12月14日，FDA发布了《行业指南：注射剂可见异物的检查》草案，该文件已完成翻译，现分享给大家，如下：

 

 

 

Inspection of Injectable Productsfor Visible Particulates

注射产品中可见颗粒的检查

  

Guidance for Industry

行业指南

 

 

 

TABLE OFCONTENTS

目录

 

 I.   INTRODUCTION

介绍

II.  STATUTORY AND REGULATORY FRAMEWORK

法律法规框架

III.  CLINICAL RISK OF VISIBLE PARTICULATES

可见颗粒物的临床风险

IV.  QUALITY RISK ASSESSMENT

质量风险评估

V.  VISUAL INSPECTION PROGRAM CONSIDERATIONS

目视检查的程序考虑

A. 100% Inspection

100%检查

1. Components and Container Closure Systems

部件和容器密封系统

2. Facility and Equipment

设施和设备

3. Process

工艺

4. Special Injectable Product Considerations

特殊注射产品的考虑

B. Statistical Sampling

统计学抽样

C. Training and Qualification

培训和确认

D. Quality Assurance Through a Life Cycle Approach

通过生命周期方法实现质量保证

E. Actions To Address Nonconformance

解决不符合问题的措施

VI.  REFERENCES

参考文献

 

 

 

 

I.  INTRODUCTION

介绍

 

Visibleparticulates in injectable products can jeopardize patient safety. Thisguidance addresses the development and implementation of a holistic, risk-basedapproach to visible particulate control that incorporates product development,manufacturing controls, visual inspection techniques, particulate identification,investigation, and corrective actions designed to assess, correct, and preventthe risk of visible particulate contamination.2The guidance also clarifies that meeting an applicable United StatesPharmacopeia (USP)3compendialstandard alone is not generally sufficient for meeting the current goodmanufacturing practice (CGMP) requirements for the manufacture of injectableproducts. The guidance does not cover subvisible particulates4 or physical defects that products are typicallyinspected for along with inspection for visible particulates (e.g., containerintegrity flaws, fill volume, appearance of lyophilized cake/suspensionsolids).

注射产品中的可见颗粒物会危及患者安全。本指南涉及开发和实施一种整体的基于风险的可见颗粒控制方法，该方法结合了产品开发、生产控制、目视检查技术、颗粒识别、调查和纠正措施，旨在评估、纠正和预防可见颗粒物污染的风险。2本指南还澄清，仅满足适用的美国药典（USP）3标准通常不足以满足注射产品的CGMP要求。本指南不包括亚可见颗粒4或物理缺陷（例如，容器完整性缺陷，灌装量，冻干粉/悬浮固体的外观）。

  

For thepurpose of this guidance:

本指南的目的：

 

Particulates refer to mobile, undissolvedparticles other than gas bubbles that are unintentionally present in aninjectable product.5They varyin nature (e.g., metal, glass, dust, fiber, rubber, polymer, mold, degradantprecipitate) and can be divided into three categories6:

颗粒是指注射产品中意外存在的气泡以外的可移动的、不溶的粒子。5它们的性质不同（例如，金属，玻璃，灰尘，纤维，橡胶，聚合物，霉菌，降解物沉淀），可分为三类：

 

• Inherentparticulates are particulates that are an innate product characteristic.

• 固有颗粒是指产品特性所固有的颗粒。

• Intrinsicparticulates are particulates that are derived from the manufacturing equipment,product formulation, or container system.

• 内部颗粒是来自生产设备、产品配方或容器系统的颗粒。

• Extrinsicparticulates are particulates that originate from the manufacturing environment andare foreign to the manufacturing process.

• 外来颗粒是源自生产环境的颗粒，是生产工艺的外来异物。 

 

Injectable products generally refer to injectablehuman drugs approved under section 505 of the Federal Food, Drug, and CosmeticAct (FD&C Act), injectable animal drugs approved under section 512 orconditionally approved under section 571 of the FD&C Act, and injectablebiological products licensed under section 351 of the Public Health ServiceAct. In some cases, the injectable product may be a drug or biological productconstituent part of a combination product, such as a drug or biological productprefilled into a syringe (see 21 CFR part 3).7

注射产品通常是指根据《联邦食品、药品和化妆品法》（FD&C法案）第505条批准的可注射人用药物，根据《联邦食品、药品和化妆品法》（FD&C法案）第512条批准或根据《联邦食品、药品和化妆品法》第571条批准的可注射动物药物，以及根据《公共卫生服务法》第351条获得许可的可注射生物制品。在某些情况下，可注射产品可以是药物或生物制品组合产品的组成部分，例如预先填充到注射器中的药物或生物制品（见21 CFR第3部分）。

 

Thecontents of this document do not have the force and effect of law and are notmeant to bind the public in any way, unless specifically incorporated into acontract. This document is intended only to provide clarity to the publicregarding existing requirements under the law. FDA guidance documents,including this guidance, should be viewed only as recommendations, unlessspecific regulatory or statutory requirements are cited. The use of the word should in Agency guidance means thatsomething is suggested or recommended, but not required.

本文件的内容不具有法律效力，除非特别纳入合同，否则无意以任何方式约束公众。本文件仅用于向公众明确法律规定的现有要求。FDA指导文件，包括本指南，应仅被视为建议，除非引用了特定的监管或法定要求。在机构指南中使用"应该"一词意味着建议或推荐了某些内容，但不是必需的。

 

 

II.  STATUTORYAND REGULATORY FRAMEWORK

法律法规框架

 

Undersection 501 of the FD&C Act, a drug product, including an injectableproduct, is deemed adulterated if:

根据FD&C法案第501条，药品(包括注射用药品)在以下情况下被视为掺假:

 

略

 

 

 

III.   CLINICALRISK OF VISIBLE PARTICULATES

可见颗粒的临床风险

 

Theclinical manifestations of adverse events caused by particulate contaminationvary and may depend on the route of administration (e.g., intravascular,intravisceral, intramuscular), patient population, and nature or class of theparticulates themselves (e.g., physical size or shape, quantity, chemicalreactivity to certain cells or tissues, immunogenicity, infectivity,carcinogenicity). Particulates in intravascular or intravisceral injectionsgenerally can cause more adverse events than those in subcutaneous orintramuscular injections. According to published case reports (Langille 2014;Doessegger et al. 2012), serious adverse events involving injectable productscontaminated with visible particulates have included:

由颗粒物污染引起的不良事件的临床表现各不相同，可能取决于给药途径（例如，血管内、内脏、肌肉注射）、患者群体以及颗粒本身的性质或类别（例如，物理大小或形状、数量、对某些细胞或组织的化学反应、免疫原性、感染性、致癌性）。血管内或内脏注射中的颗粒物通常比皮下或肌肉注射中的颗粒物引起更多的不良事件。根据已发表的病例报告（Langille 2014;Doessegger等人，2012），涉及被可见颗粒污染的注射产品的严重不良事件包括：

 

• At thesystemic level, infection and venous and arterial emboli (thrombotic ornonthrombotic).

