Who (2015) PDF

Working document QAS/15.
639
August 2015
Draft document for comment
1 SUPPLEMENTARY GUIDELINES ON
2 GOOD MANUFACTURING PRACTICES FOR HEATING,
3 VENTILATION AND AIR-CONDITIONING SYSTEMS FOR
4 NON-STERILE PHARMACEUTICAL DOSAGE FORMS
5 (August 2015)
6 DRAFT FOR COMMENT
7 Should you have any comments on the attached text, please send these to Dr S. Kopp, Dr S. Kopp,
Group Lead, Medicines Quality Assurance, Technologies, Standards and Norms
(kopps@who.int) with a copy to Ms Marie Gaspard (gaspardm@who.int) by 10 October 2015.
8 Medicines Quality Assurance working documents will be sent out electronically only and
9 will also be placed on the Medicines website for comment under “Current projects”. If you
10 do not already receive our draft working documents please let us have your email address
(to bonnyw@who.int) and we will add it to our electronic mailing list.
11
12 © World Health Organization 2015

13 All rights reserved.
14 This draft is intended for a restricted audience only, i.e. the individuals and organizations having
15 received this draft. The draft may not be reviewed, abstracted, quoted, reproduced, transmitted,
16 distributed, translated or adapted, in part or in whole, in any form or by any means outside these
17 individuals and organizations (including the organizations' concerned staff and member
18 organizations) without the permission of the World Health Organization. The draft should not be
19 displayed on any website.
20 Please send any request for permission to:
21 Dr Sabine Kopp, Group Lead, Medicines Quality Assurance, Technologies, Standards and Norms,
22 Department of Essential Medicines and Health Products, World Health Organization, CH-1211
23 Geneva 27, Switzerland. Fax: (41-22) 791 4730; email: kopps@who.int
24
25 The designations employed and the presentation of the material in this draft do not imply the
26 expression of any opinion whatsoever on the part of the World Health Organization concerning the
27 legal status of any country, territory, city or area or of its authorities, or concerning the delimitation
28 of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which
29 there may not yet be full agreement.
30 The mention of specific companies or of certain manufacturers’ products does not imply that they
31 are endorsed or recommended by the World Health Organization in preference to others of a similar
32 nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are
33 distinguished by initial capital letters.
34 All reasonable precautions have been taken by the World Health Organization to verify the
35 information contained in this draft. However, the printed material is being distributed without
36 warranty of any kind, either expressed or implied. The responsibility for the interpretation and use
37 of the material lies with the reader. In no event shall the World Health Organization be liable for
38 damages arising from its use.
39 This draft does not necessarily represent the decisions or the stated policy of the World Health
40 Organization.
Working document QAS/15.639
page 2
41 SCHEDULE FOR THE PROPOSED ADOPTION PROCESS OF

42 DOCUMENT QAS/15.639
43 SUPPLEMENTARY GUIDELINES ON GOOD MANUFACTURING
44 PRACTICES FOR HEATING, VENTILATION AND AIR-
45 CONDITIONING SYSTEMS FOR NON-STERILE
46 PHARMACEUTICAL DOSAGE FORMS.
47 PROPOSAL FOR REVISION
Discussion of proposed need for revision in view of the 29 June–

current trends in engineering and experience gained 1 July 2015
during the implementation of this guidance in inspection
during informal consultation on data management,
bioequivalence, GMP and medicines’ inspection
Preparation of draft proposal for revision by D. Smith, July-August 2015
consultant to the Medicines Quality Assurance group and
Prequalification Team (PQT)-Inspections, based on the
feedback received during the meeting and from PQT-
Inspections
Circulation of revised working document for public September 2015
consultation
Consolidation of comments received and review of 10 October 2015
feedback
Presentation to fiftieth meeting of the WHO Expert 12–16 October 2015
Committee on Specifications for Pharmaceutical
Preparations
Any other follow-up action as required …
48
49
50
51
page 3
52 BACKGROUND
53 During the consultation on data management, bioequivalence, GMP and

54 medicines’ inspection held in 2015 the possible revision of the guidance
55 for (WHO Technical Report Series, No. 961, Annex 5, 2011) was
56 discussed with the inspectors. It was suggested that in light of the new
57 developments a draft for revision be prepared. This new proposal for
58 revision was drafted based on the feedback received, the new, current
59 trends in engineering and the experience gained during the implementation
60 of this guidance in inspection.
61 At the same time, the opportunity was used to improve the graphic images
62 and make them more readable in e-version as well as in print.
63 Summary of main changes
64 Below is a list of the main changes that have been made to the WHO Technical
65 Report Series, No. 961, 2011, Annex 5: Supplementary guidelines on good
66 manufacturing practices for heating, ventilation and air-conditioning systems for
67 non-sterile pharmaceutical dosage forms.
68
69 1. The Premises section has been moved towards the beginning of the document
70 due to its important impact on HVAC designs. In addition the text has been
71 expanded and a number of sample layouts have been included.
72
73 2. The HVAC sections have been re-arranged into a more logical sequence.
74
75 3. The Commissioning, Qualification and Validation (C, Q & V) section has
76 been aligned with the proposed revisions to the Supplementary GMP
77 Validation TRS937 Annex4 guideline.
78
79 4. Significant notes were added under the new Supplementary notes on test
80 procedures section.
81
82 5. The Maintenance section has been separated out of the C, Q & V section.
83
84 6. All the diagrams have been revised (mainly to achieve better clarity).
85
86 7. Throughout the document additional notes have been added and text revised
87 to provide better understanding and avoid ambiguity.
88
89 8. A list of abbreviations has been added.
90
page 4
91 Contents
92 page
93 1. Introduction
94 2. Scope of document
95 3. Glossary
96 4. Premises
97 5. Design of HVAC systems and components
98 5.1 General
99 5.2 Air distribution
100 5.3 Recirculation system
101 5.4 Full fresh-air systems
102 5.5 Additional system components
103 6. Protection
104 6.1 Products and personnel
105 6.2 Air filtration
106 6.3 Unidirectional airflow
107 6.4 Infiltration
108 6.5 Cross-contamination
109 6.6 Displacement concept (low pressure differential,
110 high airflow)
111 6.7 Pressure differential concept (high pressure differential,
112 low airflow)
113 6.8 Physical barrier concept
114 6.9 Temperature and relative humidity
115 7. Dust control
116 8. Protection of the environment
117 8.1 General
118 8.2 Dust in exhaust air
119 8.3 Vapour and fume removal
120 9. Commissioning, qualification and validation
121 9.1 General
122 9.2 Commissioning
123 9.3 Qualification
124 9.4 Supplementary notes on test procedures
125 10. Maintenance
126 11. Abbreviations
127 References
128 Further reading
129
page 5
1301. Introduction
131
132 Heating, ventilation and air-conditioning (HVAC) play an important role in
133 ensuring the manufacture of quality pharmaceutical products. A well designed
134 HVAC system will also provide comfortable conditions for operators.
135
136 These guidelines mainly focus on recommendations for systems for
137 manufacturers of solid dosage forms. The guidelines also refer to other
138 systems or components which are not relevant to solid dosage form
139 manufacturing plants, but which may assist in providing a comparison
140 between the requirements for solid dosage-form plants and other systems.
141
142 HVAC system design influences architectural layouts with regard to items
143 such as airlock positions, doorways and lobbies. The architectural components
144 have an effect on room pressure, differential cascades and cross-
145 contamination control. The prevention of contamination and cross-
146 contamination is an essential design consideration of the HVAC system. In
147 view of these critical aspects, the design of the HVAC system should be
148 considered at the concept design stage of a pharmaceutical manufacturing
149 plant.
150
151 Temperature, relative humidity and ventilation should be appropriate and
152 should not adversely affect the quality of pharmaceutical products during
153 their manufacture and storage, or the accurate functioning of equipment.
154
155 This document aims to give guidance to pharmaceutical manufacturers
156 and inspectors of pharmaceutical manufacturing facilities on the design,
157 installation, qualification and maintenance of the HVAC systems.
158 These guidelines are intended to complement those provided in Good
159 manufacturing practices for pharmaceutical products (1) and should be read
160 in conjunction with the parent guide. The additional standards addressed by
161 the present guidelines should, therefore, be considered supplementary to
162 the general requirements set out in the parent guide.
163
1642. Scope of document
165
166 These guidelines focus primarily on the design and good manufacturing
167 practices (GMP) requirements for HVAC systems for facilities for the
168 manufacture of solid dosage forms. Most of the system design principles
169 for facilities manufacturing solid dosage forms also apply to facilities
170 manufacturing other dosage forms and other classes of products including
171 biological products, herbal medicines, complimentary medicines and
172 finishing process steps for APIs . Additional specific requirements apply
173 for handling of sterile products and hazardous products. Guidelines for
174 hazardous, sterile and biological product facilities are covered in separate
page 6
175 WHO guidelines (WHO Technical Report Series, No. 957, Annex 3,
176 Technical Report Series, No. 961 Annex 6 and working document
177 WHO/BS/2015.2253, intended to replace Technical Report Series, No.
178 822, Annex 1, 1992, respectively).
179
180 These guidelines are intended as a basic guide for use by pharmaceutical
181 manufacturers and GMP inspectors.
182 They are not intended to be prescriptive in specifying requirements and design
183 parameters. There are many parameters affecting a clean area condition and it
184 is, therefore, difficult to lay down the specific requirements for one particular
185 parameter in isolation.
186
187 Many pharmaceutical manufacturers have their own engineering design and
188 qualification standards, and requirements may vary from one manufacturer
189 to the next. Design parameters and user requirements should, therefore, be
190 set realistically for each project, with a view to creating a cost-effective
191 design, yet still complying with all regulatory standards and ensuring that
192 product quality and safety are not compromised. The three primary aspects
193 addressed in this manual are the roles that the HVAC system plays in product
194 protection, personnel protection and environmental protection (Figure 2).
195
196 Cognizance should be taken of the products to be manufactured when
197 establishing system design parameters. A facility manufacturing multiple
198 different products may have more stringent design parameters with respect
199 to cross-contamination control, compared with a single product facility.
200
201 Risk assessment studies should be an integral part of the facility design and
202 implementation, from the URS stage right through to validation, as
203 indicated in the diagram below (Figure 1). Validation protocols and criteria
204 should be justified by links to a written risk assessment.
205
page 7
206 Figure 1
207 GMP compliance sequence diagram
PROJECT
INCEPTION
C OM PL I A
P NC
GM USER
E
REQUIREMENT
SPECIFICATION
PL I A NC E
GM P
QUALIFICATION FACILITY
& LAYOUTS
VALIDATION
RISK
RISK
ASSESSMENT
ASSESSMENT
STUDIES
STUDIES
C OM
Installations
M
CO
HVAC &
PL
ENERGY
SERVICES
EFFICIENCY
DESIGNS
I
P
STUDIES A
M
N
G CE
208
209
210
page 8
211 Figure 2
212 The guidelines address the various system criteria according to the
213 sequence set out in this diagram
214
215
page 9
2163. Glossary
217
218 The definitions given below apply to terms used in these guidelines. They
219 may have different meanings in other contexts.
220
221 acceptance criteria
222 Measurable terms under which a test result will be considered acceptable.
223
224 action limit
225 The action limit is reached when the acceptance criteria of a critical
226 parameter have been exceeded. Results outside these limits will require
227 specified action and investigation.
228
229 air changes per hour
230 The volume of air supplied to a room, in m3/hr, divided by the room volume, in m3.
231
232 air-handling unit
233 The air-handling unit serves to condition the air and provide the required air
234 movement within a facility.
235
236 airflow protection booth
237 A booth or chamber, typically for purposes of carrying out Sampling or
238 Weighing, in order to provide product containment and operator protection.
239 Similar to UDAF protection but not necessarily a Grade A condition.
240
241 airlock
242 An enclosed space with two or more doors, which is interposed between
243 two or more rooms, e.g. of differing classes of cleanliness, for the purpose
244 of controlling the airflow between those rooms when they need to be
245 entered. An airlock is designed for and used by either people or goods
246 (PAL, personnel airlock; MAL, material airlock).
247
248 alert limit
249 The alert limit is reached when the normal operating range of a critical
250 parameter has been exceeded, indicating that corrective measures may need
251 to be taken to prevent the action limit being reached.
252
253 as-built
254 Condition where the installation is complete with all services connected and
255 functioning but with no production equipment, materials or personnel present.
256
257 at-rest
258 Condition where the installation is complete with equipment installed and
page 10
259 operating in a manner agreed upon by the customer and supplier, but with
260 no personnel present.
261
262 central air-conditioning unit (see air-handling unit)
263
264 change control
265 A formal system by which qualified representatives of appropriate
266 disciplines review proposed or actual changes that might affect a validated
267 status. The intent is to determine the need for action that would ensure that
268 the system is maintained in a validated state.
269
270 clean area (cleanroom)11
271 An area (or room or zone) with defined environmental control of particulate
272 and microbial contamination, constructed and used in such a way as to reduce
273 the introduction, generation and retention of contaminants within the area.
274
275 closed system
276 A system where the product or material is not exposed to the manufacturing
277 environment.
278
279 commissioning
280 Commissioning is the documented process of verifying that the equipment
281 and systems are installed according to specifications, placing the equipment
282 into active service and verifying its proper action. Commissioning takes
283 place at the conclusion of project construction but prior to validation.
284
285 containment
286 A process or device to contain product, dust or contaminants in one zone,
287 preventing it from escaping to another zone.
288
289 contamination
290 The undesired introduction of impurities of a chemical or microbial nature,
291 or of foreign matter, into or on to a starting material or intermediate, during
292 production, sampling, packaging or repackaging, storage or transport.
293
294 controlled area
295 An area within the facility in which specific environmental facility
296 conditions and procedures are defined, controlled, and monitored to prevent
297 degradation or cross-contamination of the product.
1
Note: Clean area standards, such as ISO 14644-1, provide details on how to classify air cleanliness by
means of particle concentrations, whereas the GMP standards provide a grading for air cleanliness in
terms of the condition (at-rest or operational), the permissible microbial concentrations, as well as
other factors such as gowning requirements. GMP and clean area standards should be used in
conjunction with each other to define and classify the different manufacturing environments.
page 11
298
299 controlled-not classified
300 An area where some environmental conditions are controlled (such as
301 temperature), but the area has no cleanroom classification
302
303 critical parameter or component
304 A processing parameter (such as temperature or relative humidity) that
305 affects the quality of a product, or a component that may have a direct
306 impact on the quality of the product.
307
308 critical quality attribute
309 A physical, chemical, biological or microbiological property or characteristic
310 that should be within an appropriate limit, range or distribution to ensure
311 the desired product quality.
312
313 cross-contamination
314 Contamination of a starting material, intermediate product or finished
315 product with another starting material or product during production.
316
317 design condition
318 Design condition relates to the specified range or accuracy of a controlled
319 variable used by the designer as a basis for determining the performance
320 requirements of an engineered system.
321
322 design qualification
323 Design qualification is the documented check of planning documents and
324 technical specifications for conformity of the design with the process,
325 manufacturing, GMP and regulatory requirements.
326
327 direct impact system
328 A system that is expected to have a direct impact on product quality. These
329 systems are designed and commissioned in line with good engineering
330 practice and, in addition, are subject to qualification practices.
331
332 exfiltration
333 Exfiltration is the egress of air from a controlled area to an external zone.
334
335 facility
336 The built environment within which the clean area installation and associated
337 controlled environments operate together with their supporting infrastructure.
338
339 good engineering practice
340 Established engineering methods and standards that are applied throughout
341 the project life-cycle to deliver appropriate, cost-effective solutions.
page 12
342
343 hazardous substance or product
344 A product or substance that may present a substantial risk of injury to health
345 or to the environment
346
347 HVAC
348 Heating, ventilation and air-conditioning. Also referred to as Environmental
349 control systems
350
351 indirect impact system
352 This is a system that is not expected to have a direct impact on product
353 quality, but typically will support a direct impact system. These systems are
354 designed and commissioned according to good engineering practice only.
355
356 infiltration
357 Infiltration is the ingress of air from an external zone into a controlled area.
358
359 installation qualification
360 Installation qualification is documented verification that the premises,
361 HVAC system, supporting utilities and equipment have been built and
362 installed in compliance with their approved design specification.
363
364 Most penetrating particle size (MPPS)
365 MPPS is a means of determining HEPA & ULPA filter efficiencies. The
366 MPPS is the particle size with the highest penetration for a defined filter
367 medium. (MPPS Integral overall efficiency is the efficiency, averaged over
368 the whole superficial face area of a filter element under a given operating
369 conditions of the filter. MPPS local efficiency is the efficiency, at a specific
370 point of the filter element under given operating conditions of the filter)
371
372 NLT
373 Not less than
374
375 NMT
376 Not more than
377
378 no-impact system
379 This is a system that will not have any impact, either directly or indirectly, on
380 product quality. These systems are designed and commissioned according
381 to good engineering practice only.
382
383 non-critical parameter or component
384 A processing parameter or component within a system where the operation,
page 13
385 contact, data control, alarm or failure will have an indirect impact or no
386 impact on the quality of the product.
387
388 normal operating range
389 The range that the manufacturer selects as the acceptable values for a parameter
390 during normal operations. This range must be within the operating range.
391
392 OOS
393 Out of specification
394
395 operating limits
396 The minimum and/or maximum values that will ensure that product and
397 safety requirements are met.
398
399 operating range
400 Operating range is the range of validated critical parameters within which
401 acceptable products can be manufactured.
402
403 operational condition
404 This condition relates to carrying out room classification tests with the
405 normal production process with equipment in operation, and the normal
406 staff present in the specific room.
407
408 operational qualification (OQ)
409 Operational qualification is the documentary evidence to verify that the equipment
410 operates in accordance with its design specifications in its normal operating range
411 and performs as intended throughout all anticipated operating ranges.
412
413 oral solid dosage (OSD)
414 Usually refers to an OSD plant that manufactures medicinal products such
415 as tablets, capsules and powders to be taken orally.
416
417 pass-through-hatch (PTH) or pass box (PB)
418 A cabinet with two or more doors for passing equipment or product, whilst
419 maintaining the pressure cascade and segregation between two controlled
420 zones. A passive PTH has no air supply or extract. A dynamic PTH has an
421 air supply into the chamber.
422
423 performance qualification (PQ)
424 Performance qualification is the documented verification that the process and/
425 or the total process related to the system performs as intended throughout all
426 anticipated operating ranges.
427
428
page 14
429 point extraction

