DOI: https://doi.org/10.21203/rs.3.rs-586033/v1
Background
Excess mortality in gout has mainly been attributed to cardiovascular disease (CVD). In the light of decreased CVD-mortality in general we wanted to evaluate the risk of overall mortality in gout and cause-specific contributions to mortality beyond CVD as well as temporal trends of it.
Methods
All incident cases of gout between 2006-2015 in western Sweden were identified in the healthcare database VEGA. Up to 5 population controls per case matched on age, sex and county were identified. Relevant comorbidities were identified for 5 years preceding index date in VEGA. Follow-up until the first of death, migration or study end 31 of December 2017. Cause of death was identified and cause-specific mortality grouped as: cardiovascular disease (CVD), renal disease (RD), dementia (DEM), infections, diabetes, diseases of the digestive system (DDS), lung diseases, cancer (CA) and other. Effect of gout on risk of death was calculated using COX regression on whole population and stratified by sex, adjusted for demographics and comorbidities. Death incidence rate were compared between the two time periods, 2006-2010 and 2011-2015.
Results
We identified 22055 cases of incident gout and 98 946 controls, mean age (SD) 67.3-66.3 (15.3-15.4) years and 67.6-66.5% males. All comorbidities were more common at baseline among gout cases except for dementia which was more common in controls. Overall risk for death in incident gout was increased with 3%. Gout cases had a significantly increased risk for death by CVD, RD and DDS and a significantly decreased risk for death by DEM and CA. Similar results were found when stratifying for sex. There were no significant differences in incident death rate ratios in total, and subdivided into CVD and non CVD cause of death, between cases and controls in the two time periods examined.
Conclusions
A 3% increase in all-cause death in patients with gout highlights the importance in addressing not only CVD risk factors in the management of gout patients. Furthermore, gout was associated with reduced mortality from dementia and these findings calls for further study, not least because of possible effects of urate lowering therapy on risk of dementia.
Gout is the most common inflammatory joint disease in the world [1]. It is caused by oversaturation of urate, hyperuricemia, leading to deposition and crystallization of urate in joints and tendons which triggers a pronounced immunological reaction in susceptible individuals resulting in severe pain and excessive inflammation [2]. Acute treatment of gout is available and effective, examples hereof are Non Steroidal Anti Inflammatory Drugs (NSAID), corticosteroids and colchicine. Prophylactic treatment with urate lowering therapy (ULT) is cheap, available and offers a “cure” for the disease but in spite of this it is suboptimally used in Sweden and the world [3, 4]. Gout is closely associated to a number of comorbidities; some with unclear causality such as cardiovascular disease (CVD) and cancer [5, 6] while some lead to hyperuricemia thus increasing the risk of gout, examples hereof are renal disease and obesity [7]. Neurodegenerative diseases, dementia and Parkinson, have been associated with a decreased risk in patients with gout or hyperuricemia [8, 9] although there are conflicting results [10–12]. To further complicate matters, CVD is a risk factor for dementia [13].
Gout has been associated with an increased mortality from CVD [14, 15], death overall [16] and cause specifically for in particular kidney disease, endocrine and metabolic disease [17, 18]. Comorbidities are probably the most important risk factor for death but others have been investigated. Low level of education has been shown to be related to premature mortality in the general population [19] but also in patients with gout [20]. Presence of tophi and high urate, markers of more severe gout disease, have also been shown to increase the risk for cardiovascular death [21, 22]. However, treatment with ULT has not so far been shown to have a protective effect on mortality [23]. Females represent a minority of gout patient and have to a lesser extent been studied but the mortality risk has been shown to be increased in both sex although the risk increment was higher in females [17, 18, 24]. Studies of the effects of gout disease duration on mortality risk have shown contradictory result [18, 25].
A recent registry study from Sweden including all inhabitants aged ≥ 18 years in the southern county Skåne showed a 17% increased risk of all-cause mortality. For cause-specific mortality the study showed an increased risk for death from renal disease, diseases of the digestive system, CVD and infections, whereas the risk was decreased for death from dementia [26].
