12
`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Thorax 1990;45:112-117
`Corticosteroid trials in non-asthmatic chronic
`airflow obstruction: a comparison of oral
`prednisolone and inhaled beclomethasone
`dipropionate
`
`D C Weir, R I Gove, A S Robertson, P Sherwood Burge
`
`Abstract
`One hundred and twenty seven adults
`considered on clinical grounds to have
`non-asthmatic chronic airflow obstruc-
`tion entered a randomised, double blind,
`placebo
`controlled,
`crossover
`trial
`comparing the physiological response to
`inhaled beclomethasone dipropionate 500
`jg thrice daily with oral prednisolone 40
`mg a day, both given for two weeks. One
`hundred and seven patients completed
`the study. Response was assessed as
`change in FEV, and FVC measured on the
`last treatment day, and as change in mean
`peak expiratory flow (PEF) over the final
`seven days of treatment from home PEF
`recordings performed five times daily. A
`full response to treatment was defined as
`an increase in FEV or FVC, or an increase
`in mean daily PEF over the final seven
`days of treatment, of at least 20% from
`baseline values. An improvement in one
`measurement ofat least 15%, or of 10% in
`any two measurements, was defined as a
`partial treatment response. Response to
`placebo showed a significant order effect,
`suggesting a carry over effect of active
`treatment of at least three weeks. Res-
`ponse to active treatment was therefore
`related to initial baseline values, and
`compared with placebo by considering
`responses in the first treatment phase
`only. A full response to oral prednisolone
`(16138) was significantly more common
`than to placebo (3/35). The number of full
`responses to inhaled beclomethasone (8/
`34) did not differ significantly from the
`number responding to oral prednisolone
`or placebo in the first treatment phase,
`though full and partial responses to
`inhaled beclomethasone (12/34) were
`significantly more common than those to
`placebo (4/35). When all three treatment
`phases were considered 44/107 patients
`showed a full response to one or both
`forms of corticosteroid treatment, a
`response to prednisolone (39) occurring
`more frequently than to inhaled beclo-
`methasone (26). Only 21 of the 44 respon-
`ders showed a response to both forms of
`treatment. Inhaled beclomethasone dip-
`ropionate 500 Mg thrice daily was inferior
`to oral prednisolone 40 mg per day, but
`better than placebo, in producing im-
`provement in physiological measure-
`
`ments in patients thought to have non-
`asthmatic chronic airflow obstruction. It
`was, however, an effective alternative in
`over half of those showing a response to
`prednisolone.
`
`Oral corticosteroids improve symptoms and
`lung function in some patients with severe
`chronic airflow obstruction related to cigarette
`smoking.' Published trials have in general used
`doses of oral medication which in the long term
`would have serious systemic side effects.2`4 The
`efficacy of inhaled corticosteroids in asthma is
`well established but only two previous studies
`have looked at the effect of this form of
`treatment in patients with chronic airflow
`obstruction, both with small numbers of
`patients.56 The study
`of Harding
`and
`Freedman identified a response to 400 ,g/day
`of inhaled betamethasone valerate in only three
`of six patients who showed a response to oral
`prednisolone. All responders to the inhaled
`drug were inpatients and the authors suggest
`that this may have improved compliance and
`delivery of the drug to the airways. In the
`recent study of Wardman et al with 22 out-
`patients, however, all with good inhaler tech-
`nique, the five responders to oral prednisolone
`improved to the same degree after two weeks'
`inhaled
`beclomethasone
`with
`treatment
`dipropionate 1500 yg per day. The different
`results may reflect different doses and delivery
`of the drug to the airways, but the role of
`inhaled corticosteroids in patients with severe
`chronic airflow obstruction is still unclear.
`The aim of this study was to compare the
`response to oral prednisolone 40 mg/day with
`that to inhaled beclomethasone dipropionate
`500 Mg thrice daily in outpatients with non-
`asthmatic chronic airflow obstruction.
