1. Sodium Bicarbonate for the Prevention of Contrast Induced Nephropathy: A Meta-analysis of Published Clinical Trials Vijayalakshmi Kunadian 1,2, Azfar Zaman 1, Weiliang Qiu 2 1 Freeman Hospital, Newcastle upon Tyne, United Kingdom; 2 The James Cook University Hospital, Middlesbrough, United Kingdom; 3 Channing Laboratory, Brigham and Women’s Hospital, Boston MA SCAI Annual Scientific Sessions 2009 Las Vegas The authors have nothing to disclose

2. Background Contrast induced nephropathy (CIN) is a serious but rare complication following contrast based procedures Sodium bicarbonate has been demonstrated to prevent CIN through several mechanisms

3. Objective To perform a meta-analysis of randomized clinical trials to determine if the administration of NaHCO 3 is superior to NaCl among patients with chronic renal failure undergoing catheterization and interventional procedures in preventing CIN

4. Study Selection A computerized literature search was performed on PubMed using the search terms “contrast nephropathy”, “sodium bicarbonate (NaHCO 3 )”, “sodium chloride” (NaCl) and “renal failure”. Seven published randomized clinical trials comparing NaHCO 3 vs. placebo or NaCl during diagnostic and interventional procedures requiring contrast media administration were included.

5. StudyCKDProcedureSampleTrial design Brar YesCA353NaCl vs. NaHCO 3 Maioli YesCA, PCI502NaCl vs. NaHCO 3 Brigouri YesCA, PA, PCI326NaCl+NAC vs. NaCl+NaHCO 3 vs. ascorbic acid+NAC Masuda YesEm. CA and PCI59NaCl vs. NaHCO 3 Merten YesDiagnostic and interventional 119NaCl vs. NaHCO 3 Ozcan YesCA, PCI264NaCl vs. NaHCO 3 vs. NaCl+NAC Recio- Mayoral NoUrgent or emergency CA, PCI 111NaCl+NAC vs. NaCl+NaHCO 3 RCTs Included in the Meta-analysis CKD-chronic kidney disease, CA-coronary angiography, PCI-Percutaneous coronary intervention, PA- pulmonary angiography, NAC-N-acetylcysteine

6. Outcome Measures The development of CIN following administration of NaHCO 3 and NaCl are provided. For the purpose of this meta-analysis, CIN was defined as reported in each of the individual trials. All studies describe CIN as ≥25% decrease in glomerular filtration rate (GFR) or an absolute increase of serum creatinine ≥0.5 mg/dl on days 1-5 following the procedure. Four trials (Brar, Maioli, Masuda and Recio-Mayoral) also report clinical outcomes on death, dialysis, myocardial infarction and cerebrovascular events. The occurrence of death, congestive heart failure and the need for renal replacement therapy between the two groups were evaluated from these trials.

7. Statistical Analysis DerSimonian and Laird’s random effects model was utilized to pool the odds ratios (ORs) from individual trials. Cochran’s Q test was used to evaluate heterogeneity. Light and Pillemer’s funnel plot and Egger et al.’s regression- based method were used to assess publication bias. Duval and Tweedie’s trim-and-fill method was used to adjust for any observed publication bias. All analyses were undertaken in statistical software R.

8. CharacteristicsNaCl (n=772)NaHCO 3 (n=767)P value Age70.12 ± 3.8469.76 ± 3.730.12 Gender51 (69.59%)53.52 (69.88%)0.82 Height1.68 ± 0.221.66 ± 0.220.38 Weight67.01 ± 6.7867.97 ± 7.060.04 BMI26.35 ± 2.0226.5 ± 2.410.47 Prior MI23.9 (45.16%)33.04 (51.16%)0.65 Prior CHF14.48 (25.75%)12.57 (22.09%)0.54 Prior revasc.115 (64.6%)114 (65.1%)0.92 Diabetes27.32 (36.19%)25.98 (35.72%)0.57 Hypertension71.74 (71.96%)76.72 (76.05%)0.44 Systolic BP145.27 ± 15.51143.64 ± 15.510.68 Diastolic BP72.54 ± 7.9371.66 ± 8.650.33 LVEF51.71 ± 5.852.5 ± 5.510.55 Baseline Characteristics Data are presented as weighted mean+/-SD and weighted number of events (%)

