Infections of coronary stents appear to be uncommon, but it is not clear if they are often missed, underreported, or truly rare, according to a new analysis.
In a search of multiple databases, 79 cases of coronary stent infections (CSI) were found in 65 published reports, according to Venkatakrishnan Ramakumar, MBBS, MD, department of cardiology, All India Institute of Medical Sciences, New Delhi.
Ted Bosworth/MDedge News
Dr. Venkatakrishnan Ramakumar
Over the period of evaluation, which had no defined starting point but stretched to November 2021, the 79 infections reported worldwide occurred when millions of percutaneous coronary intervention (PCI) procedures were performed. In the United States alone, the current estimated annual number of PCIs is 600,000, according to an article published in the Journal of the American Heart Association.
If the number of reported CSI cases represented even a modest fraction of those that occurred, the risk would still be almost negligible. Yet, Dr. Ramakumar insisted that there has been little attention paid to the potential for CSI, creating a situation in which many or almost all cases are simply being missed.
“We do not know how many infections have gone unrecognized,” Dr. Ramakumar said in presenting his results at the Cardiovascular Research Technologies conference, sponsored by MedStar Heart & Vascular Institute. And even if they are identified and promptly treated, there “is the potential for a publication bias,” he added, referring to the reluctance of investigators to submit and publishers to accept manuscripts with negative results.
Regardless of the frequency with which they occur, CSI is associated with bad outcomes, according to the data evaluated by Dr. Ramakumar. On the basis of in-hospital mortality, the primary endpoint of this analysis, the rate of death in patients developing CSI was 30.3%.
Successful treatment varied by hospital type
This risk was not uniform. Rather, rates of in-hospital mortality and proportion of patients treated successfully varied substantially by type of hospital. At private teaching hospitals for example, successful treatment – whether medical alone or followed by bailout surgery – was 80%. The rates fell to 40% at public teaching hospitals and then to 25% at private nonteaching hospitals.
The full-text articles included in this analysis were evaluated and selected by two reviewers working independently. A CSI diagnosis made clinically or with imaging and treatment outcomes were among criteria for the case studies to be included. Dr. Ramakumar said the study, which he claimed is the largest systematic review of CSI ever conducted, has been registered with PROSPERO, an international prospective registry of systematic reviews.
The presenting symptom was fever in 72% of cases and chest pain in the others, although there was one asymptomatic CSI reported. On angiography, 62% had a concomitant mycotic aneurysm. Intramyocardial abscess (13.9%), rupture (11.3%), and coronary fistula (7.5%) were also common findings, but no angiographic abnormalities could be identified in 53% of patients.
Following PCI, most CSI developed within 8 days (43%) or the first month (23%), but CSI was reported more than 6 months after the procedure in 19%. Complex PCI accounted for 51% of cases. Of stent types, 56% were drug eluting and 13% were bare metal.
When comparing characteristics of those who survived CSI with those who did not, most (89%) of those with a non–ST-segment elevated acute coronary syndrome ultimately survived, while survival from CSI in those with structural heart disease was only 17%.
Microbiological findings were not a criterion for study inclusion, but Staphylococcus species accounted for 65% of the infections for which positive cultures were reported. Pseudomonas accounted for 13%. Less than 4% (3.8%) tested positive for multiple pathogens. A small proportion of patients had unusual infectious organisms.
As part of this analysis, the investigators developed an artificial intelligence model to predict CSI based on patient characteristics and other variables. However, the specificity of only around 70% led Dr. Ramakumar to conclude that it does not yet have practical value.
However, he believes that better methodology to detect CSI is needed, and he proposed a diagnostic algorithm that he believes would both improve detection rates and accelerate the time to diagnosis.