Table 2.
Resting baseline physiological levelsa
| bvFTD | HCs | F | p | ηp2 | |
|---|---|---|---|---|---|
| Interbeat interval (ms) | 906.6 (156.74) | 981.9 (133.9) | 0.3 | 0.57 | 0.01 |
| Respiratory sinus arrhythmia (ms)* | 39.1 (22.9) | 62.9 (55.3) | 10.0 | 0.003 | 0.20 |
| SCL (μmhos)* | 2.0 (0.9) | 2.9 (1.7) | 4.6 | 0.037 | 0.10 |
| Respiration period (ms)* | 4035.6 (915.5) | 4750.3 (946.6) | 9.4 | 0.004 | 0.19 |
| Finger temperature (°F) | 80.7 (5.5) | 80.5 (5.6) | 0.0 | 0.97 | 0.00 |
aData are mean (SD), unless otherwise noted. Analyses of covariance (controlling for age, sex, and education) found that patients with bvFTD (n = 22) had significantly lower respiratory sinus arrhythmia, lower SCL, and shorter respiration period than the HCs (n = 22). When we added additional covariates that might influence RSA (i.e., resting interbeat interval, resting respiration period, and body mass index) to the model (while also controlling for education, sex, and age), patients continued to have lower RSA (F(1,33) = 469, p = 0.038, ηp2 = 0.13). For SCL, although including resting interbeat interval, resting respiration period, and body mass index as additional covariates caused the difference to fall to trend levels, the pattern of results was in the expected direction, with patients having lower SCL than controls (F(1,32) = 2.82, p < 0.11, ηp2 = 0.08). To examine whether the results of the original model (controlling for age, sex, and education) held in a smaller, age-matched sample, we removed patients under age 58 (n = 6) and HCs over age 70 (n = 8). In this subset of participants (17 bvFTD and 15 HCs), the patients and controls did not differ in age (F(1,31) = 3.2, p = 0.09, ηp2 = 0.09). In this smaller sample, the group differences in baseline autonomic activity remained significant. Compared with HCs, patients with bvFTD had lower RSA (F(1,27) = 5.6, p = 0.03, ηp2 = 0.17), lower SCL (F(1,27) = 4.2, p = 0.05, ηp2 = 0.14), and shorter respiration period (F(1,27) = 8.1, p = 0.008, ηp2 = 0.23). Partial eta squared (ηp2) is provided as a measure of effect size.
*p < 0.05, significant difference in the first model described above.