Hypertension and cardiac damage in pheochromocytoma and paraganglioma patients: a large-scale single-center cohort study.

BACKGROUND: Hypertension (HT) is one of the most common manifestations in patients with catecholamine-secreting neuroendocrine tumors. Although the cardiovascular manifestations of these tumors have been described, there have been no large-scale investigations of the profile of HT and changes in cardiac structure and function that occur in patients with pheochromocytomas and paragangliomas (PPGL). MATERIALS AND METHODS: In this study, we investigated the prevalence of HT and left ventricular remodeling (LVR) in a cohort of 598 patients who underwent surgery for PPGL at our center between January 2001 and April 2022. Information on demographics, reason for hospitalization, medical history, biochemical parameters, findings on echocardiography, and tumor characteristics were recorded. The LVR index was compared according to whether or not there was a history of HT. RESULTS: The average age was 47.07 ± 15.07 years, and 277 (46.32%) of the patients were male. A history of HT was found in 423 (70.74%) of the 598 patients. Paraganglioma was significantly more common in the group with HT (26.00% vs. 17.71%, P = 0.030) and significantly less likely to be found incidentally during a health check-up in this group (22.93% vs. 59.43%, P < 0.001). Among 365 patients with complete echocardiography data, left ventricular mass index (86.58 ± 26.70 vs. 75.80 ± 17.26, P < 0.001) and relative wall thickness (0.43 ± 0. 08 vs. 0.41 ± 0.06, P = 0.012) were significantly higher in patients with PPGL and a history of HT. The proportions with left ventricular hypertrophy (LVH) (19.40% vs. 8.25%, P = 0.011) and LVR (53.73% vs. 39.18%, P = 0.014) were also higher when there was a history of HT. After adjusting for age, gender, body mass index, alcohol consumption, smoking status, diabetes, stroke, creatinine level, tumor location, and tumor size, a history of HT was significantly correlated with LVH (odds ratio 2.71, 95% confidence interval 1.18-6.19; P = 0.018) and LVR (odds ratio 1.83, 95% confidence interval 1.11-3.03; P = 0.018). CONCLUSION: HT is common in patients with PPGL (70.74% in this cohort). PPGL without a history of HT is more likely to be found incidentally (59.43% in our cohort). HT is associated with LVR in PPGL patients with complete echocardiography data. These patients should be observed carefully for cardiac damage, especially those with a history of HT.

Evaluating the diagnostic efficacy of whole-body MRI versus 123I-mIBG/131I-mIBG imaging in metastatic pheochromocytoma and paraganglioma.

This study aimed to compare tumor lesion detectability and diagnostic accuracy of whole-body magnetic resonance imaging (WB-MRI) and radioiodine-labeled meta-iodo-benzylguanidine (mIBG) imaging techniques in patients with metastatic pheochromocytoma and paraganglioma (PPGL). This retrospective study included 13 patients had pheochromocytoma and 5 had paraganglioma, who were all suspected of having metastatic tumors. Each patient underwent WB-MRI and 123I-mIBG as a pretreatment screening for 131I-mIBG therapy. Two expert reviewers evaluated WB-MRI, 123I-mIBG images, and post-therapy 131I-mIBG images for the presence of metastatic lesions in the lungs, bones, liver, lymph nodes, and other organs. Diagnostic measures for detecting metastatic lesions, including sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), and receiver operating characteristics (ROC)-area under the curve (AUC), were calculated for each imaging technique. We analyzed WB-MRI images for detecting metastatic lesions, which demonstrated sensitivity, specificity, accuracy, PPV, NPV, and AUC of 82%, 97%, 90%, 96%, 86%, and 0.92, respectively. These values were 83%, 95%, 89%, 94%, 86%, and 0.90 in 123I-mIBG images and 85%, 92%, 89%, 91%, 87%, and 0.91 in post-therapy 131I-mIBG images, respectively. Our results reveal the comparable diagnostic accuracy of WB-MRI to one of the mIBG images.

Evaluation of pharmacokinetics, safety, and efficacy of [211At] meta-astatobenzylguanidine ([211At] MABG) in patients with pheochromocytoma or paraganglioma (PPGL): A study protocol.

