Are there subgroups of chronic fatigue syndrome? An exploratory cluster analysis of biological markers

Table 3 Final cluster solution–demographic, genetic and CFS criteria adherence differences

Cluster	1: Pain tolerant & good cognitions (n = 40)	2: Restored HPA dynamics (n = 15)	3: Orthostatic intolerance (n = 20)	4: Low-grade inflammation (n = 4)	5: Pain intolerant & poor cognitions (n = 33)	6: High vagal activity (n = 3)	p-value^c
Sex-no. of males (%)	7 (17.5)	4 (26.7)	7 (35.0)	1 (25.0)	10 (30.3)	1 (33.3)	0.639
Age, years-mean [95% CI]	15.7 [15.3, 16.3]	15.7 [14.9, 16.5]	15.0 [14.2, 15.7]	16.5 [15.1, 18.0]	14.8 [14.3, 15.4]	15.4 [11.2, 19.6]	0.074
BMI, kg/m²-mean [95% CI]	22.3 [21.0, 23.1]	23.2[20.4, 26.1]	20.4 [17.9, 22.6]	22.8 [11.6, 33.9]	20.7 [19.4, 22.0]	18.3 [11.4, 25.1]	0.128
Disease duration, months-mean [95% CI]	19.4 [15.9, 22.9]	25.1 [10.8, 39.5]	22.0 [16.0, 27.9]	28.0 [n.a.]	19.3 [14.8, 23.9]	30.3 [n.a.]	0.538
Fukuda-no. fulfilling criteria (%)^a	27 (71.1)	9 (60.0)	16 (80.0)	4 (100.0)	25 (78.1)	3 (100.0)	0.564
Canada-no. fulfilling criteria (%)^b	13 (35.1)	3 (20.0)	6 (30.0)	1 (25.0)	22 (68.8)	1 (33.3)	0.006
COMT [SNP rs4680]-no. (%)
AA	10 (25.0)	4 (26.7)	8 (40.0)	1 (25.0)	13 (39.4)	0 (0)	0.616
AG/GG	30 (75.0)	11 (73.3)	12 (60.0)	3 (75.0)	20 (60.6)	3 (100.0)	0.616
SLC6A4 [5-HTTLPR allele & SNP rs25531]-no. (%)
SS/SLG	11 (27.5)	3 (20.0)	2 (10.0)	3 (75.0)	9 (27.3)	1 (66.7)	0.139
SA/LALG/LALA	29 (27.5)	20 (80.0)	18 (90.0)	1 (25.0)	24 (72.7)	2 (33.3)	0.139

HPA: hypothalamus–pituitary–adrenal; BMI: Body mass index; SD: Standard deviation; ADRA2A: alpha-2A adrenergic receptor gene; COMT: Catechol-O-methyltransferase gene; ADRB2: beta-2 adrenergic receptor gene; SLC6A4: serotonin transporter (5-HTT) gene; C: Cytosine; G: Guanine; A: Adenosine; LPR: long tandem repeats; S: short allele; L: long allele
^aNumber of patients fulfilling the CDC Fukuda 1994 diagnostic criteria for Chronic Fatigue Syndrome [Fukuda 1994]
^bNumber of patients fulfilling the Canada 2003 diagnostic criteria for Chronic Fatigue Syndrome [Carruthers 2003]
^cUnadjusted p-values. The p-values are based on the Fischer exact test or one-way ANOVA, as appropriate. Only group 1–3 and 5 were used in the statistical analyses of continuous variables due to few participants in group 4 and 6

ISSN: 1479-5876