From July, 2011 to July, 2014, we enrolled 102 patients with swelling after upper limb trauma surgery for this study. The inclusion criteria were as follows: i) Patients were >18 and <75 years old; ii) patients had history of clear upper limb trauma surgery, and all had blunt trauma; iii) patients did not have any swelling associated with severe neurological and vascular damage; and iv) time of trauma was <8 h. The exclusion criteria were as follows: i) Swelling caused by other reasons, such as burns, upper limb venous or lymphatic compression, tumor swelling; ii) other serious organ disease, such as heart, kidney and brain disease; and iii) patients who refused to participate in this study.

The study was approved by the Ethics Committee of the The Third People's Hospital of Qingdao (Shandong, China). Informed consent of patients or their families was obtained. Patients were randomly divided into 3 groups (34 cases per group) according to the order of their hospitalization. We used the ‘Gu Yudong's Operative Hand Surgery’ (edition 1999) to grade the swelling degree. Group A was treated with β-sodium aescinate; group B was treated with mannitol; and group C was treated with both β-sodium aescinate and mannitol.

Group A comprised 24 men and 10 women, aged 24–73 years (average 45.3±7.6 years). Time of trauma ranged from 0.7 to 8 h (average 5.6±1.1 h). In group A, we identified 18 grade II and 16 grade III cases. In group B, we had 18 men and 16 women, aged 29–70 years (average 46.8±7.3 years). Time of trauma ranged from 0.6 to 7.6 h (average 5.4±1.5 h). There were 15 cases of grade II swelling and 19 cases of grade III swelling in group B. In group C, there were 19 men and 15 women, aged 26–69 years (average 45.8±6.3 years). Time of trauma ranged from 0.8 to 7.5 h (average 5.8±1.3 h). A total of 17 cases of grade II and 17 cases of grade III swelling were identified in group C. We compared the gender, age, time of trauma and grading of swelling of the patients in the three groups, and the differences were of no statistical significance (P>0.05).

Group A received 20 mg/time of β-sodium aescinate (trademark: Beishutai, H22025070; approved by the state and Jilin Jinsheng Pharmaceutical Co., Ltd., Jilin, China) and qd ivgtt. Group B received 20% mannitol, 125 ml/time and q12 h ivgtt. Group C received β-sodium aescinate for 10 mg/time, qd ivgtt plus mannitole (125 ml/time and qd ivgtt). Patients were monitored and the time of swelling subsidence was recorded, and were also closely monitored for complications.

Differences in AQP-1 level, clinical effects and complication occurrence before and after treatment were compared. For AQP-1 measurement 5 ml of early morning fasting venous blood was collected and the samples were sent to laboratory immediately. Double antibody sandwich enzyme-linked immunosorbent assay (ELISA) was employed using kits purchased from the Shenzhen Pinmei Biological Engineering Co., Ltd., Shenzhen, China.

To examine the clinical effects, we determined that if the time of swelling subsidence was <3 days the treatment was considered as being fully recovered and if that time ranged from 3 to 7 days it was considered as remissed. If the time of swelling subsidence was longer than 7 days then we considered the treatment as ineffective. We used the following formula for our calculations: Total efficiency = (recovery + remission)/total cases × 100%. Complications included fever, allergy, renal impairment, phlebitis, tension blisters and bone compartment syndrome.

SPSS 19.0 software (SPSS, Inc., Chicago, IL, USA) was used for data analysis. Measurement data were presented as mean ± standard deviation and analysis of variance was employed for comparisons among groups. Countable data were presented by the number of cases (%) and χ² test was used for comparisons among groups. P<0.05 was considered to indicate a statistically significant difference.

The time of swelling subsidence in group C was significantly shorter than that in the remaining groups and the difference was statistically significant (P<0.05). The recovery ratio and total efficiency in group C were significantly higher than those in other two groups and the difference was statistically significant (P<0.05; Table I).

We compared AQP-1 level in all the patients in the three groups before treatment, and the differences showed no statistical significance (P>0.05). Three and seven days after treatment, the AQP-1 level in group A and group C was decreased and the decrease intensified with time. Differences of comparison within groups were statistically significant (P<0.05). When we compared between groups, the differences were not statistically significant (P>0.05). Comparing the AQP-1 levels in group B, before and after treatment, revealed no statistical significant difference (P>0.05; Table II).

In group A, there were 3 cases with tension blisters, 1 case of allergy and 2 cases of phlebitis, and the total incidence was 17.65%. In group B, there were 4 cases of renal impairment, 2 cases of tension blisters, 1 case of allergies and 2 cases of electrolyte imbalance, and the total incidence was 23.53%. In group C, there was 1 case of tension blisters and 1 case of phlebitis, and the total incidence was 5.88%. We compared the complication incidence in all three groups, and the differences were not statistically significant (χ²=4.151, P=0.125).