430 Air extraction to remove dust with the extraction point located as close as
431 possible to the source of the dust.
432
433 pressure cascade
434 A process whereby air flows from one area, which is maintained at a higher
435 pressure, to another area at a lower pressure.
436
437 qualification
438 Qualification is the planning, carrying out and recording of tests on
439 equipment and a system, which forms part of the validated process, to
440 demonstrate that it will perform as intended.
441
442 quality critical process parameter (QCPP)
443 A process parameter which could have an impact on the critical quality
444 attribute.
445
446 relative humidity
447 The ratio of the actual water vapour pressure of the air to the saturated
448 water vapour pressure of the air at the same temperature expressed as a
449 percentage. More simply put, it is the ratio of the mass of moisture in
450 the air, relative to the mass at 100% moisture saturation, at a given
451 temperature.
452
453 standard operating procedure (SOP)
454 An authorized written procedure, giving instructions for performing
455 operations, not necessarily specific to a given product or material, but of a
456 more general nature (e.g. operation of equipment, maintenance and cleaning,
457 validation, cleaning of premises and environmental control, sampling and
458 inspection). Certain SOPs may be used to supplement product-specific
459 master and batch production documentation.
460
461 turbulent flow
462 Turbulent flow, or non-unidirectional airflow, is air distribution that is
463 introduced into the controlled space and then mixes with room air by means
464 of induction.
465
466 unidirectional airflow (UDAF)
467 Unidirectional airflow is a rectified airflow over the entire cross-sectional
468 area of a clean zone with a steady velocity and approximately parallel
469 streamlines (see also turbulent flow).
470
471 validation
page 15
472 The documented act of proving that any procedure, process, equipment,
473 material, activity or system actually leads to the expected results.
474
475 validation master plan (VMP)
476 Validation master plan is a high-level document which establishes an
477 umbrella validation plan for the entire project, and is used as guidance by
478 the project team for resource and technical planning (also referred to as
479 master qualification plan).
480
4814. Premises
482
4.1.
483 There is a close relationship between architectural design and HVAC
484 design, as they both have an impact on the functionality of the other.
485 HVAC system design influences architectural layouts with regard to items
486 such as airlock positions, doorways and lobbies. The architectural layouts
487 and building components have an effect on room pressure differential
488 cascades and cross-contamination control. The prevention of contamination
489 and cross-contamination is an essential design consideration of the HVAC
490 system. In view of these critical aspects, the design of the HVAC system
491 should be considered at the concept design stage of a pharmaceutical
492 manufacturing plant, and the design should be closely co-ordinated with the
493 architectural designers. In addition the architectural layout must ensure that
494 the material flow is in a logical sequence, excluding material flow reversals
495 where possible.
496
4.2.
497 As the efficient operation of the air-handling system and cleanliness levels
498 attained are reliant on the correct building layout and building finishes, the
499 following items should be considered.
500
501 4.2.1. Adequate airlocks, such as personnel airlocks (PAL) and/or material
502 airlocks (MAL), change rooms and passages should be provided to protect
503 passage between different cleanliness conditions. These should have supply
504 and extract air systems as appropriate.
505
506 4.2.2. Areas such as airlocks, change rooms and passages, should be
507 designed so that the required pressure cascades can be achieved.
508
509 4.2.3. Detailed diagrams depicting pressure cascades, air flow directions
510 and flow routes for personnel and materials should be prepared and
511 maintained.
512
page 16
513 4.2.4. Where possible, personnel and materials should not move from a
514 higher cleanliness zone to a lower cleanliness zone and back to a higher
515 cleanliness zone (if moving from a lower cleanliness zone to a higher
516 cleanliness zone, changing /decontamination procedures should be
517 followed).
518
519 4.2.5. The final stage of the changing room should be the same GMP
520 classification grade as the area into which it leads. Changing rooms should
521 be of a sufficient size to allow for ease of changing. Changing rooms should
522 be equipped with mirrors so that personnel can confirm the correct fit of
523 garments before leaving the changing room. Appropriate hand wash and
524 sanitizing facilities should be provided. Hand-wash basins should be
525 provided with elbow taps or sensor operated taps.
526
527 4.2.6. Door gaps around the door perimeter have a marked impact on the
528 pressure differential across the doorway. The fit of the doors should be
529 agreed upon between the architect and the HVAC designer to ensure that
530 the correct leakages are allowed for. Likewise the maintenance of doors is
531 a critical factor in room pressure control (a poorly fitting door can severely
532 compromise a room pressure differential).
533
534 4.2.7. Where the opening and closing of airlock doors could lead to cross-
535 contamination, these airlock doors should not be opened simultaneously.
536 An interlocking system and a visual and/or audible warning system should
537 be operated to prevent the opening of more than one door at a time.
538
539 4.2.8. Doors should be carefully designed to avoid un-cleanable recesses;
540 sliding doors may be undesirable for this reason. Swing doors should open
541 to the high-pressure side and be provided with self-closers. Exceptions are
542 permitted based on egress and site environmental, fire, health and safety
543 containment requirements.
544
545 4.2.9. The choice of building finishes and materials also has an impact on
546 air conditioning performance and air cleanliness. Materials should be
547 selected that will provide a well-sealed building to facilitate room pressure
548 control. Materials and paint finishes should also be non-dust and particle
549 liberating as this impacts on room cleanliness. Finishes should be easy to
550 clean and non-absorbent. To reduce the accumulation of dust and to
551 facilitate cleaning, there should be no un-cleanable recesses and a minimum
552 of projecting ledges, shelves, cupboards and equipment.
553
page 17
4.3.
554 The following diagrams illustrate some typical room and suite layouts with
555 their associated room pressures. These are purely examples and other
556 factors may dictate different room arrangements and room pressures.
557
558 Figure 3
559 Typical weigh booth layout
560
Dispensary Post-Staging
Material Flow
Room
35 Pa
Change
Room Pallet
SOB
Perforated Worktop
BIN
35 Pa BIN BIN
Table Scale
BIN BIN
Return Air Shaft

Weigh Booth
25 Pa
BIN
Airflow Protection
Plenum
BIN
Wash Bay
Floor
35 Pa
Scale
Material Flow
Dispensary Pre-Staging Room 15 Pa
561
562
563
564
565
page 18
566 Figure 4
567 Typical dispensary suite
568
Production Passage
Dispensary
Post-Staging
Room
Change Change
MAL Room MAL Room
Weigh Weigh
Return Air Shaft

