RECORD: Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes

The number of participants with cardiovascular death events (death due to cardiovascular causes or deaths with insufficient information to rule out a cardiovascular cause) and cardiovascular hospitalisation events (hospitalisation for a cardiovascular event, excluding planned admissions not associated with a worsening of the disease/condition of the participant) was recorded.

All deaths identified during the original record study and discovered after the re-adjudication efforts began were included.

IR was based on original RECORD endpoint definitions. CV death= no unequivocal non-CV cause (sudden death, death from acute vascular events, heart failure, acute MI, other CV causes, and deaths adjudicated as unknown cause). MI event=hospitalization + elevation of specific cardiac biomarkers above the upper limit of normal + cardiac ischemia symptoms/new pathological electrocardiogram findings. Stroke event=hospitalization + rapidly developed clinical signs of focal/global disturbance of cerebral function for more than 24 hours, with no apparent cause other than a vascular origin.

Independent re-adjudication was based on the Standard Data Collection for Cardiovascular Trials Initiative (draft October 2011) endpoint definitions. CV death included death resulting from an acute MI; sudden cardiac death and death due to heart failure, stroke, and to other CV causes. Deaths of unknown cause were counted as CV deaths. MI was defined as evidence of myocardial necrosis in a clinical setting consistent with myocardial ischemia. Stroke was defined as an acute episode of neurological dysfunction caused by focal or global brain, spinal cord, or retinal vascular injury.

The number of participants with a CV death (or unknown) as determined by independent re-adjudication using the original RECORD endpoint definitions was recorded. CV death was defined as any death for which an unequivocal non-CV cause could not be established. CV death included death following heart failure, death following acute myocardial infarction (MI), sudden death, death due to acute vascular events, and other CV causes. Deaths due to unknown causes were classified as "unknown deaths," but were counted as CV deaths for the analysis of this endpoint.

The number of participants with a CV (or unknown) death as determined by independent re-adjudication using the Standard Data Collection for Cardiovascular Trials Initiative (draft October 2011) endpoint definitions was recorded. CV death included death resulting from an acute myocardial infarction (MI), sudden cardiac death, death due to heart failure, death due to stroke, and death due to other CV causes. Deaths of unknown cause were counted as CV deaths.

The number of participants with an MI (fatal or non-fatal) event as determined by independent re-adjudication using the original RECORD endpoint definitions was recorded. An event of MI was defined as hospitalization plus elevation of cardiac biomarkers troponin (TN) I and/or TNT above the upper limit of normal (ULN) or creatinine kinase (CK) MB (M=muscle type; B=brain type) isoenzyme >= 2x the ULN or CK > 2x the ULN plus typical symptoms of cardiac ischemia or new pathological electrocardiogram findings, or cause of death adjudicated as MI.

The number of participants with an MI (fatal or non-fatal) event as determined by independent re-adjudication using the Standard Data Collection for Cardiovascular Trials Initiative (draft October 2011) endpoint definitions was recorded. An event of MI was defined as evidence of myocardial necrosis in a clinical setting consistent with myocardial ischemia.

Par. with a stroke (fatal or non-fatal) event as determined by independent re-adjudication using the original RECORD endpoint definitions was recorded. A stroke event=hospitalization plus rapidly developed clinical signs of focal (or global) disturbance of cerebral function lasting more than 24 hours (unless interrupted by thrombolysis, surgery, or death), with no apparent cause other than a vascular origin, including par. presenting clinical signs/symptoms suggestive of subarachnoid haemorrhage/intracerebral haemorrhage/cerebral ischemic necrosis or cause of death adjudicated as stroke.

The number of participants with a stroke (fatal or non-fatal) event as determined by independent re-adjudication using the Standard Data Collection for Cardiovascular Trials Initiative (draft October 2011) endpoint definitions was recorded. An event of stroke was defined as an acute episode of neurological dysfunction caused by focal or global brain, spinal cord, or retinal vascular injury.

Composites of participants with first cardiovascular (CV) hospitalisations and CV death or all-cause death and individual first events of acute myocardial infarction (MI) , stroke, congestive heart failure (CHF), CV death, and all-cause death.

The total number of events for individual components of cardiovascular (CV) hospitalisations and cardiovascular deaths were recorded. MI, myocardial infarction.

Participants with first cardiovascular death (death due to cardiovascular causes or deaths with insufficient information to rule out a cardiovascular cause) and cardiovascular hospitalisation (hospitalisation for a cardiovascular event, excluding planned admissions not associated with a worsening of the disease/condition of the participant) were recorded by study stratum.

