JOURNAL OF AEROSOL MEDICINE AND PULMONARY DRUG DELIVERY

Volume 30, Number 0, 2017

ª Mary Ann Liebert, Inc.
Pp. 1–11
DOI: 10.1089/jamp.2017.1379

Dry-Powder Inhaler Formulation of Rifampicin:

An Improved Targeted Delivery System
for Alveolar Tuberculosis

Tejal Rawal, M Pharm,1 Laurent Kremer, PhD,2 Iman Halloum, MSc,2 and Shital Butani, PhD1

Abstract

Background: The delivery of antitubercular drugs through direct lung targeting can lead to reduction in the
dose as well as side effects of the drug. In the present investigation, carrier (lactose)-based dry-powder inhaler
of rifampicin was prepared to achieve direct targeting of the drug into the lungs.
Methods: The dry powder inhaler formulation was prepared by simply mixing micronized rifampicin with
coarse and fine lactose preblend. Preliminary blends of the drug were prepared with various lactose grades
(Inhalac, Respitose, and Lactohale). Rotahaler and Revolizer were evaluated for the performance. The
32 factorial design was used to optimize the amount of drug (X1) and amount of fine lactose (X2). In vitro lung
deposition was carried out using Andersen Cascade Impactor. The % cell viability studies of the formulation
were carried out using murine macrophage J774 cell lines. The in vivo toxicity was determined using histo-
pathology. Further in vivo pulmonary pharmacokinetics of the developed dry-powder inhaler (DPI) formulation
was carried out in comparison to the marketed formulation in the rat lungs.
Results: Based on preliminary trials, Inhalac 230 and Inhalac 400 were selected as coarse and fine lactose grades,
respectively. Rotahaler exhibited better DPI performance with the evaluated drug blends. The mass median
aerodynamic diameter (MMAD) was in the range of 4.3–5.8 lm with the maximum fine particle fraction of
28.9%. The formulation exhibited negligible cytotoxicity on macrophage J774 cell lines with about 75%–80%
cell viability at 6- and 12-hour exposure. The histopathological examination revealed negligible toxicity of DPI
in comparison to the marketed formulation. The in vivo pulmonary pharmacokinetic studies of the DPI for-
mulation in rats showed higher drug concentration in lungs in comparison to the marketed formulation.
Conclusion: The carrier-mediated dry-powder inhaler of rifampicin could serve as an improved and efficient
system for local targeting of drugs into the lungs.

Keywords: Andersen Cascade Impactor, dry-powder inhaler, pharmacokinetics, rifampicin, tuberculosis

Introduction phages.(2–4) The present existing treatment regime consists of

high-dose combinations of oral antitubercular drugs. About

T uberculosis has become the second deadliest disease

after HIV/AIDS, about 9 million cases worldwide with
its prevalence in developing countries.(1) It is a virulent dis-
90% of the replicating bacteria are killed by isoniazid,
whereas for combating the prevailing bacteria in the persis-
tent phase, rifampicin plays a very crucial role. Since ri-
ease caused by the fatal Mycobacterium tuberculosis, which fampicin shows weak sterilizing activity, it must be taken for
mainly invades the alveolar macrophages. M. tuberculosis about 6–9 months for the complete abolition of the bacteria,
generates an ample amount of ammonia and sulfatides. These thus, leading to a large oral dose with severe side effects. In
compounds prevent the fusion of phagosomes (in which the this context, a dry-powder inhaler (DPI) formulation of ri-
Mycobacterium gets entrapped) with the lysosomes; thus fampicin would serve as an effective therapy with low dose
leading to the survival of the bacterium inside the macro- and less side effects.(5–7)

1
Institute of Pharmacy, Nirma University, Gujarat, Ahmedabad, India.
2
Centred’étude des Pathogènes et Biotechnologies pour la Santé, Montpellier, France.

1
2 RAWAL ET AL.