Life expectancy is increasing worldwide as well as in Sweden. This is to a large extent explained by a decrease in CVD mortality due to improved cardiovascular health care [27]. Decreased mortality in rheumatoid arthritis over the last decades have been shown and this may also be explained by decrease in CVD mortality [28], whereas a similar trend have not been observed in patients with gout over the same time period [29]. One explanation for this may be an increasing impact of non-CVD comorbidities on mortality in patients with gout. Following this, it is important to increase the knowledge and provide contemporary results regarding both all-cause and cause-specific mortality in gout. This may also have implications for future treatment recommendations.
In the present study, our aims were to determine the relative risks in incident gout patients in comparison to the general population 1) for overall death 2) for cause-specific mortality and 3) to examine possible temporal trends.
Data sources
The Western Swedish Health Care Register (VEGA)
The register contains information about all healthcare contacts in in- and outpatient secondary care clinics and in primary care in the WSHCR. Date of contact and primary and secondary diagnoses given by the treating physician according to the Swedish version of the International Statistical Classification of Diseases (ICD-10) are registered. VEGA was used to identify gout cases and to retrieve information about relevant comorbidities for cases and controls. For gout cases, the date of the first gout diagnosis was used as the index date/date of identification.
The Swedish Census Register
The census register holds demographic information about all registered inhabitants of Sweden including date of death and emigration. The census register was used to identify up to 5 controls for each case, matched for age, sex and place of residence on municipality level at the year of identification. Controls were assigned the same index date/date of identification as their corresponding case.
The Longitudinal Integration Database for Health Insurance and Labor Market Studies (LISA)
The LISA register is administered by Statistics Sweden and holds annual registers on all individuals 16 years of age and older. From here data on education level, income, marital status, birthplace outside Sweden and migration was retrieved.
The Swedish Prescribed Drug register
All prescribed drugs and date for dispensation by Swedish pharmacies are recorded in the Swedish Prescribed Drug register. Prescription of ULT (ATC-code M04AA and M04AB) was retrieved from here. The register has been available since 1 July 2005.
The Cause of Death register
The Swedish Cause of Death Register contains information from death certificates recorded according to the ICD- 10 system [30]. The Swedish Cause of Death Register follows recommendations from the World Health Organization for identifying the underlying cause of death [30]. It has been shown to have high validity with respect to main diagnostic categories [31, 32]. We identified the underlying cause of death for all study subjects who died between Jan 1st 2006 and Dec 31st 2017.
The cancer register
The Swedish Cancer Register, maintained by the National Board of Health and Welfare, was founded in 1958 and covers the whole population of Sweden. All data on cancer comorbidity was retrieved from here.
Study population
The Swedish healthcare system is tax-funded and offers universal access. Gout is typically diagnosed and treated in a primary care setting. All inhabitants, aged ≥ 18 years, in the Western Swedish Health Care Region (WSHCR) from January 1 2006 to December 31 2015 constituted the study population. The population of WSHCR is approximately 1.6 million (20% of the total population of Sweden)[33], and the region is considered to be representative for Sweden with regard to health status, healthcare seeking and socioeconomics.[33, 34]
To select individuals most likely to have new-onset gout, we included individuals, aged ≥ 18 years, with a diagnosis of gout in the VEGA-register during 2006–2015. The VEGA database was then searched for gout diagnoses back to January 1, 2000, and individuals with a prior diagnosis of gout were excluded. All included cases thus have a period free from gout diagnosis of at least 5 years before their index date. Gout diagnosis was defined by the presence of an ICD-10 code for gout (M10), registered at a visit to a physician in the VEGA-database. We excluded individuals with dispensation of ULT (ATC-code M04AA and M04AB) prior to first gout diagnosis. Up to 5 controls per case matched by age, year and municipality at the year of gout diagnosis was identified in the Swedish Census Registry. Controls with a prescription of ULT were excluded. The case definition has been previously validated by means of record review which showed a high validity of ICD-10 codes for gout in the VEGA-database [35]. Follow-up started from index date and ended at date of death, relocation outside of WSHCR or the end of the study (December 31, 2017), whichever occurred first.
