After adjusting for potential confounders, we found that patients with kidney stones were 1.82 times more likely to develop CKD during the 10-year follow-up period (HR = 1.82; 95% CI 1.63–2.02) compared with the general population. To the best of our knowledge, this is the first nationwide population-based cohort study to investigate the risk of CKD in patients with kidney stones.
This study focused on acquired or environmental associated kidney stones but not hereditary related cases. To confirm that the stone cohort comprised patients with acquired stones, cases presenting with a history of kidney stones for the past three years (1998–2000) were excluded. Therefore, the exposure times between the cases in the stone cohort were similar. The inclusion criterion for the incident stone cases was the appearance of ICD-9-CM diagnosis codes of kidney stone (a minimum of two times because a single appearance of the code may prove as a false positive).
CKD development requires a long period of exposure to the risk factors; therefore, we observed the patients for a period of 10 years.21 As with kidney stones, the identification of CKD also relied on the appearance (at least two times) of the ICD-9-CM diagnosis codes associated with CKD.
Potential confounders were identified under conditions that were more stringent: at least three appearance of the ICD-9-CM codes for CKD-related comorbidities. In addition to diabetes, hypertension, hyperlipidemia, and cardiovascular diseases were more common in the stone cohort than in the controls (Table 1).
Kidney stones are more prevalent in middle-aged males, with a male to female ratio of approximately 2:1. The incidence of CKD in patients with kidney stones was approximately 11.2%, which was a crude estimation based on 470 patients out of the 4197 experiencing CKD in the kidney stones cohort group.
Kidney stones are the predictors and risk factors of CKD, and their importance in the present study can be noted by their high prevalence rates (5–10%), OR (1.94), and HR (1.82), all of which were second to only those of diabetes (Tables 3 and 4).
The strength of the current study is that it is a nationwide population-based cohort study that could demonstrate the causal relationship between kidney stones and CKD. Because of the mandatory insurance policy of the NHI, the integrity and representativeness of the information is thorough compared with other regional research surveys. Moreover, Taiwan’s NHI has been in existence for more than two decades (since 1995),22,23 providing a sufficient time period to make observations; the rate of data loss is very low, unless the insurant has emigrated to another country or the insurance is suspended because of other special reasons. Zero values were found to be missing during the sampling of this study. Therefore, our data, which covers a national source of information, are highly representative and can be used as reference for developing epidemiology, public health research, health care policy, and clinical guidelines. Although several local and small-scale studies have explored the relationship between kidney stones and CKD,17–20 a nationwide population-based cohort study has not been published so far.
However, this study has some limitations. First, the NHIRD does not provide detailed information on precise relevant clinical variables such as laboratory data and imaging or pathologic findings for the kidney disease. Information about the severity of kidney stones and staging of CKD are also lacking. Second, the level of evidence derived from cohort studies is generally lower than that from randomized trials because of potential biases related to unknown confounders that cannot be adjusted for.24 Third, data on symptoms related to kidney stones, such as hematuria, colicky pain, obstruction, and infection, were unavailable, as was information on relevant therapies, such as lithotripsy or stone surgery, and medications used during treatment. These factors may have potentially affected the analysis in the current study.
Kidney stones contributed to only a small proportion of ESRD patients receiving hemodialysis,5,6 and the contribution was much lower than that of diabetes, hypertension, and hyperlipidemia. Given the slow progression of the clinical course of CKD and the minimal signs and symptoms encountered, patients often neglect the presence of this condition. Furthermore, the effect of kidney stones on CKD is probably underestimated by the clinical doctors, government health authorities, health care institutions, and policy decision makers.