PFO and Diving - part 2
Do you have a hole in your heart and if so, does it matter?

By Dr. Sawatzky

In review, when we dive, we take up inert gas in accordance with Henry's Law of solubility. If we finish the dive with a relatively small amount of additional dissolved inert gas, it slowly diffuses into the blood and exits the body through the lungs. If we have a greater inert gas load, it tends to form bubbles in the veins. These bubbles are washed back to the right heart and pumped out to the lungs. The lungs are usually excellent filters. They trap the bubbles until the gas in the bubbles diffuses into the alveoli and the bubbles disappear (about 45 minutes). In the normal resting adult, blood is only moving through about 10% of the capillaries in the lungs and the other 90% are empty. When we exercise and the amount of blood being pumped by the heart increases, more of the capillaries in the lungs open and have blood flowing through them. Therefore, in the resting or lightly working diver, if bubbles block 10% of the capillaries in the lungs, other capillaries will open and allow blood to continue flowing through the lungs as if nothing has happened. There would be no symptoms in the resting diver until over 90% of the circulation through the lungs was obstructed by bubbles. From animal experiments, we know that the lungs can trap several litres of gas without problem.

In the diver who has a PFO or a shunt in the lungs, bubbles can move through the PFO or move through the shunt and enter the arterial circulation (bypassing the lung filter). Bubbles are then pumped out to the body and the first major arteries branching off the aorta are the carotids, supplying the brain. If you send bubbles to the brain, it is the same as suffering arterial gas embolism from pulmonary over inflation and that is often very serious or fatal. Therefore, we would expect that every diver who has a PFO, and has intravascular bubbles after a dive, would suffer arterial gas embolism!

Diving medical specialists became interested in this possibility in 1989, when several divers who had conducted relatively normal dives, were presented with very serious decompression illness (DCI). These divers were investigated and it was deter mined that more of them had PFOs than would be expected in the general population. This generated a great deal of interest and a whole series of studies have since been conducted to try to quantify this increased risk.

It has been exceptionally difficult to answer this question. There are many problems. First, PFO is very common (up to 40% of the general population) and there is no reason to expect that divers have an incidence of PFO different than the rest of the population. Studies have shown that up to 40% of divers have PFOs.

Second, the injured diver frequently had many previous dives (many had hundreds and a few had more than 1,000 previous dives) and yet they had not suffered DCI previously. What was different about the dive on which they suffered DCI? Many divers who are known to have a PFO have done thousands of dives without incident.

Third, we know that a PFO only has significance if the diver has intravascular bubbles during or after the dive. Studies have shown that recreational sport divers generate relatively few bubbles after diving. In the tropics, where divers tend to push their bottom times and often use computers, up to 25% of divers will have a few bubbles after diving and a few divers will have many bubbles. Technical and commercial divers, who routinely conduct more strenuous dives (from a decompression perspective), bubble more frequently after a dive and some have many bubbles. Finally, while we were developing the Canadian Forces Air and Helium/Oxygen decompression tables at DCIEM, the divers were pushing the limits of the decompression model on every dive. Approximately 50% of the divers bubbled and many of them bubbled heavily. We did not test for PFO in these divers but there is no reason to expect that they would be different from other divers. Therefore, 30 to 40% probably had a PFO. On the first 3,234 person dives we had only 77 cases of DCI and on the hundreds of dives conducted since I analyzed the data, there have been lots of bubbles but no DCI (in general they have been doing more conservative dives). If PFO was a major risk factor in DCI, we should have seen hundreds of serious cases.

Dr. Alfred Bove, one of the senior diving cardiologist in the world, recently published a paper on the "Risk of decompression sickness with patent foramen ovale" (Undersea Biomedical Research, Volume 25, Number 3, page 175-8, 1998). In this paper, Dr. Bove calculated the risk of a diver with a PFO developing DCS from data published in three previous articles where the frequency of PFO had been determined in populations of divers, some of whom had developed DCS. These three articles contained data from 393 divers who had been screened for PFO. Of these 393 divers, 151 had suffered DCI (65/151 or 43% had a PFO). Reference frequencies of DCS were taken from a sport diving population (DAN, estimated 2.5 million dives), US Navy divers (648,488 dives), and commercial divers (43,063 dives). The sport and military divers had an incidence of DCS of approxi mately three cases per 10,000 dives (sport divers often do not report mild DCS and most military dives are short, shallow training dives where there is virtually no risk of DCS). The commercial divers had 35 cases of DCS per 10,000 dives (more strenuous dives?). The incidence of serious, type II, DCS was roughly the same in all three populations at two cases per 10,000 dives.