Outcome
The outcomes were all-cause mortality and cause-specific mortality defined as: cardiovascular disease, renal disease, dementia, infections, diabetes, diseases of the digestive system, lung diseases, neoplasms and other. For definitions based on ICD-10 codes see supplementary Table 1. We identified the underlying cause of death for all study subjects who died between start of follow-up and Dec 31st 2017. To examine temporal trends of death in incident gout, we divided the study period in two, 2006–2010 and 2011–2015. Thereafter, we calculated incident death rate ratios in total, and subdivided into CVD and non CVD cause of death, between cases and controls in the two time periods.
Confounders
The following confounders were recorded in the year of the index date for cases and controls: sex, marital status, income, education, and whether the person was born outside of Sweden. The following comorbidities were considered as possible confounders and they were identified in the VEGA database for at least 5 years prior to the year of the index date: alcohol related disorders, hypertension, ischemic heart disease, heart failure, cerebrovascular disease, diabetes mellitus, dyslipidemia, obesity, chronic kidney disease, dementia and lung diseases, Data on neoplasms was retrieved from the Swedish Cancer Registry. For definitions based on ICD-10 codes see supplementary Table 1.
Statistics
All data management and analyses were performed in SAS 9.4 and R 4.0.3. Baseline features were described as count (percent) of categorical data and as mean (standard deviation (SD)) along with median (25th and 75th percentiles) of continuous data. Incidence rate of total mortality and cause-specific mortality were calculated per 1000 person-year under assumption of Poisson distribution. Cause-specific mortality was treated through-out the study as follows: the cause of death in question was considered as event and the other causes of death were censured. 95% confidence intervals of incidence rate ratios were calculated under assumption of normal distribution for the natural logarithm of IRR. Hazard ratios were computed with Cox proportion hazard model. Impact of death of other causes on a specific cause was computed with Fine-Gray method with the death of other causes as a competing risk [36].
From the total adult population of WSHCR between 2006 and 2015, with a yearly average population of 1 264 150 individuals, we identified 22055 cases of incident gout and 98 946 controls, for details see Table 1. After certain exclusions (Table 1) the mean age (SD) was 67.1 (15.3) years in the gout cohort compared with 66.3 (15.4) in the controls and 67.6% were males in the gout cases compared with 66.5 in the controls, Table 2. Mean annual income, level of education and being born in Sweden was lower in the gout subjects, Table 2. The identified comorbidities were more common among the gout cases with the exception of dementia that was overrepresented in the non-gout control subjects, Table 2. When stratified for sex, the overall picture was similar but the incident female gout cases were older, mean (SD) age, 71.6 (15.1) years compared to males, 65.3 (15.0), supplementary Table 2.
Cases |
Controls |
|
---|---|---|
Identified in VEGA |
56 771 |
|
Identified in The Swedish Census Register |
110081 |
|
Re-used / missing personal identification number |
-78 |
0 |
= |
56693 |
110 081 |
No diagnosis of gout |
-6199 |
N/A |
= |
50494 |
110 081 |
Not VGR resident |
-117 |
0 |
= |
50377 |
110 081 |
Gout diagnosis before 2006 |
-11249 |
-69 |
= |
39128 |
110 012 |
Gout diagnosis after 2015 |
-9932 |
N/A |
= |
29196 |
