Will losing weight help?

The answer is yes! And the question, is invariably raised by patients at the sleep laboratory that are having a cpap implementation study after diagnosis of obstructive sleep apnoea.

It is a terrible cycle. We know that being overweight or obese increases the chances of having OSA and that losing weight can help but losing the weight is not always that easy when excessive levels of sleepiness are in play due to the OSA (hence the cycle).

It is always fantastic when a patient returns for a repeat diagnostic sleep study to see if their AHI has decreased after a significant weight loss. Some do it over time with diet and exercise and some undergo bariatric surgery.

A study on Polysomnography before and after weight loss in obese patients with severe osa demonstrated there was a decrease in the AHI with a decrease in BMI.

The subjects in this study all had laparoscopic adjustable gastric banding (LAGB) and prior to surgery all had severe OSA. Researchers also measured daytime sleepiness, the metabolic syndrome and quality of life (QOL) pre and post. Baseline BMI was 52.7kg/m(2).

The second PSG was conducted on average of 17.7+/-10 months after surgery and mean percentage of excess loss and weight loss were 50.1+/-15% and 44.9+/-22 kg (range 18-103 kg), respectively.

The AHI fell significantly from 61.6+/-34 to 13.4+/-13, improved sleep architecture was present with increased REM and stage 3 and 4 sleep. Daytime sleepiness, measured by Epworth Sleepiness Scale, also dropped from 13+/-7.0 to 3.8+/-3.0.

Weight can be a tricky subject to deal with but whenever faced with this question by a patient with a BMI in the overweight or obese category the answer is YES losing weight will help.

Jessica

nb: I have been asked to explain what AHI stands for. It is Apnoea Hypopnea Index that is used to assess the severity of OSA. The number of apnoeas and hypopneas per night are added up and divided by the hours of sleep. This gives a per hour score. Under 5 is normal, 5-15 is mild, 15-30 is moderate and over 30 is classified as severe.

Teeth Grinding and Sleep Apnoea

I thought this might be worth sharing – we commonly associate obstructive sleep apnoea with snoring and gasping for breath. This article talks about teeth grinding in relation to this condition and should be considered when assessing sleep disorders.

Teeth grinding on its own can lead to other health problems such as headache, jaw or ear pain, aching teeth and jaws, and strain on the joints and soft tissue of the jaw. Often people are not even aware they are grinding their teeth (some partners however are).

Recent research presented at CHEST 2009 – the 75^th annual international scientific meeting of the American College of Chest Physicians highlighted a link between night time teeth grinding (bruxism) and those that suffer from Obstructive Sleep Apnoea (OSA).

Researchers focussed on the potential influence of factors such as gender, ethnicity, gastroesophageal reflux and bruxism in relation to OSA. The small retrospective study looked at 150 men and 150 women with OSA, and was made up of 50 Caucasians, 50 African-Americans, and 50 Hispanics in each group.

Overall, 25.6% of patients were teeth grinders (almost one in four), and 35% complained of gastroesophageal symptoms and heartburn at night. Results showed a higher incidence of bruxism in men than women (43% VS 31%) and when comparing ethnic groups, Caucasians demonstrated the highest rate of bruxism (35%). The highest rate of gastroeosophageal reflux was seen in the African-American study group – 40% VS 34% in Caucasians and 31% in the Hispanic population groups.

Other factors cited which are believed to link OSA and bruxism are anxiety and caffeine. We already know that OSA (untreated or poorly controlled) can lead to altered mood including depression and anxiety. Sleep deprivation can in turn lead to a high intake of caffeine which can in turn be associated with a high risk of teeth grinding.

It is the end of an apnoeic event (usually after an arousal) that sleep bruxism is often seen. I t has been suggested that as men usually have more severe OSA, and therefore more associated arousals per night that this may explain a higher percentage of men in this group suffering from teeth grinding.

Finally, some studies are suggestive of treatments such as Continuous Positive Pressure Therapy (CPAP) for OSA in helping treat and potentially eliminate the secondary health problem of nocturnal bruxism.

We regularly receive referrals at Regional Respiratory Medicine from Oral Medicine Specialists and General Dentists who recognize how one or more symptoms might be associated with other conditions such as sleep apnoea.

