照度

1、检测点的照度强度至少应为2000 lux，USP 1790 要求为不低于2000–3750 lux；

2、对于吹灌封产品，ECA建议照明度为 10000 lux；对于半透明塑料包装或琥珀色玻璃包装应考虑增加照度，某些特殊情况下，照度提高到10000lux。

3、ECA建议应该考虑照度计的准确度，使用 CRI 指数的色彩再现最好具有 RA 值 > 90%。

（注：a. JB/T 20186-2017可见异物灯检仪中要求，设备4.2.2灯检仪的光源均匀性偏差应不大于6%，灯检仪的光源稳定性偏差应不大于4%；b. 实际执行建议结合照度计准确度，照度设置范围适当缩窄留出余量）

4、 ECA建议的检查照明的适当间隔建议为 6 个月⁠⁠⁠⁠⁠。

延伸补充观点：

1、中国药典：用无色透明容器包装的无色供试品溶液，检查时被观察供试品所在处的光照度应为1000～1500lx；用透明塑料容器包装、棕色透明容器包装的供试品或有色供试品溶液，光照度应为2000～3000lx；混悬型供试品或乳状液，光照度应增加至约4000lx；

• 药典个论对于照度的不统一性的实际执行思考：适用于中、美注册的无色透明容器包装的无色生物制品溶液，生产人员进行人工灯检，灯检光照度设置在2000~3000lx是否可以同时满足中、美药典和法规要求？

• 1.采用2000~3000lx可以同时满足中、美要求。

• 依据：①《ChP2020 三部 通则 生物制品分包装及贮运管理》人工灯检应采用日光灯（光照度应为1000~3000lx），其背景和光照度按制品的性状调整。

• 
  

• ②USP：光照强度：人工检查过程的结果受检查区域光照强度的影响。一般来说，增加照亮被检查容器的光强度将提高检查性能；建议对玻璃容器进行常规检查时，检查点的照度至少为2000–3750勒克斯。第章建议对透明玻璃容器进行日常检查时，最低光照水平不低于2000-3750勒克斯。对于塑料容器或由琥珀色玻璃制成的容器，建议提高亮度。在这种情况下，高达10000勒克斯的光照水平可能是有益的。

• 2.采用2000~3000lx不能同时满足中、美要求。

• 依据：《ChP2020 四部 通则0904 可见异物检查法》用无色透明容器包装的无色供试品溶液，检查时被观察供试品所在处的光照度应为1000～1500lx；用透明塑料容器包装、棕色透明容器包装的供试品或有色供试品溶液，光照度应为2000～30001x；混悬型供试品或乳状液，光照度应增加至约40001x。

• 3.如用2000~3000lx不能同时满足中、美要求，那么怎样才能同时满足中、美要求。

• 
  

• ①生产人员需要人工灯检两次：分别按照ChP1000～1500lx、USP2000–3750勒克斯进行灯检。

• 
  

• ②其他做法。

2、不同照度对产品质量影响，一般可以通过产品光照影响因素稳定性实验进行评估，根据《中国药典2020 四部 9001 原料药物与制剂稳定性试验指导原则》，光照影响因素稳定性实验条件一般照度为4500lx±500lx，详见识林主题词：光稳定性；

环境

Ambient conditions have a high impact on the quality of this operation. Temperature should be room temperature and not exceed 25 °C in the summer if not otherwise justified. The relative humidity and air velocity should be controlled and ensure comfortable working conditions. The noise level should therefore be below 55dB.

1、ECA对于目检环境的描述较USP1790更加具体，温度应为室温，如无其他理由，夏季不得超过 25°C。应控制相对湿度和空气速度，确保舒适的工作条件，噪音水平应低于55dB；

2、USP 1790建议降低室内灯光的亮度，来增加人员目检程序的专注度，降低外界反射对于人员检测的影响，尤其要注意安装窗户的目检房间，外界光可进入目检房间，随着时间或季节的交替，室内的光线会变化。

延伸补充观点：

中国药典：中国药典2020 四部 0904 可见异物检查法，第一法（灯检法）灯检法应在暗室中进行（注：暗室一般意义上理解为无窗户，室内照明光源关闭）。

人员

The medical check is not related to the competence of performing visual inspection activity.体检与执行目检活动的能力无关。

This is checked during the qualification of the operator and whether he has to wear his glasses is determined in this process.

