Literature search
In this study, 2035 articles were identified after duplications were removed. After screening the titles and abstracts, 1907 articles were excluded. The full texts of the remaining 128 articles were assessed. The following articles were excluded due to: reviews (55); study about nonadjuvant therapy (18); study on fresh fracture (10); conference abstract (6); duplicate publication (2); study about same therapeutic strategies (2); no predetermined outcomes reported (2); comments (2); study on children congenital anomaly (1). Finally, thirty studies were included in the analysis [15–44] (Figure 1).
The inclusion study was published from 1984 to 2020. The research is mainly located in Asia, North America, and Europe. The sample size from each study was relatively small, with a maximum of 140 patients. One study included adolescent patients [17]; however, we do not think it will impact the overall results. The interventions analyzed included physiotherapy (LIPU, extracorporeal shock wave (ESWT), electromagnetic field (EMF) therapy), and bioactive substance treatment (PRP, BMP, bone marrow aspirate (BMA, including mesenchymal stem cells)). and oral drugs (strontium ranelate (Protelos), Chinese traditional medicine (CTM)). The average follow-up period ranged from 3 months to 7.6 years (Table 1). After comprehensive consideration of the risk of bias, although all studies were RCT designs, the major factor affecting the quality was the lack of blindness of patients and intervention personnel, as well as the blindness of the outcome assessor. This may lead to a subjective influence on the healing judgment, which makes the results tend to be positive (Figure 2).
Table 1
Characteristics of the included studies.
Study | Location | Sample size | Average age | Fracture site | Interventions | Control | Follow-up |
Tang YF 2020[15] | China | 66 | 20-56 | Tibia | PRP+BMA | BMA | 16-36 Month |
Marion Mühldorfer-Fodor 2020[16] | Germany | 68 | 29(16-60) | Scaphoid | EWST+ACB | ACB | 24 Weeks |
Ahmad I 2020[17] | Pakistan | 101 | 31.46±13.9 | Tibia (major) | PROTELOS | CONTROL | 3 Months |
Samuel G 2018[18] | Indian | 40 | 34.3±9.49;39.35±10..39 | Tibia (major) | PRP | CONTROL | 36 Weeks |
Carlos AO 2017[19] | Germany | 16 | 37.14±10.22;38.88±15.36 | Humerus | PRP+ACB | ACB | 36 Weeks |
Hernigou P 2017[20] | France | 80 | 41.2±18.2 | Tibia | BMA+ACB | ACB | 7.6 (5-10) Years |
Zhang S 2017[21] | China | 47 | 38.29±7.49;36.57±8.31 | Tibia (major) | PRP+BMA | BMA | 9-15 Months |
Zhao ZC 2017[22] | China | 92 | 18-70 | Femur | PRP | CONTROL | 12 Months |
Ghaffarpasand F 2016[23] | Iran | 75 | 26.5±5.8;26.3±6.2 | Femur, Tibia | PRP | CONTROL | 12 Months |
Zhang H 2016[24] | China | 24 | 33.45;32.69 | Tibia | BMA | CONTROL | 12-34 Months |
Streit A 2016[25] | usa | 8 | 47(24-63) | Fifth metatarsal | EMF | CONTROL | 24 Weeks |
Christian von Ruden 2016[26] | Germany | 49 | 44(19-77) | Ulnar, Radial | BMP+ACB | ACB | 6 Months |
Zhai L 2016[27] | China | 63 | 39.6(23-50);38.1(20-49) | Femur, Tibia | BMA+ESWT | ESWT | 12 Months |
Shi HF 2013[28] | China | 58 | 41.1±14.5;38.4±11.6 | Femur, Tibia | EMF | CONTROL | 3 Months |
Huang ZJ 2011[29] | China | 64 | 47.3±22.2 | Ulna or radius (major) | CTM | CONTROL | 3 Months |
Schofer MD 2010[30] | USA | 101 | 42.6±14.6;45.1±11.9 | Tibia | LIUS | CONTROL | 16 Weeks |
Yuan JG 2010[31] | China | 140 | 34±2 | Humerus, Tibia | BMA | ACB | 3 Months |
Sun YP 2009[32] | China | 40 | 43±6 | Tibia (major) | BMP+ACB | ACB | 11-20 Months |
Angelo Cacchio 2009[33] | Italy | 126 | 42.7±5.9 | Tibia Femur | ESWT | CONTROL | 24 Months |
Rutten S 2008[34] | Netherlands | 13 | 42-63 | Fibula | LIUS | CONTROL | 3 Months |
G.M Calori 2008[35] | Italy | 120 | 43(19-65) | Humerus, Tibia (major) | BMP | PRP | 9 Months |
Bilic R 2006[36] | Croatia | 18 | 23±5;22±5;19±4 | Scaphoid | BMP+ACB | ACB | 24 Months |
