Respirator3, Medicine (1990) 84, 471-477
Home nebulizers: can optimal therapy be predicted by laboratory studies? B. R. O'DRISCOLL, E. A. KAY*, R. J. TAYLOR~ AND A. BERNSTEIN Departments of Thoracic Medicine, *Pharmacy and t Medical Physics, Hope Hospital, Salford M6 8HD, U.K.
Twenty patients (six severe asthma, 14 chronic obstructive pulmonary disease, COPD) were referred for consideration of domiciliary nebulized treatment. A double blind laboratory assessment demonstrated similar subjective and objective responses to nebulized salbutamol (5 rag), ipratropium bromide (IB) (0.5 rag) or a mixture of these medications in both groups of patients. The patients subsequently self-administered each treatment four times daily for one month. Fourteen patients requested long-term home nebulizer treatment (three salbutamol, four ipratropium bromide, seven mixture), and nine of these had their highest domiciliary peak flow recordings during home nebulizer treatment. However, subjective and objective laboratory assessmerits did not clearly predict the patients long-term choice of therapy in any case. There was little overall correlation between the laboratory response and the domiciliary response to treatment (Spearman correlation coefficient; subjective score, laboratory vs. home, r = 0.27, P = 0"03; peak flow response 30 rain after treatment, laboratory vs. home, r = 0.31, P < 0.02). The hospital study was also unreliable in predicting side effects during domiciliary nebulizer use. We conclude that prospective laboratory studies are of little value in the assessment of patients for home nebulizer therapy; these assessments must be made by carefully supervised domiciliary trials of nebulized treatment.
Introduction Domiciliary nebulizers are in widespread use for the treatment of patients with severe asthma or chronic obstructive pulmonary disease (COPD). However, there are no agreed principles for deciding which patients will benefit from this form of treatment or which medication should be given to each patient. It has been suggested, however, that laboratory studies might identify optimal treatment for individual patients (1,2). The object of this study was to determine whether a small number of laboratory visits could predict whether individual patients would respond to domiciliary nebulized treatment. We also wished to establish whether laboratory studies could identify optimal domiciliary nebulizer treatment for individual patients. We have conducted a double-blind study of the acute response of six asthmatic patients and 14 COPD patients to nebulized treatment. This has been compared with their response to domiciliary nebulized treatment over a 3-month period.
Patients and Methods Twenty patients were referred by their family doctor or chest physician for consideration of home nebulizer Received27 Februaryand accepted31 July 1990. Correspondence to: Dr B. R. O'Drisco[1,Hope Hospital,Salford M6 8HD. 0954-6111/90/060471 + 07 $03.00/0
treatment. All had been treated with large doses of beta agonists and ipratropium bromide by metered dose inhaler (MDI) or dry powder inhaler with limited success. Six patients had asthma without evidence of chronic bronchitis and 14 had chronic obstructive pulmonary disease as defined by the American Thoracic Society (3), (Table 1). Six of the COPD patients had evidence of significant emphysema (carbon monoxide transfer coefficient less than 72% of the predicted value). Before starting nebulizer treatment, all patients used a Wright's Mini Peak Flow Meter to record their peak flow rate (PFR) three times daily for 2 weeks on their usual treatment, then 2 weeks using a'nebuhaler' pear-shaped spacer to administer terbutaline I mg and ipratropium bromide 80/lg four times daily. All patients were admitted to hospital for 4 days at a time when their lung disease was stable. All their usual medications were continued with the exception of inhaled bronchodilators. These were omitted for 8 hours before each study. Nebulized treatment was given at 0900 on each study day in a random order, double-blind manner. Nebulized treatments consisted of: (a) saline 0.9%; (b) salbutamol 5 mg; (c) preservative-free ipratropium bromide 0.5 mg; (d) salbutamol 5 mg mixed with ipratropium bromide 0.5 mg. All treatments were made up to 4ml with 0.9% saline. All nebulized treatments were given using a Hudson Updraft nebulizer driven by oxygen at a flow rate of 9 1990Bailli~reTindall
B. R. O 'Driscoll et al. Table I
Clinical details of patients Asthmatic patients n=6
Mean age (range) years Sex Smoking status Mean FEV~(so) Mean FVC (SD) Mean KCO* as percent of predicted value (range) Inhaled steroid use (mean daily dose) Oral prednisolone use (mean daily dose) Oral theophylline use (mean daily dose) Inhaled beta agonist'l" (mean daily dose) Inhaled ipratropium bromide (mean daily dose)
Bronchitic patients n= 14
41 (20-53) 3M,3F 1 current, 3 Ex 1-371 (0.40) 2.68 1(1-00)
62 (39-74) 7M,7F 8 current, 6 Ex 0.691 (0.35) 2.03 1(0.94)
68 ( 15-103)
6 ( 1483 lag)
9 ( [666 lag)
1 (5 rag)
5 (7.5 mg)
4 (675 nag)
10 (600 mg)
6 (3367 lag)
14 ( 1035 lag)
4 (190 lag)
I0 (252 lag)
*KCO = Carbon monoxide transfer coefficient. tInhaled beta agonist: 9 salbutamol MDI, 6 fenoterol MDI, 3 terbutaline MDI, 2 salbutamol dry powder inhaler.
