Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status.

BACKGROUND: Few studies have analyzed the relationship between glucose variability (GV) and adverse health outcomes in patients with differences in glycemic status. The present study tests the hypothesis that GV predicts all-cause mortality regardless of glycemic status after simple adjustment (age and sex) and full adjustment (age, sex, cardiovascular disease, hypertension, use of aspirin, statins, GLP-1 receptor agonists, SGLT-2 inhibitors and DPP-4 inhibitors, baseline FPG and average HbA1c).
METHODS: Prospective cohort study with 795 normoglycemic patients, 233 patients with prediabetes, and 4,102 patients with type 2 diabetes. GV was measured using the coefficient of variation of fasting plasma glucose (CV-FPG) over 12 years of follow-up. The outcome measure was all-cause mortality.
RESULTS: A total of 1,223 patients (657 men, 566 women) died after a median of 9.8 years of follow-up, with an all-cause mortality rate of 23.35/1,000 person-years. In prediabetes or T2DM patients, the fourth quartile of CV-FPG exerted a significant effect on all-cause mortality after simple and full adjustment. A sensitivity analysis excluding participants who died during the first year of follow-up revealed the following results for the highest quartile in the fully adjusted model: overall, HR (95%CI) = 1.54 (1.26-1.89); dysglycemia (prediabetes and T2DM), HR = 1.41 (1.15-1.73); T2DM, HR = 1.36 (1.10-1.67).
CONCLUSION: We found CV-FPG to be useful for measurement of GV. It could also be used for the prognostic stratification of patients with dysglycemia.

Introduction

Glycemic control among dysglycemic patients is usually assessed based on glycosylated hemoglobin (HbA1c), which reflects average blood glucose over previous months but does not inform about oscillations in blood sugar over time. Two patients with the same HbA1c may experience different glycemic excursions. For this reason, measurement of glycemic variability (GV) has been proposed as a tool for glucose monitoring in patients with type 1 diabetes (T1DM) and type 2 diabetes (T2DM) with severe insulin insufficiency [1] or associated conditions such as kidney failure. GV is defined as the oscillation of blood glucose levels outside the normal range. It can be classified into short-term variability (variations within the same day or between days) and long-term variability (oscillations between different clinical visits). The former is based on determinations obtained by continuous glucose monitoring and the latter on determinations of basal plasma glucose, HbA1c, or postprandial glucose obtained at different visits to the hospital or primary care center. The most widely used long-term GV measures are standard deviation and coefficient of variation, which are easy to calculate and interpret.

Many studies have evaluated whether fluctuations in glycemia are directly related to the incidence of diabetes complications. In their meta-analysis, Nalysnyk et al [2] found a link between GV, as measured by the coefficient of variation of fasting plasma glucose, and the risk of both microvascular and macrovascular complications, as well as mortality, among patients with T2DM. Specifically, an association with the development or progression of diabetic retinopathy, cardiovascular events, and mortality was evidenced in 9 of the 10 studies included. Similar findings were reported in a second meta-analysis evaluating variability in HbA1c [3], although most studies were retrospective and were poorly adjusted for potential confounders. Glycemic fluctuations have been reported to increase oxidative stress [4], inflammatory response, and endothelial damage, all of which would lead to vascular complications [5].

By contrast, the recent meta-analysis by Alatawi and Mirghani [6] included seven studies, of which four demonstrated an association between myocardial infarction and GV and three a neutral effect. Indeed, in their meta-analysis of studies on patients with DM, Smith-Palmer et al [7] showed that the association between GV and myocardial infarction was observed only among patients with T1DM.

Although GV is observed mainly in patients with diabetes, it also affects patients with prediabetes and normal blood glucose [8]. However, few studies have analyzed the relationship between GV and adverse health outcomes (cardiovascular events and mortality) among patients with differences in glycemic status, and even fewer have been performed in southern European countries with healthier lifestyles [9]. This lack of information could be due to the lower frequency of all-cause and cardiovascular mortality in southern Europe than in central and northern Europe. However, the progressive aging of the population of southern Europe and the co-occurrence of dysglycemia at older ages require further studies in this line of research. We tested the hypothesis that GV predicts all-cause mortality regardless of glycemic status, after adjusting for established cardiovascular disease. If this hypothesis is confirmed, GV should be incorporated into the prognostic stratification of patients with dysglycemia.

Methods

Study population

We prospectively included 3,438 patients with T2DM from the first recruitment (year 2007) and 726 from the second recruitment (year 2011) of the MADIABETES cohort. These patients constitute the Spanish T2DM cohort with the highest number of person-years of follow-up. MADIABETES is also one of the few Spanish cohorts comprising primary care patients. Findings for the variables recorded—age, sex, time since diagnosis of diabetes, hypertension, dyslipidemia, and microvascular complications—are similar to those of Spanish hospital-based studies [10, 11] and the data reported by Bodicoat et al [12]. Likewise, 1,485 patients were included from the SPREDIA (Screening prediabetes and type 2 diabetes) cohort [13], which was initiated in 2010 and comprised 161 people with T2DM, 78 with previously unknown diabetes, 265 with prediabetes (impaired glucose tolerance [IGT]), and 981 with normal glycemia values. The main aim of the SPREDIA cohort study was to evaluate the performance of the Finnish Diabetes Risk Score (FINDRISC) and a simplified FINDRISC score (MADRISC) in screening for undiagnosed type 2 diabetes mellitus and dysglycemia. Given that 519 patients were excluded for having <3 fasting plasma glucose (FPG) measurements during follow-up, this analysis was based on 5,130 patients (Fig 1).

