Table 3

Mean values of VAS for pain where available

Tendinopathy	Study	VAS type (unspecified, at rest, at night, with activity)	VAS scale	GTN group		Placebo group		Treatment effect for pain (VAS 2 − VAS 1) – (VAS 4 − VAS 3)	P<0.05
Tendinopathy	Study	VAS type (unspecified, at rest, at night, with activity)	VAS scale	VAS baseline42	VAS longest follow-up42	VAS baseline (3)	VAS longest follow-up (4)	Treatment effect for pain (VAS 2 − VAS 1) – (VAS 4 − VAS 3)	P<0.05
Rotator cuff (acute)	Berrazueta et al	Unspecified	0–10	7.1	2	6	5.5	−4.6	Yes
Rotator cuff	Giner-Pascual et al 33	Unspecified	0–10	5.4	5.3	2.3	4.6	−2.4	Yes
Patellar	Steunebrink et al 35	With activity	0–10 (reverse)	4.1	6.6	5.8	7.8	−0.5*	No
Achilles	Kane et al 30	Unspecified	0–10	5.6	3.1	5.4	3	−0.1	No
Lateral epicondylitis	Paoloni et al 27	With activity	0–4	2.2	0.8	2.6	1.3	−0.1	No
	Paoloni et al 31P†	With activity	0–40‡	36	34	32	28.2	+1.8	No
	Paoloni et al 31P	At rest	0–40‡	36	35	32	29.4	+1.6	No
	Paoloni et al 31	At night	0–40‡	32	31.8	30	27.3	+2.5	No
	Ozden et al 34	Unspecified	0–10	8.1	0.7	8.8	4.9	−3.5	Yes

The statistical significance column (p<0.05) denotes whether there was a significant benefit in VAS for pain with topical GTN versus placebo.
*Due to the reverse VAS scale used (0=worst pain and 10=no pain), our subtraction was also reverse, that is, (VAS 1 – VAS 2) – (VAS 3 – VAS 4).
†In this study, only the mean VAS values of the highest strength GTN (3.6 mg/24 hours) group are presented for all pain with activity, at rest and at night; however, the lowest strength GTN group did have significantly less pain with activity at follow-up compared with placebo.
‡The VAS scale used is not stated in the article, but we assume based on the reported values and the previous article by the same group (were a VAS scale of 0–4 was used) that it is 0–40.
GTN, glyceryl trinitrate; VAS, visual analogue scale.