Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: A pilot study.

Vasoplegia observed post cardiopulmonary bypass (CPB) is associated with substantial morbidity, multiple organ failure and mortality. Circulating counts of hematopoietic stem cells (HSCs) and endothelial progenitor cells (EPC) are potential markers of neo-vascularization and vascular repair. However, the significance of changes in the circulating levels of these progenitors in perioperative CPB, and their association with post-CPB vasoplegia, are currently unexplored. We enumerated HSC and EPC counts, via flow cytometry, at different time-points during CPB in 19 individuals who underwent elective cardiac surgery. These 19 individuals were categorized into two groups based on severity of post-operative vasoplegia, a clinically insignificant vasoplegic Group 1 (G1) and a clinically significant vasoplegic Group 2 (G2). Differential changes in progenitor cell counts during different stages of surgery were compared across these two groups. Machine-learning classifiers (logistic regression and gradient boosting) were employed to determine if differential changes in progenitor counts could aid the classification of individuals into these groups. Enumerating progenitor cells revealed an early and significant increase in the circulating counts of CD34+ and CD34+CD133+ hematopoietic stem cells (HSC) in G1 individuals, while these counts were attenuated in G2 individuals. Additionally, EPCs (CD34+VEGFR2+) were lower in G2 individuals compared to G1. Gradient boosting outperformed logistic regression in assessing the vasoplegia grouping based on the fold change in circulating CD 34+ levels. Our findings indicate that a lack of early response of CD34+ cells and CD34+CD133+ HSCs might serve as an early marker for development of clinically significant vasoplegia after CPB.

1. Introduction

Vasoplegia is a well recognized post-surgical complication of cardiopulmonary bypass (CPB) [1,2]. Endothelial activation and dysfunction are considered to be pivotal in the pathophysiology of vasoplegia. Kortekaas et al reported that pre-existing endothelial cell activation, determined by increased levels of von Willebrand factor (vWF) and soluble P-selectin at baseline, predisposes patients to vasoplegia [3]. Brettner et al have suggested that deviations in endothelial injury markers in a CPB group, as compared to an off-pump surgery group, identifies a correlation between ischemia/reperfusion and the extent of endothelial activation [4]. Circulating endothelial cells have also been reported to be elevated at baseline in cardiac surgery patients compared to healthy individuals, with their numbers further increasing post CPB [5]. Additionally, impairment in endothelial barrier function is observed during the course of cardiac surgery and post-surgery, when assessed through the prism of the angiopoetin-Tie2 system [6-8].

Though multiple studies suggest that a compromised endothelium along with continued vascular damage during CPB contributes to vasoplegic syndrome, the status of vascular repair and vascular progenitors in these settings is ill-characterized. Circulating haematopoietic stem cells (HSCs: CD34+CD133+) and endothelial progenitor cells (EPCs: CD34+VEGFR2+) are known to home in for repair, to the site of damaged endothelium and the vessel wall [9]. Reduced numbers and an altered function of EPCs are associated with ischemic cardiovascular diseases [10,11]. However, it is not clear how their circulating counts vary during the unfolding of the vasoplegic syndrome in CPB.

We sought to determine the changes in the circulating counts of these progenitors and their relationship to the severity of vasoplegia in cardiac patients undergoing CPB. We also trained machine learning methods (logistic regression and gradient boosting) to assess the accuracy of the vasoplegic categorization of patients based on these markers.

2. Materials and methods

2.1 Materials

Vacutainers for blood collection K2-EDTA (Cat: 367863) and serum separation (Cat: 367812) were purchased from Becton Dickinson Biosciences. Fluorescent antibodies such as anti-human CD133-Phycoerythrin (PE) (Cat: 130-080-801) were purchased from Miltenyi Biotech Anti-human CD34-Fluorescein isothiocyanate (FITC) (Cat: 343604), anti-human VEGFR2-Allophycocyanin (APC) (Cat: 359916) and isotype control antibodies like mouse IgG2a-FITC (Cat: 400207) and mouse IgG1k-PE (Cat: 400113), were purchased from BioLegend. Other dry chemicals were purchased from Sigma-Aldrich.

2.2 Surgical procedure

Cardiac surgery was performed according to local standardized protocols. All surgical procedures were performed via a midline sternotomy under normothermic CPB. Blood cardioplegia was used for all the cardiac cases. The cardioplegic solution was a mixture of patient’s own blood and crystalloid solution. The crystalloid solution was a mixture of Plasmalyte (500 mL) and St. Thomas' solution (20 mL). The delivery was done by the use of a specialized roller pump (Sarns Medical Systems) at the ratio of 4:1 where 4 parts of blood and 1 part of the mixed cardioplegia solution were used. In lung transplant the lung preservative used was Perfadex solution (Medisan, Uppsala, Sweden) containing low-potassium (K+ 6 mmol/L) and extracellular electrolytes (Na+ 138 mmol/L; Cl- 142 mmol/L; Mg++ 0.8 mmol/L; Dextran 40 g); H2PO4- 0.8 mmol/L; Glucose (0.91 g); Osmolarity (292 mOsm/L).

2.3 Study design and sample collection

Following informed consent, subject recruitment and sample collection were done at the Cardiothoracic Surgical unit of Apollo Hospitals, Chennai. The study design was reviewed and approved by the Institute Ethics Committees of Apollo Hospital [AMH-001/01-18] as well as the Indian Institute of Technology Madras [IEC# IEC/2016/02/MDX-2/05]. A total of 19 elective cardiac surgery individuals (8 male and 11 female), age range 40–70 years (adults), and undergoing CPB, were recruited. Written consent as prescribed by the institutional ethics committees was obtained from patients a day prior to their operation. Individuals with emergency trauma surgery or any infectious diseases were excluded.

The blood samples were collected in K2-EDTA vacutainers for subsequent flow cytometry from a central venous catheter placed in the jugular vein when the patient was not on the CPB pump. When the patient was on pump, blood was collected from the inlet of the pump at specified time intervals as described below. The whole blood collected was processed within 2 hours of collection for flow cytometry. The time points for blood collection were: (1) ‘Post-Induction’ (time after the induction of anesthesia prior to the start of surgery); (2) ‘Sternotomy’ (beginning of surgery after median sternotomy); (3) ‘0 hour’ (start of the CPB pump); (4) ‘1 hour’ (1 hour on pump); (5) ‘6 hour’(end of surgery when the patient was shifted to ICU) (6) ‘24 hour’ (time point when the patient is in ICU for recovery). These time points and study design are summarized in Fig 1. To minimize inter patient bias which may contribute to variability, we considered each patient’s baseline i.e. ‘Post-Induction’ as the control for that patient.

Fig 1

Study design showing time points of sample collection.