Dr. Bove determined that the risk of DCS was 1.9 times greater in a diver who had a PFO and the risk of serious, type II DCS was 2.5 times greater than the expected incidence. He concluded that the risk of DCS in a diver who had a PFO was still too low to justify screening divers for PFO. He also concluded that the absolute risk was so low that there was no reason for a person, known to have a PFO, to stop diving or not take up diving. These findings are very interesting but are there any other studies that will help us advise divers?

Several studies have shown that divers who have serious, type II DCS and divers who have symptoms that start within 30 minutes of surfacing have a higher incidence of PFO than would be expected. Conversely, divers who develop mild, type I DCS or have symptoms that start more than 30 minutes after surfacing, have the same incidence of PFO as the general population. It has also been shown that CNS symptoms are more common in divers with PFO but spinal cord symptoms are not. Studies in animals have shown that if you inject gas into the carotid arteries supplying the brain, and if the animal has not been dived, the bubbles block relatively small arteries and have relatively minor effects on the circulation to the brain. However, if the animal is dived before the gas is injected so that the brain tissues are supersaturated with dissolved inert gas, something entirely different happens. In this circumstance, when the injected bubbles stop in the small arteries, gas moves from the tissues into the bubbles and the bubbles grow backwards through the circulation. It takes only a few seconds before the bubbles have grown so large that they completely fill the brain circulatory system. No blood is flowing through the brain!

It seems reasonable to conclude that if a diver has a PFO but does not conduct dives that result in intravascular bubbles (relatively low inert gas supersaturation), the PFO will have no significance and no effect on the risk of the diver developing DCS (DCS extremely unlikely). The primary deficiency in the data Dr. Bove used to calculate the effect of PFO on the incidence of DCS was that the bubble status of the divers was unknown. It is likely that most of the divers, on most of the dives, had no bubbles or only a few bubbles. Therefore, for most of the divers, on most of the dives, it would not matter if they had a PFO or not. Given this information, I believe we can reasonably conclude that the risk of DCS in a diver who has a PFO, and who conducts dives that result in intravascular bubbles, is greater than the risks calculated by Dr. Bove for PFO and diving in general. In addition, it would seem reasonable to assume that the effect of the PFO on the risk of the diver developing DCS in the presence of intravascular bubbles would depend on the size of the PFO and the number of bubbles.

Therefore, for the average sport diver who always stays inside the no decompression limits, the risk of forming significant numbers of intravascular bubbles is extremely low. If the diver is not forming intravascular bubbles, it does not matter if they have a PFO or not and there is no reason to screen for PFOs. The sport diver can increase their safety even more by taking a nitrox course and pretending that they are diving on air when they calculate their no decompression limits. This is highly recommended for all divers over age 40, and for divers who are younger but have an increased risk of DCS for other reasons, when they push the no decompression limits.

Most military and commercial divers, on most working dives, will also not develop large numbers of intravascular bubbles and therefore, it will not matter if they have a PFO. However, for military, commercial and technical divers who are conducting dives where the risk of intravascular bubbles is high, I believe it would be a good idea for them to undergo a simple transthoracic echocardiographic examination. This will only detect the larger PFOs but smaller PFOs should have less significance. For the diver who knows that they have a PFO, they should always plan their dives in such a fashion that the risk of forming intravascu lar bubbles is very small. I have always been very conservative when calculating my decompression profiles. Before I do much more technical diving however, I will definitely find out if I have a PFO!

In conclusion, for most divers on most dives it does not matter if they have a PFO or not and there is no reason to screen for it. For divers who are conducting dives where there is a significant risk of them forming intravascular bubbles, I believe they should be screened for PFO. If it is determined that they have a PFO, they should adjust their diving so that the risk of forming intravascular bubbles is reduced. If it is determined that a person has a large PFO or atrial septal defect, it should probably be repaired, not because of the risk of DCS, but because it is also a risk factor for stroke (a blood clot can also pass through a large PFO). For divers, a PFO can be made insignificant by simply avoiding dives that could be expected to result in intravascular bubbles.