Biologic agents for anterior cruciate ligament healing: A systematic review

Table 1 Synopsis of all clinical trials dealing with the application of biologic agents to promote anterior cruciate ligament healing

	Ref.	Study design	Methods	Results
Clinical results
+	Del Torto et al[21]	Prospective comparative study (PRFM vs control) 28 patients (14 vs 14)	ACL reconstruction with hamstring tendon graft fixed in the femoral tunnel through the RIGIDFIX system (DePuy) and in the tibial tunnel through the Bio-INTRAFIX system (DePuy) PRFM was prepared using Cascade Medical Enterprises 2 tube kit (Cascade Medical Enterprises, Wayne, NJ). PRFM clot was sutured in the proximal graft loop and it reaches the proximal tunnel once the graft is pulled in place. Distally, the PRFM clot was inserted between the four strands of the G-ST graft before the interference screw system was applied	PRFM-augmented patients showed a statistically significant higher clinical improvement at 24 mo follow-up (P = 0.032) Objective clinical evaluation both through IKDC score and with Rolimeter arthromether did not show any difference between the two groups
	Magnussen et al[20]	Retrospective comparative study (PRP vs control) 58 patients (29 vs 29)	ACL reconstruction with allograft tibial tendon, fixed with absorbable cross pin in femoral tunnel and absorbable interference screw in tibial tunnel After graft positioning, intra-articular portion was coated with PRP prepared with GPS II Platelet Concentrate Separation Kit (Biomet)	Decreased effusions at 10 ± 4 d was noted in the PRP group, but this difference disappeared by 8 ± 4 wk No differences in patient-reported outcomes were noted in the 58 patients with two-year outcome data
	Darabos et al[22]	Randomized trial (ACS vs control) 62 patients (31 vs 31)	ACL reconstruction with hamstring (30) or patellar tendon (32), fixed with BioTransFix (Arthrex) or RigidFix (Mytek) at femoral side, and with an interference screw at the tibial side ACS was produced drawing venous blood into syringes containing pre-treated glass beads, and after a period of incubation serum is extracted through centrifugation. ACS was administered with an injection regime of 4 injections on day 0 (day of surgery), day 1, day 6, and day 10	Clinical outcomes were consistently better in patients treated with ACS at all data points and for all outcome parameters, with statistically significant differences in the WOMAC stiffness subscale after 1 yr Decrease in IL-1b synovial fluid concentration was more pronounced in ACS group, with statistically significant lower values in the ACS group at day 10
	Vogrin et al[23]	Randomized trial (PRP vs control) 45 patients (22 vs 23)	ACL reconstruction with double-looped semitendinosus and gracilis tendon graft fixed with two bioabsorbable cross pins in the femoral tunnel and one bioabsorbable interference screw in the tibial tunnel PRP was produced with Magellan system (Medtronic) and applied into the femoral and tibial tunnels as well as onto the graft itself	PRP group demonstrated significantly better anteroposterior knee stability than control group: Calculated improvements in knee stability at 6 mo were 1.3 ± 1.8 mm in the control group and 3.1 ± 2.5 mm in the platelet gel group (P = 0.011)
-	Valentí Azcárate et al[24]	Randomized trial (PRP vs PRGF vs control) 150 patients (50 vs 50 vs 50)	ACL reconstruction using a patellar tendon allograft transtibial technique fixed with a RigidFix technique (DePuy Mitek,) with two biodegradable cross pins at the femoral bone and a tibial biodegradable (Byocril) interference screw PRP was produced with a double-spin procedure using a standard centrifuge and applied covering the ligament and suturing it over itself with gel in its interior, and introducing the gel obtained after activating the poor platelet concentration after implantation of the graft inside the tibial tunnel PRP was produced following Anitua’s technique (PRGF-Endoret Technology) and applied injecting it into the graft before implantation, with the biocompatible fibrin applied into the tibial tunnel at the end of surgery	No significant differences in functional results at the final follow-up of 24 m No statistically significant differences between the three groups in CRP 1 and VAS 24 h after surgery No significant differences in the range of knee motion, muscle torque, KT-1000 or IKDC score The PRGF group showed a statistically significant improvement in swelling scores 24 h after surgery compared with the PRP and control groups