110 012 |
First emigration date before diagnosis |
-216 |
-28 |
= |
28980 |
109 984 |
First immigration date 30 days after diagnosis |
-43 |
-176 |
= |
28937 |
109 808 |
ULT (M04AA, M04AB) within 5 years before diagnosis |
-6860 |
-1432 |
= |
22077 |
108 376 |
Age at diagnosis < 18 years |
-22 |
|
= |
22055 |
108 376 |
Died before Index date |
0 |
-291 |
22055 |
108 085 |
|
Control more than once |
N/A |
-9139 |
Study population |
22055 |
98 946 |
VEGA = The Western Swedish Health Care Register, VGR = Western Sweden Health Care Region, |
Gout cases, n = 22 055 |
Controls, n = 98 946 |
|
---|---|---|
Age, years, mean (SD) |
67.1 (15.3) |
66.3 (15.4) |
Male sex, n (%) |
14 926 (67.6) |
65 914 (66.5) |
Annual income, EURO, mean (SD) |
20 663 (36 930) |
21 647 (29 636) |
Education, n (%) |
||
≤ 9 years |
8 716 (39.6) |
36 480 (36.9) |
10–12 years |
8 938 (40.6) |
38 459 (39.0) |
≥ 13 years |
4 013 (18.2) |
22 391 (22.7) |
Married, n (%) |
11 718 (53.3) |
52 694 (53.4) |
Born outside Sweden, n (%) |
3 216 (14.6) |
13 611 (13.8) |
Comorbidities, n (%) |
||
Alcohol related disorders |
932 (4.2) |
2 337 (2.4) |
Hypertension |
12 590 (57.1) |
35 276 (35.7) |
Ischemic heart disease |
5 217 (23.7) |
13 265 (13.3) |
Heart failure |
4 180 (19.0) |
6 603 (6.7) |
Cerebrovascular disease |
2 269 (10.3) |
7 568 (7.7) |
Diabetes mellitus |
3 965 (18.0) |
10 550 (10.7) |
Dyslipidemia |
5 756 (26.1) |
15 859 (16.0) |
Obesity |
1 033 (4.7) |
1 642 (1.7) |
Chronic kidney disease |
964 (4.4) |
822 (0.8) |
Dementia |
545 (2.5) |
3 909 (4.0) |
Lung diseases |
3 181 (14.4) |
9 242 (9.3) |
Neoplasm |
2 608 (11.8) |
10 760 (10.9) |
Follow-up time: |
||
Mean (SD) |
5.5 (3.0) |
5.6 (3.0) |
Median (Q1; Q3) |
5.2 (3.2; 7.7) |
5.3 (3.2; 7.7) |
After a mean (SD) follow-up time of 5.5 (3.0) years for gout cases and 5.6 (3.0) years for controls, we identified 5 817 (26.4%) deaths among the gout cases, resulting in an incidence rate (IR) of 47.7per 1000 person years, which was significantly higher compared to the 20 753 (21.0%) deaths, IR 37.6 per 1000 person years, among controls, with an incidence rate ratio (IRR) (95% CI) of 1.27 (1.23–1.31), Table 3. Overall, CVD was the major cause of death in both groups, although significantly more common in gout individuals with an IRR (95% CL) of 1.56 (1.50–1.63), Table 3. Death by renal disease, infections, diabetes, diseases of the digestive system, lung diseases and other were all significantly more common in gout subjects compared to controls, Table 3. The opposite was found for death caused by dementia, IR (95% CL) 2.01 (0.50–8.01)in gout subjects compared to 3.96 (1.48–10.60) in controls, IRR 0.51 (0.45–0.58), Table 3. We have also performed a sensitivity analysis on IRRs for total deaths comparing gout cases with controls stratified by age groups, in all subjects and by sex, which showed significantly higher IRRs in gout cases in all age groups in total and divided by sex with the exception of females aged 18–40 where no significant difference was seen, see suppl table 3.
Cause of death, n (%) |
Gout cases, n = 22 055 |
Incidence rate per 1000 person-years (95% CI) |
Controls, n = 98 946 |
Incidence rate per 1000 person-years (95% CI) |
Incidence rate ratio (95% CI) |
---|---|---|---|---|---|
Total deaths |
5 817 (26.4) |
47.74 (35.95–63.40) |
20 753 (21.0) |
37.60 (27.32–51.76) |
1.27 (1.23–1.31) |
Cardiovascular disease |
2 905 (49.9) |
23.84 (15.96–35.62) |
8 406 (40.5) |
15.23 (9.22–25.17) |
1.56 (1.50–1.63) |
Renal diseases |
102 (1.8) |
0.84 (0.10–7.13) |
171 (0.8) |
0.31 (0.01–10.48) |
2.70 (2.11–3.45) |
Dementia |
245 (4.2) |
2.01 (0.50–8.01) |
