Abstract
Background and Aims:
With the evolution of liver transplant anaesthesia, judicious intraoperative fluid management has been widely recommended. However, existing literature on a weight-based restricted fluid strategy is limited. Our institution follows a weight-based, restrictive fluid approach with maintenance therapy at 4 ml/kg/h (2 ml/kg/h crystalloids + 2 ml/kg/h 4% albumin solution) and goal-directed fluid boluses (4 ml/kg 4% albumin solution).
Methods:
This retrospective study analysed 106 adult living donor liver transplant recipients managed with a protocolised restrictive fluid strategy and compared them with 148 historical controls who received conventional fluid therapy. The primary outcomes were early extubation (extubated on table or within 6 hours post-operatively) and incidence of post-operative acute kidney injury (AKI) on post-operative days (POD) 1 and 3. Secondary outcomes included intra-operative blood transfusion requirement, vasopressor use, and vascular complication rate. Data were analysed using statistical package for the social sciences version 22 and Epi Info version 7.2.1.
Results:
The restrictive group received significantly lower total fluid volumes (3284 ± 833 ml vs. 3979 ± 1524 ml; P < 0.001), had higher on-table extubation rates (83% vs. 71.3%; P = 0.031), and experienced zero AKI by POD3 compared to 13.6% in the conventional group (P < 0.001). Serum creatinine was lower on POD1 and POD3 in the restrictive group. Vasopressin use was higher (P < 0.001) in the cases, while noradrenaline and blood transfusion requirements were comparable between the groups.
Conclusion:
Our protocolised weight-based restrictive fluid regimen was associated with improved pulmonary outcomes without increased renal dysfunction. However, prospective studies with larger cohorts are required to validate our findings.
Keywords: Anaesthesia, extubation, fluid therapy, living donor liver transplant
INTRODUCTION
Since its inception, liver transplant has evolved substantially in surgical and anaesthetic techniques, thereby improving patient outcomes. With the evolution of liver transplant, one of our important struggles has been against the injudicious fluid administration; liberal fluid therapy has been discouraged owing to guarded pulmonary outcomes, ileus, cardiac complications, increased infection rates, and mortality. A restricted or goal-directed fluid strategy has therefore been widely employed in various surgical procedures to improve post-operative outcomes without increasing renal dysfunction,[1,2] except for a few clinical trials that report otherwise.[3] Patients with chronic liver disease (CLD) pose unique challenges due to a state of central hypovolaemia in addition to a fragile renal reserve, thereby requiring a fine balance of fluid therapy for optimum outcomes. Although numerous trials[4] have supported restricted fluid therapy in liver transplant, its standardised practice guidelines are still lacking.
Our institution progressively standardised a weight-based restrictive intra-operative fluid protocol to improve post-operative pulmonary outcomes and reduce transfusion requirements. Our fluid strategy comprised a continuous baseline infusion at 4 ml/kg/h (2 ml/kg/h crystalloids + 2 ml/kg/h 4% albumin solution) and goal-directed fluid boluses (4 ml/kg 4% albumin solution) as required.
In this study, we conducted a retrospective case-control analysis of adult patients who underwent living donor liver transplant (LDLT), comparing those who received restricted fluid therapy to historical controls who received conventional fluid therapy. Our primary objectives were pulmonary outcomes and the incidence of post-operative acute kidney injury (AKI). Secondary objectives included intra-operative blood transfusion requirement, vasopressor requirement, and vascular complications.
We hypothesised that a weight-based restrictive fluid strategy, supplemented with goal-directed boluses and vasopressor support, would improve perioperative pulmonary outcomes without increasing the incidence of renal or vascular complications.
METHODS
Following institutional ethics committee approval (Ethics committee/AARCE/Letter/September/2024/68), we conducted a retrospective matched case–control study of adult patients (>18 years) who underwent LDLT in a superspeciality hospital between December 2022 and November 2023 using a prospectively maintained database. Informed consent was waived due to the retrospective design. The study was registered in the Clinical Trials Registry–India (CTRI/2025/05/087220) and adhered to STrengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines.
Since restricted fluid therapy is routinely practised at our centre, the historical cohort served as the control arm.
Cases were defined as recipients who underwent LDLT between December 2022 and November 2023 and received protocolised restrictive fluid therapy. Matched historical controls were selected from patients who underwent LDLT between December 2021 and November 2022 and received conventional fluid therapy.
Restrictive fluid therapy included a continuous baseline infusion at 4 ml/kg/h (2 ml/kg/h crystalloids +2 ml/kg/h 4% albumin solution) and goal-directed fluid boluses (4 ml/kg of 4% albumin solution). Conventional fluid therapy was the historical practice of fluid management, with fluid administered at the clinician’s discretion.
Clinical data for patients, including demographic characteristics, liver disease, intra-operative details (anaesthetic and surgical management), and post-operative outcomes, were obtained from medical records and anaesthesia and intensive care unit (ICU) charts.
