BackgroundHigh-flow nasal cannula (HFNC) is an emerging therapy for respiratory failure but the extent of exhaled air dispersion during treatment is unknown. We examined exhaled air dispersion during HFNC therapy versus continuous positive airway pressure (CPAP) on a human patient simulator (HPS) in an isolation room with 16 air changes·h−1.MethodsThe HPS was programmed to represent different severity of lung injury. CPAP was delivered at 5–20 cmH2O via nasal pillows (Respironics Nuance Pro Gel or ResMed Swift FX) or an oronasal mask (ResMed Quattro Air). HFNC, humidified to 37°C, was delivered at 10–60 L·min−1 to the HPS. Exhaled airflow was marked with intrapulmonary smoke for visualisation and revealed by laser light-sheet. Normalised exhaled air concentration was estimated from the light scattered by the smoke particles. Significant exposure was defined when there was ≥20% normalised smoke concentration.ResultsIn the normal lung condition, mean±sd exhaled air dispersion, along the sagittal plane, increased from 186±34 to 264±27 mm and from 207±11 to 332±34 mm when CPAP was increased from 5 to 20 cmH2O via Respironics and ResMed nasal pillows, respectively. Leakage from the oronasal mask was negligible. Mean±sd exhaled air distances increased from 65±15 to 172±33 mm when HFNC was increased from 10 to 60 L·min−1. Air leakage to 620 mm occurred laterally when HFNC and the interface tube became loose.ConclusionExhaled air dispersion during HFNC and CPAP via different interfaces is limited provided there is good mask interface fitting.

Substantial exposure to exhaled air occurs within a 1-m region, from patients receiving NPPV via the ComfortFull 2 mask and the Image 3 mask, with more diffuse leakage from the latter, especially at higher IPAP.

Adverse effects of ambient concentrations of air pollutants on hospitalization rates for asthma are evident. Measures to improve air quality in HK are urgently needed.

A helmet with a good seal around the neck is needed to prevent nosocomial infection during NIV for patients with respiratory infections.

The inheritance, replication and perpetuation of the sperm centriole in the early human embryo are reported. Both normal monospermic and abnormal dispermic embryos (n = 127) were examined by transmission electron microscopy. Centrioles were traced from fertilization to the hatching blastocyst stage. The sperm proximal centriole is introduced into the oocyte at fertilization and remains attached to the expanding spermhead during sperm nuclear decondensation, as it forms the male pronucleus. A sperm aster is initially formed after the centriole duplicates at the pronuclear stage. At syngamy, centrioles occupy a pivotal position on opposite spindle poles, when the first mitotic figure is formed. Bipolar spindles were found in the majority of embryos, while tripolar spindles were seen in four dispermic embryos at syngamy. Two single centrioles were detected at two poles of two tripolar spindles, while two additional centrioles were located on the sides of a bipolar spindle of a dispermic embryo. Sperm tails were detected near spindle poles at syngamy and in later embryos. Typical centrioles showing the characteristic pin-wheel organization of nine triplets of microtubules were evident. During centriolar replication, the daughter centriole grows laterally from the parent and gradually acquires pericentriolar material (PCM). The two centrioles are surrounded by a halo of electron-dense PCM, which nucleates microtubules, thus making it a typical centrosome. The usual alignment of diplosomes at right angles to each other was maintained. Centrioles were detected at all stages of embryonic cleavage from the 1-cell through 8-cell stages, right up to the hatching blastocyst stage. They were closely associated with nuclei at interphase, when they were often replicating, and were prominently located at spindle poles during the first four cell cycles. In blastocysts, they were detected in trophoblast, embryoblast and endoderm cells respectively. It is evident that the sperm centrosome is the functional active centrosome in human, while the female is inactive but may contribute some centrosomal material to the zygote centrosome. It is very likely that the paternal centriole is the ancestor of the centrioles in fetal and adult somatic cells.