A tailored machine learning approach for mortality prediction in severe COVID-19 treated with glucocorticoids.

The impact of severe COVID-19 pneumonia on healthcare systems highlighted the need for accurate predictions to improve patient outcomes. Despite the established efficacy of glucocorticoids (GCs), variable patient responses are observed, and the existing clinical scores are limited in predicting non-responders. We propose a machine learning (ML) based approach to predict mortality in COVID-19 pneumonia treated with GCs.This is an ML-driven retrospective analysis involving 825 patients. We leveraged XGBoost to select the most appropriate features from the initial 52, including clinical and laboratory data. Six different ML techniques were compared. Shapley additive explanation (SHAP) values were used to describe the influence of each feature on classification. Internal validation was performed.Nine key predictors of death were identified: increasing C-reactive protein (CRP), decreasing arterial partial pressure of oxygen to fraction of inspired oxygen ratio (PaO2/FiO2), age, coronary artery disease, invasive mechanical ventilation, acute renal failure, chronic heart failure, PaO2/FiO2 earliest value, and body mass index. Random forest achieved the highest test area under the receiver operating characteristic curve at 0.938 (95% CI 0.903-0.969). SHAP values highlighted age and PaO2/FiO2 improvement as the most influential features; the latter showed a higher impact than CRP reduction over time.The proposed ML algorithm effectively predicts the risk of hospital death in COVID-19 pneumonia patients undergoing GCs. This approach can be adapted to datasets measuring similar clinical variables..

Don't forget the jumper's knee in the young sportsman: evaluation of patellar tendinopathy with a high frequency ultrasound probe.

Patellar tendinopathy, or Jumper's knee, is a painful knee condition caused by inflammation of the patella tendon. This condition is most frequently observed in subjects who play sports that require repetitive regular jumping. Jumper's knee is frequently misdiagnosed as a minor injury and many athletes, like our patient, keep on training and competing and either tend to ignore the injury or attempt to treat it themselves. However, jumper's knee is a serious condition that requires a correct and timely diagnosis, which often necessitates ultrasound investigation in order to start the most appropriate treatment.

Correlations between nailfold microvascular damage and skin involvement in systemic sclerosis patients.

OBJECTIVE: The aim of this study was to identify any correlations between microvascular damage, assessed by nailfold videocapillaroscopy and skin impairment, evaluated by three different methods, the modified Rodnan skin score (mRSS), skin high-frequency ultrasound (US) and the plicometer skin test (PST) in systemic sclerosis (SSc) patients. METHODS: Sixty-three SSc patients and 63 healthy subjects were enrolled. Nailfold videocapillaroscopy (NVC) was used to assess the nailfold capillaroscopy pattern ("Early", "Active" or "Late"), according to the Cutolo classification. All subjects were assessed by mRSS, US and PST to evaluate their dermal thickness (DT) in the seventeen skin areas of the body usually evaluated by mRSS (zygoma, fingers, hands, dorsum of hands, forearms, arms, chest, abdomen, thighs, legs, feet). Statistical evaluation was performed by nonparametric tests. RESULTS: All the three methods demonstrated progressively higher values of skin impairment in patients with "Early", "Active" or "Late" pattern of nailfold microangiopathy (for mRSS p < 0.01, US p < 0.02 and PST p < 0.02). A positive correlation was also observed in SSc patients between the three methods used to evaluate skin involvement (mRSS vs US, mRSS vs PST, PST vs US, p < 0.0001 respectively). CONCLUSIONS: This study demonstrates that there is a correlation between two of the most important aspects to classify and monitor the SSc patients, i.e. microvascular damage progression (evaluated by NVC) and skin damage (assessed by mRss, US and PST).

Long-term follow-up of nailfold videocapillaroscopic changes in dermatomyositis versus systemic sclerosis patients.

