SEARCH

SEARCH BY CITATION

Keywords:

  • computed tomography;
  • contrast media;
  • creatinine;
  • pulmonary embolism;
  • renal failure

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

Summary. Objective: To estimate the frequency of contrast nephropathy after computed tomography angiography (CTA) to rule out pulmonary embolism (PE) in the emergency department (ED) setting. Methods: We prospectively followed patients undergoing CTA for PE, while in the ED, for 45 days. Patients who refused follow-up or were receiving hemodialysis were excluded. Severe renal failure was defined as an increase in creatinine ≥ 3.0 mg dL−1 or a need for hemodialysis within the follow-up period. Patients were also followed for laboratory-defined contrast nephropathy, defined as an increase in creatinine of > 0.5 mg dL−1 or > 25%, within seven days following CTA. Results: A total of 1224 patients were followed, and 354 [29%, 95% confidence interval (CI): 26–32%] patients had paired (preCTA and post-CTA) creatinine measurements. None developed renal failure (0/1224; 0%, CI: 0–0.3%). 44 patients developed laboratory-defined contrast nephropathy, corresponding to an overall frequency of 4% (44/1224; CI: 3–5%) and 12% (44/354; 95% CI: 9–16%) among those with paired creatinine measurements. Conclusions: Following CTA for PE, the incidence of severe renal failure was very low, but the incidence of laboratory-defined contrast nephropathy (4% overall and 12% of those with paired measurements) was higher than expected.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

Between 1991 and 2002, the use of iodinated contrast-enhanced computed tomography angiography (CTA) imaging in emergency departments (EDs) and other outpatient settings increased by 234% [1] with approximately one million outpatients receiving a CTA in 2001 [2]. The rate of increase in outpatient CTA scanning has increased approximately 20–30 times faster than other computed tomography modalities [1]. In ED patients with suspected pulmonary embolism (PE), CTA has become the diagnostic modality of choice, with over 1% of all ED patients (approximately one million patients per year) receiving a CTA for this indication [1–4]. Despite this increase in the use of iodinated contrast media, the risk of contrast nephropathy remains undefined in the ED setting, even though it is recognized as one of the most important causes of renal failure in hospitalized patients [5–8]. Moreover, except for a study of patients who had CTA of the brain [9], no published study has reported on the incidence of contrast nephropathy for any type of CTA, and no study has reported on the incidence of contrast nephropathy after administration of low osmolar or non-ionic contrast material for CTA.

Prior work carried out primarily for patients undergoing cardiac catheterization has demonstrated laboratory-defined contrast nephropathy in up to 17% of patients, but there is considerable variability (up to tenfold) in the reported frequency in these studies [10,11]. Several studies have demonstrated benefits to specific prophylactic therapies designed to reduce contrast toxicity [12]. It follows that defining the frequency of contrast nephropathy from CTA in ED patients would be necessary for the implementation of prophylactic regimens in this setting.

The purpose of this study is to report the frequency associated with contrast nephropathy in a heterogeneous outpatient population undergoing CTA for PE using non-ionic, low osmolar iodinated contrast. In addition to the importance of CTA in the diagnostic evaluation of PE, this imaging modality also provides some means of estimating the risk of contrast nephropathy from other CTA imaging studies for the following reasons. (i) This is a relatively common indication for CTA in the ED [4]. (ii) The population receiving CTA of the chest for this indication is relatively diverse in terms of age, gender, comorbidities and medical acuity. (iii) The dose of contrast used for this study is within the middle range of that used for other contrast-enhanced CT modalities.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

Study design

This was a prospective, non-interventional observational, single center study designed to document the incidence.

Study setting

Patients enrolled in this study were evaluated in the ED at the Carolinas Medical Center, Charlotte, NC. This center is staffed by board-certified emergency physicians 24 h a day, has a residency in emergency medicine and treats over 110 000 patients per year. A board-certified, attending radiologist was present at the study hospital 24/7 during the study period. Approximately two patients underwent CTA to evaluate for PE each day in this ED during the study period [4].

