European Geriatric Medicine (2019) 10:863–870 https://doi.org/10.1007/s41999-019-00239-4 REVIEW Orthostatic hypotension and health outcomes: an umbrella review of observational studies Pinar Soysal1 · Nicola Veronese2 · Lee Smith3 · Gabriel Torbahn4 · Sarah E. Jackson5 · Lin Yang6 · Andrea Ungar7 · Giulia Rivasi7 · Martina Rafanelli7 · Mirko Petrovic8 · Stefania Maggi2 · Ahmet Turan Isik9 · Jacopo Demurtas10 · The Special Interest Groups in Systematic Reviews and Meta-analyses for Healthy Ageing, and Cardiovascular Medicine of the European Society of Geriatric Medicine (EuGMS) Received: 11 July 2019 / Accepted: 4 September 2019 / Published online: 7 November 2019 © European Geriatric Medicine Society 2019 Key summary points Aim To investigate potential relationships between orthostatic hypotension (OH) and negative health outcomes and mortal- ity, through an umbrella review with integrated meta-analyses. Findings Orthostatic hypotension is significantly associated with several negative outcomes in older people, but a sugges- tive evidence is available only for higher risk of coronary heart disease congestive heart failure, stroke, falls dementia, and all-cause mortality. Message Orthostatic hypotension seems to be significantly associated with several negative health outcomes in older people, even if only associations with coronary heart disease, congestive heart failure, stroke, falls, dementia, and all-cause mortality are supported by suggestive evidence. Abstract Purpose Orthostatic hypotension (OH) is associated with older age and many negative clinical outcomes in geriatric practice. We aimed to capture the breadth of outcomes that have been associated with the presence of OH and systematically assess the quality, strength and credibility of these associations using an umbrella review with integrated meta-analyses. Methods We systematically searched several major databases from their commencements through to 16th May 2019 for meta- analyses of observational studies of OH and any health-related outcome. We used these metrics to categorize the strength of evidence of significant outcomes (p < 0.05) from class I (convincing) to class IV (weak), according to the pre-established criteria. Results From 975 abstracts, seven meta-analyses of 12 outcomes were included. For each outcome, the median number of studies was four, and the median number of participants was 46,493, with a median of 3630 incident cases. There was sug- gestive (class III) evidence that OH was associated with significantly higher risk of coronary heart disease (HR = 1.32, 95% CI 1.12–1.56), stroke (HR = 1.22, 95% CI 1.08–1.38), congestive heart failure (HR = 1.30, 95% CI 1.09–1.55), all-cause mortality (RR = 1.50, 95% CI 1.24–1.81), falls (OR = 1.84, 95% CI 1.39–2.44), and dementia (HR = 1.22, 95% CI 1.11–1.35). Conclusion The current evidence base indicates that OH is significantly associated with a range of adverse cardiovascular, cognitive, and mortality outcomes in older people, although the strength of this evidence remains only suggestive. Further research in larger samples and with lower risk of bias is required to build a fuller picture of the impact of OH on health. Keywords Orthostatic hypotension · Umbrella review · Meta-analysis · Mortality · Fall · Heart failure · Heart disease · Stroke Pinar Soysal and Nicola Veronese: joint first author. Introduction Ahmet Turan Isik and Jacopo Demurtas: joint senior author. Orthostatic hypotension (OH) diagnosis is often defined Electronic supplementary material The online version of this as a drop of at least 20 mmHg in systolic BP (SBP) and/or article (https ://doi.org/10.1007/s41999 -019-00239 -4) contains 10 mmHg in diastolic BP (DBP) upon the change in position supplementary material, which is available to authorized users. (from sitting to standing) [1]. The prevalence of OH increases Extended author information available on the last page of the article with age and is estimated to be 10–30% in older adults. It is Vol.:(012 3456789) 864 European Geriatric Medicine (2019) 10:863–870 important to note that different methods used to measure OH OH, according to The Consensus Committee of the Ameri- have produced different prevalence estimates [2–4]. Reasons can Autonomic Society and the American Academy of Neu- for the increase in prevalence of OH with age include an age- rology, and (2) meta-analyses of observational studies (lon- related decrease in renin–angiotensin aldosterone level, cardiac gitudinal or case–control) that investigated the association hypertrophy, and deficiency in arterial baroreflex sensitivity of OH with any health-related outcome (e.g., cardiovascular and vasomotor control, all of which make the management