Donald Urquhart, MB, ChB Portex Unit of Anaesthesia, Intensive Care, and Respiratory Medicine Institute of Child Health Great Ormond Street Hospital for Children London UK	Indra Narang, MD Division of Respiratory Medicine Hospital for Sick Children Toronto, Canada	Adam Jaffé, MD Associate Professor Department of Respiratory Medicine Sydney Children’s Hospital and School of Women's and Children's Health University of New South Wales Sydney, Australia

BACKGROUND

In CF, salt transport abnormalities at the airway surface leads to reduced volumes of airway surface liquid, thick mucous which traps bacteria, and recurrent respiratory infections. The resultant infection-inflammation cycle causes lung damage, and the subsequent lung function deterioration ultimately results in premature death. This process may be exacerbated by periods of low oxygen levels (hypoxia). Hypoxia may occur at times when the lungs are placed under stress in CF, namely sleep, exercise, air travel, and with CF chest exacerbations. This review attempts to summarise the impact of low oxygen levels in CF, in the current context of a lack of agreed guidelines on defining hypoxia in CF or on implementing oxygen therapy.

POTENTIAL EFFECTS OF LOW OXYGEN LEVELS IN CF

Low arterial oxygen blood content (saturations) occurring on a nightly basis during sleep, or to a lesser extent repeated periods of hypoxia occurring on exercise may be bad for the child with CF, stressing the heart and circulation, affecting quality of life, as well as having theoretical effects on lung inflammation.

Low oxygen levels are associated with higher lung blood pressure. Significant associations are reported between pulmonary artery blood pressures and average oxygen levels in the bloodstream (SaO2) during sleep and exercise in adults with CF¹.

Quality of life in CF may be influenced by low blood oxygen levels (SaO2). The only study of oxygen therapy undertaken in CF² reported that, over a 12-month period, school and work attendance was maintained in 83% of the oxygen-treated group, compared with 20% in those randomised to air. There was a statistically significant difference between the groups (p<0.01). Additionally, both sleep quality, and sleep duration correlate with sleep SaO2³.

There is also laboratory evidence to suggest that low oxygen levels may contribute to the decline in lung function by switching on inflammation, as well as encouraging growth of Pseudomonas aeruginosa (PA), the main bacteria associated with CF lung disease. Hypoxia is thought to activate several important blood proteins known as cytokines, which promote inflammation. Cytokines are part of the body’s response to infection, sending signals to recruit cells to the site of infection. One of these cytokines, known as interleukin-8, attracts a type of white blood cell, the neutrophil, which is important in the CF airway where ongoing neutrophilic inflammation is seen. The neutrophil releases enzymes including elastase, which destroy lung tissue.

PA also activates these same cytokines. Low oxygen levels may exert a role on this process, as under hypoxic conditions, PA biofilms prove virtually impenetrable to antibiotics, leading to an increased and prolonged immune response.

If low oxygen levels might be bad, how do we detect them?

ASSESSING HYPOXIA IN THE CHILD WITH CF

Pulse oximetry is the main assessment tool in children, where the oxygen concentrations are measured from the skin covering the finger. Traditional methods of assessing hypoxia in children with CF are to measure resting SaO2 whilst awake in clinic, along with annual exercise testing, during which SaO2 are measured.

Exercise Hypoxia

Exercise testing protocols depend upon the facilities and levels of expertise available in each CF centre. The gold-standard cardiopulmonary exercise test involves analysing the inhaled and exhaled gases with each breath to the point of maximal exertion. The volume of oxygen breathed in during exercise, as well as the volume of exhaled carbon dioxide, are monitored with increasing exercise workload, allowing maximal oxygen uptake (VO2max) to be measured. Blood oxygen levels (SaO2) are monitored throughout the test. More usual however, is that the six-minute walk test (distance walked in 6 minutes), or the three-minute step test (stepping up and down on a single step for 3 minutes) will be performed. During these tests, heart rate and SaO2 are recorded, along with patient scores of breathlessness. The concern is that the latter tests are sub-maximal, except in those with severe CF lung disease. As such, potentially clinically significant desaturations may be missed.

Nocturnal Hypoxia

During sleep there is a change in breathing pattern and blood and air distribution in the lung, which may be associated with a reduction in SaO2 in some circumstances. Sleep studies may be done in hospital, or undertaken at home, using a portable oximeter to measure pulse and SaO2. Our own experience is with the latter method, whereby wristwatch-like oximeters are sent via the postal service with instructions for use. Patients record sleep data on consecutive nights, then return the oximeter to the sleep laboratory, where data is downloaded. Limitations of home oximetry are the inability to distinguish obstructive sleep apnoea from underventilation in the context of CF lung disease, and there is a need to offset these limitations against ease of use and the additional cost of in-patient studies. Overnight oximetry appears a useful screening test for sleep hypoxia.

