Immunomodulatory drug – Rituximab

Rituximab 500 mg/m² (maximum 1000 mg), diluted in saline to a concentration of 2 mg/ml, given twice two weeks apart, with nurse surveillance and according to local guidelines used for treating B-cell lymphomas. Infusion bags had double plastic covers to avoid content identification by nurse or patient.

Duration: 2 weeks

Versus

Placebo

An equal volume of saline, given twice two weeks apart, with nurse surveillance and according to local guidelines used for treating B-cell lymphomas. Infusion bags had double plastic covers to avoid content identification by nurse or patient.

Duration: 2 weeks

No additional infusions, or other interventions, were given during follow-up. All patients were given oral cetirizine 10 mg, paracetamol 1 g, and dexamethasone 8 mg prior to infusion.

N=30 people with CFS, diagnosed by a neurologist, according to the CDC criteria 1994 (Fukuda 1994). Pre-treatment evaluation included thorough clinical examination, standard laboratory tests and further diagnostic tests if pre-treatment evaluation revealed any relevant abnormality that could explain the severe fatigue.

Strata details: adults (18-65 years); severity mixed or unclear.

Conducted in Norway

Fatigue, cognitive, pain and other symptoms scores calculated as the mean of different self-reported symptom scales (0-6) – measures not validated.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; Post exertional Malaise (PEM) is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Immunomodulatory drug – Rituximab

Induction treatment with 2 infusions, 2 weeks apart, of rituximab (MabThera, Roche), 500 mg/m2 of body surface area (maximum of 1000 mg). In the maintenance phase, patients received a 500-mg fixed dose of rituximab at 3, 6, 9, and 12 months.

Duration: 12 months

Versus

Placebo

Induction treatment with 2 infusions, 2 weeks apart, of 500 mg/m2 of body surface area (maximum of 1000 mg) saline with added human albumin (Flexbumin [Baxalta] or Albunorm [Octapharma]), 0.4 mg/mL, to ensure no visible difference from the active comparator. In the maintenance phase, patients received a 500-mg fixed dose of saline with human albumin at 3, 6, 9, and 12 months.

Duration: 12 months

One hour before infusions, all patients received premedication with1g of oral acetaminophen, 10 mg of cetirizine, and 8mg of dexamethasone.

N=152 people with ME/CFS according to Canadian consensus criteria (Carruthers 2003). Patients where the workup uncovered other pathology as a possible cause of symptoms were excluded.

Strata details: adults (18-65 years); severity mixed or unclear (mild or mild/moderate 40%, moderate 30%, moderate/severe and severe 30%; patients with very severe ME/CFS (WHO function class IV), who were totally bedridden and in need of care were excluded).

Fatigue/fatigability (Fatigue severity scale; fatigue numeric rating scale)

Physical function (SF36 physical function; function level percentage)

Adverse events (any adverse events and any serious adverse events with possible/probable relation to intervention; suspected unexpected serious adverse reactions)

Activity levels (mean steps per 24 hours)

Conducted in Norway

PEM reporting: all participants met the Canadian criteria which has PEM as a compulsory feature.

Immunomodulatory drug – IV immunoglobulin G

High-dose intravenous IgG was administered intravenously by continuous infusion in a dosage of 2g/kg. Three infusions lasting 24 hours were administered at monthly intervals.

Duration: 3 months (3 infusions)

Versus

Placebo

A placebo solution of 10% w/v maltose was administered intravenously for 24 hours at an equivalent volume to the IgG infusion.

Duration: 3 months (3 infusions)

N=49 people with CFS, diagnosed according to Lloyd 1988 criteria.

A physical examination and standardized investigation protocol excluded other chronic infectious or immunodeficiency-related disorders.

Strata details: adults (18-65 years); severity mixed or unclear.