events, falls, depression, cognitive impairment, mortality). of postural blood pressure increasingly difficult with age [5]. OH was defined as a drop of at least 20 mmHg in systolic A number of studies have reported associations between BP (SBP) and/or 10 mmHg in diastolic BP (DBP) upon the OH and increased risk of adverse clinical outcomes, includ- change in position [1]. Both the active standing test and ing cardiovascular events and stroke [6], recurrent falls syn- head-up tilt table test for measuring blood pressure were cope and consequent injuries [7], cognitive impairment [8], accepted. impaired sleep quality [9], and depression [10]. However, no attempt has been made to synthesize the literature on the Data extraction health risks associated with OH or critically evaluate the strength of the available evidence. A better understanding Two independent investigators (PS, JD) extracted the follow- of the full spectrum of health risks associated with OH is ing information for each article: (1) first author name; (2) important for geriatric practice. OH has been shown to be year of publication; (3) journal; (4) the number of included significantly associated with older age, polyurinary inconti- studies and the total number of the people included in the nence, frailty, and functional impairment in daily life activi- review; (5) the inclusion criteria for studied population; ties. OH can, therefore, be considered as a new geriatric (6) the definition used for OH; (7) the effect size used in syndrome [11]. the review; (8) study design (case–control, longitudinal); Therefore, the present study aimed to capture the breadth (9) number of cases (i.e., people having the event of inter- of outcomes that have been shown in observational stud- est, e.g., falls) and controls (i.e., people without events) for ies to be associated with OH and systematically assess the each study; and (10) setting. Disagreements were resolved quality, strength and credibility of these associations. We through consensus with another independent reviewer (NV). used an umbrella review with integrated meta-analyses [12] We then extracted the study-specific estimated relative risk to combine evidence from a wide range of outcomes and for each health outcome (risk ratio [RR], odds ratio [OR], populations. hazard ratio [HR], mean difference [MDs]), along with the associated 95% confidence interval (CI). If two meta-anal- yses were available for the same outcome, we included the Materials and methods largest in terms of studies. The present umbrella review followed a structured protocol Outcomes (available upon request from the corresponding author) that was pre-registered in PROSPERO as CRD 42019126423. Any health-related outcome (e.g., cardiovascular events, (https ://www.crd.york.ac.uk/prosp ero/displ ay_recor falls, depression, cognitive impairment, mortality and oth- d.php?Record ID=12642 3). ers) was included. Data sources and searches Methodological quality of systematic reviews We searched several databases (Epistemonikos, MEDLINE The methodological quality of the included meta-analyses through Ovid, CINAHL, EMBASE, Cochrane library and was assessed using ROBIS. The ROBIS is completed in three JBI Database of Systematic Reviews and Implementation phases: (1) assess relevance (optional), (2) identify concerns Reports) from their inception through to 16th May 2019. with the review process, and (3) judge risk of bias. Phase 2 The search strategy used in MEDLINE is reported, as an covers four domains through which bias may be introduced example, in Supplementary Table 1. Moreover, we hand into an each systematic review of the umbrella review: study searched the reference lists of included articles. No language eligibility criteria; identification and selection of studies; data restrictions were applied. collection and study appraisal; and synthesis and findings. Phase 3 assesses the overall risk of bias in the interpretation Study selection of review findings and whether this considered limitations identified in any of the phase 2 domains. Signaling questions In this umbrella review, we included: (1) systematic reviews are included to help judge concerns with the review process with meta-analyses that included people with a diagnosis of (phase 2) and the overall risk of bias in the review (phase 3); 1 3 European Geriatric Medicine (2019) 10:863–870 865 these questions flag aspects of review design related to the Meta‑analyses of included studies potential for bias and aim to help assessors judge risk of bias in the review process, results, and conclusions. Each item can Table 1 summarizes the main findings of our umbrella be scored from low to high risk of bias [13]. review. For each outcome, the