PGPeriklis Giannakis

Periklis Giannakis

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Periklis Giannakis has a total of 79 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Statistics in Brief: A Surgeon’s Guide to Creating and Interpreting Risk Calculator Tools

    by Periklis Giannakis, Jashvant Poeran, Alejandro Gonzalez Della Valle, Junying Wang

    Was not provided for this article.

  • Open Access

    Should we use robotics in total joint arthroplasty—and if yes, how?

    by Periklis Giannakis, Jashvant Poeran, Philippe Beauchamp-Chalifour, Robert G. Marx, Alejandro Gonzalez Della Valle, Stavros G. Memtsoudis

    Should we use robotics in total joint arthroplasty—and if yes, how?

  • Impact of the REGAIN and RAGA Trials on Use of Neuraxial Anesthesia among Surgically Managed Hip Fracture Patients: An Interrupted Time Series Analysis

    by Periklis Giannakis, Junying Wang, Alex Illescas, Juliet E. Rowe, Crispiana Cozowicz, Alexander B. Stone, Philipp Gerner, Jiabin Liu, Stavros G. Memtsoudis, Jashvant Poeran

    Was not provided for this article.

  • Trends of In-Hospital Mortality after Inpatient Elective Total Hip and Knee Arthroplasty: A 2016 to 2023 Population-Based Study

    by Periklis Giannakis, Juliet E. Rowe, Renee Ren, Sara E. Covin, Alejandro Gonzalez Della Valle, Jashvant Poeran, Jiabin Liu, Robert G. Marx, Stavros G. Memtsoudis

  • Variation in anaesthesia-related costs and charges for total hip and knee arthroplasty cases: an observational study using US claims data

    by Renee Ren, Periklis Giannakis, Alex Illescas, Jashvant Poeran, Crispiana Cozowicz, Lisa Reisinger, Stavros G. Memtsoudis, Jiabin Liu

  • 1-year outcomes of robotic-assisted, computer-assisted, and manual total hip arthroplasty after the learning curve

    by Periklis Giannakis, Jashvant Poeran, Juliet E Rowe, Sophia T Zhuang, Crispiana Cozowicz, Jiabin Liu, Stavros G Memtsoudis, Alexander S McLawhorn, Robert G Marx, Alejandro Gonzalez Della Valle, et al.

    Introduction: While comparisons exist between manual (MA-THA), computer-assisted (CA-THA), and robotic-assisted total hip arthroplasty (RA-THA), they have not considered surgeon/hospital learning curves. Therefore, we compared surgical outcomes between MA-THA/CA-THA/RA-THA among hospitals with ⩾1 year of RA-THA experience. Methods: In this retrospective study (2016–2023, Premier Healthcare Database), we included adults undergoing elective MA-THA/CA-THA/RA-THA for non-traumatic/non-neoplastic diagnoses. Our primary outcome was any mechanical complication within 1-year follow-up; secondary outcomes included any infection within 1 year. We conducted pairwise mixed-effects logistic regression and reported odds ratios (OR) and 95% confidence intervals (CI). Results: Among 75,966 MA-THAs, 6,207 CA-THAs and 17,497 RA-THAs, odds of any mechanical complication within 1-year were: CA-THA versus MA-THA (OR 0.63, CI 0.52–0.75; p < 0.001), RA-THA versus MA-THA (OR 0.81, CI 0.73–0.90; p < 0.001) and RA-THA versus CA-THA (OR 1.44, CI 1.14–1.82; p = 0.003). Differences between RA-THA and CA-THA were primarily due to higher odds of mechanical loosening (OR 2.40, CI 1.19–4.85; p = 0.015) with RA-THA. Dislocation was less likely with RA-THA compared to MA-THA (OR 0.57, CI 0.44–0.72; p < 0.001). CA-THA (OR 0.46 CI 0.34–0.63; p < 0.001) and RA-THA (OR 0.83 CI 0.71–0.97; p = 0.022) were associated with significantly lower odds of infection within 1 year compared to MA-THA. RA-THA was associated with higher odds of infection within 1 year compared to CA-THA (OR 2.11 CI 1.43–3.09; p < 0.001). Conclusions: With up to 1-year follow-up, CA-THA was associated with the lowest odds for mechanical complications, primarily due to mechanical loosening and infection.

  • Outcome Differences Between General and Neuraxial Anesthesia for Hip Fracture by Frailty and Age in the Elderly: A Retrospective Cohort Study

    by Periklis Giannakis, Mariana Restrepo, Alexander B. Stone, Sophia T. Zhuang, Junying Wang, Crispiana Cozowicz, Alex Illescas, Juliet E. Rowe, Lisa Reisinger, Lazaros Poultsides, et al.

