Article Text
Abstract
Introduction An earlier food survey showed dietary potassium deficiency in rheumatoid arthritis (RA).
Objective To evaluate an adjunct role of oral potassium to reduce joint pain in RA.
Methods 172 consenting eligible symptomatic patients (median duration 6.5 years) on standard care were randomised into an assessor blind, parallel efficacy, controlled, prospective, multiarm single-centre study (80% power, drug trial design) of 16 weeks duration—arm A (potassium-rich vegetarian diet), arm B (arm A plus novel potassium food supplement) and arm C (control, regular diet). Standard efficacy (American College of Rheumatology recommendation) and safety and diet intake (3-day recall) were assessed at monthly intervals (protocol). Standard soft-ware package (SPSS V.20) was used for statistical analysis; analysis of variance), Mann-Whitney statistic and χ2 test.; significant p<0.05, two sided). Study arms were found matched at baseline. Background RA medication remained stable. Preset target for increased potassium intake (India standards) were mostly achieved and participants remained normokalemic.
Results 155 patients (90.1%) completed the study and several showed improvement (maximum improved measures in arm B). Potassium intervention was safe and well tolerated. Adverse events were mild; none caused patient withdrawal. On comparison, the mean change in pain visual analogue scale (−2.23, 95% CI −2.99 to −1.48) at week 16 (primary efficacy) from baseline was significantly superior in arm B (per protocol analysis). A high daily potassium intake (5–7.5 g, arm B) was significantly associated with low pain (study completion); OR 2.5 (univariate analysis), likelihood ratio 2.9 (logistic regression). Compliance (intervention), diet record and analysis, RA medication and absence of placebo were potential confounders.
Conclusion High oral potassium intake, based on a suitable vegetarian diet and food supplement, reduced joint pain and improved RA. It was a safe adjunct to standard care, Further validation studies are required.
Trial registration CTRI/2022/03/040726; Clinical Trial Registry of India.
- Musculo-skeletal health
- Nutrient deficiencies
- Nutritional treatment
- Precision nutrition
Data availability statement
Data are available upon reasonable request. Adequate data are available in the manuscript text and the supplement. An application for access to additional data for academic non-commercial purpose will be considered by the first author Dr Toktam Kianifard on receiving an application with full credentials and employment details of the research applicant.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THE TOPIC
Potassium is a vital micronutrient in health, but little is known about its role in rheumatoid arthritis (RA).
Patients suffering from symptomatic RA may consume potassium-deficient diet.
WHAT THIS STUDY ADDS
Increased potassium intake based on a suitable vegetarian diet and a novel food supplement may reduce joint pain and improve RA.
High oral potassium intake was found safe and well tolerated.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
A high potassium intake based on vegetarian diet and food is a useful adjunct to manage difficult symptomatic RA in clinical practice.
The current recommendations/guidelines on potassium intake should be revised.
Introduction
Rheumatoid arthritis (RA) is a lifelong autoimmune disease characterised by painful polyarthritis and articular deformities, and systemic complications.1 Patients often suffer from poor quality of life and productivity and reduced longevity.1 The global burden is enormous and rheumatology care is often inadequate.2 Standard treatment guidelines focus on drugs and adjuvant therapies such as diet are neglected.3 In ancient medicinal systems, diet is pivotal in the management of arthritis.4 An evidence-based therapeutic role of diet in RA seems limited.5
We reported low potassium intake in RA.6 Although an essential micronutrient, its role in RA is not described.7 8 Weber9 advocated a strong aetiological link.9 A population nutrition survey showed low potassium intake and serum potassium in RA subjects.10 A low body potassium in RA was reported.11 High potassium chloride (mixed in grape juice) intake for 28 days reduced pain and improved RA.12
The current objective was to investigate oral potassium as an adjunct to standard RA treatment. The primary hypothesis was potassium-reduced joint pain in RA. Potassium-rich vegetarian diet and a novel potassium-enriched food supplement were used as active intervention.
