Arthroplasty Protocol Reference

This document sets the evidence-based perioperative pathway for primary, unilateral total hip (THA) and total knee (TKA) arthroplasty and states the rationale and key studies behind each recommendation. It is the reference companion to the one-page Quick Reference. Detailed therapy progression remains in the TKA "Quiet Knee" and THA (posterior) minimal-restriction rehabilitation protocols.

Recommendations reflect current clinical practice guidelines (AAOS, ACR/AAHKS, AHA/ACC, SHEA/IDSA/APIC, ERAS Society) and high-quality randomized and registry evidence. Each section flags where the evidence is strong versus evolving, and which long-standing practices should be abandoned.

How to Use This Document

Each domain gives the recommendation, the evidence, and a boxed "Protocol recommendation." Boxes shaded red flag practices to stop or avoid. A curated key-reference list appears at the end. Numbers (doses, thresholds, durations) are intended to become standing order-set values.

Summary of Key Recommendations

1 · Preoperative Optimization

1.1 Patient selection & risk stratification

Body mass index. No rigid BMI value should be an absolute contraindication. The 2023 ACR/AAHKS guideline conditionally recommends proceeding without delaying for weight reduction at BMI 35–39; rigid thresholds limit access without strong predictive value. BMI ≥ 40 as a test for major complications has a positive predictive value near 6.7% — roughly 14 complication-free patients denied surgery for every 1 spared a complication (Giori, JBJS 2018). For THA, risk begins rising near BMI 30; for TKA it is relatively stable until ~40. This practice's adopted candidacy threshold is BMI < 36, with shared decision-making and optimization for patients above it.

Glycemic control. This practice's adopted target is HbA1c < 7% for elective arthroplasty. (Major societies permit up to < 8%; the AAOS threshold of < 6.5% is more stringent than most advise.) Perioperative glucose goal < 200 mg/dL.

Cardiac risk. Apply RCRI or the NSQIP calculator per the 2024 AHA/ACC perioperative guideline; TJA is low-to-intermediate risk. Stress testing only for poor functional capacity (< 4 METs) at elevated risk, and only if it would change management.

Nutrition (albumin). Check serum albumin. Albumin < 3.5 g/dL predicts wound complications (OR 2.18), infection, and mortality (OR 3.17), and is a stronger predictor of major complications than morbid obesity (Nelson, CORR 2015). Optimize, and consider delaying if < 3.0 g/dL. (Expanded in Domain 2.)

Anemia. Check hemoglobin; for Hb < 13 g/dL begin iron — IV iron preferred, reducing transfusion risk 39% (RR 0.61) and length of stay ~2 days (Scrimshire, BMJ Open 2020). Oral iron given only 1–2 weeks pre-op is minimally effective.

1.2 Modifiable risk factors

Smoking cessation. Start a program ≥ 4 weeks pre-op with pharmacologic (varenicline, NRT) plus behavioral support. Surgery need not be contingent on complete cessation.

Vitamin D. Give empiric preoperative supplementation to all candidates — routine testing is not required. Deficiency is common (13–76% of candidates) and associated with more complications and longer stay; nonselective supplementation is more cost-effective than screen-and-treat, and high-dose repletion lowered total complications from 8.6% to 4.3% (Birinci, J Arthroplasty 2024). (Regimen in Domain 2.)

S. aureus decolonization. Universal decolonization (without screening) — intranasal antiseptic + chlorhexidine bathing for 5 days — lowers overall, S. aureus, and MRSA PJI, and is more cost-effective and reliable than screen-and-treat. One program cut SSI from 1.11% to 0.34% (Sporer, J Arthroplasty 2016). (Full rationale and the screening debate in Domain 5.)

1.3 Prehabilitation

Offer structured prehabilitation, especially for higher-risk patients, but set expectations: benefits are modest and time-limited. A meta-analysis of 48 RCTs showed moderate-certainty benefit for function at 6 weeks but no difference beyond 6 months (Punnoose, JAMA Netw Open 2023). A BMJ network meta-analysis found exercise (OR 0.50) and combined exercise + nutrition + psychosocial prehab (OR 0.64) most reduced complications. Prioritize patients with low baseline function, obesity, or multiple comorbidities; home-based programs are effective.

1.4 Preoperative education

Standardized education is one of the highest-value, lowest-cost interventions available. A 5-minute opioid-counseling video cut early opioid consumption ~26–28% after TKA with no change in pain scores (Maheu, J Arthroplasty 2024); a simple pain-management card halved 2-day opioid use. All patients should receive education on pain expectations, the multimodal plan, and opioid risks.

1.5 Oral & dental health

Routine, universal dental clearance is not supported by the evidence and is not required by this practice. No randomized trial shows that mandatory pre-op dental screening lowers periprosthetic joint infection (PJI). The largest series — 8,654 matched patients — found no difference in PJI rate between those who did and did not obtain dental clearance, and an identical organism profile in the infections that did occur (Kwan, Cureus 2023). A prospective cohort of 777 patients found no overall reduction in early PJI with pre-op oral screening (1.0% vs 1.6%, p = 0.455) (Fenske, J Clin Med 2023). A systematic review concluded there is “insufficient evidence to support universal dental clearance before TJA” (Frey, World J Orthop 2019). Mechanistically the signal is weak because most PJI arises from skin flora (staphylococci, including MRSA), not oral organisms.

Screen selectively; refer for cause. Dental pathology is common (prevalence 9–30%) and often silent, so patient self-report alone is imperfect — but that argues for targeted screening, not universal clearance. Refer to a dentist any patient with a symptomatic or suspected active oral infection (tooth pain, abscess, swelling, bleeding or painful gums, a loose or fractured tooth, visibly poor dentition) or at elevated infection risk (poorly controlled diabetes, immunosuppression). Where the prospective data showed a difference it favored screening by a dentist over a cursory clinic look (Fenske 2023) — so refer rather than “clear” in the office. A short patient self-screen is embedded in the surgery guides to triage who needs a visit.

Extraction is not benign, and timing matters. Pre-op extraction is not a free risk-reduction step: in a 2,457-TJA series the extraction subgroup had a higher complication rate, and the authors concluded pre-op extraction “may represent an unnecessary step” (Sonn, J Arthroplasty 2019). Extraction produces its own transient bacteremia and an open, healing socket — both undesirable close to implantation. When invasive dental work is genuinely needed, complete it and let it heal before surgery: at least 2 weeks for the mucosa, and for extractions allow the socket to heal (≥ 3 weeks is reasonable). Do not schedule elective dental work within 2 weeks of surgery. (No trial defines an exact interval; this is expert-consensus timing.)

Protocol Recommendation — Optimization

Optimize ≥ 4 weeks pre-op: HbA1c < 7.5%, albumin ≥ 3.5 g/dL, Hb ≥ 13 (M)/12 (F) with IV iron if deficient, vitamin D ≥ 30 ng/mL, smoking cessation.

BMI ≥ 40 triggers enhanced counseling, not an automatic bar. Universal S. aureus decolonization for all. Standardized education (including opioid counseling) for every patient.

Dental: no mandatory clearance. Refer only for symptomatic/suspected active oral infection or elevated-risk patients; complete needed invasive work ≥ 2 weeks pre-op (extractions: allow the socket to heal, ≥ 3 weeks); avoid elective dental work within 2 weeks of surgery.

2 · Preoperative Nutritional Optimization

Nutritional screening is among the most cost-effective yet underused interventions in arthroplasty. Universal malnutrition screening before TKA is cost-effective at an ICER of ~$6,454/QALY (Torchia, J Arthroplasty 2023), yet nutritional labs are rarely ordered. Malnutrition prevalence paradoxically rises with BMI — obese patients are vulnerable to "hidden malnutrition," and ~76% of candidates consume less than the recommended 1.2 g/kg/day of protein.

2.1 Screening

2.2 Protein & essential amino acids (strongest intervention)

Target 1.2–1.5 g/kg/day of protein starting ≥ 2 weeks pre-op through 6 weeks post-op (1.5–1.9 g/kg for malnourished/sarcopenic patients), per ESPEN surgical guidance. TKA causes rapid quadriceps atrophy — 14.3% volume loss within 2 weeks in the placebo arm of a landmark RCT. Essential amino-acid (EAA) supplementation is the best-studied intervention: 20 g EAA twice daily reduced 6-week atrophy from 18.4% to 6.2% (Dreyer, JCI 2013), and the Ranawat-Award RCT showed EAA preserved rectus femoris area and accelerated ADL recovery, with strength advantages persisting at 2 years (Ueyama, JBJS 2023).

2.3 Carbohydrate loading

Include a maltodextrin clear carbohydrate drink (~12%; sports drinks are inadequate) — one bottle the evening before and one bottle up to 2 hours before arrival, per 2023 ASA fasting guidance. Benefits are chiefly comfort (less hunger, thirst, anxiety) with a modest reduction in overall complications (OR 0.64). Use caution in type 2 diabetes; contraindicated in type 1 / insulin-dependent diabetes.

Product and purchasing. Specify Ensure Pre-Surgery Clear Carbohydrate Drink (50 g carbohydrate per 10 fl oz bottle; meets ERAS/ASA pre-op guidance). The patient buys it themselves before surgery day — it is sold on Amazon and at Walmart — so the clinic does not need to stock or dispense it. If the patient cannot find this exact product, they skip the carbohydrate drink entirely: no juice, soda, Gatorade, or other substitute. Plain clear liquids remain permitted up to 2 hours before arrival. This is the same wording used in the patient pre-op instructions.

2.4 Vitamin D, iron, vitamin C

Protocol Recommendation — Nutrition

Screen albumin and hemoglobin/iron on every candidate. Protein 1.2–1.5 g/kg/day (EAA for malnourished/sarcopenic). Carbohydrate loading for non-diabetics. Empiric vitamin D supplementation for all (no testing); IV iron for iron-deficiency anemia; vitamin C 1,000 mg/day for 6 weeks.

Delay elective surgery for albumin < 3.0 g/dL pending repletion.

3 · Surgical Technique

3.1 THA surgical approach

Direct anterior (DAA) and posterior approaches yield equivalent mid- to long-term function (Yan, JAMA Netw Open 2023, network meta-analysis). DAA offers modest early-recovery advantages and lower dislocation risk; the posterior approach offers shorter operative times (13–18 min) and lower intraoperative complication risk (fewer periprosthetic fractures and lateral femoral cutaneous nerve injuries). A multicenter propensity-weighted analysis found no difference in overall 1-year surgical complications. Surgeon experience and training should drive approach selection — neither is clearly superior overall.

3.2 Tourniquet use in TKA

Routine tourniquet use should be reconsidered. The Cochrane review (41 RCTs) found higher serious adverse events (RR 1.73, number-needed-to-harm 48), more DVT, higher day-1 pain, and longer stay, with the only benefit ~3.7 minutes shorter operative time. A 2026 meta-analysis (51 RCTs) confirmed higher DVT risk (RR 1.60) with no difference in blood loss or transfusion. A level-I RCT with 3D-CT analysis found no difference in cement penetration and 100% 5-year implant survival with or without tourniquet (Peng, JBJS 2025) — addressing the main historical concern. With TXA, the tourniquet is even less justifiable.

3.3 Implant fixation

TKA: cemented and cementless both exceed 95% 10-year survival with equivalent PROMs; if cementless, use a porous-titanium (not beaded cobalt-chromium) tibial component. THA: cementless dominates and has excellent long-term survival, but consider a cemented femoral stem in patients > 70–75 (especially women, Dorr type-C bone, osteoporosis, or fragility-fracture history), where cementless carries higher periprosthetic-fracture risk (OR 2.40).

3.4 Bearing surface (THA)

Ceramic- or metal-on-highly-cross-linked-polyethylene (HXLPE) is preferred for most patients (wear 0.003–0.007 mm/yr). UK registry data (456,457 THAs) show ceramic bearings have lower revision for PJI and aseptic causes. Metal-on-metal bearings should not be used.

3.5 Tranexamic acid

Use TXA in all patients without contraindication (AAOS strong recommendation). TXA is safe even in patients with prior VTE, MI, seizures, stroke, renal disease, or atrial fibrillation, and a statewide registry associated it with lower VTE in TKA (HR 0.56). Combined systemic + topical dosing and multi-dose regimens are superior to a single dose.

Contraindications: CrCl < 30 mL/min (adjust/hold per pharmacy), TXA allergy, VTE within 12 months.

3.6 Robotics & navigation

Robotic-assisted TKA improves alignment precision (RR 0.43 for outliers) with modest early benefits (ROM, early pain, fewer opioid days); an RCT showed better 1-year scores and fewer ligament releases. Robotics are not required for a successful program and add ~$8,000–10,000 per case. Computer navigation alone is not recommended — the AAOS downgraded it for unclear advantage.

3.7 Surgical closure

Use a watertight, layered closure — a watertight capsular seal limits persistent wound drainage, an established periprosthetic-infection risk factor, so it is an infection-prevention step and not only cosmetic. Arthrotomy/capsule: a knotless barbed running suture closes faster than interrupted (TKA −5.94 min, THA −3.73 min in a 16-RCT meta-analysis), costs ~$283 less per case, and gives equal or higher leak resistance (more watertight), with comparable wound-complication rates. Deep dermal: running barbed suture. Skin: subcuticular monocryl, or seal with 2-octyl-cyanoacrylate + self-adhering mesh tape (Prineo/Zip) — the lowest pooled SSI (~1.5%), waterproof, best cosmesis; staples are faster but carry higher SSI in some data. Consider a triclosan-coated suture in higher-risk patients. Dress over an occlusive antimicrobial dressing (§7.3).

Protocol Recommendation — Surgical Technique

THA approach by surgeon expertise. TKA: default to no tourniquet when TXA is used (or brief cementation-only). TXA for all (systemic + topical, multi-dose). Bearings: ceramic/metal-on-HXLPE; never metal-on-metal. Cemented femoral stem for the elderly osteoporotic hip. Robotics optional; navigation-alone not recommended.

Closure: watertight barbed capsular repair + subcuticular / skin-adhesive skin closure under an occlusive antimicrobial dressing.

4 · Anesthesia & Multimodal Analgesia

4.1 Anesthesia type

Spinal (neuraxial) anesthesia is strongly preferred over general for both TKA and THA (ICAROS consensus, 94 studies). Benefits include lower mortality (OR 0.67 for THA), fewer pulmonary complications (OR 0.65–0.69), less DVT (OR 0.52–0.77), lower SSI (UK NJR, 779,491 patients), less transfusion, shorter stay, and more home discharge. In patients ≥ 75, general anesthesia independently raised functional decline (aOR 1.32) and non-home discharge (aOR 1.70). Despite this, neuraxial use is only ~38–40% nationally. Spinal should be the default unless contraindicated.

4.2 Regional nerve blocks (TKA)

Adductor canal block (ACB) is preferred over femoral nerve block — equivalent analgesia with better quadriceps strength and earlier mobilization (Cochrane 2019). The AAHKS/AAOS/ASRA guideline recommends single-shot ACB or periarticular injection. iPACK adds little when a periarticular injection is already used, but helps if it is not. Do not use femoral nerve blocks.

4.3 Periarticular injection (PAI) cocktail

Recommended cocktail — same recipe for TKA and THA. Draw up each component (exact volumes given so no calculation is needed), then add injectable normal saline to a 100 mL total:

Total volume 100 mL. Infiltrate systematically in layers — posterior capsule and periosteum first (before implantation, while access is good), then the capsule, fat pad, and subcutaneous tissue. Exclude morphine and liposomal bupivacaine (no benefit; strong recommendation against). Cautions: keep bupivacaine ≤ 2–2.5 mg/kg (reduce the dose for patients < 50 kg); ketorolac (renal impairment, bleeding, anticoagulation — count toward the daily ketorolac ceiling); epinephrine (severe CAD, arrhythmia, uncontrolled hypertension).

4.4 Systemic multimodal analgesia

4.5 Perioperative corticosteroids (dexamethasone)

Perioperative IV dexamethasone is one of the highest-value multimodal interventions (AAHKS/AAOS/ASRA strong evidence). It reduces pain and opioid use — the Gasbjerg BMJ trial (485 TKA) cut 0–48 h morphine by 10.7 mg — and is the most reliable antiemetic (PONV OR 0.21 for THA). It shortens length of stay by ~0.4 day, with an additional benefit from a second dose at 24 hours. A full repeat dose is superior to splitting the same total dose.

Safety. No RCT, meta-analysis, or large database has shown increased PJI; several show a protective association (national database aORs 0.80–0.87). The PADDI trial (NEJM 2021, 8,725 patients) found 8 mg dexamethasone non-inferior to placebo for 30-day SSI, including in diabetics. Hyperglycemia is transient and modest (~13 mg/dL in non-diabetics; resolves by 24 h) and is driven mainly by pre-op HbA1c, not by dexamethasone.

Give per protocol in diabetics with HbA1c < 8% (monitor glucose, target < 200). Use caution / single dose only if HbA1c ≥ 8%. This applies to short-course perioperative IV dexamethasone — distinct from chronic oral steroids or pre-op intra-articular injections, which do raise PJI risk.

4.6 Intraoperative ketamine

Low-dose (subanesthetic) ketamine is an opioid-sparing adjunct, most valuable in opioid-tolerant or high-baseline-pain patients (ASRA/AAPM/ASA consensus). Induction bolus 0.25–0.5 mg/kg IV, then an intraoperative infusion 0.1–0.25 mg/kg/h (common default ~0.15 mg/kg/h), stopped at or ~30 min before closure. Routine continuation into PACU is not required — reserve a low-dose infusion (~0.1 mg/kg/h) for opioid-tolerant patients. Consensus ceilings for acute pain: bolus ≤ 0.35 mg/kg, infusion ≤ 1 mg/kg/h. Cautions: poorly controlled cardiovascular disease/hypertension, active psychosis, elevated intracranial or intraocular pressure; treat psychotomimetic effects with a benzodiazepine.

Protocol Recommendation — Anesthesia & Analgesia

Spinal anesthesia as default. TKA: adductor canal block + PAI (bupivacaine 0.25% 100 mg [40 mL] + ketorolac 30 mg + epinephrine 0.5 mg + methylprednisolone 40 mg, injectable saline to 100 mL); no femoral block, no liposomal bupivacaine, no morphine. Scheduled acetaminophen + celecoxib; pregabalin optional (selective); low-dose IV ketamine intra-op (esp. opioid-tolerant); duloxetine for centrally sensitized patients. Dexamethasone 10 mg IV at induction; 16 mg PO the morning after surgery (POD 1). Opioids rescue only.

5 · Infection Prevention

5.1 Antibiotic prophylaxis — agent

Cefazolin is first-line (OR 0.59 for PJI vs non-cephalosporin alternatives). Most patients with reported penicillin allergy can safely receive cefazolin absent a history of anaphylaxis or Stevens-Johnson syndrome; reserve clindamycin 900 mg for true severe/anaphylactic allergy. Do not routinely add vancomycin — the ASAP trial (NEJM 2023, 4,113 patients) found adding vancomycin to cefazolin was not superior and was associated with more SSI in knee arthroplasty and more hypersensitivity reactions.

Dosing: 2 g IV for ≤ 120 kg, 3 g for > 120 kg, within 60 minutes before incision (whole dose in before tourniquet inflation if used); redose every 3–4 hours for long cases.

5.2 Antibiotic prophylaxis — duration (single dose is sufficient)

A single pre-op dose of cefazolin, with intraoperative redosing, is very likely sufficient. All available TJA-specific data show no PJI difference between single-dose and 24-hour/multi-dose regimens: a Norwegian registry of 301,204 arthroplasties found 1 dose performed identically to 4 (adjusted HRR 1.0), and a 20,682-TJA cohort and a cefazolin-shortage cohort agreed. The NOCOTA RCT found ≤ 24-hour prophylaxis non-inferior to longer courses. Prolonged prophylaxis (> 24 h) is associated with duration-dependent acute kidney injury and C. difficile, without additional SSI reduction. The CDC and WHO endorse single-dose; the AAOS has not yet issued a TJA-specific position.

Protocol Recommendation — Prophylaxis Agent & Duration

Cefazolin 2 g IV (3 g if > 120 kg) within 60 min of incision; redose q3–4 h. Clindamycin 900 mg only for severe/anaphylactic β-lactam allergy. Do NOT add IV vancomycin to cefazolin.

A single pre-op dose is evidence-supported; if a 24-hour regimen is retained, discontinue by 24 h. Never continue beyond 24 hours. For the ASC pathway, single-dose is the logical default.

5.3 Intraosseous (IO) vancomycin

IO vancomycin is the most promising emerging prophylaxis technique, particularly for TKA. It solves the problem that sank IV vancomycin in ASAP: IV vancomycin does not reach adequate tissue levels (cortical bone may never reach MIC), whereas IO injection into cancellous bone achieves tissue concentrations 10–15× higher than IV at half the dose. Pharmacokinetic superiority is established by multiple level-I RCTs (Coventry and Insall Award trials; Spangehl 2022; Wininger tourniquetless 2024; Harper THA 2023). Clinical outcomes in primary TKA are consistent across institutions — the largest series (1,923 TKAs) showed lower PJI at 90 days, 1 year, and 2 years, and a multicenter study reported 12-month PJI 0.1% vs 1.4% (RR 0.10); meta-analyses show OR 0.19–0.35. Benefit is even more pronounced in revision TKA.

Safety is superior to IV vancomycin: lower acute kidney injury (1.9% vs 3.3%), no red-man syndrome, and fewer wound complications (RR 0.50) — because systemic exposure is far lower (serum ~7.8 vs 19.6 µg/mL at closure). Evidence caveat: no large multicenter RCT powered for PJI has completed; clinical data are predominantly retrospective, and THA has pharmacokinetic data only.

5.4 Adjuncts NOT recommended

Intrawound vancomycin powder — despite positive retrospective data, the two most rigorous RCTs (Saba multicenter 2025/26; Mulpur double-blind TKA 2024) show no PJI reduction, and the Mulpur trial found significantly more wound complications. Do not use routinely (surgeon-discretion in selected high-risk/revision cases at most).

Do Not — Intraoperative Adjuncts

Intrawound vancomycin powder as a routine element — RCT-negative with increased wound complications (surgeon discretion in selected high-risk/revision cases at most).

Note on antibiotic-loaded bone cement: for cemented THA, ALBC is protective (RR 0.66 for PJI revision); for cemented TKA, the largest registry data show no PJI difference, and routine use is not supported for primary TKA.

5.5 Skin prep, irrigation, and ventilation

• Skin prep: alcohol-based — chlorhexidine-alcohol or povidone-iodine-alcohol (the alcohol vehicle is the key component; a 2024 JAMA RCT found PI-alcohol non-inferior to CHG-alcohol).
• Irrigation: dilute povidone-iodine (0.25–0.35%) or chlorhexidine lavage before closure (RR 0.60 vs saline). Register data (MARCQI) temper the size of this effect, but the intervention is low-risk.
• Laminar airflow: do not invest in it for infection prevention — WHO recommends against it, and multiple meta-analyses and registries show no benefit for deep SSI/PJI (relevant for community-hospital OR construction decisions).

5.6 SSI-prevention bundle

Implement a comprehensive bundle — single-center data show PJI falling from 1.43% to 0.11% with full implementation. Core elements: universal S. aureus decolonization; pre-op chlorhexidine bathing; correct antibiotic prophylaxis; alcohol-based skin antisepsis; perioperative normothermia with active prewarming (forced-air warming blanket) and warmed IV fluids; clippers only (never razors — RR 1.64 for SSI); glycemic control; and TXA.

5.7 MRSA screening — universal decolonization vs screen-and-treat

Universal decolonization without screening is the recommended strategy. It is at least as effective (one direct comparison: SSI 0.2% vs 0.8% favoring universal, while saving $717,206), treats the more common MSSA (colonization 26% vs 7.7% for MRSA), is simpler and higher-compliance, and is favored by the SHEA/IDSA/APIC compendium. The main historical argument for screening — identifying MRSA carriers for vancomycin — is undercut by ASAP (IV vancomycin does not help) and, where IO vancomycin is adopted, is eliminated because all TKA patients already receive vancomycin locally.

The legitimate counter-argument is mupirocin resistance with universal use; this is mitigated by using intranasal povidone-iodine (no prescription, no resistance concern, endorsed by SHEA/IDSA/APIC) rather than mupirocin. The STOP-SSI bundle (JAMA 2015) remains the strongest evidence for screening-guided prophylaxis, but its vancomycin effect cannot be isolated.

Protocol Recommendation — Infection Prevention

Cefazolin single pre-op dose (≤ 24 h max); IO vancomycin 500 mg for TKA and THA; universal decolonization without screening; alcohol-based skin prep; dilute-PI lavage; clippers only; full SSI bundle. Do NOT: add IV vancomycin, use intrawound vancomycin powder, or rely on laminar airflow.

6 · VTE Prophylaxis

6.1 Agent selection

Aspirin is appropriate for standard-risk patients. EPCAT II (NEJM 2018) showed aspirin non-inferior to rivaroxaban after an initial 5-day rivaroxaban course; EPCAT III (NEJM 2026) showed aspirin alone from day 0 non-inferior to rivaroxaban-then-aspirin for proximal DVT/PE (0.48% vs 0.45%). Meta-analyses of 11 RCTs find no VTE difference between aspirin and oral anticoagulants, with ~30% less total bleeding on aspirin. The countervailing signal is CRISTAL (JAMA 2022), where aspirin monotherapy from day 0 failed non-inferiority vs enoxaparin (3.45% vs 1.82%), largely driven by below-knee DVT.

Higher-risk patients (prior VTE, active malignancy, hypercoagulable state, BMI > 40) warrant a DOAC — rivaroxaban 10 mg daily or apixaban 2.5 mg BID. Combine chemoprophylaxis with intermittent pneumatic compression devices; compression stockings alone have insufficient evidence in TJA.

6.2 Risk-stratified regimen

6.3 Duration & imaging

Protocol: aspirin 81 mg BID — 2 weeks (14 days) for TKA and 4 weeks (28 days) for THA (intermittent pneumatic compression in hospital; TED stockings are not required). Do not order routine post-arthroplasty duplex ultrasonography (AAOS recommends against it); unilateral swelling/tenderness after TKA/THA is not a reliable predictor and should not by itself trigger imaging.

Protocol Recommendation — VTE Prophylaxis

Aspirin 81 mg BID + IPCD for standard-risk; DOAC for elevated-risk; mechanical-only for bleeding-risk. Duration 2 weeks (TKA) / 4 weeks (THA). TED stockings not required. Mobilize POD 0 and ≥ 3×/day. No routine duplex ultrasound. Document the prophylaxis plan in the operative note.

7 · PACU, Recovery, Medications & Rehabilitation

7.1 PACU — medications & discharge criteria

Monitor recovery of the neuraxial block (returning motor/sensory function) with standard vital-sign monitoring. Maintain normothermia (≥ 36 °C) — continue the active warming started in pre-op holding — and glucose < 180–200 mg/dL. Apply mechanical VTE prophylaxis (intermittent pneumatic compression) and mobilize the day of surgery once the block recedes and the patient is stable. Avoid routine urinary catheters, drains, labs, and imaging.

Discharge-from-PACU criteria (modified Aldrete / PADSS): stable vital signs; pain ≤ 4 on oral analgesics; minimal or controlled nausea; regressing neuraxial block with returning motor/sensory function and ability to protect the limb; not over-sedated; spontaneous void or a documented bladder plan.

7.2 Postoperative medications & timeline

POD 0–4 — medications (protocol)

Patient version (POD 0–4): take your steroid tablet (dexamethasone 16 mg) the morning after surgery; take your Tylenol and your anti-inflammatory on schedule — even when you feel okay. Celecoxib is what we prescribe, but over-the-counter naproxen or ibuprofen works if you prefer it — take it with an antacid such as Pepcid, and not at the same time as your aspirin. Aspirin twice a day to prevent clots; a stool softener while you are on the pain pill; the narcotic pain pill only for breakthrough pain, in small amounts; ice and elevate; get up and move a little, often.

POD 4–14 — medications

Continue acetaminophen 1,000 mg q8h and celecoxib 200 mg BID — yes, celecoxib is continued post-op for its opioid-sparing, anti-inflammatory benefit (typically through 2–6 weeks, per surgeon preference and renal/GI/CV tolerance). Wean and stop opioids — most patients are off by POD 5–7. Continue aspirin 81 mg BID through 2 weeks (TKA) / 4 weeks (THA). Stop the bowel regimen once off opioids; tamsulosin as needed.

Protocol Recommendation — Post-Operative Medications

Discharge every primary TKA/THA on the scheduled three-drug backbone — dexamethasone · celecoxib · acetaminophen — with opioids as rescue only.

Dexamethasone 16 mg PO on POD 1, in addition to the induction dose — given in hospital, or sent home with same-day discharges. Celecoxib 200 mg BID scheduled through 2–6 weeks; by patient preference, OTC naproxen or ibuprofen may substitute for celecoxib — add GI protection (famotidine/PPI) with any non-selective NSAID, and separate it from aspirin (NSAID ≥ 30 min after or ≥ 8 h before). Acetaminophen 1,000 mg q8h scheduled — not as-needed — through the first 2 weeks. Aspirin (or the elevated-risk anticoagulant) per the VTE protocol. Senna-docusate while on opioids. Oxycodone 5 mg rescue only (≈10–20 tabs).

7.3 Wound & dressing care

Most patients — occlusive antimicrobial (silver hydrofiber) dressing (e.g., Aquacel Ag Surgical): lower acute PJI (0.44% vs 1.71% vs gauze), far fewer blisters, fewer dressing changes, and earlier showering. It is waterproof — patients may shower from POD 1–2 — and is left undisturbed for 5–7 days, changed only when strikethrough reaches the gel-pad border rather than on a fixed schedule.

High-risk patients — closed-incision negative-pressure wound therapy (ciNPT) (single-use incisional device, e.g., PICO). Over hip/knee incisions, ciNPT reduced surgical-site complications by ~67% (RR 0.33) and SSI by ~60% (RR 0.40); the benefit concentrates in high-risk and revision cases, so apply it selectively, not routinely. Leave the device in place for its wear time (typically 5–7 days).

Define High-Risk — Consider ciNPT

Revision or prior open surgery on the joint · BMI ≥ 35–40 · diabetes (especially insulin-dependent / poorly controlled) · immunosuppression (RA, chronic steroids) · active smoking · therapeutic anticoagulation (chemoprophylaxis other than aspirin) · prior wound complication or infection · poor nutrition (albumin < 3.5 g/dL) · peripheral vascular disease · significant lymphedema · prolonged operative time.

Dressing-care timeline. POD 0–4: leave the dressing (silver or ciNPT) undisturbed; shower from POD 1–2 with it in place; do not soak (no baths, pools, or hot tubs). POD 4–14: remove/replace the silver dressing or ciNPT device at ~POD 5–7; the incision may get wet in the shower but not soaked until fully healed; watch for spreading redness, increasing drainage, or a foul smell; non-absorbable sutures/staples out at ~POD 12–14.

Staff follow-up. Nurse phone call on POD 1 and POD 2; a 24/7 triage line for the first 72 h; in-person visit at ~2 weeks for wound check and suture/staple removal.

7.4 Enhanced recovery (ERAS)

ERAS protocols reduce length of stay ~2.6 days, transfusion (OR 0.40), and 30-day mortality (OR 0.46) without increasing complications or readmissions. POWER2 demonstrated a dose-response: adherence > 62.5% was associated with fewer overall (OR 0.80) and moderate-to-severe (OR 0.62) complications. Apply the pathway universally, not selectively — a propensity-matched study cut median stay from 3 days to 1 across all age groups, including patients ≥ 80.

7.5 Discharge criteria & timing

Use functional milestones, not time-based criteria: adequate pain control on oral analgesics, safe ambulation and transfers, tolerating oral intake, spontaneous voiding, and adequate support. Do not use 90° knee flexion as a mandatory discharge criterion. Same-day discharge is safe in selected patients — a propensity-matched NSQIP analysis (574,375 procedures) found lower adverse events for outpatient than inpatient TJA, and Kaiser registry data confirm safety even in ASA ≥ 3. Target POD 0–1; discharge on POD 0–2 carries the lowest 30-day complication risk.

7.6 Rehabilitation

Home-based rehabilitation is appropriate first-line therapy for most patients. The HIHO RCT found inpatient rehab did not improve mobility over a monitored home program after TKA, and a formal-PT-vs-home RCT found no difference after THA. Telerehabilitation is non-inferior to in-person PT for TKA at roughly one-third the cost. Over half of TKA patients can self-direct rehabilitation; stratify to supervised PT by simple 2-week criteria (flexion < 90°, extension deficit > 10°, or dissatisfaction). Mobilize on the day of surgery — early PT reduces stay 1.2–1.8 days without added adverse events.

7.7 Hip precautions after THA

Routine hip precautions are unnecessary with a posterior approach and capsular repair. A large RCT (1,133 patients) found no difference in dislocation (0.88% overall) while the unrestricted group had better HOOS JR scores and faster return to daily activities, driving, and side sleeping. Eliminate routine precautions; counsel patients only to avoid the extreme combined position (deep flexion + adduction + internal rotation) early. Apply standard precautions only if no posterior repair was performed or dislocation risk is high.

Protocol Recommendation — Recovery & Rehab

Universal ERAS pathway; mobilize the day of surgery; milestone-based discharge targeting POD 0–1 (same-day for selected patients). Home-based rehab first-line with telerehab option and 2-week stratification to formal PT. No mandatory 90° flexion criterion; no routine hip precautions after posterior THA with repair.

8 · Ambulatory Surgery Center (Outpatient) Pathway

Outpatient TJA is safe in appropriately selected patients. A propensity-matched NSQIP analysis (574,375 procedures) found lower 30-day adverse events for outpatient than inpatient TJA, and ASC and hospital-outpatient outcomes are comparable. About 70% of patients are eligible when standard criteria are applied, and up to 75% of unselected patients can achieve same-day discharge with the right protocols. The perioperative recommendations in Domains 1–7 apply identically in the ASC — the differentiators are patient selection, logistics, and contingency planning.

8.1 Outpatient (same-day / ASC) criteria — ALL must be met

8.2 Inpatient criteria — small rural facility

Inpatient admission at our rural facilities is deliberately only slightly more permissive than outpatient selection — the hospital adds monitoring and time, not tertiary-level rescue capacity. Acceptable for inpatient surgery: age 75–82 · stable, optimized ASA III · BMI 36–39 · HbA1c 7–7.4 (defer and optimize at ≥ 7.5) · Hb 11–11.9 after an IV-iron attempt (defer and optimize below 11) · dependence in ADLs with pre-op discharge planning · inability to have spinal anesthesia (general OK) · absent caregiver or > 60 minutes from hospital with an extended-stay / swing-bed plan · untreated OSA with continuous monitoring — the patient remains inpatient until off narcotics · prior postoperative delirium with precautions · stable therapeutic anticoagulation with a bridging plan · chronic opioid use ≤ 90 MME with a written pain plan.

Pacemaker-dependence is an individualized site-of-surgery decision made with Anesthesia and Cardiology.

May still go outpatient after optimization or added monitoring

Stable CAD · COPD · treated OSA on CPAP · active smoking · BPH/urinary-retention history (tamsulosin) · depression/anxiety. Prior contralateral TJA is a favorable factor.

8.3 Refer to a tertiary facility

The following exceed a small rural hospital's backup capacity regardless of admission status — refer: ASA IV (or ASA III that cannot be stabilized) · age > 82 · BMI ≥ 40 · congestive heart failure · cirrhosis · ESRD/dialysis · unstable arrhythmia · symptomatic aortic stenosis · severe pulmonary hypertension · recent MI or coronary stent (< 6 months, on dual antiplatelet therapy) · active/recent VTE (< 3 months) · opioid dependence requiring specialty pain management (> 90 MME or on buprenorphine/methadone). Not meeting outpatient or inpatient criteria never cancels surgery — it directs the patient to the right facility.

8.4 Operative-day logistics & discharge

• Schedule TJA cases early — start before 11:00 AM (ideally before 9:00); later starts strongly predict failed same-day discharge. Limit to ~2 TJA cases per surgeon per ASC day; allow 4–6 h observation.
• Low-dose spinal to speed motor recovery; avoid general anesthesia where possible.
• Discharge when: pain ≤ 4 on oral analgesics, independent ambulation ≥ 30 ft and transfers, hemodynamically stable, full motor recovery, spontaneous void (or scan < 300 mL), tolerating intake, dry dressing, teach-back complete.

Failure-to-Launch & Safety Net

Have a written contingency for the common causes: urinary retention (intermittent cath; pre-op tamsulosin), orthostatic hypotension (fluids, staged mobilization, hold AM antihypertensives), prolonged motor block (low-dose spinal), PONV (dexamethasone + ondansetron), uncontrolled pain (ensure PAI/block; IV ketorolac rescue).

A written transfer agreement with a hospital within 30 minutes is required (and a CMS condition of participation). Nurse call on POD 1 and POD 2; 24/7 triage line for 72 h.

Program development: phase in — hospital same-day first (strict criteria), then ASC launch (standard criteria, track failure-to-launch < 5%), then expansion, then a mature simplified assessment. Outcomes improve with institutional experience.

9 · Practices to Abandon

Key References

2023 ACR/AAHKS Clinical Practice Guideline for the Optimal Timing of Elective Hip or Knee Arthroplasty. Arthritis Care Res. 2023.

Giori NJ et al. Risk Reduction Compared With Access to Care: BMI Eligibility Criterion for Joint Replacement. JBJS Am. 2018.

2024 AHA/ACC/… Guideline for Perioperative Cardiovascular Management for Noncardiac Surgery. J Am Coll Cardiol. 2024.

Nelson CL et al. Low Albumin Levels, More Than Morbid Obesity, Are Associated With Complications After TKA. CORR 2015.

Scrimshire AB et al. Preoperative Iron Treatment in Anaemic Patients Undergoing Elective THA/TKA: SR & Meta-analysis. BMJ Open 2020.

Torchia MT et al. Universal Screening for Malnutrition Prior to TKA Is Cost-Effective: A Markov Analysis. J Arthroplasty 2023.

Dreyer HC et al. Essential Amino Acid Supplementation in Patients Following TKA. J Clin Invest 2013.

Ueyama H et al. Perioperative EAA Facilitates Quadriceps Strength and Volume Recovery After TKA (RCT). JBJS Am 2023.

Yan L et al. Comparative Efficacy and Safety of Surgical Approaches for THA: SR & Network Meta-analysis. JAMA Netw Open 2023.

Ahmed I et al. Tourniquet Use for Knee Replacement Surgery. Cochrane Database Syst Rev 2020.

Peng FL et al. Effect of Tourniquet Use on 3D Cement Penetration and Mid-Term Stability in TKA (RCT). JBJS Am 2025.

Memtsoudis SG et al. Anaesthetic Care of Patients Undergoing Primary Hip and Knee Arthroplasty (ICAROS). Br J Anaesth 2019.

Fillingham YA et al. Efficacy and Safety of Regional Nerve Blocks in TKA (AAHKS/AAOS/ASRA). J Arthroplasty 2022.

Hannon CP et al. Efficacy and Safety of Periarticular Injection in TJA: Direct Meta-analysis. J Arthroplasty 2022.

Gasbjerg KS et al. Dexamethasone as an Analgesic Adjuvant After TKA (RCT). BMJ 2022.

Corcoran TB et al. Dexamethasone and Surgical-Site Infection (PADDI trial). N Engl J Med 2021.

Peel TN et al. Trial of Vancomycin and Cefazolin as Surgical Prophylaxis in Arthroplasty (ASAP). N Engl J Med 2023.

Bratzler DW et al. Clinical Practice Guidelines for Antimicrobial Prophylaxis in Surgery. Am J Health-Syst Pharm 2013.

de Jonge SW et al. Effect of Postoperative Continuation of Antibiotic Prophylaxis on SSI: SR & Meta-analysis. Lancet Infect Dis 2020.

Lutro O et al. How Many Doses of Antibiotic Prophylaxis in Primary Arthroplasty (301,204 cases, Norway). Acta Orthop 2025.

Young SW et al. Higher Tissue Concentrations of Vancomycin With IO vs Systemic Prophylaxis in TKA (Coventry Award, RCT). CORR 2014.

Spangehl MJ et al. Higher Tissue Vancomycin With Low-Dose IO Despite Limited Tourniquet Duration in TKA (RCT). J Arthroplasty 2022.

Park KJ et al. Superior Clinical Results With Intraosseous Vancomycin in Primary TKA. J Arthroplasty 2025.

Parkinson B et al. IO Regional Prophylactic Antibiotics Decrease PJI in Primary TKA: Multicenter Study. CORR 2021.

Saba BV et al. Wound Complication and Infection Rates After Vancomycin Powder and Dilute Povidone-Iodine (RCT). J Arthroplasty 2025.

Mulpur P et al. Efficacy of Intrawound Vancomycin in Primary TKA (double-blind RCT). J Arthroplasty 2024.

Calderwood MS et al. Strategies to Prevent Surgical Site Infections in Acute-Care Hospitals: 2022 Update (SHEA/IDSA/APIC). Infect Control Hosp Epidemiol 2023.

Stambough JB et al. Universal Decolonization Protocol in Primary TJA: Reduced Cost and SSI. J Arthroplasty 2017.

Schweizer ML et al. Bundled Intervention and SSI (STOP-SSI). JAMA 2015.

Anderson DR et al. Aspirin or Rivaroxaban for VTE Prophylaxis after Hip or Knee Arthroplasty (EPCAT II). N Engl J Med 2018.

Shivakumar S et al. Rivaroxaban Then Aspirin vs Aspirin Alone after THA/TKA (EPCAT III). N Engl J Med 2026.

Sidhu VS et al. Aspirin vs Enoxaparin for Symptomatic VTE after Hip/Knee Arthroplasty (CRISTAL). JAMA 2022.

Falck-Ytter Y et al. Prevention of VTE in Orthopedic Surgery Patients (ACCP). Chest 2012.

Ripollés-Melchor J et al. ERAS Adherence and Postoperative Complications in THA/TKA (POWER2). JAMA Surg 2020.

Buhagiar MA et al. Inpatient vs Monitored Home-Based Rehabilitation After TKA (HIHO, RCT). JAMA 2017.

Austin MS et al. Formal Physical Therapy After THA Is Not Required (RCT). JBJS Am 2017.

Lan RH et al. Contemporary Outpatient Arthroplasty Is Safe: 574,375 Procedures (propensity-matched). JBJS Am 2021.

Reddy NC et al. Same-Day Discharge TJA and 90-Day Adverse Events in ASA ≥ 3. JBJS Am 2021.

Diagnosis and Prevention of Periprosthetic Joint Infections: Evidence-Based CPG. AAOS 2019.