Ketogenic diets (KDs) were initially introduced to clinical practices as alimentary approaches with the aim to control drug-resistant epilepsies. Over the decades, a large and growing body of research has addressed the antiseizure effect of various KDs, and worked out KD-based dietary regimens, including their acting factors and modes of action. KDs have also appeared in weight loss therapies. Therapy control, particularly at initiation, happens through regular blood analysis and control of urine ketone levels. However, there is a lack of fast, reliable, and preferably non-invasive methods to accomplish this. The detection of exhaled breath constituents may offer a solution. The exhaled breath contains hundreds of volatile organic compounds (VOCs), which can be modified by diet. VOC detection technology has resulted in low-cost sensors that can facilitate the self-monitoring of patients in the future if reliable breath markers are available. Therefore, it is of interest to investigate the composition of exhaled breath in children on KDs. Twenty-two pediatric patients between 4 and 18 years of age were recruited in this study. Eleven of them received a KD and suffered from epilepsy, with the exception of one child, who was admitted to a weight-reduction therapy. The control group involved 11 patients with neurological disorders but not on KD. Breath volatiles were analyzed using gas chromatography mass spectrometry (GC-MS) after preconcentration of the analytes on needle traps (NTs). We found that the breath concentrations of a number of VOCs, namely acetaldehyde, acetone, 2-methylfuran, methyl-vinyl-ketone, and 2-pentanone were significantly elevated in the breath of children on a KD in comparison to their control counterparts. Interestingly, breath ethanol was lower in patients on a KD than in non-KD patients. Association studies revealed an interrelationship among (i) lipid parameters and ketone bodies, (ii) methacrolein, methyl-vinyl-ketone, and high-density lipoprotein, as well as (iii) methyl-vinyl-ketone, acetone, and 2-pentanone, thus raising the possibility of a common metabolic source. The duration of diet was positively and negatively associated with breath acetone and breath ethanol, respectively. Some of the changes were linked to β-oxidation, but there are uncertainties in regard to metabolic sources of other metabolites. Lipid peroxidation and alteration of intestinal microbial composition may also be involved in the changes of VOC profiles during KD. Since lipids used for metabolism during KD originate from external sources, the processes occurring cannot simply be compared to and deduced from changes appearing in starvation; however, lipid mobilization is also evident in starvation. To find reliable and sensitive VOC markers that are linked to the respective ketogenic regimen, further investigations are needed to reveal the metabolic background.