Abstract
Background and Aims
Chronic pain is a potentially disabling condition affecting one in three people through impaired physical function and quality of life. While the psychosocial impact of chronic pain is already well established, little is known about the potential biological consequences. Chronic pain may be associated with an increased prevalence of cardiovascular disease, an effect that has been demonstrated across a spectrum of chronic pain conditions including low back pain, pelvic pain, neuropathic pain and fibromyalgia. The aim of this study was to review and summarize the evidence for a link between chronic pain and cardiovascular disease. We sought to clarify the nature of the relationship by examining the basis for a dose-response gradient (whereby increasing pain severity would result in greater cardiovascular disease), and by evaluating the extent to which potentially confounding variables may contribute to this association.
Methods
Major electronic databases MEDLINE, EMBASE, Psychinfo, Cochrane, ProQuest and Web of Science were searched for articles reporting strengths of association between chronic pain (pain in one or more body regions, present for three months or longer) and cardiovascular outcomes (cardiovascular mortality, cardiac disease, and cerebrovascular disease). Meta-analysis was used to pool data analysing the association between chronic pain and the three principal cardiovascular outcomes. The impact of pain severity, and the role of potentially confounding variables were explored narratively.
Results
The searches generated 11,141 studies, of which 25 matched our inclusion criteria and were included in the review. Meta-analysis (of unadjusted study outcomes) demonstrated statistically significant associations between chronic pain and mortality from cardiovascular diseases: pooled odds ratio 1.20, (95% confidence intervals 1.05–1.36); chronic pain and cardiac disease: pooled odds ratio 1.73 (95% confidence intervals 1.42–2.04); and chronic pain and cerebrovascular disease: pooled odds ratio 1.81 (95% confidence intervals 1.51–2.10). The systematic review also found evidence supporting a dose-response relationship, with greater pain intensity and distribution producing a stronger association with cardiovascular outcomes.
All of the included studies were based on observational data with considerable variation in chronic pain taxonomy, methodology and study populations. The studies took an inconsistent and incomplete approach in their adjustment for potentially confounding variables, making it impossible to pool data after adjustments for confounding variables, so it cannot be concluded that these associations are causal.
Conclusions
Our review supports a possible dose-response type of association between chronic pain and cardiovascular disease, supported by a range of observational studies originating from different countries. Such research has so far failed to satisfactorily rule out that the association is due to confounding variables. What is now needed are further population based longitudinal studies that are designed to allow more robust exploration of a cause and effect relationship.
Implications
Given the high prevalence of chronic pain in developed and developing countries our results highlight a significant, but underpublicized, public health concern. Greater acknowledgement of the potentially harmful biological consequences of chronic pain may help to support regional, national and global initiatives aimed at reducing the burden of chronic pain.
1 Introduction
Claims of a causal link between back pain and cardiovascular disease appear in studies as far back as 1950 [1]. The pathological rationale for this was based on the idea that back pain is a consequence of atherosclerotic disease of the lumbar arteries [2]. However, more recent evidence points to an effect across a spectrum of chronic pain conditions including pelvic pain, neuropathic pain and fibromyalgia [3, 4,5].
Whilst chronic pain and cardiovascular disease share common risk factors (smoking, increasing age, depression), the chronic stress associated with pain may itself be a risk factor for heart disease. A causal association is lent weight to by findings that chronic pain is associated with dysfunction of the autonomic nervous system, the inflammatory system and endothelial processes; all of which play a part in the pathophysiology of cardiovascular disease [6, 7, 8, 9, 10, 11, 12].
The aim of this study is to review the evidence for a link between chronic pain and cardiovascular disease and, where possible, quantify the strength of this association by pooling data in a meta-analysis. We seek to clarify the nature of the relationship by examining the basis for a dose-response (whereby increasing pain severity would result in greater cardiovascular disease), and by evaluating the extent to which potentially confounding variables may contribute to this association.
2 Methods
A protocol for our review was devised in line with the Meta-analysis of Observation Studies in Epidemiology (MOOSE) guidelines [13] and registered on PROSPERO (CRD: 42013006585) [14].
Major electronic databases: MEDLINE (via Ovid), EMBASE (via Ovid), Psychinfo (via Ovid), Cochrane and Web of Science were searched from 1947 to December 5th 2015 for relevant studies. Examples of the search strategies are shown in Appendix A. Additional references were identified following inspection of the bibliographies of included articles. Field experts were approached directly, and parallel searches were performed though ProQuest (a web-based information suppository), in order to access and review ‘grey literature’ (i.e.; unpublished research produced outside the traditional commercial, or academic, publishing channels) that could be relevant to our searches.
Inclusion criteria were all relevant English language studies examining the relationship between chronic pain (defined as pain with a duration lasting greater than three months) and (i) cardiac disease (myocardial ischaemia or myocardial infarction), (ii) cerebrovascular disease, (iii) cardiovascular mortality, or (iv) any combination of the above.
Exclusions were: letters, case studies, and foreign language articles (although their respective bibliographies were used to ensure full coverage of the subject matter).
The shortlisted studies were subjected to quality appraisal and risk of bias assessment using the relevant version of the Critical Appraisal Skills Programme tool for cohort [15] and case–control studies [16], and a modified Risk of Bias tool developed to assess methodological quality and risk of bias in cross-sectional studies [17]. Each study was evaluated against criteria from the appropriate checklists by the principal author (AF), and then verified by one of the co-authors (SP and/or RL); any discrepancies were settled though arbitration and consensus opinion.
Data extraction was performed independently by two authors (AF and SA) using a pre-defined tool designed to summarise the following: the demographic information for each study; the nature of the variables compared; the outcome measures used; and the confounders that were accounted for during any regression analyses. Where possible, the raw counts of the measured cardiovascular outcomes amongst exposure groups (chronic pain) and control groups (no chronic pain) were retrieved to generate unadjusted odds ratios (OR) and confidence intervals (CI) that could then be incorporated into a meta-analysis. In cases where data was not immediately available from the publication at least one attempt was made to contact the corresponding authors for access to the original study data.
A meta-analysis was performed grouping studies with similar association variables and comparable outcome measures. A decision as to which pain phenotypes (number of pain sites, severity of pain) were amenable to meta-analysis was made post hoc although the plan to pool comparable data was outlined ‘a-priori’ in our study protocol.
Data were synthesised using StataSE version 13 for Mac [18], generating a pooled estimate for the association between the relevant variables. In order to account for anticipated heterogeneity between the studies a random effects model was used, thereby applying relatively less weight to larger studies [19]. A post hoc review of the studies included in the meta-analysis allowed for identification of potential sources of heterogeneity; the impact of these was quantified with the use of sensitivity analysis.
The data synthesis was supplemented with a narrative summary evaluating the impact of pain phenotype and the role of the confounding variables (any variables that study authors had elected to account for in their respective statistical models) on the association between chronic pain and cardiovascular outcomes.
3 Results
The searches generated 11,141 studies, of which 1155 were duplicates. After a review of titles and abstracts 31 full text articles were sourced for detailed appraisal against the inclusion criteria, following which a further six articles were excluded. The remaining 25 studies (12 cohorts studies, 9 cross-sectional studies and 4 case–control studies) were subjected to formal quality appraisal as outlined in the Methods section. The selection process has been summarised in Fig. 1. The defining characteristics of the included studies are set out in Table 1.
