Since the early 1930's there has been a mistaken belief concerning what happens when the refrigerant gas, R-22, also known as chlorodifluoromethane, is exposed to high temperature heat sources. Admittedly, several compounds, including hydrogen fluoride (which then forms hydrofluoric acid when it comes into contact with water vapor in the air), hydrogen chloride (which then forms hydrochloric acid when it comes into contact with water vapor in the air) are formed during thermal decomposition. However, the focus of all the literature and all of the training in the HVAC industry has been on the fact that a chemical compound called phosgene (carbonyl chloride) is formed when R-22 is heated. The reason for the focus on phosgene will become more evident as you read this article, but suffice it to say that phosgene has been used as a chemical weapon – it is highly toxic.

R-22 was one of several refrigerant gases, collectively referred to as CFC’s, that had been invented in the mid- to late 1920s to replace the highly toxic refrigerant gases then in use. Although it was widely accepted that the CFC’s were non-toxic while still performing well as refrigerants, Underwriter’s Laboratories conducted comprehensive testing of the CFC’s in the late 1920's and early 1930's. Because of the chemical makeup of CFC’s one concern of UL was what would happen if these gases were heated to high temperatures. In answering that question as to R-22, UL anticipated that phosgene might be one of the products formed. Subsequent testing by UL verified (incorrectly) this prediction.[1] Since that time, this belief that phosgene is formed when R-22 is heated has become an incontrovertible postulate in the HVAC industry.

Phosgene was a concern for three reasons. First, as mentioned, it was used as a war gas in World War I and is extremely harmful even in very low doses. Second, very low amounts of R-22 were thought to be needed to create toxic levels of phosgene. Third, there was one critically important, but quite unique aspect of phosgene. Every other product of R- 22 thermal decomposition is highly noxious and unpleasant. When these substances come into contact with mucous membranes in the eyes, mouth, nose and throat, they produce great discomfort and pain. The detection threshold is such these compounds are detected well before they reach harmful levels and the substantial discomfort that is caused upon hitting the detection threshold causes people to flee quickly...very quickly. On the other hand, phosgene is insidious. At levels which are dangerous to humans, phosgene smells something like new mown hay...sort of sweet and musty. It is not even an unpleasant smell much less a noxious one.

Given the evidence and universal belief that R-22 produces phosgene when it comes into contact with hot surfaces, its presence in the home is viewed as a potentially dangerous and perhaps even lethal situation. And, the fact that the damage may be done by phosgene before one even knows he’s been exposed, gives rise to all kinds of potential claims. Consequently, when a component of an HVAC unit leaks refrigerant gas into a home (or other inhabited location), lawsuits against the manufacturers of HVAC equipment claiming damage due to exposure to phosgene will result. While the great weight of historical information and documentary evidence actually supports such an argument, scientifically, we now know it is wrong.

In 2003, the authors herein acted as co-defense counsel for two separate defendants in a case where a family of four, we’ll call them the Smith family, made a “R-22 = phosgene” contention. There was no dispute that the evaporator coil of the Smith’s A/C system leaked R-22, and it was possible if not probable that a small amount of R-22 actually got from the unit located in the garage into their home. With evidence of heat sources from gas stove burners, pilot lights, toaster elements, curling irons, etc. the family was convinced that they had been exposed to phosgene. The Smith’s claimed to have experienced a wide variety of symptoms: some consistent with phosgene exposure and others that could not have possibly resulted from phosgene exposure. They claimed that the “phosgene contamination” damaged their health, and permanently and irreparably damaged their home and their personal property that was in the home. The only solution was to destroy all of their personal property, raze the house to bare land, and rebuild everything.

Notwithstanding a virtual mountain of evidence against them, defense counsel, using a toxicologist, a forensic organic chemist and an article which was written by a DuPont researcher who worked with Freon routinely, were able to systematically layout not only the reason phosgene could not be obtained from heating R-22 but were able to explain the UL testing errors that created the original mistaken belief as to this phosgene being produced from heating R-22. Unfortunately, the Smiths were so heavily invested in their belief and position, while they heard the same evidence the jury heard, the Smiths could not accept either the indisputable scientific evidence or the jury’s decision.

Unfortunately, there was nothing about the Smith decision which became known outside of the actual case itself. So, when the R-22-phosgene issue arose in another case, Johnson v. American Standard, Inc., the attorneys in that matter had no knowledge as to what had happened in the Smith case. Those defense attorneys clearly did not know about organic chemistry. They had no alternative but to use HVAC technicians as expert witnesses. Since virtually the entire world of HVAC technicians all believe the R-22-phosgene connection, the only information available to defense counsel was that when you heat R-22, you produce phosgene!

