Dietary interventions in fibromyalgia: a systematic review

Figures & data

Figure 1. PRISMA flow chart: summary of evidence search and selection.

Table 1. Characteristics and results of included studies.

Reference	Study design and participants	Intervention	Outcome measures	Results
Low FODMAPs diet
[20]	UCT Female FM patients (n = 38) Age: 51 ± 10 years	Diet low in FODMAPs (LFD) 4 weeks (31 women completed the intervention)	–Questionnaires: FSQ, FIQR, IBS-SSS, EQ-5D, VAS (abdominal pain and somatic pain); –evaluation of satisfaction with diet;	Comparison before and after the intervention: –↓ pain associated with FM, fatigue, gastric pain and intestinal changes after 4 weeks (p < .01) –↓GI symptoms (p < .01) –↑ mobility and ↓ discomfort in T2 (p < .05) –no significant differences in quality of life
Gluten-free diet
[16]	RCT Diet Group n = 35 female FM patients Age: 52 years Control Group n = 40 female FM patients Age: 53 years	Gluten-free diet (GFD) VS Hypocaloric Diet (HD); 6 months	–Anthropometric data: weight, BMI, waist perimeter –Biochemical analyzes –List of NCGS symptoms (GI symptoms, extraintestinal and FM-like) –Questionnaires: FIQR, BPI, PSQI, BDI, STAI, SF-12, PGI-I	Comparison before and after the intervention: –no significant differences in pain associated with FM and GI, extraintestinal and FM-like symptoms in GFD and HD Comparison between groups: –no significant differences
Hypocaloric diet
[18]	RCT Diet Group n = 43 FM obese patients (37 female Age: 44.8 ± 13.6 years Control Group n = 43 FM obese patients (38F) Age: 46.3 ± 14.4 years	Hypocaloric diet (1200 kcal/d: 20% Prot; 50% CH; 30% Fat) (G1) VS isocaloric diet (G2) 6 months	–Anthropometric data: weight, BMI, waist perimeter –Questionnaires: FIQ, TP, BDI, PSQI –Biomarkers: IL6 and PCR	Comparison before and after the intervention: –↓ pain associated with FM (p < .05), localised pain (p < .001), fatigue (p < .05) and depression (p < .001) in G1 –↓ IL6 and PCR (p < .05) in G1 Comparison between groups: –↓ IL6 (p = .034) and PCR (p = .07) in G1 vs G2
[22]	UCT Female FM patients (n = 48) Age: 54.5 ± 8.1 years	Hypocaloric diet (weight loss programme with weekly group sessions) 5 months (31 women completed the intervention)	–Anthropometric data: weight, BMI, waist circumference –Questionnaires: FIQ, HAQ, MPI, BDI, STAI, QOL, BSQ –Food diary (5 days)	Comparison before and after the intervention: –↓ pain associated with FM (p = .00), severity (p  =  .04) and day-to-day pain interference (p < .001) –↑ body image (p < .001) and quality of life (p < .001) –↓ anxiety (p < .001) and depression (p < .001) Correlation between variables: –Positive correlation between ↓ BMI and ↓ FM-associated pain (p = .02) and day-to-day pain interference (p < .001)
Vegetarian diet
[19]	CCT Diet Group n = 18 FM female patients Age: 51 years Control Group n = 15 FM female patients Age: 52 years	Vegan Diet (raw veg, fruit, whole grains, oilseeds and legumes) (VD) VS Omnivorous Diet (OD) 3 months	–Questionnaires: TP, VAS pain, BDI, HAQ –Biomarkers: Haematocrit, ERS, total cholesterol, urinary sodium –Food diary (5 days)	Comparison before and after the intervention: –↓ pain (p < .005) but not significant of the PT (p = .07) –↑ autonomy (p = .03), sleep quality (p = .01), morning stiffness (p = .0001) –↓ total cholesterol (p < .003) and urinary Na (p = .0001) –no significant statistical differences on depression, ERS and haematocrit
[21]	UCT FM patients (n = 30), (28 female)	Raw vegan diet (raw veg, fruit, whole grains, oilseeds) 7 months (20 adults completed the intervention)	–Questionnaires: FIQR, SF36, QOL, FFQ	Comparison before and after the intervention: –↓ pain associated with FM (p < .05) –↑ vitality, mobility, emotional health and general well-being after 7 months (p < .01) –↑ general quality of life (p < .05)
Monossodium Glutamate and Aspartame-free diet
[17]	RCT Diet Group n = 36 FM female patients Age: 42.3 ± 8.4 years Control Group n = 36 FM female patients Age: 39.6 ± 8.2 years	Diet free of monosodium glutamate (G1) VS diet without dietary restrictions (G2, on waiting list) 3 months	–Questionnaire: VAS pain –Food diary (3 months)	Comparison between groups, before and after the intervention: –no significant statistical differences