Brocken
Return Air Shaft
Booth 1 Booth 2 Bulk

Airflow Airflow Store
Protection Protection
Plenum Plenum
Wash Wash
MAL MAL Bay
Bay
MAL
Dispensary Pre-Staging Room
Warehouse
Area
MAL
Material Flow
569
570
571
page 19
572 Figure 5
573 Sampling booth for small volumes
574
SOB
Return Air Shaft

575
576
577
578 Figure 6
579 Sampling Booth for larger volumes
580
581
582
page 20
583 Figure 7
584 Change rooms and ablution layouts
585
586 Figure 8
587 Compression cubicle with change room and MAL
L o e tu
SOB
w rn
R
L e Ai
ve r
l
Passage
n el
ur ev
r
Ai
et L
R ow
L
588
page 21
589 Figure 9
590 Compression cubicle without change room and MAL
591 (inclusion of airlocks dependant on risk assessment)
Lo etu
w rn
R
L e Ai
ve r
l
Passage
30 Pa
Compression Supply
Air
Cubicle Grille
15 Pa
n el
ur ev
r
Ai
et L
R ow
L
592
593
page 22
594 Figure 10
595 Wash-bay suite
Passage
Material Flow
596
597
5985. Design of HVAC systems and components

599
5.1.
600 General
601
602 5.1.1. The required degree of air cleanliness in most OSD manufacturing
603 facilities can normally be achieved without the use of high-efficiency
604 particulate air (HEPA) filters, provided the air is not recirculated or in the
605 case of a single-product facility. Many open product zones of OSD form
606 facilities are capable of meeting ISO 14644-1 Class 8 or Grade D, “at-rest”
607 condition, measured against particle sizes of 0.5 µm and 5 µm, but
608 cleanliness may not necessarily be classified as such by manufacturers.
609
610 5.1.2. A risk assessment should be carried out to determine the cleanroom
611 conditions required and the extent of validation required.
612
613 5.1.3. There are two basic concepts of air delivery to pharmaceutical
614 production facilities: a recirculation system, and a full fresh air system
page 23
615 (100% outside air supply). For recirculation systems the amount of fresh air
616 should not be determined arbitrarily on a percentage basis, but, for example,
617 by the following criteria:
618
619 • sufficient fresh air to compensate for leakage from the facility and
620 loss through exhaust air systems;
621 • sufficient fresh air to comply with national building regulations
622 (depending on occupant density, between 1 and 2.5 ACPH will often
623 satisfy occupancy requirements);
624 • sufficient fresh air for odour control;
625 • sufficient fresh air to provide the required building pressurization
626
627 5.1.4. Where automated monitoring systems are used, these should be
628 capable of indicating any out-of-specification (OOS) condition without
629 delay by means of an alarm or similar system. Sophisticated computer-
630 based data monitoring systems may be installed, which can aide with
631 planning of preventive maintenance and can also provide trend logging.
632
633 (This type of system is commonly referred to as a building management
634 system (BMS), building automation system (BAS) or system control and
635 data acquisition (SCADA) system. If these systems are used for critical
636 decision-making, they should be validated. If the BMS is not validated in
637 full (or in part for these critical parameters), an independent validated
638 environmental monitoring system (EMS) should be provided, specifically
639 for recording and alarming critical parameters. Critical parameters could
640 include for example, room temperature in production areas, humidity,
641 differential pressures, fan failure alarms, etc.)
642 ,
643 5.1.5. Failure of a supply air fan, return air fan, exhaust air fan or dust
644 extract system fan can cause a system imbalance, resulting in a pressure
645 cascade malfunction with a resultant airflow reversal.
646
647 5.1.6. A fan interlock failure matrix should be set up, such that if a fan
648 serving a high pressure zone fails, then any fans serving surrounding lower
649 pressure areas should automatically stop, to prevent an airflow reversal and
650 possible cross-contamination. This fan stop-start matrix should apply to the
651 switching on and switching off of systems to ensure that there is no flow
652 reversal causing cross-contamination.
653
654 5.1.7. Appropriate alarm systems should be in place to alert personnel if a
655 critical fan fails All critical alarms should be easily identifiable and visible
656 and/or audible to relevant personnel.
657
658
page 24
5.2.
659 Air distribution
660
661 5.2.1. The positioning of supply and extract grilles should be such as to
662 provide effective room flushing. Low-level return or exhaust air grilles are
663 usually preferred. However, where this is not possible, a higher air change
664 rate may be needed to achieve a specified clean area condition, e.g. where
665 ceiling return air grilles are used.
666
667 5.2.2. There may be alternative locations for return air. For example,
668 referring to Figure 11, Room 1 (low-level return air) and Room 2 (ceiling
669 return air)..
670
671
672 Figure 11
673 Air-handling system with high-efficiency particulate air filters in air-
674 handling unit
675
Primary Filter
HEPA Filter
Supply Air
Secondary
Filter
Fan
Cooling Coil
676
677
678 The airflow schematics of the two systems (Figures 1 1 and 1 2) indicate
679 air-handling units with return air or recirculated air, having a percentage
680 of fresh air added. Depending on product characteristics and dust loading
681 it is sometimes preferable to fit filters on return air outlets or in return air
682 ducting.
683
page 25
684 Figure 12 is a schematic diagram of an air-handling system serving

685 rooms with horizontal unidirectional flow, vertical unidirectional flow and
686 turbulent flow, for rooms 3, 4 and 5, respectively. In this case the HEPA
687 filters are terminally mounted at the rooms, and not in the AHU.
688 Terminally mounted HEPA filters can assist with preventing cross-
689 contamination from room to room in the event of a fan failure condition.
690 The decision whether to install terminal HEPA filters should be based on a
691 risk-assessment study.
692
693
694
695 Figure 12
696 Horizontal unidirectional flow, vertical unidirectional flow and
697 turbulent flow
698
Primary Filter
Supply Air
Secondary
Filter
Fan
Cooling Coil
HEPA Filters
699
700
5.3.
701 Recirculation system
702
703 5.3.1. There should be no risk of contamination or cross-contamination
704 (including by fumes and volatiles) due to recirculation of air.
705
706 5.3.2. Depending on the airborne contaminants in the return-air system
707 it may be acceptable to use recirculated air, provided that HEPA filters
708 are installed in the supply air stream (or return air stream) to remove
709 contaminants and thus prevent cross-contamination. The HEPA filters for
710 this application should have an EN 1822 classification of H13.
711
712 5.3.3. HEPA filters may not be required where the air-handling system
713 is serving a single product facility and there is evidence that cross-
714 contamination would not be possible.
715
page 26
716 5.3.4. Recirculation of air from areas where pharmaceutical dust is not
717 generated such as secondary packing, may not require HEPA filters in the
718 system.
719
720 5.3.5. HEPA filters may be located in the air-handling unit or placed
721 terminally. Where H E P A fi l t e r s a r e t e r m i n a l l y m o u n t e d t h e y
722 s h o u l d preferably not be connected to the ducting by means of flexible
723 ducting. Due to the high air pressure required for the terminal filter, this
724 connection should preferably be a rigid duct connection. Where flexible
725 ducting is used, it should be as short as possible and properly fixed to
726 withstand duct pressure. When HEPA filters are terminally mounted, it
727 should be possible to carry out filter integrity tests from within the room.
728 The filter housings will therefore require ports for measuring appropriate
729 upstream concentration (refer to ISO 14644.3) and penetration
730 concentration from within the room. In addition it should be possible to
731 monitor the filter pressure drop in individual HEPA filters.
732
733 5.3.6. Air containing dust from highly toxic processes and/or solvents or
734 flammable vapours should never be recirculated to the HVAC system.
735
736
737
5.4.
738 Full fresh-air systems
739
740
741 5.4.1. The required degree of filtration of the exhaust air depends on the
742 exhaust air contaminants and local environmental regulations. HEPA filters
743 in the exhaust system would normally only be required when handling
744 hazardous materials.
745
746 Figure 13 indicates a system operating on 100% fresh air and would
747 normally be used in a facility dealing with toxic products or solvents, where
748 recirculation of air with contaminants should be avoided.
749
750
page 27
751 Figure 13
752 Full fresh-air system
753
754
755
756
757
page 28
758 Figure 14
759 Full fresh-air system with energy recovery
760
Primary Filter
HEPA Filter
Supply Air
Secondary
Filter
Fan
Cooling Coil
Fresh Air
Exhaust
Air Fan
Primary Filter
Secondary
HEPA Filter
Filter
761
762
763 5.4.2. Energy-recovery wheels if used in multiproduct facilities should
764 have been subjected to a risk assessment to determine if there is any
765 risk of cross-contamination. When such wheels are used they should
766 not become a source of possible contamination (see Figure 14). Note:
767 Alternatives to the energy-recovery wheels, such as crossover plate heat
768 exchangers and water-coil heat exchangers, may be used in multiproduct
769 facilities.
770
771 5.4.3. The potential for air leakage between the supply air and exhaust air
772 as it passes through the wheel should be prevented. The relative pressures
773 between supply and exhaust air systems should be such that the exhaust air
774 system operates at a lower pressure than the supply system.
775
776
5.5.
777 Additional system components
778
779 5.5.1. A schematic diagram of the airflow for a typical system serving a
780 low relative humidity suite is represented in Figure 15. Air can be dried
781 with a chemical drier (e.g. a rotating desiccant wheel which is continuously
782 regenerated by means of passing hot air through one segment of the wheel).
783 Alternative methods of drying air are also available.
page 29
784 Figure 15
785 Air-handling system with chemical drying
CHEMICAL DRIER UNIT
Reactivation Chemical Drier

Air Fan Desiccant Wheel
S = Supply Air
E Fresh Air R = Return Air
Reactivation
Primary Filter
F = Fresh Air
Heater
+
E = Exhaust Air
Process
Air Fan
AIR HANDLING UNIT
Primary Filter
HEPA Filter
Supply Air
Secondary
Filter
Fan
-
Cooling Coil
F
S
R
+
S
LOW HUMIDITY
PRODUCTION
FACILITY
R R
786
787
788 5.5.2. The figure illustrates the chemical drier handling part of the fresh
789 air/return air mixture on a bypass flow. The location of the chemical drier
790 should be considered in the design phase. The practice of locating the
791 complete chemical drier unit in the production cubicle is not recommended
792 as this could be a source of contamination or cross-contamination. Examples
793 of appropriate locations for the drying wheel could include:
794
795 — full flow of fresh/return air;
796 — partial handling of fresh/return air (bypass airflow);
797 — return air only;
798 — fresh air only; or
799 — pre-cooled air with any of the above alternatives.
800
801 5.5.3. Possible additional components that may be required in air handling
802 should be considered depending on the climatic conditions and locations.
803 These may include items such as:
804
805 ‒ frost coils on fresh air inlets in very cold climates to preheat the air;
806 ‒ reheaters for humidity control
807 ‒ automatic air volume control devices
808 ‒ sound attenuators
page 30
809 ‒ snow eliminators to prevent snow entering air inlets and

