HIV Lymphadenopathy: Differential Diagnosis and Important Imaging Features

Early diagnosis of HIV cases by means of text mining and machine learning models on clinical notes

Undiagnosed and untreated human immunodeficiency virus (HIV) infection increases morbidity in the HIV-positive person and allows onward transmission of the virus. Minimizing missed opportunities for HIV diagnosis when a patient visits a healthcare facility is essential in restraining the epidemic and working toward its eventual elimination. Most state-of-the-art proposals employ machine learning (ML) methods and structured data to enhance HIV diagnoses, however, there is a dearth of recent proposals utilizing unstructured textual data from Electronic Health Records (EHRs). In this work, we propose to use only the unstructured text of the clinical notes as evidence for the classification of patients as suspected or not suspected. For this purpose, we first compile a dataset of real clinical notes from a hospital with patients classified as suspects and non-suspects of having HIV. Then, we evaluate the effectiveness of two types of classification models to identify patients suspected of being infected with the virus: classical ML algorithms and two Large Language Models (LLMs) from the biomedical domain in Spanish. The results show that both LLMs outperform classical ML algorithms in the two settings we explore: one dataset version is balanced, containing an equal number of suspicious and non-suspicious patients, while the other reflects the real distribution of patients in the hospital, being unbalanced. We obtain F1 score figures of 94.7 with both LLMs in the unbalanced setting, while in the balance one, RoBERTaBio model outperforms the other one with a F1 score of 95.7. The findings indicate that leveraging unstructured text with LLMs in the biomedical domain yields promising outcomes in diminishing missed opportunities for HIV diagnosis. A tool based on our system could assist a doctor in deciding whether a patient in consultation should undergo a serological test.

The Etiologic Landscape of Lymphoproliferation in Childhood: Proposal for a Diagnostic Approach Exploring from Infections to Inborn Errors of Immunity and Metabolic Diseases

Lymphoproliferation is defined by lymphadenopathy, splenomegaly, hepatomegaly, or lymphocytic organ and tissue infiltration. The most common etiologies of lymphoproliferation are represented by infectious diseases and lymphoid malignancies. However, it is increasingly recognized that lymphoproliferative features can be the presenting sign of rare conditions, including inborn errors of immunity (IEI) and inborn errors of metabolism (IEM). Among IEI, lymphoproliferation is frequently observed in autoimmune lymphoproliferative syndrome (ALPS) and related disorders, common variable immunodeficiency (CVID), activated phosphoinositide 3-kinase δ syndrome, and Epstein-Barr virus (EBV)-related disorders. Gaucher disease and Niemann-Pick disease are the most common IEMs that can present with isolated lymphoproliferative features. Notably, other rare conditions, such as sarcoidosis, Castleman disease, systemic autoimmune diseases, and autoinflammatory disorders, should be considered in the differential diagnosis of patients with persistent lymphoproliferation when infectious and malignant diseases have been reasonably ruled out. The clinical features of lymphoproliferative diseases, as well as the associated clinical findings and data deriving from imaging and first-level laboratory investigations, could significantly help in providing the correct diagnostic suspicion for the underlying etiology. This paper reviews the most relevant diseases associated with lymphoproliferation, including infectious diseases, hematological malignancies, IEI, and IEM. Moreover, some practical indications to orient the initial diagnostic process are provided, and two diagnostic algorithms are proposed for the first-level assessment and the approach to persistent lymphoproliferation, respectively.

Research to develop a diagnostic ultrasound nomogram to predict benign or malignant lymph nodes in HIV-infected patients

BACKGROUND

This study aimed to establish an effective ultrasound diagnostic nomogram for benign or malignant lymph nodes in HIV-infected patients.

METHODS

The nomogram is based on a retrospective study of 131 HIV-infected patients who underwent ultrasound assess at the Shanghai Public Health Clinical Center from December 2017 to July 2022. The nomogram's predictive accuracy and discriminative ability were determined by concordance index (C-index) and calibration curve analysis. A nomogram combining the lymph node US characteristics were generated based on the multivariate logistic regression results.

