death in Japan.
In this issue of the Journal, Yokoyama et al5 evaluate the impact of BMI on clinical outcomes after the onset of acute myocardial infarction (AMI) by post-hoc analysis of the data from the Japan AMI Registry (JAMIR). They found that their primary endpoint (all-cause death) was significantly higher in the Q1 group (BMI <18.5 kg/m2) than in the other 3 groups. Moreover, the secondary endpoint (composite of CVD-related death, non-fatal MI, and non-fatal stroke) was also significantly highest in the Q1 group. However, as has been often seen in previous studies, the BMI data appear to be confounded by several other factors. The authors report that obese patients had more cardiovascular risk factors, such as hypertension, hyperlipidemia, and diabetes mellitus, consistent with the results of most previous studies.6–8 However, despite such adverse clinical profiles, obese patients had better clinical outcomes. In contrast, patients with low BMI had fewer cardiovascular risk factors but the highest rates of all-cause death and CVD-related death. The Q1 group (lean patients) was more often elderly and female. In their study, advanced age was independently associated with increased mortality. Many studies have shown a clear relationship between advanced age and poor clinical outcomes after AMI, irrespective of reperfusion modality.9,10 Conversely, the average age of the Q4 group in the present study was 20 years younger than that of the Q1 group.
In addition to age and sex differences, the authors report multiple baseline differences among the groups. The Q1 group had a significantly higher proportion of patients with Killip class IV (shock status), lower average ejection fraction in the acute phase, and longer door-to-balloon time than the other groups. Additionally, the proportion of statin use during hospitalization was significantly lower. Because all these differences could significantly affect clinical outcomes after the onset of AMI and only some of them were included in the multivariate analysis, careful attention is needed before making definite conclusion on the relationship between lower BMI and clinical outcomes after the onset of AMI. Indeed, data on the relationship between BMI and clinical outcomes of patients with AMI are limited and inconsistent. Using the recent JACSS registry of patients with AMI, the authors demonstrated that the in-hospital mortality rate progressively decreased with increasing BMI. However, multivariate analysis showed that BMI itself had no effect on in-hospital death.8 Several other studies demonstrated that obese patients have a lower rate of in-hospital death than normal weight patients after primary PCI for AMI.6,7 However, all studies conclude that BMI itself does not contribute to in-hospital death. Thus, although the obesity paradox exists, in that overweight and obese patients with CVD seem to have a more favorable prognosis than leaner patients, it is fair to state that it remains unclear whether lower BMI has an effect on mortality or it just reflects several confounding factors. To address this, further studies are needed.
Although the mechanism underlying the relationship between lower BMI and higher incidence of clinical outcomes is uncertain, the difference at the time of or during hospitalization probably had the most prominent effect on the differences in clinical outcome, because most clinical events occurred within 3 months after the onset of AMI. Older age at presentation may have contributed to the survival disadvantage in the Q1 group in terms of other organ reserve and comorbidities. Moreover, less aggressive treatment because of advanced age (e.g., statins) may be a potential explanation for the worse survival rate in non-obese patients. Conversely, excess body weight may provide some protective effects for AMI-related death. A slight increase in BMI can be due to a preserved or increased lean mass, which is associated with better fitness and exercise capacity, improvement in metabolic profiles, and probably better prognosis in patients with CVD.11
Interestingly, Yokoyama et al demonstrate that BARC type 3 or 5 bleeding was an independent predictor of the primary endpoint (hazard ratio: 5.04, 95% confidence interval: 3.34–7.60, P<0.01). Furthermore, BARC type 3 or 5 bleeding was most common in the Q1 group, followed by Q2, Q3, and Q4, although it did not reach statistical significance (8%, 5%, 4%, and 2%, respectively; P=0.096). Because the primary purpose of the JAMIR is to examine ischemic and bleeding events in Japanese patients with AMI and the association between these events and antiplatelet therapy, the information on clinical events, including bleeding events, and antiplatelet therapy should be highly reliable. In this respect, Yokoyama et al provide new insights into the relationship between patients with lower BMI and future adverse events based on this reliable database. Although the antithrombotic therapy was not different among the BMI groups, similar antiplatelet strategy to younger patients with higher BMI might in turn lead to higher incidence of bleeding events, especially in elderly patients with lower BMI. Considering that BARC type 3 or 5 bleeding was independently associated with all-cause death, a less severe bleeding approach might be necessary for this category of patients.
Finally, from the clinical standpoint, the most meaningful question is whether weight change (gain, loss, or maintaining a normal weight), not baseline BMI, can affect the clinical outcomes of patients with AMI.11 Future studies are needed to determine whether increased BMI or better nutrition improves the clinical outcomes of AMI.
None.
Lecture fees from Abbott Vascular and Terumo.