`
`Methods
`SUBJECTS
`Outpatients with adult onset chronic airflow
`obstruction of at least five years' duration and
`an FEV, below 70%/ predicted were recruited
`to the trial. Patients were excluded if they had a
`respiratory
`asthma,
`diagnosis
`of
`clinical
`variability
`childhood,
`in
`in
`symptoms
`association
`with
`in
`symptoms
`except
`infections, acute attacks of wheezing and
`
`Department of
`Thoracic Medicine,
`East Birmingham
`Hospital, Birmingham
`D C Weir
`R I Gove
`A S Robertson
`P S Burge
`Address for reprint requests:
`Dr P Sherwood Burge,
`Department of Thoracic
`Medicine, East Birmingham
`Hospital, Birmingham
`B9 5ST.
`Accepted 30 October 1989
`
`MYLAN EX 1010, Page 1
`
`

`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Corticosteroid trials in non-asthmatic chronic airflow obstruction
`
`113
`
`breathlessness, or deterioration after exposure
`to specific allergens. A lack of a "fixed"
`element
`in
`the
`airflow
`obstruction
`after
`inhalation of bronchodilator also favoured
`asthma as the diagnosis. The presence of some
`reversibility of airflow obstruction in response
`to inhaled bronchodilators was deliberately not
`chosen as an exclusion criterion so that its effect
`on steroid response could be assessed. No
`patient had received oral or inhaled cortico-
`steroids in the preceding six months. All
`patients gave informed consent.
`
`MEASUREMENTS
`Spirometric indices were determined on two
`occasions during the baseline phase before any
`treatment was given, and on the final day of
`each treatment period. Patients were asked to
`refrain from inhaled bronchodilators for six
`hours before the measurements, and visits were
`performed at the same time of day. FEV, and
`FVC were measured on a dry
`bellows
`spirometer (Vitalograph), the mean of three
`technically satisfactory attempts within 100% or
`100 ml (whichever was the smaller) being used
`for subsequent analysis. Baseline FEVy and
`highest mean
`FVC were
`taken
`the
`as
`measurements recorded on the two baseline
`visits before any treatment.
`Lung volumes and single breath carbon
`monoxide gas transfer (TLCO) were determined
`once during the baseline period and at the end
`of each treatment phase. A closed circuit
`helium dilution technique was used to measure
`lung volume subdivisions, rebreathing being
`continued until the concentration of helium
`was stable or for a maximum of 20 minutes.
`Single breath TLCO was taken as the-mean of
`two satisfactory manoeuvres within 15% of
`each other.
`Airflow reversibility in response to sal-
`butamol and ipratropium bromide was deter-
`mined during the baseline period. Two doses of
`salbutamol (200 Mug and 10 mg) were adminis-
`tered sequentially, and the response was deter-
`mined 20 minutes after each dose. On a sub-
`sequent day 72 and 500 ,ug of ipratropium
`bromide were given and the spirometric res-
`ponse was assessed 25 minutes after each dose.
`The smaller dose of each drug was given by a
`metered dose inhaler and the larger dose by an
`Inspiron mini-Neb nebuliser, the drug being
`nebulised to dryness. Bronchial responsiveness
`to inhaled histamine was determined by the
`method of Cockroft et al,7 on the second
`baseline visit if the FEV, was above 0-6 1.
`After the second baseline visit patients were
`asked to measure peak expiratory flow rate
`(with a mini Wright's peak flow meter) four
`hourly during waking hours at home, and
`record the best of at least three attempts with
`the best two within 20 1/min. All baseline values
`were obtained during the two weeks before the
`first treatment phase and mean daily PEF was
`calculated over the final seven days of this
`period. Diurnal variation in PEF was calcu-
`lated from the same readings as mean daily
`maximum PEF (mean daily maximum PEF
`minus mean daily minimum PEF divided by
`mean daily PEF).
`
`Serum IgE concentrations were measured by
`a PRIST technique and skinprick
`tests
`performed with house dust, Dermatophagoides
`pteronyssinus, Aspergillus fumigatus, cat fur,
`and a control solution, a positive result being
`defined as a weal 2 mm greater than that
`obtained with the control solution in two or
`more tests.