Methods and patients
This was a non-commercial investigator- initiated study.13 The protocol was registered and adhered to the tenets of the ‘Helsinki Declaration’ (updated) and national guidelines.14 15 The period of recruitment was from 12 Feb 2014 to 02 July 2014. The last patient last visit was on 12 Dec 2014.
Study design
This was a randomised, single blind (assessor), active control, parallel efficacy, prospective drug trial study of 16 weeks duration. It was carried out in a community-based rheumatology centre (CRD). There were three treatment arms—arm A (potassium-rich vegetarian diet), arm B (arm-A diet plus potassium-enriched food supplement) and arm C (control routine diet). Each randomised participant was examined at baseline and monthly study time points till study completion.
Participants and selection
Volunteer patients from the outpatient clinic of CRD signed an informed consent and were randomised if found eligible on screening.15 Patients were required to have painful arthritis for at least 8 weeks and supervised standard care for at least 16 weeks prior to enrolment. Patients received a modest allowance for travel and meals. All study procedures and the potassium food supplement (KFS) were provided free of cost to the participants.
Inclusion criteria included (1) adult patient of RA16 (2) disease duration of at least 6 months (3) maximum pain (joints) ≥4 cm on a visual analogue scale (VAS) in the previous 24 hours (screening). Exclusion criteria included (1) daily prednisolone dose of 10 mg or more anytime during the previous 4 weeks (2) abnormal serum creatinine, blood urea nitrogen assay and serum potassium (≥5.5 mEq/L) assay (3) any medication known to effect body potassium.
Efficacy
The primary efficacy was an improvement in the pain VAS at week 16 (study completion) from baseline. There were several secondary efficacy measures of disease activity and function and safety.17
Procedures
Potassium food supplement
The entire process of manufacturing beginning from procurement of standard raw food ingredients (government accredited shops) to the final preparation of standardised mixture was carried out by the study nutritionist (TK, first author) using local resources. The composition, microbial and toxicology results and shelf life are described in online supplemental figure 2. The cost of manufacturing was mostly borne by TK and several related facilities and trained staff were provided free of cost by the CRD. No commercial/manufacturing company was engaged for this purpose. KFS is not available in the market in India or abroad. Several details are described in the patent granted by the Government of India.18
Supplemental material
One hundred and forty-two gram (one unit) of supplement contained green gram (V radiate, 25 g), cow pea (V unguiculata, 25 g), coriander seed (C sativum, 25 g), cumin seed (C cyminum, 25 g) and 42 g of oral rehydration salt (Indian pharmacopoeia, 3 g potassium chloride, 5.2 g sodium chloride, 5.8 g trisodium citrate, 27 g glucose). The latter contained 2638 mg elemental potassium (green gram 294.6 mg, cow pea 287 mg, coriander seed 247.4 mg, cumin seed 245 mg, oral rehydration salt 1564 mg).
Randomisation/enrolment
Study participants were provided a study information brochure in local language and suitably counselled by TK. A standard computer-generated randomisation schedule (1:1:1) was prepared by SS. Participants were randomised on first come first serve basis.
Diet food record
A validated questionnaire and standard household measures were used in a face-to-face interview with the participants by TK to record retrospectively a 3 successive day diet consumption at all study visits.6
Clinical assessment
This was carried out by designated rheumatologists and paramedics in a blinded manner.
A 68/66 joint count was used to assess pain/tenderness and swelling.17 The pain VAS was a 10 cm-long horizontal scale: 0 for nil pain and 10 for maximum pain. A Likert scale with five categories (asymptomatic to severe) was used to record physician and patient global assessment. A questionnaire-based RA pain score instrument recorded qualitative pain (range 0–144); higher score meant worst pain.19 The functionality was assessed by a validated modified Stanford health assessment questionnaire (HAQ, range 0–24); higher score indicated more difficulty.20 Short form 36 (SF 36), with permission from the vendor, assessed physical and mental health (quality of life); higher score meant better health.21 Improvement and disease activity were assessed by standard indices.22
Others
Routine investigations were carried out as per standard of care in RA and to ensure safety of potassium administration. Serum cortisol and spot urinary potassium assay were done.
