title banner

OR in the News

OR in the News (selected articles)

Anesthesiology News, December, 2006

December 3, 2006 By Adam Marcus

Does the type of anesthesia administered to a woman during surgery to remove breast cancer affect the odds that her disease will return or spread?

As tenuous as that link might first appear, a new study by U.S. and European researchers suggests just that: Women who undergo regional and general anesthesia during their tumor operations face one-quarter the risk for local or metastatic recurrence of women given only general anesthesia during the procedure and opioids afterwards.

The researchers stress that their study, reported in the journal Anesthesiology (2006;105:660-664), is retrospective and preliminary and must be confirmed in a prospective trial. However, the results echo animal studies that support the hypothesis that general anesthesia and opioid drugs to prevent post-surgery pain create an environment that, albeit short-lived, encourages the growth and spread of cancer cells.

Daniel Sessler, MD, Chair of Outcomes Research at the Cleveland Clinic in Ohio, and senior author of the study, was cautious about exaggerating the importance of his group’s findings. “Retrospective studies are highly suspect, and that’s why I’m not going around saying this is true,” Dr. Sessler said. “However, it is nice preliminary data that supports the need for a major outcome trial.” He and his colleagues are now organizing such a study, which will require “thousands” of patients at multiple sites.

Dr. Sessler and other experts believe the results—”strange as they might seem,” he added—are well rooted in experimental and anecdotal evidence. Studies conducted in animals and humans suggest that the type of anesthesia used during surgery, the operation itself and opioid analgesia all sap the immune system.

“What determines whether cancer surgery is successful is largely a function of host defense—that is, whether the patient’s immune system can deal with the little bit of tumor that’s left and kill it off,” Dr. Sessler said. “If we do things that impair immune function exactly at that time, it seems likely to diminish the chances of cleaning up the residual tumor.”

Stunned Defenses Give Tumor Cells A Second Chance

Breast cancer is the most common form of cancer among women and the second leading cause of cancer-related death, next to lung cancer, in the United States. The vast majority of women with the disease undergo surgery to remove their tumor or tumors, as well as part or all of the affected breast. But surgeons cannot cut away every trace of cancer, so small numbers of tumor cells remain in the operating site that eventually can seed either local or distant recurrences. These cells are called micrometastases or minimal residual disease (MRD).

Ideally, the body’s cellular and humeral immune system will detect and destroy residual disease before it can erupt. As a stopgap, though, women also frequently receive chemotherapy and radiation after tumor surgery.

At the heart of the new study is a connection, supported by a growing body of evidence, between anesthesia, opioid pain relief and the immune system’s response to cancer surgery.

“Cancer surgery as done normally is bad for the immune system because the surgery itself, the anesthesia itself and the opioids used to control pain all impair natural killer cell function,” Dr. Sessler said.

Although ranking the insults is difficult, surgery may deliver the single biggest blow to the immune system. The body responds to the stress of the procedure by impeding the performance of natural killer cells. It also suppresses chemicals that deprive tumors of their blood supply and boosts the production of molecules, including vascular endothelial growth factor (VEGF), that stimulate their growth.

Animal studies also have suggested that anesthesia weakens the activity of natural killer cells, which lead the attack against cancerous cells and pathogens. Compared to general anesthesia, regional techniques may allow more killer cells to proliferate, raising the chances that these cells will be able to catch any leftover cancer cells.

Similarly, morphine and other opioids have been shown to suppress the immune system and, in particular, natural killer cells. Regional anesthesia during surgery can reduce the need for postoperative opioids, which may in turn leave the immune system stronger.

Much of the research in this area has come from the laboratory of Shamgar Ben-Eliyahu, PhD, Professor of Physiology and Psychology at Tel Aviv University, in Israel. Over the past decade, Dr. Ben-Eliyahu, an expert in neuroimmunomodulation, has published a series of studies, many funded by the U.S. National Institutes of Health, illuminating the effects of surgery, anesthesia and analgesia on human and animal tissues.

In what he called a “crucial” article (Anesthesiology 2001;94:1066-1073), Dr. Ben-Eliyahu and colleagues showed that rats given a spinal blockade—which reduces the neuroendocrine stress response to surgery—in addition to general anesthesia developed 70% fewer lung metastases after laparotomy than animals that received general anesthesia alone.

