The cerebro-renal interaction in stroke neurology.

Neurology 2012;78:1898–1899 Chronic kidney disease (CKD), defined as a reduced glomerular filtration rate (GFR) or albuminuria, is a known, strong risk factor for stroke. Meta-analyses of cohort studies and trials indicate that proteinuria/ albuminuria increases the risk of stroke by 71%– 92%, and estimated GFR (eGFR) 60 mL/min/ 1.73 m increases the risk by 43%.1,2 CKD is also predictive of poor outcomes after stroke; reduced eGFR is independently associated with increased 1and 10-year mortalities.3,4 Since end-stage renal disease (ESRD) is another established predictor of stroke risk and poor stroke outcomes,5 stroke neurologists should fully understand the interaction between stroke and CKD/ESRD. In this issue of Neurology, Kumai et al.6 studied 3,778 patients admitted within 24 hours of onset of their first-ever ischemic stroke from a large-cohort multicenter stroke registry; of these, 1,320 (34.9%) had CKD defined as proteinuria or reduced eGFR on admission. They found that CKD patients had a 49% greater risk of neurologic deterioration, defined as a 2-point increase in the NIH Stroke Scale, during hospitalization; a 138% greater risk of in-hospital mortality; and a 25% greater risk of a modified Rankin Scale score (mRS) 2 at discharge than nonCKD patients after adjustment for potential confounding factors, including initial stroke severity. As a component of CKD, proteinuria showed a much stronger association with unfavorable outcomes than reduced eGFR; for example, patients with mild proteinuria, with an estimated amount of urine protein of 30 –100 mg/mL, had a 69% greater risk of an mRS 2 than patients without proteinuria. In contrast, reduced eGFR was not associated with stroke outcomes in the study by Kumai et al.6 Albuminuria and reduced eGFR may involve separate pathologic processes. Kumai et al. did not measure filtration function prior to stroke onset but only on admission, when it may have been affected by acute stroke damage. Measurement of eGFR and urine protein in the chronic stage of stroke might show different associations. Other potential limitations include use of semiquantitative measurement of urine protein using dipstick testing, which may result in frequent falsepositive and false-negative results. However, it is meaningful that such a handy and economical measurement can predict stroke outcomes. Also, outcomes were measured after only a short interval. Vital and functional outcomes were assessed at hospital discharge (median, 23 days after stroke onset). Because their registry has a plan to perform yearly follow-up for enrolled patients, this limitation will eventually be resolved. Renal dysfunction is a bystander of stroke, since both conditions are associated with hypertension and several traditional vascular risk factors. Additionally, albuminuria is an indicator of cephalocervical and systemic vascular dysfunction via nontraditional vascular risk factors, including endothelial dysfunction, maladaptive carotid arterial remodeling, homocystinemia, coagulation disorders, impaired endothelial release of tissue plasminogen activator, extravascular coagulation, and higher levels of inflammatory cytokines and oxidative stress. An association between albuminuria and hemorrhagic transformation of infarcts has also been noted.7 Such factors might cause both initial severity and unfavorable outcomes of stroke in patients with proteinuria. These data underscore the fact that one can no longer manage and care for stroke patients while disregarding their renal condition. CKD/ESRD is not only predictive of unfavorable outcomes after general stroke, but also of stroke outcomes after specific therapies. In our multicenter observational studies (the SAMURAI rt-PA Registry), reduced eGFR was associated with early symptomatic hemorrhagic conversion, mortality, and an mRS 4 at 3 months after IV recombinant tissue plasminogen activator therapy.8 Although thrombolysis is generally not considered safe for stroke patients undergoing hemodialysis, most experts who responded to an international survey favored its use.9 Patients with stages 4 and 5 CKD (eGFR 30 mL/min/ 1.73 m) have a high 30-day mortality when