Immediate and early post-operative sequelae of off-pump total cavopulmonary connection

Abstract

Background

Extracardiac Fontan (ECF) is currently the final operation of choice for patients with a univentricular heart. Performing this procedure without cardiopulmonary bypass (CPB) carries potential benefits. In this study, we report the early results of ECF without CPB.

Patients and methods

Between 2012 and 2015, 72 consecutive patients underwent Fontan without CPB. Their medical records were examined in detail.

Results

Mean age was 11.8 ± 5.2 (range 5 to 23, median 10) years. Intraoperative mean superior vena cava clamp time was 15.19 ± 3.8 min, and the inferior vena cava clamp time was 16.93 ± 3.31 min. There were three early deaths. No patient required conversion from off-CPB to CPB. Mean inotropic score was 4.73 ± 5.9 (range 0 to 25, median 2.5). Mean time to extubation was 9.5 ± 5.82 (range 3 to 29, median 8) hours. Pleural drainage in intensive care unit (ICU) was 551.57 ± 452.77 (median 470) ml, and mean ICU stay was 2.27 ± 3.09 (median 1.5) days. Mean daily pleural drainage after discharge from the ICU was 163.7 ± 88.01 (median 140) ml, and mean time to removal of pleural tubes was 15.76 ± 8.4 (median 14) days. Total hospital stay was 17.03 ± 8.62 (median 15) days. At an early follow-up of 2–40 (median 25) months, all survivors (n = 69) had a patent Fontan circuit with normal ventricular function on echocardiography. There were no late deaths or thromboembolic complications.

Conclusions

Off-pump ECF is a low-risk procedure that avoids the harmful effects of CPB. Post-operative course of these patients is predictable with substantial savings in costs.

Introduction

Since the original description of right heart bypass by Fontan and Baudet in 1971 [1] as palliation for children with tricuspid atresia, the procedure has undergone significant modifications. De Leval et al. [2] proposed the concept of total cavopulmonary connection (TCPC) to minimize kinetic energy loss in the Fontan circuit. Subsequently, TCPC using an extracardiac conduit (ECF) was first proposed by Marcelleti et al. [3, 4] and is currently the final palliation of choice for patients with a univentricular heart (UVH). This procedure aims to divert the total systemic return from both the cavae to the lungs, thereby improving the systemic oxygen saturation by increasing the mean pulmonary blood flow in patients with complex cyanotic congenital heart diseases not amenable to biventricular repair [2]. The ECF has a number of theoretical advantages over the traditional lateral tunnel TCPC that include a shorter bypass time and the avoidance of the need for arresting the heart and avoidance of intra- atrial suture lines. Elimination of the atrial suture lines has been shown to result in a lower incidence of arrhythmias. In addition, the design of the ECF is associated with laminar blood flow within the Fontan pathway [5–7]. Lardo et al. [5] studied fluid dynamics in sheep heart preparations and noted that fluid-power losses were consistently lower for the ECF compared to the intra-atrial tunnel configurations of the Fontan operation. Nurnberg et al. [6] studied 74 consecutive patients receiving either lateral tunnel TCPC (LTF) or ECF. The incidence of new onset supraventricular tachyarrhythmias (SVTs) was lower in patients undergoing ECF as compared to those undergoing the LTF [6]. Because the ECF involves placement of a conduit outside the heart, the operation may or may not be performed with the patient supported on cardiopulmonary bypass (CPB), without arresting the heart. CPB, however, has the potential disadvantages of derangement of ventricular and pulmonary functions, arrhythmia, delayed extubation, increased incidence of pleural effusion, increased requirement of inotropes, blood products, and the activation of inflammatory cascade in the post-operative period, more so patients with UVH [8]. To overcome these problems, the ECF may also be accomplished without using CPB in patients not requiring an associated intracardiac procedure or extensive pulmonary (PA) reconstruction.

Since 2012, the lead author (ST) has adopted a policy of performing the ECF exclusively without using CPB, if no concomitant intracardiac procedures are needed. The detailed technique of performing this procedure without using CPB and a comparison of the results of this technique in 37 patients with a group of historical controls undergoing the TCPC on CPB was published by us earlier [9]. As our experience has continued to evolve, we now report detailed intraoperative, immediate and early results.

