Abstract
Background –
Immediate extubation (IE) following pediatric liver transplantation is being increasingly performed. The aim of this study was to characterize the rate of IE at our institution and identify recipient factors predictive of IE.
Methods –
All pediatric liver transplants performed at our institution between 1/1/2015 and 12/31/2020 were reviewed. Retransplants and multi-organ transplants were excluded. IE was defined as extubation in the operating room following transplant. Backward stepwise logistic regression at a p-value threshold of 0.05 was performed to identify variables associated with IE.
Results –
IE was achieved in 58 of the 81 (72%) pediatric liver transplants. The IE cohort had significantly shorter ICU length of stay and overall hospital length of stay, though IE was not an independent predictor of post-transplant length of stay. Age < 2 years, pre-operative mechanical ventilation, and total intra-operative epinephrine and dopamine infusion requirements were significant, independent risk factors against IE. This multivariable model was highly predictive of IE (area under the curve = 0.89).
Conclusions –
We describe the highest rate of immediate extubation post-pediatric liver transplantation that has been reported to date and identified significant risk factors against successful IE.
Introduction:
Historically, all liver transplant recipients remained intubated and mechanically ventilated in the immediate post-operative period. The University of Colorado was one of the first centers to report successful immediate extubation (IE) of adult liver transplant recipients in 1997 (1). Since then, IE has been widely adopted at adult liver transplant centers across the world, with IE rates ranging from 24–80% (2–5). Benefits of IE in adult recipients have included short intensive care unit (ICU) and overall hospital length of stay, lower risk of pulmonary complications, and lower cost.
While immediate extubation (IE) has become routine for select adult liver transplant recipients, this has not yet become the standard of care in pediatric liver transplantation. Only a few pediatric liver transplant centers have reported their experience with IE in this unique patient population, with the rates of IE ranging from 30%−65% (6–11). Given the limited number of reported pediatric liver transplant recipients with IE, the outcomes of IE and predictors of successful IE in children undergoing liver transplantation are not well known. The aim of this study was to describe the rate and outcomes of IE at our pediatric liver transplant institution, and determine recipient, donor, and operative characteristics associated with IE.
Methods:
This study received exemption from institutional board review (Exemption #20-0162). All pediatric (< 18 years old) liver transplants performed at our institution between 1/1/2015 and 12/31/2020 were retrospectively reviewed. Multi-organ and re-transplants were excluded. Recipients were categorized as either immediate extubation (IE) or delayed extubation (DE) based on timing and location of post-operative extubation. IE was defined as extubation in the operating room while DE was defined as extubation in the intensive care unit (ICU). Decisions at our institution regarding whether or not to pursue IE is individualized in each case, and is a shared decision made between the anesthesia, surgical, medical, and critical care teams. Post-operatively, all recipients are transferred to the ICU for post-operative management, irrespective of extubation timing.
Categorical data are presented as quantity (percentage) and compared using Chi-square test. Continuous data are presented as mean (standard deviation), unless otherwise specified, and compared using two-tailed Student’s t-test or analysis of variance, as appropriate. Logistic regression was used to determine which recipient, donor, and operative variables were significantly associated with IE. Linear regression was used to determine the association between IE and post-transplant length of stay. Variables significant upon univariable regression were entered into a backward stepwise regression, selecting for variables with p < 0.05. P-value of < 0.05 was used as the threshold for statistical significance for all comparisons. All statistical analysis was performed using STATA 16.1 (StataCorp, College Station, TX, USA).
