. Author manuscript; available in PMC: 2019 Aug 12.

Published in final edited form as: Neurol Clin. 2018 May;36(2):311–337. doi: 10.1016/j.ncl.2018.01.011

Table 2.

Methotrexate study sensitivity analysis

Outcome Measures	Multiple Imputation MTX Versus Placebo, Mean/ Median (SE)^a	P Value^b	Last Dose/Score Carried Forward MTX Versus Placebo, Mean/Median (SE)^a	P Value^b	Highest Dose/Worst Value Carried Forward for Worsening Cases MTX Versus Placebo Mean/Median (SE)^a	P Value^b
Primary outcome
Median 9-mo prednisone AUDTC, mg	2996.6 (727.1) vs 3484.7 (645.8)	.26	3330.0 (718.8) vs 3679.0 (748.0)	.27	3330.0 (718.8) vs 3679.0 (591.1)	.20
Median prednisone daily dose, mg/d	11.9 (2.9) vs 13.8 (2.6)	.26	13.2 (2.9) vs 14.6 (3.0)	.27	13.2 (2.9) vs 14.6 (2.3)	.20
Secondary outcomes
Mean 12-mo QMG change	−1.4 (0.7) vs 0.3 (1.0)	.29	−1.6 (0.7) vs 1.4 (0.9)	.01	−1.6 (0.7) vs 1.5 (0.9)	.01
Mean 12-mo MMT change	−5.5 (0.9) vs −3.3 (1.6)	.28	−5.7 (0.9) vs −3.0 (1.6)	.16	−5.7 (0.9) vs −2.6 (1.6)	.11
Median 12-mo MG-QOL change	−4.6 (4.5) vs −3.7 (4.8)	.82	−3.0 (2.0) vs −2.0 (1.5)	.18	−3.0 (1.9) vs −1(1.4)	.15
Mean 12-mo MG-ADL change	−1.2 (0.5) vs −0.3 (0.6)	.21	−1.2 (0.5) vs 0.48 (0.5)	.02	−1.2 (0.5) vs 0.5 (0.5)	.02
Mean 12-mo MGC change	−4.6 (0.9) vs −1.3 (1.1)	.09	−4.7(0.9) vs −1.1 (1.1)	.02	−4.7 (0.9) vs −0.9 (1.1)	.01

Abbreviations: AUDTC, area under the dose-time curve; MG-ADL, Myasthenia Gravis Activities of Daily Living Scale; MGC, Myasthenia Gravis Composite score; MG-QOL, Myasthenia Gravis Quality of Life Scale; MMT, manual muscle testing; MTX, methotrexate; QMG, Quantitative Myasthenia Gravis score; SE, standard error.

Mean (SE) were used as summary statistics if normality assumption was satisfied; otherwise, median (SE) were used as summary statistics (prednisone 9-mo AUDTC, prednisone daily dose, and MG-QOL). The SEs for medians were estimated by bootstrapping.

The 2-sample t test was used if normality assumption was satisfied; otherwise, the Wilcoxon rank-sum test was used. For the primary outcome, a significance level of .05 was used, and for the secondary outcomes, .01 was used to adjust for multiple comparisons.

Data from Dimachkie MM. Idiopathic inflammatory myopathies. J Neuroimmunol 2011;231(1-2):32–42 and Barohn RJ, Dimachkie MM. Tratamientos inmuno-moduladores. In: Mazia C, ed. Miastenia Gravis Y Problemas Relacionados. Buenos Aires: Inter-Médica; 2017:273–89.