Patients and methods

This is a prospective analysis of 72 patients who underwent an off-pump ECF between August 2012 and December 2015. Informed consent was obtained from all patients, and the study protocol was duly approved by the Institutional ethics committee (IEC/NP-119/11.04.2014, RP-50/2014).

Patient demographics are listed in (Table 1). Thirty-one of these patients had undergone a prior bidirectional Glenn (BDG), and 41 patients underwent the primary ECF without any previous palliation. Out of the entire cohort, six patients underwent coil embolization of major aortopulmonary collaterals in the cardiac catheterization laboratory prior to the ECF.

Table 1.

Patients’ preoperative diagnoses

Diagnoses	Number
DORV, VSD, PS	26
TA, VSD, PS	19
DILV, VSD, PS	4
cc-TGA, VSD, PS	11
d-TGA, VSD, PS	5
AVSD, Hypoplastic RV, PS	2
SI, DC, SV, PS	2
SI, DC, DORV, VSD, PS	1
SI, DC, AVSD, PS	1
SV, HYPOPLASTIC RV, PS	2

DORV double outlet right ventricle, VSD ventricular septal defect, PS pulmonary stenosis, TA tricuspid atresia, DILV double inlet left ventricle, cc-TGA corrected transposition of great arteries, d-TGA d-transposition of great arteries, AVSD atrioventricular septal defect, SI situs inversus, RV right ventricle, DC dextrocardia

Surgical procedure

All operations were performed by a single surgeon (ST). ECF was performed in patients weighing at least 15 kg so that an adequate-sized extracardiac conduit could be used. The surgical technique for performing the off-pump TCPC has been detailed in our prior publication [9]; however, it is briefly detailed as under (Figs. 1, 2, 3, and 4).

Fig. 1.

Figure showing purse string sutures for placement of shunts to decompress the vena cavae. (Reproduced with permission from: Talwar S et al. Extra cardiac Fontan without cardiopulmonary bypass: techniques and early results. Indian J Thorac Cardiovasc Surg 2013: 29(3):174–183)

Fig. 2.

A temporary shunt is established between the superior vena cava (SVC) and right atrium (RA). (Reproduced with permission from: Talwar S et al. Extra cardiac Fontan without cardiopulmonary bypass: techniques and early results. Indian J Thorac Cardiovasc Surg 2013: 29(3):174–183)

Fig. 3.

Completed anastomosis between the superior vena cava (SVC) and the right pulmonary artery (RPA). A clamp has been placed across the under-surface of the RPA, and the transected distal main pulmonary artery and the graft (G) are being sutured to it. Note the position of the clamp as it allows unhindered flow from SVC to pulmonary arteries. (Reproduced with permission from: Talwar S et al. Extra cardiac Fontan without cardiopulmonary bypass: techniques and early results. Indian J Thorac Cardiovasc Surg 2013: 29(3):174–183)

Fig. 4.

A curved Crafoord clamp (C) has been applied at the inferior vena cava (IVC) which is transected and the cardiac end is over sewn. SVC superior vena cava, RPA right pulmonary artery, G graft, S stay suture used to retract the heart. For description, see text. (Reproduced with permission from: Talwar S et al. Extra cardiac Fontan without cardiopulmonary bypass: techniques and early results. Indian J Thorac Cardiovasc Surg 2013: 29(3):174–183)

Surgical approach was through a median sternotomy. After opening the pericardium, the cardiac anatomy was assessed and purse string sutures were placed on the aorta and right atrial appendage as a precautionary measure so that if needed, emergency CPB could be instituted. The right and left pulmonary arteries (PA) were mobilized up to the hila of the lungs. The superior vena cava (SVC) was dissected free from its surrounding attachments, and both the SVC and the PA were looped separately. Heparin was administered in a dose of 3 mg/kg to achieve full systemic heparinization and an activated clotting time (ACT) of more than 250 s so that if required, CPB could be instituted immediately. An SVC to right atrium (RA) shunt was created using two-angled venous cannulae, and the BDG was constructed after clamping and transecting the SVC in the standard fashion, following which the cardiac end of the SVC was closed. The BDG circuit flow was now established. Following the BDG, the angled venous cannula in the SVC was taken out, leaving the RA cannula in situ.