2 185 (10.5) |
3.96 (1.48–10.60) |
0.51 (0.45–0.58) |
Infections |
362 (6.2) |
2.97 (0.95–9.26) |
1 193 (5.8) |
2.16 (0.57–8.20) |
1.37 (1.2–1.55) |
Diabetes |
202 (3.5) |
1.66 (0.36–7.60) |
500 (2.4) |
0.91 (0.12–7.10) |
1.83 (1.55–2.15) |
Diseases of the digestive system |
183 (3.2) |
1.50 (0.30–7.43) |
534 (2.6) |
0.97 (0.13–7.10) |
1.55 (1.31–1.84) |
Lung diseases |
254 (4.4) |
2.08 (0.54–8.10) |
912 (4.4) |
1.65 (0.36–7.59) |
1.26 (1.10–1.45) |
Neoplasms |
1 006 (17.3) |
8.26 (4.17–16.33) |
4 650 (22.4) |
8.43 (4.29–16.55) |
0.98 (0.9–1.05) |
Other |
558 (9.6) |
4.58 (1.83–11.44) |
2 202 (10.6) |
3.99 (1.50-10.64) |
1.15 (1.05–1.26) |
CI = confidence intervals |
All-cause mortality was significantly increased in the gout group with a hazard ratio (HR) of 1.24 (95% CI 1.21–1.28) when adjusting for age and sex but decreased to 1.03 (95% CI 1.00-1.06) in the fully adjusted model, Table 4. Stratifying by sex, the relative impact of gout on all-cause mortality was higher and significantly increased only in females compared to males in the fully adjusted model, HR 1.10 (95% CI 1.05–1.15) versus 0.99 (95% CI 0.95–1.03) respectively, Table 4. Death attributed to CVD was significantly higher in gout cases overall, HR 1.13 (95% CI 1.08–1.18), a finding seen for both male and females, Table 4. Death due to renal disease and diseases of the digestive system were significantly higher in gout overall, HR 1.53 (95% CI 1.16–2.01) and 1.27 (95% CI 1.07–1.52) respectively, but when stratifying for sex renal disease was only significantly increased in males with gout, HR 1.59 (95% CI 1.15–2.20) while death due to digestive diseases only was significantly increased in females, HR 1.43 (95% CI 1.08–1.89) respectively, Table 4. Death attributed to dementia was on the other hand significantly lower in gout cases overall 0.65 (95% CI 0.57–0.75), both for males 0.64 (95% CI 0.53–0.78) and for females 0.67 (95% CI 0.56–0.82), Table 4. Furthermore, this reduced risk in gout cases was seen for all of the four most common specific causes of death categorized under dementia: unspecified dementia, Alzheimer disease, vascular dementia and Parkinson disease, suppl table 4. On the other hand, dementia as a comorbidity at baseline was relatively more common in the control subjects in two of the three largest death cause groups, CVD and infections, but not in cancer compared to cases; this held true both in total and stratified by sex, suppl table 5. A lower risk in gout patients was also seen for death caused by lung diseases and neoplasms overall, HR 0.85 (95% CI 0.73–0.98) and 0.90 (95% CI 0.83–0.96) respectively, but only significantly in males when stratifying for sex, HR 0.80 (95% CI 0.66–0.98) and 0.86 (95% CI 0.79–0.94) respectively, Table 4. For information on specific diagnosis of death in each category, see suppl table 4.
Causes of death |
Cause-specific mortality |
Cause-specific mortality, Fine-Gray method |
||||||||
---|---|---|---|---|---|---|---|---|---|---|
Overall |
Male |
Female |
Overall |
Male |
Female |
|||||
Model 1 |
Model 2 |
Model 2 |
Model 2 |
Model 1 |
Model 2 |
Model 2 |
Model 2 |
|||
Cardiovascular disease |
1.54 (1.47–1.60) |
1.13 (1.08–1.18) |
1.07 (1.01–1.13) |
1.23 (1.14–1.32) |
1.54 (1.48–1.61) |
1.17 (1.12–1.23) |
1.13 (1.06–1.20) |
1.25 (1.16–1.34) |
||
Renal diseases |
2.62 (2.05–3.35) |
1.53 (1.16–2.01) |
1.59 (1.15–2.20) |
1.32 (0.79–2.20) |
2.47 (1.93–3.15) |
1.55 (1.15–2.08) |
1.64 (1.16–2.31) |
1.30 (0.72–2.36) |
||
Dementia |
0.51 (0.45–0.58) |
0.65 (0.57–0.75) |
0.64 (0.53–0.78) |
0.67 (0.56–0.82) |
0.45 (0.40–0.52) |
0.68 (0.59–0.78) |
0.70 (0.57–0.84) |
0.67 (0.55–0.82) |