BACKGROUND: Pheochromocytoma, or paraganglioma (PPGL), is a tumor that arises from catecholamine-producing chromaffin cells of the adrenal medulla or paraganglion. Systemic therapy, such as the combination of cyclophosphamide, vincristine, and dacarbazine or therapeutic radiopharmaceuticals such as [131I] meta-iodobenzylguanidine (MIBG), may be administered in cases of locally advanced tumors or distant metastases. However, the current therapies are limited in terms of efficacy and implementation. [211At] meta-astatobenzylguanidine (MABG) is an alpha-emitting radionuclide-labeled ligand that has demonstrated remarkable tumor-reducing effects in preclinical studies, and is expected to have a high therapeutic effect on pheochromocytoma cells. METHODS: We are currently conducting an investigator-initiated first-in-human clinical trial to evaluate the pharmacokinetics, safety, and efficacy of [211At] MABG. Patients with locally unresectable or metastatic PPGL refractory to standard therapy and scintigraphically positive [123I] MIBG aggregation are being recruited, and a 3 + 3 dose escalation design was adopted. The initial dose of [211At] MABG is 0.65 MBq/kg, with a dose escalation in a 1:2:4 ratio in each cohort. Dose-limiting toxicity is observed for 6 weeks after a single bolus dose of [211At] MABG, and the patients are observed for 3 months to explore safety and efficacy profiles. The primary endpoint is dose-limiting toxicity to determine both maximum tolerated and recommended doses. The secondary endpoints include radiopharmacokinetics, urinary radioactive excretion rate, urinary catecholamine response rate, objective response rate, progression free survival, [123I] MIBG scintigraphy on reducing tumor accumulation, and quality of life. TRIALS REGISTRATION: jRCT2021220012 registered on 17 June 2022.

A Comparison of Near-Infrared Autofluorescence Findings in Benign Versus Malignant Adrenal Tumors.

BACKGROUND: Many adrenal tumors are deemed radiologically indeterminate and surgically removed. Adrenal tissue, like parathyroid glands, exhibits near-infrared autofluorescence (NIRAF) properties. This study was designed to investigate the potential of NIRAF to differentiate benign versus malignant adrenal tumors. METHODS: Patients undergoing adrenalectomy between October 2021 and May 2023 were prospectively studied. Adrenalectomy specimens were inspected with NIRAF imaging. Specimen autofluorescence (AF) characteristics were recorded. Comparisons were made between different tumor types and a logistic regression model was constructed to differentiate benign versus malignant tumors. A receiver operating characteristic curve was used to identify an optimal AF threshold differentiating benign versus malignant tumors. RESULTS: A total of 108 adrenal specimens were examined: adrenocortical adenomas/other benign lesions (n = 72), pheochromocytomas (n = 18), adrenocortical neoplasms of uncertain behavior (n = 4), and malignant tumors (n = 14). A significant difference in normalized AF intensity was identified when comparing adrenocortical adenomas (3.08 times background) with pheochromocytomas (1.95, p = 0.001) and malignant tumors (1.11, p < 0.0001). The Area Under the Curve differentiating benign vs malignant tumors was 0.87, with an optimal normalized AF threshold at 1.93. CONCLUSIONS: Different adrenal pathologies exhibit diverse AF properties. These findings suggest a potential intraoperative utility of NIRAF in predicting benign versus malignant nature for radiologically indeterminate adrenal tumors.

Computed tomography-guided percutaneous cryoablation of hereditary adrenal pheochromocytoma in three patients.

OBJECTIVES: Pheochromocytomas (PHEO) are neuroendocrine tumors rarely diagnosed in children. We are reporting on the management challenges of three adolescent patients who present with hereditary PHEO. CASE PRESENTATION: The index patient and his male sibling presented with bilateral PHEO, while a third patient presented with a unilateral PHEO, all associated with von Hippel-Lindau (VHL) syndrome. The patients were treated with computed tomography (CT)-guided percutaneous cryoablation (CRA) of the adrenal lesions, with varying degrees of success. CONCLUSIONS: CT-guided percutaneous CRA of hereditary PHEO has not been reported in the pediatric population and may represent a novel treatment strategy that reduces the risk of intraprocedural complications and adrenal insufficiency (AI).