median number of studies was four, and the median number of participants was 46,493, Statistical analysis with a median of 3630 incident cases. All the studies focused on the general population as the For each meta-analysis, we re-calculated the summary effect population of interest, and all were cohort studies. Four out- size and its 95% CI, using random-effects models [14]. Next, comes related to cardiovascular diseases, four were cognitive the 95% prediction interval was estimated which further outcomes and the other four outcomes regarded falls and accounts for between-study effects and estimates the certainty mortality, including specific cause deaths. of the association if a new study addresses that same asso- Supplementary Table 3 reports the assessment of the ciation [15]. For the largest study of each meta-analysis, we quality of the meta-analyses included, showing that these evaluated whether this was statistically significant. Hetero- works (with the exception of two) had a low risk of bias, geneity was estimated using the I2 metric, with values > 50% according to the ROBIS. Supplementary Table 4 shows the indicative of high heterogeneity, and values > 75% suggesting main results of included primary studies of each meta-anal- very high heterogeneity [16, 17]. In addition, we calculated ysis. The excluded studies with reason are shown in Sup- the evidence of small study effects. In this regard, we used the plementary Table 5. regression asymmetry test [18], using a p value < 0.10 [19]. Overall, 10/12 studies (83%) reported significant sum- Finally, we applied the excess of significance test [20] which mary results (p < 0.05), as shown in Table 1. Half of the evaluates whether the number of studies with nominally sig- outcomes (6/12) reported significant heterogeneity, as I2 nificant results (i.e., with p < 0.05) among those included in a ≥ 50% and, of them, two reported a very high heterogeneity meta-analysis is too large based on the power that these data (I2 ≥ 75%). For one outcome (falls), we observed a small sets have to detect effects at α = 0.05. The number of expected study effect, while the excess significance bias was present ‘positive’ (E), i.e., statistically significant studies, was com- in 3/12 outcomes included. The largest study, in terms of pared with the observed (O) number of statistically significant participants, was statistically significant for five outcomes. studies through a χ2-based test [20]. A p value < 0.10 was No outcome included 95% prediction intervals excluding the considered indicating of excess statistical significance. null, i.e., not statistically significant. Sensitivity analysis in which these analyses were repeated Based on the above-mentioned criteria, none of the out- restricted to prospective observational studies with convinc- comes presented convincing (class I) or highly suggestive ing (class I) or highly suggestive (class II) evidence only was (class II) evidence. Six outcomes presented suggestive planned, but none met these criteria. evidence (class III): OH was associated with significantly higher risk of coronary heart disease (HR = 1.32, 95% CI Grading the evidence 1.12–1.56), stroke (HR = 1.22, 95% CI 1.08–1.38), con- gestive heart failure (HR = 1.30, 95% CI 1.09–1.55), falls Using the results of analyses described in “Statistical analy- (OR = 1.84, 95% CI 1.39–2.44), dementia (HR = 1.22, 95% sis”, associations that presented nominally statistically sig- CI 1.11–1.35), and all-cause mortality (RR = 1.50, 95% CI nificant random-effects summary estimates (i.e., p < 0.05) 1.24–1.81) (Table 1). were categorized into convincing, highly suggestive, sug- gestive, or weak evidence (class I–IV), following a grading scheme that has already been applied in various fields of Discussion medicine [21–33]. These criteria are fully reported in Sup- plementary Table 2. This umbrella review summarized the findings of seven pre- vious meta-analyses of the association between OH and 12 independent outcomes. Suggestive (i.e., class III) evidence Results for associations between OH and risk of coronary heart dis- ease, stroke, congestive heart failure, all-cause mortality, Literature search falls and dementia was found. We initially identified 975 papers. Of these, 22 full texts Cardiovascular disease (CVD) were screened and finally seven meta-analyses [34–40], which included 12 different outcomes, were included as While we identified significant associations between OH and reported in Fig. 1. several cardiovascular outcomes (coronary heart disease, 1 3 866 European Geriatric Medicine (2019) 10:863–870 Fig. 1 Prisma flow diagram Records identified through Duplicates were identified and database searching in PubMed, removed