Usp 44

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 9

á1228.

1ñ DRY HEAT DEPYROGENATION


INTRODUCTION
Dry heat is the method most frequently used for the depyrogenation of heat stable
materials. Dry heat depyrogenation is
dependent upon two parameters: time and temperature, which equates to a thermal input.
As a result, dry heat depyrogenation
processes can be easily monitored/controlled and are highly reproducible.
Depyrogenation processes typically operate at a
range of temperatures from approximately 170° up to about 400°.
The most prevalent pyrogenic agents in parental manufacturing that are of concern
relative to patient safety are bacterial
endotoxins, found in the outer cell walls of Gram-negative bacteria. The destruction of
bacterial endotoxins (depyrogenation)
by dry heat has been studied extensively and has been shown to follow first order
kinetics. The well-defined kinetics of
inactivation makes it possible to predict the efficacy of dry heat processes operating at
different times and temperatures by
understanding the total thermal input (FD).
The range of temperatures used for dry heat depyrogenation overlaps the upper range of
temperatures used for dry heat
sterilization (see Dry Heat Sterilization á1229.8ñ). This is because bacterial endotoxins
are more resistant to the effects of dry heat
than the most heat-resistant bacterial spores. This chapter provides an overview of the
process of dry heat depyrogenation, its
control, and validation.
TECHNOLOGIES USED FOR DEPYROGENATION BY DRY HEAT
Although all dry heat depyrogenation processes rely strictly on time of exposure and
temperature to assure effectiveness,
the equipment used typically falls into two categories: the dry heat “batch” oven and
continuous tunnel systems. Batch ovens
are routinely used for the depyrogenation of product containers, most often glass, but also
other heat stable product contact
parts or laboratory equipment. Continuous tunnels, on the other hand, are used primarily
to depyrogenate glass product
containers.
Batch Ovens
Circulating heated air is used to heat the load items, which may be individually covered
or wrapped in a material that is
unaffected by the temperature used, or placed in a lidded container for protection during
pre- and post-process handling. When
depyrogenation and sterilization are to be achieved in the same process, air supplied to
the oven is passed through one or more
high efficiency particulate air (HEPA) filter(s) to maintain sterility within the oven after
completion of the dwell period. These
forced air ovens typically operate at a positive air pressure differential relative to the
surrounding room. This design results in
particulate air quality that can meet ISO 5 requirements to reduce particulate matter and
microbial contamination risk
throughout processing.
In order to ensure sufficient lethality and process control, oven control probe(s) must
maintain a predefined temperature
for a predefined time period prior to cooling. The limited heat transfer capacity of air
requires that items in the oven be placed
in fixed locations confirmed acceptable during the cycle development/validation effort.
Caution should be exercised in defining
variable load patterns as minimum load sizes may result in inadvertent slower heating of
the load and greater temperature
variability. Smaller facilities may use a single door oven, but the principles of operation
and validation are the same as with
larger double door production units. The important batch oven process variables are set-
point temperature, duration of dwell
period, load type and configuration, airflow characteristics, and container size.
Continuous Tunnels
The use of tunnels for dry heat depyrogenation of glass containers on a moving conveyor
allows for substantially higher
throughput and packing densities than the batch process, reduces handling, and is
ordinarily integrated with a washing and
filling system. Tunnels typically use forced heated air systems or radiant IR systems that
recirculate air through a battery of HEPA
filters. Load items in tunnels are typically fed directly from an integrated container
washing system.
Depyrogenation tunnels have separate zones for heating and cooling, allowing for
continuous in-feed and discharge at
temperatures appropriate for production purposes. The tunnel is maintained at constant
airflow and temperature conditions
during use, and as glass passes through the tunnel it is heated to depyrogenating
temperatures and cooled before exiting.
Although the conditions within the tunnel are essentially constant and well controlled, the
temperature of the glass as it passes
through the tunnel on the conveyor will change with its location. Dwell time is controlled
by adjusting the conveyor speed,1
which in the depyrogenation tunnel is the process parameter that governs exposure time.
The air in the tunnel is most commonly heated using electrical coils but other heat
sources, such as infrared or high-pressure
steam, have been used. For energy conservation, heated air in depyrogenation tunnels is
often recirculated. The important
continuous tunnel process variables are heating zone temperature, cooling zone
temperature, belt speed, and container mass
per unit.
DRY HEAT DEPYROGENATION FUNDAMENTALS
Dry heat depyrogenation uses air first to heat and then to cool the items. The limited heat
capacity of dry air results in
relatively slow heating and cooling of the load items. Variability in temperature
distribution in dry heat ovens and tunnels is
typically much higher than that observed in moist heat systems. The limited heat capacity
of air requires that items in ovens be
placed in the same locations as confirmed acceptable in the cycle development/validation
effort. Packing and thermal mass will
also play critical roles in temperature management. Caution must also be exercised with
varying load mass and distribution as
in some instances (resulting from oven design, air flow characteristics, and control probe
position) minimum load sizes may
result in process variability.
DEPYROGENATION PROCESS CONTROL
Process temperatures in dry heat depyrogenation are controlled by calibrated temperature
sensors placed at specific locations
within the equipment. The exposure portion of the process is designed to attain a
minimum dwell time at a predefined minimum
temperature ensuring that depyrogenation conditions are adequately uniform. The defined
dwell time is determined by using
measurement devices (e.g., thermocouples) directly in contact with the items to be
depyrogenated during development. The
inactivation of bacterial endotoxins by dry heat involves the control of only two
parameters: time and temperature.
The simplicity of process control for these parameters provides a predictable
depyrogenation effect. Once validated, a clear
understanding of the inputs to the system, meaning resident endotoxin load on incoming
materials, and outputs of the system,
meaning reduction of resident levels of endotoxin to safe levels, is more meaningful than
spiking with unnaturally high levels
of challenge material and looking for a prescriptive requirement for log reduction. The
dosimetric measurement for dry heat
depyrogenation processes is the FD unit. An FD = 1 is defined as the depyrogenation
effect achieved by 1 min of heating at 250°.
The FD-value enables the integration of temperature over the process duration (time). By
convention, the rate at which
depyrogenation destruction rate (D-value) varies as a function of temperature change is
defined as the z-value. The z-value for
dry heat depyrogenation has been shown to be in the range of 45°–55°. For the purposes
of this chapter, 50° is used as a
standard z-value. Other values may be used. (1,2)
The FD approach is used as a means to compare dry heat depyrogenation effects
produced by processes that operate at
varying temperatures. Basic mathematics can be used to calculate the depyrogenation
effect produced at temperatures other
than 250° to determine equivalence to that provided at 250°.
Using a reference temperature of 250° and an assumed z-value of 50°, the FD calculation
can be determined:

