Renal Dysfunction as an Independent Predictor of Outcome

After Aneurysmal Subarachnoid Hemorrhage

A Single-Center Cohort Study
Brad E. Zacharia, MD; Andrew F. Ducruet, MD; Zachary L. Hickman, MD; Bartosz T. Grobelny, BA;
Luis Fernandez, MD; J. Michael Schmidt, PhD; Reshma Narula, BA; Lauren N. Ko; Margot E. Cohen;
Stephan A. Mayer, MD; E. Sander Connolly, Jr, MD

Background and Purpose—Acute kidney injury occurs in 1% to 25% of critically ill patients with small increases in
creatinine adversely affecting outcome. We sought to determine the burden of acute kidney injury in patients with
aneurysmal subarachnoid hemorrhage and whether this dysfunction affects outcome.
Methods—Between 1996 and 2008, 787 consecutive patients with aneurysmal subarachnoid hemorrhage were enrolled in
our prospective database. Demographics, serum creatinine levels, and discharge modified Rankin scores were recorded,
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017

and changes in creatinine clearance were calculated. A multiple logistic regression was performed using known
predictors for poor outcome after aneurysmal subarachnoid hemorrhage in addition to burden of contrast-enhanced
imaging and change in creatinine clearance.
Results—One hundred seventy-nine (23.1%) patients were at risk for renal failure during their hospitalization. In a
multivariate model, those patients who developed risk for renal failure were twice as likely to have a poor 3-month
outcome (OR, 2.01; P⫽0.021). Survival curves comparing those not at risk, those at risk (increasing severity classes
Risk, Injury, and Failure, and the 2 outcome classes Loss and End-Stage Kidney Disease [RIFLE] R), and those with
renal injury or failure (RIFLE I and F) demonstrated that risk of death increases significantly as one progresses through
the RIFLE classes (log rank, P⬍0.0001).
Conclusions—In a large, consecutive series of prospectively enrolled patients with aneurysmal subarachnoid hemorrhage,
we demonstrate, using the newly defined RIFLE classification for risk of renal failure, that even seemingly insignificant
decreases in creatinine clearance are associated with significantly worse 3-month outcomes. This study highlights the
importance of close surveillance of renal function and stresses the value of renal hygiene in the aneurysmal subarachnoid
hemorrhage population. (Stroke. 2009;40:2375-2381.)
Key Words: aneurysm 䡲 outcomes 䡲 renal disease 䡲 SAH

A neurysmal subarachnoid hemorrhage (aSAH) affects

nearly 30 000 individuals each year in the United States.
Although early surgical or endovascular protection of rup-
tured aneurysms and aggressive postoperative management
have improved outcomes, it remains a devastating disease
with mortality approaching 50% and less than 60% of
survivors returning to functional independence.1–3 Cerebral
vasospasm and associated delayed neurological deficits ac-
counts for approximately one third of aSAH death and
disability, and research has focused heavily on reducing this
burden. With improvements in aSAH management, the med-
ical (nonneurological) complications are playing a more
prominent role in outcome after aSAH.2,4,5 The most frequent
medical complications include pulmonary edema and pneu-
monia,6 cardiac arrhythmias,7 electrolyte disturbances,3 and
hematologic abnormalities.8,9 These complications can rival
the frequency of mortality from neurological complications
such as rebleeding and vasospasm.10 Although cardiopulmo-
nary complications after aSAH have been explored at length,
little has been done to evaluate the effect of renal dysfunction
in this population.
Acute renal failure occurs in 1% to 25% of critically ill
patients, and even small increases in serum creatinine (sCr)
can adversely affect outcome.11 Renal dysfunction in patients
with subarachnoid hemorrhage has been previously reported
as occurring in 0.8% to 7% of patients, although a multitude
of definitions were used.2 The Acute Dialysis Quality Initia-
tive group has recently devised a multilevel classification of
acute renal failure based on changes in sCr or urine output in
which the worst of each criterion is used (Figure 1). The

Received December 15, 2008; final revision received February 12, 2009; accepted February 24, 2009.
From the Departments of Neurological Surgery (B.E.Z., A.F.D., Z.L.H., B.T.G., R.N., L.N.K., M.E.C., E.S.C.) and Neurology (L.F., J.M.S., S.A.M.),
Columbia University, College of Physicians & Surgeons, New York, NY.
Correspondence to Brad E. Zacharia, MD, Department of Neurological Surgery, Columbia University, College of Physicians & Surgeons, 710 West
168th Street, New York, NY 10032. E-mail bez2103@columbia.edu
© 2009 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.108.545210