PsychInfo, Embase (n = 7) (n = 982) Records after duplicates were removed (n = 975) Records screened Records excluded based on (n = 975) title/abstract (n =953) Publications excluded (n =15) Full-text articles 11 Wrong study design assessed for eligibility 2 Wrong patient population (n =22) 1 Abstract 2 Poster abstract Meta-analyses included in umbrella review (n = 7) 12 outcomes stroke, congestive heart failure), none reached the cutoff for can be found in middle-age subjects and those with hyperten- class I or II evidence. sion and diabetes mellitus at baseline. These results highlight Several hypotheses may be helpful in explaining the rela- the predictive effect of OH for future congestive heart failure tionship between OH and increased CVD risk. First, patients in both the low-risk population and the high-risk population with OH are likely to have increased blood pressure variabil- with known congestive heart failure risks. On the other hand, ity related to body posture, and a large proportion of thoracic polypharmacy, in particular cardiovascular drugs including blood volume may be displaced to lower limbs due to gravity antianginals, antiarrhythmics, antihypertensive such as calcium during orthostasis [41]. Thus, both myocardial and cerebral channel blockers and α-blockers, is strongly associated with ischemia may occur frequently as a result of OH. Moreover, OH in patients with CV [45]. Therefore, careful medication subsequent acute change of hemodynamic and organ perfusion review is needed to improve orthostatic blood pressure changes status may trigger a coronary heart disease or stroke event. Sec- in routine clinical practice. ond, it has been suggested that OH is associated with higher arterial stiffness [42] and activated systematic inflammation Falls [43], which have both been involved in the pathogenesis of subclinical atherosclerosis, leading to cardiovascular disease Despite some studies failing to find a consistent association [43, 44]. Xin et al. in their analysis [38] stated that a significant between OH and falls, the present review found suggestive association between OH and congestive heart failure incidence evidence for this association meaning that this association 1 3 Included Eligibility Screening Identification European Geriatric Medicine (2019) 10:863–870 867 is less significant than expected. There are several possible explanations for the association between OH and falls. OH might cause an acute drop in cerebral oxygenation because of an impaired cerebral autoregulation, resulting in dizziness and falls [46]. Alternatively, OH might cause brain atrophy, microbleeds, and white matter brain lesions, resulting in falls [47]. OH might also cause falls through impaired muscle microcirculation, as one study found an association of OH with muscle ischemia [48]. Conversely, falls might cause OH by fear of falls, with consequent behavioral changes includ- ing lower physical activity levels, resulting in decondition- ing and muscle loss [49]. However, current evidence does not support this, as OH was not found to be associated with physical activity behavior. Dementia Suggestive evidence was found for an association between OH and dementia, but the association was not confirmed for vascular dementia or Alzheimer’s disease. The most frequently proposed mechanism linking OH to dementia is the recurrent transient brain hypoperfusion hypothesis [50]. Previous research has shown that cerebral blood flow is decreased in OH by electroencephalography [50], besides decreased brain perfusion during orthostatic pressure was demonstrated by the method of single‐photon emission computed tomography [51]. Cerebral hypoper- fusion may lead to leukoaraiosis underlying the neurode- generation process in dementia [52]. OH was traditionally thought to be detrimental only if compensatory mechanisms are inadequate. When cerebral autoregulation is impaired, it reacts less efficiently to compensate for a drop in cerebral perfusion pressure and fails to maintain adequate cerebral blood flow which may cause ischemic cerebral damage [53]. However, one recent study reported no relationship between OH and cognitive impairment related with leukoaraiosis, subtle brain microstructural damage, or cerebral blood flow [54]. OH and cognitive function are complicated and affected by multiple factors. The autonomous nervous sys- tem has been reported to be essential for orthostatic reflex and dysfunction of this system usually results in OH [55]. Some pathologies such as diabetes, alpha‐synucleinopathies, and sarcoidosis are common causes for autonomic neuropa- thy, and OH is prevalent among these diseases [56, 57]. On the other hand, in a recent study, it was demonstrated