L-shaped association of serum 25-hydroxyvitamin D concentrations with cardiovascular and all-cause mortality in individuals with osteoarthritis: results from the NHANES database prospective cohort study

Osteoarthritis (OA) is a complex chronic disease that currently affects an estimated 250 million people worldwide [1]. OA involves the spine and peripheral joints, resulting in severe disability [2, 3], and numerous studies have shown that individuals with OA have a higher risk of cardiovascular disease (CVD) and death than the general population [4‐6]. Therefore, identifying modifiable factors to prevent OA complications and reduce mortality, particularly the risk of cardiovascular death, is critical.

25-hydroxyvitamin D [25(OH)D]—the primary storage form of vitamin D—is a kind of sterol and fat-soluble substance that promotes the absorption of calcium and phosphate [7]. Serum 25(OH)D deficiency is a common risk factor for a variety of diseases, including skeletal disease, CVD, hypertension, autoimmune disease, obesity, diabetes, chronic kidney disease, and depression [8‐11]. In patients with OA, lower levels of serum 25(OH)D are associated with greater knee pain and increased progression of radiographic OA [12]. Although epidemiological studies have demonstrated that vitamin D deficiency is related to an increased risk of CVD and mortality in the general population [13], the relationships between vitamin D status and all-cause and cause-specific mortality in OA patients have not been reported and require further clarification. Furthermore, randomized clinical trials and meta-analyses have found no beneficial effects of vitamin D supplementation on overall mortality and CVD outcomes [13‐15]. Because there is no agreement on optimal serum 25(OH)D concentrations in different populations, determining the optimal vitamin D concentration in OA patients is critical.

To fill these knowledge gaps, our primary objective was to prospectively investigate the association of serum 25(OH)D concentrations with all-cause and cause-specific mortality in a nationally representative sample of American adult patients with OA. The secondary goal was to recommend vitamin D supplementation in patients with OA by determining optimal serum 25(OH)D concentrations.

To our knowledge, this is the first prospective study with a relatively large sample size to examine the link between serum 25(OH)D levels and all-cause and CVD mortality in OA patients. We discovered L-shaped relationships between serum 25(OH)D levels and all-cause and CVD mortality in OA patients in this cohort study, implying that lower serum 25(OH)D levels were significantly associated with higher all-cause and CVD mortality within a certain range. These findings could lead to clinical and dietary recommendations for lowering all-cause and CVD mortality in OA patients.

Similar to other studies [24, 25], our data revealed that 64.32% of OA patients with serum 25(OH)D insufficiency, indicating that vitamin D deficiency was widespread in OA patients. In our research, OA patients with serum 25(OH)D < 25.00 nmol/L had higher all-cause and CVD mortality compared with those with serum 25(OH)D ≥ 25.00 nmol/L, and the L-shaped relationships were seen between serum 25(OH)D concentrations and the all-cause and CVD mortality, which was similar to the large group investigation from the general population in the UK [26]. This study finding was consistent with previous reports that indicated low serum 25(OH)D concentration was associated with an increased risk of all-cause and CVD death [26‐31]. The association was generally nonlinear, with mortality decreasing with an increase in 25(OH)D up to a certain point, beyond which mortality did not decline further. Concentration of 25(OH)D < 54.40 nmol/L in OA patients increased the risk of premature death, and 25(OH)D concentration < 27.70 nmol/L had a significant risk of CVD death. Fan et.al reported that the concentration of 25(OH)D < 60.00 nmol/L could increase the risk of CVD death and premature death [26], which seemed to indicate that the demand for vitamin D in different populations is inconsistent. However, there is no clear benefit of vitamin D supplementation for all-cause and CVD mortality in the general population [14, 15]. Because there was no further reduction in mortality when serum vitamin D concentration reached a certain threshold, L-shaped relationships between serum 25(OH)D concentrations and all-cause and CVD mortality could partially explain this phenomenon.

It is well known that vitamin D played an essential role in bone health and the pathophysiology of several chronic diseases, but the optimal concentration of serum 25(OH)D is still controversial. According to the Endocrinology Society and the International Osteoporosis Foundation, the optimal concentration of serum 25(OH)D in the general adult population should be at least 75.00 nmol/L for better health outcomes (including mortality) [23, 32, 33]. However, the American Institute of medicine believed that based on its key role in bone health, 50.00 nmol/L was sufficient [34]. A meta-analysis of 14 prospective cohort studies demonstrated that the risk of death reduced nonlinearly with the increase of serum 25(OH)D, and the optimal concentration was about 75.00–87.50 nmol/L [35]. The results were inconclusive due to differences in sample size, follow-up time, and target population. In our current study, we found thresholds for CVD mortality of 27.70 nmol/L and all-cause mortality of 54.40 nmol/L in the OA population. Whether serum 25(OH)D concentration ≥ 54.40 nmol/L can decrease the overall premature death and cardiovascular risk of the OA population needs to be confirmed in future clinical trials.