The next step in the operation was the placement of the extracardiac conduit. It has always been our practice to construct the anastomosis of the conduit to the pulmonary artery before the IVC end as this serves to minimize the inferior vena cava (IVC) clamp time. After completion of the conduit-PA anastomosis, an angled cannula was introduced low down into the IVC and connected to the RA cannula, allowing passive lower body venous decompression during the IVC anastomosis. In all patients, a central venous line was placed in the femoral vein and before transection of the IVC, a test clamp was placed on the IVC to test for rise in the IVC pressure (acceptable range 15–20 mmHg) and for assessment of any haemodynamic instability. The patient was placed in a mild Trendelenburg position to aid lower body drainage. Mild systemic hypotension was treated with volume administration in the upper extremity veins. After, clamping the IVC, it was transected; the cardiac end was closed, and the conduit was then anastomosed to the caudal IVC end. A fenestration was not performed routinely but was reserved for patients with borderline PAs, ventricular dysfunction, very high central venous pressures, or when excessive pleural drainage was anticipated.

After completion of the operation, the hemodynamic status was evaluated intra-operatively. If the Fontan pressure was consistently above 18 mmHg, with a transpulmonary gradient of 12 mmHg or more, a fenestration was placed between the conduit and the right atrial free wall. This was accomplished without CPB, with a direct 4- to 5-mm side-to-side anastomosis between the graft and the right atrium in two patients.

Data collection and follow-up

In all patients, data was collected preoperatively for demographics which included age and weight at Fontan operation, cardiac morphology, mean PA pressure (mmHg), size of the right and left PA, preoperative saturation, systemic ventricular function, ejection fraction, and ventricular end-diastolic pressure (mmHg). Intra-operatively, data was collected about the size of extracardiac conduit, SVC and IVC clamp time (min), and whether a fenestration was required or not. In the intensive care unit (ICU), inotropic score was calculated using the formula described by Wernowsky et al. [10]: inotropic score = dose of dopamine [mg/kg/min] + dobutamine [mg/kg/min] + 100 × epinephrine [mg/kg/min] [10].

Early deaths, inotropic score, time to extubation (hours), ICU stay (days), duration and amount of chest tube drainage, and systemic oxygen saturation (as measured by pulse oximetry) were recorded. Prior to discharge from the hospital, all patients underwent transthoracic echocardiography.

After discharge from the hospital, the patients were called for follow-up at an interval of 1, 3, and 6 months. Follow-up information collected included copies of physician notes, resting electrocardiogram, ambulatory rhythm monitoring (if any arrhythmias were detected on ECG), and echocardiogram to assess the flow in the Fontan pathway.

Statistical analysis

Statistical analysis was performed using the Stata 14.0 software (StataCorp LP, College station, TX, US). Values are presented as mean ± standard deviation (SD) with median and interquartile rage (IQR) for quantitative variables and n (%) for qualitative variables. For purposes of analysis, the immediate post-operative period was defined as the time from surgery to hospital discharge. Follow-up postoperative period was defined as the time from initial hospital discharge to latest follow-up.

Results

Median age was 10 years (mean 11.79 ± 5.16 years, range 5–23 years), and median weight was 24 kg (mean 30.21 ± 13.4 kg, range 15–75 kg). Pre-operative saturation was a mean 78 ± 8.6% (median 78%, range 50–93%). The morphologic characteristic of the dominant ventricle was left (n = 43), right (n = 26), or indeterminate (n = 3). Primary ECF without a prior palliation was performed in 41 patients (57%), and completion Fontan was performed in 31 patients (43%) who had undergone a prior BDG. A polytetrafluoroethylene conduit (Gore-Tex Stretch Vascular Graft; W. L. Gore and Associates, Inc., Flagstaff, Ariz) was used in all patients. Median conduit size was 20 mm (mean 19.8 ± 1.75 mm, range 18–24 mm). Conduit size was 18 mm or larger in all patients and 20 mm or larger in 65% (n = 47).

The mean SVC clamp time was 15.19 ± 3.8 (median 15, range 8–26, IQR 12–18) minutes, and the mean IVC clamp time was 16.93 ± 3.31 (median 17, range, 10–30, IQR 15–18) minutes (Table 2). None of these patients required conversion from off-CPB to CPB at the time of ECF.

Table 2.