DPI are chiefly composed of micronized drug admixed formulations are desirable. Furthermore, the micronization
with the carrier particles of larger size to aid in flow and may generate positive charge on rifampicin molecules,(17)
dispersion. The drug particles in the micronized form have a which may enhance their interaction with the negatively
large surface area that readily agglomerates. In this milieu, charged sialic acid released by alveolar macrophages for
the carrier particles reduce agglomeration and improve flow. significant targeting into the lungs.(18,19)
The desired aerodynamic particle size for the drug to settle The purpose of this study was to develop carrier-mediated
in alveoli (which is the main target site) is 1–5 lm. Many of DPI of rifampicin with improved aerodynamic properties for
these particles escape mucociliary clearance and deposit into better lung targeting. As the formulation is directly admin-
the lower airways, where the air velocity is very low; thus istered into the lungs, the in vitro cytotoxicity using mac-
experiencing sedimentation under the influence of gravita- rophage J774 cell lines and in vivo cytotoxicity through
tional forces. The particles of size greater than 5 lm undergo histopathology were performed to understand the optimum
impaction in the oropharyngeal region from where they are dose administration. Further in vivo pulmonary pharmaco-
further removed by swallowing. The desired aerodynamic kinetic studies in the suitable animal models were done to
particle size for the drug to settle in alveoli (which is the determine the potential of developed DPI formulation in
main target site) is 1–5 lm. comparison to the marketed formulation.
Many of these particles escape mucociliary clearance and
deposit into the lower airways where the air velocity is very Materials and Methods
low; thus experiencing sedimentation under the influence of
gravitational forces. The particles of size greater than 5 lm Materials
undergo impaction in the oropharyngeal region from where Rifampicin and hydroxypropyl methylcellulose capsules
they are further removed by swallowing. The particles less (rotacaps) were obtained as a gift sample from Cipla Ltd.
than 1 lm also reach to alveoli; however, it may favor rapid Mumbai. Rotahaler and Revolizer were purchased from a
absorption of drug into the systemic circulation from the local vendor. Lactose grades (Inhalac 230 and Inhalac 400)
periphery of the alveoli, where proximity to the blood supply were obtained as a gift sample from Meggle, Germany.
and the large surface area facilitate transport.(8–10) Other grades (Respitose SV003, Respitose SV010, Re-
Several mechanisms are involved in the generation of spitose ML001, Respitose ML006, and Lactohale LH 230)
aerosol. When the patient inhales, the airflow through the were obtained as a gift sample from DMV-Fonterra Ex-
device generates turbulence; air penetrates the powder thus cipients GmbH & Co. The cellophane membrane was pur-
leading to the fluidization of the particles. The fluidized chased from HiMedia, India. All the chemicals and reagents
particles get separated from the carrier particles and are fur- used were of analytical grade. The protocol for the ani-
ther carried deep into the lungs, whereas the carrier particles mal study was approved by the Institutional Animal
of larger size get trapped into the oropharyngeal region. Ethics Committee held at the Institute of Pharmacy, Nirma
Lactose is an excipient that is widely used in DPI as a carrier University (Project No. IP/PCEU/PHD/19/009).
for the micronized drug particles due, in part, to desirable
features, including biocompatibility, biodegradability etc.(6,11)
Particle size distribution, density,
However, the dosing limitation of the carrier-mediated
and aerodynamic diameter
drug delivery strategy needs to be considered, as the carrier
molecules constitute major portion of the formulation. The particle size distribution of micronized rifampicin was
Since, the dose of the drug in the DPI remains less, patient’s analyzed by Malvern Mastersizer (Hydro 2000 MU). Bulk
inhalation force plays a key role in the drug administration and tapped volumes were determined by the method as per
through DPI. If the patient’s inhalation force is less, most of the United States Pharmacopoeia (USP).(20) The bulk and the
the formulation may remain in the upper airways, owing to tapped densities were further calculated. The flow properties
less amount of the drug reaching the site of action.(9) of the powder were then determined by calculating Hausner
Rifampicin is an essential component of antitubercu- ratio. Theoretical estimates of aerodynamic diameter were
lar treatment stemming from its high affinity to the DNA- obtained from the particle size data and the tapped density
dependent RNA polymerase enzyme of the bacteria, which values as per given equation:
is mainly responsible for the mRNA transcription. Ri- rffiffiffiffiffiffiffiffiffiffiffiffi
fampicin has been chosen as the model drug because of its q
high lipophilicity as the lipophilic drugs show more reten- Dae ¼ DV0:5 · (1)
q1Xk
tion in the lung tissues, which is desirable for local targeting.
A variety of formulations were prepared with rifampicin, where q1 = 1 g cm-3 and q is the tapped density, Dae and
such as nanoparticles or microparticles, for its localization DV0.5 are the aerodynamic and volume mean diameters and
into the lungs. However, the aerosolized form of rifampicin k is the dynamic shape factor.
with the carriers has not been much addressed until date,
which may in future serve to be a promising formulation for
Preformulation studies
locally targeting the drug into the lungs.(12–14)
Many researchers are developing nanoparticles, lipo- Drug excipient compatibility studies by Fourier transform
somes etc. to deliver the drug at the target site in the lungs. infrared spectroscopy. In DPI formulation, drug remains in
However, these formulations have limitations of poor drug intimate contact with the carriers, which could affect the
entrapment, poor stability, and high cost.(15,16) Considering stability of the drug. Hence, the knowledge of drug–
the developing countries, where patients cannot afford higher excipient interactions becomes necessary. The drug–lactose
cost of treatment, low-cost conventional dry-powder inhaler compatibility studies were done using Fourier transform
LACTOSE-BASED DRY-POWDER INHALER OF RIFAMPICIN 3

infrared spectroscopy (FTIR) ( Jasco FTIR 6100, Japan). Table 1. Composition of Batches as Per 32
The powder blend and KBr were thoroughly mixed in 1:1 Full Factorial Design
ratio and FTIR spectra were recorded in the range of 4000–
400 cm-1 after 72-hour storage at room temperature. Independent variables
X2 (mg)
Interference study of capsule shell and lactose. The
coarse and fine lactose blend (placebo) was filled in a rotacap X1 (Amount Coarse Fine
Batch code of drug in mg) lactose lactose
and was dissolved in the solvent that is, water:acetonitrile
(55:45). Similarly, the blend of the drug and the lactose filled A1 50 97.5 2.5
in another rotacap was also dissolved in the same solvent. A2 50 95.0 5.0
Then the analysis was done using high performance liquid A3 50 92.5 7.5
chromatography (HPLC) ( Jasco PU-2080 Plus, Japan). The B1 100 97.5 2.5
column used was Inertsil ODS 3 V (250 · 4.6 mm · 5 l). The B2 100 95.0 5.0
mobile phase comprising of isocratic solution of buffer pH B3 100 92.5 7.5
2.27:acetonitrile in the ratio of 55:45 was eluted at a flow C1 150 97.5 2.5
rate of 1 mL/min at 230 nm. C2 150 95.0 5.0
C3 150 92.5 7.5
Formulation development
Preliminary studies. Numerous preliminary batches were bulk and tapped densities were calculated. The experiment
prepared with different coarse lactose grades such as Inhalac was done in triplicate. The flow properties of the powder were
120, Inhalac 230, Respitose SV010, Respitose SV 003, Re- further determined by calculating Hausner ratio as below:
spitose ML 001, and fine lactose grades such as Inhalac 250,
Inhalac 400, Respitose ML006, and Lactohale LH 230 and Tapped density
Hausner ratio ¼ (2)
were evaluated for flow properties, emitted dose (ED), fine Bulk density
particle dose (FPD), content uniformity, and blend uniformity.
Among the devices, functionality of Rotahaler and Revolizer
were evaluated and the selection was done on the basis of the Blend uniformity. An accurately weighed quantity of the
minimum amount of drug left in the capsule and device. blend was evaluated for drug content. The blend was dis-
solved in 100 mL of the solvent and was allowed to sonicate
Preparation of rifampicin DPI. Preblends of Inhalac 230 for 10 minutes with intermittent shaking until it dissolved
and Inhalac 400 were prepared by modifications done in the completely. The drug content was determined using HPLC
reported method.(21) The fine lactose was first added to half after filtration through a 0.45-lm syringe filter. The analysis
the quantity of the coarse lactose and mixed for 2 minutes. was done in triplicate.
Furthermore, the remaining amount of coarse lactose was
added and mixed for 2 minutes. Then the blend was mixed Content uniformity. Content from 10 capsules of each
thoroughly in a vial for 10 minutes using vortex mixer (Eie formulation were separately dissolved in about 100 mL of
Instruments Pvt. Ltd., Ahmedabad). After storage of this phosphate buffer pH 7.4 and were sonicated for 10 minutes.
preblend for24 hours, micronized drug was added to the Then samples were analyzed after suitable dilution as per
preblend and then the above mixture was mixed using vortex method described in blend uniformity. Homogeneity of the
mixer for 15 minutes. The prepared mixture equivalent to sample was considered if not more than one capsule content
50 mg rifampicin was filled in the rotacaps. All the batches was outside the limits 85%–115% of the mean rifampicin
were analyzed for various parameters like flow properties, content and none was outside 75%–125% of the mean ri-
blend uniformity, content uniformity, ED, and FPD. fampicin content. The analysis was done in triplicate.