81min -~ until the nebulizer chamber was dry. The following measurements were made before treatment and 15, 30, 60,120, 180 and 240 minutes after completing treatment: pulse rate; peak flow rate (Wright's peak flow meter); FEV I and FVC (Vitalograph dry bellows spirometer). The best of two technically satisfactory measurements was recorded. Patients were asked to record any tremor or palpitations on a numerical scale from one to seven (1 =absent; 7 = worst possible severity) at the completion of each study day. They also recorded subjective benefit from each treatment on a similar scale (1 = n o benefit; 7 = best possible benefit). Each patient was loaned a Porta Neb compressor and a System 22 Acorn nebulizer and given full instructions in their use. They self-administered treatments (b), (c) and (d) four times daily for one month each, starting with salbutamol or ipratropium in random order and finishing with the combined treatment. The domiciliary study was open because no placebo was available for ipratropium bromide preservativefree vials. Each patient recorded his peak flow rate three times daily using a Wright Mini Peak Flow Meter (on awakening, half-an-hour after the first nebulized treatment and at bedtime). Symptoms (tremor, palpitatlons and breathlessness) were recorded on diary cards on a weekly basis (1-7 numerical scale). At the end of each month, subjective benefit was recorded
for that month's treatment (I = n o benefit, 7=best possible benefit). This study was approved by the district ethics committee and each patient gave written informed consent. Statistical analysis was by analysis of variance with repeated measures (for spirometric measures during acute bronchodilator study) and Wilcoxon signed rank test (subjective data). Laboratory and domiciliary measurements were compared using Spearman's rank correlation coefficient. Results
All 20 .patients completed the double-blind hospital assessment. Nebulized salbutamol, ipratropium bromide and their combination produced subjective and objective improvement in both asthmatic and COPD patients (Figs 1 and 2). Although all active treatments were superior to nebulized saline, statistical analysis showed no significant difference between the three active treatments in either group of patients. Nineteen patients completed 3 months of domiciliary nebulizer treatment (one asthmatic patient declined home nebulizer treatment after the hospital assessment since she preferred treatment with terbutaline and ipratropium bromide inhalers delivered via a nebuhaler). Of those who completed the trial, 14 elected to continue long-term home nebulizer
450 6 ,~t~hmotlr
"T .5 E
200 14 b r o n c h i t i c poflen~=
Time ofter nebulized theropy (rnin)
Fig. 1 Bronchodilator response (PFR) after four nebulized treatments given in a double blind manner in the laboratory. Open circles= saline; closed circles= salbutamol 5 mg; open squares= ipratropium bromide0'5 mg; closed squares= mixture of salbutamol and ipratropium bromide. The FEV] and FVC responses were similar to the PFR response. Asthmaticgroup: salbutamol vs. salineP < 0.01, IB vs. saline P = 0.05, mixturevs. salineP = 0.01, salbutamolvs. IB vs. mixtureNS. COPD group: salbutamol vs. saline NS, IB vs. saline NS, mixture vs. saline P=0.01, salbutamol-vs.IB vs. mixtureNS.
