RACE611 CLINICAL EPIDEMIOLOGY AND EVIDENCE-BASED MEDICINE Theraputic study

Transcription

1 RACE611 CLINICAL EPIDEMIOLOGY AND EVIDENCE-BASED MEDICINE Theraputic study Master of Science Program in Medical Epidemiology and Doctor of Philosophy Program in Clinical Epidemiology Section for Clinical Epidemiology & Biostatistics Faculty of Medicine Ramathibodi Hospital Mahidol University - r ama.org/ Academic Year 2016 Semester 1

2 REFERENCES 1. Fletcher RH, Fletcher SW, Wagner EH. Clinical Epidemiology. The Essentials. 4 th Ed. Baltimore: Lippincott Williams & Wilkins, 2005: Haynes RB, Sackett DL, Guyatt GH, Tugwell P. Clinical epidemiology. How to do clinical practice research, 3 rd Ed. Philadephia: Lippincott Williams & Wilkins, 2006: Heneghan C, Badenoch D. Evidence-based medicine toolkit. 2 nd Ed. Massachusetts: Blackwell Publishing, 2006: Levine M, Haslan D, Walter S, et al. Harm. Guyatt G, Rennie D. Users guides to the medical literature. Essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:

3 Critical appraisal for therapeutic study Appraising an article about therapy Scenario (Appendix 4) You went home to visit your 60-year old mother. She had no history of coronary heart disease, cerebrovascular disease, or other chronic illness. She just returned from visiting her younger brother who had no history of any chronic illness but his doctor prescribed aspirin to him. She asks you whether she should received aspirin to prevent heart attacks Unsure if there was any protective effects, you promise yourself that you will check the medical literature before your next visit. After going through MEDLINE, you find an article entitled, A Randomized Trial of Low-Dose Aspirin in Primary Prevention of Cardiovascular Disease in Woman, which seemed like the best article that would help answer your question. A. Read the article and critically appraise its validity using Appraisal form for an Article on Therapy. Discuss the rationale for each criterion. B. After appraising study validity, decide if you want to go on and read the rest of the article. C. What would you advise your mother on your next visit?

4 Assignments A. What is your clinical question? P: I: C: O: B. What are your search term? C. Read the article and critically appraise its validity using the Appraisal Guides for an Article on therapy (Appendix 1). D. Appraise the results of the study, discussing the rationale for each in worksheet for therapy study (Appendix 2). E. Create critical appraisal topic (CAT) from this study (Appendix 3). READING: 1. Haynes RB, Sackett DL, Guyatt GH, Tugwell P. Clinical epidemiology. How to do clinical practice research, 3 rd Ed. Philadephia: Lippincott Williams & Wilkins, 2006: Levine M, Haslan D, Walter S, et al. Harm. Guyatt G, Rennie D. Users guides to the medical literature. Essentials of evidence-based clinical practice. Chicago: AMA Press, 2002:

5 Appendix (include articles assign for reading) Appendix 1: Guideline for Critical Appraisal on Therapeutic study Levine M, Haslan D, Walter S, et al. Harm. Guyatt G, Rennie D. Users guides to the medical literature. Essentials of evidence-based clinical practice. Chicago: AMA Press, 2002: THERAPY EBM Elective CEU USERS GUIDES FOR AN ARTICLE ABOUT THERAPY Are the results valid? Did experimental and control groups begin the study with a similar prognosis? Were patients randomized? Was randomization concealed (blinded or masked)? Were patients analyzed in the groups to which they were randomized? Were patients in the treatment and control groups similar with respect to known prognostic factor? Did experimental and control groups retain a similar prognosis after the study started? Were patients aware of group allocation? Were clinicians aware of group allocation? Were outcome assessors aware of group allocation? Was follow-up complete? What are the results? How large was the treatment effect? How precise was the estimate of the treatment effect? How can I apply the results to patient care? Were the study patients similar to the patient in my practice? Were all clinically important outcomes considered? Are the likely treatment benefits worth the potential harm and costs

