Improvement Trajectories in Patient-Reported Outcomes Between Males and Females After Anterior Cruciate Ligament Reconstruction

Patient-reported outcomes (PROs) are critical for monitoring recovery and the effectiveness of an intervention for improving pain, symptoms, function, and quality of life (QOL) after joint injury.¹ Worse PROs after anterior cruciate ligament reconstruction (ACLR) have been associated with deleterious biological joint tissue changes,^2,3 suggesting that PROs are an easily implemented and important clinical indicator of the underlying development of posttraumatic osteoarthritis (PTOA), which is a common long-term condition in those with ACLR.⁴ Therefore, understanding the factors related to changes in PROs after ACLR is important for developing the most effective interventions to maximize long-term clinical outcomes. Previous authors have demonstrated that females were at higher risk of sustaining anterior cruciate ligament (ACL) injury,⁵ and recent health insurance data suggested that females were also more likely than males to be diagnosed with PTOA within 5 years of ACLR.⁶ Females differed in their symptomatic presentation and overall response to musculoskeletal diseases compared with males⁷; therefore, sex should be considered a factor in clinical rehabilitation outcomes, including self-reported progress. Further, whether sex influences improvements in PROs over time after ACLR remains unclear. Thus, understanding the effect of sex on changes in PROs after ACLR is important for developing sex-specific rehabilitation strategies to maximize long-term health outcomes after ACLR.

Approximately 43% of patients continued to describe clinically relevant knee-related symptoms at 2 years after ACLR, and the prevalence of knee-related symptoms at the 6-year follow-up examination was relatively similar to that at the 2-year follow-up (ie, 39%).⁸ Similarly, QOL assessed with the Knee injury and Osteoarthritis Outcomes Score (KOOS) improved between the preoperative and 2 years post-ACLR timepoints, yet KOOS scores remained stable at the 2-, 6-, and 10-year post-ACLR follow-up examinations.⁹ Overall, these data suggest that the greatest improvements in PROs occurred in the first 2 years after ACLR and then plateaued after the 2-year post-ACLR follow-up examination. Still, whether males and females demonstrate similar trajectories of improvement after ACLR is unknown.

Conclusions drawn from earlier studies evaluating sex differences in PROs after ACLR are ambiguous. Researchers¹⁰ who conducted a systematic review of 13 studies found no clinically meaningful PROs differences between the sexes; nonetheless, the included studies were performed at various timepoints between 2 and 7 years post-ACLR, and the authors used inconsistent survey measures. Similarly, investigators¹¹ of a separate systematic review observed no conclusive sex differences for health-related QOL in a subanalysis of 5 studies examining data collected from cross-sectional studies between 5 and 16 years post-ACLR. However, these systematic reviews included cross-sectional studies of various sample sizes and a variety of PROs to measure QOL at different timepoints post-ACLR, which may have limited the ability to detect sex differences in the trajectories over time. Conversely, others^12,13 have determined that females consistently scored worse on the KOOS Sport and Recreation subscale preoperatively and at 1, 2, and 5 years post-ACLR, indicating that sex may influence changes in PROs over time. Hence, we need to evaluate the influence of sex on the association between PROs and time post-ACLR at clinically relevant timepoints post-ACLR using a consistent PRO measure. Understanding the influence of sex on the trajectory of improvement in PROs over time is important to identify patients who are at risk of long-term, suboptimal QOL and develop personalized intervention strategies to optimize their outcomes.

The purpose of our study was to assess the effect of sex on the association between months post-ACLR and KOOS QOL scores in a cross-sectional sample of individuals with ACLR. We hypothesized that sex would influence the association between the KOOS QOL score and time post-ACLR, with males demonstrating a stronger association between a better KOOS QOL score at more months post-ACLR than females. The secondary purpose was to characterize sex differences in the KOOS QOL score at 6, 12, 24, 36, 48, and 60 months post-ACLR. We proposed that females would report lower mean KOOS QOL scores than males at each timepoint.

