DIAGNOSTIC PITHINESS OF DATA OF THE NUTRITIVE STATUS IN PATIENTS WITH ACUTE RESPIRATORY DISTRESS SYNDROME DURING NUTRITION SUPPORT Girsh A.O., Maksimishin S.V.
Omsk State Medical University,
City Clinical Hospital of Emergency Medical Care No.1, Omsk, Russia
Realization of nutritive support in patients with acute respiratory distress syndrome (ARDS) has some features and difficulties [1, 2]. Moreover, the diagnostic value of estimation criteria for protein and energy deficiency in patients with ARDS has not been estimated yet, resulting in obstacles for correct and adequate estimation of efficiency of techniques of clinical nutrition. As result, the objective of this study was to estimate pithiness of data of the nutritive status in patients with acute respiratory distress syndrome when carrying out various schemes of nutritious support.
MATERIALS AND METHODS
The article presents the results of the studies of 198 patients (mean age of 29.7 ± 4.1) with ARDS (the table 1) treated in the ICU. ARDS was diagnosed after 39 ± 6 hours in all patients on the basis [3]: oxygenation index (OI = РаО2/FiO2), computer imaging of thoracic organs (presence of non-homogenous infiltration in the lungs, especially in posterior-inferior parts), a decrease in pulmonary tissue pneumatization by the type of frosted glass, a decrease in mean value of optical density and an increase in optical density in any chosen sections), X-ray examination (presence of bilateral infiltrates in the lungs to the right and to the left), a risk factor (traumatic shock of degrees 2-3), acute initiation (within 72 hours), absence of clinical signs of left ventricular insufficiency. The inclusion criteria were: 1) age of 18-40; 2) mild, moderate or severe ARDS with need for stay in ICU; 3) artificial lung ventilation (ALV) for patients with mild, moderate and severe ARDS according to clinical guidelines of All-Russian public organization – Federation of Anesthesiologists and Intensivists; 4) more than 72 hours in ICU; 5) presence of clinical and laboratory signs of nutritive insufficiency of any severity. The exclusion criteria were: 1) contraindications for nutritive support (absence of risk of development or signs of nutritive insufficiency, refractory shock syndrome); 2) intolerance of media for enteral and parenteral nutrition; 3) severe metabolic acidosis (pH of arterial blood < 7.2); 4) severe arterial hypoxemia not corrected with ALV (PaO2 < 60 mm Hg).
Table 1
Degrees of intensity of insufficient nutrition and used schemes of nutritive therapy for patients with ARDS
|
Groups of patients (n, %) with consideration of used nutritive therapy and values of nutritive status |
Mild ARDS (200 mm Hg < IO = PaO2/FiO2 ≤ 300 mm Hg), n (%) |
Average ARDS (100 mm Hg < IO = PaO2/FiO2 ≤ 200 mm Hg), n (%) |
Severe ARDS (IO = PaO2/FiO2 ≤ 100 mm Hg), n (%) |
|
Enteral nutrition |
|||
|
group 1 (n = 75, 100 %) – Nutricomp Immun (B. Braun, Germany) |
22 (29.3 %)
|
28 (37.4 %) |
25 (33.3 %) |
|
Parenteral nutrion |
|||
|
group 2 (n = 66, 100 %) – three in one system Nutriflex 70/180 lipid (B. Braun, Germany) |
21 (31.8 %)
|
23 (34.9 %) |
22 (33.3 %) |
|
Mixed (enteral and parenteral) nutrition |
|||
|
group 3 (n = 57, 100 %) – Nutricomp Immun (B. Braun, Germany) + three in one system Nutriflex 70/180 lipid (B. Braun, Germany) |
19 (33.1 %) |
18 (31.6 %) |
20 (37 %) |
|
Total: 198 (100 %) |
62 (31.3 %) |
69 (34.9 %) |
67 (33.8 %) |
|
Nutritive status values |
|
||
|
Albumin, g/l |
24 (23; 25) |
23 (22; 24) |
22 (20; 24) |
|
Transferrin, g/l |
1.5 (1.4; 1.6) |
1.4 (1.3; 1.5) |
1.3 (1.2; 1.5) |
|
Lymphocytes, cells per ml3 |
0.8 (0.7; 0.9) |
0.8 (0.7; 0.9) |
0.7 (0.5; 0.9) |
|
Energy requirements, kcal |
3076 (3010; 3247) |
3213 (3170; 3426) |
3580 (3396; 3624) |
The clinical nutrition was initiated for all patients 74.5 ± 4.3 hours after admission to ICU. The indications for nutritive (enteral and/or parenteral) support were [1]: 1) critical condition (ARDS after traumatic shock); 2) ALV > 48 hours; 3) presence of hypermetabolism syndrome, which determines the development of severe nutritive insufficiency. Severe nutritive insufficiency was registered in all patients participating in the study (the table 1). Enteral nutrition was conducted for patients with preserved gastrointestinal functions [4-7]. Parenteral nutrition was realized for patients with gastrointestinal dysfunction with impossibility for enteral nutritive support [1, 4-7]. Mixed nutrition was prescribed for patients with evident catabolism and preserved gastrointestinal functions [1, 6].