6 What was the results? EBM ElectiveCEU ESTIMATING THE SIZE OF THE TREATMENT EFFECT Outcome + - a b Treated (Y) Control (X) c d Risk of Outcome: Y = a/(a + b) X = c/(c + d) Relative Risk, or Risk Ratio (RR), is the ratio of risk in the treated group (Y) to the risk in the control group (X): RR = Y/X Relative Risk Reduction (RRR) is the percent reduction in risk in the treated group(y) compared to controls (X): RRR = 1- RR = 1 Y/X x 100% or RRR = [(X Y)/X] x 100% Absolute Risk Reduction (ARR) is the difference in risk between control group (X) and the treated group (Y): ARR = X Y Number Needed to Treat (NNT) is the inverse of the ARR: NNT = 1/ARR = 1/(X Y) Please look for future tools and EBM information at the Users Guides Web site,

7 Appendix 2: Therapeutic worksheet A. Are the results valid? 1. Did experimental and control groups begin the study with a similar prognosis? * Were patients randomized? * Was randomization concealed (blinded or masked)? * Were patients analyzed in the groups to which they were randomized? * Were patients in the treatment and control groups similar with respect to known prognostic factor? 2. Did experimental and control groups retain a similar prognosis after the study started? * Were patients aware of group allocation? * Were clinicians aware of group allocation? * Were outcome assessors aware of group allocation? * Was follow-up complete? B. What are the results? * How large was the treatment effect? * How precise was the estimate of the treatment effect? C. How can I apply the results to patient care? * Were the study patients similar to the patient in my practice? * Were all clinically important outcomes considered? * Are the likely treatment benefits worth the potential harm and costs?

8 Appendix 3: Critical appraisal topic (CAT) for therapeutic study Clinical Question: Citation: A. Study Characteristics: 1. Patients included 2. Interventions Compared 3. Outcomes Monitored B. Validity Criteria: 1. Were patients randomized? 2. Was randomization concealed? 3. Were patients analyzed in the groups to which they were randomized? 4. Were patients in treatment and control groups similar at baseline? 5. Were patients aware of group allocation? 6. Were clinicians aware of group allocation? 7. Were outcome assessors aware of group allocation? 8. Was follow-up complete? C. Results [choose appropriate tables(s)]: Outcome Rc Rt RR RRR ARR NNT P Outcome Mean (c) Mean (t) Mean diff. P D. Applicability: 1. Are the study patients similar to the patients in my practice? 2. Were all clinically relevant outcomes reported? 3. Are the likely treatment benefits worth the harm and costs? Author s Conclusion: Reviewer s Conclusion: Reviewer: Date:

9 Appendix Reading article

10 Acetylcysteine In Diabetes (AID): A randomized study of acetylcysteine for the prevention of contrast nephropathy in diabetics Louis C. Coyle, DO, Antonio Rodriguez, MD, Robert E. Jeschke, MD, Anabela Simon-Lee, MD, Kevin C. Abbott, MD, MPH, and Allen J. Taylor, MD Washington, DC Background Patients with diabetes mellitus (DM) are at increased risk of contrast-associated nephropathy irrespective of their baseline creatinine (Cr). We tested the efficacy of N-acetylcysteine (NAC) relative to hydration in unselected patients (irrespective of baseline Cr) with DM. Methods We conducted a randomized open-label study comparing hydration alone (combined oral and rapid intravenous hydration, n = 69) to NAC plus hydration (similar hydration protocol plus NAC 600 mg BID 4 doses, n = 68) in diabetic patients (mean age 65 F 10 years, 65% men) undergoing elective coronary angiography. The primary end point was the mean change in serum Cr measured up to 96 hours postangiography. Results Baseline Cr was 1.14 F 0.43 mg/dl (Cr z1.3 mg/dl in 37 subjects). Baseline characteristics including blood urea nitrogen, Cr, and contrast volume were similar between the 2 groups. The mean Cr change in the NAC group was 0.14 F 0.47 versus 0.08 F 0.11 mg/dl in the hydration only group ( P = NS). Contrast-associated nephropathy, defined as a z0.5 mg/dl increase in Cr, was significantly more common in the NAC group, 9.2% versus 1.4%, P =.043. Similar results were found in the subgroup of participants with either an increased baseline serum Cr (z1.3 mg/dl) or in those receiving high contrast volumes (N100 ml). Conclusions N-Acetylcysteine provides no benefit over an aggressive hydration protocol in patients with DM undergoing coronary angiography. (Am Heart J 2006;151:1032.e e12.) Individuals with diabetes mellitus (DM) are at increased risk for contrast-associated nephropathy (CAN) 1,2 and thereby constitute an appropriate target population for efforts at prevention of this important complication with substantial associated morbidity and mortality risk. 2 Preventative therapies primarily include limitation of contrast exposure, administration of fluid, and selection of low osmolality, nonionic contrast media. 3 However, because these measures provide incomplete protection from CAN, interest has emerged in a number of adjunctive short-term pharmacotherapies. Among these, N-acetylcysteine (NAC) has been of considerable interest after an initial report by Tepel et al 4 From the Cardiology Service, Walter Reed Army Medical Center, Washington, DC. The opinions or assertions herein are the private views of the authors and are not to be construed as reflecting the views of the United States Army, United States Navy, or the Department of Defense. Submitted May 24, 2005; accepted February 5, Reprint requests: Allen J. Taylor, MD, Chief, Cardiology Service, Walter Reed Army Medical Center, 6900 Georgia Ave, NW, Building 2, Room 3L28, Washington, DC allen.taylor@na.amedd.army.mil /$ - see front matter n 2006, Published by Mosby, Inc. doi: /j.ahj showed a reduction in the increase in serum creatinine (Cr) with NAC compared with hydration alone. After this, a number of small studies have been conducted, which have been quantitatively summarized by several different meta-analyses, 5-8 with the summary risk ratio suggesting that NAC may be modestly effective in preventing contrast-associated increases in serum Cr. In general, these studies have been conducted in individuals with preexisting renal insufficiency. The purpose of this study is to extend our understanding of the potential of NAC to prevent CAN to unselected patients with DM, regardless of their level of serum Cr. Methods This study was a randomized, open-label, single center trial approved by the Department of Clinical Investigation of Walter Reed Army Medical Center, Washington, DC, and funded by the clinical care provided under the Army Medical Department of the Department of Defense. Between April 2001 and December 2002, 169 potential subjects were identified, of whom 137 provided consent for the study and entered the intervention. Eligible patients were men and women older than 18 years with a diagnosis of DM scheduled to undergo coronary angiography. Exclusion criteria were the need for emergency