Defining characteristics of studies included in review.
| Article | Population description | Sample ‘n’; ‘n’ included (%) | Age (average) % Female | Pain Measures (Prevalence) | CV Measures (Prevalence) | Association unadjusted/adjusted for non-modifiable confounders only (95% CI) | Association adjusted for remaining confounders (95% CI) | Confounders accounted | Quality |
|---|---|---|---|---|---|---|---|---|---|
| Cross sectional studies | |||||||||
| Conway et al. [27] United Kingdom | Nationally representative population sample (from Scottish Family Health study) | 12881 | Not given Not given | CP (36.3%) Severe CP [GCPS III/IV] (6.1%) | Rose Angina CVD [Self reported cardiac disease + use of cardiac drugs] | CP and Rose Angina, OR 3.42 (2.98–3.92); CP and CVD, OR 1.46 (1.30–1.64). Severe CP and Rose Angina OR 4.4 (3.7–5.2) | Age, Gender | Low | |
| Cote et al. [39] Canada | Nationally representative population sample | 2184; 1131 (46.2%) | 20–69 53.4% | Severe neck pain [GCPS III/IV] (4.8%) | CVD [Disorders with moderate/severe impact on health] (4.1%) | Severe neck pain vs. Grade 0 neck pain and CVD, OR 5.66 (1.29–24.87) | Age, sex, education, headache, digestive disorders, LBP, smoking, vehicular collision with neck injury | Low | |
| Ha et al. [28] South Korea | Nationally representative population sample | 31705; 23632 (43.7%) | 20–89 56.1% | CLBP [LBP >3/12] (16.6%) | CVD [Cardiac infarction and angina] (2.1%) CVA (2.0%) | CLBP and CVD, OR M 2.09 (1.32–3.30) F 2.43 (1.64–3.60); CLBP and CVA, OR M 1.74 (1.02–2.97), F 2.14 (1.41–3.24) | CLBP and CVD, OR M 2.16 (1.34–3.49) F 2.26 (1.51–3.38); CLBP and CVA, OR M 1.53 (0.91–2.58), F 1.95 (1.28–2.97) | Age, income, education, BMI, exercise habits, occupation | Moderate |
| Makela et al. [30] Finland | Nationally representative population sample (from Min-Finland Health Study) | 8000; 7217 (90.2%) | >29 54.5% | Chronic neck pain [>3/12] (M 9.5% F 13.5%) | CVD [Abnormal clinical findings, [including HTN] (28.8%) | Chronic neck pain and CVD, OR 1.37 (1.16–1.61) | Chronic neck pain and CVD, OR 1.18 (0.98–1.42) | Age, sex, obesity, DM, smoking, physical stress at work, any mental disorder, other (various) chronic MSK pain, any respiratory disorder. | Low |
| Ohayon and Stingl [3] Germany | Residents randomly sampled from telephone directory | 3786; 3011 (79.5%) | 15–100 51.4% | CP exc NeP [pain >3 days a week for >3/12] (18.4%) CNeP (6.5%) | Heart disease CVA | CP (excluding NeP) and Heart disease OR 1.6 (1.0–2.4); NeP and Heart disease, OR 2.5 (1.5–4.2). CP (excluding NeP) and CVA OR 1.1 (0.6–2.0); NeP and CVA, OR 2.5 (1.3–4.7) | Age, Sex | Moderate | |
| Parsons et al. [41] United Kingdom | Adults randomly selected from 8 General practices in Scotland and England | 45994; 15288 (33.3%) | >24 (56) 57.0% | Any Pain [>1/day] Severe CP [GCPS IV] | Angina (3.5%) Heart attack (3.7%) | Severe CP vs. Grade 0 CP and Heart attack, OR 3.87 (2.58–5.80); Severe CP vs. Grade 0 CP and Angina, OR 3.57 (3.40–5.32) | Severe CP vs. Grade 0 CP and Heart attack, OR 2.47 (1.43–4.28); Severe CP vs. Grade 0 CP and Angina, OR 3.17 (1.71–5.85) | Age, sex, BMI, DM, HTN, smoking, activity, depression, alcohol consumption | Moderate |
| Ryan et al. [31] England | Nationally representative population sample | 8947; 5354 (59.8%) | >44 54.3% | Chronic MSK pain [ICD coded chronic MSK disease with current pain] (21.2%) | CVD [All ICD causes of circulatory disease, including HTN] | CMP and CVD >65 age, OR 2.39 (1.95–2.94); CMP and CVD 45–65 age, OR 1.71 (1.36–2.14) | CMP and CVD >65 age, OR 1.83 (1.45–2.30); CMP and CVD 45–65 age, OR 1.27 (0.98–1.66) | Age, gender, social status, BMI, smoking, subjective sedentary behaviour, anxiety, depression, diet, alcohol MSK medication, endocrine and metabolic conditions | Low |
| Svensson et al. [40] Sweden | National census of men aged 40–47 | 940; 720 (76.2%) | 40–47 0.0% | Severe CLBP [LBP occurring daily or >2x/week] (16%) | Angina (6.0%) Claudication (7.4%) | Prevalence ofangina if severe LBP 9.5% vs. no LBP 3.3% (p 0.05–0.1); Prevalence of claudication if LBP 16.2% vs. no LBP 3.6% (p <0.001) | Only claudication remained significantly associated with severe LBP after further adjustment | Smoking, physical activity, worry/tension, fatigue, perception of stress, shortness of breath on exertion | Low |
| Von Korff et al. [32] United States | Nationally representative population sample (from National comorbidity survey) | 9282; 5692 (61.3%) | >17 58.2% | Chronic spinal [back or neck] pain (19%) | Heart attack (4.0%) CVA (4.3%) | Heart attack if chronic spinal pain, OR 0.9 (0.6–1.4) vs. general population; Stroke if chronic spinal pain vs. general population, OR 1.5 (1.1–2.1) | Age, sex, education, ethnicity. | Low | |
| Case–control studies | |||||||||
| Ablinet al. [34] Israel | 93 patients post cardiac catheterization, 51 healthy volunteer controls | 144 | 42–83 54.8% | FM [ACR 1990] Point tenderness FIQ Score | CAD [Angiographic] Composite cardiac severity score [LVF and CAD] | Prevalence of FM if CAD (33.3%) vs. no CAD (6%) | Tenderness and CAD, r = 0.46 (p = 0.019); FIQ and CAD, r = 0.354 (p = 0.005). Composite cardiac score and FIQ/tenderness, r = 0.354(p 0.005) | Age, gender, BMI, DM, HTN, smoking, angiotensin converting enzyme inhibitor, insulin and diuretic use | Moderate |
| Cho et al. [35] Korea | Females with FM from specialist clinics, 20 age-matched healthy controls | 50 | 42–66 (48) 100.0% | FM [ACR 1990] FIQS core | Global LV Strain [On 2 Dimensional echocardiography] | Measure of LV strain if FIQ>50 vs. FIQ<50 vs. no FM, −18.6 vs. −22.7 vs.−22.82 (±3.09,1.46, 9.62) | LV strain and FIQ >50, r = 0.68 (p<0.01) | None. However, controls were matched for age, gender and depression | Moderate |
| Kelleret al. [29] Taiwan | 9269 patients diagnosed with BPS/ICC 46345 matched controls (from Insurance bank) | 55614 | >17 (50) 81.8% | BPS/ICC [ICD coded] | IHD Stroke | BPS/IC and IHD, OR 1.85 (1.61–2.23); BPS/IC and Stroke, OR 1.68 (1.57–1.79) | Sex, age group, monthly income and geographic region | Moderate | |
| Pontari et al. [4] United States | 463 men with CPPS/CP and 121 asymptomatic age-matched controls | 584 | 20–88 (43) 0.0% | CPPS/CP [Pain in the pelvic region >3/12] | CVD [Self report, raised cholesterol, CAD, arrhythmia, HTN] | CP/CPPS and CVD, OR 7.15 (p = 0.009) | Income and psychiatric disease. However, controls were also matched for age, gender, smoking, education and employment | Moderate | |
| Cohort studies Andersson [23] Sweden | Random population sample from two municipalities in Sweden | 1812; 1609 (88.8%) | 25–74 50.4% | CP [>3/12] (55.2%) CWP [Pain in >4 distinct body areas >3/12] (9.4%) | CV deaths [ICD coded] over 13.5 years | CRP and CVD, HR 1.27 (0.76–2.10), CWP and CVD, HR 2.17 (1.12–4.21) | CWP and CVD, HR no longer significant after further adjustment (results not shown) | Age, gender, education, living alone, club membership, employment, smoking, physical activity, BMI, insomnia, anxiety and perception of stress | Moderate |