Lisa

Home Diagnostic studies

Yes, that's me. What and Why you may ask. Well this was taken on Friday when I visited Val, the practice nurse at the Horsham Physicians room. I had travelled there to train Val on how to use the home sleep study diagnostic equipment, the somte psg.

To enhance the training I decided to set the equipment up on myself to demonstrate how easy it would be for a patient to use. To have a home diagnostic study the patient needs to have a half hour appointment. During this time the staff member will explain to them the procedure for setting up the equipment. The patient then takes home the equipment and a step by step instruction manual. It is advised they begin the set up process approximately 30-45 mins prior to wishing to retire for the night. The somte psg is preset to begin recording at the patients bed time and the controls are then locked so there is no fear of the device turning off while the patients is asleep.

The following morning they simply take off the equipment (this only takes a few minutes) and then return it.

Home vs Laboratory Studies
In-laboratory diagnostic sleep studies are considered the gold standard in the diagnosis of OSA. Home studies are appropriate if OSA is highly likely on clinical grounds, there are no significant co-morbidity's and the home environment and patient are suitable.

Advantages of home sleep studies are the patient does not need to spend the night in a sleep laboratory (this usually entails being admitted through a hospital), reduced travel, can be more cost effective for patient and reduced wait time.

It is important to remember that in-laboratory diagnostic sleep studies are considered the gold standard in the diagnosis of OSA. A study comparing the results of in laboratory sleep studies vs home diagnostic sleep studies found in-laboratory studies were superior from a technical point of view. Patient preference slightly favoured in-laboratory studies (55% vs 45%) and the researchers found no difference polysomnographic indexes.

Home sleep studies definitely have there place in the diagnosis of OSA and are a great option for many patients. Home sleep studies will now be available through the Horsham physicians rooms on a weekly basis. Appointments and enquiries can be made through our rooms.

Jessica

Occupational asthma; a hazard for the saw-doctor

I encountered a new case this morning of what I suspect is occupational asthma related to Cobalt exposure. I have never met someone known to be exposed to this heavy metal before, and never had cause to evaluate the literature regarding respiratory disease caused by Cobalt.

A brief literature search yields an array of clinical case reports, lab-rat studies and OH&S documents. Cobalt, it turns out, has been reported to be associated with interstitial lung disease and asthma. The latter is more commonly associated – and, I suspect, underdiagnosed.

Interestingly, the mechanism by which Cobalt causes asthma may be entirely allergic (ie the same mechanism by which grasses and pollens cause asthma). Cases have reported evidence of sensitivity on skin-prick testing to solutions of cobalt chloride, and nasal inhalation of the same mixture has been reported to be followed by increased markers of allergic inflammation in the blood – all of which might be consistent with IgE mediated allergy.

An OH&S report from British Columbia, Canada in 1999 makes interesting reading. This paper refers specifically to people in the occupation which my patient pursues; saw doctors (or saw filers) in timber mills. Saw tips are often made of tungsten carbide.

This paper quotes estimates of up to 5% of cobalt-exposed workers as developing sensitivity to cobalt resulting in asthma. Far fewer develop interstitial lung disease. The authors conducted a study in the filing rooms of eight sawmills, seven of which used tungsten carbide saw tips, and seven of which used stellite tips (some used both). Cobalt was present in both sorts of saw tip – at concentrations of up to 30% and up to 50% respectively. Employees working in the filing rooms at the mills wore, for the purpose of the study, filter cassettes and personal sampling pumps to catch airborn metals. Lung function tests were performed before and after work, and questionnaires were also filled out. The results were interesting, and pertinent to my patient.

1. Levels of airborne metals were low. Only 62 of 278 air samples had measurable cobalt levels. Measurable levels actually occurred more often if the saw blade was tungsten carbide than if it was stellite. Wet grinding (running water over the blade while grinding) made no difference, and in fact may have made things worse. (Because wet grinding was thought to fix the problem of airborne exposure, ventilation was usually worse in rooms where wet grinding was used, and cobalt got from the grinder coolant into the air. Cobalt levels in coolant from the wet grinders were higher for TC grinders then satellite grinders).

2. No current workers in the mills (118 in total) were diagnosed as having current lung disease related to cobalt. They were, however, twice as likely to report coughing up phlegm, and three times s likely to report coughing, phlegm and wheezing which was worse at work than industrial workers from different environments. (Sounds a bit like asthma to me). TC filers doing wet grinding had significantly higher rate of abnormally low lung function.

