September 20, 2017

GRAPEVINE, TX — A small randomized study intrigued observers here at the Heart Failure Society of America 2017 Scientific Meeting, even if it was mostly a wash for its primary end point, health-related quality-of-life scores in patients discharged after a heart-failure hospitalization.

But the intervention, regular home delivery of specially prepared, low-sodium, high-potassium meals for a month, seemed in the greatly underpowered trial to improve a composite score for symptoms and physical limitation and reduce both the risk of rehospitalization and number of hospitalization days.

The positive secondary findings fell short of significance in the trial, called Geriatric Out-of-Hospital Randomized Meal Trial in Heart Failure (GOURMET-HF), but support a "food-as-medicine" concept that assumes improved nutrition can have important physiologic effects, Dr Scott L Hummel (University of Michigan and VA Ann Arbor, MI) told | Medscape Cardiology.

It's often hard to secure funding for trials with special diets like this intervention, but "it's something we ought to formally study for the hard outcomes that payers care about and patients care about," he said.

The pilot GOURMET-HF study was aimed at showing the strategy's feasibility for sodium reduction in patients with a history of heart-failure hospitalization as a prelude to a larger randomized trial.

Restriction of sodium is a cornerstone of traditional dietary recommendations for patients with heart failure, but some studies in both chronic and acute decompensated heart failure (ADHF) have associated it with malnutrition, Hummel said. Sodium restriction correlates highly with calorie restriction, he explained.

At baseline in the trial, patients who reported they were on a low-sodium diet tended to have micronutrient deficiencies and were consuming "fewer calories than you'd expect they'd need for their daily activities, even being sedentary," he said.

In an interview, Dr Lynn Warner Stevenson (Brigham and Women's Hospital, Boston, MA) agreed. When patients are told by their physicians to eat a low-sodium diet, "often they just quit eating. It's a question of maintaining a low-sodium diet but also maintaining nutrition."

Given GOURMET-HF, "Hopefully we can get both of them at the same time," she said. "I think it's particularly good for the elderly, that's why I'm excited about it."

Stevenson also said the results are "encouraging for feasibility and the opportunity to have fewer heart-failure readmissions."

Two Potassium-Intake Levels

The trial randomized 66 patients aged at least 55 discharged from a hospitalization for ADHF, excluding those with high potassium levels or low glomerular filtration rate (GFR), at three major centers. Half were assigned to usual care and half to usual care plus once-weekly home delivery of the prepackaged meals, which were compatible with the sodium-restricted Dietary Approaches to Stop Hypertension (DASH-SRD) diet.

The GOURMET-HF version of the DASH diet called for the meals to provide no more than 1500-mg sodium and 4500-mg potassium per day, the latter reduced to 3000 mg/day in patients at risk for hyperkalemia such as those with renal insufficiency or treatment with potassium-sparing diuretics or aldosterone antagonists.

They stayed on the diet for 4 weeks and were followed for a total of 12 weeks for the primary end point of Kansas City Cardiomyopathy Questionnaire (KCCQ) summary score for health-related quality of life.

Patients had a variety of different meals from which to choose, and the investigators ensured they were appealing. "We actually did some taste-testing ourselves," said Hummel.

Patients were discouraged from supplementing the meals with other food, although it was allowed, he said. "We recognized that people were going to eat other things sometimes, that they'll go to a wedding or a Super Bowl party," for example.

Patients adhered to the DASH meals a mean of 23 days out of the 4 weeks; 76% were adherent to the meals at least 80% of the days, Hummel reported.

Serum levels of potassium and creatinine were not significantly different between the groups, he said.

KCCQ summary scores went up significantly (P<0.001) and similarly in both groups between hospital discharge and the 4-week mark. The KCCQ clinical summary scores, reflecting HF symptom status and physical limitations, Hummel said, also rose in both groups, but by 18 points in the intervention group ("one standard deviation") and only 10.7 points in the control group (P=0.053).

Adverse Events, Clinical Outcomes for the DASH Diet v Usual Care: Number of Patients (Number of Events)

End points DASH-SRD, 30 d Usual care, 30 d DASH-SRD, 84 d Usual care, 84 d
Possible diet-related adverse events a   3 (3) 0 (0) 4 (4) 2 (2)
All-cause rehospitalization or death 4 (4) 9 (12) 11 (15) 14 (23c)
HF rehospitalization 4 (4) 9 (11)b 7 (8) 13 (18)
a. One each of hyperkalemia, symptomatic hypertension, acute kidney injury
b. P=0.055
c. Includes one death

The number of HF hospitalizations showed "a strong trend toward benefit" for the intervention group (P=0.055), he said, as did number of in-hospital days: 17 in the DASH-diet group and 55 in the usual-care group (P=0.06).

As a discussant following Hummel's presentation, Dr Christopher M O'Connor (Duke University, Durham, NC) agreed that GOURMET-HF showed a large trial would be feasible but also expressed reservations about the difference in clinical outcomes.

The usual-care group started to outstrip the intervention group for clinical events fairly soon after randomization, he noted, before diet changes would be expected to make a difference. He pointed to several imbalances in the two groups' baseline features: the usual-care group had a higher mean BMI and mean heart rate and contained more patients with diabetes, numerically, although short of significant differences.

"My guess is that's what's driving the difference in clinical events," he said. "The sicker patients got randomized to the usual-care arm."

With | Medscape Cardiology, Hummel speculated on how such a dietary intervention might be implemented in clinical practice, should a sufficiently large trial show it makes a clinical difference. With cost-effectiveness an issue, perhaps it would be used in higher-risk patients, although GOURMET-HF was too small to determine whether the most initially malnourished patients were the ones who benefited the most.

"But that seems to be a reasonable way of looking at it," he said. It could be for people who are malnourished, with physical challenges or few transportation options, for example. "If you knew someone had these kinds of problems, you might intervene."

GOURMET-HF was funded by the National Institute on Aging and supported by PurFoods.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from | Medscape Cardiology, 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.