The study, published in the journal Advances in Therapy, drew on a global federated health research network called TriNetX, which pools de-identified electronic medical records from healthcare organisations worldwide. The researchers, led by Aikaterini Eleftheriadou of the Frimley Knee Unit in the United Kingdom and the University of Liverpool, identified 73,951 patients who underwent primary ACL reconstruction between 5 August 2005 and 4 August 2025 across 53 healthcare organisations. Of these, 990 patients had diabetes, either type 1 or type 2, corresponding to a prevalence of just 1.3 percent among individuals undergoing the procedure.

Anterior cruciate ligament injury is among the most common and consequential ligamentous injuries of the knee, and an estimated 400,000 reconstructions are performed annually in the United States alone, with numbers rising steadily over the past two decades. Revision surgery currently stands at around three to four percent, and revisions are linked to poorer outcomes than primary procedures, including higher failure rates and lower patient-reported scores. While biological, anatomical and technical risk factors for re-rupture are increasingly well understood, the impact of comorbidities such as diabetes has remained underinvestigated, even as the condition now affects roughly ten percent of the global population and continues to climb.

Diabetes is well known to raise perioperative risks across many surgical specialties. It is associated with surgical site infections, delayed recovery, greater healthcare utilisation, and increased hospital admission and readmission. Previous smaller studies had suggested that patients with diabetes undergoing ACL reconstruction face higher rates of postoperative complications, and one retrospective analysis reported that patients with a preoperative HbA1c of 6.7 percent, approximately 50 mmol/mol, or higher were substantially more likely to require repeat surgical intervention for complications. What remained unclear was whether these short-term risks translate into a genuinely higher likelihood of the reconstructed ligament failing and needing revision.

To answer that question, the team constructed their cohorts with considerable technical care. Because TriNetX does not contain a specific procedure code for revision ACLR, and the CPT code 29888 used for the reconstruction does not specify laterality, the researchers combined procedure codes with laterality-specific diagnosis codes. Patients were required to have a unilateral ACL tear diagnosis, for example ICD-10-CM S83.511 for the right knee, while excluding the corresponding code for the opposite knee. Patients with bilateral ACL injuries were therefore excluded at cohort creation. This meant that any further ACL reconstruction recorded during follow-up could be confidently interpreted as a revision of the previously operated, ipsilateral knee rather than a new injury on the other side.

To minimise confounding, the investigators used propensity score matching, pairing each patient with diabetes with a nearest-neighbour patient without diabetes based on age, sex and body mass index, with a calliper of 0.1 pooled standard deviations and without replacement. After matching, 988 patients were analysed in each cohort, with baseline characteristics well balanced. Time-to-event outcomes were evaluated using Kaplan-Meier survival analysis, estimated at daily intervals, with individuals lost to follow-up censored at their last recorded entry to avoid bias. Cox proportional hazards regression then produced hazard ratios with 95 percent confidence intervals, and the proportional hazards assumption was checked using chi-squared tests and visual inspection of the survival curves for crossing.

The headline result was striking in its neutrality. Revision ACL reconstruction occurred in 30 of the 988 matched patients with diabetes and in 38 of the 988 matched patients without diabetes. The hazard ratio was 0.77, with a 95 percent confidence interval of 0.47 to 1.26, indicating no statistically significant difference in revision risk between the cohorts. Mean follow-up was approximately 2.8 years for patients with diabetes and 2.4 years for those without. In plain terms, the survival curves of the two groups were statistically indistinguishable over five years, and the point estimate even leaned slightly toward lower revision risk in the diabetes group, though the confidence interval comfortably spans 1.0.

The team also ran a sensitivity analysis designed to isolate the effect of diabetes from the complexity of the surgery itself. Many ACL reconstructions are performed alongside additional soft tissue knee procedures, such as meniscectomy, meniscal repair, meniscal transplantation, or additional ligament reconstruction, identified using CPT codes 29881, 29882, 29883, 29868, 27427 and 27429. Excluding patients who underwent these concomitant operations yielded 243 matched patients per cohort. Between one and ten revision events occurred in each group, and again no statistically significant difference emerged, with a hazard ratio of 0.45 and a wide confidence interval of 0.08 to 2.45. The direction of the finding was consistent with the primary analysis, although the small number of events limited statistical power, making this result supportive rather than confirmatory.

The results align with a growing body of evidence. A recent retrospective cohort of 18,797 primary ACL reconstructions from an integrated healthcare system registry reported no significant difference in revision rates after adjustment, despite lower crude revision rates among patients with diabetes. A multicentre prospective study in the United States found no difference in ipsilateral or contralateral knee surgery over two years, and a 2026 systematic review and meta-analysis, pooling 751 patients with diabetes and 42,831 without, reported no significant difference in revision rates. Notably, the new study’s 990 patients with diabetes represent, to the authors’ knowledge, the largest individual diabetes cohort evaluating revision ACLR to date. However, the same meta-analysis confirmed that diabetes does consistently raise other postoperative risks, including infection, surgical site infection, venous thromboembolism and readmission, and a small study of 19 patients suggested that poorly controlled diabetes, rather than the diagnosis alone, may drive adverse outcomes such as arthrofibrosis and repeat surgery.

The authors are careful to spell out the limitations inherent in federated real-world data research. Propensity matching could not account for diabetes duration, severity or glycaemic control, and data on graft type and surgical technique were unavailable. A residual imbalance in body mass index remained between the matched cohorts, with mean values of 32.9 versus 31.2 kilograms per square metre, though recent evidence does not consistently support an association between elevated BMI and revision risk. Surgeons may also direct patients with poorer diabetes control toward nonoperative management, creating a healthy-user bias in which the surgical diabetes cohort represents a better-optimised subgroup. Coding accuracy, regional data predominance and incomplete five-year follow-up add further caveats. Still, the message for clinical practice is meaningful: a diagnosis of diabetes alone need not deter patients or surgeons from ACL reconstruction, though individualised risk assessment, attention to glycaemic control and counselling about the broader spectrum of postoperative complications remain essential. Prospective studies capturing HbA1c, diabetes duration, graft type and patient-reported outcomes are the next step toward truly personalised preoperative risk stratification.

Subject of Research: The Impact of Diabetes on Risk of Revision Following Anterior Cruciate Ligament Reconstruction: A Real-World Data Study

Article Title: The Impact of Diabetes on Risk of Revision Following Anterior Cruciate Ligament Reconstruction: A Real-World Data Study

Article References: The Impact of Diabetes on Risk of Revision Following Anterior Cruciate Ligament Reconstruction: A Real-World Data Study. (n.d.). https://doi.org/10.1007/s12325-026-03815-w

Image Credits: AI Generated

DOI: 10.1007/s12325-026-03815-w

Keywords: diabetes, ACL reconstruction, orthopedics, revision surgery, real-world evidence, knee surgery, Impact, Risk, Revision, Following, Anterior, Cruciate