Evidence was obtained suggesting that 6 g/day of adjunctive L-lysine treatment is a sufficient dose for increasing blood L-lysine levels above the nutritional, naturally occurring levels, without inducing adverse side-effects. In addition, there was a significant decrease in the positive PANSS scores and WCST over the whole study period. These two measures represent different functional abilities in the individual; PANSS is used to measure psychosis severity and WCST cognitive functionality. Thus, evidence was obtained for symptom improvement in both of these domains; however, accounts for placebo effects, such as training on WCST, cannot be excluded.
Given the limited number of patients enrolled in the study, difficulties arise as to the possibilities of making reliable conclusions about the efficacy of L-lysine treatment. However, it should be noted that despite the low number of participants statistically significant effects of treatment on outcome measures were in fact obtained. Furthermore, albeit the relatively small changes in PANSS and WCST scores observed, statistical analysis showed fairly high effect sizes and observed powers for these changes (positive PANSS (n = 10): effect size (η) = 0.76, observed power (1-β) = 0.99; WCST number of errors (n = 10): effect size (η) = 0.64, observed power (1-β) = 0.77) indicating that they were of significant magnitude and detectable also in small sample sizes. Needless to say, it is also possible that a higher dose of L-lysine or a longer treatment period may have provided more clear-cut results. The reason that two participants (one male and one female) did not get an elevation in their blood concentration of L-lysine is unknown. In order to assess adherence to the L-lysine treatment the L-lysine containers were carefully labelled with dates and empty containers were collected weekly. However, there was no control for actual intake of the study medication beyond the assessed blood L-lysine concentrations. Metabolic individual differences, due to i.a. intense physical activity or gender, or simply non-adherence to L-lysine treatment, cannot be ruled out as one of the contributing factors to these findings.
Cautions should be taken when interpreting the data. Both the PANSS and the WCST scores seemed to improve over repeated testing, and consequently this placebo and/or training effect could explain the effects obtained. The study design presently used did not provide a satisfactory control for such effects, but the statistical measures, that is, analysis of interaction between treatment order and test session on outcome measure, that were taken to address this issue confirmed the possibility of placebo and training effects. Indeed, the WCST has been shown to be sensitive to training effects (see, for example, ) and should be used with care. Ideally, a test less sensitive to training and with higher test-retest reliability may be advisable for future studies, as well as a pre-testing cognitive training session (before the baseline assessment) in order to control for training artefacts.
The fact that the patients enrolled in the study were in a stable phase of their illness and were all treated with atypical antipsychotics may also have made it more difficult to detect beneficial effects of L-lysine treatment on cognitive functioning and symptom severity. In this respect, special attention may be paid to the self-reported improvements of positive symptoms, attention and memory capacity by three of the patients, since these potential effects of the treatment would be important but unfortunately could not be addressed by statistical analysis. It should also be noted that the patient who experienced improved memory capacity continuously trained his memory using a computerized training program.
There is an ongoing debate in the literature as to how to disentangle training effects of repeated cognitive testing, from "real" drug effects, as well as how to establish consensus regarding a reliable cognitive test battery [36–38]. The cognitive tests chosen for the purpose of the present study are some of the most frequently applied in research and several of them have been shown to be resistant to training effects [39, 40]. The initiatives taken to pharmacologically enhance cognition are plentiful; however, the results of such studies are diverse and few studies have been able to demonstrate replicable cognitive enhancement in patients with schizophrenia . One contributing factor to these shortcomings, may be that schizophrenia encompasses abnormal progressive loss in brain volume , possibly counteracting any cognitive enhancing potential. However, cognitive enhancing therapy was recently demonstrated to counteract gray matter loss in patients with schizophrenia .
Cognition is a highly specific and complex phenomenon that develops from early life into adulthood, depending on both biological pre-disposition as well as environmental input . Perhaps the lack of development of cognitive enhancing agents is due to an overly optimistic belief in cognitive enhancement being viable without concurrent cognitive training. The plasticity of the brain is a complex, time- and stimulus-dependent process by which the strengthening of synapses has to be modulated in order for learning to occur . Thus, it may in fact not be possible to find improvements in cognition by solely adding cognitive enhancing compounds without simultaneous cognitive training. Therefore, it is interesting that L-lysine treatment in combination with training was perceived to improve memory capacity in one patient. Moreover, adjunctive L-lysine treatment needs to be investigated in a larger placebo-controlled, double-blinded study further limiting possible training effects.
Taken together, preclinical and clinical studies have provided evidence for the involvement of a brain NO imbalance in schizophrenia. Furthermore, preliminary in vivo studies in rats have indicated that NO production is decreased by L-lysine . Consequently, L-lysine may have a beneficial potential in the treatment of the disorder. In addition, targeting the NO pathway may represent a new therapeutic approach that is pharmacologically fundamentally different from that of the traditional antipsychotics and, hence, may be beneficial in targeting symptoms that are currently treatment resistant.