Inappropriate Peak Inspiratory Flow Rate with Dry Powder Inhaler in Chronic Obstructive Pulmonary Disease

Optimal peak inspiratory flow rate (PIFR) is crucial for optimizing dry powder inhaler (DPI) effectiveness for chronic obstructive pulmonary disease (COPD). This study provide an insight that there was a substantial proportion of improper PIFRs (not only insufficient but also excessive) among COPD patients using DPIs. We enrolled 138 COPD patients from a medical center in Taiwan and measured PIFRs against different internal resistances of DPIs. Proportion of excessive, optimal, suboptimal, and insufficient PIFRs were 2%, 54%, 41%, 3%, respectively, against medium-high resistance; 2%, 77%, 20%, 1%, respectively, against medium resistance; 27%, 63%, 9%, 1%, respectively, against medium-low resistance; and 42%, 57%, 1%, 0%, respectively, against low resistance (p < 0.01). Although most PIFRs against medium-high (54%), medium (77%), medium-low (63%) and low (57%) resistance were optimal, a substantial proportion of PIFRs against low resistance were excessive (42%, p < 0.01), irrespective of age, body-mass index, dyspnea severity score, and COPD severity. Insufficient PIFRs were infrequent, but suboptimal/insufficient PIFRs were most prevalent in patients older than 75 years than in younger patients (36% vs. 56%, p = 0.036) against medium-high resistance. Regularly monitoring PIFRs against the specific resistance of the DPIs and instructing patients to employ a proper inspiration effort may help to optimize the effects of DPIs.

recorded. According to GOLD guideline, we classified COPD patients into GOLD Group ABCD according to frequency of exacerbation and mMRC or CAT scoring system with the greater score. PIFRs were measured against four-degrees of internal resistances (low, medium-low, medium, and medium-high) using the In-Check Dial G16 (Clement-Clarke International Ltd, Harlow, UK), a handheld inspiratory flow device simulates different internal resistances of DPIs 11 . For low resistance devices, measured PIFRs were classified to excessive (≥100 L/ min), optimal (50-99 L/min), suboptimal (30-49 L/min), and insufficient (<30 L/min) 12 , and for those other than low resistance devices, measured PIFRs were classified into four categories, excessive (≥90 L/min), optimal (60-89 L/min), suboptimal (30-59 L/min), and insufficient (<30 L/min) 5,6,10 . The institution board of National Taiwan University Hospital (201905058RINB) approved this study and written informed consent was waived by the ethics committee due to the retrospective nature of the study. All methods were carried out in accordance with relevant guidelines and regulations. Statistical analysis. Categorical variables were compared using a chi-square test or Fisher's exact test, as appropriate. Differences in continuous variables were analyzed using the Mann-Whitney U test or ANOVA. The data are presented as numbers (percentages), median (range), mean ± standard deviation, and p = 0.05 is considered a statistically significant difference. Linear regression analysis was used to investigate the relationship between PIFR and FEV1, FVC, age and body height. The statistical analyses were performed using STATA version 14 software (StataCorp LLC, TX).

Results
A total of 138 patients with stable COPD underwent PIFR measurement during the study. The median age was 72 (37-91) years, and most participants were men (131, 94.9%). The median height was 164 (146-180) cm while the median weight was 63 (40-102) kg. Mean forced expiratory volume in one second (FEV1) was 1.53 ± 0.49 L, and the mean percentage of predicted value was 70.9 ± 21.8%. Table 1 shows the demographic parameters.