Study intervention
A potassium-rich vegetarian diet and a novel KFS were active interventions (arm A and arm B). Routine diet was used as active control (arm C) and no placebo was used. All participants in the study were advised to continue their routine schedule of three meals per day. They were not to fast or consume diet other than that advised in the study.
Diet/supplement targetThe potassium-rich diet was to provide at least 3500 mg of elemental potassium daily based on the recommended daily allowance (RDA).23 24 The addition of KFS in arm B was meant to further increase the daily potassium intake to about 5.5 g.
Diet Brochure
A special diet brochure in the local language was prepared by TK and explained and provided to each participant in arms A and B. A copy is enclosed in online supplemental figure 1. It provided guidance on selection (and quantity) of food items (multiple choice) to prepare a meal for daily consumption by the participant. The meal was essentially a balanced vegetarian diet as per India recommendations.23 24 Community food habits, cost and availability were duly considered in the brochure. Non-vegetarian food items were discouraged.
Potassium food supplement (arm B)
The dose was three heaped tablespoon (provided) taken two times a day with a glass of water immediately after a meal. This amount corresponded to 1.7–2 g of elemental potassium. Each participant received a 5-week supply of KFS at baseline and every monthly follow-up visit.
Control diet
Participants in arm C continued the ongoing routine diet as per their preference
Compliance
Participants were to strictly follow the allocated diet and any other advise provided in the study. A telephonic remiinder was given every 10–14 days. The unused portion of the KFS was returned and measured at each follow-up visit. The participants were aware that their monthly urinary sample was assayed for potassium
Concurrent medication
Participants continued previous standard RA medication under supervision of a CRD rheumatologist. The medication was to be kept stable but if necessary a change could be made based on clinical judgement. The pain medication was to be used on a need basis for severe/intolerable pain. Other co-morbid disorders were treated by the primary care physician.
Statistical plan and analysis
There was no prior data available to guide sample size. As the primary objective was to evaluate an adjunct role, a modest effect size for pain relief by the therapeutic potassium intervention (arm B) was considered; 10% superior to the control.25 Based on the sample size tables in the latter publication, and an expected 20% drop out rate, 171 subjects were required (80% power, significant p<0.05, two-tailed). There were 57 participants in each study arm.
The daily diet data were analysed by TK in a blinded manner based on standard ‘Food Composition Tables’ (uncooked and cooked foods).23 24 The latter was adjusted for the KFS intake in Arm B. The dietary results at week 16 were used for efficacy and compliance.
An intention to treat (ITT, last observation carried forward) and per protocol (completer) analysis was carried out. Standard statistical software package (IBMSPSS V.20, V.2015 and 2018) was used; parametric (one way analysis of variance (ANOVA)), non-parametric (Mann-Whitney statistic, KW signed rank test) and χ2 test (categorical data) and Bonferroni’s correction for repeated measures. Unless stated, all p values in the current report pertain to ANOVA.
Though not intended for the current report, results of some regression models are shown in online supplemental tables 6–8.
Observations and results
One hundred and seventy-two patients were randomised and 155 (90.1%) completed the study (per protocol analysis) (figure 1). Seventeen (9.9%) patients were withdrawn prematurely but none was due to an AE.
The study arms were found matched at baseline that included RA disease activity measures. There was some difference in the use of methotrexate and sulfasalazine among the arms (table 1). Overall, the RA was moderately painful and active (table 5). Fifty-one patients (29.6%) recorded comorbid disorders and important being diabetes (10), hypertension (27), ischaemic heart disease (7), chronic acid-peptic disorders (34), haemorrhoids (7) and hypothyroidism (8); number of patients shown in parenthesis. None suffered from clinically apparent extra-articular complications (RA).
