Table 1 Main characteristics and findings of ten studies comparing compressive cryotherapy with cryotherapy alone after a knee surgery

Schröder and Pässler (1994)

RCT

1

44 Patients (all underwent ACL reconstruction surgery under arthroscopy)

G1 (n = 21) 15 M/6 W Mean age 24.2 y

G2 (n = 23) 18 M/5 W Mean age 24.8 y

G1: CCD (continuously up to hospital discharge)

G2: Ice bags (three times/day)

1. Pan intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Range of motion-ROM (in degrees)

4. Knee function (knee score of Noyes and McGinniss)

5. Use of analgesic medication (total doses, mg/kg) of oral tilidine, IM pethidine and piritramide

6. Blood loss (in ml)

7. Adverse events Outcomes were measured on POD 1, 2, 3, 6, 14, 28

G1 had significant differences in: ROM on all days (P < 0.01); VAS pain scale on the 6th day (P < 0.01); knee edema on the 3rd and 6th days (P < 0.035), knee function (P < 0.025) and used less oral tilidine and IM piritramide (P < 0.04)

There were no adverse events

Li et al. (2010a)

RCT

3

140 Patients (all underwent arthroscopic surgery)

G1 (n = 70)

G2 (n = 70)

Total mean age 36.6 y

G1: CCD (continuously up to POD2)

G2: Conventional ice pack therapy (cycles (1 h treatment and 1 h pause) for 1 day)

1. Pain intensity (VAS)

2. Use of analgesic medication (celecoxib)

G1 had significantly lower VAS pain scores and less use of pain medications on 6, 12, 24, 36, 48 h after operation (P < 0.05)

There were no adverse events

Li et al. (2010b)

RCT

3

140 Patients (all underwent arthroscopic surgery)

G1 (n = 70)

G2 (n = 70)

Total mean age 36.6 y

G1: CCD (continuously up to POD2)

G2: Conventional ice pack therapy (cycles (1 h treatment and 1 h pause) for 1 day)

Edema (knee circumference measured using tape)

G1 had significantly lower VAS pain scores and less use of pain medications on 6, 12, 24, 36 and 48 h after operation (P < 0.05)

There were no adverse events

Demoulin et al. (2012)

RCT

3

66 Patients (all underwent primary unilateral TKA surgery)

G1 (n = 22)

8 M/14 W

Mean age 72.0 y

G2 (n = 22)

9 M/13 W

Mean age 68.1 y

G3 (n = 22)

9 M/13 W

Mean age 71.2 y

G1: GCD (3 sessions (90 s)/day from POD2 to hospital discharge)

G2: Traditional ‘‘gel pack’’ therapy (5 sessions (20 min)/day from POD2 to hospital discharge)

G3: CCD (5 sessions (20 min)/day from POD2 to hospital discharge)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Range of motion-ROM (in degrees)

4. Cutaneous temperature

Comparison including VAS, edema, cutaneous temperature and ROM between G2 and G3 on POD7 remained non-significant (P > 0.05). There were no adverse events

Waterman et al. (2012)

RCT

3

36 Patients (all underwent ACL

reconstruction surgery)

G1 (n = 18)

15 M/3 W

Mean age 28.7 y

G2 (n = 18)

15 M/3 W

Mean age 30.9 y

G1: CCD (3 sessions (30 min)/day for 6 weeks)

G2: Conventional ice pack therapy (3 sessions

(30 min)/day for 6 weeks)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Use of analgesic medication (not reported)

4. Knee function (Lysholm score)

5. Quality of life (SF-36)

Outcomes were measured on POW 1, 2, and 6

G1 had significantly lower VAS pain scores (P < 0.0001) and

discontinued use of pain medications, by 6 weeks

(P = 0.0008)

There were no adverse events

Tian et al. (2013)

RCT

3

64 Patients (all underwent arthroscopy surgery)

G1 (n = 32)

7 M/25 W

Mean age 54.21 y

G2 (n = 32)

12 M/20 W

Mean age 56.44 y

G1: CCD (continuous treatment for 2 days after operation)

G2: Conventional ice pack therapy (cycles (30 min treatment and 8 h pause) for 2 days after operation)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Knee function (HSS score)

4. Comfort degree

G1 had significantly lower VAS pain scores and lighter edema on POD1 and 2 (P < 0.05)

G1 had higher comfort degree than G2 (P < 0.05). There were no adverse events

Xie et al. (2013)

RCT

1

40 Patients (all underwent ACL reconstruction surgery under arthroscopy)

G1 (n = 20)

16 M/4 W

Mean age 29.1 y

G2 (n = 20)

15 M/5 W

Mean age 28.2 y

G1: CCD (continuous treatment for 3 days after operation)

G2: Physiological saline ice pack therapy (4 sessions (45 min)/day for 3 days after operation)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

G1 had significantly lower VAS pain scores and lighter edema on POD1, 2 and 3 (P < 0.05)

There were no adverse events

Xu et al. (2013)

RCT

3

60 Patients (all underwent TKA surgery)

G1 (n = 30)

11 M/19 W

Mean age 65.8 y

G2 (n = 30)

10 M/20 W

Mean age 67.1 y

G1: CCD (continuous treatment for 2 days after operation)

G2: Hypertonic saline crushed ice bag (cycles (20 min treatment and 2 h pause) for 3 days after operation)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Range of motion-ROM (in degrees)

4. Postoperative drainage

5. Cutaneous temperature

G1 had significantly lower VAS pain scores and lighter edema on POD1 and 2 (P < 0.05)

G1 had better ROM on POD1, 2 and 3 (P < 0.05)

Cutaneous temperature in G1 was higher than G2 (P < 0.05)

There were no adverse events

Xie et al. (2014)

RCT

1

40 Patients (all underwent non-ACL reconstruction surgery under arthroscopy)

G1 (n = 20)

16 M/4 W

Mean age 36.8 y

G2 (n = 20)

17 M/3 W

Mean age 35.5 y

G1: CCD (continuous treatment for 2 days after operation)

G2: Physiological saline ice pack therapy (4 sessions (45 min)/day for 2 days after operation)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Range of motion-ROM (in degrees)

G1 had significantly lower VAS pain scores and lighter edema on POD1 and 2 (P < 0.05)

G1 had higher ROM than G2 (P < 0.05).

There were no adverse events

Wang et al. (2014)

RCT

3

32 Patients (20 patients underwent ACL reconstruction surgery and 12 patients underwent PCL reconstruction surgery under arthroscopy)

G1 (n = 16)

4 M/12 W

Mean age 50.72 y

G2 (n = 16)

7 M/9 W

Mean age 57.13 y

G1: CCD (continuous treatment for 2 days after operation)

G2: Conventional ice pack therapy for 2 days after operation)

1. Pain intensity (VAS)

2. Edema (knee circumference measured using tape)

3. Knee function (Lysholm score)

4. Comfort degree

G1 had significantly lower VAS pain scores and slighter edema on POD1 and 2 (P < 0.05)

G1 had higher comfort degree than G2 (P < 0.05). There were no adverse events