Fitness to Fly

Flying increases the risk of hypoxia in CF. Barometric pressure and partial pressure of oxygen fall with altitude - airline passengers breathe air with an oxygen concentration of 15% instead of the usual sea-level 21%. Methods used to predict in-flight hypoxia have been to perform a pre-flight hypoxic challenge (exposing the patient to low oxygen concentrations for a given time-period), or to predict desaturation on the basis of baseline blood gas tests or lung function. Evidence is conflicting as to which method most usefully predicts in-flight hypoxia in children, and the British Thoracic Society guidelines suggest that children with CF undergo pre-flight assessment which may include “hypoxic challenge testing in addition to spirometric tests”⁴.

CF chest exacerbations

Children with CF face challenges to their lung reserve at times of CF chest exacerbations, when ventilation to blood supply mismatching may be exaggerated, and hypoxia may ensue. Hospital admission provides an opportunity for SaO2 monitoring.

How do we interpret the meaning of tests used to assess oxygen levels in CF?

DEFINING HYPOXIA IN CF

There is a lack of objective evidence for clear definition of hypoxia in CF, and some of the issues pertaining to this across given clinical scenarios are outlined below.

Sleep Hypoxia in CF

An e-mail survey of UK paediatric CF centres (K Southern, University of Liverpool – Personal Communication), concluded that less than a quarter of centres have a definition for sleep hypoxia in CF. Various definitions are used and published methods of quantifying sleep hypoxia in CF include measuring the minimum blood oxygen levels during sleep, the average sleep oxygen levels, percentage of time spent with oxygen levels below 90%, and the lowest hourly average oxygen level recorded during sleep. A lack of definition for nocturnal hypoxia hampers the description of the prevalence of CF hypoxia in clinical practice, and may have led to both under-recognition and under-treatment of this potentially important clinical entity.

Exercise Hypoxia in CF

Exercise hypoxia in CF is defined as a fall in blood oxygen levels (SaO2) during exercise of 4% or more from the SaO2 at the beginning of exercise. This is a definition also used in healthy subjects. Clearly, if this represents a fall in SaO2 from 95% to 88%, one could deem this to represent a likely significant physiological drop in arterial blood oxygen concentration, though the definition holds less well in a very fit child whose SaO2 fall from 99% to 95% at the end of exercise. Further work is needed to establish clinically significant parameters.

Fitness to Fly

In-flight hypoxia in children with CF is defined as a fall in blood oxygen levels to below 90% at any time during the flight, necessitating the need for supplemental oxygen on long distance flights.⁴

Clearly there are various methods for investigating hypoxia in children with CF, which probably all assess slightly different aspects. At present it is unknown which is the more relevant assessment with the greatest clinical impact. Given that most of us spend 8 hours or more asleep each day, then the most prolonged period of low oxygen levels is likely to occur at night.

Is there any way of predicting likelihood of sleep hypoxia from routine clinical measures?

Lung Function (Forced Expiratory Volume in 1 second, FEV1)

Two studies suggest that %predicted FEV1 below 65% might predict sleep hypoxia in CF^7,8. This cut-off appears highly sensitive (most CF subjects with sleep hypoxia have FEV1<65%predicted), but poorly specific (a number of CF subjects with FEV1<65%predicted do not have sleep hypoxia). It may be useful in clinical practice however, to perform a sleep study in CF patients with a FEV1 below 65% predicted.

Exercise SaO2

It is previously reported that hypoxia in CF occurs more frequently during sleep than exercise⁸, suggesting that sleep studies may be indicated for all CF patients with exercise hypoxia. The mechanisms of hypoxia during exercise are likely to differ from those during sleep, borne out by poor correlations between the fall in SaO2 during exercise and sleep SaO2 measurements⁶.

How common is sleep hypoxia in the child with CF?

PREVALENCE OF SLEEP HYPOXIA IN CF

At Great Ormond Street Hospital, we studied 41 children with CF aged 8 to 16 years (Median 12.7). All individuals underwent measures of resting SaO2, followed by home oximetry on consecutive nights. The definition of sleep hypoxia used was that of SaO2 <93% for >25% of sleep time, as per the above paediatric definition used in Liverpool. Six of our 41 subjects (15%) met the criteria for sleep hypoxia using this definition. Bearing in mind that only 1-2% children with CF in the UK receive oxygen therapy⁵, then the possibility of under-recognition and undertreatment of sleep hypoxia in CF appears possible.