Results:
Study Cohort
58 of the 81 (72%) pediatric liver transplant recipients transplanted during the study period were immediately extubated (IE) in the operating room prior to transfer to the ICU. An additional 7 recipients were extubated in the ICU within 24 hours of leaving the operating room. No reintubations occurred within 48 hours in the IE cohort. Excluding reoperations, only one recipient in the IE cohort was reintubated on post-operative day (POD) 7 for hypoxemia in the setting of refractory very severe hepatopulmonary syndrome (12). Of the 23 recipients with delayed extubated (DE), 20 were not immediately extubated due to their clinical status, 2 had failed extubation attempts in the OR, and 1 did not have documentation of why IE did not occur. These recipients remained intubated for an average of 4.9 days (± 6.7 days) and a median of 2 days (range 1–21 days). Excluding reoperations, only one patient in DE cohort was reintubated. This recipient was extubated on POD1 but was reintubated the same day due to respiratory failure and was subsequently found to have pneumonia. This recipient also developed fungal sepsis and remained intubated for 11 days
Recipient, Donor, and Operative Characteristics by Extubation Timing
The characteristics of the IE and DE cohorts are summarized and compared in Table 1a–c. IE recipients were less commonly younger than 2 years of age, less commonly had acute liver failure, more commonly were non-hospitalized prior to transplant, had lower match model for end-stage liver disease (MELD)/pediatric end-stage liver disease (PELD) scores, were less likely to be mechanically ventilated pre-transplant, and less frequently had encephalopathy at time of transplant (Table 1a). IE recipients were also more likely to receive a whole liver graft and have a female donor (Table 1b). Lastly, IE recipients received less volume per kg of packed red blood cells (pRBC), fresh frozen plasma (FFP), platelets, 25% albumin, 5% albumin, epinephrine, and dopamine intra-operatively in comparison to DE recipients (Table 1c). Notably, there was no significant difference in operative time between the two groups. Those who experienced IE were also less likely to be on one or more vasopressors at the end of the operation than the DE cohort (52% vs. 74% in DE, p = 0.07).
Table 1a.
Recipient characteristics by timing of extubation
| Variable | Delayed extubation (n = 23) | Immediate extubation (n = 58) | P-value |
|---|---|---|---|
| Female | 11 (48%) | 28 (48%) | 1.0 |
| Non-Hispanic Caucasian | 13 (57%) | 35 (60%) | 0.8 |
| Age group | 0.003 | ||
| < 2 years | 18 (78%) | 21 (36%) | |
| 2–10 years | 2 (9%) | 19 (33%) | |
| > 10 years | 3 (13%) | 18 (31%) | |
| Weight < 30 kg | 20 (87%) | 42 (72%) | 0.2 |
| Diagnosis | 0.04 | ||
| Biliary atresia | 12 (52%) | 23 (40%) | |
| Acute liver failure | 5 (22%) | 3 (5%) | |
| Metabolic | 2 (9%) | 7 (12%) | |
| Other | 4 (17%) | 25 (43%) | |
| Prior abdominal surgery | 13 (56%) | 32 (55%) | 0.9 |
| Condition | 0.02 | ||
| Non-hospitalized, | 10 (43%) | 43 (74%) | |
| Hospitalized, non-ICU | 6 (26%) | 9 (16%) | |
| Hospitalized, ICU | 7 (30%) | 6 (10%) | |
| MELD/PELD | 0.04 | ||
| ≤ 30 | 6 (26%) | 28 (38%) | |
| > 30 | 8 (35%) | 18 (31%) | |
| Status 1B | 4 (17%) | 10 (17%) | |
| Status 1A | 5 (22%) | 2 (3%) | |
| Pre-operative mechanical ventilation | 6 (26%) | 1 (2%) | <0.001 |
| Encephalopathy | 4 (17%) | 2 (3%) | 0.03 |
| Ascites | 8 (35%) | 14 (24%) | 0.3 |
Table 1c.