	Vadalà et al[25]	Randomized trial (PRP vs control) 40 patients (20 vs 20)	ACL reconstruction with hamstrings (Out-In technique), fixed with Swing-Bridge device on femoral side and Evolgate screw on tibial side PRP was produced with Fast Biotech kit (MyCells PPT-Platelet Preparation Tube) and applied as follows: 5 mL between peripheral part of the graft and the femoral tunnel wall; 5 mL in semisolid pattern above the graft; 5 mL of liquid and semisolid PRP on the tibial side	Physical examination as well as the evaluation scales used showed no differences between the two groups at 14.7 mo of follow-up
	Nin et al[26]	Randomized trial (PRP vs control) 100 patients (50 vs 50)	ACL reconstruction with patellar tendon allograft, fixed with two biodegra-dable cross pins in femur and a tibial biodegradable interference screw Ligament was covered with PRP (produced with standard centrifuge) and sutured over itself with PRP in its interior. The rest of the gel was introduced after implantation of the graft inside the tibial tunnel	The results did not show any statistically significant differences between the groups for inflammatory parameters, magnetic resonance imaging appearance of the graft, and clinical evaluation scores after 18 mo
	Ventura et al[27]	Randomized trial (PRP vs control) 20 patients (10 vs 10)	ACL reconstruction with quadruple hamstring tendon graft, with a femoral transcondylic fixation (BioTransFix) and tibial interference screw (BioRCI, Smith and Nephew) PRP was obtained according to the GPS Biomet-Merck technique (Biomet) and applied in femoral and tibial tunnels	There were no significant differences concerning clinical score and examination between the two groups 6 mo after ACL surgery
Tunnel enlargement
+	Starantzis et al[28]	Randomized trial (PRP vs control) 51 patients (25 vs 26)	ACL reconstruction with hamstring tendons (Semitendinosus and Gracilis) as a quadrupled graft, using distal fixation in the femur (Crosspin Linvatec or Endobutton Linvatec) and tibial fixation with a biodegradable interference screw (Linvatec) plus bone bridge suture anchoring Half of the PRP (produced using the Biomet GPS III kit) was added between the strands of the graft and left to form a clot before the graft fixation. Then, the remaining 3 mL was injected into the femoral tunnel using an introducer	The morphology of the dilated tunnels was conical in both groups There was a statistical significant difference in the mid distance of the tunnels between the two groups 1 yr after surgery No significant difference of the Lysholm scores between the two groups during the observation period was detected
	Darabos et al[22]	Randomized trial (ACS vs control) 62 patients (31 vs 31)	ACL reconstruction with hamstring (30) or patellar tendon (32), fixed with BioTransFix (Arthrex) or RigidFix (Mytek) at femoral side, and with an interference screw at the tibial side ACS was produced drawing venous blood into syringes containing pre-treated glass beads, and after a period of incubation serum was extracted through centrifugation. ACS was administered with an injection regime of four injections on day 0 (day of surgery), day 1, day 6, and day 10	Bone tunnel enlargement measured with CT scans was significantly less (6 mo: 8%, 12 mo: 13%) in ACS group than in control group (6 mo: 31%, 12 mo: 38%)