2.4 Grouping of subjects based on severity of vasoplegia

In the absence of a standard consensus for the definition of clinical vasoplegia [12,13], we choose to categorize our recruited subjects based on the total dose of norepinephrine (NE) infused in these subjects. In most cardiothoracic units, low levels of NE (usually total dose <0.02mg/kg body weight), are given in all patients post-surgery. This dosage is increased only if a patient further develops vasoplegia. We thus reasoned that increased use of pressor beyond 0.02mg/Kg body weight would identify clinical worsening of vasoplegia. Hence, we choose a cutoff dose of total NE above 0.02mg/kg body weight to group the subjects into two groups. Among the recruited 19 subjects, 11 subjects received NE <0.02mg/Kg body weight and were classified as a clinically insignificant vasoplegia group or group 1 (G1). The remaining 8 subjects who received NE >0.02mg/Kg body weight were termed as a clinically significant vasoplegia group or group 2 (G2).

2.5 Enumeration of circulating progenitor cells via flow cytometry

For enumeration of circulating vascular progenitors in whole blood, cells were fixed and erythrocytes lysed using fluorescence-activated cell sorter (FACS) lysing solution. Blocking was performed with 5% Fetal Bovine Serum (FBS) and 0.2% Bovine Serum Albumin (BSA) for 30 min followed by staining with CD133-PE, CD34-FITC and VEGFR2-APC antibodies for 40 minutes at 4°C. Corresponding isotype immunoglobulin G (IgG) 1-PE and IgG2a-FITC and IgG2aκ-APC antibodies were used as controls. A minimum of 500,000 events were acquired and scored as per EUROSTAR guidelines using a FACS Aria flow cytometer (Becton Dickinson) [14]. Data were analyzed using the Flowjo software program (Version 7.6.1, Tree Star Inc.). Following appropriate gating, CD34+, CD133+, VEGFR2+, EPC (CD34+VEGFR2+), HSC (CD34+CD133+) and CD133+VEGFR2+ cells (S1 Fig) were enumerated from lympho-monocyte fraction and are represented as the number of cells per million lympho-monocytic events.

2.6 Statistical analysis of clinical data

Progenitor cell counts are expressed as mean ± SEM of the number of cells counted per 1 million lymphomonocyte events. Clinical data are also expressed either as mean ± SEM or Median. That the data were normally distributed was confirmed through a Kolmogorov-Smirnov test. Comparisons within and across groups were performed using paired and unpaired Student’s t-test, respectively. For non-parametric data, the Mann-Whitney test and Wilcoxon matched pair test were employed.

2.7 Mathematical model to assess the characterization of study subjects

We used two machine learning techniques, logistic regression and gradient boosting, as prediction tools. Logistic regression is a machine learning tool used for binary classification problems. In linear regression, a linear function of inputs is fitted to match output on training data. The output is a real number.

In logistic regression, the output value is transformed using a cost function known as a logistic or sigmoid function to return a value between zero and one, which can be interpreted as a probability value. This output can be thresholder to predict a value of 0 or 1. We used the implementation in scikit-learn [15].

Gradient boosting is a decision-tree-based algorithm for classification problems [16]. A decision tree is a tree of possible choices to be made given the input data. In a classification and regression tree (CART), unlike a traditional decision tree, the leaves indicate possible scores for outcomes, rather than binary choices. In gradient boosting, an ensemble of trees is used where each tree is trained to minimize errors of preceding trees. This has become a very popular machine-learning tool for tabular data in recent years, here we use the XGBoost package [17].

Based on input parameters, logistic regression and gradient boosting were used to predict whether a patient belongs to G1 or G2. Predictions were validated using “leave one out”—each patient was left out once, and the machine-learning model was trained on the remaining patients and tested on the left-out patient. The significance of these predictions was assessed using p-values, i.e. the probability of a similar or better fit happening by chance. To calculate these, we compared the results of 10,000 randomizations. Random data were generated by shuffling the vasoplegia categorization (that is, redistributing the category labels among the patients, ensuring that there are still the same number of patients as are there in G1 and G2. Here, the p-value is the fraction of random data-sets with a similar or lower mean absolute error (MAE; the average of the absolute value of the error over all individuals) to the real data. Similar results (not shown) were obtained with a more aggressive randomization, where for each patient, we retain the vasoplegia categorization but pick a random number between the minimum and maximum of the original graph as the log fold-change value.

3. Results

3.1 Characteristics of study subjects

A drop in the Mean Arterial Pressure (MAP) from pre-CPB to post-CPB despite administration of inotropes and presence of hyper-lactemia indicates the onset of vasoplegia [1,18]. Hence, we measured intra-arterial systolic blood pressure (SBP), diastolic blood pressure (DBP) and MAP in all 19 study subjects. We observed a significant decrease, not only in the MAP post-CPB at 24 hours but also in DBP at 24 hours compared to the pre-surgery state (Fig 2A). Additionally, levels of lactate in blood were significantly higher after 1 hour on the pump, peaking at 6 hours and remaining significantly high even at 24 hours compared to post-induction (Fig 2B). These parameters were similar for all 19 subjects in our study group, except for the difference in dosage of noradrenaline or norepinephrine (NE) infusion post-CPB.

Fig 2

Characterization of study subjects.

(A) Graph showing blood pressure changes between Pre-Operation and Post-Operation in all the study individuals (n = 19). (B) Lactate concentration from blood at different time points in all the individuals (n = 19) measured in mmol/L. Data shown as Mean ± SEM. ** p ≤ 0.01, ***p ≤ 0.001, **** p ≤ 0.0001 vs respective control. Mann-Whitney’s test and Wilcoxon matched pair test were employed.

Table 1 shows the baseline anthropometric and clinical characteristics and intake of medication such as Angiotensin II converting enzymes inhibitors (ACEI) and Angiotensin receptor blockers (ARBs) of the clinically classified groups G1 and G2. As shown in Table 1, there was no significant difference in age, gender ratio, BMI, heart and respiratory rate, DBP, SBP, and MAP as well as in the % intake of ACEIs, ARBs, statins, steroids except for beta-receptor blockers (β- blockers) between the two groups. The intake of β- blockers is almost two fold in G2 subjects. The percentage of existent comorbidities in recruited subjects such as diabetes, hypertension, dyslipidemia and renal failure, along with cardiac morbidities such as heart failure, right ventricular dysfunction (RV dysfunction) and pulmonary artery hypertension (PAH) are shown in Table 1. The percentage of RV dysfunction in G2 subjects is higher compared to G1 subjects. A significant increase in urea for G2 is in concordance with observed renal failure in 25% subjects of this group.

Table 1

Anthropometric and clinical characteristics of CPB individuals at baseline.