    BACKGROUND: Frailty and age are major outcome drivers in hip fracture surgery, but their interaction has not been considered in neuraxial versus general anesthesia comparisons. Using the Hospital Frailty Risk Score (HFRS), we examined this interaction. METHODS: In this retrospective cohort study (2016-2023, Premier Healthcare Database), adults undergoing hip fracture surgery with neuraxial or general anesthesia were included and stratified by age quantiles (≤71, 72–86, ≥87 years) and HFRS (low, intermediate/high). The primary outcome was an in-hospital composite of mortality and major system complications. Intensive care unit (ICU) admission and high opioid use or prolonged length of stay (LOS) ≥75th percentile were also assessed. We used mixed-effects models and reported odds ratios (OR) and 95% confidence intervals (CIs). RESULTS: Among 623,122 patients, neuraxial (versus general) anesthesia was associated with lower odds of the composite outcome in patients ≥87 years with intermediate/high frailty (OR, 0.88 and 95% CI, 0.83–0.94; P < .001). Overall, neuraxial anesthesia was linked to higher odds of respiratory complications (OR, 1.06 and 95% CI, 1.01–1.10; P = .03), driven by patients ≤71 years with intermediate/high frailty, and lower odds of renal failure (OR, 0.87 and 95% CI, 0.83–0.92; P < .001), primarily among those ≥72 years with intermediate/high frailty. Neuraxial anesthesia was also associated with higher cardiac complication odds (OR, 1.07 and 95% CI, 1.02–1.12; P = .008), particularly in patients aged 72 to 86 years with intermediate/high frailty. Mortality odds were lower overall (OR, 0.83 and 95% CI, 0.74–0.93; P = .003), driven by patients ≥87 years with intermediate/high frailty. Neuraxial anesthesia was associated with higher odds of prolonged LOS in patients aged 72 to 86 years with low frailty (OR, 1.16 and 95% CI, 1.04–1.31; P = .035), but lower odds in those ≥87 years with intermediate/high frailty (OR, 0.92 and 95% CI, 0.87–0.97; P = .012). It was associated with lower odds of high opioid use overall and within each subgroup. ICU admission odds were higher in patients ≤71 years with intermediate/high frailty (OR, 1.16 and 95% CI, 1.05–1.29; P = .019) but lower in those ≥87 years with intermediate/high frailty (OR, 0.82 and 95% CI, 0.75–0.90; P < .001). Neuraxial anesthesia was linked to higher odds of discharge home (OR, 1.08 and 95% CI, 1.04–1.12; P < .001), except among ≥87-year-old intermediate/high frailty patients, where odds were lower (OR, 0.90 and 95% CI, 0.81–0.99; P = .041). CONCLUSIONS: Neuraxial versus general anesthesia showed modest overall benefits after hip fracture, varying by age–frailty subgroup, supporting frailty-guided anesthetic decisions clinically. Level of evidence: III. Retrospective cohort study


  • Open Access

    Does upper limb dominance influence outcomes in adults undergoing primary anatomic or reverse total shoulder arthroplasty? A systematic review

    by Periklis Giannakis, Mitchell A. Johnson, Sara E. Covin, Olivia M. Jochl, Sophia T. Zhuang, Jashvant Poeran, Robert G. Marx, Samuel A. Taylor, Lawrence V. Gulotta

  • What Is the Association Between Adding Genicular Nerve Block to Adductor Canal Block and Infiltration Between the Popliteal Artery and Capsule of the Knee and Perioperative Outcomes After Primary TKA?

    by Periklis Giannakis, Sydney Yu, Alex Illescas, Juliet E. Rowe, Jiabin Liu, Robert G. Marx, Alejandro Gonzalez Della Valle, Stavros G. Memtsoudis, Jashvant Poeran, Alexandra Sideris, et al.