that the prevalence of OH, in older patients with Alzheimer’s disease, was similar to those with dementia of Lewy body, an alpha‐synucleinopathy [58]. All‑cause mortality OH represents a condition of impaired hemodynamic home- ostasis, where compensatory neuroendocrine mechanisms 1 3 Table 1 Evidence of the association between orthostatic hypotension Outcome [reference] No. of Cases Sample Type of Mean effect size p I2 Small study Excess significance Largest study 95% Prediction Level of studies size metric (95% CI) effect bias signifcant intervals evidence CHD [40] 7 5719 158,446 HR 1.32 (1.12–1.56) 0.001 65.4 No Yes No 0.81–2.15 III Stroke [40] 7 3657 158,446 HR 1.22 (1.08–1.38) 0.002 20.2 No Yes Yes 0.95–1.57 III Congestive HF [38] 4 3603 51,270 HR 1.30 (1.09–1.55) 0.004 56.5 No No Yes 0.66–2.56 III All-cause mortality [39] 10 NA 65,174 RR 1.50 (1.24–1.81) 0.00004 93.4 No NA Yes 0.75–3.00 III Falls [36] 15 2185 6323 OR 1.84 (1.39–2.44) 0.00002 73.2 Yes Yes No 0.68–5.01 III Dementia [34] 4 NA 41,972 HR 1.22 (1.11–1.35) 0.00009 0 No NA No 0.98–1.53 III Alzheimer [34] 2 NA 12,977 HR 1.18 (1.02–1.35) 0.02 0 NA NA No NA IV Vascular dementia [34] 3 NA 30,469 HR 1.40 (1.04–1.89) 0.03 0 No NA No 0.20–9.66 IV MMSE [34] 4 NA 3966 MD − 0.347 (− 0.560 to 0.001 23 No NA Yes − 1.01–0.31 IV − 0.134) MCI [34] 5 NA 12,969 OR 1.20 (1.001–1.43) 0.048 58.9 No NA No 0.71–2.01 IV CV mortality [37] 3 NA 51,013 RR 1.20 (0.73–2.00) 0.47 91.7 No NA No 0–655.7 NS Non CV mortality [37] 3 NA 51,013 RR 1.20 (0.96–1.50) 0.11 38.6 No NA Yes 0.14–9.93 NS CHD congenital heart disease, CV cardiovascular, HF heart failure, HR hazard ratio, MCI mild cognitive impairment, MD mean difference, MMSE mini-mental state examination, NA not appli- cable, OR odd ratios, RR relative risk 8 68 European Geriatric Medicine (2019) 10:863–870 are intermittently activated. These mechanisms may trigger Compliance with ethical standards the activation of other biologic effectors, e.g., platelets or the coagulation cascade, potentially promoting the occur- Conflict of interest All authors declare no conflict of interest. rence of cardio- or cerebrovascular events that can contribute Ethical approval It was not requested being a revision of already pub- to a higher mortality risk [43, 44]. Moreover, wide swings lished literature. This article does not contain any studies with human in blood pressure and supine hypertension associated with participants or animals performed by any of the authors. OH may provoke intermittent ischemic bouts and increased afterload, leading to permanent end-organ damage such as Informed consent No patients were included in this review. left ventricular hypertrophy and decreased renal function Sponsor’s role None. [5]. Baroreflex dysfunction, a marker of autonomic nervous system imbalance implicated in the pathogenesis of OH [59, 60], is characterized by enhanced sympathetic activity and withdrawal of parasympathetic control, and has long been References recognized as an important mediator of increased cardiovas- 1. American Autonomic Society, The American Academy of Neurol- cular morbidity and mortality [61–63]. ogy (1996) Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy. Limitations The Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Neurology 46:1470 2. Low PA (2008) Prevalence of orthostatic hypotension. Clin Auton The results of this study should be considered in light of its Res 18:8–13 limitations; some related to the umbrella review method and 3. Soysal P, Aydin AE, Okudur SK, Isik AT (2016) When should some to those of the individual studies included. Consider- orthostatic blood pressure changes be evaluated in elderly: 1st, ing that meta-analyses included studies with significantly 3rd or 5th minute? Arch Gerontol Geriatr 65:199–203 4. Aydin AE, Soysal P, Isik AT (2017) Which is preferable for ortho- differing designs, populations and other basic characteris- static hypotension diagnosis in older adults: active standing test tics, large heterogeneity might arise. However, a common or head-up tilt table test? Clin Interv Aging 12:207 estimate of heterogeneity (I2 < 50%) was used as one of the 5. Low PA, Tomalia VA (2015) Orthostatic hypotension: mecha- criteria for having convincing outcomes, even if the use of nisms, causes, management. J Clin Neurol 11:220–226 2 6. Verwoert GC, Mattace-Raso FU, Hofman A, Heeringa J, Stricker the same I is still discussed. Moreover, meta-analyses have BH, Breteler MM, Witteman JC (2008) Orthostatic hypotension important limitations and their results may also depend on and risk of cardiovascular disease in elderly people: the Rotterdam choices made about what estimates to select from each study study. J Am Geriatr Soc 56:1816–1820 