2375
 2376 Stroke July 2009
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017
acronym RIFLE stands for the increasing severity classes
Risk, Injury, and Failure, and the 2 outcome classes Loss and
End-Stage Kidney Disease. RIFLE has become the most
widely used definition of acute renal failure.12,13 Thus, the
concept of acute kidney injury (AKI) is not just acute renal
failure; it encompasses the entire spectrum from severe to
mild conditions, recognizing that small changes in kidney
function are associated with significant changes in short- and
possibly long-term outcomes.11
Patients with aSAH are particularly vulnerable to multi-
system organ dysfunction. Even good-grade patients experi-
ence lengthy intensive care unit stays, often undergo opera-
tive intervention, and are subjected to potentially harmful
fluid management to prevent and treat delayed neurological
deficits. Furthermore, these patients undergo a significant
number of contrast radiographic studies, including CT an-
giography, CT perfusion, and catheter-based digital subtrac-
tion angiography. Radiocontrast material has been closely
associated with renal dysfunction and a number of groups
have demonstrated significant increases in mortality in pa-
tients with radiocontrast-associated AKI.14 –17 The combina-
tion of these factors predisposes these patients to AKI.
This report focuses on the incidence and effect of mild renal
dysfunction in one of the largest clinical series of patients with
aSAH presented to date in an effort to test the hypothesis that
small changes in creatinine clearance (CrCl) are associated
with long-term outcome in this patient population.
Materials and Methods
Patient Population
The Columbia University Subarachnoid Hemorrhage Outcomes
Project prospectively enrolled 924 consecutive patients with sub-
arachnoid hemorrhage admitted to the Neurological Intensive Care
Unit between August 1, 1996, and May 1, 2008. The diagnosis of
subarachnoid hemorrhage was established on the basis of admission
CT imaging or by xanthochromia of the cerebrospinal fluid. Only
patients with documented cerebral aneurysms (787) were included in
the study.
Figure 1. RIFLE criteria. Adapted from Bellomo
et al. Critical Care. 2004;8:R204 –R212. (BioMed
Central). URL:http//ccforum.com/content/8/4/R204.
Clinical Management
The management of patients with aSAH at our institution has been
described in detail previously.18 Briefly, while the patient was in the
neurological intensive care unit, transcranial Doppler sonography
was performed daily or every other day, and all patients received oral
nimodipine. A ventricular catheter was placed in all patients with
ventriculomegaly or intraventricular hemorrhage and a decreased
level of consciousness that could not be attributed to causes other
than hydrocephalus. CT scanning was performed to evaluate all
instances of clinical deterioration. CT angiography and perfusion
were used as adjuncts when there was clinical suspicion for vaso-
spasm. All patients were administered 0.9% saline and supplemental
5% albumin to maintain central venous pressure at ⬎8 mm Hg, and
those with clinical deterioration from delayed cerebral ischemia were
treated with hypertensive hypervolemic therapy to maintain systolic
blood pressure at ⬎200 mm Hg. When significant clinical symptoms
persisted despite hypertensive hypervolemic therapy, balloon angio-
plasty of vasospastic vessels was attempted.
Clinical Assessment
Admission clinical status was evaluated using the Hunt–Hess (H&H)
grading scale and Glasgow Coma Score. Complete medical history
was recorded on admission. Premorbid functional status was as-
sessed using the modified Rankin Scale (mRS). Admission labora-
tory values, including sCr were recorded. CrCl was determined with
the Cockcroft-Gault equation:
(140-Age)*Weight in Kg*(0.85 if female)
CrCl ⫽
(72*sCr)
Admission and worst CrCl during the index hospitalization were
calculated and a percentage change was computed. Delayed ischemic
neurological deficits and new infarcts secondary to cerebral vaso-
spasm were recorded by the primary team. Finally, the number of
contrast radiographic studies, predominantly contrast-enhanced CT
scans of the brain and digital subtraction cerebral angiograms, was
recorded. Basic laboratory values were assessed daily, including at a
minimum serum chemistries and hematology.
Outcome Measures
Survival and functional outcome was assessed at discharge or 14
days (whichever occurred first), 3 months, and 12 months using
the mRS.
 Zacharia et al Outcomes in Renal Dysfunction After Aneurysmal SAH 2377