FD
= accumulated destruction
t1 = process start time
t2
= process end time
T = temperature at each time increment
Δt = time interval between temperature measurements
Summing the instantaneous temperature contributions over the entire depyrogenation
process allows for the calculation of
the overall process efficacy or FD delivered over the course of the process. Many
commercial data loggers are equipped with
software that enables them to make this calculation and integrate the total FD
accumulated during a process. The FD calculation
is used during initial validation, validation maintenance, and change control. The
mathematical principles of the FD calculation
are essentially the same as those used to calculate lethality (F0) values in moist heat
sterilization. FD values are used to confirm
process consistency over time as correlation to endotoxin destruction is rarely possible.
VALIDATION
Because dry heat depyrogenation is appropriate only for heat stable materials, a high
margin for safety is always attainable.
Times and temperatures used for the purpose of destroying challenge materials can result
in extreme challenges to material
integrity and stability. Attention to depyrogenation processes, including an understanding
of the resident endotoxin load on
incoming materials and reduction to levels needed to assure patient safety, should take
place during drug product development,
prior to validation.
Equipment Qualification
Equipment Qualification (EQ) is a predefined program that focuses on the processing
equipment to confirm that it has been
properly installed and operates as intended prior to evaluation of the process. In some
companies, EQ may be separated into
installation qualification (IQ) and operational qualification (OQ), or combined together
under a joint terminology of installation operational qualification (I/OQ). Equipment
qualification provides a baseline for preventive maintenance and change control
assuring reproducibility of equipment operation over time.
Empty Chamber Temperature Distribution for Ovens
The oven should be evaluated for empty chamber temperature distribution. This is
assessed by measurement of temperature
at each corner of oven, near the controlling probe(s) and other locations as justified.
Differences in the cycle dwell period can
be discounted in this evaluation, as only the shortest dwell period need be evaluated. The
evaluation is best performed over
the last few minutes of the dwell period once the system has fully stabilized. The
acceptance criteria for this test vary with the
oven’s design and operating conditions; however, temperature distribution is typically
substantially less uniform than observed
in autoclaves and may be ±15° or more. Depyrogenation ovens that are located at floor
level may have even greater ranges in
temperature. The temperature distribution measurement may be of value in the evaluation
of changes to the oven.
Empty Temperature Distribution in Tunnels
While these studies are often done, they are actually of limited value. Unloaded
depyrogenation tunnels will always produce
far more variability in temperature distribution than will a fully loaded tunnel. Therefore,
for depyrogenation tunnels,
temperature studies under fully loaded conditions only are indicated. Important to the
proper operation of the dry heat tunnel
is the establishment of the required air flow balance between the tunnel and the adjoining
areas. Improper air flow can cause
uneven heating across the load being processed. The temperature distribution
measurement may be of value in the evaluation
of changes to the tunnel.
Component Mapping
The ability of dry heat to penetrate load items and to bring them to the required
temperature should be determined.
Load items that are complex, of significant mass, with enclosed volumes and product
contact surfaces that must be
depyrogenated, should be subjected to component mapping to determine internal cold
spots. All load items should be
prepared, wrapped (if that is the practice), and oriented in a manner consistent with how
they will be processed. Glass typically
enters the depyrogenation process (whether tunnel or oven) wet and must be evaluated
wet to properly determine the effect
on overall thermal input. Mapping of glass components to be processed in tunnels is not
necessary; all monitoring of
temperature in tunnels is accomplished with probes in contact with the bottom of the
container.
Load Mapping in Ovens
Fixed loading patterns are necessary in oven depyrogenation because of the limited heat
capacity of the air; fully packed
conditions because of their greater mass ordinarily result in the best process temperature
uniformity. Load mapping assures
that items placed throughout the load attain the desired depyrogenation conditions.
Identification of cold zones within the
oven should be established during depyrogenation cycle development. Information from
the load mapping is used to adjust
cycle timing to assure appropriate efficacy across the entire load. It may be possible to
validate maximum and minimum loads
(as determined by either the number of items or their mass).
