A propensity-matched analysis of 13,250 patients, presented at the 2026 American Academy of Orthopaedic Surgeons (AAOS) Annual Meeting, found that being on testosterone replacement therapy in the year before a total knee replacement is associated with significantly higher rates of infection, blood clots, kidney injury, and implant failure requiring revision surgery.
Knee replacement is the most common major joint replacement performed in the United States, which makes this the arthroplasty finding most men on testosterone are likely to encounter personally. If you are on TRT and have a knee replacement on the calendar, this is the data point that should change your preoperative conversation.
Key Takeaways
The 2026 AAOS study matched 13,250 total-knee-replacement patients and compared testosterone users to non-users with similar age, BMI, and comorbidities
At one year, periprosthetic joint infection was 2.4% in TRT users versus 0.9% in non-users — a 2.7-fold increase — and sepsis showed the same gap
At five years, infection reached 4.3% versus 1.9% and revision surgery rose to 4.1% versus 2.7%
Findings match the 2026 hip and shoulder replacement signals — the pattern now replicates across all three major joints
Current best practice: disclose TRT 4-6 weeks before surgery and consider a documented 3-6 month preoperative pause with hematocrit normalization
This does not change the overall TRT safety picture established by TRAVERSE — it changes the perioperative protocol
What the Knee Study Actually Showed
The study — titled "Preoperative Testosterone Replacement Therapy Is Associated with Increased Complication Risk After Total Knee Arthroplasty: A Propensity-Matched Analysis of 13,250 Patients" — pulled patients from a large national electronic health record database. Researchers identified patients undergoing primary total knee arthroplasty (TKA) who had documented testosterone use within 12 months before surgery, then matched them 1:1 against non-users with similar age, body mass index, smoking status, and comorbidity burden. The matching is what lets the analysis isolate the testosterone signal from the confounders that usually muddy hormone-outcome research.
The team tracked outcomes at three windows: 90 days, one year, and five years.
90-Day Knee Replacement Complications (TRT vs Non-Users)
Complication
TRT Users
Non-Users
Pulmonary embolism
1.6%
1.2%
Pneumonia
3.3%
1.9%
Acute kidney injury
4.2%
2.9%
Sepsis
1.9%
1.1%
One-Year Knee Replacement Complications
Complication
TRT Users
Non-Users
Pulmonary embolism
2.6%
2.0%
Deep vein thrombosis
4.5%
3.3%
Cardiac events
3.0%
2.4%
Pneumonia
6.0%
4.0%
Acute kidney injury
7.9%
5.2%
Sepsis
2.4%
0.9%
Periprosthetic joint infection
2.4%
0.9%
Periprosthetic fracture
0.7%
0.2%
Aseptic loosening
1.0%
0.5%
Instability
0.6%
0.3%
Revision surgery
1.6%
1.0%
Five-Year Knee Replacement Complications
Complication
TRT Users
Non-Users
Periprosthetic joint infection
4.3%
1.9%
Periprosthetic fracture
1.6%
0.6%
Aseptic loosening
2.7%
1.3%
Instability
1.7%
0.8%
Revision surgery
4.1%
2.7%
The numbers that stand out are the one-year periprosthetic joint infection rate (2.4% vs 0.9%, a 2.7-fold increase) and the matching one-year sepsis rate (also 2.4% vs 0.9%). By five years, the infection gap widens further — 4.3% versus 1.9% — and revision surgery, the outcome patients fear most, climbs to 4.1% versus 2.7%.
A 90-day complication is frightening but usually treatable. The part of this dataset worth thinking hardest about is the five-year prosthetic complication panel — because every hardware-related outcome was elevated, and periprosthetic joint infection (PJI) more than doubled.
PJI is one of the most consequential complications in orthopedics. It frequently requires multi-stage revision surgery, weeks to months of intravenous antibiotics, and in some cases permanent loss of function in the joint. An infection that seeds years after the implant goes in is the worst-case timeline, and that is exactly the pattern the five-year data describe.
The mechanisms are not nailed down, but the leading candidates are reasonable:
Elevated hematocrit and blood viscosity. Testosterone increases red blood cell production, often pushing hematocrit into the upper 40s or low 50s. Above roughly 52%, microvascular flow worsens and clot risk rises — which fits the elevated pulmonary embolism and deep vein thrombosis rates seen here. It may also impair microcirculation at the bone-implant interface, where good blood flow is what clears low-grade bacterial colonization. See our coverage of TRT polycythemia and hematocrit management.
Immune modulation. Testosterone affects neutrophil function and biofilm clearance. The persistent five-year PJI signal suggests sustained androgen exposure may impair the body's ability to control low-grade implant colonization before it becomes a clinical infection.
Bone remodeling at the implant interface. Periprosthetic fracture, loosening, and instability were all elevated at five years. These are mechanical failures of the bone-implant junction. Testosterone has complex effects on bone — net beneficial in most contexts, but possibly disruptive at the high-stress micro-motion zones around an implant.