Statistical Analysis Table 1. Baseline Characteristics of the 787 Study Patients

Where appropriate, continuous variables were dichotomized based
Characteristic No. of Patients (%)
on clinical cut points or median values. The association of mild renal
dysfunction (a decrement in CrCl of ⬎25%) was assessed in a Age, years 55.1⫾15
univariate analysis using Student t test for analysis of continuous Sex
variables and ␹2 for dichotomized variables. Nonparametric tests
were used when appropriate. Variables found to be significant on Male 221 (28.2)
univariate analysis were entered into a multiple logistic regression Female 562 (71.8)
model based on the clinical relevance of each variable. A model was
Pre-existing renal disease 19 (2.5)
constructed using a forward-stepwise multiple logistic regression,
including all clinically relevant variables with P⬍0.25 in univariate Diabetes mellitus 57 (7.4)
analysis examining the effect of mild renal dysfunction on functional Premorbid mRS
outcome at 3 months after controlling for known predictors of
0 713 (91.9)
outcome, including, age, H&H grade, clinical vasospasm, pre-
existing history of diabetes mellitus or renal failure, and premorbid 1–3 58 (7.5)
mRS. Tests for interactions were performed for all clinically signif- 4–5 5 (0.6)
icant variables in the multivariate model. Date of death after aSAH
was determined as accurately as possible based on medical records Admission H&H
and 3- and 12-month follow-up. A Kaplan–Meier curve was created Grade I–II 348 (44.8)
and analyzed by log-rank test. Results are reported as mean⫾SD, and Grade III–V 429 (55.2)
significance was set at P⬍0.05 for all analyses. Data analysis was
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017

performed with commercially available statistical software (JMP Admission sCR, mg/dL 0.82⫾0.6
Version 7; SAS Inc). Admission CrCl, mL/min 108.2⫾49.0
Peak sCr, mg/dL 1.03⫾0.9
Results Clinical vasospasm 159 (20.2)
Table 1 depicts baseline characteristics of 787 consecutive
patients with aSAH that were included in this analysis. The New infarct secondary to vasospasm 96 (12.7)
mean age of patients at admission was 55.1⫾15 years. Five Total contrast-enhanced imaging studies 2.3⫾1.6
hundred sixty-two (71.8%) of the patients were women. Nine- Decrease in CrCl ⱖ25% 179 (23.1)
teen (2.5%) of the patients had pre-existing renal disease, and Disposition at discharge
57 (7.4%) had diabetes mellitus. Premorbid mRS was less Home 410 (52.1)
than or equal to one in 726 (94.9%), and 182 (24.4%) of the Acute rehabilitation 171 (21.7)
patients presented with poor-grade H&H (Grades IV to V).
Skilled nursing facility 87 (11.1)
Average admission sCr was 0.82⫾0.6 mg/dL and peak sCr
Hospital 18 (2.3)
was 1.03⫾0.9 mg/dL in these patients. Average admission
CrCl was 107.7⫾49.1 mL/min. Patients received an average Deceased 100 (12.7)
of 2.3⫾1.6 contrast-enhanced imaging studies. One hundred 14-day mRS
fifty-nine (20.2%) patients experienced clinical deterioration 0 43 (5.5)
secondary to cerebral vasospasm and 96 (12.7%) developed a 1–3 338 (43.1)
new infarct secondary to vasospasm. The majority (73.8%) 4–6 403 (51.4)
were discharged either to home or an acute rehabilitation
center. Finally, almost half of the patients (48.7%) were
discharged with a favorable mRS (0 to 3). discharged with a good mRS (OR, 0.5; P⬍0.0001) and have
One hundred seventy-nine (23.1%) patients were at risk for a good 3-month mRS (OR, 0.5; P⫽0.004). There was a trend
renal failure during their hospitalization as defined by the toward worse outcome at 12 months in the at-risk group (OR,
RIFLE criteria. Table 2 compares demographic, medical 0.6; P⫽0.0712).
characteristics, and outcome between patients at risk and Univariate analysis of the primary outcome, poor mRS (4
patients not at risk for renal failure. Patients at risk for renal to 6) at 3 months is shown in Table 3. Patients with poor
failure were more likely to have a poor-grade admission outcome were older (63⫾15 versus 52.4⫾13.9 years;
H&H (31.3% versus 21.9%; P⫽0.0124). There were no P⬍0.0001), less likely to have a premorbid mRS ⱕ1 (87.7%
significant differences, however, in premorbid mRS, age, sex, versus 96.9%; P⫽0.0002), and more likely to have diabetes
incidence of clinical vasospasm or new infarct from vaso- mellitus (16.2% versus 4.1%; P⬍0.0001). Patients with a
spasm, and history of renal disease or diabetes mellitus. poor outcome were 7 times more likely to have a poor-grade
Those who developed risk for renal failure had better baseline admission H&H (48.4% versus 11.8%; OR, 7; P⬍0001),
renal function as evidenced by admission sCr (0.7⫾0.3 more than twice as likely to experience a clinical deteriora-
versus 0.86⫾0.7 mg/dL; P⬍0.0001) and CrCl (121.8⫾57.2 tion due to vasospasm (32.4% versus 17.2%; OR, 2.32;
versus 105.2⫾44.6 mL/min; P⫽0.0013), but went on to P⫽0.0003), almost 4 times more likely to develop a new
develop higher peak sCr (1.35⫾1.1 versus 0.93⫾0.8 mg/dL; infarct secondary to vasospasm (27.4% versus 8.8%; OR, 3.9;
P⬍0.0001). There was no difference in the number of P⬍0.0001), and twice as likely to develop risk for renal
contrast-enhanced imaging studies between these 2 groups. failure (32.5% versus 18.1%; OR, 2.2; P⫽0.004). Admission
Risk of in-hospital death was 4 times higher in those patients CrCl was significantly lower in those with poor outcome
at risk for renal failure (OR, 4.0; P⬍0.0001). Furthermore, (96.9⫾54 versus 114.4⫾46.8 mL/min; P⬍0.0001) and ad-
those at risk for renal failure were half as likely to be mission (0.98⫾1.24 versus 0.77⫾0.35; P⫽0.0125) and peak
 2378 Stroke July 2009

