Article Text

Download PDFPDF

Original research
Risk of acute kidney injury associated with neuroimaging obtained during triage and treatment of patients with acute ischemic stroke symptoms
  1. Shelby L Hall1,2,
  2. Stephan A Munich1,2,
  3. Marshall C Cress1,2,
  4. Leonardo Rangel-Castilla1,2,
  5. Elad I Levy1,2,3,4,
  6. Kenneth V Snyder1,2,3,4,5,
  7. Adnan H Siddiqui1,2,3,4,6
  1. 1Department of Neurosurgery, Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, Buffalo, New York, USA
  2. 2Department of Neurosurgery, Gates Vascular Institute at Kaleida Health, Buffalo, New York, USA
  3. 3Department of Radiology, Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, Buffalo, New York, USA
  4. 4Toshiba Stroke and Vascular Research Center, University at Buffalo, State University of New York, Buffalo, New York, USA
  5. 5Department of Neurology, Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, Buffalo, New York, USA
  6. 6Jacobs Institute, Buffalo, New York, USA
  1. Correspondence to Dr Adnan H Siddiqui, University at Buffalo Neurosurgery, 100 High Street, Suite B4, Buffalo, NY 14203 USA; asiddiqui{at}


Background Combining non-contrast CT (NCCT), CT angiography (CTA), and CT perfusion (CTP) imaging (referred to as a CT stroke study, CTSS) provides a rapid evaluation of the cerebrovascular axis during acute ischemic stroke. Iodinated contrast-enhanced CT imaging is not without risk, which includes renal injury. If a patient's CTSS identifies vascular pathology, digital subtraction angiography (DSA) is often performed within 24–48 h. Such patients may receive multiple administrations of iodinated contrast material over a short time period.

Objective We aimed to evaluate the incidence of acute kidney injury (AKI) in patients who underwent a CTSS and DSA for evaluation of acute ischemic symptoms or for stroke intervention within a 48 h period between August 2012 and December 2014.

Methods We identified 84 patients for inclusion in the analysis. Patients fell into one of two cohorts: AKI, defined as a rise in the serum creatinine level of ≥0.5 mg/dL from baseline, or non-AKI. Clinical parameters included pre- and post-imaging serum creatinine level, time between CTSS and DSA, and type of angiographic procedure (diagnostic vs intervention) performed.

Results Four patients (4.7%) experienced AKI, one of whom had baseline renal dysfunction (defined as baseline serum creatinine level ≥1.5 mg/dL). The mean difference between baseline and peak creatinine values was found to be significantly greater in patients with AKI than in non-AKI patients (1.65 vs −0.09, respectively; p=0.0008).

Conclusions This study provides preliminary evidence of the safety and feasibility of obtaining CTSS with additional DSA imaging, whether for diagnosis or intervention, to identify possible acute ischemic stroke.

  • Stroke
  • CT
  • CT Angiography
  • CT perfusion
  • Angiography

Statistics from


In the routine diagnosis and treatment of acute ischemic stroke, basic imaging of the brain and cerebrovasculature has become common practice. Routine evaluation of patients presenting with stroke-like symptoms typically includes non-contrast CT (NCCT) of the head.1 With the recent publication of five trials2–6 comparing endovascular and routine medical treatment for patients with acute anterior circulation occlusion, the importance of precise identification of the occlusion site during initial patient triage was highlighted. This localization is efficiently obtained by performing contrast CT angiography (CTA) and perfusion imaging studies. By combining NCCT imaging with CTA from the aortic arch to skull vertex and CT perfusion (CTP) imaging, the entire cerebrovascular axis can be visualized during acute ischemic stroke events.7 Although rapid and effective, iodinated contrast-enhanced CT imaging is not without risk, the primary of which is renal injury. This may be of particular concern in patients who are elderly, diabetic, and those with baseline renal dysfunction.8 Recent studies suggest these imaging strategies to be safe (associated with a low incidence of contrast nephropathy) in the evaluation of acute ischemic stroke.7–9

Patients who present to our institution with symptoms suggestive of acute ischemic stroke routinely undergo investigation with a stroke CT protocol (herein referred to as CT stroke study (CTSS)), which consists of NCCT scan of the brain, CTP of the head, and CTA of the head and neck. If a favorable penumbral pattern is identified in conjunction with an occlusion site that is amenable to endovascular intervention, patients are taken immediately to the interventional suite for possible intervention. Patients in whom no acute occlusion is identified but whose CTSS is concerning for non-occlusive vascular pathology (eg, carotid stenosis or intracranial atherosclerotic disease) typically undergo further investigation with digital subtraction angiography (DSA) within the next 24–48 h. Therefore, at our institution, patients routinely receive multiple administrations of iodinated contrast material over a relatively short time period. In this study we aimed to evaluate the incidence of acute kidney injury (AKI) in patients who underwent a CTSS and a DSA within a 48 h period.


