Association between urinary N-acetyl-β-glucosaminidase activity-urinary creatinine concentration ratio and risk of disability and all-cause mortality.

BACKGROUND: Recent studies have suggested that chronic kidney disease is associated with cardiovascular disease, dementia, and frailty, all of which cause disability and early death. We investigated whether increased activity of urinary N-acetyl-β-glucosaminidase (NAG), a marker of kidney injury, is associated with risk of disability or all-cause mortality in a general population.
METHODS: Follow-up data from the Hidaka Cohort Study, a population-based cohort study of members of a Japanese rural community, were obtained via questionnaires completed by participants or their relatives. Multivariable analyses were used to investigate relations between urinary NAG activity-urinary creatinine concentration ratio and risk of disability or all-cause mortality.
RESULTS: A total of 1182 participants were followed up for a median of 12.4 years. The endpoints were receipt of support under the public long-term care insurance program, and all-cause mortality. A total of 122 participants (10.3%) were reported to be receiving long-term care and 230 (19.5%) had died. After adjustment for cardiovascular risk factors along with physical activity, and using the quartile 1 results as a reference, the odds ratio (OR) for disability was 2.12 [95% confidence interval (95% confidence interval [CI]), 1.04-4.33; p = 0.038) and the hazard ratio (HR) for all-cause mortality was 1.65 (95% CI, 1.05-2.62; p = 0.031) in participants with urinary NAG/creatinine ratio in quartile 4. Similar results were obtained in participants without proteinuria: OR for disability, 2.46 (95% CI, 1.18-5.16; p = 0.017); and HR for all-cause mortality, 1.62 (95% CI, 1.00-2.63; p = 0.049).
CONCLUSIONS: Increased urinary NAG/creatinine ratio was associated with risk of disability or all-cause mortality in a general population.

Introduction

Chronic kidney disease (CKD) is a growing health burden worldwide, and evidence increasingly suggests that it contributes not only to the risk of cardiovascular disease and death [1-3] but also to cognitive impairment and frailty [4-6]. Approximately 13 percent of the Japanese population aged ≥20 years has been reported to have CKD, and its prevalence increases with age [7]. Therefore, there are a considerable number of individuals with CKD among elderly Japanese. However, few studies have investigated whether decreased kidney function is independently associated with risk of disability in elderly individuals in a general population.

Values for estimated glomerular filtration rate (eGFR) represent serum creatinine–based estimates of kidney function and have been widely used in clinical and epidemiological settings. Decreased eGFR may also indicate decreased kidney function as a result of past kidney injury or simply as a result of aging. Therefore, to estimate current kidney injury, a number of biomarkers have been developed [8]. Among these, urinary N-acetyl-β-glucosaminidase (NAG) activity has been used to estimate tubular injury. NAG, a renal-tubular enzyme present in normal urine, is increased in a wide variety of kidney diseases [9-11]; therefore, it has for many years been regarded as a marker of renal tubular injury [8]. However, NAG activity is also closely related to development of kidney disease and enzymuria in patients with predominantly glomerular disorders [11-13]. Urinary NAG activity, reported as a normalized ratio (urinary NAG activity–urinary creatinine concentration ratio, abbreviated here as urinary NAG/creatinine ratio) to control for variation in urine flow rate, is related to urinary protein–creatinine concentration ratio [12, 14]. However, in patients with tubulointerstitial disease, enzymuria is common even in the absence of proteinuria [12]. Based on these findings, determination of urinary NAG activity may be useful for detecting both global kidney injury and non-proteinuric active kidney disease in large populations.

High urinary NAG activity has been reported to be associated with future risk of all-cause mortality and hospitalization in patients with heart failure, and measuring urinary NAG activity in addition to eGFR and urinary albumin excretion has been shown to improve prediction of these events [15]. In a population study, urinary NAG/creatinine ratio has also been shown to be associated with risk of myocardial infarction, ischemic stroke, and all-cause mortality, independently of urinary albumin excretion and cardiovascular risk factors [16]. However, in contrast to the clinical trial results [15], the NAG/creatinine ratio did not add any significant benefit to the baseline risk-prediction model. In patients with diabetes, we found conflicting results from several studies. Baseline urinary NAG activity was shown by two groups to independently predict both macro- and microalbuminuria in patients with type 1 diabetes [13, 17]; however, urinary NAG activity has been reported by another group to have no predictive value for the development of diabetic nephropathy (although this finding was based on data from only a small number of patients) [18]. Furthermore, urinary NAG activity/creatinine ratio has been found to be associated with increased carotid intima-media thickness and carotid atherosclerosis, independently of albuminuria, in patients with type 2 diabetes [19].

In response to the rapid aging of Japan’s population, in 2000 the Japanese government began implementation, via the Long-Term Care Insurance Act [20], of a public long-term care (LTC) insurance program to deliver improved support for the country’s older citizens. To test the hypothesis that urinary NAG activity may be associated with development of disability or death (from any cause) in the elderly, we used data from the Hidaka Cohort Study, a population-based study of a Japanese rural community [21-23], combined with data from the public LTC insurance program, to investigate the relation between baseline urinary NAG/creatinine ratio and risk of disability or all-cause mortality in a general Japanese population.

Participants and methods

Study population

The present study was a follow-up to the Hidaka Cohort Study, a population-based study of risk factors for cardiovascular disease, cancer, diabetes, and death in members of a Japanese rural community [21-23]. Data were collected from 2155 participants of that study, all of whom were aged ≥20 years at the time of the baseline survey in 1993.

The inclusion criterion for the present study was age ≥65 years at initiation of the follow-up study at the end of October 2005 and 1438 participants were eligible for this study. All participants of the age specified would have been eligible to apply for support from the Japanese government’s public LTC insurance program. Recipients of support from this program are considered to have age-related impairment and were therefore defined in this study as having a disability (for details, see Exposure and outcomes).

Participants with a history of cardiovascular disease or cancer recorded at the time of the baseline survey were excluded from subsequent analyses. Additionally, we also excluded participants who had been reported to be bedridden or in need of nursing care at baseline (See Fig 1).

Fig 1

Enrollment of participants in the present follow-up study of the Hidaka Cohort Study.

The study used data from those eligible for support under the Japanese government’s public long-term care insurance program or who were reported to have died.

Ethics statement

The study was carried out in accordance with the Helsinki Declaration and approved by the institutional ethical review board of Hidaka Medical Center. Informed consent was obtained from all participants or their families.

Exposure and outcomes

The primary exposure was kidney injury, as determined by increased urinary NAG/creatinine ratio. The primary outcomes were disability (based on necessity for LTC) and all-cause mortality.

Regarding disability, under the public LTC insurance program, LTC services are provided across various settings from home- and community-based to institutional, ranging from the loan of aids for daily living and home visits from care assistants and nurses for insured persons able to live more or less independently, to day care, respite care, and continuous residential care for those with greater need for support. Thus, there are six grades of disability, and the number of available services is determined accordingly. In each case, the necessity of the requested service must be recognized by the Certification Committee for Insurance, based on their assessment of functional and cognitive status on documents, provided by family physicians, in which the extent of participants’ impairments and the effects on their daily activities are described. Members of the Certification Committee also visit participants to assess their disability in accordance with uniform national standards; therefore, any person receiving a service via the public LTC insurance program is considered to have a disability; therefore, any participant in the present study who was receiving LTC under the program was likewise classified as having a disability. Participants who died during the follow-up period were also regarded as having an event.

Data collection and laboratory tests

The baseline survey was conducted between July and August 1993. The following data were collected: demographics, past medical history, history of diabetes mellitus, history of hypertension, smoking status, whether or not the participants were bedridden, self-reported physical activity (categorized as low, moderate, or severe), body mass index, blood pressure, laboratory test results (including those for cardiovascular risk factors), and urinalysis results, as described for previous studies [21-23]. Most of the blood samples were drawn within 8 h of the participants’ most recent meal; thus, the samples were obtained mainly when they were in a non-fasting state.

Urinary NAG activity was determined with the use of the synthetic substrate sodio m-cresolsulfonphthaleinyl (MCP) N-acetyl-β-d-glucosaminide [24]. MCP N-acetyl-β-d-glucosaminide reacts with NAG to generate MCP and N-acetylglucosamine. The MCP released can be determined in alkaline solution at 580 nm with a spectrophotometer by subtracting the absorbance of a MCP N-acetyl-β-d-glucosaminide substrate blank. The results obtained with this method correlate highly with those of the conventional fluorimetric method, which uses another substrate, 4-methylumbelliferyl N-acetyl-β-d-glucosaminide [24]. Spot urine samples were also used.

To avoid the dilution effects of urine samples due to variation in participants’ water intake, urinary NAG activity was divided by urinary creatinine concentration to give a urinary NAG/creatinine ratio. Both serum and urinary creatinine concentration were measured by the enzymatic method.

