Principles of Physiology and Respirator Performance

Available evidence demonstrates that today's respirators provide their expected level of protection when properly selected, used, and maintained.

• By Larry Janssen
• Jun 01, 2003

EMERGENCY response workers frequently are required to wear respiratory protection to prevent the inhalation of toxic air contaminants. However, it is known there is a wide range of tolerance to the stresses of work among the working population. An individual's size, age, and fitness are among the conditions that influence the performance of his cardiorespiratory system and ability to perform the heavy work often required in emergency response.

Respirator manufacturers must address these physiological requirements and limitations when designing their products. This article will address the physiological and performance issues surrounding two respirator design criteria:

• Air flow rates used for testing air purifying respirator (APR) filters;
• Minute volume and peak air flow capabilities of powered air purifying respirators (PAPR), supplied air respirators (SAR), and self-contained breathing apparatus (SCBA).

Because there is a wide range of ventilation (V_e) requirements, aerobic capacity (V0₂ max), and peak inspiratory flow rates (PF_I) among the working population, respirators must accommodate the needs of as many individuals as possible, given the constraints of technical feasibility, comfort, and cost. It will be shown that modern respirators meet workers' physiological demands, and at the same time provide high levels of protection.

Air Flow Rates for Filter Testing
Factors that affect filter performance: Particulate respirator filter tests conducted by approval agencies are intended to assure that all filters of a given type will have a stated minimum level of efficiency when tested under uniform laboratory conditions. To assure that approved particulate respirators will perform adequately in the workplace, filter test conditions are chosen to approximate a "worst case" situation.

Table 1 describes several conditions that are known to affect filter performance and the test criteria the National Institute for Occupational Safety and Health uses to approve N, R, and P series particulate respirators.1 It is important to recognize that it is unlikely that one or more of these conditions will occur at any given time in a given workplace. As a result, filters can be expected to perform much more efficiently in the workplace than they do in the laboratory testing.

Physiological considerations: The rate at which air flows during respiration is constantly changing. Because PF_I may exceed V_e by four or more times, it has been argued that filters should be tested at PF_I rates to provide acceptable protection.2 In other words, although the current test flow rate of 85 Lpm represents a very high work rate not sustainable by most individuals (Table 2), it has been suggested that filters should be tested at more than 350 Lpm.

This table assumes all individuals have the same maximal aerobic capacity of 42 ml/kg/min (average 20-30 year old male weighing 70 kg). For each 10 kg over or under 70 kg, add or subtract 10 percent from the values for minute ventilation (V_e) and work (Kcal/min) shown. Fitness is evaluated by the percentage of VO₂ max that the individual can sustain for different intensities of work. Note: High and low fitness are evaluated as found in the normal workplace and do not include aerobic athletes.

Filter performance modeling: A filter performance model based on single-fiber filtration mechanisms can be used to illustrate why there is no need to test filters at the suggested high flow rates.3 The mechanisms by which filters collect particles of various sizes are briefly summarized as follows:

• Impaction: Effective in removing particles approximately 1 m m or larger. The inertia of these large particles causes them to impact on the filter media. Because particle inertia is increased with velocity, impaction removal is favored by higher air velocity.
• Interception: Aids in removing particles down to approximately 0.6 m m. Particles in an airstream passing around a piece of filter media make contact with the media and are removed from the airstream.
• Diffusion: Most effective in removing particles less than 0.2 m m. These very small particles exhibit a random movement known as Brownian Motion. This random activity may cause the particles to "bounce" into the filter media. Since the probability that one of these particles will strike the filter media increases as the time the particle remains in the filter increases, diffusion removal improves with lower air velocity.

• Some filters, commonly called "electrostatics," use electrical charges on the filter media to enhance removal of submicrometer particles. These charges attract airborne particles of opposite charge. "Electrostatic" filters also use the mechanical processes of impaction and interception to remove larger particles.

The filter performance model incorporates characteristics of filter design, challenge aerosol and air flow that interact to affect filter efficiency. It allows these characteristics to be changed individually and predicts both particle count and particle mass penetrations. Thus, it is easy to predict the effect of each individual change on the efficiency of the filter.

Table 3 summarizes the output of the model for a filter with typical design characteristics under various challenge conditions. Only the variables listed were changed in each case.