Selection bias is the real caveat. This is observational data, not a randomized trial. Men who choose TRT may differ from non-users in ways propensity matching cannot fully capture — more baseline visceral adiposity, undiagnosed cardiometabolic disease, or other factors that drove them to seek hormone optimization in the first place. That is the strongest argument against over-reading the signal.
How This Fits the 2026 Surgical Picture
The knee study does not stand alone. It is one of three propensity-matched 2026 analyses that landed on the same conclusion across three different joints. We covered the hip and shoulder studies in detail in TRT before hip and shoulder surgery.
Joint
Study Size
1-Year Infection Signal
5-Year Revision Signal
Knee
13,250 matched patients (AAOS 2026)
2.4% vs 0.9% (2.7x)
4.1% vs 2.7%
Hip
Propensity-matched cohort (J. Arthroplasty 2026)
Elevated
4.1% vs 2.3%
Shoulder
1,369 matched pairs (JSES 2026)
4.8% vs 2.4% (2.0x)
7.9% vs 4.5%
Three independent databases, three different joints, three propensity-matched designs, three signals in the same direction. Any single observational study can be confounded; three convergent studies across distinct procedures are much harder to explain away. That replication is the reason this matters more than a one-off finding would.
The closely related 2025 rotator cuff repair literature adds the only timing data we have: patients whose last testosterone use was within 6 months of surgery had higher revision and opioid-use rates than those who stopped 6-12 months before. It is the one published study that quantifies a preoperative washout window — and the answer it suggests is "longer is safer."
What This Means for Your Knee Surgery Plan
If you are on TRT and have an elective knee replacement scheduled, the data argue for three concrete adjustments.
1. Disclose Your TRT at the Surgical Consult
A lot of men do not mention TRT during preoperative consults — especially if they got it through a telehealth clinic and do not think of it as a "real" prescription. It is one, and your surgical team needs to know about it the same way they need to know about a blood thinner.
Log into your clinic portal, screenshot your protocol (dose, frequency, duration of use), and bring it to the consult. If your prescriber will not provide records on request, that is a TRT clinic red flag worth addressing independently of the surgery question.
2. Get Your Hematocrit in Range Before Surgery
Most surgical centers want hematocrit below 50% — some want below 47% — for elective procedures. Given how much of this risk signal plausibly traces back to red cell mass and blood viscosity, this is the single most actionable lever. If you are running 51-54% on TRT, you have three options:
Reduce your TRT dose for 6-12 weeks preoperatively, coordinated with your prescriber
Therapeutic phlebotomy (blood donation) on a schedule your surgeon approves
Pause TRT entirely for a defined washout window — typically 3-6 months
The right lever depends on your symptoms, baseline hematocrit, and surgical timing. This is a conversation for your prescriber, not just your surgeon.
3. Build a Documented Pause-and-Resume Plan
A defensible preoperative protocol for TRT users facing knee replacement, synthesizing the 2025-2026 literature:
Time Point
Action
6 months pre-op
Surgical consult; flag TRT status; align prescriber and surgeon
3-6 months pre-op
Begin TRT taper or pause; monitor symptoms
6 weeks pre-op
Repeat CBC; confirm hematocrit below the surgical threshold
2 weeks pre-op
Final labs; confirm hematocrit, kidney function, coagulation
Surgery day
Document TRT pause duration in the chart
4-8 weeks post-op
Surgical clearance check before resuming TRT
8+ weeks post-op
Resume once wound healing is complete and infection risk has passed
The 3-6 month pause is the inconvenient part. You will likely feel libido and energy decline during the washout, especially in the final weeks. That is the trade-off the current data argue for. For the monitoring infrastructure you will want during resumption, see the TRT bloodwork schedule.
How to Talk to Your Surgeon
Many orthopedic surgeons heard the AAOS knee data when it was presented in early 2026, but few have a written protocol for it yet. Bring the reference with you. Three questions to ask at your preoperative consult:
What is your protocol for patients on testosterone replacement therapy?
What hematocrit threshold do you want before clearing me for surgery?
How long before surgery should I pause TRT, and when can I safely resume?
If your surgeon dismisses the question, it is not necessarily a deal-breaker — but it is worth pulling your TRT prescriber into the conversation. Some larger clinics have built perioperative protocols and can coordinate directly with the surgical team. For how to identify clinics that take this kind of coordination seriously, see how to choose a TRT clinic and the best online TRT clinics.
What This Data Cannot Tell You
Four limitations to keep in mind.
Observational, not randomized. Propensity matching reduces confounding but cannot eliminate it. The men who chose TRT may differ from non-users in ways no database captured.
No knee-specific washout-window data. Only the rotator cuff study has quantified preoperative timing. The 3-6 month window suggested here is extrapolated, not directly tested for knee arthroplasty. The study did define exposure as use within 12 months of surgery, which is the most conservative reading of "how far out is far enough."
All TRT modalities pooled. The analysis lumped injections, gels, pellets, and oral together. Whether risk differs by route is unknown — pellets in particular tend to run hematocrit higher than weekly injections, which may matter.