Patients aged <18 years; patients with hepatic encephalopathy (Grade 2 or more at the time of transplant), acute liver failure, recent history of hepatorenal syndrome/AKI (<1 week), baseline serum creatinine >1.5 mg/dL, chronic kidney disease, or hepatopulmonary syndrome/portopulmonary hypertension; and those who underwent a combined procedure with liver transplant were excluded from the study. Patients who required intra-operative massive blood transfusion, continuous renal replacement therapy (CRRT), and post-operative mechanical ventilation/CRRT for graft dysfunction or neurological complications or patients with incomplete data were excluded from the analysis. After applying pre-defined exclusion criteria, 106 cases and 146 controls were included in the final analysis [Figure 1].
Figure 1.
Flow diagram for selection of study population. LDLT: Living donor liver transplant; AKI: Acute kidney injury; H/O: History of; HRS: Hepatorenal syndrome; HPS: Hepatopulmonary syndrome; PoPH: Portopulmonary hypertension; n: Number of patients; CRRT: Continuous renal replacement therapy; HE: Hepatic encephalopathy; CKD: Chronic kidney disease
The primary outcomes included pulmonary outcomes, that is, immediate and early extubation rate (extubated within 6 hours post-operatively), duration of mechanical ventilation, and pulmonary complications such as pulmonary oedema and pleural effusion requiring drainage within 72 hours post-operatively, and incidence of AKI on post-operative days (POD) 1 and 3 as per Kidney Disease: Improving Global Outcomes (KDIGO) criteria (≥0.3 mg/dL rise in serum creatinine or ≥1.5-fold increase from baseline). Serum creatinine values were recorded on POD0/POD1 and POD3. Need for renal replacement therapy was also noted. Secondary outcomes included intra-operative packed red blood cells (PRBCs) (units) transfused, vasopressor requirement (noradrenaline-µg/kg/min and vasopressin-units), and incidence of vascular complications.
Selection and information bias were minimised by pre-defined eligibility criteria and the use of a systematically maintained transplant database, while confounding was addressed by matching between the groups.
During the pre-operative fasting period, patients received intravenous fluids at 40–60 mL/h. All patients underwent standard general anaesthesia care and monitoring. Intra-operatively, continuous baseline fluid infusion at 4 ml/kg/h (2 ml/kg/h crystalloids +2 ml/kg/h 4% albumin solution) with intermittent boluses guided by goal-directed parameters (4% albumin solution at 4 ml/kg) was administered as a fluid bolus over 5–10 minutes.
Goal-directed parameters measured using the EV1000 (Edwards Lifesciences) system included stroke volume variation (SVV), systemic vascular resistance (SVR), and pulse pressure variation (PPV). SVV >15% associated with hypotension (mean arterial pressure <65 mmHg) was managed with fluid boluses after excluding reversible surgical causes such as cava compression and rotation of the liver. Additional fluid boluses were administered in the event of sudden blood loss, if urine output was <0.5 ml/kg/h for two consecutive hours, and at the time of cava clamping and reperfusion, at the discretion of the anaesthesiologist. Continuous albumin (20%) infusion at 10 ml/h was given in all cases.
Noradrenaline was used as the primary and vasopressin as a second-line vasopressor. Transfusion of PRBC, fresh frozen plasma (FFP), cryoprecipitate, and apheresis platelets was guided by serial arterial blood gas (ABG) analysis/thromboelastogram (TEG) values in combination with clinical assessment of bleeding.
Pre-operative fasting, anaesthesia-induced vasodilatation, and third-space losses (ascites, pleural effusion drained) were not replaced with fluids unless haemodynamic instability warranted fluid replacement. At the end of surgery, the total volume of fluid transfused (crystalloids + colloids), the average infusion rate (ml/kg/h), and the doses of noradrenaline (µg/kg/min) and vasopressin (units) administered were calculated. Sodium bicarbonate was administered when base excess was more than (-) 5 with pH <7.30 (in the presence of normocarbia). Based upon TEG values, low-dose heparin infusion was started in the post-reperfusion phase once adequate haemostasis had been achieved. Lactate levels at closure were also documented. Unless metabolic acidosis was significant with pH <7.25, all patients fulfilling institutional extubation criteria, that is, fully awake patients generating adequate tidal volume with no signs of respiratory distress and haemodynamically stable (without escalating vasopressor requirement), were given an extubation trial without taking lactate levels and their correction into consideration once satisfactory surgical vascular/biliary anastomosis was ensured. For the retrospective conventional fluid group, these same criteria guided extubation decisions as part of standard institutional practice, which had largely remained unchanged over the study period.
Data were entered into Microsoft Excel data sheets and were analysed using Statistical Package for the Social Sciences (SPSS) 22 (International Business Machines SPSS Statistics, Somers, New York, United States of America) and Epi-Info 7.2.1 (CDC, Atlanta) software. Categorical data were represented as frequencies and proportions, while continuous data as mean + standard deviation (SD). The Chi-square test was used to assess the significance of qualitative data. Kolmogorov–Smirnov and Shapiro–Wilk tests were used to assess the normality of the continuous data. An independent t-test was used to assess the difference between the means of two quantitative variables. The Mann–Whitney U test was used for non-parametric data between two groups. A P value of < 0.05 was considered statistically significant after assuming all the rules of statistical tests.