Table 2. Demographic and Outcome Comparisons for Not-at-Risk and At-Risk Subgroups
Not at Risk for At Risk for
Variable Renal Failure Renal Failure OR† P Value
Age, years 54.6⫾14.8* 56.5⫾15.8* N/A NS
Female 420 (71.0) 137 (76.5) N/A NS
Premorbid Rankin ⱕ1 562 (95.6) 162 (92.1) N/A NS
Poor-grade admission H&H (Grades IV–V) 130 (21.9) 56 (31.3) 1.6 0.0124
Pre-existing renal disease 16 (2.7) 3 (1.7) N/A NS
Pre-existing diabetes mellitus 39 (6.7) 17 (9.8) N/A NS
Admission sCr, mg/dL 0.86⫾0.70 0.70⫾0.30 N/A ⬍0.0001
Admission CrCl, mL/min 105.2⫾44.6* 121.8⫾57.2* N/A 0.0013
Peak sCr, mg/dL 0.93⫾0.80* 1.35⫾1.10* N/A ⬍0.0001
Total no. of contrast-enhanced imaging studies 2.3 (1.6) 2.4 (1.7) N/A NS
In-hospital death 50 (8.4) 48 (26.8) 4.0 ⬍0.0001
Clinical vasospasm 111 (18.7) 44 (24.6) N/A NS
New infarct secondary to vasospasm 65 (11.5) 28 (16) N/A NS
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017

14-day outcome
mRS ⱕ3 (good outcome) 316 (53.3) 61 (34.3) 0.5 ⬍0.0001
3-month outcome
mRS ⱕ3 (good outcome) 362 (80.6) 80 (67.8) 0.5 0.004
12-month outcome
mRS ⱕ3 (good outcome) 359 (90.0) 79 (83.2) 0.6 0.0712
Results are shown as no. of patients with percentage of cohort in parentheses, except as noted.
*Mean⫾SD.
†OR of at-risk versus not-at-risk subgroups.
N/A indicates not applicable; NS, nonsignificant.

