Alexander J. Rodríguez

Effect of vitamin D supplementation on measures of arterial stiffness: a systematic review and meta‐analysis of randomized controlled trials

Low vitamin D has been associated with poor arterial compliance in observational studies. Arterial stiffness has prognostic value for cardiovascular disease risk. The aim of this systematic review was to clarify the literature surrounding the use of vitamin D to ameliorate arterial stiffness.

Effects of vitamin D supplementation on inflammatory markers in heart failure: a systematic review and meta-analysis of randomized controlled trials

Vitamin D is reported to have anti-inflammatory properties; however the effects of vitamin D supplementation on inflammation in patients with heart failure (HF) have not been established. We performed a systematic review and meta-analysis examining effects of vitamin D supplementation on inflammatory markers in patients with HF. MEDLINE, CINAHL, EMBASE, All EBM, and Clinical Trials registries were systematically searched for RCTs from inception to 25 January 2017. Two independent reviewers screened all full text articles (no date or language limits) for RCTs reporting effects of vitamin D supplementation (any form, route, duration, and co-supplementation) compared with placebo or usual care on inflammatory markers in patients with heart failure. Two reviewers assessed risk of bias and quality using the grading of recommendations, assessment, development, and evaluation approach. Seven studies met inclusion criteria and six had data available for pooling (n = 1012). In meta-analyses, vitamin D-supplemented groups had lower concentrations of tumor necrosis factor-alpha (TNF-α) at follow-up compared with controls (n = 380; p = 0.04). There were no differences in C-reactive protein (n = 231), interleukin (IL)-10 (n = 247) or IL-6 (n = 154) between vitamin D and control groups (all p > 0.05). Our findings suggest that vitamin D supplementation may have specific, but modest effects on inflammatory markers in HF.

Pain, quality of life and safety outcomes of kyphoplasty for vertebral compression fractures: report of a task force of the American Society for Bone and Mineral Research†

The relative efficacy and harms of balloon kyphoplasty (BK) for treating vertebral compression fractures (VCF) are uncertain. We searched multiple electronic databases to March 2016 for randomized and quasi‐randomized controlled trials comparing BK with control treatment (nonsurgical management [NSM], percutaneous vertebroplasty [PV], KIVA VCF treatment system [Benvenue Medical, Inc., Santa Clara, CA, USA], vertebral body stenting, or other) in adults with VCF. Outcomes included back pain, back disability, quality of life, new VCF, and adverse events (AEs). One reviewer extracted data, a second checked accuracy, and two rated risk of bias (ROB). Mean differences and 95% confidence intervals (CIs) were calculated using inverse‐variance models. Risk ratios of new VCF and AE were calculated using Mantel‐Haenszel models. Ten unique trials enrolled 1837 participants (age range, 61 to 76 years; 74% female), all rated as having high or uncertain ROB. Versus NSM, BK was associated with greater reductions in pain, back‐related disability, and better quality of life (k = 1 trial) that appeared to lessen over time, but were less than minimally clinically important differences. Risk of new VCF at 3 and 12 months was not significantly different (k = 2 trials). Risk of any AE was increased at 1 month (RR = 1.73; 95% CI, 1.36 to 2.21). There were no significant differences between BK and PV in back pain, back disability, quality of life, risk of new VCF, or any AE (k = 1 to 3 trials). Limitations included lack of a BK versus sham comparison, availability of only one RCT of BK versus NSM, and lack of study blinding. Individuals with painful VCF experienced symptomatic improvement compared with baseline with all interventions. The clinical importance of the greater improvements with BK versus NSM is unclear, may be due to placebo effect, and may not counterbalance short‐term AE risks. Outcomes appeared similar between BK and other surgical interventions. Well‐conducted randomized trials comparing BK with sham would help resolve remaining uncertainty about the relative benefits and harms of BK.

Effect of weight loss induced by energy restriction on measures of arterial compliance: A systematic review and meta-analysis

We read with interest the comprehensive meta-analysis of the effect of weight loss induced by energy restriction, with and without adjunctive therapies, on various measures of arterial compliance, by Petersen and colleagues published in Atherosclerosis [1]. Overall, modest weight loss was associated with some improvement in the cardio-ankle vascular index, b-stiffness index, arterial compliance and distensibility, distal oscillatory compliance, proximal capacitance compliance, systemic arterial compliance and reflection time. Therewas no significant effect on augmentation index, augmentation pressure, pulse pressure or strain. We appreciate the varied nature of the literature in this area, which means that studies of different weight loss strategies (for example, diet only, diet and exercise, weight loss drugs or bariatric surgery) were grouped together to increase statistical power. However, in an attempt to explore this obvious source of heterogeneity, the authors conducted a series of sub-analyses in which the studies in each outcome were stratified by intervention type. For a number of the outcomes explored (distensibility, distal oscillatory compliance and proximal capacitance compliance), the effect of combined diet and exercise-induced weight loss was associated with improvements in these markers, whereas a diet-only approach showed no significant improvement. We would be interested in the Author’s opinions on the impact exercise is having on these measures of arterial stiffness, as in another sub-analysis, combined diet and exercise produced an even stronger reduction in pulse pressure ( 0.73 [ 0.99, 0.47]) compared to diet alone ( 0.47 [ 0.82, 0.11]), suggestive of an additive effect of exercise and that it could possibly be potentiating the beneficial effects of diet on weight loss. A Japanese study demonstrated that the change in cardio-ankle