    Background: Limited evidence exists on the functional and analgesic efficacy of adding a genicular nerve block to an adductor canal block (ACB) plus infiltration between the popliteal artery and posterior knee capsule (IPACK) for TKA. Filling this knowledge gap would be important because if this approach were to be effective, patients could benefit from an additional opioid-free, motor-sparing analgesic modality. Questions/purposes: (1) Is the addition of a genicular nerve block to an ACB/IPACK associated with better pain management during hospitalization? (2) Is the addition of a genicular nerve block to an ACB/IPACK associated with better pain management in the postanesthesia care unit (PACU)? (3) Is the addition of a genicular nerve block to an ACB/IPACK associated with earlier time to physical therapy (PT) clearance? (4) Is the addition of a genicular nerve block to an ACB/IPACK associated with lower incidence of opioid refills within 90 days postoperatively? Methods: This was a retrospective, propensity score–matched cohort study including patients undergoing TKA from January 2021 to December 2024 at a high-volume academic institution specializing in inpatient and outpatient arthroplasty. All adults undergoing primary elective unilateral TKA for primary osteoarthritis with administration of an ACB/IPACK, with or without a genicular nerve block, were assessed for eligibility (n = 20,648). After excluding patients with American Society of Anesthesiologists physical status of > III; patients receiving general anesthesia, additional peripheral nerve blocks, or acupuncture; and patients discharged to a facility, a total of 10,156 patients undergoing inpatient TKA and 2814 patients undergoing outpatient TKA were considered eligible for propensity score matching. After a 1:1 propensity score match on baseline demographic characteristics, psychosocial factors, active opioid prior to admission, perioperative and anesthesia details, and first PT visit pass or fail, 2803 and 1102 patients from each group were analyzed for inpatient and outpatient TKAs. The median (IQR) age was 71 years (64 to 76) for inpatient TKAs and 65 years (60 to 71) for outpatient TKAs; 64% and 48% were female, respectively, and no differences among groups within the same setting were observed after matching. Aside for the comparison of interest, the same standardized perioperative pain management protocol was followed for all patients. For our first aim and second aim, we evaluated median and highest pain scores and opioid consumption. For our third aim, we assessed PT clearance, and for our fourth aim, we assessed incidence of opioid refills within 90 days. A difference in pain of ≥ 2 on the numeric rating scale (NRS) and a difference in opioid consumption of ≥ 10 oral morphine milligram equivalents (OMEs) were considered clinically meaningful. For the inpatient setting, a 4-hour median difference in time to PT clearance was considered clinically meaningful, and for the outpatient setting, any statistically significant difference with a median time to PT clearance of < 12 hours for either group was considered clinically meaningful. Results: During hospitalization, we found no difference between the intervention and control group in terms of highest pain and median pain. In terms of total opioid consumption, we found no clinically important difference between the intervention group and control group (median [IQR] NRS score 55 [30 to 82] versus 52 [30 to 80]; p = 0.001) in the inpatient setting and no difference in the outpatient setting. In the PACU, we found no difference in terms of highest pain and median pain in the inpatient setting and no clinically important difference for highest pain (median [IQR] NRS score 7 [5 to 8] versus 7 [6 to 8]; p = 0.03) in the outpatient setting. In the inpatient setting, no clinically important difference was observed between the intervention and control group for opioid consumption (median [IQR] OMEs 53 [30 to 82] versus 48 [25 to 75]; p < 0.001); no difference was observed in the outpatient setting. We found no clinically meaningful difference between the intervention group and control group in terms of time to PT clearance in the inpatient setting (median [IQR] 21 hours [18 to 25] versus 22 hours [19 to 41]; p < 0.001) and the outpatient setting (median [IQR] 13 hours [4 to 16] versus 14 hours [4 to 16]; p = 0.001). We found no difference in terms of incidence of opioid refills within 90 days. Conclusion: The associations we found do not justify the addition of a genicular nerve block in patients undergoing inpatient and outpatient TKA with a comprehensive multimodal analgesia protocol given the modest, at best, effect size observed, its potential financial cost and time cost, and risk for adverse events. Our study suggested that there are subpopulations in the general TKA population that could potentially benefit from the addition of genicular nerve block, but until future evidence identifies these subpopulations, our data suggest that the addition of a genicular nerve block was not associated with any clinically meaningful benefit in the general TKA population. Level of Evidence: Level III, therapeutic study.

  • Open Access

    Injury patterns and cumulative injury burden among U.S. competitive fencers: A survey

    by Katharine Holmes, Mary Rojas, Periklis Giannakis, Jashvant Poeran, Lindsay Bottoms, Alexis Colvin

    Background: Fencing is a highly asymmetrical sport that combines both repetitive upper-extremity and lower-extremity actions. Although fencing related injuries have been described in clinical- and competition-based cohorts, population level data capturing both training and competition exposures and cumulative injury burden remains limited. Objective: To characterize injury patterns, mechanisms, and anatomical distribution among adult competitive fencers and to examine associations between training related exposures and reported injury burden. Methods: Adult competitive fencers registered with USA Fencing were invited to complete an anonymous web-based survey capturing demographics, training and competition exposures, and self-reported fencing related injuries. Injury burden was defined as experiencing three or more lifetime fencing-related injuries among injured respondents. Multivariable logistic regression was used to examine associations between training exposures and injury burden with continuous predictors modeled using restricted (natural) cubic splines to allow for non-linear relationships. Descriptive analyses, correlation analyses, and Poisson regression were performed as sensitivity analyses. Results: Among 303 respondents, 270 (89.1%) reported at least one fencing related injury, accounting for 571 total injuries. Overuse injuries predominated and most frequently involved the knee, ankle and dominant upper extremity with gradual-onset, non-contact mechanisms accounting for the majority of the injuries. Upper-extremity injuries were significantly more likely to occur on the dominant side. In multivariable analyses, years of fencing experience demonstrated a significant non-linear association with higher injury burden, while weekly training volume showed a non-linear association that approached statistical significance. Age at starting fencing, competition frequency, and sex were not independently associated with injury burden. Sensitivity analyses using Poisson regression yielded qualitatively similar findings. Conclusions: Among adult competitive fencers, higher injury burden is most strongly associated with cumulative training exposure, particularly years of fencing experience, with additional contribution from weekly training volume. Injury patterns are characterized by overuse and pronounced dominant-side upper-extremity involvement, consistent with the sport’s asymmetrical biomechanical demands. These findings underscore the importance of monitoring cumulative exposure and addressing asymmetrical loading to mitigate recurrent injury burden in fencing.