-	Vadalà et al[25]	Randomized trial (PRP vs control) 40 patients (20 vs 20)	ACL reconstruction with hamstrings (Out-In technique), fixed with Swing-Bridge device on femoral side and Evolgate screw on tibial side PRP was produced with Fast Biotech kit (MyCells PPT-Platelet Preparation Tube) and applied as follows: 5 mL between peripheral part of the graft and the femoral tunnel wall; 5 mL in semisolid pattern above the graft; 5 mL of liquid and semisolid PRP on the tibial side	The use of PRP did not seem to be effective in preventing tunnel enlargement at 14.7 mo of follow-up
	Mirzatolooei et al[29]	Randomized trial (PRP vs control) 46 patients (23 vs 23)	ACL reconstruction with single-bundle quadrupled autograft of hamstrings, fixed with a cross-pin in femoral tunnel and a bio-absorbable interference screw in tibial tunnel Graft was immersed in the PRP solution (produced with Double syringe system, Arthrex) for five minutes before implantation; 2 mL of PRP was injected into the femoral tunnel and 1.5 mL into the tibial tunnel at the end of the surgery	Despite slightly less tunnel widening in the PRP group, there were no significant differences at any of the sites of measurement between immediately after surgery and three months post-operatively
	Silva et al[30]	Randomized trial (4 groups) 40 patients (10 control vs 10 PRP in FT vs 10 PRP in FT and intra-articular at 2- and 4 wk vs 10 PRP activated with thrombin in FT)	Double-bundle arthroscopic ACL reconstruction with autologous hamstring tendons, fixed with two Endobuttons for the AMT and PLT in the femur and two bioabsorbable interference screws in the tibia PRP (produced with GPS III Kit, Biomet) was placed between the strands of the graft in each femoral tunnel	At 3 mo postoperatively, all tunnels had enlarged compared to the diameter of the drill and most tunnels enlarged more in the midsection than at the aperture in a fusiform manner The use of growth factors during and after surgery did not show any influence in the tunnel enlargement (P = 0.563)
	Orrego et al[31]	Randomized trial (4 groups) 108 patients (27 control vs 26 PC vs 28 BP vs 27 PC + BP)	ACL reconstructions with quadruple semitendinosus-gracilis graft, fixed with a biodegradable transfixing pin proximally and a biodegradable inter-ference screw distally; BP was placed by interference fit at the femoral tunnel Five milliliter PRP (produced with Biomet GPS II kit, Biomet) was added between the graft strands before passing it into the tunnel. After fixation, 1 mL of PRP was injected into the femoral tunnel between the graft strands	The use of PC did not show any significant effect in the tunnel widening evolution at 6 mo follow-up The use of a BP effectively prevented tunnel widening The BP and PC combination did not show a synergic effect as compared to PC or BP individually
ACL graft-bone interface/integration
+	Rupreht et al[32]	Randomized trial (PRP vs control) 41 patients (21 vs 20)	ACL reconstruction with double-looped semitendinosus and gracilis tendon autograft, fixed with two bioabsorbable cross pins in femoral tunnel and one bioabsorbable interference screw in tibial tunnel. PRP was applied after autograft positioning, into the femoral and tibial tunnels (1 mL in each of them), and onto the graft itself (3 mL)	A gradual increase in the percentage of the tunnel wall consisting of tunnel wall cortical bone (TCB) during the follow-up was observed. At 6 mo the mean percentage of TCB was significantly higher (P = 0.003) in the PRP group than in the control group
	Vogrin et al[33]	Randomized trial (PRP vs control) 41 patients (21 vs 20)	ACL reconstruction with double-looped semitendinosus and gracilis tendon graft fixed with two bioabsorbable cross pins in the femoral tunnel and one bioabsorbable interference screw in the tibial tunnel PRP was produced with Magellan system (Medtronic) and applied into the femoral and tibial tunnels as well as onto the graft itself	After 4-6 wk, there was a significantly higher level of vascularization in the osteoligamentous interface in PRP group (0.33 ± 0.09 vs 0.16 ± 0.09, P < 0.001) In the intra-articular part of the graft, there was no evidence of revascularization in either group
