If you look at the EPA site, you will find that flammability is only ONE of the things EPA looks at in approving a refrigerant. Also, different refrigerants are approved for different purposes (marine, commercial chillers, etc.). Something not approved for one use is sometimes on a different approved list, as the hazards/pollution index/accident probability factors add up differently for different applications.
Also, nobody claims the EPA as the final authority on environmental issues (Some of their radiation regulations, for example, are ludicrous - coal burning power plants pump more long half life alpha and gamma radioactivity out their stacks than an n plant collects for burial. The n plant can't release anything, the coal plant is exempt! 50% of US power comes from coal, 17% from N plants. So which one is the environmental radiation source?) However, on refrigerants the EPA states their rulings are made solely on the basis of data PROVIDED BY THE MANUFACTURER, not by their own testing. The flammability, Hydrocarbon loss rates through non-barrier hoses designed for r12 only, the compressor life span effects, etc. are all from the refrigerant manufacturer, and therefore probably present a 'best case' estimate.
Finally, there is a large difference in flammability hazard between liguid gasoline and a hydrocarbon vapor under high pressure. Gasoline must be vaporized and mixed with air to explode, Hollywood notwithstanding. That vaporization must fall within a relatively narrow fuel/air ratio, otherwise it will be too lean or rich to explode. That vaporization occurs inside a nice safe metal cylinder head. Running tubing and connectors with highly pressurized hydrocarbons which may vaporize under the engine compartment is a different risk factor than relatively low pressure fuel lines from the fuel pump. A low pressure spray is less likely to produce an explosive air/fuel mixture than a high pressure refrigerant leak, which may have a lower flashpoint, greater volatility, and wider ignition temperature and fuel/air range than gasoline, as well as a different 'bang/cc' factor - it may have much more punch than gasoline! I have seen less than 1cc of toluene liquid produce a vapor mix in a fume hood accidently ignite from static electricity, despite safety use of safety cans and grounding straps. That 1 cc shattered the safety glass of the chemical fume hood sash, split the fume hood ducting, and was heard throughout the building.
Keep cool, but think safe.
2003 Firemist Red/grey leather SL 500
2015 Palladium Silver/black mbtex GLK 350
1987 Smoke Silver/burgundy mbtex 300E Sportline (SOLD)
Click to see 87 300E