Psychological status (Hamilton Depression Scale & Zung Self-Rating Depression Scale)

Adverse events (phlebitis & constitutional symptoms)

Conducted in Australia

‘Quality of life visual analogue scales modified to include 10 aspects of physical and neuropsychiatric symptomatology typical of CFS (fatigue, headaches, myalgia, concentration impairment and functional activity)’ were completed, but a single value for QoL was reported for each group. This outcome was not extracted due to lack of information on how the overall score was derived, the range, or the direction of scales PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – Lloyd 1988 criteria were excluded from the diagnostic criteria review on the basis there was unclear methodology for the development of the criteria and have therefore been downgraded here for indirectness. The study states that the criteria emphasize the same features as the criteria published subsequently by the Centers for Disease Control [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Immunomodulatory drug – IV immunoglobulin G

Patients were scheduled to receive a total of six infusions of IV IgG. Infusions were initiated at a rate of 0.5 mL/kg/hour and increased as per the IV IgG package insert to a maximum of 4 mL/kg/hour.

Duration: once per month for 6 months

Versus

Placebo

Participants in the placebo group received the same course of IV but IgG was replaced with an exactly correlating volume of a 1% albumin solution as placebo.

Duration: once per month for 6 months

N=30 people with CFS, diagnosed according to the diagnostic criteria of CDC /Holmes 1988 after thorough medical, psychometric, and psychiatric evaluations did not establish another explanation for chronic fatigue.

Strata details: adults (18-65 years); severity mixed or unclear.

Physical functioning (on the Medical Outcome Study Short Form, a precursor to SF36)

Psychological status (mental health on the Medical Outcome Study Short Form)

Adverse events (major adverse events)

Conducted in USA

PEM reporting: 96.4% of participants had “post-exertional fatigue (prolonged)” but as there was no additional information the committee was not satisfied that this symptom was PEM)Serious population indirectness – Holmes 1988 criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear; unclear if symptom described in study is PEM [PEM reanalysis].

Immunomodulatory drug – rintatolimod (Ampligen)

Patients initially received a 200 mg IV dose of rintatolimod twice weekly for two weeks. Following this, a 400 mg dose of rintatolimod was administered twice weekly for 40 weeks.

Total duration: 42 weeks.

Versus

Placebo

Patients initially received a 200 mg IV dose of placebo (physiological saline) twice weekly for two weeks. Following this, a 400 mg dose of saline placebo was administered twice weekly for 40 weeks.

Total duration: 42 weeks.

N=234 people with CFS, diagnosed according to the CDC criteria 1988 (Holmes 1988) & 1994 CDC criteria. A complete medical history to confirm diagnosis was performed at baseline. A complete physical examination and laboratory tests were also performed. Only subjects with Karnofsky Performance Score values ranging from 20 to 60 were eligible.

Strata details: adults (18-65 years); severity mixed or unclear.

Participants were stratified according to treadmill duration (≥9 minutes vs >9 minutes) then randomised.

Quality of life (Vitality Score subscale)

Adverse events (Serious Adverse Events with possible/probable relation to intervention)

Physical functioning (Karnofsky Performance Score & Activities of Daily Living)

Exercise performance measure (treadmill exercise duration)

Conducted in USA.

SD or CIs were not reported for quality of life and physical functioning outcomes and therefore are not analysed.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – Holmes 1988 criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Immunomodulatory drug – IV immunoglobulin G

Participants received 3 infusions, each lasting 24 hours at monthly intervals. Three dose arms:

1. IV IgG (Intragram) at 0.5 g/kg (n=22)
2. IV IgG (Intragram) at 1.0 g/kg (n=28)
3. IV IgG (Intragram) at 2.0 g/kg (n=23)

Placebo

Identical placebo solution given via IV (n=26). Participants received 3 infusions each lasting 24 hours at monthly intervals.

IV IgG dose arms are pooled for analysis.

N=99 people with CFS, diagnosed according to Schluederberg criteria.

Strata details: adults (18-65 years); severity mixed or unclear.

Conducted in Australia

Study reported quality of life outcomes (QAL, POMS depression, confusion, fatigue & energy) and physical function (Karnofsky scale but no analysable data were reported for these outcomes. Results are reported narratively in the clinical evidence table.