creatinine was significantly higher (1.31⫾1.6 versus
0.89⫾0.44 mg/dL; P⫽0.0003). Patients with a poor 3-month
mRS had a greater number of contrast-enhanced imaging
studies (2.8⫾2.2 versus 2.2⫾1.4; P⫽0.0201). These 2 groups
did not differ with regard to sex or history of renal disease.
Significant and clinically relevant factors from the univar-
iate analysis were entered in a multiple logistic regression
model to determine independent predictors of poor mRS at 3
months, the results of which are shown in Table 4. The model
revealed 7 independent predictors of poor outcome at 3
months. Age and poor admission H&H, known predictors of
poor outcome after aSAH, were associated with ORs of 2 per
10-year increase (P⬍0.0001) and 6.92 (P⬍0.0001), respec-
tively. Per unit increase in the total number of contrast-
enhanced imaging studies, the odds of a poor 3-month mRS
were increased by 1.27 (P⫽0.0013). History of diabetes
mellitus was associated with an almost tripling of the odds of
a poor 3-month outcome (OR, 2.78; P⫽0.021), and a pre-
morbid mRS ⱕ1 was significantly protective (OR, 0.28;
P⫽0.0184). The development of a new infarct secondary to
vasospasm portended a significantly worse 3-month outcome
(OR, 3.6; P⫽0.0002). Finally, those patients who developed
risk for renal failure were twice as likely to have a poor
discharge mRS (OR, 2.01; P⫽0.021). Of note, there were no
significant interactions between the included variables.
We then performed a second multiple regression using all
significant factors from prior analysis and stratified patients
into 3 groups based on RIFLE classification (not at risk, at
risk, and injury/failure). In this analysis, decline of renal
function sufficient to be classified into each advanced group
was associated with an increase in the odds of poor outcome
(OR, 1.11 and 3.96; P⫽0.0038). All previously significant
factors remained so in this model.
Kaplan–Meier curves comparing survival in those not at
risk, those at risk (RIFLE R), and those with renal injury or
failure (RIFLE I and F) are shown in Figure 2. Median
survival was greater than the observation period for all
groups. However, although the time to 75% survival in those
not at risk was still greater than the observation period, it was
only 90 days for the at-risk group and 38 days for the injury
and failure group. Log-rank test demonstrates a significant
(P⬍0.0001) difference among the 3 survival curves.
Discussion
With improvements in neurocritical care, focus has shifted to
examining the role of nonneurological complications on
outcome after aSAH.2,4,5 In fact, these complications can rival
the frequency of morbidity and mortality from neurological
complications.10 Recent studies of patients with brain trauma
and aSAH have demonstrated that nearly 80% of these
patients develop dysfunction of at least one nonneurological
organ system.19,20 Not surprisingly, nonneurological organ
dysfunction correlates with the severity of neurological im-
pairment. The majority of these studies pay particular atten-
tion to cardiopulmonary dysfunction, but the burden of renal,
hematologic, and hepatic injuries remains largely unstudied.
It is thus paramount that we understand the effect of neuro-
logical insults on the remainder of the body and vice versa.
 Zacharia et al Outcomes in Renal Dysfunction After Aneurysmal SAH 2379

Table 3. Univariate Analysis of Outcome at 3-Month Follow-Up

Good 3-Month Poor 3-Month
Variable Rankin Rankin OR† P Value
Age, years 52.4⫾13.9* 63⫾15* N/A ⬍0.0001
Female 320 (72.1) 95 (75.4) N/A NS
Premorbid Rankin ⱕ1 434 (96.9) 107 (87.7) 0.23 0.0002
Poor-grade admission H&H 53 (11.8) 61 (48.4) 7 ⬍0.0001
Pre-existing renal disease 10 (2.3) 7 (6) N/A NS
Pre-existing diabetes 18 (4.1) 19 (16.2) 4.6 ⬍0.0001
Admission Cr, mg/dL 0.77⫾0.35* 0.98⫾1.24* N/A 0.0125
Admission CrCl, mL/min 114.4⫾46.8* 96.9⫾54* N/A ⬍0.0001
Peak creatinine 0.89⫾0.44* 1.31⫾1.6* N/A 0.0003
CrCl decrease ⱖ25% 80 (18.1) 38 (32.5) 2.2 0.0040
Clinical vasospasm 77 (17.2) 41 (32.4) 2.32 0.0003
New infarct secondary to vasospasm 38 (8.8) 34 (27.4) 3.9 ⬍0.0001
Total contrast-enhanced imaging studies 2.2⫾1.4* 2.8⫾2.2* N/A 0.0201
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017

Results are shown as no. of patients with percentage of cohort in parentheses, except as noted.
*Mean⫾SD.
†OR of poor versus good 3-month Rankin.
N/A indicates ; NS, nonsignificant.