Comments on Li et al.: Meta-analysis of hypertension and osteoporotic fracture risk in women and men

Dear Editor, We read with interest the meta-analysis by Li et al. on the association between hypertension and osteoporotic fractures [1]. There is growing interest in understanding the potential for interaction between bone and the vasculature. In the main analysis, hypertension (heterogeneously reported throughout the included studies and not defined by the authors) was associated with an odds ratio of 1.33 (95% confidence interval 1.25–1.40) for all fractures in men and women. However, several potential issues suggest these results should be interpreted with caution. Importantly, the authors state that that they Banalyzed 10 articles encompassing 28 independent studies, 1,430,431 participants, and 148,048 osteoporotic fracture cases.^ However, in the main analysis (Figure 2), it appears that effect estimates based on different fracture types from studies reporting data on any fracture, as well as on specific fracture types (e.g., hip or vertebral), were treated as independent studies. That is to say a hip or vertebral fracture would also have been included as an Bany fracture^ resulting in multiple case reporting for the one event. Indeed, for the study of Yang et al. [2], their methodology states that BAll types of fragility fracture were classified into three main groups: any fracture (at all sites), hip fracture, and clinical vertebral fracture^ confirming our concern. It is therefore not appropriate to say that analyses of different types of fractures within the same study are Bindependent studies.^ This warrants clarification. Furthermore, the methods stated that odds ratios from all studies were pooled together. However, some of the included studies were longitudinal in design and thus reported hazard ratios and these hazard ratios were also pooled [3]. Including hazard ratios in the main analysis is not appropriate because they are an examination of time to an event. In short, incidence is not the same as prevalence. Also, both crude and adjusted estimates were included in the main analysis, which again may not be appropriate as crude estimates are subject to confounding. However, some of these limitations have been addressed in sub-group analyses (e.g., stratifying by study type) which overall appear to support the main finding that fractures are highly prevalent in elderly people with hypertension. Finally, it appears reference 13 is incorrectly cited as the data from Table 1 cannot be found in the source listed in the reference list [4]. The data reported in Table 1 appears to be from a longitudinal study by an author of the same name reporting hazard ratios, rather than odds ratios [2]. The incorrectly cited Yang paper from the list of references (a retrospective case-control study) instead explored whether a history of osteoporotic fractures influenced the risk of hypertension [4], not the reverse hypothesis (that hypertension predicts increased fracture risk) as tested by Li et al. Overall, the article by Li et al. has highlighted that more research is needed to provide a greater understanding of potential biological mechanisms and in clarifying the bidirectional nature of the relationship between hypertension and osteoporosis. A response to these comments is available at https://doi.org/10.1007/ s00134-017-4245-3.

At what price decreased mortality risk

Dear Editor, There is continued interest in the future of vertebral augmentation procedures in clinical practice following the highly publicised ‘sham’ trials in 2009 [1, 2]. Because of this, the popularity of vertebral augmentation has been in decline, despite evidence that vertebral augmentation increases survival [3]. Procedure uptake and survival were investigated by Ong et al. in their study of Medicare data on mortality outcomes following these procedures [4]. Ten-year mortality risk was lower in those undergoing percutaneous vertebroplasty (VP) and balloon kyphoplasty (BKP) compared to non-surgical management (NSM). This finding, whilst supportive of a role for vertebral augmentation in clinical practice, may be biased by underlying socioeconomic factors that may ultimately govern decisions about opting for surgical or non-surgical management. Given the lack of information regarding these factors (owing to limited data richness fromMedicare data), wewould be interested in the author’s comments on how costs of treatments potentially influenced the outcomes noted. Is it quite possible that opting for surgical over non-surgical management may be a proxy for wealth, access to certain services and healthcare utilisation more generally? In support of this hypothesis, patients with lower socioeconomic status as well as those from southern states were at an increased mortality risk compared to those of higher socioeconomic status. Also, patients from other presumably wealthier regions (e.g. north-east states) had increased survival. A previous study indicated that use of domestic help following hip fractures had socioeconomic disparities supportive of a view that factors that promote beneficial outcomes may be accessed by those most able to afford them [5]. It is alarming though that reductions in mortality risks in the post-2009 (sham trials) era were attenuated. We speculate that the reduction in popularity of these procedures and subsequent increases in costs made these procedures unavailable to those who may not have the means to support themselves adequately post-surgery [6]. It was noted that length of stay in hospital was shorter for patients opting for NSM and that discharge to home was higher in these patients (avoiding inhospital costs?). As Ong et al. noted ‘home discharge rates appear to reflect a shifting of the NSM patients from inpatient to other facilities, and do not reflect recovery of the patients’. Discharge to care facilities was higher amongst BKP and VP patients, where the ratio of patients discharged to home relative to NSM was 2.27 (2.20, 2.35) and 1.86 (1.80, 1.93) for BKP and VP respectively. These data again are potentially reflective of greater healthcare utilisation amongst those more able to afford care. Overall, Ong et al. demonstrate that surgical augmentation of vertebrae to treat VCF may have a future in clinical practice. Subsequent trials have demonstrated what type of patient is most likely to benefit from such procedures based mostly on clinical symptoms prior to surgery [7]. Additional attention needs to be paid to other patient factors such as ability to access services that contribute to treatment success in the short and long terms. Understanding of these factors may result in more favourable outcomes which may promote procedure uptake.