To identify nailfold videocapillaroscopy (NVC) changes in patients with dermatomyositis (DM) during a 3-year follow-up and to compare the NVC findings between DM and systemic sclerosis (SSc) patients at their first visit. Retrospective study of 24 DM and 24 SSc patients, matched for age and disease duration at first NVC. Capillaroscopic patterns/scores and clinical parameters had been yearly assessed. Nineteen out of 24 DM patients (79%) showed a NVC "scleroderma-like pattern." No statistically significant variation of all the capillaroscopic scores was observed during the 3-year follow-up. By comparing DM patients with or without anti-Jo-1 positivity, no statistically significant difference of the scores of the main capillary parameters was observed at baseline between the groups. Comparing at baseline DM with SSc patients, the giant capillary and microhemorrhage scores were significantly higher in SSc than those in DM patients (p = 0.04 and p = 0.05, respectively), while capillary density, ramification (abnormally shaped capillaries, expression of angiogenesis), and disorganization scores were higher in DM patients (p = 0.05, p = 0.002, p = 0.004, respectively). The absolute number of ramified capillaries was significantly higher in DM patients (p = 0.002), while the absolute capillary number was significantly higher in SSc patients (p = 0.05) at baseline. This pilot study demonstrates, for the first time, over long-term, that the capillaroscopic manifestations of DM persist in contrast to the progressive changes described in SSc patients, and the anti-Jo-1 positivity does not seem to modify the NVC pattern.

Correlations between blood perfusion and dermal thickness in different skin areas of systemic sclerosis patients.

OBJECTIVE: To identify possible correlations between skin blood perfusion (BP) and dermal thickness (DT) in different skin areas of systemic sclerosis (SSc) patients. METHODS: Sixty-two SSc patients, according to 2013 EULAR/ACR criteria, and 62 healthy subjects (CNT) were enrolled. Skin BP was analysed by laser speckle contrast analysis (LASCA) at the level of dorsum of the middle phalanx of the third fingers, dorsal aspect of the hands and zygoma. DT was assessed by both skin high frequency ultrasound (US) and modified Rodnan skin score (mRSS) in the same above reported areas. All patients were studied also by nailfold videocapillaroscopy (NVC) to assess the proper pattern of microvascular damage ("Early", "Active", or "Late"). RESULTS: At the level of finger dorsum a statistically significant negative correlation was observed in SSc patients between skin BP and both ultrasound-DT (p=0.0005 r=0.43) and mRSS (p=0.0007 r=0.42), but not at the level of hand dorsum and zygoma. No statistically significant correlation was present between skin BP and ultrasound-DT at any level in CNT. In detail, SSc patients, compared to CNT, showed a statistically significant lower BP only at level of fingers (median PU 72.6 vs 136.1 respectively, p<0.0001) and a statistically significant higher ultrasound-DT at the level of dorsum of 3th finger bilaterally (median mm 0.9 vs 0.7, p<0.0001), dorsum of hands (median mm 0.9 vs 0.7, p<0.0001) and zygoma (median mm 0.8 vs 0.7, p<0.0001). A significant positive correlation between ultrasound-DT and mRSS was observed in SSc patients at level of the three areas (dorsum of fingers p<0.0001 r=0.51; dorsum of hands p=0.03 r=0.27; zygoma p=0.0001 r=0.45). A progressive decrease of skin BP and increase of ultrasound-DT was found correlated with the progression of the severity of NVC patterns. CONCLUSIONS: This study demonstrates for the first time in SSc patients a significant inverse relationship between skin BP, measured by LASCA, and DT, evaluated by both US and mRSS, at the level of dorsum of the middle phalanx of the third fingers.

Assessment of treatment effects on digital ulcer and blood perfusion by laser speckle contrast analysis in a patient affected by systemic sclerosis.

Laser speckle contrast analysis (LASCA) is a good tool to evaluate the variation in peripheral blood perfusion during long-term follow-up and is able to safely monitor digital ulcer evolution in scleroderma patients. It evaluates blood perfusion in different areas within the skin lesions and surrounding them during standard treatment.

Subclinical dermal involvement is detectable by high frequency ultrasound even in patients with limited cutaneous systemic sclerosis.

