In early publications, bacteremia was reported to occur in 4% to 18% of patients undergoing cardiac catheterization (1,2). However, in these studies, blood for culture was obtained from the intravascular catheter or the vessel from which the catheter was removed. Sande et al. (3) determined the true frequency of bacteremia during and after cardiac catheterization and the frequency of fever by obtaining 214 blood cultures from 106 patients from a vein in the arm on the side opposite to that of the catheter. All venous samples were sterile; therefore, no bacteremia could be demonstrated during cardiac catheterization.

The mortality and morbidity associated with cardiac catheterization were analyzed over a period of 9 years (1971-1979) by Gwost et al. (4). No infection and only two pyrogenic reactions occurred in 1,771 patients. Furthermore, there were only three cases of bacteremia or bacterial endocarditis after 12,367 catheterization procedures reported in an early cooperative study (5). Ricci et al. (6) found only five documented infections after review of 7,690 medical records of cardiac catheterizations over a 40-month period.

Between 1980 and 1988, 12,251 arterial punctures for cardiac catheterization, PTCA, or pure diagnostic intraarterial angiography (IAA) were performed by Würsten et al. (7). The only infectious complications documented were prolonged healing of a wound in the groin and a severe graft infection necessitating ligation of the common femoral artery.

In a retrospective study from January 1991 to December 1998, Munoz et al. (8) found a bacteremia rate of 0.11% in 22,006 invasive nonsurgical cardiologic procedures (0.24% after PTCA, 0.6% after diagnostic cardiac catheterization, 0.8% after electrophysiologic studies). Multivariate analysis identified the presence of congestive heart failure (OR 21.0; CI 95% 6.8-66.0) and age older than 60 years (OR 1.9; CI 95% 1.9-6.3) as independent risk factors for a blood stream infection after nonsurgical cardiologic procedures.

The incidence of bacteremia and other infections associated with cardiac catheterization, therefore, seems to be very low. Even if synthetic vascular grafts have to be catheterized, the infection rate is very low. Mohr et al. (9) investigated 109 percutaneous catheterizations of synthetic vascular grafts in 89 patients to determine the risk of major complications. Ninety-six catheterizations were performed through the inguinal portion of the aortofemoral graft. There were no instances of graft infection, and only one superficial infection developed at a cutaneous puncture site, which did not involve the graft.

A large prospective study to assess the frequency of bacteremia due to cardiac catheterization and PCI was performed by Banai et al. (10) who examined patients undergoing a total of 960 interventional procedures. They analyzed blood cultures, withdrawn from the arterial sheath immediately after arterial puncture and at the end of the procedure. A third blood culture was withdrawn from a peripheral
vein 4 hours later. The incidence of positive blood cultures immediately after the procedure was 7.3% after diagnostic catheterization and 4.6% after PCI (balloon angioplasty and stent implantation). After 4 hours, positive blood cultures were noted in 3.9% and 4.1%, respectively. However, only four cases of bacteremia were considered to be clinically significant (1 × Staphylococcus aureus, 3 × Klebsiella spp.). Moreover, all of these cases were related most likely to an intravenous line and not to the cardiac procedure itself. The data of Banai et al. suggest that clinically significant bacteremia is a rare complication of cardiac catheterization and also of PCI (explaining the low incidence of coronary stent infection; see below).

A further prospective study (11) investigated the frequency of bacteremia after more complex PCI in 147 consecutive patients. Procedures included balloon angioplasty, cutting balloon angioplasty, rotational coronary atherectomy, and single and multiple stent implantations. All procedures were performed via the femoral route. Blood cultures were taken immediately at the end of the procedure from the side arm of the arterial sheath. A second blood culture was taken after 12 hours prior to sheath removal. 17.7% of the patients had bacteria isolated from the first blood culture with coagulase-negative staphylococci being the most common microorganisms (57.7%). After 12 hours, 12.0% of the patients yielded microorganisms in the blood culture; 70.5% of them were coagulase-negative staphylococci. Four of the 147 patients developed a temperature >38°C; all had negative blood cultures. No patient developed clinical evidence of septicemia, endarteritis, or endocarditis during hospital stay or after discharge. The authors suggested that the relatively high incidence of bacteremia was due to the complex nature of PCI. However, the statistical analysis did not show a difference in the number of devices used in those patients with and in those without bacteremia.

PTCA today is one of the most common procedures in nonsurgical invasive cardiology.

No infection was reported during early and long-term follow-up (at least 1 year) of 3,079 patients after coronary angioplasty in the early 1980s (12). In a prospective study, 164 PTCAs resulted in one S. aureus infection (0.6%) that could be related to the procedure (13).

Malanoski et al. (14) identified a risk of 0.25% for S. aureus bacteremia (SAB) among 1,944 PTCA procedures performed during 25 months at one institution. Cleveland and Gelfand (15) summarized the reported cases of invasive staphylococcal infections associated with PTCA and described three more patients with invasive staphylococcal disease after PTCA of which two patients had received single intravenous antibiotic prophylaxis with cefazolin. This may well be explained by the fact that the predominant risk of infection may not have been the procedure itself but more likely the retention of the femoral sheath for more than 24 hours.

