Five-Factor Score Helps Predict Postural Hypotension, Falls, and Cognitive Impairment in Elderly

Deborah Brauser

June 15, 2016

PARIS, FRANCE — A simple algorithm of variables may help clinicians to identify elderly patients with postural hypotension as well as help to predict future falls and even cognitive dysfunction, new research suggests[1].

Examining data from more than 1300 participants in the InChianti study, all of whom lived in Chianti, Italy, investigators found that a newly improved, five-point Detecting Risk of Postural Hypotension (DROP) score helped them to determine those with systolic postural dysfunction, defined as a >20-mm-Hg fall in supine blood pressure on standing, through 9 years of follow-up. The five predictive factors were being older than age 65, a fall in the past 12 months, a previous stroke, having hypertension, and having Parkinson's disease.

Added together, this five-factor DROP score also helped to predict increased mortality, cognitive impairment based on changes in Mini-Mental State Exam (MMSE) scores, and future falls in the next 2 to 3 years.

"We showed that the likelihood of postural hypotension can be predicted from factors in existing medical history," lead author Dr Christopher Clark (University of Exeter Medical School, UK) told attendees here at the European Society of Hypertension (ESH) 2016 Annual Meeting.

Dr Christopher Clark

"This project is clinician driven because we have colleagues who aren't detecting this phenomenon of drops in blood pressure, so they're not responding to it," he later added to heartwire from Medscape. "We want to get out the message that 'here's a test that might help in predictions and that will make your lives easier.' "

At last year's ESH meeting, Clark reported "very preliminary" trial results showing that a total DROP score made up of the variables of age >65 years, a recent fall, a previous stroke, and having hypertension or angina could predict postural hypotension.

He noted that the new analysis was more "rigorous" and included cohort comparisons, which could explain why the variable of angina last year was replaced with Parkinson's-disease status in the new report. Interestingly, both last year and this year did not show diabetes as a top-five postural hypotension predictor. "It was one of the univariate associations, but, quite surprisingly, it dropped out of the multivariable model," he said.

Session comoderator Dr Dariusz Gasecki (Medical University of Gdańsk, Poland) said that the study "was well-planned, and the findings are very important for clinical practice."

"What I find exciting was that Parkinson's disease was also involved in this predictive score. I'll keep that in mind and may also use the Mini-Mental score as a predictor," Gasecki said to heartwire . "I think some indices of the brain could really help us."

Costly Falls

Clark noted that more than three million older people in the UK fall each year, costing the National Health Service around £2.3 billion. "And postural hypotension is a clear risk factor for these falls." However, this condition isn't routinely looked for in clinical practice.

In a survey they're currently conducting of 90 general practices in Southwest England, "we found that when people have symptoms, they're very likely to have their postural blood pressure checked. But we're not routinely checking those with diabetes and we're only checking about a third of the [nonsymptomatic] elderly."

Clark reported that in addition to completing more analysis on their earlier assessments, the researchers wanted to also investigate whether future falls and cognitive impairment could be predicted in this patient population.

The original population-based InChianti study was created to assess causes of walking difficulties in the elderly, with baseline examinations conducted in 1998—and roughly 9 years of total follow-ups conducted at 3-year time points. BP was measured while the patients were lying down and at 1 and 3 minutes after standing up.

For the current analysis, the investigators randomized 1317 participants into two study cohorts: a "derivation cohort" (n=649; mean age 68.5 years; preexisting diabetes, 12.3%) and a validation cohort (n=668; mean age 68.2 years; preexisting diabetes, 11.4%). "This allowed us to work with one-half of the data set to develop weighted prediction scores and then test those DROP scores in the other half," explained Clark.

Candidate predictor variables assessed included age, sex, drugs (including antiarrhythmics and anticholinesterase inhibitors), medical history, frailty, and MMSE scores.

At the end of 9 years of follow-up, 8.6% of the first cohort and 6.7% of the second had systolic drops of >20 mm Hg at 1 minute.

Five Strongest Multivariable Associations With the Presence of Postural Hypotension*

Variable Prevalence (%) Odds Ratio 95% CI
Age (>65 years) 100 3.5 1.1–11.6
Hypertension 38 2.2 1.2–4.0
Past-year fall 22 1.8 1.0–3.3
Previous stroke 6 2.3 1.0–5.2
Parkinson's disease 1 9.4 2.4–37.1
*Combined cohorts

The DROP score gave 1 point for each of these variables; thus, a potential score could range from 0 to 5. When researchers examined the performance of the score, the odds ratio (OR) for postural hypotension rose as the score increased (OR per unit increment in DROP score, 1.8; 95% CI 1.3–2.5; P <0.001).

In addition, the score correlated with all-cause mortality. The incremental hazard ratio per unit increase in DROP score was 1.80 (95% CI 1.6–2.0, P<0.001).

The score also predicted falls in the second and third years of follow-up (P<0.01 for trend) and predicted mean reductions in MMSE scores (P<0.001).

"Clearly, further work is needed to validate this DROP score in other relevant populations," said Clark. "And we intend to model the impact of it in reducing postural hypotension," he noted, adding that a new study on implementation is now planned.

In addition, the researchers plan to further examine the data for associations with systolic drops of >20 mm Hg at 3 minutes.

"Very Good Idea"

Gasecki noted that although more data is needed, the findings can be helpful to clinicians now. "I think many people aren't so aware that postural hypotension is important. I'm glad the investigators will be assessing pressure at 3 minutes, but in some patients I think you need 10 minutes to assess properly. Some patients have delayed reactions."

Still, "let's go ahead and do at least this preliminary test for our patients."

His comoderator Dr Cristina Sierra (Hospital Clinic and University of Barcelona, Spain) agreed. She added that postural hypertension is very frequent, especially in older patients, in clinical practices in her country. "So I think this DROP score is a very good idea," she told heartwire .

The study was funded by the National Institute on Aging and by the National Institute for Health Research. The study authors, Gasecki, and Sierra reported no relevant financial relationships.

Follow Deborah Brauser on Twitter: @heartwireDeb. For more from, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.