Chronic Kidney Disease and ICD Related Complications
Since 1996, all recipients of an ICD or CRT-D in the Leiden University Medical Centre (LUMC) were registered in the departmental cardiology information system (EPD-Vision, LUMC). Clinical characteristics at baseline, data on implant procedure and follow-up were noted in this system. For the current analysis, all patients receiving their first ICD or CRT-D between 1996 and 2012 for primary prevention (left ventricular ejection fraction <35%) or secondary prevention (survival of cardiac arrest or syncope due to suspected ventricular arrhythmia) were included. Device replacements were included only if the patients received their initial ICD between 1996 and 2012.
Patient groups were created in accordance with the National Kidney Foundation classification system using the Modification of Diet in Renal Disease (MDRD) formula. A GFR >90 mL/min/1.73 m was defined as normal kidney function, GFR 30–90 mL/min/1.73 m as mild to moderate CKD and a GFR <30 mL/min/1.73 m as advanced CKD. Patients without available GFR measurements around the time of device implant were excluded from this study.
All ICD recipients were followed-up 2 months postimplant and every 3–6 months thereafter. Patients were encouraged to visit the outpatient clinic for wound examination in any case of hematoma or when in doubt regarding the wound healing process. Furthermore, patients could contact the department in case of questions at any time. Data on follow-up were collected until July 1, 2012, until a new implantation if this took place in 2012, or until the moment patients were lost to follow-up or died.
All defibrillator systems used were implanted transvenously and without thoracotomy. Implanted systems were manufactured by Biotronik (Berlin, Germany), Boston Scientific (Natick, MA, USA), Medtronic (Minneapolis, MN, USA), or St. Jude Medical/Ventritex (St. Paul, MN, USA). In the CRT-D devices, left ventricular (LV)-lead placement was performed after obtaining a coronary sinus venogram. Subsequently, the LV lead was inserted and positioned using an 8F guiding catheter. LV lead positioning was preferred in a lateral or posterolateral vein. The right atrial and right ventricular leads were positioned conventionally.
In patients using oral anticoagulants, periprocedural target international normalized ratio (INR) for patients without mechanical valves was 1.5–2.0, and in case of a mechanical valve, target INR was 2.0–2.5. Flucloxacillin was administered immediately before device implant and 4 hours after the procedure. In case of generator replacement without lead replacement, antibiotics were only administered before device implant.
Ischemic heart disease was defined as the presence of significant coronary artery disease (a diameter of stenosis of at least 50% in at least 1 coronary artery). Hypertension was defined as a documented history for hypertension or a repeatedly measured office blood pressure of >140/90 mmHg.
All adverse events as registered in EPD-Vision were analyzed retrospectively by M.S.B. and M.K.B. Adverse events were categorized as pneumothorax, pocket hematoma, lead related complications, or device infection. Pneumothorax was defined as all cases of pneumothorax present on routine chest X-ray after device implantation. Hematoma was defined as all cases of hematoma severe enough to be documented in the patient file or requiring surgical intervention after device implantation. Since most hematomas occur immediately after device implant, hematoma was likely to be discovered during routine postoperative wound control or at the routine appointment 2 months postimplant. Lead related complications were defined as all cases of lead failure or lead dislocation requiring the implantation or relocation of an atrial, LV of right ventricular (RV) lead. Device infection was defined as all cases of device explantation due to suspected infection regardless of the presence of a positive lead or device culture.
For the initial data presentation, patient groups were created according to kidney function (GFR > 90, GFR 30–90, GFR < 30 mL/min/1.73 m). Continuous data were described by their mean and SD, and analyzed using an independent Student's t-test. Categorical data were described as proportions (percentage) and compared using a χ-test for independence. For the analysis of differences between the GFR groups, the patients with normal kidney function served as controls.
For the different complications, various models were built to take into account that each complication has different characteristics. The inclusion of variables in these models was determined by univariate analysis (P < 0.1), biological considerations, and potential confounding effects.
Hematoma may occur after every device procedure, so a generalized estimating equations model was used. In this model, every implantation is a new risk factor for hematoma, and the fact that patients sometimes receive several consecutive devices is taken into account.
Pneumothorax occurs directly after first implant (or after lead replacement), so a standard binary logistic regression analysis was used.
Lead problems can occur at any given moment during follow-up, independent of the implantation of a new device; therefore, a standard Cox regression analysis was used to investigate the time to first lead problem.
The occurrence of first infection was analyzed using Cox regression analysis with second and third device implant as a time-dependent covariate. This was done to adjust for the increased risk for infection attributed by device replacement.
All statistical analyses were performed using SPSS (version 20.0, SPPS Inc., Chicago, IL, USA).