PEM reporting: the percentage of participants with PEM was not reported. Serious population indirectness – unclear criteria used.

Schluederberg 1992 publication was not included in the diagnostic criteria review as it presented a review of the CDC 1988 criteria rather than an original set of criteria [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Antidepressants – serotonin-norepinephrine reuptake inhibitors – Duloxetine hydrochloride

Duloxetine hydrochloride at 30 mg once a day for 1 week, then 60 mg one a day for a following 3 weeks. The dose then increased to 90mg per day for next 4 weeks (as tolerated). If highest doses not tolerated doses could be as low as 60mg per day.

Duration: 12 weeks.

Versus

Placebo

Identical placebo given in same way as study drug. Duration: 12 weeks.

N=60 people with CFS, diagnosed according to the CDC criteria 1994 (Fukuda 1994)

Strata details: adults (18-65 years); severity mixed or unclear.

Quality of life (eight SF36 subscales)

Fatigue (MFI-20, five subscales)

Psychological status (HADS anxiety & depression)

Pain (Brief Pain Inventory for severity & interference)

Symptom scales (Clinical Global Impression of Severity & Improvement; CDC symptom inventory)

Adverse events

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Antidepressants – MAOIs – Moclobemide

Moclobemide is a reversible inhibitor of monoamine oxidase (RIMA). Treatment was initially given as 150 mg tablets to be taken twice daily after meals. After 1 week the dose was increased to 2 tablets in morning and 1 tablet at night for a total dose of 450mg/day. This was increased to 600mg/day if tolerated.

Duration: 6 weeks.

Versus

Placebo

increased to 2 tablets in morning and 1 tablet at night for a total dose of 450mg/day. This was increased to 600mg/day if tolerated.

Duration: 6 weeks.

Concurrent care: Intermittent night dosages of benzodiazepines were allowed for insomnia.

N=90 people with CFS, diagnosed according to Lloyd 1988 criteria.

Strata details: adults (18-65 years); severity mixed or unclear.

Physical functioning (Karnofsky Performance Index)

Psychological status (Profile Of Mood States, POMS fatigue, vigor and depression)

Symptom scales (Globally improved cases (patient-rated))

Conducted in Australia.

Results reported are standard units of improvement (pre-treatment score-posttreatment score/SD of mean pre-treatment score)

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness - Lloyd 1988 criteria were excluded from the diagnostic criteria review on the basis there was unclear methodology for the development of the criteria and have therefore been downgraded here for indirectness [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Antidepressants – selective serotonin reuptake inhibitors – Fluoxetine

Fluoxetine 20 mg u.i.d. Duration: 12 weeks.

Versus

Other - Amisulpride

Amisulpride (a substituted benzamide) is an atypical antipsychotic. Given at 25 mg b.i.d.

Duration: 12 weeks.

Quality of life (SF-12)

Fatigue (Fatigue Severity Scale)

Psychological status (HADS anxiety & depression)

Pain (on VAS)

Adverse events (FIBSER – global burden)

Symptom scales (Clinical Global Impression of severity, CGI-S)

Conducted in Italy.

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness – 1994 CDC criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis.

Antidepressants – selective serotonin reuptake inhibitors – Fluoxetine

Fluoxetine 20 mg once daily. Duration: 8 weeks.

Versus

Placebo

Placebo given once daily. Duration: 8 weeks.

N=107 people with CFS, diagnosed according to Oxford Criteria (Sharpe 1991) criteria.

Strata details: adults (18-65 years); severity mixed or unclear.

Fatigue (Checklist Individual Strength (CIS) fatigue)

Psychological status (Beck Depression Inventory)

Adverse events (tremor & perspiration)

Symptom scales (self-reported global improvement)

Conducted in Netherlands.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness - Oxford criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

This four-arm study compared an antidepressant, graded exercise and placebos of both:

1. Fluoxetine & exercise control
2. Graded exercise & placebo
3. Fluoxetine & graded exercise
4. Placebo & exercise control

Fluoxetine at a fixed daily dose of 20 mg.