Fig 1

Flow chart of study participants.

FPG: fasting plasma glucose; NG: normoglycemia; Pre-DM: prediabetes; T2DM: type 2 diabetes mellitus.

Study variables

Patient follow-up started in 2007 and ended in 2019. Mortality data from the year 2020 were not included since values were substantially higher than expected because of COVID-19. As a dynamic cohort, not all patients completed the 12 years of follow-up, and the median follow-up for assessment of mortality was 9.8 years. Follow-up was terminated because of one of the following three circumstances: end of follow-up (12/31/2019), loss to follow-up through change of residence, and date of death. Mortality data were obtained from the Ministry of Health (National Institute of Deaths Registry [INDEF]), which includes the date of death but not its cause. There were no losses to follow-up with respect to mortality, because, regardless of whether the patient moved to a new city, mortality is recorded at national level and is based on the patient’s identification data. These include the national identity card number, which is unique for each Spanish citizen.

In addition, demographic, anthropometric, clinical, and laboratory data were obtained from the family physician responsible for each patient and, when necessary, from the electronic primary care clinical records (AP-Madrid® software), which have been validated for research purposes [14] and are widely used in morbidity and mortality studies [15].

Participants not previously diagnosed with DM at baseline underwent a standard 75-g oral glucose tolerance test (OGTT), which was performed according to World Health Organization (WHO) recommendations [16]. The test includes FPG and glucose measurements over time; the glucose oxidase method was used to determine blood glucose levels. In patients who underwent an OGTT, the FPG was taken after eight hours of fasting and before glucose intake. General practitioners requested FPG measurements under conditions of usual clinical practice, and patients with less well-controlled disease tended to have a significantly higher number of annual blood glucose measurements. When this was the case, participating physicians included only the last blood glucose measurement for the year in the data collection notebook. The median number of glucose measurements was eight. GV was only measured when at least three glucose measurements were collected.

Glycemic status was defined using the oral glucose tolerance test (OGTT) [17] as follows: normoglycemia (OGTT <140 mg/dl), prediabetes (OGTT 140–199 mg/dl), and newly diagnosed DM (participants with no previous diagnosis of DM at baseline and an OGTT ≥200 mg/dl).

Pre-existing cardiovascular disease was defined as a history of myocardial infarction, stroke, or peripheral vascular disease.

Statistical analyses

As previously mentioned, the coefficient of variation of FPG (CV-FPG) was obtained in patients with at least three FPG values during follow-up (80.9%, 87.5%, and 93% in patients with normoglycemia, IGT, and T2DM, respectively) and calculated, for each patient, as the ratio of the standard deviation to the mean FPG multiplied by 100 (CV = SD/mean x 100, in %). A certain degree of GV is reasonable in subjects with normal glucose tolerance and even more so in those with diabetes or impaired glucose regulation in blood. Therefore, it is crucial to identify the limit beyond which GV acquires pathological significance (association with mortality). For this reason, patients were categorized according to quartiles of CV-FPG, both in the overall sample and in the subsamples of patients with T2DM and patients with T2DM plus IGT, given that CV-FPG varies with each glycemic status category. The values of these quartiles for the total sample were as follows: Q1: ≤9.047; Q2: 9.048 to 15.232; Q3: 15.233 to 24.438: Q4: ≥24.439. For the subsample of patients from MADIABETES and SPREDIA (T2DM patients), the values were as follows: Q1: ≤12.3287; Q2: 12.3288 to 18.6246; Q3: 18.6247 to 27.0836; Q4: ≥27.0836.

Data are presented as proportions, means (standard deviation, SD), or, in the case of variables that did not conform to a normal distribution, medians (interquartile range). The t test was used to compare two means, whilst the χ2 test was used for two or more proportions. An analysis of variance (ANOVA) was performed to compare continuous variables among the four quartiles of CV-FPG.

The mortality rate was calculated by considering the total number of deaths during follow-up divided by the total number of person-years.

Univariate survival analysis was performed using the Kaplan-Meier method and log-rank test. Multivariate survival analysis was conducted using Cox regression. In the first analysis with all subjects, the hazard ratios (HRs) and 95% confidence interval (CI) were calculated based on the following models: model 1, adjusted for age and sex; model 2, further adjusted for history of cardiovascular disease; model 3, further adjusted for glycemic status, hypertension, and use of statins, aspirin and antidiabetic-drugs in patients with T2DM; and model 4, further adjusted for baseline FPG in all samples and for baseline FPG plus average HbA1c (when at least two measurements were taken) in the T2DM sample. The interaction between CV-FPG and sex was assessed using a likelihood ratio test of their product terms in the full model for each glycemic status.

Lastly, a sensitivity analysis was performed excluding participants who died during the first year of follow-up to avoid the possible influence of the severity of underlying illnesses.

The analyses were performed with SPSS version 21.0 (IBM Corp., Armonk, NY, USA); a 2-sided p value < 0.05 was considered statistically significant.

Ethics statement

The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Ramón y Cajal Hospital (Madrid) for the MADIABETES cohort (approval identification code:2017/335) and the Institutional Review Board of Carlos III Hospital (Madrid) for the SPREDIA cohort (approval identification code: P07/2012). A written informed consent was obtained from all subjects involved in the study.

Results

During follow-up, a total of 1,223 patients (657 men, 566 women) died, with an all-cause mortality rate of 23.35/1,000 person-years (26.07/1,000 in men and 20.83/1,000 in women). The mortality rates according to glycemic status are shown in Fig 2. Table 1 shows the baseline socio-demographic, anthropometric, and clinical findings for survivors and patients who died after a median of 9.8 years of follow-up. Compared with the survivors, patients who died were more likely to be male, older, ex-smokers, and hypertensive. They also had dyslipidemia and more frequently had a history of cardiovascular disease. They also had a lower mean body mass index, higher mean systolic blood pressure, and higher mean FPG.