The study recommended that automated saw-grinding machines should be fitted in complete enclosures and vented to outside air. They also recommended that any hard-metal tool grinder with respiratory symptoms should be carefully evaluated by their family doctor, and should have respiratory function tests and a chest xray.

My patient has a long history of asthma, predating his current employment. In my opinion he should not, unfortunately, have been employed in an environment where he was exposed to a substance known to cause occupational asthma. I trust the recent advent of visiting OH&S specialists in his work environment signals that mill operators are waking up to their responsibility to their workers in this regard.

Andrew

Diffusing Capacity of the Lung for Carbon Monoxide - DLCO

In an effort to explain a little about the tests we perform in our lab I have previously discussed Spirometry, Bronchial Provocation and MIPS/MEPS and thought a blog about DLCO may be interesting.
The primary function of the lungs is to exchange gases. The two processes involved are the exchange of gas between the blood and the alveoli and the ventilation of the alveoli with air.
Carbon dioxide (CO2) is very soluble in water and moves rapidly across the alveolar membranes so that the pressure of CO2 in the alveoli is virtually identical to that in the blood. The rate-limiting step in expiring CO2 is the ventilation of the lung with atmospheric air (which contains virtually no CO2).
For oxygen, which is poorly soluble in water, the diffusion across the alveolar membrane is much slower and so oxygen uptake is more sensitive to alveolar disease than is CO2 removal. Although ventilation is also important for replenishing alveolar O2 supplies, the rate-limiting step in O2 exchange is usually exchange across the alveolar membrane.

The most popular method of measuring the diffusing capacity (DCO) is the single breath method whereby the patient takes a vital capacity breath of a mix of 0.3% Carbon monoxide, O2 21% in N2 and a tracer gas delivered via the automated valve from the spirometer followed by a 10 second breath hold then exhales. The first 750ml of dead space contaminated gas or washout is discarded and the next liter collected and analyzed in the alveolar sampling device. During exhalation the concentration of both Carbon monoxide (CO) and tracer gas fall with dead space washout. The tracer gas shows a plateau as alveolar gas is exhaled. CO will show a similar pattern but with a lower concentration because of diffusion during the breath hold.

The inert tracer gas we use in our lab, Methane (CH4) measures dilution of the inhaled CO and estimates the alveolar volume (VA). It distinguishes between the dilution of CO with air in the lungs and the uptake of CO into the blood. The change in CH4 concentration reflects the dilution of the gas remaining in the lungs (i.e. residual volume). This change is used to determine the initial CO concentration before diffusion and to determine the lung volume at which the breath hold occurred. The Krogh constant (KCO or DLCO/VA is an attempt to adjust the DLCO for differences in lung volume. It follows that when a KCO does not correct to normal there is likely to be an abnormality of the lung tissue.

Carbon monoxide (CO) is used to measure diffusing capacity (DCO). CO has similar physical properties to oxygen in terms of it solubility and ability to diffuse across the alveolar membrane into the plasma. The CO effectively follows the same path as oxygen, even combining and competing for the same binding sites on the Hb molecule as oxygen to form carboxyhaemoglobin (COHb) Reported in milliliters of CO/minute/mm of Hg, dry i.e. STPD. The affinity of CO for hemoglobin is approximately 210 times that of oxygen. The CO transferred across the alveolar wall is retained within the circulation and not exhaled.

The uptake of CO is determined by two things.
The diffusion properties of the blood-gas barrier
The rate of combination of CO with blood
The diffusion properties of the alveolar membrane depend on its thickness and area. Thus, the diffusing capacity is reduced by diseases in which the thickness is increased, including diffuse interstitial fibrosis, sarcoidosis and asbestosis and in a patient who has had a pneumonectomy where the surface area is reduced.

The rate of combination of CO with blood is reduced when the number of red cells in the capillaries is reduced as in anaemia and where the capillary blood volume is reduced in pulmonary embolism. At our lab we correct the DLCO for haemoglobin.

We know that the diseased lung tends to empty unevenly therefore the expired liter of gas that is analyzed for CO is probably not representative of the whole lung. For this reason the diffusing capacity is sometimes referred to as the transfer factor (TLCO) to emphasize that it is more a measure of the lung’s overall ability to transfer gas into the blood than a specific test of diffusion characteristics. In spite of this, the test has a definite place in the pulmonary function laboratory and is frequently useful in assessing the severity and type of lung disease.

