• Users Online: 3152
  • Print this page
  • Email this page

 
Table of Contents
ORIGINAL ARTICLE
Year : 2022  |  Volume : 11  |  Issue : 3  |  Page : 157-161

Effect of hyperglycaemia on outcome of critically ill patients with and without diabetes mellitus admitted to medical intensive care unit


1 Department of Medicine, Narayana Medical College, Nellore, India
2 Department of Medicine, Sri Venkateswara Institute of Medical Sciences, Tirupati, India
3 Department of Endocrinology and Metabolism, All India Institute of Medical Sciences, Mangalagiri, Andhra Pradesh, India

Date of Submission22-Oct-2021
Date of Decision19-Apr-2022
Date of Acceptance21-Apr-2022
Date of Web Publication12-Jul-2022

Correspondence Address:
Sameeraja Vaddera
Assistant Professor, Department of Medicine, Sri Venkateswara Institute of Medical Sciences, Tirupati 517 507, Andhra Pradesh
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/jcsr.jcsr_60_21

Rights and Permissions
  Abstract 


Background: Hyperglycaemia among critically ill patients is associated with nosocomial infections, multi-organ dysfunction and prolonged hospitalisation. Sparse data are available regarding the effect of hyperglycaemia on mortality in medical intensive care unit (MICU) patients in India.
Methods: A prospective study was conducted in MICU at a tertiary teaching hospital in Southern India during the period of March 2018–June 2019. Patients were classified as having/not having diabetes mellitus. Blood glucose was monitored in all patients with. Age, severity of critical illness, comorbidities and laboratory variables were recorded. The association of hyperglycaemia with mortality was studied.
Results: Patients with diabetes mellitus were older, had more number of comorbidities like hypertension, coronary artery disease compared to those without diabetes mellitus. On logistic regression analysis mean blood glucose did not emerge as an independent determinant of mortality in patients with and without diabetes mellitus. The acute physiology and chronic health evaluation II (APACHE II) score was found to be an independent determinant of mortality in patient with (P = 0.0001) and without (P = 0.0001) diabetes mellitus.
Conclusion: Our observations suggest that in critically ill patients with or without diabetes mellitus admitted to MICU, hyperglycaemia was not a predictor of mortality.

Keywords: Critically ill, hyperglycaemia, mortality


How to cite this article:
Lakshmi T S, Katyarmal D T, Vaddera S, Vaikkakara S, S. Sarma K V. Effect of hyperglycaemia on outcome of critically ill patients with and without diabetes mellitus admitted to medical intensive care unit. J Clin Sci Res 2022;11:157-61

How to cite this URL:
Lakshmi T S, Katyarmal D T, Vaddera S, Vaikkakara S, S. Sarma K V. Effect of hyperglycaemia on outcome of critically ill patients with and without diabetes mellitus admitted to medical intensive care unit. J Clin Sci Res [serial online] 2022 [cited 2022 Aug 12];11:157-61. Available from: https://www.jcsr.co.in/text.asp?2022/11/3/157/350740




  Introduction Top


Hyperglycaemia is commonly encountered in the medical intensive care unit (MICU) and has been observed to be associated with poor outcome in both children and adults.[1] The risk of mortality in critically ill patients who require intensive care for more than 5 days is 20%.[1] It was hypothesised that poorly controlled sugars among critically ill patients required prolonged ICU stay and are associated with nosocomial infections, multi-organ dysfunction and other complications. This highlights the importance of the requirement of studies on the effect of hyperglycaemia on outcome of critically ill patients. Existing studies focused mainly on management and tight glycaemic control in critically ill ICU patients. Most of these studies were conducted on surgical patients. Sparse data are available from India assessing the effect of hyperglycaemia on mortality in adults admitted to the MICU. The present study was undertaken to assess the effect of hyperglycaemia on mortality and morbidity of the patients admitted in the medical ICU at our tertiary care centre.


  Material and Methods Top


After obtaining approval from the Institutional Ethics Committee (Approval number 741 dated April 2, 2018), a prospective study was conducted on consecutive patients admitted to the MICU at Sri Venkateswara Institute of Medical Sciences, Tirupati, Andhra Pradesh, from March 2018–June 2019. Written informed consent was obtained from all study participants or the next responsible attendants if the patient was unconscious.

A thorough history followed by detailed clinical examination was done in all patients. The severity of critical illness at the time of initial admission was assessed using acute physiology and chronic health evaluation II (APACHE II) score.[2] Diabetes mellitus (DM) and pre-diabetes conditions were diagnosed and classified as per the American Diabetes Association criteria 2017.[3] Regular capillary blood glucose levels were measured daily using the same glucometer (SD Codefree, SD BIOSENSOR, Cat. No. 01GM11). The estimated glucose levels were calculated from glycosylated haemoglobin (HbA1c level using the following formula: 28.7 × HbA1c – 46.7.[4] The mean blood glucose mean blood glucose (MBG) levels over the period of stay in the MICU was calculated from capillary blood glucose readings. An intermediate target level for blood glucose of 140–180 mg/dL was aimed to be achieved.[5] The median interquartile range (IQR) difference between the estimated blood glucose (EBG) and MBG was 15.35 (-12.9–41.925). Mortality was considered as the end-point for assessing the outcome. In the 'worst case scenario analysis,' the discharge against medical advice (DAMA) patients were considered to have died.

Statistical analysis

All data were recorded in a pre-designed structured pro forma. Data entry was carried out using Microsoft Excel (Microsoft Corp, Richmond, USA). Normally distributed data were summarised as mean ± standard deviation (SD) and as median interquartile range if data is skewed. Categorical variables were compared using the Chi-square test and student t-test and presented as percentages. Continuous variables were compared using independent sample t-test; or Mann–Whitney U test as appropriate. Based on average capillary blood sugar during the hospital stay, all patients admitted to MICU were divided into quartiles and mortality is analysed with Chi-square test among the quartiles of overall patients, diabetic patients and of non-diabetic patients. In all the patients, the effect of blood glucose level on outcome was predicted by multiple linear regression method taking number of days on mechanical ventilation and length of hospital stay as dependent variables. Multivariable analysis was done taking mortality and length of hospital stay (LOS) separately as dependant variables, the impact of average blood glucose independent of other patient variables like gender, age and APACHE II score by logistic regression method. All P values were two-tailed; P < 0.05 was considered statistically significant. The statistical software IBM SPSS Statistics Version 20 (IBM Corp Somers NY, USA), Stata/IC 12 for Windows (StataCorp LP, Texas, USA) and MedCalc Version 11.3.0forWindows2000/XP/Vista/7 (MedCalcSoftwarebvba, Belgium) were used.


  Results Top


During the study period, 315 patients were screened; of these, 286 were included in the study. The reasons for exclusion were: < 18 years (n = 5); hospital stay < 24 h (n = 4); malignacy (n = 8); human immunodeficiency virus (n = 1);unwilling to participate (n = 11).

Of the 286 included patients with DM, 140 (49%) were not known to have DM; the remaining 51% were known to have DM. Among the 146 patients who were known to have DM, 31 (21.2%) were receiving corticosteroids; 21/140 (15%) (P = 0.225). Among patients without DM (n = 140), 5 had stress-induced diabetes. Patients in the diabetic group are at a higher age than the patients in the non-diabetic group with statistical significance (P < 0.0001). Male predominance was seen in in patients with and without diabetes mellitus (91 vs. 55 and 80 vs. 60). Hypertension, chronic obstructive pulmonary disease, coronary artery disease and other comorbidities were observed more commonly among diabetic patients than in non-diabetic patients. Comparison of baseline demographic characteristics of patients with and without DM is shown in [Table 1].
Table 1: Comparison of baseline demographic parameters among patients admitted to medical intensive care unit with and without diabetes mellitus

Click here to view


Including DAMA in the worst outcome, mortality was 24.8%, of which 53.5% (n = 38) in diabetic patients and 46.5% (n = 33) in non-diabetic patients with no statistical significance (P = 0.63). HbA1C and EBS were higher in diabetics when compared to non-diabetics with high statistical significance of P < 0.0001. The mean duration of hospital stay and mean duration of mechanical ventilator support were similar in both diabetic and non-diabetic patients with no statistical significance: P = 0.641 for mean hospital stay and 0.401 for mean duration of mechanical ventilation [Table 2]. A statistically significant positive correlation was noted between HbA1C and MBS in patients admitted to MICU as r value is 0.4 in diabetic and 0.27 in non-diabetic patients with a P < 0.05 0.01 [Figure 1].
Figure 1: Correlation between HbA1C and mean blood sugar in patients admitted to medical intensive care unit
HbA1C = Glycosylated haemoglobin


Click here to view
Table 2: Comparison of various parameters among patients admitted to medical intensive care unit with and without diabetes mellitus

Click here to view


In patients with diabetes mellitus, a negative correlation was observed between MBS and duration of hospital stay (r = 0.02), duration of mechanical ventilator support (r = 0.09). In non-diabetic patients, a positive correlation was observed between MBS and duration of hospital stay (r = 0.78) and duration of mechanical ventilator support (r = 0.94).

In patients without diabetes mellitus, the mean capillary blood sugar was not an independent determinant of duration of hospital stay [Table 3] and duration of mechanical ventilator support [Table 4] by multiple linear regression model. patients with diabetes mellitus, the MBS did not emerge as an independent determinant of duration of hospital stay [Table 5] and duration of mechanical ventilator support [Table 6].
Table 3: Multiple linear regression model for prediction of duration of hospital stay in patients wiithout diabetes mellitus

Click here to view
Table 4: Multiple linear regression model for prediction of duration of mechanical ventilator support in patients without diabetes mellitus

Click here to view
Table 5: Multiple linear regression model for prediction of duration of hospital stay in with diabetes mellitus

Click here to view
Table 6: Multiple linear regression model for prediction of duration of mechanical ventilator support in patients with diabetes mellitus

Click here to view


However, APACHE II score was an independent determinant of mechanical ventilator support duration in diabetic patients with a statistically significant P value of 0.0001 [Table 6].

Logistic regression analysis showed that MBS is not an independent determinant of mortality both in diabetic (P > 0.05) and non-diabetic (P > 0.05) patients. However, the APACHE II score is an independent determinant of mortality in diabetic (P = 0.0001) [Table 7] and non-diabetic (P = 0.0001) patients [Table 8].
Table 7: Comparison of mortality among patients without diabetes mellitus by logistic regression method

Click here to view
Table 8: Comparison of mortality among patients with diabetes mellitus by logistic regression method

Click here to view



  Discussion Top


The majority of patients requiring admission to MICU in the present study belonged to the sixth decade of life with a mean age of 57.8 ± 15.5 years; this can be due to the higher prevalence of DM in that age group. A similar demographic trend was evident in a studies from Kolkata[6] and Bengaluru,[7] the mean age of presentation was 61 ± 16.7 and 58.1 ± 12.3 years.

In the present study, men outnumbered women with a ratio of 1.4:1. A similar trend is noticed in studies conducted in Kolkata where male to female ratio is 1.4:1 and in Bengaluru with a ratio of 1.8:1. We had observed that diabetes mellitus was more common among males; the observation was similar to the globally reported results.[8]The mortality in critically ill depends on several factors and hyperglycaemia is one such factor. Mortality rate observed in the present study was 53.5% (n = 38) in patients with diabetes mellitus and 46.5% (n = 33) in patients without diabetes mellitus (P = 0.630). In a study from Israel,[10] glucose control was demonstrated to be independently associated with decreased mortality (odds ratio [OR] 0.286, 95% confidence interval [CI] 0.086–0.951, P = 0.041). In our study, analysis by quartiles revealed no statistical significance in mortality with a P > 0.05 in both diabetic and non-diabetic patients. In a study,[9] the authors had divided patients into quartiles based on SD of blood sugar levels. Higher SD quartile was associated with increased in-hospital mortality in patients without DM and in patients with DM. In a study[6] conducted in India, higher mortality was observed with statistical significance (OR: 2.023, 95% CI: 1.483–2.758) in the highest quartiles of SD.

In the present study, multiple linear regression analysis showed no statistical significant association between average blood sugar level and length of hospital stay (LOS) both in diabetic and non-diabetic patients. In a study from Israel,[10] patients with and without DM had longer LOS in patients with increased glycaemic variability after adjustment for age, sex, smoking, alcohol, body mass index and comorbidities. Higher SD quartile was associated with longer LOS in patients without DM and in patients with DM.

Taking mortality and LOS separately as dependent variables, the impact of average blood glucose independent of other patient variables like age, gender, APACHE II score, mean capillary blood sugar did not emerge as an independent determinant of mortality and LOS both in diabetic and non-diabetic patients. However, the APACHE II score emerged as an independent determinant of mortality both in diabetic and non-diabetic patients with a statistically significance P = 0.0001. In a study[11] conducted in Australia, a higher APACHE II score was significantly associated with mortality (P < 0.001).

In the present study, n = 71 (24.8%) patients had poor outcome (died n = 71); DAMA n = 26). Mortality is observed in n = 38 (53.5%) in diabetic patients and n = 33 (46.5%) in non-diabetic patients with no statistical significance showing P > 0.05. Patients with a higher APACHE II score had higher mortality (P < 0.0001). Of 286 patients included in the study both diabetic and non-diabetic patients were divided into quartiles, mortality, duration of hospital stay and duration of mechanical ventilator support were compared showing no significance between the quartiles (P > 0.05).

In the present study, hyperglycaemia per se does not seem to solely affect the mortality in the patients admitted to MICU. However, mortality is influenced by several other factors mainly by clinical diagnosis and severity of illness as depicted by high mortality in patients with high APACHE II score at the time of admission.

The present study is a single centre study, whether the observations can be extrapolated to other hospitals/institutions need to be evaluated further. The mortality rate in this study depends on the diagnosis and critical condition of the patient along with hyperglycaemia.

Financial support and sponsorship

Nil.

Conflicts of interest

The authors are faculty members/residents of Sri Venkateswara Institute of Medical sciences, Tirupati, of which Journal of Clinical and Scientific Research is the official Publication. The article was subject to the journal's standard procedures, with peer review handled independently of these faculty and their research groups.



 
  References Top

1.
Pan C, Shang S, Kirch W, Thoenes M. Burden of diabetes in the adult Chinese population: A systematic literature review and future projections. Int J Gen Med 2010;3:173-9.  Back to cited text no. 1
    
2.
Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: A severity of disease classification system. Crit Care Med 1985;13:818-29.  Back to cited text no. 2
    
3.
American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2014;37:S81-90.  Back to cited text no. 3
    
4.
Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, et al. Translating the A1C assay into estimated average glucose values. Diabetes Care 2008;31:1473-8.  Back to cited text no. 4
    
5.
Preiser JC, Devos P. Clinical experience with tight glucose control by intensive insulin therapy. Crit Care Med 2007;35:S503-7.  Back to cited text no. 5
    
6.
Todi S, Bhattacharya M. Glycemic variability and outcome in critically ill. Indian J Crit Care Med 2014;18:285-90.  Back to cited text no. 6
[PUBMED]  [Full text]  
7.
Bansal B, Carvalho P, Mehta Y, Yadav J, Sharma P, Mithal A, et al. Prognostic significance of glycemic variability after cardiac surgery. J Diabetes Complications 2016;30:613-7.  Back to cited text no. 7
    
8.
Kautzky-Willer A, Harreiter J, Pacini G. Sex and gender differences in risk, pathophysiology and complications of type 2 diabetes mellitus. Endocr Rev 2016;37:278-316.  Back to cited text no. 8
    
9.
Sharif K, Ghadir S, Jakubowicz D, Amital H, Bragazzi NL, Watad A, et al. Improved outcome of patients with diabetes mellitus with good glycemic control in the cardiac Intensive Care Unit: A retrospective study. Cardiovasc Diabetol 2019;18:4.  Back to cited text no. 9
    
10.
Akirov A, Diker-Cohen T, Masri-Iraqi H, Shimon I. High glucose variability increases mortality risk in hospitalized patients. J Clin Endocrinol Metab 2017;102:2230-41.  Back to cited text no. 10
    
11.
Egi M, Bellomo R, Stachowski E, French CJ, Hart G. Variability of blood glucose concentration and short-term mortality in critically ill patients. Anesthesiology 2006;105:244-52.  Back to cited text no. 11
    


    Figures

  [Figure 1]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]



 

Top
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Material and Methods
Results
Discussion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed244    
    Printed2    
    Emailed0    
    PDF Downloaded19    
    Comments [Add]    

Recommend this journal