This retrospective study was approved by the local Institutional Review Board. Patients who had received both a CTSS and DSA within the same 48 h period for evaluation of acute ischemia or for stroke intervention between August 2012 and December 2014 were identified from our clinical database. In accordance with the CTSS protocol, 130 mL of an iodinated contrast agent (80 mL for CTA and 50 mL for CTP) is administered. The typical agent used is Omnipaque 350 (GE Healthcare). If there is concern that the patient has compromised renal function, Visipaque 320 (GE Healthcare) is substituted. This is typically considered to be the case at our institution if the patient has a glomerular filtration rate (GFR) <50ml/min/1.73m2 or a serum creatinine level ≥1.5 mg/dL. The decision for this substitution is based on a multidisciplinary discussion between the emergency room CT technician, nurse, and emergency room physician. A similar process is undertaken in the angiography suite as well, with the inclusion of the angiographer rather than the emergency room physician.

Laboratory studies for baseline serum creatinine, blood urea nitrogen (BUN), and GFR that were obtained immediately preceding the CTSS were recorded from our clinical database. For the purpose of identifying AKI due to contrast administration associated with diagnostic studies evaluating acute ischemic stroke symptoms, the patients’ renal function was monitored for an average of 7 days. Serum creatinine and BUN levels were drawn immediately post-CTSS and DSA. They subsequently were drawn at the discretion of the treating physicians throughout the patients’ hospital admissions. As is customary, patients in whom normal values were obtained subsequently underwent less frequent laboratory evaluations than those in whom there was concern for the development of renal dysfunction. Baseline renal dysfunction was defined as a baseline serum creatinine level of ≥1.5 mg/dL,8 whereas AKI was defined as a rise of ≥0.5 mg/dL in serum creatinine level above the baseline value.10 ,11

Statistical analysis

To test for statistical differences we used the Wilcoxon–Mann–Whitney test for continuous variables and the Fisher exact test for categorical values. p<0.05 was considered statistically significant


We identified 111 patients who underwent CTSS and DSA imaging for stroke intervention and/or evaluation of ischemic symptoms within the specified 48 h time period. Of these 111 patients, 13 did not have any post-procedure laboratory studies and were not included in our statistical analysis. Given that we aimed to study the effect of multiple contrast administrations in the setting of acute stroke evaluation, eight patients were excluded because they did not have both imaging studies completed within 48 h. Five patients were excluded because the amount of contrast agent administered during DSA was not available. An additional patient was found to have subarachnoid hemorrhage and was therefore excluded. A total of 84 patients were included in the statistical analysis.

Of the 84 patients who met our inclusion criteria, only four patients (4.7%) who experienced a rise in the serum creatinine level of ≥0.5 mg/dL were identified and therefore met the criteria for AKI. Continuous and categorical variables for the AKI and non-AKI cohorts are compared in table 1. According to our definition of baseline renal dysfunction, 11 of the 84 patients presented with underlying renal dysfunction, only one of whom experienced AKI. Each of the remaining three patients with AKI had a normal baseline serum creatinine level (<1.5 mg/dL). The mean peak creatinine value for the AKI group was significantly greater than that for the non-AKI group (3.00 mg/dL and 0.94 mg/dL, respectively; p=0.0009, table 2). Analysis showed the mean baseline creatinine values between the two cohorts to be statistically significant (p=0.0459). To account for this difference in baseline creatinine value, we also analyzed the mean difference between baseline and peak creatinine (Δ creatinine) values, which was found to be significantly greater in patients with AKI than in non-AKI patients (1.65 vs −0.09, respectively; p=0.0008).