There is controversy regarding the relationship between vitamin D and cancer mortality [36]. A meta-analysis of 12 prospective cohorts reported that higher baseline serum 25(OH)D levels were associated with a lower risk of cancer mortality [36, 37]; however, other prospective studies revealed no association between serum vitamin D levels and cancer mortality [38, 39]. There was no significant relationship between serum 25(OH)D levels and cancer mortality in OA patients in our study. Larger prospective studies are needed to investigate the link between vitamin D and cancer mortality.

The identification of populations at higher risk of all-cause mortality in patients with OA might have significant public health implications [4]. Based on the stratified analysis, we observed the advantage in all-cause mortality from higher serum 25(OH)D (≥ 54.40 nmol/L) in patients with OA, particularly in elderly patients (≥ 60 years old) and White patients. Previous studies demonstrated that the prevalence of severe vitamin D deficiency was 15- to 20-fold higher in African Americans compared to European Americans [40]. Our study also revealed that 44.00% of non-Hispanic Blacks with serum 25(OH)D below 25.00 nmol/L in the OA patients, however, vitamin D deficiency appeared to be more harmful in Whites patients compared with non-Whites patients. A study of the White population in the United States found that vitamin D deficiency was associated with an increased risk of stroke death, but not with Blacks [41]. One possible explanation is that Blacks have adapted to be relatively resistant to the negative effects of 25(OH)D deficiency [42]. Furthermore, similar to previous research [43], our study supported a higher risk of death in the elderly population with low vitamin D. As a result, when managing OA patients, we should consider monitoring vitamin D levels during clinical review and supplementing accordingly, particularly in an elderly patient (≥ 60 years old), and White patients.

Several possible mechanisms could explain the link between lower 25 (OH) D levels and an increased risk of death. Low vitamin D levels were associated with a variety of diseases which contributed to deaths, such as obesity and dyslipidemia, hypertension, and diabetes mellitus [44, 45]. Low levels of vitamin D were also associated with CVD, such as coronary artery calcification, increased carotid intima-media thickness, and elevated serum triglyceride levels [46‐48]. Secondary hyperparathyroidism associated with low level of vitamin D is reported to increase the risk of cardiovascular mortality. Secondary hyperparathyroidism activated renin-angiotensin-aldosterone system, elevated blood pressure, activated systemic inflammatory response and macrophage inflammatory response, promoted cholesterol accumulation in the vascular wall, and aggravated atherosclerosis [49‐51]. In the present study, we further discovered that the lower 25(OH)D levels were primarily associated with increased CVD mortality. This might be because participants with lower 25(OH)D levels seemed to have poor profiles (high cardiovascular risk, comorbidity, unhealthy lifestyle, and low education). Therefore, future studies are needed to clarify the potential mechanism of vitamin D in the prevention of all-cause and CVD mortality.

There are some benefits in this study. First, we used a nationally representative sample of adults with OA in the United States, which had a relatively large sample size and helped generalize our results. Furthermore, the number of cardiovascular events and deaths in long-term follow-ups provided sufficient strength for the analysis in the present study. Second, by adjusting for socioeconomic status, diet and lifestyle factors, comorbidity, and other potential confounding factors, we could improve the effectiveness of the conclusion. Finally, the serum 25 (OH) D concentrations in the NHANES database were measured using a standard method, ensuring the data analysis reliability.

There are some limitations to our study. First, since this was an observational study, cause and impact could not be determined. Second, although a single measurement of serum 25 (OH) D was a reasonable proxy for evaluating vitamin D status [52], the current study measured serum 25 (OH) D concentrations only once at baseline, which would underestimate the correlation of interest [53]. Third, our study, like other observational studies, could not rule out residual or unknown confounding or accidental confounding effects due to measurement errors and unmeasured variables (i.e., psychosocial stress or genetic susceptibility).

After multivariable adjustment, lower serum 25(OH)D levels were significantly and nonlinearly associated with higher risks of all-cause and CVD mortality in American adults with OA. We further discovered L-shaped associations between serum 25(OH)D levels and all-cause and CVD mortality plateaued at 54.40 nmol/L for all-cause mortality and at 27.70 nmol/L for CVD mortality. The thresholds of 27.70 and 54.40 nmol/L might represent intervention targets for lowering the risk of premature death and cardiovascular disease. These results highlight the potential advantages of monitoring and evaluating vitamin D status in the prevention of CVD and mortality among adults with OA.