Duration: 6 months.

Versus

Graded exercise

Subjects were instructed to carry out their preferred aerobic activity (usually walking/jogging, swimming or cycling), for 20 minutes, at least three times per week. The intensity of the activity was initially set at a level which utilised oxygen at approximately 75% of the subject’s tested functional maximum. Exercise intensity was increased when there was a consistent recorded reduction of 10 beats per minute in post-exercise heart rate for one week and two points on the perceived exertion scale.

This group also received a placebo fluoxetine capsule of similar taste and appearance, taken daily. Duration: 6 months.

Versus

Placebo or exercise control

Fluoxetine placebo: a capsule of similar taste and appearance, taken daily for 6 months. Exercise control consisted of a placebo exercise programme in which participant activity diaries were reviewed by a physiotherapist. Subjects were not offered any specific advice on how much exercise they should be taking but were told to do what they could when they felt capable and to rest when they felt they needed to.

N=136 people with CFS, diagnosed according to Oxford Criteria (Sharpe 1991).

Strata details: adults (18-65 years); severity mixed or unclear.

Fatigue (14-item Chalder fatigue scale)

Psychological status (depression on the Hospital Anxiety and Depression Scale)

Exercise performance measure (functional work capacity/VO2 peak)

Conducted in United Kingdom.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness - Oxford criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Nasal corticosteroids – Flunisolide

Nasal (not oral) corticosteroid (Flunisolide) self-administered with two sprays twice daily.

Duration: 4 weeks - 8 weeks (see comments).

Versus

Placebo

Saline spray, two sprays daily.

Duration: 4 weeks - 8 weeks (see comments).

N=28 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994). All participants also had rhinitis

Strata details: adults (18-65 years); severity mixed or unclear.

Sleep quality (Epworth Sleepiness Scale & Functional Outcomes of Sleep Questionnaire)

Fatigue (Chronic Fatigue Syndrome Severity Rating)

Conducted in USA.

This was a hybrid parallel/cross-over trial design. There were 4 groups of 7 who were treated as follows: active throughout; placebo throughout; active then placebo; placebo then active. Thus 21 had the active treatment at one point. In the analysis the results from these 21 people were aggregated without any apparent adjustments for some having had the other treatment (with the possibility of carryover effects).

PEM reporting: the percentage of participants with PEM was not reported.

Very serious population indirectness – 1994 CDC criteria used; PEM is not a compulsory feature and all participants had rhinitis [original analysis]; percentage of participants with PEM unclear & all participants had rhinitis [PEM reanalysis]

Oral corticosteroids – Hydrocortisone Hydrocortisone pills, dose of 16 mg per square metre of body surface per day (20-30mg every morning at 8am and 5mg every day at 2pm).

Duration: 12 weeks.

Versus

Placebo

Identical placebo at same doses as hydrocortisone group. Duration: 12 weeks.

N=70 people with CFS, diagnosed according to the CDC criteria 1988 (Holmes 1988). CDC criteria 1994 (Fukuda 1994) were also met.

Strata details: adults (18-65 years); severity mixed or unclear.

Psychological status (Beck Depression Inventory; Profile of Mood States seven subscales; Symptom checklist 90-R three subscales; Hamilton Depression Scale)

Adverse events (adverse reaction)

Activity levels (activity scale)

Symptom scales (Wellness Scale & Sickness Impact Profile)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness – Holmes 1988 and 1994 CDC criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Oral corticosteroids – Fludrocortisone

Initial dose of fludrocortisone acetate was 0.1mg via 1 tablet orally. Dose doubled if no AEs reported after 2 weeks of treatment.

Duration: 6 weeks.

Versus

Placebo

Identical tablets taken at same dosing regimen as study drug. Duration: 6 weeks.