`Inhaler technique was checked at each visit
`and corrected
`patients
`All
`necessary.
`as
`continued their usual bronchodilator treatment
`unchanged during the trial, and were instruc-
`ted to maintain a constant timing between
`doses and PEF readings.
`
`DESIGN
`The trial was a randomised, double blind,
`double dummy, crossover study designed to
`compare inhaled beclomethasone dipropionate
`500 jug thrice daily, oral prednisolone 40
`mg/day, and placebo. Each treatment was given
`for two weeks followed by a two week washout
`period before the next treatment period. The
`first treatment period was preceded by a four
`week baseline period. Patients attended the
`laboratory on days 1 and 14 of the baseline
`period for initial investigations, and on the last
`day of each treatment period for subsequent
`assessments.
`
`ANALYSIS
`A full response to treatment was defined as an
`improvement in absolute values of FEV, or
`FVC on the final treatment day, or mean PEF
`over the last seven days of treatment, of at
`least 200o when compared with baseline. An
`improvement of at least 1500
`in any one
`measure or at least 100,) in any two measures
`was defined as a partial treatment response.
`Baseline data were compared by means of a
`paired or unpaired Student's t test for normally
`distributed data or a Wilcoxon signed rank or
`rank sum test for data not normally distributed.
`Treatment order effect was assessed by a logit
`regression on proportions, the GLIM statis-
`tical package being used.8 Active treatment
`response rates were compared by McNemar's
`test, and the responses to the treatments given
`during the first phase by a x2 test.
`All predicted values are derived from pub-
`lished equations.9
`
`Results
`Of the 127 patients who entered the study, 107
`completed
`protocol.
`the
`Eleven
`patients
`defaulted at subsequent visits:
`six had an
`infective exacerbation of their disease, one died
`of an unrelated cause during the run in period,
`and two had complications during the oral
`prednisolone phase (exacerbation of chronic
`duodenal ulceration in one, left ventricular
`failure in the other). The mean age (63 years)
`and the FEV, and FVC (390% and 70% predic-
`ted) of those withdrawn were similar to the
`mean values in patients who completed the
`study. The baseline lung function characteris-
`tics of the study group are given in table 1 and
`details of smoking and atopy in table 2.
`There was a significant order effect in the
`
`MYLAN EX 1010, Page 2
`
`

`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`114
`
`Weir, Gove, Robertson, Burge
`
`Table 1
`Baseline lungfunction characteristics of the
`study group (mean (SD) values unless otherwise
`indicated)
`
`Table 3 Number offull responders (partial responders
`in parentheses) to each treatment according to the phase
`in which it was administered
`
`(25F) Number
`107 (25F)
`Age: years
`62 9 (9 0)
`FEV,: litres
`1.19 (0*47)
`(o predicted
`44 2 (16 7)
`Forced vital capacity: " predicted
`79 2 (19-0)
`Peak flow (PEF): U) predicted
`52 4 (17 9)
`Total lung capacity: 00 predicted
`121 5 (19 4)
`Residual volume: ", predicted
`173-6 (52 0)
`Transfer coefficient: "O predicted
`77 8 (31 7)
`Diurnal variation in PEF: U) mean
`19 8 (9 9)
`PC20 (mg/ml): geometric mean (range)
`0 81 (0 03-16 0)
`FEV,: indices of reversibility of airflow obstruction
`As ",0 initial FEV,, in response to:
`10 mg salbutamol
`18 0 (15 9)
`200 pg salbutamol
`11 4 (11 2)
`500 pg ipratropium bromide
`20 1 (17 0)
`72 pg ipratropium bromide
`13 1 (11 7)
`As 0,) (predicted minus initial) FEV,, in response to:
`10 mg salbutamol
`15 5 (16 8)
`200 pg salbutamol
`10-9 (13 7)
`500 pg ipratropium bromide
`19 2 (26 1)
`72 pg ipratropium bromide
`12 5 (16 4)
`PC,0-provocative concentration of histamine causing a 20%
`fall in FEVy.