RESULTS

Predictors contained in the ultrasound diagnostic nomogram included age (OR 1.044 95%CI: 1.014-1.074 P = 0.004), number of enlarged lymph node regions (OR 5.445 95%CI: 1.139-26.029 P = 0.034), and color Doppler flow imaging (CDFI) grades (OR 9.614 95%CI: 1.889-48.930 P = 0.006). The model displayed good discrimination with a C (ROC) of 0.775 and good calibration.

CONCLUSIONS

The proposed nomogram may result in more-accurate diagnostic predictions for benign or malignant lymph nodes in patients with HIV infection.

Ultrasound findings in Kaposi sarcoma patients: overlapping sonographic features with disseminated tuberculosis

BACKGROUND

Focused Assessment with Sonography for HIV-associated TB (FASH) is a diagnostic tool for extra-pulmonary tuberculosis (TB) in symptomatic patients with advanced HIV. As Kaposi's sarcoma (KS) is also prevalent in this patient population, changes due to KS may mimic TB findings and clinical interpretation of target FASH findings can be challenging. We aimed to describe sonographic findings in patients with KS.

METHODS

We performed a prospective observational study at Lighthouse clinic at Kamuzu Central Hospital, Lilongwe, Malawi, in consecutive patients with newly diagnosed KS, without known diagnosis of TB, referred for paclitaxel treatment. All patients underwent FASH and abdominal ultrasound to assess for effusions and changes in liver and spleen, as well as systematic sonographic assessment for lymphadenopathy.

RESULTS

We included 30 patients. We found inguinal lymph nodes using ultrasound in 20 patients; in 3 (10%) additionally abdominal lymph nodes were found. Pathological effusions were seen in eight patients (27%): pericardial effusion in one (3%), pleural effusion in six (20%) and ascites in four (13%) patients. We found focal spleen lesions in three (10%) patients. Most of these lesions were echogenic, but in one patient, we saw hypoechoic lesions with an echogenic center. In three (10%) patients an unusual "sponge-like pattern" of the splenic vasculature was found. Six (20%) patients had echogenic focal lesions in the liver resembling hemangiomas, individual lesions showing a hypoechoic center. In two patients echogenic portal fields were seen.

CONCLUSIONS

The majority of patients with newly diagnosed KS demonstrate sonographic features of disease, predominantly lymphadenopathy. Effusions were observed in a significant minority, as well as focal lesions in liver or spleen, which commonly resemble hemangiomas, but hypoechoic lesions were also observed and can easily be mistaken for extra-pulmonary TB. A 'sponge-like pattern' of the spleen should not be confused with micro-abscesses. In conclusion, this case series illustrates the diverse nature of ultrasound features in patients with KS, which can be difficult to distinguish from other opportunistic diseases, including TB.

Clinical value of 18F-FDG PET/CT in the management of HIV-associated lymphoma

HIV is still a major public health problem. At present, HIV-associated lymphoma remains the leading cause of deaths among people living with HIV, which should be paid more attention to. ¹⁸F-fluorodeoxglucose (FDG) PET/CT has been recommended in the initial staging, restaging, response assessment and prognostic prediction of lymphomas in general population. HIV-associated lymphoma is, however, a different entity from lymphoma in HIV-negative with a poorer prognosis. The ability to accurately risk-stratify HIV-infected patients with lymphoma will help guide treatment strategy and improve the prognosis. In the review, the current clinical applications of ¹⁸F-FDG PET/CT in HIV-associated lymphoma will be discussed, such as diagnosis, initial staging, response evaluation, prognostic prediction, PET-guided radiotherapy decision, and surveillance for recurrence. Moreover, future perspectives will also be presented.

Clinical features and 18F-FDG PET/CT for distinguishing of malignant lymphoma from inflammatory lymphadenopathy in HIV-infected patients

Background

It is vital to distinguish between inflammatory and malignant lymphadenopathy in human immunodeficiency virus (HIV) infected individuals. The purpose of our study was to differentiate the variations in the clinical characteristics of HIV patients, and apply ¹⁸F-FDG PET/CT parameters for distinguishing of malignant lymphoma and inflammatory lymphadenopathy in such patients.