810 blocking airflow
811 ‒ dust eliminators on air inlets in arid and dusty locations
812 ‒ moisture eliminators in humid areas with high rainfall
813 ‒ fresh air precooling coils for very hot or humid climates.
814
815
8166. Protection
817
818 6.1. Products and personnel
819
820 6.1.1. Areas for the manufacture of pharmaceuticals, where pharmaceutical
821 starting materials and products, utensils, primary packing materials and
822 equipment are exposed to the environment, should be defined as “clean
823 areas”, “clean zones”, “controlled areas” or “cleanrooms”.
824
825 6.1.2. The achievement of a particular clean area condition depends on a
826 number of criteria that should be addressed at the design and qualification
827 stages. A suitable balance between the different criteria will be required in
828 order to create an efficient clean area.
829
830 6.1.3. Some of the basic criteria to be considered which affects room
831 cleanliness should include:
832
833 • building finishes and structure• dust control and containment
834 • air filtration
835 • air change rate or flushing rate
836 • room pressure
837 • location of air terminals and directional airflow
838 • temperature
839 • relative humidity
840 • material flow
841 • personnel flow
842 • gowning procedures
843 • equipment movement
844 • process being carried out (open or closed system)
845 • outside air conditions
846 • occupancy
847 • type of product
848 • cleaning standard operating procedures (SOPs).
849
850 6.1.4. Air filtration and air change rates should be set to ensure that the
851 defined clean area condition is attained.
852
page 31
853 6.1.5. The air change rates should be determined by the manufacturer and
854 designer, taking into account the various critical parameters using a risk
855 based approach with due consideration of capital and running costs and
856 energy usage. Primarily the air change rate should be set to a level that will
857 achieve the required clean area condition.
858
859 6.1.6. Air change rates are normally determined by the following
860 considerations (could normally vary between 6 and 20 air changes per hour):
861
862 • area condition required: whether a specific room cleanliness
863 condition is in fact required and whether the room condition is
864 rated for an “at rest” condition or an “operational” condition (air
865 change rate should be selected on need rather than tradition);
866 • the product characteristics (e.g. odours, hygroscopicity, etc.);
867 • the quality and filtration of the supply air;
868 • particulates generated by the manufacturing process;
869 • particulates generated by the operators;
870 • configuration of the room and air supply and extract locations;
871 • sufficient air to achieve containment effect and to clean up the area;
872 • sufficient air to cope with the room heat load;
873 • sufficient air to balance extract rates;
874 • sufficient air to maintain the required room pressure.
875
876 6.1.7. If a cleanroom classification is specified, the manufacturer should
877 state whether this is achieved under “as-built” (Figure 16), “at-rest” (Figure
878 17) or “operational” (Figure 18) conditions.
879
880 6.1.8. Room classification tests in the “as-built” condition should be
881 carried out on the bare room, in the absence of any equipment or personnel.
882
883 6.1.9. Room classification tests in the “at-rest” condition should be carried
884 out with the equipment operating where relevant, but without any operators.
885 Because of the amounts of dust usually generated in a solid dosage facility,
886 the clean area classifications would be rated for the “at-rest” condition.
887
888 6.1.10. Room classification tests in the “operational” condition are
889 normally carried out during the normal production process with equipment
890 operating, and the normal number of personnel present in the room. When
891 qualifying for the operational condition details of the process operating,
892 number and positions of staff should be stipulated for each room, to enable
893 future qualifications to duplicate the same conditions.
894
895 6.1.11. Room clean-up or recovery tests are performed to determine
896 whether the installation is capable of returning to a specified cleanliness
page 32
897 level within a finite time, after being exposed briefly to a source of
898 airborne particulate challenge. Room “clean-up” or “recovery” tests should
899 demonstrate a change in particle concentration by a factor of 100 within
900 the prescribed time (as per ISO 14644-3 clause B.12) (3). The guidance
901 time period for clean-up or recovery is about 15 to 20 minutes.
902 In some instances it is not possible to increase the concentration by a factor
903 of 100 (such as for an ISO 14644 Class 8 condition) as the high particle
904 concentration can damage the particle counter. In this instance the particle
905 decay method can be used as per ISO 14644-3 clause B.12.3.2.
906
907 6.1.12. Materials and products should be protected from contamination
908 and cross-contamination during all stages of manufacture (see also section
909 6.5 for cross-contamination control).
910
911 Note: contaminants may result from inappropriate premises (e.g. poor design,
912 layout or finishing), poor cleaning procedures, contaminants brought in by
913 personnel, poor manufacturing process and a poor HVAC system.
914
915 Figure 16
916 “As-built” condition
917
918
919
920
page 33
921 Figure 17
922 “At-rest” condition
923
924
925 Figure 18
926 “Operational” condition
927
page 34
928
929
930 6.1.13. Airborne contaminants should be controlled through effective
931 ventilation and filtration.
932
933 6.1.14. External contaminants should be removed by effective filtration of
934 the supply air (see Figure 19 for an example of a shell-like building
935 layout to enhance containment and protection from external
936 contaminants).
937
938 6.1.15. Internal contaminants should be controlled by dilution and
939 flushing of contaminants in the room, or by displacement airflow (See
940 Figures 20 and 21 for examples of methods for the flushing of airborne
941 contaminants).
942
943 6.1.16. Airborne particulates and the degree of filtration should be
944 considered critical parameters with reference to the level of product
945 protection required.
946
947 6.1.17. Personnel should not be a source of contamination.
948
949 6.1.18. The level of protection and air cleanliness for different areas should
950 be determined according to the product being manufactured, the process
951 being used and the product’s susceptibility to degradation (Table 3).
page 35
952 Figure 19
953 Shell-like containment control concept
954
Personnel Movement
Personnel Movement
Waste Exit
955
956
957
958 6.2. Air filtration
959
960 Note: The degree to which air is filtered plays an important role in the
961 prevention of contamination and the control of cross-contamination.
962
963 6.2.1. The type of filters required for different applications depends on
964 the quality of the ambient air and the return air (where applicable) and
965 also on the air change rates. Table 4 gives the recommended filtration
966 levels for different levels of protection in a pharmaceutical facility.
967 Manufacturers should determine and prove the appropriate use of
968 filters.
969
970 6.2.2. Filter classes should always be linked to the standard test method
971 because referring to actual filter efficiencies can be very misleading (as
972 different test methods each result in a different efficiency value for the
973 same filter). (Referring to filter classifications such as an 85% filter or a 5 µm
974 filter are not valid classifications and should not be used, as this can lead to
975 the incorrect filter being installed. Only the EN 779 and EN 1822 or ISO
976 29463 classifications, as per Tables 1 and 2, should be used.)
page 36
977
978
979 Table 1
980 Comparison of filter test standards
Eurovent
Eurovent 4/5
4/5
ASHRAE
Eurovent ASHRAE ASHRAE
52.1 EN 779 & EN 1822
4/5 rating 52.2 52.1
BS6540 Part
BS6540
1
Part 1
ISO 29463
Average
Average MPPS integral
(supersede dust spot
Merv rating arrestance overall EN Rating
d) efficiency
Am (%) efficiency (%)
Em (%)
99.999995 U17 75E

99.99995 U16 65E
EU 14 99.9995 U15 55E
EN 1822: 2009
EU 13 Merv 18 99.995 H14 45E
EU 12 Merv 17 99.95 H13 35E
EU 11 99.5 E12 25E
EU 10 95 E11 15E
EU 9 Merv 16 >95 85 E10
EU 9 Merv 15 95 F9
EU 8 Merv 14 90 F8
Merv 13 >98 85 MPPS = F7
most
EU 7 >98 80
penetrating
Merv 12 >95 75 particle Size
EU 6 >95 70 F6
Merv 11 >95 65
>95 60
Merv 10 >95 55
EU 5 Merv 9 >95 50 F5
EN 1822: 2009
Merv 8 >95 45
>95 40
Merv 7 >90 35
EU 4 >90 30 G4
Merv 6 90 25
EU 3 Merv 5 85 20 G3
80 <20
Merv 4 75
EU 2 Merv 3 70 G2
Merv 2 65
EU 1 Merv 1 <65 G1
981
page 37
982 Table 2
983 Comparison of ISO and EN Filter Standards
ISO 29463
Global or Integral Local Values for EN 1822
Values for MPPS MPPS
MPPS
Collection Collection Integral
Filter Penetrat Penetrat EN
Efficiency Efficiency Overall
Class ion (%) ion (%) Rating
(%) (%) Efficiency
(%)
≥ ≤
ISO 75 E ≥ 99.9999 ≤ 0.0001 99.999995 U17
99.999995 0.000005
≥ ≤
ISO 70 E ≥ 99.9999 ≤ 0.0001 - -
99.99999 0.00001
≥ ≤ ≤
ISO 65 E ≥ 99.99975 99.99995 U16
99.99995 0.00005 0.00025
ISO 60 E ≥ 99.9999 ≤ 0.0001 ≥ 99.9995 ≤ 0.0005 - -
≥
ISO 55 E ≤ 0.0005 ≥ 99.9975 ≤ 0.0025 99.9995 U15
99.99995
ISO 50 E ≥ 99.999 ≤ 0.001 ≥ 99.995 ≤ 0.005 - -
ISO 45 E ≥ 99.995 ≤ 0.005 ≥ 99.975 ≤ 0.025 99.995 H14
ISO 40 E ≥ 99.99 ≤ 0.01 ≥ 99.95 ≤ 0.05 - -
ISO 35 E ≥ 99.95 ≤ 0.05 ≥ 99.75 ≤ 0.25 99.95 H13
ISO 30 E ≥ 99.9 ≤ 0.1 - -
ISO 25 E ≥ 99.5 ≤ 0.5 99.5 E12
ISO 20 E ≥ 99 ≤1 - -
ISO 15 E ≥ 95 ≤5 95 E11
The above all tested for MPPS (most penetrating
particle size)
984 Note: The filter classifications referred to above relate to the EN
985 1822:2009 and EN 779: 2002 test standards (EN 779 relates to filter
986 classes G1 to F9 and EN1822 relates to filter classes E10 to U17).
page 38
987
988 6.2.3. In selecting filters, the manufacturer should have considered other
989 factors, such as particularly contaminated ambient conditions, local
990 regulations and specific product requirements. Good prefiltration extends
991 the life of the more expensive filters downstream.
992
993 6.2.4. Filters have an impact on the cleanroom class or Level of
994 Protection. The different levels of protection and recommended filters
995 grades are given in Tables 3 and 4 below.
996
997 Table 3
998 Examples of levels of protection (based on ISPE oral solid dosage
999
1000 (OSD) guideline criteria)
Level Condition Example of area
Level 1 General Area with normal housekeeping and
maintenance where there is no potential for
product contamination, e.g. warehousing.
Level 2 Protected Area in which steps are taken to protect

the pharmaceutical starting material or
product from direct or indirect
contamination or degradation, e.g.
secondary packing, warehousing, first
stage change rooms.
Level 3 Controlled Area in which specific environmental

conditions are defined, controlled and
monitored to prevent contamination or
degradation of the pharmaceutical starting
material or product, e.g. where product,
starting materials and components are
exposed to the room environment; plus
equipment wash and storage areas for
equipment product contact parts.
1001
1002
page 39
1003 Table 4
1004
1005 Levels of protection and recommended filtration
Level of protection Recommended filtration
Level 1 Primary filters only (e.g. EN 779 G4 filters)
Level 2 Protected areas operating on 100% outside air:

primary plus secondary filters (e.g. EN 779 G4
plus F8 or F9 filters)
Level 3 Production facility operating on recirculated