McCready et al. (16) noted that septic complications after cardiac catheterization and PTCA are quite uncommon, but they described nine cases of septic complications after this procedure resulting in two deaths. Their study suggests that repeated puncture of the same femoral artery and the femoral artery sheath being left in for more than 1 day are risk factors for septic complications (15,16). Cardiac abscess after PTCA has also been described in a patient in whom a problematic and repeated procedure probably led to a direct colonization and subsequent infection of an intimal dissection of the right coronary artery (17).

Siddiqui and Lester (18) described a case of septic arthritis and bilateral endogenous endophthalmitis associated with PTCA. Several cases of septic endarteritis with S. aureus after PTCA have been described by different authors (19,20). In another case report, an epidural abscess occurred in a patient after PTCA. The explanation was that the residual arterial sheath, whose tip was near the aortic bifurcation, was injected with an infected bolus, thus facilitating infection through the lumbar radicular arteries (21). The absence of specific signs may easily cause a delay in recognizing the infection (18,21,22). As mentioned above, it has been considered that retention of the sheath for more than 24 hours could be a risk factor for infection. However, prospective studies are not available on this issue (15,23). Salinas et al. (24) reported a case of infective coronary aneurysms 7 days after a balloon angioplasty.

Endocarditis following PTCA is such a rare complication that there are merely individual case reports. Wang et al. (25) reported a case of Candida parapsilosis endocarditis. Barbetseas et al. (26) reported a case of a patient with infective endocarditis of a prosthetic valve in the aortic position after receiving PTCA. Shibata et al. (27) reported a case of a 73-year-old man who developed infectious endocarditis caused by S. aureus after PTCA. A postmortem examination revealed multiple myocardial microabscesses and myocardial infarction resulting from an embolic vegetation.

Although infections after PTCA are rare, some have resulted in death (28). Infections may become evident several weeks after the procedure (14). Even ultrasonography and computed tomography may fail to reveal vascular infection, and, when there is clinical suspicion of infection, it may be prudent to initiate early surgical exploration (29). Some authors recommend use of the contralateral inguinal site if PTCA is to follow a recent catheterization (15, 16 and 17,30,31), whereas others found no correlation between ipsilateral inguinal puncture and infectious complications (13). At present, this question cannot be answered, because none of the cited studies have the statistical power to be able to detect significant differences between contralateral and ipsilateral repuncture.

Risk factors for bacteremia and other infectious complications during cardiac catheterization, mainly during PTCA, are age older than 60 years, congestive heart failure, duration of procedure, number of catheterizations at the same site, difficult vascular access, and an arterial sheath in place for more than 1 day (8,16,32,33,34). The most common microorganisms that cause PTCA-related bacteremia are S. aureus, coagulase-negative staphylococci, and group B streptococci (32).

Cutting Balloon Angioplasty Kobeiter et al. (35) monitored the long-term results in 19 patients undergoing cutting balloon angioplasty over a period of 32 months. They observed no case of infection. The authors postulate a randomized trial to be necessary to confirm the favorable results.

Gunther et al. (36) described in 1993 the first case of lethal complications resulting from a myocardial abscess near the stent in the right coronary artery. A second case of fatal outcome resulting from stent infection with Pseudomonas aeruginosa, which led to infective mitral endocarditis and saccular aneurysm of the coronary artery, was described by Leroy et al. (37). Studies in a swine model suggest that metallic stents have the potential of becoming infected after bacterial challenge unlike arteries that have undergone angioplasty without stenting (38). Seven stent-artery complexes implanted in the iliac arteries of 14 swine were culture positive after an intravenous bacterial challenge with S. aureus, whereas only 1 of 14 arteries that underwent angioplasty were positive for S. aureus (p = .03) (38). The pathophysiology behind the stent infection is unknown; perhaps the stents or endothelial injury served as a nidus for bacterial adherence. Bouchart et al. (39) suggested that multiple repeat procedures through the same arterial sheath may increase the risk for bacterial infection of the coronary stent.

Dieter et al. (40) reported the case of a patient who developed an infected aortic aneurysm after placement of a coronary artery stent. They pointed out that infectious complications have been rare, but that the associated mortality is alarmingly high (41). The authors analyzed four cases of stent infections, two with P. aeruginosa and two with S. aureus as causative microorganisms. Despite aggressive measures (surgical removal of the infected stent and artery complex), three of the four patients died. The authors pointed out, that the clinician must be sensitive to fever, return of angina pectoris, and bacteremia in patients who have undergone coronary stent placement. With regard to the generally low infection rate, the prophylactic use of antibiotics is not recommended.

Recently also Schoenkerman et al. (42) described stent infections as a rare, but dramatic sequela of coronary stent implantation. They reported three cases of infections, two with mycotic aneurysms and one with purulent pericarditis. Two of these three cases were associated with drug-eluting stents (DES). In all cases, S. aureus was the causative microorganism. The authors suggested the possibility that recent infection with S. aureus within 16 days prior to PCI may be an additional risk factor for the complication of coronary stent infection.