Optimization of rifampicin DPI. The 32 full factorial ex- Determination of device resistance. The device was
perimental design was applied for the optimization of var- loaded with an empty capsule and attached to dosage unit
ious formulations and process parameters for the preparation sampling apparatus, which was further connected to a high-
of the DPI as per Table 1. The amount of drug (X1) and capacity vacuum pump (Erweka DFM 2000) through a critical
amount of fine lactose (X2) were selected as the independent flow controller (Copley Scientific Limited, Model ADL/579).
variables. Each factor was set to three levels. Nine formu- At each pressure drop value, the flow rate was recorded.(23,24)
lations were prepared according to the design. Flow prop- Device resistance was determined by the formula:
erties, density, ED, FPD, blend uniformity, and content pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
uniformity were taken as response parameters. Design ex- Pressure drop
Device resistance ¼ (3)
pert 7.0.0 software was used for data treatment and design Flow rate
space generation.
Mass balance and in vitro deposition studies. Before
Characterization
carrying out the in vitro deposition studies, it is necessary to
Flow properties of the micronized drug and the blend. determine the mass balance, which means the balance be-
The flow properties were determined by the method as per tween entry and exit of a material in a system. Mass balance
USP.(22) After determination of bulk and tapped volume, the in terms of Andersen Cascade Impactor (ACI) refers to the
4 RAWAL ET AL.

Table 2. Particle size Distribution of Drug, 5 lm. The FPF was determined by using Copley Inhaler
Coarse and Fine Lactose Testing Data Analysis Software (CITDAS, Version 3.10).
All the experiments were performed in triplicate.
Particle size distribution
Lactose grades d0.1 (lm) d0.5 (lm) d0.9 (lm) Antimicrobial activity. Antimicrobial studies were con-
ducted using agar well diffusion method to determine the
Rifampicin 1.09 3.65 9.67 retention of rifampicin activity in DPI formulation. As Ba-
Coarse lactose 45.00 90.00 130.00 cillus subtilis shows high susceptibility to rifampicin, B.
(Inhalac 230) subtilis ATCC 6633 strain was taken as an indicator mi-
Fine lactose 1.20 7.50 25.00
(Inhalac 400) croorganism on the nutrient agar medium.(27) The wells
were filled with different concentrations of rifampicin DPI
solutions (0.25 and 0.5 mg/mL) and pure rifampicin solution
total mass of active pharmaceutical ingredient recovered (0.5 mg/mL). Plates were incubated at 37C for 24 hours to
from an impactor.(25) After the determination of mass bal- determine the zone of inhibition.
ance, the in vitro deposition studies were performed by an
eight-stage ACI (Thermo Fisher Scientific) containing Scanning electron microscopy. The surface morphology
stainless steel collection plates and fine particle fraction of micronized rifampicin, the combination of coarse and fine
(FPF) and mass median aerodynamic diameter (MMAD) lactose, and the DPI formulation was observed using the
were determined. For imitating the throat of a human, throat LEO 1530 Scanning Electron Microscopy (SEM) (Oberko-
piece referred to as ‘‘USP throat’’(26) was used. It was chen, Germany). The plasma deposition method was used
connected to the top of the ACI. The mouthpiece adaptor employing a gold coating unit to make the sample surface
was fabricated as per Rotahaler design. The induction port conductive to the scanning electron beam. The gold coating
and preseparator were used to complete the impactor for of about 40–100 A was done under the argon atmosphere.
DPI. The preseparator was filled with 15 mL of solvent. The The SEM of gold-coated samples was carried out at high
rotacap filled with the powder blend equivalent to 50 mg vacuum with specimen working distance of 17.4 mm and an
rifampicin was further loaded into the Rotahaler. Inspira- accelerating voltage of 30 kV.
tion of single dose of formulation (n = 3) was done for 8
seconds at a flow rate of 30 L/min. Cell viability assay. Cytotoxicity was evaluated using
After performing the experiment, the particles on each murine macrophages J774 cell lines grown in Dulbecco’s
stage were washed with the HPLC solvent (water:acetonitrile modified Eagle’s medium, supplemented with 10% heat-
in the ratio 55:45) and volume was made up to 20 mL using a inactivated fetal bovine serum with 5% CO2. Cells were
calibrated standard volumetric flask. Rifampicin content was seeded in 96-well plates at a density of 1 · 105 cells/well and
determined using HPLC. The cumulative percentage of dose incubated overnight at 37C to allow cells to adhere. The
from stages 2 to 5 depicts the FPF of the formulation which medium was then removed and cells were incubated with
represents the cutoff diameter of the particles less than 180 lL of the fresh medium containing two rifampicin

