Dosimetrically administered nebulized morphine for breathlessness in very severe chronic obstructive pulmonary disease: a randomized, controlled trial

Chronic obstructive pulmonary disease (COPD) is the second, most frequent chronic respiratory entity [1]. In the last year of life of patients with severe and very severe COPD, breathlessness occurs in up to 98% [2]. There is evidence for the use of systemic, oral or parenteral, opioids to reduce the symptom of chronic breathlessness, in particular morphine [3]. Although there have been sporadic reports on central respiratory depression after systemic use of this compound, appropriately administered morphine is considered a relatively safe medication [3]. Nevertheless, an alternative route of delivery by nebulization was proposed to reduce other burdensome side effects of systemic morphine, such as constipation or dizziness. Rationale for this approach was supported by both in vitro and in vivo studies which showed beneficial effects of opioids delivered directly to the bronchial tree [4]. Hitherto, the effectiveness of nebulized morphine in breathlessness was demonstrated only in a few uncontrolled studies and case reports [5‐7]. Notably, beneficial effect of nebulized morphine was not confirmed in two systematic reviews [8, 9], whereas the most recent review showed a modest benefit [10], resulting from just one positive randomized trial [11], with low to moderate evidence across the available trials. Hence, in contrast to systemic delivery, nebulized morphine has not been considered a standard treatment.

Recent studies have shown that opioid receptors are localized within epithelium of human trachea and large bronchi on unmyelinated C nerve fibers and pulmonary neuroendocrine cells (PNEC) [12]. It was proposed that morphine, acting directly on PNECs and C-fibers, might limit neurogenic inflammation and afferent signal propagation to central nervous system, decreasing the sensation of breathlessness [12]. Hence, nebulization might be a clinically effective route of morphine administration provided the drug particles reach the large bronchi. The standard nebulizers, however, perform inhalation poorly, as a large majority of the dose is lost, while the remaining dose is dispersed in the bronchial tree in an unpredictable manner. Recently, we demonstrated that the large bronchi could be easily targeted by a dosimetrically operated nebulizer [13].

The aim of the present study was to compare clinical effectiveness of morphine and 0,9% NaCl, both delivered by the same inhalation system, calibrated to target large airways, in patients with very severe COPD and chronic breathlessness in a double blind randomized study. The null hypothesis was that there was no clinically and statistically significant difference in breathlessness between nebulized morphine and 0,9% NaCl. The primary endpoint was the intensity of breathlessness measured by visual analogue scale (VAS) with daily activity and the secondary endpoints were the most effective dose of morphine, exercise tolerance measured by Wilcock’s test and treatment safety.

Out of the 30 patients primary screened for the study, 5 declined participation, and 8 died before entering the trial. Due to the observed, bigger than expected, differences in VAS scores between the two study arms [27], the trial needed to be stopped, ethically, after 10 of 11 admitted patients completed study protocol Table 1 i.e. after reaching calculated sample size. One patient developed infective COPD exacerbation on the second day of trial during 0.9% NaCl nebulization phase and was excluded from the analysis. Two patients were included into the study in spite of higher values of FEV1% (35,5% and 31% – as assessed, consecutively, 3 and 5 months before the study). At the time of screening they were unable to perform dynamic spirometry due to the steady progress of the disease. Standard COPD treatment remained unchanged during the study and each patient received maximum COPD treatment.

FEV1, forced expiratory volume in one second; BMI, body mass index; LTOT, long term oxygen therapy; M, male; F, female; HT, hypertension; CAD, coronary artery disease; DM, diabetes mellitus; LPR, laryngo-pharyngeal reflux; PAF, paroxysmal atrial fibrillation; TR, tricuspid regurgitation; MR, mitral regurgitation; OSA, obstructive sleep apnea; GERD, gastroesophageal reflux disease.

Mean VAS changes for morphine and 0.9% NaCl periods were 25.4 mm (standard deviation (SD): 9.0; median: 23,0; range: 14.0 to 41,5; confidence interval (CI): 95%) and 6.3 mm (SD: 7.8; median: 6.8; range: −11,5 to 19,5; CI: 95%), respectively. In both groups (either starting with morphine or 0.9% NaCl) breathlessness gradually decreased during morphine phase (p < 0.0001 for both: day-to-day and minute-to-minute analysis) and a statistically significant breathlessness reduction was already achieved on the second day of the period (p < 0.002) Figs. 2 and 3. Morphine dose, however, was raised further to achieve 20 mm VAS drop. 7 out of 10 patients required raising morphine dose to 3 mg on day 3, remaining 3 patients met aforementioned clinical criterion on day 4 at a dose of 5 mg. Statistically significant and sustained decrease in breathlessness during morphine nebulization started in thirtieth minute (p = 0.005) after nebulization and peaked four hours after nebulization Fig. 3. All study patients expressed their willingness to continue morphine nebulization at home.

During NaCl nebulization, a statistically significant (p < 0.0001 for day-to-day analysis and p < 0.00001 for minute-to-minute analysis) drop in breathlessness intensity started on the third day (p = 0.04) in those patients who started their NaCl nebulization after four days of morphine nebulization. However, this drop (mean VAS change of 9.8 mm) did not meet a preset clinical significant cut-off of 20 mm. This improvement was not seen in Group 1 (p = 0.926 for day-to-day analysis and p = 0.908 for minute-to-minute analysis), where NaCl nebulization preceded morphine nebulization Fig. 4. The mechanism for this change is not immediately apparent.

