Table 4. Clinical and Adverse Events in a Study of the Association Between Levothyroxine Treatment and Thyroid-Related Symptoms Among Adults Aged 80 Years and Older With Subclinical Hypothyroidisma.
| Outcomes | No. (%) | Event Rate per 100 Person-Years | Estimated Risk Difference (95% CI)b | ||
|---|---|---|---|---|---|
| Levothyroxine (n = 112) | Placebo (n = 139) | Levothyroxine (n = 112) | Placebo (n = 139) | ||
| Prespecified secondary outcomes | |||||
| Clinical outcomes | |||||
| Fatal or nonfatal cardiovascular event | 7 (6.3) | 14 (10.1) | 4.52 | 7.64 | Hazard ratio, 0.61 (0.24 to 1.50) |
| Death from any cause | 5 (4.5) | 4 (2.9) | 2.99 | 2.02 | Hazard ratio, 1.39 (0.37 to 5.19) |
| Adverse events | |||||
| Cardiovascular deathc | 0 | 1 (0.7) | 0.00 | 0.51 | |
| Serious adverse events | |||||
| Events | 53 | 61 | |||
| Participants with >1 serious adverse eventd | 33 (29.5) | 40 (28.8) | −0.01 (−0.04 to 0.01) | ||
| Adverse event of special interest | |||||
| New-onset atrial fibrillation | 4 (3.6) | 6 (4.3) | 2.57 | 3.19 | 0.00 (−0.02 to 0.03) |
| Heart failure | 3 (2.7) | 6 (4.3) | 1.90 | 3.21 | 0.01 (−0.03 to 0.05) |
| Fracture | 4 (3.6) | 5 (3.6) | 2.53 | 2.68 | 0.00 (−0.04 to 0.03) |
| Hypothyroidisme | 0 (0.0) | 0 (0.0) | |||
| Withdrawal | |||||
| Permanent discontinuation of trial regimenf | 38 (33.9) | 43 (30.9) | 27.34 | 24.84 | −0.04 (−0.15 to 0.05) |
| Withdrawal from follow-upg | 10 (8.9) | 9 (6.5) | 5.98 | 4.56 | −0.03 (−0.09 to 0.02) |
| ThyPRO hyperthyroid symptoms scoresh | 10.9 (11.3) | 9.1 (10.8) | 9.6 (9.4) | 9.3 (10.4) | Adjusted difference, −0.50 (−2.62 to 1.63)e |
Adverse events were recorded and reported until the end of the study. Preplanned secondary outcomes effects were estimated using Cox proportional hazard regression models adjusted for sex, dose at randomization, study site and study. Adverse events event rates were estimated using Cox proportional hazard regression models with, where possible, adjustment for study site, study, dose at randomization, sex, and age, presented as risk differences and 95% CIs, obtained through bootstrap resampling in 1000 iterations.
Unless otherwise noted.
For outcomes with too few events to run regression models, event rates and a log-rank test P value were reported. Serious adverse events were all undesired medical events involving a participant, which are not necessarily associated with the treatment, that are fatal, threaten the life of the participant, make hospital admission or an extension of the admission necessary, cause persistent or significant invalidity or work disability, manifest themselves in a congenital abnormality or malformation, or could, according to the researchers, have developed to a serious undesired medical event but were prevented because of premature interference.
Analysis adjusted for study site, sex, dose at randomization, and age.
Defined as thyrotropin level of 20 mIU/L or higher during trial laboratory measurements.
Analysis adjusted for study site, sex, and dose at randomization.
Adjusted difference was estimated in linear regression models predicting change from baseline to 12-month visit (95% CI) with study site, sex, and randomization dose as stratification variables and study as random effect.
For scale definitions see Table 1 footnotes.