Vanessa

PET scanning, in lung cancer and in Ballarat

Professor Alex Pitman from Lake Imaging , at St John of God Hospital in Ballarat, graced us with his presence in Hamilton last night, and provided us with an engaging and informative outline of PET scanning.

PET stands for Positron Emission Tomography. This mode of whole-body scanning is very useful in cancer medicine, having a role in the diagnosis and staging of a wide ranger of tumours. For our purposes, it is an indispensable part of the management of almost all lung cancer.

How does PET scanning work? Now I am no expert in this field. However…. a PET scan is a nuclear-medicine test. It involves the injection of a ‘tracer’ into the patient. In Australia, this tracer is universally F-18 fluoro deoxy glucose, or FDG. Essentially this is F-18 fluorine attached onto a glucose molecule. FDG tracks around the body in the same way as glucose; ie, it is taken up by organs to be metabolized. Tumours take up lots of glucose, but metabolize it pretty poorly – so FDG tracer hangs around in them. The F-18 fluorine bit of FDG is radioactive, and emits positrons, which collide with and annihilate electrons, releasing photons. After the contrast injection, and a bit of a rest, a patient goes throught the PET scanner. The photons are detected by the PET scanner, allowing mapping of the tumour (and the body in general). A PET scan is combined with a CT scan, which allows better visual localization of the tumour within the body.

When do we use it?

1. PET scanning is used in combination with CT scanning and anatomical biopsy for the evaluation of solitary pulmonary nodules. If a nodule is too difficult to biopsy without considerable risk to the patient, a PET scan is a useful way of ‘ruling in’ a malignant diagnosis. In Australia, if a nodule is ‘hot’ on PET scan then there is over 90% likelihood of it being a cancer. (In other countries, fungal infections and TB in particular also often turn up and are hot on PET scan. I have seen a patient with a fungal infection in Australia also undergo surgery because a lesion was ‘hot’ on PET scan, so we need to be a bit careful).

2. PET scanning is also used for staging non-small cell lung cancer prior to surgery. In a patient with lung cancer a PET scan is useful if combined with CT and mediastinal biopsy procedures to evaluate the possibility of malignant involvement of mediastinal lymph nodes prior to surgical treatment. Indeed, a recently published small study has suggested that PET combined with diagnostic CT scan, without mediastinal biopsy sampling, has a very high negative predictive value (NPV) (ie if PET/CT is clearly negative then there is not going to be cancer, NPV in this study 97%) for mediastinal spread, They suggest that PET/CT should be relied on to indicate if a patient with NSCLC is anatomically suitable for surgical resection without the patient needing an invasive mediastinoscopy or endobronchial ultrasound biopsy. However, a study published a couple of weeks earlier suggested that there is considerable risk of false negative and false positive results if PET is relied on alone to evaluate the likelihood of malignant involvement of the mediastinal nodes.

The PET scan operators would very much like us to use PET to re-stage cancer after therapy, and to monitor disease in lung cancer. Medicare will not reimburse PET in those situations (but will for the evaluation of a solitary nodule or for the staging of cancer prior to treatment). The clinical reality is, however, that the benefit of treating recurrent non-small cell lung cancer with radiotherapy and chemotherapy is not sufficient that it can clearly be regarded as a good thing to diagnose recurrent disease early and treat it. That is to say, even if a PET scan picks up recurrent disease, we don’t necessarily know what to do about it. There are some new chemotherapeutic agents that may change the playing field in that regard, but are yet to have their early promise confirmed. If and when they do, the indication for PET scan will become even more broad. For the time being I think that the use of restage lung cancer after treatment routinely would be expensive and indulgent.

Undoubtedly PET scanning has an important, and increasing, role in management of patients with lung cancer. It is terrific for us that there is now a PET scanner in Ballarat. The service they have provided us to date has been excellent. For the benefit of our patients, however, evaluation of the PET/CT in lung cancer still needs to occur in a multidisciplinary environment, with radiologist, nuclear medicine specialist, pathologist, oncologist, thoracic surgeon, respiratory physician, radiation oncologist all present and all able to discuss the patient’s situation. It is difficult to assemble such a team outside of the major metropolitan areas. The team at St Vincent’s Hospital / Peter Macallum in Melbourne have provided a great service for our Victorian patients to date, and continue to do so.