Fig 2

Mortality rates according to glycemic status.

NG: normoglycemia; IGT: impaired glucose tolerance; T2DM: type 2 diabetes mellitus.

Table 1

Baseline socio-demographic, anthropometric, and clinical data of study participants overall and by survival status.

Variables	All patients (N = 5,130)	Dead (N = 1,223)	Survivors (N = 3,907)	p value
Age, mean (SD)	66.6 (101)	74.5 (8.7)	64.1 (9.2)	<0.001
Sex male, n (%)	2,474 (48.2)	657 (53.7)	1,817 (46.5)	<0.001
Never smoked, n (%)	2,619 (51.1)	694 (56.7)	1,925 (49.3)	<0.001
Ex-smoker, n (%)	1,386 (27)	346 (28.3)	1,040 (26.6)
Active smoker, n (%)	1,125 (21.9)	183 (15)	942 (24.1)
BMI, mean (SD)	30.2 (5.1)	29.7 (5.2)	30.4 (5.1)	<0.001
Baseline SBP, mean (SD)	131.7 (11.3)	133 (10.8)	131.2 (11.4)	<0.001
Baseline DBP, mean (SD)	75.7 (6.9)	73.4 (6.5)	76.4 (6.8)	<0.001
T2DM, n (%)	4,102 (80.0)	1,199 (98)	2,903 (74.0)	<0.001
Prediabetes (IGT), n (%)	233 (4.5)	8 (0.7)	225 (5.8)
Normoglycemia, n (%)	795 (15.5)	16 (1.3)	779 (19.9)
History of CVD, n (%)	849 (16.5)	400 (32.7)	449 (11.5)	<0.001
Hypertension, n (%)	3,596 (70.1)	1,015 (83)	2,581 (66.1)	<0.001
Dyslipidemia, n (%)	2,669 (52)	678 (55.4)	1,991 (51)	0.006
Baseline FPG level, mean (SD)	127.9 (31)	134.8 (30.2)	125.8 (31)	<0.001

BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; T2DM: type 2 diabetes mellitus; IGT: impaired glucose tolerance; CVD: cardiovascular disease; FPG: fasting plasma glucose

In Table 2 we summarize the baseline treatments received by study participants, finding significantly higher use of statins, aspirin and insulin among patients who had died than among survivors.

Table 2

Baseline treatments of study participants overall and by survival status.

Variables	All patients (N = 5,130)	Dead (N = 1,223)	Survivors (N = 3,907)	p value
Statin use, n (%)	3,142 (61.2)	819 (67)	2,323 (59.5)	<0.001
Aspirin use, n (%)	2,611 (50.9)	844 (69)	1,767 (45.2)	<0.001
Metformin use, n (%)	2,128 (47.8)	464 (48.2)	1,664 (47.7)	0.784
Sulfonylurea use, n (%)	864 (19.4)	202 (21)	662 (19)	0.163
Insulin use, n (%)	982 (22.1)	354 (36.8)	628 (18)	<0.001
GLP-1_receptor agonist use, n (%)	31 (0.6)	3 (0.2)	28 (0.7)	0.063
SGLT-2 inhibitor use, n (%)	48 (0.9)	1 (0.1)	47 (1.2)	<0.001
DPP-4 inhibitor use, n (%)	182 (3.5)	51 (4.2)	131 (3.4)	0.178

GLP-1: glucagon like-peptide 1; SGLT-2: sodium-glucose co-transporter-2; DPP-4: dipeptidyl peptidase 4.

Table 3 presents the baseline characteristics of all study participants according to the quartiles of the CV-FPG. A significant linear trend (p<0.05) across the baseline quartiles was observed for the following variables: male sex, smoking status, dyslipidemia, hypertension, glycemic status, cardiovascular disease, and use of statins and aspirin.

Table 3

Baseline factors of 5,130 subjects with differences in glycemic status grouped by CV-FPG quartile.

Variables	CV-FPG Quartile
	1 (lowest)	2	3	4 (highest)	p value
Range	≤9.047	9.048–15.232	15.233–24.438	≥24.439
N	1,282	1,283	1,283	1,282
Anthropometric and clinical variables
Male sex, n (%)	570 (44.5)	628 (48.9)	635 (49.5)	641 (50)	0.01*
Age, mean (SD)	64.7 (8.3)	66.8 (9.6)	67.7 (10.2)	67.2 (11.7)	<0.01
BMI, mean (SD)	29.0 (4.9)	30.4 (4.9)	30.5 (5.0)	30.7 (5.4)	<0.01
Baseline SBP, mean (SD)	129.2 (12.2)	131.8 (10.8)	132.3 (10.5)	133.1 (11.2)	<0.01
Baseline DBP, mean (SD)	76.1 (7.4)	76.0 (6.7)	75.4 (6.6)	75.3 (6.8)	0.01
Smoking, n (%)	364 (28.4)	273 (21.3)	231 (18.0)	257 (20.0)	<0.01*
Hypertension, n (%)	680 (53)	934 (72.8)	995 (77.6)	987 (77)	<0.01*
Dyslipidemia, n (%)	633 (49.4)	675 (52.6)	650 (50.7)	711 (55.5)	<0.01*
Glycemic status
Normoglycemia, n (%)	609 (47.5)	156 (12.2)	21 (1.6)	9 (0.7)	<0.01*
Prediabetes (IGT), n (%)	148 (11.5)	65 (5.1)	14 (1.1)	6 (0.5)
Type 2 DM, n (%)	525 (41.0)	1,062 (82.8)	1,248 (97.3)	1,267 (98.8)
Cardiovascular Disease
Previous myocardial infarction, n (%)	59 (4.6)	108 (8.4)	153 (11.9)	161 (12.6)	<0.01*
Previous stroke, n (%)	33 (2.6)	72 (5.6)	98 (7.6)	107 (8.3)	<0.01*
Primary prevention, n (%)	1,181 (92.1)	1.091 (85.0)	1,028 (80.1)	981 (76.5)	<0.01*
Medication profile
Statin use, n (%)	574 (44.8)	798 (62.2)	873 (68.0)	897 (70.0)	<0.01*
Aspirin use, n (%)	335 (26.1)	626 (48.8)	794 (61.9)	856 (66.8)	<0.01*

