fMRI/PET have shed new light in our understanding of consciousness and have shown that the vegetative brain is more complex than previously envisioned [46–69]. The activation of cerebral structures to complex language stimuli or to mental imagery suggests that the VS brain may retain some capacity for higher levels of neuro-functioning [48–52, 61].

fMRI and PET improve the diagnostic accuracy of VS and help in differentiating VS from MCS [46–53, 68]. Hannawi et al. [47] reported reduced activity in the cingulate gyrus in all VS and MCS subjects and noted that in VS, the reduction affected bilateral medial dorsal thalamic nuclei, while in MCS, the reduction was limited to the right medial dorsal nucleus, thus separating the two states [47]. fMRI/PET may also be helpful in the prognostication of recovery [18, 49, 56, 57]. The presence of activation of primary sensory cortices, higher-order associative areas, and subcortical areas to either mental imagery tasks or high-level language stimuli is often associated with some recovery.

Whether fMRI/PET can establish consciousness and evaluate if the patient feels pain remain controversial. Neuroimaging responses to complex language stimuli or to mental imagery in some subset of VS patients are similar to the response of normal controls [13, 50–52, 56–58, 64]. Does this correlation indicate “conscious awareness”? Similarly, neuroimaging activation of cerebral regions to painful stimuli implies pain perception [54, 66–69].

There is a consensus that the utilization of complex language stimuli activates in some VS/MCS patients’ brain regions is associated with language processing (bilateral superior temporal planes, superior temporal gyri, inferior frontal gyri, and left posterior inferior temporal cortex) [50, 55–59, 65]. Similarly, the use of mental imagery paradigms in some VS patients activates brain structures that are similar to the activation in normal controls [51–53]. There are two possible interpretations of these data. Proponents of “functional neuroimaging” believe that brain activation demonstrates “neuronal correlates of speech comprehension” [56], a state of consciousness [50] and a retained capacity for “covert cognition.” They conclude that neuroimaging activation is a signature of some conscious processing [13, 21, 48–52, 56–60, 65, 69, 71, 72]. The alternative interpretation is based on the demonstration of reduced and disrupted brain activation and on clinical observations. Hannawi et al. [47] in a meta-analysis of 13 studies (272 patients, 259 healthy controls) noted “…consistently reduced activity in patients with DoC in bilateral medial dorsal nucleus of the thalamus, left cingulate, posterior cingulate, precuneus, and middle frontal and medial temporal gyri”. Stender [70] showed a decreased of global cerebral metabolism (indicating decreased neuronal activity) in VS patients compared to normal controls (Fig. 13.2). Many studies have shown that brain connectivity is disrupted in VS/MCS [46, 50, 64, 68, 79]. Dehaene and Changeux [73] suggest that conscious awareness occurs “when incoming information is made globally available to multiple brain systems through a network of neurons with long-range axons densely distributed in prefrontal, parieto-temporal, and cingulate cortices.” It is therefore conceivable that fMRI activation to language and imagery in some VS subjects simply reflects the preservation of some modules, while these modules have disrupted connectivity and lack the integrative network processes necessary for normal cognitive functions [55, 74].

Furthermore, 46/121 (38 %) patients in MCS who by definition are supposed to have some minimal consciousness showed no response or limited activation on fMRI, suggesting absence of any awareness [55]. This dissociation between behavioral observation and neuroimaging should set a cautionary note on any of our interpretations. The other issue closely correlated to conscious awareness is the issue of pain perception. Are behavioral responses to pain observed in VS patients (crying, withdrawing from the stimulus, and changes in heart rate) an expression of emotional primordial reflexes or indication of some awareness? Can the information gathered by neuroimaging answer this question? 60 % of VS and 19 % of MCS showed none or minimal cerebral activation to nociceptive stimuli. The remainder of VS/MCS patients showed fMRI/PET complex cerebral activation to nociceptive stimuli, demonstrating some neuronal processing of pain [66–69]. Whether “…are images of increased blood oxygen level–dependent signal (fMRI) or oxygen/glucose uptake (PET) equivalent to conscious perception, or indicative of awareness, or are they simply document-retained modular function in the absence of the integrative processes necessary for consciousness…” remains an open question. VS subjects are unable to communicate their feelings and wishes and to manifest volition. Present technology no matter how sophisticated cannot read or interpret a person’s mind. Are we willing to accept neuroimaging tests as surrogates for communication among us and VS subjects? Controversial interpretation of data does not offer sufficient evidence to proxy a machine as a “sentient communicator”; thus, the issue of residual consciousness and pain perception in VS remains an unresolved Gordian knot [54, 55, 74]. The lay press has become concerned by some of our claims and editorializes against “people who take pretty brain-scan images and claim they can use them to predict what product somebody will buy, what party they will vote for, whether they are lying or whether a criminal should be held responsible for his crime” [75] or what they think [76]. David Brooks of the New York Times said:“…an important task these days is to harvest the exciting gains made by science and data while understanding the limits of science and data. The next time somebody tells you what a brain scan says, be a little skeptical…” [75].

The care of VS/MCS is affecting their family [96–98]. The distress and burden are “high for all caregivers” [96]. Leonardi et al. [97] suggest that there is a need for comprehensive support strategies for caregivers to “diminish the level of burden.” Caregivers often report a loss of income estimated to be “less than 17,000 euros per year” [97].

The health-care cost of caring for VS/MCS patients needs to be evaluated particularly in the present era of limited resources. Former Colorado Gov. Richard Lamm back in 1987 framed the issue succinctly: “We refuse to even discuss the reality that in an aging society with exploding numbers of elderly and exponentially rising health-care costs and myriad other unmet needs, the level of treatment for a 90-year-old might be different from that for a 9-year-old. Should we not put our limited dollars into more quality of a healthy life rather than more quantity of a sick life–more to the recoverable young rather than the terminally ill?” [99]. The cost of care of a woman that survived for 17 years in VS was estimated in 1981 at $6,104,590 ($ 61,028 per year) [100]. Kaufman and Lipton [101] obtained in 1992 the hospital bill of 13 patients in VS. Their mean hospital length of stay was 197 days and the averaged cost was $170,000. According to the US government data, the average costs for long-term care in 2010 were $205 per day or $74,820 per year [102]. The prevalence of VS in adults in the United States is estimated between 4 and 16.8/100,000 [25]. The US population in 2014 was 318,857,056; then the estimated number of VS/MCS patients is between 12,754 and 51,017 with a cost to the health care between one to four billion dollars/year. The MSTF [3, 32] similarly estimated the total annual cost for caring of VS patients in the United States to be between $1 and $7 billion. This is a huge burden to the health budget that cannot be overlooked. It may be time to set aside preconceived ideas and look pragmatically to the issues of VS in the contest of all others health needs and ask ourselves if the cost/benefit ratio is too much into the negative.