Diet
The results of dietary analysis and potassium intake are shown in tables 2 and 3. At baseline, the potassium intake was deficient in each of the study arms; Indian RDA is 3225 mg for women, and 3750 mg for men.24 However, the potassium intake was substantially increased in the active intervention arms at week 16 with a several fold rise in arm B (median 5648 mg, range 4365–7545 mg); 84% of participants in arm B consumed 5 g or more daily. Several other nutrients were also found increased in the active intervention arms; the measures in arm C seemed stable (table 2).
Less than 5% of participants recorded consumption of eggs, red meat and fish, or other Western style food such as pizza or French Fries.
Safety and tolerability
Adverse events (AE) were reported by 14 patients in arm A (table 4), 16 patients in arm B and 11 patients in arm C (p=0.67, χ2 statistic). They were mild and mostly related to abdominal complaints. Three patients in arm B required reduction in the daily dose of KFS for relief from mild to moderate abdominal pain/discomfort. Laboratory investigations (routine haematology, metabolic renal and hepatic) remained within normal range. All participants remained normokalemic at all study time points (online supplemental table 1); correspondingly electrocardiography showed normal pattern (table 4).
Efficacy
Primary efficacy: pain (VAS) was reduced at all predetermined study time points (figure 2). The mean change at week 16 from baseline was statistically significantly (p=0.039) as per protocol analysis but not by an intention to treat analysis (p=0.17) (table 5 and online supplemental table 2). The mean change in pain VAS in arm B (95% CI −2.99 to –1.48) was significantly superior to that in arm C (adjusted p=0.02) and arm A (adjusted p=0.04) (table 4).
Secondary efficacy: several variables improved to varying extent although it seemed numerically superior in the majority in arm B (table 4).
RA medication
The background RA medication seemed mostly stable (table 6, online supplemental file 1, table 3). However, there was a conspicuous increase in the number of patients being treated with methotrexate in arm C at week 16 (online supplemental table 3).
Other outcomes
Clinical
At baseline, the mean systolic/diastolic blood pressure (mm Hg) was 131/79 in arm A, 129/80 in arm B and 125/79 in arm C; correspondingly at week 16, it was 117/80, 116/80 and 126/82. The blood pressure was recorded as a routine procedure by the study nurse (sitting position).
Serum cortisol (morning) assay (online supplemental table 1): though not significantly different, the increase was maximum in arm B at week 16
Correlation (online supplemental table 4): potassium intake and pain VAS showed a significant inverse correlation (r=−0.19). Urine potassium assay did not correlate with oral potassium intake or serum potassium.
Univariate analysis (online supplemental table 6): daily potassium intake of 5 g or more (OR 3.14) was significantly associated with low pain VAS (≤ 4 cm).
Logistic (step forward) regression (online supplemental table 7): a daily potassium intake of 5 g or more (likelihood ratio 2.87) and methotrexate use (likelihood ratio 16.1) were significant predictors of low pain VAS when adjusted for several clinical, medication and diet-related variables.
Discussion
The reduction in joint pain was substantial and significantly superior in patients of chronic symptomatic RA on standard care who consumed high potassium (arm B, potassium-rich vegetarian diet plus a novel potassium-enriched food supplement) as compared with only potassium-rich vegetarian diet (arm A) or a routine diet (arm C) in this randomised, assessor blind, controlled study of 16-week duration. The reduction in pain in arm A and arm B was much more than what was reported as a minimal clinical important difference.26 Of 155 (90%) patients completed the study. Over 80% patients in arm B consumed 5.2 g or more of elemental potassium daily (range 4365–7545 mg). The AE was generally mild, and none led to a patient withdrawal.