plus ambient air, where potential for cross-
contamination exists: Primary plus secondary
plus tertiary filters (e.g. EN 779 G4 plus F8 plus
EN 1822 H13 filters) (for full fresh air system,
without recirculation, G4 and F8 or F9 filters are
acceptable)
1006
1007 6.2.5. Materials for components of an HVAC system should be selected
1008 with care so that they do not become a source of contamination. Any
1009 component with the potential for liberating particulate or microbial
1010 contamination into the airstream should be located upstream of the final
1011 filters.
1012
1013 6.2.6. Where possible ventilation dampers, filters and other services should
1014 be designed and positioned so that they are accessible from outside the
1015 manufacturing areas (service voids or service corridors) for maintenance
1016 purposes.
1017
1018 6.2.7. Directional airflow within production or primary packing areas
1019 should assist in preventing contamination. Airflows should be planned in
1020 conjunction with operator locations, so as to minimize contamination of the
1021 product by the operator and also to protect the operator from dust inhalation.
1022 Different airflow patterns are indicated in Figures 20 & 21 below.
1023
1024
page 40
1025 Figure 20
1026 Turbulent dilution of dirty air
1027
1028
1029
1030
1031 Low-level extract is ideal for dust suppression purposes, but is not
1032 essential. (Low-level extract is essential for Grade A, B & C classified
1033 areas.)
1034
1035
page 41
1036 Figure 21
1037 Unidirectional displacement of dirty air
1038
1039
1040
1041
1042
1043 6.2.8. HVAC air distribution components should be designed, installed
1044 and located to prevent contaminants generated within the room from being
1045 spread.
1046
1047 6.2.9. Supply air diffusers should be selected with care taking consideration
1048 of, e.g. room requirements and positions of equipment and operators in the
1049 room. Supply air diffusers of the high induction type (e.g. those typically
1050 used for office-type air-conditioning) should where possible not be used
1051 in clean areas where dust is liberated. Air diffusers should be of the non-
1052 induction type, introducing air with the least amount of induction so as to
1053 maximize the flushing effect. In rooms where the process results in high
1054 dust liberation; perforated plates or low induction swirl diffusers with
1055 low level extract or return should be used (to contain the dust at the lower
1056 level of the room) (see Figures 22–24 for illustrations of the three types of
1057 diffuser). In cases where dust liberation is low, ceiling return air grilles may
1058 be acceptable.
1059
1060
1061
page 42
1062 6.2.10. Induction and certain swirl diffusers induce room air vertically
1063 up to the diffuser to mix with the supply air. These diffusers create good
1064 dilution of contaminants in the room and may be used in rooms where there
1065 is low dust liberation. However, if used in rooms where excessive dust is
1066 generated, the distribution of dust in the room could be hazardous for the
1067 operators in the room, as dust is drawn up into the supply air stream and
1068 then spread throughout the room. Airflow patterns for different diffuser
1069 types are indicated in figures 22, 23 & 24 below.
1070
1071 Figure 22
1072 Induction diffuser
1073
1074
1075
1076
page 43
1077 Figure 23
1078 Perforated plate diffuser
1079
1080
1081
page 44
1082 Figure 24
1083 Swirl diffuser
1084
1085
1086
1087 6.3. Unidirectional airflow
1088
1089 6.3.1. Unidirectional airflow (UDAF) should be used for weighing booths
1090 or sampling booths to provide operator and product protection and should
1091 also have a slight air in-flow from the room to enhance containment. Dust
1092 containment at the weigh booth should be demonstrated by smoke airflow
1093 pattern tests, or other appropriate tests. UDAF can also be used to provide
1094 protection of other dusty processes.
1095
1096 6.3.2. Sampling of materials such as starting materials, primary packaging
1097 materials and products, should be carried out in the same environmental
1098 conditions that are required for the further processing of the product.
1099
1100 6.3.3. In a weighing booth situation, the aim of the UDAF is to provide
1101 dust containment and operator protection.
1102
1103 Example: in Figure 25 the dust generated at the weighing station is
1104 immediately extracted through the perforated worktop, thus protecting the
1105 operator from dust inhalation, but at the same time protecting the product
1106 from contamination by the operator by means of the vertical unidirectional
1107 airflow stream.
page 45
1108 Figure 25
1109 Operator protection at weighing station
1110
1111 6.3.4 The unidirectional flow velocity should be such that it does not
1112 disrupt the sensitivity of balances in weighing areas. Where necessary the
1113 velocity may be reduced to prevent inaccuracies during weighing,
1114 provided that sufficient airflow is maintained to provide containment.
1115 Conventional unidirectional airflow systems, where a Grade A condition is
1116 required, have a guidance airflow velocity of 0.36 to 0.54 m/s. However, in
1117 a weigh booth or sampling booth a lower velocity can be used as a Grade
1118 A condition is not required. It is often necessary to reduce velocities to a
1119 lower level in order not to influence balance readings. The airflow velocity
1120 and directional flow should still ensure product containment. For this type
1121 of application it is sometimes better to refer to the unit as an airflow
1122 protection booth (APB) rather than a UDAF, in order to avoid confusion,
1123 with a Grade A requirement. To assist with containment for weighing and
1124 sampling operations there should be a slight inflow of air into the UDAF
1125 protected zone from the surrounding room in order to prevent dust
1126 escaping. Thus the amount of air extracted from below the UDAF/APB
1127 should exceed the amount of air supplied. (Unless the hazardous or sterile
1128 nature of the product dictates otherwise.)
page 46
1129 6.3.5 The position in which the operator stands relative to the source of
1130 dust liberation and airflow should be determined to ensure that the operator
1131 is not in the path of an airflow that could lead to contamination of the
1132 product (Figure 26).
1133 Figure 26
1134 Operator protection by horizontal airflow
1135
1136
1137
1138 6.3.6 Once the system has been designed and qualified with a specific
1139 layout for operators and processes, this should be maintained in accordance
1140 with an SOP.
1141
1142 6.3.7 There should be no obstructions in the path of a unidirectional flow
1143 air stream that may cause the operator to be exposed to dust.
1144
1145 Figure 27 illustrates the incorrect use of a weighing scale which has a solid
1146 back. The back of the weighing scale should not block the return air path
1147 as this obstructs the airflow and causes air to rise vertically carrying dust,
1148 resulting in a hazardous situation for the operator.
1149
1150 Figure 28 illustrates a situation where an open bin is placed below a vertical
1151 unidirectional flow distributor. The downward airflow should be prevented
1152 from entering the bin, and then being forced to rise again, as this would
1153 carry light dust up towards the operator’s face. In such an occurrence it
page 47
1154 may be necessary to add a partial cover over the bin to limit the entry of
1155 air. Point extraction could also be used but this can result in the excessive
1156 loss of product.
1157
1158 Figure 2 9 shows that a solid worktop can sometimes cause deflection of
1159 the vertical unidirectional airflow resulting in a flow reversal. A possible
1160 solution would be to have a 100 mm gap between the back of the table and
1161 the wall, with the air being extracted here.
1162
1163 Figure 27
1164 Operator subject to powder inhalation due to obstruction
1165
1166
1167
1168
1169
1170
page 48
1171 Figure 28
1172 Operator subject to powder contamination due to airflow reversal in
1173 bin
1174
1175
1176
1177
page 49
1178 Figure 29
1179 Operator subject to powder inhalation due to worktop obstruction
1180
1181
1182 6.3.8 The manufacturer should select either vertical or horizontal
1183 unidirectional airflow (Figure 30) and an appropriate airflow pattern to
1184 provide the best protection for the particular application.
1185
1186
page 50
1187 Figure 30
1188 Diagram indicating horizontal and vertical unidirectional flow
1189
1190
1191
1192
1193
1194
page 51
1195 6.3.9 Return or exhaust air grilles in rooms or at weigh or sampling booths
1196 should preferably be of the perforated grille types, which are easy to clean.
1197 Return/exhaust air filters can either be installed at the room terminal or in
1198 the air-handling unit. Maintenance and cleaning of filters and ducts should
1199 be addressed to ensure constant airflow.
1200
1201
6.4.
1202 Infiltration
1203
1204 6.4.1. Air infiltration of unfiltered air into a pharmaceutical plant should
1205 not be a source of contamination.
1206
1207 6.4.2. Manufacturing facilities should normally be maintained at a positive
1208 pressure relative to the outside, to limit the ingress of contaminants. Where
1209 facilities are to be maintained at negative pressures relative to the ambient
1210 pressure, special precautions should be taken. Refer to the WHO
1211 guidelines for hazardous products, for further guidance on negative
1212 pressure facilities.
1213
1214 6.4.3. The location of the negative pressure facility should be carefully
1215 considered with reference to the areas surrounding it, particular attention
1216 being given to ensuring that the building structure is well sealed.
1217
1218 6.4.4. Negative pressure zones should, as far as possible, be encapsulated
1219 by surrounding areas with clean air supplies, so that only clean air can
1220 infiltrate into the controlled zone.
1221
6.5.
1222 Cross-contamination
1223
1224 6.5.1. Where different products are manufactured at the same time, in
1225 different areas or cubicles, in a multiproduct OSD manufacturing site,
1226 measures should be taken to ensure that dust cannot move from one cubicle
1227 to another.
1228
1229 6.5.2. Correct directional air movement and a pressure cascade system
1230 can assist in preventing cross-contamination. The pressure cascade should
1231 be such that the direction of airflow is from the clean corridor into the
1232 cubicles, resulting in dust containment.
1233
1234 6.5.3. For cubicles where dust is liberated, the corridor should be
1235 maintained at a higher pressure than the cubicles and the cubicles at a
1236 higher pressure than atmospheric pressure.
1237
page 52
1238 6.5.4. Containment can normally be achieved by application of the

1239 displacement concept (low pressure differential, high airflow), or the
1240 pressure differential concept (high pressure differential, low airflow), or the
1241 physical barrier concept.
1242
1243 6.5.5. The pressure cascade regime and the direction of airflow should be
1244 appropriate to the product and processing method used.
1245
1246 6.5.6. Highly potent products should be manufactured under a pressure
1247 cascade regime that is negative relative to atmospheric pressure.
1248
1249 6.5.7. The pressure cascade for each facility should be individually
1250 assessed according to the product handled and level of protection required.
1251
1252 6.5.8. Building structure should be given special attention to accommodate
1253 the pressure cascade design.
1254
1255 6.5.9. Ceilings and walls, close fitting doors and sealed light fittings should
1256 be in place, to limit ingress or egress of air.
1257
6.6.
1258 Displacement concept (low pressure differential, high airflow)
1259
1260 Note: This method of containment is not the preferred method, as the
1261 measurement and monitoring of airflow velocities in doorways is difficult.
1262 This concept is commonly found in production processes where large
1263 amounts of dust are generated.
1264
1265 6.6.1. Under this concept the air should be supplied to the corridor,
1266 flow through the doorway, and be extracted from the back of the cubicle.
1267 Normally the cubicle door should be closed and the air should enter the
1268 cubicle through a door grille, although the concept can be applied to an
1269 opening without a door. (With the door open the pressure differential will
1270 effectively be zero.)
1271
1272 6.6.2. The velocity should be high enough to prevent turbulence within the
1273 doorway resulting in dust escaping.
1274
1275 6.6.3. This displacement airflow should be calculated as the product of
1276 the door area and the velocity through the doorway, which generally
1277 results in fairly large air quantities.
1278
1279 Note: Although this method of containment may still exist on older facilities,
1280 it is not the preferred method, as the measurement and monitoring of doorway
1281 velocities is difficult. In addition, simultaneously maintaining the correct
1282 room pressure and the correct room air change rate is often not achieved.
page 53
1283
1284
6.7.
1285 Pressure differential concept (high pressure differential, low airflow)
1286
1287 Note: The pressure differential concept may normally be used in zones where
1288 little or no dust is being generated. It may be used alone or in combination
1289 with other containment control techniques and concepts, such as a double
1290 door airlock.
1291
1292 6.7.1. The high pressure differential between the clean and less clean
1293 zones should be generated by leakage through the gaps of the closed doors
1294 to the cubicle.
1295
1296 6.7.2. The pressure differential should be of sufficient magnitude to ensure
1297 containment and prevention of flow reversal, but should not be so high as to
1298 create turbulence problems.
1299
1300 6.7.3. In considering room pressure differentials, transient variations, such
1301 as machine extract systems, should be taken into consideration.
1302
1303 6.7.4. A pressure differential of 15 Pa is often used for achieving
1304 containment between two adjacent zones, but pressure differentials of
1305 between 5 Pa and 20 Pa may be acceptable. Where the design pressure
1306 differential is too low and tolerances are at opposite extremities, a flow
1307 reversal can take place. For example, where a control tolerance of ± 3 Pa
1308 is specified, the implications of adjacent rooms being operated at the upper
1309 and lower tolerances should be evaluated. It is important to select pressures
1310 and tolerances such that a flow reversal is unlikely to occur.
1311
1312 6.7.5. The pressure differential between adjacent rooms could be
1313 considered a critical parameter, depending on the outcome of risk analysis.
1314 The limits for the pressure differential between adjacent areas should be
1315 such that there is no risk of overlap in the acceptable operating range, e.g.
1316 5 Pa to 15 Pa in one room and 15 Pa to 30 Pa in an adjacent room, resulting
1317 in the failure of the pressure cascade, where the first room is at the maximum
1318 pressure limit and the second room is at its minimum pressure limit.
1319
1320 6.7.6. Low p r e s s u r e d i f f e r e n t i a l s m a y b e a c c e p t a b l e
1321 w h e n a i r l o c k s (pressure sinks or pressure bubbles) are used to
1322 segregate areas.
1323
1324 6.7.7. The effect of room pressure tolerances are illustrated in Figure 31.
1325 If one room is at the higher side of the tolerance and the other at the lower
1326 side of the tolerance, it could result in either a high or a low pressure
1327 differential. When setting tolerances it is also important to specify if the
1328 tolerance is applicable to the absolute room pressures or the pressure
1329 differentials. In the diagram below the tolerances have been based on a ±
page 54
1330 3 Pa tolerance on absolute room pressures, resulting in pressure