• 全身感染以及静脉和动脉栓塞（血栓性或非血栓性）。

• Microscopic emboli, abscesses, and granulomas in visceralorgans.

• 内脏器官的微栓塞、脓肿和肉芽肿。

• Phlebitis, inflammatory reactions, granulomas, andinfections at injection sites.

• 静脉炎、炎症反应、肉芽肿和注射部位感染。

 

Furthermore,different patient populations may have different risks for developing adverseevents after exposure to injectable products contaminated with particulates.Risk factors include age (e.g., pediatric and elderly patients), personal orfamily history of thrombophilia, major surgery, cancer, trauma, underlying infection,autoimmune disease, diabetes-associated late-stage vasculitis, obesity, andsmoking.12

此外，不同的患者群体在暴露于被颗粒物污染的注射产品后，可能有不同的不良事件风险。危险因素包括年龄（例如，儿童和老年患者）、易栓症的个人或家族史、大手术史、癌症、创伤、潜在感染、自身免疫性疾病、糖尿病相关性晚期血管炎、肥胖和吸烟。

 

Applicantsshould consider these clinical risk factors when developing their qualitytarget product profile and in establishing an appropriate control strategy andacceptance criteria for visible particulates.13

申请人在制定其质量目标产品概况以及为可见颗粒物建立适当的控制策略和接受标准时，应考虑这些临床风险因素。

 

 

IV.  QUALITY RISK ASSESSMENT

风险评估

 

Visibleparticulates can have a negative effect on overall product quality. To ensureproduct quality and to limit clinical risk, manufacturers should conduct a riskassessment during product development.14During this risk assessment, manufacturers shouldidentify the typical visible particulates that could contaminate the injectableproduct and characterize their size ranges, quantity, and composition;determine risks for each type; and provide a visual description (e.g., photographsor drawings of typical defects) to be used for training purposes.15Manufacturersshould also consider the potential sources of particulates, appropriateanalytical methods to monitor them, and mitigation strategies to prevent theirpresence in the final product.

可见颗粒会对整体产品质量产生负面影响。为确保产品质量及减少临床风险，制造商应在产品开发期间进行风险评估。14在此风险评估期间，制造商应确定可能污染注射产品的典型可见颗粒，并表征其尺寸范围，数量和成分;确定每种类型的风险;并提供用于培训目的的视觉描述（例如，典型缺陷的照片或图纸）。15制造商还应考虑颗粒物的潜在来源、监测颗粒物的适当分析方法以及防止其存在于最终产品中的缓解策略。

 

Differentconsiderations are relevant depending on the category of particulates found duringthe risk assessment:

根据风险评估期间发现的颗粒物类别，不同的考虑因素是相关的：

 

• Inherent particulates areassociated with specific products or their formulations—such as proteinaceousparticulates, liposomes, or agglomerates—and are considered part of the qualitytarget product profile. Their presence should not be cause for rejection ofindividual units or product batches if they are a property of the approved productand product release specifications are met. For hard-to-inspect productscontaining inherent particulates, such as suspensions or emulsions, manufacturersshould develop supplemental testing methods to ensure adequate detection ofvisible particulates (see section V, Visual Inspection Program Considerations).In addition, manufacturers should monitor time-dependent changes duringstability testing that may lead to increases in size or number beyond theapproved acceptance criteria.

• 固有颗粒与特定产品或其配方（如蛋白质颗粒、脂质体或附聚物）相关，并被视为目标产品质量概况的一部分。如果它们是已批准产品的属性并且满足产品放行标准，则它们的存在不应导致拒绝单个单元或产品批次。对于含有固有颗粒（如混悬液或乳液）的难以检测的产品，制造商应开发补充测试方法，以确保充分检测可见颗粒（参见第五节，目视检查程序考虑）。此外，制造商应在稳定性测试期间监控与时间相关的变化，这些变化可能导致固有颗粒的尺寸或数量增加，超出批准的接受标准。

   

• Intrinsic particulates can be related to the manufacturingprocess. Such particulates could come from components, containers and closures(e.g., glass vials, rubber stoppers), and product contact processing equipment(e.g., tubing, filters, gaskets). Manufacturers should control suchparticulates before the actual manufacturing process through careful selectionand quality control of components, containers and closures, packagingmaterials, and manufacturing equipment. Additionally, manufacturers shouldconduct studies to determine whether their manufacturing processes generateparticulates. Similarly, manufacturers should study and understand the impactof handling, washing, and sterilization processes on manufacturing equipment (i.e.,wear and tear) that could lead to particulate generation over time. Suchprocess development studies can minimize intrinsic particulates by informingselection of the appropriate handling, washing, and sterilization proceduresand establishing equipment life spans. Manufacturers should also evaluatetrends in reject data at designated manufacturing facilities and use a lifecycle management approach to monitor and control process-related intrinsicparticulates in their final products.

