It has recently been observed by many clinicians that children with LKS may respond to treatment with steroids or ACTH. This has greatly heightened interest in LKS by both parents and physicians alike.

Part of the problem is in defining the syndrome. To begin with one must consider the entire autistic spectrum disorder. Although not everyone would agree, Asperger's Syndrome (high functioning), Pervasive Developmental Disorder (severe impairment), and Landau-Kleffner Syndrome are in this broad category. If we strictly limit the definition of LKS to those children who loose language and develop autistic characteristics after a seizure or seizures, LKS would remain a rare entity. However, we soon came to note a pattern where a child whose developement was normal or nearly normal would, after some sort of insult (usually between ages 12 and 24 months), loose language and demonstrate autistic characteristics. The insult could be a seizure but could involve many major to minor illnesses or traumas. Thus epilepsy may be just one of many precipitants of our syndrome of interest.

The insults, in our experience, include events like: the chicken pox, an ear infection(s), a bad cold, falling from changing table to floor with no obvious serious injury, a planned (non-emergent) surgical procedure. One parent explained that their child was taken to a day care facility in the morning and was quite him/herself but when picked up in the evening never spoke again and became autistic. Nothing of note had happened during the day.

Other families report language and behavior change with NO obvious precipitant. In our opinion it is the sequence of normal development followed by communication and behavior change (loss of previously attained function) between ages 10 months and 36 months that constitutes the syndrome we all find interesting and we will continue to call LKS for the time being. Children who are very different as infants, who demonstrate autistic behaviors early on and never loose previously gained function are probably not in the LKS category even if they have communication problems and/or seizures. However, this may prove to be an incorrect formulation. By analogy to the picture of infantile spasms (with EEG hypsarrhythmia) where ACTH and steroids may reverse this suddenly appearing seizure disorder, these same medications have been shown in the LKS (loss of function) group to be effective in some children.

It is simple to understand that parents of child carrying a diagnosis of autism/PDD would want to know if the child has LKS as this may open theraputic possibilities not otherwise available. From the physician's standpoint the issue is how to identify which children in the autistic spectrum disorder are likely to respond to steroid or ACTH (or more recently anticonvulsant) therapy. Even within LKS not all children will respond to such treatments. Part of the problem may be that the full LKS picture may not be recognized until years after the time of functional loss, thereby delaying investigation for possible treatment.

From our prelilminary experience with what we refer to as the 4 Hz FMAER, those children with LKS (by our definition) fail to demonstrate a brain response at 4 Hz to the FMAER as Stefanatos suggested. Children with PDD who do not show a loss of function as described above typically have normal FMAER. We have also demonstrated that in patients where steroids and anticonvulsants appear to work, the 4 Hz response is restored.

It is too early to say for sure whether the FMAER is the best way to select patients as candidates for treatment BUT so far this may be the case! At CH we include the FMAER within our qEEG study when LKS is suspected.

QEEG evaluation forms a central part of our neurophysiological study of children with putative LKS. First 32 channels are gathered in waking and sleep to look for spikes and ESES. If spikes are noted, we can locate the source in dephth. If spikes are very focal and especially if they arise from the left temporal lobe, then surgical procedures can and have been helpful if medical treatment fails. Next we look for other signs of regional brain abnormality via EEG slowing as measured by spectral analysis. Finally the standard click AER is performed in addition to the FMAER. Preliminarily, we have found that when both the FMAER and click AER are quite abnormal, the chance of response to treatment is poorer than when the FMAER is abnormal but the click AER only mildly impaired. However all of this stems from our initial observations. None-the-less it seems that a full qEEG study should form an integral part of an LKS evaluation.