The trusted source for
healthcare information and
A group of investigators at Columbia-Presbyterian Medical Center in New York City that is examining the relationship between infection and immune responses involved with left-ventricular assist devices (LVADs) proposes both are specific to the implanted devices. The investigators did not focus on where the devices were surgically implanted and whether their drive lines — the tubing that drives the pump— jut to the side (three or nine o’clock) or proceed straight downward (six o’clock). (See related article on LVADs, p. 115.)
But O.H. Frazier, MD, at the Texas Heart Institute in Houston, says the site where LVADs are placed has much to do with whether they get infected. The problem, he says, is often with the way the pumps are placed surgically, not with the pumps themselves.
"The Lancet article (1999; 354:550-555) is a bit misleading," he says. "I think it contains a lot of soft data. The study has to do with surgical contamination and violation of certain surgical principles." says Frazier, chief of cardiopulmonary transplantation, co-director of the Cullen Cardiovascular Research Laboratories, and director of Surgical Research at the Texas Heart Institute. "I don’t want to be overly critical of the study, but no infection to speak of was seen in the initial group of patients receiving LVADs; it just was seen with extraperitoneal placement of the pumps. [The authors’] conclusions, that there are any cellular immunity implications of the pump itself, are soft. There are implications from chronic infection that can result from the chronic contamination that result from the extraperitoneal placement of the pump."
When Frazier first started working with LVADs, he placed the devices interabdominally. He says he has gone back to placing them in that position, as opposed to extraperitoneally, and you "nearly never see infection." He says the article discounts the relation of surgical technique to infection. (The study group is placing the devices extraperitoneally behind the posterior rectus sheath.) "When we first started working with these devices, we didn’t see infection, and we had the pumps in as long as the authors did. Whenever you have trauma in an area of contaminated blood, you get infection. The investigators got bacterial infections that were overtreated, and they got secondary candida with the antibiotics."
Frazier started working with an interabdominal LVAD in the 1970s. The device was envisioned then as a long-term permanent pump, not as a bridge to transplant. He and others wrote the original 1976 Request For Application to the Food and Drug Administration of the National Heart, Lung, and Blood Institute for a permanent, long-term implantable LVAD. "That’s where the pumps manufactured by Thermo Cardiosystems Inc. [TCI] and Novacor came from," he says. "I worked on the TCI pump [now called HeartMate] since I was a medical student with Dr. DeBakey."
The initial pump his team worked on in the 70s was an interabdominal pump. Then the team thought it might fit outside the abdomen, and their physicians placed two in that position in 1986. Both developed hematomas and got infected. The patients’ blood had to be thinned in order to put them on the heart-lung machine, and they bled. "The pump worked all right, but there was hematoma for that reason," he says, "so after the first two, we didn’t put any more outside the abdomen."
TCI’s device was the first pump designed as a bridge to transplant approved by the FDA in 1992. Frazier published the study upon which the FDA approval was based.1 "There were no infections in those patients," he says. "I put the pump inside the abdomen, and the drive line came out at six o’clock." He says the drive line was stiff, and he pulled it from outside the abdomen, then straight down the left side, and out. Since the pump itself was inside the abdomen, it was protected from infection.
"When I took those patients to transplant, it was very hard to cut the drive line out," says Frazier. "I could literally lift the patient off the table; the drive line was so well adhered. I had absolutely no infections."
The Novacor pump was placed by a physician at Stanford around that same time. "The Novacor pump is a much longer, bigger, heavier pump," Frazier says. "Initially it was placed in the abdominal cavity, but it was so long, it eroded into the transverse colon. The surgeon had to do a colostomy on that patient. From that point on, the Novacor LVAD was placed extraperitoneally. It was too long to place inside the belly." The surgeons developed a technique of making a long space for it extraperitoneally. The pump took up too much of the left side of the abdomen, and its drive line couldn’t exit on the left side. "The drive line came out on the right side," he says. "It was tunneled to a long distance of tissue from the body of the pump at 3 o’clock."
As the TCI pump began expanding to more and more uses, surgeons found it much easier to put it in the extraperitoneal space, rather than in the abdomen. The drive line had to be brought out at 6 o’clock because it was too stiff to turn when placed in an extraperitoneal position. It came directly out of the bottom of the pump. "The drive line went outside the body very close to where the pump was, about two inches away. So you could only place it on the left side," he says. "As a result, blood collected and got infected. They all got infected. The explanation for the infection is very logical: Trauma and contamination equals infection."
Novacor has managed to bring the drive line out through longer tunnels on the right side, and as a result, Novacor has seen fewer infections, says Frazier.
A third device made by Thoratec was designed to be used as a temporary, intermediate pump and is placed outside the body. "But it has the advantage that it can be used for small patients," says Frazier. "There are size limitations for the pumps that are placed inside the body. The Thoratec device has a low infection rate." (See photos of the devices, inserted in this issue.)
(Frazier has no affiliation with the companies mentioned in this article.
Silviu Itesci, MD, and Mehmet C. Oz, MD, both cardiothoracic surgeons at Columiba-Presbyterian Medical Center in New York City and two of the Lancet study’s authors, take issue with Frazier’s commentary. Itescu states, "I am not sure what the relevance of [LVAD placement] is to the infection rate, but I would emphasize the exposure of the device to circulating elements of blood as being the primary etiology of increased susceptibility to infection." Oz adds, "I may be Bud Frazier’s biggest fan, but he is incorrect. The TCI database revealed infections in both groups. How we define infections among centers varies, but over the long run, they are clearly the Achilles’ heel of this procedure.")
1. Frazier OH. Thermo Cardiosystems’ left ventricular assist device. Ann Thorac Surg 1992; 54(5):1,019-1,020.