Selection of participants

We studied adult (age > 17) ED patients selected for CTA of the chest to evaluate for suspected PE between April 2003 and October 2005. Patients were first selected for evaluation for PE based upon a wide range of symptoms or signs including shortness of breath, chest pain, syncope, respiratory distress, unexplained hypoxemia, tachycardia, etc. Patients were then selected for CTA based upon a clinical protocol that incorporates pretest probability assessment and a D-dimer assay [4]. Patients with a high pretest probability or positive D-dimer underwent CTA to evaluate for PE. The only absolute contraindication to CTA at our institution during the study period was a known, severe allergy to iodinated contrast material. We had no institutional protocol that precluded CTA based upon the baseline creatinine measurement and no protocol to institute prophylactic therapies.

Patients were enrolled by ED physicians 24 h per day with assistance from research personnel approximately 16 h per day, six days per week. Patients who were unable or unwilling to provide written informed consent for follow-up, or who were already receiving chronic hemodialysis, were excluded. Clinical data including age, gender, race and comorbidities were collected in real time at the bedside using electronic data forms completed by the evaluating ED physician. The method of operation and content of these forms has been published [13]. Study enrollment as a percentage of all ED patients who underwent CTA of the chest to rule out PE was determined by audits of administrative logs of all imaging procedures, as previously described [4].

Creatinine measurements

Blood for baseline creatinine measurements was obtained in the ED by standard venipuncture, usually within an hour of patient arrival. If the patient had had a baseline creatinine measurement within the previous 30 days, we allowed this as a baseline measurement. The peak serum creatinine measured between two and seven days after CTA (a ‘paired’ value) was used to define the subsequent change in creatinine for outcomes. Creatinine measurements were performed on serum (BD Vacutainer® SSTTM; BD Biosciences, Franklin Lakes, NJ, USA) using a Food and Drug Administration approved laboratory device (Beckman CX6; Ramsey, MN, USA). The decision to order a baseline or follow-up creatinine was at the discretion of the patient's physician.

CTA protocol

Patients underwent CTA of the chest using image acquisition parameters that we have previously reported [14]. Images were obtained following the administration of 120 mL of Iopamidol (Isovue-200®; Bracco Diagnostics, Princeton, NJ, USA) i.v. contrast via power injector through an antecubital peripheral venous catheter over approximately 30 s.

Clinical follow-up

We followed patients using a combination of telephone survey and review of paper and electronic medical records, as we have previously described [13]. We prospectively asked patients to provide primary and alternative contact telephone numbers. At 45 days following enrollment, five separate attempts at telephone contact were made by blinded researchers. We also performed medical record review for all patients. Creatinine measurements and dates of measure were obtained from the paper and electronic medical records.

Outcome measures

The primary outcomes were the development of one or both of the following after CT angiography to evaluate for PE: (i) development of severe acute renal failure within 45 days, defined as the need for admission or re-presentation where follow-up creatinine measurements revealed an absolute increase in creatinine of ≥ 3.0 mg dL−1 or the patient required hemodialysis [8]; (ii) development of laboratory-defined contrast nephropathy, defined as an increase in creatinine > 25% of baseline or an absolute increase of > 0.5 mg dL−1 within seven days [10]. We also report on the frequency of an increase in creatinine from < 1.2 to ≥ 1.2 mg dL−1

Data analysis

Data were downloaded from the SQL server and transferred into spreadsheet format (Microsoft®excel; Seattle, WA, USA) to prepare it for statistical analysis using statsdirect v3.3 software (Cheshire, UK). Frequencies are reported as proportions with 95% confidence interval (CI) from the Clopper–Pearson method.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

A total of 1224 patients were followed, representing more than 90% of all patients who underwent CT angiography to rule out PE in our ED during the study period. None of the 1224 patients developed the primary outcome of severe acute renal failure or the need for hemodialysis during the 45-day follow-up period (0%, CI: 0–0.3%). Baseline creatinine measurements were available for 766 patients (63%). The baseline creatinine measurement was obtained at the index visit in 702 (91%) of these patients. A second creatinine measurement (a ‘paired’ value) was also available for a total of 354/1224 patients (29%). Data comparing the clinical characteristics of those with and without paired creatinine measurements are reported in Table 1. Male patients, patients with diabetes mellitus, hypertension, coronary artery disease or congestive heart failure, patients with an elevated baseline creatinine and patients with venous thromboembolism observed with CTA were more likely to have paired creatinine measurements. 25% (CI: 21–30%) of those with paired measurements had a baseline creatinine ≥ 1.2 mg dL−1.

