A new study suggests that it may not be uncommon for patients who are in a minimally conscious state to be misdiagnosed as being in a persistent vegetative state. The study underscores the necessity of using standardized and objective diagnostic criteria in diagnosing coma. However, it also leaves some important questions unanswered.
As background it is essential to understand a bit about consciousness and coma, for not all comas are created equally. In order to be conscious a person requires at least one hemisphere of the brain be mostly functioning and they require a functioning brainstem. The cortical hemispheres contain the gray matter – that part of the brain that thinks. So it makes sense that a certain minimal amount of gray matter is necessary to generate consciousness. As gray matter is damaged or inhibited from functioning one’s level of consciousness decreases until it descends beyond that fuzzy boundary into unconsciousness. When such unconsciousness is persistent we call that coma.
But interestingly the cortex by itself cannot generate consciousness. It requires constant prodding by a diffuse region in the brainstem (that primitive part at the base of the brain that connects the brain and the spinal cord) called the brainstem activating system. This region sends a constant barrage of electrical signals through the thalamus (the relay center of the brain) and then onto the cortex. Without this constant stimulation the cortex will lapse into sleep and coma.
It is possible, therefore, that someone’s cortex, with all of their memories and personality, can be intact but they are permanently asleep because their brainstem is damaged. This is a very different kind of coma than that which results from diffuse cortical damage – although the end result may be very similar and of no consequence to the individual.
Therefore when considering coma, generally or in an individual, the exact type and severity of coma needs to be carefully defined before any meaningful statements or predictions can be made.
A minimally conscious state (MCS) is one in which a person has severe brain damage and is only able to produce subtle and intermittent signs of consciousness. This needs to be distinguished from a persistent vegetative state (PV), in which there are, by definition, no signs of consciousness. Although in a PV state patients may exhibit many automatic behaviors, like opening and closing the eyes, wandering gaze, and facial movements, that can be mistaken for conscious behavior. PV states carry a grim prognosis with a small chance of any meaningful recovery that diminishes further over time. Very rarely a patient in a PV state may improve into a MCS. A MCS also has a poor prognosis, but it is better than a PV state.
In addition to prognosis there are implications for management. Patients in a MCS may be able to perceive, if dimly, contact with their loved ones. They may also experience pain and therefore require pain medication.
In 2004 Giacino et al published the CRS-R – Coma Recovery Scale – Revised, which outlines a standardized procedure for diagnosing MCS. The procedure is fairly elaborate – it looks for the signs of consciousness that are in the definition of MCS, and requires that any findings are reproducible. One putative advantage also is that it takes into account that a patient in a MCS may have intermittent consciousness, and therefore repeated testing after attempting to arouse the patient is required.
Now Giacino and colleagues have published a study where they compare the CRS-R to a so-called clinical consensus diagnosis – where physicians, nurses, physical therapists, etc. base their diagnosis on the standard neurological exam for coma. What they found is that of 44 patients with a consensus diagnosis of PVS, 18 (41%) met criteria for MCS according to the more elaborate CRS-R exam.
This figure, 41%, is consistent with prior studies that show many patients with a diagnosis of PVS have signs of minimal consciousness if they are examined more thoroughly. While it is disappointing that the misdiagnosis rate remains so high, such results are not surprising. The diagnosis of PVS is partially based on the demonstration of a negative – the lack of any sign of even subtle and transient consciousness. Demonstrating a negative is always only as reliable as the thoroughness of the search, and it is not surprising that a more thorough search is therefore finding more patients with subtle signs of consciousness.
My one quibble with the way the study is currently presented (it is not yet in its final form) is that we are not told how many of the patients were diagnosed by neurologists vs intensive care medical doctors or physiatrists. I admit my bias as a neurologist that neurologists tend to do more thorough neurological exams. Perhaps this factor did not matter, but the information is not currently available.
More importantly, while the study informs us that 10% of the patients diagnosed with MCS by consensus criteria emerged from MCS during the study, we are not given any comparison between patients diagnosed with PVS and those diagnosed with MCS by the CRS-R criteria. If, as is likely, the CRS-R is finding the most subtle cases of MCS – those closest to being in a PVS, it would be very interesting to learn if there was any difference in the clinical outcome of these two groups.
The comparison being made is not to a highly reliable gold standard, but between two methods of clinical assessment, one more sensitive than the other (by definition, since it found more cases of MCS). However, is the CRS-R as specific, or is it finding false positives. Even if it is not finding false positives, what is the difference in clinical outcome between those in a PVS and those barely in a MCS? I am not arguing that there is no difference – just that the study as currently published does not give that information. I suspect that if there were a significant difference they would have reported it.
As technology advances the distinction in these forms of clinical diagnosis may make more, and less, difference. Less in that other forms of diagnosis may become available, such as using functional MRI scanning to assess cortical activity. Even now the clinical exam must be put into the context of brain imaging. But advances in fMRI and similar technology may free us from having to infer brain activity from examination alone, which has limits even when highly standardized.
Further, the distinction between PVS and MCS may become more meaningful as treatments for MCS advance. Already there are reported cases of using deep brain stimulation to increase the level of arousal in patients in a MCS. This is still a crude intervention, but has lead to meaningful clinical improvements. As brain-computer chip interfaces improve we will likely see the day when neural prosthetics could boost brain function sufficiently to wake the comatose.
In the final analysis I think this study makes a reasonable case for using more elaborate and standardized diagnostic procedures to more reliably diagnose PVS vs MCS. More evidence is needed, however, to assess the clinical relevance of this distinction.