FIG. 1. FTIR spectra of (a) pure rifampicin and (b) overlay spectra of rifampicin and its
mixture with lactose. FTIR, Fourier transform infrared spectroscopy.
LACTOSE-BASED DRY-POWDER INHALER OF RIFAMPICIN 5

samples (pure rifampicin and dry-powder formulation) at Table 4. Criterion for Device Resistance Selection
0.125, 0.25, 0.5 mg/mL concentration for 6, 12, and
24 hours. At these time points, 20 lL of Cell Titer Blue Level of
Reagent (Promega) was added to each well. Then cells were Resistance values resistance
incubated for another 2 hours at 37C, to reduce resazurin to 0.020–0.022 Low
resorufin. Fluorescence was measured at 540 nm excitation 0.023–0.034 Medium
and 590 nm emission. The percentage of cell survival was Above 0.034 High
determined in comparison to the control wells. The experi-
ment was performed in triplicate.(27,28)
of the extracted lungs of DPI-treated animals in comparison to
marketed formulation-treated group was carried out.
In vivo pulmonary pharmacokinetic studies. Lungs were
homogenized in phosphate-buffered saline (PBS) and ace- Stability studies. The selected formulation was subjected
tonitrile was added with vortex mixing for protein precipi- to stability studies as per ICH guideline at room temperature
tation. The standard curve was plotted by spiking known (25 – 2C/60% – 5% relative humidity (RH)) and acceler-
amounts of rifampicin for the analysis of extracted lungs. ated conditions (40 – 2C/75% – 5% RH) up to 6 months.
Animal protocols were approved by the Institutional Animal The blend uniformity values were determined after 3 and 6
Ethics Committee, Nirma University. The in vivo pulmonary months. The studies were done in triplicate.
pharmacokinetic studies of DPI formulation were compared
with the marketed formulation of rifampicin to investigate
Statistical analysis. Each batch of animals was prepared
the local action of dry-powder formulation. Two groups of
three times, and data values were expressed as mean – stan-
18 male Wistar rats (220–250 g) each were used for the
dard deviation. A difference between means was considered
study. Rats were placed in individual cages at constant
significant, if p-value is equal to or less than 0.05.
temperature and were given free access to water ad libitum.
Rats were kept on fasting overnight before experimentation,
mouth of rats of Group I (conventional DPI) were sealed, Results and Discussion
and animals were allowed to inhale about 313.56 mg/Kg of DPI is widely used to deliver the drug to the lungs and is
formulation. formulated using various carriers like lactose, mannitol,
The second group animals were orally administered the trehalose etc. Among all, lactose is widely used due to its
marketed formulation in solution form (1 mg/mL) through well-explored toxicity profile, physical and chemical sta-
oral cannula. Three animals from both groups were eutha- bility, compatibility profile, ease of availability, and low
nized at 1, 3, 6, 9, 12, and 24 hours after administration. The price.(30) The fine particles of the drug adhere onto the
extracted lungs from experimental rats were homogenized in surface of coarse lactose and upon inhalation, coarse lactose
10 mL PBS. The lung tissues were deproteinized by the remains in the upper respiratory tract, whereas the fine
acetonitrile treatment and rifampicin released in the super- particles of the drug enters deep into the lungs. However, it
natant was analyzed by HPLC. The mean pulmonary phar- is difficult for a complete dose of the drug to detach from the
macokinetic parameters Cmax, tmax, AUC0–24, and t1/2 were coarse lactose, which leads to dose inaccuracy.
calculated as reported elsewhere.(29) It is reported that the coarse lactose should be mixed with
small quantities of fine lactose so that fine lactose occupy
the high adhesion sites (active sites) on the surface of coarse
In vivo toxicity studies. One animal from Group I was
lactose.(31–36) This preblend was then mixed with the fine
allowed to inhale the formulation after sealing its mouth. particulate drug. This results in binding of the drug on to the
Twenty-four hours postinhalation, lungs were extracted and low adhesion sites (less active sites) on the coarse lactose
were stored in 10% formalin solution to conduct histopatho- which facilitates an easy detachment of the drug upon in-
logical studies to determine toxic effects. The histopathology halation. In the present study, DPI of rifampicin was de-
veloped using various combinations of drug, coarse and fine
Table 3. Blend Uniformity and Content lactose grades.
Uniformity of Design Batches
Particle size distribution and aerodynamic
Blend Content diameter of rifampicin
Batch code uniformity (%) uniformity (%)
In DPI, the particle size of the drug and the carrier play a
A1 106.8 – 2.03 101.75 – 3.63 vital role in delivering the drug to the target site and hence the
A2 95.85 – 2.08 94.69 – 1.34
A3 90.40 – 2.13 91.53 – 2.60 Table 5. Device Resistance Determination
B1 93.70 – 1.03 93.57 – 3.76
B2 89.20 – 2.65 92.92 – 2.89 Pressure Flow rate Device
B3 85.20 – 2.82 88.56 – 2.63 (KPa) (L/min) resistance
C1 91.60 – 4.03 93.21 – 5.67
C2 90.50 – 3.81 91.02 – 4.35 1 30.00 0.0333
C3 88.80 – 5.03 88.90 – 2.38 2 41.90 0.0340
3 50.67 0.0337
All values are expressed as mean – SD; n = 10. 4 60.00 0.0337
SD, standard deviation.
6 RAWAL ET AL.

Table 6. Parameters Measured Using Andersen Cascade Impactor for Design Batches
Mass median Geometrical
Fine particle aerodynamic diameter standard
Batch code Mass balance (%) fraction (FPF) (%) (MMAD) (lm) deviation (GSD)
A1 106.50 – 0.14 22.60 – 0.87 5.70 – 0.23 2.10 – 0.03
A2 110.00 – 0.18 23.50 – 1.17 4.50 – 0.06 2.10 – 0.03
A3 110.80 – 0.26 22.90 – 0.10 5.60 – 0.06 2.10 – 0.03
B1 112.40 – 0.39 28.90 – 0.49 5.10 – 0.23 2.10 – 0.04
B2 112.50 – 0.21 27.60 – 0.31 4.50 – 0.17 2.00 – 0.03
B3 113.00 – 0.52 26.80 – 0.36 4.10 – 0.26 2.00 – 0.03
C1 109.30 – 0.98 25.80 – 0.56 4.70 – 0.06 2.10 – 0.05
C2 109.10 – 1.11 23.90 – 1.57 4.10 – 0.25 2.00 – 0.05
C3 108.00 – 1.09 23.80 – 2.35 5.10 – 0.26 2.00 – 0.07
All values are expressed as mean – SD; n = 3.
FPF, fine particle fraction