Reply to: systematic review and meta-analysis for the association of bone mineral density and osteoporosis/osteopenia with vascular calcification in women

Dear Editor, We read with interest the study by Zhang and Feng on the association between bone mineral density (BMD) and vascular calcification in post-menopausal women. The results of their meta-analysis of four studies (n = 1666 vascular calcification cases and n = 1190 controls) demonstrated that both hip and spine BMD were significantly lower in women with vascular calcification. Further, in those with vascular calcification, there was an approximate four and a half-fold increased likelihood of having osteoporosis, and an approximate two-fold increased likelihood of having osteopenia. The authors conclude that patients with vascular calcification ‘have lower lumbar spine and hip BMD and increased risk for developing osteoporosis or osteopenia’. While it is possible that aspects of increased vascular calcification may contribute to declines in bone density (through common risk factors such as age, smoking, low levels of physical activity and chronic inflammation), this conclusion potentially ignores evidence that vascular calcification can also result from dysregulated bone metabolism in ageing. A longitudinal study in postmenopausal women has shown that progression of aortic calcification is greater in those who demonstrate bone loss, compared to those who experience no bone loss. In a Japanese sample of older men and women, ectopic calcium deposition in mitral and aortic valves was related to severe bone loss attributed to osteoporosis. Furthermore, studies conducted in patients with chronic kidney disease (a condition characterized by bone mineral derangement) demonstrate that vascular calcification is a common feature in this population. Additionally, important catabolic bone metabolites such as phosphate (hyperphosphatemia), calcium (hypercalcemia) and fibroblast growth factor 23 are all elevated in chronic kidney disease, further suggestive of a role for bone dysregulation. Thus, vascular calcification may well be a consequence of age-related declines in bone density, in addition to the possibility that cardiovascular diseases, including vascular calcification, contribute to bone loss. Other consequences of ageing may also explain the lower BMD of those with vascular calcification, particularly given that the included cross-sectional cohort study by El Maghraoui et al. (which contributed over 50% of participants to the metaanalysis), demonstrated that participants with abdominal aortic calcification were approximately 10 years older than those without calcification. A meta-regression accounting for age was not performed in the article by Zhang and Feng and thus one may speculate that the findings may be confounded by the effects of age. A previous study (not included in the present metaanalysis) of Japanese-American women has conversely reported no association between BMD and aortic calcification, and thus prospective studies investigating associations between bone loss and vascular calcification are needed to clarify the relationship between these conditions. Nevertheless, Zhang and Feng’s study highlights the increasing research interest in mechanisms that may lead to both osteoporosis and cardiovascular disease in ageing populations.

Lower muscle tissue is associated with higher pulse wave velocity: A systematic review and meta-analysis of observational study data

Muscle loss and arterial stiffness share common risk factors and are commonly seen in the elderly. We aimed to synthesise the existing literature on studies that have examined this association. We searched electronic databases for studies reporting correlations or associations between a measure of muscle tissue and a measure of arterial stiffness. Meta‐analysis was conducted using Fishers Z‐transformed r‐correlation (rZ) values. Pooled weighted rZ and 95% confidence intervals were calculated in an inverse‐variance, random‐effects model. Heterogeneity was assessed by the inconsistency index (I2). Study quality was assessed on a checklist using items from validated quality appraisal guidelines. 1195 records identified, 21 satisfied our inclusion criteria totalling 8558 participants with mean age 52±4 years (range 23‐74). Most studies reported an inverse relationship between muscle tissue and arterial stiffness. Eight studies had data eligible for meta‐analysis. Muscle tissue was inversely associated with pulse wave velocity in healthy individuals [rZ=−.15 (95% CI −0.24, −0.07); P=.0006; I2=85%; n=3577] and in any population [rZ=−.18 (−0.26, −0.10); P<.0001; I2=81%; n=3930]. In a leave‐one‐out sensitivity analysis, the results remained unchanged. Lower muscle tissue was associated with arterial stiffness. Studies were limited by cross‐sectional design. Cardiovascular risk monitoring may be strengthened by screening for low muscle mass and maintaining muscle mass may be a primary prevention strategy.