Complying with Revised Weighing Guidelines

The latest revisions to the United States Pharmacopeia (USP) General Chapters <41> Balances and <1251> Weighing on an Analytical Balance became official on Dec. 1, 2013 (1, 2).  These revisions aim to ensure weighing accuracy and eliminate unnecessary overtesting for US pharmaceutical manufacturers and suppliers. The chapters attempt to simplify previous descriptions and reflect current state-of-the-art weighing practices. Revisions include modified balance test procedures, which focus on the determination of minimum weight, and the execution of routine testing.

The new requirements of General Chapter <41> are mandatory. The revised chapter now states that accurate weighing must be performed using a calibrated balance. It also details a relatively demanding balance assessment, describing accuracy and repeatability tests with new acceptance limits. This applies for balances used to weigh analytes for quantitative measures.

For accuracy testing, use of a test weight of 5–100% of the balance capacity is now applicable. Accuracy, defined by USP as the systematic error, must be determined at higher loads close to the balance capacity (i.e., largest systematic error), as it cannot be detected reliably below 5%. Hence, accuracy testing below 5% is no longer allowed (see Figure 1).

In contrast, repeatability testing and minimum-weight determination should be carried out using a small weight below 5% of the balance capacity (see Figure 2). In this lower range, the repeatability is the dominant source of error and is virtually independent of the test weight. With higher weights (i.e., above 5%), the repeatability is a weak function of the test weight and so the repeatability will increase, leading to higher minimum weight values.

For more information on balance testing, a video tutorial can be viewed at: www.mt.com/lab-usp-tutorial.

Advisory Chapter <1251> provides additional clarification and extends the scope to any balance used for an analytical procedure. These major revisions focus on balance qualification and operation. Installation qualification (IQ), operational qualification (OQ), and performance qualification (PQ) recommendations are given. PQ, for example, should be carried out using a risk-based approach.

For the first time, the term “minimum weight” is officially introduced in a pharmaceutical compendium. This requirement is clearly applied to the sample weight, not including the tare vessel. Every balance has an individual minimum weight, depending on the type and performance of the load cell, the location, and the environmental conditions. The minimum weight of every balance should be periodically assessed for conformance. During a webinar in December 2013, it was encouraging to note that the majority of balance users (75% of those surveyed) are already familiar with the term minimum weight (3).

It is also important to note that USP recommends how to establish the minimum weight for applications outside of USP <41>, with the phrase “If not subject to the requirements of General Chapter <41>, the minimum weight value may vary depending on the required weighing tolerance and the specific use of the balance” (1).

Chapter <1251> goes on to provide practical recommendations for the installation qualification and operation of balances. Amongst others, it describes several environmental factors that can influence the performance of a balance, such as air currents (drafts), temperature variations (e.g., direct sunlight), vibrations, and electrostatic effects. These factors can all affect the minimum weight of a balance at a specific time.

Crucially, weighing net samples sufficiently above the minimum weight is recommended to take balance performance fluctuations into account. The implication is that a safety factor should be used when weighing, as shown in Figure 3. The use of a safety factor ensures that the smallest net weight is sufficiently larger than the minimum weight determined at a particular time by a particular person.

Applying a safety factor minimizes the risk of out-of-calibration, out-of- specification (OOS) or out-of-tolerance (OOT) results. The safety factor is the quotient of the smallest net weight and the experimentally determined minimum weight. It accounts for environmental influences and variations in operators, samples, and statistical factors. The safety factor applied should depend on the risk. It should be higher for more critical applications and unstable environments.

Another consequence of the USP revisions is that a daily balance check, typical in the pharmaceutical industry, is no longer a requirement. Instead, the type and frequency of balance checks should be determined by the risk and process tolerance of the application. The more critical the impact of an incorrect result, the more testing should be carried out, to reduce the risk of OOS or OOT results. In any case, it should contain at least a regular calibration with determination of measurement uncertainty and periodic sensitivity and repeatability testing. Performing the right tests at the right intervals will ensure quality results. During a webinar in December 2013, a majority of balance users (77% of those surveyed) admitted to still carrying out balance checks daily or before every use (3). This result indicates that many companies could potentially benefit from eliminating unnecessary testing, which would bring significant time and cost savings.