这些检查是在操作员资质确认过程中进行的，由此确定操作员是否必须戴眼镜进行目检。

The use of corrective lenses to achieve the desired visual acuity is permitted.

1、ECA和USP 1790均规定了注射剂目检人员应该接受视力和色觉的定期检查，其中USP 1790提到了视力检查应包括远视力和近视力检查。

2、ECA和USP 1790均规定了操作员可以戴眼镜进行目检，ECA给出了是否需要佩戴的测试检查方法。

延伸补充观点：

中国药典：按照药典要求，人员远距离和近距离视力测验，均应为4.9及以上（矫正视力应为5.0及以上），应无色盲，还应注意人员从相同形状事物中区分微小差别的能力。（注：实践中需注意，药典采用的是标准对数视力表按照五分记录法，5.0即为正常标准视力值，如果采用 “国际标准视力表”小数记录法需进行换算）。

A bracketing approach for similar products is acceptable. Following the training via training sets there should be a side to side training with an experienced operator. The trainee should have the chance to ask questions with the experienced operator performing a parallel 100 % inspection of the inspected units of the trainee. 对于类似产品采用括号法是可以接受的。通过训练集接受培训之后，还应接受与熟练目检员并行操作的培训。在对被检单元进行并行100%目检时，受训人员应有机会向熟练目检员提问。

Following successful completion of the side by side training the initial qualification should be performed with a qualification set. This set should focus on critical and major defects and also contain a limited number of minor defects. The number of units to be inspected should represent the number of objects which are inspected in the routine process from one eye break to the next eye break. 成功完成并行培训后，应使用确认集进行初始资质确认。该确认集应着重于关键缺陷和主要缺陷，同时也包含有限数量的次要缺陷。被检单元数应代表常规目检过程中相邻两次休息之间接受目检的物品的数量。

There is sometimes the expectation for operators to be qualified taking into account different stress factors, including fatigue at the end of shift. It can be doubted that a fatigue effect can be simulated in an examination situation with the impact for the operator, whether he/she can keep his/her job. With this there is always nervousness and some people have even examination anxiety which leads to release of Adrenalin in the human body, which counters a fatigue effect. Additionally, when performing initial qualification, the fatigue at the end of shift can't be simulated anyway, as the operators are not yet allowed to perform a "normal" visual inspection workday. 有时会期望操作员资质确认考虑不同的压力因素，包括班次结束时的疲劳。值得怀疑的是，在考核情境（决定操作员能否保住工作）中是否能够模拟疲劳效应。在考核情境下，人们总是会感到紧张，有些人甚至会有考试焦虑，这会导致人体释放肾上腺素，以抵消疲劳效应。此外，在进行初始资质确认时，无法模拟班次结束时的疲劳，因为操作员此时还不被允许执行“正常”的目检工作。

A set of product specific containers with real product, with 10-20% but less than 30% of the containers containing randomly distributed defects. The sets should be cleaned after usage and routinely checked for defects at least every 6 months. There should be a logbook for each test set.

一组装有真实产品的容器，其中10-20％（但少于30％）的容器有随机分布的缺陷。该测试集应包含所有已知缺陷，使用后应清洁测试套件，并至少每6个月定期检查缺陷。每个测试集都应设立日志。

Qualification should be performed for each product type and package that the inspector will encounter. A bracketed or matrix approach can be used to simplify qualification of products with similar physical or visual characteristics such as container type and size, dosage form, product viscosity, color, and others (product families or groups). It is common to initially train and qualify personnel on clear solutions in clear containers (if produced at the facility) and then expand their expertise to the inspection of more difficult dosage forms or presentations.

Acceptance criteria should reflect the different risk categories (i.e., critical, major, and minor) of these defects when combined into a single test set, and a single acceptance criterion should not be used with a set with multiple defect risk categories. Inspector fatigue may be addressed in the qualification process by testing under worst-case conditions (e.g., at the end of a typical inspection shift).

7.5 Test Sets

The number of defective units in each test set should be limited to approximately 10% to prevent rejection bias (62).

