Accuracy and consistency of online large language model-based artificial intelligence chat platforms in answering patients' questions about heart failure

BACKGROUND

Heart failure (HF) is a prevalent condition associated with significant morbidity. Patients may have questions that they feel embarrassed to ask or will face delays awaiting responses from their healthcare providers which may impact their health behavior. We aimed to investigate the potential of large language model (LLM) based artificial intelligence (AI) chat platforms in complementing the delivery of patient-centered care.

METHODS

Using online patient forums and physician experience, we created 30 questions related to diagnosis, management and prognosis of HF. The questions were posed to two LLM-based AI chat platforms (OpenAI's ChatGPT-3.5 and Google's Bard). Each set of answers was evaluated by two HF experts, independently and blinded to each other, for accuracy (adequacy of content) and consistency of content.

RESULTS

ChatGPT provided mostly appropriate answers (27/30, 90%) and showed a high degree of consistency (93%). Bard provided a similar content in its answers and thus was evaluated only for adequacy (23/30, 77%). The two HF experts' grades were concordant in 83% and 67% of the questions for ChatGPT and Bard, respectively.

CONCLUSION

LLM-based AI chat platforms demonstrate potential in improving HF education and empowering patients, however, these platforms currently suffer from issues related to factual errors and difficulty with more contemporary recommendations. This inaccurate information may pose serious and life-threatening implications for patients that should be considered and addressed in future research.

Management of Pulmonary Hypertension in the Context of Heart Failure with Preserved Ejection Fraction

PURPOSE OF REVIEW

To review the current evidence and modalities for treating pulmonary hypertension (PH) in heart failure with preserved ejection fraction (HFpEF).

RECENT FINDINGS

In recent years, several therapies have been developed that improve morbidity in HFpEF, though these studies have not specifically studied patients with PF-HFpEF. Multiple trials of therapies specifically targeting the pulmonary vasculature such as phosphodiesterase (PDE) inhibitors, prostacyclin analogs, endothelin receptor antagonists (ERA), and soluble guanylate cyclase stimulators have also been conducted. However, these therapies demonstrated lack of consistency in improving hemodynamics or functional outcomes in PH-HFpEF. There is limited evidence to support the use of pulmonary vasculature-targeting therapies in PH-HFpEF. The mainstay of therapy remains the treatment of the underlying HFpEF condition. There is emerging evidence that newer HF therapies such as sodium-glucose transporter 2 inhibitors and angiotensin-receptor-neprilysin inhibitors are associated with improved hemodynamics and quality of life of patients with PH-HFpEF. There is also a growing realization that more robust phenotyping PH and right ventricular (RV) function may hold promise for therapeutic strategies for patients with PH-HFpEF.

Safety and Activity of Immune Checkpoint Inhibitors in People Living With HIV and Cancer: A Real-World Report From the Cancer Therapy Using Checkpoint Inhibitors in People Living With HIV-International (CATCH-IT) Consortium

PURPOSE

Compared with people living without HIV (PWOH), people living with HIV (PWH) and cancer have traditionally been excluded from immune checkpoint inhibitor (ICI) trials. Furthermore, there is a paucity of real-world data on the use of ICIs in PWH and cancer.

METHODS

This retrospective study included PWH treated with anti-PD-1- or anti-PD-L1-based therapies for advanced cancers. Kaplan-Meier method was used to estimate overall survival (OS) and progression-free survival (PFS). Objective response rates (ORRs) were measured per RECIST 1.1 or other tumor-specific criteria, whenever feasible. Restricted mean survival time (RMST) was used to compare OS and PFS between matched PWH and PWOH with metastatic NSCLC (mNSCLC).

RESULTS

Among 390 PWH, median age was 58 years, 85% (n = 331) were males, 36% (n = 138) were Black; 70% (n = 274) received anti-PD-1/anti-PD-L1 monotherapy. Most common cancers were NSCLC (28%, n = 111), hepatocellular carcinoma ([HCC]; 11%, n = 44), and head and neck squamous cell carcinoma (HNSCC; 10%, n = 39). Seventy percent (152/216) had CD4+ T cell counts ≥200 cells/µL, and 94% (179/190) had HIV viral load <400 copies/mL. Twenty percent (79/390) had any grade immune-related adverse events (irAEs) and 7.7% (30/390) had grade ≥3 irAEs. ORRs were 69% (nonmelanoma skin cancer), 31% (NSCLC), 16% (HCC), and 11% (HNSCC). In the matched mNSCLC cohort (61 PWH v 110 PWOH), 20% (12/61) PWH and 22% (24/110) PWOH had irAEs. Adjusted 42-month RMST difference was -0.06 months (95% CI, -5.49 to 5.37; P = .98) for PFS and 2.23 months (95% CI, -4.02 to 8.48; P = .48) for OS.