Because this is a retrospective study, all cases from the specified period that met the inclusion criteria were included. However, for statistical adequacy and to ensure scientific validity, sample size calculation was performed based on pilot data suggesting a 15% difference in early extubation rates between the groups (70% vs 55%) and assuming a 1:1.4 allocation ratio, 80% power, and 5% two-sided alpha error, a minimum sample size of 96 in the restrictive group and 134 in the conventional group was calculated using a two-proportion comparison formula. A total of 106 and 146 patients were ultimately included. Baseline characteristics, including age, gender, weight, body mass index (BMI), Child Turcotte Pugh (CTP), and Model for End Stage Liver Disease Sodium (MELD-Na) scores, were matched between the groups to ensure comparability.
RESULTS
This case-control analysis included 106 patients in the restrictive fluid group (cases) and 148 patients in the conventional fluid group (controls). Baseline characteristics, including age, gender, weight, BMI, CTP, and MELD-Na scores, were comparable between the groups. The mean age of cases and controls was 44.59 ± 10.64 years and 44.02 ± 10.46 years, respectively (P = 0.668). Males predominated in both groups (cases: 66.0%; controls: 75.3%), whereas females accounted for 34.0% and 24.7%, respectively (P = 0.105). The mean weight was 74.10 ± 15.46 kg in cases and 71.85 ± 15.14 kg in controls (P = 0.247), and mean BMI was 25.84 ± 4.97 and 24.80 ± 4.81 kg/m² in cases and controls, respectively (P = 0.094) [Table 1]. The mean CTP score was 9.52 ± 2.34 in cases and 10.07 ± 2.47 in controls (P = 0.075). The mean MELD-Na scores were 18.12 ± 6.08 in cases and 19.64 ± 7.53 in controls (P = 0.087) [Table 1]. Diabetes mellitus was significantly more common among controls (16.00%) compared to cases (6.60%) (P = 0.038). The incidence of all other comorbidities was comparable between the groups [Table 1].
Table 1.
Baseline characteristics of the study population
| Parameter | Restrictive (n=106) | Conventional (n=148) | P |
|---|---|---|---|
| Age (years) | 44.59±10.64 | 44.02±10.46 | 0.668 |
| Male (%) | 66.0% | 75.3% | 0.105 |
| Female (%) | 34.0% | 24.7% | |
| BMI (kg/m2) | 25.84±4.97 | 24.80±4.81 | 0.094 |
| MELD Na | 18.12±6.08 | 19.64±7.53 | 0.087 |
| CTP Score | 9.52±2.34 | 10.07±2.47 | 0.075 |
|
Comorbidities | |||
| Nil | 61 (57.5%) | 98 (65.3%) | 0.21 |
| HTN | 12 (11.32%) | 10 (6.67%) | 0.279 |
| DM | 7 (6.60%) | 24 (16%) | 0.038* |
| CAD - Critical/Non-Critical | 5 (4.72%) | 5 (3.33%) | 0.814 |
| Smoker | 7 (6.60%) | 5 (3.33%) | 0.358 |
| COPD | 0 (0.0) | 1 (0.7%) | 1.000 |
| Bronchial Asthma | 3 (2.8%) | 0 (0.0%) | 0.070 |
| Psoriasis | 1 (0.9%) | 0 (0.0%) | 0.414 |
| Sarcoidosis | 1 (0.9%) | 1 (0.7%) | 1.000 |
| Hypothyroidism | 6 (5.7%) | 5 (3.3%) | 0.371 |
| HOCM | 1 (0.9%) | 0 (0.0%) | 0.414 |
| Obesity (BMI >30 kg/m2) | 3 (2.8%) | 2 (1.3%) | 0.652 |
| Tobacco | 1 (0.9%) | 1 (0.7%) | 1.000 |
| Solitary Kidney | 1 (0.9%) | 2 (1.3%) | 1.000 |
| H/O Pulmonary Koch’s | 2 (1.9%) | 3 (2.0%) | 1.000 |
| Hyperthyroidism | 1 (0.9%) | 0 (0.0%) | 0.414 |
| Seizure disorder | 0 (0.0%) | 1 (0.7%) | 1.000 |
| Pulmonary Hypertension | 0 (0.0%) | 1 (0.7%) | 1.000 |
| Sjogren Syndrome | 0 (0.0%) | 1 (0.7%) | 1.000 |
| Ulcerative Colitis | 0 (0.0%) | 1 (0.7%) | 1.000 |
| Cerebrovascular accident | 0 (0.0%) | 0 (0.0%) | 1.000 |
| Patent Foramen Ovale | 1 (0.9%) | 0 (0.0%) | 0.414 |
| Retransplant | 1 (0.9%) | 0 (0.0%) | 0.414 |
BMI: Body mass index; CTP: Child Turcotte Pugh; MELD Na: Model for end stage liver disease Sodium; DM: Diabetes mellitus; HTN: Hypertension; CAD: Coronary artery disease; COPD: Chronic obstructive pulmonary disease; HOCM: Hypertrophic obstructive cardiomyopathy; n: Number of patients; H/O: History of. *Bold P<0.05 signifies statistically significant parameter
Total fluid administration was significantly lower in cases (3284.20 ± 833.00 ml) than in controls (3978.66 ± 1524.38 ml; P < 0.001). Fluids per kg per hour were also lower in cases (5.03 ± 1.01) versus controls (5.80 ± 2.40, P = 0.002). Mean total albumin use was higher in cases (419.86 ± 120.97) compared to controls (381.01 ± 102.93) (P = 0.006) [Figure 2,Table 2].