Table 1.   Comparison of patients with and without paired creatinine measurements
CharacteristicWith paired measurements n = 354Without paired measurements n = 870Difference
  1. *Standard deviation (SD) and 95% confidence interval (CI) shown in parentheses; percentage of the 412 patients with a baseline creatinine, but without a paired measurement.

Mean age, years (SD)*53 (18)47 (17)
Male gender, % (CI)42 (36–47)28 (25–31)13 (7–20)
African American race, % (CI)48 (42–53)51 (48–55)−3 (−9–3)
Caucasian race, % (CI)50 (44–55)45 (41–48)5 (−1–11)
Baseline creatinine ≥ 1.2 mg dL−1, % (CI)25 (21–30)19 (15–23)6 (1–12)
Diabetes mellitus, % (CI)22 (18–26)12 (10–15)10 (5–15)
Hypertension, % (CI)48 (42–53)36 (33–39)12 (6–19)
Coronary artery disease, % (CI)14 (10–18)8 (6–10)6 (2–10)
Congestive heart failure, % (CI)9 (6–13)6 (4–8)3 (0–7)
Venous thromboembolism positive, % (CI)46 (41–51)4 (0–3)42 (37–48)

Forty-four patients who had a paired measurement developed laboratory-defined contrast nephropathy within 45 days. This corresponds to an overall 45-day incidence of at least 4% (44/1224; CI: 3–5%) of the total population and 12% (44/354; 95% CI: 9–16%) of those with paired creatinine measurements. Ninety patients who had an elevated baseline creatinine ≥ 1.2 mg dL−1 also had a subsequent creatinine measurement within seven days. Five of these patients (6%; CI: 2–12%) developed contrast nephropathy compared with 15% (39/264; CI: 10–20%) of those who had a normal baseline creatinine. Eight of the 90 patients (9%; CI: 4–17%) with an elevated baseline creatinine had a decrease in creatinine by 0.5 mg dL−1 or a decrease of more than −25% of baseline following CTA. An increase in creatinine from a baseline of < 1.2 mg dL−1 to a post-CTA maximum of ≥ 1.2 mg dL−1 was observed in 32/264 (12%; CI: 8–17%) patients. Fig. 1 shows the plot of all creatinine concentrations in the 354 patients with paired measurements. Overall, 40% (CI: 35–45%) had an increase in serum creatinine, 36% (CI: 31–41%) had a decrease and 24% (CI: 20–30%) had no change. We also observed that patients with a history of coronary artery disease developed the outcome of laboratory-defined nephropathy at a relatively high frequency (24%; CI: 13–39%).

image

Figure 1.  Serum creatinine concentrations before and after contrast-enhanced computed tomography angiography of the chest. Dotted line indicates the threshold for an elevated creatinine (≥ 1.2 mg dL−1).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

In this paper, we report on the incidence of contrast nephropathy after administration of non-ionic contrast for CTA to evaluate for PE in an outpatient population. This study has several novel but plausible findings. First, the observed incidence of laboratory-defined contrast nephropathy (4%; CI: 3–5%) was higher than expected. Secondly, laboratory-defined contrast nephropathy occurred at a relatively high frequency among those with coronary artery disease (24%; CI: 13–39%). Finally, this study also demonstrates the unexpected finding that laboratory-defined contrast nephropathy occurred at a lower than expected frequency among those with an elevated baseline creatinine concentration (6%; CI: 2–12%).

