Posted first on Pulse May 8 2015
By Dr. Jan De Backer, CEO FLUIDDA
“The same principles that keep us going are the ones that could bring us down”
During our training as aerospace engineers my classmates and I were taught the importance of redundancy. The two terms I remember vividly were “fail safe” and “safe life”. The term “fail safe” meant we had to include so-called back-up systems in our designs. A commercial airliner, for instance, has multiple autopilots in case one or more malfunction. The same holds true for the power supply onboard. The term “safe life” on the other hand meant that the part under consideration had to be designed to last the duration of the design life without repair or interruption. This design principle was reserved for parts that were very difficult to examine or replace during the lifetime of the aircraft and therefore had to be over-dimensioned.
Shortly after graduating from engineering college I saw the opportunity to apply methods and principles from the aerospace world into the medical field, pulmonary medicine to be precise. By studying the human body extensively I realized that whoever created us must have had the same courses on redundancy! Our human body is a remarkable combination of “fail safe” and “safe life”. By default we are designed with parts that are suppose to last us a lifetime. Apart from a number of exceptions most of us will have to get by with one heart, one set of lungs, one liver, etc. However within those organs the fail safe principles seem to be applied. The capillary system that oxygenates our organs is made up of many many vessels and we don’t have just one big airway but a branching structure serving a large number of independent alveoli. This “fail safe” measure ensures that when one of our veins, arteries or airways is blocked the other parts can take over to preserve the vital functions.
When the “defect” is severe enough, the patient will present with symptoms. In case of significant bronchoconstriction (narrowing of the airways as occurs for example during an asthma attack), the patient will have difficulties breathing and will make a wheezing sound. However when the initial onset of disease or trauma is mild, often the patient remains sub-symptomatic since the healthier, unaffected regions compensate for the loss of functionality of the diseased areas. To make things worse, with progressing disease our body tends to compensate even more, so as long as there are enough healthy zones the function is maintained. When the unaffected areas become so few that they lose the ability to compensate, the patient will start to experience symptoms followed by a, in many cases rapid, decline in function. That is when our biological redundancy has become dangerous.
The level of danger depends on the sensitivity of the diagnostic tools at our disposal to detect early signs of disease. Andreas Vesalius already realized in the 16th century that we cannot just rely on symptoms, patient related outcomes or other derivative measures to understand pathophysiology but that a direct view at the organ is required. We can argue that these principles still hold true today but that the implementation and success of tools that adhere to these principles differs greatly from one therapeutic area to another. In respiratory medicine we rely on a “stress test” to assess the state of the patient’s respiratory system. The most commonly used parameter in this test is the “Forced Expiratory Volume in One Second” or FEV1. This is simply the amount of air a patient can exhale in one second. This value is subsequently compared to reference values of corresponding healthy volunteers and expressed as a % predicted. So if a patient has an FEV1 of 100% predicted, the lung function is considered normal. We inherently assume that when certain areas in the lung are obstructed this forced maneuver will indicate this. At the same time we believe that when this parameter is stable the state of the respiratory system has not changed. These assumptions however disregard the biological redundancy as described above in that our organs are designed with excess capacity and back up systems. It is therefor fair to assume that minor or even moderate airway obstructions might not be detected even when stressing the system. At the same time it could very well happen that a stable FEV1 is the result of compensation by healthy lung zones for the initial and progressive failure of a diseased lung area rather than a sign that nothing changed inside the respiratory system.
The animation below illustrates this point based on real patient-specific data:
The animation tells the story of a patient suffering from Cystic Fibrosis, a genetic disease affecting children and adolescents. Cystic Fibrosis is characterized by excessive mucus production in the lung that is very hard to clear so patients run the risk of literally drowning in their own mucus. The patient in this study was intensely followed for a 2-year period (2007-2009) and several FEV1 measurements were performed as indication by the red dots on the top. At the beginning and after 2 years a high-resolution CT scan was taken and Functional Respiratory Images (FRI) measurements were performed. FRI provides, in contrast to the FEV1, regional information about airway structure and function.
We can see from the animation that the FEV1 remained quite stable, around 80% predicted, over the 2 year period. However when we look at the airway and lung lobe geometry we can see drastic changes. After two years we observed a near full collapse of the left upper lobe (the red zone on the right hand side of the screen) and a significant alteration of the airways leading to that lobe. So while the FEV1 suggested that the patient’s respiratory system was stable, internal changes were occurring that were not being detected by the “stress test”. It appears that due to this patient’s biological redundancy, lung function was preserved at the expense of the healthy reserve.
This case clearly demonstrates that we must be careful relying only on derivative measures such as FEV1 to assess the health of an organ, in this case the lung. Whenever possible we should consider using methods that provide a direct view so we can establish how much of the anatomical structure is still healthy and how much of the “reserves” have been used. It is an illusion to think that one parameter will tell the whole story yet the majority of clinical trials still use the primary endpoint approach. We, as a field, must move towards an environment where we weigh the evidence of multiple parameters. And use the composite endpoint to determine the best treatment for an individual patient. By doing so we can mitigate the dangers of our biological redundancy.