Immediate postoperative variables

Parameters	Mean ± S.D.	Range	Median (IQR)
SVC clamp time (min)	15.19 ± 3.8	8–26	15 (12, 18)
IVC clamp time (min)	16.93 ± 3.31	10–30	17 (15, 18)
Conversion to on-pump	NIL	NIL	NIL
Inotropic score	4.73 ± 5.9	0–25	2.5 (0, 8)
Time to extubation (hours)	9.5 ± 5.82	3 to 29	8 (6, 12)
Chest tube drainage (in ICU) (ml)	641.57 ± 895.96	30 to 1200	470 (250, 762)

ICU intensive care unit, IQR interquartile range

The mean inotropic score 4.73 ± 5.9 (median 2.5, IQR 0–8, range 0 to 25), indicating a minimal need for inotropes. Mean time to extubation was 9.5 ± 5.82 h (median 8, range 3 to 29, IQR 6–12) hours. Mean ICU stay was 2.27 ± 3.09 days (median 1.5 days, range 1 to 20, IQR 1–25) days. The mean chest tube drainage in the ICU was 551.57 ± 452.77 ml (median 470 ml, range 30 to 1200 ml, IQR 250–762 ml) (Table 3).

Table 3.

Early postoperative patients’ variables

Parameters	Mean ± S.D.	Range	Median (IQR)
ICU stay (days)	2.27 ± 3.09	1 to 20	1.5 (1, 2)
Daily chest tube drainage post-ICU discharge (ml)	163.7 ± 88.01	50–550	140 (100, 185)
Time to chest tube removal (days)	15.76 ± 8.4	4 to 64	14 (10, 19)
Total hospital stay (days)	17.03 ± 8.62	7 to 64	15 (12, 20)
Saturation at discharge (%)	99.3 ± 1.31	93 to 100	100% (99, 100)

ICU intensive care unit, IQR interquartile range

There were three early deaths. One sudden death occurred due to presumed cerebrovascular accident on post-operative day 4 in the ward, and two deaths occurred on day 3 and 7 in the ICU. The first of these two patients had severe unexplained ventricular dysfunction. The patient who died on day 7 had bleeding from the IVC to conduit anastomosis at the time of initial surgery. This bleeding was triggered inadvertently by the wall sucker-induced trauma to the IVC while checking hemostasis. At that time, multiple sutures were needed to repair an accidental tear in the IVC following attempts to control the bleeding. Conduit thrombosis was noticed in this patient 8 h following the initial surgery, and the IVC to conduit anastomosis was revised under CPB and circulatory arrest. Subsequently, this patient developed multi-organ dysfunction and died on the 7th post-operative day. Other than this patient, institution of CPB was not needed in any patient.

No patient developed Fontan failure. There were no peri-operative cerebral events, hepatic dysfunction (as assessed by liver enzymes), cardiac ischemia, or arrhythmias. In all patients, aspirin (5 mg/kg/day) and oral warfarin were started on the first post-operative day. The dose of warfarin was titrated to achieve an international normalized ratio (INR) of 2–3. Apart from the one patient discussed above, no patient developed thrombosis of the conduit, and thromboembolic and bleeding complications were not observed.

Mean post-operative saturation was 99.33 ± 1.32% (median 100%, IQR 99–100, range 93 to 100%).

Mean daily pleural drainage after discharge from the ICU was 163.7 ± 88.01 (median 140) ml. Forty-three patients (59.7%) had chest tube drainage less than 500 ml, and 67 patients (93.1%) had chest tube drainage less than 1000 ml. Chest tubes were removed in 15.77 ± 8.4 days (median 14, IQR 10–19 days).

Mean hospital stay was 17.03 ± 8.4 days (median 15 days, IQR 12–20, range 7 to 64 days). Thirty-nine patients (54.17%) had hospital stays less than 2 weeks.

These patients incurred an average expenditure Rs. 32,030.74 ± 9118.09 (median 34,500) compared to Rs. 68,518.52 ± 6516.96 (median 65,000) in patients undergoing TCPC on CPB. The difference in costs is due to avoidance of disposables used for CPB and represents the actual amount billed to the patients. However, we did not calculate the costs related to ICU and hospital stay and costs of drugs and manpower. It is, however, expected that these costs should also be less to the hospital in patients undergoing off-CPB ECF because of reduced ICU and hospital stay, as has been shown by us earlier.

