Characteristics of the study population
Included participants and those lost to follow-up are shown in figure 1.
In total, 46 patients with PsA were included and started VLED treatment, whereof 41, 39 and 35 patients attended the six, 12- and 24-month visits respectively. The 11 patients (8 women and 3 men) lost to follow-up during the whole study were significantly younger than the patients who continued in the study (p=0.005). The demographics and the medication of the completers are shown in table 1. Treatment with cs/bDMARDs was unchanged compared to baseline in 89% (31/35) at the 12-month visit and in 74% (26/35) at the 24-month visit.
Table 1 Age, sex and medication of the patients with psoriatic arthritis (PsA) at baseline (BL) and the 12- months (M12) and 24-months (M24) visits.
|
BL
(N=39)
|
M12
(N =39)
|
M24
(N =35)
|
Women, n(%)
Men, n(%)
|
|
25 (64.1)
14 (35.9)
|
21 (60.0)
14 (40.0)
|
Age, years (IQR)
|
|
56 (49–63)
|
59 (51–56)
|
NSAIDs, n (%)
|
27 (69.2)
|
25 (64.1)
|
21 (60.0)
|
TNFi all, n (%)
TNFi in monotherapy
TNFi + csDMARD
|
15 (38.5)
4 (10.3)
11 (28.2)
|
16 (41.0)
4 (10.3)
12 (30.7)
|
14 (40.0)
3 (8.6)
11 (31.4)
|
Ustekinumab monotherapy, n (%)
|
1 (2.6)
|
0
|
0
|
Secukinumab monotherapy, n (%)
|
0
|
1 (2.6)
|
2 (5.7)
|
csDMARD without biologic, n (%)
Methotrexate
Sulfasalazine
Apremilast
Methotrexate + Sulfasalazine
|
17 (43.6)
11 (28.2)
2 (5.1)
1 (2.6)
3 (7.7)
|
16 (41.0)
11 (28.3)
1 (2.6)
1 (2.6)
3 (7.7)
|
14 (40.0)
10 (28.6)
1 (2.9)
2 (5.7)
1 (2.9)
|
Prednisolone, n (%)
|
3 (7.7)
|
2 (5.1)
|
2 (5.7)
|
Anti-hypertensives, n (%)
|
17 (43.6)
|
14 (35.9)
|
14 (40.0)
|
Lipid lowering therapy, n (%)
|
6 (15.4)
|
5 (12.8)
|
5 (14.3)
|
Oral anti-diabetics, n (%)
|
1 (2.6)
|
2 (5.1)
|
2 (5.7)
|
Allopurinol, n (%)
|
2 (5.1)
|
2 (5.1)
|
2 (5.7)
|
csDMARD = conventional synthetic Disease Modifying Anti-Rheumatic Drug, IQR= inter-quartile range, NSAID = non-steroidal anti-inflammatory drug, TNFi = tumour necrosis factor inhibitor
A total of 52 controls were included and started VLED treatment, whereof 42 and 39 controls came to the six- and 12-months visits respectively. A comparison of baseline characteristics between the PsA patients and controls is shown in table 2.