CONCLUSION

Among PWH, ICIs demonstrated differential activity across cancer types with no excess toxicity. Safety and activity of ICIs were similar between matched cohorts of PWH and PWOH with mNSCLC.

Pan-cancer (ca) analysis of the safety and efficacy of immune checkpoint inhibitors (ICI) in patients (pts) living with HIV (PLWH): Results from the international CATCH-IT consortium

2649

Background: PLWH and ca are inadequately represented in clinical trials evaluating ICI especially in the setting of low CD4 counts (ct) and elevated HIV viral loads (VL). We assembled an international cohort of PLWH and ca treated with ICI to evaluate toxicity profiles and clinical outcomes. Methods: We retrospectively collected data on 204 PLWH and ca receiving ≥ 1 cycle of ICI between 2015-2021 at 14 academic medical centers in the US and Europe. Immune-related adverse events (irAEs) were graded per the Common Terminology Criteria for Adverse Events (CTCAE) V5.0. Baseline CD4 ct, CD8 ct and HIV VL were collected within 3 months (mo) of ICI initiation when available. Fisher’s exact test was performed to compare categorical variables. Median (med) Overall Survival (OS) and Objective Response Rate (ORR) were calculated for 186 pts treated in the metastatic (met) setting. Results: Among 204 PLWH treated with ICI, 174 (85%) were cis-gender males. 61 (31%) were Black and 34 (18%) were Hispanic/Latinx. Pts were treated with pembrolizumab (n=93), nivolumab (n=71), atezolizumab (n= 20), nivolumab and ipilimumab (n=13), durvalumab (n=6), or avelumab (n=1). Med number of prior lines of systemic therapy was 1 (range: 0-5). Among pts with available baseline data, 36/133 (27%) had CD4 ct <200 cells/µL while 12/136 (9%) had VL ≥400 copies/mL. irAEs of any grade occurred in 43 (21%) pts and 13 (7%) were grade ≥3 while 19 (9%) required steroids. Pts with CD4 ct <200 cells/µL experienced fewer irAEs than pts with CD4 ct ≥200 cells/µL (2/36 vs 26/97; p<0.01). The incidence of any grade irAEs was similar between pts with CD4/CD8 ratio <0.4 vs ≥0.4 (8/54 vs 18/72; p=0.16) and between pts with HIV VL ≥400 vs <400 copies/mL (1/12 vs 28/124; p=0.46). Clinical outcomes are shown in the table below. Among 29 pts with met non-small cell lung ca (NSCLC) with available CD4 ct, the ORR of pts with CD4 ct <200 cells/µL was 13% (95% CI:0-53) vs 38% (95% CI:18-62) in pts with CD4 ct ≥ 200 cells/µL (1/8 vs 8/21; p=0.38). Conclusions: In the largest dataset to our knowledge, we demonstrate tolerability and activity of ICI among PLWH regardless of CD4 ct and HIV VL levels. CD4 ct <200 cells/µL may be associated with a lower incidence of irAEs. An analysis of a larger cohort is underway. [Table: see text]

Patients Who Do Not Receive Aspirin Because of Allergy Have an Increased Risk of Venous Thromboembolism Following Total Joint Arthroplasty

BACKGROUND

Patient-reported allergies to aspirin or nonsteroidal anti-inflammatory drugs (NSAIDs) may preclude certain patients from receiving aspirin as venous thromboembolism (VTE) prophylaxis after total joint arthroplasty (TJA). The purpose of the study was to (1) determine whether the use of non-aspirin agents due to a self-reported aspirin or NSAID allergy was associated with a higher incidence of VTE and (2) determine the rate of true allergic reactions in patients who reported an allergy but still received aspirin.

METHODS

Prospectively collected data from 45,171 patients who underwent primary TJA between 2000 and 2019 were reviewed. Patients who reported an allergy to aspirin or NSAIDs were identified (n = 823). Using a validated VTE risk calculator, each patient was assigned a risk score based on 26 comorbidities. Bleeding complications, VTEs that occurred within 90 days postoperatively, and allergic reactions were collected as end points. Multivariable logistic regression was performed to determine the factors associated with VTE.

