Introduction: The annual report for the National Audit of Cardiac Rhythm Management (CRM) Devices describes procedural activity and complication rates; it is reported through the National Institute of Cardiovascular Outcomes Research (NICOR). NICOR sets a cut off of 15 months following implant to attribute device related complications. However, device related revisions can occur after 15 months and we sought to investigate these and compare them with the NICOR group.
Methods: Our devices database was searched for all device procedures during 2018, and all recorded complications. A comparison was made between patients within, and outside of the 15-month reporting period. Complications were classified as wound related, infection, pneumothorax, haematoma, lead displacement and ‘other’ lead complications such as parameter changes or integrity faults. The time from the first device implant to detection of complication was recorded, however some patients underwent further procedures within this period. Statistical analysis was performed with Chi-squared and t-test.
Results: 2,393 CRM device procedures were performed at our centre over 2018. A total of 174 (7.28%) complications were recorded, involving 169 patients. 115 (4.81%) complications were recorded across 110 patients, within the 15-month NICOR period, and 59 (2.47%) occurred after 15 months.
NICOR mean age was higher than the non-NICOR group (69.8 ± 16.5 versus 63.8 ± 19.0 years, p=0.04) but the proportion of males was similar (57.3 versus 64.4%, p=0.37). All-cause mortality was similar 5five (4.5%) versus 3 (5.1%), p=0.88. As the data was dichotomised by time from implant, mean time from first implant to recorded complication was shorter in the NICOR group compared with the non-NICOR group (1.48 ± 3.09 versus 9.89 ± 5.89 years, p<0.0001).
Within the NICOR group (n=115), lead displacements made up the largest proportion, 45 (39.12%), followed by ‘other’ lead complications 16 (13.91%), wound revisions 16 (13.91%), infection 13 (11.30%), pneumothorax 5 (4.35%), haematoma 9 (7.83%) and pericardial effusion 6 (5.22%). By contrast, the non-NICOR complications (n=59) were dominated by ‘other’ lead complications, 37 (61.71%) which primarily related to integrity issues and parameter changes. There was 1 lead displacement beyond 15 months (1.69%), 8 (13.56%) wound revisions and 12 (20.34%) very late infections. There were no reported pneumothoraxes, haematomas or pericardial effusions after 15 months.
Discussion: A considerable proportion of complications requiring intervention present many years after the initial implant, and certainly outside of the NICOR reporting window. Although reported locally, this may reflect a significant burden of morbidity related to CRM devices that goes unreported at a national level.
Conclusion: Late device related complications falling outside of the NICOR reporting window are an important source of patient comorbidity and they deserve our attention and further investigation.