The incidence of biopsy-proven IgA nephropathy is associated with multiple socioeconomic deprivation

Transcription

1 clinical investigation & 2013 International Society of Nephrology The incidence of biopsy-proven IgA nephropathy is associated with multiple socioeconomic deprivation Emily P. McQuarrie 1, Bruce Mackinnon 1, Valerie McNeice 2, Jonathan G. Fox 1 and Colin C. Geddes 1 1 Glasgow Renal and Transplant Unit, Western Infirmaryon behalf of the Scottish Renal Biopsy Registry, Glasgow, UK and 2 Glasgow Centre for Population Health, Glasgow, UK Chronic kidney disease is more common in areas of socioeconomic deprivation, but the relationship with the incidence and diagnosis of biopsy-proven renal disease is unknown. In order to study this, all consecutive adult patients undergoing renal biopsy in West and Central Scotland over an 11-year period were prospectively analyzed for demographics, indication, and histologic diagnosis. Using the Scottish Index of Multiple Deprivation, 1555 eligible patients were separated into quintiles of socioeconomic deprivation according to postcode. Patients in the most deprived quintile were significantly more likely to undergo biopsy compared with patients from less deprived areas (109.5 compared to 95.9 per million population/year). Biopsy indications were significantly more likely to be nephrotic syndrome, or significant proteinuria without renal impairment. Patients in the most deprived quintile were significantly more likely to have glomerulonephritis. There was a significant twofold increase in the diagnosis of IgA nephropathy in the patients residing in the most compared with the least deprived postcodes not explained by the demographics of the underlying population. Thus, patients from areas of socioeconomic deprivation in West and Central Scotland are significantly more likely to undergo native renal biopsy and have a higher prevalence of IgA nephropathy. Kidney International (2013) 85, ; doi: /ki ; published online 11 September 2013 KEYWORDS: chronic kidney disease; clinical epidemiology; glomerulonephritis; IgA; nephropathy; renal biopsy Chronic kidney disease (CKD) is common and is associated with a significant burden of morbidity and mortality. 1,2 Despite this, the underlying etiology of many primary renal diseases remains poorly understood, with a resultant paucity of preventive measures. Living in a deprived area increases the likelihood of an individual experiencing health inequalities and is associated with an increased prevalence of cardiovascular disease, stroke and malignancy. 3 It is widely accepted that health is determined via a complex set of interactions between genetic endowment and the social and physical environment. Health behavior, in response to these factors, acts as an intermediary. 4 Socioeconomic deprivation is closely associated with various health-damaging behaviors and risk factors for ill-health, including smoking, excessive alcohol intake, poor diet, and obesity. Evidence exists to support a clustering of patients with CKD in areas of deprivation, 5 and acceptance rates onto renal replacement therapy programs are significantly associated with underlying deprivation in the areas studied. 6 Globally, studies have shown an association between CKD and measures of deprivation. 7 9 Attempts to address the issue of the relationship between the incidence of renal diseases and deprivation are potentially limited by the confounding factors of access to healthcare and the definitions of deprivation and renal disease used. Scotland has robust and validated methods of tracking deprivation, healthcare, which is free at the point of access, and an established native renal biopsy registry. 10 As such, we aimed to assess the impact of deprivation on the incidence of native renal biopsy and the diagnosis made as a result. First, we hypothesized that patients facing socioeconomic challenges were less likely to access renal services and ultimately, therefore, less likely to undergo native biopsy. Second, we postulated that the clustering of lifestyle risk factors, of which deprivation status is a surrogate marker, could be associated with a greater risk of developing renal disease. Correspondence: Emily P. McQuarrie, Glasgow Renal and Transplant Unit, Western Infirmary, Dumbarton Road, Glasgow, G11 6NT, UK. Emily.mcquarrie@nhs.net Received 27 August 2012; revised 22 May 2013; accepted 29 May 2013; published online 11 September 2013 RESULTS Baseline demographics A total of 1818 native biopsies were performed in the 11 years between 2000 and A total of 263 patients were excluded (75 had no postcode available, 188 had a postcode outside of Greater Glasgow and Clyde (GG&C) or Forth Valley Health 198 Kidney International (2014) 85,