||
Infections |
1.34 (1.19–1.51) |
1.11 (0.98–1.26) |
1.05 (0.90–1.23) |
1.21 (0.98–1.48) |
1.25 (1.11–1.40) |
1.13 (1.00-1.28) |
1.10 (0.94–1.28) |
1.19 (0.97–1.46) |
||
Diabetes |
1.79 (1.52–2.11) |
0.94 (0.79–1.12) |
0.89 (0.71–1.11) |
1.03 (0.78–1.36) |
1.69 (1.43–1.99) |
0.98 (0.82–1.17) |
0.96 (0.76–1.20) |
1.02 (0.77–1.35) |
||
Diseases of the digestive system |
1.52 (1.29–1.80) |
1.27 (1.07–1.52) |
1.18 (0.94–1.49) |
1.43 (1.08–1.89) |
1.44 (1.22–1.71) |
1.30 (1.09–1.56) |
1.24 (0.98–1.55) |
1.41 (1.06–1.87) |
||
Lung diseases |
1.24 (1.08–1.42) |
0.85 (0.73–0.98) |
0.80 (0.66–0.98) |
0.90 (0.71–1.12) |
1.17 (1.02–1.34) |
0.86 (0.75-1.00) |
0.83 (0.68–1.01) |
0.89 (0.71–1.11) |
||
Neoplasms |
0.95 (0.89–1.02) |
0.90 (0.83–0.96) |
0.86 (0.79–0.94) |
0.97 (0.86–1.11) |
0.90 (0.84–0.96) |
0.91 (0.84–0.97) |
0.88 (0.80–0.96) |
0.98 (0.86–1.11) |
||
Other |
1.12 (1.02–1.23) |
1.03 (0.93–1.14) |
1.02 (0.91–1.15) |
1.04 (0.88–1.23) |
1.06 (0.96–1.16) |
1.05 (0.95–1.16) |
1.06 (0.94–1.19) |
1.03 (0.87–1.22) |
||
All-cause mortality |
1.24 (1.21–1.28) |
1.03 (1.00-1.06) |
0.99 (0.95–1.03) |
1.10 (1.05–1.15) |
|
|||||
adjusted hazard ratios, 95% confidence intervals, and with competing risk Cox proportional hazard model with the method proposed by Fine and Gray, CI = confidence intervals Model 1: adjusted for sex and age at baseline Model 2: adjusted for sex and age at baseline, marital status, income, education, born outside of Sweden, alcohol related disorders, hypertension, ischemic heart disease, heart failure, cerebrovascular disease, diabetes mellitus, dyslipidemia, obesity, chronic kidney disease, dementia, lung diseases, any neoplasm |
In the sensitivity analysis, where possible impact of competing causes of death on our results was evaluated with the method proposed by Fine and Gray [36], all point estimates were in the same direction and of similar magnitude, see Table 4.
There were no significant differences in incident death rate ratios in total, and subdivided into CVD and non CVD cause of death, between cases and controls in the two time periods examined, 2006–2010 versus 2011–2015, see supplementary table 4.
In this study, we found significantly increased total number of deaths in the gout cases. After adjusting for age, sex, birthplace outside Sweden and relevant comorbidities we found a 3% increased risk in all-cause mortality among the gout cases. Furthermore, the gout patients had a significantly increased risk for death caused by CVD, renal disease, and diseases of the digestive system and a significantly decreased risk for death caused by dementia, lung diseases and neoplasms. Similar results were found in men and women when stratifying for sex. We found no significant temporal trends of death rate between the first and the last part of the study period.
CVD
As expected CVD was the main cause of death for patients with gout in our study in line with previous studies [14, 15, 18, 22, 37]. Not only do the patients have an extensive CVD morbidity at time of first gout diagnosis, but in addition gout is treated with acute medication with potentially cardiotoxic medicines such as NSAIDs or corticosteroids over time. This emphasizes the need for CVD screening in gout patients but also an increased use of ULT to diminish the need for acute treatment. Furthermore, gout comes with episodes of severe inflammation which may have negative impact on CVD. This is supported by results from the CARES trial in 2017 where treatment with the anti-inflammatory, anti-gout, interleukin-1 inhibitor canakinumab decreased the risk of recurrent cardiovascular events [38].