-	Del Torto et al[21]	Prospective comparative study (PRFM vs control) 28 patients (14 vs 14)	ACL reconstruction with hamstring tendon graft fixed in the femoral tunnel through the RIGIDFIX system (DePuy) and in the tibial tunnel through the Bio-INTRAFIX system (DePuy) PRFM was prepared using Cascade Medical Enterprises 2 tube kit (Cascade Medical Enterprises). PRFM clot was sutured in the proximal graft loop and it reaches the proximal tunnel once the graft is pulled in place. Distally, the PRFM clot was inserted between the four strands of the G-ST graft before the interference screw system was applied	MRI evaluation considering graft–tunnel interface and graft signal intensity provided similar results between the two examined groups, without any statistically significant difference. In the majority of the cases, a good signal quality of the graft and a scarce film of synovial fluid at the graft–tunnel interface were observed
	Silva et al[17]	Randomized trial (BMC vs control) 43 patients (20 vs 23)	ACL reconstruction with double-looped semitendinosus and gracilis tendon autograft fixed with a Toggleloc Ziploop (Biomet) in femoral tunnel and a bioabsorbable interference screw (Biomet) in tibial tunnel Bone marrow was harvested from the iliac crest and concentrated to obtain 3 mL BMC. 1.5 mL of BMC concentrate was injected inside the femoral end of the graft itself before graft positioning, and the remaining 1.5 mL BMC injected within the tunnel around the graft, from the bottom down to the entrance of the tunnel	Adult non-cultivated BMC did not seem to accelerate graft-to-bone healing in ACL reconstruction: No difference in the signal-to-noise ratio of the inter-zone on MRI between the experimental and the control group 3 mo after surgery
	Valentí Azcárate et al[24]	Randomized trial (PRP vs PRGF vs control) 150 patients (50 vs 50 vs 50)	ACL reconstruction using a patellar tendon allograft transtibial technique fixed with a RigidFix technique (DePuy Mitek) with two biodegradable cross pins at the femoral bone and a tibial biodegradable (Byocril) interference screw PRP was produced with a double-spin procedure using a standard centrifuge and applied covering the ligament and suturing it over itself with gel in its interior, and introducing the gel obtained after activating the poor platelet concentration after implantation of the graft inside the tibial tunnel PRP was produced following Anitua’s technique (PRGF-Endoret Technology) and applied injecting it into the graft before implantation, with the biocompatible fibrin applied into the tibial tunnel at the end of surgery	No statistically significant differences were noted between groups in intensity, thickness, or uniformity of graft at 6 mo MRI
	Figueroa et al[34]	Comparative study (PRP vs control) 50 patients (30 vs 20)	ACL reconstruction with hamstring tendons fixed with a cross-pin in femoral tunnel and a bio-absorbable interference screw in tibial tunnel PRP was produced with Magellan Magellan system (Medtronic) and applied under arthroscopy in both the tibial (3 mL) and femoral (3 mL) tunnels with a long needle syringe, and directly applied in the intra-articular graft portion (4 mL)	No statistically significant benefit in the PRP group in terms of integration assessment at 6 mo follow-up
	Silva et al[35]	Randomized trial (4 groups) 40 patients (10 control vs 10 PRP in FT vs 10 PRP in FT and intra-articular at 2- and 4 wk vs 10 PRP activated with thrombin in FT)	Double-bundle arthroscopic ACL reconstruction with autologous hamstring tendons, fixed with two Endobuttons for the AMT and PLT in the femur and two bioabsorbable interference screws in the tibia PRP (produced with Mini GPS III Kit, Biomet) was placed between the strands of the graft in each femoral tunnel	The graft integration was not complete at 3 mo after surgery in the PL and AM femoral tunnel, and the use of PRP isolated or with thrombin seemed not to accelerate tendon integration