No resumption guidance. When and how to safely resume TRT after knee replacement has no randomized evidence. The 4-8 week post-op window above is consensus practice, not trial-derived.
Bottom Line
The 2026 AAOS analysis of 13,250 patients found that preoperative testosterone replacement therapy is associated with higher rates of infection, blood clots, kidney injury, and five-year implant failure after total knee replacement — replicating the same signal seen in the 2026 hip and shoulder studies. The evidence is observational, but the convergence across three joints is hard to dismiss.
For most men on TRT, this changes the perioperative protocol, not the decision to be on testosterone in the first place. A documented 3-6 month preoperative pause, hematocrit normalization, and coordination between your prescriber and surgical team is the safer path given current evidence. Randomized data on washout and resumption are still missing — until they arrive, treat this as a planning problem, not a panic signal.
Pausing TRT for a few months is rough but recoverable. A revision knee replacement is not.
References
Omurzakov A, Omurzakov AM, Chalmers BP, et al. Preoperative Testosterone Replacement Therapy Is Associated With Increased Complication Risk After Total Knee Arthroplasty: A Propensity-Matched Analysis of 13,250 Patients. Presented at the 2026 Annual Meeting of the American Academy of Orthopaedic Surgeons (AAOS); March 2026. The Journal of Arthroplasty. 2026. PMID: 41177190.
Preoperative Testosterone Replacement Therapy Is Associated With Increased Complication Risk After Total Hip Arthroplasty: A Propensity-Matched Analysis of Real-World Data. The Journal of Arthroplasty. 2026. PMID: 41519493.
Preoperative testosterone replacement therapy is associated with increased rates of periprosthetic joint infection, acute kidney injury, and emergency department utilization after total shoulder arthroplasty: a propensity-score matched analysis. Journal of Shoulder and Elbow Surgery. 2026. PMID: 40902712.
Preoperative testosterone replacement therapy: a potential risk-factor for complications and reoperation after rotator cuff repair. Sports Health. 2025. PMID: 41459022.
Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular Safety of Testosterone-Replacement Therapy (TRAVERSE). N Engl J Med. 2023;389:107-117.
AAOS 2026 Annual Meeting Press Release. New research links testosterone therapy with serious health risks after total knee replacement surgery.
What did the 2026 AAOS knee replacement testosterone study find?
A propensity-matched analysis of 13,250 patients presented at the 2026 American Academy of Orthopaedic Surgeons (AAOS) Annual Meeting found that men on testosterone replacement therapy within 12 months before total knee arthroplasty had significantly higher complication rates. At one year, periprosthetic joint infection was 2.4% versus 0.9% in non-users — a 2.7-fold increase — and sepsis was also 2.4% versus 0.9%. At five years, periprosthetic joint infection reached 4.3% versus 1.9% and revision surgery rose to 4.1% versus 2.7%.
Should I stop TRT before a knee replacement?
There is no formal guideline yet, but the converging 2025-2026 data on knee, hip, and shoulder replacement all point the same direction: preoperative testosterone raises complication risk. Most surgical teams now recommend disclosing TRT status at least 4-6 weeks before surgery and consider a documented 3-6 month pause combined with hematocrit normalization. Coordinate the decision between your surgeon, anesthesiologist, and TRT prescriber rather than stopping on your own.
How long before knee surgery should I stop testosterone?
The only timing-specific data comes from the 2025 rotator cuff repair literature, where stopping testosterone within 6 months of surgery was associated with worse outcomes than stopping 6-12 months out. Extrapolating cautiously, a 3-6 month preoperative pause is a defensible starting point for elective knee replacement. The knee study itself defined exposure as testosterone use within 12 months of surgery, so a washout longer than that is the conservative target.
Does this mean testosterone therapy is unsafe?
No. The knee data describe perioperative risk for a specific group — patients undergoing major joint replacement. It does not change the overall cardiovascular safety profile established by the TRAVERSE trial, nor does it apply to routine non-surgical TRT use. The takeaway is that knee replacement planning should account for testosterone status the same way it accounts for blood thinners, GLP-1 agonists, or other risk modifiers.
Why would testosterone raise knee replacement infection rates years later?
The five-year periprosthetic joint infection signal — 4.3% versus 1.9% — is the most striking finding because infections can seed years after the implant goes in. The mechanisms are not proven, but candidates include immune effects on biofilm clearance, elevated red cell mass and blood viscosity impairing microcirculation at the bone-implant interface, and altered bone remodeling around the implant from sustained androgen exposure.
What labs does my surgeon need before a knee replacement on TRT?
At minimum: a complete blood count focused on hematocrit (most surgeons want it below 50% before elective surgery), a comprehensive metabolic panel for kidney function, a coagulation panel, and a documented testosterone pause date. Some centers add a hematology consult if hematocrit has run above 52%. Bring a screenshot of your protocol — dose, frequency, and duration of use — to the preoperative consult.