1、ECA和USP 1790 对于目检人员的培训均要求通过分阶段的方式，先从了解缺陷品开始，然后由有经验的分析员手把手的现场教学和平行测试，再到使用确认集进行资质确认。

2、ECA和USP 1790中均规定了对于类似的产品的培训和考核可以采用括号法。

3、ECA规定了确认集（目检人员资质确认标准品）的数量应该能够代表日常目检时每次检测的数量。

4、ECA要求确认集的缺陷品占比约10-20%（但少于30％），USP 1790要求确认集的缺陷品占比约10%。

5、ECA和USP 1790均提到了对于操作员资质确认应该考虑不同的压力因素，应基于最差条件，例如：在完成一个班次的目检后； ECA指南对于此要求有不同的解读，认为人员在考核情境下，人们总是会感到紧张，有些人甚至会有考试焦虑，这会导致人体释放肾上腺素，以抵消疲劳效应，并且在进行初始资质确认时，无法模拟班次结束时的疲劳，因为操作员此时还不被允许执行“正常”的目检工作。

延伸补充观点：

1、ECA Visual Inspection Group 《Questions & Answers Document Version 2.0》：MI4 Q：Should the set of samples for the qualification (training) of the staff be taken directly form the production rejects or should it be produced artificially based on typical defects?

• 问：用于员工资质确认（培训）的样品组应该直接从生产不良品中提取，还是应该根据典型缺陷人工制备？

• A：Usually, the training set should be taken directly from the production rejects. Unfortunately, some failures occur only very rarely so that you have to produce some of them yourself. Finally, you have a set with failures that are typical for the production that has to undergo a formal release procedure described in an SOP. This guarantees that the failures are typical and that you can use them for training.

• 答：通常，培训集应该直接从生产的不合格品中获取。但是不幸的是，有些缺陷很少发生，所以你必须自己制造一些。最后，您有一组典型的缺陷样品，这些缺陷样品必须符合批准的SOP中的要求。这保证了缺陷样品是典型的，并且您可以将其用于培训。

2、FDA 行业指南：注射产品可见异物检查：For personnel qualification and automated inspection systems validation, a mixture of good injectable product units and defective units containing visible particulates should be used (Melchore 2011). This test set should be prepared and approved by quality assurance staff. Manufacturers should develop libraries of defective units from samples collected throughout the product life cycle, samples created to simulate production defects, or samples purchased to be representative of the types of particulates likely to occur for the drug product and itsmanufacturing process. Quality assurance staff should review the library of defective samples and compare the samples to established standards for proper classification. The library should contain examples from the lower limits of visual detection determined in the threshold studies. If a new particulate matter defect is identified, it should be analyzed to determine its source and added to the training library. 对于人员资质确认和自动设备验证的，应使用合格的产品和标准缺陷品的组合。该“确认集”（有的叫测试集）由QA准备和批准。从产品的生命周期中收集缺陷样品，委外制备的缺陷样品或者药品及其生产过程中可能产生的异物类型而购买的样品中建立缺陷样品库。QA应检查缺陷样品库，并将样品与批准的标准进行比较，从而对缺陷样品进行分级。该缺陷库中应包含检测阀值研究中确定的视觉检查下限的缺陷样品。如果在生产过程中发现了新的缺陷样品，应该对该缺陷进行分析并确认其来源，并将该类型的缺陷加入到“培训集”中。

3、关于人工确认集10%的缺陷率的科学依据补充：Studies by Wolfe et al (2007) and Rich et al (2008) indicate that the probability of detection during human visual inspection is influenced by the defect rate. PoD increases with increasing defect rate. Ideally test sets would be prepared with a defect rate equivalent to that observed in routine production inspection — typically 1–2% (Shabushnig, 2014). It is not uncommon to require 100 defect examples which would result in test sets of 10,000–20,000 in size. Such large sets are not practical, especially when multiple test sets are often required to represent the range of products and containers manufactured at a given location. As such, a practical compromise can be made by not exceeding a defect rate of 10% resulting in test sets of approximately 1,000 when 100 defect examples are included. This limits the degree of bias and better reflects true performance in routine inspection. 研究表明，人类视觉检查期间的检测概率受到缺陷率的影响。PoD随着缺陷率的增加而增加。理想情况下，准备的测试集的缺陷率与常规生产检验中观察到的缺陷率相当—通常为1–2%（Shabushnig, 2014）。需要100个缺陷示例并不罕见，这将导致测试集的大小为10,000–20,000。这样大的测试集是不实际的，尤其是当经常需要多个测试集来代表在给定位置生产的产品和容器的范围时。因此，当包含100个缺陷示例时，可以通过不超过10%的缺陷率得到大约1000个测试集来进行实际折衷。这限制了偏差的程度，并更好地反映了日常检查的真实性能。

Routine requalification must be performed at least every 12 months. After a second failure the operator must undergo a repeated eye test and a subsequent new qualification with the training kit and a subsequent new qualification via a qualification kit. 必须至少每 12 个月进行一次例行重新鉴定。第二次失败后，操作员必须再次接受视力测试，然后使用培训套件进行新的资格认证，最后通过资格认证套件进行新的资格认证。

PERCENT RECOVERY QUALIFICATION METHOD: A test set is prepared with bracketing of each particle type with a one-seeded unit being near-threshold (just above 70% PoD) and the other larger with a higher PoD. Typically, not less than 80% or 90% of the units containing visible particles are expected to be detected from the qualification test set.