While no patient developed severe acute renal failure in this study, the incidence of laboratory-defined contrast nephropathy was higher than expected for this population of mixed comorbidities and risk factors. A study of CTA of the brain, published in 1984, reported a 2% rate of contrast nephropathy [9]. However, this study used high osmolar, ionic contrast, which many experts believe to be more nephrotoxic than current agents [12]. The PIOPED II study reports that one patient out of 1095 receiving CTA developed contrast nephropathy indicated by an elevated creatinine following CTA [15]. However, this study provides no other data comparing baseline and post-CTA creatinine measurements and does not discuss the type of contrast used. Because no other comparable studies of CTA populations have been published, we considered the results of coronary angiography studies to permit some context for our results. The reported rate of contrast nephropathy following coronary angiography is highly variable, but most studies found an incidence of between 2% and 10% [10]. We expect that populations undergoing coronary angiography may have higher rates of other medical conditions, which may predispose to contrast nephropathy compared with our population, even with some degree of oversampling for these conditions (Table 1) [10,11]. Accordingly, our finding of a similar incidence of contrast nephropathy in our outpatient population was unexpected. Patients undergoing coronary angiography also receive a lower overall dose of contrast, given in small subsequent doses. This potentially important difference in administration has never been studied as an independent predictor variable.

The biological and clinical significance of a transient, asymptomatic > 25% increase or absolute increase of > 0.5 mg dL−1 in serum creatinine concentration after CTA remains uncertain. None of the 1224 patients developed severe acute renal failure or required hemodialysis, and most of the 44 (4%) patients who developed laboratory-defined contrast nephropathy had two serum creatinine measurements that were in the normal range. Analysis of prior literature reveals that, at most, 4% of patients with laboratory-defined contrast nephropathy required temporary dialysis [10]. These data may suggest that the renal injury from CTA is minimal. However, mechanistic studies show that iodinated contrast molecules are clearly toxic to renal tubular epithelial cells via several mechanisms, including osmotic stress and cell membrane peroxidation from reactive oxygen species [16]. Moreover, it is well recognized that patients can suffer a 50% decrease in glomerular filtration rate without a significant rise in serum creatinine. Furthermore, patients with positive outcomes may have remained asymptomatic and unrecognized for the duration of this study. Thus, it could be argued that our observation of a ≥ 25% increase in serum creatinine in at least 4% of all patients who underwent CTA represents the ‘tip of the iceberg’. From a more conservative point of view, our data suggest that at least 4% of patients undergoing CTA experience substantial renal tubular damage, rendering them more vulnerable to additional chronic insults (e.g. systemic hypertension and diabetes mellitus), thereby accelerating the development of permanent renal failure and attendant complications [10].

The study includes the following limitations. First, this study was an observational study and did not control for patient who received preCTA and post-CTA serum creatinine measurements. However, no patient developed severe acute renal failure or required hemodialysis following CTA. Previous studies also demonstrate that emergency physicians can accurately predict patients likely to have an elevated baseline creatinine or impaired renal function [17,18]. We did not control for any treatments ordered by the evaluating physician and it may be that physicians have employed effective prophylactic measures such as prestudy i.v. saline or other treatments. Secondly, this study did not record or control for all published potential risk factors for contrast nephropathy. A single published study suggests that women are more likely to develop contrast nephropathy [19]. Women were less likely to have paired creatinine measurements in this study. Finally, we do not report a mortality rate as an outcome.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

The overall risk of severe acute renal failure from i.v. contrast for CTA to rule out PE was very low. The risk of contrast nephropathy was higher than expected: at least 4% overall and 12% of those with paired creatinine measurements. This study calls into question the perceived safety of CTA, especially among those with coronary artery disease, and emphasizes the inadequacy of relying exclusively on baseline creatinine measurement to gage the risk of contrast nephropathy.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References