`
`Table 2
`
`Smoking and atopy in the 107 subjects
`
`Number of:
`Current smokers
`Ex-smokers
`Lifelong non-smokers
`Mean (SD) cigarette consumption
`(cigarette years)
`Number with chronic bronchitis*
`Serum IgE (kU/l): geometric mean (range)
`Number with positive skin test responses
`
`*Medical Research Council definition.
`
`41
`54
`12
`
`761 (547)
`77
`74 (3-4500)
`46
`
`response to placebo, in that the placebo res-
`ponse rate was greater when placebo had been
`preceded by active treatment (X% = 5 06,
`p < 0-05: table 3); this was not seen with the
`response to prednisolone or to beclomethasone
`dipropionate (X2 = 0.75 and x2 = 0 02).
`Because of this order effect, response to treat-
`ment was defined with respect to
`initial
`baseline values before any trial treatment had
`been given, and the two active treatments were
`compared with placebo by a parallel group
`analysis of the first treatment phase data.
`
`ANALYSIS OF DATA AS A PARALLEL GROUP STUDY
`USING THE FIRST TREATMENT PHASE ONLY
`On entry into the study patients randomised to
`receive placebo, beclomethasone, or predn-
`isolone for the initial treatment phase, did not
`differ in terms of baseline physiological charac-
`teristics (table 4). The number of patients
`showing a full response to prednisolone (16/38)
`
`Table 4
`Baseline characteristics of the groups according to the treatment received in
`the first phase (mean (SD) unless otherwise indicated)
`
`Placebo
`
`35 (7F)
`64-2(81)
`1-15 (048)
`42-8 (14 2)
`
`Number
`Age:years
`FEV,: litres
`0° predicted
`Reversibility of airflow obstruction
`(as % baseline) in response to:
`IO mg salbutamol
`500 pg ipratropium bromide
`Cigarette consumption:
`Cigarette years
`Number of:
`Current smokers
`Ex-smokers
`Lifelong non-smokers
`There are no significant differences between groups.
`
`16 3 (18-3)
`20 0 (18 5)
`712 (522)
`11
`19
`5
`
`Beclomethasone
`34 (1 IF)
`608(10-1)
`1 21 (0 53)
`45.1 (18 2)
`
`20-2 (16 3)
`22 6 (19 5)
`620 (486)
`10
`20
`4
`
`Prednisolone
`
`38 (7F)
`637(85)
`1 20 (0 43)
`44.7 (17 7)
`
`17 7 (13 3)
`18 0 (12 9)
`896 (566)
`7
`28
`3
`
`Phase I
`
`Phase 2
`
`Phase 3
`
`Placebo
`Beclomethasone
`Prednisolone
`
`3 (1)/35
`7 (1)/31
`8 (4)/34
`10 (4)/41
`16 (1)/38
`12 (1)/35
`**p < 0 05 for treatment order effect.
`
`12 (0)/41**
`8 (2)/32
`11 (3)/34
`
`was significantly greater than the number
`showing a similar response to placebo (3/35,
`zi' = 10 64;
`p < 0 005). A full
`response
`frequently
`occurred
`with
`beclo-
`more
`methasone (8/34) than with placebo, though
`this was not significant (X2 = 2 22; NS). When
`partial responses are included in the analysis
`the response rate for both inhaled beclo-
`methasone (12/34) and oral prednisolone (17/
`38) was significantly greater than that for
`placebo
`(4/35:
`x2 =551, p < 002;
`and
`Xi2 = 9-86, p < 0 002 respectively). There was
`no significant difference in the response to
`either active drug in this analysis either for full
`responders (X2 = 2 79, p < 0 1) or when full
`partial
`and
`responders
`considered
`were
`(Z2 = 0 67).
`
`ANALYSIS OF DATA FROM ALL THREE TREATMENT
`PHASES IN THE CROSSOVER STUDY
`When all
`three
`treatment
`phases
`were
`considered to allow within subject comparison
`of response, 21 patients showed a full response
`to both prednisolone and beclomethasone. A
`further 18 patients responded to prednisolone
`only and five to beclomethasone only. Six of the
`18 prednisolone only responders showed a
`partial response to beclomethasone and one of
`the
`five beclomethasone
`only
`responders
`showed a partial response to oral prednisolone.