RZE QUALIFICATION METHOD: The previously determined test set RZE is used to assess the results obtained by the inspector to be qualified. Each inspector to be qualified would then inspect the blinded set, and the RZE (the average PoD for all units containing particles >70% PoD) is calculated for each trial. The acceptance criterion is to achieve an RZE during the qualification that is equal to or better than the RZE originally determined during the threshold study.

%PoD = (Number of times rejected)/(Number of times inspected) × 100

RZE = the average of all PoD values for all units in the test set with an individual %PoD >70%

When assessing inspector performance, a limit is also needed for false rejection. A target of

Inspectors should be requalified at least annually. Requalification includes a test of visual acuity and testing with at least one product/test set configuration. A single successful inspection of the test set is sufficient for requalification. Requalification may also be necessary if poor performance is observed during routine inspection or if the inspector has been away from the inspection operation for an extended period (e.g., more than 3 months). If an inspector fails the requalification test, a retraining process should be initiated to identify the root cause and allow the inspector to receive additional instruction. After this process has been completed, the inspector may attempt to meet the acceptance criteria one additional time. If the inspector fails, he or she may attempt to qualify again after a specified time period.

1、对于人员确认的方法和标准，1790列举了两种，POD或RZE，并列出了详细的计算公式。

2、ECA和1790均要求考察误检率，1790建议的误检率低于5%，ECA建议5-10%。

3、ECA和1790均规定人员定期再确认的周期至少为每年；ECA规定了人员资质确认第二次测试失败后需要重新进行视力检查和测试集考核，1790规定每次的再确认均需要视力测试和测试集考核。

4、1790规定了除定期再确认外，当日常发现人员检测表现差或人员脱离检测较长（例如3个月以上）时也需要考虑再次确认。

5、1790规定了人员再确认至少选择一个产品，至少进行一次成功的测试集考核。

延伸补充观点：

EU GMP Annex 1：8.31人工检查时，应在有适当照明和环境的受控条件下进行。检查频率应经适当控制和确认。执行检查的操作员应至少每年接受目检资质确认（如果是佩戴眼镜的员工，在接受资质确认时也应佩戴矫正镜片）。确认应使用来自生产商缺陷库的适当样本组，并考虑最差条件（例如，检查时间、传送产品给操作员的传送带系统的线速度、容器尺寸或疲劳程度），并且应包括对视力检查的考虑。应尽可能避免操作员的注意力分散，并应在检查时经常进行适当的休息。

操作

The post inspection recovery time for visual Inspection staff is of essential importance. The maximum time for continuous inspection activity between breaks and the total maximum inspection time for a shift/workday must be defined. 检查后的恢复时间对于目检人员来说至关重要。必须定义休息之间连续检查活动的最大时间以及一个班次/工作日的总最大检查时间。 Industry practice is 20-60 minutes of inspection. A current good practice goal is 30 minutes of inspection followed by a break of at least 5 minutes continued for a total maximum duration goal of no longer than 4 hours. All Maximum uninterrupted inspection activity has to be qualified. 行业惯例是20至60分钟的检查。当前良好的做法是检查 30 分钟，然后休息至少 5 分钟，总最长持续时间目标不超过 4 小时。所有最大不间断检查活动均须合格。

Inspecting for extended periods of time can cause inspector fatigue and a decrease in inspection performance. On the basis of industry experience (46), it is recommended that inspectors be given a documented break from performing inspection at least every hour. This break should allow time to rest the eyes and mind and may be achieved with a short rest (e.g., 5 min) or a longer meal break. This need for regular breaks may also be met through rotation to a non-inspection function, such as material handling or documentation.