Parameters	Group 1 (n = 11)	Group 2 (n = 8)
Age (yrs)	57 ± 2	54 ± 2
Sex (M/F)	5/6	3/5
BMI	25.51± 1.02	26.46 ± 2.38
Heart Rate (per min)	83.45 ± 3.32	89 ± 6.27
Systolic BP (mmHg)	121.8 ± 2.26	115.0 ± 4.62
Diastolic BP (mmHg)	78.18 ± 2.26	72.5 ± 2.50
Mean Arterial Pressure (mmHg)	92.59 ± 2.53	86.67 ± 2.88
Respiratory Rate (No. of breaths/min)	20.18 ± 0.68	22.13 ± 1.23
LVEF	61.3±2.9	61.7±1.9
Urea (mg/dl)	24.80 ± 2.19	45.29 ± 6.7**
Creatinine (mg/dl)	0.74 ± 0.05	0.94 ± 0.12
Comorbidities % prevalence
Diabetes (%)	45.0	37.0
Hypertension (%)	45.0	25.0
Dyslipidemia (%)	9.0	12.0
Renal failure (%)	0	25.0
Heart failure (%)	27.27	37.5
Right Ventricular dysfunction (%)	9.09	37.5
PAH (%)	36.36	50.0
Medication intake (% per group)
ACEI (%)	27.27	35.1
ARB (%)	27.27	12.5
Statins(%)	27.3	25
Steroids (%)	18.2	37.5
Beta blockers (%)	36.4	75
Types of surgery	No. of patients
Mitral valve replacement/Repair	2	5
Aortic valve replacement/Repair	2	1
Double valve repair	1	1
Two procedure (CABG+ Valve replacement)	3	0
Double lung transplant	3	0
Double lung transplant+CABG	0	1
Pre operative risk score
Euro SCORE II (Median % estimated mortality)	1.25	3.05

LVEF—Left ventricular ejection fraction, Intake of Angiotensin II converting enzymes inhibitor (ACEI) and Angiotensin receptor blocker (ARB) are expressed in percentage, CABG- Coronary artery bypass graft; Euro SCORE II- European System for Cardiac Operative Risk Evaluation II; Data expressed in Mean± S.E.M unless otherwise specified **p<0.0.

The post-operative clinical outcomes for the two groups are shown in Table 2. The cardiac function measured as left ventricular ejection fraction (LVEF) post CPB, showed normal values of more than 55% for both the groups (Table 2). However, MAP was maintained above 70mmHg for both groups with infusion of vasopressors (Table 2). Intraoperative parameters such as fluid balance have an effect on the post CPB outcomes like vasoplegia, organ dysfunction and mortality [1,19]. Hence, we determined the intraoperative fluid balance change (FBC) during CPB in terms of the difference in volume of fluid intake and output (I & O) for each subject [20] (calculated from the perfusionist chart). The fluid balance change is expressed as positive, negative or zero fluid balance and the percentage of subjects under each fluid balance state in a group is shown in Table 2. The percentage of positive fluid balance subjects is higher in group 2 (75%) compared to group 1 (36.4%). (A positive fluid balance, also known as fluid overload, is correlated to higher mortality and organ dysfunction [21,22]). Moreover, neither group had a trigger Hematocrit (Hct) for transfusion, (Median %Hct Group 1 = 34; Group 2 = 28). The higher incidence of positive fluid balance in G2 reflects the higher incidence of vasoplegia in that group, although it is difficult to establish causality.

Table 2

Clinical and Biochemical characteristics of individuals at intra and post-CPB.

Post-CPB (at ICU) measurements
Clinical parameters		Group 1 (n = 11)	Group 2 (n = 08)
CPB duration (min) (Median)		225.2 ± 41.04	206.1 ± 45.09
Cross Clamping time (min) (Median)		198.8 ± 45.64	161 ± 45.73
Complete flow rate (lit/min) (Median)		3.90 ± 0.16	3.97 ± 0.17
Average flow rate(lit/min) (Median)		2.89 ± 0.12	2.77 ± 0.14
Heart Rate (per min) at 24 hrs		77.82 ± 4.14	95.00 ± 5.57 *
Systolic BP (mmHg) at 24 hrs		124.1 ± 4.50	102.3 ± 4.74 **
Diastolic BP (mmHg) at 24 hrs		59.55 ± 2.00	55.63 ± 1.36
Mean Arterial Pressure (mmHg) at 24 hrs		77.64 ± 1.82	71.63 ± 1.96 *
Respiratory rate (No. of breaths/min)		20.36 ± 0.88	22.00 ± 1.30
LVEF (%) (Median)		59.63 ± 1.78	57.86 ± 2.14
Fluid balance change (At end of CPB)
Positive (%)		54.55	75.0
Negative (%)		36.36	25.0
Zero (%)		9.09	0
Duration of ventilation (hrs)	CMV	12.30 ± 1.10	18.13 ± 3.07
	CPAP	2.37 ± 0.46	22.25 ± 16.59 **
ICU Stay (days)		4.50 ± 0.61	10.63 ± 4.46
Hospital stay (days)		9.81 ± 0.90	23.25 ± 5.9*
Mortality (number and %)		0 & 0%	2 & 10.5%
Biochemical measurements
Urea (mg/dl)		33.27 ± 5.34	36.86 ± 3.77
Creatinine (mg/dl)		0.73 ± 0.08	0.96 ± 0.09*
EGFR (ml/min/1.73m2)		116.9 ± 7.41	93.32 ± 12.74
Hemoglobin (%)		10.03 ± 0.40	10.33 ± 0.40
ESR (mm/hr)		53.60 ± 7.58	40.00 ± 9.33
WBC (103/mm3)		14.02 ± 2.06	20.62 ± 2.82 *
Lymphocytes (%)		10.55 ± 2.08	5.12 ± 0.85 *
Monocytes (%)		7.09 ± 1.03	7.12 ± 1.00
Neutrophils (%)		80.18 ± 3.46	86.88 ± 1.69
Platelet count (103/mm3)		187.2 ± 11.23	192.6 ± 39.17
Post-operative risk score (ICU score)
SOFA Score (Median score)/estimated % Mortality		7/18.2	8/26.3
APACHE II Score (Median score)/estimated % Mortality		9/7.6	13/7.75

EGFR—Estimated Glomerular Filteration Rate; ESR—Erythrocyte Sedimentation Rate; WBC—White Blood Cells; CMV—Continuous mandatory ventilation; CPAP—Continuous positive airway pressure; ACEI—Angiotensin Converting Enzymes inhibitor; ARB—Angiotensin II receptor blocker;LVEF—Left ventricular ejection fraction; SOFA Score-Sequential organ Failure Assessment II;APACHE II -Acute physiology and Chronic Health Evaluation II.Data expressed in Mean ± S.E.M unless otherwise specified. LVEF, Hemoglobin and Death Rate are expressed in %.

*p<0.05 and

**p<0.01.

There was a significant difference in post CPB characteristics such as the duration of ventilation (p<0.01) and the duration of ICU (p = 0.056) and hospital stay (p<0.05) which were all higher in case of G2 subjects, confirming the severity of clinical vasoplegia in G2 compared to G1 subjects. Among the measured biochemical parameters there was a significant increase in WBC count and a significant decrease in lymphocyte count for G2 compared to G1 subjects (Table 2).