P= 0"03 I
P =0'02 i__]11,
Subjective responses to nebulized treatments given in the laboratory. (1 =no benefit, 7 = best possible benefit). Statistical comparisons by Wilcoxon signed rank test. S = saline, IB =ipratropium bromide. Mix=mixture ors and IB. Fig. 2
treatment, three requested the 'nebuhaler' regimen for long-term use and two preferred their previous metered dose inhaler treatment (Table 2). The laboratory measurements correlated poorly with the patients choice of treatment (Table 2). No patient had a superior subjective and objective response (in the laboratory) to their preferred long-term therapy. The subjective laboratory response was predictive of the patient's long-term choice of treatment in only two cases and there was little overall correlation between the subjective response to each treatment in
the laboratory and the subjective response to the same medication during the domiciliary study (Fig. 3). The objective laboratory measurements were also of little value in predicting the patients long-term choice of treatment. The mean pretreatment peak flow rate in the laboratory was almost identical to the mean pretreatment domiciliary peak flow rate (Fig. 4). This confirms that these patients were reliabie witnesses; however, there was little correlation between the PFR response to each nebulized treatment in the laboratory and the response to the same medication when given at home (Fig. 5). This remained true even after correction for the rise in baseline PFR which occurred during successful domiciliary treatment. Three patients did not demonstrate any significant (> 15%) increase in PFR, FEV~ or FVC after any of the laboratory treatments. Two of these 'nonresponders' (patients 7 and 13) had their highest domiciliary peak flow rates while using nebulized treatment and two of them (patients 11 and 13) elected to continue long-term domiciliary nebulizer treatment. Compared with the laboratory studies, home peak flow charts correlated well with the patient's choice of treatment (Tables 2 and 3). Of 14 patients who requested long-term home nebulizer .treatment, nine had their highest domiciliary peak flow rates during nebulizer treatment. Furthermore, three of the five patients who decided against long-term home nebulizer treatment had a poor domiciliary peak flow response to this treatment. Laboratory studies had identified only one of these patients as a'nonresponder' to nebulized treatment.
Table 2 Patients' laboratory and domiciliary responses to treatment Best laboratory P F R rise*
Asthmatic patients 1 Salb 2 Mix 3 IB 4 Salb 5 IB 6 Mix COPD patients 7 Nonresponder (see text) 8 Mix 9 IB [0 Salb li Nonresponder 12 Mix 13 Nonresponder 14 Salb 15 Mix 16 Mix 17 Mix 18 Mix 19 IB 20 Mix
Best laboratory subjective response
Highest domiciliary peak flow recordings
Final choice of treatment
Was laboratory test predictive?
Salb Salb Mix IB Salb ~ IB Salb = IB = Mix
Not done (see text) Nebulizer (Mix) Nebulizer (IB) Nebulizer (Salb) Nebulizer (Salb = Mix) Nebulizer (Mix)
Nebuhaler Nebulizer (Mix) Nebulizer (Mix) Nebulizer (Mix) Nebulizer (Mix) Nebulizer (Salb)
Not tested Yes Yes No No No
Nebulizer (IB =Mix)
Metered dose inhaler
IB Salb Mix Salb = Mix Salb = IB = Mix Salb=IB IB = Mix Salb=IB = M i x IB = Mix IB = Mix Salb -- IB = Mix Salb = IB = Mix Salb-- IB
All equal Nebuhaler Nebuhaler = IB Nebuhaler = Salb Nebulizer (Mix) Nebulizer (Mix) Nebulizer (Mix) Nebulizer (IB) Nebulizer (IB) Nebuhaler Nebuhaler Metered dose inhaler All equal
Nebulizer (IB) Nebuhaler Nebuhaler Nebulizer (Salb) Nebulizer (IB) Nebulizer (IB) Nebulizer (Mix) Nebulizer (IB) Metered dose Inhaler Nebulizer (Mix) Nebuhaler Nebulizer (Mix) Nebulizer (Salb)
Yes No No No No No No No No Yes No No No
Home peak flow recordings (morning post treatment levels) were averaged over each treatment period, average recordings differing by less than 51 rain- ' were considered equal. Per cent P F R increase at 30 min. Salb ---salbutamol; IB = ipratropium bromide; Mix = combined treatment.
4o0 '6o Joo
Be~eline P F R I rain-i o~ h o m e
Subjeclive score of home
Fig. 3 Relationship between the subjective response to each treatment in the laboratory and at home. Closed circles = salbutamol 5mg; open squares=ipratropium bromide 0.5 rag; closed squares =mixture of salbutamol and ipratroplum bromide. (r = 0.27, P = 0.03).