Median follow-up was 29 months. At follow-up, no patient has any arrhythmias on ECG or thromboembolic complications and on follow-up transthoracic echocardiography, the Fontan circuit was patent in all patients with normal systemic ventricular function. A study comprising of ambulatory 24-h continuous Holter monitoring in patients undergoing the Fontan is in progress to elucidate the incidence and to identify the risk factors and subsets with a higher incidence of arrhythmias.

Discussion

Management of patients with single ventricular physiology has continued to evolve. Many institutions have adopted various approaches with improved early clinical outcomes. Staging the Fontan operation with the BDG has undoubtedly contributed significantly to the improved early outcome of these patients. Studies by De Leval et al. [2] have added further guidance in optimizing the geometry of the cavopulmonary connections.

The use of an extracardiac IVC to PA connection has a number of theoretical advantages. It avoids extensive atrial suture lines, exposure of the atrium to higher venous pressures, and theoretically better preservation of kinetic energy in the Fontan circuit^. [11]. In addition, because the connection is created outside the heart, aortic cross-clamping can be entirely avoided. These theoretical advantages have prompted the application of the ECF at our institution.

A higher duration of CPB is considered as a risk factor for adverse early outcomes in patients undergoing the Fontan procedure [10, 12]. In Feb 2012, we introduced the off-pump ECF at our institution [9]. Subsequently, this operation was successfully established in our institution as a feasible surgical procedure associated with overall favorable haemodynamics and a smooth and predictable post-operative course. In line with the improving worldwide results in the early postoperative course after the Fontan operation, our patients were mostly transferred to the intermediary care unit after a short period of mechanical ventilation on the first or sometimes the second postoperative day. Also, as described by us in an earlier publication [13], the duration of the ICU stay, pleural effusions, and total hospital stay were significantly shorter in the off-pump Fontan group as compared to the on-pump TCPC. Postoperative mortality and significant early morbidity, including early arrhythmias, need for prolonged ventilation and prolonged early postoperative pleural effusions and ascites, were also decreased [13]. Although this is not related to the technique of off-pump Fontan, we did not observe any instances of Fontan failure in our cohort.

Many authors have described that the use of CPB in patients with UVH leads to increased occurrence of pulmonary injury, higher systemic inflammatory response syndrome, fluid retention, myocardial dysfunction, thromboembolic events, and the necessity for blood transfusions. These studies reveal a direct association between the duration of CPB and worse early outcomes [14–18].

CPB is known to produce a transient elevation of pulmonary artery pressures and pulmonary dysfunction that may lead to a protracted ICU course and a higher incidence of pleural effusion. Preoperative low-oxygen saturation, increased pulmonary artery pressure, and usage of CPB are among the factors associated with prolonged pleural effusion [8]. The off-pump ECF can be technically challenging in certain subsets of patients with anomalies of situs and hetreotaxy and in patients with a midline IVC with the entire cardiac mass lying over it. However, with increasing experience, it can be successfully accomplished in nearly all anatomical subsets. However, we would still prefer to perform the procedure on CPB if extensive PA plasty is needed and in high-risk Fontan candidates where we would prefer the intra-extracardiac Fontan modification described by Jonas [19], for ease of fenestration and it needed, a Fontan takedown. Other than these subsets, we would prefer to perform the Fontan without CPB as we believe that the avoidance of CPB definitely has positive effects on the outcome of Fontan operation in the early postoperative period.

Post-operative pleural effusions following the Fontan operation are multifactorial, and many inflammatory, hydrostatic, and hormonal mechanisms contribute to their development [20, 21]. Inflammatory changes, resulting in capillary leakage and subsequent fluid retention, may consequently lead to increased pulmonary vascular resistance and decreased pulmonary blood flow [22, 23]. Decreased ventricular compliance contributes to the development of persistent pleural effusion and ascites or even of early Fontan failure, especially in borderline Fontan candidates [14, 24]. Thus, overall CPB time has a negative impact. Also, use of an oxygenator and plastic tubing through which the blood passes seems to make an important impact through the activation of the complement system and induction of a systemic inflammatory reaction [20–24]. Contact activation by alloplastic material stimulates the release of reactive oxygen species and granulocyte contents including elastase.