*p value for linear trend across baseline CV-FPG quartiles.

CV-FPG: coefficient of variation of fasting plasma glucose; BMI: body mass index; SBP: systolic blood pressure; DBP: diastolic blood pressure; T2DM: type 2 diabetes mellitus; IGT: impaired glucose tolerance

As shown in Fig 3, crude mortality was higher in the top quartile of CV-FPG than in the other quartiles (p< 0.001) during the 12-year follow-up period.

Fig 3

Crude all-cause mortality according to quartiles of the coefficient of variation of fasting plasma glucose levels (5,130 subjects with normoglycemia, prediabetes, or T2DM).

Table 4 shows all-cause mortality in total study participants according to quartiles of CV-FPG. Compared with patients in the lowest quartile, age- and sex-adjusted HRs (95%CI) in the third and highest CV-FPG quartiles were 1.31 (1.07–1.60) and 1.87 (1.54–2.26), respectively. However, findings for the second quartile were not significant (HR [95%CI] = 1.03 [0.83–1.27]). After further adjustment for history of cardiovascular disease (model 2), the effect of CV-FPG was attenuated, although it remained statistically significant for the third and fourth quartiles (HR [95%CI] = 1.23 [1.01–1.50] and 1.71 [1.41.2.08], respectively). Further adjustment (model 3 and 4) revealed a significant effect only for the highest quartile.

Table 4

Hazard ratios of all-cause mortality grouped by CV-FPG in patients with normoglycemia, IGT and T2DM.

Normoglycemia, IGT and T2DM (n = 5,130)	CV-FPG quartile
	1 (lowest)	2	3	4 (highest)
N	1,282	1,283	1,283	1,282
All-cause mortality, n (%)	137 (11.2)	238 (19.5)	365 (29.8)	483 (39.5)
Person-years	11,370	13,270	14,020	13,710
Mortality rate (per 1,000 person-years)	12.05	17.94	26.03	35.23
Model 1	1	1.03 (0.83–1.27)	1.31 (1.07–1.60)*	1.87 (1.54–2.26)**
Model 2	1	0.98 (0.79–1.21)	1.23 (1.01–1.50)*	1.71 (1.41–2.08)**
Model 3	1	0.84 (0.68–1.05)	1.04 (0.85–1.27)	1.44 (1.18–1.76)**
Model 4	1	0.85 (0.70–1.05)	1.01 (0.82–1.24)	1.13–1.71)*

Model 1: adjusted for age and sex. Model 2: adjusted for age, sex, and history of cardiovascular disease. Model 3: adjusted for variables in model 2 plus glycemic status, hypertension, use of aspirin and statins. Model 4: Model 3 plus baseline FPG

*p < 0.05

**p < 0.001.

CV-FPG: Coefficient of variation of fasting plasm glucose

Table 5 shows the results of the main analysis only in patients with T2DM; the results are very similar, with significant HRs for the fourth quartile in all models and the third quartile in model 1. No significant interaction with sex was found (p = 0.957).

Table 5

Hazard ratios of all-cause mortality grouped by CV-FPG in T2DM patients.

T2DM (n = 4,102)	CV-FPG quartile
	1 (lowest)	2	3	4 (highest)
N	1,025	1,026	1,026	1,025
All-cause mortality, n (%)	227 (22.1)	249 (24.3)	310 (30.2)	413 (40.3)
Person-years	10,573	11,345	11,253	10,887
Mortality rate (per 1,000 person-years)	21.47	21.95	27.55	37.94
Model 1	1	0.91 (0.76–1.09)	1.19 (1.01–1.41)*	1.52 (1.25–1.86)**
Model 2	1	0.82 (0.66–1.02)	1.02 (0.83–1.25)	1.42 (1.16–1.73)**
Model 3	1	0.82 (0.66–1.02)	1.02 (0.83–1.25)	1.41 (1.16–1.73)**
Model 4	1	0.82 (0.66–1.02)	0.99 (0.80–1.22)	1.37 (1.11–1.68)*

Model 1: adjusted for age and sex. Model 2: adjusted for age, sex, and history of cardiovascular disease. Model 3: adjusted for variables in model 2 plus hypertension, use of aspirin and statins. Model 4: Model 3 plus use of GLP-1 receptor agonists, SGLT-2 inhibitors and DPP-4 inhibitors, baseline FPG and average HbA1c (when at least two measurements) *p < 0.05

**p < 0.001.

CV-FPG: Coefficient of variation of fasting plasm glucose; GLP-1: glucagon like-peptide-1; SGLT-2:sodium-glucose co-transporter-2; DPP-4: dipeptidyl peptidase 4

Lastly, Table 6 (prediabetes [IGT] or T2DM) also shows the significant effect of the fourth quartile on all-cause mortality in all models.

