V SAGLIMBENE1,2, G WONG1,3,4, N BONDONNO5, M RUOSPO2,6, SC PALMER7, K CAMPBELL8, V GARCIA LARSEN9, P NATALE2, A TEIXEIRA-PINTO1, L GARGANO2, AM MURGO2, DW JOHNSON8,10, M TONELLI11, R GELFMAN2, E CELIA2, T ECDER2, A BERNAT2, D DEL CASTILLO2, D TIMOFTE2, M TÖRÖK2, A BEDNAREK-SKUBLEWSKA2,12, J DUŁAWA2,13, P STROUMZA2, M HANSIS2, E FABRICIUS2, C WOLLHEIM2, J HEGBRANT2, JC CRAIG1,3, GFM STRIPPOLI1,2,14
1University of Sydney, Sydney, New South Wales; 2Diaverum Medical-Scientific Office, Lund, Sweden; 3Children’s Hospital at Westmead, Sydney, New South Wales; 4Westmead Hospital, Sydney, New South Wales; 5University of Western Australia, Perth, Western Australia; 6Amedeo Avogadro University of Eastern Piedmont, Novara, Italy; 7University of Otago Christchurch, Christchurch, New Zealand; 8University of Queensland at Princess Alexandra Hospital, Woolloongabba, Queensland; 9Johns Hopkins Bloomberg School of Public Health, Baltimore, US; 10University of Queensland, Woolloongabba, Queensland; 11University of Calgary, Calgary, Canada; 12Medical University of Lublin, Lublin, Poland; 13Medical University of Silesia, Katowice, Poland; 14University of Bari, Bari, Italy
Aim: To evaluate the association between fruit intake and mortality for adults treated with haemodialysis.
Background: High fruit intake is associated with reduced risk of cardiovascular disease in the general population. It is unclear whether this association occurs in patients on haemodialysis, in whom high fruit intake is generally discouraged due to risks associated with hyperkalaemia.
Methods: Using data from the DIET-HD study, a prospective, longitudinal study in 9757 adults treated with haemodialysis in Europe and South America, fruit intake (portions/day) was measured by the GA2LEN food frequency questionnaire. Adjusted Cox regression analyses clustered by country were conducted to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for cardiovascular and all-cause mortality.
Results: During a median follow up of 1.5 years (8108 person-years), there were 1214 deaths including 515 attributable to cardiovascular causes. There was no association between fruit intake (one portion (70 g) increase per day) and cardiovascular (HR 1.00, 95% CI 0.97-1.02) and all-cause mortality (0.99, 0.97-1.00). Apple consumption (≥ 1 apple per day versus none) was associated with lower risks of cardiovascular (0.67, 0.52-0.88) and all-cause (0.83, 0.69-0.98) mortality. Higher consumption (one fruit increase per day) of avocados (2.75, 1.28-5.91) and apricots (1.84, 1.26-2.69) was associated with increased cardiovascular mortality.
Conclusions: There was no overall association between fruit consumption and short-term all-cause and cardiovascular mortality for people treated with haemodialysis. Daily apple consumption may be associated with lower mortality in this clinical setting.