Methods

This retrospective cross-sectional study included 59 consecutive HIV-infected patients who underwent whole-body ¹⁸F-FDG PET/CT. Of these patients, 37 had biopsy-proven HIV-associated lymphoma, and 22 with HIV-associated inflammatory lymphadenopathy were used as controls. The determined parameters were the maximum of standard uptake value (SUV_max), SUV_max of only lymph nodes (SUV_LN), the most FDG-avid lesion-to-liver SUV_max ratio (SUR_max), laboratory examinations and demographics. The optimal cut-off of ¹⁸F-FDG PET/CT value was analyzed by receiver operating characteristic curve (ROC).

Results

Considering the clinical records, the Karnofsky Performance Status (KPS) scores in patients with inflammatory lymphadenopathy were obviously higher than those in patients with malignant lymphoma (P = 0.015), whereas lymphocyte counts and lactate dehydrogenase (LDH) were obviously lower (P = 0.014 and 0.010, respectively). For the ¹⁸F-FDG PET/CT imaging, extra-lymphatic lesions, especially digestive tract and Waldeyer’s ring, occurred more frequently in malignant lymphoma than inflammatory lymphadenopathy. Furthermore, the SUR_max and SUV_LN in malignant lymphoma were markedly higher than those in inflammatory lymphadenopathy (P = 0.000 and 0.000, respectively). The cut-off point of 3.1 for SUR_max had higher specificity (91.9%) and relatively reasonable sensitivity (68.2%) and the cut-off point of 8.0 for the SUV_LN had high specificity (89.2%) and relatively reasonable sensitivity (63.6%).

Conclusion

Our study identified the distinctive characteristics of the clinical manifestations, the SUR_max, SUV_LN and detectability of extra-lymphatic lesions on ¹⁸F-FDG PET, and thus provides a new basis for distinguishing of malignant lymphoma from inflammatory lymphadenopathy in HIV-infected patients.

Multilocular Thymic Cyst in a Patient with Untreated HIV/AIDS: Case Report and Review of the Literature

BACKGROUND

Multilocular thymic cysts (MTCs) in adults with human immunodeficiency virus (HIV) are rarely reported.

CASE PRESENTATION

We describe a case of symptomatic MTC in a male with untreated HIV. A presumptive diagnosis was established based on radiographic imaging and biopsy. Pathologic diagnosis and exclusion of malignancy were ultimately confirmed the following thymectomy. In conjunction with starting antiretroviral therapy, the patient recovered well post-operatively with a resolution of his presenting symptoms.

CONCLUSION

Our case report and review of the literature serve to highlight MTCs as an important clinical entity occurring in persons with HIV.

The Value of Nutritional Status in the Prognostic Analysis of Patients with AIDS-Related Lymphoma

Objective

Many studies have suggested that indexes of nutritional status, such as body mass index (BMI), serum albumin (ALB), serum pre-albumin (PA), and hemoglobin, may be used as risk factors for the prognosis of HIV or lymphoma. Therefore, this study aimed to retrospectively analyze and explore the value of nutritional status in the prognostic assessment of patients with AIDS-related lymphoma (ARL).

Methods

In this retrospective study, the clinical data of 69 patients with ARL were collected. All patients had a definite diagnosis of non-Hodgkin lymphoma by pathological examination and met the requirements of the Hematopoietic and Lymphocytic Tissue Tumor Classification (2016) established by the World Health Organization. Patients who did not receive standard chemotherapy, those with incomplete medical records, and those with an unclear pathological diagnosis were excluded. The patients were divided into two groups (survival and death) according to the prognostic outcome, and their clinical characteristics and prognoses were discussed by relevant statistical methods.

Results

During the three-year follow-up period, 20 (28.99%) patients died, and 49 (71.01%) survived. The one-year cumulative survival rate was 78.26%. A univariate analysis found that the prognosis was associated with the International Prognostic Index (IPI) score, BMI, ALB, PA, and CD4 T lymphocyte count. The Cox risk proportional regression analysis showed that the IPI score, BMI, and PA were the independent risk factors for survival; their combination had a greater ability to forecast the clinical outcome (area under the curve = 0.874, P < 0.001).

Conclusion

In this study, at the time of the visit, the patients with ARL tended to be in the advanced stages of disease and, therefore, at high risk of mortality. Therefore, their nutritional status might be of great value to the prognostic assessment. The combination of BMI, PA, and IPI scores could be used for risk stratification and better screening of high-risk patients.