Table 1

Continuous and categorical variables compared between study cohorts

Table 2

Summary statistics and Wilcoxon–Mann–Whitney test for continuous variables*

The mean amount of contrast material used for DSA (AKI=173.75 mL, non-AKI=159.55 mL; p=0.6059) and CTSS (AKI=130 mL, non-AKI=128.75 mL; p=0.7807) was similar and did not predict the development of AKI. Similarly, the contrast agent used for these imaging protocols did not render an additional risk of developing AKI (AKI: Omnipaque=2 cases, Visipaque=2 cases; non-AKI: Omnipaque=43 cases, Visipaque=37 cases; p=1.000, tables 3 and 4). The type of DSA performed (diagnostic vs interventional) also was not a statistically significant factor in predicting the development of AKI (table 4). Lastly, the mean time between CTSS and DSA studies (AKI: 6 h 9 min (±8:33), non-AKI: 12 h 55 min (±14:53); p=0.9749) showed no statistically significant difference between the two cohorts (table 5).

Table 3

Frequency and percentage of categorical variables (total n=84 patients)

Table 4

Contingency table and Fisher exact test for categorical variables*

Table 5

Descriptive statistics for time elapsed (h:min:s) from CTSS to DSA by group*


Increasingly, the evaluation and triage of patients presenting with acute ischemic stroke relies on advanced imaging, including CTA and CTP. These studies require administration of iodinated contrast material. With the recent publication of five trials demonstrating the benefit of endovascular therapy for acute stroke,2–6 more of these patients can be expected to undergo endovascular evaluation and/or intervention, subjecting them to additional contrast. In this light, we aimed to evaluate the safety of multiple contrast administrations encountered during routine triage of patients presenting with acute stroke symptoms at our comprehensive stroke center.

Review of the current literature demonstrates preliminary support for the safety and efficacy of CT imaging studies in patients suspected of acute ischemic stroke.9 ,12–14 Smith et al7 reported on the safety and feasibility of a CT protocol for acute stroke, stating that no instances of renal injury were observed in their study of 53 consecutive patients undergoing CT protocol (NCCT, CTA, CTP) for suspected clinical stroke or transient ischemic attack. However, the authors do not clearly demonstrate the criteria they used to define renal injury. Josephson et al8 further studied this topic, retrospectively reviewing the data for 1075 patients receiving routine CTA and CTP at a single institution. The authors reported that 52 of 1075 patients (4.8%) included in the analysis had a rise in creatinine of >0.5 mg/dL. In four of these patients the cause of renal injury was considered to be contrast administration. These previous studies evaluated the risk of kidney injury associated with non-invasive stroke imaging (eg, CTA and CTP). As already mentioned, with a greater number of patients expected to be triaged to endovascular therapy, we aimed to investigate whether the addition of DSA performed in the acute period portends an increased risk of contrast-induced nephropathy.

In the present retrospective study of patients receiving CTSS and DSA within 48 h, only 4 of the 84 patients (4.7%) experienced AKI by the criteria described above (a rise of ≥0.5 mg/dL in serum creatinine). This compares favorably with the aforementioned studies reported in the literature.7 ,8 Given that previous studies assessed only AKI associated with CTA and/or CTP, we conclude from our analysis that the addition of DSA to the care of patients with acute stroke for either diagnostic or therapeutic purposes does not portend a greater risk of kidney injury. Additionally, there was no statistically significant difference in the interval of time between the CTSS and DSA in patients with and without AKI, suggesting that postponement for theoretical renal recovery is unnecessary.

We defined AKI as a rise of 0.5 mg/dL from the baseline creatinine level. In this way, we aimed to address new renal injury regardless of baseline function. Of the four patients who experienced AKI according to this criterion, only one had baseline renal dysfunction. We also studied the change in creatinine level as a means to control for any baseline renal dysfunction and found a significantly greater change in patients who developed AKI compared with those who did not (1.65 vs −0.09). We also monitored the patients’ renal function for an average of 7 days. In this way, we aimed to assess renal injury associated with contrast administration at the time of presentation. Delayed renal injury as a sequel of subsequent events (eg, hypotension, nephrotoxic medications) was not captured in the present study. Although patients presenting with acute ischemic stroke often have underlying medical comorbidities that may predispose them to the development of renal failure/renal injury, we did not find any statistically significant effect of either hypertension (p=1.0) or diabetes mellitus (p=0.127) for the development of AKI.