Patients told not to make any dietary changes (including salt intake) during study.

N=25 people with CFS, diagnosed according to the CDC criteria 1988 (Holmes 1988) and CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-65 years); severity mixed or unclear.

Quality of life (SF36)

Cognitive function (inability to concentrate, forgetfulness and confusion all on VAS; reaction time)

Psychological status (positive and negative effects scale, PANAS)

Pain (muscle pain and joint pain on VAS)

Sleep quality (unrefreshing sleep on VAS)

Adverse events (adverse events, adverse events causing withdrawal from the study)

Activity levels (distance until exhausted)

Exercise performance measures (time on treadmill)

Symptom scales (headaches, painful lymph nodes and sore throat on VAS)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – Holmes 1988 and 1994 CDC criteria used; PEM is not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Oral corticosteroids – Fludrocortisone

Fludrocortisone starting at a dose of 0.025 mg/day (1 capsule) for a week, then 0.05 mg/day (2 capsules) for the following week, and eventually increased to 0.1 mg/day (4 capsules) for remaining 7 weeks. Total duration: 9 weeks.

Versus

Placebo

Identical capsules containing methylcellulose only were given exactly as the study drug in the same dose increments. Total duration: 9 weeks.

Patients advised to drink at least 2L of fluid per day and to keep normal NaCl intake to their usual levels. Both groups also had KCl tablets 10mEq/day for duration of treatment. If AEs emerged, doses were reduced to previously tolerated levels.

N=70 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-50 years); severity mixed or unclear.

All participants also had neurally-mediated hypotension.

Fatigue (Wood Mental Fatigue Inventory; POMS vigour and fatigue subscales)

Physical functioning (SF-36 physical function subscale)

Psychological status (Beck Depression Inventory & SF-36 mental health subscale)

Adverse events (adverse effects)

Activity levels (Duke Activity Status)

Symptom scales (Wellness Score)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.

Very serious population indirectness–1994 CDC criteria used; PEM is not a compulsory feature and all participants also had neurally-mediated hypotension [original analysis]; percentage of participants with PEM unclear and all participants had neurally-mediated hypotension [PEM reanalysis]

Sympathomimetic/central antihypertensive drugs – Clonidine

Clonidine IV infusion 2.5 micro-g/kg in 10ml normal saline over 5 minutes. One-off treatment.

Versus

Placebo

IV infusion of 10ml normal saline over 5 mins. One-off treatment.

Heparinised cannula used for infusion.

N=10 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994)

Strata details: adults (18-60 years); severity mixed or unclear.

Conducted in United Kingdom

Crossover: randomised order for clonidine/placebo with washout of 2 weeks.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Sympathomimetic/central antihypertensive drugs – Clonidine

Tablets containing 25µg of clonidine hydrochloride were enclosed in orange opaque, demolition-resistant lactose capsules. Clonidine dosages were chosen to yield plasma concentrations within the lower range of what is considered clinically effective. One-half of the dose was given during the first 3 days to minimize introductory adverse effects. After 8 weeks of the full dose, the dose was halved for 1 additional week to avoid rebound effects, after which treatment was discontinued.

Versus

Placebo

Empty capsules were used as placebo comparators. Duration: 8 weeks.

N=120 CFS patients (excluding 68 healthy controls), diagnosed according to the following: 3 months of unexplained, disabling chronic/relapsing fatigue of new onset, and no accompanying symptoms.

Strata details: young people (12-18 years); severity mixed or unclear.

Fatigue (Chalder Fatigue Questionnaire)

Physical functioning (functional disability inventory)

Cognitive function (digit span backward test total)

Pain (BPI average pain score)

Sleep quality (KSQ insomnia score)

Adverse events (self-reported)

Activity levels (steps per day measured by accelerometer)

Symptom scales (CFS symptom inventory hypersensitivity score)

Conducted in Norway

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – PEM not compulsory feature of diagnosis [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Sympathomimetic/central antihypertensive drugs – Methylphenidate

10 mg twice daily (8am and 2pm). Taken for 1 month.