RESULTS

The overall incidence of aspirin or NSAID allergy was 1.8%. Only 80 of 267 patients (30%) reported an allergy to only aspirin and still received aspirin. Compared with patients who received aspirin without a history of allergy (n = 17,648), patients who received non-aspirin thromboprophylaxis due to allergy (n = 383) had a higher incidence of VTE (2.87% versus 0.24%, p < 0.001). On multivariable regression, these patients had an 8-fold increase in VTE risk (adjusted odds ratio, 7.94; 95% confidence interval, 2.86 to 22.07; p < 0.001). The incidence of true allergic reactions to aspirin among those with a reported allergy was 1.8% (number needed to harm = 55). No patients developed anaphylaxis or severe hypersensitivity reactions.

CONCLUSIONS

Patients with a self-reported allergy to aspirin or NSAIDs were at a significantly increased risk for VTE if they received non-aspirin thromboprophylaxis agents following TJA. Future research should evaluate the usefulness of preoperative allergy testing when selecting a VTE prophylactic agent in patients with self-reported allergies.

LEVEL OF EVIDENCE

Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Interim Spacer Exchange for Treatment of Periprosthetic Joint Infection: Almost Half the Patients Subsequently Fail

BACKGROUND

Interim spacer exchange may be performed in patients undergoing 2-stage exchange for periprosthetic joint infection. Several studies have demonstrated that interim spacer exchange is associated with poor outcomes. This study investigated the survivorship and risk factors for failure in patients with an interim spacer exchange.

METHODS

Two institutional databases identified 182 patients who underwent spacer exchange from 2000 to 2017. Primary outcomes included progression to reimplantation, treatment success, and mortality. Bivariate analysis was performed to evaluate risk factors associated with failure. Kaplan-Meier curves using host and local grades were generated to evaluate for primary outcomes and differences in survivorship.

RESULTS

The overall failure rate was 49% in patients with a spacer exchange. Most patients (60%) failed before 2 years. Higher comorbidity scores, elevated erythrocyte sedimentation rate, and non-White race were more prevalent in patients who failed. Negative cultures at the time of exchange were more prevalent in patients who did not fail. Failure rate was higher in immunocompromised conditions, and those who had revision prior to exchange. After considering clinically relevant variables, advanced host grade C was the single factor associated with treatment failure. Although survivorship curves were not significantly different between extremity local grades, higher host grades were associated with treatment failure.

CONCLUSION

Almost 1 out of 2 patients with spacer exchange were found to fail the intended 2-stage revision arthroplasty. Benefits of delivering additional antibiotic load with a new spacer should be balanced against poor outcomes in patients with the aforementioned risk factors.

Red Cell Distribution Width: Commonly Performed Test Predicts Mortality in Primary Total Joint Arthroplasty

INTRODUCTION

Mortality after total joint arthroplasty (TJA) has been thoroughly explored. Short and long-term mortality appear to be correlated with patient comorbidities. Red Cell Distribution Width (RDW) is a commonly performed test that reflects the variation in red blood cell size. This study investigated the utility of RDW, when combined with comorbidity indices, in predicting mortality after TJA.

METHODS

Using a single institutional database, 30,437 primary TJA were identified. Patient demographics (age, gender, body mass index (BMI), pre-operative hemoglobin, RDW, and Charlson Comorbidity Index(CCI)) were queried. The primary outcome was 1-year mortality after TJA. Anemia was defined as hemoglobin <12g/dL for women and <13 g/dL for men. The normal range for RDW is 11.5-14.5%. A preliminary analysis assessed the bivariate association between demographics, preoperative anemia, RDW, CCI, and all-cause mortality within 1-year after TJA. A multivariate regression model was conducted to determine independent predictors of 1-year mortality. Finally, ROC curves were used to compare AUC of RDW, CCI and the combination of both in predicting 1-year mortality.

RESULTS

The mean RDW was 13.6% ± 1.2. Eighteen percent of patients had pre-operative anemia. The mean CCI was 0.4 ± 0.9. RDW, anemia, CCI, and age were significantly associated with a higher incidence of 1-year mortality. RDW, CCI, age, and male sex were found to be independent risk factors for 1-year mortality. RDW (AUC = 0.68) was a better predictor of mortality compared to CCI (AUC = 0.66). The combination of RDW and CCI (AUC = 0.76) predicted 1-year mortality more accurately than CCI or RDW alone.