Table 2 Comparison between PsA patients and controls at baseline
|
Patients
N=39
|
Controls
N=39
|
p-value
|
Women, n(%)
Men, n(%)
|
25 (64.1)
14 (35.9)
|
29 (74.4)
10 (25.6)
|
0.462
|
Age
years
|
55
(48–62)
|
56
(48–60)
|
0.617
|
Body weight
kg
|
106.0
(93.5–112.5
|
107.0
(96.5–122.2)
|
0.358
|
Body height
cm
|
168.0
(161.0–176.0)
|
165.0
(161.0–170.0)
|
0.180
|
BMI
kg/m2
|
35.2
(33.9–37.9)
|
38.5
(36.9–41.7)
|
<0.001
|
Waistline
cm
|
115.0
(112.0–122.0)
|
117.0
(107–126.5)
|
0.635
|
BP systolic
mmHg
|
127.5
(115.0–136.0)
|
122.0
(115.0–137.0)
|
0.503
|
BP diastol
mmHg
|
77.5
(70.0–82.5)
|
77.0
(70.0–85.0)
|
0.736
|
CRP
mg/L
|
5
(3–9)
|
4
(2–6)
|
0.137
|
Hb
g/L
|
144
(131–150)
|
141
(135–149)
|
0.881
|
WBC
109/L
|
6.1
(5.3–7.7)
|
6.6
(5.6–7.6)
|
0.719
|
PLT
109/L
|
270
(202–299)
|
266
(225–302)
|
0.924
|
S-TC
mmol/L
|
5.5
(4.6–6.2)
|
4.9
(4.1–5.6)
|
0.034
|
S-LDL
mmol/L
|
3.6
(3.0–4.4)
|
3.2
(2.54.0)
|
0.08
|
S- HDL
mmol/L
|
1.4
(1.1–1.6)
|
1.4
(1.0–1.6)
|
0.867
|
S- TG
mmol/L
|
1.6
(1.2–2.4)
|
1.7
(1.2–2.1)
|
0.512
|
HbA1c mmol/mol
|
35.5
(32.0–37.2)
|
37.5
(35.0–44.0)
|
0.003
|
S-glucose
mmol/L
|
5.7
(5.3–6.2)
|
6.2
(5.8–7.1)
|
0.003
|
Creatinine
µmol/L
|
75
(64–84)
|
68
(62–79)
|
0.240
|
S-urate
µmol/L
|
330
(280–400)
|
350
(283–384)
|
0.928
|
ALT
µkat/L
|
0.52
(0.41–0.68)
|
0.50
(0.42–0.68)
|
0.806
|
Anti-hypertensives, n (%)
|
17 (43.6)
|
16 (41.0)
|
0.819
|
Lipid lowering therapy n (%)
|
6 (15.4)
|
11 (28.2)
|
0.170
|
Oral anti-diabetics n (%)
|
1 (2.6)
|
4 (10.3)
|
0.358
|
Insulin
n (%)
|
0 (0)
|
1 (2.6)
|
1.00
|
Allopurinol,
n (%)
|
2 (5.1)
|
1 (2.6)
|
1.00
|
Mets
n (%)
|
30
(76.9)
|
28
(71.8)
|
0.614
|
ALT = alanine transaminase, BMI = Body Mass Index, BP = blood pressure, CRP = C- reactive protein, Hb = haemoglobin HbA1c = glycosylated hemoglobin, HDL = high-density lipoprotein cholesterol, LDL = low-density lipoprotein cholesterol, PLT = platelet count, S– =serum, TC = total cholesterol, TG = triglycerides, Mets = metabolic syndrome, WBC = White Blood Cell count
Body weight and disease activity after 12 months
There were 39 patients with PsA attended the 12-month visit. At baseline their median weight was 106.0 (IQR 93.5–112.5) kg and at 12 months 87.5 (80.6–95.5) kg (p<0.001). In median, the patients had lost 16.1 (10.5–22.8) % of their baseline weight. In total, 77% (N=30/39) of the patients had a weight loss of 10% or more compared to baseline, and 17.9% (N=7/39) a weight loss of 5-10%. The majority of patients (34/39) had however regained some weight between the six- and 12-month visits, in median 3.9 (IQR 1.5–6.6) kg.