Figure 2.
Comparison of parameters between the two groups. FFP: Fresh frozen plasma; SDP: Single donor platelets; PRBC: Packed red blood cells; GRWR: Graft recipient weight ratio
Table 2.
Comparison of perioperative parameters between two groups
| Parameter | Cases |
Controls |
P | ||
|---|---|---|---|---|---|
| Mean±SD | Median [IQR] | Mean±SD | Median [IQR] | ||
| Ascites (ml) | 4548.57±4179.27 | 3500 [1000 to 7000] | 4729.90±3411.44 | 4000 [2000 to 7000] | 0.758 |
| Pleural Effusion (ml) | 1466.67±833.76 | 1000 [850 to 2300] | 2440.00±2283.73 | 1600 [1400 to 2500] | 0.174 |
| Total Fluids (ml) | 3284.20±833.00 | 3225 [2700 to 3600] | 3978.66±1524.38 | 3500 [3000 to 4600] | <0.001* |
| Total Albumin (ml) | 419.86±120.97 | 400 [350 to 500] | 381.01±102.93 | 370 [300 to 450] | 0.006* |
| Total Fluids (ml/kg/h) | 5.03±1.01 | 5 [4 to 6] | 5.80±2.40 | 5 [4 to 7] | 0.002* |
| PRBC (units) | 2.46±2.60 | 2 [0 to 4] | 2.99±3.02 | 2 [1 to 4] | 0.149# |
| FFP (units) | 1.71±2.02 | 2 [0 to 3] | 2.15±2.50 | 2 [0 to 4] | 0.136# |
| Cryoprecipitate | 0.55±1.62 | 0 [0 to 0] | 0.25±1.09 | 0 [0 to 0] | 0.085# |
| Platelets apheresis | 0.02±0.13 | 0 [0 to 0] | 0.13±0.55 | 0 [0 to 0] | 0.044* # |
| Intra-operative Urine Output (ml) | 854.95±396.44 | 750 [600 to 1000] | 899.59±591.36 | 800 [550 to 1100] | 0.502# |
| Inj. Noradrenaline (μg/kg/min) | 0.09±0.07 | 0 [0 to 0] | 0.10±0.14 | 0 [0 to 0] | 0.524# |
| Vasopressin (total units) | 5.41±5.79 | 3.30[0.72 to 9.25] | 1.71±2.98 | 0[0 to 2.40] | <0.001* |
| Cold Ischaemia (min) | 40.09±11.58 | 37 [33 to 45] | 48.27±16.65 | 46 [37 to 55] | <0.001* |
| GRWR | 1.14±0.28 | 1 [1 to 1] | 1.15±0.31 | 1 [1 to 1] | 0.873 |
| Last Lactate | 5.06±2.28 | 5 [3 to 7] | 4.47±2.13 | 4 [3 to 6] | 0.041 |
| Total Duration of Surgery (min) | 483.07±86.74 | 473 [424 to 535] | 606.55±128.47 | 591 [525 to 688] | <0.001 * |
GRWR: Graft recipient weight ratio; FFP: Fresh frozen plasma; PRBC: Packed red blood cells; IQR: Interquartile range; Inj: Injection; SD:Standard deviation. *Bold P<0.05 signifies statistically significant parameter
Ascites, pleural effusion, FFP and cryoprecipitate transfusion, intra-operative urine output, and graft-to-recipient weight ratio (GRWR) were comparable between the groups (P > 0.05). Surgical duration was shorter in cases (483.07 ± 86.74 min) versus controls (606.55 ± 128.47 min, P < 0.001). The cold ischaemia time was also lower in cases (40.09 ± 11.58 min) than in controls (48.27 ± 16.65 min; P < 0.001). Lactate levels were higher in cases (5.06 ± 2.28) than in controls (4.47 ± 2.13, P = 0.041) [Figure 2,Table 2].
The immediate extubation rate was significantly higher in cases (83.0%) than in controls (71.3%; P = 0.031) [Table 3]. Patients extubated within 6 hours were comparable between the groups (16.7% in cases and 16.3% in controls). The mean duration of mechanical ventilation was also comparable between the groups (10.30 ± 4.43 hours in cases and 10.59 ± 7.84 hours in controls) (P = 0.071) [Table 3]. Only one patient in the study group required post-operative intercostal chest drain (ICD) for massive pleural effusion.
