Summary
This study aimed to elucidate the perioperative body composition alterations and their prognostic utility for hospital length of stay in low risk, normal- and increased-body mass index, off-pump coronary artery bypass grafting (OPCAB) patients. All body composition parameters assessed by bioelectrical impedance analysis (BIA) were altered in a time-dependent fashion. Patients with longer hospitalization had significantly higher baseline content of fat mass (FM%) and the fat mass index (FMI), and significantly lower content of fat free mass (FFM%) (p = 0.011, p = 0.04 and p = 0.012, respectively). High FM% values had 15-fold, low FFM% values had 13-fold and high FMI values had 7-fold higher risk for longer stay in the hospital (p = 0.001, p = 0.001 and p = 0.005, respectively). The combined panel of three variables (higher FM%, lower FFM% and higher FMI) had 16-fold higher risk of longer hospitalization (adjusted OR = 16.40, 95% CI: 3.52–76.34; p = 0.0004). A higher preoperative FM and lower FFM content might be potential indicators of prolonged hospitalization in normal weight or overweight patients subjected to OPCAB.
Introduction
Nutritional status assessment is mainly aimed at identifying malnourishment. However, in patients referred for coronary artery bypass grafting (CABG), the percentage of overweight and obese patients systematically increases [1]. Higher indices of central adiposity are associated with an increased risk of coronary artery disease (CAD) and cardiovascular mortality, irrespectively of the body mass index (BMI) measures, even among those with normal weight [2, 3].
The CABG procedure triggers a systemic inflammatory response that is far more intense if extracorporeal circulation is employed [4]. Contrary to the classic surgery, off-pump CABG (OPCABG) results in a lesser extent of endothelial activation and increased microvascular permeability. In effect, less plasma proteins leak, and less fluid accumulates in the interstitial space [5].
Bioelectrical impedance analysis (BIA) allows for noninvasive, functional and cheap estimation of body compartments. It has been widely studied for assessment of body composition and hydration status in patients undergoing cardiac surgery [6–12]. Compartmental differentiation with BIA allows for superior detection of subtle membrane dependent fluid imbalances, over the classic anthropometric measurements.
Cardiopulmonary bypass employment during CABG and the patient’s initial state of malnutrition significantly affect body composition assessment results. Therefore, this pilot study was carried out on a group of OPCAB patients without clinical signs of malnutrition. Very few BIA reports link the postoperative BIA values to the clinical outcome after OPCAB, while most relevant reports focus on the effects of preoperative malnutrition on morbidity and mortality [13–17]. This study aimed to elucidate the utility of body composition parameters, namely fat mass (FM), fat free mass (FFM) and body water content, in refining in-hospital prognostic accuracy of OPCAB patients.
Aim
We hypothesize that BIA might be a potential indicator of prolonged hospitalization in normal- and increased-BMI patients subjected to OPCAB.
Material and methods
Study population
Fifty patients (40 men, 80%; 10 women, 20%) were consecutively recruited from all patients scheduled for elective OPCAB at the Department of Cardiac Surgery at the Medical University of Warsaw between October 2018 and July 2019. The study protocol was approved by the Institutional Review Board (KB/144/2018). All provided written informed consent.
The inclusion criteria were as follows: (i) elective OPCAB, (ii) age 18-80 years, (iii) left ventricle ejection fraction (LVEF) more than 40%, (iv) bioimpedance testing performed preoperatively and postoperatively, (v) BMI ≥ 22 kg/m2, (vi) Mini Nutritional Assessment questionnaire (MNA) total score ≥ 17, (vii) serum albumin levels > 3.5 g/dl, (viii) provided written informed consent.
Exclusion criteria were as follows: (i) cardiovascular implantable electronic device (contraindication to bioimpedance testing), (ii) clinical instability requiring emergent surgery, (iii) complicated intraoperative course, (iv) intraoperative insertion of intra-aortic balloon pump (IABP), (v) history of congenital heart abnormality or previous cardiac surgery, (vi) cognitive impairment hindering cooperation.