The number of participants with first cardiovascular or microvascular events (renal, foot, eye) were recorded.

Failure of glycaemic control was defined as two consecutive HbA1c values of ≥8.5 percent, or HbA1c ≥8.5percent at a single visit, after which the subject was either moved to the post-randomised treatment phase or triple therapy was started.

The number of participants with addition of a third oral agent or switch to insulin from randomised dual combination treatment were recorded.

The number of participants starting insulin at any time during the study was recorded.

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in HbA1c was calculated as the value at Month 60 minus the Baseline value.

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in fasting plasma glucose was calculated as the value at Month 60 minus the Baseline value.

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in insulin and pro-insulin was calculated as the value at Month 60 minus the Baseline value.

Number of responders, i.e., participants meeting glycaemic targets (HbA1c less than or equal to 7 percent, FPG less than or equal to 7 mmol/L)

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in HOMA beta-cell function and insulin sensitivity was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in TC, LDL cholesterol, HDL cholesterol, triglycerides, and FFAs was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in TC:HDL cholesterol and LDL cholesterol:HDL cholesterol was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in Apo-B was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in urinary albumin creatinine ratio was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in body weight was calculated as the value at Month 60 minus the Baseline value.

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in alanine aminotransferase was calculated as the value at Month 60 minus the Baseline value.

Model adjusted (adjusted for any imbalances in the baseline values between within stratum treatment groups) change from baseline in waist circumference was calculated as the value at Month 60 minus the Baseline value.

Model adjusted (adjusted for any imbalances in the baseline values between within treatment groups) change from baseline in SBP and DBP was calculated as the value at Month 60 minus the Baseline value.

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in C-Reactive Protein was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in fibrinogen was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The model adjusted (adjusted for any imbalances in the baseline [BL] values between within stratum treatment groups) ratio to BL in plasminogen activator inhibitor-1 (PAI-1) antigen was calculated as the ratio of the Month 60 value to the BL value and was expressed as percent change from BL. For each treatment group, the model-adjusted mean change from BL at Month 60 was determined on the log scale. This mean was then back transformed to give a geometric mean (GM) of the ratio of the Month 60 value to BL on the original scale. The GM was expressed as a percentage (100*[GM^-1]).

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up (OFU) was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the OFU. The neoplasms/cancer events of bladder, breast, colon, liver, pancreatic, prostate cancer, and melanoma were pre-specified as cancers of interest for the OFU. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up (OFU) was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the OFU. The neoplasms/cancer events of bladder, breast, colon, liver, pancreatic, prostate cancer, and melanoma were pre-specified as cancers of interest for the OFU. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The OFU was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the OFU. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The OFU was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the OFU. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. The following bone fractures were grouped and were identified as potentially high morbidity bone fractures: hip, pelvis, upper leg, vertebral (lumbar spine, thoracic spine, cervical spine, spine - site unknown).

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. The following bone fractures were grouped and were identified as potentially high morbidity bone fractures: hip, pelvis, upper leg, vertebral (lumbar spine, thoracic spine, cervical spine, spine - site unknown).

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant.

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. The indicated fracture outcome was pre-specified in the CRF and included "Unknown" as a category. Fracture events with missing outcome data were reported as "Data unavailable."

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. The indicated fracture outcome was pre-specified in the CRF and included "Unknown" as a category. Fracture events with missing outcome data were reported as "Data unavailable."

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. A bone fracture event is defined as one or more fractured bones occurring on the same date and that had the same Higher Level Group Term (HLGT) for fracture location, per participant. The indicated fracture outcome was pre-specified in the CRF and included "Unknown" as a category. Fracture events with missing outcome data were reported as "Data unavailable."

The observational follow-up was designed to collect data concerning cancer and bone fractures in RECORD participants during a 4-year period after the end of the main RECORD study. At the end of the main study, all study medication was stopped. Participants were not provided with study medication in the observational follow-up; instead, anti-diabetic treatment was prescribed at the investigator's discretion. An SAE is defined as any event that is fatal; life threatening; disabling/incapacitating; results in hospitalization (excluding elective surgery or routine clinical procedures); prolongs a hospital stay; is associated with a congenital abnormality; cancer; is associated with an overdose. In addition, any event that the investigator regards as serious or that would suggest any significant hazard, contraindication, side effect, or precaution that may be associated with the study procedures should be reported as an SAE.