July 14, 2010
This week we have been preparing the nursing staff for the introduction of the CPAP, (see Yiwens blogs) and testing devices that the neonatal ward nurses think are broken. We also went into town to purchase a multimeter so that we can repair the old bililights, and try to figure out what is wrong with the oxygen sensor. We asked Dr. Molyneux if she already had one, but she said no, but wouldn’t we please buy one for the hospital, so we did.
We have been using the flow meter on the oxygen sensor to measure the flow out of each flow splitter on the oxygen concentrators in the neonatal ward. Each oxygen concentrator has a total flow meter, with a variable knob so that the flow can be set, up to 5 LPM (liters per minute). The output of the concentrator is then split by a flow splitter. After taking one off, we saw that these are just 4 nozzles that screw into a machined metal block with channels in it for the directing the air to the nozzles and another place to screw into the concentrator. Three of the nozzles are rated for 0.5 LPM and the fourth is rated for 1 LPM. In theory, this is the maximum amount of flow that can pass through the nozzles. The 0.5 LPM is most suitable for very small babies, like most in the high risk neonatal ward where the oxygen is used, and after splitting, provides about 30-35% oxygen, assuming that the oxygen concentrator is working properly and producing a main supply of approx. 93% oxygen. This seems counterintuitive at first that the oxygen concentration after splitting would be different from the main oxygen concentration supply.
The big problem is, however, that using the flow meter on our sensor to test the flow coming out of the 0.5 and 1 LPM nozzles, we get no flow at all on any of the machines. On one I noticed that there seemed to be a hissing noise, as if air was escaping from the splitter. It wasn’t screwed in very well and if I held it up to the screw hole the flow jumped in the 0.5LPM nozzles to 0.25LPM, however the flow in the 1 LPM nozzle was unaffected. This result was not consistent over the other machines. On the others the splitters were screwed in fine, but still no flow of air was sensed. Another problem was that several of the machines seemed incapable of achieving a total flow rate of above 2 LPM. On one Devilibiss model that sits on the ground, as opposed to up on the wall, I opened the back of the oxygen concentrator to see a missing outer filter and a filthy inner filter. The general insides were pretty dusty too. After acquiring a new outer filter and a replacement inner filter and cleaning things up, the flow did increase, but only after fiddling with the flow splitter! So it seems like something about the splitter affects how good the total flow is and what amount of flow comes out of the nozzles.
After this we tested two of the concentrators by sticking the oxygen tubes into a glass of water to see if there are any bubbles. There were, so that means there is flow coming out of them, even though our flow meter doesn’t show any (?!) but in some cases it isn’t very strong. On one machine that is split into 4 1LPM nozzles, not all nozzles produce the same bubbling rate. And if you compare the bubble rate on this machine to the bubble rate of the 1LPM nozzle on the one with 3 0.5LPM nozzles and 1 1LMP nozzle, the rates are different, so something is still amuck, but atleast the air delivery isn’t a placebo as we were initially concerned.