CONCLUSION

RDW appears to be a useful parameter that, when combined with CCI, can predict the risk for 1-year mortality after TJA.

Aspirin Is Safe for Venous Thromboembolism Prophylaxis for Patients With a History of Gastrointestinal Issues

BACKGROUND

The safety of acetylsalicylic acid (ASA, aspirin) in patients with prior history of gastroesophageal reflux or peptic ulcer disease remains unclear. The purpose of this study was to determine the safety of ASA for venous thromboembolism (VTE) prophylaxis after total joint arthroplasty in patients with prior history of gastrointestinal (GI) issues.

METHODS

This was an institutional, retrospective cohort study of 19,044 patients who underwent primary total hip and total knee arthroplasty from 2013 to 2019. We divided the patients into two cohorts based on the presence or absence of pre-existing GI issues. Patient demographics, VTE prophylaxis, and postoperative complications were collected. The primary outcome measure was GI bleed.

RESULTS

In our series, 3090 patients had a preoperative GI issue and 15,954 did not have a GI issue. ASA was the most common mode of VTE prophylaxis (89%), followed by Coumadin (4.7%), direct oralanticoagulants (4.2%), low-molecular-weight heparin (1.7%), and others (0.4%). In the cohort of patients given ASA, there was no significant difference in postoperative GI bleeding between those with (2/1781, 0.11%) and without preoperative GI issues (8/7,628, 0.10%, P = 1.0). In the overall cohort, history of preoperative GI issues was associated with an increased risk of postoperative GI bleeding (0.32% vs 0.11%, P = .031). In logistic regression analysis, ASA was associated with a protective effect against GI bleed (OR = 0.09, 95% CI 0.01-0.40, P = .003).

CONCLUSION

ASA is safe for VTE prophylaxis after total joint arthroplasty in patients with history of GI issues and is not associated with an increased risk of postoperative GI bleeds.

Canceled Total Joint Arthroplasty: Who, What, When, and Why?

BACKGROUND

Unexpected cancelation of scheduled total joint arthroplasty (TJA) procedures creates patient distress and disruption for the clinical team. The purpose of this study is to identify the etiology and fate of cancelations for scheduled TJAs.

METHODS

A consecutive series of 11,670 patients at a single institution from 2013 to 2017 was reviewed in March 2020. All patients who were scheduled for a primary total hip arthroplasty or total knee arthroplasty and subsequently canceled were identified. The etiology of cancelation and time to rescheduling were recorded.

RESULTS

Of the 505 (4.3%) canceled patients, 209 (42%) were due to medical reasons. Three hundred ninety-one patients (77%) eventually underwent their procedure at a mean delay of 165 days (19-1908). Only 53 (25%) patients canceled for a medical reason underwent further diagnostic or therapeutic intervention for their medical condition. When compared to patient-driven cancelations, those canceled for medical reasons had a higher mean Charlson Comorbidity Index (0.82 vs 0.39, P < .001), were canceled closer to the scheduled surgery date (8.55 vs 18.1 days, P < .001), and were more likely to eventually undergo surgery (86% vs 73%, P = .004).

CONCLUSION

Canceled elective TJA surgeries are most often due to a medical concern, however only a minority of these patients undergo intervention for that medical condition. To minimize the risk of cancelation, healthcare providers may consider early referral of medically complex patients to the patient's primary care physician. After cancelation, patients should have a clearly defined path to return to the operative schedule to prevent further delays.

African American Patients Have Improved Functional Gains and Comparable Clinical Outcomes to Caucasian Patients After Total Hip and Knee Arthroplasty

BACKGROUND

With the increasing popularity of alternative payment models, minorities who use more postacute care resources may face difficulties with access to quality total hip arthroplasty (THA) and total knee arthroplasty (TKA) care. The purpose of this study is to compare differences in perioperative complications and functional outcomes between African American and Caucasian patients undergoing THA and TKA.

METHODS

We reviewed a consecutive series of all primary THA and TKA patients at our institution from 2015 to 2018. Demographics, comorbidities, 90-day complications, readmissions, Veterans Rand 12-Item Health Survey (VR-12), Hip disability Osteoarthritis Outcome Score (HOOS), and Knee injury and Osteoarthritis Outcome Scores (KOOS) were compared between African American and Caucasian patients. A multivariate analysis was performed to control for confounding variables.