1、ECA规定了黑背景至少观察5秒，白背景至少观察5秒；1790规定每个背景约5秒，对于较大或较复杂的容器可能需要更多的观察时间，同时1790指出延长的观察时间可以导致分析员疲劳，继而导致人员检测能力的下降。

2、ECA规定使用半自动系统时，检查时间可能相对较短。

3、ECA中规定单次最大观察时间的业界一般要求是20-60分钟，最佳实践建议观察30分钟，休息至少5分钟，目检的最长持续观察时间不超过4小时；USP建议最长观察时间为一小时，然后休息5分钟或者更长时间进餐等，定期休息也可以通过其它非目检的操作来实现，例如物料准备、记录书写等。

延伸补充观点：

中国药典：将供试品置遮光板边缘处，在明视距离（指供试品至人眼的清晰观测距离，通常为25cm），手持容器颈部，轻轻旋转和翻转容器（但应避免产生气泡），使药液中可能存在的可见异物悬浮，分别在黑色和白色背景下目视检查，重复检查，总检查时限为20s。

半自动化目检

通过以下方式确保对受检单元的全面检查：

1. To avoid disturbances of the inspector an inspection booth should exclude external light sources.

1. 为了避免对检查员造成干扰，目检间应屏蔽外部光源。

2. At least a full 360°rotation of the unit in the inspection field facilitates inspection.

2. 受检单元在目检视野区至少可以360°旋转，以方便检查。

3. Inspection of the top and the bottom of the unit ensures complete inspection of the unit.

3. 检查受检单元的顶部和底部，确保全面目检。

4. For liquid products particle inspection can be supported by pre-rotating the objects before inspection to get the particles moving within the solution.

4. 对于液体产品的颗粒物目检，在检查前旋转受检物品以使颗粒物在溶液内移动。

5. Due to the nature of a semi-automated inspection the inspection field is typically illuminated at around 8.000 lux or higher. Difficult to inspect objects, e.g. brown glass vials or colored solutions, may require illumination above 10.000 lux.

5. 由于半自动化目检的性质，目检视野区的照度通常在8,000 lux或更高。难以目检的物品，例如棕色玻璃瓶或有色溶液，可能需要10,000 lux以上的照明。

6. For an undisturbed visual inspection, the background of the booth should be black. Reflecting surfaces should be avoided.

6. 为了目检不受干扰，目检间的背景应为黑色。应避免光反射表面。

7. Magnification should be used in order to bring the visual focus point of the operator's eyes to the point of inspection using an ergonomic sitting situation of the operator.

7. 应使用放大功能，以便使操作员在采用符合人体工程学的坐姿时眼睛的视觉焦点能对准检查点。

Risk Management Principles via assessments should be used during qualification of all the variable parameters noted above. Pre-start up tests should be established and then implemented in a controlled manner to be used periodically. Functional control kits should be used to ensure variable parameters are within the validated state before commencement of the inspection process, whereas potential drifts of the equipment (e.g., Illumination) is more suited to be verified periodically, e.g. every six months.

在对上述所有可变量进行确认时，应进行评估并运用风险管理原则。应建立启动前测试，然后以受控的方式定期测试。应使用功能控制套件来确保在开始目检之前可变量处于验证状态，设备的潜在漂变（例如照明）更适合定期确认，例如每六个月一次。

The most important parameters are: 最重要的参数是： Conveyor Speed 传送带速度 Illumination 照度 pre-spinning / if applicable 预旋转/如适用 Rotation within the inspection field to ensure 360 degree inspection of the unit 在检查区域内旋转，确保对设备进行 360 度检查 Validation of the semi-automated inspection process must show at least equivalence to manual inspection results. 半自动化检查过程的验证必须至少与手动检查结果相当

For lyophilized products inspectors are qualified at maximum inspection speed and may inspect during routine operation at maximum or any lower inspection speed.

对于冻干产品，目检员以最大目检速度通过资质确认，常规操作中的目检速度可以≤最大目检速度。

For liquid products inspection the qualified speed must be routinely used at all times due to the influence of speed on the inspection validity.

对于液体产品，由于目检速度会影响目检的有效性，常规操作中的目检速度应始终为通过资质确认的目检速度。

After an inspection process of an uninterrupted and predefined duration, operators require an eye break. Semi-automated visual inspection requires greater concentration by the operator. For that, times of uninterrupted inspection are typically shorter and times of breaks may differ from the numbers given in chapter 2.3 (Operation of the Manual Visual Inspection).

