Oral vitamin B12 for patients suspected of subtle cobalamin deficiency: a multicentre pragmatic randomised controlled trial

Table 3 Different outcomes in the oral B12 treatment group and the placebo group after one and four months

	Oral B12		Placebo		Impovement from baseline between oral B12 and placebo*
Biological marker	1 month Mean (SD); N	4 months Mean (SD); N	1 month Mean (SD); N	4 months Mean (SD); N	1 month* Δ (CI95%; p-value)	4 months* Δ (CI95%; p-value)
MMA (μmol/l)	0.23 (0.08); 22	0.41 (0.30); 19	0.37 (0.14); 23	0.35 (0.16); 22	- 0.13 (CI95% -0.19 to -0.06; p < 0.001)	0.03 (CI95% -0.12 to 0.17; p = 0.686)
Serum cobalamin (pM/l)	263.4 (89.8); 25	202.6 (56.3); 23	154.5 (41.1); 24	162.9 (39.8); 23	101.6 (CI95% 60.1 to 143.2; p < 0.001)	35.0 (CI95% 6.4 to 63.5; p = 0.018)
Hcys (μmol/l)	16.5 (6.1); 22	17.1 (7.5); 19	13.9 (4.3); 23	15.6 (5.8); 22	0.04 (CI95% -1.2 to 1.3; p = 0.950)	-1.0 (CI95% -4.0 to 2.0; p = 0.502)
Hematocrite (% red cells)	39.6 (4.1); 26	40.1 (4.0); 26	39.7 (4.6); 24	39.4 (4.6); 22	-0.4 (CI95% -1.7 to 0.8; p = 0.502)	0.5 (CI95% -1.0 to 2.1; p = 0.475)
MCV (fl)	89.8 (6.9); 26	89.0 (7.0); 26	92.8 (7.0); 24	92.6 (7.6); 22	-0.4 (CI95% -2.2 to 1.4; p = 0.674)	-0.1 (CI95% -2.3 to 2.2; p = 0.950)
MMSE (score 0-30)	-	27.8 (2.3); 26	-	28.1 (2.2); 21	-	-0.4 (CI95% -1.3 to 0.6; p = 0.432)

* Least square means were computed using linear regression with robust standard error to take heteroscedasticity into account. Each value at 1 or 4 months was introduced separately as the dependent variable, with group allocation and baseline value as independent variables. Interpretation of these magnitudes is nevertheless restricted. Variance is not constant across baseline values (heteroscedasticity). Effects are more important for patients with initial higher levels of deficiency. Hcys = homocysteine, MMA = methylmalonic acid, MMSE= Mini Mental State Examination.

ISSN: 2731-4553