efficacy of formulation. The particle size distribution of the observed in the case of the capsule containing the blend of
micronized drug was determined by Malvern Mastersizer. drug and lactose. It revealed that interference was not ex-
The particle size distribution of rifampicin and grades of hibited by the capsule shell and lactose.
coarse and fine lactose are presented in Table 2. In spite of the
particle size distribution, aerodynamic diameter can be better Preliminary trials
correlated with the drug deposition in lungs. It is reported that
Several preliminary batches were prepared using various
the aerodynamic diameter in the range of 1–5 lm is desirable
ratios of coarse lactose grades like Inhalac 120, Inhalac 230,
for the drug particles to get deposited deep into the lungs. The
Respitose SV010, Respitose SV 003, Respitose ML 001,
aerodynamic diameter calculated for micronized rifampicin
and fine lactose grades such as Inhalac 250, Inhalac 400,
was 3.325 lm, which is within the required range.
Respitose ML006, and Lactohale LH 230. Out of these,
Inhalac 230 was selected as coarse lactose in combination
Preformulation studies with Inhalac 400, a fine lactose grade based on their good
Drug excipient compatibility studies. Drug and excipient flow properties. So this combination was chosen as the
compatibility studies were conducted by FTIR. The FTIR carrier for further optimization.
spectra of rifampicin and mixture of rifampicin and lactose
have been shown in Figure 1, which indicated the presence of Optimization
identical peaks (3333, 2935, 1726, and 1649 cm-1) in both Quality by design (QbD) is a well-known concept being
rifampicin as well as its mixture with the excipients thus used in many industries to build quality in the product. The
showing the compatibility of the drug and the excipients. USFDA made it mandatory to use these concepts in phar-
maceutical formulation development and design of ex-
Interference study of capsule shell and lactose. The re- periments (DOE) is one of the tools of QbD. However, QbD
ported HPLC method was used with some modifications.(37) is a newer concept, but researchers have been using DOE
Chromatogram of the capsule containing lactose was a since long specially for formulation development. Various
straight line, whereas the characteristic peak of the drug was optimization designs can be used based on the outcome of

FIG. 2. Comparative in vitro deposition studies of the design and the optimized batches.
LACTOSE-BASED DRY-POWDER INHALER OF RIFAMPICIN 7

FIG. 3. Overlay plot showing design space.

the preliminary trials. There are a few excipients required in X1 and amount of fine lactose as X2. Nine batches were pre-
the DPI, and in the present investigation, we have observed pared and evaluated extensively for following parameters.
that the amount of drug, fine lactose, and coarse lactose were
important factors affecting the performance of the DPI. Hence
Flow properties of the micronized drug and the blend.
32 full factorial design was used taking the amount of drug as
The micronized rifampicin showed higher value of the
Hausner ratio (1.58), whereas the design batches showed the
Hausner ratio between 1.21–1.25, except batch C3. As per the
Chapter 1174, powder flow of USP [16], the micronized ri-
fampicin, can be characterized in the category of ‘‘exceed-
ingly poor flow’’ and the design batches in the ‘‘fair flow’’
properties, except batch C3, which showed the Hausner ratio
of 1.33 depicting the ‘‘poor flow.’’ This might be due to the
presence of more amounts of the drug and the fine lactose.
Large amounts of fine lactose might have bound to both high
as well as low adhesion sites of the coarse lactose, thereby not
allowing the rifampicin to bind to the coarse lactose. This
might have led to the poor flow of the blend. However, the
developed formulation fulfilled the requirement of minimum
flow properties for filling the blend into the capsule shell.

Blend uniformity. As per Chapter 601, aerosols, nasal

sprays, metered-dose inhalers, and DPIs of USP,(20) the drug
content for a DPI must be within 85%–115% of the label
claim. The results have been shown in Table 3. All the
batches were found to be within the assay limits. However,
the batches A3, B3, and C3 containing higher proportion of
fine lactose showed poor blend uniformity. This might be
due to the fact that as the fine lactose was increased, it does
not allow the drug to properly bind to the coarse lactose.
Additionally, these batches showed poor flow properties,
which might also contribute to the poor blend uniformity.
The complete polynomial regression model for the blend
uniformity of the design batches was described by the fol-
lowing equation:
FIG. 4. In vitro antimicrobial studies of pure rifampicin
and developed DPI formulation on Bacillus subtilis ATCC YBU ¼ 88:76 þ 3:69X1  4:61X2 þ 3:40X1 X2
6633. The representative experiment is a mean of three in- (4)
dependent experiments. DPI, dry-powder inhaler. þ 4:62 X21 þ 0:90X22
8 RAWAL ET AL.

FIG. 5. Analysis of optimized formulation through SEM, (a) pure rifampicin, and (b)
developed DPI formulation. SEM, scanning electron microscopy.

The above equation shows that the amount of drug (X1) A3, B3, and C3, the average value of the content uniformity
had a positive impact and the amount of fine lactose (X2) had was low. This indicates that increase in the proportion of the
a negative impact on the blend uniformity. As percent of the fine lactose negatively affected the content uniformity.
drug in the total content increased, the lower amount of it
adhere to the capsule shell as well as the device thus resulting Device resistance calculation. The device resistance is a
in the increase in blend uniformity. The increase in the fine parameter which is inversely proportional to the flow rate. If
lactose leads to poor flow properties, which might be re- the device resistance is less, the flow rate required will be
sponsible for its negative effect on the blend uniformity. more. Since in tuberculosis, the inspiratory flow rate of a
patient reduces, so in this perspective a device with medium
Content uniformity. As per the USP, the amount of to high resistance is required. The criterion for the device
active pharmaceutical ingredient should not be outside resistance has been shown in Table 4. The device used in the
80%–120% of the label claim for more than one of the 10 present work was Rotahaler. Rotahaler was operated at
capsules, none of the capsules should be outside 75%–125% different pressure values so as to set different flow rates as
of the label claim, and the average should not be outside shown in Table 5. The device resistance at different flow
85%–115% of the label claim. The average of the ten cap- rates varied from 0.033 to 0.034, which indicated that Ro-
sules of each batch was found to be within 85%–115% of tahaler fall in the upper border of medium-level resistance.
the label claim (Table 3). But again, in the case of batches The operating flow rate used for the developed rifampi-
cin DPI for further ACI analysis was 30 L/min, which was
achieved at 1 KPa pressure.