经过一段预定时长的持续目检后，操作员需要让眼睛得到休息。半自动化目检需要操作员更加集中注意力。因此，持续目检时间通常较短，休息时间可能与第 2.3 章（人工目检操作）中的不同。

CRITICAL PROCESS PARAMETERS FOR SEMI-AUTOMATED INSPECTION Light intensity must be controlled, as with MVI. The rate of inspection is controlled by the speed of the roller/conveyor or some equipment that allows the inspector to call for a group of containers each time. Spin speed for liquid products and rotation rate for all containers should be established during validation/qualification and maintained within the validated range for routine inspection. The background color is controlled by the color of the rollers selected and the color of the background seen through the spaces between the rollers. Qualification of inspectors and validation of the inspection equipment should be based on comparison with the compendial single-container manual-inspection process with an expectation that alternative methods such as semi-automated inspection demonstrate equivalent or better performance.

1、ECA指南中对于半自动目检的要求更加具体和详细，更具有指导性。

2、ECA和1790 均要求半自动化目检程序应与人工目检等效或者更优。

3、半自动目检的关键控制参数：传送带速度、照度、预混合、360°观察等。

4、1790中指出半自动目检的背景颜色控制可以通过滚轴的颜色和滚轴间的空隙颜色控制。

5、ECA规定半自动化目检需要操作员高度集中注意力，为此，不间断检查的时间通常较短。

6、ECA规定，对于冻干产品，目检员以最大目检速度通过资质确认，常规操作中的目检速度可以≤最大目检速度

7、ECA规定，对于液体产品，由于目检速度会影响目检的有效性，常规操作中的目检速度应始终为通过资质确认的目检速度。

确认/验证

注：颗粒的目检遵循概率检测原则，取决于被检药品溶液中颗粒的大小和可能的物理运动。需按照“Knapp 统计”原理估计颗粒的检出率。Julius Knapp 证明，一组无缺陷样品瓶（合格品）中混入缺陷样品瓶的比例不超过25％，颗粒检出率的统计学估计最准确。引入新目检系统或自动化目检系统时，准备一个测试集（其中包含含颗粒的样品瓶（不合格品）），进行人工目检-机器目检的对比，检出率放大10倍（10X）。当重复10次机器目检的结果和由此产生的测试统计值与重复10 次人工目检的的结果和统计值一样好时，即为成功。这就是所谓的 Knapp 测试。

➣ LIGHT OBSCURATION METHODS

➣ IMAGING METHODS

日常操作

再确认

再验证

1、ECA和1790均要求自动检查需要和人工目检进行对比，证明其等效或者更优，ECA规定了具体的测试方法和标准Knapp测试，同时1790指出了自动目检可能会有更高的误检率。

2、1790列举了自动检查的两种技术，光阻法和照相法，并描述了方法的适用范围。

3、ECA规定了自动目检机日常确认的要求，测试前后。

4、ECA规定了自动目检机定期回顾的要求，每年或最多两年一次。

5、ECA规定了自动目检机再验证的要求，建议3-5年一次。

冻干产品目检

冻干药品也必须经过100%目检。然而，由于只能检出冻干粉饼表面的颗粒物，而不能检出冻干粉饼本身存在的颗粒物，因此必须进行额外的不溶性微粒检查。为此，复溶一定量的样品，并目检复溶溶液中是否含有颗粒物。结果应在预先界定的粒子限值范围内。由于该补充检查本质上具有破坏性，因此只有在仅检查有限数量的样品的情况下才能接受进行该检查。根据取样计划从批次中抽取样品（例如根据 DIN ISO 2859 的 S-3 和/或 S-4）。

1、由于冻干产品的特殊性，ECA和1790均规定了冻干产品100%目检后需要进行复溶后的补充测试，取样计划建议为ISO 2859 的 S-3 和/或 S-4，1790中对于批量200-100000建议的取样量为20支（基于AQL 0.65%）。

2、1790除冻干产品外，对于较难目检的产品（剂型、包装及给药系统等）列出了推荐的补充测试建议，例如：粉针产品、乳剂、混悬剂、细胞及基因治疗制剂，茶色/琥珀色玻璃、半透明塑料包装产品，大容量制剂，组合产品等。

3、1790中规定，对于包装类型限定目视检查时，可以采取替代的补充测试方法，例如转移、过滤、离心、过筛等。

延伸补充观点：

对于冻干产品，需要注意的是在复溶过程中，应在洁净的环境中进行，避免造成假阳性的情况，更多可以参考PDA TR79 Particulate Matter Control in Difficult to Inspect Parenterals 2018

应该至少定义两类特定于产品的缺陷。定义更多的缺陷类别可能是合适的，例如：

Critical defects: may cause a lack of sterility, container integrity or cause harm to patients

关键缺陷：可能导致无菌性问题、容器完整性问题或导致患者伤害

Major defects: may alter the content or the function of the product

主要缺陷：可能会影响产品的内容物或功能

Further defect classes can be:

更多缺陷类别可以是：

Minor defects: Defects that do not affect patient health or product functionality

次要缺陷：不影响患者健康或产品功能性的缺陷

If a company performs additional visual inspection for specific cosmetic appearance defects for products intended to be placed to culturally sensitive markets it is recommended not to include them into the GMP defect classification.