1331 differential variances of between 21 and 9 Pa. . For a room pressure
1332 differential of 15 Pa and a tolerance based on ± 3 Pa of differential
1333 pressure, then the resultant variances would only be between 12 and 18 Pa.
1334
1335 Figure 31
1336 Examples of pressure cascades
1337
Air Leakage
Air Leakage
Air Leakage
Image of room pressure gauge

indicating colour coded normal,
alert & action parameters
Air Leakage
Air Leakage
Air Leakage
Air Leakage
Air Leakage
Air Leakage
1338
1339
1340 6.7.8. The pressure control and monitoring devices used should be
1341 calibrated and qualified. Compliance with specifications should be regularly
1342 verified and the results recorded. Pressure control devices should be linked
1343 to an alarm system set according to the levels determined by a risk analysis.
1344
1345 6.7.9. Manual control systems, where used, should be set up during
1346 commissioning, with set point marked, and should not change unless other
1347 system conditions change.
1348
1349 6.7.10. Airlocks can be important components in setting up and maintaining
1350 pressure cascade systems and also to limit cross-contamination.
1351
1352 6.7.11. Airlocks with different pressure cascade regimes include the
1353 cascade airlock, sink airlock and bubble airlock (Figures 32–34):
1354
1355 • cascade airlock: higher pressure on one side of the airlock and
1356 lower pressure on the other;
1357 • sink airlock: lower pressure inside the airlock and higher pressure on
page 55
1358 both outer sides;

1359 • bubble airlock: higher pressure inside the airlock and lower
1360 pressure on both outer sides.
1361
1362 Figure 32
1363 Example of cascade airlock
1364 (In most cases the internal pressure of the airlock is not critical. The
1365 pressure differential between the two outer sides is the important criteria.)
1366
1367
1368
1369
page 56
1370 Figure 33
1371 Example of sink airlock
1372
1373
1374
page 57
1375
1376 Figure 34
1377 Example of bubble airlock
1378
Air Leakage Air Leakage
15 Pa Material Airlock
15 Pa
30 Pa
Bubble Airlock
1379
1380
1381 Note: The diagrams above and the differential pressures shown here are
1382 for illustration purposes only. Pressures indicated in these examples are
1383 absolute pressures, whereas the local pressure indication would most likely
1384 be pressure differential from room to room.
1385
1386 6.7.12. Doors should open to the high pressure side, so that room pressure
1387 assists in holding the door closed and in addition be provided with self-
1388 closers. Should the doors open to the low pressure side, the door closer
1389 springs should be sufficient to hold the door closed and prevent the pressure
1390 differential from pushing the door open. There should be a method to
1391 indicate if both doors to airlocks are open at the same time, or alternatively
1392 these should be interlocked. The determination of which doors should be
1393 interlocked should be the subject of a risk assessment study.
1394
1395 6.7.13. Central dust extraction systems should be interlocked with the
1396 appropriate air-handling systems, to ensure that they operate simultaneously.
1397
1398 6.7.14. Room pressure differential between adjacent cubicles, which are
1399 linked by common dust extraction ducting, should be avoided.
1400
page 58
1401 6.7.15. Air should not flow through the dust extraction ducting or return
1402 air ducting from the room with the higher pressure to the room with the
1403 lower pressure (this would normally occur only if extract or return systems
1404 were inoperative). Systems should be designed to prevent dust flowing back
1405 in the opposite direction in the event of component failure or airflow failure.
1406
1407 6.7.16. Adequate room pressure differential indication should be provided
1408 so that each critical room pressure can be traced back to ambient pressure
1409 (by summation of the room pressure differentials), in order to determine the
1410 room actual absolute pressure, such as a pressure gauge installed to
1411 indicate the pressure differential from a central corridor to the outside.
1412 Room pressure indication gauges should have a range and graduation scale
1413 which enables the reading to an appropriate accuracy. Normal operating
1414 range, alert and action limits should be defined and displayed at the point of
1415 indication. A colour coding of these limits on the gauge may be helpful.
1416
1417 Room pressure indication may be either analogue or digital, and may be
1418 represented as either pressure differentials or absolute pressures.
1419 Whichever system is used any out-of-specification condition should be
1420 easily identifiable.
1421
1422 6.7.17. Material pass-through-hatches (PTH) or pass boxes (PB) can also
1423 be used for separating two different zones. PTHs fall into two categories,
1424 namely a dynamic PTH or a passive PTH. Dynamic PTHs have an air
1425 supply to or extraction from them, and can then be used as bubble, sink or
1426 cascade PTHs.
1427
1428 6.7.18. Room pressure differential tolerances should always be set with a
1429 maximum and minimum tolerance. Setting tolerances as NMT or NLT can
1430 easily lead to an OOS condition.
1431
6.8.
1432 Physical barrier concept
1433
1434 6.8.1. Where appropriate, an impervious barrier to prevent cross-
1435 contamination between two zones, such as closed systems, pumped or
1436 vacuum transfer of materials, should be used.
1437
1438
6.9.
1439 Temperature and relative humidity
1440
1441 6.9.1. Where appropriate, temperature and relative humidity should be
1442 controlled, monitored and recorded, where relevant, to ensure compliance
1443 with requirements pertinent to the materials and products and provide a
1444 comfortable environment for the operator where necessary.
1445
page 59
1446 6.9.2. Maximum and minimum room temperatures and relative humidity
1447 should be appropriate. Alert and action limits on temperatures and
1448 humidities should be set, as appropriate.
1449
1450 6.9.3. The operating band, or tolerance, between the acceptable minimum
1451 and maximum temperatures should not be made too close. Tight control
1452 tolerances may be difficult to achieve and can also add unnecessary
1453 installation and running costs.
1454
1455 6.9.4. Cubicles, or suites, in which products requiring low relative humidity
1456 are processed, should have well-sealed walls and ceilings and should also
1457 be separated from adjacent areas with higher relative humidity by means of
1458 suitable airlocks.
1459
1460 6.9.5. Precautions should be taken to prevent moisture migration that
1461 increases the load on the HVAC system.
1462
1463 6.9.6. Humidity control should be achieved by removing moisture from
1464 the air, or adding moisture to the air, as relevant.
1465
1466 6.9.7. Dehumidification (moisture removal) may be achieved by means of
1467 either refrigerated dehumidifiers or chemical dehumidifiers.
1468
1469 6.9.8. Appropriate cooling media for dehumidification such as low
1470 temperature chilled water/glycol mixture or refrigerant should be used.
1471
1472 6.9.9. Humidifiers should be avoided if possible as they may become a
1473 source of contamination (e.g. microbiological growth). Where humidification
1474 is required, this should be achieved by appropriate means such as the injection
1475 of steam into the air stream. A product-contamination assessment should be
1476 done to determine whether pure or clean steam is required for the purposes
1477 of humidification.
1478
1479 6.9.10. Where steam humidifiers are used, chemicals such as corrosion
1480 inhibitors or chelating agents, which could have a detrimental effect on
1481 the product, should not be added to the boiler system. Only appropriate
1482 additives should be added to the boiler system.
1483
1484 6.9.11. Humidification systems should be well drained. No condensate
1485 should accumulate in air-handling systems.
1486
1487 6.9.12. Other humidification appliances such as evaporative systems,
1488 atomizers and water mist sprays, should not be used because of the potential
1489 risk of microbial contamination.
1490
1491 6.9.13. Duct material in the vicinity of the humidifier should not add
1492 contaminants to air that will not be removed by filtration further downstream.
1493
page 60
1494 6.9.14. Air filters should not be installed immediately downstream of

1495 humidifiers, as moisture on the filters could lead to bacterial growth.
1496
1497 6.9.15. Cold surfaces should be insulated to prevent condensation within
1498 the clean area or on air-handling components.
1499
1500 6.9.16. When specifying relative humidity, the associated temperature
1501 should also be specified.
1502
1503 6.9.17. Chemical driers using silica gel or lithium chloride are acceptable,
1504 provided that they do not become sources of contamination.
1505
15067. Dust control
1507
1508 7.1. Wherever possible, dust or vapour contamination should be
1509 removed at source. Point-of-use extraction, i.e. as close as possible to the
1510 point where the dust is generated, should be employed. Spot ventilation or
1511 capture hoods may be used as appropriate. The HVAC system should not
1512 serve as the primary mechanism of dust control.
1513
1514 7.2. Point-of-use extraction should be either in the form of a fixed,
1515 high-velocity extraction point or an articulated arm with movable hood or a
1516 fixed extraction hood.
1517
1518 7.3. Dust extraction ducting should be designed with sufficient transfer
1519 velocity to ensure that dust is carried away, and does not settle in the ducting.
1520 Periodic checks should be performed to ensure that there is no build-up of
1521 the dust in the ducting.
1522
1523 7.4. The required transfer velocity should be determined: it is dependent
1524 on the density of the dust (the denser the dust, the higher the transfer
1525 velocity should be, e.g. 15–20 m/s).
1526
1527 7.5. Airflow direction should be carefully chosen, to ensure that the
1528 operator does not contaminate the product, and also so that the operator is
1529 not put at risk by the product.
1530
1531 7.6. Point extraction alone is usually not sufficient to capture all of
1532 the contaminants, and general directional airflow should be used to assist
1533 in removing dust and vapours from the room.
1534
1535 7.7. Typically, in a room operating with turbulent airflow, the air should
1536 be introduced from ceiling diffusers, located at the door entry side of the
1537 room and extracted from the rear of the room at low level to help give a
1538 flushing effect in the room. Correct flushing of the rooms may be verified
1539 by airflow visualization smoke tests.
1540
page 61
1541 7.8. When dealing with particularly harmful products, additional steps,
1542 such as handling the products in glove boxes or using barrier isolator
1543 technology, should be used.
1544
15458. Protection of the environment
1546
8.1.
1547 General
1548
1549 8.1.1. It should be noted that protection of the environment is not addressed
1550 in this guideline, and discharges into the atmosphere should be compliant
1551 with relevant local and national environmental legislation and standards.
1552
1553 8.1.2. Dust, vapours and fumes could be possible sources of contamination;
1554 therefore, care should be taken when deciding on the location of the inlet
1555 and exhaust points relative to one other.
1556
8.2.
1557 Dust in exhaust air
1558
1559 8.2.1. Exhaust air discharge points on pharmaceutical equipment and
1560 facilities, such as from fluid bed driers and tablet-coating equipment, and
1561 exhaust air from dust extraction systems, carry heavy dust loads and should be
1562 provided with adequate filtration to prevent contamination of the ambient air.
1563
1564 8.2.2. Where the powders are not highly potent, final filters on a dust
1565 exhaust system should be fine dust filters with a filter classification of F9
1566 according to EN 779 filter standards.
1567
1568 8.2.3. Where reverse-pulse dust collectors are used for removing dust from
1569 dust extraction systems, they should usually be equipped with cartridge
1570 filters containing a compressed air lance, and be capable of continuous
1571 operation without interrupting the airflow.
1572
1573 8.2.4. Alternative types of dust collectors (such as those operating with a
1574 mechanical shaker, requiring that the fan be switched off when the mechanical
1575 shaker is activated) should be used in such a manner that there is no risk
1576 of cross-contamination. There should be no disruption of airflow during a
1577 production run as the loss of airflow could disrupt the pressure cascade.
1578
1579 8.2.5. Mechanical-shaker dust collectors should not be used for applications
1580 where continuous airflow is required, in order to avoid unacceptable
1581 fluctuations in room pressures, except in the case where room pressures are
1582 automatically controlled.
1583
1584 8.2.6. When wet scrubbers are used, the dust-slurry should be removed by
1585 a suitable means, e.g. a drainage system or waste removal contractor.
1586
page 62
1587 8.2.7. The quality of the exhaust air should be determined to see whether the
1588 filtration efficiency is adequate with all types of dust collectors and wet
1589 scrubbers.
1590
1591 8.2.8. Where necessary, additional filtration may be provided downstream
1592 of the dust collector.
1593
8.3.
1594 Vapour and fume removal
1595
1596 8.3.1. Vapour should be extracted at the point of generation. When planning
1597 the system for the extraction of residual vapours, the density of the vapour
1598 should be taken into account. If the vapour is lighter than air, the extract
1599 grilles should be at a high level, or possibly at both high and low levels.
1600
1601 8.3.2. The systems for fume, dust and effluent control should be designed,
1602 installed and operated in such a manner that they do not become possible
1603 sources of contamination or cross-contamination, e.g. an exhaust-air
1604 discharge point located close to the HVAC system fresh air inlet.
1605
1606 8.3.3. Fumes should be removed by means of wet scrubbers or dry
1607 chemical scrubbers (deep-bed scrubbers).
1608
1609 8.3.4. Wet scrubbers for fume removal normally require the addition of
1610 various chemicals to the water to increase the adsorption efficiency.
1611
1612 8.3.5. Deep-bed scrubbers should be designed with activated carbon filters
1613 or granular chemical adsorption media. The chemical media for deep-bed
1614 scrubbers should be specific to the effluent being treated.
1615
1616 8.3.6. The type and quantity of the vapours to be removed should be
1617 known to enable the appropriate filter media, as well as the volume of media
1618 required to be determined.
1619
1620
1621 9. Commissioning, qualification and validation
1622
9.1.
1623 General
9.2.
1624
1625 9.2.1. The heating, ventilation and air-conditioning (HVAC) system plays
1626 an important role in the protection of the product, the personnel and the
1627 environment.
1628
1629 9.2.2. For all HVAC installation components, subsystems or parameters,
1630 critical parameters and non-critical parameters should be determined.
1631
1632 9.2.3. Some of the parameters of a typical HVAC system that should be
1633 qualified include:
page 63
1634
1635 • room temperature and humidity;
1636 • supply air and return air quantities;
1637 • room pressure, air change rate, flow patterns, particle count and clean-up
1638 rates;
1639 • unidirectional flow velocities and HEPA filter penetration tests; and
1640 • critical alarms, etc.
1641
1642 9.3. Commissioning
1643
1644 9.3.1. Commissioning should involve the setting up, balancing,
1645 adjustment and testing of the entire HVAC system, to ensure that the
1646 system meets all the requirements, as specified in the user requirement
1647 specification, and capacities as specified by the designer or developer. The
1648 commissioning plan should start at the early stages of a project so that it
1649 can be integrated with qualification and verification procedures.
1650
1651 9.3.2. Acceptable tolerances for all system parameters should be specified
1652 and agreed by the user prior to commencing the physical installation.
1653 These tolerances should be specified in the User Requirement
1654 Specifications
1655
1656 9.3.3. Acceptance criteria should be set for all system parameters. The
1657 measured data should fall within the acceptance criteria.
1658
1659 9.3.4. System installation records should provide documented evidence
1660 of all measured capacities of the system.
1661
1662 9.3.5. The i n s t a l l a t i o n r e c o r d s should include items such as the
1663 design and measured figures for airflows, water flows, system pressures
1664 electrical amperages, etc. These should be contained in the operating
1665 and maintenance manuals (O & M manuals). The installation records of
1666 the system should provide documented evidence of all measured capacities
1667 of the system.
1668
1669 9.3.6. Typical information that should be contained in the O&M
1670 manuals is the following:
1671
1672
• system description;
1673
• operating instructions,
1674
• trouble shooting;
page 64
1675
• commissioning data schedules;
1676
• maintenance instructions;
1677
• list of equipment suppliers;
1678
• spare parts lists;
1679
• equipment capacity and data schedules;
1680
• supplier’s literature;
1681
• control system operation;
1682
• electrical drawings;
1683
• as-built drawings;
1684
• maintenance records.
1685
1686 9.3.7. O & M manuals, schematic drawings, protocols and reports should
1687 be maintained as reference documents for any future changes and upgrades
1688 to the system. As-built drawings should be available and should be kept up
1689 to date with all the latest system changes. Any changes from the originally
1690 approved system should be covered by change control documentation and
1691 risk assessment studies where deemed necessary.
1692
1693 9.3.8. Training should be provided to personnel after installation of the
1694 system, and should include how to perform operation and maintenance.
1695
1696 9.3.9. Commissioning should be a precursor to system qualification and
1697 validation.
9.4.
1698
9.5.
1699 Qualification
9.6.
1700
1701 9.6.1. Manufacturers should qualify HVAC systems using a risk-based
1702 approach. The basic concepts of qualification of HVAC systems are set out
1703 in Figure 35 below.
1704
1705
page 65
1706 Figure 35
1707 Qualification is a part of validation
1708
1709 9.6.2. The qualification of the HVAC system should be described in a