• 内部颗粒可能与制造过程有关。这些颗粒可能来自组件、容器和瓶盖（例如，玻璃瓶、橡胶塞）和产品接触加工设备（例如，管道、过滤器、垫圈）。制造商应在实际生产工艺之前通过对组件，容器和瓶盖，包装材料和制造设备的仔细选择和质量控制来控制此类颗粒。此外，制造商应进行研究，以确定其制造过程是否产生颗粒。同样，制造商应研究和了解加工、清洗和灭菌过程对制造设备的影响（即磨损），这些影响可能导致颗粒物随时间推移而产生。这种工艺开发研究可以通过选择适当的加工，清洗和灭菌程序以及确定设备寿命来最大限度地减少固有颗粒。制造商还应在指定的制造工厂评估废品数据的趋势，并使用生命周期管理方法来监测和控制其最终产品中与过程相关的固有颗粒。

   

• Intrinsic particulates can also be related to the formulation or stabilityof the product or its container closure (e.g., particulates formed because ofprecipitation of active pharmaceutical ingredients, glass delamination, orprotein-silicone oil interaction). These types of particulates can form afterproduct release and can change in size or number when the product is stored.Manufacturers should study the risk of this type of intrinsic particulateforming under accelerated or stressed conditions in the product developmentphase to determine particulate characteristics and any time-dependentparticulate formation or growth that can occur. In addition, an analyticalmethod suitable for characterizing and monitoring product-specific particulatesshould be developed. A robust product design achieved through formulationoptimization and container closure screening during development is critical toreduce the formation of product-related intrinsic particulates. Informationobtained from these studies can be used to support product-specific inspectionprocesses (e.g., particulate seeding for test kits with known product-specificintrinsic particulates, particulate identification, and rejectionclassification).

• 内部颗粒还可以与产品或其容器密闭系统的配方或稳定性有关（例如，由于活性药物成分的沉淀，玻璃分层或蛋白质 - 硅油相互作用而形成的颗粒）。这些类型的颗粒物可以在产品放行后形成，并且其大小或数量在产品储存期间会发生改变。制造商应在产品开发阶段研究在加速或破坏条件下形成这种类型的固有颗粒的风险，以确定颗粒特性以及可能发生的任何时间依赖性颗粒的形成或发展。此外，应开发一种适用于表征和监测产品特异性颗粒的分析方法。通过开发过程中的配方优化和容器密闭系统筛选实现稳健的产品设计对于减少与产品相关的内在颗粒的形成至关重要。从这些研究中获得的信息可用于支持特定于产品的检测过程（例如，具有已知产品特异性固有颗粒的检测试剂盒的颗粒筛选，颗粒鉴定和剔除分类）。

   

• Extrinsic particulates arisefrom sources other than the formulation’s components, the containers andclosures, or the manufacturing equipment’s product contact surfaces. Theseparticulates, derived from materials not intended to be in contact with theinjectable product, can negatively affect product quality and could indicatepossible microbial contamination or another CGMP issue. Their presence in thefinal product can occur because of poor conditions in the manufacturingfacility (e.g., poor environmental control; equipment design, age, andmaintenance; facility location, construction, and maintenance; material andpersonnel flows). Manufacturing facilities must be CGMP compliant and ofappropriate design to support the manufacture of injectable products (see 21CFR part 211, subpart C; § 211.63; and part 4).

• 外来颗粒来自配方组分、容器和瓶盖或制造设备的产品接触面以外的来源。这些颗粒来自不打算与注射产品接触的材料，可能会对产品质量产生负面影响，并可能表明可能存在微生物污染或其他CGMP问题。它们存在于最终产品中可能是由于制造设施中的恶劣条件（例如，环境控制不良;设备设计，老化和维护;设施位置，施工和维护;物料和人员流动）。制造设施必须符合CGMP标准，并具有适当的设计，以支持注射产品的制造（参见21 CFR第211部分，子部分C;§ 211.63;和第4部分）。

 

Manufacturersshould not rely on downstream adjustments during manufacturing to justify apoorly designed product or process. Instead, quality should be built into themanufacturing process, starting with the development phase and continuingduring scale-up, process qualification studies, and commercial manufacturing.16Successfulmanagement of visible particulates also includes vigilant assessment of thestate of control, early detection of poor process performance, and effectiveprocess improvement throughout the product’s life cycle.

制造商不应在制造过程中依靠下游调整来证明设计不佳的产品或工艺的合理性。相反，质量应该内置于制造过程中，从开发阶段开始，在放大生产、工艺确认研究和商业制造期间继续进行。16对可见颗粒物的成功管理还包括对受控状态的警惕评估，早期发现不良工艺性能，以及在整个产品生命周期内有效改进工艺。

 

Proactively addressing risk is animportant part of a life cycle approach to visible particulate control. Formalrisk assessments conducted during product development contribute to processunderstanding and form a foundation for knowledge management. Their resultsshould be used to establish adequate product-specific production controls andclearly defined in-process alert and action limits for particulates. Thresholdstudies should be conducted to determine the characteristics (e.g., size,shape, color) of visible particulates that can be reproducibly detected bytrained personnel. These threshold studies can also be the basis forestablishing particulate standards that will be used to establish inspectionprocedures, help avoid inspection bias, and allow manufacturers to verify theirmanufacturing processes are in a state of control.

主动应对风险是可见颗粒控制的生命周期方法的重要组成部分。在产品开发过程中进行的正式风险评估有助于过程理解，并为知识管理奠定基础。其结果应用于建立适当的产品特定生产控制，并明确定义颗粒的工艺警戒限和行动限。应进行阈值研究，以确定可见颗粒的特征（例如，大小，形状，颜色）可以由经培训的人员可重复地检测到。这些阈值研究也可以作为建立颗粒标准的基础，这些标准将用于建立检查程序，帮助避免检查偏差，并允许制造商确认其制造过程是否处于受控状态。

 

 

V.   VISUALINSPECTION PROGRAM CONSIDERATIONS

目视检查程序的考虑点

 

Visualinspection can be viewed as part of a larger program to ensure that injectableproducts are essentially free of visible particulates.17Duringinjectable product development, manufacturers should establish procedures forinspecting the product, statistical sampling plan(s), acceptance/rejection criteria,and procedures for evaluating inspection results. Inspection procedures carriedover from another site or another product may not always be suitable for a newproduct.