The presence or absence of urinary protein was determined by dipstick urinalysis. Estimated glomerular filtration rate (eGFR) was calculated based on the revised equations for eGFR from serum creatinine for Japanese people [25].

The follow-up study

In December 2005, questionnaires were mailed to participants aged ≥ 65 years to ask whether they were receiving a service under the public LTC insurance program. Relatives were asked if relevant participants had died (S1 File). Completed questionnaires were collected between January and March 2006. Therefore, the mean and median follow-up periods for total study population were 11.4 years and 12.4 years, respectively. We also asked in the follow-up study if the participants had needed nursing care at the time of the baseline survey in 1993 (S2 File).

Statistical analyses

Continuous variables, expressed as means and standard deviations or medians and interquartile ranges, were compared by means of analysis of variance or the Kruskal–Wallis non-parametric test. Categorical variables, expressed as proportions, were compared using the chi-square test or the logistic regression model. We also used the Tukey test or Steel–Dwass test to compare data for the continuous variables between the groups. Data for urinary NAG/creatinine ratio were divided into quartiles, and risk of disability or death for quartiles 2, 3, and 4 relative to that for quartile 1 were calculated by means of the multiple logistic regression model for estimating the risk of disability and the Cox proportional hazards model for estimating all-cause mortality. p < 0.05 was considered to indicate a statistically significant difference.

To estimate risk of disability, we used the multiple logistic regression model rather than the Cox proportional hazards model, because data on the exact date on which participants entered the public LTC insurance program had not been collected. Furthermore, this program was implemented from April 2000 and the baseline survey was conducted in 1993; the LTC service was not available between 1993 and March 2000.

We focused on cardiovascular risk factors, because the results of recent studies have shown that most cardiovascular risk factors, such as smoking habit, hypertension, diabetes mellitus, obesity, and dyslipidemia, are also risk factors for CKD [26]. Therefore, these cardiovascular risk factors are potential confounders for the association between kidney injury and subsequent risk of disability or all-cause mortality. Physical activity is another confounder, because participants with low daily activity could have potential disability or disease. Therefore, we added physical activity as a covariate to the multivariable model in addition to the cardiovascular risk factors. We used these risk factors in the multivariable models as covariates, including age, sex, current smoker status, body mass index, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol (HDL cholesterol), history of diabetes mellitus, and physical activity.

Urinary NAG activity has been reported to be closely associated with urinary protein concentration [12, 14]. Therefore, to investigate whether determination of urinary NAG activity would be useful for participants without proteinuria, we carried out multivariable analyses for the subgroup of participants without proteinuria. Urinary protein was recorded as a dichotomous variable. The dipstick urinalysis results plus or more than plus were recorded as ‘Present’, and minus or plus minus as ‘Absent’.

SPSS 11.01J software for Windows (SPSS, Japan, Tokyo, Japan) and was used to perform the statistical analyses. Additionally, we used R version 4.1.2 (the R Foundation for Statistical Computing, Vienna, Austria) software for the Steel–Dwass test.

Results

Of the 2155 participants comprising the baseline population of the Hidaka Cohort Study, 1438 participants aged ≥ 65 years at the initiation of the follow-up study at the end of October in 2005 were eligible for this study. A total of 1393 questionnaires were collected, giving a response rate of 96.9%. After excluding participants in bedridden (n = 13) or in need of nursing care (n = 12), with a history of cardiovascular disease or cancer (n = 135) at the time of the baseline survey, lost to follow-up (n = 45), or for whom data were missing (n = 51), data from 1182 participants were included in subsequent analyses (Fig 1). A total of 830 (70.2%) were not receiving LTC and were not reported to have died; they are subsequently reported as the no disability and alive (NA) group. Of the remainder, 122 (10.3%) were receiving LTC and 230 (19.5%) were reported to have died.

Table 1 summarizes the baseline characteristics of the 1182 participants whose data were included in the present study. Participants in the Disability or the Had died group were significantly older than those in the NA group (mean age ± standard deviation, 63.9 ± 6.3 years for NA group, 72.1 ±6.7 years for Disability group, and 75.7 ± 8.4 years for Had died group, respectively; p < 0.001) and had decreased eGFR (p < 0.001), reflecting their higher proportion of CKD patients (p < 0.001). The considerable proportion of CKD patients in the total cohort was consistent with the proportion reported previously [7]. Notably, over forty percent of participants in the Had died group had CKD.

Table 1

Baseline characteristics of the participants not receiving long-term care and not reported to have died (NA group), participants receiving long-term care (Disability group), and participants reported to have died (Had died group).

Characteristic	Total cohort	NA group	Disability group	Had died group	pb
No. of participants	1182	830	122	230	NA
Observation period, years	12.4 (12.4–12.5)	12.5 (12.4–12.5)	12.5 (12.4–12.5)	6.8 (4.4–10.1)	ND
Age, years	67.1 ± 8.4	63.9 ± 6.3	72.1 ±6.7**	75.7 ± 8.4**	< 0.001
Female,%	58.4	59.6	70.5*	47.4**	< 0.001
History of hypertension, %	42.3	40.5	55.7**	41.7	0.006
History of diabetes, %	6.3	6.1	7.4	6.1	0.865
Current smoker, %	21.5	19.8	14.8	31.3**	< 0.001
CKDc, %	22.4	16.4	27.9**	41.3**	< 0.001
Presence of Af, %	1.6	0.8	1.6	4.3**	0.001
Physical activity, %			** d	** d	< 0.001b
Low	37.7	29.4	45.9	63.5	NA
Moderate	36.0	42.0	33.6	15.7	NA
High	26.2	28.6	20.5	20.9	NA
Proteinuria, %	4.5	2.3	8.2	10.4	< 0.001
Body mass index, kg/m2	22.5 ± 3.1	22.9 ± 3.0	22.1 ± 3.0*	21.6 ± 3.2**	< 0.001
SBP, mmHg	138 ± 22	137 ± 21	144 ± 21**	139 ± 23	0.001
DBP, mmHg	78 ± 12	78 ± 12	79 ± 13	77 ± 12	0.333
HbA1c, %	5.2 (5.0–5.6)	5.2 (5.0–5.6)	5.2 (5.0–5.6)	5.2 (4.9–5.6)	0.444
TC, mg/dL	202 ± 37	206 ± 35	205 ± 40	188 ± 37**	< 0.001
HDL cholesterol, mg/dL	58 ± 14	57 ± 14	59 ± 14	59 ± 15	0.249
Non-HDL cholesterol, mg/dL	144 ± 39	148 ± 37	146 ± 41	129 ± 39**	< 0.001
Triglyceride, mg/dL	93 (69–134)	98 (72–139)	87 (64–123)*	85 (63–124)**	< 0.001
BUN, mg/dL	16.1 (13.5–19.1)	16.0 (13.4–18.6)	16.5 (13.6–20.2)	16.6 (13.7–19.8)	0.054
Serum creatinine, mg/dL	0.7 (0.6–0.8)	0.7 (0.6–0.8)	0.7 (0,6–0.8)	0.8 (0.7–1.0)**	< 0.001
eGFR, mL/min/1.73 m2	71 (61–78)	73 (64–78)	65 (58–74)**	63 (52–73)**	< 0.001
Urinary creatinine, g/L	0.9 (0.6–1.2)	0.9 (0.6–1.3)	0.8 (0.6–1.1)	0.9 (0.5–1.2)	0.556
Urinary NAG activity, U/L	3.9 (2.5–6.3)	3.7 (2.5–5.6)	4.6 (3.1–7.7)**	4.9 (2.8–8.2)**	< 0.001
Urinary NAG/creatinine ratio, U/g L	4.6 (3.3–6.7)	4.3 (3.1–6.1)	5.9 (4.0–7.9)**	5.8 (4.1–8.2)**	< 0.001

Af, atrial fibrillation; BUN, blood urea nitrogen; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HbA1c, glycosylated hemoglobin A1c; HDL, high-density lipoprotein; NAG, N-acetyl-β-d-glucosaminidase; NA, not applicable; ND, not determined; SBP, systolic blood pressure; TC, total cholesterol; urinary NAG/creatinine ratio, urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio.

a Values expressed as mean ± standard deviation or median (interquartile range).

b Analysis of variance or the Kruskal–Wallis non-parametric test for continuous variables and the chi-square test for categorical variables were used to estimate p values for differences among the NA, Disability, and Had died groups.

c The percentage of participants with eGFR <60 mL/min/1.73 m2.

d Expressing the p value for differences of low, moderate, and high physical activity categories between the NA and Disability groups or between the NA and Had died groups based on the chi-square test.

* p < 0.05 for the difference between the NA and Disability groups or the NA and Had died groups (Tukey, Steel–Dwass test, and the chi-square test).