2 EP McQuarrie et al.: Deprivation and native renal biopsy clinical investigation Biopsies PMP Year Figure 1 Biopsy incidence over time. The rise in incidence in 2006/ 2007 coincides with the introduction of chronic kidney disease (CKD) guidelines in primary care. PMP, per million population. Board (FVHB)), leaving 1555 eligible patients (94.2 biopsies per million population per year (PMP/yr)). An increase in the number of biopsies being performed was seen in 2006 and 2007 (Figure 1). This coincided with the introduction of estimated glomerular filtration rate reporting and the adoption of the Quality Outcomes Framework for primary care physicians in the United Kingdom, changes that increased the diagnosis and referral of patients with CKD. At the time of biopsy, 58.5% of patients were male, mean age was 56.3 (s.d. 17.1) years, mean estimated glomerular filtration rate was 41.4 (s.d. 32.2) ml/min per 1.73 m 2, and median urinary protein/creatinine ratio was 234 mg/mmol (interquartile range ). Biopsy indications Biopsy indications were as follows: chronically reduced function (not nephrotic syndrome), 30.2 PMP/yr; acute renal failure, 17.7 PMP/yr; nephrotic syndrome, 15.4 PMP/yr; moderate-to-severe proteinuria (41 g per 24 h) with normal renal function with or without microscopic hematuria (not nephrotic syndrome), 10.2 PMP/yr; and mild proteinuria (p1 g per 24 h) with normal renal function with or without microscopic hematuria, 3.4 PMP/yr (Table 1). Biopsy diagnoses The biopsy diagnoses were as follows: glomerulonephritis (GN), 40.0 PMP/yr; vasculitis, 12.1 PMP/yr; interstitial nephropathy, 6.1 PMP/yr; chronic ischemia, 6.0 PMP/yr; diabetic nephropathy, 5.1 PMP/yr; amyloid and myeloma, 5.0 PMP/yr; acute tubular necrosis, 3.4 PMP/yr; systemic lupus erythematosis, 3.4 PMP/yr; and others/nondiagnostic, 7.3 PMP/yr (Table 1). Incidence of native renal biopsy and deprivation Patients in the most deprived quintile within GG&C and FVHB were significantly more likely to undergo native biopsy compared with patients from less deprived areas (Table 2, Po0.0001, w 2 test). A total of biopsies PMP/yr were performed in the most deprived areas as compared with 95.9 biopsies PMP/yr in the least deprived areas. Indications for native renal biopsy, baseline demographics, and deprivation Indications for biopsy in patients from deprived areas were more likely to be nephrotic syndrome (24.2 vs PMP/yr; w 2, P ¼ 0.001) or significant proteinuria without renal impairment (16.1 vs. 5.7 PMP/yr; w 2, P ¼ 0.003; Table 1). Patients from deprived areas had significantly higher baseline proteinuria (median urinary protein/creatinine ratio 317 vs. 257 mg/mmol; Kruskall Wallis, P ¼ 0.02) but similar baseline renal function at the time of biopsy. Diagnosis of native renal biopsy and deprivation Patients in the most deprived quintile were significantly more likely to have GN (57.0 vs PMP/yr; w 2, P ¼ ). There was a twofold increase in the diagnosis of IgA nephropathy in the most deprived patients as compared with the least deprived patients (20.0 vs.9.1 PMP/yr; one-way analysis of variance, Po0.0001; Table 1). There was no significant difference in the biopsy-proven incidence of the other main primary glomerulopathies according to deprivation category (Table 1). Biopsy-proven diabetic nephropathy was more common in patients from the most deprived quintile (9.3 vs. 6.0 PMP/yr; one-way analysis of variance, P ¼ 0.01) than in less deprived patients (Table 1). Demographics according to deprivation Table 3 details the demographics of the whole population (Table 3a) and of adult patients (age 416 years) undergoing biopsy (Table 3b) in each of the deprivation quintiles. 11 The whole population in quintile 1 (the most deprived) contained slightly fewer men and fewer people of working age than the other quintiles. Therefore, the underlying population in quintile 1 does not appear skewed to favor demonstrating a difference in the likelihood of glomerulonephitis. In particular, young men were not overrepresented in deprivation quintile 1. Patients who underwent a biopsy in quintile 1 were not younger or more likely to be male, but male patients who were biopsied were more likely to be of working age than those biopsied in other quintiles. DISCUSSION We have demonstrated that patients from deprived areas in Scotland are significantly more likely to undergo native renal biopsy, suggesting a higher prevalence of renal disease. Such patients were significantly more likely to be found to have IgA nephropathy, at renal biopsy. This is the first time that a definitive link has been shown between socioeconomic deprivation and biopsy-proven renal disease. This study is uniquely placed to address the association between deprivation and the incidence of biopsy-proven renal diseases for several reasons. The area-based deprivation status of the population studied was based on wellestablished and systematic methods. 