9. CharacteristicsNaCl (n=772)NaHCO 3 (n=767)P value NAC84 (47.2%)80 (45.7%)0.82 β-blocker55.5 (76.96%)57.86 (79.47%)0.74 ACE inhibitors39.54 (51.56%)39.99 (52.81%)0.89 ARB21.01 (23.47%)23.03 (25.76%)0.19 Statin39.21 (74.88%)31.23 (71.28%)0.7 Diuretic33.41 (37.57%)34 (38.31%)0.75 CCB19.67 (22.31%)26.94 (29.29%)0.51 CA55.08 (43.05%)65.44 (51.01%)0.13 PCI41.86 (42.31%)34.59 (38.35%)0.37 CA, ad hoc PCI30 (27.02%)18 (16.67%)0.07 Contrast vol.157.39 ± 35.7157.22 ± 41.40.35 Baseline Characteristics Data are presented as weighted mean+/-SD and weighted number of events (%), ARB-angiotensin receptor blocker, CCB-calcium channel blocker

10. Development of CIN: All Patients Brar21/15824/165 Maioli25/25029/252 Brigouri2/10811/111 Masuda2/3010/29 Merten1/608/59 Ozcan4/8812/88 Recio-Mayoral1/5612/55 Total750759 Favours NaHCO 3 Favours NaCl OR 0.33 (0.16, 0.69), P=0.003 Duval and Tweedie’s trim-and-fill adjustment for publication bias showed there was an attenuated OR of 0.55 (95% CI 0.27-1.10; P=0.09) among all patients.

11. Development of CIN: Renal Failure Patients Brar21/15824/165 Maioli25/25029/252 Brigouri2/10811/111 Masuda2/3010/29 Merten1/608/59 Ozcan4/8812/88 Total694704 Favours NaHCO 3 Favours NaCl OR 0.41 (0.20, 0.82), P=0.01 Study NaHCO 3 NaCl Duval and Tweedie’s trim-and-fill adjustment for publication bias showed there was an attenuated odds ratio of 0.70 (95% CI 0.35-1.43; P=0.33).

12. Brigouri2/10811/111 Recio-Mayoral1/5612/55 Total164166 Development of CIN: Among Patients Who Received NAC Favours NaHCO 3 Favours NaCl OR 0.12 (0.04, 0.42), P=0.0008 Study NaHCO 3 NaCl NAC-N-acetylcysteine

13. Brar2/1754/178 Maioli1/2501/252 Brigouri1/111 Masuda1/303/29 Merten0/600/59 Ozcan1/88 Recio-Mayoral1/563/55 Total770.5772.5 Need for Renal Replacement Therapy Favours NaHCO 3 Favours NaCl OR 0.56 (0.22, 1.41), P=0.22 StudyNaHCO 3 NaCl

14. Brar4/1757/178 Maioli4/2503/252 Masuda0/302/29 Recio-Mayoral1/564/55 Total511514 Clinical Endpoint: Death Favours NaHCO 3 Favours NaCl OR 0.60 (0.26, 1.41), P=0.24 Study NaHCO 3 NaCl

15. Masuda11/3011/29 Recio-Mayoral1/562/55 Total8684 Clinical Endpoint: Heart Failure Favours NaHCO 3 Favours NaCl OR 0.85 (0.32, 2.24), P=0.74 Study NaHCO 3 NaCl

16. Limitations The results observed here may not be applicable to all patients in clinical practice. The present study remains subject to the inherent caveats of a meta-analysis including publication-bias. Patient level data was not available in this study. The dose of NaHCO 3 varied across the trials.

17. Conclusions The present meta-analysis demonstrates that based on currently available randomized trials, the administration of NaHCO 3 is superior to administration of NaCl alone in the prevention of CIN among patients with moderate to severe CKD undergoing diagnostic and interventional procedures requiring contrast media.

18. The use of NaHCO 3 however, did not result in significant benefit in terms of reductions in death, heart failure and the requirement for renal replacement therapy. The results of this study should be considered in the context of publication bias inherent to meta-analysis. Adequately powered studies are required to determine the beneficial effect of NaHCO 3 in preventing CIN and improving short-term and long-term clinical outcomes among patients who undergo coronary diagnostic and interventional procedures requiring contrast media. Conclusions