Annals of Emergency Medicine
Volume 54, Issue 1 , Pages e1-e5, July 2009

Update: Prevention of Contrast-Induced Nephropathy in the Emergency Department

  • Richard Sinert, DO

      Affiliations

    • Corresponding Author InformationAddress for correspondence: Richard Sinert, DO, Department of Emergency Medicine, Box 1228, SUNY-Downstate Medical Center, 450 Clarkson Avenue, Brooklyn, NY 11203; 718-245-2973, fax 718-245-4799
    • ,
  • Christopher I. Doty, MD

Department of Emergency Medicine, SUNY-Downstate Medical Center, Brooklyn, NY

published online 17 October 2008.

Article Outline

Study objective

Some studies have suggested that there exist therapies that can prevent contrast-induced nephropathy, which are practical in an emergency setting. This evidence-based emergency medicine (EBEM) critical appraisal reviews the literature, including additional studies appearing since the publication of an earlier EBEM review in 2007.

Methods

The updated search for randomized controlled trials from 2006 to 2008 complemented the previous search from 1966 to 2005. The methodological quality of the studies was assessed. Qualitative methods were used to summarize the study results.

Results

The search identified 2 studies not included in the previously published review of prophylactic therapies against contrast-induced nephropathy, yielding a total of 4 blinded, randomized, placebo-controlled trials involving bicarbonate and ascorbic acid. The present study of bicarbonate found a similar benefit as the previously reviewed bicarbonate trial. The early termination of both bicarbonate studies presents significant analytical concerns. The present study of ascorbic acid failed to reproduce the prophylactic effect of ascorbic acid observed in the previously reviewed trial. This difference in outcomes of the 2 ascorbic acid trials is related to variations in the volume of contrast and procedural hydration between the studies.

Conclusion

Although bicarbonate should still be considered a low-risk prophylactic agent, it appears that other factors (type of contrast agent, volume of contrast, and procedural hydration protocol) have an important influence on the risk of contrast-induced nephropathy. For the emergency physician, limiting exposure to contrast agents and adequate precontrast hydration are still the first line of defense against contrast-induced nephropathy.

 

In 2007, we published an evidence-based emergency medicine (EBEM) review of therapies to prevent contrast-induced nephropathy, which were practical in an emergency setting.1 We performed an updated review on this topic using the same search strategies and selection criteria described in our previous review.1 We searched MEDLINE, using the PubMed interface, between January 2006 and April 2008 and EMBASE from January 2006 to April 2008 with the Ovid Technologies interface and identified 2 eligible randomized controlled trials by Masuda et al2 and Boscheri et al3 that were not included in our original review. We also screened the bibliographies of the 2 articles and did not find any additional relevant studies. We also searched relevant databases within the Cochrane Library, particularly the Central Register of Controlled Trials, for new studies and protocols through April 2008 and found 0 articles.

The studies added in this update by Masuda et al2 investigated prophylactic administration of bicarbonate, whereas Boscheri et al3 tested ascorbic acid. For comparison, we included studies from our previous review by Merten et al4 and Spargias et al,5 who also investigated bicarbonate and vitamin C, respectively. According to Masuda et al,2 the impetus to restudy the Merten et al4 bicarbonate prophylaxis protocol was to narrow the inclusion criteria to patients undergoing emergency cardiac catheterizations. Boscheri et al3 tested ascorbic acid with better control over the volume of contrast media in the intervention and comparison groups.

From Table 1, we can see a comparison of the characteristics across the 4 studies. Important differences between the studies can be seen in the areas of inclusion criteria, contrast agents (type and volume), and procedural hydration. The Masuda et al2 study was the only one to narrow the inclusion criteria to only those patients undergoing emergency cardiac catheterizations, whereas the other studies included emergency and nonemergency cases. Both bicarbonate studies used a nonionic low-osmolality contrast medium, whereas Spargias et al5 allowed a mixture of contrast agents and Boscheri et al3 only used a nonionic iso-osmolar agent. The volume of contrast agents administered in the study by Spargias et al5 was almost twice that of the other studies (≈280 mL versus <140 mL). The Boscheri et al3 study used the largest volume (250 mL/hour Boscheri et al3 versus 125 mL/hour Spargias et al5 versus 3mL/kg Merten et al4 and Masuda et al2) of pre- and postprocedural hydration compared with the other studies.

