Longitudinal serum total cholesterol patterns and subsequent clinical relapse in ulcerative colitis

Table 1 Baseline characteristics of patients according to clinical relapse, n (%)/median (interquartile range)

Characteristics	All patients (n = 452)	Non-relapse (n = 253)	Relapse (n = 199)	HR (95%CI)	P value
Age (years)	49 (37-58)	49 (38-60)	48 (37-56)	0.994 (0.984-1.005)	0.177
Male	236 (52.2)	133 (52.6)	103 (51.8)	0.960 (0.727-1.268)	0.773
Age at diagnosis (years)	39 (30-49)	39 (30-49)	38 (28-48)	0.994 (0.984-1.005)	0.283
Disease duration (months)	80.0 (33.5-149.0)	78.0 (33.0-152.0)	84.0 (34.0-144.0)	0.999 (0.998-1.001)	0.522
Disease extent
Proctitis	182 (40.3)	109 (43.1)	73 (36.7)	Ref	0.331
Left-sided colitis	130 (28.8)	67 (26.5)	63 (31.7)	1.285 (0.874-1.890)	0.2901
Extensive colitis	140 (31.0)	77 (30.4)	63 (31.7)	1.176 (0.800-1.729)	0.6931
BMI (kg/m2)	22.94 (20.94-25.08)	23.11 (20.93-24.92)	22.83 (20.96-25.28)	1.009 (0.965-1.054)	0.700
Current medication
5-ASA, topical	86 (19.0)	51 (20.2)	35 (17.6)	0.935 (0.649-1.347)	0.719
5-ASA, oral	186 (41.2)	92 (36.4)	94 (47.2)	1.363 (1.032-1.801)	0.029
5-ASA, both	156 (34.5)	97 (38.3)	59 (29.6)	0.720 (0.531-0.976)	0.034
Steroid	6 (1.3)	2 (0.8)	4 (2.0)	2.312 (0.858-6.225)	0.097
Azathioprine	28 (6.2)	12 (4.7)	16 (8.0)	1.451 (0.870-2.419)	0.154
Biologic therapy	36 (8.0)	19 (7.5)	17 (8.5)	1.088 (0.662-1.789)	0.740
Statin use history	57 (12.6)	31 (12.3)	26 (13.1)	1.023 (0.677-1.545)	0.915
Laboratory findings
WBC (/μL)	6130 (5155-7320)	6080 (5130-7280)	6160 (5230-7450)	1.003 (0.996-1.011)2	0.409
Hemoglobin (g/dL)	14.1 (13.0-15.2)	14.0 (13.1-15.2)	14.1 (13.0-15.0)	1.001 (0.999-1.002)	0.315
Platelet count (× 103/μL)	247.0 (214.0-291.5)	242 (204.0-277.0)	263.0 (225.0-302.0)	1.005 (1.003-1.007)	< 0.001
ESR (mm/hour)	11.0 (5.0-20.0)	11.0 (5.0-20.0)	12.0 (7.0-20.0)	1.009 (1.000-1.019)	0.060
Albumin (g/dL)	4.6 (4.4-4.8)	4.6 (4.4-4.8)	4.6 (4.4-4.8)	0.847 (0.623-1.152)	0.290
CRP (mg/dL)	0.05 (0.03-0.11)	0.04 (0.03-0.11)	0.05 (0.03-0.11)	1.513 (0.949-2.412)	0.082
Baseline TC (mmol/L)	4.64 (4.09-5.20)	4.60 (4.06-5.15)	4.71 (4.14-5.20)	1.123 (0.962-1.361)	0.163
PRO-2
PRO-2 remission	306 (67.7)	179 (70.8)	127 (63.8)	Ref
No PRO-2 remission	146 (32.3)	74 (29.2)	72 (36.2)	1.337 (1.001-1.786)	0.049
Endoscopic finding
MES 0	205 (45.4)	133 (52.6)	72 (36.2)	Ref
MES 1	247 (54.6)	120 (47.4)	127 (63.8)	1.684 (1.261–2.249)	< 0.001
Histologic finding
Geboes score < 3.1	316 (69.9)	191 (75.5)	125 (62.8)	Ref
Geboes score ≥ 3.1	136 (30.1)	62 (24.5)	74 (37.2)	1.687 (1.265-2.249)	< 0.001
Nancy index < 2	315 (69.7)	190 (75.1)	125 (62.8)	Ref
Nancy index ≥ 2	137 (30.3)	63 (24.9)	74 (37.2)	1.659 (1.244-2.212)	< 0.001

¹P values indicate pairwise comparisons using the Bonferroni correction.

²For white blood cell count, Cox proportional hazards regression was performed using values scaled per 100/μL increase.

Categorical variables are expressed as n (%). Hazard ratios and P values were estimated using Cox proportional hazards regression analysis. 5-ASA: 5-aminosalicylic acid; BMI: Body mass index; WBC: White blood cell count; ESR: Erythrocyte sedimentation rate; CRP: C-reactive protein; PRO-2: 2-item patient-reported outcome; MES: Mayo endoscopic subscore; HR: Hazard ratio; Ref: Reference; TC: Total cholesterol.

Table 2 Subsequent relapse risk according to maximal serum total cholesterol levels at 1 year, 2 years, and 3 years after endoscopic improvement

Landmark time	Cutoff group	Patients		Univariable analysis		Multivariable analysis
		Non-relapse	Relapse	HR (95%CI)	P value	Adjusted HR (95%CI)	P value
1-year	< 4.53 mmol/L	82	29	Ref		Ref
	≥ 4.53 mmol/L	166	121	1.73 (1.15-2.60)	0.008	1.85 (1.11-3.07)	0.018
2-year	< 4.53 mmol/L	58	14	Ref		Ref
	≥ 4.53 mmol/L	181	90	1.88 (1.07-3.30)	0.028	2.05 (1.05-4.01)	0.036
3-year	< 5.56 mmol/L	148	28	Ref		Ref
	≥ 5.56 mmol/L	87	32	1.92 (1.16-3.18)	0.012	2.16 (1.11-4.22)	0.024

Hazard ratios and 95%CIs were estimated using Cox proportional hazards models for landmark analysis. The analyses were restricted to patients who were relapse-free at each landmark time point. Multivariable models were adjusted for age, sex, baseline serum total cholesterol, infliximab use, statin use, and Mayo endoscopic subscore 0/1. HR: Hazard ratio; Ref: Reference.