12 Biopsy practice is standard among nephrologists in the unit studied, 10 although the number of biopsies being performed has increased (Figure 1). This coincided with the introduction of primary Kidney International (2014) 85,

3 clinical investigation EP McQuarrie et al.: Deprivation and native renal biopsy Table 1 Biopsy indication, baseline demographics of patients biopsied, biopsy diagnosis, and GN subclassification by deprivation category, reported as biopsies per million population per year (PMP/yr) Deprivation quintile 1 N ¼ 282,281 2 N ¼ 292,334 3 N ¼ 299,218 4 N ¼ 306,685 5 N ¼ 300,548 P-value Biopsies per million population per year Indications ARF (n ¼ 312) F CRF (n ¼ 531) F QP41g N func (n ¼ 179) U Nephrotic (n ¼ 271) U QPo1g N func (n ¼ 59) F Baseline demographics upcr (mg/mmol) 310 ( ) 235 (67 642) 202 (74 462) 204 (91 556) 257 (85 556) 0.02 e egfr (ml/min per 1.73 m 2 ) 42.3(32.7) 40.7(34.2) 40.8(29.0) 43.3(32.2) 39.5(33.5) NS y Overall diagnoses GN (n ¼ 703) U Vasculitis (n ¼ 212) F IN (n ¼ 108) F Ischemia (n ¼ 105) F ATN (n ¼ 60) F SLE (n ¼ 60) F DMN (n ¼ 90) U Amyloid (n ¼ 88) F GN subclassification MCN (n ¼ 64) F MGN (n ¼ 172) F IGAN (n ¼ 216) U MCGN (n ¼ 39) F FSGS (n ¼ 146) F Abbreviations: ARF, acute renal failure; ATN, acute tubular necrosis; CRF, chronic renal failure; DMN, diabetic nephropathy; egfr, estimated glomerular filtration rate; FSGS, focal segmental glomerulosclerosis; GN, glomerulonephritis; IGAN, IgA nephropathy; IN, interstitial nephritis; MCGN, mesangiocapillary glomerulonephritis; MCN, minimal change nephropathy; MGN, membranous glomerulopathy; Nephrotic, nephrotic syndrome; NS, not significant; QPo1g N func, o1 g/day proteinuria and normal renal function; QP41g N func, 41 g/day proteinuria and normal renal function; SLE, systemic lupus erythematosus; upcr, urinary protein/creatinine ratio. Columns: deprivation quintile (1¼ most deprived, 5 ¼ least deprived) with overall underlying population in each category. P is estimate of significance: F comparison by w 2, y one-way analysis of variance (ANOVA), or e Kruskall Wallis test as appropriate. Estimates of significance relate to analysis of difference across all five groups. upcr is median (interquartile range) and egfr is mean (s.d.). Bold values are statistically significant values. Table 2 Biopsy incidence by deprivation category Deprivation category Population Total biopsies Biopsies PMP/yr 1 282, , , , , Abbreviation: PMP/yr, per million population per year. Deprivation category: 1 ¼ most deprived and 5 ¼ least deprived. Population indicates the number of people in the deprivation category. Total number of biopsies performed and the number of biopsies performed per million population per year. Po0.001 across all groups by w 2 test. care registers of CKD, leading to an increase in referral. 13 Finally, the National Health Service is free at the point of access, reducing bias associated with referral and access to renal biopsy. Therefore, differences seen are more likely to reflect true variation in renal disease incidence related to deprivation status without associated bias. Our hypothesis that individuals living in more deprived areas were less likely to be investigated with a renal biopsy proved to be incorrect. Instead, we found an increased Table 3a Age and gender distribution of the whole population contained within each deprivation quintile Q1 (most deprived) Q2 Q3 Q4 Q5 (least deprived) Mean % F age 416y % M age 416y % F pension age % M pension age % Total pension age % F working age % M working age % Total working age Abbreviations: F, female; M, male; Q, quintile; y, year. Pension age for men is 465 years and for women is 460 years. Working age for women is years and for men is years. incidence of biopsy in deprived areas, which was largely driven by patients undergoing biopsy to investigate nephrotic syndrome or significant proteinuria. It is widely accepted that there are multiple barriers to accessing healthcare, beyond finance, for the most deprived. 14 If this is the case in our cohort, then the actual differences in incidence of renal disease 200 Kidney International (2014) 85,