Relationship between morning (pretreatment) PFRin hospital (mean of four recordings) and at home (mean of 14 recordings while patients were using their usual inhaled treatment). (r = 0'9, P < 0.001). As r e g a r d s side effects d u r i n g the acute study, iprat r o p i u m b r o m i d e and saline h a d no effect o n pulse rate, w h e r e a s s a l b u t a m o l o r the m i x t u r e o f s a l b u t a m o l and
-2|IBO f =
"C. ~ ~
40 2o 0
~' i:~ 9 ~e
I I J I L J i I L I 5 I0 15 20 25 50 35 40 45 50
Percent PFR rise 30 minules ofler nebulizer of home
Fig. 5 Relationship between hospital PFR rise and domiciliary PFR rise 30 mins after each nebulized treatment. (Single recording in hospital, mean of 28 recordings at home). Closed circles=salbutamol 5mg; open squares=ipratroplum bromide 0.5 rag; closed squares=mixture of salbutamol and ipratropium bromide. (r=0.31, P< 0.02).
ipratropium produced a small rise in pulse rate (salbutamol 2-7 beats rain-l, mixture 10 beats m-~) which was maximal at 15 rain. The occurrence of tremor or palpitation during the acute study did not correlate well with the occurrence of the same symptoms during domiciliary nebulizer use (five true-positive predictions, five false-positive predictions, five false-negative predictions). Two patients (patients 8 and 15) were unable to tolerate salbutamol due to tremor during the long-term study, but only one of these patients had complained of tremor during the acute study. One patient (No. 9) developed angina (for the first time) while using nebulized salbutamol four times daily at home, but he did not experience this symptom on either of the two occasions when salbutamol was given in the laboratory. Two patients (9 and 11) were unable to tolerate ipratropium bromide during the domiciliary study (because of an unpleasant taste) but neither experienced any side effects during the laboratory study.
We found double-blind laboratory studies to be of little value in selecting long-term nebulizer therapy for individual patients with asthma or chronic obstructive pulmonary disease who had been referred for consideration of this form of treatment. These findings are in agreement with a preliminary report which found that laboratory dose response studies failed to predict which patients would respond to domiciliary nebulizer
treatment (4). Although the present study utilized a different nebulizer chamber for laboratory and domiciliary studies, the manufacturers of both systems report an almost identical particle size, and high doses of both drugs were given. Therefore we do not believe that any slight differences in nebulizer characteristics would have influenced the results. Like many hospi. tals, our inpatient and outpatient services have separate equipment suppliers, therefore the present study is likely to reflect actual clinical practice when 'nebulizer trials' are conducted. Although some patients receive domiciliary nebulizer treatment unnecessarily, and it has been suggested that much of the benefit may be a placebo effect (5,6), there are undoubtedly some patients with severe airflow obstruction who derive substantial subjective and objective benefit from this form of treatment (7,8). Some such patients have been identified even in trials where the overall results were not favourable towards domiciliary nebulized treatment (5) and the majority of patients (9 of 14) who chose long-term home nebulizer treatment in the present study had their highest domiciliary peak flow rates during nebulizer treatment. Our laboratory assessments failed to predict most of the adverse effects which occurred during long-term home nebulizer use. However, it seems sensible to give the first dose of each nebulized treatment under hospital supervision since some patients may develop tachycardia or angina after high doses of nebulized beta agonists. Furthermore, ipratropium bromide nebulizer solution may cause bronchospasm in occasional cases, although this might not occur with the newer isotonic and preservative-free solutions (9). The patients' final choice of treatment was influenced by a number of factors, including minor side effects and their judgement of the efficacy of each treatment. Many of the subtle advantages and disadvantages of each treatment only became apparent over a period of days or weeks of domiciliary treatment. Several of the patients who selected a mixture of salbutamol and IB nebulizer solutions for long-term treatment commented that the bronchodilator effect seemed to last longer with combined treatment and such an effect has been described in previous studies (2,10). Having assessed these patients carefully, we conclude that little information of any real value was gained from the preliminary double-blind studies. The laboratory studies did not clearly predict the patient's final choice of treatment in any individual case and there was little correlation between subjective and objective measurements in the laboratory and those made during domiciliary treatment. Such an imprecise guide would be of little value to a practicing clinician
B. R. O'Driscoll et al.
Table 3 Mean peak flow rates (morning post treatment) of individual patients during each period of domiciliary treatment
Patients who chose nebulizer for long-term treatment 2 3 4 5 6 8 11 ~2 13" 14 15 17 19"* 20 Mean
555 283 176 293 145 102 61 153 118 78 90 70 373* 124
560 304 172 325 124 100 126 190 133 120 86 84 365 120
568 329 216" 349 243 NT 126 189 135 NT 98 75 250 /25
589 417" 182 278 256 102 91 194 129 177 I05' 64 252 NT
602* 381 206 349* 294* NT 94 202* 140' 199' NT 80 221 87
Declined nebulizer Unableto tolerate nebulizers 305 184
Patients who chose nebuhaler for long-term treatment 1 9 10 18
291 126 263 192
295* 135" 305 232*
Patients who chose metered doseinhalerforlong-termtreatment 7 16 ..