Table 6

Hazard ratios of all-cause mortality grouped by CV-FPG in patients with IGT and T2DM.

IGT and T2DM (n = 4,335)	CV-FPG quartile
	1 (lowest)	2	3	4 (highest)
N	1,084	1,084	1,084	1,083
All-cause mortality, n (%)	205 (18.9)	246 (22.7)	325 (30.0)	431 (39.8)
Person-years	10,684	11,825	11,889	11,503
Mortality rate (per 1,000 person-years)	19.19	20.80	27.34	37.47
Model 1	1	0.89 (0.72–1.11)	1.13 (0.92–1.39)	1.61 (1.32–1.96)*
Model 2	1	0.86 (0.69–1.07)	1.08 (0.88–1.32)	1.49 (1.22–1.82)*
Model 3	1	0.86 (0.69–1.06)	1.07 (0.87–1.31)	1.48 (1.22–1.81)*
Model 4	1	0.85 (0.68–1.06)	1.03 (0.83–1.27)	1.41 (1.15–1.73)*

Model 1: adjusted for age and sex. Model 2: adjusted for age, sex, and history of cardiovascular disease. Model 3: adjusted for variables in model 2 plus hypertension, use of aspirin and statins. Model 4: Model 3 plus baseline FPG *p < 0.001.

CV-FPG: Coefficient of variation of fasting plasm glucose

The results of the sensitivity analysis were consistent with previous findings. Therefore, the results for the highest quartile in the most adjusted model were as follows: overall, HR (95%CI) = 1.54 (1.26–1.89); dysglycemia (prediabetes and T2DM), HR = 1.41 (1.15–1.73); and T2DM, HR = 1.36 (1.10–1.67).

Discussion

To our knowledge, this is the first large cohort study performed in southern Europe to investigate the association between GV and all-cause mortality in patients with differences in glycemic status. This is the main difference with respect to previous, similar studies [18], given that we studied the GV in different metabolic situations (normoglycemia, impaired glucose tolerance [IGT], and diabetes mellitus).

Our findings show that the highest degree of GV, expressed as the highest quartile of the CV-FPG, behaves as a long-term predictor of all-cause mortality in patients with T2DM and the subgroup with prediabetes or T2DM for any adjusted model. In addition, the third quartile proved to be a predictor of mortality exclusively for the age- and sex-adjusted model in patients with T2DM.

Although there are different ways of measuring GV, the CV-FPG and mean amplitude of glycemic excursions (MAGE) are considered the most useful for research purposes [19]. Other studies have found similar results to ours with the CV-FGP. For example, the Verona Diabetes Study [20] enrolled 1,409 T2DM patients aged 56–74 years with a 10-year follow-up of mortality. In the multivariate analysis, the relative risk for all-cause death associated with the highest versus lowest tertile of the CV-FPG was 1.68 (95%CI, 1.29–2.18). The crude Kaplan-Meier analysis showed that survival was longer in patients in the lower tertile of CV-FPG (p = 0.001) than in patients of the other two tertiles, for whom differences were not significant. A subsequent analysis in 1,319 T2DM patients of the Verona Diabetes Study [21] showed differences between the age groups. The group aged >65 years had an adjusted HR for CV-FPG of 1.56 (1.17–2.08), and the younger group had a non-significantly adjusted HR of 1.34 (0.79–2.27). These differences are not surprising, given that the lower incidence of mortality and the smaller size in the younger group might have diminished the statistical power and thus precluded identification of significant results. Therefore, in our study, we preferred to adjust for age and not to stratify.

In a retrospective cohort study of 5,008 T2DM patients from Taiwan [22], the fully adjusted HR of all-cause mortality for the highest versus lowest tertile of annual CV-FPG was 5.53 (95%CI, 3.85–7.94). The association was considerably stronger than in our study, probably because participants were exclusively patients with T2DM treated in hospital, thus leading to selection bias.

In critically ill patients receiving intravenous insulin the coefficient of GV was independently associated with 30-day mortality (OR = 1.23 for every 10% increase, p<0.001), even after adjustment for hypoglycemia, age, disease severity, and comorbidities. The association was observed both in non-diabetics (OR = 1.37, p<0.001) and in diabetics (OR = 1.15, p = 0.001) [23]. The effect on mortality has been attributed to hypoglycemia, especially in critically ill patients, although our study ruled out this possibility, as it was adjusted for the presence of hypoglycemia (defined as <60 mg/dL). However, in chronic T2DM patients, the effect of GV on all-cause mortality is due to its association with 8-iso prostaglandin F2α, a marker of oxidative stress and a potential mediator of organ dysfunction [4].

The association between mortality and GV could be due to an increase in the incidence of cancer, given the known association between marked GV and a dose-dependent high risk of future malignancies among people without diabetes [24]. Furthermore, both dysglycemia and overt atherosclerosis increase the risk of cancer [25].

A retrospective Chinese study [26] of 8,871 patients with T2DM followed for 7 years showed no association between CV-FPG and all-cause mortality after adjustment for baseline FPG. Furthermore, after stratifying by HbA1c, the HR of the highest CV-FPG quartile was only significant among those with HbA1c >7% (HR = 1.63; 95%CI, 1.25–2.13). In contrast, our study included baseline FPG in the fully adjusted models; therefore, the significant results were independent of the degree of glycemic control.

