LOS ANGELES – A prospective, multicenter study validated the "spot sign" on CT angiography as a common finding in patients with intracerebral hemorrhage that is predictive of significant hematoma growth and early death.

Close to one-third of these patients who present early to an emergency department have spot signs on CT angiogram (CTA), and nearly 60% of them die within the first 3 months, Dr. Andrew M. Demchuk reported at the International Stroke Conference.

The study enrolled 268 patients with an intracerebral hemorrhage (ICH) less than 100 mL in size who arrived at an emergency department within 6 hours of symptom onset. At baseline, patients at 11 centers in six countries underwent non–contrast enhanced CT (NCCT) imaging and a first-pass CTA in the head of the Circle of Willis through the entire hematoma volume. A follow-up NCCT scan was performed at 24 hours, with clinical follow-up at 24 hours and at 3 months.

A neuroradiologist at the University of Toronto who did not know follow-up information interpreted the CT angiograms for each patient, using a predetermined definition of the spot sign. Separately, investigators at the University of Calgary (Alta.) performed hematoma volumetric analyses of each ICH and intraventricular hemorrhage (IVH), using quantitative tomography without knowing any of the NCCT results.

At baseline, 80 patients (30%) had a spot sign, according to Dr. Demchuk of the University of Calgary and his associates. They noted that there may have been a slight predilection for more spot signs in patients with lobar ICH, but the spot sign was seen frequently regardless of ICH location. The incidence of a spot sign varied from 25% to 41% in different regions of the brain.

The median time from onset of symptoms to baseline CT imaging was 138 minutes.

A quarter of patients with spot signs deteriorated rapidly, he noted. Four died before the 24-hour NCCT, seven underwent surgery before the 24-hour CT, and nine were treated with off-label recombinant activated factor VII (rFVIIa). In a previous trial, hemostatic therapy with rFVIIa reduced the risk of hematoma expansion but did not improve overall outcomes (N. Engl. J. Med. 2008;358:2127-37). Another three treatment trials are now underway that base rFVIIa treatment on CTA findings, noted Dr. Demchuk, director of the Calgary Stroke Program for Alberta Health Services.

In the current study, called PREDICT (Predicting Hematoma Growth and Outcome in Intracerebral Hemorrhage Using Contrast Bolus CT), 27% of 228 patients who had a 24-hour follow-up NCCT available had spot signs at baseline. The size of ICH or IVH at baseline in patients with the spot sign was roughly double the size of the hematoma in those without the spot sign.

In multiple analyses that used 10 different definitions of hematoma growth, expansion was significantly more likely to occur in patients with a spot sign. "There’s a lot of debate in the literature, still, in terms of what’s the best growth criteria. It doesn’t matter what growth criteria you use. All of them had a much higher frequency in the setting of a spot sign than without a spot sign," Dr. Demchuk said at the conference, which was sponsored by the American Heart Association.

Among 176 patients with 3-month follow-up data, clinical outcomes were significantly worse in those with spot signs at baseline. Early neurologic worsening was seen in 38% of those with spot signs and 13% of those without. Mortality was greater in patients with spot signs (59%) than in those without (26%). The difference in mortality was early and dramatic, with most of the divergence between groups occurring within 30 days.

The study defined a spot sign according to six criteria: The shape is spotlike, serpiginous, or linear. The spot is located within the margin of a parenchymal hematoma without connection to an outside vessel, and is greater than 1.5 mm in diameter in at least one dimension. The density of the spot sign is at least double the density of the hematoma. Single or multiple spot signs could be present, and these must not be caused by calcific deposition (hyperdensity in the same location on NCCT).

Having more than one spot sign predicted even greater growth. Patients with more than one spot sign had triple the growth in hematoma, compared with patients without a spot sign, Dr. Demchuk said. The use of a "spot sign score" calculation, which has been proposed by previous researchers, did not correlate with hematoma growth in this study.

The spot sign did not seem to be as helpful in predicting hematoma growth in patients on warfarin, especially those with an elevated international normalized ratio (INR) at baseline, he added. The study included about 15 patients on warfarin who had an INR greater than 1.5 at baseline. Hematoma expansion in this subgroup was three times more likely to occur than in patients who were not on warfarin, whether a spot sign was present or not.

The study was funded by the Canadian Stroke Consortium and by Novo Nordisk, which is developing rFVIIa. Dr. Demchuk did not receive any personal financial support from Novo Nordisk.

Spot Sign Predicts Intracerebral Hemorrhage Growth