After discharge from the ICU, we noted median total chest tube drainage of 470 ml; 43 patients (59.7%) had total chest tube drainage less than 500 ml, and 67 patients (93.1%) had chest tube drainage less than 1000 ml. This drainage over a period of 15 days ± 8.39 (median 14, IQR 10–19 days) is acceptable and has contributed to our improving results when ECF is performed without CPB.

Thromboembolic complications are a well-recognized source of morbidity and mortality in patients with Fontan physiology [25–27]. Predisposing factors include atrial dilation, atrial tachyarrhythmias, and sluggish flow. Also, several clotting factor abnormalities have been reported, including decreased levels of protein C, protein S, and antithrombin III. Increased platelet reactivity has also been recognized [28, 29]. The pathophysiology of thrombogenesis in patients with ECF Fontan remains to be elucidated. Due to the use of prosthetic material, anticoagulation was used in all of our patients and no thromboembolic events were reported. It has also been hypothesized that extracardiac conduits would result in a lower incidence of thromboembolism, owing to the avoidance of intracardiac prosthetic material. But this remains controversial. In a systematic review by Marrone et al. [30] that included 1075 patients with an ECF from 20 studies, 5.2% of patients had a thromboembolic event over a mean follow-up that ranged from 2 to 144 months. We, however, believe that a longer follow-up period is required to comment on the conduit thrombosis in patients undergoing ECF.

Atrial tachyarrhythmias are a leading source of morbidity in patients undergoing the Fontan operation [30]. Sustained tachycardia often leads to more frequent and prolonged recurrences, which are not tolerated by patients with Fontan physiology. They may result in reduction in ventricular systolic function, increase in atrioventricular valve regurgitation, atrial thrombosis, congestive heart failure, and syncope [31]. In addition to improving hemodynamics, the principal objective of the ECF is to minimize atrial arrhythmias. In our study, none of our patients developed reported symptoms of palpitations or signs or ECG evidence of arrhythmias.

However, a potential disadvantage of the ECF is related to the lack of growth potential of the conduit incorporated into the circuit [32]. In our experience, this issue has been effectively dealt with by waiting to perform the operation until the patient reaches a weight of about 15 kg (about 3–5 years of age), so that a near adult IVC-sized conduit (≥ 18 mm in 97% of our cases) can be inserted. This strategy is to accommodate the patient’s future exercise demands. Although long-term data are not available, to our knowledge, none of our patients who underwent ECF on CPB or off-CPB has required conduit replacement for growth-related pathway obstruction. Delaying the Fontan operation until 3 to 5 years of age incurs minimal additional risk. Worsening cyanosis does not generally develop until later, when patients start using more extensive lower body exercise and the head/lower body size ratio decreases further.

Substantial savings in costs can be achieved with the use of an off-pump procedure with the average hospital bill of these patients being Indian rupees 32,030.74 ± 9118.09 (median 34,500) compared to Indian rupees 68,518.52 ± 6516.96 (median 65,000) in patients undergoing TCPC on CPB (1 USD = Rupees 65) because of avoidance of CPB and its attendant morbidity and reduced amount and duration of pleural effusions.

Study limitations

The follow-up of our patients is short and is limited to studying their clinical course and echocardiographic findings. We propose to perform serial CT angiograms at longer follow-up and compare these to the on-CPB patients. Additionally, a study is in progress to determine the incidence of arrhythmias and to identify the risk factors responsible for these all patients undergoing the Fontan operation at our institute. We did not study the differences in serum inflammatory markers in patients undergoing ECF with or without CPB. A study is in progress to address this issue.

Another potential limitation of this paper is the oversimplification of the costs of the off-pump CPB presented in this paper. Primarily, the costs are lower because of the elimination of the costs of the disposables used in the CPB circuit. The costs of the ICU and hospital stay, drugs, and manpower costs have not been taken into account while presenting this data. However, we believe that the reduction in the ICU and hospital stay as demonstrated by us earlier [13] would translate into an overall reduction in costs in patients undergoing off-CPB ECF.

Conclusion

The early outcomes following off-pump ECF are acceptable. It is a safe and reproducible procedure that provides an adequately functioning conduit.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Statement of human rights/ethical approval

All procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. For this study, formal consent was obtained.

Informed consent

Informed consent was obtained from all individual participants included in the study.