High-Specific-Activity 131 I-MIBG for the Treatment of Advanced Pheochromocytoma and Paraganglioma.

PATIENTS AND METHODS: The primary endpoints were objective response rate (ORR) and disease control rate (DCR). Secondary endpoints were duration of response, blood pressure control, safety, overall and progression-free survival rates, MIBG uptake, and correlations with genetic background. RESULTS: The study included 25 patients. Twenty-four patients had distant metastases, 17 (68%) had hormonally active tumors, and 13 (52%) had previously received antineoplastic treatment. In 24 evaluable patients, the ORR was 38%, including 2 patients with complete response, and the DCR was 83%; median time to response was 12.5 months (95% confidence interval, 4.6-25.1). Twelve patients had sporadic disease, among whom the ORR was 25% and DCR was 83%. Twelve patients had hereditary disease ( SDHB , VHL , RET ); among these, the ORR was 50%, and DCR was 83%. Plasma metanephrines normalized in 30% of patients and improved by greater than 50% in 46%. Sixteen patients had hormonally active tumors and hypertension; in 9 (56%) of these, blood pressure normalized, leading to discontinuation of antihypertensive therapy.The most common adverse events were grades 1-2 nausea/vomiting and transient bone marrow suppression. One patient developed premature ovarian failure. Reversible grades 3-4 myelosuppression were seen in 7 patients (28%). One patient had fatal pneumonitis. CONCLUSIONS: HSA- 131 I-MIBG is associated with a high DCR in patients with MPPGL, regardless of underlying genetic mutation.

CT-based radiomics research for discriminating the risk stratification of pheochromocytoma using different machine learning models: a multi-center study.

OBJECTIVES: The purpose of this study was to explore and verify the value of various machine learning models in preoperative risk stratification of pheochromocytoma. METHODS: A total of 155 patients diagnosed with pheochromocytoma through surgical pathology were included in this research (training cohort: n = 105; test cohort: n = 50); the risk stratification scoring system classified a PASS score of < 4 as low risk and a PASS score of ≥ 4 as high risk. From CT images captured during the non-enhanced, arterial, and portal venous phase, radiomic features were extracted. After reducing dimensions and selecting features, Logistic Regression (LR), Extra Trees, and K-Nearest Neighbor (KNN) were utilized to construct the radiomics models. By adopting ROC curve analysis, the optimal radiomics model was selected. Univariate and multivariate logistic regression analyses of clinical radiological features were used to determine the variables and establish a clinical model. The integration of radiomics and clinical features resulted in the creation of a combined model. ROC curve analysis was used to evaluate the performance of the model, while decision curve analysis (DCA) was employed to assess its clinical value. RESULTS: 3591 radiomics features were extracted from the region of interest in unenhanced and dual-phase (arterial and portal venous phase) CT images. 13 radiomics features were deemed to be valuable. The LR model demonstrated the highest prediction efficiency and robustness among the tested radiomics models, with an AUC of 0.877 in the training cohort and 0.857 in the test cohort. Ultimately, the composite of clinical features was utilized to formulate the clinical model. The combined model demonstrated the best discriminative ability (AUC, training cohort: 0.887; test cohort: 0.874). The DCA of the combined model showed the best clinical efficacy. CONCLUSION: The combined model integrating radiomics and clinical features had an outstanding performance in differentiating the risk of pheochromocytoma and could offer a non-intrusive and effective approach for making clinical decisions.

Cannabinoid hyperemesis and pheochromocytoma hypertensive urgency: a case report.

