CKD and Acute and Long-Term Outcome of Patients with Peripheral Artery Disease and Critical Limb Ischemia

Abstract

Background and objectives: Despite the many studies showing an association between CKD and a high risk of ischemic events and mortality, the association of CKD with peripheral arterial disease (PAD) still has not been well described.

Design, setting, participants, & measurements: This large cohort study assessed the association of CKD, even in the earlier stages, with morbidity, short- and long-term outcome, and costs among patients with PAD.

Results: We identified 41,882 patients with PAD who had an index hospitalization between January 1, 2009, and December 31, 2011. Of these, 8470 (20.2%) also had CKD (CKD stage 2: n=2158 [26%]; stage 3: n=3941 [47%]; stage 4: n=935 [11%]; stage 5: n=1436 [17%]). The ratio of women to men was 1:1.2. Compared with patients without known CKD, those with CKD had higher frequencies of coronary artery disease (1.8-fold higher; P<0.001), chronic heart failure (3.3-fold higher; P<0.001), and Rutherford PAD categories 5 and 6 (1.8-fold higher; P<0.001); underwent significantly fewer revascularizations (0.9-fold fewer; P<0.001); had a nearly two-fold higher amputation rate (P<0.001); had higher frequencies of in-hospital infections (2.1-fold higher; P<0.001), acute renal failure (2.8-fold higher; P<0.001), and sepsis (1.9-fold higher; P<0.001); had a 2.5-fold higher frequency of myocardial infarction (P<0.001); and had a nearly three-fold higher in-hospital mortality rate (P<0.001). In an adjusted multivariable Cox regression model, CKD remained a significant predictor of long-term outcome of patients with PAD during follow-up for up to 4 years (until December 31, 2012; median, 775 days; 25th-75th percentiles, 469-1120 days); the hazard ratio was 2.59 (95% confidence interval, 2.21 to 2.78; P<0.001). The projected mortality rates after 4 years were 27% in patients without known CKD and 46%, 52%, 72%, and 78% in those with CKD stages 2, 3, 4, and 5, respectively. Lengths of hospital stay and reimbursement costs were on average nearly 1.4-fold higher (P<0.001) in patients who also had CKD.

Conclusions: This analysis illustrates the significant and important association of CKD with in-hospital and long-term mortality, morbidity, amputation rates, duration and costs of hospitalization, in-hospital treatment, and complications in patients with PAD.

Keywords: acute kidney injury; amputation; chronic kidney disease; cohort studies; hospital mortality; hospitalization; humans; length of stay; outcome; peripheral arterial disease.

Figure 1.

CKD and survival and amputation. During follow-up, CKD was a high significant independent predictor of long-term mortality and amputation (P<0.001). Cox regression analysis adjusted for additional comorbidities and baseline parameters is presented for death (A) and amputation (B). Highly significant differences were observed among the CKD stages (P<0.001).