Discussion: The following observations can be made from Table 3:

1. The filter is much more efficient under plausible workplace conditions than it is under NIOSH test conditions. This observation is consistent with the "worst case" assumptions about the NIOSH test conditions.
2. Increasing the flow rate to 400 Lpm significantly increases count penetration of the "NIOSH" aerosol but has a much smaller effect on mass penetration. This is expected because particle mass increases with the cube of diameter. That is, one 1 m m particle has the same mass as 1,000 0.1 m m particles. As such, a significantly higher number of particles penetrated the filter, but because they are small particles they have very little mass.
3. There is a very small difference in mass penetration of the workplace aerosol at 400 Lpm versus 30 Lpm. Again, this is expected because workplace aerosols are much larger than laboratory test aerosols. It is also evident in this case that the small particles with very little mass are those that penetrate more readily at the higher flow rates. This is explained by the fact that diffusion removal is less efficient at higher flow rates, while impaction efficiency for larger particles (with much greater mass) is increased. This is a very significant point because exposure limits for the great majority of particulate air contaminants are based on the mass of the contaminant inhaled, i.e., mass dose. An increase in the penetration of small particles has a minimal effect on the contaminant dose inhaled.

In addition to the fact that high flow rates have little effect on mass penetration, it is important to recognize that 400 Lpm is an unrealistically high flow rate to model filter performance in a workplace. Physiologists generally assume a 3:1 or 4:1 ratio of PF_I to V_e . Therefore, it follows from Table 2 that 400 Lpm represents a PF_I value that is only achieved by a few individuals under extremely heavy, non-sustainable work activity. Further, the duration of the peak flow is no more than a few tenths of a second per breath.4,5 Because most exposure limits represent time-weighted average (TWA) exposures averaged over eight hours (or 15 minutes for short term exposure limits [STEL]), a brief increase in penetration of very small particles contributes little to overall dose.

Finally, it is important to consider what is known about respirator performance in the workplace. Workplace protection factor (WPF) studies measure the protection provided by a properly functioning respirator when correctly used by a properly fitted and trained worker. All sources of facepiece penetration, including filter penetration and faceseal leakage, are taken into account. WPF studies have been conducted with half-facepiece negative pressure respirators using dust/mist and dust/fume/mist filters. The approval tests for each of these respirator categories used an air flow rate of 32 Lpm and particle sizes outside the most penetrating range.6

Nonetheless, 5^th percentile WPF values above 10 (the value necessary for half-facepieces) were consistently found.7 It is interesting that the 5^th percentile WPFs for respirators using these filters were not significantly different from those found for half-facepieces with HEPA filters, which were tested with a most penetrating aerosol at 32 and 85 Lpm.

Therefore, it is clear that there is no need to conduct filter tests at flow rates above 85 Lpm.

Minute Volume and Peak Air Flow Capabilities for PAPR, SAR, and SCBA
Physiological considerations: PAPR, continuous flow and pressure demand SAR, and pressure demand SCBA are generally classified as positive-pressure devices. That is, they are intended to maintain a slightly greater than ambient pressure in the facepiece, hood, or helmet during both inhalation and exhalation. Because contaminants are unlikely to migrate from ambient pressure to a higher pressure, positive-pressure respirators are assumed to provide more protection than negative-pressure respirators.

In order to maintain positive pressure at all times, a respirator's air flow capability must exceed both V_e and PF_I (neglecting the respirator's dead volume). It has been suggested that, because PF_I values in excess of 400 Lpm can occur, positive-pressure respirators should be able to supply air at this rate to prevent negative pressure excursions in the breathing zone.8

Performance considerations for tight-fitting respirators: Numerous reports of tight-fitting positive pressure respirators allowing negative pressure excursions in the wearer's breathing zone (known as overbreathing) exist in the literature.9,10,11 Most of these reports involved respirators designed and approved in the 1970s and early 1980s. Partly as a result of these reports, higher air flow performance criteria were adopted for SCBA used in firefighting.12 These criteria require a minute volume of at least 103 ± 3 Lpm with a peak flow capability in excess of 300 Lpm. It was believed that the respiration requirements of 98 percent of firefighters would be satisfied at this level of performance. That is, it was thought positive pressure would be maintained essentially all the time for 98 percent of firefighters. At least two studies conducted with pressure-demand SCBA that meet the higher air flow requirements have reported negative pressure excursions in the facepiece at high work rates.13,14

Campbell, et al. collected data on pressure inside the facepiece of SCBA during actual firefighting activities.13 They used this information to develop a sophisticated mathematical model to estimate the effect of overbreathing on the protection provided by the SCBA. Their model takes into account the duration and frequency of overbreathing, protection provided when positive pressure is maintained as well as when overbreathing occurs, and additional factors. The authors concluded that the effect of overbreathing is not significant and recommended that the assigned protection factor (APF) remain at 10,000.