** p < 0.01 for the difference between the NA and Disability groups or the NA and Had died groups (Tukey, Steel–Dwass test, and the chi-square test).

Participants in the Had died group were more likely to have current smoker status (p < 0.001), atrial fibrillation (p = 0.001), proteinuria (p < 0.001), and low physical activity (p < 0.001) than those in the NA and Disability groups. Other significant differences in physical activity were found between the NA and Disability groups (p = 0.001) and between the NA and Had died groups (p < 0.001). Participants in the Disability group were also more likely to have had a history of hypertension and had significantly higher systolic blood pressure (< 0.001 for both variables). We also found increased urinary NAG activity and urinary NAG/creatinine ratio in both the Disability group and the Had died group (p < 0.001 for both variables).

Table 2 shows baseline characteristics and risk factors according to urinary NAG/creatinine ratio quartile. We found significantly higher values for age (p < 0.001), proportion of participants with a history of diabetes (p < 0.001), proportion of participants with proteinuria (p < 0.001), systolic blood pressure (p = 0.001), and glycosylated hemoglobinA1c (p = 0.001), and significantly lower physical activity (p = 0.001), between urinary NAG/creatinine ratio quartiles. By contrast, we found no differences among them in terms of eGFR (p = 0.584).

Table 2

Baseline characteristics and risk factors according to urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio (urinary NAG/creatinine) quartile.

Baseline characteristic	Urinary NAG/creatinine quartile				pb
	1	2	3	4
No. of participants	295	296	295	296	NA
Urinary NAG/creatinine	0.3–3.3	3.3–4.6	4.6–6.7	6.7–34.2	NA
Age, years	63.6 ± 7.2	66.1 ± 7.3**	68.7 ± 8.8**	69.8 ± 8.6**	< 0.001
Female, %	52.5	60.8*	61.0*	59.5	0.097
History of hypertension, %	38.6	39.5	42.0	49.0*	0.047
History of diabetes, %	2.4	4.7	6.4*	11.5**	< 0.001
Current smoker, %	19.7	21.3	22.4	22.6	0.812
Presence of Af, %	0.3	1.4	1.7	3.0*	0.135
Physical activity, %
Low	28.1	37.2	39.0	46.6	reference
Moderate	42.7	37.8*	33.9**	29.7**	< 0.001c
High	29.2	25.0*	27.1	23.6**	0.010c
Proteinuria, %	1.4	4.1	3.7	8.8**	0.001
Body mass index, kg/m2	22.8 ± 2.9	22.3 ± 2.9	22.4 ± 3.3	22.5 ± 3.2	0.222
SBP, mmHg	134 ± 20	136 ± 22	138 ± 22	142 ± 22**	< 0.001
DBP, mmHg	78 ± 11	78 ± 13	78 ± 12	78 ± 12	0.996
HbA1c, %	5.2 (4.9–5.5)	5.2 (5.0–5.6)	5.2 (4.9–5.6)	5.4 (5.0–5.8)**	0.001
TC, mg/dL	203 ± 36	202 ± 35	203 ± 38	200 ± 38	0.572
HDL-C, mg/dL	57 ± 14	59 ± 14	58 ± 14	57 ± 14	0.068
Non-HDL-C, mg/dL	146 ± 38	143 ± 36	145 ± 40	143 ± 41	0.634
Triglyceride, mg/dL	98(60–137)	92(61–123)	93(64–122)	92(61–122)	0.133
BUN, mg/dL	16.5 (13.8–19.7)	16.2 (13.8–19.1)	15.7 (13.4–18.9)	15.6 (13.3–18.9)	0.151
Serum creatinine, mg/dL	0.7 (0.7–0.9)	0.7 (0.6–0.8)	0.7 (0.6–0.8)*	0.7 (0.6–0.8)	0.027
eGFR, mL/min/1.73 m2	67 (63–77)	72 (62–77)	72 (61–79)	70 (58–79)	0.584
Urinary creatinine, mg/dl	1.1(0.7–1.4)	0.9(0.6–1.2)	0.8(0.5–1.1)	0.8(0.5–1.1)	ND
Urinary NAG activity, U/L	2.6(1.7–3.5)	2.9(0.8–3.6)	4.5(2.9–6.2)	7.5(4.5–10.5)	ND

Af, atrial fibrillation; BUN, blood urea nitrogen; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HbA1c, glycosylated hemoglobin A1c; HDL-C, high-density lipoprotein-cholesterol; NAG, N-acetyl-β-d-glucosaminidase; NA, not applicable; SBP, systolic blood pressure; TC, total cholesterol.

a Values expressed as mean ± standard deviation or median (interquartile range).

b Analysis of variance or the Kruskal–Wallis non-parametric test for continuous variables and the logistic regression model for categorical variables were used to estimate p values for differences among quartiles.

c p values for moderate physical activity quartiles or high physical activity quartiles compared with low physical activity quartiles were estimated using the logistic regression model.

* p < 0.05 for the difference between the Q2, Q3, or Q4 group versus the Q1 group (Tukey or Steel–Dwass test, and the logistic regression model).

** p < 0.01 for the difference between the Q2, Q3, or Q4 group versus the Q1 group (Tukey or Steel–Dwass test, and the logistic regression model).

Table 3 shows correlations between baseline urinary NAG/creatinine ratio and eGFR and other cardiovascular risk factors and kidney injury markers. Urinary NAG/creatinine ratio correlated significantly with age, systolic blood pressure, and glycosylated hemoglobin A1c, suggesting a high degree of correlation between it and traditional cardiovascular risk factors. eGFR correlated significantly with age, systolic blood pressure, blood urea nitrogen, urinary creatinine, and urinary NAG activity.

Table 3

Correlations between baseline urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio (urinary NAG/creatinine ratio), estimated glomerular filtration rate (eGFR), and other variables.

Variable	Urinary NAG/creatinine ratio		eGFR
	R	p	R	P
Age	0.31	<0.001*	–0.39	<0.001*
Body mass index	–0.03	0.357	–0.02	0.539
SBP	0.11	<0.001*	–0.09	0.003*
DBP	–0.01	0.760	0.00	0.913
HbA1c	0.12	<0.001*	0.02	0.574
TC	–0.03	0.269	–0.01	0.769
HDL cholesterol	–0.01	0.700	0.07	0.011*
Non-HDL cholesterol	–0.03	0.369	–0.04	0.144
Triglyceride	–0.07	0.016*	–0.05	0.118
BUN	–0.06	0.041*	–0.28	<0.001*
Serum creatinine	–0.05	0.097	–0.75	<0.001*
eGFR	–0.03	0.352	1.00	NA
Urinary creatinine	–0.17	<0.001*	–0.11	<0.001*
Urinary NAG activity	0.58	<0.001*	–0.11	<0.001*
Urinary NAG/creatinine ratio	1.00	NA	–0.03	0.352

BUN, blood urea nitrogen; DBP, diastolic blood pressure; HbA1c, glycosylated hemoglobin A1c; HDL, high-density lipoprotein; NAG, N-acetyl-β-d-glucosaminidase; SBP, systolic blood pressure; TC, total cholesterol.

* p<0.05.

Table 4 summarizes the results of univariate and multivariable-adjusted analyses for subsequent risk of disability. The results of univariate analyses with continuous variables showed this risk to be significantly higher in participants with higher NAG/creatinine ratio (odds ratio [OR], 1.15; 95% confidence interval [CI], 1.09–1.21; p < 0.001). We also found a significant increased risk of disability in participants with proteinuria than in those without proteinuria (OR, 3.81; 95% CI, 1.73–8.40; p < 0.001). Also using the results for quartile 1 as a reference, we found a significantly lower risk in participants with eGFR in quartile 3 than in those with eGFR in quartile 1 (OR, 0.28; 95% CI, 0.16–0.51; p <0.001). The ORs of urinary NAG/creatinine ratio in quartiles 3 and 4 were significantly higher: 3.29 (95% CI, 1.70–6.34; p <0.001) and 4.48 (95% CI, 2.36–8.51; p < 0.001), respectively.

Table 4

Univariate and multivariable-adjusted odds ratios (ORs) for subsequent disability, by kidney injury marker.