The predictive risk assessment score (EuroSCORE II, SOFA II score and APACHE II) is provided in Tables 1 and 2. The EuroSCORE II (European System for Cardiac Operative Risk Evaluation II) score, used as a pre-operative index of patients’ risk profile [23,24] in terms of estimated percentage mortality (% mortality), is shown in Table 1. This score takes into account patient related factors (anthropometric and comorbidities), cardiac related factors (pre-operative LV function, recent myocardial infarction and pulmonary hypertension) and operation related factors (urgency and weight of intervention). The SOFA II (Sequential organ Failure Assessment II) score is used to determine level of organ dysfunction and mortality risk in ICU patients (Table 2). APACHE II (Acute Physiology and Chronic Health Evaluation II) score estimates ICU mortality (Table 2) [25-27]. The categories of post-operative ICU parameters taken into account during calculation of SOFA II and APACHE II are vitals (temperature, mean arterial pressure, heart rate, respiratory rate) and blood chemistry (urea, creatinine, bilirubin and electrolytes); oxygenation parameters (PaO2, FiO2 and mechanical ventilation); and hematology (Hematocrit, WBC). The value signifying the most extreme deviation from the normal for each physiological variable within a 24 hour period post operation is used for scoring.

As shown in Table 1, the pre-operative risk score, EuroSCORE II predicts twice the estimated % mortality for group 2 (3.05%) compared to group 1 (1.25%), indicating the presence of higher risk for group 2 subjects in developing clinical complications upon cardiac surgery. The post-operative ICU score SOFA II indicate higher % mortality (26.3%) in group 2 compared to group 1 (18.2%) while the APACHE II score showed similar score values and estimated % mortality among the groups (Table 2). This establishes the need to carry out studies which determine biomarkers, especially vascular markers which can provide independent predictions, and may also complement risk scores to determine the severity of post-operative clinical outcomes.

3.2 Circulating levels of progenitor cells

Changes in the circulating levels of progenitors were enumerated in the individuals using flow cytometry. As can be seen in Fig 3A, for the combined 19 on-pump cases, we observed an early increase in CD34+ cells followed by CD133+ and VEGFR2+ cells. The CD34+ cells count peaked at 1 hour and remained elevated at 6 hours (a 25-fold increase compared to post induction), but gradually decreased by 24 hours (Fig 3A). There was a 15-fold increase in CD133+ cell counts which peaked at 6 hours compared to post induction, later dipping at 24 hours (Fig 3A). In contrast, VEGFR2+ cells had a late response to CPB, which increased gradually post-CPB from 6 hours onwards and remained high at 24 hours (statistically non-significant) (Fig 3A). With regard to double positive cells, EPCs rose from 1 hour onwards and remained elevated at 24 hours (Fig 3B) whereas HSC peaked at 6 hours but dipped by 24 hours (Fig 3B). CD133+VEGFR2+ cells did not show any observable change in numbers upon CPB in the study subjects.

Fig 3

Fold changes in circulating counts of measured progenitor cells for all 15 on pump cases (4 were considered outlier by statistical package for the social sciences).

(A) Line graph showing fold change in single positive cell counts for CD34+, CD133+& VEGFR2+ cells and (B) Line graph summarizing changes in double positive cell counts for HSCs (CD34+CD133+) and EPCs (CD34+VEGFR2+). Data is plotted as mean ± SEM. *p ≤ 0.05, ** p ≤ 0.01 vs post induction, $ p ≤ 0.05, $ $ p ≤ 0.01 vs sternotomy and # vs 6 hours. Mann-Whitney’s test and Wilcoxon matched pair test were employed accordingly.

3.3 Circulating levels of progenitor cells and levels of lactate in G1 versus G2 subjects

We then sought to determine how changes in circulating progenitors differed between G1 and G2. As can be seen from Fig 4A and 4B, the G1 individuals had an early increase in the number of CD34+ from 0 hour onwards, whereas this increase was delayed in G2 individuals. In G2 individuals, an increase in the cell number with regard to CD34+ and HSCs was observed only at 6 hours. EPCs started rising from 1 hour onwards in G1 individuals and remained elevated till 24 hours. This response was blunted in G2 (Fig 4C). The other cell types, i.e. VEGR2+ cells and CD133+, did not show any statistically significant trend for G1 and G2 subjects.

Fig 4

Circulating counts of progenitor cells in individuals undergoing CPB at various intra-operative time points in G1 and G2 study individuals.

Line graph showing fold change in cell counts for (A) CD34+ (B) HSC (CD34+CD133+) and (C) EPC (CD34+VEGFR2+) in vasoplegic groups namely G1 and G2. Data is plotted as mean ± SEM. *p ≤ 0.05, ** p ≤ 0.01 vs G2 at that point. Mann-Whitney’s test and Wilcoxon matched pair test were employed accordingly.

3.4 Assessing vasoplegia grouping for individual patients using machine learning

The log-fold change of CD34+ marker for each patient was plotted on the y-axis at different time points with respect to three baselines: induction (B1), sternotomy (B2) and 0 hour (P0) against the case number shown arranged in increasing order of amount of nor-epinephrine dosage given to patients on the x-axis (as in Fig 5A–5C respectively). We observed that, on using CD34+ as a marker, the 0-hour log fold change with respect to baselines B1 i.e induction (Fig 5A) and B2 i.e sternotomy (Fig 5B) and 24 hours fold change with respect to baseline B2 (Fig 5C) show the clearest visual trend in representing the difference between G1 and G2.

Fig 5

Bar graph showing log fold change of CD34+ marker for each patient, identified by case number, computed at different time points with respect to three baselines: Induction, sternotomy and 0 hr.

(A) Log fold change at on-pump 0hr with respect to induction. (B) Log fold change for on-pump 0hr time point with respect to sternotomy. (C) Shows 24hr log fold change with respect to sternotomy. Case numbers for patients in G1 (insignificant vasoplegia) are shown in green and those for patients in G2 (significant vasoplegia) are shown in red. The status of individual patients as diabetic (d), hypertensive (h) or having renal failure (r) is also indicated together with each bar. (D), (E), (F) show machine learning groupings for G1 and G2 using logistic regression and (G), (H), (I) show the same based on gradient boosting. Correctly classified patients are shown, by case number, in green boxes, and misclassified patients in red boxes. Corresponding p-values are shown.

Patients were then classified using logistic regression (Fig 5D–5F) and gradient boosting (Fig 5G–5I) algorithms on the above-mentioned log fold change values for the CD34+ marker. Grouping on each patient were made by leaving that patient out and training on other patients. Gradient boosting outperformed logistic regression in general, with both methods mis-classifying only two patients in the best case (Fig 5D and 5G). The performance of the classifiers was significant (p<0.05) for Fig 5D, 5E, 5G and 5H (0 hour compared to baselines B1 and B2).

