O. Ifudu et al., CORRELATES OF VASCULAR ACCESS AND NONVASCULAR ACCESS-RELATED HOSPITALIZATIONS IN HEMODIALYSIS-PATIENTS, American journal of nephrology, 16(2), 1996, pp. 118-123
Four hundred and thirty randomly selected hemodialysis patients, aged
20 years and over, were studied to identifiy risk factors for vascular
access and nonvascular access-related hospitalizations in the immedia
tely preceding 1 year. Risk estimates for hospitalization were assesse
d using a multinominal logistic analysis model. We measured functional
status, utilizing a 14-point Karnofsky scale, and in a separate analy
sis of covariance, in which Karnofsky score was the outcome, we examin
ed the relationships of age, gender, ethnicity, renal diagnosis, and h
ospitalization. Individual comparisons were adjusted for multiple comp
arison bias by Tukey's Honest Difference method. There were a total of
508 hospitalizations of which 322 (63%) lasted greater than or equal
to 1 week. Two hundred and sixty (60%) patients were hospitalized at l
east once; 105 (24.4%) for access problems only, 115 (27%) for a nonac
cess problem only, and 40 for access and nonaccess-related problems. A
ccess-related problems, accounted for 48% of all hospitalizations. The
risk of hemodialysis vascular access morbidity was increased in women
(p < 0.028) and white (p < 0.048) hemodialysis patients. Neither diab
etic nor elderly hemodialysis patients were at greater risk for access
hospitalization than their respective counterparts, thought a greater
proportion of the access hospitalizations in the elderly (greater tha
n or equal to 64 years) lasted greater than or equal to 1 week (p < 0.
0006). More access-related hospitalizations in blacks (64.5%), lasted
for greater than or equal to 1 week than in whites (40.6%) (p < 0.001)
. Hispanics (p < 0.043), whites (p < 0.002), and the older patients (p
< 0.054) were at greater risk for nonaccess hospitalization than blac
ks and younger patients, respectively. Even after adjusting for age, r
ace, and diabetes, each decrease of one unit in the modified Karnofsky
score was associated with a 3-4% increased risk for all types of hosp
italization (p < 0.001)-poor functional status is associated with incr
eased risk for all hospitalizations. We conclude that the risk for hem
odialysis vascular access morbidity is increased in women and white he
modialysis patients. Poor functional status is associated with increas
ed risk for all hospitalizations.