When is O2 Therapy Needed?
Currently, the “standard” indication for supplemental oxygen therapy is a PaO2 < 60 mm Hg or a SaO2 of < 90% [Resp Care 47: 717, 2002], however arterial oxygen may be as low as 40 mm Hg in supine patients without producing hyperlactemia – this has been shown in COPD [NEJM 274: 878, 1966] and ARDS patients [Crit Care Med 12: 75, 1984]. Oxygen has also been shown to vasoconstrict non-pulmonary arterial beds and reduce cardiac output [JCI 41: 126, 1962].
Hemodynamic and Physiologic Responses to O2
The hemodynamic response to controlled oxygen administration was investigated in 35 patients who presented with acute decompensation of COPD (measured under inspiratory fractions in oxygen of 0.21 and 0.28). Response to oxygen therapy was characterized by a decrease in cardiac output and an increase in oxygen delivery because of a sharp increase in arterial oxygen content. DO2/VO2 increased in each patient (p < 0.001). Further analysis individualized 2 types of hemodynamic response to oxygen therapy- in 15 patients, oxygen therapy resulted in a significant increase in DO2 without change in cardiac output, whereas in the other 20 patients, oxygen therapy induced a significant decrease in cardiac output (p < 0.001) without change in DO2. Patients in whom cardiac output was unaffected by oxygen administration presented with a lower mixed venous oxygen tension (less than 35 mmHg) and a lower coefficient of oxygen delivery (less than 3.5). Overall, increasing arterial PO2 from 61 to 83 mm Hg was shown to have no effect on oxygen delivery (insignificant decrease from 12.8 to 12.1 mL/kg/min [p > 0.05, Am Rev Respir Dis 124: 26, 1981]). Lastly, oxygen does not necessarily protect against myocardial ischemia [Kavanagh BP et al. Anesth Analg 76: 950, 1993]. Thus, the common indications for oxygen may need to be revisited.
Delivery of O2
Nasal cannulas can deliver up to 45% O2 at flow rates of 6L/min in patients with normal ventilatory pattern (RR from 10 -> 40/min reduces FIO2 by 48%).
Oxygen face masks can deliver an FIO2 up to 0.60 at 10 L/min, but these are dependent on the patient’s respiratory rate. Note that at least 5 L/min is required to clear exhaled gas from the mask.
Partial Rebreather Masks
Partial rebreather mask-reservoir bags can deliver an FIO2 up to 0.75 at 7 L/min, and non-rebreathers can deliver an FIO2 of 1.0 at 10 L/min. The problem with bag masks are that 1) NG tubes cannot be used concurrently because they have to fit tightly and 2) aerosolized bronchodilator therapy cannot be used with them.
High-flow masks can adjust the FIO2 by changing the size of entrainment ports on the side. These masks can deliver FIO2 up to 0.50 but their advantage is they aren’t dependent on ventilation rate (ideal in COPD where an increase in FIO2 may lead to decreased respiratory effort).
In the normal lung, oxygen has not been show to induce toxicity if kept at an FiO2 of 0.60 or less [Crit Care Clin 6: 749, 1990; Intensive Care Med 18: 139, 1992], although in the injured lung may only tolerate an FiO2 of 50% [Crit Care Med 15: 598, 1987]. Studies have shown that 6-12 hours of 100% oxygen in humans can cause tracheobronchitis and decreased VC [Intensive Care Med 3: 134, 1988]. The consensus is that an FIO2 of 0.60 for longer than 48 hours is toxic. Marino believes that, as most ICU patients are anti-oxidant deficient, any amount of oxygen should be considered toxic. There are three approaches to limiting oxygen injury:
Approaches to Limiting O2 Injury
1) Only use O2 when indicated
2) Minimize FIO2 by using PEEP if necessary
3) Consider antioxidant support – while not all data support this, there are several studies which suggest benefit in critically ill patients, especially selenium (Marino recommends 70/55 ug/day in men/women).
Appropriate use of respiratory pharmacologics mandates monitoring of certain physiologic variables. Peak expiratory flow (best out of 3) is useful, although only when the patient is putting forth a maximal effort (often they do not). QPeak values are particularly useful in monitoring bronchodilators (20 min. before and after administration) and/or following asthmatics. Nml. for 20 yr. men 690L/min for women 460 and subtract 15-20% for elderly. On patients who are ventilated, peak inspiratory pressure is also useful for assessing bronchodilators. Also, auto-PEEP should respond to bronchodilators in ventilated patients.
There doesn’t seem to be a difference between MDI and nebulizers if they are used properly (most particles are lost in the equipment or pharynx), but always use a spacer for MDI. MDI can be used in the vent and are cheaper.
B-agonists can be used for acute asthma, if they fail after 3 x 20 min add steroids. Side effects of B-agonists include tachy, tremor, decreased K+, temporary decreased PO2 at high doses.
Ipratropium MDI can increase bronchodilation when combined with albuterol and should be used in asthma exacerbations.
Steroids (after 3 failed B-agonist doses of 4-8 puffs each) can take 6 – 8 hours to work. Recommendations are either 80 mg methylprednisolone IV or 60 mg prednisone PO, there is no difference [JAMA 260: 527, 1988]. In patients w/ paralytics and on the vent, steroids can sometimes cause a global myopathy/rhabdomyolysis in addition to traditional proximal myopathy.
Lastly, Mg given 2 g IV over 20 minutes in patients with asthma exacerbation [Lancet 361: 2114, 2003] improves dilation over albuterol alone (but was not compared to albuterol + ipratropium).
Mucomyst (NAC) can be given as an aerosol (avoid if possible, irritating and can promote coughing and/or bronchospasm) or injected directly into the airways. Tastes awful and can promote n/v, direct instillation can help.