Power analysis

For each time point, we calculated the sample means and standard deviations separately for the two groups (I = non-vasoplegic or insignificantly vasoplegic; II = vasoplegic), and the pooled standard deviation. We calculated the effect size given as Cohen's d = (difference in means)/(pooled standard deviation). We calculated the statistical power for the observed effect size, as well as for effect sizes of 1.0, 1.5 and 2.0, at a desired significance of 0.05 for 8 non-vasoplegic and 7 vasoplegic patients, by simulating 100 data sets per effect size. For the 0-hour vs induction and 0 hour vs sternotomy time points, we find a large effect size and a moderately high statistical power, as shown in Table 3 below.

Table 3

Power analysis.

	0 hour vs induction	0 hour vs sternotomy	24 hour vs sternotomy
Mean I (Non-Vasoplegic)	1.23	0.68	0.37
Mean II (Vasoplegic)	-0.51	-0.50	-0.09
Standard deviation I	1.08	0.53	0.42
Standard deviation II	1.05	0.96	0.43
Pooled Std Deviation	1.06	0.76	0.42
Cohen's d effect size	1.63	1.56	1.10
Power (actual effect size)	0.62	0.55	0.20
Power (effect size 1)	0.20	0.20	0.20
Power (effect size 1.5)	0.55	0.55	0.55
Power (effect size 2)	0.77	0.77	0.77

Propensity matching

Given the small dataset, a thorough propensity matching on all parameters is unfeasible. Instead, a Euroscore II value was calculated for each patient based on their clinical parameters, and these were matched. Out of eight non-vasoplegic and seven vasoplegic patients, we found a subset of five good matches per group, as given in the supplementary table (S1 Table) and supplementary figure (S2 Fig). Using these matched subsets of 5 vs 5 patients led to comparably good predictions as with the full set. Since the Euroscore II is comparable between the two groups, there is no significance difference in the risk profiles between the two groups.

4. Discussion

In this study, we report differences in time dependent changes in levels of circulating progenitors and EPCs in CPB surgery in individuals exhibiting clinically significant and insignificant vasoplegia. We have demonstrated (as in Fig 3A and 3B) that CD34+ cells are the first one to respond to CPB, followed by CD133+, HSCs, EPCs and VEGFR2+ cells. This part of the study supports the conclusions of previous studies investigating the mobilization of progenitor cells in circulation during and after CPB. Our findings on CD34+ and HSCs are comparable with previous studies done by Dotsenko et al and Scheubel et al [28,29]. They reported an increase in CD34+ cell counts at the end of surgery which contained to remain elevated 24 hours post-operation. Similarly, the rise in circulating HSCs peaks at 6 hours post-operatively and decreases after 24 hours. Increased EPCs and VEGFR2+ levels at the end of the surgery have also been reported in previous studies [30-32]. Notably, the kinetics of progenitor cells and lactate levels follow a trend similar to that seen in Figs 2B, 3A and 3B, which suggests hypoxia driven mobilization of progenitor cells [33].

To the best of our knowledge this study represents the first attempt to associate changes in numbers of progenitors with the severity of vasoplegia. We have shown that individuals who do not develop clinically significant vasoplegia after surgery (G1) showed significant early increase in the number of CD34+ and HSCs. This rise was not observed in the individuals who developed significant vasoplegia (G2). Similarly, a blunted response of EPCs was observed in G2 individuals in comparison to G1.

Upon categorization based on the severity of vasoplegia, the early response of CD34+ and HSC counts in G1, as seen in Fig 4A and 4B, might indicate the body's attempt to limit vascular damage by inducing repair mechanism against the insult caused due to CPB. This early response is compromised in G2 individuals. Prior studies have claimed that CD34+ cells promote therapeutic angiogenesis through paracrine signaling in response to myocardial ischemia and have the potential to ensure engraftment of transplanted cells [34]. Recently, Cogle et al have shown that the increase in the number of CD34+ cells in bone marrow is directly associated with improved functional outcomes after acute myocardial infarction (AMI) in the Timing in Myocardial Infarction Evaluation (TIME) and Late TIME clinical trials [35]. In another study, Fadini et al have found that CD34+ cells have a strong negative correlation with cardiovascular risk [36]. Similarly, HSCs are known to promote neovascularization on tissue engineered construct and have shown significant correlation with autologous white cell count and engraftment kinetics in myeloma patients [37,38]. Based on these studies and our observations, we speculate that endothelial injury caused by CPB, if not accompanied by an early mobilization of sufficient circulating progenitor cells, may result in impaired vascular repair, thereby contributing to the severity of vasoplegia. However, the reason for delayed response of CD34+ and HSCs cells in G2 individuals is still unknown. Given the variability of vasoplegia even among patients within different risk groups, it would be reasonable to assume that pre-operative profiles may influence the ability to mobilize circulating progenitors and HSCs.

We also observed a depression in EPC count in G2 patients as seen in Fig 4C. The reason for decreased circulating counts of EPCs in G2 could be due to compromised migration, homing and reparative potential of these cells as shown by Ruel et al. They demonstrated that the migratory effect of EPC was compromised in on-pump CPB patients [32]. Further, Lei Du et al have reported that EPCs start homing to the injured site after 4 hours from the end of the surgery, reporting a drop in the counts of circulating EPCs after 4 hours accompanied by a corresponding increase in their numbers in lung and kidney tissues [39].

To support our findings, we employed machine learning techniques to evaluate our categorization of groups G1 and G2 with respect to the CD34+ marker. We found that the log fold change values of CD34+ markers (as seen in Fig 5A–5C) reflected our findings in terms of an increasing trend in G1 and decreasing trend in G2. Among the algorithms we utilized, namely logistic regression and gradient boosting, we found that gradient boosting performed better in classifying patients into G1 and G2 as seen in Fig 5G–5I. We found that in most patients, the clinical classification agreed with the grouping from gradient boosting. However certain patients, namely patients 6, 8 and 9 in our study, were frequently misclassified. Of these, patient 9 had Pre-CPB renal failure, while patients 6 and 8 had borderline levels of vasoplegia as indicated by norepinephrine dosage. While these results are promising, a study on a larger cohort would enable us to refine our predictive methods for better understanding.

In summary, this pilot study shows that a distinctive pattern of EPC response characterizes patients likely to develop clinically significant vasoplegia. Further studies will help clarify the profile and the use of the EPC response as a biomarker to dictate the threshold to start pressors (or low dose steroids).

Study limitations

There are certain limitations of the current study. For instance, due to small sample size, certain observations did not reach statistical significance. The other major limitation of the study is the non-availability of the blood sample beyond 24 hours. This limited us to observations of the kinetics of progenitor cells only up to 24 hours. Additionally, we did not measure the levels of secretory endothelial markers during perioperative time points to assess endothelial damage. Further studies with large cohorts tracking the changes both between and within risk stratified groups undergoing cardiopulmonary bypass will be needed to determine if a particular response profile could be used to predict the occurrence of significant vasoplegia.

