Table 3.

Primary End Point and Its Individual Components Among the 1091 Asymptomatic and the 1170 Symptomatic Patients After Excluding Patients Aged ≥ 80 Years According to Treatment Group in the Periprocedural Period.^*

	Periprocedural Period†
	CAS No. of Events (Rate±SE‡)	CEA No. of Events (Rate±SE‡)	Percent Absolute Treatment Effect (95% CI)‡	HR (95% CI)§	P∥	Treatment by Symptomatic Status Interaction P
MI Endpoint
Asymptomatic	5 (0.9±0.4)	12 (2.2±0.6)	−1.3 (−2.8 to −0.2)	0.43 (0.15 to 1.24)	0.12	0.91
Symptomatic	6 (1.0±0.4)	13 (2.2±0.6)	−1.2 (−2.6 to 0.3)	0.47 (0.18 to 1.23)	0.13
Stroke Endpoint (any stroke within periprocedural† period)
All Stroke end points
Asymptomatic	13 (2.4±0.7)	8 (1.5±0.5)	0.9 (−0.7 to 2.5)	1.64 (0.68 to 3.97)	0.27	0.68
Symptomatic	30 (5.1±0.9)	15 (2.6±0.7)	2.6 (0.4 to 4.8)	2.12 (1.14 to 3.93)	0.018
Major Stroke
Asymptomatic	3 (0.5±0.3)	2 (0.4±0.3)	0.2 (−0.6 to 1.0)	1.50 (0.25 to 8.96)	0.66¶	0.92
Symptomatic	5 (0.9±0.4)	3 (0.5±0.3)	0.3 (−0.6 to 1.3)	1.70 (0.41 to 7.13)	0.47¶
Minor Stroke
Asymptomatic	10 (1.8±0.6)	6 (1.1±0.4)	0.7 (−0.7 to 2.1)	1.74 (0.63 to 4.79)	0.29	0.70
Symptomatic	25 (4.3±0.8)	12 (2.1±0.6)	2.2 (0.2 to 4.2)	2.19 (1.10 to 4.36)	0.026
Stroke±Death end point (any stroke or death within periprocedural† period)
Asymptomatic	13 (2.4±0.7)	8 (1.5±0.5)	0.9 (−0.7 to 2.5)	1.64 (0.68 to 3.97)	0.27	0.56
Symptomatic	33 (5.6±1.0)	15 (2.6±0.7)	3.1 (0.8 to 5.3)	2.34 (1.27 to 4.31)	0.006
Primary Endpoint (any stroke, MI or death within periprocedural† period)
Asymptomatic	17 (3.1±0.7)	20 (3.7±0.8)	−0.6 (−2.7 to 1.6)	0.87 (0.46 to 1.66)	0.67	0.28
Symptomatic	37 (6.3±1.0)	28 (4.8±0.9)	1.5 (−1.1 to 4.2)	1.38 (0.85 to 2.26)	0.19

^*For death, stroke and MI outcomes, patients may have > 1 event (examples, fatal stroke events are included in both death and stroke outcomes, individuals may have both an ipsilateral and a subsequent nonipsilateral stroke, etc.).

^†Periprocedural period is defined per protocol as 30 days postprocedure for all patients receiving assigned therapy within 30 days from randomization, or 36 days after randomization for all patients not receiving assigned treatment within 30 days.

^‡Treatment difference CAS–CEA. Because of the relatively short periprocedural period that minimizes censoring, event rates and absolute differences in event rates are calculated as a proportion of patients with events.

^§Relative treatment (CAS/CEA) efficacy is defined by the HR adjusted for age, symptomatic status and sex. For the periprocedural period all patients were censored at the end of their periprocedural period.

^∥P is defined by significance of HR (per protocol).

^¶Univariate proportional hazards model is used because of the small no. of events.