Study results can be influenced by the method used to measure GV. In this regard, the ADVANCE trial [27] in T2DM patients analyzed the visit-to-visit GV using the SD of HbA1c (SD-HbA1c) and of glucose (SD-FPG). There were significant linear associations between SD-HbA1c and combined macro-and microvascular events, major macrovascular events, and all-cause mortality after adjusting for mean HbA1c during the first 24 months and other confounders. SD-FPG, adjusted for mean FPG during the first 24 months and other factors, was also continuously associated with combined macro- and microvascular events, major macrovascular events, and major microvascular events, but not for all-cause mortality. In our case, we could not compare the results with variability in HbA1c, because this measurement was recorded in very few patients with normoglycemia.

In contrast, in an observational analysis of the ALLHAT study [28] (4,982 hypertensive participants, 35.3% diagnosed with DM and 25.6% with a history of cardiovascular disease), the fully-adjusted HR (95% CI) for all-cause mortality was 2.22 (1.22–4.04) for the highest versus lowest quartile of SD-FPG (≥26.4 vs. <5.5 mg/dL). A Taiwanese retrospective cohort study in T2DM patients [29] followed for at least 2 years showed that variability in HbA1c, as measured using SD-HbA1c or CV-HbA1c, was a significant risk factor for all-cause mortality, yielding a higher HR with SD-HbA1c than with CV-HbA1c (1.99 vs. 1.06, both with p<0.05) after full adjustment for use of statins, as in the present study.

Our study is characterized by a series of strengths, including its prospective design and the large number of patients with diabetes, prediabetes, and normoglycemia, as well as its long-term follow-up. In addition, to our knowledge, ours is the first study to examine the relationship between variability in FPG and all-cause mortality in patients with differences in glycemic status in southern European countries. This aspect is especially relevant, given the possible lower effect of GV on all-cause mortality in countries with healthier lifestyles [9] and better glycemic control than other countries participating in the EUROASPIRE IV survey [30].

However, our study is also subject to a series of limitations. First, given that we included patients with differences in glycemic status, the analyses could not be adjusted for duration of diabetes, mean HbA1c, diabetic nephropathy, diabetes treatments, or microalbuminuria, as in other studies. Second, we did not have information on the cause of death, which would have enabled us to verify that mortality is, to a large extent, accounted for by cardiovascular disease, given the known association between GV and macrovascular complications. Third, we did not record hypoglycemia episodes and were therefore unable to assess their association with mortality. Fourth, we could not study GV measured with CV-HbA1c, given that few persons with normoglycemia or IGT had at least three HbA1c measurements during follow-up. Fifth, as data were from two sources, namely, the MADIABETES and SPREDIA cohorts, they may have been subject to a certain degree of heterogeneity.

Lastly, given the observational nature of the present study, individuals with higher GV and lower GV were dissimilar. Therefore, to obtain an accurate picture of the association between GV and all-cause mortality, it was necessary to adjust for differences in both groups in the multivariate analysis. Propensity score matching (PSM) would be a more appropriate alternative that would yield less biased results than standard methods such as Cox regression. However, one of the drawbacks of PSM is the loss of sample in terms of size. In addition, PSM should not be used in practice because our sample size was insufficiently large. Given that propensity scores can only control for observed confounders, they cannot be counted upon to balance unobserved covariates.

Conclusion

Our results and those of other, similar studies show that the prognostic stratification of patients with some degree of dysglycemia should incorporate measurement of GV. CV-FPG proved useful for measuring GV in our study. However, the best method for assessing GV under conditions of daily clinical practice remains to be defined. In addition, it remains unclear whether the consequences of GV for mortality can justify using drugs for control of GV, especially in prediabetic patients.

20 Apr 2022

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,

Venkata Naga Srikanth Garikipati, 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 provide additional details regarding participant consent. In the Methods section,

please ensure that you have specified (1) whether consent was informed and (2) what type you obtained

(for instance, written or verbal). If your study included minors, state whether you obtained consent from parents

or guardians. If the need for consent was waived by the ethics committee, please include this information.

3. Thank you for stating the following financial disclosure:

“This study was funded by Instituto de Salud Carlos III through projects “PI15/00259” and “PI18/01025” and co-funded by the European Regional Development Fund, “A way of shaping Europe””

Please state what role the funders took in the study.  If the funders had no role, please state: ""The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.""

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

4. Thank you for stating the following in your Competing Interests section:

“The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.”

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state ""The authors have declared that no competing interests exist."", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

5. 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.

6. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

7. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

8. One of the noted authors is a group or consortium [Abanades-Herranz JC]. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

[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: Yes

Reviewer #2: Yes

**********

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

Reviewer #1: No

Reviewer #2: Yes

**********

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: Yes

**********

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: No

Reviewer #2: 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 have found in their study that CV-FPG is useful for measurement of GV. Although the finding are potential, reviewer has few concerns:

-data representation should be re-arranged. For instance, few table are too long to read. Split them into two.

- Consult the statistician to confirm the statistical tests used in the study.

- Check for the grammar and typo error.

-Please report the other medical underlyig conditions and how they affect the finding?

Reviewer #2: The current study “Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status” contributes in highlighting the significance to include glycemic variability (GV) as one of the factors to monitor the prognosis of different glycemic status patients.

The finding do suggest how GV could behave as a long-term predictor of all-cause mortality in individuals with certain underlying conditions with variation included .It is rightly understood, with the limitation of the method to satisfy all the variables not limited to duration of the disease, diabetes treatments, or microalbuminuria etc, which would prevent this to be used with confidence.

But the authors do make a valid case for inclusion of GV analysis through their proposed methods so as to derive information for managing the outcome of the diabetic condition of the patients.