Does oxygen work as a treatment for low oxygen levels in CF?

OXYGEN THERAPY IN CF

In the only long-term oxygen trial in CF², twenty-eight CF subjects were randomised to receive either air or oxygen therapy, and followed up for up to 3 years. Although no differences in mortality or hospitalisation were found, 83% of those in oxygen maintained school/work attendance at 12 months compared with only 20% of the air group (p<0.01). Of note in the study design is that night-time oxygen therapy flows were decided by the flow needed to normalise daytime blood gas tests. Since daytime measures may be poorly sensitive predictors of sleep hypoxia, it is possible that patients in the oxygen arm of the study may have remained hypoxic and treatment effect may have been underestimated.

Caveats exist to starting oxygen therapy in CF. Firstly; children with CF already carry a heavy burden of care. The only trial of long-term oxygen therapy in CF highlights this as, of 146 subjects approached to take part, only 28 entered the study². Oxygen may be poorly tolerated, has household safety implications, and may be perceived as palliative rather than active therapy. Secondly, oxygen therapy may blunt respiratory drive, although no carbon dioxide retention was seen after one year of oxygen therapy in CF². Finally, oxygen may itself cause toxicity and activate lung inflammation.

CONCLUSION

The lack of a clear definition of hypoxia in CF prevents a uniform approach of assessment for CF patients during sleep and exercise. It appears that daytime SaO2 are of little value in assessing sleep hypoxia. Hypoxia may occur during sleep in those with a normal exercise test, often because the type of exercise test undertaken or patient volition results in a sub-maximal effort. It appears therefore, that a sleep study is needed to confidently rule in/rule out sleep hypoxia in CF, and lung function may be the best daytime predictor of nocturnal desaturation. Limiting sleep studies to those with reduced FEV1 (i.e. FEV1<65% predicted) may prove an effective screening tool in the detection of sleep hypoxia, but this remains to be confirmed.

The potential effects of hypoxia on the pulmonary circulation, sleep quality, quality of life and inflammation in CF, suggests that earlier use of oxygen therapy in those in whom low oxygen levels are demonstrated appears worthy of further exploration. Our prevalence data suggests that some children with CF and hypoxia may be under-treated. A uniform approach to defining hypoxia needs to be developed, along with guidelines for prescribing oxygen therapy in children with CF. Any change in practice must be evidence-based and the need for adequately-powered randomised, controlled trials of oxygen therapy in CF is apparent.

1. Fraser KL, Tullis DE, Sasson Z, Hyland RH, Thornley KS, Hanly PJ. Pulmonary hypertension and cardiac function in adult cystic fibrosis: role of hypoxemia. Chest 1999; 115: 1321-1328.

2. Zinman R, Corey M, Coates AL, Canny GJ, Connolly J, Levison H, Beaudry PH. Nocturnal home oxygen in the treatment of hypoxemic cystic fibrosis patients. J Pediatr 1989; 114: 368-377.

3. Milross MA, Piper AJ, Norman M, Dobbin CJ, Grunstein RR, Sullivan CE, Bye PT. Subjective sleep quality in cystic fibrosis. Sleep Med 2002; 3: 205-212.

4. British Thoracic Society Standards of Care Committee. Managing passengers with respiratory disease planning air travel: British Thoracic Society recommendations. Thorax 2002 57: 289-304.

5. Balfour-Lynn IM, Primhak RA, Shaw BN. Home oxygen for children – who, how and when? Thorax 2005; 60: 76-81.

6. Frangiolias DD, Wilcox PG. Predictability of oxygen desaturation during sleep in patients with cystic fibrosis. Chest 2001; 119: 434-441.

7. Versteegh FG, Bogaard JM, Raatgever JW, Stam H, Neijens HJ, Kerrebijn KF. Relationship between airway obstruction, desaturation during exercise and nocturnal hypoxaemia in cystic fibrosis. Eur Respir J 1990; 3: 68-73.

8. Coffey MJ, Fitzgerald MX, McNicholas WT. Comparison of oxygen desaturation during sleep and exercise in patients with cystic fibrosis. Chest 1991; 100: 659-662.

ACKNOWLEDGEMENT:
Don Urquhart received financial assistance from the Xtreme Everest Research Group at University College London to carry out research into hypoxia in Cystic Fibrosis. For more information, contact: d.urquhart(at)ich.ucl.ac.uk or UrquhD(at)gosh.nhs.uk.

[click here]
for a printable version of this article