3.1 Demographics data of patients
The average age of patients in the intervention and control groups was 57.82 ± 11.29 years and 60.56 ± 10.56 years, respectively, with no significant difference between the two groups (p = 0.541). The average BMI of all patients was 22.5 kg/m², ranging from 18.5 to 23.9, with no significant differences observed between the groups (p > 0.05). Regarding RP types, one patient (9.1%) had primary RP, while ten patients (90.90%) had secondary RP: four with scleroderma (36.4%), one with mixed connective tissue disease (9.10%), and five with other diseases (45.50%). Ulcer distribution among patients was as follows: five had ulcers in both hands (45.5%), four had ulcers only in the right hand (36.4%), and two had no ulcers (18.20%). The average duration of ulcers was 17.33 ± 21.25 weeks. Post-surgery, slight bruising and swelling were observed in the first week, but no severe discomfort or complications occurred during the follow-up period. Additional characteristics of the patient groups are detailed in Table 1.
3.2 VAS pain score
In the HDFG-BTX group, preoperative scores ranged from 2.2 to 9 (average 5.33 ± 2.04), decreasing postoperatively to 0.2 to 1.4 (average 0.84 ± 0.82). This resulted in an average reduction of 4.49 ± 1.41 (Wilcoxon signed-rank test, Z = -2.67, p = 0.008), an 84.2% pain reduction (p = 0.018). The HDFG group showed preoperative scores from 2 to 7 (average 4.27 ± 1.53), dropping to 0.4 to 3.0 postoperatively (average 1.57 ± 0.92). This was a reduction of 2.70 ± 1.41, representing a 63.2% pain decrease (Z = -2.94, p = 0.003). The greater pain reduction in the HDFG-BTX group was statistically significant compared to the HDFG group (Mann-Whitney U test, Z = -2.36, p = 0.018). These findings are detailed in Table S1 and shown in Figure 3a-b.
A repeated measures ANOVA assessed treatment effects over time at five points (7 days, 14 days, 1 month, 3 months, 12 months). There was a significant main effect of time (F = 67.898, p < 0.001) and a significant interaction between time and treatment (F = 5.964, p < 0.001). Post-hoc Bonferroni analyses highlighted significant differences in VAS pain score reductions at specific times within each group. At 3 and 12 months, the VAS score reduction in the HDFG-BTX group was significantly greater than in the HDFG group (p = 0.025 and p = 0.011, respectively) (Figure 4a-c).
3.3 McCabe cold sensitivity scores
After the treatment, all patients reported significant relief in the symptom of finger sensitivity to cold. In the HDFG-BTX group, the mean McCabe cold sensitivity score prior to treatment was 272.73 ± 75.38. Following treatment, this mean score decreased significantly to 75.00 ± 48.73. This change reflects a substantial mean reduction of 197.73 ± 49.31, equating to a 72.5% decrease in cold sensitivity (Z = -2.70, p = 0.003). Similarly, the average McCabe cold sensitivity score in the HDFG group was 258.33 ± 51.54 before treatment. After treatment, this average decreased to 136.11 ± 50.17, indicating a significant reduction in cold sensitivity. The average decrease of 122.22 ± 51.54 corresponds to a 47.3% reduction in sensitivity (Z = -2.95, p = 0.007) (Figure 3c). Comparatively, the reduction in McCabe cold sensitivity scores was greater in the HDFG-BTX group than in the HDFG group (Figure 3d), with this difference being statistically significant (Z = -3.31, p = 0.001). This significant difference highlights the superior efficacy of the combined HDFG and BTX-A treatment in alleviating cold sensitivity symptoms in patients with RP, as evidenced by a greater reduction in McCabe cold sensitivity scores compared to HDFG alone.
To further evaluate treatment effects over time, a repeated measures ANOVA was conducted to analyze the reduction in McCabe cold sensitivity scores across five time points (7 days, 14 days, 1 month, 3 months, 12 months). The analysis demonstrated a significant main effect of time (F = 131.78, p < 0.001) and a significant interaction between time and treatment (F = 12.50, p < 0.001). Post-hoc analyses with the Bonferroni method revealed notable differences in McCabe cold sensitivity scores reduction at specific time points within each group. Notably, at 3 months and 12 months, the reduction in scores for the HDFG-BTX group was significantly greater than that for the HDFG group (p = 0.001 for both groups) as shown in Figure 4e-f.