There are however, some concerns with the current draft of the manuscript. The authors should follow a consistent mode of representation. E.g. at some places Fig. is used while at some places Figure is written. Kindly adhere to one form of writing.

**********

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

[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.

2 Jun 2022

We thank the editor and the two reviewers for their comments on our manuscript entitled “Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status “. Below is our response to each point raised by the academic editor and reviewers.

We hope that we satisfyingly addressed them, and that the manuscript will be now suited for publication.

Sincerely,

Salinero-Fort MA

On behalf of all authors,

Academic editor:

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.

File naming was edited to comply with the style requirements. We hopefully have no divergences from the style requirements now.

2. Please provide additional details regarding participant consent. In the Methods section, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

In methods section include the following paragraph:

“The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Ramón y Cajal Hospital (Madrid) for the MADIABETES cohort (approval identification code:2017/335) and the Institutional Review Board of Carlos III Hospital (Madrid) for the SPREDIA cohort (approval identification code: P07/2012). A written informed consent was obtained from all subjects involved in the study”.

The e-mail of Institutional Review Board of Ramón y Cajal Hospital (Madrid) is: ceic.hrc@salud.madrid.org

3. Thank you for stating the following financial disclosure:

“This study was funded by Instituto de Salud Carlos III through projects “PI15/00259” and “PI18/01025” and co-funded by the European Regional Development Fund, “A way of shaping Europe””

Please state what role the funders took in the study. If the funders had no role, please state: ""The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.""

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

We confirm that:

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

4. Thank you for stating the following in your Competing Interests section:

“The authors declare that they have no known competing financial interests or personal relationships that could influence the work reported in this paper.”

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state ""The authors have declared that no competing interests exist."", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

We confirm that:

The authors have declared that no competing interests exist

5. 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.

There are restrictions on the availability of data for the MADIABETES study, due to the signed consent agreements around data sharing, which only allow access to external researchers for research following the project purposes.

Requestors wishing to access the MADIABETES data used in this study can request it to the MADIABETES Steering Committee: estudios.fiibap@salud.madrid.org

The request will then be passed to members of the MADIABETES Steering Committee for deliberation.

However, the minimal data set is available from the Institutional Repository of the Regional Health System of the Community of Madrid (https://repositoriosaludmadrid.es/)

https://repositoriosaludmadrid.es/handle/20.500.12530/326/browse?type=author&order=ASC&rpp=20&authority=e55f8892-3241-4a96-a0d6-640436d5c75e&value=SALINERO-FORT%2C+MIGUEL+A.

6. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

There are restrictions on the availability of data for the MADIABETES study, due to the signed consent agreements around data sharing, which only allow access to external researchers for research following the project purposes.

Requestors wishing to access the MADIABETES data used in this study can request it to the MADIABETES Steering Committee: estudios.fiibap@salud.madrid.org

The request will then be passed to members of the MADIABETES Steering Committee for deliberation.

7. Please note that in order to use the direct billing option the corresponding author must be affiliated with the chosen institute. Please either amend your manuscript to change the affiliation or corresponding author, or email us at plosone@plos.org with a request to remove this option.

Corresponding author:

Salinero-Fort MA1-5

1. Foundation for Research and Biomedical Innovation of Primary Care of the Community of Madrid (FIIBAP) [payment institution]

2. The Hospital La Paz Institute for Health Research (IdiPAZ).

3. Health Services and Chronic Conditions Research Network (REDISSEC), Madrid, Spain.

4. General Subdirectorate of Research and Documentation. Department of Health, Madrid, Spain.

5. Alfonso X El Sabio University, Madrid, Spain.

8. One of the noted authors is a group or consortium [Abanades-Herranz JC]. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

The MADIABETES consortium is currently led by Dr. Iriarte-Campo V, a researcher at the Foundation for Research and Biomedical Innovation of Primary Care of the Community of Madrid (FIIBAP). His e-mail is: victor.iriarte@salud.madrid.org

Dr. Abanades-Herranz since 2020 is not the leader or head of the group.

Reviewer #1: The authors have found in their study that CV-FPG is useful for measurement of GV. Although the finding are potential, reviewer has few concerns:

-data representation should be re-arranged. For instance, few table are too long to read. Split them into two.

As suggested by the reviewer, the previous table 1 is now broken down into two new tables and the previous table 3 is now broken down into three new tables. This will surely make the tables easier to read.

- Consult the statistician to confirm the statistical tests used in the study.

The normality of the distribution of all the continuous variables was checked using the Kolmogorov-Smirnov Test and all of them were normally distributed, so parametric tests could be used.

Also, Valentín Hernández-Barrera, an expert statistician of Preventive Medicine and Public Health Teaching and Research Unit, Health Sciences Faculty, Universidad Rey Juan Carlos, Madrid, Spain (ORCID: 0000-0001-5790-1959; E-mail: valentin.hernandez@urjc.es) confirms that the statistical tests used are correct and appropriate. Several previous studies, with similar designs to ours, have used the same statistical methods applied in our investigation:

• Kim C, Sohn J-H, Jang MU, Kim S-H, Choi M-G, Ryu O-H, et al. (2015) Association between Visit-to-Visit Glucose Variability and Cognitive Function in Aged Type 2 Diabetic Patients: A Cross-Sectional Study. PLoS ONE 10(7): e0132118.doi:10.1371/journal.pone.0132118

• Wang A, Liu X, Xu J, Han X, Su Z, Chen S, Zhang N, Wu S, Wang Y, Wang Y. Visit-to-Visit Variability of Fasting Plasma Glucose and the Risk of Cardiovascular Disease and All-Cause Mortality in the General Population. J Am Heart Assoc. 2017 Nov 29;6(12):e006757. doi: 10.1161/JAHA.117.006757. PMID: 29187392; PMCID: PMC5779006.