Methods
Patient Population
Since 1996, all recipients of an ICD or CRT-D in the Leiden University Medical Centre (LUMC) were registered in the departmental cardiology information system (EPD-Vision, LUMC). Clinical characteristics at baseline, data on implant procedure and follow-up were noted in this system. For the current analysis, all patients receiving their first ICD or CRT-D between 1996 and 2012 for primary prevention (left ventricular ejection fraction <35%) or secondary prevention (survival of cardiac arrest or syncope due to suspected ventricular arrhythmia) were included. Device replacements were included only if the patients received their initial ICD between 1996 and 2012.
Patient groups were created in accordance with the National Kidney Foundation classification system using the Modification of Diet in Renal Disease (MDRD) formula. A GFR >90 mL/min/1.73 m was defined as normal kidney function, GFR 30–90 mL/min/1.73 m as mild to moderate CKD and a GFR <30 mL/min/1.73 m as advanced CKD. Patients without available GFR measurements around the time of device implant were excluded from this study.
Follow-up
All ICD recipients were followed-up 2 months postimplant and every 3–6 months thereafter. Patients were encouraged to visit the outpatient clinic for wound examination in any case of hematoma or when in doubt regarding the wound healing process. Furthermore, patients could contact the department in case of questions at any time. Data on follow-up were collected until July 1, 2012, until a new implantation if this took place in 2012, or until the moment patients were lost to follow-up or died.
Device Implantation and Settings
All defibrillator systems used were implanted transvenously and without thoracotomy. Implanted systems were manufactured by Biotronik (Berlin, Germany), Boston Scientific (Natick, MA, USA), Medtronic (Minneapolis, MN, USA), or St. Jude Medical/Ventritex (St. Paul, MN, USA). In the CRT-D devices, left ventricular (LV)-lead placement was performed after obtaining a coronary sinus venogram. Subsequently, the LV lead was inserted and positioned using an 8F guiding catheter. LV lead positioning was preferred in a lateral or posterolateral vein. The right atrial and right ventricular leads were positioned conventionally.
In patients using oral anticoagulants, periprocedural target international normalized ratio (INR) for patients without mechanical valves was 1.5–2.0, and in case of a mechanical valve, target INR was 2.0–2.5. Flucloxacillin was administered immediately before device implant and 4 hours after the procedure. In case of generator replacement without lead replacement, antibiotics were only administered before device implant.
Definition of Variables
Ischemic heart disease was defined as the presence of significant coronary artery disease (a diameter of stenosis of at least 50% in at least 1 coronary artery). Hypertension was defined as a documented history for hypertension or a repeatedly measured office blood pressure of >140/90 mmHg.
Complications
All adverse events as registered in EPD-Vision were analyzed retrospectively by M.S.B. and M.K.B. Adverse events were categorized as pneumothorax, pocket hematoma, lead related complications, or device infection. Pneumothorax was defined as all cases of pneumothorax present on routine chest X-ray after device implantation. Hematoma was defined as all cases of hematoma severe enough to be documented in the patient file or requiring surgical intervention after device implantation. Since most hematomas occur immediately after device implant, hematoma was likely to be discovered during routine postoperative wound control or at the routine appointment 2 months postimplant. Lead related complications were defined as all cases of lead failure or lead dislocation requiring the implantation or relocation of an atrial, LV of right ventricular (RV) lead. Device infection was defined as all cases of device explantation due to suspected infection regardless of the presence of a positive lead or device culture.
Statistical Analysis
For the initial data presentation, patient groups were created according to kidney function (GFR > 90, GFR 30–90, GFR < 30 mL/min/1.73 m). Continuous data were described by their mean and SD, and analyzed using an independent Student's t-test. Categorical data were described as proportions (percentage) and compared using a χ-test for independence. For the analysis of differences between the GFR groups, the patients with normal kidney function served as controls.
For the different complications, various models were built to take into account that each complication has different characteristics. The inclusion of variables in these models was determined by univariate analysis (P < 0.1), biological considerations, and potential confounding effects.
Hematoma may occur after every device procedure, so a generalized estimating equations model was used. In this model, every implantation is a new risk factor for hematoma, and the fact that patients sometimes receive several consecutive devices is taken into account.
Pneumothorax occurs directly after first implant (or after lead replacement), so a standard binary logistic regression analysis was used.
Lead problems can occur at any given moment during follow-up, independent of the implantation of a new device; therefore, a standard Cox regression analysis was used to investigate the time to first lead problem.
The occurrence of first infection was analyzed using Cox regression analysis with second and third device implant as a time-dependent covariate. This was done to adjust for the increased risk for infection attributed by device replacement.
All statistical analyses were performed using SPSS (version 20.0, SPPS Inc., Chicago, IL, USA).
Source...