11 1032.e10 Coyle et al American Heart Journal May 2006 Table I. Baseline characteristics of the 137 study patients Table II. Procedural variables Hydration, N=69 Hydration + NAC, N = 68 Hydration, N=69 Hydration + NAC, N = 65 P Male, n (%) 47 (68.1) 42 (61.8) Age, mean F SD 63.3 F 9.2 y 66.7 F 10 y Hypertension, n (%) 60 (87.0) 62 (91.2) Congestive 8 (11.6) 8 (11.8) heart failure, n (%) DM, n (%) 69 (100%) 68 (100%) Medications n (%) Angiotensin-converting 45 (65.2) 45 (66.2) enzyme inhibitor Insulin 18 (26.1) 18 (26.5) Oral agent 35 (50.7) 29 (42.6) Metformin 31 (41.9) 39 (42.6) Thiazolidinedione 7 (10.1) 5 (7.4) Angiography indication, n (%) Angina/chest pain 42 (60.9) 44 (64.7) Recent acute 5 (7.2) 6 (8.9) coronary syndrome Others 22 (31.9) 18 (26.5) Baseline Cr (mg/dl) Mean 1.10 F F Cr range Baseline glomerular 76 F F filtration rate (ml/kg per minute) Systolic blood pressure 149 F F (mm Hg) Diastolic blood pressure 75 F F (mm Hg) Left ventricular end-diastolic 20 F 9 18 F 8.36 pressure (mm Hg) Preangiogram 1172 F F hydration (ml) Postangiogram 1971 F F hydration (ml) Contrast volume (ml) 98 F F procedures, anticipated inability to complete the hydration protocol, or a history of CAN. Baseline serum Cr was measured during the precatheterization evaluation, and glomerular filtration rate was calculated using the Modification of Diet in Renal Disease Study Group equation. 9 Randomization and end points Patients were randomized in a 1:1 fashion using a random number generator with concealment of allocation to either hydration only (control arm) or hydration plus NAC (intervention arm). This was an open-label study because of the inability to blind the intervention (administration of NAC) due to its unique taste. All participants received an oral prehydration protocol consisting of 1 L of clear fluids over the 10 hours before angiography. At the initiation of the coronary angiographic procedure, intravenous hydration (0.45N saline) at a rate of 300 ml/h for 6 hours was begun in the catheterization laboratory. Patients received a 1-L bottle and instructed to drink the full contents of clear liquids over the 10 hours before coronary angiography. Intravenous and oral fluids administered either during or after the procedure were tracked by study personnel. Patients assigned to the NAC arm received NAC 600 mg every 12 hours by mouth for 2 doses before and 2 doses after angiography. All patients underwent invasive coronary angiographic procedures with the selection and volume of contrast at the discretion of the operator. The prespecified primary end point was the maximal change in serum Cr between 48 and 96 hours after angiography. The prespecified secondary end point was the proportion of participants with contrast associated nephropathy (CAN) defined as an increase in serum Cr of z0.5 mg/dl. Statistical analysis Assuming a baseline Cr of 1.4 F 0.6 mg/dl, for an a of.05 and power of 0.80, a sample size of 65 per group was required to detect a mean difference in the change in serum Cr of 0.3 mg/dl, similar to the effect size by Tepel et al. 4 A secondary end point was the proportion of participants with CAN, although the trial was not powered to the secondary end point. Comparisons of baseline characteristics and end point data between the 2 study groups were performed using a t test for independent continuous variables and a v 2 test for categorical variables. Normality of the data was tested with a 1-sample Kolmogorov-Smirnov test. Data are presented as mean F SD. A 2-tailed P value of V.05 was considered significant. Results The characteristics of the 137 participants are shown in Table I, demonstrating even allocation of baseline characteristics and medications between the 2 study groups. There was no significant difference between the 2 groups for any measured baseline variables including Cr, calculated glomerular filtration rate, blood pressure or left ventricular end-diastolic pressure. The hydration and contrast results are shown in Table II. The NAC group received slightly more preprocedural fluids and slightly less contrast (88 F 61 vs 98 F 65 ml, P =.34), but these differences were not statistically significant. Most of the patients in each group received nonionic low-osmolar contrast. Nonionic contrast was administered in 73.8% of NAC subjects compared with 68.1% of hydration subjects ( P = NS). Low or isoosmolar contrast was administered to 86.2% of NAC subjects compared with 81.2% of hydration subjects ( P = NS). The observed changes in Cr were unrelated to the type of contrast (nonionic or low osmolar) administered. Specifically, the changes in serum for those administered high- versus low-osmolar contrast (0.10 F 0.15 vs 0.12 F 0.33 mg/dl, P =.80) or ionic versus nonionic contrast (0.07 F 0.15 vs 0.13 F 0.35 mg/dl, P =.27) were similar.

