Table 4.

Summary Statistics for Secondary End-Points (2 and 24 Hours Post-Treatment)

Group	Time Point	Mean ± SD	Mean ± SD Change From Baseline	50% Pain Relief Rate	P Value* vs Placebo	Pain-Free Rate	P Value* vs Placebo
Active (n = 18)	Before	5.5 ± 1.5
	15 minutes after	1.2 ± 1.0	−4.3 ± 1.5	17	P < .001	9	P = .063
				94%	P* < .001	50%	P* = .51
	2 hours	1.3 ± 1.9	−4.2 ± 2.3	16	P < .001	9	P = .018
				89%	P* < .001	50%	P* = .14
	24 hours	1.2 ± 2.5	−4.3 ± 2.6	15	P = .13	14	P = .114
				83%	P* = 1.00	78%	P* = .91
	24 hours sustained			14	P = .001	8	P = .035
				78%	P* = .008	44%	P* = .28
Placebo (n = 12)	Before	4.7 ± 2.4
	15 minutes after	3.6 ± 2.0	–1.2 ± 1.4	2		2
				17%		17%
	2 hours	4.9 ± 3.0	0.2 ± 2.1	2		1
				17%		8%
	24 hours	1.9 ± 2.5	–2.8 ± 4.4	7		6
				58%		50%
	24 hours sustained			2		1
				17%		8%

^*The P^* values are Bonferroni-adjusted P values for 8 tests.

Patients with data available for both 2 and 24 hours only. “24 Hours Sustained” denotes improvements at both 2 and 24 hours.

Note that rescue medication use following the 15-minute post-measurement and before 24 hours in the placebo group was higher (76%) compared with the active treatment group (17%).

Placebo rates with imputation using LOCF for 15 minutes after, 2 hours, 24 hours, 24 hours sustained:

50% pain relief: 3 (18%) P < .001, 3 (18%) P < .001, 9 (53%) P = .053, 3 (18%) P < .001.

Pain-free: 2 (12%) P = .015, 1 (6%) P = .004, 7 (41%) P = .027, 1 (6%) P = .009.

The corresponding Bonferroni adjusted for 8 tests P values are for 50% pain relief: P^* < .001. P^* < .001, P^* = .42, and P^* = .003, respectively; and for pain-free: P^* = .12, P^* = .031, P^* = .22, and P^* = .073, respectively.

LOCF = last observations carried forward; SD, standard deviation.