| Dreyer et al. [24] Denmark | Patients referred to Copenhagen Hospital rheumatology service | 1361; 1269 (83.2%) | >18 100.0% | FM [ACR 1990] | Death from IHD and CVA [ICD coded] over an average 3.9 years | Deaths from IHD too few to allow analysis. Deaths from CVA if FM; SMR 3.8 (1.2–8.8)times higher than expected for Danish women | None | Moderate | |
| Heliovaara et al. [22] Finland | Nationally representative population sample | 8000; 7217(90.2%) | >28 54.5% | CLBP [Pain for 3/12 + signs on clinical examination] (17%) | Any CV mortality or Coronary mortality [ICD 8 coded] over 12–14 years | CV Mortality if CLBP, RR 1.0 (0.8–1.2); Coronary Mortality if CLBP, RR 1.0 (0.8–1.3) | Age, sex, smoking | Moderate | |
| Kadam et al. [33] United Kingdom | Patients registered with a semi-urban general practice in North-East England | 3968; 2606 (57.9%) | 18–75 53.4% | CRP [<4 areas, for >3/12] (25.9%) CWP [ACR 1990](8%) | CVD [from primary care codes, may include HTN] over 3 years (14.3%) | CRP and CVD, RR 1.0 (0.8–1.4); CWP and CVD, RR 1.65(1.1–2.4) | Age, sex, social deprivation, anxiety and depression | Moderate | |
| Lindgren and Bergman [36] Sweden | Nationally representative population sample | 3928; 2425 (58.0%) | 20–74 (47) 53.2% | CRP [<4 areas, for >3/12] (24.5%) CWP [ACR 1990] (12.5%) | CVD [ICD 10 codes for stroke and IHD] for 8.9 years (6.3%) | CRP and CVD, OR 1.6 (1.0–2.4); CWP and CVD, OR 1.9 (1.2–3.1) | Age, sex, SE status, smoking, follow-up time | Moderate | |
| Smith et al. [21] United Kingdom | Nationally representative population sample (from Contraception Study) | 11797; 10073 (85.4%) | (56) 1 100.0% | CP [>3/12] (38.4%) | CV mortality [ICD 8 coded, including mortality from HTN] over 7 years (1.1%) | CP and CV Mortality, OR 0.95 (0.63–1.44) | Age, social class, smoking | Moderate | |
| Su et al. [38] Taiwan | 61,612 patients with FM and 184,834 matched controls (from insurance database) | 246446 | (44) 1 59.30% | FM [ICD 9 codes for Myositis/Myalgia + treatment >3/12] | CAD: Acute coronary syndrome, angina, IHD, and coronary atherosclerosis | FM and CAD, IRR of 1.64 (95% CI 1.61–1.68) (age and sex matched) | FM and CAD, HR of 1.47 (95% CI 1.43–1.51) | Age, gender, occupation, monthly income, “comorbidities of traditional cardiovascular risk factors”, depression, and anxiety | Moderate |
| Torrance et al. [20] Scotland | Patients from 29 general practices in Scotland (50% selected based on analgesic prescription) | 6940; 5855 (84.3%) | >24 52.7% | CP [>3/12] (12.1–36.3%) Severe CP [GCPS III/IV] | All circulatory system deaths [ICD 10 coded] over ten years | CP and Circulatory system mortality, HR 1.26 (1.0–1.59); Severe CP and Circulatory system mortality, HR 2.11 (1.59–2.80) | CP and Circulatory system mortality, HR 0.86 (0.65–1.14); Severe CP and Circulatory system mortality, HR 1.68 (1.20–2.35) | Age, sex, education, housing, long term illnesses | Moderate |
| Zhu et al. [25] Australia | Elderly Australian women (from national Osteoporosis study) | 1484 | 70–85 100.0% | Frequent LBP [daily pain over previous year] (21.8%) | Coronary events [death/new diagnosis of IHD or angina] over 5 years (8.4%) | Frequent LBP and coronary heart events, HR 2.13 (1.35–3.34) when compared to infrequent LBP. Adjusted for Age | Severe LBP and coronary heart events, HR 2.13 (1.32–3.44) when compared to infrequent LBP | Age, BMI, DM, HTN, smoking, analgesic use, raised cholesterol, CVD | Moderate |
| Matched cohort studies | |||||||||
| Chen et al. [26] Taiwan | 1291 females with BPS/IC, and 3760 matched controls | 4512 | >17 (47) 100.0% | BPS/ICC [ICD coded] | CVD [ICD coded for IHDx and unspecified CVD] over 3 years | BPS/IC and CVD, HR 2.28 (1.58–3.27) | BPS/IC and CVD, HR 1.65 (1.09–2.48) | Age, income, DM, HTN, High cholesterol, geographical region, chronic kidney disease, urinary tract infection, bladder outlet obstruction, (n) physician visits | Moderate |
| Mease et al. [37] United States | 764786 FM patients and 707721 matched controls (from insurance register) | 1472507 | Not given Not given | FM medicated FM unmedicated | CVD TIA CAD | FM medicated vs. no FM and CVD, IRR 1.19 (p < 0.001); FM untreated vs. no FM and CVD IRR 1.35 (p < 0.001). FM medicated vs. FM unmedicated and CVD, IRR 2.02 (p < 0.001) | Risks remained significant after adjustment for confounders | Not clear. However, control group were matched for age and gender | Low |
| Tsai et al. [5] Taiwan | 3420 with FM and 10260 matched controls (from insurance database) | 13680 | >19 55.9% | FM [ICD 9 codes for Myositis/ Myalgia + treatment >3/12] | CAD events [new PCI and CABG procedures] over 8.6 years (1.2%) | FM and CAD, HR 2.11 (1.48–3.00) | FM and CAD, HR 1.97 (1.36–2.85) | Age, sex, obesity, DM, HTN, anxiety, depression, NSAIDS, opioids, CV drugs, number of cardiology visits | Moderate |
-
Abbreviations: American college of Rheumatology (ACR); Bladder pain syndrome/interstitialcystitis(BPS/IC); Body mass index (BMI); Cardiovascular (CV); Cardiovascular disease (CVD); Cerebrovascular accident (CVA); Confidence intervals (CI); Coronary artery disease (CAD); Chronic pain (CP); Chronic regional pain (CRP); Chronic widespread pain (CWP); Diabetes Mellitus (DM); Fibromyalgia (FM); Fibromyalgia impact questionnaire (FIQ); Grading chronic pain Scale (GCPS); Hazard ration (HR); Hypertension (HTN); Incidence risk ratio (IRR); International classification of diseases (ICD); Ischaemic heart disease (IHD); Left ventricular failure (LVF); Musculoskeletal (MSK); Myocardial infarction (MI); Non-Steroidal anti-inflammatory drugs (NSAIDS); Odds ration (OR); Osteoarthritis (OA); Percutaneous coronary intervention (PCI); Relative risk (RR); Standardized mortality ration (SMR).
Flow diagram detailing article selection.
3.1 Chronic pain and cardiovascular mortality: (total ‘n’ = 23,148) [20, 21, 22]
One study reported a borderline significant association between chronic pain and circulatory system deaths, using unadjusted data from a population of primary care patients in Scotland: Hazard Ratio (HR) 1.26, 95% confidence Intervals (CI) 1.0’1.59 [20]. However, the relationship became statistically insignificant following the addition of potentially confounding variables in the analysis: HR 0.86, (95% CI 0.65–1.14). Two remaining studies did not present crude outcomes, but reported statistically insignificant associations between chronic pain and mortality from cardiovascular diseases after adjustment for potentially confounding variables; Odds ratio (OR) 0.95 (95% CI 0.63–1.44) from a cohort of women enrolled in a National Contraceptive Study in Great Britain [21]; relative risk (RR) 1.0 (95% CI 0.8–1.2) amongst participants from a nationally representative sample from the population of Finland [22].
A meta-analysis, using the unadjusted or least adjusted outcomes from these studies, demonstrates a statistically significant association between chronic pain and mortality from cardiovascular diseases (Fig. 2): pooled OR 1.20 (95% CI 1.05–1.36), I2 3.7% (p = 0.35). A sensitivity analysis was performed removing one study that had made adjustments for confounding variables [21] (age, smoking and social class): the pooled estimate [OR 1.25 (95% CI 1.08–1.41), I2 0% (p = 0.58)] remained significant and fairly similar to the initial results.
