Subscribe to eTOC

Study Helps Clarify Risk of Hemorrhage in Cerebral Cavernous Malformation Patients

A new study attempts to answer: what is the risk of a hemorrhage in patients diagnosed with intracerebral cavernous malformation (pictured here)?

In the largest study to date of patients with intracerebral cavernous malformations, one in five who were initially diagnosed with a hemorrhage experienced another one within the first year, but the risk fell by almost one-half during the second year and continued to decrease for the next five-ten years.

Researchers at the Mayo Clinic in Rochester, MN, retrospectively analyzed the medical records and MRI data of 292 patients initially diagnosed with intracerebral cavernous malformations (ICM/CCM) between 1989 and 1999. In 74 patients with confirmed hemorrhage, the risk of a second hemorrhage fell from 18 percent during the first year to just over 9 percent during the second, and to 3 percent within five years, they reported in the Feb. 1 online issue of Neurology.

“An initial diagnosis of ICM with hemorrhage was the largest predictor of a secondary event, with a five-fold increase in risk during the first year,” the lead study author Kelly D. Flemming, MD, a researcher in the Mayo Clinic department of neurology, told Neurology Today in an e-mail. The median time from an initial to second hemorrhage was eight months.

“When we look at those patients who initially presented with hemorrhage, the risk was highest in the first year at 18.3 percent, but fell to 9.22 percent in the second year and significantly declined for the next three years. This is an important finding,” she said.


: “An initial diagnosis of ICM with hemorrhage was the largest predictor of a secondary event, with a five-fold increase in risk during the first year.”

The investigators also found an annual rate of hemorrhage of 6.19 percent in patients who presented with symptoms associated with hemorrhage and 2.8 percent when they were not related to hemorrhage. In 33 percent of the subjects hemorrhage was an incidental finding.

The most common symptoms were focal deficits, seizures, headaches without focal deficits, and neurologic symptoms without focal neurologic deficits.

“Our main finding is that depending on how a patient presents, their hemorrhage risk is different,” she said. “If someone had an incidental finding of ICM, their risk was very low. I always tell patients that, optimistically, there is 99.67 percent chance per year that everything will go well.”

The team also reported that the risk was greater for men than women, and for patients with ICM at multiple sites. However, in contrast to other smaller studies, they did not find pregnancy to be associated any increased risk.

“There are several limitations with our study,” said Dr. Flemming. “For one, patients were retrospectively identified and I personally did not have the chance to interview them and do in depth family history or assess their medications or medical condition. We had to use just the information from the medical records. In addition, some patients did not return the mailed survey about how they were doing, which limited the overall follow up time.”

“Our data is important for patient counseling, especially when comparing natural history to the risk/benefit ratio of potential treatments, such as surgery and radiosurgery,” Dr. Flemming said.

The new findings represent “the next step in furthering our understanding of the untreated clinical course of CCM,” wrote Rustam-Al-Shahi Salman, PhD, a senior clinical fellow and honorary consultant neurologist at Western General Hospital, Edinburgh, UK, and Gordon D. Murray, PhD, with the university's Center for Population Health Sciences, in an accompanying editorial.

Nonetheless, they wrote, CCM analysis of long-term outcomes presents a challenge, especially because patients with multiple ICMs caused by genetic mutations may have a different clinical course yet not all patients undergo genetic testing. He said risk-benefit analysis is not beyond doubt, but the size of risk from conservative management, where re-bleeds do occur but after the first year or two are usually not very severe, versus the risk of complications from surgery, which can be very severe, makes the argument against surgery for most patients with ICM.

Jay Preston Mohr, MD, the Daniel Sciarra Professor of Neurology at Columbia University Neurological Institute, said that because the study was based on sufficient population data and confirmed rates observed in other studies, it “should become the new standard for measuring and predicting hemorrhage risk after an initial event in ICM patients.”

However, a critical piece of the risk analysis remains missing, he told Neurology Today in a telephone interview.

“The biggest problem with better understanding of ICM is the paucity of reports on the severity of these hemorrhages and nowhere in this paper could I find any data on hemorrhage severity. Even though it was not their target, this is the most important bit of information clinicians need when counseling patients.”

Dr. Flemming told NeurologyToday that the severity data has not yet been analyzed.

Dr. Mohr is the principal investigator of ARUBA, a randomized clinical trial of intervention options in unruptured brain arteriovenous malformations (AVM). ARUBA will examine deaths, stroke incidence, including hemorrhage confirmed by imaging, as well as clinical impairment, in 400 patients.


said that because the study was based on sufficient population data and confirmed rates observed in other studies, it “should become the new standard for measuring and predicting hemorrhage risk after an initial event in ICM patients.”

“What the authors of this new paper have reported is very similar to what we see in AVM patients,” he said. “If a patient gets past the second year they tend to stabilize, with a low risk of hemorrhage.”

Issam A. Awad, MD, professor of surgery, neurosurgery, and neurology at the University of Chicago Pritzker School of Medicine, said the new study is important because it emphasizes the importance of natural history in guiding treatment decisions in the absence of clinical trials, and because the findings confirm those reported in smaller studies over the past 20 years.

“The fact that the results are in alignment, for the most part, with other smaller studies offers reassurance about the assumptions that guide current clinical management,” he told Neurology Today in a telephone interview.

Even so, Dr. Awad pointed out that statistical risk analysis in patients with multiple lesions remains complicated. “While the risk of bleeding may be slightly greater in patients with multiple lesions, the bleeding risk per lesion may lower.”

Dr. Flemming said she was surprised that multiple ICMs were associated with increased risk, but admitted the gender differences the team found were possibly due to bias.

“Neurologists should be very cautious in educating and monitoring patients who present with hemorrhage and do not undergo surgery for recurrent hemorrhage,” she added. However, neurologists can reassure patients where cavernous malformations are an incidental finding that the risk of hemorrhage is low.

Neurologists need to carefully monitor patients over the first year or so after hemorrhage, she added, especially for any sudden focal neurologic deficit or headache that could potentially represent a new hemorrhage. Patients should also be told to watch for any sudden weakness, numbness or lack of coordination of a limb, as well as sudden, severe headache, sudden difficulty with balance or coordination, or any sudden double vision or other changes in vision.


• Flemming, KD, Link MJ, Brown MD, Christianson TJH. The prospective hemorrhage risk of intracerebral cavernous malformations. Neurology 2012;E-pub 2012 Feb. 1.
    • Al-Shari Salman R, Murray GD. Editorial: The next step in understanding the prognosis of cerebral cavernous malformations. Neurology 2012; E-pub 2012 Feb. 1.