对于拟投放到文化敏感市场的产品，如果企业进行特定外观缺陷的额外目检，建议不要将其纳入GMP缺陷类别。

1、对于缺陷品分类，ECA和1790的规定一致，建议按照关键缺陷、主要缺陷和微小缺陷，从缺陷对于产品质量和病人用药风险角度来分级。

2、对于拟投放到文化敏感市场的产品，如果企业进行特定外观缺陷的额外目检，ECA建议不要将其纳入GMP缺陷类别。

3、1790 详细列出了异物标准品、异物标准品制作及测试集等相关要求。

延伸补充观点：

1、对于新建车间、新的产品品种或者剂型等，在初期因为生产批次少，不能收集足够的缺陷样品，或者有些缺陷类型很少发生，所以不得不自己制备或者委外制备。如果是自己制备，那么首先必须了解产品的特性，以及可能存在的缺陷类型有哪些，比如产品里有纤维、胶塞碎片、玻璃碎片等等。那么就需要选择一些合适的材料进行模拟，不仅仅要考虑这些颗粒，还需要考虑产品的特性，如是悬浊液还是透明液体还是冻干产品，容器的装量等等，都尽可能的与实际产品保持一致，才能具有代表性。具体的做法可以文献Considerations for Design and Use of Container Challenge Sets for Qualification and Validation of Visible Particulate Inspection，国内可以参考JB/T 20135安瓿注射液异物自动检查机的附录A 试瓶制作部分。

2、BPOG: BioPhorum Operations Group Ltd在2023年1月发布《100%目检并不意味着100%缺陷检测》，旨在纠正一个普遍的误解：即认为对生物制药产品的100%目检等同于100%的缺陷检测。尽管产品经过了全面的目检，但由于目检过程本身的概率性质，并不能保证能够系统性地识别所有缺陷。目检缺陷的关键性和可检测性，这两者并不直接相关，因此即使进行了100%的目检，也不能保证100%的缺陷被检测出来。因此100%的目检并不等同于100%的缺陷检测，而应该被视为一个概率性过程，并且需要结合风险管理和质量控制策略来优化目检过程。

Reinsertion should only be conducted using a procedure that has been approved by the quality organization and addresses key parameters such as the inspection conditions (e.g., same as primary inspection or modified to enhance detection of a specific defect type), the number of times reinspection may be performed (this should be limited and justified), and the acceptance criteria (e.g., same as primary inspection or tightened). If reinspection is required often, consideration should be given to improving the sensitivity of the primary inspection process or of the manufacturing controls to prevent defects in the upstream process as determined by root cause analysis.

1、ECA和1790均要求基于历史数据进行缺陷趋势分析，建立警戒线和行动线。

2、ECA建议趋势分析应反映和参考工艺步骤的工艺能力指数 (CpK)，并给出了建议的行动线参考值和趋势分析的方法。

3、ECA中建议在自动化检测中定义“灰色弹出通道”，可能有助于分离检测结果不明确的容器。

4、对于重复目检，ECA和1790均要求限制，不超过2次，重复目检需要论述和批准后执行。

批放行

对于放行决策，需要评估两个标准：

• 1. Trending analysis of the 100% batch inspection (see chapter 6) and

• 1. 100%批目检的趋势分析（见第 6 章）和

• 2. The result of the AQL manual inspection.

• 2. AQL人工目检的结果。

The results of the 100% visual inspection, done as part of the manufacturing process, are an integral part of the batch documentation, specifying the types of defects found (fibre, turbidity, crack, etc.) as well as a classification of the defect such as critical, major or minor. Acceptance criteria must be pre-defined for these defect classes of all defects found during the 100% inspection process (see above). For the automated in-line inspection these classifications needs to be done on a technical level (side-wall defect, crimping defect,…) and cannot be done using the classification such as critical, major or minor (see above).