In vitro deposition studies by Andersen cascade impac-

tor. Before carrying out the in vitro deposition, the mass
balance was determined.(38) The mass balance within the
limits of 85%–115% is acceptable. All the batches showed a
mass balance of around 100%–115%, thereby revealing no
drug loss in the system (Table 6). Furthermore, for a drug to
act deep into the lungs, proper deposition is required. ACI

FIG. 6. Cell viability studies of pure rifampicin and de-

veloped DPI formulation on macrophage J774 cell lines. FIG. 7. Comparative in vivo pulmonary pharmacokinetic
The representative experiment is a mean of three indepen- studies of developed DPI formulation and marketed for-
dent experiments. mulation.
LACTOSE-BASED DRY-POWDER INHALER OF RIFAMPICIN 9

FIG. 8. Photomicrographs of rat lungs showing histopathological changes, (A) Control,

(B) developed DPI formulation, (C) Marketed formulation.

serves as one of the best tool to perform in vitro deposition Scanning electron microscopy. The SEM images
studies due to its several merits. First, it serves as a very (Fig. 5) show that the fine lactose has occupied most of the
effective device for the determination of mass median active sites of the coarse lactose (A). In the dry-powder
aerodynamic diameter (MMAD) of the inhalation formula- formulation (B), after the adherence of fine lactose to the
tion, which is a key factor influencing deposition of the drug active sites, remaining sites had been covered by the drug.
in various regions of the lungs and the respiratory tract. Hence, it can be depicted from the SEM images that upon
Second, ACI chiefly determines the fraction of the drug inhalation, the drug will be easily detached from the lactose
which gets deposited deep in the lungs (also known as FPF), and will go deep into the lungs.
thus providing the best resolution for the particles in the
desired size range for DPI, that is, 1–5 lm. ACI funda-
mentally functions on the principle of inertial impaction Cell viability assay. The drug delivery to the lungs for
(separation based on the inertial differences, which is a tuberculosis is to have a local targeting of the drug for the
function of particle size and velocity). It basically consists improved clinical outcome. In that context, the dose for the
of eight stages with a cutoff diameter between 0.7 and lung delivery can be directly taken up considering the dose per
10 lm.(39) The particles of the desired size range of 1–5 lm kilogram body weight to the organ weight, which is 10 mg/kg
reside in the stages 2–5 of the ACI, as the cutoff diameter of in the case of rifampicin. Since the weight of the human lungs
these stages lie between 1.1 and 4.7 lm. The results of FPF is *1 kg, considering 20% of the drug will reach the target
and MMAD values for all the batches have been shown in site, 50 mg of the drug can be taken to formulate DPI. This
Table 6. The in vitro deposition of all the DPI formulation would help to attain local minimum inhibitory concentration
from stages 2 to 5 has been shown in Figure 2 and the overlay over24 hours. However, lung clearance needs to be studied.
plot showing design space has been shown in Figure 3. The To check the safety of the formulation, % cell viability studies
polynomial equation obtained for the FPF was: were performed using macrophage J774 cell lines. The cyto-
toxicity test results, expressed as % cell viability (survival) v/s
incubation time, have been shown in Figure 6. The observed
YFPF ¼ 18:29 þ 2:06X1 þ 0:99X2  0:35 X1 X2
(5) effect of dry-powder formulation on cell viability was similar
 3:01X12 þ 0:32X22 as that of the pure rifampicin. Negligible cytotoxicity with
about 75%–80% of cell viability was observed after 6 and
The above equation shows that both factors, that is, the 12 hours exposure of 0.125 mg/mL dose.
amount of drug and amount of fine lactose had a positive However, higher concentrations, that is, 0.25 and 0.5 mg/
impact on the FPF. This might be due to the fact that as the mL, showed some toxicity. Considering the slow drug re-
amount of fine lactose increased, it bound effectively with lease profile and volume of the fluid available in the lungs,
the active sites of the coarse lactose, and thus preventing the
binding of the drug to such sites. This might lead to ease of
detachment of the drug from the coarse lactose leading to Table 7. Stability Studies of the Optimized Batch
high FPF. Blend
Time Study Conditions uniformity
Antimicrobial activity. Next, we decided to determine Initial 109.10 – 0.39
antimicrobial activity of rifampicin–DPI formulation by an 3 months Long-term 25C – 2C/ 107.80 – 1.13
agar well diffusion method. Our results show nearly the study 60% – 5% RH
same antimicrobial activity as that of pure rifampicin. The 6 months 106.20 – 0.74
zone of inhibition estimated by pure rifampicin solution, 0.5 Initial 109.10 – 0.39
and 0.25 mg/mL of rifampicin DPI were calculated to be 3 months Accelerated 40C – 2C/ 102.80 – 0.95
20.4 – 0.67, 18.4 – 0.41, and 17.2 – 0.20 mm, respectively. In study 75% – 5% RH
6 months 99.80 – 0.74
brief, our results indicate that rifampicin antimicrobial ac-
tivity remained intact when delivered through DPI formula- All values are expressed as mean – SD; n = 3.
tion (Fig. 4). RH, relative humidity.
10 RAWAL ET AL.