• Xu D, Fang H, Xu W, Yan Y, Liu Y, Yao B. Fasting plasma glucose variability and all-cause mortality among type 2 diabetes patients: a dynamic cohort study in Shanghai, China. Sci Rep. 2016 Dec 22;6:39633. doi: 10.1038/srep39633. PMID: 28004765; PMCID: PMC5177938.

Lastly, Hans DeVries J (Academic Medical Center at the University of Amsterdam, Amsterdam, the Netherlands) in his article entitled “Glucose Variability: Where It Is Important and How to Measure It” (Diabetes. 2013 May;62(5):1405-8. doi: 10.2337/db12-1610. PMID: 23613566), says: “When diabetes investigators want to assess glucose variability, I would recommend coefficient of variation and mean absolute glucose (MAG)”.

- Check for the grammar and typo error.

An expert native translator has reviewed the manuscript and confirms that there are no grammatical errors. Please find enclosed a certificate from the translator.

-Please report the other medical underlyig conditions and how they affect the finding?

Our study has incorporated the main variables that can affect glycemic variability in the multivariate analysis. Study's covariates are as follows: Age, sex, cardiovascular disease, hypertension, use of aspirin, statins, GLP-1 receptor agonists, SGLT-2 inhibitors and DPP-4 inhibitors, baseline FPG and average HbA1c (when at least two measurements)

The studies listed below have used the same or similar control variables as our study.

• Cardoso CRL, Leite NC, Moram CBM, Salles GF. Long-term visit-to-visit glycemic variability as predictor of micro- and macrovascular complications in patients with type 2 diabetes: The Rio de Janeiro Type 2 Diabetes Cohort Study. Cardiovasc Diabetol. 2018 Feb 24;17(1):33. doi: 10.1186/s12933-018-0677-0. PMID: 29477146; PMCID: PMC6389075.

• Orsi E, Solini A, Bonora E, Fondelli C, Trevisan R, Vedovato M, Cavalot F, Gruden G, Morano S, Nicolucci A, Penno G, Pugliese G; Renal Insufficiency and Cardiovascular Events (RIACE) Study Group. Haemoglobin A1c variability is a strong, independent predictor of all-cause mortality in patients with type 2 diabetes. Diabetes Obes Metab. 2018 Aug;20(8):1885-1893. doi: 10.1111/dom.13306. Epub 2018 Apr 19. PMID: 29582548.

An alternative option would have been to use propensity score matching, but this was discarded for two reasons:

1) No studies have ever used the propensity score to measure the effect of glycemic variability on mortality as can be demonstrated by the following search strategy in PubMed: "glycemic variability" AND "propensity score matching": 0 Results..

2) In the limitations section, we have added the following explanation about the not use of propensity score matching.

“Given the observational nature of the present study, individuals with higher GV and lower GV were dissimilar. Therefore, to obtain an accurate picture of the association between GV and all-cause mortality, it was necessary to adjust for differences in both groups in the multivariate analysis. Propensity score matching (PSM) would be a more appropriate alternative that would yield less biased results than standard methods such as Cox regression. However, one of the drawbacks of PSM is the loss of sample in terms of size. In addition, PSM should not be used in practice because our sample size was insufficiently large. Given that propensity scores can only control for observed confounders, they cannot be counted upon to balance unobserved covariates”

Reviewer #2: The current study “Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status” contributes in highlighting the significance to include glycemic variability (GV) as one of the factors to monitor the prognosis of different glycemic status patients.

The finding do suggest how GV could behave as a long-term predictor of all-cause mortality in individuals with certain underlying conditions with variation included .It is rightly understood, with the limitation of the method to satisfy all the variables not limited to duration of the disease, diabetes treatments, or microalbuminuria etc, which would prevent this to be used with confidence.

But the authors do make a valid case for inclusion of GV analysis through their proposed methods so as to derive information for managing the outcome of the diabetic condition of the patients.

The methodological approach to studying the effect of glycemic variability on mortality is similar to that proposed by all the studies published to date. The ideal would have been to ensure that both groups: those exposed to greater glycemic variability and those exposed to lesser glycemic variability, were similar in all the variables predictive of mortality. This approach would have needed to perform a propensity score matching, which was discarded because it would have meant a considerable reduction in the study sample. To date, no article like ours has done so.

There are however, some concerns with the current draft of the manuscript. The authors should follow a consistent mode of representation. E.g. at some places Fig. is used while at some places Figure is written. Kindly adhere to one form of writing.

Thank you for this comment. Sorry for the lack of consistency. Following your suggestion, the text has been completely reviewed and all these errors edited.

Submitted filename: 2022_06_01_Rebutal letter.docx

Click here for additional data file.

6 Jul 2022

Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status.

PONE-D-21-35463R1

Dear Dr.Salinero-Fort,

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,

Venkata Naga Srikanth Garikipati, 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 #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 #2: Yes

Reviewer #3: Yes

**********

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

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 #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 #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 #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 #2: Yes: Manju Narwal

Reviewer #3: No

**********

15 Jul 2022

PONE-D-21-35463R1

Glycemic variability and all-cause mortality in a large prospective southern European cohort of patients with differences in glycemic status.

Dear Dr. Salinero-Fort:

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

Dr. Venkata Naga Srikanth Garikipati

Academic Editor

PLOS ONE