Vs

Placebo

Crossover: The same 60 patients took both drugs, but in a random order. Thus about half would have had the study drug in the first period, whilst the other half would have had the placebo first. A washout period of 1 week was used before each patient took the alternative compound in the second period of 4 weeks. Patients who stopped the treatment during the first period but who returned after 4 weeks were allowed to start therapy with the second compound

N=60 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-50 years); severity mixed or unclear.

Quality of life (SF-36 physical and mental subscales)

Fatigue (CIS fatigue total score)

Psychological status (HADS depression and HADS anxiety)

Adverse events (six categories)

Conducted in Belgium.

Crossover: 1 week (half- life of drug = 2 hours, so likely to be appropriate).

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Sympathomimetic/central antihypertensive drugs – Methylphenidate

5mg methylphenidate daily for week 1 and 10mg twice daily for weeks 2 to 12. Mitochondrial modulator (nutritional supplement) given as 4 tablets twice daily. The combination of these two agents is called KPAX002. KPAX002 is comprised of a low dosage of methylphenidate hydrochloride, combined with nutrients believed to modulate mitochondrial function.

Duration: 12 weeks

Versus

Placebo

Placebo version of KPAX002 treatment.

Duration: 12 weeks

N=135 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults; severity mixed or unclear.

Fatigue (CIS fatigue total score; fatigue on VAS)

Cognitive function (concentration disturbance on VAS)

Adverse events (AEs leading to discontinuation; serious AEs - pyelonephritis)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Amphetamines – Dexamphetamine

Dexamphetamine 5mg twice daily for first week. Dose increased to 10mg twice daily if indicated at start of 2nd week. Increment repeated if appropriate at start of 3rd week.

This dose continued for a further 4 weeks.

Duration 6 weeks.

Versus

Placebo

Identical doses and strategies for increase as study drug. Duration 6 weeks

N=20 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (17-72); severity mixed or unclear.

Quality of life (SF36 physical and mental)

Fatigue (Severity Scale)

Sleep quality (sleep latency)

Adverse events (anorexia)

Conducted in Australia.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Amphetamines - Lisdexamphetamine.

Lisdexamfetamine given as a flexible morning dose (progressing from 30, through 50, and then to 70 mg/day) provided no serious AEs occurred. Duration 6 weeks

Versus

Placebo

Identical placebo given in same incremental doses. Duration 6 weeks.

N=30 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-60); severity mixed or unclear.

Note: most participants have executive functioning impairment. Not downgraded for indirectness.

Fatigue (Fatigue Severity Scale)

Cognitive function (Behaviour Rating Inventory of Executive Function, BRIEF)

Psychological status (Hamilton Anxiety Scale)

Pain (McGill Pain Questionnaire)

Symptom scales (Clinical Global Improvement, severity)

Adverse events (headache, dry mouth, insomnia; discontinuation due to adverse events)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Modafinil – two dose arms

Modafinil (200mg)

Modafinil (400mg) dose increased slowly at 3 day intervals starting at 100mg until required dose reached. Duration: 20 days.

The two dose arms were pooled for analysis.

Versus

Placebo

Identical doses of placebo. Duration 20 days

N=14 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-70 years); severity mixed or unclear.

Quality of life (SF36)

Fatigue (Chalder physical and mental fatigue scales)

Adverse events

Conducted in United Kingdom.

Two intervention arms at different dose – pooled for analysis.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Antiviral drug – Acyclovir

Valganciclovir 900 mg (two 450 mg tablets) twice daily for 21 days followed by 900 mg once daily to complete 6 months.

Versus

Placebo

Identical appearing placebo 900 mg (two 450 mg tablets) twice daily for 21 days followed by 900 mg once daily to complete 6 months

N=30 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18+ years); severity mixed or unclear.

Inclusion criteria included suspected viral onset and elevated antibody titers.