Table 1. Summary of the 4 studies to prevent contrast-induced nephropathy, including 2 appraised in a previous review.
Study CharacteristicsMerten et al,4 2004 (n=119), BicarbonateMasuda et al,2 2007 (n=59), BicarbonateSpargias et al,5 2004 (n= 231), Ascorbic AcidBoscheri et al,3 2007 (n=143), Ascorbic Acid
PopulationAvg.⁎ age = 67.9 yearsAvg.⁎ age = 75.5 yearsAvg.⁎ age = 65.5 yearsAvg.⁎ age = 71.0 years
Avg.⁎ Cr = 1.80 mg/dLAvg.⁎ Cr = 1.32 mg/dLAvg.⁎ Cr = 1.45 mg/dLAvg.⁎ Cr = 1.74 mg/dL
Procedures: Emergency and nonemergency cardiac catheterizations, CTs and other procedures (brachytherapy, renal and peripheral arteriograms, aortograms, embolization)Procedures: Only emergency cardiac catheterizationsProcedures: Emergency and nonemergency cardiac catheterizationsProcedures: Emergency and nonemergency cardiac catheterizations
Exclusion criteria: Severe or recent ESRD, uncontrolled HTN, pregnancy, or recent use of dopamine, mannitol or fenoldopamExclusion criteria: Severe or recent ESRD, pregnancy, or recent exposure to radiographic contrast media mannitol, fenoldopam, NAC or nonstudy sodium bicarbonateExclusion criteria: Severe or recent ESRD.Exclusion criteria: Severe or recent ESRD, pregnancy, or recent exposure to NAC or radiographic contrast media, MI <3 months, cardiogenic shock, vasopressors, ejection fraction <25%
Contrast agent: 100% nonionic low osmolalityContrast agent: 100% nonionic, low osmolalityContrast agent: 59% Nonionic iso-osmolar 37% Low osmolar/nonionic 4% Low osmolar/ionicContrast agent: 100% nonionic iso-osmolar
IV fluid volume:IV fluid volume:IV fluid volume:IV fluid volume:
Intervention: 135±76 mLIntervention: 112±89 mLIntervention: 287±148 mLIntervention: 99±46 mL
Comparison: 133±62 mLComparison: 120±61 mLComparison: 261±128 mLComparison: 112±67 mL
InterventionNaHCO3 154 mEq/L NaHCO3in D5W 1000 mL at 3 mL/kg per hour for 1 hour precontrast and 1 mL/kg per hour during the contrast exposure and 6 hours postcontrastNaHCO3 154 mEq/L NaHCO3in D5W 1000 mL at 3 mL/kg per hour for 1 hour precontrast and 1 mL/kg per hour during the contrast exposure and 6 hours postcontrastAscorbic acid 3 g oral, 2 hours precontrast, and 2 g the night of and the morning after the procedure. NS at 50–125 mL/hour starting at least 2 hours precontrast and continued for 6 hours.Ascorbic acid 1g 20 minutes before dye and NS 500 mL 2 hours before, and NS 500 mL during/after over 6 hours
ComparisonHydration: 154 mEq/L of NaCl in 5% D5W 3 mL/kg per hour for 1 hour precontrast and 1 mL/kg per hour during the contrast exposure and 6 hours postcontrastHydration—NS 3 mL/kg per hour for 1 hour precontrast and 1 mL/kg per hour during the contrast exposure and 6 hours postcontrastHydration: NS at 50–125 mL/hour starting at least 2 hours precontrast and continued for 6 hours.Hydration—NS 500 mL 2 hours before, and NS 500 mL during/after over 6 hours
Outcome measuresCIN: increase in serum creatinine level of 0.5 mg/dL or an increase in serum creatinine level by 25% at 48 hours postcontrastCIN: increase in serum creatinine level of 0.5 mg/dL or an increase in serum level creatinine by 25% at 48 hours postcontrastCIN: increase in serum creatinine level of 0.5 mg/dL or an increase in serum creatinine level by 25% at 48 hours postcontrastCIN: increase in serum creatinine level by 25% at 48 hours postcontrast

CT, Computed tomography; ESRD, end stage renal disease; HTN, hypertension; NAC, N-acetylcysteine; MI, myocardial infarction; NaHCO3, sodium bicarbonate; NS, normal saline solution; CIN, contrast-induced nephropathy; NaCl, sodium chloride.