Forest plot forthe association between chronic pain and cardiovascular mortality (n = 23,148).
3.2 Chronic widespread pain and cardiovascular mortality (total ‘n’ = 2878) [23, 24]
A Swedish population study reported increased mortality from cardiovascular causes amongst respondents with chronic widespread pain compared to those without pain, adjusted for age and sex, HR 2.17 (95% CI 1.12 –4.21), an association that lost statistical significance after further adjustment for life-style confounding variables including smoking and physical activity (adjusted HR not reported) [23]. A Danish study comparing Fibromyalgia patients from a Rheumatology clinic and the general population reported significantly increased mortality from cerebrovascular accidents in the pain group; Standardized mortality ratios (SMR) 3.8 (95% CI 1.2–8.8) [24]. However, the authors did not adjust the statistical models to account for the influence of potentially confounding variables.
3.3 Chronic pain severity and cardiovascular mortality (total ‘n’ = 7035) [20, 25]
One study measured the impact of pain severity, qualified using the Grading of Chronic Pain Scale (GCPS), on the association with cardiovascular mortality. The authors demonstrated that severe pain (GCPS grades III–IV) remained significantly associated with mortality from cardiovascular diseases even after correction for confounding variables HR 1.68 (95% CI 1.20–2.35) [20]. Another study demonstrated that an increased frequency of symptoms amongst Australian women reporting chronic lower back pain was associated with an increase in the occurrence of coronary events (including death from coronary heart disease) over a 5-year followup period; HR for daily back pain 2.13 (1.32–3.44) when compared to infrequent/no back pain, after adjustments for confounding variables [25].
3.4 Chronic pain and cardiovascular diseases (total ‘n’ = 118,497) [3, 4, 26, 27, 28, 29, 30, 31, 32]
Almost all of the studies reporting associations between chronic pain and cardiac disease (n = 7), cerebrovascular disease (n = 4) or both combined (n = 3) demonstrated statistically significant associations between the two respective variables [3, 4, 26, 27, 28, 29, 30, 31, 32]. Effect sizes ranged from OR 1.37 (95% Cl 1.16-1.61) for the association between chronic neck pain and cardiovascular disease amongst a nationally representative sample of the population of Finland, to OR 7.15 (p = 0.009) for the association between chronic pelvic pain syndrome and self-reported cardiovascular diseases amongst a cohort of men from the United States [4, 30]. Following the introduction of confounding variables into the analyses, statistical significance was lost entirely in only one study [30], and in subsets of participants in three other studies (those aged below 65 in one study, male participants reporting stroke in another study and participants reporting cardiac disease but not stroke in a third study) [28, 31, 32]. Full details of the study outcomes are presented in Table 1.
Two studies that used matched controls in their analysis [4, 26] were excluded from pooling, as it was not possible to reliably generate comparable outcome measures for comparison. A metaanalysis using the least adjusted outcomes from the remaining studies demonstrates a statistically significant association between chronic pain and cardiac disease: pooled OR 1.73 (95% Cl 1.42–2.04), I2 87.0% (p <0.001) as well as cerebrovascular disease: pooled OR 1.81 (95% Cl 1.51–2.10), I2 83.3% (p < 0.001) (Figs. 3 and 4).
Forest plot for the association between chronic pain and cardiac disease (n= 116,007).
Forest plot forthe association between chronic pain and cerebrovasculardisease (n=103,126).
In most studies hypertension was adjusted for as a potentially confounding variable. However, in three studies hypertension was measured as a primary cardiovascular outcome (thereby contributing to ‘cardiovascular disease’ burden) [30, 31, 33]. A sensitivity analysis removing these studies did not affect the significance of the pooled outcomes: cardiac disease pooled OR 1.75 (95% CI 1.30–2.21), I2 87.7% (p < 0.001); cerebrovascular disease pooled OR 1.94 (95% CI 1.39–2.50), I2 79.8% (p = 0.002).
3.5 Chronic widespread pain and cardiovascular diseases (total ‘n’ = 1,737,858) [5, 33, 34, 35, 36, 37, 38]
All seven studies demonstrated statistically significant associations between chronic widespread pain or fibromyalgia and cardiovascular outcomes, before and after correction for potentially confounding variables [5, 33, 34, 35, 36, 37, 38]. In two studies participants reporting chronic widespread pain and chronic regional pain were compared independently against pain free cohorts [33, 36]. In both instances the outcome measure was greater for the widespread pain cohort than for the regional pain cohorts, and in one study statistical significance was lost for the regional pain cohort (but not the widespread pain cohort) after adjustments were made for confounding variables [33].
3.6 Chronic pain severity and cardiovascular diseases (total ‘n’ = 1,506,715) [3, 25, 27, 37, 39, 40, 41]
Various methods were used to quantify the severity of pain phenotypes: higher Von Korff pain grade [27, 39], presence of neuropathic symptoms [3], scores on numerical pain rating scales [41], frequency of symptoms [25, 40], and qualification based on medication taken for symptoms [37]. In all instances the strength of association was greater in the ‘more severe’ chronic pain group, although in one instance neither outcome was statistically significant after correction for confounding variables [40]. The relevant study results are summarised in Table 2.
Summary of severity measures used.
| Study | Severity measure | High severity outcome measure (95% Cl) | Low severity outcome measure (95% Cl) | Control group |
|---|---|---|---|---|
| Conway | Von Korff Pain Grade (I–IV) | Grades III–IV CP and Rose Angina OR 4.4 (3.7–5.2) | All CP and Rose Angina, OR 3.42 (2.98–3.92) | No chronic pain |
| Cote | Von Korff Pain Grade (I–IV) | Grades III–IV chronic neck pain and CVD, OR 5.66 (1.29–24.87) | Grades II chronic neck pain and CVD 1.21 (0.33–4.47) | No chronic pain |
| Mease | Medicated FM vs. non medicated FM | FM subjects taking medication and CVD, lRR 2.02 (p < 0.001) | N/A | Unmedicated FM subjects |
| Ohayon | Neuropathic pain vs. non neuropathic pain | NeP and Heart disease, OR 2.5 (1.5–4.2); NeP and CVA, OR 2.5 (1.3–4.7) | Non NeP CP and Heart disease OR 1.6 (1.0–2.4); Non NeP CP and CVA OR 1.1 (0.6–2.0) | No chronic pain |
| Parsons | Numeric pain score (1–5) | Chronic pain score 5/5 and Heart attack, OR 3.87 (2.58–5.80); or Angina, OR 3.57 (3.40–5.32) | Chronic pain score 3/5 and Heart attack, OR 1.50 (0.95, 2.38); or Angina, OR 2.01 (1.30–3.09) | Chronic pain score 1/5 |
| Svensson | Frequency of pain symptoms | Prevalence of angina if weekly LBP 9.5% | Prevalence of angina if monthly LBP 8.5% | no LBP 3.3% |
| Zhu | Frequency of pain symptoms | Daily LBP and coronary heart events, HR 2.13 (1.35–3.34) | Monthly LBP and coronary heart events, HR 1.78 (1.15–2.75) | Infrequent LBP (<1/month) |
3.7 Exclusions [42, 43, 44, 45, 46, 47]
Three studies were excluded from the review because the duration of pain symptoms did not meet the criteria for chronicity: pain in the past month lasting greater than one week [42]; pain in the past month lasting a day or longer [43]; or any pain, ache or tenderness on movement in one or more joints during the last month [44]. One study (an Abstract) reported data already included in the review under a different first author [45]. Two studies, addressing risk of cardiovascular diseases amongst participants meeting the American College of Rheumatology (ACR) definition for fibromyalgia, were excluded as both the fibromyalgia and the control groups were deemed to have a similar prevalence of non-specific chronic pain conditions [46, 47].