`In total, 44/107 (41 %) patients showed a full
`response to prednisolone or beclomethasone
`and a further six a partial response. The
`significantly
`response
`rate
`was
`greater
`with
`prednisolone
`than
`beclomethasone
`(Mcnemar's test 2-71,p < 005).
`% change in
`FEV, with BDP
`150 -,
`
`100_
`
`50_.
`
`0-
`
`.
`
`0
`0
`
`w .
`
`: 0
`
`I
`
`0
`
`0
`
`0
`
`. 0
`
`ft
`
`11
`
`4.
`
`0 fto
`.
`
`*so 0
`so
`
`4~ L, ,
`* 0
`"ft .:
`*so .w
`-M. : 0
`0ft0.
`
`-
`
`000
`
`-50
`
`d
`5d
`100
`% change in FEVy with prednisolone
`Figure I
`Percentage change in FEV, from baseline
`values in individual patients after treatment with
`prednisolone and beclomethasone (BDP). Slope of least
`squares regression line = 0 72 (SE 0 06).
`
`150
`
`MYLAN EX 1010, Page 3
`
`

`
`-20 _w1
`
`The responses of individual patients to oral
`prednisolone and inhaled beclomethasone are
`shown in figures 1-3 for FEV,, FVC, and PEF.
`The slope of the least squares regression line
`for each plot is significantly different from the
`line of identity, indicating a greater effect of
`prednisolone on each measurement.
`A full response to an active treatment
`occurred on 65 occasions. The measurements
`in which a response was seen on these occasions
`are shown in the Venn diagram (fig 4). A full
`response was seen in all three measurements on
`only seven occasions and in two of the
`measurements on a further 17 occasions; in
`most cases (41) a full response was seen in only
`one measurement.
`In the 44 patients who had a full response to
`prednisolone or beclomethasone the change in
`I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`the
`showing
`the
`greatest
`measurement
`response was expressed as a percentage of the
`value,
`predicted
`patient's
`measure
`a
`independent of baseline FEV, (fig 5). The
`responders remained distinct from the non-
`responders.
`The mean reversibility in FEV, in response
`to 10 mg inhaled salbutamol was 18%, expres-
`sed as a percentage of the prebronchodilator
`If expressed in terms of
`value (table
`1).
`potential reversibility-that is, as a percentage
`of the predicted minus the prebronchodilator
`value-the mean improvement was 15 5%.
`Only 13 patients showed an increase in this
`measurement of over 500/0, indicating that
`most of the patients had relatively fixed airflow
`obstruction. Response to prednisolone or
`beclomethasone or both in the 13 "reversible"
`patients (4/13) was similar to that seen in the
`"irreversible" patients (40/94; x2 = 0-64 NS).
`Cigarette consumption in the patients varied
`from zero to 2520 cigarette years, with a mean
`of 761 cigarette years (table 2). Twelve patients
`claimed to be life long non-smokers. Eighty one
`patients admitted to a cigarette consumption in
`excess of 400 cigarette years. Full and partial
`responses to prednisolone or beclomethasone
`or both were similar in this group of heavier
`smokers (39/81) and in the remaining patients
`(11/26; x2 = 0.57; NS).
`
`1
`1
`50
`0
`% change in FVC with prednisolone
`
`Discussion
`The finding of a significant order effect with
`placebo treatment complicated the analysis of
`the data from this trial. The analysis of the data
`FVC on the first treatment phase removes the
`confounding influence of the order effect and
`shows that both active treatments are superior
`physiological
`in producing a
`placebo
`to
`response in these patients. In this analysis oral
`prednisolone produced a response in more
`subjects than inhaled beclomethasone, though
`the difference was not significant. Previous
`studies have not always commented on an order
`effect, though some have used a single blind
`design that would avoid this problem.'6 Two
`smaller studies with a crossover design found
`no treatment order effect,'01' possibly because
`of the smaller numbers of patients.