The MNA is a simple screening tool for classifying subjects into three groups: well-nourished (≥ 23 points), at risk of undernutrition (≥ 17 points) and undernutrition (< 17 points).
The median postoperative hospital length of stay (LOS) was 8.79 days. Therefore, we grouped the cohort accordingly: < 9 days of hospitalization (n = 29, 58%) and ≥ 9 days of hospitalization (n = 21, 42%). The demographics and baseline clinical characteristics are presented in Table I.
Table I
Demographic and clinical characteristics of patients
| Variable | Overall (n = 50; 100%) | Hospital LOS < 9 days (n = 29; 58%) | Hospital LOS ≥ 9 days (n = 21; 42%) | P-value |
|---|---|---|---|---|
| Age, median (IQR) [years] | 69 (61–72.25) | 70 (61.50–72.50) | 68 (60.5–72.5) | 0.708 |
| Female sex, n (%) | 10 (20%) | 4 (13.8%) | 6 (28.6%) | 0.197 |
| Baseline characteristics: | ||||
| BMI, mean ±SD [kg/m2] | 29 ±0.54 | 28.93 ±4.05 | 29.42 ±3.52 | 0.659 |
| Normal weight (BMI 18.5–25 kg/m2), n (%) | 10 (20) | 7 (24) | 3 (14.3) | 0.390 |
| Overweight (BMI ≥ 25 and < 30 kg/m2), n (%) | 19 (38) | 10 (34.5) | 9 (42.9) | 0.547 |
| Obese (BMI ≥ 30 kg/m2), n (%) | 21 (42) | 12 (41.4) | 9 (42.9) | 0.917 |
| CCS 1, n (%) | 2 (4) | – | 2 (9.5) | – |
| CCS 2, n (%) | 28 (56) | 18 (62) | 10 (47.6) | 0.310 |
| CCS 3, n (%) | 20 (40) | 11 (37.9) | 9 (42.9) | 0.726 |
| LVEF, median (IQR) (%) | 55.00 (50.75–60.00) | 60.00 (54.50–61.50) | 55.00 (45.50–60.00) | 0.080 |
| EuroSCORE II, median (IQR) (%) | 1.21 (0.80–1.85) | 1.2 (0.78–1.80) | 1.3 (0.80–2.04) | 0.467 |
| Diabetes mellitus, n (%) | 20 (40) | 14 (48.3) | 6 (28.6) | 0.160 |
| COPD, n (%) | 4 (8) | 1 (3.4) | 3 (14.3) | 0.163 |
| CKD STAGE > 3, n (%) | 18 (36) | 9 (31) | 9 (42.9) | 0.390 |
| Hypertension, n (%) | 44 (88) | 26 (89.7) | 18 (85.7) | 0.672 |
| AF, n (%) | 5 (10) | 4 (13.8) | 1 (4.8) | 0.293 |
| Hyperlipidemia*, n (%) | 29 (58) | 18 (62.1) | 11 (52.4) | 0.493 |
| Previous MI, n (%) | 24 (48) | 14 (48.3) | 10 (47.6) | 0.963 |
| Previous PCI, n (%) | 15 (30) | 8 (27.6) | 7 (33.3) | 0.662 |
| MNA, mean ± SD [total points] | 24.24 ±2.94 | 24.52 ±2.91 | 23.86 ±3.00 | 0.439 |
| β-blockers, n (%) | 45 (90) | 24 (82.8) | 21 (100) | 0.056 |
| ACEI, n (%) | 35 (70) | 20 (69) | 15 (71.4) | 0.851 |
| Sartans, n (%) | 7 (14) | 4 (13.8) | 3 (14.3) | 0.960 |
| CCB, n (%) | 21 (42) | 14 (48.3) | 7 (33.3) | 0.291 |
| Statins, n (%) | 44 (88) | 24 (82.8) | 20 (95.2) | 0.180 |