3.8 Quality
A breakdown of the quality scoring for each study is included under Appendix B. None of the included studies were considered to be ‘High’ Quality, although in general the longitudinal (cohort and case control) studies fared better than the cross-sectional/population studies. The main weakness across both groups of studies was the inadequate inclusion of potentially confounding variables in the statistical modelling, with only two studies including general analgesic use as a potential confounding risk factor [25, 31] and only one study specifically addressing the impact of non-steroidal anti inflammatory drugs (NSAIDs) [5], as demonstrated in Table 3.
Summary of confounding variables accounted for in the review papers.
| Authors | Age | Sex | Status | Anxiety | Depression | Smoking | BMI | Diabetes | HTN | Mobility | NSAIDS | Quality | Study Showed a link? |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ablin | Yes | Yes | No | No | No | Yes | Yes | Yes | Yes | No | No | Moderate | Yes |
| Andersson | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No | Yes | No | Moderate | Yes, with confounding |
| Chen | Yes | N/A | Yes | No | No | No | No | Yes | Yes | No | No | Moderate | Yes |
| Cho | M | M | No | No | M | No | No | No | No | No | No | Moderate | Yes |
| Conway | Yes | Yes | No | No | No | No | No | No | No | No | No | Low | Yes |
| Cote | Yes | Yes | Yes | No | Yes | No | No | No | No | No | No | Low | Yes |
| Dreyer | No | No | No | No | No | No | No | No | No | No | No | Moderate | Yes |
| Ha | Yes | No | Yes | No | No | No | Yes | No | No | Yes | No | Moderate | Yes |
| Heliovaara | Yes | Yes | No | No | No | Yes | No | No | No | No | No | Moderate | No |
| Kadam | Yes | Yes | Yes | Yes | Yes | No | No | No | No | No | No | Moderate | Yes |
| Keller | Yes | Yes | Yes | No | No | No | No | No | No | No | No | Moderate | Yes |
| Lindgren | Yes | Yes | Yes | No | No | Yes | No | No | No | No | No | Moderate | Yes |
| Makela | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | No | No | Low | Yes, with confounding |
| Mease | M | M | No | No | No | No | No | No | No | No | No | Low | Yes |
| Ohayon | Yes | Yes | No | No | No | No | No | No | No | No | No | Moderate | Yes |
| Parsons | Yes | Yes | No | No | Yes | Yes | Yes | Yes | Yes | Yes | No | Moderate | Yes |
| Pontari | M | M | Yes | Yes | No | M | No | No | No | No | No | Moderate | Yes |
| Ryan | Yes | Yes | Yes | Yes | Yes | Yes | Yes | Yes | No | Yes | Yes | Low | Yes |
| Smith | Yes | N/A | Yes | No | No | Yes | No | No | No | No | No | Moderate | No |
| Su | Yes | Yes | Yes | Yes | Yes | No | No | No | No | No | No | Moderate | Yes |
| Svensson | No | N/A | Yes | Yes | No | Yes | No | No | No | Yes | No | Low | Yes, with confounding |
| Torrance | Yes | Yes | Yes | No | No | No | No | No | No | No | No | Moderate | Yes, with confounding |
| Tsai | M | M | No | M | M | No | No | M | M | No | M | Moderate | Yes |
| Von Korff | Yes | Yes | Yes | No | No | No | No | No | No | No | No | Low | Yes, with confounding |
| Zhu | Yes | N/A | Yes | No | No | Yes | Yes | Yes | Yes | Yes | Yes | Moderate | Yes |
-
M, matched control groups; N/A, not applicable.
Abbreviation: BMI (body mass index), HTN (hypertension), NSAIDS (non-steroidal anti-inflammatory drugs.)
4 Discussion
4.1 Summary of results
We identified 25 studies that explore the relationship between chronic pain and cardiovascular disease and/or cardiovascular mortality. Pooling of available data has demonstrated statistically significant associations between chronic pain and mortality from cardiovascular diseases, pooled OR 1.21 (95% Cl 1.05–1.36); chronic pain and cardiac disease, OR 1.77 (95% Cl 1.47–2.07); and chronic pain and cerebrovascular disease, OR 1.81 (95% Cl 1.51–2.10).
4.2 Pathophysiological mechanisms
There are theoretical models that could explain an independent association between chronic pain and cardiovascular disease. The biological response to acute pain includes activation of the sympathetic nervous system which may contribute to a (protective) reduction in pain sensitivity that has been demonstrated both clinically and experimentally [6]. The balance of this relationship appears to reverse as pain shifts from acute to chronic, with persistent and maladaptive responses of baroreceptors to ascending nociceptive signals leading to sustained increases in blood pressure and greater cardiovascular morbidity and mortality [7]. Although in humans, for whom the development of hypertension is multifactorial, the evidence for alterations in blood pressure due to chronic pain remain epidemiological, with little support from clinical studies.
4.3 Dose-response
The biological credibility of a model in which chronic pain predisposes to cardiovascular disease through stress or inflammation would be strengthened by evidence of a dose-response relationship. In our analyses, all associations between chronic pain phenotypes and cardiovascular outcomes (cardiac disease, cerebrovascular disease and cardiovascular mortality) appeared to be stronger (i.e. larger effect size), and the associations in these subgroups appeared to be more independent of confounding effects, when subgroups with increasing pain distribution and/or increasing pain severity were selectively analysed.
4.4 Strengths
A recent meta-analysis demonstrated increases in cardiovascular related deaths amongst populations with chronic pain (pooled OR 1.09, 95% CI 0.84–1.41) and chronic widespread pain (pooled OR 1.17, 95% CI 0.89–1.63) that were not statistically significant [48]. However, the authors included three studies (of five in the meta-analysis) where pain symptoms in the exposure groups were not exclusively ‘chronic’, and would likely bias the pooled results towards the Null Hypothesis. We have constructed our review in accordance with guidance published by the Meta-analysis of Observation Studies in Epidemiology (MOOSE) consortium [13], using standardized definitions for chronic pain phenotypes that had been agreed (a priori) in an attempt to limit bias that may arise during retrieval, analysis and interpretation of available evidence. To our knowledge this is the first review article that attempts to rigorously address the impact that the chronicity of pain symptoms may have on the risk of cardiovascular disease and cardiovascular mortality.
4.5 Limitations
All the data collated in our study were extracted from observational studies, as the hypothesis under question does not lend itself to analysis through the more robust methodology of randomised controlled trials. We are therefore restricted in our ability to draw firm conclusions regarding causality and directionality between the two variables. Studies varied markedly in design and methodology, often using non-standardised definitions of chronic pain and widespread pain, and selecting cardiovascular outcomes that were not always comparable. For example, some studies included hypertensive disease as a cardiovascular outcome [4, 21, 30, 31, 33] while others included hypertension as a confounding variable [25, 26, 34, 41]. These limitations are reflected in our quality appraisal and risk-of-bias assessment for both longitudinal and cross-sectional studies in our review. Through a search of the literature we have identified 11 variables that may be independently associated with chronic pain and cardiovascular disease: age, gender, socio-economic status, smoking, body mass index, depression, anxiety, mobility, hypertension, diabetes and use of non-steroidal anti-inflammatory drugs (NSAID) [49, 50, 51]. Cooccurrence of one or more of these risk factors may account for part of the associations identified in our meta-analysis of unadjusted outcomes. While most of the authors acknowledged the potential for confounding in their analyses there was little consistency with regards to the choice of variables included in the statistical models, with none of the authors attempting to adjust for all 11 variables outline above, and only three studies acknowledging the potential role of NSAIDs. This heterogeneity has made it impossible to perform a meaningful pooling of data for comparison after adjustment for confounders. It is likely that confounding plays a significant role in the association between chronic pain and mortality from cardiovascular disease, with all three of the reported associations losing statistical significance after adjustment [20, 21, 22]. Studies addressing the risk of cardiovascular morbidity (cardiac and cerebrovascular disease) were less affected by confounding, providing a firmer basis for claims of an independent link between these variables and chronic pain, and studies focussing on subsets of participants with severe or widespread chronic pain were the most likely to demonstrate a statistically significant association after adjustment.