Intra-operative characteristics by timing of recipient extubation
| Variable | Delayed extubation (n = 23) | Immediate extubation (n = 58) | P-value |
|---|---|---|---|
| Operative time, hours | 6.98 (2.36) | 6.75 (1.31) | 0.6 |
| Packed red blood cells, mL/kg | 74.55 (82.31) | 15.25 (23.10) | < 0.001 |
| Fresh frozen plasma, mL/kg | 19.88 (35.40) | 1.92 (6.02) | < 0.001 |
| Platelets, mL/kg | 9.14 (16.93) | 1.04 (5.47) | 0.002 |
| Cryoprecipitate, mL/kg | 0.77 (2.80) | 0.69 (4.25) | 0.9 |
| Albumin 25%, mL/kg | 1.45 (2.62) | 0.20 (1.07) | 0.003 |
| Albumin 5%, mL/kg | 30.61 (24.10) | 18.81 (14.42) | 0.008 |
| Crystalloid, mL/kg | 63.49 (60.82) | 53.85 (38.79) | 0.4 |
| Total norepinephrine, mcg/kg | 23.67 (49.47) | 18.37 (37.29) | 0.6 |
| Total epinephrine, mcg/kg | 19.14 (34.84) | 3.22 (8.21) | 0.002 |
| Total vasopressin, units/kg | 0.16 (0.40) | 0.07 (0.15) | 0.1 |
| Total dopamine, mcg/kg | 1174.11 (1587.35) | 456.16 (794.68) | 0.008 |
| Total phenylephrine, mcg/kg | 18.15 (35.29) | 25.39 (79.16) | 0.7 |
| ≥1 | 17 (74%) | 30 (52%) |
Table 1b.
Donor characteristics by timing of recipient extubation
| Variable | Delayed extubation (n = 23) | Immediate extubation (n = 58) | P-value |
|---|---|---|---|
| Female donor | 7 (30%) | 37 (64%) | 0.007 |
| Non-Hispanic Caucasian donor | 17 (74%) | 43 (74%) | 1.0 |
| Donor age group | 0.4 | ||
| < 3 years | 4 (17%) | 12 (21%) | |
| 3–18 years | 8 (35%) | 19 (33%) | |
| 19–30 years | 3 (13%) | 15 (26%) | |
| > 30 years | 8 (35%) | 12 (21%) | |
| Donor body mass index, kg/m2 | 20.38 (5.31) | 20.78 (3.88) | 0.7 |
| Cold ischemia, hours | 5.72 (3.30) | 5.38 (2.77) | 0.6 |
| Graft type | 0.005 | ||
| Deceased whole | 9 (39%) | 38 (66%) | |
| Deceased partial | 6 (26%) | 2 (3%) | |
| Living | 8 (35%) | 18 (31%) |
Predictors of Immediate Extubation
The following variables were significantly associated with IE upon univariable logistic regression: recipient age < 2, pre-transplant ICU hospitalization, status 1A, pre-operative mechanical ventilation, female donor, deceased donor partial graft, pRBC transfusion volume, FFP transfusion volume, platelet transfusion volume, albumin 25% and 5% volume, total epinephrine administered, and total dopamine administered intra-operatively (Table 2a). In the overall multivariable model, age < 2 years, pre-operative mechanical ventilation, and greater intra-operative epinephrine and dopamine infusion requirements were significant and independent risk factors against IE (Table 2b). This multivariable model was highly predictive of IE with an area under the curve (AUC) of 0.89.
Table 2a.
Univariable Logistic Regression for Immediate Extubation
| Variable | Reference | Odds Ratio | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|---|
| Recipient Factors: | |||||
| Female | Male | 1.02 | 0.39 | 2.68 | 1.0 |
| Non-Hispanic Caucasian | Other race/ethnicity | 1.17 | 0.44 | 3.11 | 0.8 |
| 2–10 years | 1.58 | 0.24 | 10.60 | 0.6 | |
| Weight < 30 kg | Weight > 30 kg | 0.39 | 0.10 | 1.51 | 0.2 |
| Other | 3.26 | 0.92 | 11.56 | 0.07 | |
| Previous abdominal surgery | 0.95 | 0.36 | 2.51 | 0.9 | |
| Hospitalized, ICU | 0.20 | 0.05 | 0.72 | 0.01 | |
| Status 1A | 0.09 | 0.01 | 0.55 | 0.01 | |
| Pre-operative mechanical ventilation | 0.05 | 0.01 | 0.44 | 0.007 | |
| Encephalopathy | 0.17 | 0.03 | 1.00 | 0.05 | |
| Ascites | 0.60 | 0.21 | 1.70 | 0.3 | |
| Donor Factors: | |||||
| Female | Male | 4.03 | 1.43 | 11.36 | 0.008 |
| Non-Hispanic Caucasian | Other race/ethnicity | 1.01 | 0.34 | 3.04 | 1.0 |
| 19–30 years | 3.33 | 0.72 | 15.37 | 0.1 | |
| BMI, kg/m2 | 1.02 | 0.91 | 1.15 | 0.7 | |
| Cold ischemia time, hours | 0.96 | 0.81 | 1.13 | 0.6 | |