BACKGROUND: This report presents a case of cannabinoid-induced hyperemesis syndrome causing repeated violent retching in a patient with a large (8 cm) adrenal pheochromocytoma resulting in hypertensive urgency. CASE PRESENTATION: A 69-year-old white male patient with a previously diagnosed pheochromocytoma presented to the emergency department with nausea and vomiting and was found to have hypertensive urgency. Computed tomography scan did not show any acute abdominal pathology and history was inconsistent with a gastrointestinal etiology. Patient had a history of daily cannabinoid use for many years and repeated self-limited hyperemesis episodes, and thus a diagnosis of cannabinoid-induced hyperemesis syndrome was made. It was concluded that the likely explanation for the hypertensive urgency was from physical compression of his adrenal tumor during the episodes of retching resulting in a catecholamine surge. The patient was given antiemetics and admitted to the intensive care unit for blood pressure management. Blood pressure was initially controlled with phentolamine and a clevidipine infusion, then transitioned to oral doxazosin and phenoxybenzamine. Hyperemesis and abdominal pain resolved after 24 hours, and his blood pressure returned to baseline. The patient was discharged with the recommendation to stop all cannabis use. On follow-up, his blood pressure remained well controlled, and he subsequently underwent adrenalectomy for tumor removal. CONCLUSION: Hyperemesis can cause hypertensive events in patients with pheochromocytoma by increasing abdominal pressure, leading to catecholamine release.

Approach to the Patient With Bilateral Adrenal Masses.

Bilateral adrenal masses, increasingly encountered in clinical practice, manifest across diverse contexts, including incidental discovery, malignancy staging, and targeted imaging after hormonal diagnosis of adrenal disorders. The spectrum encompasses various pathologies, such as cortical adenomas, macronodular adrenal disease, pheochromocytomas, myelolipomas, infiltrative disorders, and primary and secondary malignancies. Notably, not all masses in both adrenal glands necessarily share the same etiology, often exhibiting diverse causes. Recently, the European Society of Endocrinology and the European Network for the Study of Adrenal Tumors updated guidelines, introduced a 4-option schema based on imaging, aiding in targeted hormonal testing and management. This "Approach to the Patient" review delves into the latest advancements in imaging, biochemical, and genetic approaches for the diagnostic and management nuances of bilateral adrenal masses. It provides insights and a contemporary framework for navigating the complexities associated with this clinical entity.

Head-to-head comparison between [68Ga]Ga-DOTA-NOC and [18F]DOPA PET/CT in a diverse cohort of patients with pheochromocytomas and paragangliomas.

PURPOSE: To compare the detection ability of 68Ga-labelled DOTA-l-Nal3-octreotide ([68Ga]Ga-DOTA-NOC) and 6-[18F]fluoro-L-3,4-dihydroxyphenylalanine ([18F]DOPA) in patients with phaeochromocytomas and paragangliomas (PPGLs) of different origins and gene mutations, such as germline succinate dehydrogenase complex genes (SDHx). METHODS: Eighty-five patients with histopathologically confirmed PPGLs who underwent both [68Ga]Ga-DOTA-NOC and [18F]DOPA PET/CT from March 2017 to June 2023 were enrolled in this retrospective study. For comparative analyses, PPGLs were classified as phaeochromocytoma (PCC), sympathetic paraganglioma (sPGL), and head/neck paraganglioma (HNPGL). Detection rates were analyzed on per-patient and per-lesion bases and compared using the Chi-square/Fischer's exact test. RESULTS: Among 85 patients with PPGLs (48 males; 43 years ± 17 [SD]), the patient-based detection rates of [68Ga]Ga-DOTA-NOC and [18F]DOPA PET/CT were 87.1% (74/85) and 89.4% (76/85), respectively (p = 0.634), and the lesion-based detection rates were 80.8% (479/593) and 71.2% (422/593), respectively (p < 0.001). Only one patient with a recurrent PCC presented double-negative imaging, while 66 patients exhibited double-positive imaging. The remaining patients were either [68Ga]Ga-DOTA-NOC-negative/[18F]DOPA-positive (n = 10) or [68Ga]Ga-DOTA-NOC-positive/[18F]DOPA-negative (n = 8). In subgroup analyses, [68Ga]Ga-DOTA-NOC PET/CT detected significantly more metastases of sPGL (91.1%, 236/259) and SDHx-related PPGL (89.6%, 86/96) than [18F]DOPA PET/CT (48.6%[126/259] and 50.0%[48/96], respectively; both p < 0.001). However, [18F]DOPA showed significantly higher detection rates of PCC in both primary/recurrent and metastatic lesions (94.3%[50/53] vs. 62.3%[33/53] and 87.9%[174/198] vs. 69.2%[137/198], respectively; both p < 0.001). Regarding metastases in different organs, [68Ga]Ga-DOTA-NOC PET/CT detected more lesions than [18F]DOPA PET/CT in bone (96.2%[176/183] vs. 66.1%[121/183]; p < 0.001) and lymph nodes (82.0%[73/89] vs. 53.9%[48/89]; p < 0.001) but less lesions in peritoneum (20%[4/20] vs. 100%[20/20]; p < 0.001). CONCLUSION: [68Ga]Ga-DOTA-NOC and [18F]DOPA are complementary in diagnosing PPGL under the appropriate clinical setting. [68Ga]Ga-DOTA-NOC should be considered as the ideal first-line tracer for detecting metastases of sPGL and SDHx-related tumours, whereas [18F]DOPA may be the optimal tracer for evaluating non-SDHx-related PCC, especially in detecting primary lesions and monitoring recurrence.