	Orrego et al[31]	Randomized trial (4 groups) 108 patients (27 control vs 26 PC vs 28 BP vs 27 PC + BP)	ACL reconstructions with quadruple semitendinosus-gracilis graft, fixed with a biodegradable transfixing pin proximally and a biodegradable interference screw distally; BP placed by interference fit at the femoral tunnel 5 mL PRP (produced with GPS II kit, Biomet) was added between the graft strands before passing it into the tunnel. After fixation, 1 mL of PRP was injected into the femoral tunnel between the graft strands	The use of PC did not show any significant effect in the osteoligamentous interface at 6 mo follow-up
ACL graft remodeling
+	Seijas et al[36]	Randomized trial (PRP vs control) 98 patients (49 vs 49)	ACL reconstruction with autologous patellar tendon grafts with 9 mm bone plugs and fixed with hydroxyapatite screws in femur and tibia 8 mL of PRP was produced with PRGF technique (BTI Systems Vitoria, Spain) and percutaneously injected into the suprapatellar joint after portal suture	More patients in the PRP group than controls attained higher stages of remodeling at month 4 (P = 0.003), month 6 (P = 0.0001), and month 12 (but NS P = 0.354)
	Rupreht et al[37]	Randomized trial (PRP vs control) 41 patients (21 vs 20)	ACL reconstruction with double-looped semitendinosus and gracilis tendon autograft, fixed with two bioabsorbable cross pins in femoral tunnel and one bioabsorbable interference screw in tibial tunnel PRP was applied after autograft positioning, into the femoral and tibial tunnels (1 mL in each of them), and onto the graft itself (3 mL)	MRI measurements indicated a reduced extent of edema during the first postoperative month as well as an increased vascular density and microvessel permeability in the proximal tibial tunnel at 1 and 2.5 postoperative months as the effect of the application of PRP
	Radice et al[38]	Comparative study (PRP vs control) 50 patients (25 vs 25)	ACL reconstructions with BPTB autograft (15 vs 10) or Harmstring (10 vs 15). Fixation in BPTB autograft with metallic interference screws, in hamstring autograft with metallic or bioabsorbable cross-pin in the femur and a bioabsorbable screw with a metallic staple in the proximal tibia PRP (produced with GPS III Kit, Biomet) was administered with the help of a sutured and compressed Gelfoam; 5 mL PRP was added homogeneously so as to completely cover the graft	ACL reconstruction with the use of PRPG achieved complete homogeneous grafts assessed by MRI, in 179 d compared with 369 d for ACL reconstruction without PRPG. This represented a time shortening of 48% with respect to ACL reconstruction without PRPG
	Sánchez et al[39]	Comparative study (PRP vs control) 37 patients (22 vs 15)	ACL reconstruction with hamstring tendons, fixed with transcondylar screw proximally and PRGF-treated bone plug and two metal staples distally 6 mL PRP was produced with BTI System II (BTI Biotechnology Institute) and injected within the tendon graft fascicles with several punctures performed along the graft length, graft soaked in PRP until implantation and the remaining aliquots applied at the portals during suturing	PRGF resulted in more mature tissue than controls at histology (P = 0.024) Histologically evident newly formed connective tissue enveloping the graft was present in 77.3% of PRGF-treated grafts and 40% of controls Overall, arthroscopic evaluations were not statistically different between PRGF and control groups (P = 0.051)
	Orrego et al[31]	Randomized trial (4 groups) 108 patients (27 control vs 26 PC vs 28 BP vs 27 PC + BP)	ACL reconstructions with quadruple semitendinosus-gracilis graft, fixed with a biodegradable transfixing pin proximally and a biodegradable inter-ference screw distally; BP placed by interference fit at the femoral tunnel 5 mL PRP (produced with Biomet GPS II kit, Biomet) was added between the graft strands before passing it into the tunnel. After fixation, 1 mL of PRP was injected into the femoral tunnel between the graft strands	The use of PC had an enhancing effect on the graft maturation process evaluated only by MRI signal intensity at 6 mo follow-up