In a series of 787 consecutively treated patients with patients admitted with congestive heart failure. Lassnigg et al26
aSAH, we demonstrate that a seemingly inconsequential and Loef et al27 explored this phenomenon in cardiac and
decrement in renal function can adversely affect outcome noncardiac surgery patients, respectively, and concluded that
independent of other known predictors. To our knowledge, elevations as small as 25% in sCr increase the risk of short-
this is the first study focusing on the effects of renal and long-term mortality. Recently, Chertow and colleagues
dysfunction as defined by the RIFLE criteria in the aSAH extended these findings to a large cohort of hospitalized
population. The need for a system to standardize and classify patients and showed that an increase in sCr of 0.3 to 0.4
renal dysfunction was fulfilled by the recently developed mg/dL is associated with a 70% increase in the risk of death.23
RIFLE criteria. In addition, the need to classify the severity of We used the RIFLE criteria to determine which patients
the syndrome rather than only considering its most severe were at risk for renal failure based on a decrement in
form was emphasized.11 In light of this new classification, the estimated CrCl of ⬎25%. Consistent with previous reports in
link between mild AKI and outcome is just being elucidated. the literature, 23.1% of patients developed risk for renal
Epidemiological studies have found that 10% to 17% of failure during their hospitalization.13,21,22 As might be ex-
patients in an intensive care unit achieved a maximum RIFLE pected, these individuals were older and more frequently
class of R, 5% to 27% I, and 3.5% to 28% F.13,21,22 These poor-grade H&H. Interestingly, however, there was no dif-
increasing RIFLE classes were associated with a nearly linear ference in premorbid Rankin score, incidence of clinical
increase in hospital mortality.11 Several groups have exam- vasospasm, or history of renal disease or diabetes mellitus
ined the association between small changes in sCr concen- between these groups. Most importantly, those at risk for
tration in a number of unique clinical settings.23 Studies by renal failure were 4 times more likely to experience an
Gottlieb et al24 and Smith et al25 demonstrated an increase in in-hospital death than those not at risk. These results are
mortality and length of stay with small increases in sCr in consistent with published data from Hoste and colleagues
who demonstrated an OR of hospital mortality of 2.5 for
Table 4. Independent Predictors of Poor 3-Month Outcome those patients with RIFLE R.13
Covariates OR 95% CIs P Value There was significantly worse discharge and 3-month
outcome and a trend toward worse 12-month outcome in
Age 2* 1.05–1.09 ⬍0.0001
those at risk for renal failure in our aSAH cohort. In fact,
Total scans 1.27† 1.1–1.48 0.0013
those at risk were nearly half as likely to have a good outcome
Poor admission H&H 6.92 N/A ⬍0.0001 at 12 months than those not at risk. In a more parsimonious
Premorbid mRS ⱕ1 0.28 N/A 0.0184 multiple logistic regression model controlling for known
Risk of renal failure 2.01 N/A 0.021 predictors of outcome, being at risk for renal failure was
New infarct secondary 3.6 N/A 0.0002 associated with a doubling of the likelihood of poor outcome
to vasospasm at 3 months. Importantly, across all H&H grades, a decrement
Diabetes 2.78 N/A 0.0216 in CrCl ⱖ25% is associated with a significantly worse
*Per 10-year increase. 3-month outcome. Survival analysis revealed that patients at
†Per unit increase. risk for renal failure were 60% as likely to be alive at 1 year
N/A indicates. than patients not at risk. As demonstrated by other groups,
 2380 Stroke July 2009
our analyses revealed a significant relationship between
worse outcome and increasing RIFLE classification. RIFLE
R was associated with a doubling of the odds of poor
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017
outcome, and becoming RIFLE I or F was associated with an
almost quadrupling of the risk of poor outcome.
Although the results are convincing, we recognize that
there are several limitations of our study. Retrospective
analysis of prospectively collected data has many of the
methodological shortcomings of purely retrospective studies
and, as such, it remains difficult to accurately assess causal-
ity. We attempted to control for disease severity, but given