Discussion
We investigated the PIFRs against different internal resistances of DPIs among stable COPD patients and found correlations between PIFRs and FEV1, FVC, and age. Moreover, among stable COPD patients, we found that more PIFRs measured against medium resistance were optimal, more PIFRs against medium-high resistance were suboptimal, while a majority of PIFRs against low resistance were excessive.
Forced inspiration could provide faster acceleration rates, which increases the deaglommeration of particles before the dose leaves the device. Therefore, the guideline of inhalation therapies suggests "instruct the patient to inhale forcefully from the beginning" 1 . This forced inspiration method provides a simple and precise instruction to the patients. This method is easy to remember, readily found in most prescribing information for DPIs and allows patients to perform the inhalation consistently in daily life 5 . More importantly, without the maximum inhalation effort, the relationship between the flow rate and the pressure drop will be inconsistent 4 . A PIFR between 60 and 90 L/min was suggested as optimal by several previous studies [13][14][15] , and 30 L/min is considered the minimum effective PIFR, making the range of 30-60 L/min a debatable area 6,10,11,16 . Thus, in this study, this flow range of 30-60 L/min was classified as suboptimal while those less than 30 L/min as insufficient. Pavkov et al. studied 26 patients using Breezhaler, a low resistance device and found that a consistent fine particle mass can be achieved at rate of 50-100 L/min. In this way, the flow range of 50-100 L/min was classified as optimal while 30-49 L/min as suboptimal for low resistance devices 12 . High prevalence of suboptimal PIFRs were reported in previous studies which accounts for approximately 20-78% of studied population 5,13,[17][18][19][20][21] . Similarly, our study www.nature.com/scientificreports www.nature.com/scientificreports/ found the proportion of suboptimal or insufficient PIFRs range from 3-44% among stable COPD patients with different severity.
However, although most studies emphasized on suboptimal PIFRs as a major problem of inappropriate DPI usage, we further found a substantial proportion of PIFRs were excessive when measured against medium-low www.nature.com/scientificreports www.nature.com/scientificreports/ and low resistance (27% and 42%, respectively). Excessive inspiratory flow rates also have negative impact on drug deposition in respiratory tract 2 . Usmani et al. used inhaled technetium-99m-labeled monodisperse albuterol aerosols and compared the respiratory deposition at slow and fast inspiratory flow rates, and found that faster inspiratory flows yield more oropharyngeal and central lung deposition (regardless of particle size), decreased total lung deposition, and lesser clinical effectiveness 9 . We found that patients tend to have excessive PIFRs as measured against medium-low or low resistance, while suboptimal or insufficient PIFR values resulted when measured against medium-high resistance. This remained true when patients were subdivided by BMI, age, COPD group or COPD stage, emphasizing that the prevalence of excessive or suboptimal peak inspiratory flow rates could be highly correlated with an internal resistance of DPIs per se.
Several pulmonary function parameters were also found to have association with PIFRs. Mahler et al. measured PIFR against medium-low resistance in COPD GOLD stage 3 and 4 patients (mean FEV1 of 0.92 ± 0.26 L) and found that PIFRs are associated with FVC and inspiratory capacity 19 . Duates et al. demonstrated a significant correlation between PIFRs and severity of air trapping, represented by the ratio of residual volume over total lung capacity (RV/TLC) 22 . We found values of PIFRs correlated positively with FEV1 and FVC. This finding was different from Janssens et al. where PIFRs were not statistically correlated with FEV1. This possibly is because they measured PIFRs against zero resistance and the smaller study population (26 COPD patients) in their study 13 .
Similar to previous studies 13,18-20 , we found a negative correlation between age and PIFRs (p < 0.05). Furthermore, as measured against medium-high resistance, patients more than 75 years have a higher prevalence of suboptimal or insufficient PIFR (36% vs. 56%, p = 0.036). In contrast, the prevalence of suboptimal or insufficient PIFRs did not significantly correlate with age among medium, medium-low, and low resistance of DPIs. It was similar to the study conducted by Kawamatawong et al. that older patients had a higher proportion of suboptimal PIFR breathing against Turbuhaler than Accuhaler (19.3% vs 9.3%) 18 .
There are some limitations to our study. First, our COPD population is predominantly male, which may not be representative of the spectrum of the COPD population worldwide, although our papulation was similar to a random cross-sectional national survey in Taiwan which showed males accounted for 78.9% of the COPD population 23 . However, this male predominant papulation could lead to female gender and short stature, two important predictors of reduced PIFRs in the previous studies, did not have significant roles in this study 17,19,20,22,24 . Second, as patients may not exert maximal inspiratory effort every time in daily life, there may be a difference between in-office evaluation and daily practice at home. The very small number of participants with BMI < 18.5 (n = 7), GOLD stage 4 (n = 3) and mMRC grade 0 (n = 4) substantially led to great variations and pseudo-higher PIFRs in patients with BMI < 18.5 and GOLD stage 4 and pseudo-lower PIFRs in those with mMRC grade 0 (all the differences were not statistically significant). Therefore, there was great limitation and the interpretation of those findings should be very cautious. Fourth, this is a cross-sectional study and all patients had received effective treatment and pulmonary rehabilitation program without inspiratory muscle training for a period of time. A further prospective study design was needed in the future to address the effects of treatment and pulmonary rehabilitation on the PIFR. Finally, our study population is mainly stable COPD patients without acute exacerbation in previous 3 months. The result cannot be applied to those experiencing acute exacerbation of chronic obstructive disease or who just recover from acute illness.
Our study identified that PIFRs are correlated with FEV1, FVC, and age. We also revealed a substantial proportion of improper PIFRs (excessive and suboptimal) as measured against different resistances of DPIs, but insufficient PIFRs were infrequent (<5%). The highest probability of optimal PIFR was measured against medium resistance. The lower the internal resistance, the higher the probability of excessive PIFR was noted, and suboptimal PIFRs were more often noticed when measured against higher internal resistance. Both excessive and suboptimal PIFRs may negatively impact drug deposition, therefore, regularly monitoring PIFR against the resistance of specific DPI and instructing patients to employ a proper inspiration effort may help to optimize the effects of DPIs.