12 American Heart Journal Volume 151, Number 5 Coyle et al 1032.e11 Figure 1 Figure 2 Bar graph showing the maximal change in serum Cr 48 to 96 hours after contrast administration among diabetic subjects in the AID study treated with either hydration alone or hydration and NAC. Box plot showing the distribution of values in the maximal change in serum Cr 48 to 96 hours after contrast administration among diabetic subjects in the AID study treated with either hydration alone or hydration and NAC. There was no difference between the 2 groups for the primary end point. The mean maximal change in serum Cr measured 48 to 96 hours after angiography in the hydration group was 0.08 F 0.11 and 0.15 F 0.42 mg/dl in the NAC group (nonparametric P =.80) (Figure 1). For the secondary end point of the proportion of individuals with CAN, 1 (1.4%) of 69 participants in the hydration only group developed CAN compared with 6 (9.2%) of 65 in the NAC group ( P =.043) (Figure 2). Among the 7 subjects who developed CAN, the mean contrast administration was significantly higher (143 F 57 vs 91 F 63 ml, P =.033), but their pre- and postangiography hydration volumes were similar to the other study subjects. The calculated glomerular filtration rate decreased slightly in both groups (hydration 5.3 F 8.5 vs NAC 5.5 F 11.8 ml/kg per minute, P =.91). Nonprespecified subgroup analyses were performed on questions of interest that emerged from the primary data analysis. Among participants receiving greater than 100 ml of contrast, the change in serum Cr was similar in the hydration (0.11 F 0.12 mg/dl, n = 35) and NAC groups (0.24 F 0.61 mg/dl, n = 29, P =.25). Among participants with baseline Cr z1.3 mg/dl, the change in serum Cr was similar in the hydration (0.12 F 0.16 mg/dl, n = 18) and NAC groups (0.29 F 0.72 mg/dl, n = 19, P =.34). We also performed exploratory logistic regression to test the relationships among CAN, baseline Cr, and volume of contrast administration. In this analysis, CAN was predicted by the volume of contrast administration (OR 1.015/mL contrast, 95% CI , P =.022) and treatment with NAC (OR 9.36, 95% CI , P =.053). No study-related adverse events were noted from the NAC or the hydration protocol. Discussion The potential of NAC to reduce the risk of CAN has been a topic of intense, recent interest, manifested by the number of small randomized controlled studies on this topic. 5-8 This is likely, in part, due to the absence of effective adjunctive pharmacotherapies for this important angiographic complication. However, it seems likely that the potential for benefit from NAC and the absence of data indicating potential harm also have contributed to interest in NAC before definitive demonstration of meaningful clinical benefit on the incidence of CAN and its attendant morbidity and mortality. The present study, the first to show potential harm of this pharmacotherapy, and the emerging literature on the confounding effects of NAC on serum Cr levels 10 lead us to question the application of NAC for the prevention of CAN. Multiple small randomized controlled clinical trials have been conducted, with many, but not all of them, showing positive results when outcomes were measured as the change in serum Cr or the occurrence of CAN. Multiple meta-analyses of this topic have been reported 5-8 and, in general, have suggested that the evidence to date shows a modest benefit of NAC on CAN. However, the degree of instability in these analyses and the potential for publication bias within the literature are notable. In the most recent meta-analysis by Pannu et al, 8 the absence of any small trials with