and how to report them in the meta-analysis [64]. Applying 7. Atkins D, Hanusa B, Sefcik T, Kapoor W (1991) Syncope and orthostatic hypotension. Am J Med 91:179–185 the criteria suggested by the ROBIS for evaluating the qual- 8. Yap PLK, Niti M, Yap KB, Ng TP (2008) Orthostatic hypoten- ity of meta-analyses, we observed the presence of a high sion, hypotension and cognitive status: early comorbid markers risk of bias in two out of the seven meta-analyses included. of primary dementia? Dement Geriatr Cogn Disord 26:239 This evidence is mainly associated with the second phase in 9. Gupta V, Lipsitz LA (2007) Orthostatic hypotension in the elderly: diagnosis and treatment. Am J Med 120:841–847 which a high risk of bias in eligibility and selection of stud- 10. Schneider L, Sloane RB, Staples FR, Bender M (1986) Pretreat- ies and synthesis and findings of evidence. ment orthostatic hypotension as a predictor of response to nortrip- tyline in geriatric depression. J Clin Psychopharmacol 6:172–176 1 1. Chen L, Xu Y, Chen X, Lee W-J, Chen L-K (2019) Association Conclusions between orthostatic hypotension and frailty in hospitalized older patients: a geriatric syndrome more than a cardiovascular condi- tion. J Nutr Health Aging 23:318–322 In summary, OH seems to be significantly associated with 1 2. Ioannidis JP (2009) Integration of evidence from multiple meta- several negative health outcomes in older people, even if analyses: a primer on umbrella reviews, treatment networks and multiple treatments meta-analyses. CMAJ 181:488–493 only the association with coronary heart disease, congestive 13. Whiting P, Savović J, Higgins JP, Caldwell DM, Reeves BC, Shea heart failure, stroke, falls, dementia and all-cause mortality B, Davies P, Kleijnen J, Churchill R (2016) ROBIS: a new tool is supported by a suggestive evidence. However, the present to assess risk of bias in systematic reviews was developed. J Clin review does not allow to draw firm conclusions whether OH Epidemiol 69:225–234 14. Avelino-Silva TJ, Jaluul O (2017) Malnutrition in hospitalized can be considered as a risk factor for other medical con- older patients: management strategies to improve patient care and ditions. For instance, it is not clear whether patients with clinical outcomes. Int J Gerontol 11:56–61 OH benefit from antihypertensive treatments to the same 1 5. Stroud M, Duncan H, Nightingale J (2003) Guidelines for enteral extent as those without. Future prospective studies aiming feeding in adult hospital patients. Gut 52:vii1 16. Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a at investigating this relationship on larger cohorts of patients meta-analysis. Stat Med 21:1539–1558 and with less biases are necessary to reinforce the observed 17. Veronese N, Notarnicola M, Osella A et al (2018) Menopause associations in this umbrella review. does not affect fatty liver severity in women: a population study 1 3 European Geriatric Medicine (2019) 10:863–870 869 in a Mediterranean area. Endocr Metabol Immune Disord Drug 3 5. Min M, Shi T, Sun C, Liang M, Zhang Y, Wu Y, Sun Y (2018) Targets 18(5):513–521. https: //doi.org/10.2174/187153 03186 6618 The association between orthostatic hypotension and dementia: a 04231 01755 meta-analysis of prospective cohort studies. Int J Geriatr Psychia- 1 8. Egger M, Davey Smith G, Schneider M, Minder C (1997) Bias try 33:1541–1547 in meta-analysis detected by a simple, graphical test. BMJ 3 6. Mol A, Hoang PTSB, Sharmin S, Reijnierse EM, van Wezel RJ, 315:629–634 Meskers CG, Maier AB (2018) Orthostatic hypotension and falls 19. Carvalho AF, Kohler CA, Brunoni AR, Miskowiak KW, Her- in older adults: a systematic review and meta-analysis. J Am Med rmann N, Lanctot KL, Hyphantis TN, Quevedo J, Fernandes BS, Dir Assoc 20(5):589–597 Berk M (2016) Bias in peripheral depression biomarkers. Psy- 3 7. Ricci F, Fedorowski A, Radico F, Romanello M, Tatasciore A, chother Psychosom 85:81–90 Di Nicola M, Zimarino M, De Caterina R (2015) Cardiovascu- 20. Ioannidis JP, Trikalinos TA (2007) An exploratory test for an lar morbidity and mortality related to orthostatic hypotension: a excess of significant findings. Clin Trials 4:245–253 meta-analysis of prospective observational studies. Eur Heart J 2 1. Aromataris E, Fernandez R, Godfrey CM, Holly C, Khalil H, 36:1609–1617 Tungpunkom P (2015) Summarizing systematic reviews: meth- 38. Xin W, Lin Z, Li X (2013) Orthostatic hypotension and the risk odological development, conduct and reporting of an umbrella of congestive heart failure: a meta-analysis of prospective cohort review approach. Int J Evid Based Health Care 13:132–140 studies. PLoS One 8:e63169 22. Belbasis L, Savvidou MD, Kanu C, Evangelou E, Tzoulaki I 39. Xin W, Lin Z, Mi S (2014) Orthostatic hypotension and mortality (2016) Birth weight in relation to health and disease in later life: risk: a meta-analysis of cohort studies. Heart 100:406–413 an umbrella review of systematic reviews and meta-analyses. 