A majority of the disease activity variables were still significantly improved at the 12-month follow-up compared to baseline, including swollen/66 and tender/68 joints count, Leeds enthesitis index, BSA, CRP, VAS for global health and fatigue, HAQ, BASFI, DAS28-CRP and DAPSA. Criteria for MDA was met by 28.2% (11/39) at baseline and 38.5% (15/39) at the 12-months follow-up (p=0.008). However, no significant difference was found in VAS pain and DLQI at 12 months follow-up, compared to baseline. (Table 3 and Fig 2)
Table 3 Body weight, BMI and measures of disease activity and function before weight loss treatment at baseline (BL) and after 12-months (M12) and 24-months (M24) in patients with psoriatic arthritis (PsA). Values are median and inter-quartile range (IQR).
|
BL
N=39
|
M12
N=39
|
M24
N=35
|
BL vs M12
p-value
N=39
|
BL vs M24
p-value
N=35
|
Body weight
kg
|
106.0
(93.5–112.5)
|
87.5
(80.6–95.5)
|
92.7
(85.7–100.7)
|
<0.001
|
<0.001
|
BMI
kg/m2
|
35.2
(33.9–37.9)
|
30.5
(28.0–32.9)
|
32.3
(30.3–35.4)
|
<0.001
|
<0.001
|
Tender joints 68
count
|
4
(1–14)
|
3
(0-6)
|
2
(0-7)
|
0.001
|
<0.001
|
Swollen joints 66
count
|
0
(0–1)
|
0
(0–1)
|
0
(0-0)
|
0.015
|
0.003
|
Leeds enthesitis index, count
|
1
(0–4)
|
0
(0–2)
|
1
(0–3)
|
<0.001
|
0.002
|
BSA
%
|
0.75
(0–2)
|
0.25
(0–1)
|
0.25
(0–1.5)
|
0.018
|
0.194
|
CRP
mg/L
|
5
(3–9)
|
2
(1–5)
|
3
(2–5)
|
0.009
|
0.011
|
Hb
g/L
|
144
(131–150)
|
143
(131–150)
|
142
(135–149)
|
0.296
|
0.537
|
WBC
109/L
|
6.1
(5.3–7.7)
|
5.7
(5.0–7.4)
|
5.9
(4.7–7.8)
|
0.133
|
0.346
|
PLT
109/L
|
270
(202–299)
|
262
(203–308)
|
273
(220–332)
|
0.368
|
0.334
|
VAS Pain
mm
|
35
(15–60)
|
30
(14–46)
|
35
(15–55)
|
0.073
|
0.799
|
VAS Global health, mm
|
40
(20–70)
|
30
(17–52)
|
40
(20–60)
|
0.023
|
0.282
|
VAS Fatigue
mm
|
50
(30–70)
|
30
(20–60)
|
60
(30–80)
|
0.021
|
0.064
|
HAQ
score
|
0.63
(0.13–1.00)
|
0.25
(0–0.63)
|
0.25
(0.13-0.88)
|
0.002
|
0.033
|
BASFI
score
|
2.6
(1.4–5.2)
|
1.6
(0.57–3.6)
|
2.1
(0.9–4.2)
|
<0.001
|
0.131
|
DLQI
score
|
1
(0–4)
|
1
(0–4)
|
1
(0–4)
|
0.380
|
0.982
|
DAS-28CRP
score
|
3.1
(2.1–3.7)
|
2.4
(1.8–3.3)
|
2.4
(1.7–2.9)
|
<0.001
|
<0.001
|
DAPSA
score
|
16.1
(6.2–23.3)
|
9.5
(4.8–18.1)
|
9.7
(5.3–19.0)
|
<0.001
|
<0.001
|
MDA
n (%)
|
11 (28.2)
|
15 (38.5)
|
16 (45.7)
|
0.008
|
0.016
|
BASFI = Bath Ankylosing Spondylitis Functional Index, BMI = Body Mass Index, BSA = Body Surface Area, CRP = C- reactive protein, DAPSA = Disease Activity in PSoriatic Arthritis, DAS28CRP = Disease Activity Score using 28 joint counts based on CRP, DLQI = Dermatology Life Quality Index, ESR = erythrocyte sedimentation rate, HAQ = Health Assessment Questionnaire, Hb = haemoglobin, MDA = Minimal Disease Activity, PLT = platelet count, VAS = Visual Analogue Scale, WBC = White Blood Cell count
Subgroup analyses showed that the women with PsA demonstrated a greater reduction of CRP (∆CRP) compared to the men (p=0.041), and the men a greater reduction in swollen joints (∆swollen/66) than the women (p=0.047), but there was no significant difference in change in BMI, DAPSA, DA28CRP, BSA, HAQ between the sexes. PsA patients on a bDMARD had a greater reduction of tender joints (∆tender/68; p=0.015) than those without, but otherwise the change in disease activity was similar between the groups at 12 months. The weight loss and change in CRP was also comparable in the PsA patients and controls at the six and 12-month visits (table 4).