The parameters of nutrition insufficiency severity were determined [1]: albumin (g/l) and transferrin (g/l) with the automatic analyzer Hitachi 902, Roche Diagnostics (Switzerland). SysmexXT 4000i hematologic analyzer (Symex, USA) was used for calculation of absolute number of lymphocytes (thousand per mcl) in venous blood. Energetic requirement was estimated on the basis of indirect calorimetry with MPR 6-03 (Triton, Russia) with function of metabolimeter. Body mass index was calculated – BMI (kg/m2) = actual body mass (kg) / height (m2). The diagnostic value of the studied indices determining the nutritive modality degree was estimated with the following criteria. These criteria show the operation characteristics of the studied parameters, which were calculated with four-field tables (the table 2) [8]: 1. Sensitivity (%) (Se – proportion of patients with this symptom (the positive result) or incidence of the symptom in patients) = A/(A +C) x 100 %. 2. Specificity (%) (Sp – incidence of absence of the symptom in healthy individuals) = D/(B + D) x 100 %. Efficiency of the method (%) (the rate of false-positive results) of screening = 100 – SP (%). 4. Predictive value positive (PVP – determined as incidence of coincidence with a disease) = A/(A + B). 5. Predictive value negative (%) (PVN – determined as incidence of coincidence with absence of a disease = D/(C + D).
Table 2
Estimation of diagnostic pithiness of data with use of four-field table
|
Outcomes |
Result of use of reference test for calculation of operational characteristics in qualitative estimation of value |
|
|
Poor |
A (true positive) |
B (false positive) |
|
Favorable |
C (false negative) |
D (true negative) |
The study was conducted on the basis of approval from the bioethical committee of City Clinical Hospital of Emergency Medical Care No.1 and corresponded to the ethical standards of Helsinki Declare – Ethical Principles for Medical Research with Human Subjects 2000, and the Rules for Clinical Practice in the Russian Federation confirmed by the Order of Health Ministry of RF, 19 June 2003, No.266.
RESULTS
The results of the study show that the key diagnostic criterion of nutritive insufficiency in patients with ARDS is energy requirement (the tables 3, 4, 5). Actually, this criterion is highly specific and sensitive, but also with low incidence of false-positive results, and with predictive value of positive results and low predictive value of negative results. Therefore, the use of the energy requirement value for estimation of severity of nutritive insufficiency is quite substantiated and statistically significant. The low percentage of substantiation of qualitative diagnosis of protein-energy insufficiency for patients with ARDS and the low incidence of coincidence with the studied disease, as compared to energy requirement value, were common for such parameters as albumin, transferrin and absolute number of lymphocytes (the table 3, 4, 5). Certainly, the use of these parameters as the main diagnostic signs of nutritive insufficiency in patients with ARDS is quite substantiated and statistically significant. Therefore, the use of this value for diagnosis of nutritive modality in patients with ARDS should be combined with estimation of energy requirement with aim of more objective estimation. BMI was irrelevant for intensity of protein-energy insufficiency in patients with ARDS (the tables 3, 4, 5). It was associated with the fact that this criterion had the low sensitivity and specificity in estimation of nutritive modality in patients with ARDS, and also had the high rate of false-positive results, low predictive value of positive results and high negative predictive value. As result, the use of BMI for estimation of nutritive insufficiency is quite inadequate and untrue.
Table 3
Findings of protein and energy deficiency in patients with enteral nutrition
|
Nutritive status in patients with mild ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
1 g |
1 g |
1 g |
1 g |
1 г/g |
||
|
Albumin, g/l |
58.4 |
57.8 |
42.2 |
55.3 |
56.1 |
|
|
Transferrin, g/l |
61.3 |
60.9 |
39.1 |
59.7 |
58.4 |
|
|
Cell lymphocytes per mm3 |
55.7 |
56.4 |
43.6 |
57.2 |
54.5 |
|
|
Energy requirements, kcal |
87.8 |
85.8 |
14.2 |
83.5 |
81.4 |
|
|
BMI, kg/m2 |
33.7 |
36.2 |
63.8 |
34.5 |
35.7 |
|
|
Nutritive status in patients with moderate ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
2 g |
2 g |
2 g |
2 g |
2 g |
||
|
Albumin, g/l |
57.6 |
56.3 |
43.7 |
53.6 |
54.3 |
|
|
Transferrin, g/l |
60.6 |
61.4 |
38.6 |
58.2 |
56.9 |
|
|
Cell lymphocytes per mm3 |
53.6 |
54.7 |
45.3 |
55.4 |
53.7 |
|
|
Energy requirements, kcal |
86.2 |
87.7 |
13.3 |
85.8 |
84.6 |
|
|
BMI, kg/m2 |
32.3 |
34.5 |
65.5 |
33.7 |
34.8 |
|
|
Nutritive status in patients with severe ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
3 g |
3 g |
3 g |
3 g |
3 g |
||
|
Albumin, g/l |
56.4 |
55.9 |
44.1 |
54.7 |
55.2 |
|
|
Transferrin, g/l |
62.3 |
63.5 |
36.5 |
59.6 |
58.4 |
|
|
Cell lymphocytes per mm3 |
51.3 |
52.2 |
47.8 |
53.7 |
52.9 |
|
|
Energy requirements, kcal |
88.1 |
89.3 |
10.7 |
86.1 |
87.5 |
|
|
BMI, kg/m2 |
30.4 |
31.5 |
68.5 |
32.4 |
31.6 |
|
Table 4
Findings of protein and energy deficiency in patients with parenteral nutrition
|
Nutritive status in patients with mild ARDS |
Criteria |
||||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
|||
|
1 g |
1 g |
1 g |
1 g |
1 g |
|||
|
Albumin, g/l |
57.2 |
56.3 |
43.7 |
54.6 |
55.4 |
||
|
Transferrin, g/l |
60.7 |
61.2 |
38.8 |
57.4 |
56,5 |
||
|
Cell lymphocytes per mm3 |
54.6 |
55.7 |
44.3 |
56.5 |
55.1 |
||
|
Energy requirements, kcal |
85.4 |
84.8 |
15.2 |
82.1 |
83.2 |
||
|
BMI, kg/m2 |
32.4 |
34.5 |
65.5 |
32.7 |
34.6 |
||
|
Nutritive status in patients with moderate ARDS |
Criteria |
||||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
|||
|
2 g |
2 g |
2 g |
2 g |
2 g |
|||
|
Albumin, g/l |
56.5 |
55.8 |
44.2 |
55.7 |
54.8 |
||
|
Transferrin, g/l |
61.4 |
62.1 |
37.9 |
56.2 |
55.7 |
||
|
Cell lymphocytes per mm3 |
55.6 |
54.3 |
45.7 |
54.9 |
57.3 |
||
|
Energy requirements, kcal |
87.2 |
86.9 |
13.1 |
85.7 |
86.5 |
||
|
BMI, kg/m2 |
34.7 |
33.1 |
66.9 |
31.4 |
32.8 |
||
|
Nutritive status in patients with severe ARDS |
Criteria |
||||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
|||
|
3 г/g |
3 г/g |
3 г/g |
3 г/g |
3 г/g |
|||
|
Albumin, g/l |
55.7 |
54.3 |
45.7 |
52.8 |
53.4 |
||
|
Transferrin, g/l |
60.2 |
61.7 |
38.3 |
55.9 |
54.2 |
||
|
Cell lymphocytes per mm3 |
54.1 |
53.7 |
46.3 |
52.8 |
53.7 |
||
|
Energy requirements, kcal |
88.1 |
87.4 |
12.6 |
87.1 |
88.2 |
||
|
BMI, kg/m2 |
33.3 |
32.5 |
67.5 |
32.5 |
31.3 |
||
Table 5
Findings of protein and energy deficiency in patients with mixed nutrition
|
Nutritive status in patients with mild ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
1 g |
1 g |
1 g |
1 g |
1 g |
||
|
Albumin, g/l |
53.8 |
52.9 |
47.1 |
51.6 |
52.1 |
|
|
Transferrin, g/l |
61.4 |
62.3 |
37.7 |
57.1 |
56.9 |
|
|
Cell lymphocytes per ml3 |
52.7 |
51.3 |
48.7 |
50.4 |
51.2 |
|
|
Energy requirements, kcal |
89.6 |
88.7 |
11.3 |
89.2 |
87.4 |
|
|
BMI, kg/m2 |
31.6 |
32.1 |
67.9 |
30.3 |
30.7 |
|
|
Nutritive status in patients with moderate ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
2 g |
2 g |
2 g |
2 g |
2 g |
||
|
Albumin, g/l |
52.4 |
51.7 |
48.3 |
50.6 |
51.1 |
|
|
Transferrin, g/l |
60.5 |
61.7 |
38.3 |
56.8 |
55.9 |
|
|
Cell lymphocytes per ml3 |
53.4 |
53.6 |
46.4 |
52.7 |
53.8 |
|
|
Energy requirements, kcal |
88.1 |
87.4 |
12.6 |
88.8 |
89.2 |
|
|
BMI, kg/m2 |
30.3 |
31.6 |
68.4 |
31.2 |
30.9 |
|
|
Nutritive status in patients with severe ARDS |
Criteria |
|||||
|
Sensitivity (%) |
Specificity (%) |
Price of technique (%) |
Prediction of favorable outcome (%) |
Prediction of poor outcome (%) |
||
|
3 g |
3 g |
3 g |
3 g |
3 g |
||
|
Albumin, g/l |
51.7 |
52.1 |
47.9 |
51.5 |
50.4 |
|
|
Transferrin, g/l |
61.2 |
62.4 |
37.6 |
55.7 |
55.1 |
|
|
Cell lymphocytes per ml3 |
52.9 |
51.7 |
48.3 |
51.5 |
52.6 |
|
|
Energy requirements, kcal |
89.3 |
88.7 |
11.3 |
89.4 |
89.8 |
|
|
BMI, kg/m2 |
31.5 |
30.7 |
69.3 |
31.8 |
32.1 |
|
DISCUSSION
The high diagnostic significance of such parameter as energy requirement for estimation of protein-energy deficiency in patients with ARDS was determined by its pathogenetic features [2, 3], namely by development of systemic inflammatory response, which significantly increases energy losses [1, 6]. Moreover, ARDS cause the negative energy balance by means of significant breakdown and oxidation of glycogen and protein, and by means of intense lipolysis [4, 5].
The second (according to significance) estimation criterion of nutritive condition was transferrin. It was associated with its level in vascular bed and with short period of half-life (8 days) as compared to albumin (20 days) [1]. It makes transferrin more sensitive criterion (as compared to albumin) in relation to depletion of visceral pool of protein [4, 8]. However the transferrin level highly depends on plasma level of iron and on organ and system disorders, particularly on renal and hepatic insufficiency [1]. Moreover, significance of estimation of transferrin is limited in hypoferric anemia, which exists before or appears during the course of a disease, resulting in compensatory increase in the blood even in conditions of protein deficiency [5].
The time course of changes in serum albumin is insufficiently reliable for estimation of adequacy of nutritive (both enteral and perenteral) support and monitoring of time course of changes in nutritive modality [4, 5]. Moreover, the plasma level of albumin depends on presence of organ and system disorders, particularly, hepatic insufficiency [1, 7].
Absolute amount of peripheral blood lymphocytes can be used for estimation of both nutritive status and cellular link of immunity [1], which is the most significant response to infection 72 hours after disease onset [4]. Moreover, the synthesis of immune system cells, which are necessary for adequate response to immune inflammation, requires the absence of evident protein and energy deficiency [5]. Absolute amount of peripheral blood lymphocytes can be used as a secondary parameter for estimation of nutritive status [1]. It is associated with the fact that the absolute amount of lymphocytes in peripheral blood is influenced by presence of renal and hepatic failure, electrolytic disorders, infection, hypoalbuminemia, metabolic stress, chronic diseases, and use of drugs for immune suppression action [4, 5].
Low value of BMI for estimation of intensity of protein-energy insufficiency is related to the fact that critically ill patients, particularly with ARDS, have the fast and high body weight gain, which is determined by fluid retention during fluid load [1, 4, 5].
CONCLUSION
1. Correct diagnostic estimation of nutritive status in patients with ARDS requires for estimation of energy requirement as the value with high sensitivity and specificity, with use of adequate estimation of efficiency of clinical nutrition in such patients.
2. Transferrin, serum albumin and absolute amount of peripheral blood lymphocytes, which are less informative, should be used for diagnostic estimation of intensity of protein-energy deficiency in patients with ARDS as the secondary criteria in combination with energy requirement.
3. BMI cannot be used for patients with ARDS for estimation of nutritive modality due to its low diagnostic value.
Information on financing and conflicts of interests
The study was conducted without sponsorship. The authors declare the absence of any clear or potential conflicts of interests relating to publication of this article.