40. Xin W, Mi S, Lin Z, Wang H, Wei W (2016) Orthostatic hypoten- BMC Med 14:147 sion and the risk of incidental cardiovascular diseases: a meta- 2 3. Bellou V, Belbasis L, Tzoulaki I, Evangelou E, Ioannidis JP analysis of prospective cohort studies. Prev Med 85:90–97 (2016) Environmental risk factors and Parkinson’s disease: an 41. Smit AA, Halliwill JR, Low PA, Wieling W (1999) Pathophysi- umbrella review of meta-analyses. Parkinsonism Relat Disord ological basis of orthostatic hypotension in autonomic failure. J 23:1–9 Physiol 519(Pt 1):1–10 24. Dinu M, Pagliai G, Casini A, Sofi F (2017) Mediterranean diet and 4 2. Mattace-Raso FU, van der Cammen TJ, Knetsch AM, van den multiple health outcomes: an umbrella review of meta-analyses of Meiracker AH, Schalekamp MA, Hofman A, Witteman JC (2006) observational studies and randomized trials. Nutr Metabol Car- Arterial stiffness as the candidate underlying mechanism for pos- diovasc Dis 27:e21 tural blood pressure changes and orthostatic hypotension in older 25. Kyrgiou M, Kalliala I, Markozannes G, Gunter MJ, Paraskevaidis adults: the Rotterdam study. J Hypertens 24:339–344 E, Gabra H, Martin-Hirsch P, Tsilidis KK (2017) Adiposity and 43. Fedorowski A, Ostling G, Persson M, Struck J, Engstrom G, cancer at major anatomical sites: umbrella review of the literature. Nilsson PM, Hedblad B, Melander O (2012) Orthostatic blood BMJ 356:j477 pressure response, carotid intima-media thickness, and plasma 26. Li X, Meng X, Timofeeva M, Tzoulaki I, Tsilidis KK, Ioannidis fibrinogen in older nondiabetic adults. J Hypertens 30:522–529 PA, Campbell H, Theodoratou E (2017) Serum uric acid levels 4 4. Fan XH, Wang Y, Sun K, Zhang W, Wang H, Wu H, Zhang and multiple health outcomes: umbrella review of evidence from H, Zhou X, Hui R (2010) Disorders of orthostatic blood pres- observational studies, randomised controlled trials, and Mendelian sure response are associated with cardiovascular disease and randomisation studies. BMJ 357:j2376 target organ damage in hypertensive patients. Am J Hypertens 27. Theodoratou E, Tzoulaki I, Zgaga L, Ioannidis JPA (2014) Vita- 23:829–837 min D and multiple health outcomes: umbrella review of sys- 45. Pepersack T, Gilles C, Petrovic M, Spinnewine A, Baeyens H, tematic reviews and meta-analyses of observational studies and Beyer I, Boland B, Dalleur O, De Lepeleire J, Even-Adin D, Van randomised trials. BMJ 348:g2035 Nes MC, Samalea-Suarez A, Somers A, Working Group Clinical 2 8. Veronese N, Solmi M, Caruso MG, Giannelli G, Osella AR, Pharmacology, Pharmacotherapy and Pharmaceutical Care, Bel- Evangelou E, Maggi S, Fontana L, Stubbs B, Tzoulaki I (2018) gian Society for Gerontology and Geriatrics (2013) Prevalence of Dietary fiber and health outcomes: an umbrella review of system- orthostatic hypotension and relationship with drug use amongst atic reviews and meta-analyses. Am J Clin Nutr 107:436–444 older patients. Acta Clin Belg 68:107–112 29. Machado MO, Veronese N, Sanches M et al (2018) The associa- 4 6. Mager DR (2012) Orthostatic hypotension: pathophysiology, tion of depression and all-cause and cause-specific mortality: an problems, and prevention. Home Healthc Nurse 30:525–530 umbrella review of systematic reviews and meta-analyses. BMC 47. Aoki M, Tanaka K, Wakaoka T, Kuze B, Hayashi H, Mizuta K, Ito Med 16:112 Y (2013) The association between impaired perception of vertical- 30. Veronese N, Demurtas J, Celotto S et al (2018) Is chocolate con- ity and cerebral white matter lesions in the elderly patients with sumption associated with health outcomes? An umbrella review of orthostatic hypotension. J Vestib Res 23:85–93 systematic reviews and meta-analyses. Clin Nutr 38(3):1101–1108 48. Humm AM, Bostock H, Troller R, Z’Graggen WJ (2011) Mus- 31. Veronese N, Demurtas J, Pesolillo G et al (2019) Magnesium cle ischaemia in patients with orthostatic hypotension assessed and health outcomes: an umbrella review of systematic reviews by velocity recovery cycles. J Neurol Neurosurg Psychiatry and meta-analyses of observational and intervention studies. Eur 82:1394–1398 J Nutr. https ://doi.org/10.1007/s00394 -019-01905 -w 4 9. Chisholm P, Anpalahan M (2017) Orthostatic hypotension: patho- 32. Köhler CA, Evangelou E, Stubbs B et al (2018) Mapping risk physiology, assessment, treatment and the paradox of supine factors for depression across the lifespan: an umbrella review of hypertension. Intern Med J 47:370–379 evidence from meta-analyses and Mendelian