`Although in our study the order effect was
`seen only with placebo it might have occurred
`to some extent with active treatment. The lack
`
`PEF
`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Corticosteroid trials in non-asthmatic chronic airflow obstruction
`
`115
`
`*
`
`*
`
`0
`4 0
`
`**
`
`:. ._ . .
`..*
`
`@
`
`*
`
`II
`
`20
`-20
`0
`40
`60
`80
`% change in mean PEF with prednisolone
`
`% change in
`FVC with BDP
`100
`
`50._
`
`0-
`
`I
`
`.
`
`:
`.,
`
`-
`0 . -..
`
`.:.
`
`...
`
`'*
`
`'a
`.6
`
`.
`
`.,L. ..
`I
`
`..a
`
`Percentage
`Figure 2
`change in mean daily peak
`expiratory flow (PEF)
`from baseline value in
`individual patients after
`treatment with
`prednisolone and
`beclomethasone (BDP).
`Slope of least squares
`regression line = 0-46
`(SE 0 05).
`
`Figure 3
`Percentage
`change in forced vital
`capacity (FVC) from
`baseline value in
`individual patients after
`treatment with
`prednisolone and
`beclomethasone (BDP).
`Slope of least squares
`regression line = 0 62
`(SE 007).
`
`Figure 4
`Venn diagram
`showing the measurements
`in which response occurred
`in the 65 individual
`responses to treatment.
`
`% change in
`mean PEF
`with BDP
`100
`
`-
`
`80 1
`
`60 1
`
`40 4
`
`20 _
`
`0-
`
`-40
`
`-50
`
`-50
`
`FEV1
`
`MYLAN EX 1010, Page 4
`
`

`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Weir, Gove, Robertson, Burgc
`
`minute walking distances improved signifi-
`cantly with both active treatments in the
`steroid responsive patients, whereas no effect
`was seen with treatment in the non-responders.
`Visual analogue scores showed a wide variation
`and, although they improved in all response
`groups with treatment, the changes were not
`significant.
`A further possible criticism of our results is
`that because of the "soft" entry criteria we
`inadvertently included patients with missed
`asthma in the study population. The criteria
`were chosen to reflect clinical practice and only
`where the physician was unsure of the benefit
`of steroid treatment-that is, where asthma was
`not present-was a patient entered. Those who
`had had respiratory disease in childhood were
`excluded, which eliminated many patients with
`asthma. Not all patients were current smokers
`or ex-smokers, although most had smoked
`heavily. Many patients showed a degree of
`reversibility in response to inhaled broncho-
`dilators that was within the "asthmatic" range
`of 20 0o or more of the prebronchodilator FEV,.
`This, however, is a misleading measure of
`reversibility
`in patients with a low pre-
`bronchodilator FEV,. A better indication of
`reversibility is obtained by considering the
`reversibility as a percentage of the predicted
`FEV, minus the prebronchodilator value (table
`1). Most of our patients had largely irreversible
`airflow obstruction. Response to one or both
`corticosteroid treatments was not related to
`past cigarette consumption or to reversibility in
`response to inhaled salbutamol, suggesting
`that our patients were predominantly non-
`asthmatic. The degree of response to cortico-
`steroids shown in figure 5 shows a unimodal
`distribution, again suggesting that the patients
`came from a single disease group. Hence we
`believe that most of the patients were not
`asthmatic, and that our findings are relevant to
`patients diagnosed as having chronic airflow
`obstruction in clinical practice.
`The trial of Wardman et al6 showed that
`1500 pg beclomethasone/day was comparable
`to oral prednisolone 30 mg in patients similar to
`ours, in terms both of the number showing a
`response and of the degree of improvement
`seen in the measure of lung function. Like
`Harding and Freedman,5 we have not found
`this to be the case. The likeliest explanation for
`this is differences in selection of patients and
`perhaps in deposition ofaerosol in the lung. We
`attempted to optimise inhaler technique in our
`group by checking and correcting technique at
`each visit, but possibly improved delivery of
`the drug to the airways by means of a spacing
`device would have produced more responders
`to beclomethasone. Our results would suggest,
`however, that inhaled beclomethasone 500 pg
`thrice daily should be considered an effective
`treatment in over half of patients with non-
`asthmatic airflow obstruction who show a
`response to oral prednisolone 40 mg daily.