Gating strategy applied for whole blood.

(TIF)

Click here for additional data file.

Propensity matched subgroups of 5 patients under each group (Group I and Group II).

(TIF)

Click here for additional data file.

Propensity matching.

(TIF)

Click here for additional data file.

27 Jul 2020

PONE-D-20-15232

Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: An observational study

PLOS ONE

Dear Dr. Thangaraj,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Specifically, the patient population should be expanded and a control group should be added. Furthermore, more detailed statistical analysis, such as a multivariate analysis and a propensity score matching, are needed to confirm the strenght of the results.

Please submit your revised manuscript by Sep 10 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Antonio Cannatà

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 include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[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: Partly

Reviewer #2: Partly

Reviewer #3: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: I Don't Know

Reviewer #3: 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

Reviewer #3: 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: Yes

Reviewer #2: Yes

Reviewer #3: 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: Dr. Thangaraj,

We would like to thank you and your colleagues for submitting your work to PLOS ONE. This type of translational research is much welcomed to this journal. Cardiothoracic surgeons constantly deal with the vasoplegic patient after cardiopulmonary bypass and therefore, attempts to finding biomarkers to help determine which patients may have a vasoplegic response can result in many rescued patient. This type of study is particularly difficult to pull off, due to multiple confounding factors regarding patients so I commend you on the attempt. However, there are multiple concerns I have regarding your study.

1. This study is severely limited by the sample size of patients. In order to make any meaningful generalizations, sample sizes of 11 and 8 are inadequate. Please do a power analysis if possible.

2. The patient population is not adequately described in multiple areas. There is a range for the ages of the patients which is 40-70. This is a very wide range and there are multiple studies showing that age affects inflammation. The study has medication intake per group, but statins are not included and statins are showed to be anti-inflammatory. Lopressor is a very popular medication which has also been shown to be anti-inflammatory. Also, preoperative comorbidities like COPD, chronic liver disease, steroid use, etc can have such a major impact on outcome. Did any of the patients have heart failure? What was the PREOPERATIVE LVEF of the patients? Was there pulmonary artery hypertension or RV dysfunction?

At this point, I have no clue what types of patients are being operated on.

3. Please use a score to determine risk assessment of the cardiac surgery patients. This will give us an idea of how sick the patients are that were included in the study. In the US, we use a the STS-PROM. Whatever score is suitable for you, please include. If could very well be the patients who had an increase pro-inflammatory response were patients who were sicker prior to surgery.

4. Operative details were left out. What type of operations did these patients undergo? I don't believe a 40 year old undergoing a CABG and a 70 year old undergoing CABG will react differently. A 70 year old is expected to have level of CAD, while a 40 year old is likely to not have CAD. I would assume the 40 year old may have increased proinflammatory markers prior to surgery, does this affect your study? What about a 40 yo undergoing a MVrepair for flailed mitral disease whose relatively healthy vs. a 40 year old with endocarditis? We see these patients all the time.

5. More on operative detail - What were the CPB times for the patients? What was the cross-clamp times? What cardioplegia did you use? What MAPs were maintained in the OR? What was the flow for the patients? All of these things that happen in the OR affects pro-inflammatory markers and lactic acidosis, etc. If you have a patient undergoing a cross-clamp time of 180 minutes vs. 80 minutes, I would assume that the longer cross clamp time would have more pro-inflammatory markers. We certainly see that these patients are usually sicker coming out of the operating room. You may want to look into the affects of CPB time or cross clamp time on the markers you discussed.

6. Post-operative course is not described passed 24 hours. First, the study has a 10% mortality rate, which is exceedingly high depending on the operation and how sick the patient is prior to surgery. Once again, this is a reason patient population needs to be more adequately described and a risk assessment score needs to be included in the paper.

7. Norepi is the drug that was used as a consensus for clinical vasoplegia. How was it determined to place this patient on norepi vs. epi. What about norepi vs. vasopressin? What about clinical fluid status? Maybe the patient is not vasoplegic, but dry and required fluids? Once again, patient population needs to be described.

Once again, I commend you on a very difficult study. The results of the study were definitely interesting. However, without the information stated above, I cannot agree that there is applicability to these findings. There are many pro- and anti-inflammatory aspects of the patients that are confounding your study. A multivariable analysis may need to be done. I am happy to re-review the study once more information and work is done. I encourage the authors to look into some of my suggestions as I believe it would make this work clinically more applicable.

Reviewer #2: The subject of the paper is scientifically relevant and the the work performed correctly, as to my expertise. Nevertheless, the population enrolled to the study is small, considered the following sub-categorization in two groups. The differences observed by the authors wil certainly be reinforced by inclusion of 10-20 more patients and scientific soundness of the paper would benefit from including other groups of patients - off-pump cardiac surgery as a "control group" and ECMO patients.

Reviewer #3: I congratulate the author for the great job.

The authors analysed vasoplegia observed post-cardiopulmonary bypass and the correlation with CD34+ cells and CD34+CD133+ hematopoietic stem cells.

This is an important topic in cardiac surgery, source of important comorbidity and post-operative mortality.

The paper is well written and the conclusion are definitely supported.

Regards

**********

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

Reviewer #3: Yes: Cristina Barbero, MD, PhD

[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.

23 Sep 2020

Response to Reviewers

Editor comment: Furthermore, more detailed statistical analysis, such as a multivariate analysis and a propensity score matching, are needed to confirm the strength of the results.

Ans: Given the small data set size, it is unfeasible to perform propensity matching on a large number of parameters. We have instead performed a basic propensity matching of the two groups of patients using Euroscore II (or Euro SCORE II) values, and find similar results to the full set. Results are shown in a separate heading under propensity matching under section 3.4 and included in the revised manuscript as Supplementary table 1, (S1 Table) and supplementary figure S2 (S2_fig).

Reviewer #1

Query #1: This study is severely limited by the sample size of patients. In order to make any meaningful generalizations, sample sizes of 11 and 8 are inadequate. Please do a power analysis if possible.

Ans: The objective of this study was to observe if the endothelial repair response as indicated by change in circulating progenitor cell mobilization was different in patients with and without clinically defined vasoplegia. In that regard the fold change experienced was by individual patients with respect to their own baseline. The subsequent trajectory of the response was characterized by both quantitative (change in number of circulating progenitor cells) and qualitative (change in types of circulating progenitor cell markers) methods. The analysis was made to see if this quantum varied between patients who developed vasoplegia and those that did not (Hence, each patient serves as their own control). If there were indications that it was the case then we planned to do a bigger study as this was more in the nature of a pilot study.

We have performed a power analysis which is now included in results section 3.4. The effect size (Cohen's d) for our data is 1.63 (i.e., the means of the two groups are 1.63 standard deviations apart), which is a large effect. Based on simulating random data with the same distribution (same effect size), for 8 non-vasoplegic and 7 vasoplegic patients at a significance threshold of 0.05, the statistical power is 0.6. The results of the power analysis are shown in table-3 in the revised manuscript.

Query #2: The patient population is not adequately described in multiple areas. There is a range for the ages of the patients which is 40-70. This is a very wide range and there are multiple studies showing that age affects inflammation. The study has medication intake per group, but statins are not included and statins are showed to be anti-inflammatory. Lopressor is a very popular medication which has also been shown to be anti-inflammatory. Also, preoperative comorbidities like COPD, chronic liver disease, steroid use, etc. can have such a major impact on outcome. Did any of the patients have heart failure? What was the PREOPERATIVE LVEF of the patients? Was there pulmonary artery hypertension or RV dysfunction? At this point, I have no clue what types of patients are being operated on.

Ans: We partly agree with the reviewer. As we have given most critical information required to understand the patient population which might affect the post CPB outcomes (Please refer to Table-1 in the revised manuscript). We felt that ACEI (Angiotensin-converting-enzyme inhibitor) and ARBs (Angiotensin II receptor blockers) have most significant effect on post-operative vasodilatory syndromes; however, as suggested by the reviewer, we have now added the details of intake of beta blocker (e.g. Lopressor a brand name of Metoprolol in USA, which is selective β1 receptor blocker) as well as steroids and statins in table 1 of the revised manuscript.

We have included the pre OP co-morbidities heart failure, RV dysfunction and PAH as suggested by the reviewer. However, the idea was to see if prediction of vasoplegia from EPC behavior alone was possible and if so, how many patients could be accurately identified as developing vasoplegia for a given EPC profile independent of co-morbidity or risk score profile.

Moreover, in the revised manuscript we have included objective risk scores which takes in to account the pre-operative co-morbidities (Please refer to answer to query # 3 for more details).

Query #3: Please use a score to determine risk assessment of the cardiac surgery patients. This will give us an idea of how sick the patients are that were included in the study. In the US, we use the STS-PROM. Whatever score is suitable for you, please include. If could very well be the patients who had an increase pro-inflammatory response were patients who were sicker prior to surgery.

Ans: We have now included risk assessment scores Euro SCORE II, SOFA II and APACHE II used in our hospital set up in table 2 of result section 3.1 in the revised manuscript. The results of the score calculation is appropriately discussed in the revised manuscript. Euro SCORE II (European System for Cardiac Operative Risk Evaluation II) is used as a pre-operative indicator of patients’ risk profile[1,2]. SOFA II (Sequential organ Failure Assessment II) is used to determine level of organ dysfunction and mortality risk in ICU patients. APACHE II (Acute physiology and Chronic Health Evaluation II) score estimates ICU mortality[3-5]. From Euro SCORE II calculation it was observed that the group 2 subjects had two-fold % estimated mortality prediction compared to group 1. Post-operative ICU score, SOFA II indicates higher % mortality (26.3%) in group 2 compared to group 1 (18.2%) and APACHE II showed similar estimated % mortality for group 1 (7.6%) and group 2 (7.75%) (Table-2). Hence, the enumeration and identification of circulating progenitors in our study subjects in a temporal manner could add new parameter in the prediction of the severity of post CPB clinical state.

Query #4. Operative details were left out. What type of operations did these patients undergo? I don't believe a 40-year-old undergoing a CABG and a 70-year old undergoing CABG will react differently. A 70-year old is expected to have level of CAD, while a 40-year old is likely to not have CAD. I would assume the 40-year-old may have increased pro-inflammatory markers prior to surgery; does this affect your study? What about a 40-year undergoing a MV repair for failed mitral disease whose relatively healthy vs. a 40-year-old with endocarditis? We see these patients all the time.

Ans: We have now included the type of operation conducted in the subject group in table 1 of the revised manuscript as asked by the reviewer.

Patient characteristics like pre-operative co-morbidities and operation related factors were discussed while calculating the objective risk scores Euro SCORE II. We have now included pre-operative risk prediction score Euro SCORE II in the revised manuscript (Table-1).

Query #5: More on operative detail - What were the CPB times for the patients? What were the cross-clamp times? What cardioplegia did you use? What MAPs were maintained in the OR? What was the flow for the patients? All of these things that happen in the OR affects pro-inflammatory markers and lactic acidosis, etc. If you have a patient undergoing a cross- clamp time of 180 minutes vs. 80 minutes, I would assume that the longer cross clamp time would have more pro- inflammatory markers. We certainly see that these patients are usually sicker coming out of the operating room. You may want to look into the effects of CPB time or cross clamp time on the markers you discussed.

Ans: We do agree with the reviewer that longer cross clamp time would have more pro-inflammatory markers. Other intra-operative parameters such as CPB time, flow rates and MAPs too could affect the patient outcome. We have now included duration of CPB and Cross clamp time as well as flow rate in table 2. However, we did not find significant difference in these parameters among the 2 groups as shown in Table-2 of the revised manuscript.

Blood cardioplegia was used for all the cardiac cases. The cardioplegic solution was a mixture of patient’s own blood and crystalloid solution. The crystalloid solution was a mixture of Plasmalyte (500 mL) and St. Thomas' solution (20 mL). The delivery was done by the use of a specialized roller pump (Sarns Medical Systems) at the ratio of 4:1 where 4 parts of blood and 1 part of the mixed cardioplegia solution was used. In lung transplant the lung preservative used was Perfadex solution (Medisan, Uppsala, Sweden) containing low-potassium (K+ 6 mmol/L) and extracellular electrolytes (Na+ 138 mmol/L; Cl- 142 mmol/L; Mg++ 0.8 mmol/L; Dextran 40 g);H2PO4- 0.8 mmol/L; Glucose (0.91 g); Osmolarity (292 mOsm/L). All these details are now included in methods section 2.2 of the revised manuscript.

The MAP of 60 to 80 mmHg was maintained in the OR for all the surgery cases which were included in our study group.

We examined this question and found no correlation between CPB time or cross clamp time with the markers. This is evident in the bar plot shown here for the reviewers' benefit. (Shown in the attached file ' Response to Reviewers)

Query #6: Post-operative course is not described passed 24 hours. First, the study has a 10% mortality rate, which is exceedingly high depending on the operation and how sick the patient is prior to surgery. Once again, this is a reason patient population needs to be more adequately described and a risk assessment score needs to be included in the paper.

Ans: Vasodilatory syndrome post cardiac surgery occurs between 6 to 24 hour. The study was designed to observe circulating progenitor cell response during the operation and in the first 24 hours. However, we have included post-operative milestones such as total hours ventilated and duration of hospital stay.