| HbA1c, median (IQR) (%) | 6.9 (6.5–7.63) | 6.9 (6.45–7.45) | 6.6 (6.5–9.75) | 0.672 |
| Hemoglobin, mean ± SD [g/dl] | 13.7 ±0.22 | 13.53 ±1.43 | 14.01 ±1.78 | 0.303 |
| Hematocrit, mean ± SD (%) | 40.56 ±0.66 | 39.83 ±3.98 | 41.57 ±5.37 | 0.194 |
| Platelets, median (IQR) [× 103/μl] | 224 (193–268) | 221 (180.5–248.5) | 243 (205.5–294.5) | 0.371 |
| Creatinine, median (IQR) [mg/dl] | 1.01 (0.85–1.24) | 1.01 (0.90–1.22) | 1.01 (0.76–1.126) | 0.623 |
| eGFR baseline**, mean ± SD [ml/ min/1.73m²] | 71.13 ±20.88 | 69.75 ±19.50 | 73.04 ±23.00 | 0.588 |
| Postoperative parameters: | ||||
| Number of coronary grafts, median (IQR) | 2.00 (2.00–4.00) | 2.00 (2.00–3.00) | 2.00 (2.00–3.00) | 0.875 |
| ICU, median (IQR) [h] | 47.50 (43.5–51.25) | 46.00 (29.00–48.00) | 50.00 (47.50–70.50) | 0.001 |
| Mechanical ventilation, median (IQR) [h] | 6.50 (5–9) | 6.00 (4.00–8.00) | 8.0 (6.00–13.00) | 0.006 |
| Troponin 6h after OPCAB, median (IQR) [ng/ml] | 1.14 (0.56–2.06) | 0.98 (0.38–2.11) | 1.42 (0.90–2.44) | 0.160 |
| Troponin 18 h after OPCAB, median (IQR) [ng/ml] | 0.94 (0.32–2.13) | 0.78 (0.25–1.81) | 1.14 (0.61–4.36) | 0.079 |
| CK-MB 6 h after OPCAB, median (IQR) [ng/ml] | 5.15 (3.38–8.40) | 5.1 (2.55–8.30) | 5.2 (4.6–12.6) | 0.169 |
| CK-MB 18h after OPCAB, median (IQR) [ng/ml] | 5.65 (3.40–12.45) | 4.7 (2.65–7.85) | 6.30 (4.75–29.00) | 0.046 |
| MI, n (%) | 5 (10) | 2 (6.9) | 3 (14.3) | 0.390 |
| Stroke/TIA, n (%) | 0 (0) | 0 (0) | 0 (0) | – |
| Acute kidney injury, n (%) | 2 (4%) | 0 (0%) | 2 (9.5%) | 0.090 |
| Chest drainage volume 24 h after OPCAB, median (IQR) [ml] | 715 (570–883.75) | 700 (560–855) | 720 (580–985) | 0.694 |
| Total chest drainage volume after OPCAB, median (IQR) [ml] | 1030 (853.75–1383.75) | 970 (735–1335) | 1300 (903–1585) | 0.111 |
| Hemoglobin day-7, median (IQR) [g/dl] | 9.95 (9.25–11.15) | 10 (8.90–11.30) | 9.90 (9.65–11.2) | 0.821 |
| Hematocrit day-7, median (IQR) (%) | 30.45 (28.13–33.23) | 29.70 (27.05–33.25) | 31.10 (29.20–33.55) | 0.637 |
| Creatinine day-7, median (IQR) [mg/dl] | 0.97 (0.81–1.12) | 0.97 (0.83–1.10) | 0.97 (0.80–1.25) | 0.844 |
| eGFR** day-7, mean ± SD [ml/min] | 73.06 ±21.79 | 74.15 ±17.80 | 71.59 ±26.74 | 0.686 |
Data are expressed as mean and standard deviation (SD), median and interquartile range (IQR), or number (n) and percentage (%) of patients depending on the data distribution.