A significant improvement (p < 0.05) in Wilcock’s test, independent of the substance used, was observed in both groups Fig. 5 with the exception of number read per breath during NaCl nebulization (p = 0.06). We did not detect statistically significant difference between influence of 0.9% NaCl and morphine on the test. The mean “number of numbers read” increase was 12.7 for the morphine and 8.1 in NaCl period (p = 0,09), respectively, whereas mean “number read per breath” change was 4.3 and 3.7 (p = 0,28), respectively. Aforementioned data were obtained for 9 (“number of numbers read”) and 8 (“number read per breath”) patients. One patient, due to visual impairment, could not perform Wilcock’s test properly and in another collection of necessary “number read per breath” was impossible due to severe breathlessness.

Morphine nebulization was well tolerated and there was no significant side effects, apart from a bitter taste (8 patients) and transient, mild dizziness immediately after the nebulization (2 patients). Mean respiratory rate during NaCl period equaled 20.2 breaths/min, during morphine period – 20.18 (p = 0.74). No changes in heart rate, blood pressure and SpO₂ and no significant decrease in spirometric parameters was observed between periods Figs. 6 and 7.

Our study showed a reduction of chronic breathlessness accompanying severe COPD by dosimetric nebulization system. We were able to demonstrate in the randomized setting that inhaled morphine at a dose 3–5 mg decreases breathlessness by more than 20 mm in the VAS, with minimal side effects and that this improvement is sustained for at least 24 h after 1 dose. We were able to meet this high threshold despite the evidence that VAS change greater than 10 mm might already be clinically significant in chronic breathlessness [26]. Nonetheless, we adopted 20 mm in view of earlier, unsuccessful nebulized morphine trials. Despite positive VAS results we did not detect significant difference in Wilcock’s test. Taking into consideration that Wilcock’s test results correlate with FVC [19] this can be explained by the lack of clinically significant changes in results of static spirometry. An apparent limitation of this study was inability to achieve effective blinding, since this requirement could not be accomplished due to a strong bitter taste of nebulized morphine. Moreover, trial design changed after publication of study by Johnson et al. [26] which allowed us to properly calculate sample size.

Our positive results differ most likely due to the differences in the methods of nebulization. Virtually all previous studies used opioids delivered by jet nebulizers widely known for their unreliable drug delivery, or did not specify the equipment used explicitly. Indeed, up to 70% of the drug delivered by jet nebulizer is deposited inside the apparatus and up to 20% is lost into the environment leaving barely 10% of set dose that reaches the lungs [28]. Moreover, aerosol is produced in a constant fashion, resulting in a considerable variability in drug deposition among consecutive nebulizations delivered by the same jet nebulizer [28]. Furthermore, only one group [29] among the randomised controlled studies analyzed by systematic reviews [8‐10] chose mass median aerodynamic diameter (MMAD) suitable for deposition in the large airways (3,1–4,9 μm). The aforementioned might explain why despite using a wide range of nebulized morphine doses (1 mg–50 mg [30, 31]) previous researchers were mostly unable to achieve positive results.

This study used dosimetric nebulizer (PNEUMONEB®) coupled with BCTS-S head. PNEUMONEB® analyzes patients breathing pattern in a real time and delivers drug aerosol bolus in the third quarter of the inspiration. This approach minimizes drug losses to the environment and to the inner surface of the device [28], increases drug deposition in the trachea and large bronchi [32], where PNECs and C-fibers are located, and ensures the repeatability between consecutive drug deliveries [33]. BCTS-S head produces large aerosol particles with MMAD of 4.6 μm [13], further increasing morphine deposition in the trachea and large bronchi [32]. This method may increase drug deposition in lungs up to 60% of the dose [13], of which a significant portion reaches opioid receptors. It is worth underlining that pharmacokinetics of nebulized morphine delivered by PNEUMONEB® in cancer patients [34] is substantially different from that of morphine delivered by other routes [35‐39]. Although bioavailability of nebulized morphine was estimated in earlier, jet nebulizer studies at less than 10% [40, 41], its effective doses in our study (3 and 5 mg) were close to ones given parenterally [42]. However, it is worth mentioning that plasma levels of morphine and its metabolites after nebulization with PNEUMONEB® with BCTS-S head [34] are lower than after intravenous [36] delivery or after nebulization with AERx® [43], a system designed to deliver drugs systemically through alveoli.

Further work is then required, using PNEUMONEB®, correlated simultaneously with plasma levels of morphine and its metabolites, in order to understand where, centrally or locally, the opioid is having its dominant effect.

The role of dynamic hyperinflation was not considered during this study, however respiratory rates and measured lung volumes remained stable throughout the study. Its contribution to breathlessness needs to be considered in future studies.

Our study showed an apparent reduction of chronic breathlessness intensity now accompanying severe COPD with morphine delivered by dosimetric nebulization system which ensured delivery of drug to the desired level of the airways. Treatment was effective and safe in all participating patients. In consequence, all of them expressed willingness to continue morphine nebulizations at home. In the majority of patients the effective dose was 3 mg. This positive effect was most probably achieved through direct morphine action on its receptors located in the epithelium of the trachea and large bronchi.