A much simpler approach using information from Burgess and Crutchfield supports this conclusion.14 In this study, firefighters wearing SCBA exercised on a treadmill at 80 percent of their aerobic capacity. It must be emphasized this is a work rate that most individuals could not sustain for more than 10-15 minutes at a time. Facepiece pressure was monitored and V_e was determined for each subject. The authors found that in the worst case, a subject experienced negative pressure in the facepiece approximately 5.75 percent of the time.

It is important to realize the wearer still receives protection from the SCBA even during periods of negative pressure. It is reasonable to assume the APF of 100 listed for negative pressure (demand) SCBA15 during negative pressure excursions. This is a conservative approach because the duration of negative-pressure excursions in a pressure-demand SCBA is considerably less than the duration of negative pressure in the facepiece of demand SCBA.11 Using this information, it is possible to calculate a "protected" exposure level for the wearer. The APF for demand SCBA is applied during overbreathing and the APF for pressure-demand SCBA is used when positive pressure is maintained.

Example:
A firefighter is wearing an SCBA in an atmosphere containing 100 ppm carbon monoxide (CO). Specific tasks require an effort of 80 percent VO₂ max. The firefighter is able to maintain this work rate for 15 minutes. During these tasks, negative pressure in the facepiece occurs 6 percent of the time (a total of 0.9 minute). The exposure limit for CO is 25 ppm as an eight-hour TWA.16

The resulting protected exposure is not appreciably different than the protected exposure (0.01 ppm) if overbreathing had not occurred. The worker is well protected in either case. The difference in exposure narrows further if an eight-hour TWA is calculated, assuming a reasonable number of periods of high work rate that result in overbreathing.

Example:
The firefighter in the above situation performs four periods of work at 80 percent VO₂ max during an eight-hour shift. If it is assumed the CO concentration remains constant and the firefighter is exposed the entire shift (both extreme assumptions), the protected eight-hour TWA exposure would be 0.02 ppm

Thus, it can be seen that a plausible number of negative-pressure excursions have a negligible effect on the protection provided by a tight-fitting positive-pressure respirator. This notion is supported by a WPF study on a full facepiece PAPR.17 While facepiece pressure was not measured in this study, the authors estimated the work rate was moderate to high. It is therefore likely that some overbreathing occurred. The 5^th percentile WPF of 1335 found in this study is consistent with the APF of 1000 for tight-fitting PAPR.15

Performance considerations for loose-fitting respirators: There are fewer published studies that have reported negative-pressure excursions in loose-fitting PAPR or SAR. The APF of 25 for loose-fitting facepiece PAPR was in part based on a simulated workplace protection factor (SWPF) study by da Rosa et al., which reported overbreathing at 80 percent of the subjects' maximum work rate.18

A more recent SWPF study also found negative-pressure excursions with hooded loose-fitting facepiece PAPR and SAR when subjects ran in place.19 The authors reported no consistent relationship between the pressure measurements and the SWPF measurements. Mean air flow rates were above the required six CFM for all the devices, but flow rates as low as 4.66 and 5.27 CFM were found for one hooded and one loose-fitting facepiece PAPR, respectively.

Interestingly, the 5^th percentile SWPF measurements remained above 150,000 for these two devices. Fifth percentile SWPF values ranged from 86,000 to more than 250,000 for all but one of the remaining devices. The SAR with the highest mean and minimum air flows did not have the highest measured SWPFs, and the poorest performing device did not have the lowest mean or minimum air flow. The results of this study appear to indicate that:

1. The effect of occasional negative-pressure spikes on the protection provided by loose-fitting respirators is not great, and
2. Design characteristics other than air flow rate influence the performance of loose-fitting respirators.

Finally, it should be noted that recent WPF studies on a loose-fitting facepiece PAPR and SAR with hoods or helmets support the current APFs of 25 and 1000, respectively.20,21

Summary and Conclusions
Several respirator performance criteria are set to satisfy the physiological requirements of the worker. In particular, V_e and PF_I must be understood and used appropriately in the design process. Filtration principles and the nature of workplace aerosols must also be understood to determine appropriate test conditions for particulate respirator filters. Current filter test criteria assure that significant aerosol penetration will not occur in the workplace.

It is unlikely that existing positive-pressure respirators can assure positive pressure in the breathing zone at all times for all wearers. High, unsustainable work rates and activities such as running in place have been shown to cause negative-pressure excursions under laboratory and workplace conditions. Because these excursions are brief and infrequent for today's respirators, both logic and performance measurements indicate they have a negligible effect on exposure.