1710 validation master plan (VMP), or a subsection of the VMP.
1711
1712 9.6.3. The validation master plan should define the nature and extent of
1713 testing and the test procedures and protocols to be followed.
1714
1715 9.6.4. Stages of the qualification of the HVAC system should include
1716 design qualification (DQ), installation qualification (IQ), operational
1717 qualification (OQ) and performance qualification (PQ). The relationship
1718 between the development stage of a project and the qualification/validation
1719 stage is given in the V-Diagram (Figure 36) below.
1720
page 66
1721 Figure 36
V-Model for Direct Impact Systems
PQ Test Plan
User Requirement Performance
Specification Qualification
(incl. UAT)
OQ Test Plan
Functional Design Operational
Qualification
Specification Qualification
Design
(incl. FAT)
DQ Test Plan Detail Design and IQ Test Plan

Installation
Configuration
Qualification
Specifications
(incl. PDI)
Build & Project

Implementation
1722
1723 9.6.5. Critical and non-critical parameters for all HVAC installation
1724 components, subsystems and controls should be determined by means of a
1725 risk analysis.
1726
1727 9.6.6. Any parameter that may affect the quality of the pharmaceutical
1728 product should be considered a critical parameter.
1729
1730 9.6.7. All critical parameters should be included in the qualification
1731 process.
1732
1733 Note: A realistic approach to differentiating between critical and
1734 noncritical parameters, systems or components is required, to avoid
1735 making the validation process unnecessarily complex.
1736 Example
1737 • The humidity of the room where the product is exposed should be
1738 considered a critical parameter when a humidity-sensitive product
1739 is being manufactured. The humidity sensors and the humidity
1740 monitoring system should, therefore, be qualified. Components or
1741 equipment such as the heat transfer system, chemical drier or
1742 steam humidifier, which is producing the humidity-controlled air, is
page 67
1743 further removed from the product and may not require operational
1744 qualification.
1745 • A room cleanliness classification is a critical parameter and,
1746 therefore, the room air-change rates and high-efficiency particulate
1747 air (HEPA) filters should be considered critical parameters and
1748 components, and therefore require qualification. Items such as the
1749 fan generating the airflow and the primary and secondary filters
1750 are considered non-critical components, and may not require
1751 operational qualification.
1752
1753 9.6.8. Non-critical systems and components should be subject to
1754 verification by good engineering practice (GEP) and may not necessarily
1755 require full qualification.
1756
1757 9.6.9. A change control procedure should be followed when changes are
1758 planned to the HVAC system, its components and controls, that may affect
1759 critical parameters.
1760
1761 9.6.10. The design condition, normal operating ranges, operating range
1762 and alert and action limits should be defined and be realistic. Alert limits
1763 should be based on system capability.
1764 .
1765 9.6.11. All parameters should fall within the design condition range
1766 during system operational qualification. Conditions may go out of the
1767 design condition range during normal operating procedures but they should
1768 remain within the operating range.
1769
1770 9.6.12. Out-of-limit results (e.g. alert or action limit deviations) should be
1771 recorded and form part of the batch manufacturing records, and their
1772 impact should be investigated.
1773
1774 9.6.13. The relationships between design conditions, operating range and
1775 qualified acceptance criteria are given in Figure 37. There should be SOPs
1776 to determine action to be taken when alert and action limits are reached.
1777
page 68
1778 Figure 37
1779 System operating ranges
1780
1781
1782 9.6.14. A narrow range of relative humidities coupled with a wide range of
1783 temperatures is unacceptable as changes in temperature will automatically
1784 give rise to variations in the relative humidity.
1785
1786 9.6.15. Some of the typical HVAC system parameters, based on a risk
1787 assessment, that should be qualified for a pharmaceutical facility may
1788 include:
1789
1790 — temperature
1791 — relative humidity
1792 — supply air quantities for all diffusers
1793 — return air or exhaust air quantities
1794 — room air-change rates
1795 — room pressures (pressure differentials)
1796 — room airflow patterns
1797 — unidirectional flow velocities
1798 — containment system velocities
1799 — HEPA filter penetration tests
1800 — room particle counts
1801 — room clean-up or recovery rates
1802 — microbiological air and surface counts where appropriate
1803 — operation of de-dusting
1804 — warning/alarm systems where applicable.
1805
1806 9.6.16. The maximum time interval between tests and re-qualification
1807 should be defined by the manufacturer. The type of facility under test and
1808 the product level of protection should be considered.
1809
page 69
1810 Note: Table 5 gives intervals for reference purposes only. The actual test
1811 periods may be more or less frequent, depending on the product and process
1812 and subject to risk assessment.
1813 Table 5
1814 Strategic tests to demonstrate continued compliance
1815 (Time intervals given for re-qualification are for reference purposes only.
1816 The actual tests required will depend on specific facility requirements)
Test parameter Max. time Test procedure

interval
between tests
(all classes)
Particle count test 6 months Dust particle counts to be carried

(verification of (ISO 5) out and result printouts produced.
cleanliness)
12 months No. of readings and positions of
(>ISO 5) tests to be in accordance with ISO
14644-1 Annex B.2
Air pressure Log of pressure differential
difference (to verify 12 months readings to be produced - critical
absence of cross- plants should be logged daily,
contamination) preferably continuously. In
accordance with ISO 14644-3
Airflow volume Airflow readings for supply air and
return air grilles to be measured
(to verify air change 12 months and air change rates to be
rates)
calculated. In accordance with
ISO 14644-3 Annex B.4
Airflow velocity Air velocities for containment
systems and unidirectional flow
(to verify 12 months protection systems to be measured.
unidirectional flow or In accordance with ISO 14644-3
containment Annex B.4
conditions)
page 70
HEPA filter leakage Filter penetration tests to be carried

tests out by a competent person to
12 months demonstrate filter media, filter seal
(to verify filter
and filter frame integrity.. In
integrity)
Annex B.6
Containment leakage Demonstrate that contaminant is
(to verify absence of maintained within a room by
cross-contamination) means of:
12 months • airflow direction smoke tests
• room air pressures. In
Annex B.13
Recovery Test to establish time that a
12 months cleanroom takes to recover from a
(to verify clean-up contaminated condition to the
time)
specified cleanroom condition. In
Room temperatures Demonstrate that room
(to verify temperature temperatures at determined
tolerance adherence) 12 months locations comply with specified
tolerances. In accordance with
ISO 14644-3 Annex B.8.2
Warehouse and store Demonstrate that store

temperatures temperatures are uniform within
specified tolerances
(to verify temperature
mapping conditions) In accordance with WHO 45th
36 months report (WHO Technical Report
Series, No. 961), Annex 9 and
WHO 49th report (WHO Technical
Report Series, No. 992), Annex 5
plus Supplements 1 to 16
Room Humidities Demonstrate that room humidities
at determined locations comply
(To verify humidity 12 months with specified tolerances. In
tolerance adherence) accordance with ISO 14644-3
Annex B.9.2
page 71
1817 9.6.17. Requalification should also be done when any change, which could
1818 affect system performance, takes place.
1819
1820 9.6.18. Room clean-up or recovery tests are performed to determine
1821 whether the installation is capable of returning to a specified cleanliness
1822 level within a finite time, after being exposed briefly to a source of
1823 airborne particulate challenge.
1824
1825 Room clean-up or recovery tests should demonstrate a change in particle
1826 concentration by a factor of 100 within the prescribed time (as per ISO
1827 14644-3 clause B.12) (3). The guidance time period for clean-up or
1828 recovery is about 15 to 20 minutes.
1829 In some instances it is not possible to increase the concentration by a

1830 factor of 100 (such as for an ISO 14644 Class 8 condition) as the high
1831 particle concentration can damage the particle counter. In this instance the
1832 particle decay method can be used as per ISO 14644-3 clause B.12.3.2.
1833 9.6.19. If energy-saving procedures such as reducing the airflow during

1834 non-production hours are used, precautionary measures should be in place
1835 to ensure that the systems are not operated outside the defined relevant
1836 environmental conditions.
1837
1838 These precautionary measures should be based on a risk assessment to

1839 ensure that there is no negative impact on the quality of the product.
1840 Qualification tests should be carried out to demonstrate that there are no
1841 flow reversals, loss of room pressurisation cascade, temperature, humidity
1842 excursions, etc.
1843 9.6.20. Additional documents that should be included in the qualification

1844 manuals should include system airflow schematics, room pressure cascade
1845 drawings, zone concept drawings, air-handling system allocation drawings,
1846 particle count mapping drawings, etc.
1847
1848
page 72
1849 9.7. Supplementary notes on test procedures