Antibody titers were required to fit one of the following schema:

(i)

HHV-6 IgG ≥ 1:640, EBV VCA IgG ≥ 1:640, and EBV EA IgG ≥ 1:160 or

(ii)

HHV-6 IgG ≥ 1:320, EBV VCA IgG ≥ 1:1,280 and EBV EA IgG ≥ 1:160.

Fatigue (multidimensional fatigue inventory, MFI-20)

Adverse events (treatment-related adverse events)

Conducted in USA.

Other outcomes reported but insufficient information for analysis: general symptom scores, sleep, psychological status, cognitive function

PEM reporting: 96.7% of participants had PEM.

Very serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature and requirement for participants to have suspected viral onset and elevated antibody titers [original analysis]; serious population indirectness – requirement for participants to have suspected viral onset and elevated antibody titers [PEM reanalysis].

Antiviral drugs – IV Acyclovir

IV acyclovir (500mg per square metre) infused over a period of 60 minutes in 150ml of saline every 8 hrs for 7 days of hospitalisation

Versus

Placebo

IV placebo (500mg per square metre) infused over a period of 60 minutes in 150ml of saline every 8 hrs for 7 days of hospitalisation.

N=27 people with CFS, diagnosed according to CDC criteria 1988 (Holmes 1988)

Strata details: adults (mean age 34.1 years); severity mixed or unclear.

Psychological status (Profile of Mood States – 6 subscales)

Adverse events (reversible renal failure)

Activity levels (rest in hours/day)

Symptom scales (Wellness score)

Conducted in USA.

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness – Holmes 1988 criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

5-HT3 antagonists – Ondansetron

Ondansetron (8 mg tablets). 16mg per day in 2 doses. Duration 10 weeks.

Versus

Placebo

Identical placebo - 2 tablets taken per day.

Duration 10 weeks

N=67 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (range of mean age – 34.7 to 35.8 years); severity mixed or unclear.

Fatigue (Checklist Individual Strength (CIS) fatigue)

Adverse events (constipation & malaise)

Activity levels (Actometer)

Adverse events

Symptom scales (Sickness Impact Profile)

Conducted in Netherlands.

PEM reporting: the percentage of participants with PEM was not reported.Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Galantamine hydrobromide

Galantamine hydrobromide, 3 x 2.5mg per day or 3 x 5mg per day or 3 x 7.5mg per day or 3 x 10mg per day. Duration: 16 weeks.

Versus

Placebo

Placebo 3 x daily. Titration details not clear.

Duration 16 weeks

N=434 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata criteria: adults (18-65 years); severity mixed or unclear.

Conducted in USA.

Other outcomes reported but insufficient information for analysis: fatigue, cognitive function, and sleep quality – results reported narratively in clinical evidence table.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Galantamine hydrobromide

Galantamine hydrobromide 10mg 3x daily, reached by a schedule of escalating dosage.

Duration 8 weeks.

Versus

Placebo

Placebo 3 x daily. Duration 8 weeks.

N=49 people with CFS, not diagnosed according to a consensus-based set of criteria.

Strata criteria: adults (range of mean ages 43.44 to 44.46 years); severity mixed or unclear

Fatigue (on VAS)

Cognitive function (memory on VAS)

Pain (myalgia on VAS)

Sleep quality (sleep disturbance on VAS)

Return to school/work (work capacity/satisfaction on VAS)

Adverse events

Conducted in Iceland.

In placebo group an optional cross-over design was added to parallel group RCT design - patients could cross-over after 2 weeks if failed to improve or had symptoms worsening.PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – unclear criteria for diagnosis [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Antihistamines - Terfenadine.

60mg terfenadine twice daily. Duration 2 months

Versus

Placebo

Placebo twice daily. Duration 2 months.

Participants were permitted to take oral contraceptives, antibiotics, vitamins, aspirin, NSAIDS, beta adrenergic agonists; not permitted to take antihistamines, decongestants, tricyclic antidepressants or ENT anti-inflammatory agents.

