1.1 Methods

This study was a single-center, retrospective cohort study including patients who elected to undergo ACLR and nearly 800 patients following ACLR were screened for inclusion. The current study included data from participants who suffered complications following ACLR and matched participants who underwent primary, unilateral, and uncomplicated ACLR with no complications and participated in a post-operative test battery between 5- and 12-months post-surgery. Inclusion criteria for post-operative complications included infection, cyclops lesions, arthrofibrosis, deep vein thrombi, loose bodies, fractures, wound dehiscence, and other post-operative complications following ACLR. Inclusion criteria for matched participants were primary, isolated, and uncomplicated ACLR. Matched participants were excluded if they had any prior lower extremity surgeries, surgical or post-operative complications, or any other lower-extremity injury within six months. All participants were instructed to follow a similar rehabilitation protocol. Participants were screened for subsequent injury within two years of surgery via medical chart review and follow-up phone calls. This study was approved by an institutional review board for human subjects research and all participants provided informed consent.

1.2 Data processing

Patient reported outcome scores were compared via group means. The LSI was calculated as (involved limb/uninvolved limb) ∗ 100 % for strength and hop testing outcomes.²³ Peak torque was also normalized to the patient's body mass (Nm/kg) for strength testing.

1.3 Statistical analyses

Between group differences in age, height, mass, and time post-surgery were analyzed via independent samples t-tests. Patient reported outcomes, LSI variables, and normalized peak torque were compared across groups using analyses of covariance (ANCOVA) with time post-surgery and graft type as model covariates. Dependent variables included isokinetic knee extension and flexion LSI and normalized peak torque, hop test LSI. Paired samples t-tests were conducted as post-hoc tests when indicated. Alpha level was set at p ≤ 0.05. Statistical analyses were completed in Stata (College Station, TX).²⁴ Effect sizes were interpreted as small (η2 = 0.01), medium (η2 = 0.06), and large (η2 = 0.14) effects.²⁵

1.4 Results

Of the original 788 patients with ACLR reviewed, a total of 13 participants met all criteria to be included in the cohort of patients with complications. These participants were randomly matched, 2:1 to 26 ACLR control participants on the basis of age, sex, and pre-surgical activity level (Fig. 1). Of the total sample 1.6 % of patients experienced a post operative complication, with the majority being cyclops lesions (Table 1). The groups were similar in age, height, and time post-surgery (Table 1). Further descriptions of the postoperative complications and their treatments can be Table 2. No participants following complications suffered an additional ACL injury, while six participants in the uncomplicated group suffered a subsequent injury (4 ipsilateral, 2 contralateral). Given the low number of participants with subsequent injuries, no statistical analyses were conducted.

Close

Fig. 1 Consort flow diagram of inclusion.

Export to PPT

Raw differences in patient reported outcomes, strength, hop test outcomes between participants with complications and matched participants can be seen in Table 3. The ANCOVA models indicated no significant differences in patient-reported outcomes, strength, or hop tests when controlling for time post-surgery and graft type. No post hocs were indicated. Normalized strength measures for participants following ACLR are displayed in Fig. 2.

Close

Fig. 2 Normalized strength metrics of complications group.

Export to PPT

1.5 Discussion

This study investigated commonly used metrics for tracking recovery following ACLR and decision-making regarding return to physical activity and sport. The most important findings of this study are that patients with post-operative complications following ACLR exhibited minimal and non-statistically significant differences in both PROMs and strength and hop testing LSIs compared to primary, isolated, and uncomplicated cases. In this study, IKDC, KOOS symptoms, pain, quality of life, activities of daily living, and sport scores were comparable between the complicated and uncomplicated groups, as were LSI for knee extension and flexion strength, single leg hop, triple hop, and 6-m timed hop.