1850
1851 9.7.1. General
1852
1853 9.7.1.1. Tests should be carried as described in ISO 14644-3.
1854 However below are some supplementary notes and aspects that provide
1855 additional guidance.
1856
1857 9.7.2. Airflow measurements
1858
1859 9.7.2.1. The ISO 14644 method - "B.4.3.3 Supply airflow rate
1860 calculated from filter face velocity" - should not be used to measure the
1861 airflow at diffuser outlets. The diffuser air directional blades or swirl
1862 outlets result in highly inaccurate measurements.
1863
1864 9.7.2.2. The cone and anemometer method is more accurate. Other
1865 methods can be used such as volume flow regulators with built in orifice
1866 and pressure differential ports, whereby airflow can be read off a graph
1867 from the corresponding pressure differentials.
1868
1869 9.7.3. Non-viable air particle counts
1870
1871 9.7.3.1. Particle count test results should be calculated using the
1872 UCL (upper confidence level) formulas as described in ISO 14644-3, if
1873 there are up to nine locations. The practice of using the average value of
1874 all particle count readings as the pass criteria is not acceptable.
1875
1876 9.7.3.2. Ensure that the test certificate states the condition under
1877 which the test was taken i.e. “as built”, “at-rest” or “operational”. The
1878 operational condition should be clearly defined for each room. (For
1879 example: number of staff, staff locations, manner of equipment operating,
1880 etc.)
1881
1882 9.7.3.3. ISO 14644-1 clause B.4.3.42 states that when only one test
1883 location is determined by formula, “take a minimum of three single sample
1884 volumes (B.4.2) at that location.” A more representative result would be
1885 obtained by taking a single sample at each of three different locations in
1886 that room. The actual locations should be based on a risk assessment.
1887
1888 9.7.3.4. In addition to determining the number of the sampling
1889 locations based on the area of the clean room, a risk assessment should
page 73
1890 determine if additional sample locations are warranted. Consider aspects

1891 such as personnel and/or production activities and air flow dead spots.
1892
1893 9.7.3.5. Where a UDAF is located within a room the UDAF and its
1894 background environment should be considered separately in terms of
1895 sampling location calculations, and should be individually certified.
1896
1897 9.7.3.6. The mapping drawing indicating test location should be
1898 included with the test certificate, and the same mapping locations should
1899 be used for future tests for comparative purposes.
1900
1901 9.7.4. HEPA filter integrity tests
1902
1903 9.7.4.1. Filter media, frame and seal should be tested for each filter
1904 and results for media, frame and seal penetration reflected separately on
1905 the test certificates.
1906
1907 9.7.4.2. When HEPA filters are terminally mounted at the room, it
1908 should be possible to carry out filter integrity tests from within the room.
1909 The filter housings will therefore require ports for measuring appropriate
1910 upstream concentration and penetration concentration from within the
1911 room. In addition it should be possible to measure the filter pressure drop
1912 in individual HEPA filters, also preferably from within the room. These
1913 pressure drops should be recorded on the filter test certificate as an
1914 indication of the filter life. (The practice of measuring the appropriate
1915 upstream concentration from the ceiling void or at the air handling plant-
1916 room, and then measuring the filter penetration concentration in the room
1917 is unacceptable. The time lag between measuring the upstream
1918 concentration and the penetration concentration could mean that by the
1919 time the room penetration is measured, the upstream concentration is no
1920 longer the required concentration.)
1921
1922 9.7.4.3. The test procedure should not compromise the quality of
1923 the product.
1924
1925
1926 10. Maintenance
1927
10.1.
1928 Maintenance records, maintenance procedures and Operating &
1929 Maintenance Manuals should be sufficient to indicate that the company
1930 has control over the HVAC systems. There should be a planned
1931 preventive maintenance programme, procedures and records for the HVAC
page 74
1932 system. Records should be kept for a sufficient length of time should they
1933 be required for any product defect analysis.
1934
10.2.
1935 O&M manuals, schematic drawings, protocols and reports should be
1936 maintained as reference documents for any future changes and upgrades
1937 to the system. These documents should be kept up to date, containing
1938 any system revisions made.
1939
10.3.
1940 The O&M manuals should typically contain the following information:
1941 system description, operating instructions, trouble shooting,
1942 commissioning data, maintenance instructions, list of equipment suppliers,
1943 spare parts list, equipment data/capacity schedules, supplier’s literature,
1944 control system description, electrical drawings and as-built drawings.
1945
10.4.
1946 Maintenance personnel should receive appropriate training, and training
1947 records should be kept.
1948
10.5.
1949 HEPA filters should be changed either by a specialist or a trained person,
1950 and then followed by installed filter leakage testing.
1951
10.6.
1952 Any maintenance activity should be assessed critically to determine any
1953 impact on product quality including possible contamination.
1954
10.7.
1955 Maintenance activities should normally be scheduled to take place outside
1956 production hours, and any system stoppage should be assessed with a view
1957 to the possible need for requalification of an area as a result of an
1958 interruption of the service.
1959
1960
1961 References
1962
1963 1. Good manufacturing practices for pharmaceutical products: main
1964 principles. In: WHO Expert Committee on Specifications for
1965 Pharmaceutical Preparations Thirty-seventh report. Geneva, World
1966 Health Organization, 2003 (WHO Technical Report Series, No. 908),
1967 Annex 4. http://whqlibdoc.who.int/trs/WHO_ TRS_908_eng.pdf;
1968 Quality assurance of pharmaceuticals. A compendium of guidelines
1969 and related materials. Volume 2, Second updated edition. Good
1970 manufacturing practices and inspection. Geneva, World Health
1971 Organization, 2007; and Quality assurance of pharmaceuticals. A
1972 compendium of guidelines and related materials. Geneva, World
1973 Health Organization, 2015 (CD-ROM) (in print).
1974
1975 2. Expert Committee on Specifications for Pharmaceutical
1976 Preparations.Fortieth report. Geneva, World Health Organization,
1977 2005 (WHO Technical Report Series, No. 937)
page 75
1978 http://whqlibdoc.who.int/trs/WHO_TRS_937_eng.pdf.
1979
1980 3. Technical supplements to Model guidance for the storage and

1981 transport of time- and temperature-sensitive pharmaceutical
1982 products. WHO Expert Committee on Specifications for
1983 Pharmaceutical Preparations. Forty Ninth report. Geneva, World
1984 Health Organization, 2015 (WHO Technical Report Series, No. 992),
1985 Annex 5.
1986 4. Model guidance for the storage and transport of time- and
1987 temperature-sensitive pharmaceutical products (jointly with the
1988 Expert Committee on Biological Standardization). WHO Expert
1989 Committee on Specifications for Pharmaceutical Preparations.
1990 Forty-fifth report. Geneva, World Health Organization, 2011 (WHO
1991 Technical Report Series, No. 961), Annex 9.
1992 Further reading
1993
1994 Quality assurance of pharmaceuticals. A compendium of guidelines and
1995 related materials, Volume 1. Geneva, World Health Organization, 1997.
1996 Quality Assurance of Pharmaceuticals. A compendium of guidelines and
1997 related materials, Volume 2, Second updated edition. Good manufacturing
1998 practices and inspection. Geneva, World Health Organization, 2007.
1999 http://
2000 www.who.int/medicines/areas/quality_safety/quality_assurance/productio
2001 n/ en/index.html; and Quality Assurance of Pharmaceuticals. A
2002 compendium of guidelines and related materials. Geneva, World Health
2003 Organization, 2015 (CD-ROM) (in print).
2004
2005 World Health Organization. Supplements and updates available at:
2006 www.who.int/medicines.
2007
2008 ASHRAE handbook 1999. HVAC Applications, SI edition. Atlanta, GA,
2009 ASHRAE, 2007. http://www.ashrae.org/technology/page/548.
2010
2011 ASHRAE handbook 2000. HVAC Systems and Equipment. Atlanta, GA,
2012 ASHRAE, 2008. http://www.ashrae.org/technology/page/548.
2013
2014 Daly BB. Woods practical guide to fan engineering. Colchester, Woods of
2015 Colchester Ltd. Third impression, June 1985. Cambridge, Cambridge
2016 University Press. www.flaktwoods.com.
2017
page 76
2018
2019 European Commission. The rules governing medicinal products in the
2020 European Community, Volume IV. Good manufacturing practice for
2021 medicinal products. European Commission, Brussels, 2005.
2022 http://www.cen.eu/cenorm/ sectors/sectors/healthcare/index.asp.
2023
2024 ISPE Baseline® pharmaceutical engineering guides, Volume 2. Oral solid
2025 dosage forms, Second Edition / November 2009, International Society for
2026 Pharmaceutical Engineering. http://www.ispe.org/.
2027
2028 ISPE Baseline® pharmaceutical engineering guides for new and
2029 renovated facilities, Volume 5. Commissioning and qualification, 1st ed.
2030 Tampa, Fl, International Society for Pharmaceutical Engineering, 2001.
2031 http://www.ispe.org/.
2032
2033 International Cleanroom Standards, ISO 14644. Geneva, International
2034 Organization for Standardization. http://www.iso.org/iso/standards_
2035 development.htm.
2036
2037 Luwa. Introduction to high efficiency filtration. Bulletin 50.10.10, Sheet
2038 020. Pharmaceutical Inspectorate Convention/Pharmaceutical Inspection
2039 Co-operation Scheme. Guide to Good Manufacturing Practice for
2040 Medicinal Products. PH 1/97 (Rev. 3), 15 January 2002.
2041
2042 PIC/s GMP Guide (PE 009)
2043 http://www.picscheme.org/publication.php?id=4.
2044
2045 ICH Q9: “Quality Risk Management”, November 2005
2046 http://www.ich.org.
2047
2048 World Health Organization. Guidelines on quality risk management,
2049 Geneva, World Health Organization, 2013 (WHO Technical Report Series,
2050 No. 981), Annex 2).
2051 ***

Who (2015) PDF

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Who (2015) PDF

Încărcat de

Drepturi de autor:

Formate disponibile

Working document QAS/15.

12 © World Health Organization 2015

41 SCHEDULE FOR THE PROPOSED ADOPTION PROCESS OF

Discussion of proposed need for revision in view of the 29 June–

53 During the consultation on data management, bioequivalence, GMP and

63 Summary of main changes

207 GMP compliance sequence diagram

429 point extraction

Return Air Shaft

Dispensary Pre-Staging Room 15 Pa

Return Air Shaft

Booth 1 Booth 2 Bulk

Return Air Shaft

5985. Design of HVAC systems and components

684 Figure 12 is a schematic diagram of an air-handling system serving

Reactivation Chemical Drier

809 ‒ snow eliminators to prevent snow entering air inlets and

99.999995 U17 75E

ISO 45 E ≥ 99.995 ≤ 0.005 ≥ 99.975 ≤ 0.025 99.995 H14

ISO 40 E ≥ 99.99 ≤ 0.01 ≥ 99.95 ≤ 0.05 - -

ISO 35 E ≥ 99.95 ≤ 0.05 ≥ 99.75 ≤ 0.25 99.95 H13

ISO 30 E ≥ 99.9 ≤ 0.1 - -

ISO 25 E ≥ 99.5 ≤ 0.5 99.5 E12

Level 2 Protected Area in which steps are taken to protect

Level 3 Controlled Area in which specific environmental

Level 2 Protected areas operating on 100% outside air:

Level 3 Production facility operating on recirculated

1238 6.5.4. Containment can normally be achieved by application of the

1330 3 Pa tolerance on absolute room pressures, resulting in pressure

Image of room pressure gauge

1358 both outer sides;

Air Leakage Air Leakage

1494 6.9.14. Air filters should not be installed immediately downstream of

1709 9.6.2. The qualification of the HVAC system should be described in a

V-Model for Direct Impact Systems

DQ Test Plan Detail Design and IQ Test Plan

Build & Project

Test parameter Max. time Test procedure

Particle count test 6 months Dust particle counts to be carried

HEPA filter leakage Filter penetration tests to be carried

Warehouse and store Demonstrate that store

1829 In some instances it is not possible to increase the concentration by a

1833 9.6.19. If energy-saving procedures such as reducing the airflow during

1838 These precautionary measures should be based on a risk assessment to

1843 9.6.20. Additional documents that should be included in the qualification

1849 9.7. Supplementary notes on test procedures

1890 determine if additional sample locations are warranted. Consider aspects

1980 3. Technical supplements to Model guidance for the storage and

S-ar putea să vă placă și