BACKGROUND: The aim of the study was to detect by skin high-frequency ultrasound (US) possible subclinical skin involvement in patients affected by limited cutaneous systemic sclerosis (lcSSc), in those skin areas apparently not affected by the disease on the basis of a normal modified Rodnan skin score (mRSS). Differences in dermal thickness (DT) in comparison with healthy subjects were investigated. METHODS: Fifty patients with lcSSc (age 62 ± 13 years (mean ± SD), disease duration 5 ± 5 years) and 50 sex-matched and age-matched healthy subjects (age 62 ± 11 years) were enrolled. DT was evaluated by both mRSS and US at the usual 17 skin areas (zygoma, fingers, dorsum of the hands, forearms, upper arms, chest, abdomen, thighs, lower legs and feet). Non-parametric tests were used for the statistical analysis. RESULTS: Subclinical dermal involvement was detected by US even in the skin areas in patients with lcSSc, who had a normal local mRSS. In addition, statistically significantly higher mean DT was found in almost all skin areas, when compared to healthy subjects (p < 0.0001 for all areas). In particular, DT was significantly greater in patients with lcSSc than in healthy subjects in four out of six skin areas with a normal mRSS (score = 0) (upper arm, chest and abdomen), despite the clinical classification of lcSSc. CONCLUSIONS: This study strongly suggests that subclinical dermal involvement may be detectable by US even in skin areas with a normal mRSS in patients classified as having lcSSc. This should be taken into account during SSc subset classification in clinical studies/trials.

Microvascular damage evaluation in systemic sclerosis: the role of nailfold videocapillaroscopy and laser techniques.

Microvascular damage and a decrease in peripheral blood perfusion are typical features of systemic sclerosis (SSc) with serious clinical implications, not only for a very early diagnosis, but also for disease progression. Nailfold videocapillaroscopy is a validated and safe imaging technique able to detect peripheral capillary morphology, as well as to classify and to score any nailfold abnormalities into different microangiopathy patterns. Capillaroscopic analysis is now included in the ACR/EULAR classification criteria for SSc. The decrease in peripheral blood perfusion is usually associated with microvascular damage in SSc, which may be studied by different methods. Several of these make use of safe laser technologies. This paper focuses on these new clinical aspects to assess SSc microvascular impairment.

Capillaroscopy 2016: new perspectives in systemic sclerosis.

Systemic sclerosis (SSc) is an autoimmune disorder of unknown aetiology characterized by early impairment of the microvascular system. Nailfold microangiopathy and decreased peripheral blood perfusion are typical clinical aspects of SSc. The best method to evaluate vascular injury is nailfold videocapillaroscopy, which detects peripheral capillary morphology, and classifies and scores the abnormalities into different patterns of microangiopathy. Microangiopathy appears to be the best evaluable predictor of the disease development and has been observed to precede the other symptoms by many years. Peripheral blood perfusion is also impaired in SSc, and there are different methods to assess it: laser Doppler and laser speckle techniques, thermography and other emerging techniques.

Correlations between skin blood perfusion values and nailfold capillaroscopy scores in systemic sclerosis patients.

OBJECTIVES: To correlate blood perfusion (BP) values assessed by laser speckle contrast analysis (LASCA) in selected skin areas of hands and face with nailfold capillary damage scores in systemic sclerosis (SSc) patients. METHODS: Seventy SSc patients (mean SSc duration 6 ± 5 years) and 70 volunteer healthy subjects were enrolled after informed consent. LASCA was performed at different areas of the face (forehead, tip of nose, zygomas and perioral region) and at dorsal and volar regions of hands. Microvascular damage was assessed and scored by nailfold videocapillaroscopy (NVC) and the microangiopathy evolution score (MES) was calculated. RESULTS: SSc patients showed a significantly lower BP than healthy subjects at fingertips, periungual areas and palm of hands (p<0.0001), but not at the level of face and dorsum of hands. A gradual decrease of BP at fingertips, periungual and palm areas, was found in SSc patients with progressive severity of NVC patterns of microangiopathy ("early", "active", or "late") (p<0.01). A negative correlation was observed between MES and BP values, as well as between loss of capillaries and BP, at the same areas (p<0.001 and p<0.01, respectively). Patients with diffuse cutaneous SSc (dcSSc) showed lower BP than those with limited cutaneous SSc (p<0.04). CONCLUSIONS: LASCA detects a significant reduction of BP only in those areas usually affected by Raynaud's phenomenon (fingertips, periungual and palm areas), especially in dcSSc patients, and BP values significantly correlate with the nailfold capillaroscopy scores of microangiopathy.

Kikuchi-Fujimoto's disease associated with systemic lupus erythematous: difficult case report and literature review.

Kikuchi-Fujimoto's disease (KFD), or histiocytic necrotizing lymphadenitis, is a benign and self-limiting disease of unknown aetiology. KFD tends to affect a young population under 30 years of age and predominantly females. KFD is a rare pathology and its association with systemic lupus erythematosus (SLE) is not frequent. Herein, we present the case of a male Italian patient with SLE in association with KFD with 5 years of follow-up, where a differential diagnosis from infection or lymphoproliferative disease was problematic.