METHODS

Participants

Participants were recruited from a variety of sources, including the community, recreational clubs, intercollegiate athletics, and a local orthopaedic clinic, to maximize the generalizability of our cohort. We included all participants from the aforementioned studies who were between 16 and 35 years of age and within 6 to 60 months of primary, unilateral ACLR. Excluded were individuals with a history of lower extremity orthopaedic surgery before or after ACLR and those who required multiligament reconstruction at the time of ACLR, were pregnant or planned to become pregnant, or had radiographic evidence of knee OA at the time of testing. All participants had completed formal physical therapy and returned to unrestricted physical activity at the time of study enrollment. Demographic information including age, sex, body mass index (BMI), and months post-ACLR are provided in Table 1. Bins were created to visualize sex-specific frequencies of months post-ACLR at common clinical timepoints (Figure 1).

Table 1 Descriptive Statistics by Sex (Mean ± SD)

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

Primary Analysis: Time Post-ACLR and Sex

In the primary model, a significant interaction existed between time post-ACLR and sex (β = −0.46, P < .01). More months post-ACLR were moderately associated with a better KOOS QOL score for males (R² = 0.29, β = 0.69, P < .001), whereas females exhibited a weaker association (R² = 0.04, β = 0.23, P = .02; Figure 2). The interaction between time post-ACLR and sex remained associated with the KOOS QOL score after controlling for age (P = .004) and BMI (P = .003; Table 2).

View Figure

• Download as Powerpoint
• Download Figure

Table 2 Regression Model Comparison for Sensitivity Analysis β (Standard Error)

View Fullscreen

Secondary Analysis: Sex Differences in the KOOS QOL Score at 6, 12, 24, 36, 48, and 60 Months Post-ACLR

Males demonstrated greater estimated KOOS QOL scores at 36 months (80.37 ± 2.94 versus 70.47 ± 2.11, respectively; P < .01), 48 months (88.63 ± 4.13 versus 73.20 ± 3.01, respectively; P < .01), and 60 months (96.90 ± 5.42 versus 75.93 ± 4.03, respectively; P < .01) post-ACLR than females (Table 3; Supplemental Figure 2). No differences were observed at 6, 12, or 24 months post-ACLR, but it is noteworthy that females had greater estimated KOOS QOL scores than males at 6 months (females = 63.64 ± 2.00 versus 59.71 ± 2.26, respectively; P= .19) and 12 months (65.01 ± 1.68 versus 63.84 ± 1.96, respectively; P = .65; Table 3). Males displayed greater estimated KOOS QOL scores than females at 24 months (72.10 ± 2.06 versus 67.74 ± 1.54, respectively; P = .09; Table 3).

Table 3 Model-Estimated Marginal Means (Standard Error) for Knee injury and Osteoarthritis Outcome Score Quality of Life Scores at Selected Timepoints

View Fullscreen

DISCUSSION

Consistent with our primary hypothesis, males showed greater improvements in KOOS QOL with increasing time post-ACLR, whereas females demonstrated a weak association between improvements in KOOS QOL score with increasing time post-ACLR, suggesting a plateau in QOL score over time. Furthermore, females exhibited lower KOOS QOL scores at 36 months post-ACLR, indicating that these data were collected well after most individuals had completed formal rehabilitation and returned to unrestricted activity.¹⁷ We chose to focus our primary analysis on KOOS QOL score, as KOOS QOL is the most responsive subscale,¹⁴ yet our post hoc findings provided overall support for the results of the primary analysis. Specifically, males displayed greater improvements than females on the KOOS Sport and Recreation and Symptoms subscales with more time post-ACLR. These differences in PROs suggest that males and females had different long-term recovery trajectories after ACLR. Previous researchers¹⁸ assessing PROs in patients with arthroscopic meniscal repair noted the minimally important clinical difference for the KOOS QOL score was 16.9. Our model-estimated marginal means indicated sex differences of 15.43 and 20.67 on the KOOS QOL subscale at 48 and 60 months post-ACLR, respectively. Therefore, the KOOS QOL sex differences in our study tended to be most clinically meaningful at the later timepoints post-ACLR. This cross-sectional analysis is important in justifying the need to evaluate sex differences using longitudinal studies and to understand the unique combination of biological and functional mechanisms that may contribute to these sex differences. Our results signal that the biological variable of sex must be considered in future analyses of PROs after ACLR. Finally, future researchers may seek to develop specific interventions to optimize long-term QOL and improve women's health after ACL injury and ACLR.