目视检查可以看作是一个更大的计划的一部分，以确保注射产品基本上没有可见颗粒。在可注射产品开发过程中，制造商应建立产品检验程序、统计抽样计划、验收/拒收标准和检验结果评价程序。从另一个地点或另一种产品延续下来的检验程序可能并不总是适用于新产品。

 

Duringprocess scale- up or transfer to contract manufacturers, the visual inspectionmethods should be assessed to confirm they are still appropriate and valid atthe new scale or manufacturing site. The visual inspection program should allowfor appropriate adaptations based on knowledge gained throughout the product’slife cycle. For example, the inspection procedures and/or analytical andstatistical methods may need revision if the batch size, manufacturing process,or conditions change.

在工艺放大或转移至合同制造商的过程中，应评估目视检查方法，以确认它们在新的规模或制造现场仍然合适和有效。目视检查程序应允许根据在整个产品生命周期中获得的知识进行适当的调整。例如，如果批量大小、制造工艺或条件发生变化，则检查程序和/或分析和统计方法可能需要修订。

 

Inaddition to inspection, a visible particulate control program should includethe training and qualification of operators and investigation of discrepancies,including root cause analysis, corrective actions, and preventive actions.

除检查外，可见颗粒物控制计划还应包括操作人员的培训和确认以及偏差调查，包括根本原因分析，纠正措施和预防措施。

 

Trainedand qualified personnel, automated inspection technology, or a combination of bothshould be used to inspect each unit of injectable product for visibleparticulates (hereinafter 100% inspection).In addition, the quality unit should sample each batch for acceptance quality limit (AQL) testing. 18A visualinspection program should ensure that any visible particulates present in thebatch at the time of release are only those that have a low probability ofdetection because they are of a size approaching the visible detection limit.This section covers 100% inspection, statistical sampling, training andqualification, quality assurance through a life cycle approach, and actions toaddress nonconformance.

应使用经培训和确认的人员，自动检测技术或两者的组合来检测每个可注射产品的可见颗粒（以下简称100%检测）。此外，质量部门应对每批产品进行抽样，以进行验收质量限值（AQL）测试。18目视检查程序应确保放行时批次中出现的任何可见颗粒仅是那些由于其大小接近可见检测极限而被检测到的概率较低的颗粒。本节包括100%检查，统计抽样，培训和确认，通过生命周期方法进行质量保证以及解决不符合情况的措施。

 

 

A.  100%Inspection

100%检查

 

Manufacturersshould conduct 100% inspection during the stage at which there is the greatestlikelihood that visible particulates will be detected in the final container(e.g., before labeling to maximize container clarity). Manufacturers shouldensure that the equipment used and the physical environment where visualinspection will be performed are designed to minimize variability and maximizedetectability in the inspection process.

制造商应在最终容器中最有可能检测到可见颗粒的阶段进行100%检查（例如，在贴标之前，以最大限度地提高容器的透明度）。制造商应确保所使用的设备和将要进行目视检查的物理环境以最大限度地减少检查过程中的可变性和提高可检测性。

 

Important factors to considerfollow.

需要考虑的重要因素如下。

 

1.   Componentsand Container Closure Systems

组件和容器密闭系统

 

Visibleparticulate contamination could be traced to components or container closuresystems. To ensure visible particulate control, manufacturers must have writtenprocedures for the receipt, identification, storage, handling, sampling,testing, and approval or rejection of components and product containers(including devices and device components that contact injectable products) (§211.80; see also part 4). Such procedures must ensure that components andcontainers and closures are tested or examined and approved, as appropriate,before use in manufacturing(§211.84). Containers and closures must not alter the product’s safety, identity,strength, quality, or purity (§§ 211.94(a) and 600.11(h); see also part 4).

可见颗粒污染可能追溯到组件或容器封闭系统。为确保可见颗粒物的控制，制造商必须有书面程序来接收、鉴定、储存、处理、取样、测试以及批准或拒绝组件和产品容器（包括接触注射产品的设备和设备组件）（§ 211.80;另见第4部分）。此类程序必须确保在用于制造之前，组件、容器和封盖在适当情况下经过测试或检查和批准（§ 211.84）。容器和封盖不得改变产品的安全性、特性、强度、质量或纯度（§§ 211.94（a）和 600.11（h）;另见第 4 部分）。

 

2.   Facilityand Equipment

设施和设备

 

To complywith CGMP requirements, manufacturing facilities must be designed, constructed,and outfitted with equipment to prevent injectable products from beingcontaminated with particulates. Applicable CGMP regulations include:

为了符合CGMP要求，必须设计，建造和配备设备，以防止注射产品被颗粒物污染。适用的CGMP法规包括：

 

• Buildings and facilities (§§ 211.42 through 211.58and 600.11).

• Equipment design, size, and location (§ 211.63).

• Equipment construction (§§ 211.65 and 600.11).

• Equipment cleaning and maintenance (§§ 211.67 and600.11).

 

Inspections can be conductedmanually and/or using a range of automated inspection techniques:

检测可以人工和/或使用一系列自动检测技术进行：

 

• For manual inspections, the inspectionstation should have a backdrop of one or more solid colors (e.g., black andwhite) to provide adequate contrast and to allow maximum visibility of productcontents. The light intensity of the inspection station is also critical toachieving maximum visibility. Manufacturers should consider container color,size, and shape as well as product characteristics when determining the idealintensity.

• 对于人工检查，检查工作站应具有一种或多种纯色（例如，黑白）的背景，以提供足够的对比度并允许产品内容物的最大可见性。检查工作站的光强度对于实现最大可见度也至关重要。制造商在确定理想强度时，应考虑容器的颜色、尺寸和形状以及产品特性。

• During semi-automated inspections, a machinerotates the product at a constant rate past a trained inspector’s field ofvision. Rejected products are removed mechanically or by hand.

• 在半自动检测过程中，机器以恒定的速度旋转产品，以通过经培训的检查员的视野。不合格品通过机械或手动剔除。

• Automated inspection technology can beused as part of an investigation in the inspection process for injectableproducts, as a replacement for manual inspection, or as an additional qualityassurance step. Automated inspection technology can use different wavelengthsand sensors to detect hard-to-see particulates in sterile powder, suspensions,or light-protected injection products for which visual inspection is notcompletely effective.