⁎Weighted average.

Table 2 compares the potential sources of bias between the studies under review. Concealment and blinding were not evident in the study by Masuda et al.2 Intention-to-treat analysis was also not performed in the studies by Masuda et al2 and Boscheri et al.3 Merten et al4 and Spargias et al5 lost 3% and 7% of their patients to follow-up, respectively. Merten et al4 accounted for their patients lost to follow-up by including their baseline data into the analysis as the last observation carried forward. They found that reanalyzing their data this way was not informative and did not affect their results. Spargias et al5 conducted a sensitivity analysis of patients lost to follow-up by assuming their changes in postprocedural renal function reflected the change observed in the overall study population No significant differences in their results were noted in this sensitivity analysis. What is not covered as a potential source of bias in Table 2 is the effect of early termination of both bicarbonate studies. Merten et al4 estimated a total sample size of 260 patients required to adequately power (80%; α=0.05) the study to detect a change in the incidence of contrast-induced nephropathy from 15% to 5%. Masuda et al,4 using similar calculations (14% to 2%), estimated a sufficient sample size of 160 patients. Both the Merten et al4 (n=137; 18 excluded after randomization) and the Masuda et al2 (n=61; 2 excluded after randomization) trials were terminated early because an interim analysis determined that the evidence in favor of the intervention made it unethical to continue to recruit patients into the comparison arms. This may have led to an overestimation of the treatment effect. This bias arises because random fluctuations toward greater and lesser treatment effects occur naturally in randomized clinical trials, if the decision to interrupt the trial and analyze the data occurred during a “random high,” the resulting estimate of the treatment effect maybe misleading.6, 7

Table 2. Susceptibility to bias of the 4 studies to prevent contrast-induced nephropathy, including 2 appraised in a previous review.
Potential BiasMerten et al,4 2004, BicarbonateMasuda et al,2 2007, BicarbonateSpargias et al,5 2004, Ascorbic AcidBoscheri et al3 2007, Ascorbic Acid
RandomizationSingle-center randomization assigned by the pharmacy according to a computer-generated scheduleComputer-generated random numbers determined randomization
Single center

Randomization was in blocks of 10

Single-center randomization technique not mentioned
ConcealmentAdequateAdequateAdequateAdequate
BlindingTreatment assignment blinded to patients, providers and assessorsNo blindingDouble blinded, but the identities of those who were blinded not specifiedDouble blinded to patients and investigators
Intention to treatYesTwo patients (1 for each group) were excluded after randomizationYesYes
Baseline comparisonsWell balanced for risk factors for CINWell balanced for risk factors for CINWell balanced for risk factors for CINWell balanced for risk factors for CIN
CointerventionsSame hydration protocols in both intervention and comparison groupsSame hydration protocols in both intervention and comparison groupsSimilar hydration protocols in both intervention and comparison groupsSimilar hydration protocols in both intervention and comparison groups
Completeness of follow-up7% patients lost to follow-up were excluded from analysis. No change in the primary result when patients lost to follow-up were added back into the analysisAll patients accounted for except for 1 patient in each group who was excluded after being randomized3% patients lost to follow-up were excluded from analysis. No change in the primary result when patients lost to follow-up were added back into the analysisAll patients accounted for

Table 3 compares the results of the studies. There is significant heterogeneity of the event rates in the comparison groups across the studies. The Masuda et al2 comparison group had a significantly higher incidence of contrast-induced nephropathy (34.5%) than did the comparable groups within the other studies (20.4%,5 13.6%,4 and 4.3%,3 respectively). We believe the likely explanation for the higher incidence of contrast-induced nephropathy in the Masuda et al2 study is multifactorial. The Masuda et al2 patients were most likely sicker because only emergency cardiac catheterization cases were included, whereas the other studies included both emergency and nonemergency patients. Masuda et al2 used a nonionic low-osmolar contrast agent with a higher risk of kidney injury compared to the iso-osmolar medium used in the ascorbic acid studies.8 Finally, the bicarbonate compared to the vitamin C studies had significantly smaller-volume pre- and postprocedural hydration protocols. Furthermore, only two thirds of the patients in the Masuda et al2 study received the full amount of fluid before the procedure because some patients were sent to catheterization before the full hour.