Table 3.
Pulmonary outcomes between two groups
| Group |
P | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cases |
Controls |
|||||||||
| Count | % | Count | % | |||||||
| Immediate Extubation | No | 18 | 17.0% | 43 | 28.7% | 0.031* | ||||
| Yes | 88 | 83.0% | 107 | 71.3% | ||||||
| Delayed extubation | <6 h | 3 | 16.7% | 7 | 16.3% | 0.970 | ||||
| >6 h | 15 | 83.3% | 36 | 83.7% | ||||||
|
| ||||||||||
| Group | Count | Mean | SD | Median (IQR) | ||||||
|
| ||||||||||
| Duration of Mechanical Ventilation (hours) | Cases | 18 | 10.30 | 4.43 | 11 (10-14) | 0.071 | ||||
| Controls | 43 | 10.59 | 7.84 | 9 (8-12) | ||||||
| Reasons for delayed extubation | Cases | • Poor respiratory effort (50%) – most common • High lactates with acidosis (22.2%), • Sarcopenia (11.1%), • Surgical reasons (11.1%) • High inotropes (5.6%). |
||||||||
| Post-operative respiratory support (Number of patients) | Cases | • HFNO - 07 • NIV- 03 • Oxygen support (Face mask/Nasal prongs beyond 72 hours) - 25 |
||||||||
| Pulmonary complication (Number of patients) | Cases | • Need for ICD - 01 | ||||||||
HFNO: High-flow nasal oxygen; NIV: Non-invasive ventilation; ICD: Intercostal chest drain; SD: Standard deviation; *Bold P<0.05 signifies statistically significant parameter
AKI was defined using KDIGO criteria, based solely on changes in serum creatinine. Post-operative creatinine was evaluated on PODs 0, 1, and 3 (pre-operative value as baseline). Baseline creatinine values were identical between groups (0.74 ± 0.20; median 0.70, P = 0.791). On POD0, mean creatinine was comparable (cases 0.82 ± 0.23 vs. controls 0.83 ± 0.25, P = 0.718). On POD1, mean creatinine was significantly lower in cases (0.73 ± 0.25) than in controls (0.83 ± 0.31, P = 0.005), a trend that persisted on POD3 (0.62 ± 0.23 vs. 0.75 ± 0.35, P = 0.001) [Table 4]. Incidence of AKI: On POD1, AKI occurred in 15.1% of cases versus 24.5% of controls, a difference that did not reach statistical significance (P = 0.068). By POD3, no case patients developed AKI (0.0%) compared with 13.6% of controls (P < 0.001) [Table 4]. No patient required renal replacement therapy.
Table 4.
Renal outcomes between two groups
| Serum Creatinine | Group |
||||||
|---|---|---|---|---|---|---|---|
| Cases |
Controls |
P | |||||
| Mean | SD | Median | Mean | SD | Median | ||
| Baseline | 0.74 | 0.20 | 0.70 | 0.74 | 0.20 | 0.70 | 0.791 |
| POD0 | 0.82 | 0.23 | 0.80 | 0.83 | 0.25 | 0.80 | 0.718 |
| POD1 | 0.73 | 0.25 | 0.70 | 0.83 | 0.31 | 0.80 | 0.005* |
| POD3 | 0.62 | 0.23 | 0.60 | 0.75 | 0.35 | 0.70 | 0.001* |
|
| |||||||
| AKI |
Cases
|
Control
|
|||||
| Count | % | Count | % | ||||
|
| |||||||
| POD1 | Yes | 16 | 15.1% | 36 | 24.5% | 0.068 | |
| No | 90 | 84.9% | 112 | 75.6% | |||
| POD3 | Yes | 0 | 0.0% | 20 | 13.6% | <0.001* | |
| No | 106 | 100% | 128 | 86.4% | |||
*Bold P<0.05 signifies statistically significant parameter. SD: Standard deviation; POD: Post-operative day
66.9% of cases (n = 71) required PRBC transfusion (2.46 ± 2.60 units) versus 77.02% of controls (n = 114; 2.99 ± 3.02 units), though the difference was statistically insignificant (P = 0.149). Apheresis platelet use was significantly lower in cases (0.02 ± 0.13) than in controls (0.13 ± 0.55) (P = 0.044). 95% cases required noradrenaline infusion (0.09 ± 0.07 µg/kg/min) versus 0.10 + 0.14 µg/kg/min in controls, statistically non-significant (P = 0.524). Vasopressin was infused in 83% cases with a mean of 5.41 ± 5.79 units versus 1.71 ± 2.98 units in controls (P < 0.001) [Figure 2, Table 2].
Among the vascular complications in the case group, only one patient required a redo of the hepatic artery anastomosis. Nineteen patients were initiated on heparin infusion after arterial anastomosis based on TEG values.
