The Vacuum Integrity Testing of Lyophilizers

1 Vacuum Integrity TestingJANUARY/FEBRUARY 2005 PHARMACEUTICAL ENGINEERING1 Copyright ISPE 2005 The Vacuum Integrity Testing ofLyophilizersby Charles D. Dern, PEThis articlesummarizes andclarifies termsand issuesrelated to thevacuum integritytesting equipment that manufac-tures Large Volume Parenterals(LVPs), the current Good Manufactur-ing Practices (cGMPs) state that: Equipment shall be constructed so that con-tact components, including process materials,drug products, or the drug product contact areaof containers or closures, shall not affect thesafety, identity, strength, quality, or purity ofthe Large Volume Parenteral drug product.

2 1 Because of the nature of the vapor pressure ofice, both the primary and secondary dryingphases of the lyophilization cycle must takeplace in a Vacuum in order to effect the sublima-tion and desorption of water or other solventout of the product. In turn, because the lyo-philization process occurs in an evacuated ves-sel, both designers and users of Lyophilizers arepresented with unique challenges in maintain-ing the sterility of the product in a these challenges are the measurementof system tightness and the establishment ofan inleakage criterion that maintains a reason-able assurance of product sterility.

3 With this inmind, the Vacuum Integrity Test is an impor-tant part of any Factory Acceptance Test (FAT),Site Acceptance Test (SAT), and/or OperationQualification (OQ).Basic DefinitionsBefore exploring practical issues, some basicdefinitions are essential. One can measure therelative tightness of evacuated vessels by one oftwo criteria: rate of rise or leak rate. Rate ofrise is the amount of pressure change in anevacuated vessel over a given period, ,milliTorr per minute (mTorr/min) or milliBarper second (mBar/sec).

4 2 For example, if oneevacuates a vessel to 100 mTorr ( mBar),closes the isolation valve to the Vacuum pump,and then observes that after one minute, thepressure is 102 mTorr ( mBar), then therate of rise is quite simply 2 mTorr per minute( mBar/min). Mathematically the formulais:Rate ofFinish Pressure - Start PressureRise=_____Elapsed TimeHowever, rates of rise, no matter how carefullydone, are not an accurate basis for comparingtightness among vessels of various sizes. Thisis because rates of rise do not account for thevolumes of the vessels in question.

5 If a 10 ft3 Figure 1. A Typicalresearch fromThe Official Journal of ISPEPHARMACEUTICAL ENGINEERING January/February 2005, Vol. 25 No. 1 Vacuum Integrity Testing2 PHARMACEUTICAL ENGINEERING JANUARY/FEBRUARY 2005 Copyright ISPE 2005(2831 L) vessel and a 100 ft3 (2831 L) vessel have the samerate of rise, a greater amount of gas must leak into the 100 ft3vessel to raise the pressure the same amount, in fact, 10 timesas much. To do an accurate comparison, therefore, one mustaccount for the respective volumes of the vessels.

6 This isaccomplished by a leak rate. Obtaining a leak rate involvesmultiplying rate of rise by the system volume. Thus, if a rateof rise is expressed in millitorr per minute (mTorr/min.), thena leak rate is expressed as millitorr cubic feet per minute(mTorr-ft3/min.) The general formula is:Leak(Finish Pressure - Start Pressure) VolumeRate=_____Elapsed TimeorLeakRate= Rate of Rise VolumeFor example, assume that vessels of 10 ft3 and 100 ft3 both areevacuated to 100 mTorr ( mBar) and are maintained ata constant temperature.

7 At this pressure, the 10 ft3 vessel willcontain standard cubic feet (SCF) ( L) of gas andthe 100 ft3 vessel will contain SCF ( L) of further that each vessel has an identical leak thatallows SCF ( L) of gas in one minute into eachvessel. At the end of one minute: The 10 ft3 vessel contains SCF ( L) of gas andis at a pressure of 176 mTorr ( mBar) for a rate of riseof 76 mTorr/min (0,101 mBar/min). The 100 ft3 vessel contains SCF ( L) of gas andis at a pressure of mTorr ( mBar) for a rate ofrise of mTorr/min ( mBar/min).

8 Both chambers have the same leak yet the smaller chamberhas the greater rate of rise. However, if the rates of rise aremultiplied by the respective chamber volumes, one obtains:10 ft3 76 mTorr/min = 760 mTorr-ft3/min( L mBar/min = mBar-L/min)and100 ft3 mTorr/min =760 mTorr-ft3/min(2831 L mBar/min = mBar-L/min)The vessels have identical leak rates. Even though the 100 ft3vessel has 10 times the evacuated volume of the 10 ft3 vessel,as long as the vessels are at the same pressure and haveidentical leaks, virtually the same amount of gas will enterinto each vessel over a limited range.

9 This is because theorifice of each leak sees approximately the same obvious advantage of leak rate over rate of rise is thatthose who own Lyophilizers of various sizes can specify asingle master acceptance criterion (although the actual testrequires that one measure a rate of rise). Figures 1 and 2 ofresearch and production Lyophilizers respectively, show justhow size can vary among systems. Yet, despite their sizedifferences, both systems can reasonably be held to the sameleak rate for Vacuum IntegrityThe actual Testing for Vacuum Integrity is the same timestraightforward and not so straightforward.

10 It is straightfor-ward in that the basic test sequence is simple: chill condens-ing plates (to protect Vacuum pumps), evacuate system, stopevacuation, allow system to stabilize, and measure rate ofrise. It is not so straightforward for several reasons: theproblem of real leaks and virtual leaks, the influence ofsystem temperature, and the lack of an industry-establishedacceptance A. Equivalent rates of rise of given volumes for a leak rateof 1 10-2 mBar- : Volume:mBar/mTorr/Volume: Volume:mBar/ Integrity TestingJANUARY/FEBRUARY 2005 PHARMACEUTICAL ENGINEERING3 Copyright ISPE 2005 Figure 2.