Table 3. Results of the 4 studies to prevent contrast-induced nephropathy, including 2 appraised in a previous review.
OutcomesMerten et al,4 2004, BicarbonateMasuda et al,2 2007, BicarbonateSpargias et al,5 2004, Ascorbic AcidBoscheri et al,3 2007, Ascorbic Acid
Risk for CINTreatment1/601.7%2/306.7%11/1189.3%5/746.8%
Comparison8/5913.6%10/2934.5%23/11320.4%3/694.4%
Relative risk of CIN (95% CI)0.12(0.02–0.72)0.19(0.05–0.81)0.46(0.24–0.88)1.55(0.3–6.3)
Number needed to treat (95% CI)8.4(6.8–40)4(2.1–12.0)9.1(5.4–52.7)NA

NA, Not applicable (because the effect was not statistically significant; RR includes 1.0.)

In contrast, the comparison group in the study by Boscheri et al3 had the lowest incidence of contrast-induced nephropathy (4.3%) observed. These patients received the smallest volume of only a nonionic iso-osmolar contrast agent and were given the largest volume of pre- and postprocedural hydration.

Table 3 also compares the effectiveness of the interventions in preventing contrast-induced nephropathy. The 2 studies using bicarbonate compared to the trial of ascorbic acid by Spargias et al5 had the greatest reduction in relative risk (88% and 81% versus 54%) of kidney injury. Not surprisingly, the study by Boscheri et al,3 with the lowest incidence of kidney injury in their comparison group, failed to show a benefit to their intervention (ascorbic acid). The large confidence intervals observed by Boscheri et al3 reflect the study's relatively small sample and effect sizes.

In conclusion, from this updated review the most effective intervention appears to be bicarbonate, although both these studies have significant limitations related to bias from the low volumes of pre- and postprocedural hydration and early study termination. What is clear from these trials is that factors other than their primary interventions (bicarbonate or ascorbic acid) had an important effect on the risk of contrast-induced nephropathy. Independent risk factors for contrast-induced nephropathy include the type of contrast agent (low versus iso-osmolar), volume of contrast (>140 mL),9 and procedural hydration protocols.10, 11 For the emergency physician faced with the decision to expose a patient with renal insufficiency to intravenous contrast media, bicarbonate may represent a relatively low-risk prophylactic agent. However, withholding or limiting the volume of intravenous contrast, selecting a less nephrotoxic contrast agent, and appropriate levels of hydration should also be considered.12, 13

Back to Article Outline

Critically Appraised Topic (CAT): What interventions are available to prevent the renal failure and death associated with contrast-induced nephropathy? 2008 Update 

QuestionIn patients with baseline renal insufficiency, which interventions are effective in preventing contrast-induced nephropathy and its complications, given the resource and time constraints in a busy emergency department (ED)?
Reviewed bySinert RH, Doty C
Date of searchMay 2008
Expiration dateApril 2010
Clinical bottom lineEvidence from randomized trials shows that the intervention (bicarbonate) under review is appropriate to an ED setting and decreased the risk of contrast-induced nephropathy. The case for the effectiveness of ascorbic acid was less certain. Independent risk factors for contrast-induced nephropathy include the type of contrast agent (low versus iso-osmolar), volume of contrast (>140 mL), and procedural hydration protocols
Search strategyMEDLINE, EMBASE, the Cochrane Library from the dates of origin through April 2008.
CitationsPrimary: 2 fully reported additional randomized trials
(Updated review) Masuda M, Yamada T, Mine T, et al. Comparison of usefulness of sodium bicarbonate versus sodium chloride to prevent contrast-induced nephropathy in patients undergoing an emergency coronary procedure. Am J Cardiol. 2007;100:781-786.
(Updated review) Boscheri A, Weinbrenner C, Botzek B, et al. Failure of ascorbic acid to prevent contrast-media induced nephropathy in patients with renal dysfunction. Clin Nephrol. 2007;68:279-286.
(Previous review) Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA. 2004;291:2328-2334.
(Previous review) Spargias K, Alexopoulos E, Kyrzopoulos S, et al. Ascorbic acid prevents contrast-mediated nephropathy in patients with renal dysfunction undergoing coronary angiography or intervention. Circulation. 2004;110:2837-2842.
Primary study characteristicsStudy Population
Patients with baseline renal insufficiency over age 18 years receiving radiocontrast agents.
Interventions
Bicarbonate: 1 New study
Ascorbic acid: 1 New study
Outcome Measures
Contrast-induced nephropathy was defined by either a proportional increase of 25% in serum creatinine level or an absolute increase in serum creatinine level by 0.5 mg/dL from baseline 48 hours after contrast exposure. In addition, we reviewed each trial for changes in the rates of renal failure requiring inhospital dialysis and death between intervention and control groups.
Critical appraisalFair to good quality was observed in all trials.
Results
Intervention# StudiesRelative Risk of Contrast-Induced Nephropathy Intervention vs Comparison (95% CI)
Bicarbonate (updated review)10.19(0.05–0.81)
Bicarbonate (previous review)10.12(0.02–0.72)
Ascorbic acid (updated review)11.55(0.3–6.3)
Ascorbic acid (previous review)10.46(0.24–0.88)

