Cardiac resynchronisation therapy defibrillators (CRT-D) or implantable cardioverter defibrillators (ICD) are used to prevent sudden death in people who have, or have high risk of, dangerously abnormal heart rhythms such as atrial fibrillation. In particular, the devices are often used in the management of heart failure. Heart failure occurs when the heart becomes too weak or stiff to properly pump blood around the body. Single chamber (SC) and dual chamber (DC) ICDs are a battery-operated device placed under the skin with a one (SC) or two (DC) wires connected to the heart to track the heart rate. When an abnormal heart rhythm is detected, the ICD delivers an electric shock to restore a normal heart rate. A CRT-D differs to an ICD in that it also sends signals to the lower chambers of the heart to help synchronize the heartbeat and improve cardiac function.
We will use anonymous data from GP and hospital records to quantify the occurrence of hospitalization and cardiovascular events, particularly heart attack and stroke, among people with SC ICD, DC ICD and CRT-D devices. Furthermore, we will compare the ability of SC ICD and DC ICD to diagnose atrial fibrillation, with a focus on any difference in the time to diagnosis. We will also compare the effectiveness of DC ICD or CRT-D in preventing heart failure related hospitalizations.
Evaluation of hospitalizations and cardiovascular events in people with SC ICD, DC ICD and CRT-D devices will help guide patient management and improve device implantation choices.
CRT-D or ICD devices provide effective clinical management of major cardiac arrhythmias, including among people with heart failure. However, the differential benefit between the devices has been largely unexplored using real-world data.
Our overall aim is to investigate the effect CRT-D and ICD devices have on hospitalization rates and cardiovascular events. We will use linked data for GP practices, hospital attendances and death registrations to conduct two cohort studies. Our first study will include all patients with a SC ICD or DC ICD. We will describe the incidence of atrial fibrillation, major adverse cardiovascular events (MACE), all-cause mortality and hospitalization activity for those with a SC ICD or DC ICD with a focus of outcomes at 12 and 36 months after device implantation. We will also compare the time to atrial fibrillation diagnosis between people with a SC ICD and a DC ICD using Cox proportional hazard regression models. Our second study will include heart failure patients who are subsequently fitted with a DC ICD or CRT-D. Again, we will describe the incidence of MACE, all-cause mortality and hospitalization activity, but here with a focus on heart failure-related attendances, at 12 and 36 months. We will also compare the risk of non-elective in those with a DC ICD and CRT-D using Cox proportional hazard regression models.
Analysis 1 & 2
• Major adverse cardiovascular events (MACE). Primary outcome: composite MACE defined as MI, stroke and CVD death. Secondary outcome: separately (a) MI, (b) stroke and (c) CVD death. MI and stroke will be identified using CPRD Aurum (see codelists in appendix) and HES APC and OP (MI = ICD-10 I21, I22 or I23 or stoke (ICD-10 = I60-I64) and deaths will additionally be identified using ONS death registrations where the primary cause is from the I chapter (Analysis 1 Objective 3 & 7 and Analysis 2 Objective 4).
• All-cause mortality to be identified using ONS death record (Analysis 1 Objective 4 and Analysis 2 Objective 6)
• Hospitalization activity to be identified using HES APC, OP and EC and categorised as admission, outpatients and emergency, respectively (Analysis 1 Objective 5 and Analysis 2 Objective 3 & 7). We will define a hospital admission as a hospital provider spell. A hospital provider spell is the total continuous stay of a patient using a hospital bed on premises controlled by a health care provider during which medical care is the responsibility of one or more consultants, or the patient is receiving care under one or more nursing episodes in a ward. A hospital provider spell starts with a hospital provider admission and ends with a hospital provider discharge. The total number of hospital provider spells in a year are those spells in which patients have been discharged within that year, regardless of the year of admission. Hospitalization activity will be classed as all-cause (any), CVD (using primary diagnosis field and the IX chapter of ICD-10 [I codes]), or in Analysis 2 as HF-related (defined below)
Analysis 1 only:
• Atrial fibrillation (AF) to be identified using CPRD Aurum (see codelists in appendix) and HES (ICD-10 code I48) (Objective 1 & 6)
Analysis 2 only:
• HF admission to be identified using HES APC (Objective 1 & 7-9). Classified as elective or non-elective. We will include HF or respiratory condition where HF is a likely contributory using ICD-10 codes = I110, I130, I32, K761, I50, J81, J00-J47, J80-84, J85-86, J90, J91, J96, J98, J99.
Jennifer Davidson - Chief Investigator - Health iQ Ltd ( UK ) t/a CorEvitas
Jennifer Davidson - Corresponding Applicant - Health iQ Ltd ( UK ) t/a CorEvitas
Archie Farrer - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Boglarka Kovacs - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Gulsah Akin Unal - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Helen Mebrahtu - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Judith Ruzangi - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Mico Hamlyn - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Seth Jarvis - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Simon Wan Yau Ming - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
Gulsah Akin Unal - Collaborator - Health iQ Ltd ( UK ) t/a CorEvitas
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation