Background

Diabetes is a complex, chronic disease with multifactorial risk factors. Although glycemic control is an important factor for risk reduction, the prevention and early detection of complications are other factors to reduce both morbidity and mortality.

The global prevalence of diabetes mellitus in adults aged from 20 to 79 years is estimated at 8.8%, with a higher incidence in high-middle-income countries than in low-income countries with the peak incidence at age 65–69 years for men and 75–79 years for women [1].

Type 2 diabetes mellitus can be considered as an age-related disorder; although aging has a minimal effect on insulin action directly, multiple factors in an older person may contribute to hyperglycemia; aging has been thought of as an independent factor associated with low-grade inflammation [2]. As chronic inflammation plays an important role in the development and progression of type 2 diabetes mellitus (T2DM), the role of the inflammatory process in the occurrence and progression of diabetes complications has been demonstrated before when high doses of sodium salicylate administration were found to decrease glycosuria in diabetics [3]. Other inflammatory markers found to be implicated are TLC count, CRP, and interleukin 6 (IL-6) [4].

N/L ratio was found to be associated with an elevation of pro-inflammatory cytokine [5, 6]. So the use of the N/L ratio as a cheap, rapid inflammatory marker for early detection of microvascular diabetic complications (retinopathy, nephropathy, and neuropathy) and rapid early intervention has been very effective and represents a great advance in diabetes management that reduce morbidity incidence as amputation, blindness, and renal failure and also the risk of mortality [7].

Methodology

Aim of the work

Our study aims to prove that NLR could be used as a predictor of microvascular complications during follow-up of elderly patients with type 2 diabetes.

Study design and population

This case-control study was carried out in Kasr Elainy Hospital, Cairo University, Egypt, in the period extending from May 2018 to May 2019. The study included 60 patients of more than 65 years old of both sex, on the presentation full history with full clinical examination taken. The population of the study was divided into two groups: Group A: 30 patients with type 2 diabetes with hypertension and Group B: 30 patients with type 2 diabetes without hypertension.

Lab investigations

  1. 1.

    Blood test (CBC with differential): (NLR) is estimated by dividing the absolute neutrophil ratio by the absolute lymphocyte ratio.

  2. 2.

    Urine test, albumin/creatinine ratio. All patients were surveyed on age, sex, duration of diabetes, chronic disease, family history of diabetes, smoking, and drugs that could affect inflammation like anti-dyslipidemia agents, (NSAIDs), opioids’ analogs, hormonal contraceptives, steroids, immunosuppressive, and immune-modulators drugs.

  3. 3.

    Patients with type 2 diabetes mellitus according to ADA diagnostic criteria, from both sex ≥ 65 years old were included.

  4. 4.

    ADA criteria: HbA1c ≥ 6.5, or FPG ≥ 126 mg/dl (7.0 mmol/L), or 2-h PG ≥ 200 mg/dl (11.1 mmol/L), or a random plasma glucose ≥ 200 mg/dl (11.1 mmol/L) in patients with classic symptoms of hyperglycemia, or hyperglycemic crises.

  5. 5.

    Any patient with acute or chronic infections and inflammatory conditions for the last 2 weeks, chronic heart failure, history of myocardial infarction or coronary artery disease, hematological disorders, autoimmune diseases, cancers, severe liver or kidney disorders, and active smokers were excluded.

Statistical analysis

Data were analyzed using SPSS software version 21 (IBM SPSS, Armonk, NY, USA) and presented as median and range or mean and standard deviation. Comparisons between patients’ groups were analyzed using an independent t test and one-way ANOVA. Pearson’s χ2 was used to determine the significance of associations between categorical variables. A person’s test was used to detect the strength of concordance between two variables. Receiver operating characteristic (ROC) curves were used to assess the diagnostic power of the NL ratio for diabetic complications. P < 0.05 (two-tailed) was considered to indicate a statistically significant difference.

Results

More than 70 patients were seen, but some were not included for many reasons (evidence of current infection, evidence of other chronic infections ex. HCV, abnormal lab test ex. abnormal low platelet count) (Table 1).

Table 1 Patient characteristics
Full size table

There is no statistically significant difference between Groups 1 and 2 respecting the following variables: age, gender, family history, duration of diabetes or hypertension, BMI, patients on single or double oral hypoglycemic drugs (OHD), and patients on aspirin, statin, or insulin.

Regarding the laboratory hematological indices, there was a statistically significant difference in Hb level between Group 1 (12.5 ± 0.9) and Group 2 (13.3 ± 1.3), with a P value of 0.011, while other studies of hematological indices did not differ statistically (P>0.05).

There is no significant difference between the hematological indices of Retinopathy, neuropathy, and nephropathy in the two groups of patients with P values of 0.195, 0.424, and 0.492, respectively (Table 2).

Table 2 Laboratory data of the patients’ groups
Full size table

In Group 1, there is a significant association between neutrophil-lymphocyte ratio and retinopathy (P=0.006, Fig. 1), while there is no statistically significant association between NL ratio and neuropathy (P=0.905, Fig. 2). There is a significant association between neutrophil-lymphocyte ratio and macro-albuminuria (P= 0.042, Fig. 3).