4 EP McQuarrie et al.: Deprivation and native renal biopsy clinical investigation Table 3b Age and gender distribution of adult patients who underwent native renal biopsy in each deprivation quintile Bx Q1 Bx Q2 Bx Q3 Bx Q4 Bx Q5 N ¼ 350 N ¼ 330 N ¼ 336 N ¼ 308 N ¼ 231 Mean Mean age (y) (s.d.) 55.0 (17.2) 56.5 (17.2) 56.0 (16.4) 55.8 (17.4) 58.9 (17.1) 56.4 %Fage416y %M4age 16y % F pension age % M pension age % Total pension age % F working age % M working age % Total working age Abbreviations: F, female; M, male; Q, quintile; y, year. Mean age at biopsy is presented as mean (s.d.). Pension age for men is 465 years and for women is 460 years. Working age for women is years and for men is years. according to the degree of deprivation may be even more marked. The higher level of proteinuria at biopsy in the most deprived quintile hints at this, but this study was not designed to address access to healthcare. There was a nonsignificant increase in biopsy incidence in the least deprived quintile. In the developed world, relative deprivation is a surrogate marker for a number of factors associated with poor health. In Scotland, patients from deprived areas are more likely to smoke and to smoke heavily. 3 Smoking was shown to be associated with progression of IgA nephropathy in a Japanese population, 15 but no studies have addressed smoking and the risk of developing renal diseases. Drug and alcohol misuse is also more common in areas of multiple deprivation, although average quantity of alcohol consumed is not correlated with deprivation in Scotland. 16 Other factors associated with living in an area of socioeconomic deprivation are the consumption of a poor diet (low in fruit and vegetables, high in saturated fat and sodium), obesity, and reduced levels of exercise. 3 The impact of genetic and racial factors should also be considered, but the vast majority of patients in our study are of white Caucasian ethnicity. 17 Areas of multiple deprivation are largely clustered in inner city regions in Scotland, raising the possibility of environmental factors as a possible confounder. The relative Scottish Index of Multiple Deprivation (SIMD) ranking is derived from seven domains; although it would be interesting to see which of these domains has most impact on CKD, the way the ranking is calculated does not allow this. Previous studies have reported an association between deprivation and CKD, both in the United Kingdom and the United States. 5,7 The biggest etiological drivers of CKD incidence are hypertension, diabetes, and vascular disease. It is recognized that hypertension, diabetes, and vascular disease are more common in patients from deprived areas, and the association with chronic kidney disease is perhaps unsurprising. However, these associations do not explain the finding of an association between primary glomerulopathies and deprivation. IgA nephropathy is the most commonly diagnosed primary glomerulopathy in adults and encompasses a spectrum of disease, both clinically and histopathologically. It typically presents in young or middle-aged adults and is more common in men. We have demonstrated that the association between deprivation and IgA nephropathy seen in this study is not explained by age or gender differences among the underlying population from which it was sampled. The association between the diagnosis of diabetic nephropathy and deprivation is difficult to interpret. Renal biopsy is only performed in these patients if the clinical scenario suggests an alternative explanation for proteinuria. Differences seen may therefore not represent a true difference in the incidence of diabetic nephropathy. Living in an area of socioeconomic deprivation has been shown to be associated with adverse outcomes once an individual has developed established kidney disease. In Australia and France, patients from deprived areas are at an increased risk of requiring renal replacement therapy, 18,19 and in the United Kingdom, patients have a lower chance of receiving a living donor kidney transplant. 6 It is therefore important to address the issues of disease causation, diagnosis, and treatment aggressively in this group of patients. In contrast, living in a less deprived area is associated with a significant survival advantage of up to 20 years in Scotland. 3 Limitations As this is an observational study, we can only report associations and cannot demonstrate causation. The possibility of sampling bias skewing the apparent finding of a higher rate of diagnosis of IgA nephropathy in patients from deprived areas cannot be excluded, although there was no significant difference in the rate of diagnosis of many other conditions studied. The SIMD is a validated, robust, population-level indicator of deprivation, which is used as a proxy for individual deprivation in this study. However, not everyone living in a deprived area will be deprived and vice versa. Furthermore, the SIMD provides a measure of deprivation relative to other datazones in Scotland, but it does not easily enable comparison with measures of absolute or relative deprivation in other countries. The fact that the cohort of Kidney International (2014) 85,