The Johnson matter wound its way through the court system for years, eventually being decided by the California Supreme Court. It was dismissed by the trial court using a somewhat new legal challenge called the “sophisticated user” doctrine. The Appellate Court upheld the trial court. Finally, the California Supreme Court was asked to review the decision of the lower court. This case was decided April 3, 2008. It began when an HVAC technician did work on a commercial A/C system which required doing some brazing work [2] on a component of the system. Using standard protocol, the technician removed the R-22 refrigerant gas from the system before he began brazing, however trace amounts remained. While brazing, he was hit with a very pungent, potent and unpleasant odor that he contended was phosgene. He alleged the highly toxic gas caused him permanent lung damage. Since California workers compensation law prevented him from suing his employer, his lawyer creatively sued the manufacturer of the A/C system (American Standard) claiming that it owed the plaintiff a duty to warn him about exposure to phosgene if he heated a part of the system that contained R-22.

The manufacturer defended the lawsuit with the sophisticated user doctrine. The plaintiff’s expert sought to establish that the plaintiff was exposed to phosgene which caused him lung damage. The defendant’s expert not only agreed with the plaintiff’s expert but attempted to also prove that phosgene was produced by heating R-22 AND that this was common knowledge among the HVAC technicians of the world. Thus, there was no need to warn Johnson of something he (and every other HVAC person) obviously knew or should have known! The court agreed with the defendants’s position, finding that everyone in the HVAC business knows about the R-22/phosgene connection. The problem is it simply isn’t true.

While the defendant’s position was legally sound it is scientifically wrong and dangerously so. The next time that American Standard has to deal with a phosgene case, it may not be a Johnson-type sophisticated used case, but a homeowner case. The difference will be that in the Smith case, notwithstanding the overwhelming evidence against our position, we were able to demonstrate to the jury that phosgene is not a thermal decomposition product of R-22. That will not be a viable option for American Standard (it admitted R-22 can create phosgene), or perhaps most other HVAC manufacturers. They will be stuck with American Standard’s position in the Johnson case and will not know how to challenge it. Furthermore, there is the implication that even the California Supreme Court has now validated the finding concerning the phosgene/R-22 connection.

Consider where this leaves us. There is a 75 year old mistake that appears to have no legitimate prospect for being corrected and even worse, seems to be gaining proponents and is being re-enforced, both in the HVAC trades and in the legal arena. In an already litigious society, this simply encourages more law suits. It also victimizes innocent people. When a homeowner is concerned about what he may have been exposed to when he is told his A/C unit had to be recharged with Freon by a technician performing routine maintenance on the system, he goes to the Internet and he can find a variety of horror stories about phosgene poisoning from a leaky A/C system. For example, see the following website: http://www.toxichome.net/the_real_story.htm. The family who has this website, the Veerkamps, were charged a large amount of money by lawyers and consultants to tilt at windmills.

Consider a nationally-known company who, on its website, makes the following statement:

“The most dangerous and most common cause of indoor air contamination comes from leaking refrigerants contacting a heat source that can convert the freon into deadly phosgene gas (mustard gas). Carbon sooting frequently evidences the presence of this reaction. Almost all air conditioners and many refrigerators leak freon. There are many sources of heat in the home (heat strips, pilot lights, electric range and oven, toasters, etc.), which are capable of converting the freon into phosgene gas.”

This company offers to “test” your home for these substances for fees ranging from $300 to $5,000. The entire general public who have either A/C units or refrigerators are at risk in the sense that they are told that ordinary household appliances may expose them to a horrible war gas and that they can have their homes tested for its presence by spending $5,000, when, in fact, none of the information quoted above is accurate, correct or truthful. This 75 year old mistake is causing people to spend substantial sums of money looking for the nonexistent while simultaneously creating the potential of creating the next area of mass tort litigation in the United States.

ENDNOTES

[1] It is not the intent of the authors to be critical of the UL investigators who made these findings. Based upon the then state of knowledge of electrochemical bonding energies and the unavailability of tools such as gas chromatographs and mass spectrometers routinely used by chemists of today, the UL investigators came to conclusions that were completely appropriate in the 1930s. It is only with the advantages of contemporary increases in knowledge and modern diagnostic devices which make our hindsight 20/20 enabling us to “see” and understand the mistake made by the original investigators.

[2] Brazing is a process that uses a propane or butane type torch and is more or less a process that lies between soldering and welding. Most significantly, it involves creating high temperatures of the objects being brazed.

Kent Seitzinger John Broghammer

Dear Editor,

I would like to thank Dr Collinson (1) for his article about the utility of ischemia modified albumin (IMA) in chest pain patients. However the results may be overstated regarding IMA in patients with chest pain, particularly in patients with acute myocardial infarction (AMI).

First of all excluding ST segment elevation MI patients could create a selection bias. Because some of these patients have normal cardiac troponin (cTn) levels initially although they have a diagnostic ECG for ST segment elevation myocardial infarction.

The second point to emphasize is the negative predictive value of both a negative cardiac troponin and IMA was stated as %100. However the contribution of IMA to this rate is minimal because IMA was positive in only two acute myocardial infarction patients with a negative cTn (2/37). So it seems that cTn was able to exclude the great majority of AMI patients. And also a negative test might not be able to allow exclude AMI because 19 patients had a positive cTn without a positive IMA (19/37). And also the suggestion in the discussion section that a positive IMA alone will need a follow up to confirm AMI may be controversial. Because of the 342 study participants with a positive IMA, only two of them were AMI as well as a negative cTn (2/342). It seems IMA contributes little to cTn in diagnosing AMI.