RESULTS

Of the 5284 patients included in the study, 1041 were African American (24.5%). Although African American patients had lower preoperative HOOS/KOOS (33.5 vs 45.1, P < .001) and mental VR-12 scores (37.8 vs 51.5, P < .001) compared with Caucasian patients, there was no clinical difference at 1 year in HOOS/KOOS (50.2 vs 50.4), mental VR-12 (55.0 vs 52.6), or physical VR-12 scores (39.5 vs 39.8). When controlling for demographics and medical comorbidities, African American race was associated with increased rehabilitation facility discharge (odds ratio, 1.69; P < .001) but no difference in readmissions or complications.

CONCLUSION

Although African American patients had lower preoperative functional scores, they made improved postoperative gains when compared with Caucasian patients. Although there was no difference in postoperative complications, further studies should assess social causes for the increase in rehabilitation utilization rates in minority patients.

High Bundled Payment Costs Following Total Hip and Knee Arthroplasty in African Americans

BACKGROUND

Concerns exist that minorities who utilize more resources in an episode-of-care following total hip (THA) and knee arthroplasty (TKA) may face difficulties with access to quality arthroplasty care in bundled payment programs. The purpose of this study is to determine if African American patients undergoing TKA or THA have higher episode-of-care costs compared to Caucasian patients.

METHODS

We queried Medicare claims data for a consecutive series of 7310 primary TKA and THA patients at our institution from 2015 to 2018. We compared patient demographics, comorbidities, readmissions, and 90-day episode-of-care costs between African American and Caucasian patients. A multivariate regression analysis was performed to identify the independent effect of race on episode-of-care costs.

RESULTS

Compared to Caucasians, African Americans were younger, but had higher rates of pulmonary disease and diabetes. African American patients had increased rates of discharge to a rehabilitation facility (20% vs 13%, P < .001), with higher subacute rehabilitation ($1909 vs $1284, P < .001), home health ($819 vs $698, P = .022), post-acute care ($5656 vs $4961, P = .008), and overall 90-day episode-of-care costs ($19,457 vs $18,694, P = .001). When controlling for confounding comorbidities, African American race was associated with higher episode-of-care costs of $440 (P < .001).

CONCLUSION

African American patients have increased episode-of-care costs following THA and TKA when compared to Caucasian patients, mainly due to increased rates of home health and rehabilitation utilization. Further study is needed to identify social variables that can help reduce post-acute care resources and prevent reduction in access to arthroplasty care in bundled payment models.

Characterizing Femoral and Acetabular Bone Loss in Two-Stage Revision Total Hip Arthroplasty for Infection

BACKGROUND

The process of infection and multiple procedure-related factors in 2-stage exchange arthroplasty may lead to severe bone loss that can complicate subsequent reimplantation. The primary purpose of this study is to quantitatively characterize the extent of bone loss on the femoral and acetabular side prior to reimplantation during 2-stage revision hip arthroplasty for infection.

METHODS

This retrospective case series includes 47 patients with planned 2-stage exchange arthroplasty for treatment of infected total hip replacement. Demographic, clinical, and surgical information, and complications during the 2-stage process were collected on all patients. The radiographs of all patients were reviewed prior to first-stage explantation and post-reimplantation or latest follow-up to characterize bone loss using the Paprosky classification systems.

RESULTS

Of the 47 patients with planned 2-stage exchange, 10 patients (21%) were not reimplanted. Following explant and spacer placement, 32 patients (68%) experienced progressive femoral bone loss, and 20 (43%) experienced progressive acetabular bone loss. Patients who experienced progression of both femoral and acetabular bone (14, 30%) had a longer time between resection and reimplantation (164 ± 128 vs 88 ± 26 days, P = .03). A reimplantation time of greater than 90 days was associated with an increased risk of combined bone loss progression (15% vs 53%, P = .01). For patients who underwent reimplantation (79%), increased bone loss led to high rates of diaphyseal femoral fixation (97%) and revision acetabular components (38%).

CONCLUSION

Increased time to reimplantation in 2-stage exchange arthroplasty for prosthetic hip infections is associated with a high rate of acetabular and femoral bone loss, and increased complexity of reimplantation.