BDI: Beck Depression Inventory; BMI: Body Mass Index; BPI: Brief Pain Inventory; BSQ: Body Shape Questionnaire; CCT: Controlled Clinical Trial; EQ-5D: EuroQol-5Dimension; FFQ: Food Frequency Questionnaire; FIQ: Fibromyalgia Impact Questionnaire; FIQR: Fibromyalgia Impact Questionnaire Revised; FSQ: Fibromyalgia Survey Questionnaire; HAQ: Health Assessment Questionnaire; IBS-SSS: Irritable Bowel Syndrome-Symptom Severity Survey; MPI: Multidimensional Pain Inventory; NCGS: Non Coeliac Gluten Sensitivity; PGI-S: Patient Global Impression Scale Severity; PGI-I: Patient Global Impression Scale Improvement; PSQI: Pittsburgh Sleep Quality Index; QOL: Quality Of Fife Survey; RCT: Randomized Controlled Trial; SF-36: Short-Form Healthy Survey; STAI: State Trait Anxiety Inventory; TP: Tender Points; UCT: Uncontrolled Clinical Trial; VAS: Visual Analogic Scale.

Table 2. Cochrane risk of bias of included studies.

	Shapiro [22]	Slim [16]	Marum [20]	Vellisca [17]	Kaartinen [19]	Senna [18]	Donaldson [21]
Randomization sequency generation	X	+	X	?	X	+	X
Allocation concealment	X	?	X	?	X	+	X
Blinding of participants and personal	X	X	X	X	X	X	X
Blinding of outcomes assessment	X	X	X	X	X	X	X
Incomplete outcome data	X	+	X	+	+	+	+
Selective reporting	+	+	+	+	+	+	+
Other bias	X	?	?	?	?	?	?

+: Low risk of bias; X: High risk of bias; ?: Unclear risk of bias.

Table 3. GRADE Profile for studies evaluating the impact of dietary interventions on PRO and inflammation in FM patients.