Table 4 Comparison between PsA patients and controls: change in bodyweight, BMI, CRP and urate at the 6-months (M6) and 12-months (M12) visits
|
Patients
N=39
|
Controls
N=39
|
p-value
|
M6
|
|
|
|
Weight loss since baseline.
kg
|
19.0
(15.6–27.0)
|
23.0
(17.5–29.5)
|
0.280
|
Weight loss/baseline weight
%
|
18.7
(15.2–26.7)
|
21.7
(15.5–26.4)
|
0.600
|
∆ BMI
kg/m2
|
7.1
(5.3–9.6]
|
8.0
(6.3–10.7)
|
0.213
|
∆ CRP
mg/L
|
1
(1–4)
|
2
(0–3)
|
0.562
|
∆ urate
µmol/L
|
40
(-2.0–81.5)
|
40.5
(6.8–77.5)
|
0.954
|
|
|
|
|
M12
|
|
|
|
Weight loss since baseline,
kg
|
16.1
(10.5–22.8)
|
16.6
(9.7–22.9)
|
0.885
|
Weight loss/baseline weight
%
|
16.3
(10.5–22.4)
|
15.7
(9.8–21.2)
|
0.853
|
∆BMI
kg/m2
|
5.7
(3.6–8.5)
|
6.0
(3.7-8.2)
|
0.657
|
∆ CRP
mg/L
|
2
(0–4)
|
2
(0–3)
|
0.952
|
∆ urate
µmol/L
|
28
(-6.5–68.5)
|
35.5
(-3.3–58.8)
|
0.775
|
BMI = Body Mass Index , CRP = C- reactive protein
Body weight and disease activity after 24 months
In total, 35 patients were examined at the 24-month visit. The median weight was 92.7 (IQR 85.7–100.7) kg and the median weight loss 7.4 ( IQR 5.1–14.0) % of the baseline weight. Forty percent (N=14/35) of the patients still had a weight loss of 10% or more compared to baseline, and 37.1% (N=13/35) had a weight loss of 5-10%. All patients had however regained some weight since the 12-month visit, in median 6.3 (IQR 3.4–9.3) kg.
At 24 months the swollen/66 and tender/68 joints count, Leeds enthesitis index, CRP, HAQ, DAS28-CRP and DAPSA scores were still significantly lower compared to baseline. (Table 2 and fig 2) Criteria for MDA was met by 45.7 % (16/35) at the 24 -month visit (p=0.016, compared to baseline). There were however no significant differences in BSA, DLQI, BASFI and VAS global health, pain and fatigue between baseline and 24 months.
Subgroup analyses showed that the women demonstrated a greater reduction in VAS Global health (p=0.007) and DAS28CRP (∆DAS28CRP) (p=0.011) than the men, but there was no significant difference in the change in BMI, CRP, swollen/tender joints, DAPSA, BSA, HAQ between the sexes. Patients treatment with a bDMARD had a greater reduction in tender joints (∆tender/68; p=0.029) and DAPSA (∆DAPSA; p=0.048) compared to patients without a bDMARD at 24 months. No comparisons between PsA patients and controls could be made at the 24-month visit.