DISCUSSION
Fluid management in liver transplant is often complex owing to the multitude of patients and surgery-related factors. Traditionally, liberal fluid administration was standard practice to compensate for anaesthesia-induced vasodilation, third-space losses (especially ascites), and intra-operative haemodynamic instability. However, with improved understanding of the pathophysiology of CLD, refined surgical techniques, and the widespread adoption of enhanced recovery after surgery (ERAS) guidelines,[5] liberal fluid administration has given way to a restrictive fluid strategy.
In CLD, central hypovolaemia persists despite fluid resuscitation due to splanchnic vasodilatation. Therefore, increased fluid loading may not necessarily achieve haemodynamic stability; rather, it can paradoxically worsen portal hypertension, thereby increasing blood loss, especially during the dissection phase.[6] In the neo-hepatic phase, fluid overload can lead to graft congestion.[7] Fluid overload has been associated with increased post-operative pulmonary complications,[8,9] prolonged mechanical ventilation, and higher re-intubation rates. Bozbas et al.[10] demonstrated that pulmonary complications significantly delayed extubation and prolonged post-anaesthesia care unit stay.
Different restricted fluid strategies have been employed over the years – fixed weight-based protocols, hypovolaemic phlebotomy,[11] low central venous pressure (CVP),[12,13] or net zero balance (especially during the dissection phase) – which have shown beneficial effects on pulmonary outcomes and blood transfusion requirement without increasing the risk of renal dysfunction. However, only a few studies have defined weight-based restrictive fluid regimens in liver transplant,[14,15,16] particularly in the context of LDLT.[4]
Reydellet et al.[14] described a crystalloid infusion rate of 4–6 ml/kg/h, with colloid boluses for fluid challenges, and reported improved post-operative outcomes. Sahmeddinni et al.[15] used 5 ml/kg/h and observed improved pulmonary outcomes without increased renal dysfunction compared with a 10 ml/kg/h regimen. Our findings align with these studies as restricted fluid therapy in our cohort was associated with improved post-operative pulmonary outcomes, with an immediate extubation rate of 83% and no re-intubations. Delayed extubation patients had a mean ventilation duration of 10.3 ± 4.43 hours.
Post-operative renal dysfunction is one of the major concerns with fluid restriction, carrying significant implications for graft and patient outcomes.[17,18,19,20] While intra-operative factors such as vasopressor use, blood product transfusions, and reperfusion injury contribute to renal dysfunction,[21,22] we observed that a moderately restrictive strategy did not increase the incidence of AKI. On POD1, AKI occurred in 15.1% of cases versus 24.5% of controls, and by POD3, it had declined to 0% in cases and 13.6% in controls (P < 0.001). Notably, no patients required renal replacement therapy. Our findings corroborate those of Morkane et al.,[6] who reported that euvolaemic or moderately restrictive fluid strategies in deceased-donor liver transplantation (DDLT) reduced transfusions and facilitated early extubation without increasing the risk of AKI. The higher incidence of AKI in controls, despite a non-restrictive fluid strategy, reinforces our conclusion that renal dysfunction in LDLT results from a complex interplay of multiple perioperative factors and is not solely attributable to fluid restriction.
LDLT presents unique challenges, including prolonged surgical duration, partial graft, and complex vascular/biliary anastomosis. Unlike fixed-volume protocols, our institutional approach combined continuous maintenance crystalloid/colloid infusion (4 ml/kg/h) with goal-directed boluses, adapting to the dynamic surgical phases while maintaining intra-vascular volume stability and compensating for the ongoing losses, thereby minimising complications associated with strict fluid restriction while preserving its benefits such as reduced portal hypertension and lower PRBC requirements (mean 2.46 ± 2.60 units in cases vs 2.99 ± 3.02 in controls, though statistically non-significant, P = 0.149).[23,24]
CLD is characterised by reduced SVR; therefore, stable haemodynamics require an optimum balance of fluids and vasopressors. Excessive fluid restriction may be accompanied by an increased use of vasopressors and a sustained state of extreme low perfusion, making the patient prone to vascular complications (ischaemic or thrombotic).[25] Our protocol maintained haemodynamic stability, with comparable noradrenaline requirements between groups (0.09 ± 0.07 vs 0.10 ± 0.14 µg/kg/min), and only one case required arterial re-anastomosis. However, with a change in practice, vasopressin was often introduced as a second vasopressor even before noradrenaline had reached its maximal dose, particularly to manage post-reperfusion portal hyperperfusion, which could explain the significantly increased vasopressin use observed in cases, while data on vascular complications in controls were incomplete, preventing statistical analysis.
Metabolic acidosis, particularly common during the anhepatic phase, may be compounded by fluid restriction. Our continuous basal infusion helped minimise acid–base disturbances due to hypovolaemia; in our study, 65 patients required sodium bicarbonate infusion (mean 70.6 ± 80.92 ml). Additionally, while fluid overload can cause bowel oedema[8] and delay gut recovery,[25,26] heterogeneous biliary reconstruction in our cohort limited the assessment of gut-specific outcomes.
Our case-control analysis supports the feasibility of a weight-based restrictive fluid strategy in LDLT, demonstrating improved pulmonary outcomes without increased renal or vascular risks. These results align with trials supporting restrictive strategies in abdominal surgery while providing LDLT-specific insights.