Kidney injury marker	Univariate analysis		Multivariable model 1c		Multivariable model 2d
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Urinary NAG/creatinine ratioe	1.15 (1.09–1.21)	<0.001*	1.10 (1.04–1.17)	0.002*	1.10 (1.04–1.17)	0.002*
eGFRe	0.97 (0.96–0.99)	<0.001*	1.00 (0.98–1.02)	0.981	1.00 (0.98–1.02)	0.964
Urinary proteinf	3.81 (1.73–8.40)	0.001*	3.66 (1.48–9.04)	0.005*	3.65 (1.48–9.02)	0.005*
eGFR quartile
1	1 (reference)		1 (reference)		1 (reference)
2	0.49 (0.30–0.81)	0.005*	1.26 (0.69–2.31)	0.449	1.27 (0.69–2.33)	0.438
3	0.28 (0.16–0.51)	<0.001*	0.74 (0.38–1.43)	0.371	0.76 (0.39–1.46)	0.406
4	0.45 (0.27–0.75)	0.002*	1.65 (0.87–3.12)	0.124	1.70 (0.90–3.23)	0.104
Urinary NAG/creatinine ratio quartile
1	1 (reference)		1 (reference)		1 (reference)
2	1.76 (0.87–3.58)	0.12	1.27 (0.59–2.74)	0.537	1.27 (0.59–2.73)	0.544
3	3.29 (1.70–6.34)	<0.001*	1.96 (0.95–4.02)	0.067	1.96 (0.95–4.02)	0.068
4	4.48 (2.36–8.51)	<0.001*	2.14 (1.05–4.36)	0.036*	2.12 (1.04–4.33)	0.038*

CI, confidence interval; eGFR, estimated glomerular filtration rate; urinary NAG/creatinine ratio, urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio.

a Odds ratios were estimated by means of the multiple logistic regression model.

b Data for participants reported to have died (n = 230) were excluded from the analysis.

c Multivariable model 1 adjusted for traditional cardiovascular risk factors including age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, high-density lipoprotein cholesterol, and history of diabetes mellitus.

d Multivariable model 2 was adjusted for physical activity in addition to the traditional cardiovascular risk factors used in multivariable model 1.

e Included as a continuous variable.

f Risk for participants with proteinuria relative to that for participants without proteinuria.

* p < 0.05.

In multivariable-adjusted model 1, which included the traditional cardiovascular risk factors including age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, high-density lipoprotein-cholesterol (HDL cholesterol), and history of diabetes mellitus, urinary NAG/creatinine ratio was also significantly associated with risk of disability (OR, 2.14; 95% CI, 1.05–4.36; p = 0.036 for quartile 4 vs quartile 1; OR, 1.10; 95% CI, 1.04–1.17; p = 0.002 when urinary NAG/creatinine ratio was treated as a continuous variable). To avoid the confounder of underlying disability at the time of the baseline survey, we added physical activity as a covariate to multivariable model 2 in addition to the traditional cardiovascular risk factors. Even after adjustment for these covariates, the association of urinary NAG/creatinine ratio with risk of disability was similar to that found under model 1 (OR, 2.12; 95% CI, 1.04–4.33; p = 0.038).

Table 5 shows univariate and multivariable-adjusted hazard ratios (HRs) for subsequent all-cause mortality by kidney injury marker. The results of univariate analysis showed all the kidney injury markers to be significantly associated with risk of all-cause mortality: HR, 3.12 (95% CI, 2.04–4.76; p <0.001) for presence of urinary protein; HR, 0.32 (95% CI, 0.22–0.47; p <0.001) for eGFR quartile 3 vs quartile 1; and HR, 3.78 (95% CI, 2.44–5.86; p <0.001) for urinary NAG/creatinine ratio quartile 4 vs quartile 1. Even after adjustment for traditional cardiovascular risk factors including age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, HDL cholesterol, and history of diabetes mellitus, urinary NAG/creatinine ratio was significantly associated with risk of all-cause mortality. We obtained a similar result with multivariable model 2, which included physical activity as a covariate.

Table 5

Univariate and multivariable-adjusted hazard ratios (HRs) for subsequent all-cause mortality, by kidney injury marker.

Kidney injury marker	Univariate analysis		Multivariable model 1b		Multivariable model 2c
	HR (95% CI)	p	HR (95% CI)	p	HR (95% CI)	p
Urinary NAG/creatinine ratiod	1.09 (1.07–1.12)	<0.001*	1.07 (1.04–1.10)	<0.001*	1.07 (1.03–1.10)	<0.001*
eGFRd	0.97 (0.96–0.98)	<0.001*	1.00 (0.99–1.01)	0.474	1.00 (0.99–1.01)	0.662
Urinary proteine	3.12 (2.04–4.76)	<0.001*	2.01 (1.30–3.09)	0.002*	1.97 (1.28–3.03)	0.002*
eGFR quartile
1	1 (reference)		1 (reference)		1 (reference)
2	0.37 (0.26–0.53)	<0.001*	0.77 (0.54–1.11)	0.166	0.77 (0.53–1.10)	0.152
3	0.32 (0.22–0.47)	<0.001*	0.68 (0.46–1.02)	0.059	0.73 (0.49–1.08)	0.116
4	0.37 (0.26–0.53)	<0.001*	1.00 (0.67–1.50)	0.994	1.05 (0.70–1.58)	0.800
Urinary NAG/creatinine ratio quartile
1	1 (reference)		1 (reference)		1 (reference)
2	1.91 (1.19–3.09)	0.008*	1.36 (0.84–2.22)	0.213	1.34 (0.82–2.18)	0.239
3	2.81 (1.79–4.42)	<0.001*	1.34 (0.84–2.15)	0.218	1.30 (0.81–2.09)	0.271
4	3.78 (2.44–5.86)	<0.001*	1.72 (1.09–2.72)	0.019*	1.65 (1.05–2.62)	0.031*

CI, confidence interval; eGFR, estimated glomerular filtration rate; urinary NAG/creatinine ratio, urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio.

a Estimated by means of the Cox regression model.

b Multivariable model 1 was adjusted for age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, high-density lipoprotein cholesterol, and history of diabetes mellitus.

c Multivariable model 2 was adjusted for physical activity in addition to the variables used in multivariable model 1.

d Included as a continuous variable.

e Risk for participants with proteinuria relative to that for participants without proteinuria.

* p < 0.05.

High urinary NAG/creatinine ratio was therefore associated with both higher risk of disability (see Table 4) and all-cause mortality (see Table 5). Furthermore, these risks (especially that for disability) increased in a stepwise fashion in line with increased urinary NAG/creatinine ratio.

Table 6 shows multivariable-adjusted ORs for disability and HRs for all-cause mortality adjusted for cardiovascular risk factors and kidney injury markers. Among the potential risk factors, only kidney injury markers were independently associated with subsequent risk of disability. We also found current smoker status, low total cholesterol, urinary protein, and urinary NAG/creatinine ratio to be independently associated with risk of all-cause mortality. Therefore, among various risk factors, only kidney injury markers, urinary protein, and urinary NAG/creatinine ratio were significantly associated with both subsequent risk of disability and all-cause mortality.

Table 6

Multivariable-adjusted odds ratios (ORs) for disability and hazard ratios (HRs) for all-cause mortality adjusted for cardiovascular risk factors and kidney injury markers.

Variablea	Disability		All-cause mortality
	OR (95% CI)	p	HR (95% CI)	p
Male sex	0.69 (0.37–1.28)	0.236	1.34 (0.95–1.89)	0.091
Age	1.18 (1.14–1.23)	<0.001*	1.13 (1.11–1.16)	<0.001*
History of diabetes	1.03 (0.44–2.40)	0.944	1.04 (0.60–1.80)	0.890
Current smoker status	1.23 (0.59–2.55)	0.580	1.85 (1.30–2.63)	0.001*
Body mass index	0.95 (0.88–1.03)	0.183	0.98 (0.94–1.03)	0.423
SBP	1.01 (1.00–1.02)	0.173	1.00 (0.99–1.00)	0.181
TC	1.00 (0.99–1.01)	0.729	0.99 (0.99–1.00)	<0.001*
HDL cholesterol	1.01 (0.99–1.03)	0.236	1.01 (1.00–1.02)	0.197
Physical activity		0.483b		0.016b*
Low	1 (reference)	NA	1 (reference)	NA
Moderate	0.82 (0.49–1.37)	0.440	0.62 (0.42–0.92)	0.018*
High	0.71 (0.39–1.28)	0.252	0.66 (0.46–0.95)	0.027*
eGFR	1.00 (0.98–1.02)	0.940	1.00 (0.99–1.01)	0.949
Urinary protein	3.32 (1.32–8.34)	0.011*	1.74 (1.11–2.72)	0.015*
Urinary NAG/creatinine	1.10 (1.03–1.17)	0.003*	1.06 (1.03–1.09)	<0.001*

CI, confidence interval; eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; NA, not applicable; SBP, systolic blood pressure; TC, total cholesterol; urinary NAG/creatinine, urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio.

a All variables were included in the logistic regression model and the Cox proportional regression model to estimate the risk associated with each variable.

b Expressed as p value among three categories of physical activity.

*p < 0.05.

We found presence of urinary protein to be an independent risk factor for risk of disability and all-cause mortality. Therefore, we investigated whether urinary NAG/creatinine ratio could predict subsequent risk of disability and all-cause mortality in participants without proteinuria. We performed a multivariable analysis adjusted for age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, HDL cholesterol, history of diabetes mellitus, and physical activity in participants without proteinuria.