3.4 Fingertip ulcer healing time
A total of 14 out of 20 hands (70%) had ulcers before treatment, and among them, 13 hands healed after treatment. One patient underwent distal finger amputation 2 weeks after the operation due to pre-existing fingertip necrosis. All ulcers were completely healed within one month after the operation and did not recur during the 1-year follow-up period. Figure 5a presents representative images depicting ulcer healing before and after treatment. Follow-up results revealed that hand ulcers in the HDFG-BTX group exhibited shorter healing times compared with the HDFG group (Figure 5b). The ulcer healing time was 14.25 ± 2.49 days for the HDFG-BTX group and 25.6 ± 4.34 days for the HDFG group. This difference was statistically significant (Z = -2.93, p = 0.002) as determined by the Mann-Whitney U test, suggesting that HDFG-BTX was more effective than HDFG in promoting ulcer healing.
3.5 Blood supply of the hands
Interestingly, following treatment with HDFG-BTX, significant improvements were reported not only in pain relief and sensitivity to cold but also in hand skin color, soft tissue texture, elasticity (Supplementary Video 2). Figure 5c illustrates representative images of the hand's appearance before and after treatment. Infrared thermal imaging was utilized to assess blood flow in the hand, revealing a significant improvement in blood supply after treatment with HDFG-BTX. Figure 5d presents a representative image of the hand's blood supply before and after treatment. Notably, evident insufficiency of blood supply was observed before surgery, particularly in the fingers with ulcers. However, one year post-surgery, the blood supply to the fingers was essentially restored. Significant differences in the percentage of blood perfusion units in the hands of the HDFG-BTX group before and after treatment are shown in Figure 5e.
3.6 Flow cytometry analysis of ADSCs-RP
We measured the number of SVF cells in high-density fat tissue. The number of isolated SVF cells per milliliter of fat tissue was 3.87 ± 1.63 × 105, ranging from 1.67 to 6.82 × 105 cells. Flow cytometry analysis demonstrated that the surface markers of ADSCs were highly positive in ADSCs-RP, including CD105 (PE 95.09%), CD90 (FITC 96.39%), CD73 (APC 99.94%), while surface markers, such as CD45/CD19/CD34/CD11b/HLA-DR were negative (PE 1.73%) (Figure 5f-g). The results suggested that typical stromal stem cells, which have a powerful therapeutic effect, can still be found in autologous high-density fat from patients with RP.
3.7 Multiple linear regression and one-way ANCOVA analysis for evaluation of treatment
To assess the impact of individual and combined factors on post-treatment outcomes, multiple linear regression analysis was conducted. The variables included in the model were treatment group, age, gender, and ulcer presence and its number pre-operation. The regression model revealed that the treatment group was a significant predictor of the reduction in VAS pain scores (R2=0.49, p<0.01) and McCabe cold sensitivity scores (R2=0.63, p<0.001), with the HDFG-BTX group showing significantly better outcomes compared with the HDFG group. Specifically, the HDFG-BTX treatment resulted in greater reductions in pain and cold sensitivity scores, indicating a superior therapeutic effect. In contrast, age, gender, ulcer presence and number of ulcers were not significant predictors in the model, suggesting that these factors did not significantly influence the post-treatment outcomes within the context of this study. These findings reinforce the effectiveness of the HDFG-BTX combination therapy in enhancing clinical outcomes for patients with Raynaud’s phenomenon, positioning it as a potentially superior treatment modality.
To further validate these results, we conducted a one-way ANCOVA to control for potential confounding variables. This analysis confirmed that the treatment group was the sole significant factor affecting difference in post-treatment outcomes, with a significant effect observed (p = 0.006, F = 10.244). This underscores the enhanced efficacy of HDFG-BTX therapy in reducing VAS pain and McCabe cold sensitivity scores compared with HDFG alone. Importantly, the adjustment for covariates did not alter the significance of the treatment group’s impact, reinforcing the robustness of the observed therapeutic benefits.