| Living | 0.53 | 0.18 | 1.61 | 0.3 | |
| Intra-Operative Factors: | |||||
| Operative time, hours | 0.92 | 0.69 | 1.23 | 0.6 | |
| Packed red blood cells, mL/kg | 0.97 | 0.95 | 0.99 | < 0.001 | |
| Fresh frozen plasma, mL/kg | 0.94 | 0.90 | 0.99 | 0.01 | |
| Platelets, mL/kg | 0.92 | 0.85 | 0.98 | 0.02 | |
| Cryoprecipitate, mL/kg | 0.99 | 0.88 | 1.12 | 0.9 | |
| Albumin 25%, mL/kg | 0.67 | 0.49 | 0.93 | 0.02 | |
| Albumin 5%, mL/kg | 0.97 | 0.94 | 0.99 | 0.02 | |
| Crystalloid, mL/kg | 1.00 | 0.99 | 1.01 | 0.4 | |
| Total norepinephrine, mcg/kg | 1.00 | 0.99 | 1.01 | 0.6 | |
| Total epinephrine, mcg/kg | 0.95 | 0.91 | 0.99 | 0.03 | |
| Total vasopressin, units/kg | 0.25 | 0.04 | 1.63 | 0.1 | |
| Total dopamine, mcg/kg | 1.00 | 1.00 | 1.00 | 0.01 | |
| Total phenylephrine, mcg/kg | 1.00 | 0.99 | 1.01 | 0.7 | |
| ≥ 1 pressor at end of surgery | Off pressors at end of surgery | 0.38 | 0.13 | 1.10 | 0.07 |
Table 2b.
Multivariable Logistic Regression for Immediate Extubation
| Variable | Adjusted Odds Ratio | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|
| Age < 2 years | 0.06 | 0.01 | 0.41 | 0.003 |
| Pre-operative mechanical ventilation | 0.008 | 0.0004 | 0.15 | 0.001 |
| Total epinephrine, mcg/kg | 0.92 | 0.85 | 0.99 | 0.03 |
| Total dopamine, mcg/kg | 0.99 | 0.99 | 0.99 | 0.02 |
Post-hoc analysis aimed at understanding why deceased donor partial grafts and female donors were associated with lower and greater odds of IE, respectively, upon univariable analysis was performed. This analysis demonstrated that recipients of deceased donor partial grafts were more likely to be younger than 2 years of age (whole 32% vs. deceased partial 88% vs. living 65%, p = 0.002) and received a greater amount of dopamine infusion (whole 620 ± 902 mcg/kg vs. deceased partial 1786 ± 1940 mcg/kg vs. living 387 ± 980 mcg/kg, p = 0.006). Recipients of female donor grafts were significantly less likely to be on mechanical ventilation pre-operatively (0% vs. 19% of male donor recipients, p =0.003).
Clinical Outcomes
Seven deaths and 2 retransplants occurred in this cohort, with a median follow up time of 2.18 years (range 0 days – 5.82 years). One-year patient survival was 91% in both the IE and DE groups (Figure 1, p = 0.5). One-year graft survival was 87% in the IE cohort and 91% in the DE cohort (Figure 2, p = 0.3). There were no significant differences between the IE and DE cohorts in regards to the following complications: hepatic artery thrombosis (12% vs. 4%, p = 0.3), portal vein thrombosis (3% vs. 13%, p = 0.1), bile leak (10% vs. 9%, p = 0.8), reoperation (28% vs. 30%, p = 0.8), or ventilator-associated pneumonia (0% vs. 4%, p = 0.1). Excluding 5 recipients who died prior to transfer out of the intensive care unit (ICU), ICU length of stay was shorter (6 ± 16 days) in the IE group than the DE group (13 ± 15 days) (p = 0.09). Excluding 6 recipients who died prior to discharge, overall hospital length of stays was similar in the IE groups (16 ± 21 days) and DE group (22 ± 18 days) (p = 0.2). One recipient in IE group developed refractory very severe hepatopulmonary syndrome requiring an extremely prolonged course of extra-corporeal membrane oxygenation support (ECMO). This recipient had a 121 day ICU length of stay and 159 day overall hospital length of stay. When excluding this outlier in addition to those who had early in-ICU or in-hospital death and therefore artificially shorter length of stay, the IE cohort had significantly shorter ICU length of stay (4 ± 4 vs. 13 ± 15 days, p < 0.001) and overall hospital length of stay (13 ± 7 vs. 22 ± 18 days, p = 0.002), respectively.