*denotes.highest mean PFR for each individual patient. **Patient 19 had recently completed a course of steroid tablets, his PFR declined progressively during the trial but he found that nebulizer treatment provided superior symptomatic relief. NT = Not tested.
who was asked to evaluate a patient for home nebulize'r treatment. Accordingly, we must disagree with pre'cibus suggestions that laboratory studies will identify Optimal therapy for such patients (1,2). Furthermore;:thepresent findings must cast some doubt on the conclusions o f Jenkins et al. (5) who found that domi~qi~tr~, treattnen~: with nebulized salbutamol was no be~t~4 f o r m o s t patients than treatment with of salbutamol delivered by a metered dose inhaler together with nebulized saline. The authors of that study used laboratory studies to predict the 'optimal' domiciliary treatment for each patient, and those who did not respond well to a metered dose inhaler were excluded from the Study, whereas these are the very patients whom many physicians would consider for domiciliary nebulizer treatment. In addition, it is possible that nebulized saline may have a therapeutic effect by loosening mucus plugs in small airways. Such an effect
has been demonstrated in bronchiectatic patients (11) and the patients in the present study had a modest bronchodilator response following the inhalation of nebulized saline in the laboratory (Fig. 1). Since it is impossible to say whether this response is a placebo response or a physiological response to nebulized saline, we would suggest that home nebulizer trials which use nebulized saline as a placebo treatment must be interpreted with some caution. We would recommend that patients who are referred for consideration of home nebulizer therapy are given equipment on loan or hire from the hospital to assess various nebulized treatments under careful supervision over a period of several weeks. For safety reasons we would recommend that the first dose of each medication should be given in hospital but assessment of response must follow prolonged domiciliary use. The patient can then decide if he/she wishes to
purchase the necessary equipment. Although home peak flow recordings are useful in monitoring a patient's response, some patients may achieve benefit which can not be measured by simple spirometry (12), therefore we would recommend that the final choice of treatment should lie with the patient.
Acknowledgements We thank the nursing staff on the Medical Investigation Ward, Hope Hospital for their invaluable assistance during this study. Unit dose vials of preservative free ipratropium bromide were a gift from Boehringer lngelheim.
References 1. Cochrane GM, Prior JG, Rees PJ. Home Nebulisers for airflow limitation. Br MedJ 1985; 290: 1608-1609. 2. Brown IG, Chan CS, Kelly CA, Dent AG, Zimmerman PV. Assessment of the clinical usefulness of nebulised ipratropium bromide in patients with chronic airflow limitation. Thorax 1984; 39: 272-276. 3. American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1987; 136: 225-243.
4. Hadfield JW, Tattersfield AE. Domiciliary nebuliser therapy: a double blind dose response assessment. Thorax 1986; 41:247 (abstract). 5. Jenkins SC, Heaton RW, Fulton TJ, Moxham J. Comparison of domiciliary nebulised salbutamol and salbutamol from a metered dose inhaler in stable chronic airflow limitation. Chest 1987; 91: 804--807. 6. Gunawardena KA, Smith AP, Shankleman J. A comparison of metered dose inhalers with nebulisers for the delivery of ipratropium bromide in domiciliary practice. Brit J Dis Chest 1986; 80:170-178. 7. Wilson RSE, Connellan SJ. Domiciliary nebulised salbutamol solution in severe chronic airway obstruction. Thorax 1980; 35: 873-876. 8. McGivern DV, Ward M, Revill S, Sechiari A, McFarlane J, Davies D. Home nebulisers in severe chronic asthma. Brit J Dis Chest 1984; 78: 376-382. 9. Beasley CRW, Rafferty P, Holgate ST. Bronehoconstrictor properties of preservatives in ipratropium bromide (Atrovent) nebuliser solution. Br Med J 1987; 294:
1197-1198. 10. Gross N J, Skorodin MS. Role of the parasympathetic system in airway obstruction due to emphysema. New Engl J Med 1984; 331:421--425. 11. Sutton PP, Gemmell HG, Innes M, Davidson J, Smith FW, Legge JS, Friend JAR. Use of nebulised saline and nebulised terbutaline as an adjunct to chest physiotherapy. Thorn:: 1988; 43: 57-60. 12. Chrystyn H, Mulley BA, Peake MD. Dose response relation to oral theophylline in severe chronic obstructive airways disease. Br Med J 1988; 297: 1506-I 510.