we can assume that the maximum concentration of rifam- No signs of severe toxicity were seen in the histopatholog-
picin attained in lungs would be safe. Furthermore, con- ical studies. Thus, from the above studies it can be con-
sidering the higher number of human alveolar macrophages cluded that rifampicin DPI can serve as a safe formulation to
dispersed on the total surface area of alveoli compared with provide better local drug concentration in the lungs in
that of the cell culture,(40) the formulation can be considered comparison to the marketed formulations.
as noncytotoxic. However, in vivo pulmonary pharmacoki-
netic studies were carried out to measure rifampicin concen-
tration in the lungs. Acknowledgments
The authors would like to thank the Institute of Pharmacy,
In vivo pulmonary pharmacokinetic studies. The primary Nirma University, Ahmedabad, for providing them all the
objective of the in vivo studies was to demonstrate drug facilities to carry out their research work. The authors ac-
concentrations in the lungs after administration of developed knowledge the Department of Science and Technology,
DPI in comparison with oral marketed product. Adminis- Government of India for financial support vide reference no.
tration of dose equivalent to 50 mg human dose in the form SR/WOS-A/LS-1119/2015 (G) under Women Scientist
of DPI gave higher drug concentration in lungs in compar- Scheme. The authors kindly thank Zydus Cadila Healthcare
ison to 600 mg human dose of marketed product. Thus, we Ltd., Ahmedabad, for granting them permission to work on
can assume a lower concentration of drug in blood after the Andersen Cascade Impactor and Lupin Ltd. Aurangabad
use of DPI.(25,41) for providing them rifampicin as gift sample. They would
However, an additional blood analysis could have been also thank Dr. Rajeev Tyagi and Mr. Rajesh Parmar for
done to understand the safety of the formulations. The helping them in carrying out pulmonary pharmacokinetic
comparative drug concentration in the lungs and time pro- studies. They would also like to thank Meggle Germany
filing of all the formulations tested are shown (Fig. 7). A and DFE Pharma for providing them various inhalable
marked increase of 1.5-fold was seen in Cmax of Group I lactose grades.
treated with the developed DPI formulation (970.76 –
60.83 lg/mL) in comparison to Group II treated with mar-
keted formulation (677.03 – 47.61 lg/mL). Furthermore, Author Disclosure Statement
marked increment of t1/2 and AUC0–24 was seen in Group I No competing financial interests exist. The authors are
(10.26 – 1.15 and 8783.33 – 100.66) when compared with responsible for the overall content and writing of the article.
Group II (3.33 – 0.58 and 4140.51 – 84.77). The increased t1/
2 in Group I may be due to improved local retention of
rifampicin in the form of conventional DPI, which indicates References
the potential use of this lung-targeted DPI formulation to 1. Du Toit LC, Pillay V, and Danckwerts MP: Tuberculosis
deliver rifampicin for an effective pulmonary administration. chemotherapy: Current drug delivery approaches. Respir
Res. 2006;7:118.
In vivo toxicity studies. The administration of conven- 2. Alamelu R: Immunology of tuberculosis. Indian J Med Res.
tional DPI (Group-I) produced mild changes in the pathol- 2004;120:213–232.
ogy of treated lungs (Fig. 8B) in comparison to normal lung 3. Sharma SK, and Mohan A: From an incurable scourage to a
(Fig. 8A) showing hardly any toxicity. Whereas Group-II curable disease-journey over a millenium. Tuberculosis.
administered with the marketed formulation showed severe 2013;137:455–493.
toxicity in peribronchial infiltration of inflammatory cells, 4. Fogel N: A disease without boundaries. Tuberculosis.
hyperplasia of lymphoid tissue, and interalveolar septal 2015;95:527–531.
thickening (Fig. 8C). Thus, it is inferred from our results 5. Chan ED, and Iseman MD: Current medical treatment for
that local targeting of rifampicin through conventional DPI tuberculosis. BMJ. 2002;325:1282–1286.
is nontoxic and safe to use. However, repeated dose toxicity 6. Telko MJ, and Hickey AJ: Dry powder inhaler formulation.
Respir Care. 2005;50:1209–1227.
studies are indeed required to assess the long-term safety of
7. Sia IG, and Wieland ML: Current concepts in the manage-
the formulation.
ment of tuberculosis. Mayo Clin Proc. 2011;86:348–361.
8. Frijlink H, and De Boer AH: Dry powder inhalers for
Stability studies. The results for the stability studies of pulmonary drug delivery. Expert Opin Drug Deliv. 2004;1:
the formulation are shown in Table 7. It can be concluded 67–86.
from the studies that the developed DPI was stable even after 9. Paranjpe M, and Goymann CCM: Nanoparticle mediated
6 months as the blend uniformity was within the limit (85%– pulmonary drug delivery: A review. Int J Mol Sci. 2014;15:
115%) at room temperature as well as accelerated conditions. 5852–5873.
10. Dal Negro RW: Dry powder inhalers and the right things to
Conclusion remember: A concept review. MultidiscipRespir Med. 2015;
10:13.
The rifampicin DPI formulation was developed system- 11. Pilcer G, Wauthoz N, and Amighi K: Lactose characteris-
atically and evaluated extensively. The optimized formula- tics and the generation of the aerosol. Adv Drug Deliv Rev.
tion of DPI when evaluated using Rotahaler, showed 2012;64:233–256.
desired MMAD (1–5 lm) and higher FPF (28.9%). The 12. Campbell EA, Korzheva N, Mustaev A, Murakami K, Nair
in vivo pulmonary pharmacokinetic studies showed the S, Goldfarb A, and Darst SA: Structural mechanism for
improved localized retention of rifampicin in the DPI- rifampicin inhibition of bacterial RNA polymerase. Cell.
treated group in comparison to the marketed formulation. 2000;104:901–912.
LACTOSE-BASED DRY-POWDER INHALER OF RIFAMPICIN 11