Filters could no doubt be designed to pass extreme test criteria. Similarly, it may be possible to design positive-pressure respirators to maintain positive pressure 100 percent of the time for all users. However, it is likely that cost and size would increase and comfort could decrease. Further, available evidence demonstrates that today's respirators provide their expected level of protection when properly selected, used, and maintained. As such, there is no demonstrated need for radical design changes.

References
1. "Approval of Respiratory Protective Devices," Code of Federal Regulations Title 42 (1), Part 84. 2000. pp 553-614.

2. Berndtsson, G: "Today's Filter Paradox: Too Good to Be True Protection." Paper presented at the American Industrial Hygiene Conference and Exposition, Toronto, ON, Canada. June 1999.

3. Hinds W.C.: Aerosol Technology. New York: John Wiley and sons, 1982. pp. 172-182.

4. Office of Scientific Research and Development: Inspiratory and Expiratory Air Flow Measurements on Human Subjects With and Without Resistance at Several Work Rates by L.Silverman, B. Lee, T. Plotkin, L. Amory, and A. R. Yancey (OSRD No. 5732). Office of Scientific Research and Development, 1945.

5. Wallart, J.C.: "The Effect of Speech on Peak Flow Values at Varied Levels of Work Load." Paper presented at the International Society for Respiratory Protection 8th International Conference. Amsterdam, The Netherlands. September 1997.

6. "Respiratory Protective Devices; Tests for Permissibility; Fees," Code of Federal Regulations Title 30, Part 11, Subpart K. 1994.

7. Nelson, T.J.: The assigned protection factor of 10 for half-mask respirators. Am. Ind. Hyg. Assoc. J. 56: 717-724 (1995).

8. Backman, L: "Airflow Requirements." Paper presented at the International Society for Respiratory Protection 9th International Conference. October 1999.

9. Raven, P.B., O. Bradley, D. Rohm-Young, F. McClure, and B. Skaggs: Physiological response to "pressure-demand" respirator wear. Am. Ind. Hyg. Assoc. J. 43(10): 773-781 (1982).

10. Dahlback, G.O. and L. Novak: Do pressure-demand breathing systems safeguard against inward leakage? Am. Ind. Hyg. Assoc. J. 44(5): 336-340 (1983).

11. Air Force School of Aerospace Medicine: Physiological Limits of Firefighters by L.G. Myhre, R.D. Holden, F.W. Baumgardner, and D. Tucker (ESL-TR-79-06). Tyndall Air Force Base FL: Air Force Engineering and Services Center, 1979.

12. National Fire Protection Association: Standard on Open-Circuit Self-Contained Breathing Apparatus for Fire Fighters (NFPA 1981). Quincy, MA: National Fire Protection Association, 1987.

13. Campbell, D. L., G.P. Noonan, T.R. Merinar and J.A. Stobbe: Estimated workplace protection factors for positive-pressure self-contained breathing apparatus. Am. Ind. Hyg. Assoc. J. 55(4):322-329 (1994).

14. J.L. Burgess and C.D. Crutchfield: Quantitative respirator fit tests of Tucson fire fighters and measurement of negative pressure excursions during exertion. Appl. Occup. Environ. Hyg. 10(1): 29-36 (1995).

15. American National Standards Institute: American National Standard for Respiratory Protection (ANSI Z88.2). New York: American National Standards Institute, 1992.

16. American Conference of Governmental Industrial Hygienists: 2001 TLVs and BEIs: Threshold Limit Values for Chemical Substances and Physical agents and biological exposure indices. Cincinnati, OH: American Conference of Governmental Industrial Hygienists, 2001.

17. Colton, C.E. and H.E. Mullins and C.R. Rhoe: "Workplace Protection Factors for a Powered Air Purifying Respirator." Paper presented at the American Industrial Hygiene Conference and Exposition. May 1990.

18. da Rosa, R.A., C.A. Cadena-Fix, and J.E. Kramer: Powered-air purifying respirator study. JISRP 8(2): 15-36 (1990).

19. Cohen, H.J., L.H. Hecker, D.K. Mattheis, J. S. Johnson, A.H. Biermann, and K.L Foote: Simulated workplace protection factor study of powered air-purifying and supplied air respirators. Am. Ind. Hyg. Assoc. J. 62: 595-604 (2001).

This article originally appeared in the June 2003 issue of Occupational Health & Safety.