Correlations between nailfold microangiopathy severity, finger dermal thickness and fingertip blood perfusion in systemic sclerosis patients.

OBJECTIVE: The aim of this study was to identify possible correlations between nailfold microangiopathy severity, finger dermal thickness (DT) and fingertip blood perfusion (FBP) in systemic sclerosis (SSc) patients. METHODS: Fifty-seven SSc patients and 37 healthy subjects were enrolled. All patients were evaluated by nailfold videocapillaroscopy (NVC) to classify and score the severity of microangiopathy. Both modified Rodnan skin score (mRss) and skin high-frequency ultrasound were used to detect finger DT. Laser Doppler flowmetry (LDF) was employed to detect FBP. RESULTS: A positive correlation was found between nailfold microvascular damage severity and both ultrasound-DT (p=0.028) and mRss values (p<0.0001). In particular, both ultrasound-DT and mRss were found progressively higher in patients with 'Early', 'Active' or 'Late' NVC pattern of microangiopathy. A negative correlation was observed between nailfold microvascular damage severity and FBP (p<0.0001), showing the lowest FBP of the patients with more advanced NVC patterns. A negative correlation was observed between FBP, and both ultrasound-DT (p=0.007) and mRss values (p=0.0002). SSc patients showed a higher ultrasound-DT at the level of the fingers, as well as a lower FBP than healthy subjects (p<0.0001). CONCLUSIONS: This study demonstrates a relationship between nailfold microangiopathy severity, DT and FBP in SSc patients.

Prospective capillaroscopy-based study on transition from primary to secondary Raynaud's phenomenon: preliminary results.

The objective of this prospective study was to investigate the transition from primary (PRP) to secondary (SRP) Raynaud's phenomenon (RP), in a large cohort of patients affected by isolated RP. A total of 2065 patients with RP were investigated by clinical interview, laboratory examinations, and nailfold videocapillaroscopy (NVC). Patients with negative NVC at first visit were yearly followed to monitor either the appearance of specific morphological alterations at NVC, or clinical manifestations of an underlying disease. Capillary abnormalities at NVC were scored, as well as the qualitative patterns of microangiopathy (Early, Active and Late). NVC was found negative at first visit in 1500 subjects; among them, 412 patients were evaluable and they were followed for a mean time of 5±4 years (range 2-13 years). Sixty-eight patients (16%) achieved a diagnosis of SRP during follow-up, showing normal or not specific capillary alterations at NVC 4% of patients (the diagnosis was undifferentiated connective tissue diseases), Early scleroderma-pattern 57%, Active scleroderma-pattern 7%, Late scleroderma-pattern 12%, and scleroderma-like pattern 18% of patients. The time of transition from normal/not specific capillary alterations to Early scleroderma-pattern was 4.4±3.8 years. Enlarged capillaries (diameter between 20 and 50 microns) and mild reduction of capillary density were found the more frequent markers at first NVC visit in patients who progressed to a scleroderma pattern (P=0.01). This study demonstrates in a large cohort, that almost 16% of patients initially diagnosed as affected by RP with negative NVC may transit to SRP during a mean follow-up of 4.4 years. PRP patients showing major notspecific alterations of nailfold capillaries at first NVC should be strictly monitored at least once a year since at higher risk of transition to SRP.

Coexistence of osteopoikilosis with seronegative spondyloarthritis and Raynaud's phenomenon: first case report with evaluation of the nailfold capillary bed and literature review.

Osteopoikilosis (OPK) is a rare autosomal dominant bone disorder characterized by numerous hyperostotic areas that tend to localize in periarticular osseous regions. It is usually asymptomatic and is often diagnosed incidentally during X-rays. OPK may be an isolated finding or associated with other pathologies, e.g. skin manifestations, rheumatic and/or skeletal disorders. We report a literature review and, for the first time, the coexistence of OPK with seronegative spondyloarthritis and Raynaud's phenomenon in a 48-year old female. To the best of our knowledge, this is the first case of OPK studied by videocapillaroscopy, demonstrating the absence of specific microvascular abnormalities of nailfold capillaries.