Work supported by grants R01 HL074384 and R41/42 HL074415 from NIH/NHLBI to J. A. Kline.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Disclosure of Conflict of Interests
  9. References
  • 1
    Kalra MK, Maher MM, D'Souza R, Saini S. Multidetector computed tomography technology: current status and emerging developments. J Comput Assist Tomogr 2004; 28: S2S6.
  • 2
    Bhargavan M, Sunshine JH. Utilization of radiology services in the United States: levels and trends in modalities, regions, and populations. Radiology 2005; 234: 82432.
  • 3
    Kline JA, Jones AE. Availability of technology to evaluate for pulmonary embolism in academic emergency departments in the United States. J Thromb Haemost 2003; 11: 22402.
  • 4
    Kline JA, Webb WB, Jones AE, Hernandez J. Impact of a rapid rule-out protocol for pulmonary embolism on the rate of screening, missed cases, and pulmonary vascular imaging in an urban U.S. emergency department. Ann Emerg Med 2004; 44: 490503.
  • 5
    Levy EM, Viscoli CM, Horwitz RI. The effect of acute renal failure on mortality. A cohort analysis. JAMA 1996; 19: 148194.
  • 6
    Briguori C, Tavano D, Colombo A. Contrast agent-associated nephrotoxicity. Prog Cardiovasc Dis 2003; 45: 493503.
  • 7
    McCullough PA, Wolyn R, Rocher LL, Levin RN, O'Neill WW. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 1997; 103: 36875.
  • 8
    Hou SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT. Hospital-acquired renal insufficiency: a prospective study. Am J Med 1983; 74: 2438.
  • 9
    Cramer BC, Parfrey PS, Hutchinson TA, Baran D, Melanson DM, Ethier RE, Seely JF. Renal function following infusion of radiologic contrast material. A prospective controlled study. Arch Intern Med 1985; 145: 879.
  • 10
    Gami AS, Garovic VD. Contrast nephropathy after coronary angiography. Mayo Clin Proc 2004; 79: 2119.
  • 11
    Quader MA, Sawmiller C, Sumpio BA. Contrast-induced nephropathy: review of incidence and pathophysiology. Ann Vasc Surg 1998; 12: 61220.
  • 12
    Asif A, Garces G, Preston RA, Roth D. Current trials of interventions to prevent radiocontrast-induced nephropathy. Am J Ther 2005; 12: 12732.
  • 13
    Kline JA, Mitchell AM, Runyon MS, Jones AE, Webb WB. Electronic medical record review as a surrogate to telephone follow-up to establish outcome for diagnostic research studies in the emergency department. Acad Emerg Med 2005; 12: 112733.
  • 14
    Richman PB, Wood J, Kasper DM, Collins JM, Petri RW, Field AG, Cowles DN, Kline JA. Contribution of indirect computed tomography venography to computed tomography angiography of the chest for the diagnosis of thromboembolic disease in two United States emergency departments. J Thromb Haemost 2003; 1: 6527.
  • 15
    Stein PD, Fowler SE, Goodman LR, Gottschalk A, Hales CA, Hull RD, Leeper KV, Popovich J, Quinn DA, Sos TA, Sostman HD, Tapson VF, Wakefield TW, Weg JG, Woodard PK. Multidetector computed tomography for acute pulmonary embolism. JAMA 2006; 354: 231727.
  • 16
    Katholi RE, Woods WT, Taylor GJ, Deitrick CL, Womack K.A, Katholi CR, McCann WP. Oxygen free radicals and contrast nephropathy. Am J Kidney Dis 1998; 32: 6471.
  • 17
    Olsen JC, Salomon B. Utility of the creatinine prior to intravenous contrast studies in the emergency department. J Emerg Med 1996; 14: 5436.
  • 18
    Tippins RB, Torres WE, Baumgartner BR, Baumgarten DA. Are screening serum creatinine levels necessary prior to outpatient CT examinations? Radiology 2000; 216: 4814.
  • 19
    Iakovou I, Dangas G, Mehran R, Lansky AJ, Ashby DT, Fahy M, Mintz GS, Kent KM, Pichard AD, Satler LF, Stone GW, Leon MB. Impact of gender on the incidence and outcome of contrast-induced nephropathy after percutaneous coronary intervention. J Invasive Cardiol 2003; 15: 1822.