** eGFR by MDRD = 175 × SCr (mg/dl)-1.154 × age -0.203 × 0.742 (if woman) (same ethnicity correction factors).
ACEI – angiotensin-converting-enzyme inhibitors, AF – atrial fibrillation, BMI – body mass index, CCB – calcium channel blockers, CCS – Canadian Cardiovascular Society Angina Grade, CKD – chronic kidney disease, COPD – chronic obstructive pulmonary disease, eGFR – estimated glomerular filtration rate, Hb – hemoglobin, HbA1c– glycated hemoglobin, ICU – intensive care unit, IQR – interquartile range, LVEF – left ventricular ejection fraction, MI – myocardial infarction, MNA – Mini Nutritional Assessment form, OPCAB – off-pump coronary artery bypass grafting, PAD – peripheral artery disease, PCI – percutaneous intervention, SCr – serum creatinine, TIA – transient ischemic attack.
Data collection
Upon admission, anthropometric measurements (age, gender, body weight and height), laboratory tests, LVEF and clinical status data were registered. Patients were stratified with MNA and EuroSCORE II.
Body composition assessment
Body composition assessment was conducted at two time points: 1 day before surgery (D0) and on the sixth postoperative day (POD6). Tanita Segmental Body Composition Analyzer BC-418 MA (Tanita, Tokyo, Japan) was used for all the tests. Tanita is a segmental BIA device that measures impedance of the body segments with eight electrodes to estimate FM, FFM and total body water (TBW) content. The device uses high-frequency electrical current (50 kHz, 90 μA) and measures the impedance of the trunk and four limbs, analyzing a total of five different body segments [13].
Skeletal muscle % estimation
The muscle mass percentage at a distinct time point (D0 and POD6) was calculated by subtracting the TBW% from the FFM%. FFM% constitutes the percentage sum of skeletal and visceral muscle mass, bone mass, and TBW% – in other words, everything but the fat mass %. Considering that the magnitude of bone mass and visceral muscle mass change within a seven-day period is negligible, the calculated difference (Δ) corresponds to skeletal muscle mass percentage only. The Δ in muscle loss or gain was computed by subtracting the measurements on D0 from those of POD6.
Clinical outcome evaluation
Hospitalization data, such as duration of mechanical ventilation, length of intensive care unit (ICU) and hospital stay, pre- and postoperative laboratory tests, chest drainage volume as well as the in-hospital major adverse cardiac and cerebrovascular event rate were registered for every participant. Postoperative clinical characteristics are presented in Table I.
Statistical analysis
All categorical variables were presented as a number and percentage. Depending on the normality of distribution as assessed by the Shapiro-Wilk test, continuous variables were shown as mean ± standard deviation (SD) or median and interquartile range (IQR). The χ2 test was used for dichotomous variables and Student’s t-test or the Mann-Whitney test for unpaired samples.
Regression analysis was used to evaluate the relationship between baseline BIA and hospital LOS parameters. In order to identify prognostic indicators of prolonged hospital stay, univariate analyses were first performed using clinical and nutritional status indices as independent variables and hospital LOS as the response variable. All factors exhibiting univariate significance at a p-value < 0.05 were employed in a multivariate logistic regression. Analysis was adjusted for EuroSCORE II. The odds ratios (OR) and 95% confidence intervals (95% CI) of the entered factors were calculated accordingly. The discriminatory capacity of the model was tested using the area under the receiver operating characteristic curve (ROC). All tests were two-sided with the significance level of p < 0.05. Calculations were performed using SPSS version 29.0.1.0 (IBM Corporation, Armonk, USA). Box plots and ROC curves were illustrated by Adobe Illustrator (version 27.5, USA).
Results
Characteristics of the groups
The final study group consisted of 50 patients with a median age of 69 (IQR, 61–72.25) years; 10 (20%) patients were female. Twenty-one (42%) patients were obese (BMI > 30 kg/m2), 19 (38%) patients were overweight (BMI ≥ 25 kg/m2 and < 30 kg/m2), and 10 (20%) had a normal BMI (18.5–25 kg/m2). The mean BMI of the study population was 29.36 ±3.70 kg/m2. According to the MNA score, none of the patients was malnourished (< 17 points). All the patients were considered low risk, as depicted by EuroSCORE II % operative risk of the entire cohort: 1.21 (0.80–1.85). Among the 50 patients, 29 (58%) of them stayed at the hospital shorter than 9 days, and 21 (42%) of them stayed 9 days or longer.
No differences were found between the groups in terms of anthropometric (BMI), standard assessment of nutritional status (MNA), and baseline clinical characteristics (Table I).
Apart from the length of ICU stay and mechanical ventilation length, which are derivatives of the original division related to the length of hospitalization, and the level of CK-MB 18 h postoperatively (which was borderline significant, p = 0.046), other postoperative characteristics did not significantly differ between the groups (Table I). These findings confirm the clinical preoperative and postoperative uniformity of study groups assessed using standard clinical and anthropometric parameters.
Body composition analysis
After grouping patients according to the length of hospitalization, preoperative and postoperative BMI did not differ significantly between the groups. Baseline body composition parameters differed significantly between patients with hospital LOS < 9 days and ≥ 9 days in terms of percentage composition of FM, FFM, TBW and skeletal muscle. In particular, FFM%, TBW% and skeletal muscle (%) were lower upon admission in the group of patients with a prolonged hospital stay (p = 0.012, p = 0.011, p = 0.012, respectively). In contrast to the preoperative values, the BIA showed no significant differences in body composition measurements between the compared groups after the OPCAB procedure (Figure 1, Table II).
Table II
Body composition analysis characteristics of patients by the length of hospital stay
Figure 1
Baseline values of box-plot alterations and receiver operating characteristic (ROC) curves of body composition estimates according to hospital length of stay. Baseline values of box plots and receiver operating characteristic (ROC) curves: A – FM (%) baseline values of box plots for hospital length of stay comparison; B – FM (%) baseline ROC curve for prediction of increased length of stay in the hospital; C – FFM (%) baseline values of box plots for hospital length of stay comparison; D – FFM (%) baseline ROC curve for prediction of increased length of stay in the hospital; E – FMI baseline values of box plots for hospital length of stay comparison; F – FMI baseline ROC curve for prediction of increased length of stay in the hospital; G – Combined panel of parameters (FM (%), FFM (%) and FMI) baseline ROC curve for prediction of increased length of stay in the hospital
AUC – area under curve, CI – confidence interval, FM – fat mass, FMI – fat mass index, FFM – fat-free mass, LOS – length of stay, ROC – receiver operating characteristic.
The magnitude of percentage change (Δ) in FM%, FFM%, TBW% and skeletal mass (%) was determined by subtracting the preoperative values measured on D0 from the postoperative parameters on POD6, for both < 9 days and ≥ 9 days of hospitalization groups, separately, in an effort to assess the body composition changes before and after elective OPCAB. There were no significant differences in the magnitude of the change (Δ) of the compared parameters of body composition between the study groups (Table II).
Patients with longer hospitalization had significantly higher FM and FMI, and significantly lower FFM baseline parameters (p = 0.011, p = 0.04 and p = 0.012, respectively) (Figures 1 A, C, E). ROC curve analysis was performed to evaluate the potential of baseline body composition parameters for prediction of hospitalization length. Area under the curve (AUC) for FM% was 0.751, for FFM% was 0.749 and for FMI was 0.714 (Figures 1 B, D, F). The ROC curve showed that pooling the 3 parameters together (FM%, FFM% and FMI) yielded a higher AUC than the value of each individual biomarker, as AUC was 0.767 (95% CI: 0.62–0.91) (Figure 1 G).
Based on the best FM%, FFM% and FMI accuracy cut-off determined by ROC curve analysis, the study population was divided into two subgroups: low FM% values (mean ± standard deviation (SD): 22.62 ±0.93) including 23 (51%) patients and high FM% values (34.17 ±0.99) including 22 (49%) patients; low FFM% values (mean ± standard deviation (SD): 66.05 ±0.96) including 23 (51%) patients and high FFM% values (77.66 ±0.94) including 22 (49%) patients; low FMI values (mean ± standard deviation (SD): 6.09 ±0.37) including 22 (49%) patients and high FMI values (10.76 ±0.39) including 23 (51%) patients (Table III, Figure 1).