Figure 1.
Recipient Survival By Timing of Extubation (p = 0.5)
Figure 2.
Graft Survival By Timing of Extubation (p = 0.3)
Predictors of ICU and Hospital Length of Stay
To assess if IE was associated with shorter ICU and overall hospital length of stay independent of other markers of disease severity, a multivariable linear regression was performed. The outlier described above and in-ICU or in-hospital mortalities were excluded from the regressions of ICU and overall hospital length of stay, respectively. IE was significantly negatively associated with ICU length of stay (B = −9.08, 95% CI −13.52 - −4.65, p < 0.001). The following variables were also found to be associated with ICU length of stay upon univariable regression: “other” diagnosis, pre-operative non-ICU and ICU hospitalization, status 1A, pre-operative mechanical ventilation, ascites, female donor, deceased donor partial graft, volumes of pRBC, FFP, platelets, and albumin 5%, and epinephrine, vasopressin, and dopamine total requirement (Table 3a). Upon multivariable linear regression, IE was no longer significantly associated with ICU length of stay (p > 0.05). Rather, pre-operative mechanical ventilation, deceased donor partial graft, and pRBC transfusion volume were found to be independent predictors of ICU length of stay (Table 3b).
Table 3a.
Univariable Linear Regression for Intensive Care Unit Length of Stay
| Variable | Reference | Coefficient | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|---|
| Immediate Extubation | Delayed Extubation | −9.08 | −13.52 | −4.65 | < 0.001 |
| Recipient Factors: | |||||
| Female | Male | 1.66 | −2.74 | 6.06 | 0.5 |
| Non-Hispanic Caucasian | Other race/ethnicity | 1.08 | −3.37 | 5.54 | 0.6 |
| 2–10 years | 4.77 | −0.35 | 9.90 | 0.07 | |
| Weight < 30 kg | Weight > 30 kg | 1.98 | −3.17 | 7.13 | 0.4 |
| Other | −5.76 | −10.53 | −1.00 | 0.02 | |
| Previous abdominal surgery | 1.74 | −2.71 | 6.18 | 0.4 | |
| Hospitalized, ICU | 6.94 | 1.37 | 12.50 | 0.02 | |
| Status 1A | 9.47 | 1.81 | 17.13 | 0.02 | |
| Pre-operative mechanical ventilation | 9.78 | 1.96 | 17.59 | 0.02 | |
| Encephalopathy | 3.25 | −4.85 | 11.35 | 0.4 | |
| Ascites | 5.99 | 1.12 | 10.87 | 0.02 | |
| Donor Factors: | |||||
| Female | Male | −5.50 | −9.73 | −1.26 | 0.01 |
| Non-Hispanic Caucasian | Other race/ethnicity | −0.62 | −5.53 | 4.30 | 0.8 |
| 19–30 years | 1.38 | −5.50 | 8.25 | 0.7 | |
| BMI, kg/m2 | −0.35 | −0.85 | 0.15 | 0.2 | |
| Cold ischemia time, hours | 0.72 | −0.02 | 1.46 | 0.06 | |
| Living | −0.96 | −4.61 | 2.69 | 0.6 | |
| Intra-Operative Factors: | |||||
| Operative time, hours | 0.42 | −1.01 | 1.86 | 0.6 | |
| Packed red blood cells, mL/kg | 0.10 | 0.06 | 0.14 | < 0.001 | |
| Fresh frozen plasma, mL/kg | 0.21 | 0.10 | 0.32 | < 0.001 | |
| Platelets, mL/kg | 0.48 | 0.19 | 0.77 | 0.001 | |
| Cryoprecipitate, mL/kg | 0.37 | −0.17 | 0.92 | 0.2 | |
| Albumin 25%, mL/kg | 1.07 | −0.21 | 2.35 | 0.1 | |
| Albumin 5%, mL/kg | −0.05 | −0.10 | 0.00 | 0.046 | |
| Crystalloid, mL/kg | 0.12 | −0.01 | 0.25 | 0.08 | |
| Total norepinephrine, mcg/kg | 0.03 | −0.04 | 0.09 | 0.4 | |
| Total epinephrine, mcg/kg | 0.18 | 0.08 | 0.28 | < 0.001 | |
| Total vasopressin, units/kg | 14.48 | 4.29 | 24.66 | 0.006 | |
| Total dopamine, mcg/kg | 0.00 | 0.00 | 0.01 | 0.001 | |
| Total phenylephrine, mcg/kg | 0.01 | −0.03 | 0.04 | 0.7 | |
| ≥ 1 pressor at end of surgery | Off pressors at end of surgery | 1.12 | −3.32 | 5.56 | 0.6 |
Table 3b.