As it is known, cardiac troponins have a high sensitivity and specificity in diagnosing AMI, however these high sensitivity and specificity decline in the early hours of the symptoms onset. The critical question here is “How much IMA adds in diagnosing AMI in the early hours of symptoms onset to cTn, ECG, likelihood or risk stratifications made by different societies and the coagulation and inflammation markers studied so far like d-dimer, C-reactive protein and IL-6?”. As it was mentioned in the manuscript, the median time of symptoms onset was 6 hours in the study population. These findings do not inform us about the utility of IMA in the early hours of AMI which was the most critical challenge. Furthermore 37 AMI patients as a pathologic group may be too small to make suggestions for the diagnosing power of IMA in AMI.

References

1. Collinson PO, Gaze DC, Bainbridge K, et al. Utility of admission cardiac troponin and ‘Ischemia Modified Albumin’ measurements for rapid evaluation and rule out of suspected acute myocardial infarction in the emergency department. Emerg Med J. 2006;23:256-261.

Dear Sir,

Please find enclosed our article submitted to European Journal of Anaesthesia, which was presented as a poster in Euro anaesthesia conference in Copenhagen on 31 May 2008 and also published in their May supplement, including the images.( A.Arora, K.Kada, A.Ferguson; Look again! Delayed traumatic rupture of diaphragm radiologically simulating a pneumothorax; Eur J Anaesthesiology 2008; 25 (suppl 44): 13AP1-3).

Our case is similar to yours except for the fact that a CT scan was done based on clinical finding and X-ray before considering drainage and hence it avoided the complication described in your presentation.

Background and Goal of Study:

Diaphragmatic injury with herniation is a well-documented complication of both penetrating and blunt trauma. It occurs in approximately 3% of abdominal and 1.5% of thoracic injuries with a 2:1 ratio of penetrating to blunt trauma. It is left-sided in 70-80% of cases. Diagnosis requires a high index of suspicion as many cases are missed at first presentation.

We describe a case of delayed traumatic diaphragmatic hernia radiologically simulating tension pneumothorax.

Materials and Methods:

6 months prior to presentation a 52 year old bricklayer had fallen from a ladder. Chest X-ray at that time demonstrated undisplaced rib fractures with no pneumothorax and follow-up was not thought necessary.

He came to hospital on this occasion with a one month history of dull upper abdominal pain. Repeat chest X-ray showed unilateral radiolucent hemithorax on the left with a few broken ribs on that side and mediastinal shift to the right; a scaphoid abdomen was noted on examination.

After review of the film, needle thoracocentesis was deferred and a CT scan performed. This showed a diaphragmatic hernia containing transverse colon. He underwent thoraco-abdominal repair of diaphragmatic hernia.

Results and Discussion:

The injury is uncommon, occurring in about 4% of victims of major closed chest or abdominal trauma.(1-4) Large pressure differentials between the chest and abdomen, generated at the time of injury, cause tearing of the dome of the diaphragm, more commonly on the left as the liver protects the right side by spreading the forces more evenly. The injury is indicative of severe trauma and injuries that accompany such trauma often take priority at presentation, resulting in injury to the diaphragm being overlooked.

In some series this has been overlooked in up to roughly two thirds of patients. The patient may present months or years later in one of two ways. The first way is with mild symptoms typically in the lower chest, or epigastric pain occurring intermittently and often brought on by taking food; physical signs are minimal. This is the so called "latent" stage.

The clinical picture may change dramatically and the patient present in the so called "late" stage with severe symptoms of pain, breathlessness, and intestinal obstruction or gastrointestinal haemorrhage.

Examination may show that the patient is in shock. There may also be evidence of mediastinal shift, and bowel sounds may be audible in the chest. Nearly all such patients will have presented previously with minor "latent" symptoms.

Treatment at any stage of the illness requires an operation. Reduction of herniated contents and closure of the defect is undertaken at the time of diagnosis through a laparotomy incision in the case of an acute onset and by an appropriate thoracotomy in chronic cases.

Surgery performed during the "latent" stage carries a perioperative mortality of under 10%4. When surgery is undertaken in the presence of strangulation, in the "late" stage, there is a mortality of between 20% and 80%.(3)

Conclusion:

As most patients will present with minor symptoms before presenting with a severe illness the diagnosis must always be borne in mind. In all cases an accurate history (particularly trauma), clinical alertness, and a chest radiograph are the key to diagnosis.

References:

1. Alyono D, Perry JF Jr. Impact of speed limit. 1. Chest injuries, a review of 966 cases. J Thorac Cardiovasc Surg 1982; 83:519-22.

2. McCollum C, Anyarnwy CH, Umeh BUO, Swarup AS. Management of traumatic rupture of the diaphragm. BrJ3 Surg 1987; 74:181-3.

3. Payne JH, Yellin AE. Traumatic diaphragmatic hernia. Arch Surg 1982; 117: 18-24.

4. De la Rocha AG, Creel RJ, Mulligan GWN, Burns CM. Diaphragmatic rupture due to blunt abdominal trauma. Surg Gynecol Obstet 1982; 154:175- 80