Using PGIC to rate the PsA patients’ overall impression of change in status since study start, 54% of the patients reported “very much or much improved”, 23% “minimally improved”, 17 % “no change” and 6% “worse”. (Fig 3)
Change in disease activity and self-reported function in relation to weight loss
At 12-months follow-up the change in BMI compared to baseline (∆BMI) was significantly correlated with ∆DAS28CRP (rS=0.526, p=0.001), ∆DAPSA (rS=0.383, p=0.017), ∆CRP (rS=0.455, p=0.004), ∆BASFI (rS=0.455, p=0.004) and ∆VAS for global health (rS=0.483, p=0.002) in the PsA patients. (Fig 4a-d) At 24-months follow-up ∆BMI was correlated with ∆HAQ (rS=0.466, p=0.005) and ∆VAS for fatigue (rS=0.455, p=0.006). (Fig 4e) PGIC was also significantly associated with ∆BMI (rS= -0.412: p=0.014) at 24 months.
Change in variables associated with the metabolic syndrome
Data regarding the PsA patients is given in table 5. At the six-month visit blood pressure, HbA1c and serum levels of TC, LDL, TG and glucose were significantly reduced, and HDL was significantly increased, compared to baseline. After 12-months there were still significant reductions in HbA1c, TG and glucose and an increase in HDL, and after 24-months a lowered TG and higher HDL remained. Criteria for MetS were met by 76.9 % (30/39) at baseline, 35.9 % (14/39) after six months, 38.5 % (15/39) after 12-months, and by 60 % (21/35) after 24-months.
Table 5 Parameters associated with the metabolic syndrome at baseline (BL) and after weight loss at 6 months (M6), 12 months (M12) and 24 months (M24) in patients with psoriatic arthritis (PsA) and obesity. Values are median and inter-quartile range (IQR).
|
BL
N=39
|
M6
N=39
|
M12
N=39
|
M24
N=35
|
BL-M6
p-value
N=39
|
BL-M12
p-value
N=39
|
BL-M24
p-value
N=35
|
BMI
kg/m2
|
35.2
(34.0-37.8)
|
29.7
(26.2-31.4)
|
30.5
(28.0-33.0)
|
32.3
(30.2-35.4)
|
<0.001
|
<0.001
|
<0.001
|
Waistline
cm
|
115
(112–122)
|
95
(8–102)
|
97.5
(90–105)
|
107
(96–113)
|
<0.001
|
<0.001
|
<0.001
|
BP systolic
mmHg
|
127.5
(115–136)
|
120
(102.5–130)
|
121
(113.5–140)
|
124
(115–145)
|
<0.001
|
0.410
|
0.701
|
BP diastol
mmHg
|
77.5
(70–82.5)
|
70
(62.5–77.5)
|
72.5
(65–82.5)
|
75
(70–86)
|
<0.001
|
0.073
|
0.891
|
S-TC
mmol/L
|
5.5
(4.6–6.2)
|
4.9
(4.4–5.7)
|
5.4
(4.7–5.9)
|
5.6
(4.6–6.0)
|
0.019
|
0.508
|
0.227
|
S-LDL
mmol/L
|
3.6
(3.0–4.4)
|
3.2
(2.6–3.9)
|
3.4
(2.8–4.0)
|
3.9
(2.8–4.5)
|
0.003
|
0.468
|
0.596
|
S- HDL
mmol/L
|
1.4
(1.1–1.6)
|
1.4
(1.2–1.6)
|
1.65
(1.4–2.0)
|
1.6
(1.3–1.7)
|
0.001
|
<0.001
|
0.005
|
S- TG
mmol/L
|
1.6
(1.2–2.4)
|
1.1
(0.9–1.6)
|
1.2
(1.0–1.7)
|
1.4
(1.0–2.1)
|
<0.001
|
<0.001
|
0.007
|
HbA1c mmol/mol
|
35.5
(32.0–37.2)
|
32.5
(29.8–36.0)
|
34
(31–36)
|
35
(32–38)
|
<0.001
|
0.002
|
0.203
|
S-glucose
mmol/L
|
5.7
(5.3–6.2)
|
5.4
(5.1–5.8)
|
5.5
(5.0–5.9)
|
5.6
(5.2–6.2)
|
<0.001
|
0.006
|
0.055
|
Creatinine
µmol/L
|
75
(64–84)
|
70
(62–86)
|
72
(60–81)
|
74
(58–83)
|
0.066
|
0.041
|
0.009
|
ALT
µkat/L
|
0.52
(0.41–0.68)
|
0.45
(0.36-0.62)
|
0.42
(0.31–0.55)
|
0.45
(0.31–0.58)
|
0.143
|
0.015
|
0.075
|
Mets
n (%)
|
30
(76.9)
|
14
(35.9)
|
15
(38.5)
|
21
(60)
|
<0.001
|
<0.001
|
0.118
|
The change in serum levels of TG (∆TG) was correlated with ∆DAPSA (rS=0.532, p= 0.002) and ∆DAS28CRP (rS=0.398, p=0.027) at the 12 months visit. In addition, ∆LDL was correlated with ∆DAPSA (rS=0.437, p= 0.014) at the 24-month visit.