4.6 Future work
What are now needed are well-designed longitudinal studies, using standardised definitions of chronic pain and chronic widespread pain, allowing for more robust exploration of a cause and effect relationship. Demonstration of dose-response gradients (for associations based purely on observational data) are metrics that will serve to increase the validity of study outcomes, as outlined by the “Recommendation, Assessment, Development, and Evaluation” (GRADE) Working Group [52]; futures studies should strive to include pain-severity measure that allow such comparisons. Further research in support of a causal link between chronic pain and cardiovascular diseases should also be powered to account for the role of all of the potentially confounding variables outlined above, in particular to assess the impact of NSAIDs. Demonstration of an association independent of NSAID use would call for re-evaluation of the risk-benefit of these drugs, which are often excluded from chronic pain treatment pathways on the basis of their potentially unfavourable cardiovascular and cerebrovascular effect profiles.
5 Conclusions
Our review supports a possible dose-response type of association between chronic pain and cardiovascular disease, supported by a range of observational studies originating from different countries. Such research has, so far, failed to satisfactorily rule out that the association is due to confounding variables. There is mounting evidence that chronic stress or inflammation may provide a biologically plausible pathway from pain to cardiovascular damage. What is now needed are further population based longitudinal studies that are designed to allow more robust exploration of a cause and effect relationship.
Highlights
Chronic pain is associated with an increased occurrence of cardiovascular disease.
Increasing pain severity produces a stronger association with cardiovascular outcomes.
The extent to which these associations are due to confounding variables remains uncertain.
DOI of refers to article: http://dx.doi.org/10.1016/j.sjpain.2016.09.005.
-
Conflicts of interest: Dr. Langford reports personal fees from Grunenthal, grants and personal fees from Napp/Mundipharma, personal fees from Pfizer, personal fees from Astrazeneca, personal fees from BioQuiddity, personal fees from The Medicines Co, all outside the submitted work. No other conflicts of interests declared from remaining authors.
-
Funding sources: None declared.
-
Perspective: This review provides further support for an association between chronic pain and cardiovascular disease that appears to be dose-dependent, but available data does not yet allow us to conclude that these associations are causal. These findings highlight the potentially harmful biological consequences of chronic pain.
-
Author contributions: Dr Fayaz was the principal author of the article, but all remaining authors contributed to the design, write-up and review of the paper. All authors have read and approved the paper in the format submitted for publication.
Acknowledgements
Thanks are expressed to Tim Reeves, Research support Librarian at Imperial College for assisting with the formulation of the preliminary search strategies.
References
[1] Gyntelberg F. One year incidence of low back pain among male residents of Copenhagen aged 40–59. Dan Med Bull 1974;21:30–6.Search in Google Scholar
[2] Kauppila LI, Tallroth K. Postmortem angiographic findings for arteries supplying the lumbar spine: their relationship to low-back symptoms. J Spinal Disord 1993;6:124–9.Search in Google Scholar
[3] Ohayon MM, Stingl JC. Prevalence and comorbidity of chronic pain in the German general population. J Psychiatric Res 2012;46:444–50.Search in Google Scholar
[4] Pontari MA, McNaughton-Collins M, O’Leary MP, Calhoun EA, Jang T, Kusek JW, Landis JR, Knauss J, Litwin MS. A case–control study of risk factors in men with chronic pelvic pain syndrome. BJU Int 2005;96:559–65.Search in Google Scholar
[5] Tsai PS, Fan YC, Huang CJ. Fibromyalgia is associated with coronary heart disease: a population-based cohort study. Reg Anesth Pain Med 2015;40:37–42.Search in Google Scholar
[6] Dworkin BR, Filewich RJ, Miller NE, Craigmyle N, Pickering TG. Baroreceptor activation reduces reactivity to noxious stimulation: implications for hypertension. Science (New York, NY) 1979;205:1299–301.Search in Google Scholar
[7] Sacco M, Meschi M, Regolisti G, Detrenis S, Bianchi L, Bertorelli M, Pioli S, Magnano A, Spagnoli F, Giuri PG, Fiaccadori E, Caiazza A. The relationship between blood pressure and pain. J Clin Hypertens 2013;15:600–5.Search in Google Scholar
[8] Clark AK, Staniland AA, Malcangio M. Fractalkine/CX3CR1 signalling in chronic pain and inflammation. Curr Pharm Biotechnol 2011;12:1707–14.Search in Google Scholar
[9] da Cunha Ribeiro RP, Roschel H, Artioli GG, Dassouki T, Perandini LA, Calich AL, de Sa Pinto AL, Lima FR, Bonfa E, Gualano B. Cardiac autonomic impairment and chronotropic incompetence in fibromyalgia. Arthr Res Ther 2011;13:R190.Search in Google Scholar
[10] Furlan R, Colombo S, Perego F, Atzeni F, Diana A, Barbic F, Porta A, Pace F, Malliani A, Sarzi-Puttini P. Abnormalities of cardiovascular neural control and reduced orthostatic tolerance in patients with primary fibromyalgia. J Rheumatol 2005;32:1787–93.Search in Google Scholar
[11] Shankar N, Thakur M, Tandon OP, Saxena AK, Arora S, Bhattacharya N. Autonomic status and pain profile in patients of chronic low back pain and following electro acupuncture therapy: a randomized control trial. Indian J Physiol Pharmacol 2011;55:25–36.Search in Google Scholar
[12] Kalezic N, Noborisaka Y, Nakata M, Crenshaw AG, Karlsson S, Lyskov E, Eriksson PO. Cardiovascular and muscle activity during chewing in whiplash-associated disorders (WAD). Arch Oral Biol 2010;55:447–53.Search in Google Scholar
[13] Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12.Search in Google Scholar
[14] Fayaz A, Panesar SS, Langford RM, Donaldson LJ. Assessing the relationship between chronic pain and cardiovascular disease: a protocol for a systematic review. PROSPERO Int Prospect Regist Syst Rev 2013.Search in Google Scholar
[15] Programme CAS. CASP checklists for cohort studies. Oxford: CASP; 2014. http://media.wix.com/ugd/dded87e37a4ab637fe46a0869f9f977dacf134.pdf.Search in Google Scholar
[16] Programme CAS. CASP checklists for case control studies. Oxford: CASP; 2014. http://media.wix.com/ugd/dded8763fb65dd4e0548e2bfd0a982295f839e.pdf.Search in Google Scholar
[17] Hoy D, Brooks P, Woolf A, Blyth F, March L, Bain C, Baker P, Smith E, Buchbinder R. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol 2012;65:934–9.Search in Google Scholar
[18] StataCorp. Stata Statistical Software: Release 13. College Station, T. S. L.; 2013.Search in Google Scholar
[19] Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. A basic introduction to fixed-effect and random-effects models for meta-analysis. Res Synth Methods 2010;1:97–111.Search in Google Scholar
[20] Torrance N, Elliott AM, Lee AJ, Smith BH. Severe chronic pain is associated with increased 10 year mortality. A cohort record linkage study. Eur J Pain (London, England) 2010;14:380–6.Search in Google Scholar
[21] Smith BH, Elliott AM, Hannaford PC. Pain and subsequent mortality and cancer among women in the Royal College of General Practitioners Oral Contraception Study. Br J Gen Pract 2003;53:45–6.Search in Google Scholar
[22] Heliovaara M, Makela M, Aromaa A, Impivaara O, Knekt P, Reunanen A. Low back pain and subsequent cardiovascular mortality. Spine 1995;20:2109–11.Search in Google Scholar