100%目检是生产过程的一部分，其结果是批文件记录的组成部分，其中指定了发现的缺陷的类型（纤维、浑浊、裂缝等）以及缺陷的类别，例如严重、主要或次要缺陷。对于100%目检中发现的所有缺陷，必须按照缺陷类别预先界定可接受标准（见上文）。对于自动在线目检，不能使用诸如关键缺陷、主要缺陷或次要缺陷的缺陷类别，而需从技术层面（侧壁缺陷、轧盖缺陷等）进行分类（见上文）。

For the AQL manual inspection a randomized sampling of the 100% inspected batch should be performed according to a pre-determined AQL procedure. AQL manual inspection can be carried out by production staff under a quality oversight or the quality unit.

对于 AQL人工目检，应根据预先确定的AQL 程序对100%目检批次进行随机抽样。AQL人工目检可由生产人员在质量监督下进行或由质量部门进行。

1、对于放行决策，ECA和1790均需要评估两个标准： 100%批目检的趋势分析。 AQL人工目检的结果。

2、对于建议的AQL最低标准，ECA和1790基本一致，均要求严重：AQL 样品中 0 个缺陷 Major: ≤ 0.65 AQL 主要：≤ 0.65 合格标准 Minor: ≤ 4.0 AQL 微小：≤ 4.0。

3、批放行的取样方案，1790建议选择一般抽样水平II。

4、对于超出 AQL 限制，ECA和1790规定一致，调查后可能要对整个批次进行 100% 重新检验，然后再进行第二次 AQL 人工检验，该过程可以重复，但必须通过调查证明其合理性，并且最多不应执行超过 2 次，建议对重复的 AQL 测试设置更严格的 AQL 限制。

5、对于可指定原因的高误检率（合格品误检为不合格品）的情况，例如气泡、特殊包装或胶塞差异，1790允许二步目检法，此种情况多对适用于自动目检机，对于这两步的检测，均需再分别进行AQL测试，结合两步测试的结果重新计算异物的POD。

6、ECA指出790中规定异物的检出应满足AQL的标准，建议的AQL为0.65，此AQL标准未基于异物的关键性，如果异物属于严重缺陷，则AQL应失败，因此，如果发现存在颗粒物，应进行调查和评估，该评估应包括异物的性质、大小和检测率。

已分销产品问题

Upon receipt, suspect containers should be subjected to the same inspection conditions and methodology used in the release inspection. Particle(s) verified in the returned or re-evaluated supply must be carefully characterized by an analytical forensic process to determine their source and likely cause. Single particles of typical product-contact materials are unlikely to present a concern. Multiple particles, large particle sizes, and any particles indicative of physical or chemical change are significant events and should be subject to further investigation. Rare instances of particulate material falling into the gray zone should be expected given the probabilistic nature of the inspection process and should not routinely trigger further evaluation of retention samples.

The particle detection threshold should be determined for a specific inspection method and product/package combination incorporating a variety of particle types and densities that are typically found in the manufacturing environment. For example, the detection threshold for routine, reliable detection (≥70% probability) of a single spherical particle in a clear solution contained in a 10-mL vial utilizing diffuse illumination between 2000 and 3000 lux is often near 150 μm in diameter (5). Units returned from distribution may be false positive, may contain particle(s) larger than the acceptance threshold that were missed, may contain particle(s) in the gray zone (i.e., less than the detection threshold), or may have suffered a physicochemical change that resulted in a visible change. Ideally, there will be no visible particles in the containers released to market; however, there is always a low probability that this may occur.

1、当对已分销到市场的产品的颗粒物产生担忧时，USP 790 建议的取样量为20支，如果未检测到异物，则认为基本无可见异物，如果检测到异物，需要测试更多的样品来收集相关的信息来判定异物的风险，需要注意的是，该评估仅适用于放行测试（100%目检后）成功完成并符合标准的情况。

2、对已分销到市场的产品的颗粒物产生担忧时，ECA建议按照符合 DIN ISO 2859 的实际抽样计划检查一定数量的样品，对于非关键缺陷，标准为零，并不实用，应避免使用。

3、对于异物的投诉，ECA和1790 均建议比较好的作法是分离投诉样本的颗粒，并确定其大小并在调查中描述颗粒的性质，基于异物的大小及种类进行评估。

4、该部分中的内容中，1790 提出了对于特定产品的目检检测方法需要进行不同种类异物检测限的研究。