Vardhman Mahavir Medical College & Safdarjung Hospital

Clinical guidelines recommend endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) as the initial diagnostic tool for lung cancer staging. However, despite the availability of mediastinal forceps biopsy and cryobiopsy, the optimal diagnostic approaches for other mediastinal conditions remain unclear.

We searched multiple databases and sources up to February 21, 2025, and employed single-arm, pairwise, and network meta-analytical approaches to comprehensively evaluate EBUS-based biopsies for mediastinal diseases in terms of efficacy and safety.

Fifteen prospective studies including 1,316 participants evaluated five EBUS-based mediastinal biopsy strategies (EBUS-TBNA, forceps biopsy, cryobiopsy, and the combinations of EBUS-TBNA with forceps biopsy or cryobiopsy) were involved. Concomitant EBUS-TBNA enhanced the efficacy of both forceps biopsy and cryobiopsy. EBUS-TBNA plus cryobiopsy yielded the best diagnostic outcome, showing significant benefits over EBUS-TBNA (OR 4.01, 95% CrI 3.05-5.33), forceps biopsy (OR 2.75, 95% CrI 1.94-3.92), cryobiopsy (OR 1.80, 95% CrI 1.33-2.45), and EBUS-TBNA plus forceps biopsy (OR 1.81, 95% CrI 1.20-2.72). A similarly favourable safety profile was observed in all EBUS-based biopsy methods.

EBUS-TBNA is the diagnostic cornerstone for mediastinal lesions, with EBUS-TBNA plus cryobiopsy being most effective. All EBUS-guided biopsies demonstrated a favourable safety profile.

Colistin is a last-resort antibiotic against multidrug-resistant (MDR) Klebsiella pneumoniae. Reliable susceptibility testing is critical but challenging due to colistin's complex pharmacodynamics and testing limitations. This study evaluates the performance of the Colistin Agar Test (CAT) against the gold standard Broth Microdilution (BMD) method.

Study was conducted after confirming stability of colistin in agar plate. A total of 150 non-duplicate clinical isolates of K. pneumoniae (103 colistin intermediate, 47 resistant) were tested using both BMD and CAT. Antimicrobial susceptibility testing (AST) of these isolates was performed using VITEK-2. Quality control was ensured using CLSI-recommended strains. Statistical parameters including Categorical Agreement (CA), Essential Agreement (EA), Major Error (ME), Very Major Error (VME), sensitivity, specificity, predictive values, and Cohen's Kappa were calculated using BMD as reference.

MICs ranged from 0.125-16 µg/mL (BMD) and 0.125-8 µg/mL (CAT). CA and EA for CAT were 96 % and 87.3 % respectively. ME and VME were 4.8 % and 2.1 %. Sensitivity, specificity, PPV, and NPV of CAT were 97.9 %, 95.1 %, 90.1 %, and 98.9 % respectively. Cohen's Kappa showed almost perfect agreement (0.91).

CAT demonstrated high agreement with BMD and showed promise as a screening tool for colistin susceptibility in K. pneumoniae. While BMD remains the gold standard, CAT offers a feasible alternative for routine use, particularly in resource-limited settings. However, resistant results by CAT should be confirmed with BMD to avoid false resistance reporting and potential therapeutic exclusion of colistin.

Objective Late-onset neonatal sepsis (LONS) remains a significant cause of neonatal morbidity and mortality, especially in developing countries. Our objective was to identify vitamin D deficiency in LONS, its association with maternal vitamin D deficiency, and its relationship with microbial profile and mortality in late-onset sepsis (LOS). Study design We conducted an analytical observational study with two matched groups (Group A and Group B) over 29 months, from November 2018 to March 2021, in the level III outborn neonatal unit at a tertiary-care centre in New Delhi, India. A total of 320 neonates were enrolled (160 septic, 160 non-septic). Outborn neonates aged 3 to 28 days were screened; those with LOS confirmed by sepsis screen or culture were enrolled as cases, while age-matched non-septic neonates were selected as controls. Epidemiological profiles, vitamin D status, and clinical outcomes, including sepsis severity and mortality, were compared between groups. Results Neonatal 25(OH) vitamin D levels in Group A (20.95±18.37 ng/mL) were significantly lower than those in Group B (25.09±16.21 ng/mL) (p < 0.001). Mothers of septic neonates had significantly lower 25(OH) vitamin D levels (25.0±16.21 ng/mL) than mothers of the non-septic group (29.86±14.13 ng/mL) (p = 0.001). Vitamin D deficiency was significantly more common in the sepsis group (40.6%) compared to the non-septic group (20.6%) (p < 0.001). Gram-negative and fungal sepsis (Acinetobacter, E. coli, Klebsiella, and Candida spp.) were associated with severe vitamin D deficiency. Mortality was 23.8% in the sepsis group, with significantly lower mean vitamin D levels among non-survivors (13.9±11.9 ng/mL vs. 23.3±19.6 ng/mL, p < 0.001). Receiver operating characteristic (ROC) analysis identified a 25(OH)D cutoff of 20.85 ng/mL (AUC 0.64) for predicting LOS. Conclusions Neonatal and maternal vitamin D deficiency is associated with an increased risk of LONS, particularly due to gram-negative and fungal pathogens, and correlates with higher mortality.