Table 7 shows the ORs and HRs for disability and all-cause mortality, stratified by urinary NAG/creatinine ratio quartiles in participants without proteinuria. The participants with urinary NAG/creatinine ratio in quartile 4 were at higher risk for disability (OR, 2.46; 95% CI, 1.18–5.16; p = 0.017) and all-cause mortality (HR, 1.62; 95% CI, 1.00–2.63; p = 0.049). These risks increased in a stepwise fashion in line with increased NAG/creatinine ratio.

Table 7

Multivariable-adjusted odds ratios (ORs)a and hazard ratios (HRs) for subsequent disability and all-cause mortality according to urinary N-acetyl-β-d-glucosaminidase activity–urinary creatinine concentration ratio (urinary NAG/creatinine ratio) quartile in participants without proteinuria.

Urinary NAG/creatinine ratio quartile	Disability		All-cause mortality
	OR (95% CI)	p	HR (95% CI)	p
1	1 (reference)	NA	1 (reference)	NA
2	1.09 (0.48–2.46)	0.843	1.33 (0.80–2.22)	0.274
3	1.99 (0.94–4.23)	0.074	1.41 (0.87–2.31)	0.164
4	2.46 (1.18–5.16)	0.017*	1.62 (1.00–2.63)	0.049*

CI, confidence interval; NA, not applicable.

a Both ORs and HRs were estimated by means of the logistic regression model and the Cox regression model adjusted for age, sex, systolic blood pressure, current smoker status, body mass index, total cholesterol, high-density lipoprotein-cholesterol, history of diabetes mellitus, and physical activity.

* p < 0.05.

Discussion

In this community-based cohort study of 1182 persons who were apparently healthy at the time of the baseline survey, increased urinary NAG/creatinine ratio was found to be associated with subsequent risk of disability or all-cause mortality, independent of traditional cardiovascular risk factors including age. The risk increased in a stepwise fashion from the lowest to the highest urinary NAG/creatinine ratio quartile. This relation remained significant even in participants without proteinuria. Furthermore, among various cardiovascular risk factors, only kidney injury markers were associated with both the risk of disability and the risk of all-cause mortality in this cohort, in which over twenty percent of participants had CKD. Therefore, we should focus on CKD to reduce future risk of disability and death.

A number of studies have been carried out to investigate the relation between CKD and cognitive impairment and frailty. However, the mechanism by which CKD leads to subsequent disability is not yet fully understood. One possible mechanism may be failure of appropriate blood pressure control. A previous study revealed that even in the early stage of CKD, there are considerable increases in nocturnal blood pressure [27]. However, during the follow-up period, nocturnal hypertension had not been well-recognized by most physicians because ambulatory blood pressure monitoring had not yet been approved for use in routine clinical practice during the follow-up period. Therefore, participants with potentially decreased kidney function might not have received treatment to achieve appropriate blood pressure control. By contrast, participants who were recognized as having daytime hypertension at the time of the baseline survey might have received appropriate treatment, and therefore baseline high blood pressure may no longer be associated with subsequent risk of disability and all-cause mortality.

Further support for the idea that CKD is closely related to aging is provided by the finding that patients with CKD have decreased expression of Klotho and increased fibroblast growth factor 23 levels in accordance with CKD stage (1 to 5) [28, 29]. The results of studies using animal models have suggested that Klotho, which as a coreceptor for fibroblast growth factor 23 plays a critical role in phosphate excretion by the kidney, is closely related to accelerated aging and early death [30-33]. This evidence is supported by results of human studies, which have shown increased blood fibroblast growth factor-23 levels to be independently associated with increased risk of vascular and nonvascular mortality and disability in a general population [34, 35]. Furthermore, low plasma Klotho concentrations have been shown to be independently associated with disability [36], cognitive decline [37], and mortality [38] in older, community-dwelling persons.

Chronic kidney disease has been defined as having decreased eGFR, therefore eGFR has been used to assess the severity of CKD [39]. However, eGFR was determined with the use of serum creatinine concentration, along with age and sex as covariates. Therefore, adjusting for age and sex in a multivariable model seems to be ‘over-adjustment’. This may lead to a non-significant association between eGFR and risk of disability and all-cause mortality in multivariable models. We found additional benefit in measuring urinary NAG/creatinine ratio, because we found no differences in eGFR among the urinary NAG/creatinine ratio quartiles (see Table 2). Furthermore, even after adjustment for eGFR, urinary protein, and traditional cardiovascular risk factors, urinary NAG/creatinine ratio remained significantly associated with risk of disability and all-cause mortality. Therefore, measurement of urinary NAG/creatinine ratio may have a benefit in addition to measurement of eGFR and urinary protein.

Limitations

This study has several limitations. First, we did not evaluate disability at the baseline survey in 1993 unless the participant was bedridden, and the Long-Term Care Insurance Act was not passed until 2000. Therefore, we had no contemporary data on the participants’ disabilities at baseline. In the follow-up study, we asked participants retrospectively about their baseline disability status. However, this later information may not have been sufficient to identify participants with mild to moderate disability at the time of the baseline survey. Therefore, we added physical activity as a covariate to the multivariable models in order to adjust for this confounder.

Second, we did not collect data on participants’ disability status but simply asked if they were using the public LTC insurance program. Therefore, participants with a disability but who did not use the service might not have been recognized as having a disability. This may have contributed to reducing the strength of the association between baseline urinary NAG/creatinine ratio and risk of disability in this study.

Third, although urinary protein concentration has commonly been used as a marker of kidney injury and shown to be closely related to urinary NAG/creatinine ratio [12, 14], only the presence or absence of urinary protein was determined in the present study. Therefore, it was not possible to investigate the relation between baseline urinary protein concentration and subsequent risk of disability or all-cause mortality. However, because the great majority of the total cohort did not have proteinuria, and in these participants, urinary NAG/creatinine ratio was associated with subsequent risk of disability and death, determination of urinary NAG activity may predict this risk in most members of a general population.

Fourth, urinary NAG/creatinine ratio was determined only once, at the baseline survey, and > 12 years had passed before the data for disability or all-cause mortality were collected. This may have resulted in underestimation of the relation between urinary NAG/creatinine ratio and subsequent disability or all-cause mortality [40].

Conclusions

Increased urinary NAG/creatinine ratio was associated with both risk of subsequent disability and risk of all-cause mortality, independent of traditional cardiovascular risk factors. The results of our study support the findings of previous studies in suggesting that chronic kidney injury is closely related to aging. For prediction of risk of disability or all-cause mortality, urinary NAG/creatinine ratio was useful even in persons without proteinuria. Therefore, urinary NAG activity may be one of the most useful markers of kidney injury in terms of predicting subsequent risk of disability or all-cause mortality in a general population. This is especially important in aging populations such as that of Japan.

Screening questionnaire concerning long-term care service use Part 1.

(DOC)

Click here for additional data file.

Screening questionnaire concerning long-term care service use Part 2.

(DOCX)

Click here for additional data file.

(XLSX)

Click here for additional data file.

1 Nov 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Yan Li, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please upload an English language copy of the questionnaire as a supplementary file.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

4. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please delete it from any other section.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors evaluated whether increased urinary N-acetyl-β-glucosaminidase activity to creatinine ratio (NCR) sampled from spot urine was predictive of disability or death in a total of 1,209 elderly Japanese. The authors performed multivariable logistic regression and found that subjects with the highest quartile of NCR had increased risk of disability, death, and combination of the two.

Overall, the paper contained valid information on NCR and adverse outcome. However, there are several significant issues that need to be addressed.

Major comments

My first major concern is about the description of the study, which is different from previously published papers (reference 19-21).

1. The baseline of the current study, unlike the what the authors wrote in the abstract, the methods and the results that "this study started in December 2005", was collected at year 1993.

2. Follow-up was conducted more than ten years later, around 2005. The endpoints including disability and death were collected at this period.

The authors should 1) clarify the exact inclusion criteria, e.g., aged >= 65 years at the time of follow-up. 2) correct all related sentences and provide the duration of the follow-up. 3) conduct additional analyses including Cox regression for all-cause mortality, and if possible, for the composite outcome including all-cause mortality and disability, when exact date of disability was possible. 4) modify the study flow chart accordingly (figure 1).

Secondly, given the long-term care act was implemented around year 2000, and that the disability status could not be determined at baseline ("apparently healthy"), the conclusion that NCR PREDICTED risk of disability could not be drawn. Unless additional information was provided, baseline NCR was ASSOCIATED with disability in the cross-sectional survey conducted at follow-up. Meanwhile, were all subjects screened for eligibility of LTC or did they need to apply for it?

Thirdly, additional statistical analysis concern should be addressed apart from the Cox regression as mentioned in the first major comment.