5 clinical investigation EP McQuarrie et al.: Deprivation and native renal biopsy patients is large compared with other epidemiological studies and comes from a well-defined population is a strength of the study, but it is difficult to know whether the observations are generalizable to other populations. The population in the areas studied is predominantly urban and white Caucasian, which again limits generalizability. The observations provide targets for future research to determine which of the factors associated with socioeconomic deprivation have a causal role in the development of biopsy-proven renal diseases. CONCLUSIONS We have demonstrated for the first time a clear association between the incidence of biopsy-proven renal disease, particularly IgA nephropathy, and socioeconomic deprivation. The fact that individuals in whom presentation of IgAN is most common are not overrepresented in the studied areas of multiple deprivation suggests a difference in disease incidence rather than an epiphenomenon. This novel association deserves further study in order to understand environmental or genetic factors associated with the development of primary renal diseases, particularly IgA nephropathy. This may help prevent the development of CKD, with its associated burdens of renal replacement therapy, cardiovascular disease, and mortality. MATERIALS AND METHODS The Glasgow Renal and Transplant Unit serves a geographically defined population of 1.5 million people across West and Central Scotland. All consecutive adult native renal biopsies (one per patient) performed between 2000 and 2010 were prospectively collected. Baseline demographics, biopsy indications, and diagnoses were recorded. The SIMD 20 is a relative ranking of all areas of Scotland, produced by the Scottish Government every 2 years. We utilized the SIMD The SIMD uses a multidimensional model to assess deprivation, using 38 indicators across seven domains. Each domain is weighted according to its importance to the overall concept of multiple deprivation and robustness of the indicators comprising the domain. The domains are weighted as follows: employment (12), income (12), crime (2), housing (1), health (including standardized mortality ratio, alcohol, and drug misuse) (6), education (6), and geographical and convenience access (4). In all, 6505 geographical areas (datazones), based on postcode, with an average of 750 people living in each, are ranked. This ranking provides a relative measure of deprivation, whereby one postcode area can be ranked relative to another, but cannot be used to determine how much more deprived one area is compared with another. 21 All datazones within the two Scottish Health Boards being studied (Greater Glasgow and Clyde and Forth Valley) were then further sorted according to their SIMD ranking and separated into health board specific quintiles of deprivation (Deprivation Category) for analysis, with 1 being the most deprived and 5 being least deprived. The postcode of the patient undergoing biopsy was then used to determine his or her health board specific deprivation category. As such, all patients were aligned with a deprivation category based on the area in which they live. Biopsy indications were prospectively collected from 2003 and were classified as acute renal failure, nephrotic syndrome, mild proteinuria (p1 g per 24 h) with normal renal function with or without microscopic hematuria, moderate-to-severe proteinuria (41 g per 24 h) with normal renal function with or without microscopic hematuria, not nephrotic syndrome, chronically reduced function, not nephrotic syndrome, and others. Biopsy diagnoses were classified as glomerulonephritis, vasculitis, systemic lupus erythematosis,, interstitial nephropathy (including chronic pyelonephritis), acute tubular necrosis,, chronic ischemia (histological changes of chronic ischemia including glomerulosclerosis, tubular atrophy, hypertensive changes in arteries, or evidence of cholesterol emboli in