Multivariable Linear Regression for Intensive Care Unit Length of Stay
| Variable | Adjusted Coefficient | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|
| Pre-operative mechanical ventilation | 7.98 | 2.51 | 13.46 | 0.005 |
| Deceased donor partial graft | 17.86 | 11.92 | 23.80 | < 0.001 |
| Packed red blood cells, mL/kg | 0.06 | 0.03 | 0.09 | < 0.001 |
Next, the linear regression was repeated for overall hospital length of stay. IE was negatively associated with hospital length of stay (B = −9.01, 95% CI −14.71 - −3.0, p = 0.002). Additionally, age < 2 years, “other” diagnosis, MELD/PELD > 30, ascites, female donor, deceased donor partial graft, and volume of pRBC, FFP, albumin, crystalloid, epinephrine, vasopressin, and dopamine required were also associated with hospital length of stay (Table 4a). In the multivariable model, IE was no longer independently associated with hospital length of stay (p > 0.05). The significant, independent predictors of hospital length of stay were age < 2 years, MELD/PELD > 30, deceased donor partial graft, vasopressin requirement, and dopamine requirement (Table 4b).
Table 4a.
Univariable Linear Regression for Overall Post-Transplant Hospital Length of Stay
| Variable | Reference | Coefficient | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|---|
| Immediate Extubation | Delayed Extubation | −9.01 | −14.71 | −3.30 | 0.002 |
| Recipient Factors: | |||||
| Female | Male | 1.84 | −3.55 | 7.22 | 0.5 |
| Non-Hispanic Caucasian | Other race/ethnicity | 0.10 | −5.38 | 5.57 | 1.0 |
| 2–10 years | 11.05 | 5.05 | 17.05 | < 0.001 | |
| Weight < 30 kg | Weight > 30 kg | 7.67 | 1.50 | 13.84 | 0.02 |
| Other | −8.50 | −14.29 | −2.70 | 0.005 | |
| Previous abdominal surgery | 4.30 | −1.06 | 9.66 | 0.1 | |
| Hospitalized, ICU | 5.18 | −2.05 | 12.41 | 0.2 | |
| Status 1A | 6.10 | −3.95 | 16.14 | 0.2 | |
| Pre-operative mechanical ventilation | 6.03 | −4.64 | 16.70 | 0.3 | |
| Encephalopathy | −3.62 | −14.35 | 7.11 | 0.5 | |
| Ascites | 7.02 | 1.16 | 12.88 | 0.02 | |
| Donor Factors: | |||||
| Female | Male | −5.39 | −10.67 | −0.12 | 0.045 |
| Non-Hispanic Caucasian | Other race/ethnicity | −1.60 | −7.58 | 4.38 | 0.6 |
| 19–30 years | 2.42 | −6.21 | 11.05 | 0.6 | |
| BMI, kg/m2 | −0.32 | −0.93 | 0.29 | 0.3 | |
| Cold ischemia time, hours | 0.73 | −0.18 | 1.63 | 0.1 | |
| Living | 1.83 | −2.78 | 6.43 | 0.4 | |
| Intra-Operative Factors: | |||||
| Operative time, hours | 0.88 | −0.86 | 2.63 | 0.3 | |
| Packed red blood cells, mL/kg | 0.13 | 0.08 | 0.17 | < 0.001 | |
| Fresh frozen plasma, mL/kg | 0.23 | 0.10 | 0.37 | 0.001 | |
| Platelets, mL/kg | 0.67 | 0.32 | 1.01 | < 0.001 | |
| Cryoprecipitate, mL/kg | 0.51 | −0.16 | 1.17 | 0.1 | |
| Albumin 25%, mL/kg | 1.58 | 0.04 | 3.12 | 0.04 | |
| Albumin 5%, mL/kg | −0.03 | −0.09 | 0.03 | 0.3 | |
| Crystalloid, mL/kg | 0.17 | 0.01 | 0.33 | 0.04 | |
| Total norepinephrine, mcg/kg | 0.07 | −0.01 | 0.14 | 0.08 | |
| Total epinephrine, mcg/kg | 0.23 | 0.11 | 0.35 | < 0.001 | |
| Total vasopressin, units/kg | 19.93 | 7.74 | 32.13 | 0.002 | |