`The contribution of the following is gratefully
`acknowledged: Dr D Geddes for help with the
`trial design, Allen and Hanburys Ltd for finan-
`inhalers,
`cial support and providing the
`
`RESPONDERS
`
`PARTIAL
`RESPONDERS
`
`NON
`RESPONDERS
`
`FVC
`
`PEF
`
`FEVI
`
`FVC
`
`PEF
`
`FEV,
`
`FVC
`
`PEF
`
`0
`
`0
`
`0.
`
`0
`
`*
`
`*
`Aft.
`
`--
`
`0
`
`*
`
`A"
`
`*@
`
`0
`000
`- m
`
`0
`
`*0
`
`0 0 0
`
`0
`
`@0
`00
`
`@000
`00
`
`0
`
`*:
`
`0
`
`@00
`@0*
`00
`00
`
`0
`
`0
`
`00
`
`FEVI
`
`*
`
`0*
`
`*
`
`0*
`
`0
`
`40
`
`30
`
`20
`
`10 -
`
`116
`
`a)
`
`_)
`'a)
`Co
`
`o-0
`0.
`'a
`U)
`
`Q C
`
`"I0-0
`UL)
`
`o 0
`
`0)
`
`U0
`
`Change from baseline after treatment in the measurement that showed the
`Figure S
`greatest percentage change. The change is expressed as a percentage of the predicted
`valuefor the patientfor that measurement. The horizontal lines represent the mean
`response of all the patients who showed a positive change.
`
`of any detectable order effect for the response
`to prednisolone and beclomethasone suggests
`that the action of the second active treatment is
`more powerful than the carry over effect of the
`initial treatment. We feel justified therefore in
`using data from all three treatment phases to
`further compare individual responses to predn-
`isolone and beclomethasone. When this was
`done prednisolone was superior to beclometh-
`asone in producing a response, though over half
`of the patients showing a response to predn-
`isolone also responded to inhaled beclometh-
`asone. The reason why five patients responded
`to inhaled beclomethasone only is not clear.
`Our study may be criticised for the response
`criteria adopted. Response was expressed in
`terms of percentage change from the baseline
`value, a criterion used in previous trials.6 101 13
`The validity of such a definition is questionable
`when the absolute value of the variable studied
`is low. In these circumstances small changes
`that are within the error of measurement of the
`variable may assume undue significance. Only
`40O of our responders, however, showed a
`change in FEV, or FVC that was within the
`950/0
`term
`short
`for
`confidence
`limits
`in FEV, and FVC published
`variability
`recently.'4 Previous studies of longer term
`variability in spirometric indices in similar
`patients suggest that our criteria are reason-
`able."5 16 Expressing change as a percentage of a
`measure independent of the baseline-that is,
`the predicted value-did not suggest that any
`responders had been misclassified (fig 5). One
`partial responder and two non-responders may
`have been wrongly classified.
`Symptomatic change was not determined
`formally in all patients. Visual analogue scales
`for five symptoms and six minute walking
`distances were, however, determined in the
`first 83 patients recruited to the trial.'7 Six
`
`MYLAN EX 1010, Page 5
`
`

`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Corticosteroid trials in non-asthmatic chronic airflow obstruction
`
`117
`
`Dr R Holder (University of Birmingham) for
`statistical advice, and Ms J Shepherd and Dr R
`Cayton and the staff of the respiratory function
`laboratory, East Birmingham Hospital, for
`practical assistance.
`
`1 Rudd RM. Corticosteroids in chronic bronchitis. Br Med J
`1984;288:1553-4.
`2 Stokes TC, Shaylor JM, O'Reilly JF, Harrison BDW.
`Assessment of steroid responsiveness in patients with
`chronic airflow obstruction. Lancet 1982;ii:345-8.