Renal disease
Death caused by renal disease was significantly increased in the gout patients in line with previous studies [18, 39]. Decreased renal secretion of urate by genetic predisposition and / or kidney disease is a major cause of hyperuricemia, which in a proportion will lead to gout [40]. At time of diagnosis, the gout patients in this study had an increased occurrence of multiple risk factors for kidney damage such as hypertension, diabetes and obesity. In addition, they also have an increased use of acute medication with potentially nephrotoxic NSAIDs for gout attacks. With appropriate and increased use of ULT, such need for acute treatment could be diminished. The possible negative effect of gout upon kidney disease is still an unresolved matter where further research effort is needed [41].
Infections
In the present study, death due to infections was not increased in gout patients. The risk was found increased in the recent population based registry study from the southern county Skåne, Sweden [26] and in a in a cohort study from Taiwan comparing 6000 (1400 female) gout patients with the general population [37]. A population based cohort study using data from the UK Clinical Practice Research Datalink that included 130 000 gout patients and 250 000 controls from 1987 to 2014 showed that patients with gout did not have an increased risk of infection-related mortality [42] while a Dutch prospective cohort study with crystal proven gout showed an increased risk for death due to infectious disease as well as cancer and CVD [22]. Frequency of gout increases with age and so do comorbidities. A recent study from the US by Singh et al have shown an increase in serious infections leading to hospitalization over the last two decades in gout patients. The most common infections seen were pneumonia and sepsis, with age and comorbidities as the major identified contributing factors [43]. In the present study, pneumonia and sepsis were the major infectious causes of death in both cases and controls. These findings emphasize the EULAR guidelines regarding vaccination from 2019 where it is advised to consider pneumococcal vaccination for the majority of patients with inflammatory rheumatic diseases [44].
Dementia
In our study, gout protected against death by dementia which was also showed in the aforementioned population based registry study from the southern county Skåne, Sweden [26]. Register studies have shown a decreased risk for dementia in gout individuals [45], metaanalysis of cross-sectional studies have shown protective effect of (increasing) urate on risk for Alzheimers dementia [9] while longitudinal studies show conflicting results [10–12]. In the present study, we lack data on urate levels. Obesity is strongly associated to both gout and hyperuricemia and was more common in the gout subjects in our study. The obesity paradox refers to the dual, age-dependent, effects on dementia, where overweight and obesity in middle age are associated with an increased future dementia risk in old age but when examined later in life higher BMI is linked to better cognition and decreased mortality [46] which may influenced the results in our study. It should be noted that our capture of obesity at baseline likely lacked sensitivity and we did not have data on BMI. Our findings are however supported by the observation that diagnosed dementia was less common in gout cases vs controls already at start of follow-up. The possible protective effect of hyperuricemia on dementia development needs to be further examined and related to the treatment target for urate suggested by current ACR and EULAR treatment guidelines [47, 48].
Lung disease
Gout has not been found to be associated to smoking [49, 50] although exposure to air pollution or inorganic dust may have a role in gout pathogenesis [51, 52]. Nevertheless, lung diseases were a more common comorbidity in the gout population of our study and in spite of this we saw a significant protective effect of gout on death by lung diseases in males which haven’t been shown in earlier studies [17, 26]. In the current study we have no data on smoking, however in 2017 we performed a questionnaire study on 800 gout patients in the same region of western Sweden as this study was performed and there we found no differences in frequency of current smoking compared to non-gout controls but a significantly higher degree of former smokers among gout patients in both males and females [53]. The two most common lung related causes of death in the current study in both cases and controls were chronic obstructive pulmonary disease and interstitial pulmonary disease but small numbers limit further analysis.
Cancer
Prevalence of neoplasms were higher at time of first gout diagnosis compared to non-gout controls in our study. In spite of this, we saw a a protective effect of gout on death attributed to cancer which haven’t been shown in earlier population based studies [17, 26]. Gout is characterized by hyperuricemia which has potent antioxidant properties possibly protecting against cancer. However, gout is also closely associated to insulin resistance, obesity and increased alcohol intake which are all well known carcinogenic risk factors. An increased risk for some cancers have been shown in gout patients, particularly urological cancers, cancers of the digestive system and lung cancer [54]. In the current study, the four major causes of death by cancer were cancer of the prostate, lung, pancreas and large intestine / colon in both cases and controls. This mirrors the incidence and mortality of cancer in Sweden [55].