Table III
ROC curve results of discrimination for preoperative BIA parameters in terms of prolonged hospital LOS
High FM%, low FFM% and high FMI values were significantly associated with a crude 14-fold, 12-fold and 7-fold (respectively) increased hospital LOS in patients after the OPCAB procedure in univariate logistic regression analysis. The panel of these three variables as one combined variable showed a stronger significant association with increased hospital LOS as the crude impact was 15-fold (Table IV).
Table IV
Univariate and multivariate logistic regression analysis of prolonged hospital LOS predictors
For subsequent adjustments, we used a multivariate logistic regression model to identify independent variables for increased hospitalization. All three studied baseline variables (FM%, FFM% and FMI) showed independent prediction potential for increased hospitalization in the adjusted by the EuroSCORE II model. High FM% values had 15-fold, low FFM% values had 13-fold and high FMI values had 7-fold higher risk of experiencing longer stay in the hospital (p = 0.001, p = 0.001 and p = 0.005, respectively). Importantly, the combined panel of three variables (higher FM%, lower FFM% and higher FMI) had 16-fold higher risk of longer hospitalization (adjusted OR = 16.40, 95% CI: 3.52–76.34; p = 0.0004) (Table IV).
Discussion
Currently, multiple clinical indices and risk stratification models are used to predict the cardiac surgery outcome, but no highly specific tool is available [18, 19]. Considering that CABG promotes profound body composition changes and highly attenuates metabolically dependent FM and FFM, perioperative assessment of relevant parameters could compliment the prognostic accuracy of the established clinical risk algorithms [14, 20].
Prolonged hospital LOS after cardiac surgery has been proven to be associated with unfavorable outcomes [15, 21]. Perioperative management of coronary artery bypass patients aims to minimize hospital LOS as a means of preventing hospital associated complications and costs optimization [22]. New markers of prolonged LOS are needed to allow for better screening and preprocedural optimization of the low-risk CABG patients.
Fat mass
The most frequently observed consequence of CABG surgery is considerable reduction of adipose tissue mass and fat free mass during the recovery period. The demand for oxygen and energy increases in the postoperative period and this leads to upregulation in fat metabolism [23]. Hypermetabolism is caused by endocrine, metabolic and immunological factors (mainly insulin, glucagon, cortisol, catecholamines and cytokines TNF-α and IL-1). In patients undergoing surgery, lipolysis begins to dominate, and fatty acids become the primary source of energy, which leads to development of insulin resistance and a reduction of the typical anabolic effect of insulin [23].
The vast majority of patients in our study experienced a reduction in adipose tissue mass after OPCAB. These findings contrast with those of van Venrooij et al., where no changes in fat mass were reported after the CABG procedure [14]. There is a paucity of data on how CABG surgery impacts adipose mass; however, the relationship between changes in fat mass and clinical outcomes becomes more apparent. In our study, patients with hospitalization time ≥ 9 days had significantly higher FM content than patients hospitalized for a shorter period. Due to the lack of data on caloric and nutrient consumption in the study group determination of the exact cause for post-operative fat mass decline in our cohort is problematic. In future studies more consideration of this occurrence is warranted.
Fat-free mass
A postoperative FFM% increase stands against the definition of malnutrition, but rather should be attributed to the body water constituent – a significant constitutent of FFM (FFM = skeletal muscle mass + body cell mass + total body water + bone mineral mass) [24]. Considering that the parallel changes of TBW% and FFM% reflected mild overhydration (2.11 ±2.60% vs. 1.02 ±1.28%, respectively), a relatively small gain of FFM can be assumed. The obtained perioperative fat-free mass index (FFMI) value ranges stand in the context of previously reported findings [15, 16].