In our population of patients, the types of post-operative complications varied as did the subsequent interventions for these complications. Interventions ranged from courses of oral antibiotics or anticoagulants for postoperative wound infections and DVT to follow up surgeries such as excision of cyclops lesions, loose body removals, and manipulations under anesthesia. Although patients received various interventions based on their specific complications, the overall focus of this study was on whether any post-surgical complications would alter the course of a patient's recovery and rehabilitation following ACLR. While it may have been expected that complications requiring follow up surgeries may have a larger impact on strength compared to smaller complications such as those treated with oral medications, this was not the case in our cohort. Overall, both groups demonstrated strength LSI below the clinically desired threshold of 90 %, but the additional post-surgical complications did not appear to affect strength recovery as participants approached time for clearance to return to sport.²⁶Fig. 2 shows that in our cohort there were no differences in normalized strength outcomes among the different types of post-operative complications. Further investigation into the differences in outcomes between the different types of complications was limited by the low incidence of each type of complication.

The overall lack of differences in PROMs and LSI outcomes between patients who had post-operative complications following ACLR and those who had primary, isolated, and uncomplicated cases, while surprising, is useful data for this population of patients. Similar research in patients following a single revision ACLR has found a similar lack of differences in outcomes.²⁷ While patients with post-operative complicating often have problems that need to be addressed through second surgeries or longer-term deficits, these results suggest that this population is not necessarily at an increased risk for worse patient reported outcomes or delayed limb strength recovery in the early post-operative period compared to patients without these complications. These findings also suggest that this group may not have different requirements when it comes to rehabilitation or return to sport measures and decision making compared to patients with uncomplicated ACLRs.

Several studies have looked at the incidence of various complications associated with ACLR but have not evaluated outcomes in these populations.^9,8 This is likely due to the relatively low number of patients with complications within single sites. Rates of perioperative complication in ACLR have been reported to range from 1 % to 15 %, primarily involving deep vein thrombosis (DVT), hemarthrosis, effusion, synovitis, infection, or arthrofibrosis.^3,4,5,6,7 In our sample group, 13 of the 788 ACLR cases screened for inclusion qualified as a complication following ACLR. Our complications rate of 1.6 % was similar to previous studies. In addition to these more common complications, a previous study also investigated reports of rarer complications associated with ACLR including complications related to the fixation device, fractures, rare vascular injuries, and nerve injuries but did not evaluate outcome measures in these cases.⁸

Little research has reported on outcomes of patients with complications following ACLR.²⁸ One previous study reported on outcomes of an entire cohort of patients who underwent ACLR, including both complicated and uncomplicated cases, with no isolation of outcomes for patients who had post-operative complications and no comparison in outcomes between those with post-operative complications and those without.²⁸ The IKDC scores at 12 months post-op for the cohort in this study were 83.3 ± 13.6 which was consistent with the IKDC scores for both the uncomplicated and complicated groups in our study.²⁸

Evaluating cases of complications following ACLR is important as it has been reported that post-operative complications accompanying ACLR can lead to worse outcomes including increased rates of arthritis, chronic pain, reinjury, additional surgeries, and longer rehabilitation times.⁹ Additionally, it has been reported that this population of patients is less likely to have a full recovery and return to their preinjury level activity.⁹ In this respect, return to sport measures including LSI and PROMS such as IKDC and KOOS would be useful to characterize the outcomes for this population of patients.

1.6 Clinical implications

The results of this study indicate that outcomes in patients with complications following ACLR may be comparable to outcomes in patients with uncomplicated ACLR in a post-operative period greater than six months post-surgery. As many current rehabilitation and return to sport guidelines were created based on data from uncomplicated ACLRs, there currently exists no specific guidelines for the complicated ACLR population.^14,18,16,17 This data shows that during the recovery period that patients with complications following ACLR may expect similar self-reported function, strength, and hopping ability to those with uncomplicated ACLRs. These results suggests that previously established guidelines based on uncomplicated ACLRs may be extrapolated and implemented successfully in this population.

1.7 Limitations

Complications following ACLR are rare, as such the study sample is small. While this minimized statistical power, it is representative of a sample accrued over multiple years in a large hospital system. We attempted to minimize bias through random matching and covarying our statistical models to assure that we would have the best opportunity to find any differences, if they existed between groups. However, the effect sizes for all the LSIs and PROMs investigated in this study were small or medium indicating that regardless of sample size there is little clinical difference in the outcomes between groups. Differences in the types of post-operative complications and the treatment of those complications with interventions ranging from short courses of oral medications to the need for follow up surgeries may alter the long-term outcomes of these patients. Due to the small incidence of each individual complication in out cohort it was not possible to look at how each type of complication may affect patients differently.