Then, in a study published in Neuroimmunomodulation (2004;11:255-260), Dr. Ben-Eliyahu and fellow investigators demonstrated that the opioid fentanyl inhibited the activity of natural killer cells in rats inoculated with tumor cells and increased the rodents’ risk for metastatic cancer. The tumor burden rose with the dose of fentanyl the animals received.

Dr. Ben-Eliyahu said the newest research adds an important clinical piece to the puzzle he has been completing. Led by Aristomenis K. Exadaktylos, MD, an anesthesia research fellow at Mater Misericordiae University Hospital in Dublin, Ireland, the researchers compared rates of recurrence and metastasis in 129 women who had undergone breast cancer surgery at that institution between September 2001 and December 2002. Of those, 50 received paraverterbral anesthesia—consisting of a 0.2-mL/kg bolus of 0.25% levobupivicaine—before induction of general anesthesia. For the general regimen, anesthesia was induced with 0.5 mcg/kg of fentanyl and 1.5-3.0 mg/kg of propofol (Diprivan, AstraZeneca), after which the women were given a laryngeal mask airway through which was administered 2% to 3% sevoflurane in nitrous oxide and oxygen.

Every woman also was given a 100-mg suppository of diclofenac after induction but before surgery, along with 0.05-mg/kg boluses of morphine during the operation at the discretion of the anesthetist.

Over the next three years, three women (6%) who received paravertebral anesthesia developed recurrent or metastatic tumors, compared with 19 of the 79 (24%) who received general anesthesia alone. At 36 months after surgery, 94% of women in the first group remained free of disease, compared with 77% of the others (P=0.007). Paravertebral anesthesia also seemed to delay the return of breast cancer (P =0.013) in the women who went on to suffer recurrences.

The researchers found no statistically significant differences in tumor grade, disease progression, adjuvant treatment, age and other factors among women in the two study groups. However, they noted, women who received general anesthesia alone tended to have slightly larger tumors, smaller tumor margins and higher rates of receiving chemotherapy than the other women—which could have influenced the outcome of the study.

Study Method Faulted

Critics of the research pointed to its retrospective nature and the inability to control for factors that might have biased the results.

“It’s very unclear whether the groups of patients receiving the blockade were equivalent to the patients who didn’t,” said E. Andrew Ochroch, MD, Director of Thoracic Anesthesiology at the University of Pennsylvania Health System in Philadelphia and co-author of an editorial accompanying the Anesthesiology article. The study “seems to be pushing us in the direction that the blockade is very, very good, but I believe that’s a false assumption. There’s not enough data to show that.”

Still, Jonathan Moss, MD, Professor of Anesthesiology and Critical Care at the University of Chicago Hospitals, called Dr. Sessler’s study “intriguing” and said it could have far-reaching implications for the care of patients with cancer if confirmed. “What happens during the course of your hospital surgery and anesthesia can really affect your well-being two years out,” Dr. Moss said.

Support for a direct connection between analgesia and tumor growth has come from recent studies demonstrating that opioids in clinical doses potentiate endothelial cell migration and proliferation, two components of angiogenesis that improve gastrointestinal recovery from the effects of surgery. In an article recently published in the journal Microvascular Research, Dr. Moss and colleagues showed that methylnaltrexone, a peripheral opioid receptor antagonist being developed by Wyeth and Progenics for postoperative ileus and constipation from opioid use, blocks opiate- and VEGF-induced angiogenesis in human lung and skin cells. “An effect of opioids on tumor proliferation could also explain the findings in [Dr. Sessler’s] study,” Dr. Moss said. “In addition to the opiates we routinely give, endogenous opioids go up quite substantially during surgery, and maybe even into the postoperative period.”

To be sure, the new research raises a host of questions. One of them is the relative impact on immune function of anesthesia, analgesia and the physical act of surgery. Another, the optimal time frame for quashing MRD and minimizing the odds of recurrence. Although Dr. Ben-Eliyahu cannot be certain, he believes this window is extremely small. “I cannot tell you whether it is a week or a month—nobody can,” he said. “My gut feeling is that it’s a matter of days. All our animal studies show that it is a very short period of opportunity.”

Whatever the case, Dr. Moss predicted, the study is certain to trigger additional interest in the intersection between anesthesia, surgery, pain relief and cancer. “I think people are going to be scurrying around to look at the data.”

Adam Marcus

Anesthesiology News, Issue 12/2006 | Volume 32:12