## Table 3 |
||||||

Median spindle frequency distribution across AHI groups, scalp locations and night
sections |
||||||

Non-OSA (7) |
Mild OSA (11) |
Mod OSA (10) |
KW H |
df |
||

Location | Night Period | |||||

Frontal | total^{∗∗∗} |
12.6 (1.6) | 12.4 (1.5) | 12.1 (1.3)^{c,f} |
51.704 | 2 |

I^{∗} |
12.4 (1.6) | 12.3 (1.6) | 12.1 (1.3)^{a,d} |
8.135 | 2 | |

II^{∗∗} |
12.3 (1.5) | 12.3 (1.4) | 12.1 (1.3)^{b} |
12.339 | 2 | |

III^{∗∗∗} |
12.8 (1.4) | 12.5 (1.6)^{c} |
12.0 (1.2)^{c,f} |
48.252 | 2 | |

Central | total^{∗∗∗} |
13.5 (1.4) | 13.1 (1.7)^{c} |
12.5 (2.1)^{c,f} |
180.116 | 2 |

I^{∗∗∗} |
13.3 (1.9) | 13.0 (2.2) | 12.5 (1.9)^{c,f} |
51.716 | 2 | |

II^{∗∗∗} |
13.6 (1.2) | 13.3 (1.8)^{a} |
12.7 (2.1)^{c,f} |
46.849 | 2 | |

III^{∗∗∗} |
13.5 (1.3) | 13.1 (1.4)^{c} |
12.3 (2.3)^{c,f} |
85.931 | 2 | |

Parietal | total^{∗∗∗} |
13.7 (0.9) | 13.6 (1.6) | 12.9 (2.3)^{c,f} |
194.36 | 2 |

I^{∗∗∗} |
13.6 (1.0) | 13.7 (2.0) | 12.6 (2.1)^{c,f} |
66.639 | 2 | |

II^{∗∗∗} |
13.7 (0.9) | 13.8 (1.5) | 13.3 (1.8)^{c,f} |
82.901 | 2 | |

III^{∗∗∗} |
13.8 (0.9) | 13.4 (1.4)^{c} |
12.5 (2.4)^{c,f} |
73.002 | 2 |

AHI, Apnea-Hypopnea Index; Mod, moderate OSA; Location, scalp EEG topography; total,
(I+II+III) sample; I, II and III, initial, middle and final night sections, respectively.
Data presented as median (interquartile range). KW H, Kruskal-Wallis statistic; df,
degrees of freedom. P-value significance as follows: ^{∗}<0.05, ^{∗∗}<0.01, ^{∗∗∗}<0.001. Dunn’s post-hoc multiple comparisons tests as follows: *a*<0.05, *b*<0.01, *c*<0.001 for contiguous comparisons; *d*<0.05, *e*<0.01, *f*<0.001 for non-contiguous comparisons. Compared to non-OSA subjects, moderate OSA
patients had significantly lower median spindle frequency in all locations and night
sections under study. It should be noticed that frequency distributions were, in the
large majority, non-unimodal, indicating that variations in frequency medians were
only partially informative, and resulted from variability in the proportions of at
least two spindle populations.

Schönwald * et al.*

Schönwald * et al.* *BMC Neuroscience* 2012 **13**:89 doi:10.1186/1471-2202-13-89