Common questions about the USP revisions
Since the USP announced these revisions, some questions have arisen.

Q: The USP chapters focus primarily on analytical balances. Are these revisions applicable to balances in a production or manufacturing environment?

A: The scope of USP is quality control of drugs. If the procedure referred to in a USP monograph requires materials to be ‘accurately’ weighed, a balance that weighs ‘accurately’ (meeting 0.10% requirements for repeatability and accuracy) is needed. If not, the balance repeatability and accuracy should meet other defined weighing tolerances commensurate with the requirements of the application. This aspect does not apply to production scales. However, production scales can also have a critical impact on the quality of drugs and thus need to meet GMP requirements. GMP requires scales to be consistently accurate and to be periodically calibrated and tested. Thus, the same scientific principles of managing weighing equipment--summarized in GWP (Good Weighing Practice, Mettler Toledo)--apply for production and quality control. The difference is that USP <41> explicitly requires a weighing tolerance of 0.10%, whereas the weighing tolerance in production can be chosen and defined based on specific quality and process requirements. Most pharmaceutical companies typically use a weighing tolerance limit between 0.1% and 2%.  

For all applications, weighing should be carried out using a calibrated balance that meets the requirements for repeatability and accuracy.

Q: What does USP mean when they say that balance checks with external weights can be replaced partially using automatic or manually triggered adjustment by means of built-in weights?

A: With some limitations, the internal test can be considered as an external sensitivity test. It detects systematic deviations and adjusts for it. However, since the internal weights are not traceable, an external, traceable weight must be used periodically. In practice, ‘partially’ means that if the internal test is performed on a daily basis, the external frequency can be extended to once a week or once a month, for example. It is recommended to activate the in-built automatic adjustment function using the internal weights and to program (or trigger manually) the balance for a daily adjustment. Depending on the risk of the application, further measures may be recommended.

Q: Do we need to establish the minimum weight for each balance or can we just set a nominal minimum weight for all five-place balances?

A: Every balance has an individual minimum weight, depending on the type and performance of the load cell, the location and the environmental conditions. This means that 10 different balances of an identical model may have 10 different minimum weights. In theory, it is possible to set the smallest net weight to a relatively high threshold to cover a family of identical balances. However, this would eventually mean that weighing range is lost with those balances that perform better, and weighing more material could mean wasted sample and, therefore, higher costs. Nevertheless, if the smallest net weight for a series of instruments is set to 50 mg, for example, then the minimum weight of every balance must still be periodically assessed for conformance.

Q: What is a safety factor and how is it determined?

A: The safety factor (see Figure 3) is the quotient of the smallest net weight and the determined minimum weight; it accounts for environmental influences and variations in operators, samples, and statistical factors. The safety factor depends on the risk. It should be higher for more critical applications and unstable environments. We would recommend measuring the current, actual safety factor through a GWP Verification and then defining it based on the results and the risk analysis.

Commercially available programs, like Mettler Toledo’s Good Weighing Practice, can help pharmaceutical manufacturers and suppliers to comply with the new USP guidelines by putting the guidelines into practice in a structured and effective way. Services can, for example, perform risk assessments and provide test methods and frequencies based on specific risks.

References
1. USP General Chapter <41>, “Balances,” Second supplement to USP36-NF 31 (US Pharmacopeial Convention, Rockville, MD, June 1, 2013).
2. USP General Chapter <1251>, “Weighing on an Analytical Balance,” Second supplement to USP36-NF 31 (US Pharmacopeial Convention, Rockville, MD, June 1, 2013).
3. G. Martin, M. Huber, and J. Ratcliff, “Ensuring USP Compliance with Revised Chapters on Weighing,” webinar (Dec. 3, 2013).

About the Authors
Joanne Ratcliff, PhD, is communications project manager for Laboratory & Weighing Technologies, Joanne.Ratcliff@mt.com, and Martin Huber, PhD, is head of the GWP competence center, Martin.Huber@mt.com, both at Mettler Toledo.