[23] Andersson H. Increased mortality among individuals with chronic widespread pain relates to lifestyle factors: a prospective population-based study. Disabil Rehabil 2009;31:1980–7.Search in Google Scholar
[24] Dreyer L, Kendall S, Danneskiold-Samsoe B, Bartels EM, Bliddal H. Mortality in a cohort of Danish patients with fibromyalgia: increased frequency of suicide. Arthr Rheum 2010;62:3101–8.Search in Google Scholar
[25] Zhu K, Devine A, Dick IM, Prince RL. Association of back pain frequency with mortality, coronary heart events, mobility, and quality of life in elderly women. Spine 2007;32:2012–8.Search in Google Scholar
[26] Chen HM, Lin CC, Kang CS, Lee CT, Lin HC, Chung SD. Bladder pain syndrome/interstitial cystitis increase the risk of coronary heart disease. Neurourol Urodyn 2014;33:511–5.Search in Google Scholar
[27] Conway LC, Smith BH, Hocking LJ, McGilchrist MM, Dominiczak AF, Morris A, Porteous DJ, Goebel A, Goodson NJ. The prevalence of rose angina is increased in people reporting chronic pain: results from a cross-sectional general population study. Rheumatology (United Kingdom) 2014;53:i68.Search in Google Scholar
[28] Ha IH, Lee J, Kim MR, Kim H, Shin JS. The association between the history of cardiovascular diseases and chronic low back pain in South Koreans: a cross-sectional study. PLOS ONE 2014;9:e93671.Search in Google Scholar
[29] Keller JJ, Chen YK, Lin HC. Comorbidities of bladder pain syndrome/interstitial cystitis: a population-based study. BJU Int 2012;110:E903–9.Search in Google Scholar
[30] Makela M, Heliovaara M, Sievers K, Impivaara O, Knekt P, Aromaa A. Prevalence, determinants, and consequences of chronic neck pain in Finland. Am J Epidemiol 1991;134:1356–67.Search in Google Scholar
[31] Ryan C, McDonough S, Kirwan J, Leveille S, Martin D. An investigation of association between chronic musculoskeletal pain and cardiovascular disease in the Health Survey for England (2008). Eur J Pain 2014;18:740–50.Search in Google Scholar
[32] Von Korff M, Crane P, Lane M, Miglioretti DL, Simon G, Saunders K, Stang P, Brandenburg N, Kessler R. Chronic spinal pain and physical-mental comorbidity in the United States: results from the national comorbidity survey replication. Pain 2005;113:331–9.Search in Google Scholar
[33] Kadam UT, Thomas E, Croft PR. Is chronic widespread pain a predictor of all-cause morbidity? A 3 year prospective population based study in family practice. J Rheumatol 2005;32:1341–8.Search in Google Scholar
[34] Ablin JN, Beilinson N, Aloush V, Elkayam O, Finkelstein A. Association between fibromyalgia and coronary heart disease and coronary catheterization. Clin Cardiol 2009;32:E7–11.Search in Google Scholar
[35] Cho KI, Lee JH, Lee HG, Kim SM, Kim TI. Assessment of myocardial function in patients with fibromyalgia and the relationship to chronic emotional and physical stress. Sunhwangi 2010;40:74–80.Search in Google Scholar
[36] Lindgren H, Bergman S. Chronic musculoskeletal pain predicted hospitalisation due to serious medical conditions in a 10 year follow up study. BMC Musculoskelet Disord 2010;11:127.Search in Google Scholar
[37] Mease PJ, Duh MS, Vekeman F, Lefebvre P, Laliberte F, Boerstoel-Streefland M, Jiang W, Zimetbaum PJ. Evaluation of cardiovascular diseases (CVDs) risk for patients with fibromyalgia syndrome in the United States. Pharmacoepidemiol Drug Saf 2010;19:S284–5.Search in Google Scholar
[38] Su C-H, Chen J-H, Lan J-L, Wang Y-C, Tseng C-H, Hsu C-Y, Huang L. Increased risk of coronary heart disease in patients with primary fibromyalgia and those with concomitant comorbidity – a Taiwanese population-based cohort study. PLoS One 2015;10.Search in Google Scholar
[39] Cote P, Cassidy JD, Carroll L. The factors associated with neck pain and its related disability in the Saskatchewan population. Spine 2000;25:1109–17.Search in Google Scholar
[40] Svensson HO, Vedin A, Wilhelmsson C, Andersson GB. Low-back pain in relation to other diseases and cardiovascular risk factors. Spine 1983;8:277–85.Search in Google Scholar
[41] Parsons S, McBeth J, Macfarlane G, Hannaford P, Symmons D. Self-reported pain severity is associated with a history of coronary heart disease. Eur J Pain 2015;19:167–75.Search in Google Scholar
[42] McBeth J, Symmons DP, Silman AJ, Allison T, Webb R, Brammah T, Macfarlane GJ. Musculoskeletal pain is associated with a long-term increased risk of cancer and cardiovascular-related mortality. [Erratum appears in Rheumatology (Oxford). 2009 Apr;48(4):459]. Rheumatology 2009;48:74–7.Search in Google Scholar
[43] Macfarlane GJ, McBeth J, Silman AJ. Widespread body pain and mortality: prospective population based study. BMJ 2001;323:662–5.Search in Google Scholar
[44] Macfarlane GJ, Jones GT, Knekt P, Aromaa A, McBeth J, Mikkelsson M, Heliovaara M. Is the report of widespread body pain associated with long-term increased mortality? Data from the Mini-Finland Health Survey. Rheumatology (Oxford, England) 2007;46:805–7.Search in Google Scholar
[45] Chen J, Su C. Increased risk of coronary heart diseases in patients with fibromyalgia especially in those with concomitant comorbidity – a nationwide population-based cohort study. Int J Rheum Dis 2015;18:70–1.Search in Google Scholar
[46] Hesler BD, Dalton JE, Singh H, Chahar P, Saager L, Sessler DI, Turan A. Association between fibromyalgia and adverse perioperative outcomes. Br J Anaesth 2014;113:792–9.Search in Google Scholar
[47] Wolfe F, Hassett AL, Walitt B, Michaud K. Mortality in fibromyalgia: a study of 8,186 patients over thirty-five years. Arthr Care Res (Hoboken) 2011;63:94–101.Search in Google Scholar
[48] Smith D, Wilkie R, Uthman O, Jordan JL, McBeth J. Chronic pain and mortality: a systematic review. PLOS ONE 2014;9:e99048.Search in Google Scholar
[49] van Hecke O, Torrance N, Smith BH. Chronic pain epidemiology and its clinical relevance. Br J Anaesth 2013;111:13–8.Search in Google Scholar
[50] Ryan CG, McDonough S, Kirwan JP, Leveille S, Martin DJ. An investigation of association between chronic musculoskeletal pain and cardiovascular disease in the Health Survey for England (2008). Eur J Pain (London, England) 2013.Search in Google Scholar
[51] Bhala N, Emberson J, Merhi A, Abramson S, Arber N, Baron JA, Bombardier C, Cannon C, Farkouh ME, FitzGerald GA, Goss P, Halls H, Hawk E, Hawkey C, Hennekens C, Hochberg M, Holland LE, Kearney PM, Laine L, Lanas A, Lance P, Laupacis A, Oates J, Patrono C, Schnitzer TJ, Solomon S, Tugwell P, Wilson K, Wittes J, Baigent C. Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet 2013;382:769–79.Search in Google Scholar
[52] Grading quality of evidence and strength of recommendations. BMJ 2004;328:1490.Search in Google Scholar
Appendix A. Search terms used for review
Search terms used for Embase (via Ovid)Embase Classic + Embase 1947 to 2015 July 17
Chronic Pain.ab,ti.
widespread pain.ab,ti.
fibromyalgia.ab,ti.
(chronic adj5 pain).ti,ab.
exp Chronic Pain/
exp Fibromyalgia/
cardiovascular disease.ti,ab.
myocardial infarction.ti,ab.
angina.ti,ab.
stroke.ti,ab.
ischaemic heart disease?.ti,ab.
atherosclerosis.ti,ab.
cardiovascular morbidity.ti,ab.
cardiovascular mortality.ti,ab.
exp heart muscle ischaemia/
exp heart infarction/
exp cerebrovascular accident/
exp cardiovascular disease/
7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18
1 or 2 or 3 or 4 or 5 or 6
19 and 20
8691 results generated Medline (via Ovid)
Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE(R) 1946 to Present
Chronic Pain.ab,ti.
widespread pain.ab,ti.
fibromyalgia.ab,ti.