Digestive system
The increased risk of death from diseases of the digestive system in our study is supported by previous studies[17, 26]. Diagnosed alcohol related disorders were significantly higher at time of first gout diagnosis compared to the controls in our study. This was reflected in alcoholic liver disease being the only significantly increased cause-specific death related to diseases of the digestive system in the gout cases in the study.
Diabetes
Diabetes was one of the more common comorbidities at baseline in the gout patients of our study but in spite of this we could not find an elevated risk of death by diabetes which has been shown in earlier studies [17, 56].
General risk factors
Higher income is associated with decreased mortality in the Nordic countries [57] and lower level of education is a risk factor for increased mortality in the general population as well as in gout patients [19, 20]. The gout patients in our study had lower income and educational level compared to controls and they were also more often born outside Sweden. Increased mortality risk due to CVD and cancer has been shown in Swedish immigrants [58, 59]. The mortality risk increment in women compared to men with gout must be interpreted in the context of factors such as a lower overall mortality in women, higher age of gout onset and consequently more comorbidities at start and during follow-up [60].
There are some limitations to this study. First, as in all register studies there is a risk for misclassification of diagnosis. However, we have in a previous study found the gout diagnosis valid to a high degree [61]. Secondly, obesity and alcohol abuse was in the study only defined with ICD-10 codes at baseline and we lack data specifically on BMI and alcohol intake, at baseline and during follow-up, which will lead to an underestimation of these risk factors. Third, we lack data on smoking. Fourth, urate may have a role in many of the mortality causes in the study, not least dementia, but we lack data on urate levels.
There are also some strengths to this study. The study is population based and the sample size is quite large minimizing the risk for selection bias. Data on cause of death, comorbidities and cancer were collected from three different registers which all have almost complete coverage of the population. The study was performed on incident cases of gout minimizing the risk for survival bias. Furthermore, sensitivity analysis was performed and showed no impact of competing causes of death on our results.
To conclude, in this study we have shown a 3% increase in all-cause death in patients with gout highlighting the importance in addressing not only CVD risk factors but also infections, diseases of the digestive system and renal diseases in the management of patients with gout. Furthermore, gout was associated with reduced mortality from dementia and these findings calls for further study, not least because of possible effects of urate lowering therapy on risk of dementia
cardiovascular disease (CVD)
hazard ratio (HR)
incidence rate (IR)
incidence rate ratio (IRR)
International Statistical Classification of Diseases 10 (ICD-10)
Longitudinal Integration Database for Health Insurance and Labor Market Studies (LISA)
Non Steroidal Anti Inflammatory Drugs (NSAID)
Standard deviation (SD)
The Western Swedish Health Care Register (VEGA)
urate lowering therapy (ULT)
Western Swedish Health Care Region (WSHCR)
Ethics approval and consent to participate
Ethical approval for the study was granted from the Ethical Review Board of Gothenburg, Sweden. Patient consent was waived, as data were derived from administrative registers that do not require such consent
Consent for publication
Not applicable
Availability of data and materials
The datasets analysed during the current study are available from the corresponding author on reasonable request.
Competing interests
The authors declare that they have no competing interests.
Funding
Funding for the study was received from the following sources, which did not influence either the design of the study, the collection and analysis of the data, or the preparation of the manuscript: Reumatikerfonden and Gothenburg University. All researchers assigned as authors state their complete independence from the funders with regard to this study.
Authors' contributions
All authors have contributed substantially in the process of completing this study and had full access to the data, specified as follows: MD contributed to the conception and design of the study, as well as managing and interpretation of data, and drafting and revising the manuscript. TZS contributed to interpretation of data, was responsible for statistical work and participated in drafting and revising the manuscript. LTHJ contributed to the conception and design of the study, as well as interpretation of data, and drafting and revising the manuscript. All authors have approved the manuscript and agree to be accountable.
Acknowledgements
Not applicable