Dietary interventions compared to usual treatment in Fibromyalgia patients
Certainty assessment							Summary of findings
Participants (studies) Follow-up	Risk of bias	Inconsistency	Indirectness	Imprecision	Publication bias	Overall certainty of evidence	Study event rates (%)		Relative effect (95% CI)	Anticipated absolute effects
							With usualtreatment	With Dietaryinterventions		Risk withusualtreatment	Risk difference with Dietary interventions
Pain in FM (CRITICAL OUTCOME, assessed with Questionnaires: FIQ, FIQR; Scale from 0–100)
RCTs
161 (2 studies) (6 months) [18,32]	serious^a^,^b	serious^c	not serious	serious^d	none	⨁◯◯◯ VERY LOW	83	78	–	–	The mean FIQ in intervention group was 4.27 lower (7.3 lower to 1.24 lower)
Observational
82 (3 studies) (5–7 months) [21,22]	serious^a^,^e^,^f^,^g^,^h^,^i	not serious	not serious	serious^d	none	⨁◯◯◯ VERY LOW	–	82	–	–	–
Pain in FM (CRITICAL OUTCOME, assessed with VAS; Scale from 1–10)
RTCs
105 (2 studies) (3 months) [17,19]	serious^a^,^e^,^i^,^j^,^k	serious^l^,^m	serious^n	serious^d	none	⨁◯◯◯ VERY LOW	51	54	–	–	The mean VAS in intervention group was 0.16 lower (0.58 lower to 0.26 higher)
Fatigue in FM (CRITICAL OUTCOME, assessed with FIQ and FIQR; Scale from 0–100)
RCTs
161 (2 studies) (6 months) [18,32]	serious^a^,^b	serious^c	not serious	serious^d	none	⨁◯◯◯ VERY LOW	83	78	–	–	not pooled
Observational
82 (3 studies) (1–7 months) [20,21^,22]	very serious^a^,^e^,^f^,^g^,^h^,^i	not serious	not serious	serious^d	none	⨁◯◯◯ VERY LOW	–	82	–	–	–
Quality of Sleep in FM (IMPORTANT OUTCOME, assessed with PSQI; Scale from 0–21)
RCTs
161 (2 studies) (6 months) [18,32]	serious^a^,^b	serious^c	not serious	serious^d	none	⨁◯◯◯ VERY LOW	83	78	–	–	The mean PSQI in intervention group was 1.22 lower (2.05 lower to 0.4 lower)
Anxiety in FM (IMPORTANT OUTCOME, assessed with STAI, EQ-5D)
Observational
62 (2 studies) (1–5 months) [20,22]	very serious^a^,^e^,^f^,^i	not serious	not serious	serious^d	none	⨁◯◯◯ VERY LOW	–	62	–	–	–
Depression in FM (IMPORTANT OUTCOME, assessed with BDI; Scale from 0–50)
RCTs
194 (3 studies) (3–6 months) [16,18,19]	very serious^a^,^b^,^e^,^j	serious^c^,^l^,^m	not serious	serious^d	none	⨁◯◯◯ VERY LOW	98	96	–	–	The mean BDI in intervention group was 0.86 lower (1.79 lower to 0.07 higher)
Observational
31 (1 study) (5 months) [22]	serious^a^,^e^,^f^,^i	not serious	not serious	serious^d	none^o	⨁◯◯◯ VERY LOW	–	31	–	–	–
Quality of Life in FM (CRITICAL OUTCOME, assessed with SF-12, HAQ)
RCTs
108 (2 studies) (3–6 months) [16,19]	serious^a^,^b^,^e	serious^c^,^l^,^m	not serious	serious^d	none	⨁◯◯◯ VERY LOW	55	53	–	–	not pooled
Quality of Life in FM (CRITICAL OUTCOME, assessed with EQ-5D, SF-36, HAQ)
Observational
82 (3 studies) (4 weeks to 7 months) [20,21,22]	very serious^a^,^e^,^f^,^g^,^h	not serious	not serious	serious^d	none	⨁◯◯◯ VERY LOW	–	82	–	–	–
Gastrointestinal symptoms in FM (IMPORTANT OUTCOME, assessed with IBS-SSS)
Observational
31 (1 study) (1 month) [20]	serious^a^,^e^,^f	not serious	not serious	serious^d	none^o	⨁◯◯◯ VERY LOW	–	31	–	–	-
Inflammation in FM (IMPORTANT OUTCOME, assessed with IL6 biomarkers)
RCTs
86 (1 study) (6 months) [18]	not serious^a	not serious	not serious	serious^d	none	⨁⨁⨁◯ MODERATE	43	43	–	–	The mean IL6 in intervention group was 0.7 lower (1.31 lower to 0.09 lower)
Inflammation in FM (IMPORTANT OUTCOME, assessed with PCR biomarker)
RCTs
86 (1 study) (6 months) [18]	not serious^a	not serious	not serious	serious^d	none	⨁⨁⨁◯ MODERATE	43	43	–	–	The mean PCR in intervention group was 0.6 lower (1.06 lower to 0.14 lower)

RCTs: Randomized Controlled Trials; CI: Confidence interval.