1. The ranges for NCR quartiles, numbers and rates of events should be provided.

2. The Relative Risk should be Odds Ratio in logistic regression. A raw calculation from the data provided in the supplemental file also confirmed the need to replace the term RR with OR.

3. Analysis for continuous NCR should be conducted.

4. Re-classification of predicted risk after adding the NCR into base model should be analyzed, even if non-significant. This is particularly important given the fact that the age of subjects in DD group on average was 10 years older than NA group and reached the healthy life expectancy of 74.1 years.

Fourthly, introduction and discussion section should be revised to give readers a clear and objective view on the current finding.

1. The population included in the current study were not all CKD patients, indeed most of them were free of CKD. First paragraph of introduction focused on CKD, which is not the population, nor the disease need to diagnose in the population. The authors may, for instance, give a summary of kidney biomarkers and population risk here.

2. Details on the act of long-term care and its use in the determination of disability should be put into methods section. Disability, on the other hand, could be discussed here as a general term.

3. The conclusion in the first paragraph of discussion that subjects with proteinuria could not benefit from NCR test is underpowered given there were only 56 subjects with proteinuria. The conclusion therefore should be downplayed.

4. The comparison of current study with existing evidence on NCR and adverse outcome should be discussed firstly. Multiple studies focused on this topic could be retrieved from PubMed search, e.g., the paper by Solbu et al. published in J Am Soc Nephrol (27: 533–542, 2016, doi: 10.1681/ASN.2014100960), compared NCR with albumin-to-creatinine ratio in risk prediction of mortality in a low-risk population.

5. The relationship among NCR/CKD, aging and cardiovascular risk should be clearly illustrated. In the second paragraph of discussion the authors made a bold statement that CKD is related to physiological change of aging rather than that of CVD, which is not appropriate as in the following paragraph the authors explained mechanisms involving in such procedure, i.e., lack of BP control, whereas hypertension is one of the most important CV risk factors.

6. Is there any interplay between NCR and Klotho and FGF23? Please provide pathophysiological and/or epidemiological links here given the current study population was only general but not CKD population. Alternatively, replace them with other evidence on the pathophysiological importance of NCR.

7. Given the current guidelines from KDIGO and other organizations all recommend the use of ACR and eGFR in CKD staging, whether NCR levels were indicative of CKD stage should be addressed before further discussion on CKD.

8. eGFR is an established risk marker of CKD but it should be used in conjunction with others when eGFR is not low (<60ml/min/1.73m2). Current data suggested the population included here is not particularly of impaired renal filtration. Relevant discussion on eGFR in comparison to NCR should be shortened and organized.

Minor comments

To make it less confusing, only keep the number of included subjects in the abstract and delete the total number screened.

Line 127 of page 6, "an" MCP should be "a" MCP.

In the methods section, please specify the name and version of statistical software.

In the methods section, please specify the regression models used for the evaluation.

Table 4 has duplicated results with Table 3 (the "entire cohort"). Similarly Figure 2 is a graphical presentation of some results presented in Table 3. Such items should be removed.

Conclusion "public LTC insurance services" should be replaced by disability or disability determined by public LTC insurance services.

Reviewer #2: In this manuscript, the author has demonstrated that the activity of urinary N-acetyl-β-glucosaminidase (NAG), a marker of kidney injury, is associated with subsequent risk of disability or early death in a general population. However, there exists several problems.

1、 What is the specific date on which to determine whether the events occurred or not？Please describe the medium and IQR of the follow-up period.

2、 The association between the quartile groups of NAG groups and disability or death should be evaluated using the Kaplan-Meier survival method and compared using log-rank statistics.

3、 How about the relationships between urinary NAG and other clinical variables?

4、 The association between the NAG and cardiovascular mortality and renal mortality should better be evaluated.

5、 What is your criteria for selecting independent variables in your multivariable-adjusted model？

6、 The usage of anti-hypertensive drugs, anti-diabetic drug, statins and comorbid conditions should be recorded and adjusted.

7、 In order to determine whether risk prediction models were improved by addition of the NAG, C-index should be calculated for the demographic, eGFR, and cardiovascular risk factor model for each outcome.

8、 Q1 needs to be placed in the table, and the specific values of Q1-Q4 need to be marked.

9、 NAG is one of the urine kidney injury biomarkers, not equals to CKD. These two concepts should not be confused.

10、 In line 277-279, you mentioned that “The association between urinary NAG/creatinine ratio and disability or death, which was found to be independent of cardiovascular risk factors (including age), suggests that CKD is related to the physiological changes of aging rather than those of cardiovascular disease. ”, while I am so confused about this sentence. How can I come to a conclusion that “CKD is related to the physiological changes of aging” from your data? The discussion should be re-arranged.

11、 In your opinion, what is the probably reason that urinary NAG/creatinine ratio can not predict subsequent disability or death in patients with proteinuria?

12、 I think Table 2 and Figure 2 maybe not necessary.

Reviewer #3: The description of the study design in this article is incomplete and lacks specific follow-up methods. There are also doubts about the statistical methods used in this article to illustrate the relationship between urinary NAG/creatinine ratio and the risk of death and disability. So I don't think this article has reached the standard for receiving manuscripts. Questions are as follows:

1. What is the collection method of the outcome event data? At what time were the collection points?

2. Since the beginning of the study in 2005, what was the median follow-up time for all participants? How many subjects were lost to follow-up?

3. The statistical method used in the analysis of survival data in this paper used multivariate logistic regression analysis instead of the Cox risk regression model (Table 2 and Table 3), which ignored the impact of censored data.

4. In Table 4, participants with proteinuria were only 56, which could produce an unstable model.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

29 Dec 2021

Reviewer #1: The authors evaluated whether increased urinary N-acetyl-β-glucosaminidase activity to creatinine ratio (NCR) sampled from spot urine was predictive of disability or death in a total of 1,209 elderly Japanese. The authors performed multivariable logistic regression and found that subjects with the highest quartile of NCR had increased risk of disability, death, and combination of the two.

Overall, the paper contained valid information on NCR and adverse outcome. However, there are several significant issues that need to be addressed.

Major comments

My first major concern is about the description of the study, which is different from previously published papers (reference 19-21).

1. The baseline of the current study, unlike the what the authors wrote in the abstract, the methods and the results that "this study started in December 2005", was collected at year 1993.

Response

Thank you for pointing this out. We have specified the observation period in the results section of the abstract as follows: ‘A total of 1182 participants were followed up for a median of 12.4 years’ (line 30).

2. Follow-up was conducted more than ten years later, around 2005. The endpoints including disability and death were collected at this period.

The authors should 1) clarify the exact inclusion criteria, e.g., aged >= 65 years at the time of follow-up. 2) correct all related sentences and provide the duration of the follow-up. 3) conduct composite outcome including all-cause mortality and disability, when exact date of disability was possible. 4) modify the study flow chart accordingly (figure 1).

Response

1. The inclusion criterion is now clarified in the Study population subsection of Participants and Methods: ‘The inclusion criterion for the present study was age >65 years at initiation of the follow-up study at the end of October 2005’ (lines 96–97).

2. The dates of the baseline survey and follow-up study are now specified in the Data collection and laboratory tests and The follow-up study subsections, respectively, of Participants and Methods: ‘The baseline survey was conducted between July and August 1993’ (line 130) and ‘Completed questionnaires were collected between January and March 2006’ (line 155-156). The follow-up periods are also specified: ‘Therefore, the mean and median follow-up periods for total study population were 11.4 years and 12.4 years, respectively’ (lines 156–157).

3. Participants were not asked their exact date of entry into the LTC insurance program. Therefore, the Cox proportional hazards regression model could not be used to estimate the risk of disability. Instead, we conducted an additional analysis using another Cox regression model to investigate the relation between baseline urinary NAG/creatinine ratio and risk of all-cause mortality. This is explained in the Statistical analyses subsection thus: ‘To estimate risk of disability, we used the multiple logistic regression model rather than the Cox proportional hazards model, because data on the exact date on which participants entered the public LTC insurance program had not been collected.’ (lines 169-171).

4. The participant selection flow chart (Fig 1) has been modified accordingly.

Secondly, given the long-term care act was implemented around year 2000, and that the disability status could not be determined at baseline ("apparently healthy"), the conclusion that NCR PREDICTED risk of disability could not be drawn. Unless additional information was provided, baseline NCR was ASSOCIATED with disability in the cross-sectional survey conducted at follow-up. Meanwhile, were all subjects screened for eligibility of LTC or did they need to apply for it?

Response

We agree with your point and have therefore added information to the analysis. At the baseline survey, we identified participants who were bedridden, and in the follow-up study, we asked participants if they had needed nursing care at the time of the baseline study. Participants who were bedridden or needed nursing care at baseline have been excluded from the analysis. Additionally, we have added baseline physical activity to the model as a covariate, to adjust for potentially underlying disability and disease in participants with low physical activity (‘Physical activity is another confounder, because participants with low daily activity could have potential disability or disease...’; lines 178–180).