the absence of other pathology to explain the clinical features), diabetic nephropathy, amyloid and myeloma, and others/non-diagnostic. All samples were examined using light microscopy, immunofluorescence, and electron microscopy where possible. Patients with GN were then classified further, focusing on the five main primary GNs: IgA nephropathy, minimal change nephropathy, membranous nephropathy, focal segmental glomerulosclerosis, and mesangiocapillary glomerulonephritis. In cases where a biopsy revealed more than one diagnosis, the diagnosis most relevant to the indication for renal biopsy was recorded. Statistics The biopsy rate per million population per year was calculated to allow comparison with other populations internationally, but statistical comparisons between SIMD quintiles in this study were not based on this derived figure. Comparisons between SIMD quintiles were conducted using w 2 test or one-way analysis of variance. These comparisons were based on actual numbers of biopsies performed in each group or actual values seen, and did not take into account person-years. Estimates of significance relate to differences across all groups, unless post hoc analysis was conducted, in which case this is specified. Ethics The Scottish Renal Biopsy Registry has multisite ethical approval for data collection and epidemiological analysis. DISCLOSURE All the authors declared no competing interests. REFERENCES 1. Go AS, Chertow GM, Fan D et al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: Gerstein HC, Mann JF, Yi Q et al. Albuminuria and risk of cardiovascular events, death, and heart failure in diabetic and nondiabetic individuals. JAMA 2001; 286: The Scottish Government: Equally Well: Report of the Ministerial Task Force on Health Inequalities Volume ; Dahlgren G, Whitehead M. Tackling inequalities in health: what can we learn from what has been tried? Working paper prepared for the King s Fund International Seminar on Tackling Inequalities in Health, September 1993 Ditchley Park, Oxfordshire King s Fund. 5. Hossain MP, Palmer D, Goyder E et al. Social deprivation and prevalence of chronic kidney disease in the UK: workload implications for primary care. QJM 2012; 105: Udayaraj UP, Ben Shlomo Y, Roderick P et al. Socio-economic status, ethnicity and geographical variations in acceptance rates for renal replacement therapy in England and Wales: an ecological study. J Epidemiol Community Health 2010; 64: Coresh J, Wei GL, McQuillan G et al. Prevalence of high blood pressure and elevated serum creatinine level in the United States: findings from 202 Kidney International (2014) 85,

6 EP McQuarrie et al.: Deprivation and native renal biopsy clinical investigation the third National Health and Nutrition Examination Survey ( ). Arch Intern Med 2001; 161: Al Qaoud TM, Nitsch D, Wells J et al. Socioeconomic status and reduced kidney function in the Whitehall II Study: role of obesity and metabolic syndrome. Am J Kidney Dis 2011; 58: Hossain MP, Goyder EC, Rigby JE et al. CKD and poverty: a growing global challenge. Am J Kidney Dis 2009; 53: McQuarrie EP, Mackinnon B, Young B et al. Centre variation in incidence, indication and diagnosis of adult native renal biopsy in Scotland. Nephrol Dial Transplant 2009; 24: The Scottish Government. Scottish Neighbourhood Statistics, 2013; McConnachie A, Weir C. Evaluation of Statistical Techniques in the Scottish Index of Multiple Deprivation document by the University of Glasgow 2005; Aghaie-Jaladerany H, Cowell D, Geddes CC. The early impact of the United Kingdom Chronic Kidney Disease (CKD) guidelines on the number of new attendances at renal clinics. Scott Med J 2007; 52: Watt G. The inverse care law today. Lancet 2002; 360: Yamamoto R, Nagasawa Y, Shoji T et al. Cigarette smoking and progression of IgA nephropathy. Am J Kidney Dis 2010; 56: Watt G. Alcohol problems in very deprived areas. Br J Gen Pract 2011; 61: Registrar General for Scotland. Scotland s census 2001: Key Statistics for Council areas and Health Board areas 2003; Cass A, Cunningham J, Wang Z et al. Social disadvantage and variation in the incidence of end-stage renal disease in Australian capital cities. Aust NZ J Public Health 2001; 25: Couchoud C, Guihenneuc C, Bayer F et al. Medical practice patterns and socio-economic factors may explain geographical variation of end-stage renal disease incidence. Nephrol Dial Transplant 2012; 27: The Scottish Government: Scottish Index of Multiple Deprivation 2009: General Report 2009; Scottish Government. Data Sources and Suitability SIMD Overall Index, Kidney International (2014) 85,