Strengths and implication
The real to life management of RA and the dietary nature, safety and tolerability of the potassium intervention were the core strength. RA is a difficult to treat disorder and the medication is potentially toxic.1 2 Medicinal use of potassium is fraught with drug toxicity.8 We used diet and a food supplement to administer potassium in the current study as per recommendations.27 28 However, the potassium content of food supplements in Europe (500–1000 mg) and USA (<100 mg) seems inadequate to meet the daily requirements and needs to be revised.28 29 The physiological adjustment to high potassium intake is rapid in the healthy state.27–29 The baseline potassium intake in the current study was inadequate (table 2) and this has been previously reported both in RA and healthy populations.6 8 10 28 29 The current potassium intervention was based on preset targets (see methods) and the target for arm B was considered therapeutic by in-house expert consensus. Encouragingly, arm B showed substantial pain reduction. Importantly, all study participants remained normokalemic and further supported our contention of safety with diet based and enriched potassium intake (online supplemental table 1). The clinical benefit of higher potassium intake in painful inflammatory disorders such as RA should be further investigated.
Several unique problems in our setting such as poor socioeconomic status and difficult logistic complicate management of RA.2 19 Standard conventional disease-modifying anti-rheumatic drugs (DMARD) and steroids are preferred.1–3 Effective pain management is critical to any successful outcome. Despite standard of care in RA and in particular with modern potent biologic DMARDS, residual pain remains a vexing issue.1 3 30 On the other hand, despite impressive scientific evidence and universal community concerns, rheumatologists neglect diet in medical practise.2–4 6 31 32
Pain is a core issue in RA and a predictor of psychosocial health and disease outcome and is often found to be disproportionate to the clinical assessment.1 3 17 22 26 30 33 34 Pain VAS is a reliable and valid easy to perform measure.26 The substantial reduction in pain VAS in arm B was consistent with definite improvement in several other measures; significant improvement in function also shown in HAQ (table 5). In our opinion, the latter was likely to be associated with the increased potassium intake.
Diet interventional studies are difficult and often complicated by lack of adequate control.35 The current study was statistically designed. Chronic RA is a heterogenous disease. However, the study arms were well matched (table 1). The background RA medication remained almost unchanged (table 6, online supplemental file 1). Importantly, the compliance to dietary intervention seemed fair (tables 2 and 3).
The inadvertent lowering of mean systolic blood pressure, although modest, in the active potassium intervention arms (but not in the control arm) was consistent with the beneficial effect of potassium on blood pressure. In retrospect, this also meant that participants were compliant for active diet intervention. Cardiovascular morbidity and premature mortality are major concerns in RA.36 Several studies have reported the benefit of potassium on blood pressure and other cardiovascular function8 28 37
Limitations
Compliance to dietary intervention and protocol, RA medication bias, unblinded nature of intervention and difficulties in recording diet and analysis were of special concern in the current study. Despite varying potassium intake, there were no significant differences in the urinary potassium assay between the arms and this may be due to intense renal compensation (online supplemental table 1).8 37 Dietary recall can be problematic and ‘food composition tables’ may underestimate diet analysis.38 Participants in the current study were aware of their dietary intervention and this is likely to increase expectation of benefit (placebo effect). We did not use placebo in the current study. The current 16-week study period was considered sufficient to substantially mitigate a placebo effect. Clinical assessments were carried out in a blinded manner.
Erythrocyte sedimentation rate (ESR) and CRP (C reactive protein) are acute phase reactants and were found substantially increased at randomisation baseline (table 1). There was a significant reduction in ESR in group B, which was consistent with pain improvement, but the CRP response was lacklustre (table 5). This is rather unusual and we have no definite explanation. However, in personal experience, the author (AC) has noticed this uncommonly in patients of chronic symptomatic RA with systemic complications and on prolonged therapy with conventional DMARD and low-dose steroids. Such a discordance (ESR-CRP) has been reported in RA, lupus and certain other disorders.39 40
The background medication might have introduced some bias that was not recognised. But this seems unlikely in view of a comprehensive analysis of use of DMARD and steroids (table 6, online supplemental file 1, table 3). Interestingly, the overall consumption of analgesics in the current study was modest and there were no significant differences between the arms (online supplemental table 3). However, patients of chronic RA often do not report proper use of analgesics in the experience of the author (AC).