Although the mechanisms causing sex differences in PROs after ACLR remain unclear, authors^19–23 have described differences in the functional, psychological, and biological responses to ACLR between the sexes, which may contribute to the sex differences we identified in PROs. Adequate quadriceps function is critical for physical performance, as diminished quadriceps function impairs function in daily activities and complex, sport-specific tasks after ACLR.²⁴ Quadriceps weakness is also associated with worse PROs^25,26 and deficits in clinical outcomes such as hop testing^27,28 after ACLR. Females had poorer quadriceps function than males in the first 12 months post-ACLR,^19,20 and their performance on the single-legged hop in the ACLR limb between 1 and 2 years post-ACLR was worse than in males.²¹ Increased difficulty for females in regaining sufficient quadriceps strength at the same clinical timepoints as males and possible sex differences in the psychological response post-ACLR²² may contribute to greater interlimb asymmetries in functional performance and influence the ability to return to sport.²⁹ Therefore, future investigators may pursue methods to specifically optimize strength in females after ACLR to ensure that they achieve similar functional performance benchmarks as males. Additionally, biological responses to ACL injury and ACLR differ between males and females, which may also contribute to worse KOOS QOL scores after ACLR for females. In particular, higher urinary concentrations of biomarkers of type II cartilage breakdown (CTX-II) are associated with the development and progression of knee OA.³⁰ Higher concentrations of urinary CTX-II in females at the time of ACLR are more strongly associated with worse KOOS QOL scores at 2 years post-ACLR than in males.²³ Further, we should continue to evaluate differences in the biological response to injury between sexes to determine if sex-specific biological factors may improve outcomes after ACLR in females. Sex-specific progressions have been advocated in functional rehabilitation for patients with total knee arthroplasty³¹; similarly, females may benefit from extended rehabilitation post-ACLR to address persistent KOOS QOL deficits.

Our cross-sectional study was unique, as we assessed sex differences in KOOS QOL scores at different timepoints throughout the first 60 months post-ACLR; however, several limitations should be noted to inform future research. The KOOS was initially validated in a sample of 9 men and 12 women who underwent ACLR.¹⁴ To our knowledge, no inherent sex biases have been reported for the KOOS QOL subscale; nonetheless, earlier authors³ suggested that the biological process of pain and experience of pain differs between the sexes² and may contribute to the sex differences in self-reported KOOS QOL scores we identified. Differences in how males and females experience and report pain after ACLR are understudied and would benefit from additional examination. Males exhibited greater BMI than females (Table 1); yet when included as a covariate, BMI did not influence the relationship between months post-ACLR and KOOS QOL scores. Male participants (22.48 ± 4.41 years) were older than females (20.20 ± 2.80 years), and younger age has been associated with a greater risk of reinjury within 5 years.³³ The age difference between the sexes was consistent with prior literature³⁴ that supported males typically being older at the time of ACL injury; after controlling for age, no change in the interaction effect was found. Further, earlier investigators determined that females participated in less physical activity after ACLR than healthy control individuals.³⁵ Future authors should assess the sex-specific effect of physical activity on PROs after ACLR. Finally, our study was cross-sectional, and we did not longitudinally evaluate how the KOOS QOL score changed over time in each person. We also acknowledge that our sample consisted of 61% female ACLR patients, and the sample was specifically skewed toward more female participants in the range of 25 to 36 months post-ACLR. A longitudinal cohort is needed if we are to understand sex differences in long-term QOL that may be time dependent. Future researchers should identify mechanisms that influence sex-specific recovery progress post-ACLR.

A higher KOOS QOL score was more strongly associated with greater months post-ACLR in males than in females. Females had lower KOOS QOL scores between 3 and 5 years post-ACLR. These differences in PROs suggest that males and females have different trajectories for recovery after ACLR, with males more likely to continue to improve over time, whereas females may tend to plateau. Clinicians should consider that rehabilitation progress may differ between males and females, with the possibility that females require extended rehabilitation or different clinical interventions after ACLR to maximize improvements in long-term QOL.