• 自动检测技术可用作注射产品检测过程中调查的一部分，作为手动检测的替代品，或作为额外的质量保证步骤。自动检测技术可以使用不同的波长和传感器来检测无菌粉末、悬浮液或光保护注射产品中难以看到的颗粒，这些颗粒的目视检测并不完全有效。

       

Regardlessof the technique—manual, semi-automated, or automated—the inspectionenvironment should be free from distractions and extraneous light, and theinspection rate should be qualified and should allow for thorough visualinspection. Manufacturers can operate independent inspection stations asseparate units or units that are connected in a series. Some inspection equipmentdoes not require controlled separate facilities for visible particulateinspection.

无论采用何种技术（人工、半自动或自动），检测环境都应不受干扰和无关光线的影响，并且检测率应符合要求，并应允许进行彻底的目视检查。制造商可以将独立的检测站作为单独的单元或串联连接的单元进行操作。某些检测设备不需要受控的单独设施进行可见颗粒物检测。

 

Formanual and semi-automated inspections, the inspection environment should beergonomically designed for inspector comfort.

对于人工和半自动检测，检测环境应符合人体工程学设计，以确保检测人员的舒适性。

 

Forsemi-automated and automated inspections, equipment must be routinelycalibrated, inspected, or checked in accordance with a written program designedto ensure proper performance, and records of those calibration checks andinspections must be maintained (§ 211.68). Equipment should also be properlyqualified. See section V.C, Training and Qualification, for more information.

对于半自动和自动检查，必须根据旨在确保适当性能的书面程序对设备进行日常校准或检查，并且必须保留这些校准检查的记录（§  211.68）。设备也应经过适当的确认。有关详细信息，请参阅第 V.C 节"培训和确认"。

 

Whencompared with manual inspection, automated inspection technology may improvedetectability of visible particulates because machine variability is generallyeasier to control than the variability individual personnel can bring to tasksperformed repetitively over time. In some cases, the technology can detecthigher levels of specific visible particulates. In others, it can detectparticulates at the lower end of the visual inspection range with greaterstatistical reliability when compared with manual and semi-automated inspectionof the same product (Melchore 2010).

与人工检查相比，自动检测技术可以提高可见颗粒物的可检测性，因为机器的可变性通常比人员随时间推移重复执行任务带来的可变性更容易控制。在某些情况下，该技术可以检测更高水平的特定可见颗粒。在其他情况下，与同一产品的人工和半自动检测相比，它可以检测视觉检测范围下端的颗粒物，具有更高的统计可靠性（Melchore 2010）。

 

Automatedinspection technology may allow manufacturers to better control productquality. Manufacturers may need to adjust in-process action and alert limits ifthey change from manual to automated inspection. Adjustments should be based onstatistical process and batch data analysis obtained during evaluation andvalidation of automated inspection equipment.

自动检测技术可以让制造商更好地控制产品质量。如果制造商从人工检测更改为自动检测，则可能需要调整过程中的行动限和警戒限。调整应基于在自动检测设备评估和验证过程中获得的统计过程和批次数据分析。

 

Among theautomated inspection technologies currently in use (e.g., high-speed industrialcamera, visible diode array, X-ray, near-field radar, ultraviolet and nearinfrared spectroscopy), each has its advantages and disadvantages but, ifproperly implemented, all can substantially improve the accuracy of visualinspection.

在目前使用的自动检测技术（例如，高速工业相机、可见二极管阵列、X射线、近场雷达、紫外和近红外光谱）中，每种技术都有其优点和缺点，但如实施得当，所有这些都可以大大提高目视检测的准确性。

 

3. Process

工艺

 

Manufacturersshould conduct inspection feasibility studies for visible particulatedetectability, unit inspection duration, illumination, and fatigue time frame.These studies should be scientifically based and analyzed using appropriate statisticalmethodology. Depending on thestudy results, manufacturers mayneed to adjust particulate standards or inspection processes or, in some cases,change equipment to improve accuracy and reduce patient risk.

制造商应进行可见颗粒可检测性、单位检查持续时间、照明和疲劳时间框架的检查可行性研究。这些研究应以科学为基础，并使用适当的统计方法进行分析。根据研究结果，制造商可能需要调整颗粒标准或检查过程，或者在某些情况下更换设备以提高准确性并降低患者风险。

 

Manufacturersmust implement written procedures for production and process controls (§211.100), which should cover each aspect of the visual inspection process. Suchprocedures should cover handling of the units (e.g., swirling, inversion,distance from light), maximum length of the inspection period without a restbreak, and disposition and documentation of products that were rejected basedon the results of the visual inspection.

制造商必须实施书面的生产和过程控制程序（§ 211.100），其中应涵盖目视检查过程的各个方面。这些程序应包括对被检查单元的处理（例如，旋转、反转、与光的距离）、无休息不间断检查的最长时长，以及根据目视检查结果对被拒绝的产品进行处置和记录。

 

Acomplete program19for the control and monitoring of particulatematter must include written procedures for production and process control,sampling and testing of in-process materials, and control of microbiologicalcontamination that are designed to minimize the occurrence of visibleparticulates, identify affected batches of injectable product, and facilitateinvestigation to determine the sources of visible particulates (§§ 211.100,211.110, and 211.113).

用于控制和监测颗粒物的完整程序19必须包括生产和工艺控制，工艺物料的取样和测试以及微生物污染控制的书面程序，旨在最大限度地减少可见颗粒的发生，识别受影响的注射产品批次，并促进调查以确定可见颗粒的来源（§§ 211.100， 211.110 和211.113）。

 

Writtenprocedures should also cover how to conduct 100% inspections to ensure batchesare essentially free of visible particulates. All records must be documented inaccordance with applicable regulatory requirements (§ 211.188(b)(5); see also §600.12). Adequate written procedures can contribute to a more thorough understandingof the potential sources and quantity of visible particulates, leading toimprovements in process design. The increased level of understanding would alsopromote a more robust particulate control program and higher qualityinvestigations (see § 211.192).