Kaufmann et al. (43) reviewed all published cases of coronary stent infections, that is, all patients who presented with symptoms of infection within the first 4 weeks after PCI. They stated that, although the implantation of medical devices in general represents one of the most important risk factors for healthcare-associated infections, the reports of coronary stent infections are exceedingly rare. The authors summarized 10 case reports. One patient received a DES; all other patients had a bare metal stent (BMS) implanted. In seven patients, S. aureus was the causative microorganism, and in four the stent was completely or partially removed. Three patients died despite initiation of antibiotic and surgical treatment. The very low number of published cases suggests that coronary stent infections represent a rather uncommon complication of PCI. But the exact incidence of infection is unknown.

According to Kaufmann et al. (43), fever is the hallmark of coronary stent infection. Chest pain is present in only half of the patients. They concluded that blood cultures should be taken in all patients presenting with fever within the first (four) weeks after stent implantation even in the absence of chest pain, ECG abnormalities, or elevation of cardiac enzymes.

Infections in Drug-Eluting Stents Kaufmann et al. (43) speculated that DES, introduced in the first decade, may predispose more to infection than BMS because of their immunomodulating and antiproliferative effects. They stated, however, that this seems not to be the case. There are only a few documented cases of stent infection after DES insertion. Recently, Lee et al. (44) found in their retrospective study on 1,023 consecutive patients who underwent PCI with DES a mortality of 9.4%. Of the 96 patients who died, no one died from PCI-related infection.

Aoki et al. (45) reviewed all published case reports of coronary artery aneurysms after DES implantation since 2004 and compared them with published case reports of aneurysms after BMS implantation. They stated that the incidence of coronary artery aneurysms after DES implantation is low within the first 9 months (aneurysms have been reported from 3 days to up to 4 years) with an incidence of 0.2% to 2.3%. This is a rate similar to the aneurysm rate after BMS implantation (0.3% to 3.9%). The clinical course is variable ranging from natural resolution to lifethreatening complications.

Gonda et al. (46) reported the unique case of a DES infection with MRSA in a 75-year-old male 11 months after the procedure. In the cardiac magnetic resonance angiography, they found a nonenhancing fluid collection surrounding the DES, suggesting the presence of an abscess, which was confirmed in the following surgical intervention.

In an early study published in 1978, the overall complication rate for 100 consecutive patients treated with the IBP was 23%, with five patients developing surgical site infections and two developing septicemia (47). Surgical site problems contributed to the death of one patient. Another patient died with septicemia and an infected aortotomy closure site 4.5 months after the original procedure.

Goldberg et al. (48) compared the percutaneous and surgical techniques of IBP insertion in 101 patients. In the percutaneous group (51 patients), no infection developed, but, in the surgical group, three patients developed sepsis with bacteremia (including one patient who required vein patch repair of the femoral artery and one patient who developed a surgical site infection requiring debridement).

An outbreak of Pseudomonas cepacia bacteremia associated with a contaminated water reservoir of an IBP was reported by Rutala et al. (49).

Forty-five patients who died after insertion of an intraaortic balloon device were studied at necropsy (50). Thirty-six percent had one or more complications related to the use of the device, one of which was a local surgical site infection not suspected during life. In two other patients in whom the balloon was implanted for septic shock, there was no evidence of bacterial seeding of either the balloon catheter or the prosthetic introducer graft. In a study of 240 consecutive percutaneous intraaortic balloon counterpulsations, Eltchaninoff et al. (51) identified
only one case of S. aureus bloodstream infection and one superficial infection. In the retrospective study by Meco et al. (52), 7 of 116 patients (6%) requiring postoperative IBP support had infection of the insertion site.

Yang et al. (53) investigated 112 used intraaortic balloons for physical integrity, gas leakage, mechanical performance, surface chemistry and morphology, and physical stability. These devices were all used clinically only once, and the duration of use in vivo ranged from 6 to 312 hours. Macroscopic examination of the balloons and the outer catheters revealed no obvious change in either shape or color. No discernible abrasions or cracks were observed. However, 61% of the balloons were creased and 40% of the central lumina and 21% of the sheaths showed visible bending flaws. Moreover, 65% of the balloons and 38% of the central lumina were contaminated by visible residual organic debris. The authors concluded that the presence of residual organic debris that cannot be eliminated is an indication that such intraaortic balloons should not be reused.

In their prospective study, Crystal et al. (54) found an incidence of fever of 47%, true bacteremia of 15%, and sepsis of 12% in 60 patients treated with an intraaortic balloon counterpulsation pump. The authors suggested evaluating the benefit of antibiotic prophylaxis. However, no studies addressing this issue have been published so far.

Laser thermal-assisted balloon angioplasty is used in the treatment of patients with advanced peripheral vascular disease and in high-risk patients who are poor candidates for operative reconstructions. White et al. (55) followed 28 patients, including 27 who had advanced peripheral vascular disease, for 3 years after laser thermal-assisted balloon angioplasty. Eighteen patients were successfully recanalized, but five amputations were required within 1 month and another six were needed between 8 and 12 months. Early amputations were necessitated by failure of wound healing. Whether this was due to infection, however, was not mentioned in the report.