Andrew

Screening type II diabetes patients for OSA

Previously, Andrew blogged about a Japanese study that looked at the effectiveness of a screening tool for OSA. This screening tool compared patients BP, gender, BMI and presence or absence of snoring (these results were then scored) and their AHI (apnoea hyponea index) to determine if there was a correlation and if so a what point. A verified screening tool would be a signifcant break through and it is something we are interested in examining in our population group.

Recently, a Romainian study looked at an OSA screening tool for patients with Type II diabetes and a BMI greater than 30. There were 80 patients in this study. The researchers then assessed the patients Epworth Sleepiness Scale (ESS). The ESS is a questionnaire the patient answers about how likely they would doze off or fall asleep in a variety of situations. A score is derived from these answers. Clinically a score of 13 or above is deemed significant but in this project the researchers referred all patients with a score greater than 10 to have a overnight sleep study. This amounted to 20% of the population group and of these ALL were diagnosed with obstructive sleep apnoea. Of these 33.3% had moderate OSA (AHI = 15-30 events/h of sleep), and 58.3% had severe OSA (AHI >/=30 events/h of sleep). The researchers claimed 'prevalence of OSA in the sampled population was high'. This is true but it would have been interesting to know how many patients missed being diagnosed because they scored under 10 on the ESS.
Anecdotal evidence within our practice demonstrates there are many patients with very low levels of self reported daytime tiredness (e.g a low ESS) who are diagnosed with obstructive sleep apnoea.

Jessica

Compulsive behaviours and restless legs syndrome

Restless legs syndrome (RLS) is a common neurological disorder, affecting up to 15% of people in our community. In up to 2.5% of our community the condition is so severe that quality of life is adversely affected.

The cause of this condition is unknown. Central dopaminergic pathways (dopamine is a neurotransmitter, a chemical carrying messages between nerves) have been implicated in this disorder, and ‘dopaminergic’ medications are first-line drugs when it comes to treating RLS. Dopaminergic neurological pathways beyond my understanding are also important in mechanisms of behavioural reward and reinforcement – ie you perform an action, it feels good so you do it again. Dopaminergic medications interfere with these pathways, and in some people this leads to reduced ability to control impulses towards the performance of personally destructive behaviours.

In the last five years or so, two relatively – new dopaminergic medications have become available in Australia for treatment of restless legs syndrome. The first of these, ropinirole, is available on private script only. The most recent, pramipexole, is available on the PBS for patients who score more than 21/40 on the International Restless Legs Syndrome Rating Scale. With the availability of pramipexole, previously used in the USA for treatment of RLS (and also used in Parkinson’s disease) has come increasing interest in the question of whether these medications really do cause reduced impulse control, and correspondingly increased compulsive behaviour. Is this true? Is it a problem?

A colleague of mine in at the Melbourne Sleep Disorders Centre has done some excellent, recent, local epidemiological work in this field. Similar work from the Mayo Clinic was published in Sleep journal last month. The findings were similar to those my colleague presented at the Australasian Sleep Association conference in Melbourne in October. They suggest that the phenomenon of reduced impulse control in patients with RLS on dopaminergic medications is real, and is a problem.

The recently - published article surveyed 100 consecutive patients with RLS who had been treated with dopaminergic medications. (Not all of these patients had been treated with pramipexole or ropinorole. Other medications included levodopa, bromocriptine, pergolide, cabergoline, rotigotine patch and apomorphine). They also surveyed 275 patients with obstructive sleep apneoa, as a control group, who had no symptoms of RLS. A smaller group of patients with RLS on no dopaminergics was also included (52 patients).

A variety of compulsive behaviours were screened for, using a mix of commonly used and validated screening tools (for compulsive gambling) and home-made questionnaires. Information was sought about compulsive gambling, shopping, eating, hypersexuality and ‘punding’. I had no idea about this last verb, but it refers to repetitive, complex, stereotypical but purposeless actions – for example grooming, cleaning, hording, operating technical equipment with no goal. A self – completed questionnaire was followed by a telephone interview to try to heighten the rigour of the process (and make it more specific for real problems).

