Acute Renal Failure Is Associated With Higher Death and Disability in Patients With Acute Ischemic Stroke
Analysis of Nationwide Inpatient Sample
Background and Purpose—Acute renal failure (ARF) in setting of acute ischemic stroke (AIS) is associated with worse outcome. We sought to determine the prevalence of ARF and effect on outcomes of patients with AIS.
Methods—Data from all patients admitted to US hospitals between 2002 and 2010 with a primary discharge diagnosis of ischemic stroke and secondary diagnosis of ARF were included. The effect of ARF on rates of intracerebral hemorrhage and discharge outcomes was analyzed after adjusting for potential confounders using logistic regression analysis.
Results—Of 7 068 334 patients with AIS, 372 223 (5.3%) had ARF during hospitalization. Dialysis was required in 2364 (0.6%) of 372 223 patients. Patients with AIS with ARF had higher rates of moderate to severe disability (41.3% versus 30%; P<0.0001), intracerebral hemorrhage (1.0% versus 0.5%; P<0.0001), and in-hospital mortality (8.4% versus 2.9%; P<0.0001) compared with those without ARF. After adjusting for confounding factors, patients with AIS with ARF had higher odds of moderate to severe disability (odds ratio, 1.3; 95% confidence interval, 1.3–1.4; P<0.0001), intracerebral hemorrhage (odds ratio, 1.4; 95% confidence interval, 1.3–1.6; P<0.0001), and death (odds ratio, 2.2; 95% confidence interval, 2.0–2.2; P<0.0001).
Conclusions—ARF in patients with AIS is associated with significantly higher rates of moderate to severe disability at discharge and in-hospital mortality.
There is increased recognition of acute renal failure (ARF) in hospitalized patients with guidelines such as the Kidney Disease Improving Global Outcomes consensus guidelines specifically addressing the issue of ARF. ARF is defined by an increase in the serum creatinine level of ≥0.3 mg/dL within 48 hours; a rise in serum creatinine level by 1.5× the baseline value within the previous 7 days; or a urine volume of <0.5 mL/kg of body weight per hour for 6 hours.1 ARF can be seen in 14% of patients with ischemic stroke.2 ARF is associated with increased hospital mortality among patients with acute ischemic stroke (AIS).2 Our study aimed at examining the effect of ARF on stroke outcomes at the time of discharge in a nationally representative sample of patients with AIS.
We used the data files from National Inpatient Sample from 2002 to 2010 for our analysis. A comprehensive synopsis on National Inpatient Sample data is available at http://www.hcup-us.ahrq.gov.
We used the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) primary diagnosis codes 433–437.1 to identify the patients admitted with ischemic stroke. We also used ICD-9-CM secondary diagnosis codes 584 and 593.9 to identify patients with ARF. We excluded all the cases with pre-existing renal failure. We also excluded all the cases with ICD-9-CM secondary diagnosis codes of chronic kidney disease III (585.3), IV (585.4), V (585.5 and 585.6), unspecified (585.9), and renal transplant (556.1–556.9, 996.81, and V42.0). We divided patients with AIS into with and without ARF. An extended Methods is provided in the online-only Data Supplement.
A total of 372 223 (5.3%) patients admitted with AIS had ARF. Dialysis was required in 2364 (0.6%) of 372 223 patients. Patients with ARF were older (mean age, 74±28 versus 72±31 years; P<0.0001) and predominantly men (52%; P<0.0001). Comorbid conditions, such as hypertension, diabetes mellitus, atrial fibrillation, and congestive heart failure, were significantly higher in patients with ARF (Table I in the online-only Data Supplement). In-hospital complications, such as pneumonia, deep venous thrombosis, pulmonary embolism, urinary tract infection, sepsis, myocardial infarction, and gastrointestinal bleeding, were also significantly higher in patients with ARF. Patients with ARF were more likely to receive procedures such as intubation, tracheotomy, blood transfusion, and gastrostomy tube placement. Length of stay (6±19 versus 4±10; P<0.0001) and mean hospital charges ($38 613±137 427 versus $24 474±67 582; P<0.0001) were higher in patients with ARF as compared with those without ARF.
Patients with ARF had higher incidence of moderate to severe disability (41.3% versus 30%; P<0.0001), intracerebral hemorrhage (1% versus 0.5%; P<0.0001), and in-hospital mortality (8.4% versus 2.9%; P<0.0001).
Moderate to severe disability (odds ratio, 1.3; 95% confidence interval, 1.3–1.4; P<0.0001), intracerebral hemorrhage (odds ratio, 1.4; 95% confidence interval, 1.3–1.6; P<0.0001), and in-hospital mortality (odds ratio, 2.2; 95% confidence interval, 2.0–2.2; P<0.0001) were statistically significant after adjusting for age, sex, race/ethnicity, congestive heart failure, diabetes mellitus, hypertension, dyslipidemia, chronic lung disease, atrial fibrillation, nicotine dependence, pneumonia, myocardial infarction, gastrointestinal bleeding, and sepsis (Table II in the online-only Data Supplement).