(chronic adj5 pain).ti,ab.
exp Chronic Pain/
exp Fibromyalgia/
cardiovascular disease.ab,ti.
myocardial infarction.ab,ti.
angina.ab,ti.
stroke.ab,ti.
ischaemic heart disease?.ab,ti.
atherosclerosis.ab,ti.
cardiovascular morbidity.ab,ti.
cardiovascular mortality.ab,ti.
exp Myocardial Ischaemia/or exp Coronary Disease/or exp Myocardial Infarction/
exp Stroke/
exp cardiovascular disease/
1 or 2 or 3 or 4 or 5 or 6
7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17
18 and 29
2024 results generated Cochrane search terms
“chronic pain*”:ab,ti
“widespread pain*”:ab,ti
fibromyalgia:ab,ti
MeSH descriptor [Chronic Pain] this term only
MeSH descriptor [Fibromyalgia] this term only
1 or 2 or 3 or 4 or 5
“cardiovascular disease*”:ab,ti
“myocardial infarction”:ab,ti
angina:ab,ti
stroke*:ab,ti
“heart disease*”:ab,ti
“cardiovascular mortality”:ab,ti
“cardiovascular morbidity”:ab,ti
atherosclerosis:ab,ti
MeSH descriptor [Cardiovascular Diseases] this term only
MeSH descriptor [Stroke] this term only
MeSH descriptor [Myocardial Infarction] this term only
7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17
6 and 18
Citations returned = 35 Web of Science search terms
“chronic pain” or “widespread pain” or “severe pain” or “persistent pain” or “back pain” or fibromyalgia
“cardiovascular disease$” or “myocardial infarction” or “angina pectoris” or stroke$ or atherosclero* or “Coronary Disease$” or “heart disease$” or “heart failure$” or “left ventricular dysfunction” or “metabolic syndrome”
1 and 2
Citations returned= 67
Appendix B. Quality scores for included studies
Risk of Bias assessment for Cross-sectional studies.
| Article | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | Bias | Quality[a] | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Conway | ? | ? | ? | ? | H | L | ? | L | L | ? | L | M | High | Low | Data from abstract only, many parameters not outlined fully |
| Cote | L | L | L | H | M | L | H | L | L | L | H | H | High | Low | Not clearwhat CVD were but Hypertension included separate. Not clear what pain duration was, Huge CI |
| Ha | L | L | H | H | H | L | L | ? | L | L | H | M | Moderate | Moderate | Low inclusion rate from survey population, overall lower outcome prevalence than expected |
| Makela | L | L | L | L | L | L | H | L | L | L | L | H | High | Low | Overall reasonable quality, but outcome likely to include HTN and confounders included OA/CLBP |
| Ohayaon | L | H | H | L | H | H | H | L | L | L | L | M | Moderate | Moderate | Sampling method restricted to telephone users and responders, inadequate confounding |
| Parsons | L | L | L | H | M | L | H | L | L | L | L | M | Moderate | Moderate | Low survey response rate, the period of pain is not fully defined |
| Ryan | L | H | L | ? | H | L | H | L | L | L | L | H | High | Low | Outcomes restricted to older age groups, surrogates for CP, CVD outcomes include HTN |
| Svensson | L | H | H | L | H | L | H | L | L | H | H | H | High | Low | Inadequate confounding, males only, restricted age group, data irregularities, not necessarily CP |
| Von Korff | L | L | H | H | H | L | H | L | H | L | L | M | High | Low | Selection biased towards mental illness, chronicity based on self-assessment, confounders unclear |
| Scoring tool | |
|---|---|
| 1 | Was the study’s target population a close representation of the national population? |
| 2 | Was the sampling frame a true or close representation of the target population? |
| 3 | Was some form of random selection used to select the sample? |
| 4 | Was the likelihood of non-response bias minimal? |
| 5 | External validity score |
| 6 | Were data collected directly from the subjects (as opposed to a proxy)? |
| 7 | Was an acceptable case definition used in the study? |
| 8 | Was the study instrument that measured the parameter of interest shown to have reliability and validity? |
| 9 | Was the same mode of data collection used for all subjects? |
| 10 | Was the length of the shortest prevalence period for the parameter of interest appropriate? |
| 11 | Were the numerator and denominator for the parameter of interest appropriate? |
| 12 | Internal validity score |
| Key | |
|---|---|
| High | H |
| Low | L |
| Moderate | M |
| Can’t tell | ? |
Critical Appraisal Skills Programme tool for Case Control studies
| Article | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | Score | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ablin | Y | Y | Y | Y | N | N | ? | ? | Y | Y | Y | Moderate | Exposure not qualified, inadequate confounding, numerical inconsistencies, CI not presented |
| Cho | Y | Y | Y | Y | Y | N | ? | ? | Y | N | Y | Moderate | Inadequate confounding, small and homogeneous population |
| Keller | Y | Y | Y | Y | N | N | N | N | Y | N | Y | Moderate | Inadequate confounding, third party data collection, homogeneous population |
| Pontari | Y | Y | Y | Y | Y | N | ? | ? | Y | Y | Y | Moderate | Inadequate confounding, outcome includes hypertension |
| Scoring tool | |
|---|---|
| 1 | Did the study address a clearly focused issue? |
| 2 | Did the authors use an appropriate method to answer their question? |
| 3 | Were the cases recruited in an acceptable way? |
| 4 | Were the controls selected in an acceptable way? |
| 5 | Was the exposure accurately measured to minimise bias? |
| 6 | Was the outcome accurately measured to minimise bias? |
| 7 | Have the authors taken account of the potential confounding factors |
| 8 | Do you believe the results? |
| 9 | Can the results be applied to the local population? |
| 10 | Do the results of this study fit with other available evidence? |
Critical Appraisal Skills Programmes tool for Cohort studies.
| Article | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | Score | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Andersson | Y | Y | Y | Y | N | Y | Y | N | N | Moderate | Inadequate confounding, pain prevalence did not vary significantly with age or gender |
| Chen | Y | Y | Y | Y | N | Y | Y | N | Y | Moderate | Inadequate confounding, low incidence of cardiovascular events |
| Dreyer | Y | Y | N | Y | N | N | N | Y | Y | Moderate | Inadequate confounding, controls not defined, low incidence of cardiovascular events |
| Heliovaara | Y | Y | Y | N | N | Y | Y | Y | Y | Moderate | Inadequate confounding, ICD codes not quoted, therefore cardiovascular outcomes unclear |
| Kadam | Y | Y | Y | ? | N | Y | N | Y | Y | Moderate | Inadequate confounding, outcomes not clearly defined, may include Hypertension |
| Lindgren | Y | Y | Y | Y | N | Y | Y | Y | Y | Moderate | Inadequate confounding |
| Mease | Y | Y | N | ? | N | ? | N | Y | Y | Low | Inadequate confounding, controls not defined, study group identified by third party diagnosis |
| Smith | Y | Y | Y | ? | N | Y | Y | N | N | Moderate | Inadequate confounding, female only, outcomes include mortality from HTN |
| Su | Y | Y | N | Y | Y | Y | Y | Y | Y | Moderate | Exposure generated from insurance records (not direct from subjects), measure for FM not standard |
| Torrance | Y | Y | Y | ? | N | Y | Y | Y | Y | Moderate | Inadequate confounding, outcomes not clearly defined, may include Hypertension related deaths |
| Tsai | Y | Y | N | Y | Y | Y | Y | Y | Y | Moderate | Exposure generated from insurance records (not direct from subjects) but supplemented with medication history |
| Zhu | Y | Y | Y | Y | Y | Y | Y | N | Y | Moderate | Female only population, elderly - results may not be translatable to wider population |
| Scoring tool | |
|---|---|
| 1 | Did the study address a clearly focused issue? |
| 2 | Was the cohort recruited in an acceptable way? |
| 3 | Was the exposure accurately measured to minimise bias? |
| 4 | Was the outcome accurately measured to minimise bias? |
| 5 | Have the authors identified all important confounding factors? |
| 6 | Was the follow up of subjects complete? |
| 7 | Do you believe the results? |
| 8 | Can the results be applied to the local population? |
| 9 | Do the results of this study fit with other available evidence? |
© 2016 Scandinavian Association for the Study of Pain
