Update on the Burden of CKD

Recent studies confirm that the global burden of CKD is high but point to interesting trends and methodologic challenges. Data on CKD burden have grown dramatically over the past years, building on the foundation set by standardizing the definition and staging of CKD in 2002. I will briefly summarize recent landmark studies that estimate the global burden of CKD and ESRD better than ever before and key recent papers on CKD prevalence trends. These data will be discussed in the context of the methodologic challenges that we must surpass to attain a good understanding of CKD burden.

The global prevalence of CKD was estimated in two meta-analyses.^1,2 Their results were remarkably similar, although both note major limitations and methodologic heterogeneity in the studies reviewed. A meta-analysis of 44 country prevalence studies estimated the worldwide prevalence of CKD at 13.4% (95% confidence interval [95% CI], 11.7% to 15.1%).² A meta-analysis of 33 prevalence studies estimated worldwide prevalence of CKD by sex at 10.4% in men (95% CI, 9.3% to 11.9%) and 11.8% in women (95% CI, 11.2% to 12.6%).¹ Interestingly, prevalence estimates were approximately 15% higher in low- and middle-income countries compared with high-income countries. Prevalence data in low- and middle-income countries were expanded by a study of convenience samples from 12 countries in six world regions.³ They found a prevalence of 14.3% (95% CI, 14.0% to 14.5%) in general populations and 36.1% (95% CI, 34.7% to 37.6%) in high-risk populations. This study also confirmed that awareness of CKD remains very low (6% in the general populations and 10% in high-risk populations).

Studies of CKD prevalence in the United States and Europe recently provided some interesting insights. In the United States, several decades of rising CKD prevalence^4,5 seem to have reached a plateau in the mid-2000s.⁶ Interestingly, this reflected a mixed picture—decrease in CKD prevalence at older age in the face of a continued rise in high-risk groups, including blacks and patients with diabetes. Among United States patients with diabetes, trends are mixed as well.⁷ Albuminuria prevalence is decreasing, particularly among white patients with diabetes. This may be due to the higher rates of treatment with angiotensin-converting enzyme inhibitors, statins, and hypoglycemic medications. However, the prevalence of low eGFR among United States patients with diabetes continued to rise from 1988 to 2014. This may be the result of younger age of onset, improved cardiovascular survival, or other factors. Given the rising prevalence of diabetes and obesity, these results indicate that the net effect of improved treatment versus higher prevalence is uncertain, and the recent report of stabilizing overall prevalence should be taken with caution. Conclusions about trends must be tempered by realizing that, unfortunately, even a 0.04 mg/dl higher mean serum creatinine due to calibration issues can contribute to meaningful changes (5% lower eGFR, 23% higher stage G3 CKD, and 12% higher overall CKD prevalence estimate for the United States).⁴

European studies of CKD burden were recently reviewed, and their results were aggregated.^8,9 Overall, the results support a high prevalence of CKD similar to that of the United States. However, the studies also showed marked variation in prevalence across countries from 3.3% in Norway to 17.3% in northwest Germany. Much of the variation persisted, despite stratification by diabetes and hypertension and attempts to standardize definitions and assays. It is unclear how much of the variation is explained and how much is due to residual methodologic variation. The same group nicely summarized the limitations of the existing literature reporting prevalence of CKD in the European adult general population. The majority of studies did not report the sampling frame used, and response rates varied from 10% to 87%. Additional variation was seen in the creatinine assay (67% Jaffe and 13% enzymatic, with 29% calibrated to isotope dilution mass spectrometry), the equation used to estimate GFR (52% Chronic Kidney Disease Epidemiology Collaboration and 75% Modification of Diet in Renal Disease), definition of CKD (92% used eGFR<60), inclusion of urinary markers (in 60% of studies), and reporting of 95% CIs (39% of studies). The 2013 Kidney Disease Improving Global Outcomes guidelines keep the same definition of CKD, but few of the studies to date incorporate the updated staging, which includes cause, GFR category, and albuminuria category as parallel measures. In summary, the literature is improving, but inferences on the basis of published data are still quite limited.

Our understanding of RRT globally, both provided and needed, was advanced by a recent Lancet publication.¹⁰ Overall, this work estimates that, in 2010, 2.6 million people received RRT worldwide, whereas the number of patients needing RRT is between 4.9 and 9.7 million. It confirmed the very strong relationship between RRT prevalence and per capita income and projected that, with rising global incomes, worldwide use of RRT will more than double to 5.4 million people by 2030, with the most growth in Asia (1.0 million to a projected 2.2 million) and most rapid relative increase projected in Latin America. Data from Asia are still limited, but some studies provide interesting insights. In China, an analysis of 35 million patients hospitalized in tertiary (class 3) Chinese hospitals, which typically have over 500 beds and provide specialist health services, from 2010 to 2015 shows that the percentage of hospitalizations with CKD related to diabetes now exceeds the percentage related to GN.¹¹ This paralleled population-based data. In complete registry data on RRT among 3 million urban insured employees in Nanjing,¹² trends from 2005 to 2014 revealed that, although RRT incidence rates decreased, survival improved, prevalence increased, and importantly, the base urban insured population doubled. Hence, they conclude that there is a real threat of ESRD confronting China during its process of health care transition. In India, the Million Deaths Study,¹³ which used an enhanced verbal autopsy tool between 2001 and 2013, found that 2.9% of deaths were due to renal failure in 2010–2013 among 15 to 69 year olds, an increase of 50% from 2001 to 2013. Diabetes was the largest factor leading to renal failure deaths. This is likely an underestimate given the nonspecific nature of renal symptoms.

The global burden of disease project estimated a 19.6% increase in CKD prevalence from 2005 to 2015 on the basis of a complex Bayesian model integrating multiple sources of data globally.¹⁴ Aging of the global population completely accounted for the CKD rise. Stratification by presumed cause of CKD shows changes of +29%, +23%, +5%, and +16% in CKD due to diabetes, hypertension, GN, and other causes, respectively. In terms of disability adjusted life years among 310 diseases, the rank of CKD rose from 30 in 1990 to 22 in 2005 and 20 in 2015. Likewise, for years lived with disability,¹⁵ the rank of CKD rose from 28 in 1990 to 25 in 2005 and 24 in 2015. This was despite the 2015 estimates being one third lower than the 2013 estimates due to a change in the data for younger ages. Furthermore, the methodology does not include cystic kidney disease and other congenital renal disease under CKD, because they are counted under their own primary disease.

In conclusion, the data now clearly show a large and rising burden of CKD globally. Our ability to firmly quantify CKD will be improved by continued efforts to standardize serum creatinine and albuminuria measures, because they strongly influence estimates of CKD prevalence and staging; integrate CKD measures into a wider range of studies; use the latest equations and staging system in all publications; establish registries and surveillance systems; and ultimately, improve CKD awareness as a foundation for research and action. To the extent that we can measure it with current methods, the growing CKD burden parallels the rising prevalence of diabetes and aging of the population. Although treatments have improved for diabetes and hypertension, the total CKD disease burden is still increasing, and the need for improved surveillance and prevention efforts is great.

Disclosures

A provisional patent (J.C., Inker, and Levey) was filed August 15, 2014: “Precise estimation of GFR from multiple biomarkers” PCT/US2015/044567. The technology is not licensed in whole or part to any company. Tufts Medical Center, Johns Hopkins University, and Metabolon Inc (Durham, NC). have a collaboration agreement to develop a product to estimate GFR from a panel of markers.