No significant differences were found for systolic and diastolic blood pressures at the 12- and 24-month visits compared to baseline, although three patients were able to stop treatment for hypertension during the study, and two patients could halve the dose of antihypertensives. Two patients stopped lipid-lowering therapy, while two other patients started lipid-lowering therapy during the follow-up. In addition, one patient started treatment with glucose-lowering medication during the study.
Data regarding the controls is given in supplementary table 1. Similar to the PsA patients, the controls demonstrated lowered serum levels of TC, LDL, TG, HbA1c, S-glucose and ALT and increased serum levels of HDL. Criteria for MetS were met by 71.8 % (28/39) of the controls at baseline, 48.7 % (19/39) after six months, 53.8 % (21/39) after 12-months
Serum urate was also significantly reduced during the study in both PsA patients and controls. (table 6). Baseline serum urate was similar in the PsA patients and controls, but significantly higher in men compared women in both the PsA patients (p=0.002) and controls (p=0.001). The decrease in serum urate (D-urate) was however not significantly correlated with D-BMI at any timepoint in the female and male patients and controls except in the PsA women at the 24-month visit (rS=0.493; p=0.032).
Table 6 Serum urate in median and inter-quartile range (IQR) at baseline (BL) and after weight loss at 6 months (M6), 12 months (M12) and 24 months (M24) in women and men with psoriatic arthritis and obesity.
S-urate
µmol/L
|
BL
|
M6
|
M12
|
M24
|
BL-M6
p-value
|
BL-M12
p-value
|
BL-M24
p-value
|
All patients
|
330
(280–400)
N=39
|
301
(244–356)
N=39
|
306
(250–348)
N=39
|
324
(264–351)
N=35
|
0.001
|
0.001
|
0.005
|
Women
|
309
(269–361)
N=25
|
259
(232–350)
N=25
|
286
(242–348)
N=25
|
294
(234–338)
N=21
|
0.023
|
0.053
|
0.033
|
Men
|
394
(324–416)
N=14
|
324
(304–367)
N=14
|
319
(294–357)
N=14
|
334
(306–368)
N=14
|
0.017
|
0.003
|
0.064
|
|
|
|
|
|
|
|
|
All controls
|
350
(283–384)
N=39
|
280
(243–356)
N=39
|
306
(267–352)
N=39
|
N.A.
|
<0.001
|
0.001
|
N.A.
|
Women
|
309
(278–366)
N=29
|
266
(230–299)
N=29
|
283
(235–321)
N=29
|
N.A.
|
0.001
|
0.002
|
N.A.
|
Men
|
400
(355–461)
N=10
|
359
(332–436)
N=10
|
372
(339–445)
N=10
|
N.A.
|
0.214
|
0.139
|
N.A.
|