Validation of the modified CT criteria for identifying non-adenomas.

PURPOSE: Although adrenal computed tomography (CT) percentage washout is a potentially powerful imaging technique for differentiating adrenal adenomas from non-adenomas, its application to non-adenomas can be problematic. Recently, modified criteria for diagnosing pheochromocytomas using adrenal CT were developed based on data from 199 patients with surgically proven pheochromocytomas and adenomas. However, these criteria have not been thoroughly validated. The purpose of this study was to validate the performance of the modified criteria for diagnosing non-adenomas including pheochromocytomas. METHODS: The conventional and modified criteria were applied to 266 patients from two cohorts who had surgically proven lipid-poor adenomas (155/266, 58.3%) and non-adenomas (111/266, 41.7%) and underwent adrenal CT. Two radiologists calculated the attenuation on each dynamic phase and percentage washout of adrenal masses. The final assessments based on the conventional and modified criteria were categorized into adenomas or non-adenomas. The diagnostic performance of each criterion for diagnosing non-adenomas was evaluated using the area under the receiver operating characteristic curve (AUC). False negatives and positives were also compared. RESULTS: The AUC for the diagnosis of non-adenomas was 0.806 for conventional criteria and 0.858 for modified criteria (p = 0.047). The false-negative rate of conventional criteria for the diagnosis of non-adenomas was 29.7%. Use of modified criteria could have reduced the false-negative rate by to 7.2%. The false-positive rate increased from 9% to 21.3% when using the modified criteria. CONCLUSION: The utilization of modified criteria has the potential to identify additional non-adenomas that would otherwise be misdiagnosed as adenomas using conventional criteria alone.

Imaging of Pheochromocytomas and Paragangliomas.

Pheochromocytomas/paragangliomas are unique in their highly variable molecular landscape driven by genetic alterations, either germline or somatic. These mutations translate into different clusters with distinct tumor locations, biochemical/metabolomic features, tumor cell characteristics (eg, receptors, transporters), and disease course. Such tumor heterogeneity calls for different imaging strategies in order to provide proper diagnosis and follow-up. This also warrants selection of the most appropriate and locally available imaging modalities tailored to an individual patient based on consideration of many relevant factors including age, (anticipated) tumor location(s), size, and multifocality, underlying genotype, biochemical phenotype, chance of metastases, as well as the patient's personal preference and treatment goals. Anatomical imaging using computed tomography and magnetic resonance imaging and functional imaging using positron emission tomography and single photon emission computed tomography are currently a cornerstone in the evaluation of patients with pheochromocytomas/paragangliomas. In modern nuclear medicine practice, a multitude of radionuclides with relevance to diagnostic work-up and treatment planning (theranostics) is available, including radiolabeled metaiodobenzylguanidine, fluorodeoxyglucose, fluorodihydroxyphenylalanine, and somatostatin analogues. This review amalgamates up-to-date imaging guidelines, expert opinions, and recent discoveries. Based on the rich toolbox for anatomical and functional imaging that is currently available, we aim to define a customized approach in patients with (suspected) pheochromocytomas/paragangliomas from a practical clinical perspective. We provide imaging algorithms for different starting points for initial diagnostic work-up and course of the disease, including adrenal incidentaloma, established biochemical diagnosis, postsurgical follow-up, tumor screening in pathogenic variant carriers, staging and restaging of metastatic disease, theranostics, and response monitoring.