However, as a historical cohort study, potential confounders (e.g. surgical experience, ischaemia time, and ICU protocols) cannot be excluded. Prospective randomised controlled trials are, therefore, needed to validate our findings. Also, the impact of fluid restriction on cardiac complications, bowel recovery, ICU and hospital stay, and the optimal fluid transfusion rate for improved outcomes warrants further investigation.
CONCLUSION
This study provides substantial evidence that a weight-based, goal-directed, restrictive fluid strategy in LDLT can yield superior clinical outcomes without compromising intra-operative stability or increasing complications. However, limitations remain due to the retrospective design and use of historical controls, with potential changes in surgical duration, technique, or perioperative care over the study period. Nonetheless, the consistent and favourable trends across renal, respiratory, and haemodynamic parameters highlight the clinical viability of this approach in liver transplantation.
Presentation at conferences/CMEs and abstract publication
The manuscript in its present form has not been presented at any conference/CMEs and has not been previously published.
Statement on data sharing
De-identified data may be requested with reasonable justification from the authors (email to the corresponding author) and shall be shared after approval as per the authors’ Institution policy.
Disclosure of use of artificial intelligence (AI)-assistive or generative tools
AI tools were solely used for the purpose of language/grammar correction. No AI assisted tools were employed for any scientific analysis or interpretation.
Declaration of use of permitted tools
The tables, figures and charts used in this manuscript are not copyrighted.
Conflicts of interest
There are no conflicts of interest.
Acknowledgements
Authors would like to thank Dr. Mahesh V, (MBBS, MD, DNB. Associate professor, Department of Community Medicine, CIMS, Chamarajanagar) and Dr. Sarika Pandey (PhD Clinical research assistant, Department of HPB Surgery and Liver Transplantation) for their invaluable support for the statistical analysis. We would also like to thank Dr. Vinod Choudhary for his immense support with the conduction of this study.
Funding Statement
Nil.
REFERENCES
- 1.Varadhan KK, Lobo DN. A meta-analysis of randomised controlled trials of intravenous fluid therapy in major elective open abdominal surgery: Getting the balance right. Proc Nutr Soc. 2010;69:488–98. doi: 10.1017/S0029665110001734. [DOI] [PubMed] [Google Scholar]
- 2.Wilms H, Mittal A, Haydock MD, van den Heever M, Devaud M, Windsor JA. A systematic review of goal directed fluid therapy: Rating of evidence for goals and monitoring methods. J Crit Care. 2014;29:204–9. doi: 10.1016/j.jcrc.2013.10.019. [DOI] [PubMed] [Google Scholar]
- 3.Myles PS, Bellomo R, Corcoran T, Forbes A, Peyton P, Story D, et al. Restrictive versus liberal fluid therapy for major abdominal surgery. N Engl J Med. 2018;378:2263–74. doi: 10.1056/NEJMoa1801601. [DOI] [PubMed] [Google Scholar]
- 4.Carrier FM, Chassé M, Wang HT, Aslanian P, Bilodeau M, Turgeon AF. Effects of perioperative fluid management on postoperative outcomes in liver transplantation: A systematic review protocol. Syst Rev. 2018;7:180. doi: 10.1186/s13643-018-0841-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Brustia R, Monsel A, Skurzak S, Schiffer E, Carrier FM, Patrono D, et al. Guidelines for perioperative care for liver transplantation: Enhanced recovery after surgery (ERAS) recommendations. Transplantation. 2022;106:552–61. doi: 10.1097/TP.0000000000003808. [DOI] [PubMed] [Google Scholar]
- 6.Morkane CM, Sapisochin G, Mukhtar AM, Reyntjens KMEM, Wagener G, Spiro M, et al. Perioperative fluid management and outcomes in adult deceased donor liver transplantation-A systematic review of the literature and expert panel recommendations. Clin Transplant. 2022;36:e14651. doi: 10.1111/ctr.14651. [DOI] [PubMed] [Google Scholar]
- 7.Brezeanu LN, Brezeanu RC, Diculescu M, Droc G. Anaesthesia for liver transplantation: An update. J Crit Care Med (Targu Mures) 2020;6:91–100. doi: 10.2478/jccm-2020-0011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Holte K, Sharrock NE, Kehlet H. Pathophysiology and clinical implications of perioperative fluid excess. Br J Anaesth. 2002;89:622–32. doi: 10.1093/bja/aef220. [DOI] [PubMed] [Google Scholar]
- 9.Hong SK, Hwang S, Lee SG, Lee LS, Ahn CS, Kim KH, et al. Pulmonary complications following adult liver transplantation. Transplant Proc. 2006;38:2979–81. doi: 10.1016/j.transproceed.2006.08.090. [DOI] [PubMed] [Google Scholar]