`3 Mandella LA, Manfreda J, Warren CPW, Anthonisen NR.
`Steroid response in stable chronic obstructive pulmonary
`disease. Ann Intern Med 1982;96:17-21.
`4 Franklin W, Michelson AL, Lowell FC, Schiller IW.
`Bronchodilators and corticosteroids in the treatment of
`obstructive pulmonary emphysema. N Engl J Med
`1958;258:774-8.
`5 Harding SM, Freedman S. A comparison of oral and inhaled
`steroids in patients with chronic airways obstruction:
`features determining response. Thorax 1978;33:214-8.
`6 Wardman AG, Simpson FG, Knox AJ, Page RL, Cooke NJ.
`The use of high dose inhaled beclomethasone dipro-
`pionate as a means of assessing steroid responsiveness in
`obstructive airways disease. Br J Dis Chest 1988;82:
`168-71.
`7 Cockroft DW, Killian DN, Mellon JJA, Hargreave FE.
`Bronchial reactivity to inhaled histamine: a method and
`
`clinical survey. Clin Allergy 1977;7:235-43.
`8 Baker RJ, Nelder JA. The GLIM system. Release 3. Oxford:
`Numerical Algorithms Group, 1978.
`9 Quanjer P, ed. Standardised lung function testing. Bull Eur
`Physiopathol Respir 1983;19(suppl 5):1-95.
`10 Mitchell DM, Gildeh P, Dimond AH, Collins JV. Value of
`serial peak expiratory flow measurements in assessing
`treatment response in chronic airflow limitation. Thorax
`1986;41:606-10.
`11 Mitchell DM, Gildeh P, Rehahn M, Dimond AH, Collins
`JV. Effects of prednisolone in chronic airflow limitation.
`Lancet 1984;ii: 193-5.
`12 Weir DC, Robertson AS, Gove RI, Burge PS. Time course
`of response to oral and inhaled corticosteroids in non-
`Thorax
`obstruction.
`airflow
`chronic
`asthmatic
`1990;45:1 17-20.
`13 Lam WK, So SY, Yu DYC. Response to oral corticosteroids
`Chest
`Br J Dis
`obstruction.
`airflow
`chronic
`in
`1983;77: 189-98.
`14 Tweeddale PM, Alexander F, McHardy GJR. Short term
`variability in FEVy and bronchodilator responsiveness in
`patients with obstructive ventilatory defects. Thorax
`1987;42:487-90.
`15 Mungall IPF, Hainsworth R. Assessment of respiratory
`function in patients with chronic obstructive airways
`disease. Thorax 1979;34:254-8.
`16 Rozas CJ, Goldman AL. Daily spirometric variability:
`normal subjects and subjects with chronic bronchitis with
`obstruction. Ann Intern Med
`and without airflow
`1982;142:1287-91.
`17 Gove RI. Predictors and features of a response to cortico-
`steroids in chronic airflow obstruction. MD thesis,
`University of London, 1986:41.
`
`MYLAN EX 1010, Page 6
`
`

`
`
`
`http://thorax.bmj.com/Downloaded from
`
`
`
`group.bmj.com on March 10, 2015 - Published by
`
`Corticosteroid trials in non-asthmatic chronic
`airflow obstruction: a comparison of oral
`prednisolone and inhaled beclomethasone
`dipropionate.
`D C Weir, R I Gove, A S Robertson and P S Burge
`
`Thorax(cid:160)
`1990 45: 112-117
`doi: 10.1136/thx.45.2.112
`
`Updated information and services can be found at:
`
` http://thorax.bmj.com/content/45/2/112
`
`These include:
`
`Email alerting
`service
`
`Receive free email alerts when new articles cite this article. Sign up in the
`box at the top right corner of the online article.
`
`Notes
`
`To request permissions go to:
`http://group.bmj.com/group/rights-licensing/permissions
`
`To order reprints go to:
`http://journals.bmj.com/cgi/reprintform
`
`To subscribe to BMJ go to:
`http://group.bmj.com/subscribe/
`
`MYLAN EX 1010, Page 7