书面程序还应涵盖如何进行100%检查，以确保批次基本上没有可见颗粒。所有记录必须根据适用的法规要求（§211.188（b）（5）;另见§ 600.12）进行记录。适当的书面程序有助于更全面地了解可见颗粒的潜在来源和数量，从而改进工艺设计。理解水平的提高还将促进更强大的颗粒物控制程序和更高质量的调查（见§  211.192）。

 

4. SpecialInjectable Product Considerations

特殊注射产品考虑

 

Large volume parenterals should undergo the same level ofinspection as small volume injectable products. In many cases, patient riskfrom particulate contamination is higher for large volume parenterals than forsmall volume injectable products because of the volume of product administeredand the potential for a patient to receive a continuous administration overmany days. Packaging and labeling of large volume parenterals (e.g., overwrapsand printing on the flexible bags) can interfere with visual inspection. Largevolume intravenous bags that have an outer bag can be particularly challengingto inspect. Manufacturers should take appropriate measures to ensure adequate100% inspection of these products. Supplemental destructive testing may also bewarranted to ensure these products are essentially free of visible particulatesif the packaging does not allow for the identification of particulates withinthe accepted visible size range.

大容量注射剂应接受与小容量注射产品相同水平的检查。在许多情况下，由于施用的产品量以及患者连续多日给药的可能性，大容量肠胃外的患者颗粒污染风险高于小容量注射产品。大容量注射剂的包装和标签（例如，在软袋上外包装和印刷）可能会干扰目视检查。具有外袋的大容量静脉注射袋可能特别难以检查。制造商应采取适当措施，确保对这些产品进行充分的100%检查。如果包装不允许在可接受的可见尺寸范围内识别颗粒，则还可以进行补充破坏性测试，以确保这些产品基本不含可见颗粒。

 

Opaque products and containers (e.g.,lyophilized powders, suspension products, tinted vials) present obviouschallenges to visual inspection. Using advanced technologies such as those describedin section V.A.2 in this guidance (e.g., X-ray spectroscopy) can help, as cansupplemental destructive testing after the 100% inspection, which providesadditional assurance of product quality. Supplemental destructive testing maynot be warranted, however, if the technology used in the 100% inspection isvalidated to meet or surpass human inspection capabilities. Manufacturersshould conduct a feasibility study to demonstrate the suitability of thetechnology selected for the specific product.

不透明的产品和容器（例如冻干粉末、悬浮产品、有色小瓶）对目视检查提出了明显的挑战。使用本指南中V.A.2节中描述的先进技术（例如X射线光谱）可以提供帮助，100%检查后的补充破坏性测试也可以提供帮助，从而为产品质量提供额外的保证。但是，如果100%检测中使用的技术经过验证，可以达到或超过人工检测能力，则可能不需要补充破坏性测试。制造商应进行可行性研究，以证明为特定产品选择的技术的适用性。

 

 

B.  StatisticalSampling

抽样程序

 

Following100% inspection, manufacturers should employ statistically sound samplingplans, validated inspection methods, and appropriate acceptance criteria toensure that each product batch meets a pre-established AQL for visibleparticulate contamination. This is consistent with USP General Chapters<1> and <790>; however, a more stringent sampling plan and acceptancecriteria may be appropriate for higher risk products.

在100%检查之后，制造商应采用统计上合理的抽样计划，经过验证的检查方法和适当的接受标准，以确保每个产品批次都符合预先建立的可见颗粒污染AQL。这与 USP 通则 <1> 和 <790> 一致;然而，对于高风险产品可能适用更严格的抽样计划和验收标准。

 

Asampling plan allows the user to make a specific statistical quality statement20about theattribute of interest (e.g., a defect) in a batch based on the sample size andsampling locations. Manufacturers should select their sampling plans inaccordance with the risk for a particular type of product defect. CGMPregulations require manufacturers to ensure that batches of injectable productsmeet appropriate specifications and statistical quality control criteria as acondition for their approval and release (§ 211.165).

抽样计划允许用户根据样本数量和抽样位置对批次中感兴趣的属性（例如，缺陷）20做出具体的统计质量陈述。制造商应根据特定类型产品缺陷的风险选择其抽样计划。CGMP法规要求制造商确保注射产品的批次符合适当的标准和统计质量控制标准，作为其批准和放行的条件（§ 211.165）。

 

Manufacturersshould quantify the following parameters with respect to design and use of samplingplans21:

制造商应量化以下有关设计和使用抽样计划的参数21：

 

• Operatingcharacteristic curves developed for each defect classification or qualityattribute that is being tested.

• 为正在测试的每个缺陷分类或质量属性开发的操作特性曲线。

• Accept/rejectcriteria, AQL, and unacceptable quality limit (also referred to as rejectable quality limit, limiting quality, or lot tolerance percent defective).

• 接受/拒绝标准、AQL 和不可接受的质量限度（也称为可拒绝质量限、限制质量或批次公差百分比缺陷）。

• Themethodology and acceptance criteria for the statistical sampling plan should considerpatient risk, particulate type, and product and container characteristics thatmay interfere with particulate visibility. For example, an adequate samplingplan with an acceptable AQL for nondestructive/destructive testing could followASTM E2234.22Firms that wish to propose an alternative minimumstandard for their specific product should ensure that there is a risk-basedjustification for the proposed standard.

• 统计抽样计划的方法和接受标准应考虑患者风险、颗粒类型以及可能干扰颗粒物可见性的产品和容器特征。例如，可以遵循ASTM E2234制定一个适当的抽样计划，一个可接受的AQL进行无损/破坏性测试。希望为其特定产品提出替代性最低标准的公司应确保拟议标准有基于风险的理由。

 

Extrinsicparticulates identified during 100% inspection or AQL of the batch—whichsuggests the presence of filth, sterility assurance issues, or other CGMPviolations—may result in product that could be considered adulterated, even ifthe statistical sampling acceptance criteria are met. Likewise, multiplevisible particulates (extrinsic or intrinsic) within a single container may beindicative of manufacturing problems and should trigger increased scrutiny ofthe batch.

在100%检查或该批次的AQL期间发现的外来颗粒物 - 这表明存在污染，无菌保证问题或其他CGMP违规行为 - 可能导致产品被视为掺假，即使符合统计抽样接受标准。同样，单个容器内的多个可见颗粒（外来或内部）可能表明生产问题，并应引发对批次的更多审查。

 

Ifretained samples are used to evaluate the suitability of product indistribution (such as in the case of product complaints), manufacturers should consideradditional factors such as historical data for the facility and/or product whenevaluating the suitability of a given product batch.