As explained in the Limitations subsection of the Discussion, ‘we did not collect data on participants’ disability status but simply asked if they were using the public LTC insurance program’ (lines 422–423); their disability status was confirmed by officials who visited them at home. ‘Therefore, participants with a disability but who did not use the service might not have been recognized as having a disability. This may have contributed to reducing the strength of the association between baseline urinary NAG/creatinine ratio and risk of disability in this study.’ (lines 423–426). It is possible that such participants did not need any government support.

Thirdly, additional statistical analysis concern should be addressed apart from the Cox regression as mentioned in the first major comment.

1. The ranges for NCR quartiles, numbers and rates of events should be provided.

Response

We have added baseline characteristic data for urinary NAG–creatinine ratio quartiles to Table 2. We have also presented the numbers of events in Table 1 and the Results (‘A total of 830 (70.2%) were not receiving LTC and were not reported to have died; they are subsequently reported as the no disability and alive (NA) group. Of the remainder, 122 (10.3%) were receiving LTC and 230 (19.5%) were reported to have died.’; lines 200–202).

2. The Relative Risk should be Odds Ratio in logistic regression. A raw calculation from the data provided in the supplemental file also confirmed the need to replace the term RR with OR.

Response

Thank you for this information, which is important for the presentation of the data. We have changed the relevant terms as advised.

3. Analysis for continuous NCR should be conducted.

Response

We have added the results of multivariable adjusted analyses for continuous NCR to Tables 4, 5, and 6. In each case, NCR was significantly associated with both outcomes.

4. Re-classification of predicted risk after adding the NCR into base model should be analyzed, even if non-significant. This is particularly important given the fact that the age of subjects in DD group on average was 10 years older than NA group and reached the healthy life expectancy of 74.1 years.

Response

Thank you for your important point about the analyses.

We have included all variables in the multivariable-adjusted models that we used in the current study, and we have described every predicted risk for each variable in Table 6.

Fourthly, introduction and discussion section should be revised to give readers a clear and objective view on the current finding.

1. The population included in the current study were not all CKD patients, indeed most of them were free of CKD. First paragraph of introduction focused on CKD, which is not the population, nor the disease need to diagnose in the population. The authors may, for instance, give a summary of kidney biomarkers and population risk here.

Response

Thank you for your helpful comment.

In the Introduction, we refer to the prevalence of CKD in the Japanese population (13% of persons aged >20 years). Additionally, we give the prevalence of CKD (defined as eGFR <60 mL/min/1.73 m2) in our cohort: 22.4% (see Table 1). We also refer to previous reports that high urinary NAG activity is associated with the risk of cardiovascular disease, heart failure, and mortality (‘High urinary NAG activity has been reported to be associated with future risk of all-cause mortality and hospitalization in patients with heart failure, and measuring urinary NAG activity in addition to eGFR and urinary albumin excretion has been shown to improve prediction of these events [15]. In a population study, urinary NAG/creatinine ratio has also been shown to be associated with risk of myocardial infarction, ischemic stroke, and all-cause mortality, independently of urinary albumin excretion and cardiovascular risk factors [16].’; lines 67–72).

2. Details on the act of long-term care and its use in the determination of disability should be put into methods section. Disability, on the other hand, could be discussed here as a general term.

Response

Thank you for this helpful comment. In the Participants and methods section, we now state how we have defined disability in the Study population subsection (‘Recipients of support from this program are considered to have age-related impairment and were therefore defined in this study as having a disability’; lines 99–101) and provide details of long-term care in the Exposure and outcomes subsection (lines 114–126).

3. The conclusion in the first paragraph of discussion that subjects with proteinuria could not benefit from NCR test is underpowered given there were only 56 subjects with proteinuria. The conclusion therefore should be downplayed.

Response

In accordance with your recommendation, we have downplayed this statement in the first paragraph of the Discussion (‘This relation remained significant even in participants without proteinuria.’; lines 375–376).

4. The comparison of current study with existing evidence on NCR and adverse outcome should be discussed firstly. Multiple studies focused on this topic could be retrieved from PubMed search, e.g., the paper by Solbu et al. published in J Am Soc Nephrol (27: 533–542, 2016, doi: 10.1681/ASN.2014100960), compared NCR with albumin-to-creatinine ratio in risk prediction of mortality in a low-risk population.

Response

In the Introduction, we refer to previous reports that focused on the relation between NCR and risk of cardiovascular diseases and mortality (lines 67–80). However, we are willing to move this information to the Discussion, if advised to do so.

5. The relationship among NCR/CKD, aging and cardiovascular risk should be clearly illustrated. In the second paragraph of discussion the authors made a bold statement that CKD is related to physiological change of aging rather than that of CVD, which is not appropriate as in the following paragraph the authors explained mechanisms involving in such procedure, i.e., lack of BP control, whereas hypertension is one of the most important CV risk factors.

Response

We refer to nocturnal hypertension as an example of ‘masked hypertension’. People with this condition may not be recognized as having hypertension in epidemiological studies based on office hypertension, leading to misclassification of the disease (lines 382–388).

We did not perform ambulatory blood pressure monitoring in this study, so we were unable to confirm this hypothesis.

6. Is there any interplay between NCR and Klotho and FGF23? Please provide pathophysiological and/or epidemiological links here given the current study population was only general but not CKD population. Alternatively, replace them with other evidence on the pathophysiological importance of NCR.

Response

NCR simply represents tubular and interstitial injury of the kidney. We were unable to find any interactions between NCR and FGF23 or Klotho. However, all of them relate to kidney function, and considerable increases in NCR and FGF23 and a decrease in Klotho in relation to decreases in kidney function have already been reported.

In addition to studies using experimental animal models, we found a number of epidemiological studies regarding relations between Klotho and FGF23 and the future risk of cardiovascular disease, death, and frailty in the general population. We have added this information to the Discussion (lines 392–401).

7. Given the current guidelines from KDIGO and other organizations all recommend the use of ACR and eGFR in CKD staging, whether NCR levels were indicative of CKD stage should be addressed before further discussion on CKD.

Response

In the Discussion, we propose using additional kidney injury markers alongside eGFR and ACR to achieve improved estimates of future risk of disability and death in the general population.

8. eGFR is an established risk marker of CKD but it should be used in conjunction with others when eGFR is not low (<60ml/min/1.73m2). Current data suggested the population included here is not particularly of impaired renal filtration. Relevant discussion on eGFR in comparison to NCR should be shortened and organized.

Response

Thank you for this insight.

Because eGFR is already an established risk marker, we have reduced discussion of it.

Minor comments

To make it less confusing, only keep the number of included subjects in the abstract and delete the total number screened.

Response

As suggested, the number in the baseline population has been deleted from the abstract; only the number of participants is specified (line 30).

Line 127 of page 6, "an" MCP should be "a" MCP.

Response

The text has been changed as requested.

In the methods section, please specify the name and version of statistical software.

Response

The statistical software is now specified: ‘SPSS 11.01J software for Windows (SPSS, Japan, Tokyo, Japan) was used to perform all statistical analyses.’ (lines 190–191).

In the methods section, please specify the regression models used for the evaluation.

Response

In the Statistical analyses subsection, we now explain why the logistic regression model was chosen for estimating disability: ‘To estimate risk of disability, we used the multiple logistic regression model rather than the Cox proportional hazards model, because data on the exact date on which participants entered the public LTC insurance program had not been collected. Furthermore, this program was implemented from April 2000 and the baseline survey was conducted in 1993; the LTC service was not available between 1993 and March 2000’ (lines 169–173).

We also used the Cox proportional hazards regression model to estimate the risk of all-cause mortality (please see our response to Reviewer #1’s second comment).

Table 4 has duplicated results with Table 3 (the "entire cohort"). Similarly Figure 2 is a graphical presentation of some results presented in Table 3. Such items should be removed.

Response

We have deleted the results for the entire cohort from the multivariable analysis results presented in Table 4. We have also deleted Figure 2.

Conclusion "public LTC insurance services" should be replaced by disability or disability determined by public LTC insurance services.

Response

As requested, ‘subsequent need for public LTC insurance services’ has been replaced by ‘disability’.

Reviewer #2: In this manuscript, the author has demonstrated that the activity of urinary N-acetyl-β-glucosaminidase (NAG), a marker of kidney injury, is associated with subsequent risk of disability or early death in a general population. However, there exists several problems.

1、 What is the specific date on which to determine whether the events occurred or not？Please describe the medium and IQR of the follow-up period.