It is also possible that the current benefits of potassium intervention were confounded by the use of balanced vegetarian diet. Plant-based diets are reported useful in RA.41 Notably, although several nutrients were found increased in the active intervention arms (table 2), only potassium intake was identified as a significant predictor for low-pain VAS in the regression analysis (online supplemental table 7).
Other studies
Several elegant studies have reported benefit of mediterranean diet (MD) in RA, which was shown to be of modest nature in the cochrane analysis.42 43 To the best of our knowledge, although predominantly vegetarian the MD was not assessed for potassium and other nutrients.41
The benefits of potassium have been mostly reported in conjunction with vegetarian diet, which per se is supposedly anti-inflammatory and in diverse medical disorders such as osteoporosis, hypertension and cardiovascular disorders and dysbiosis. 8 29 44 45 The profound sarcopenia and general debility are common in chronic RA and likely to further contribute to body potassium deficiency and deleterious consequences.8 46 Modulation of potassium ion channels was reported to control pain and immune-mediated inflammation.8 47 However, much more research is warranted.
Mechanism of action
The precise mechanism of pain reduction by potassium in the current study is not known. Increased potassium intake (especially therapeutic intervention) has been reported to increase endogenous steroids, which are potent anti-inflammatory (analgesic) agents.8 9 12 The serum cortisol assay was conspicuously increased in arm B (online supplemental table 1). Other mechanisms may operate and include neurophysiologic effects and potassium ion channels.47 48
Vegetables and fruits contain non-chloride forms of potassium, which are reportedly more beneficial for bone strength.29
Conclusion
A higher oral potassium intake derived from a suitable diet and a novel food supplement improved joint pains considerably in a substantial number of symptomatic chronic RA patients who were on background standard rheumatology care in this 16-week randomised controlled study. Patients also improved on several other efficacy measures. Oral potassium was a safe adjunct, and higher intake (5–7.5 g/day) was well tolerated. A predominantly vegetarian diet with sufficient potassium intake should be advocated in the management of RA. Pending further validation, some patients with difficult RA may also benefit from a judicious use of potassium-enriched food supplement.
Data availability statement
Data are available upon reasonable request. Adequate data are available in the manuscript text and the supplement. An application for access to additional data for academic non-commercial purpose will be considered by the first author Dr Toktam Kianifard on receiving an application with full credentials and employment details of the research applicant.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by CRD Ethics Committee Approval 04/02/2014. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
Arthritis Research Care Foundation-Centre for Rheumatic Diseases CRD), Pune, India, provided generous material and logistic help for the study. Several colleagues (rheumatologist) assisted in the study and namely Dr Nachiket Kulkarni, Dr Naisar Nahar, and Dr Kiran Adams. Several administrative, nursing and paramedic staff of CRD Pune provided wholehearted assistance and participation. We remain indebted to our patients who volunteered and consented to participate in the study despite several personal and logistic hurdles.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors (1) Concept and design: TK, AC; (2) Protocol preparation: TK, AC, MS, SS, VA, AC; (3) Data analysis and statistical report: TK, SS, AC; (4) Preparation of first draft of manuscript: TK, AC; (5) finalisation of manuscript: TK, AC, MS, SS, VA, AC; (6) Response to referee comments: TK, AC; (7) The gurantor (AC) accepts full responsibility for the work and/or conduct of the study, had access to the data, and controlled the decision to publish.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors and was part of a self-funded PhD program of Dr TK in University of Pune, Pune, India. None of the authors received any funding to participate in the research study or for preparing the manuscript for publication. Limited logistic and material support was provided by the Arthritis Research Care Foundation- CRD Pune (India), which is a non-profit making registered charitable research society.
Competing interests No, there are no competing interests.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.