RCT CONSORT CONSORT Introduction CONSORT Method RCT

Transcription

1 RCT 2018/01/09 ICU

2 RCT CONSORT CONSORT Introduction CONSORT Method RCT

3 RCT CONSORT CONSORT Introduction CONSORT Method RCT

4 /

5

6

7

8 Intervention Outcome Intervention, Outcome I O

9 Patients Intervention Patient Patient Patient Patient Patient Comparison Patient Patient Patient Patient Patient Outcome Outcome

10 Patients RCT Intervention Patient Patient Patient Patient Patient Comparison Patient Patient Patient Patient Patient Outcome? Outcome

11

12 Patients RCT Intervention Patient Patient Patient Patient Patient Comparison Patient Patient Patient Patient Patient Outcome? Outcome

13 RCT

14 N Engl J Med 2017;377:

15 RCT CONSORT CONSORT Introduction CONSORT Method RCT

16 CONSORT 1996 CONSORT Consolidated Standards of Reporting Trials: 2001 CONSORT RCT 25

17

18 RCT CONSORT CONSORT Introduction CONSORT Method RCT

19

20 Introduction rationale hypothesis Clinical Question Research Question PICO Intervention vs. Comparison

21 RQ FINER Feasible Interesting Novel Ethical Relevant

22 OK? Feasible Interesting Novel Ethical Relevant

23 RCTs are one of the possible experiments. Premise Hypothesis Outcome

24 RCTs are one of the possible experiments. Premise Hypothesis Outcome

25 RCTs are one of the possible experiments. Premise Hypothesis Outcome

26 Many other trials, in our opinion, failed either because,,,, Wrong premises Premises that were too vague and generic. If too generic, the heterogeneity of the population is an unavoidable consequence.

27 vs. The trialists spent a lot of effort in improving their internal validity, by increasing the number of rules and caveats to avoid possible biases in the conduct of the trials. This obviously makes the results valid only for the restricted patient population enrolled in the study and the generalization of the results (the external validity) is problematic.

28 EGDT: N Engl J Med 2001;345:

29

30 Hypothesis Severe sepsis, septic shock CVP ABP SvO2

31 Outcome N Engl J Med 2001;345:

32 ?

33 Premise 1, 2 May be by chance the first RCT picked a population with baseline SvO2 of 48 %. The baseline SvO2 in the remaining trials was already higher than 70 %, suggesting that oxygen imbalance was not the primary problem in the majority of the patients.

34 ?

35 Premise 3 Too generic and wrong

36 Crit Care Med 2010; 38[Suppl.]:S534 S538 The pendulum effect Single center Multi center EGDT

37 Truly negative trials are valuable because they prevent the use of interventions that are either costly but nonbeneficial or even harmful (e.g.,intensive insulin therapy and hydroxyethyl starch). Understanding an evidence base requires the ability to distinguish among these reasons so as to differentiate trials that are truly negative from those that may be falsely negative.

38 Studies published between January 2007 and May 2013 Single center

39 RCT CONSORT CONSORT Introduction CONSORT Method RCT

40

41 eligibility criteria

42 Participant eligibility criteria

43 Crit Care Med 2010; 38[Suppl.]:S534 S538 Why have so many RCTs failed? Was the right group of patients not identified? AMI ICU

44 General ICU sepsis ALI

45 Crit Care Med 2010; 38[Suppl.]:S534 S538 High Exclusion Rates Lead to Limited Clinical Applicability With such tight exclusion criteria limiting enrollment,,,, How can the patient population that is actually studied really be representative of the entire population initially considered eligible? How can the results be applicable to the general ICU population?

46 TRICC study Just 13% of 6451 patients screened were randomized.

47 VAAST study 802 (12.9%) of the 6229 patients screened underwent randomization

48 N Engl J Med 2010;362: (84%) of 2011 patients were randomized.

49 Intervention

50 Crit Care Med 2010; 38[Suppl.]:S534 S538 Why have so many RCTs failed? Were the interventions ineffective? truly negative OK

51 Crit Care Med 2010; 38[Suppl.]:S534 S538 Why have so many RCTs failed? Was it a question of severity? U shape or Is the timing of interventions correct? Shock 2007; 28: ARDS Chest 2010; 137:81 87

52 Outcome

53 Sample size α p

54 Crit Care Med 2010; 38[Suppl.]:S534 S538 Why have so many RCTs failed? Were the end points wrong? : Was the power inadequate?