如果使用留样样品来评估产品在分销中的适用性（例如在产品投诉的情况下），制造商在评估给定产品批次的适用性时应考虑其他因素，例如设施和/或产品的历史数据。

 

According to § 211.194(a)(2), “the suitability ofall testing methods used shall be verified under actual conditions of use.”Manufacturers also must validate and document tests used to ensure that eachbatch of the product conforms to final specifications for release anddistribution (§ 211.165(e)).

根据§ 211.194（a）（2），"所有使用的测试方法的适用性应在实际使用条件下进行验证。制造商还必须验证和记录用于确保每批产品符合最终放行和销售标准的测试（§ 211.165（e））。

 

 

C.   Trainingand Qualification

培训和确认

 

Onlycertified inspectors and qualified equipment should be used to inspectinjectable products for visible particulates. Personnel conducting inspections(100% inspection and AQL inspection) must be adequately trained (including, asappropriate, periodic retraining or requalification) (§§ 211.25 and 600.10(b)).

只有经认证的检查人员和已确认的设备才能用于检查注射产品的可见颗粒。进行检查（100%检查和AQL检查）的人员必须经过充分的培训（包括酌情定期再培训或再确认）（§§ 211.25和600.10（b））。

 

Formalizedtraining and qualification programs promote consistent performance byindividual inspectors or automated inspection machines and help minimizevariability among different inspectors or machines (Melchore 2011). The programcan include a combination of training materials, standard operating procedures(SOPs), on-the-job training, and testing. Inspector candidates should betrained in the relevant CGMP requirements and should have normal near visualacuity (with or without the use of corrective lenses) and no impairment ofcolor vision (Ricci et al. 1998).

正式的培训和确认程序促进了检查人员或自动检查机的一致性能，并有助于最大限度地减少不同检查人员或机器之间的可变性（Melchore 2011）。该程序可以包括培训材料，标准操作程序（SOP），岗位培训和测试的组合。准检查人员应接受相关CGMP要求的培训，并且应具有正常的近距视敏度（无论是否使用矫正镜片）并且没有色觉障碍（Ricci等人，1998年）。

 

 

Regarding inspection equipment:

关于检查设备：

 

• Thespecific backdrop and light intensity selected for manual inspection stationsshould be qualified.

• 为人工检查站选择的特定的背景和光强应进行确认。

• Semi-automatedinspection equipment should be properly calibrated and qualified at a specificvial-spin and belt speed. Lighting should also be qualified to allow foraccurate human detection of defective products.

• 半自动检测设备应在特定的瓶子旋转速度和传送带速度下进行适当校准和确认。照明应进行确认以对有缺陷的产品进行准确的人工检测。

• Automatedinspection machines should be validated to meet or surpass human inspectioncapabilities and can be qualified using training standards or artificialintelligence technology.

• 自动检测机应经过验证，以达到或超过人工检测能力，并可以使用培训标准或人工智能技术进行确认。 

• Forpersonnel qualification and automated inspection systems validation, a mixtureof good injectable product units and defective units containing visibleparticulates should be used (Melchore 2011). This test set should be preparedand approved by quality assurance staff. Manufacturers should develop librariesof defective units from samples collected throughout the product life cycle,samples created to simulate production defects, or samples purchased to berepresentative of the types of particulates likely to occur for the drugproduct and its manufacturing process. Quality assurance staff should reviewthe library of defective samples and compare the samples to establishedstandards for proper classification. The library should contain examples fromthe lower limits of visual detection determined in the threshold studies. If anew particulate matter defect is identified, it should be analyzed to determineits source and added to the training library.

• 对于人员确认和自动检测系统验证，应使用好的注射产品单元和含有可见颗粒的缺陷单元的混合（Melchore 2011）。该测试集应由质量保证人员准备和批准。制造商应从整个产品生命周期中收集的样品、为模拟生产缺陷而创建的样品或为代表药品及其制造过程中可能发生的颗粒类型而购买的样品中开发缺陷库。质量保证人员应检查有缺陷的样品库，并将样品与已建立的标准进行比较，以便进行适当的分类。该库应包含阈值研究中确定的视觉检测下限的示例。如果发现新的颗粒物缺陷，应对其进行分析以确定其来源并将其添加到培训库中。

 

 

Typically, the percentage ofdefective units in a test set should not exceed 10–20 percent, and the test setquantities should be sufficient to provide an adequate degree of confidence inthe test results. Trained inspectors should review defective units before theyare included in the test set to determine if the visible particulates in themcan be detected under normal conditions, and the identity of defective unitsshould be masked to test subjects. The quality unit should control the testsets to ensure that qualification tests are not manipulated or biased.

通常，测试集中缺陷单元的百分比不应超过 10–20%，并且测试集数量应足以在测试结果中提供足够的置信度。经培训的检查人员应在缺陷单元纳入测试集之前对其进行检查，以确定在正常条件下是否可以检测到其中的可见颗粒，并且应向测试对象掩盖缺陷单元的身份。质量单位应控制测试集，以确保确认测试不受操纵或偏颇。

 

Thequality unit should also establish and approve qualification protocols thatidentify the sample test sets, test duration, grading method for test results,documentation of test results, acceptance criteria for certification, andactions to be taken for test failures. The protocols should also specify requalificationtesting methods and frequency.

质量部门还应建立和批准确认方案，以确定样品测试集，测试持续时间，测试结果的分级方法，测试结果的文件，认证的接受标准以及测试失败应采取的行动。方案还应规定再确认测试的方法和频率。

 

 

D.  QualityAssurance Through a Life Cycle Approach

通过生命周期方法实现质量保证

 

Processperformance and product quality monitoring systems should provide informationto ensure process control throughout a product’s life cycle. Process performancemeasurements (e.g., deviations, in-process defect results, statistical processcontrol reports, equipment and facility breakdowns) provide information on thestate of control during manufacturing. Product quality indicators (e.g.,stability test results, complaints, returned product) can help determinewhether particulate matter in the product caused an event. Similarly, fieldalert reports and adverse event reports could reveal possibleparticulates-related quality issues. This information should be used to evaluatethe effectiveness of visible particulate control strategies.