	Ventura et al[27]	Randomized trial (PRP vs control) 20 patients (10 vs 10)	ACL reconstruction with quadruple hamstring tendon graft, with a femoral transcondylic fixation (BioTransFix, Arthrex) and tibial interference screw (BioRCI, Smith and Nephew) PRP was obtained according to the GPS Biomet-Merck technique (Biomet) and applied in femoral and tibial tunnels	CT highlighted a significant difference (P < 0.01) between ACL density of the two groups and showed that ACL density was similar to that of the posterior cruciate ligament in GF-treated group at 6 mo follow-up
-	Figueroa et al[34]	Comparative study (PRP vs control) 50 patients (30 vs 20)	ACL reconstruction with hamstrings fixed with a femoral cross-pin and a tibial bio-absorbable interference screw PRP was produced with Magellan Magellan system (Medtronic, Minneapolis, MN) and applied under arthroscopy in both the tibial (3 mL) and femoral (3 mL) tunnels with a long needle syringe, and directly applied in the intra-articular graft portion (4 mL)	No statistically significant benefit in the PRP group in terms of graft maturation (ligamentization) at 6 mo of follow-up
	Vogrin et al[33]	Randomized trial (PRP vs control) 41 patients (21 vs 20)	ACL reconstruction with double-looped semitendinosus and gracilis tendon graft fixed with two bioabsorbable cross pins in the femoral tunnel and one bioabsorbable interference screw in the tibial tunnel PRP was produced with Magellan system (Medtronic, Minneapolis, MN) and applied into the femoral and tibial tunnels as well as onto the graft itself	After 4-6 wk, significantly higher level of vascularization in the osteoligamentous interface in PRP group (0.33 ± 0.09 vs 0.16 ± 0.09, P < 0.001). No evidence of revascularization in the intra-articular part in either group
ACL repair
+	Centeno et al[18]	Prospective study (BMC, PRP, PL) 10 patients	Pre-injection of hypertonic dextrose solution into the ACL using fluoroscopic guidance 2-5 d prior to BMC injection in order to prompt a brief inflammatory response in the ACL Intra-ligamentous injection of autologous BMC (harvested from iliac crest and isolated through	Patients included had ACL laxity on exam, and MRI evidence of grade 1, 2, or 3 ACL tears < 1 cm retraction 7/10 patients showed improvement in objective measures of ACL integrity in their post-procedure MRIs
			centrifugation), PRP and PL (prepared from venous blood via centrifugation and recentrifugation after freezing) using fluoroscopic guidance. Remaining BMCs were injected into the joint	The mean VAS change was a decrease of 1.7 (P = 0.25), the mean LEFS change was an increase of 23.3 (P = 0.03), and mean reported improvement was 86.7%
	Seijas et al[19]	Retrospective study (PRGF-Endoret) 19 patients	PRGF-Endoret was produced using the technique described by Anitua and applied with a spine needle in both the proximal origin of the bundle and in the middle portion thereof in an amount of about 4 cc At the end of the surgery another injection of PRGF-Endoret was administered (6 cc) in the articular space.	16/19 professional soccer players with partial ACL tears returned to the same level Normal KT-1000 values in all operated cases Time to return to play: 16.2 ± 1.4 wk for Tegner 9 pts, 12.3 ± 1.1 for Tegner 10

PRP: Platelet rich plasma; ACL: Anterior cruciate ligament; PRFM: Platelet rich fibrin matrix; ACS: Autologous conditioned serum; PC: Platelet concentrate; BP: Bone plug; FT: Femoral tunnel; PLT: Posterolateral tunnel; AMT: Anteromedial tunnel; BPTB: Bone-patellar; PRGF: Plasma preparation rich in growth factors; BMC: Bone marrow concentrate; PL: Platelet lysate; VAS: Pain visual analog scale; NS: Non-significant; LEFS: Lower extremity functional scale.