55

56 2

57 34

58 47

59 (Allocation concealment)

60 allocation :,

61 ABAB AABB BABA,,,,,, concealment

62

63 Implementation enrollment assign

64 9

65 Proper use of envelopes Envelopes must be made opaque through use of a cardboard or aluminium foil insert Envelopes must be numbered, opened sequentially Reporting the use of sequentially numbered opaque sealed envelopes (SNOSE) is accepted to be the minimum standard

66 Blinding masking Unblinded RCTs can be a problem because there is a risk that patients will receive different treatment according to the physician s own beliefs. Crit Care Med 2010; 38[Suppl.]:S534 S538

67 Statistical method

68 Registration Protocol Funding

69 RCT CONSORT CONSORT Introduction CONSORT Method RCT

70 Critical Care RCT JAMA. 2017;317(13):

71 Premise MENDS study 28

72 Hypothesis DEX ventilater-free days

73 8 Investigator-initiated Chief investigator Yu Kawazoe (2003 ) Kyohei Miyamoto 2006 February 2013 January Collaborater Takeshi Morimoto Hitoshi Yamamura Akihiro Fuke Atsunori Hashimoto Makoto Ito Nobuaki Shime Kohei Kato Kenji Yamauchi Hiroyuki Koami ESICM

74 Inclusion NIV

75 Exclusion (Child-Pugh B or C) (NYHA ) Dexmedetomidine Attending physician's decision

76

77 Attending physician eligible Internet Data and Information Center for Medical Research (INDICE) INDICE UMIN electronic data capturing system

78

79 ( ) COPD ID ID

80

81 JSICM JSICM

82 Primary Outcome Co - Primary Outcome Ventilator Free Days ( ) /5/2 Ventilator Free Days (Outcome data ) 5 5 VFD VFD2

83 Sample size 28 80% 60% MENDS study 84% vs. 59% α 5% β 20% %

84 28 80% 80% 60% 72% α 5% β 20% %

85 Masking, Blinding Open label limitation

86 The main statistical analysis for efficacy Comparison between groups of 28-day mortality Measure: Death until 28 days after the initiation of mechanical ventilation. Patients who were alive at 28 days were censored at 28 days. Aim: Comparison of the occurrence of the events between groups Test: Log-rank test, Cox proportional hazard model Significance: Two-sided p values of less than 0.05 were considered statistically significant. Confidence interval: 95%confidence interval (CI) of the hazard ratio were calculated. Expression: The number of the occurrence of the events, Cumulative incidence, and hazard ratio.

87 Negative!

88 The secondary statistical analysis for efficacy 28-day mortality Measure: Death during 28 days Aim: Comparison of the occurrence of the events between groups Test: Chi-square test or Fisher exact test Significance: Two-sided p values of less than 0.05 were considered statistically significant. Confidence interval: NA. Expression: The number of the occurrence of the events, the ratio of the occurrence of the events.

89 Data and safety monitoring board (DSMB) and interim analysis Dr. Sadao Kawasaki (Department of Emergency Medicine Minami Wakayama Medical Center, Japan) Dr. Takahiro Ashikawa (Department of emergency medicine, Minami Wakayama Medical Center, Japan) Dr. Yasuhiro Iwasaki (Department of Emergency and Critical Care Medicine, Wakayama Medical University, Japan)

90 The DSMB will independently perform 1 Emergent reporting system

91 Discussion The study may have identified a clinically important benefit of dexmedetomidine an 8% reduction in 28-day mortality that did not demonstrate statistical significance. Physicians may consider an 8% difference in 28-day mortality to be clinically significant but this study was underpowered to detect this difference.

92 Funding Funding/Support: This study was supported in part by a noncontractual research grant to Wakayama Medical University provided by Hospira Japan. Role of the Funder/Sponsor: Hospira Japan participated in part of the design of this study; however, it had no role in the conduct of the study; collection, management, analysis, and interpretation of the data; preparation or approval of the manuscript; or decision to submit the manuscript for publication.

93 No earlier multicenter RCTs have investigated the effects of ANP infusion in patients with cardiac surgery associated AKI. We therefore decided to evaluate the effects of ANP on renal function and medical costs in patients with cardiac surgery associated AKI.

94 ICU ICU ICU 11 May March

95 Double-blind ANP 0.02 μg /kg/min (2 ml/h) (5% glucose, 2 ml/h Preoperative level

96 Study coordinators INDICE INDICE randomization system

97 Inclusion Age of 20 years Recent elective cardiac surgery, namely, CABG, valve surgery, thoracic aortic aneurysm surgery, or a combination thereof. AKI within 48 hours after cardiac surgery ANP Exclusion Severe hypotension or cardiogenic shock Right ventricular infarction Dehydration End-stage renal disease Administration of nonsteroidal anti-inflammatory drugs, angiotensin-converting enzyme(ace) inhibitor, or angiotensin receptor blocker (ARB) within 24 hours before the operation Administration of contrast agent within 3 days before the operation Extracorporeal membrane oxygenation.

98 The primary end point Change in renal function over the 90-day follow-up measured by serum levels of creatinine and cystatin C, and creatinine clearance or egfr.

99 Sample size and power calculation a difference in means of α 5% β 20% 87 10% 97

100

101 Reasons why,,,, approval of ethics committees difficulty in obtaining consent from patients decisions by several hospitals to prioritize patients with ventricular assist devices. many surgeons in trial sites insisted on using ANP in patients with high preoperative creatinine levels these patients were excluded before assessment for eligibility and were administered ANP during surgery.

102

103

104 Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

105 RCT CONSORT CONSORT Introduction CONSORT Method RCT

106 positive negative positive positive positive