The end result was that 17% of patients in the RLS group treated with dopaminergic agents displayed problems with impulse control. Only 6% of the OSA control group displayed similar behaviour. This 11% difference was statistically significant. In particular, compulsive shopping (9% vs 0.7%), gambling (5% vs 0.4%) and punding (7% vs1%) were behaviours where the striking discrepancy in prevalence was also statistically significant.

Twelve of the seventeen patients displaying difficulty with impulse control were taking pramipexole at the time of symptom onset, and fifteen of them had taken that medication at some time. Five were taking ropinorole at time of symptom onset.

These behaviours usually began several months after onset of therapy (mean of 9.5 months) . Their frequency may have been underreported because of study design (perhaps they should have surveyed husbands/wives/partners as well).

The case reports provided in the Sleep journal study serve to underline what is at stake here, and emphasise why it is important to aggressively screen patients on these medications for these bevhaviours – behaviours to which they will seldom readily confess. Sometimes stories communicate better than figures:

Case Report: A 47 yr old woman reported concerning behaviours after she had been taking pramipexole at 0.50mg daily for 6 months. She lost an estimated $5000 on purchases from the shopping channel (‘ugly clothes and jewelry that I didn’t even need’) and set her alarm clock for the early morning hours ‘because I just couldn’t miss a sale’. She had food binges, eating an entire pizza or a dozen donuts at one sitting ‘even though I wasn’t hungry’. Sometimes she would stay up ‘all hours of the night’ cross stitching. Finally, she reported ‘being more risky’ by performing sexual acts in public. Although present for almost 2 years, these behaviours resolved completely in 1 to 2 months after pramipexole was discontinued.

Case Report: A 59 yr old man is dealing with ongoing litigation related to inappropriate sexual behaviours involving the Internet that prompted police to raid his home, much to the shock of his wife and grandchildren. He gained more than 200pounds with food binges, his wife constantly returned unneeded purchases to the store, asnd he spent 10 to 12 hours per day on the computer in chat rooms, playing games and viewing pornography. All of the behaviours started within a year of his taking ropinirole, 4.0mg daily, and resolved quickly when he was taken off the medication.

Pramipexole can be tremendously effective in treatment of restless legs syndrome. The potential for associated problems with impulse control must not, however, be taken lightly.

Acute rhinosinusitis effect on pulmonary function

A ‘Short Report’ in the Primary Care Respiratory Journal (2009); 18(3): 224-226, caught my attention. It addresses the effect of acute rhinosinusitis on pulmonary function in patients who have no chronic airways diseases or allergies.

Previously researches had noted that Chronic Obstructive Pulmonary Disease (COPD) patients have a high frequency of nasal symptoms and abnormalities in the paranasal sinuses on CT scanning, and that frequently patients with existing asthma later get a worsening of their lower airways disease with acute or chronic sinusitis problems. The influence of acute rhinosinusitis on lower airway function in patients without chronic airway diseases or allergy had not been investigated.

This study, carried out in a general practice in Denmark by two physicians, and reported by Associate Professor Jens Georg Hanson of the Aarhus University Hospital, Denmark, looked at 25 adult patients with a diagnosis of acute rhinosinusitis who presented with facial pain, anterior discharge and a CRP (C-reactive protein) value > 10 mg/l( normal < 8 mg/l). All patients with recurrent acute rhinosinusitis, asthma, COPD, allergy, and any condition or chronic disease that could modify the CRP levels were excluded.

Spirometry was used to measure pulmonary function, with the patients being tested at their initial visit, and again 8 weeks later when the rhinosinusitis had resolved. The results showed that the acute episode was associated with a temporary but significant reduction in F25-75 and in PEF, with no significant changes in FEV1 and FVC. There was a significant correlation between CRP and F25-75, but not with CRP and PEF. There was a significant correlation between changes in F25-75 and PEF, but no correlation between pain score and F25-75 or PEF.

The researchers suggest that a possible explanation of the changes might be small airway obstruction and air trapping, but as this finding was not expected, the researchers did not test if any slight obstruction in the small airways was reversible to inhaled bronchodilators. The author speculates that the significant correlation between the inflammation expressed as elevated values of CRP and the reduction in F25-75 points to the fact that the small airways are likely to be inflamed. It was suggest that further investigations be carried out to confirm and make clearer, the findings.


I wonder what pulmonary function changes are present in patients with recurrent acute rhinosinusitis, and their lung function prognosis.

Heather