- 10.Bozbas SS, Eyuboglu FO, Ozturk Ergur F, Gullu Arslan N, Sevmis S, Karakayali H, et al. Pulmonary complications and mortality after liver transplant. Exp Clin Transplant. 2008;6:264–70. [PubMed] [Google Scholar]
- 11.Massicotte L, Hevesi Z, Zaouter C, Thibeault L, Karakiewicz P, Roy L, et al. Association of phlebotomy on blood product transfusion requirements during liver transplantation: An observational cohort study on 1000 cases. Transplant Direct. 2022;8:e1258. doi: 10.1097/TXD.0000000000001258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang B, He HK, Cheng B, Wei K, Min S. Effect of low central venous pressure on postoperative pulmonary complications in patients undergoing liver transplantation. Surg Today. 2013;43:777–81. doi: 10.1007/s00595-012-0419-y. [DOI] [PubMed] [Google Scholar]
- 13.Fayed NA, Yassen KA, Abdulla AR. Comparison between 2 strategies of fluid management on blood loss and transfusion requirements during liver transplantation. J Cardiothorac Vasc Anesth. 2017;31:1741–50. doi: 10.1053/j.jvca.2017.02.177. [DOI] [PubMed] [Google Scholar]
- 14.Reydellet L, Blasco V, Mercier MF, Antonini F, Nafati C, Harti-Souab K, et al. Impact of a goal-directed therapy protocol on postoperative fluid balance in patients undergoing liver transplantation: A retrospective study. Ann Fr Anesth Reanim. 2014;33:e47–54. doi: 10.1016/j.annfar.2013.12.016. [DOI] [PubMed] [Google Scholar]
- 15.Sahmeddini MA, Janatmakan F, Khosravi MB, Ghaffaripour S, Eghbal MH, Nickeghbalian S, et al. Restricted crystalloid fluid therapy during orthotopic liver transplant surgery and its effect on respiratory and renal insufficiency in the early post-operative period: A randomized clinical trial. Int J Organ Transplant Med. 2014;5:113–9. [PMC free article] [PubMed] [Google Scholar]
- 16.Jiang GQ, Chen P, Bai DS, Tan JW, Su H, Peng MH. Individualized peri-operative fluid therapy facilitating early-phase recovery after liver transplantation. World J Gastroenterol. 2012;18:1981–6. doi: 10.3748/wjg.v18.i16.1981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Thongprayoon C, Kaewput W, Thamcharoen N, Bathini T, Watthanasuntorn K, Lertjitbanjong P, et al. Incidence and impact of acute kidney injury after liver transplantation: A meta-analysis. J Clin Med. 2019;8:372. doi: 10.3390/jcm8030372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim WH, Oh HW, Yang SM, Yu JH, Lee HC, Jung CW, et al. Intraoperative hemodynamic parameters and acute kidney injury after living donor liver transplantation. Transplantation. 2019;103:1877–86. doi: 10.1097/TP.0000000000002584. [DOI] [PubMed] [Google Scholar]
- 19.Hilmi IA, Damian D, Al-Khafaji A, Planinsic R, Boucek C, Sakai T, et al. Acute kidney injury following orthotopic liver transplantation: Incidence, risk factors, and effects on patient and graft outcomes. Br J Anaesth. 2015;114:919–26. doi: 10.1093/bja/aeu556. [DOI] [PubMed] [Google Scholar]
- 20.Feldkamp T, Bienholz A, Paul A, Saner FH. Renal damage after liver transplantation. Biosci Rep. 2020;40:BSR20191187. doi: 10.1042/BSR20191187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Goren O, Matot I. Perioperative acute kidney injury. Br J Anaesth. 2015;115(Suppl 2):ii3–14. doi: 10.1093/bja/aev380. [DOI] [PubMed] [Google Scholar]
- 22.Durand F, Francoz C, Asrani SK, Khemichian S, Pham TA, Sung RS, et al. Acute kidney injury after liver transplantation. Transplantation. 2018;102:1636–49. doi: 10.1097/TP.0000000000002305. [DOI] [PubMed] [Google Scholar]
- 23.Lekerika N, Gutiérrez Rico RM, Arco Vázquez J, Prieto Molano L, Arana-Arri E, Martínez Indart L, et al. Predicting fluid responsiveness in patients undergoing orthotopic liver transplantation: Effects on intraoperative blood transfusion and postoperative complications. Transplant Proc. 2014;46:3087–91. doi: 10.1016/j.transproceed.2014.10.005. [DOI] [PubMed] [Google Scholar]
- 24.Wu G, Chen T, Chen Z. Effect of controlled low central venous pressure technique on postoperative hepatic insufficiency in patients undergoing a major hepatic resection. Am J Transl Res. 2021;13:8286–93. [PMC free article] [PubMed] [Google Scholar]
- 25.Gautam PL. Fluid therapy in liver transplant. In: Vohra V, Gupta N, Jolly AS, Bhalotra S, editors. Peri-Operative Anesthetic Management in Liver Transplantation. Singapore: Springer; 2023. pp. 243–49. [Google Scholar]
- 26.Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, Allison SP. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: A randomised controlled trial. Lancet. 2002;359:1812–8. doi: 10.1016/S0140-6736(02)08711-1. [DOI] [PubMed] [Google Scholar]