The loss of FFM during convalescence has been related to a fast drop in muscle protein production rates in the first 4 h after the CABG procedure [25]. It is still undetermined whether postoperative FFM restoration after CABG is linked to the fluid balance or physical activity. It could be argued that in our study group the dominant lack of FFM reduction was due to the patients’ participation in the early postoperative rehabilitation program and their resultant engagement in regular physical exercise, allowing for some regain of FFM [26]. In consequence, it is obligatory to describe the range of change of FFM during recovery from CABG surgery, which is of particular importance in the case of persistent FFM wasting after the operation [27]. Further research should examine whether such optimized early rehabilitation programs combined with the implementation of nutritional care could prevent or reverse FFM deterioration after CABG surgery [17, 28].
Hospital LOS according to BIA assessment
There is no recommended cut-off value of FM or FFM for the prediction of postoperative outcome in normal- or increased-BMI patients referred for CABG. Despite the absence of differences in the clinical course between the study groups, we were able to determine preoperative cut-off values among body composition parameters related to the length of hospitalization. FM > 29.2%, FMI > 8.37 kg/m2, FFM < 70.8% and the combined parameter of FM%, FFM% and FMI > 0.429 were independent predictors of hospital LOS ≥ 9 days. It should be assumed that cut-off values may vary depending on demographic and anthropometric conditions, especially in the presence of chronic disorders. Currently, age, gender and BMI are the major factors decisively affecting the body composition in healthy individuals [29]. EuroSCORE II enables a comprehensive assessment of perioperative risk in cardiac surgery and is causally related to metabolic and musculoskeletal changes due to surgical procedures [30]. In our study, the body composition parameter panel was adjusted for EuroSCORE II. In addition, age, gender, BMI, ejection fraction, and EuroSCORE II were not significantly associated with prolonged hospitalization in univariate analysis, and only high FM and low FFM values and their derivatives were independent predictors of prolonged hospitalization in multivariate analysis in our study.
Despite the mean values of FM%, FFM%, TBW% and skeletal muscle (%) that were significantly different between the study groups before OPCAB, the magnitude of change of these measurements did not significantly differ between the groups. The body composition parameter patterns registered before the surgery were significantly reflected in the postoperative results. The clinical outcome is a result of many prognostic factors, the awareness of which, in both the preoperative and postoperative period, determines the occurrence of complications and the final outcome. Body composition monitoring is one of the commonly available technologies that enables individualization and optimization of the therapeutic process in cardiac surgery. BIA can be used not only in perioperative management but also in risk stratification.
Limitations
As this was a pilot study, the main limitation is the small number of patients. We aimed to detect body composition changes based on hospital LOS. The hospital LOS is affected by numerous factors apart from nourishment status. Patients’ nutritional assessment was only evaluated by applying the MNA questionnaire and subjects were predominantly male. BIA was not compared with other techniques for evaluation of fluid content and the limitations of this study comprise the specified restrictions of BIA implementation. Touch-type electrodes were utilized in this study, despite adhesive electrodes being recommended by ESPEN (European Society for Clinical Nutrition and Metabolism). Many factors affect body composition parameters in the postoperative period and their independent influence requires a larger patient population and a multi-frequency current body composition analyzer to allow for intra- and extracellular fluids. These results have been proven relevant only in a uniform group of OPCAB patients and should be further refined in more diverse cohorts. The single-center analysis restricted utility of these results in broader practice, but it indicates the direction for further research.
Conclusions
Preoperative high FM, FMI and low FFM content are independent predictors of prolonged hospital length of stay in normal- and increased-BMI patients after OPCAB. Moreover, combining these three parameters with the ROC curve into one panel can increase their prediction power, and help clinicians in early stratification of patients and prediction of the hospitalization length for the CABG operation. We recommend using BIA to optimize perioperative management of elective cardiac surgery patients. Further research on larger, more diverse cohorts, focused on preoperative body composition assessment, could better identify risk factors associated with the in-hospital outcome.