the retrospective nature of the analyses, we are unable to
account for all possible confounders. It is possible that the
decrease in CrCl is simply indicative of disease severity, and
these patients are more likely to receive nephrotoxic drugs,
contrast-enhanced imaging studies, experience hypotension
and hypoxia, and have concomitant injury in other organ
systems. Published studies, however, have demonstrated a
consistently high relative risk associated with AKI despite
adjustment for comorbid conditions.23 In addition, our anal-
yses demonstrate that AKI increased the odds of a poor
outcome at all H&H levels. Although residual confounding
could lessen the magnitude of the risk estimates, additional
covariates would be unlikely to extinguish the increased odds
of poor outcome. Furthermore, we lacked sufficient data to
identify the etiology or type of renal failure and were unable
to quantify urine output.
Regardless, we feel that our study findings are of signifi-
cant clinical interest. A decrease of 25% in creatinine clear-
ance can represent an increase in sCr of as little as 0.2 mg/dL,
a change that rarely prompts alteration of the treatment
course. As we continue to improve the neurological manage-
ment of patients with aSAH, there is a shift toward increasing
disease severity and consequently more critically ill patients.
These poor-grade patients with aSAH are invariably managed
in an intensive care unit setting, undergo surgical or endo-
vascular procedures, have diminished mental capacity, and
are frequently intubated. These factors combine to place these
patients at a remarkably high risk for medical complications.
The cardiopulmonary contribution to mortality and morbidity
in these patients has been extensively examined. Administra-
tion of hyperosmolar therapy and nephrotoxic antibiotics as
well as the frequent number of contrast-enhanced radiograph-
ic images and endovascular interventions, however, places
Figure 2. Survival after aSAH stratified by RIFLE
risk classification.
these patients at particular risk for renal complications. This
study highlights the importance of early recognition of renal
risk and prompts clinicians to practice renal hygiene. That is,
avoidance of redundant contrast-enhanced imaging studies,
adequate hydration and renal protection strategies, and frequent
screening of medication lists for potentially nephrotoxic drugs
and dose adjustment for those with renal impairment.
Conclusion
In a large, consecutive series of prospectively enrolled pa-
tients with aSAH, we demonstrate, using the newly defined
RIFLE classification for risk of renal failure, that even
seemingly insignificant decreases in creatinine clearance are
associated with significantly worse 3-month outcomes.
Whether renal dysfunction is merely a marker of disease
severity or an independent predictor of poor outcome cannot
be elucidated from retrospective analysis. Nevertheless, this
study highlights the importance of close surveillance of renal
function and stresses the value of renal hygiene in the aSAH
population.
Disclosures
None.
References
1. Johnston SC, Selvin S, Gress DR. The burden, trends, and demographics
of mortality from subarachnoid hemorrhage. Neurology. 1998;50:
1413–1418.
2. Solenski NJ, Haley EC Jr, Kassell NF, Kongable G, Germanson T,
Truskowski L, Torner JC. Medical complications of aneurysmal sub-
arachnoid hemorrhage: a report of the multicenter, cooperative aneurysm
study. Participants of the multicenter cooperative aneurysm study. Crit
Care Med. 1995;23:1007–1017.
3. Wijdicks EF, Kallmes DF, Manno EM, Fulgham JR, Piepgras DG. Sub-
arachnoid hemorrhage: neurointensive care and aneurysm repair. Mayo
Clin Proc. 2005;80:550 –559.
4. Holmes JM. The medical management of subarachnoid haemorrhage.
BMJ. 1958;1:788 –790.
5. Weir B. Medical aspects of the preoperative management of aneurysms:
a review. Can J Neurol Sci. 1979;6:441– 450.
6. Schell AR, Shenoy MM, Friedman SA, Patel AR. Pulmonary edema
associated with subarachnoid hemorrhage. Evidence for a cardiogenic
origin. Arch Intern Med. 1987;147:591–592.
7. Marion DW, Segal R, Thompson ME. Subarachnoid hemorrhage and the
heart. Neurosurgery. 1986;18:101–106.
8. Parkinson D, Stephensen S. Leukocytosis and subarachnoid hemorrhage.
Surg Neurol. 1984;21:132–134.
9. Spallone A, Mariani G, Rosa G, Corrao D. Disseminated intravascular
coagulation as a complication of ruptured intracranial aneurysms. Report
