Conference Coverage

Ventricular assist devices linked to sepsis

View on the News

Comment by Daniel Ouellette, MD, FCCP

Dr. Daniel Oullette

The young, recently graduated clinic nurse came to see me before I went in to see my patient. "I can't register a pulse or blood pressure", she said, assuming that her failure to do so indicated a lack of clinical acumen on her part. "You won't", I told her. I took her with me into the exam room and watched her listen with her stethoscope in amazement to the mechanical hum of the LVAD device over the patient's chest.

Modern technology saves our patients' lives, but there is always another side to the coin. Reports that LVAD devices are associated with a high incidence of bloodstream infections is important for future clinical practice. The fact that the causes and risk factors for these infections are unknown make this phenomena one of high interest.



– Back in 2008, there was only one case.

“Bloodstream infection is a serious infection, so I thought, ‘Let’s see what’s happening,’ ” explained Ira Das, MD, a consultant microbiologist at Queen Elizabeth Hospital.

Ira Das, FRCPath, a Consultant Microbiologist at Queen Elizabeth Hospital in Birmingham, U.K. Damian McNamara/Frontline Medical News

Dr. Ira Das

A total of 129 ventricular assist devices (VADs) were implanted in 118 patients between January 2008 and December 2016 at this institution. The researchers found 23 patients with a VAD (19.5%) had developed a microbiology-confirmed bloodstream infection. The 25 cases of sepsis in this group included two patients who each experienced episodes associated with different microorganisms.

Coagulase-negative staphylococci were the most common cause, present in 32% of the 25 cases. Sepsis was caused by Enterococcus faecium in 12%, Candida parapsilosis in 8%, and Staphylococcus aureus in 2%. Another 4% were either Enterococcus faecalis, Serratia marcescens, Pseudomonas aeruginosa, C. guilliermondii, or C. orthopsilosis. The remaining 16% of bloodstream infections were polymicrobial.

Less certain was the source of these infections.

“In the majority of cases, we didn’t know where it was coming from,” Dr. Das said at the annual meeting of the American Society for Microbiology. In 6 of the 25 cases, VAD was confirmed to be the focus of infection, either through imaging or because a failing component of the explanted device was examined later. An intravascular catheter was the source in another 5 patients, and in 14 cases, the source remained a mystery.

“Some of these infections just might have been hard to see,” Dr. Das said. “If the infection is inside the device, it’s not always easy to visualize.”

The study supports earlier findings from a review article that points to a significant infection risk associated with the implantation of VADs (Expert Rev Med Devices. 2011 Sep;8[5]:627-34). That article’s authors noted, “Despite recent improvements in outcomes, device-related infections remain a significant complication of LVAD [left ventricular assist device] therapy.”

In a previous study of people with end-stage heart failure, other investigators noted that, “despite the substantial survival benefit, the morbidity and mortality associated with the use of the left ventricular assist device were considerable. In particular, infection and mechanical failure of the device were major factors in the 2-year survival rate of only 23%” (N Engl J Med. 2001 Nov 15;345[20]:1435-43).

Similarly, in the current study, mortality was higher among those with sepsis and a VAD. Mortality was 39% – including eight patients who died with a VAD in situ and one following cardiac transplantation. However, Dr. Das cautioned, “It’s a small number, and there are other factors that could have contributed. They all go on anticoagulants so they have bleeding tendencies, and many of the patients are in the ICU with multiorgan failure.”

Infection prevention remains paramount to minimize mortality and other adverse events associated with a patient’s having a VAD. “We have to make sure that infection control procedures and our treatments are up to the optimal standard,” Dr. Das said. “It’s not easy to remove the device.”

Of the 129 VADs implanted, 68 were long-term LVADs, 11 were short-term LVADs, 15 were right ventricular devices, and 35 were biventricular devices.

The study is ongoing. The data presented at the meeting were collected up until December 2016.

“Since then, I’ve seen two more cases, and – very interestingly – one was Haemophilus influenzae,” Dr. Das said. “The patient was on the device, he was at home, and he came in with bacteremia.” Again, the source of infection proved elusive. “With H. influenzae, you would think it was coming from his chest, but the chest x-ray was normal.”

The second case, a patient with a coagulase-negative staphylococci bloodstream infection, was scheduled for a PET scan at the time of Dr. Das’ presentation to try to identify the source of infection.

Dr. Das had no relevant disclosures.

Recommended Reading

Idarucizumab reversed dabigatran completely and rapidly in study
MDedge Cardiology
Optimal age for cardiovascular disease screening remains elusive
MDedge Cardiology
Allopurinol and ventricular arrhythmias: Is there a link?
MDedge Cardiology
Counsel low-risk patients case by case on diet, exercise
MDedge Cardiology
Nearly half of patients who stop taking opioids for 6 months resume use later
MDedge Cardiology
ANCA-associated vasculitis appears to increase risk of stroke, death
MDedge Cardiology
Improving diet over time reduces mortality
MDedge Cardiology
Watchman device for AF patients ineligible for oral anticoagulation gains support from 1-year registry outcomes
MDedge Cardiology
Plasma biomarker distinguishes ARDS, acute heart failure
MDedge Cardiology
Study: No increased mortality with ACA-prompted readmission declines
MDedge Cardiology