TABLE 2.

BMI and inflammatory markers of mothers with very low milk production, moderate milk production, and controls¹

	Milk production groups2
	VL (n = 23)	MOD (n = 20)	Control (n = 18)	P value
BMI,3 kg/m2	39.6 [31.3–41.8]a	29.0 [25.3–31.2]b	26.1 [21.6–28.2]b	<0.001
Obesity, % BMI ≥ 30 kg/m2	78	40	22	0.003
Inflammatory markers
Serum hsCRP,4 % >5 mg/L	57	15	17	0.004
Serum TNF-α,5 pg/mL	6.3 [5.8–7.0]	6.1 [5.7–6.7]	5.8 [5.3–6.5]	0.31
Milk TNF-α,6 % detected	67	32	33	0.04
Milk TNF-α concentration of detected samples,7 pg/mL	1.60 [0.88–2.13]	0.98 [0.77–2.49]	1.11 [0.70–1.42]	N/A

¹Values are median [IQR] or percentages. Group differences were analyzed by nonparametric Kruskal–Wallis rank sum tests. Group median values in a row with different letters were significantly different (P < 0.05). Proportional differences were assessed by Pearson chi-square test. CRP, C-reactive protein; hsCRP, high-sensitivity CRP; MOD, moderate milk production group; VL, very low milk production group.

²Milk production groups were defined as VL (<300 mL/d), MOD (≥300 mL/d), and control (externally recruited controls who were exclusively breastfeeding infants with healthy weight gain).

³BMI measured at the baseline time point for VL and MOD groups.

⁴Serum samples were unavailable for 2 participants: VL, n = 21; MOD, n = 20; control, n = 18.

⁵One outlier of serum TNF-α > 20 pg/mL was removed for this analysis: VL, n = 23; MOD, n = 19; control, n = 18.

⁶Milk samples were unavailable for TNF-α analysis for 3 participants: VL, n = 21; MOD, n = 19; control, n = 18. The lower limit of detection was 0.5 pg/mL.

⁷Milk TNF-α concentrations are provided for the samples with a detectable value (>0.5 pg/mL): VL, n = 14; MOD, n = 6; control, n = 6.