randomization stud- 5 0. Elmstahl S, Rosen I (1997) Postural hypotension and EEG vari- ies. J Psychiatr Res 103:189–207 ables predict cognitive decline: results from a 5-year follow-up of 33. Smith L, Luchini C, Demurtas J et al (2019) Telomere length and healthy elderly women. Dement Geriatr Cogn Disord 8:180–187 health outcomes: an umbrella review of systematic reviews and 5 1. Toyry JP, Kuikka JT, Lansimies EA (1997) Regional cerebral meta-analyses of observational studies. Ageing Res Rev 51:1–10 perfusion in cardiovascular reflex syncope. Eur J Nucl Med 34. Iseli R, Nguyen VTV, Sharmin S, Reijnierse EM, Lim WK, Maier 24:215–218 AB (2019) Orthostatic hypotension and cognition in older adults: a systematic review and meta-analysis. Exp Gerontol 120:40–49 1 3 870 European Geriatric Medicine (2019) 10:863–870 5 2. Brown WR, Thore CR (2011) Review: cerebral microvascular 59. Xin W, Lin Z, Mi S (2014) Orthostatic hypotension and mortality pathology in ageing and neurodegeneration. Neuropathol Appl risk: a meta-analysis of cohort studies. Heart 100:406–413 Neurobiol 37:56–74 6 0. Robertson D (2008) The pathophysiology and diagnosis of ortho- 5 3. Liu H, Zhang J (2012) Cerebral hypoperfusion and cognitive static hypotension. Clin Auton Res 18(Suppl 1):2–7 impairment: the pathogenic role of vascular oxidative stress. Int J 6 1. Sabbah HN (2012) Baroreflex activation for the treatment of heart Neurosci 122:494–499 failure. Curr Cardiol Rep 14:326–333 5 4. Foster-Dingley JC, Moonen JEF, de Ruijter W, van der Mast RC, 62. Schwartz PJ, La Rovere MT (1998) ATRAMI: a mark in the van der Grond J (2018) Orthostatic hypotension in older persons quest for the prognostic value of autonomic markers. Auto- is not associated with cognitive functioning, features of cerebral nomic tone and reflexes after myocardial infarction. Eur Heart J damage or cerebral blood flow. J Hypertens 36:1201–1206 19:1593–1595 55. Flachenecker P, Wolf A, Krauser M, Hartung H-P, Reiners K 6 3. La Rovere MT, Bigger JT Jr, Marcus FI, Mortara A, Schwartz (1999) Cardiovascular autonomic dysfunction in multiple sclero- PJ (1998) Baroreflex sensitivity and heart-rate variability in sis: correlation with orthostatic intolerance. J Neurol 246:578–586 prediction of total cardiac mortality after myocardial infarction. 56. Lim SY, Lang AE (2010) The nonmotor symptoms of Parkinson’s ATRAMI (autonomic tone and reflexes after myocardial infarc- disease: an overview. Mov Disord 25(Suppl 1):S123–S130 tion) investigators. Lancet 351:478–484 57. Gaspar L, Kruzliak P, Komornikova A et al (2016) Orthostatic 6 4. Ioannidis JP (2016) The mass production of redundant, mislead- hypotension in diabetic patients: 10-year follow-up study. J Dia- ing, and conflicted systematic reviews and meta-analyses. Milbank betes Complicat 30:67–71 Q 94:485–514 5 8. Isik AT, Kocyigit SE, Smith L, Aydin AE, Soysal P (2019) A comparison of the prevalence of orthostatic hypotension between Publisher’s Note Springer Nature remains neutral with regard to older patients with Alzheimer’s disease, Lewy body dementia, jurisdictional claims in published maps and institutional affiliations. and without dementia. Exp Gerontol 124:110628. https ://doi. org/10.1016/j.exger. 2019.06.001 A ffiliations Pinar Soysal1 · Nicola Veronese2 · Lee Smith3 · Gabriel Torbahn4 · Sarah E. Jackson5 · Lin Yang6 · Andrea Ungar7 · Giulia Rivasi7 · Martina Rafanelli7 · Mirko Petrovic8 · Stefania Maggi2 · Ahmet Turan Isik9 · Jacopo Demurtas10 · The Special Interest Groups in Systematic Reviews and Meta-analyses for Healthy Ageing, and Cardiovascular Medicine of the European Society of Geriatric Medicine (EuGMS) * Pinar Soysal 6 Department of Cancer Epidemiology and Prevention dr.pinarsoysal@hotmail.com Research, Albert Health Services, Calgary, AB, Canada 7 1 Department of Geriatric Medicine, Faculty of Medicine, Department of Geriatrics, Azienda Ospedaliero-Universitaria Bezmialem Vakif University, Adnan Menderes Bulvarı Careggi and University of Florence, Florence, Italy (Vatan Street), Fatih, 34093 Istanbul, Turkey 8 Department of Internal Medicine, Section of Geriatrics, 2 National Research Council, Neuroscience Institute, Aging Ghent University, Ghent, Belgium Branch, Padua, Italy 9 Unit for Aging Brain and Dementia, Department of Geriatric 3 The Cambridge Centre for Sport and Exercise Sciences, Medicine, Faculty of Medicine, Dokuz Eylul University, Anglia Ruskin University, Cambridge, UK Izmir, Turkey 10 4 Institute for Biomedicine of Aging, Friedrich Alexander Primary Care Department Azienda USL Toscana Sud Est, University Erlangen-Nürnberg, Nuremberg, Germany 58100 Grosseto, Italy 5 Department of Behavioural Science and Health, University College London, London, UK 1 3