In addition to serum creatinine, BUN and GFR are additional laboratory values used to assess kidney function and potential injury. Because urea and creatinine are freely filtered by the glomerulus, significant changes in either of these laboratory values are readily seen in the GFR, more specifically in an inversely parabolic relationship.15 In other words, a fall in the rate of filtration (GFR) is normally associated with a rise in serum levels of BUN and creatinine. Because BUN is much more easily affected by dietary and physiologic conditions not related to renal function, creatinine levels are a much more reliable indicator of renal function15 and were the primary parameter measured in this study.

There are several limitations to this study. Given the retrospective nature of our analysis, inherent potentially confounding factors and biases exist. Our study necessarily excluded 13 patients due to unavailability of post-imaging laboratory studies (in addition to eight patients in whom the CTSS and DSA were not performed within our specified 48 h window, five patients because the volume of contrast medium used during their DSA was unavailable, and one patient found to have subarachnoid hemorrhage). Although serum creatinine levels are a good predictor of renal function, creatinine clearance is thought to be a more accurate indication of kidney function.15 These data were not available for analysis in our study. Instead, we used a conservative cut-off value of a rise in serum creatinine level of 0.5 mg/dL to define cases of AKI, and this change in serum creatinine is generally accepted as indicative of AKI.10 ,11 A complete demographic analysis was not obtained for this study, which may potentially confound the results as there are preliminary data suggesting that African-Americans have a higher risk of AKI after administration of iodinated contrast material.16 Although the long-term sequelae of contrast-induced AKI were not evaluated in this study, no patient who developed AKI required hemodialysis as a result of their renal injury.

This study provides preliminary evidence for the safety and feasibility of obtaining a CTSS (NCCT, CTA, CTP) with additional DSA imaging, whether for interventional or diagnostic purposes, to identify and treat possible acute ischemic stroke.


The authors thank Tingting Zhuang from the University at Buffalo Department of Biostatistics for statistical analysis and Debra J Zimmer from the University at Buffalo Department of Neurosurgery for editorial assistance.


View Abstract


  • Contributors Conception and design: SAM, SLH, MCC. Data acquisition: all authors. Data interpretation: all authors. Drafting the manuscript: SAM, SLH, MCC. Critically revising the manuscript: all authors. Final approval of the manuscript: all authors.

  • Competing interests EIL: Shareholder/ownership interests: Intratech Medical, Blockade Medical, NeXtGen Biologics. Principal investigator: Covidien US SWIFT PRIME trials. Honoraria for training and lecturing: Covidien. Consultant: Pulsar, Blockade Medical. Other financial support: Abbott for carotid training for physicians. Advisory boards: Stryker, NeXtGen Biologics, MEDX. AHS: Grants: National Institutes of Health/NINDS/NIBIB, University at Buffalo (none related to present study). Financial interests: Hotspur, Intratech Medical, StimSox, Valor Medical, Blockade Medical, and Lazarus Effect. Consultant: Codman & Shurtleff, Concentric Medical, ev3/Covidien Vascular Therapies, GuidePoint Global Consulting, Penumbra, Stryker, Pulsar Vascular, MicroVention, Lazarus Effect, Blockade Medical. Speakers’ bureau: Codman & Shurtleff. Speakers’ bureau: National Steering Committee–Penumbra 3D Separator Trial, Covidien SWIFT PRIME trial, MicroVention FRED trial. Advisory boards: Codman & Shurtleff, Covidien Neurovascular. Honoraria: Abbott Vascular, Codman & Shurtleff, Penumbra, Snyder, Boston Scientific. Research and stockholder: Cordis: research and financial interest; EndoTex: research and financial interest; Medtronic: research and consultant support; Abbott Vascular: research and consultant support; ev3: research and consultant support; Toshiba: research and consultant support; Micrus: research and consultant support and financial interest; Zimmer: research and consultant support; Access Closure: financial interest and stockholder; Niagara Gore Medical: stockholder; EPI: research and financial interest; Primus: financial interest; Guidant: research; Kerberos: research.

  • Ethics approval Ethics approval was received from the University at Buffalo Health Sciences Institutional Review Board (Project No. 735471-1).

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Data may be available for sharing on a per-request basis.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.