Incidence of ARF was ≈5.3% in our study, which was lower compared with previous studies of patients with stroke.2 The variation in frequency of ARF is probably explained by the different definitions used in previous studies. Patients with AIS can experience ARF in several ways: hemodynamic variations in blood pressure, prerenal/acute tubular injury from volume depletion, contrast-induced nephropathy, obstructive renal failure, and thromboembolic phenomenon with tissue-type plasminogen activator. Our study was not directed toward identifying the mechanistic explanation for the observed occurrence. We also observed that dialysis was required in 2364 patients who developed ARF. The proportion of patients requiring dialysis was similar to those reported in previous studies among critically ill patients.3,4
We observed higher odds of in-hospital complications, moderate to severe disability, intracerebral hemorrhage, and in-hospital mortality in patients with AIS who developed ARF. The association persisted after adjustment for age, sex, and other confounders. Our study findings are similar to Tsagalis et al5 who also demonstrated that ARF was related to higher mortality. Khatri et al,2 studying patients with both ischemic and hemorrhagic stroke, showed that ARF in the setting of ischemic stroke was related to increased mortality. Our study found similar findings using data at a national level that is devoid of biases introduced by patient demographics and local institutional practices.
The mechanism underlying the higher rate of adverse outcomes in patients with ARF is unclear. Patients with AIS and ARF have higher rates of underlying comorbidities, such as hypertension, diabetes mellitus, atrial fibrillation, and congestive heart failure. The higher rate may be secondary to higher rates of underlying comorbidities with subsequently higher rates of in-hospital complications such as pneumonia, deep venous thrombosis, pulmonary embolus, urinary tract infection, sepsis, myocardial infarction, and gastrointestinal bleed. The higher rates of adverse outcomes may be related to procedural complications as patients with ARF had higher rates of in-hospital procedures such as gastrostomy, mechanical ventilation, transfusion, and tracheostomy. There is a possibility that ARF directly contributes to worsening of neurological ischemic injury. Published data from animal models suggest that ARF leads to inflammation and functional changes in brain.6 However, this has not been evaluated in the setting of ischemic stroke.
Our study is based on National Inpatient Sample data set and limitations are described previously.7 National Inpatient Sample data set depends on the accuracy of diagnoses and procedure codes. Diagnosis of AIS has a true positive rate of ≤84% in previous population-based studies.8 ICD-9-CM codes for ARF have a sensitivity of 35.4%, specificity of 97.7%, positive predictive value of 47.9%, and negative predictive value of 96.1%. ICD-9-CM codes for ARF requiring dialysis had positive predictive value of 94.0% and negative predictive value of 90%.9 The low sensitivity of ARF may result in failure to identify patients with minor or unrecognized ARF. Therefore, the results may be reflective of patients with moderate to severe ARF and the death and disability may consequently be higher than if patients with minor ARF were included. Because of the high specificity of codes for both ischemic stroke and ARF, the study accurately depicts the relationship between the 2 acute diseases but may underestimate the prevalence of both in the data set. Another limitation of our study was that we did not adjust for stroke severity using National Institutes of Health Stroke Scale score in our model because the information was not available in the data. Our study was able to assess only in-hospital outcomes based on discharge destination. However, discharge destination seems to correlate strongly with modified Rankin scale at 3 and 12 months and has been described elsewhere previously.10
In summary, ARF is associated with higher odds of moderate to severe disability and in-hospital mortality among patients with AIS. Further studies are required to determine whether the rate of ARF can be reduced in patients with ischemic stroke with subsequent reduction in death and disability.
Sources of Funding
Dr Qureshi is supported by National Institutes of Health U01-NS062091-01A2.
↵* Drs Khursheed and Daimee are joint third authors and contributed equally.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.114.004672/-/DC1.
- Received January 2, 2014.
- Revision received January 31, 2014.
- Accepted February 3, 2014.
- © 2014 American Heart Association, Inc.
- Tsagalis G,
- Akrivos T,
- Alevizaki M,
- Manios E,
- Theodorakis M,
- Laggouranis A,
- et al
- Liu M,
- Liang Y,
- Chigurupati S,
- Lathia JD,
- Pletnikov M,
- Sun Z,
- et al
- Broderick J,
- Brott T,
- Kothari R,
- Miller R,
- Khoury J,
- Pancioli A,
- et al
- Waikar SS,
- Wald R,
- Chertow GM,
- Curhan GC,
- Winkelmayer WC,
- Liangos O,
- et al