13. Sethuraman VV, and Hickey AJ: Powder properties and 30. Rahimpour Y, and Hamishekhar H: Lactose engineering for
their influence on dry powder inhaler delivery of an anti better performance in dry powder inhalers. Adv Pharm
tubercular drug. AAPS Pharm Sci Tech. 2000;3:1–10. Bull. 2012;2:183–187.
14. Son YJ, and McConville JT: A new respirable form of 31. Zeng XM, Martin GP, Tee SK, and Marriott C: The role of
rifampicin. Eur J Pharm Biopharm. 2011;78:366–376. fine particle lactose on the dispersion and deaggregation of
15. Singh D, Harikumar SL, and Nirmala: Nanoparticles: An salbutamol sulphate in an air stream in vitro. Int J Pharm.
overview. J Drug Deliv Ther. 2013;3:169–175. 1998;176:99–110.
16. Akbarzadeh A, Sadabady RR, Davaran S, Joo SW, Zar- 32. Islam N, Stewart P, Larson I, and Hartley P: Effect of
ghami N, Hanifehpour Y, Samiei M, Kouhi M, and Nejati- carrier size on the dispersion of salmeterol xinafoate from
Koshki K: Liposomes: Classification, preparation, and ap- interactive mixtures. J Pharm Sci. 2004;93:1030–1038.
plications. Nanoscale Res Lett. 2013;8:102. 33. Adi H, Larson I, Chiou H, Young P, Traini D, and Stewart
17. Malassiney P, Goeau-Brissonni£re O, Coggia M, and Pe- P: Role of agglomeration in the dispersion of salmeterol
chere J-C: Rifampicin loading of vascular grafts. J Anti- xinafoate from mixtures for inhalation with differing drug
microb Chemother. 1996;37:1121–1129. to fine lactose ratios. J Pharm Sci. 2008;97:3140–3152.
18. Ko JA, Park HJ, Hwang SJ, Park JB, and Lee JS: Preparation 34. Shur J, Harris H, Jones MD, Kaerger JS, and Price R: The
and characterization of chitosan microparticles intended for role of fines in the modification of the fluidization and
controlled drug delivery. Int J Pharm. 2002;249:165–174. dispersion mechanism within dry powder inhaler formula-
19. Bernkop-Schnurch A, and Dunnhaupt S: Chitosan-based tions. Pharm Res. 2008;25:1931–1940.
drug delivery systems. Eur J Pharm Biopharm. 2012;81: 35. Thalberg K, Berg E, and Fransson M: Modeling dispersion
463–469. of dry powders for inhalation. The concepts of total fines,
20. U.S. Pharmacopeia: USP 29 Section 601. Aerosols, Nasal cohesive energy and interaction parameters. Int J Pharm.
Sprays, Metered-Dose Inhaler, December 10. Available at: 2012;427:224–233.
[Link]/v29240/usp29nf24s0_c601_viewall 36. Kinnunen H, Hebbink G, Peters H, Shur J, and Price R: An
.html Last access: Jan. 20, 2017. investigation into the effect of fine lactose particles on the
21. Misaka S, Sato H, Yamauchi Y, Onoue S, and Yamada S: fluidization behaviour and aerosolization performance of
Novel dry powder formulation of ovalbumin for develop- carrier-based dry powder inhaler formulations. AAPS
ment of COPD-like animal model: Physicochemical char- PharmSciTech. 2014;15:898–909.
acterization and biomarker profiling in rats. Eur J Pharm 37. Viéitez MCD, Renedo MJ, Ruz N, Prieto MJB, and Calleja
Sci. 2009;37:469–476. I: High-performance liquid–chromatographic determination
22. U.S. Pharmacopoeia: USP 29 Section 1174. Powder flow, of rifampicin in plasma and tissues. J Chromatogr A. 2004;
December 10. Available at: [Link]/v29240/ 1031:289–294.
usp29nf24s0_c1174.html Last access: Jan. 20, 2017. 38. Dechraska J, Suwandecha T, Maliwan K, and Srichana T:
23. Clarke AR, and Hollingworth AM: The relationship be- The comparison of fluid dynamics parameters in an an-
tween powder inhaler resistance and peak inspiratory con- dersen cascade impactor equipped with and without a
ditions in healthy volunteers-implications for in vitro preseparator. AAPS PharmSciTech. 2014;15:792–801.
testing. J Aerosol Med. 1993;6:99–110. 39. Christopher D, Curry P, Doub W, Furnkranz K, Lavery M,
24. Hassoun M, Ho S, Muddle J, Buttini F, Parry M, Hammond Lin K, Lavery M, Lin K, Lyapustina S, Mitchell J, Rogers
M, and Forbes B: Formulating powder-device combinations B, Strickland H, Tougas T, Tsong Y, and Wyka B: Con-
of salmeterol xinafoate dry powder inhalers. Int J Pharm. siderations for the development and practice of cascade
2015;490:360–367. impaction testing including a mass balance failure inves-
25. Tougas T, Christopher D, Mitchell JP, Strickland H, Wyka tigation tree. J Aerosol Med. 2003;16:235–247.
B, Oort MV, and Lyapustina S: Improved quality control 40. Wallace WA, Gillooly M, and Lamb D: Intra-alveolar mac-
metrics for cascade impaction measurements of Orally In- rophage number in current smokers and non-smokers: Mor-
haled Drug Products (OIPs). AAPS PharmSciTech. 2009; phometric study of tissue section. Thorax. 1992;47:437–440.
10:1276–1285. 41. Jain DP, Jain PV, Bhalekar MR: Pharmacokinetic study of
26. Chrystyn H, Safioti G, Keegstra JR, and Gopalan G: Effect dry powder inhaler formulations of antiasthmatic drugs. Int
of inhalation profile and throat geometry on predicted lung J Pharm Sci Rev Res. 2013;23:319–327.
deposition of budesonide and formoterol (BF) in COPD:
An in-vitro comparison of Spiromax with Turbuhaler. Int J Received on February 18, 2017
Pharm. 2015;491:268–276. in final form, April 13, 2017
27. Maretti E, Rossi T, Bondi M, Croce MA, Hanuskova M,
Leo E, Sacchetti F, and Iannuccelli V: Inhaled solid lipid Reviewed by:
microparticles to target alveolar macrophages for tubercu- Lin Yang
losis. Int J Pharm. 2014;462:74–82. Anthony Hickey
28. Barot KP, Jain SV, Gupta N, Kremer L, Singh S, and Ta-
kale VB: Design, synthesis and docking studies of some Address correspondence to:
novel (R)-2-(40-chlorophenyl)-3-(40-nitrophenyl)-1,2,3,5 Shital Butani, PhD
tetrahydrobenzo[4,5] imidazo [1,2-c]pyrimidin-4-ol deriv- Institute of Pharmacy
atives as antitubercular agents. Eur J Med Chem. 2014; Nirma University
83:245–255. Gujarat-382481
29. Chougule M, Padhi B, and Misra A: Nano-liposomal dry Ahmedabad
powder inhaler of tacrolimus: Preparation, characterization, India
and pulmonary pharmacokinetics. Int J Nanomedicine.
2007;2:675–688. E-mail: shital_26@[Link]