13 1032.e12 Coyle et al American Heart Journal May 2006 negative results was noted, indicating possible publication bias. Of further concern was a high degree of quantitative instability in the analysis such that the addition of a single negative trial of only 50 participants could meaningful alter the quantitative analysis. Of further concern is the absence of a demonstrated biologic mechanism for NAC as an effective treatment of pharmacoprophylaxis of CAN. Proposed mechanistic theories include several basic and animal investigations postulating antioxidant or vasodilator effects of NAC. However, recent evidence indicates that any observed changes in Cr after NAC administration may simply be confounded by the independent effect of NAC on Cr measurement. An intriguing study by Hoffmann et al 10 showed that, in the absence of contrast administration in subjects with normal Cr, NAC may decrease serum Cr with no effect on glomerular filtration rate, as measured by serum cystatin levels. Thus, it appears that NAC may interfere with the measurement of serum creatinine leading to the conclusion that studies showing a benefit of NAC after contrast administration are confounded. The present study, showing no effect of NAC on the pharmacoprophylaxis of CAN, confirms the negative results of other small randomized controlled trials on this topic and is the first to suggest that this intervention, applied broadly to patients with DM irrespective of serum Cr level, could even be harmful. In light of the data published to date on NAC, an alternative interpretation of our study is that our adverse finding with respect to CAN is a chance finding (a type I error) among the many small trials that have been conducted. Thus, we recommend that the primary interpretation of our study be from the perspective of the primary end point, with our negative results demonstrating no benefit of NAC. Within this interpretation of our data, several limitations are notable. First, the data presented here are applicable to unselected patients with DM in which the control group was treated with an aggressive oral and intravenous pre- and posthydration strategy. As previously shown by our group, this hydration strategy is safe in appropriately selected patients, and its effectiveness is similar to intravenous hydration alone. 11 In comparison, prior studies of NAC have typically used standard hydration of 1 ml/kg per minute, an amount that may be insufficient for maximal protection from contrast nephrotoxicity. 12,13 This study focused upon treatment of a high-risk group, patients with DM, although most had a normal level of serum Cr. Lastly, it should be noted that the intravenous infusion protocol in this study used 0.45N saline; whether similar results would be found using normal saline is unclear. Based upon the discordant results of many small clinical trials, and the potential that any Cr change observed in the setting of NAC may be unrelated to changes in glomerular filtration rate, we believe at this time that it is premature to apply NAC in pharmacoprophylaxis of CAN. Other potential pharmacotherapies such as intravenous administration of serum bicarbonate 14 or vitamin C 15 are under study for this indication, and ultimately, a sufficiently powered randomized clinical trial is required. Until then, proper hydration, limitation of contrast volume, and contrast selection should be the primary methods to prevent CAN. References 1. Lautin EM, Freeman NJ, Schoenfeld AH, et al. Radiocontrastassociated renal dysfunction: incidence and risk factors. AJR Am J Roentgenol 1991;157: McCullough PA, Wolyn R, Rocher LL, et al. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 1997;103: Maeder M, Klein M, Fehr T, et al. Contrast nephropathy: review focusing on prevention. J Am Coll Cardiol 2004;44: Tepel M, van der GM, Schwarzfeld C, et al. Prevention of radiographic-contrast-agent induced reductions in renal function by acetylcysteine. [see comments]. N Engl J Med 2000;343: Alonso A, Lau J, Jaber BL, et al. Prevention of radiocontrast nephropathy with N-acetylcysteine in patients with chronic kidney disease: a meta-analysis of randomized, controlled trials. Am J Kidney Dis 2004;43: Birck R, Krzossok S, Markowetz F, et al. Acetylcysteine for prevention of contrast nephropathy: meta-analysis. Lancet 2003; 362: Isenbarger DW, Kent SM, O Malley PG. Meta-analysis of randomized clinical trials on the usefulness of acetylcysteine for prevention of contrast nephropathy. Am J Cardiol 2003;92: Pannu N, Manns B, Lee H, et al. Systematic review of the impact of N-acetylcysteine on contrast nephropathy. Kidney Int 2004; 65: Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130: Hoffmann U, Fischereder M, Kruger B, et al. The value of N-acetylcysteine in the prevention of radiocontrast agent-induced nephropathy seems questionable. J Am Soc Nephrol 2004;15: Taylor AJ, Hotchkiss D, Morse RW, et al. PREPARED: Preparation for Angiography in Renal Dysfunction: a randomized trial of inpatient vs outpatient hydration protocols for cardiac catheterization in mild-to-moderate renal dysfunction. Chest 1998;114: Bader BD, Berger ED, Heede MB, et al. What is the best hydration regimen to prevent contrast media induced nephrotoxicity? Clin Nephrol 2004;62: Baker CS, Wragg A, Kumar S, et al. A rapid protocol for the prevention of contrast-induced renal dysfunction: the RAPPID study. J Am Coll Cardiol 2003;41: Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrastinduced nephropathy with sodium bicarbonate: a randomized controlled trial. JAMA 2004;291: 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: