3.1 Postmatch baseline demographics and comorbidities

After matching cohorts 1:1 based on age, sex, CCI, obesity, tobacco use, alcohol abuse, and osteoarthritis, 9582 patients (4791 patients with PI and 4791 patients without PI) undergoing THA were included. This analysis revealed that PI patients had significantly higher comorbidity rates of asthma (p < 0.001), COPD (p < 0.001), CKD (p < 0.001), pulmonary heart disease (p = 0.009), rheumatoid arthritis (p < 0.001), cancer (p < 0.001), coagulopathy (p < 0.001), and deficiency anemia (p < 0.001). Patients without PI had significantly higher rates of hypertension (p < 0.001), and diabetes mellitus (p < 0.001) (Table 1).

After 1:1 matching of TKA cohorts, 16,476 patients (8238 patients with PI and 8238 patients without PI) were included. Similar trends were seen in the matched cohorts of patients undergoing TKA. Statistically significant increased rates of asthma (p < 0.001), COPD (p < 0.001), osteoarthritis (p < 0.001), rheumatoid arthritis (p < 0.001), liver disease (p < 0.001), cancer (p < 0.001), coagulopathy (p < 0.001), and deficiency anemia (p < 0.001) were seen in the PI cohort. In contrast, there were significantly increased rates of diabetes mellitus (p < 0.001), hypertension (p < 0.001), and ischemic heart disease (p = 0.006) in the no PI cohort (Table 2).

3.2 90-day, 1-year, and 2-year postoperative outcomes

A multivariate logistic regression analysis was done on the matched cohorts, controlling for age, sex, and CCI to compare postoperative outcomes between PI and non-PI patients following THA. Within 90 days postoperative, this study found that PI patients had significantly higher incidence of DVT (OR: 1.52 [95 %CI: 1.14–2.04], p = 0.005), pneumonia (OR: 1.59 [95 %CI: 1.22–2.08], p < 0.001), UTI (OR: 1.28 [OR: 1.06–1.52], p = 0.007), and any complication (OR: 1.16 [95 %CI: 1.04–1.28], p = 0.007). At 1 year, PI patients that underwent THA were shown to have significant increased odds of PJI (OR: 1.49 [95 %CI: 1.05–2.13], p = 0.028), DVT (OR: 1.51 [95 %CI: 1.19–1.92], p < 0.001), AKI (OR: 1.33 [95 %CI: 1.11–1.59], p = 0.001), pneumonia (OR: 1.96 [95 %CI: 1.61–2.33], p < 0.001), UTI (OR: 1.33 [95 %CI: 1.18–1.52], p < 0.001), and any complication (OR: 1.32 [95 %CI: 1.20–1.45], p < 0.001). After 2 years postoperative, PI patients were associated with significantly increased odds of PJI revision (OR: 2.17 [95 %CI: 1.12–4.55], p = 0.026) (Tables 3–5).

Patients undergoing TKA were also analyzed at these timepoints. Within 90 days, PI patients had significantly higher odds of SSI (OR: 1.69 [95 %CI: 1.18–2.44], p = 0.005), AKI (OR: 1.28 [95 %CI: 1.05–1.59], p = 0.014), wound disruption (OR: 1.52 [95 %CI: 1.15–2.04], p = 0.003), pneumonia (OR: 1.82 [95 %CI: 1.43–2.27], p < 0.001), UTI (OR: 1.18 [95 %CI: 1.02–1.33], p = 0.022), hospital readmission (OR: 1.72 [95 %CI: 1.33–2.22], p < 0.001), and any complication (OR: 1.20 [95 %CI: 1.10–1.32], p < 0.001). However, patients with PI had significantly lower rates of transfusion (OR: 0.79 [95 %CI: 0.65–0.96], p = 0.018). At 1 year postoperative, PI patients were associated with significantly higher rates of PJI (OR: 1.45 [95 %CI: 1.15–1.82], p = 0.002), SSI (OR: 1.49 [95 %CI: 1.18–1.92], p = 0.001), DVT (OR: 1.32 [95 %CI: 1.10–1.56], p = 0.002), AKI (OR: 1.35 [95 %CI: 1.16–1.56], p < 0.001), wound disruption (OR: 1.67 [95 %CI: 1.32–2.17], p < 0.001), pneumonia (OR: 1.89 [95 %CI: 1.64–2.22], p < 0.001), UTI (OR: 1.25 [95 %CI: 1.14–1.37], p < 0.001), readmission (OR: 1.72 [95 %CI: 1.35–2.22], p < 0.001), and any complication (OR: 1.37 [95 %CI: 1.28–1.47], p < 0.001). After 2 years of follow up, there were no significant differences between PI and non-PI patients (Tables 6–8).

4.1 Periprosthetic joint infection: persistent risk and biological plausibility

The strong association between PI and increased PJI risk, particularly within the first postoperative year, is biologically plausible and supported by mechanistic evidence. Conditions such as CVID and XLA are characterized antibody-deficient phenotypes, where phagocytic defects compromise effective pathogen clearance.¹⁶ These immunologic deficits collectively facilitate persistent, biofilm-mediated infections once a prosthetic implant becomes colonized, emphasizing the broader interplay between host immune competence and tissue integration observed across orthopedic biologic applications.¹⁷

Case reports illustrate the clinical challenges reflected in our cohort-level findings. For example, chronic PJI culminating in knee fusion has been documented in a patient with CVID,¹⁴ and recurrent Streptococcus pneumoniae PJI has been reported in a patient with congenital asplenia, underscoring the difficulty of achieving durable eradication in immunodeficient hosts.¹⁵ Importantly, our data demonstrates that this increased risk potentially translates to a persistent need for PJI-related revision at 2 years after THA. This prolonged vulnerability may be related to anatomical and biomechanical factors unique to THA, including deeper soft-tissue envelopes and delayed recognition of early infection.¹⁸ Additionally, the increased risk of PJI observed in TJA patients occurred despite the significantly higher prevalence of diabetes mellitus among non-PI patients, a condition that can increase the risk of PJI by 43 %.¹⁹

These findings reinforce existing surgical guidelines recommending intensified infection-prevention strategies in immunocompromised populations. Practical considerations include preoperative co-management with immunology specialists, optimization of immunoglobulin replacement therapy for antibody-deficient phenotypes, and meticulous implementation of SSI-prevention bundles, such as chlorhexidine bathing, nasal decolonization where indicated, perioperative glucose control, and heightened vigilance for early signs of infection.^9,20

To date little research exists regarding the effectiveness of these interventions for TJAs for those with PI. However, much data has been collected regarding HIV. One meta analysis found that treatment with highly active antiretroviral therapy and optimizing underlying comorbidities lowered the rate of PJI following TJA.²¹ Future studies should be done to determine if similar optimization of comorbidities and treatment with immunoglobulin replacement therapy are effective for reducing PJI following TJA in patients with PI.

4.2 Respiratory and urinary vulnerabilities

PI patients also demonstrated significantly higher odds of postoperative pneumonia and UTI after both THA and TKA. These findings are consistent with hallmark vulnerabilities observed across inborn errors of immunity, where recurrent sinopulmonary and urinary infections are central clinical features.⁹ These observations have direct implications for perioperative optimization. A comprehensive approach should include vaccination review where applicable, aggressive pulmonary hygiene strategies, early mobilization, and prompt evaluation of respiratory or urinary symptoms. Judicious use of empiric and targeted antibiotics in accordance with established guidelines is critical to reducing downstream complications while minimizing antimicrobial resistance.^9,20

4.3 Infection-linked hypercoagulability

Our study also identified a significant increase in DVT risk among PI patients undergoing both THA and TKA. This relationship may be mediated, at least in part, by the interplay between infection and thrombosis. Infection-driven systemic inflammation promotes endothelial activation, platelet aggregation, and cytokine-induced shifts in coagulation pathways.²² Epidemiologic data support this mechanistic link, with prior studies demonstrating a several-fold increase in VTE risk following acute infections.²³ Similar to our PJI data, these findings were also found in spite of a significantly increased rate of diabetes mellitus, a comorbidity that can increase the risk of DVT by 45 %19. These findings highlight the importance of guideline-concordant VTE prophylaxis, early postoperative mobilization, and a low threshold for evaluating new lower-extremity symptoms or unexplained dyspnea in PI patients.

4.4 Renal and wound healing challenges

Beyond infectious complications, PI patients were at increased risk of AKI following both THA and TKA and demonstrated higher rates of wound disruption after TKA. Although causality cannot be established within the confines of this dataset, several plausible contributors exist. Management of postoperative infections often involves nephrotoxic antimicrobials, prolonged hemodynamic stress, and fluid shifts, all of which may compound existing renal vulnerabilities.²⁴ Similarly, baseline comorbidity differences observed in PI patients, such as higher rates of coagulopathy and deficiency anemia, may impair wound healing capacity and tissue integrity. Importantly, these results were found even though hypertension was significantly more prevalent in non-PI patients, as hypertension has been shown to increase the incidence of acute renal failure following TKA procedures.²⁵ From a clinical standpoint, preventive measures such as early hydration, avoidance of redundant nephrotoxins, careful antibiotic selection and dosing, and regimented wound surveillance represent low-cost, high-yield strategies that may help mitigate these risks.

4.5 Readmissions and system-level burden

Hospital readmissions were significantly more frequent among PI patients undergoing TKA, paralleling the observed increases in early infection and wound disruption. Although we lacked per-case cost data, these findings likely reflect an amplified economic burden. PJI alone is projected to exceed $1.8 billion annually in the United States by 2030, largely driven by revisions, reoperations, and prolonged episodes of care.²⁶ Given the elevated rates of 1-year PJI and PJI-related revisions following THA in our cohort, PI may contribute disproportionately to this rising financial impact. Targeted perioperative optimization strategies, multidisciplinary care pathways, and early escalation protocols tailored to PI patients may represent cost-effective strategies to reduce both clinical and economic burden.^20,26

4.6 Limitations

Several limitations should be considered when interpreting these findings. This study relied on an administrative claims database, which inherently introduces the potential for misclassification bias. Diagnoses, procedures, and complications were identified using ICD and CPT codes, and while comprehensive code sets and validated PearlDiver comorbidity flags were applied, incomplete or inaccurate coding may have led to under or overestimation of true complication rates.

Another important limitation is the lack of granular clinical detail within the dataset. Primary immunodeficiencies encompass a heterogeneous group of disorders ranging from relatively mild antibody deficiencies to severe combined immunologic defects. However, the database does not provide information on PI subtype, baseline immunologic parameters, vaccination history, or use and timing of immunoglobulin replacement therapy. As a result, our findings represent an averaged effect across diverse phenotypes, and risk profiles likely differ considerably among individual subgroups. Future studies incorporating laboratory-confirmed diagnoses and treatment status are needed to better define these nuances.

Similarly, the database does not capture perioperative management variables that may substantially influence outcomes in immunodeficient patients. We were unable to assess surgical techniques, institutional protocols for infection prevention, antibiotic prophylaxis regimens, perioperative glucose control, or standardized wound-care practices. Differences in surgeon experience and hospital volume, which are known to affect arthroplasty outcomes, also could not be evaluated. These unmeasured factors may contribute meaningfully to the associations we observed.

Despite 1:1 matching on age, sex, CCI, and key comorbidities, the possibility of residual confounding remains. Variables such as nutritional status, functional capacity, and adherence to perioperative care pathways were unavailable and may partially explain the elevated complication rates seen in the PI cohort. Additionally, while this is the largest analysis of its kind to date, the relatively low prevalence of patients with PI limited statistical power for rare outcomes, particularly long-term events like PJI-related revision beyond two years.

Finally, the retrospective nature of this study precludes any inference of causality. While our findings are biologically plausible and align with known mechanisms of impaired host defense, prospective studies incorporating detailed immunologic, surgical, and perioperative data are needed to validate these results and guide evidence-based optimization strategies for this vulnerable patient population.