The total number of cardiac surgeries carried out during the period of study was 260, out of which 2 patients died. Hence; the overall mortality rate during the period of study is 0.77%. Out of total 260 cardiac surgeries 19 patients consented to be part of the pilot study of which 2 patients died. Thus, the 10% mortality was a reflective of number of subjects consented and enrolled in the study. As mentioned in the reply to query #3, we have included the risk score in the result section of the revised manuscript for the study population.

Query #7: Norepi is the drug that was used as a consensus for clinical vasoplegia. How was it determined to place this patient on norepi vs. epi. What about norepi vs. vasopressin? What about clinical fluid status? Maybe the patient is not vasoplegic, but dry and required fluids? Once again, patient population needs to be described.

Ans: The protocol in our department is use of noradrenalin limited to patients who have moderate ejection fraction (30-50%) or normal EF (> 50%) if there is hypotension (< 60mmHg MAP), after the CVP was restored to pre surgical (at induction level) or as identified by the surgeon or anesthetist as having been the optimal filling in the OR.

In poor EF (<30%) adrenaline or dobutamine is first line and nor-adrenaline added only if warm peripheries and low systemic vascular resistance index (SVRI<1600) is documented. Increase in noradrenalin above 5mls/kg/min is not allowed unless further low SVRI by PAFC (Pulmonary artery floatation catheter) reading and good LV and RV function by echo is documented. Vasopressin is added in post-operative period only if the noradrenalin dose exceeds 10mls/kg/min.

Clinical fluid status difference between two groups is added in the revised manuscript. The results are also shown as percentage of subjects with positive, negative or zero fluid state at the end of CPB in Table-2 of the revised manuscript.

Reviewer #2:

Query #1: The subject of the paper is scientifically relevant and the work performed correctly, as to my expertise. Nevertheless, the population enrolled to the study is small, considered the following sub-categorization in two groups. The differences observed by the authors will certainly be reinforced by inclusion of 10-20 more patients and scientific soundness of the paper would benefit from including other groups of patients off-pump cardiac surgery as a "control group" and ECMO patients.

Ans: The study was designed to elucidate if there was a particular trajectory to the EPC response that was associated with the development of vasoplegia irrespective of patient profile. Our intention was not to compare different patient sub groups but to compare the circulating progenitor cell response profiles among the groups who developed vasoplegia to those who did not.

We would like to bring in the kind attention of the reviewer that the patient’s own baseline/post induction blood sample acts as patient’s control. This helps us to develop a patient’s EPC profile which is comparable between the patients. Therefore, a separate control group for comparisons is not necessary for our study design.

Reviewer #3:

I congratulate the author for the great job. The authors analysed vasoplegia observed post-cardiopulmonary bypass and the correlation with CD34+ cells and CD34+CD133+ hematopoietic stem cells. This is an important topic in cardiac surgery, source of important co morbidity and post-operative mortality. The paper is well written and the conclusions are definitely supported.

Ans: We thank the reviewer for the comments.

REFERENCES

1. Roques F, Nashef SA, Michel P, Gauducheau E, de Vincentiis C, Baudet E, et al. Risk factors and outcome in European cardiac surgery: analysis of the EuroSCORE multinational database of 19030 patients. Eur J cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg. 1999;15: 813–816. doi:10.1016/s1010-7940(99)00106-2

2. Roques F, Michel P, Goldstone AR, Nashef SAM. The logistic EuroSCORE. European heart journal. England; 2003. pp. 881–882. doi:10.1016/s0195-668x(02)00799-6

3. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13: 818–829.

4. Wong DT, Knaus WA. Predicting outcome in critical care: the current status of the APACHE prognostic scoring system. Can J Anaesth. 1991;38: 374–383. doi:10.1007/BF03007629

5. Mei YQ, Ji Q, Liu H, Wang X, Feng J, Long C, et al. Study on the relationship of APACHE III and levels of cytokines in patients with systemic inflammatory response syndrome after coronary artery bypass grafting. Biol Pharm Bull. 2007;30: 410–414. doi:10.1248/bpb.30.410

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

14 Oct 2020

PONE-D-20-15232R1

Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: An observational study

PLOS ONE

Dear Dr. Thangaraj,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Specifically, a thorough grammar revision is needed.

Please submit your revised manuscript by Nov 28 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Antonio Cannatà

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

A thorough grammar revision is needed. Please consider using an external proofreading before resubmitting the manuscript.

[Note: HTML markup is below. Please do not edit.]

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 #1: All comments have been addressed

Reviewer #2: 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 #1: Yes

Reviewer #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: 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 #1: Yes

Reviewer #2: 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 #1: Yes

Reviewer #2: 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 #1: Thank you for revising your manuscript.

I do very much appreciate the fact that a lot of the data that I requested was added to the manuscript. I think when you talk about vasoplegia, it is important to also have a good clinical picture for such a picture. With regards to this, I think I would like to see more about what to do with this information in a more clinical setting? Do we prophylactically treat these patients who are predisposed to vasoplegia? Please add something short regarding what to do with your information.

Also, please review grammar and english syntax.

I commend the authors on an interesting study and a well written paper.

Reviewer #2: The authors correctly addressed my comments and those of the other reviewer. The reason of enrolling such a small population of patients is now better outlined and is reasonable. In my opinion, the pilot nature of the study must be clear from the very title of the paper, so my suggestion is to add "a pilot study" to the title.

**********

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 #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.

28 Oct 2020

Editor comment: A thorough grammar revision is needed.

Ans: As per the suggestion by the editor, we have made a thorough grammar revision for the revised manuscript.

Reviewer #1

Query #1: Thank you for revising your manuscript.

I do very much appreciate the fact that a lot of the data that I requested was added to the manuscript. I think when you talk about vasoplegia, it is important to also have a good clinical picture for such a picture. With regards to this, I think I would like to see more about what to do with this information in a more clinical setting? Do we prophylactically treat these patients who are predisposed to vasoplegia? Please add something short regarding what to do with your information.

Also, please review grammar and English syntax.

I commend the authors on an interesting study and a well written paper.

Ans: We thank the reviewer for the comments. We have now included a short summary regarding the findings and perspective of the current pilot study in the discussion section of the revised manuscript.

Query #2: The authors correctly addressed my comments and those of the other reviewer. The reason of enrolling such a small population of patients is now better outlined and is reasonable. In my opinion, the pilot nature of the study must be clear from the very title of the paper, so my suggestion is to add "a pilot study" to the title.

Ans: We thank the reviewer for the comment. As suggested by the reviewer we have changed the title from ‘Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: An observational study’ to ‘Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: A pilot study’.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

2 Nov 2020

Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: A pilot study

PONE-D-20-15232R2

Dear Dr. Thangaraj,

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,

Antonio Cannatà

Academic Editor

PLOS ONE

9 Nov 2020

PONE-D-20-15232R2

Altered kinetics of circulating progenitor cells in cardiopulmonary bypass (CPB) associated vasoplegic patients: A pilot study

Dear Dr. Thangaraj:

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. Antonio Cannatà

Academic Editor

PLOS ONE