N=30 people with CFS, diagnosed according to CDC criteria 1988 (Holmes 1988).

Strata details: adults (mean age 36.2 years); severity mixed or unclear.

Physical functioning (modified Medical Outcome Study Short Form – physical functioning)

Psychological status (modified Medical Outcome Study Short Form – mental health)

Conducted in USA.

PEM reporting: 82.1% of participants had “post-exertional fatigue (prolonged)” but as there was no additional information the committee was not satisfied that this symptom was PEM.

Serious population indirectness – Holmes 1988 criteria used; PEM not a compulsory feature [original analysis]; unclear if symptom described in study is PEM and <95% had this symptom [PEM reanalysis].

Pro-inflammatory cytokine antagonists - Anakinra.

Anakinra (Interleukin-1 receptor antagonist) 100mg subcutaneously per day. Each participant provided with a box containing 28 syringes and supplies of drug. Patients instructed by physician on how to administer. Daily alarm used to assist compliance, along with adherence monitoring.

Duration: 4 weeks

Versus

Placebo

Identical placebo given in identical doses intramuscularly. Duration: 4 weeks

N=50 people with CFS, diagnosed according to CDC criteria 1994 (Fukuda 1994).

Strata details: adults (18-59 years); severity mixed or unclear.

Mortality

Fatigue (CIS fatigue)

Physical functioning (SF36 physical function)

Psychological status (Symptom Checklist 90)

Pain (VAS maximum pain score)

Adverse events (AEs & withdrawal due to AEs)

Symptom scales (Sickness Impact Profile)

Conducted in Netherlands.

PEM reporting: the percentage of participants with PEM was not reported.

Serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature [original analysis]; percentage of participants with PEM unclear [PEM reanalysis].

Staphylococcus toxoid preparation, Staphypan Berna (SB). Composed of undefined extracts of 2 strains of staphylococci (S. aureus and S. epidermidis), and a preservative compound thiomersal. Injection given subcutaneously in gluteal region by nurse. Drug administered in increasing doses of 0.1ml, 0.2ml, 0.3ml, 0.4ml, 0.6ml, 0.8ml, 0.9ml, and 1.0ml weekly, followed by booster doses of 1.0ml every 4 weeks with final dose given week 24. Drug kept in 1ml ampoules and packed in boxes marked with patient numbers.

Versus

Sterile water. Injection given subcutaneously in gluteal region by nurse. Administered in increasing doses of 0.1ml, 0.2ml, 0.3ml, 0.4ml, 0.6ml, 0.8ml, 0.9ml, and 1.0ml weekly, followed by booster doses of 1.0ml every 4 weeks with final dose given week 24. Drug kept in 1ml ampoules and packed in boxes marked with patient numbers. Similar in colour to active treatment.

Both active substance and placebo caused slight local pain and reaction after injection.

N=100 people with CFS (and fibromyalgia), diagnosed according to CDC criteria 1994 (Fukuda 1994) (and ACR criteria for FM). Investigations prior to study entry included physical examination, vital signs and blood parameters.

Strata details: adults (age 18-65); severity mixed or unclear (according to global assessment of illness severity measured at baseline 17% were moderately ill, 70% markedly ill, 12% severely ill, 1% most extremely ill)

Pain (Visual analogue of pain scale)

General symptom scales (Clinical global assessment of change – observer rated; clinical global assessment of severity – observer rated)

Adverse events (most frequent side effects; clinical global assessment of side effects)

Conducted in Sweden.

Comprehensive Psychopathological Rating Scale (CPRS-15) - authors selected 15 items relevant to FM/CFS from original 65-item scale which covers a broad range of psychiatric illnesses. Does not seem to be validated subscale.

PEM reporting: the percentage of participants with PEM was not reported.

Very serious population indirectness – 1994 CDC criteria used; PEM not a compulsory feature and all participants also had fibromyalgia [original analysis]; percentage of participants with PEM unclear and all participants also had fibromyalgia [PEM reanalysis].