Fig. 1
figure 1

The difference in the mean value of NLR in geriatric patients screened for retinopathy in Group 1 (T2DM with HTN) and Group 2 (only T2DM) individually

Full size image
Fig. 2
figure 2

The difference in the mean value of NLR in geriatric patients screened for neuropathy in Group 1 (T2DM with hypertension) and Group 2 (only T2DM) individually

Full size image
Fig. 3
figure 3

The difference in the mean value of NLR in geriatric patients who screened for nephropathy with A/C ratio in both Group 1 (DM + HTN) and Group 2 (only DM) individually

Full size image

In Group 2, there was no statistically significant difference relating NLR with retinopathy, neuropathy, or nephropathy with P values 0.268, 0.390, and 0.342, respectively, (Figs. 1, 2, and 3).

There is a significant correlation between nephropathy (a/c) and NL ratio in Group 1 (r = 0.473, P = 0.008). Also, there is a significant correlation between nephropathy (a/c) and NL ratio in Group 2 (r = 0.406, P = 0.026). There is a significant correlation between nephropathy (c/a) and NL ratio in all patients (r = 0.44, P < 0.001, Fig. 4). NLR shows a ~68% sensitivity and ~66% specificity, for detecting retinopathy with a cutoff value of 1.58, in all patients. NLR shows a ~70% sensitivity and ~69% specificity, for detecting retinopathy with a cutoff value of 1.6, in Group 1 patients. NLR shows a ~78% sensitivity and ~70% specificity, for detecting nephropathy with a cutoff value of 1.92, in all patients. NLR shows a ~67% sensitivity and ~71% specificity, for detecting nephropathy with a cutoff value of 1.98, in Group 1 (DM + HTN) patients (Table 3).

Fig. 4
figure 4

Showing the correlation of NLR and DN in T2DM geriatric patients

Full size image
Table 3 ROC curve analysis for diagnosis
Full size table

Discussion

Changes in human behavior and lifestyle over the last century have resulted in a dramatic increase in the incidence of diabetes worldwide [8]. NLR is a new, simple, and cheap marker of subclinical inflammation that has recently been used as a systemic inflammatory marker in chronic disease as well as a predictor of prognosis in cardiovascular diseases and malignancies [7]. Our study showed that elderly diabetic patients (with or without hypertension) with microvascular complications had a higher NLR value than those without complications. These results showed agreement with Öztürk et al., a study conducted on 242 elderly patients with type 2 DM (145 diabetic patients with complications and 97 diabetic patients without complications) and 218 control subjects; this study showed that patients with microvascular complications had higher NLR than patients without complications and controls [7].

Our study showed that NLR is significantly elevated in elderly diabetic patients with nephropathy, to diabetic patients without and this shows agreement with Ciray et al. [9,10,11].

Ateia et al.’s study on 100 patients showed that there is a significant increase in NLR and its relation to albuminuria in the group with macro-albuminuria, compared to the group of those with micro-albuminuria [12]. This comes from our results (Fig. 3).

A meta-analysis study by Liu et al. on 1911 diabetic patients showed that patients with macro-albuminuria (MA) have higher levels of NLR. This indicates that NLR contributes to the development and progression of diabetic nephropathy [13]. Regarding retinopathy, our study revealed that diabetic patients with hypertension and retinopathy had a higher statistically significant NLR. However, the relation between the NLR and retinopathy was statistically insignificant in elderly patients who have only diabetes. This was in agreement with Ciray et al., who targeted 114 patients with diabetes, and showed that NLR did not differ statistically regarding retinopathy [9]. According to Akdoğan et al., in a study that targeted 278 patients with type 2 diabetes and 107 healthy control, their NLR was not affected by retinopathy [14]. However, Öztürk et al.’s and Liu et al.’s studies found a statistically significant difference between NLR and retinopathy in elderly diabetes [7, 13].

As regards the NLR and its relation to neuropathy, our study showed no significant relation between elevated NLR and neuropathy. Onalan et al.’s study that included 100 diabetic patients and 100 controls and Chittawar et al.’s study that included 115 patients with diabetes agreed with our results [15, 16]. However, Öztürk et al.’s study showed a significant difference in relating NLR to peripheral neuropathy in elderly diabetic patients [7]. The disproportionate distribution of the sample was a significant aspect that may have impacted the reliability of our results. Regarding neuropathy, we had 53 of our 60 patients with neuropathy, and thus, only 7 participants were free of neuropathy.

Our results showed that the cutoff value for predicting the microvascular complication in the elderly diabetic for retinopathy and nephropathy were 1.58 and 1.92, respectively; however, Öztürk et al.’s study conducted in Turkey and included 242 elderly diabetes patients showed that the cutoff values for NLR are 2.4 and 2.9 for retinopathy and nephropathy, respectively [7]. This difference might be due to geographical, environmental, and lifestyle differences and also the need for a larger sample size study.

Observing our results between the elderly diabetic with hypertension and the elderly diabetic without hypertension, our study showed that hypertension has a significant effect on the increase of NLR value. This was in concordance with a study done by Imtiaz et al., which measured the NLR as a predictor of systemic inflammation in the prevalent chronic diseases in 1070 of the Asian population and found a significant association between hypertension with high NLR levels [17]. Also, Gang et al.’s study agreed with our result and demonstrated that NLR value increased in hypertensive patients and could predict even it independently by age, sex, body mass index, smoking, drinking, triglyceride, and creatinine [18].

Conclusion

NLR is a simple, cheap tool that could be used as a marker to predict microvascular complications in type 2 geriatric diabetic patients mainly in predicting nephropathy followed by retinopathy. It is advised to take the baseline NLR and follow this up, as our data showed that increased albuminuria (macro-albuminuria compared with micro-albuminuria) had a higher NLR ratio.