Ensemble Machine Learning Model Incorporating Radiomics and Body Composition for Predicting Intraoperative HDI in PPGL.

CONTEXT: Intraoperative hemodynamic instability (HDI) can lead to cardiovascular and cerebrovascular complications during surgery for pheochromocytoma/paraganglioma (PPGL). OBJECTIVES: We aimed to assess the risk of intraoperative HDI in patients with PPGL to improve surgical outcome. METHODS: A total of 199 consecutive patients with PPGL confirmed by surgical pathology were retrospectively included in this study. This cohort was separated into 2 groups according to intraoperative systolic blood pressure, the HDI group (n = 101) and the hemodynamic stability (HDS) group (n = 98). It was also divided into 2 subcohorts for predictive modeling: the training cohort (n = 140) and the validation cohort (n = 59). Prediction models were developed with both the ensemble machine learning method (EL model) and the multivariate logistic regression model using body composition parameters on computed tomography, tumor radiomics, and clinical data. The efficiency of the models was evaluated with discrimination, calibration, and decision curves. RESULTS: The EL model showed good discrimination between the HDI group and HDS group, with an area under the curve of (AUC) of 96.2% (95% CI, 93.5%-99.0%) in the training cohort, and an AUC of 93.7% (95% CI, 88.0%-99.4%) in the validation cohort. The AUC values from the EL model were significantly higher than the logistic regression model, which had an AUC of 74.4% (95% CI, 66.1%-82.6%) in the training cohort and an AUC of 74.2% (95% CI, 61.1%-87.3%) in the validation cohort. Favorable calibration performance and clinical applicability of the EL model were observed. CONCLUSION: The EL model combining preoperative computed tomography-based body composition, tumor radiomics, and clinical data could potentially help predict intraoperative HDI in patients with PPGL.

177 Lu-DOTATATE in the Treatment of Recurrent Pheochromocytoma With Multiple Metastases.

ABSTRACT: The treatment of metastatic pheochromocytoma is challenging. We report a case of a woman with recurrent pheochromocytoma with multiple metastases who achieved excellent response after 4 cycles of 177 Lu-DOTATATE therapy. She did not experience any observable adverse effects. Her disease was still stable 6 months after the fourth cycle of treatment.

A Prospective Comparative Study of 18 F-FDOPA PET/CT Versus 123 I-MIBG Scintigraphy With SPECT/CT for the Diagnosis of Pheochromocytoma and Paraganglioma.