of two cases. J Neurosurg. 1983;59:142–145.
 Zacharia et al Outcomes in Renal Dysfunction After Aneurysmal SAH
10. Haley EC Jr, Kassell NF, Torner JC. The international cooperative study
on the timing of aneurysm surgery. The North American experience.
Stroke. 1992;23:205–214.
11. Kellum JA, Bellomo R, Ronco C. Definition and classification of acute
kidney injury. Nephron Clin Pract. 2008;109:c182–187.
12. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal
failure— definition, outcome measures, animal models, fluid therapy and
information technology needs: the second international consensus con-
ference of the Acute Dialysis Quality Initiative (ADQI) group. Crit Care.
2004;8:R204 –212.
13. Hoste EA, Kellum JA. Acute kidney injury: epidemiology and diagnostic
criteria. Curr Opin Crit Care. 2006;12:531–537.
14. Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure on
mortality. A cohort analysis. JAMA. 1996;275:1489 –1494.
15. McCullough PA. Radiocontrast-induced acute kidney injury. Nephron
Physiol. 2008;109:61–72.
16. Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Palevsky PM, Fine
MJ. Incidence and outcomes of contrast-induced AKI following
computed tomography. Clin J Am Soc Nephrol. 2008;3:1274 –1281.
17. Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Sonel AF, Fine MJ,
Palevsky PM. Prevention, incidence, and outcomes of contrast-induced
acute kidney injury. Arch Intern Med. 2008;168:1325–1332.
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017
18. Wartenberg KE, Schmidt JM, Claassen J, Temes RE, Frontera JA, Ostap-
kovich N, Parra A, Connolly ES, Mayer SA. Impact of medical compli-
cations on outcome after subarachnoid hemorrhage. Crit Care Med.
2006;34:617– 623; quiz 624.
19. Gruber A, Reinprecht A, Illievich UM, Fitzgerald R, Dietrich W, Czech
T, Richling B. Extracerebral organ dysfunction and neurologic outcome
2381
after aneurysmal subarachnoid hemorrhage. Crit Care Med. 1999;27:
505–514.
20. Zygun DA, Doig CJ, Gupta AK, Whiting G, Nicholas C, Shepherd E,
Conway-Smith C, Menon DK. Non-neurological organ dysfunction in
neurocritical care. J Crit Care. 2003;18:238 –244.
21. Ostermann M, Chang RW. Acute kidney injury in the intensive care unit
according to RIFLE. Crit Care Med. 2007;35:1837–1843; quiz 1852.
22. Uchino S, Bellomo R, Goldsmith D, Bates S, Ronco C. An assessment of
the RIFLE criteria for acute renal failure in hospitalized patients. Crit
Care Med. 2006;34:1913–1917.
23. Chertow GM, Burdick E, Honour M, Bonventre JV, Bates DW. Acute
kidney injury, mortality, length of stay, and costs in hospitalized patients.
J Am Soc Nephrol. 2005;16:3365–3370.
24. Gottlieb SS, Abraham W, Butler J, Forman DE, Loh E, Massie BM,
O’Connor CM, Rich MW, Stevenson LW, Young J, Krumholz HM. The
prognostic importance of different definitions of worsening renal function
in congestive heart failure. J Card Fail. 2002;8:136 –141.
25. Smith GL, Vaccarino V, Kosiborod M, Lichtman JH, Cheng S, Watnick
SG, Krumholz HM. Worsening renal function: what is a clinically mean-
ingful change in creatinine during hospitalization with heart failure?
J Card Fail. 2003;9:13–25.
26. Lassnigg A, Schmidlin D, Mouhieddine M, Bachmann LM, Druml W,
Bauer P, Hiesmayr M. Minimal changes of serum creatinine predict
prognosis in patients after cardiothoracic surgery: a prospective cohort
study. J Am Soc Nephrol. 2004;15:1597–1605.
27. Loef BG, Epema AH, Smilde TD, Henning RH, Ebels T, Navis G,
Stegeman CA. Immediate postoperative renal function deterioration in
cardiac surgical patients predicts in-hospital mortality and long-term
survival. J Am Soc Nephrol. 2005;16:195–200.
 Renal Dysfunction as an Independent Predictor of Outcome After Aneurysmal
Subarachnoid Hemorrhage: A Single-Center Cohort Study
Brad E. Zacharia, Andrew F. Ducruet, Zachary L. Hickman, Bartosz T. Grobelny, Luis
Fernandez, J. Michael Schmidt, Reshma Narula, Lauren N. Ko, Margot E. Cohen, Stephan A.
Mayer and E. Sander Connolly, Jr
Downloaded from http://stroke.ahajournals.org/ by guest on November 20, 2017

Stroke. 2009;40:2375-2381; originally published online May 21, 2009;

doi: 10.1161/STROKEAHA.108.545210
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2009 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/40/7/2375

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Stroke is online at:

http://stroke.ahajournals.org//subscriptions/