Ricardo M 2006[37] | France | 21 | 26.7(17-42) | Scaphoid | LIUS | CONTROL | 1-4 Years |
Calori GM 2006[38] | Italy | 29 | 47±2.56;35.3±1.76 | Femur, Tibia | BMP | PRP | 9 Months |
Simonis RB 2002[39] | UK | 34 | 32(16-61) | Tibia | EMF | CONTROL | 6 Months |
Friedlaender GE 2001[40] | USA | 122 | 38±16;34±11 | Tibia | BMP+ACB | ACB | 24 Months |
Cook SD 1999[41] | USA | 30 | na | Tibia | BMP | ACB | 9 Months |
Scott G 1994[42] | UK | 23 | 43(23-87) | Femur, Tibia | EMF | CONTROL | 6 Months |
Sharrard WJW 1990[43] | UK | 51 | 34.7(18-84);45.4(18-76) | Tibia | EMF | CONTROL | 12 Weeks |
Barker AT 1984[44] | UK | 17 | 34(19-72) | Tibia | EMF | CONTROL | 48 Weeks |
Abbreviations: ACB: autologous cancellous bone; BMA: bone marrow aspirate; BMP: bone morphogenetic proteins; CTM: Chinese traditional medicine; EMF: electromagnetic fields; ESWT: extracorporeal shock wave; LIUS: Low-intensity pulsed ultrasonography; PROTELOS = Strontium ranelate; PRP: Platelet-Rich Plasma. |
*Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). |
**If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. |
From: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71 |
For more information, visit: http://www.prisma-statement.org/ |
In the classification of adjuvant therapeutic strategies, for long bone nonunion treatment, surgical stabilization was necessary, but for short bones, it was not. Therefore, surgical fixation was adopted according to actual needs and not considered in strategies. However, ACB, as the gold standard treatment, might affect adjuvant therapy and is considered a strategy. In addition, it should be noted that in the healing rate outcome, both intervention groups might achieve fracture healing during the follow-up period. In this case, the healing time is an important supplement to assess adjuvant therapy effects.
For healing rate outcome, 15 intervention strategies were analyzed, including ACB, BMA, BMP, CTM, EMF, ESWT, LIUS, PROTELOS, PRP, their combinations, and blank control (Figure 3, A). In pairwise comparisons, except ESWT (OR: 1.76; 95% CI: 0.24, 13.00) and BMA+ESWT (OR: 0.79; 95% CI: 0.15, 4.27), which were not significantly different from the blank control, the other strategies were significantly better than the control. Compared to ACB, only BMA+ACB was found to be significantly better (OR: 0.12; 95% CI: 0.03, 0.59). In the SUCRA ranking results, BMA+PRP (96%), BMA+ACB (90%), and BMA alone (82%) showed relative advantages in healing rate (Figure 4, A). A comparison-adjusted funnel plot did not show potential publication bias (Figure 5, A).
For the healing time, a total of 13 intervention strategies were analyzed, including ACB, BMA, BMP, CTM, EMF, EWST, LIUS, PRP, their combination, and blank control (Figure 3, B). In the pairwise comparison, BMP (SMD: -2.61; 95% CI: -5.01, 0.22) and LIUS (SMD: -8.85; 95% CI: -12.85, -4.86) interventions had significantly shorter healing times than the blank control. Compared with ACB, only LIUS intervention significantly shortened the healing time (SMD: -8.93; 95% CI: -14.37, -3.48). In the SUCRA ranking results, LIUS (99%), PRP+BMA (74%), and BMP (69%) had relative advantages (Figure 4, B). No publication bias was detected according to the funnel plot (Figure 5, B).
For AE outcome, a total of 7 intervention strategies were analyzed, including ACB, BMA, BMA+ACB, BMP, EMF, ESWT, PRP, and control (Figure 3, C). Compared with the control, EMF (OR: 13.21; 95% CI: 1.58, 110.40) and ESWT (OR: 4.90; 95% CI: 1.38, 17.43) had higher AE risk. In the SUCRA ranking results, BMA+ACB (95%) had relatively few side effects (Figure 4, C). Although no publication bias was detected (Figure 5, C), the AE results included malunion, infection, reoperation, and adverse drug reaction with high heterogeneity.