In this retrospective cohort study of 9,635 hypertensive patients, our key finding was the non-linear association between LDL-C and all-cause and cardiovascular mortality. After adjusting confounders of age, gender, race, marital status, education level, smoking, body mass index, systolic and diastolic blood pressure, and medication use, the RCS curve based on multivariate adjusted cox regression well revealed the correlation between them on a continuous scale. Distinguishing from the traditional impression that lower LDL-C level were better for health, we found both low and high LDL-C levels contributed to increased risk of death in hypertension population. As to the lowest risk of all-cause and cardiovascular mortality, we had preliminary view that the LDL-C concentration of slightly higher than 2.89mmol/L may be optimal in hypertensive patients according to RCS curve. These new findings may provide some reference for lipid control in hypertensive populations.
As the major culprit in development of atherosclerosis, the elevated levels of LDL-C are strongly associated with cardiovascular disease with no doubt. According to statistics from the World Health Organization report in 2021, cardiovascular disease causes 17.9 million deaths in 2019, making up 32% of total global deaths 24, so it can be easily understood that higher levels of LDL-C accompanied by higher risk of death. But when come to the lower LDL-C levels, the higher death risk seems incomprehensible. For this counterintuitive result, there are several probable explanations. First, it is hypothesized that debilitation and disease can lead to lower cholesterol levels 17,25,26 and, in this study, patients with lower levels of LDL-C had an older age (Q1: mean age of 62.2 ± 16.0, Q2: mean age of 59.5 ± 16.2) than higher level groups (Q4: mean age of 58.5 ± 15.4, Q5: mean age of 58.3 ± 14.6) indeed. Individual comorbidity profiles were not included in our study, but can be inferred from individual medication histories, the low-level groups had higher percentages of medication use. Second, although most studies had spared no effort to emphasize the benefits of lipid lowering, the long-term safety and efficacy of LDL-C lowering therapies remains a question to be further explored 27. Moreover, a number of studies have reported neurocognitive deficits, hemorrhagic stroke, and new-onset diabetes in the presence of reduced LDL-C 27–29, which may invariably increase the risk of all-cause mortality. Third, Kaysen GA et al found higher LDL-C was significantly associated with lower infection-related mortality an international retrospective cohort study 13, in other words, the risk of infectious death may increase when LDL-C at a low level, and so of all-cause mortality. Finally, as the world's second most common cause of death, cancer was related to low LDL-C levels, which is repeatedly mentioned in multiple studies 30–33. In total, reduced LDL-C levels might elevate the likelihood of mortality from the possible reason above, which and then results in increased all-cause mortality.
Consistent with our study result, some previous studies conducted in other populations have demonstrated a correlation between LDL-C levels and the risk of all-cause and cardiovascular mortality. Zhen Zhou et al. and Vale ́rie Tikhonoff et al. had explored relationships in older people. Zhen Z et al. reported there was a U-shaped relationship between untreated LDL-C level and all-cause mortality 34 and Tikhonoff V et al. found LDL-C concentration is a multifaceted risk factor in older adults 35. Chang C-H et al. demonstrated both lower and higher levels of mean LDL-C were associated with increased all-cause and cardiovascular mortality in type 2 diabetes patients through 36. Also, multiple studies in general population had got the similar result 17,37–39. Additionally, through a prospective cohort study of 108 243 individuals in Denmark, Johannesen CDL et al. found the lowest risk of all-cause mortality were at concentrations of LDL-C of 3.6-3.7 mmol/L 17. However, from our current study, lipid controlling in hypertensive populations should be even more strict.
Study Strengths and Limitations
Thanks to the ongoing NHANES project and continued data collection, we were able to build such a large sample size cohort of hypertensive people for our analysis. No individuals lost to follow-up and the cause of death of every participant was recorded on the National Death Index death certificate records. As far as our knowledge extends, relationship between low-density and all-cause and cardiovascular mortality remains controversial, a few studies explored it in general or other populations, but specifically in hypertensive populations, our study may be the first attempt to do so. Another strength of our study is that we adjusted for several confounders which may influence the accuracy of analysis results.
However, limitations also should be considered. First, the population we included were only living in United States, other countries or ethnicities may be not applicable. Second, several variables we included such as smoking status and medicine use were may cause recall bias, because they were subjective from participants. Third, we did not consider changes in LDL-C concentration over time or changes influenced by the initiation or cessation of lipid-lowering treatment throughout the observation period, this may make the findings unreliable. Finally, given the observational nature of the study, causality cannot be definitively established. Therefore, it is imperative to interpret the findings with caution, considering both potential causal and reverse relationships. Subsequent research is warranted to elucidate the possible causal link between LDL-C levels and mortality.