^a According to the nature of a dietary study, there isn't possible to blind population. As so, this was not considered a factor to downgrade studies.

^b Study does not use an independent control group (two different interventions are applied).

^c Heterogeneity of the population in pharmaceutical therapy, at baseline.

^d Sample size is always reduced in all studies; the Optimal Information Size (OIS) is never estimated.

^e Study did not apply intention-to-treat analysis, despite presenting >5% of loss of follow up and/or do not justify missing data.

^f Study has no control group.

^g Do not control possible confounding factors, as the increased Omega 3 ingestion, which is known to have an anti-inflammatory potential.

^h Do not justify missing data.

ⁱ Do not control possible confounding factors, as medical therapy.

^j Study has no randomization.

^k Study does not indicate how the randomization was done.

^l Heterogeneity of the population in BMI, at baseline.

^m Heterogeneity of the population in pain level, at baseline.

ⁿ Study evaluates only pain, ignoring other important symptoms for the patient.

^o Single study.

Table 4. Summary of author’s justification of risk of bias classification.

Bias	Author’s judgement	Support for judgement
Randomization sequence generation	High risk	Uncontrolled or unrandomized clinical trial
Allocation concealment	High risk	Open label
	Unclear risk	Authors don’t define in the study
	Low risk	Quote: “outcome assessor was unaware of allocation of patients”
Blinding of participants and personal	High risk	Open label
Blinding of outcomes assessment	High risk	Open label
Incomplete outcome data	High risk	Missing data >10% without intention-to-treat analysis
	Low risk	No missing data or, in the presence of missing data >10%, authors describe an intention-to-treat analysis
Selective reporting	Low risk	All pro-specified outcomes were reported
Other bias	High risk	Uncontrol of possible confounders, like medication or diet compliance
	Unclear risk	Authors don’t say if possible confounding domain exists

Marum AP, Moreira C, Saraiva F, et al. A low fermentable oligo-di-mono saccharides and polyols (FODMAP) diet reduced pain and improved daily life in fibromyalgia patients. Scand J Pain. 2016;13:166–172.

 PubMed Web of Science ®Google Scholar

Slim M, Calandre EP, Garcia-Leiva JM, et al. The effects of a gluten-free diet versus a hypocaloric diet among patients with fibromyalgia experiencing gluten sensitivity-like symptoms: a pilot, open-label randomized clinical trial. J Clin Gastroenterol. 2017;51:500–507.

 PubMed Web of Science ®Google Scholar

Senna MK, Sallam RA, Ashour HS, et al. Effect of weight reduction on the quality of life in obese patients with fibromyalgia syndrome: a randomized controlled trial. Clin Rheumatol. 2012;31:1591–1597.

 PubMed Web of Science ®Google Scholar

Shapiro JR, Anderson DA, Danoff-Burg S. A pilot study of the effects of behavioral weight loss treatment on fibromyalgia symptoms. J Psychosom Res. 2005;59:275–282.

 PubMed Web of Science ®Google Scholar

Kaartinen K, Lammi K, Hypen M, et al. Vegan diet alleviates fibromyalgia symptoms. Scand J Rheumatol. 2000;29:308–313.

 PubMed Web of Science ®Google Scholar

Donaldson MS, Speight N, Loomis S. Fibromyalgia syndrome improved using a mostly raw vegetarian diet: an observational study. BMC Complem Altern Med. 2001;1:7.

 PubMedGoogle Scholar

Vellisca MY, Latorre JI. Monosodium glutamate and aspartame in perceived pain in fibromyalgia. Rheumatol Int. 2014;34:1011–1013.

 PubMed Web of Science ®Google Scholar

Slim M, Calandre EP, Rico-Villademoros F. An insight into the gastrointestinal component of fibromyalgia: clinical manifestations and potential underlying mechanisms. Rheumatol Int. 2015;35:433–444.

 PubMed Web of Science ®Google Scholar