No study was sufficiently powered to detect significant improvements in mortality rate or the rate of renal failure requiring dialysis with their interventions.

Back to Article Outline

References 

  1. Sinert R, Doty CI. Prevention of contrast-induced nephropathy in the emergency department. Ann Emerg Med. 2007;50:335–345
  2. Masuda M, Yamada T, Mine T, et al. Comparison of usefulness of sodium bicarbonate versus sodium chloride to prevent contrast-induced nephropathy in patients undergoing an emergent coronary procedure. Am J Cardiol. 2007;100:781–786
  3. Boscheri A, Weinbrenner C, Botzek B, et al. Failure of ascorbic acid to prevent contrast-media induced nephropathy in patients with renal dysfunction. Clin Nephrol. 2007;68:279–286
  4. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-induced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA. 2004;291:2328–2334
  5. Spargias K, Alexopoulos E, Kyrzopoulos S, et al. Ascorbic acid prevents contrast-mediated nephropathy in patients with renal dysfunction undergoing coronary angiography or intervention. Circulation. 2004;110:2837–2842
  6. Bassler D, Ferreira-Gonzalez I, Briel M, et al. Systematic reviewers neglect bias that results from trials stopped early for benefit. J Clin Epidemiol. 2007;60:869–873
  7. Heffner AC, Milbrandt EB, Venkataraman R. Trials stopped early for benefit? (Not so fast!). Crit Care. 2007;11:305
  8. McCullough PA, Bertrand ME, Brinker JA, et al. A meta-analysis of the renal safety of isosmolar iodixanol compared with low-osmolar contrast media. J Am Coll Cardiol. 2006;48:692–699
  9. Freeman RV, O'Donnell M, Share D, et al. Nephropathy requiring dialysis after percutaneous coronary intervention and the critical role of an adjusted contrast dose. Am J Cardiol. 2002;90:1068–1073
  10. Mueller C, Buerkle G, Buettner HJ, et al. Prevention of contrast media-associated nephropathy: randomized comparison of 2 hydration regimens in 1620 patients undergoing coronary angioplasty. Arch Intern Med. 2002;162:329–336
  11. Trivedi HS, Moore H, Nasr S, et al. A randomized prospective trial to assess the role of saline hydration on the development of contrast nephrotoxicity. Nephron Clin Pract. 2003;93:C29–C34
  12. Stacul F, Adam A, Becker CR, et al. Strategies to reduce the risk of contrast-induced nephropathy. Am J Cardiol. 2006;98(Suppl):59K–77K
  13. Solomon R, Deray G. How to prevent contrast-induced nephropathy and manage risk patients: practical recommendations. Kidney Int Suppl. 2006;(100):S51–S53

 Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article that might create any potential conflict of interest. The authors have stated that no such relationships exist. See the Manuscript Submission Agreement in this issue for examples of specific conflicts covered by this statement.

 Publication date: Available online October 16, 2008.

 Reprints not available from the authors.

PII: S0196-0644(08)01638-7

doi:10.1016/j.annemergmed.2008.08.014

Annals of Emergency Medicine
Volume 54, Issue 1 , Pages e1-e5, July 2009

Article Tools