Response

The follow-up period is now mentioned in the abstract (‘A total of 1182 participants were followed up for a median of 12.4 years.’; line 30) and Participants and Methods (‘the mean and median follow-up periods were 11.4 years and 12.4 years, respectively’; lines 156–157).

2、 The association between the quartile groups of NAG groups and disability or death should be evaluated using the Kaplan-Meier survival method and compared using log-rank statistics.

Response

Participants were not asked their exact date of entry into the LTC insurance program. Therefore, we were unable to use the Kaplan–Meier survival method to estimate the incidence of disability. Instead, we used the model to estimate all-cause mortality. However, the results were equal to those of the univariate analysis obtained using the Cox proportional hazards model.

If our study had been a randomized controlled trial, it would have been helpful to present the cumulative survival rate in figure form, because all confounders had already been adjusted for at the time of randomization. However, the cohort study results require adjustment for potential confounders; simply presenting the results of univariate analysis in figure form may lead to misunderstanding of the results.

3、 How about the relationships between urinary NAG and other clinical variables?

Response

Thank you for making this important point. To confirm collinearity, we performed a correlation analysis between urinary NAG and other clinical variables (see Table 3).

4、 The association between the NAG and cardiovascular mortality and renal mortality should better be evaluated.

Response

We did not set end-stage renal failure as an endpoint. Therefore, we were unable to determine whether a high urinary NAG to creatinine ratio was associated with increased risk of end-stage renal failure.

We have previously used the Cox proportional hazards model to investigate the relation between urinary NAG activity and cardiovascular events. However, we found no association between increased urinary NAG activity and risk of cardiovascular disease.

5、 What is your criteria for selecting independent variables in your multivariable-adjusted model？

Response

We included traditional cardiovascular risk factors as covariates, because most cardiovascular risk factors are also risk factors for CKD. Furthermore, to adjust for potentially underlying disability and diseases among the participants, we also used physical activity as a covariate (lines 174–183).

6、 The usage of anti-hypertensive drugs, anti-diabetic drug, statins and comorbid conditions should be recorded and adjusted.

Response

Unfortunately, participants were not asked whether they had used these drugs during the follow-up period. Therefore, we are unable to adjust the analyses for these confounders.

7、 In order to determine whether risk prediction models were improved by addition of the NAG, C-index should be calculated for the demographic, eGFR, and cardiovascular risk factor model for each outcome.

Response

Thank you for your helpful suggestion. It would be useful if we could create a new model for predicting future disability. However, the aim of this paper was to establish an association between NAG/creatinine ratio and disability or early death. Creation of a new model for predicting future disability is beyond the scope of this paper. However, we will try to do this in the future.

8、 Q1 needs to be placed in the table, and the specific values of Q1-Q4 need to be marked.

Response

Thank you for your helpful comment. We have made the changes requested.

9、 NAG is one of the urine kidney injury biomarkers, not equals to CKD. These two concepts should not be confused.

Response

Thank you for this observation.

We regard NAG activity as a kidney injury marker and CKD as a consequence of chronic kidney injury.

10、 In line 277-279, you mentioned that “The association between urinary NAG/creatinine ratio and disability or death, which was found to be independent of cardiovascular risk factors (including age), suggests that CKD is related to the physiological changes of aging rather than those of cardiovascular disease. ”, while I am so confused about this sentence. How can I come to a conclusion that “CKD is related to the physiological changes of aging” from your data? The discussion should be re-arranged.

Response

We agree that this statement was confusing and an overstatement. We have replaced it with discussion about the mechanisms underlying CKD and aging, with reference to several relevant studies (lines 392–401).

11、 In your opinion, what is the probably reason that urinary NAG/creatinine ratio can not predict subsequent disability or death in patients with proteinuria?

Response

As mentioned in the Limitations subsection, ‘the great majority of the total cohort did not have proteinuria’ (lines 431–432); we considered the number of participants with proteinuria (n=56) to be too low to allow estimation of the risk of disability or death.

12、 I think Table 2 and Figure 2 maybe not necessary.

Response

Thank you for your comment. We have deleted Figure 2 and amended Table 2.

Reviewer #3: The description of the study design in this article is incomplete and lacks specific follow-up methods. There are also doubts about the statistical methods used in this article to illustrate the relationship between urinary NAG/creatinine ratio and the risk of death and disability. So I don't think this article has reached the standard for receiving manuscripts. Questions are as follows:

1. What is the collection method of the outcome event data? At what time were the collection points?

Response

We mailed the questionnaires to the participants or their relatives at the end of December 2005 and collected outcome data between January and March 2006. For details, please see the subsection The follow-up study of Participants and Methods; lines 153–159).

2. Since the beginning of the study in 2005, what was the median follow-up time for all participants? How many subjects were lost to follow-up?

Response

The median follow-up time was 12.4 years (see Table 1).

A total of 1438 participants who were aged ≥ 65 years at the initiation of the follow-up study at the end of October 2005 were eligible for this study. Only 45 subjects were lost to follow-up (see Results; line 198). Initially, we collected 1400 responses. However, during the Cox hazards regression analysis, we found several faults in the responses (e.g. no descriptions of the date of death). Therefore, 1393 completed questionnaires were used.

3. The statistical method used in the analysis of survival data in this paper used multivariate logistic regression analysis instead of the Cox risk regression model (Table 2 and Table 3), which ignored the impact of censored data.

Response

Thank you for your comment.

We did not collect data on the exact date of participants’ entry into the public LTC program. Therefore, we were unable to use the Cox regression model. However, if participants were able to discontinue their use of the service, they may be considered to no longer have a disability.

For data from participants who had died, it is preferable to use the Cox regression model. Therefore, in the revised manuscript, we describe how we used the Cox proportional regression model to estimate the relation between baseline urinary NAG/creatinine ratio and future risk of all-cause mortality.

4. In Table 4, participants with proteinuria were only 56, which could produce an unstable model.

Response

We agree and have revised the manuscript to present the results for risk in participants without proteinuria only.

Submitted filename: Response to Reviewers_for submission_2021_12_27.docx

Click here for additional data file.

21 Jan 2022

24 Feb 2022

Reviewer #2: Thanks for your detailed modification, and there exists minor questions as following:

1、 Why chose Kruskal–Wallis non-parametric test for continuous variables, instead of ANOVA? Were all continuous variables skewness distribution?

Response

Thank you for drawing our attention to this point.

We checked the distribution of all continuous variables and confirmed that the data for several variables follow a normal distribution; these data were then reanalyzed using ANOVA.

We have accordingly changed the sentence on lines 162–164 (page 7) to read, ‘Continuous variables, expressed as means and standard deviations or medians and interquartile ranges, were compared by means of analysis of variance or the Kruskal–Wallis non-parametric test.’

2、 In Table 1, the P value between any two comparisons should be marked, especially when compared to NA group.

Response

Thank you for your comment.

In addition to using ANOVA or the Kruskal–Wallis test to find significant differences among the NA, Disability, and Had died groups, we used the Tukey or Steel–Dwass test to compare data for the NA group versus the Disability group and the NA group versus the Had died group.

This is now stated on lines 165 and 166 (page 7) (‘We also used the Tukey test or Steel–Dwass test to compare data for the continuous variables between the groups.’), and the statistical package is named on lines 193 and 194 (page 8) (‘Additionally, we used R version 4.1.2 (the R Foundation for Statistical Computing, Vienna, Austria) software for the Steel–Dwass test.’). Relevant p values and explanatory footnotes have been added to Table 1.

3、 In Table 2, since urinary NAG/creatinine ratio quartile is a ranked variable, P-trend value and the P value for Q2\\Q3\\Q4 compared to Q1 should be listed. Therefore, as for ranked variable, Pearson chi-square test is not suitable for categorical variables to estimate P values.

Response

Thank you for making this important point.

We used the logistic regression model, instead of the Pearson chi-square test, to estimate differences among quartiles. Furthermore, we calculated p values for Q2, Q3, or Q4 versus Q1; significant differences are highlighted with asterisks in Table 2.

Use of the logistic regression model is now mentioned in the sentence on lines 164 and 165 (page 7) (‘Categorical variables, expressed as proportions, were compared using the chi-square test or the logistic regression model.’) and in Table 2, footnote c.

Submitted filename: Reviewers comments_2022_02_24.docx

Click here for additional data file.

7 Mar 2022

Association between urinary N-acetyl-β-glucosaminidase activity–urinary creatinine concentration ratio and risk of disability and all-cause mortatility

PONE-D-21-27646R2

Dear Dr. Tanaka,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Yan Li, MD, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: (No Response)

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: (No Response)

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: (No Response)

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: (No Response)

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

17 Mar 2022

PONE-D-21-27646R2

Association between urinary N-acetyl-β-glucosaminidase activity–urinary creatinine concentration ratio and risk of disability and all-cause mortality

Dear Dr. Tanaka:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Yan Li

Academic Editor

PLOS ONE