工艺性能和产品质量监控系统应提供信息，以确保在整个产品生命周期中进行工艺控制。工艺性能量度指标（例如，偏差、过程中缺陷结果、统计过程控制报告、设备和设施故障）提供有关制造过程中受控状态的信息。产品质量指标（例如，稳定性测试结果、投诉、退回的产品）可以帮助确定产品中的颗粒物是否导致了事件。同样，现场警报报告和不良事件报告可以揭示可能与颗粒物相关的质量问题。应使用这些信息来评估可见颗粒物控制策略的有效性。

 

Trends ofincreased particulate contamination, identification of new types ofparticulates, or particulates that exceed alert or action limits may indicate aflaw in product or process design. For example, inconsistent product qualitycould be caused by any one or a combination of these factors:

颗粒物污染增加的趋势、新型颗粒物的识别或超过警戒限或行动限的颗粒物的识别可能表明产品或工艺设计中存在缺陷。例如，不一致的产品质量可能是由以下任何一个因素或多种因素共同导致的：

 

• Inadequate controls of components, containers, orclosures.

• 对组件、容器或瓶盖的控制不足

• A product formulation that is not stable.

• 不稳定的产品配方

• Uncontrolled changes to the manufacturing process.

• 制造过程的不受控制的变化。

• Equipment and facilities that are not suitable fortheir intended use.

• 不适合其预期用途的设备和设施。

• Personnel practices that generate particles.

• 产生颗粒的人员实践

 

If aninvestigation reveals a flaw in product or process design, it is important toredesign the product or process to ensure reproducible product quality andreduction of particulate matter.

如果调查发现产品或工艺设计中存在缺陷，则必须重新设计产品或工艺，以确保可重复的产品质量并减少颗粒物。

 

 

E.  ActionsTo Address Nonconformance

解决不符合问题的措施

 

Manufacturersmust investigate quality discrepancies identified through the inspectionprocess, quality control testing, complaints, or as a result of a batch failureand extend their investigation to other batches that may be affected (§§ 211.192and 211.198). Such investigations should seek to identify the particulates and categorize them (intrinsic orextrinsic) because the presence of certain categories of particulates couldindicate CGMP issues or sterility failures.

制造商必须调查通过检查过程、质量控制测试、投诉或批次故障导致的质量差异，并将其调查范围扩大到可能受影响的其他批次（§§ 211.192 和 211.198）。此类调查应设法识别颗粒物并对其进行分类（内部或外来），因为某些类别的颗粒物的存在可能表明CGMP问题或无菌失败。

 

Investigationscan include tightened sampling plans, examination of particles to understandtheir origin, and evaluation of batch release impact. The investigation shoulddetermine the sources of the variation and identify appropriate correctiveactions and preventive actions. The investigations may also revealopportunities to enhance the robustness of particle detection (e.g.,improvements to the 100% inspection or AQL inspection program).

调查可以包括收严抽样计划，检查颗粒以了解其来源，以及评估批放行的影响。调查应确定变异的来源，并确定适当的纠正措施和预防措施。该调查还可能揭示增强颗粒检测稳健性的机会（例如，改进100%检测或AQL检测程序）。

 

Investigationsof manufacturing inspection outcomes should be conducted in situations such asthe following:

在以下情况下，应进行生产检查结果的调查：

 

• Individual or total defect limits are exceeded.

• 单个缺陷或缺陷总量超出限度。

• A batch fails to meet AQL limits.

• 批次无法满足 AQL 限度

 

Atypicaltrends should also be investigated. This includes examining defective unitsremoved from a batch that are within in-process specifications but outside ofstatistical (historical) trend limits for the manufacturing process ordefective units with visible particulates that have not been commonly observed.

还应调查非典型趋势。这包括检查从批次中剔除的缺陷单元，这些批次在中控的标准范围内，但超出生产过程的统计（历史）趋势限度，或者缺陷单元具有通常未观察到的可见颗粒。

 

Reinspectionof product batches may be permissible with appropriate scientific justificationand should be conducted according to approved SOPs with tightened acceptancecriteria. FDA does not recommend more than one reinspection in an attempt torelease a batch with atypical defect levels. Samples failing the AQLreinspection should be counted along with rejects from any other inspection ofthe product (e.g., such as 100% inspection and the original AQL visualinspection) in calculations to account for and reconcile all units of finalproduct in the batch.

产品批次的重新检查可能是允许的，应进行适当的科学论证和根据批准的SOP进行，并收严接受标准。FDA不建议在尝试放行具有非典型缺陷水平的批次时进行多次重新检查。在计算中，未通过 AQL 重新检查的样品应与产品的任何其他检查（例如，100%检查和原始 AQL 目视检查）的不合格品一起计数，以计算和核对批次中最终产品的所有单位。

 

Correctiveactions, such as reinspection, should be justified based on risk and havequality unit oversight and must be documented consistent with applicablewritten procedures (§ 211.100(b)).

纠正措施，如重新检查，应根据风险证明其合理性，并有质量部门监督，并且必须与适用的书面程序保持一致的记录（§ 211.100（b））。

 

Customercomplaints must be handled according to applicable CGMP regulations (§ 211.198)and should result in particulate identification whenever possible, aninvestigation into the potential source of the particulate, corrective actions(if necessary), and analysis of the batch’s retain samples for evidence ofvisible particulate contamination.

客户投诉必须根据适用的CGMP法规（§  211.198）进行处理，并应尽可能识别颗粒物，调查颗粒物的潜在来源，采取纠正措施（如有必要），并分析批次的保留样品，以获得可见颗粒污染的证据。

 

Ensuringthe effectiveness, safety, and quality of injectable products is of utmost importance.Therefore, FDA recommends the use of a holistic, risk-based approach to visibleparticulate control. This approach includes the use of a robust visualinspection program along with the implementation of other relevant CGMP measuresto help ensure that injectable products are not adulterated and are essentiallyfree of visible particulates.

确保注射产品的有效性、安全性和质量至关重要。因此，FDA建议使用基于风险的整体方法来控制可见颗粒物。这种方法包括使用强大的目视检查程序以及实施其他相关的CGMP措施，以帮助确保注射产品不掺假，并且基本上没有可见颗粒。