F D is calculated from the temperature data at all monitored locations within the load
pattern.
Load Mapping in Tunnels
Load mapping can be assessed using sets of calibrated sensors (i.e., trailing or wireless
temperature sensors) positioned within
the glass pack as it moves through the tunnel. Temperature sensors should be placed into
direct contact with the glass item at
the bottom of the container. Temperature measurements should be made on the leading
edge, the middle (highest density),
and the trailing edge of the glass pack across the width of the conveyor belt. There should
be NLT 5 temperature sensors
positioned across the belt in each section of the load. FD is calculated from the
temperature data at all monitored locations.
Studies should be performed using all container sizes to determine the lowest FD
locations. The FD results can be used to support
the selection of containers/conditions to be evaluated in the confirmation studies (see
below). There is no requirement to
perform temperature heat distribution measurements during these studies.
Confirmation of Depyrogenation
The materials/glass components to be depyrogenated should be assessed for their
incoming endotoxin content prior to the
validation study. This would include glass as received and immediately after washing. All
tested materials should be handled
and prepared using defined procedures. Materials and glass prepared in the same manner
are used in the depyrogenation
validation studies. The addition of challenge material to the load items, including a
requirement to demonstrate a 3-log
reduction, may not be required if time and temperature studies consistently indicate that
depyrogenation conditions are met.
Temperature monitoring as described above must be done simultaneously with the
depyrogenation confirmation studies. The
confirmation studies should be performed at reduced time-temperature conditions from
those utilized in routine processing
and deliver lower FD results when compared to those determined in the mapping studies
and are considered “worst case”
confirmation of depyrogenation process efficacy. There is no requirement to perform
temperature heat distribution
measurements during these studies.
OVENS
A minimum of five (5) samples should be taken in proximity at NLT 10 temperature-
monitored locations (including those
determined to be the coldest from the oven load mapping study) in the oven and tested for
endotoxin content post-processing.
The process is considered acceptable if the amount of endotoxin per sample is NMT 0.1
EU.
TUNNELS
A minimum of five (5) samples should be taken in proximity to each monitored position
within the tunnel load (including
those determined to be the coldest from the tunnel load mapping study) and tested for
endotoxin content post-processing.
The process is considered acceptable if the amount of endotoxin per sample is NMT 0.1
EU.
ROUTINE PROCESS CONTROL
As with all processes, after the dry heat depyrogenation process has been validated, it
must be subject to ongoing controls
that maintain it within the validated state at all times. Temperature and exposure time,
which are the important dry heat
depyrogenation parameters, can be used to confirm performance on a routine basis.
Where direct assessment of FD is not
possible, assuring that the temperature and exposure time conditions were met results in
an equivalent confidence that the
depyrogenation system operated in a validated state of control.
Depyrogenation á1228ñ details the general practices that are appropriate for all
depyrogenation systems. This is accomplished
by a number of related practices that are essential for the continued use of the process
over an extended period of time. The
essential practices to maintain validated status include calibration, physical
measurements, periodic endotoxin assessment on
incoming materials, ongoing process control, change control, preventive maintenance,
and periodic reassessment and training.
APPENDIX
Additional Sources of Information
• Tsuji K, Harrison S. Dry heat destruction of lipopolysaccharide: dry heat destruction
kinetics. Appl Environ Microbio. 1978;
36(5):710–714.
• Ludwig J., Avis KE. Dry heat inactivation of endotoxin on the surface of glass. J
Parenteral Sci Technol. 1990; 44(1):4–12.
• Parenteral Drug Association (PDA), technical report 3, Validation of dry heat processes
used for depyrogenation and
sterilization. 2013.
REFERENCES
1. Tsuji K, Lewis A. Dry heat destruction of lipopolysaccharide: a mathematical
approach to process evaluation. Appl Environ
Microbio. 1978; 36(5):715–719.
2. Akers MJ, Ketron KM, Thompson, BR. F value requirements for the destruction of
endotoxin in the validation of dry-heat
sterilization/depyrogenation cycles. J Parenteral Sci Technol. 1982; (36):12–6.
4
Printed on: Wed Aug 11 2021, 06:04:17 PM
(EST)
Official Status: Currently Official on 11-Aug-2021 DocId: 1_GUID-B1C4ACBB-A2C5-
4298-AD66-23570AACBD27_1_en-US
Printed by: Le Tran Official Date: Official as of 01-Aug-2016 Document Type:
GENERAL CHAPTER @2021 USPC
https://online.uspnf.com/uspnf/document/1_GUID-B1C4ACBB-A2C5-4298-AD66-
23570AACBD27_1_en-US 4/4
Official

You might also like