PURPOSE: This study aimed to compare the diagnostic performances of 18 F-FDOPA PET/CT and 123 I-MIBG scintigraphy with SPECT/CT for detection of pheochromocytoma and paraganglioma (PPGL). PATIENTS AND METHODS: We conducted a prospective, single-institution comparative study. Patients suspected of having PPGL or those showing recurrence and/or distant metastasis of PPGL were enrolled. The primary objective was to affirm the noninferiority of 18 F-FDOPA PET/CT for diagnostic sensitivity. Both 123 I-MIBG scintigraphy with SPECT/CT (at 4 and 24 hours) and 18 F-FDOPA PET/CT (at 5 and 60 minutes after radiotracer administration) were performed. The final diagnosis was established either pathologically or via clinical follow-up. Nuclear physicians, unaware of the clinical data, undertook image analysis. RESULTS: Thirty-two patients were evaluated: 14 of 21 with an initial diagnosis and 9 of 11 with recurrence/metastasis had PPGLs in their final diagnoses. In patient-based analyses, 18 F-FDOPA PET/CT (95.7%) exhibited noninferior sensitivity compared with 123 I-MIBG SPECT/CT (91.3%), within the predetermined noninferiority margin of -12% by a 95% confidence interval lower limit of -10%. Both modalities showed no significant difference in specificity (88.9% vs 88.9%). In the region-based analysis for the recurrence/metastasis group, 18 F-FDOPA PET/CT demonstrated significantly higher sensitivity compared with 123 I-MIBG SPECT/CT (86.2% vs 65.5%, P = 0.031) and superior interobserver agreement (κ = 0.94 vs 0.85). The inclusion of an early phase in dual-phase 18 F-FDOPA PET/CT slightly improved diagnostic performance, albeit not to a statistically significant degree. CONCLUSIONS: 18 F-FDOPA PET/CT demonstrated noninferior sensitivity and comparable specificity to 123 I-MIBG SPECT/CT in the diagnosing PPGL. Notably, in the assessment of PPGL recurrence and metastasis, 18 F-FDOPA PET/CT outperformed 123 I-MIBG SPECT/CT in terms of both sensitivity and interobserver agreement.

Laparoscopic excision of normotensive variant pheochromocytoma in a case of triple vessel coronary artery disease: The anaesthesia challenge.

Pheochromocytomas are catecholamine-secreting tumours arising mostly from the adrenal medulla. With the advancement in surgical and anaesthetic techniques, the incidence of severe morbidity and mortality associated with surgery is low. However, concurrent coronary artery disease and pheochromocytoma continue to be a challenge due to the risk of adverse cardiovascular events. We describe the successful management of pheochromocytoma excision in a patient with coronary artery disease.

Machine learning for differentiation of lipid-poor adrenal adenoma and subclinical pheochromocytoma based on multiphase CT imaging radiomics.

BACKGROUND: There is a paucity of research investigating the application of machine learning techniques for distinguishing between lipid-poor adrenal adenoma (LPA) and subclinical pheochromocytoma (sPHEO) based on radiomic features extracted from non-contrast and dynamic contrast-enhanced computed tomography (CT) scans of the abdomen. METHODS: We conducted a retrospective analysis of multiphase spiral CT scans, including non-contrast, arterial, venous, and delayed phases, as well as thin- and thick-thickness images from 134 patients with surgically and pathologically confirmed. A total of 52 patients with LPA and 44 patients with sPHEO were randomly assigned to training/testing sets in a 7:3 ratio. Additionally, a validation set was comprised of 22 LPA cases and 16 sPHEO cases from two other hospitals. We used 3D Slicer and PyRadiomics to segment tumors and extract radiomic features, respectively. We then applied T-test and least absolute shrinkage and selection operator (LASSO) to select features. Six binary classifiers, including K-nearest neighbor (KNN), logistic regression (LR), decision tree (DT), random forest (RF), support vector machine (SVM), and multi-layer perceptron (MLP), were employed to differentiate LPA from sPHEO. Receiver operating characteristic (ROC) curves and area under the curve (AUC) values were compared using DeLong's method. RESULTS: All six classifiers showed good diagnostic performance for each phase and slice thickness, as well as for the entire CT data, with AUC values ranging from 0.706 to 1. Non-contrast CT densities of LPA were significantly lower than those of sPHEO (P < 0.001). However, using the optimal threshold for non-contrast CT density, sensitivity was only 0.743, specificity 0.744, and AUC 0.828. Delayed phase CT density yielded a sensitivity of 0.971, specificity of 0.641, and AUC of 0.814. In radiomics, AUC values for the testing set using non-contrast CT images were: KNN 0.919, LR 0.979, DT 0.835, RF 0.967, SVM 0.979, and MLP 0.981. In the validation set, AUC values were: KNN 0.891, LR 0.974, DT 0.891, RF 0.964, SVM 0.949, and MLP 0.979. CONCLUSIONS: The machine learning model based on CT radiomics can accurately differentiate LPA from sPHEO, even using non-contrast CT data alone, making contrast-enhanced CT unnecessary for diagnosing LPA and sPHEO.