The trusted source for
healthcare information and
Abstract & Commentary
By John P. DiMarco, MD, PhD
Source: Peterson PN, et al. Gender differences in procedure-related adverse events in patients receiving implantable cardioverter-defibrillator therapy. Circulation. 2009;119:1078-1084.
This report from the implantable cardioverter defibrillator (ICD) portion of the National Cardiovascular Data Registry focuses on the relationships of gender to in-hospital, ICD-associated adverse events. The National ICD Registry (NICDR) was established in 2006 after the Centers for Medicare & Medicaid Services required that all ICD implants in Medicare recipients for primary prevention of sudden cardiac death be included in a comprehensive registry. Most hospitals report all ICD implants to the NICDR, including implants for both primary and secondary prevention indications and implants in non-Medicare recipients. The NICDR collects baseline data about each patient at the time of implant. After hospital discharge, trained reviewers analyze the hospital chart for in-hospital complications. Major adverse events prospectively defined in the NICDR include: cardiac arrest, cardiac perforation, cardiac valve injury, coronary venous dissection, hemothorax or pneumothorax, deep phlebitis, transient ischemic attack, stroke, myocardial infarction, pericardial tamponade, and AV fistula. Other adverse events of interest include: drug reaction, new conduction block, pocket hematoma, lead dislodgement, peripheral embolus, and device-related infection. Hospital length of stay is also collected.
In this report, data from 161,407 patients undergoing a first ICD implantation between January 2006 and December 2007 were analyzed. Repeat procedures for generator changes were excluded. Of these patients, 73% were men and 27% were women. The mean age was similar between men and women (67.9 vs. 67.0 years). There were also no differences in the proportion of men and women who received their ICD for a primary prevention indication (72% vs. 72.2%) or in reported left-ventricular ejection fraction (27.1% vs. 27.6%). Women, however, were more likely to be Black (10.9% vs. 17.5%), to have New York Heart Association class III or IV congestive heart failure (50.1% vs. 57.2%), and to have a nonischemic cardiomyopathy (27.2% vs. 44.3%). More women than men received biventricular ICDs (39.1% vs. 34.1%).
In the entire cohort, in-hospital adverse events were noted in 3.6%. The adverse event rate was higher in women than in men (4.4% vs. 3.3%, p < 0.001). Adverse events that were more commonly seen in women were typically procedure-related technical issues, including pneumothorax and hemothorax, coronary venous dissection, pericardial tamponade, perforation, and lead dislodgement. There was no difference between men and women in in-hospital mortality. Hospital length-of-stay was longer for patients who had a reported complication (7.1 days) compared to those without a complication (3.8 days, p < 0.001). After adjustment for various factors, women had a significantly higher odds ratio (OR) for experiencing any adverse event (OR 1.32) or a major adverse event (OR 1.71) compared to men.
Peterson et al then discussed possible reasons why women are more likely to have complications with ICD implantation than men. In the NICDR, there was no difference in age or most comorbidities between men and women. Since the increased frequency of adverse events was primarily due to technical issues, they postulate that body size may be a factor. Unfortunately, the NICDR does not collect data on body size or BMI, so this remains only a hypothesis.
Peterson et al conclude that data from the NICDR should be used to guide efforts to reduce the frequency of adverse events associated with ICD implantation in both men and women.
These data from the NICDR point out how valuable the Registry will be for analyzing the risks and benefits associated with ICD therapy. In the past, information about ICD complications was usually derived either from single-center surveys or from reports from randomized, clinical trials. However, in routine clinical practice, patients who receive ICDs tend to be older and may have more comorbidities than the patients seen in clinical trials. The size of the NICDR database and the fact that it includes over 90% of all ICD implants in the United States allows it to be an important tool for analyzing various aspects of ICD therapy.
It is important to recognize, however, that this paper deals with only in-hospital complications. Many infections, or some lead dislodgments, particularly coronary sinus leads, will occur after hospital discharge for elective procedure, and would not be captured. Other major complications of ICD therapy, such as inappropriate shocks and lead and generator failures, also usually occur later during the course of therapy. Thus, the total complication rate is likely to be substantially higher than the figure quoted here. At the same time, some of the adverse events listed, e.g., myocardial infarctions and strokes, even though they occurred during the same hospitalization, the ICD implants were probably due to the patient's underlying disease and were not strictly device related
The reasons for the gender-based differences in complications are not obvious. As an implanting physician, I agree that body habitus may make a difference in the complications that are related to vascular access and lead placement; these complications accounted for a large fraction of major gender-related differences reported in this paper. The observations reinforce the need for careful attention to technique when implanting ICDs in both men and women. The NICDR is a valuable resource toward this goal, and it should be strengthened to allow more long-term follow-up data to be collected and analyzed.