Cosgarea et al. [4] performed a retrospective analysis of 191 consecutive autologous BPTB ACL reconstructions and similarly to Shelbourne et al. [17] and Wasilewski et al. [19] showed that surgery performed within the first 3 weeks of injury had significantly higher incidence of arthrofibrosis compared to surgery performed later than 3 weeks from injury (21 % vs. 9 %, respectively). However, an important finding of their study was that incidence of arthrofibrosis decreased from more than 20 % to less than 3 % when postoperative rehabilitation protocol was changed from bracing in 45° flexion for 7 days before the initiation of passive extension to bracing in full extension immediately after surgery. They therefore concluded that although surgery within 3 weeks from injury may place a knee at increased risk for arthrofibrosis, postoperative splinting in full extension with immediate protected weightbearing ambulation rather than splinting the knee in flexion position is the single most important factor in preventing arthrofibrosis.

Mayr et al. [12] performed a retrospective analysis of risk factors for arthrofibrosis after ACL reconstruction in 223 patients, 75 % of which had their reconstruction with autologous BPTB graft. They also demonstrated that incidence of arthrofibrosis was increased in cases where reconstruction was performed within 4 weeks from injury, but that irritated knee (swelling, effusion, hyperthermia) and lack of full ROM before surgery were more important risk factors for the development of arthrofibrosis than time interval from injury to surgery. In other words, when surgery was performed later than 4 weeks from injury but the knee was irritated, there was as increased risk for the development of arthrofibrosis compared to when surgery was performed within the first 4 weeks from the injury.

The first prospective randomized clinical trial that investigated the incidence of arthrofibrosis in “early” versus “delayed” ACL reconstruction was performed by Meighan et al. [13]. They studied a small series of athletic patients that underwent ACL reconstruction using autologous quadrupled hamstrings graft and used similar postoperative rehabilitation protocols for both groups. The “early” reconstruction group had surgery within 2 weeks from injury, and the “delayed” group had surgery between 8 and 12 weeks from injury. Although loss of knee motion was more pronounced at 2 weeks after the operation in the “early” group, at 1-year follow-up, there were no differences in knee motion, nor there were differences between the groups in relation to IKDC, Lysholm, and Tegner scores and examination of quadriceps and hamstrings muscle power and torque.

Bottoni et al. [2] performed another prospective clinical trial for the same purpose using a larger sample. The reconstructions were performed with autologous hamstring autograft. “Early” reconstruction patients had their surgery within the first 17 days after the injury, and “late” reconstruction patients had their surgery at 6 or more weeks after the injury. Both reconstruction groups followed similar supervised rehabilitation protocols with early mobilization and emphasis on maintenance of extension. The investigators found comparable knee flexion and extension in both groups. Furthermore, no clinical differences were observed between the two groups in relation to knee stability and Lysholm and Tegner scores. The authors concluded that delaying surgery for some arbitrary period of time due to the concern of increased risk of arthrofibrosis is not necessary, although they did not recommend performing ACL reconstruction acutely.

The outcomes of “early” versus “delayed” ACL reconstruction were also investigated in a systematic review and meta-analysis by Smith et al. [18]. There were overall six studies fulfilling inclusion criteria in which ACL reconstructions were performed with either autologous patellar tendon or hamstring grafts. “Early” reconstruction was considered surgery performed within 1 month from the injury. This meta-analysis could not identify any significant differences in the incidence of arthrofibrosis or in any functional outcome score or activity level outcome scores between reconstructions performed “early” compared to those performed “late.” Of note, the authors noted the methodological limitations in previous investigations, including limited statistical power and lack of sufficient prospective, randomized data.