| Total dopamine, mcg/kg | 0.00 | 0.00 | 0.01 | 0.001 | |
| Total phenylephrine, mcg/kg | 0.01 | −0.03 | 0.06 | 0.6 | |
| ≥ 1 pressor at end of surgery | Off pressors at end of surgery | 0.63 | −4.81 | 6.06 | 0.8 |
Table 4b.
Multivariable Linear Regression for Overall Post-Transplant Hospital Length of Stay
| Variable | Adjusted Coefficient | 95% Confidence Interval | P-Value | |
|---|---|---|---|---|
| Age < 2 years | 5.18 | 1.15 | 9.20 | 0.01 |
| MELD/PELD > 30 | 4.60 | 0.60 | 8.61 | 0.03 |
| Deceased donor partial graft | 17.97 | 10.08 | 25.85 | < 0.001 |
| Total vasopressin, units/kg | 12.77 | 3.29 | 22.24 | 0.009 |
| Total dopamine, mcg/kg | 0.002 | 0.0004 | 0.004 | 0.02 |
Discussion:
In this single center retrospective review of 81 pediatric liver transplant recipients, 72% of recipients were immediately extubated in the operating room. After excluding a single outlier placed on ECMO post-transplant, pediatric recipients with IE had a significantly shorter ICU and hospital length of stay, though IE was not independently associated with this outcome. Age < 2 years, pre-operative mechanical ventilation, and increasing intra-operative epinephrine and dopamine transfusion requirements were identified as significant risk factors against IE.
Our experience demonstrates that the majority of first-time, liver-only pediatric liver transplant recipients can be extubated immediately post-transplant. An IE rate of 72% is the highest reported percentage among pediatric liver transplant centers and is similar to the highest reported series of IE in adult liver transplants (5–11). IE can be safely performed in this patient population, as aside from for reoperations, only 1 recipient in the IE group required reintubation for respiratory failure on POD7 in the setting of refractory hepatopulmonary syndrome. Furthermore, the survival and complication rates were similar across recipients with IE and DE.
Though several previous studies have described baseline differences in recipient, donor, and intra-operative characteristics between pediatric liver transplant recipients with IE and DE, this is one of the first to identify independent risk factors against IE using a multivariable logistic regression model that accounts for confounding effects (6–11). Our data suggests that recipients who are younger than 2 years, had respiratory failure prior to their transplant, and/or required high amounts of pressor support intra-operatively are unlikely to be immediately extubated, while the majority of recipients without these risk factors can be safely and successfully extubated in the OR. These data may be used to develop standardized protocols for IE following pediatric liver transplantation and guide pre-operative family counseling regarding the likelihood of IE.
Interestingly, we found that recipients of deceased donor partial and female donor grafts were less likely to be extubated in the OR, which are two variables which perhaps would not be intuitively linked to DE. Further assessment, however, demonstrated that these associations were confounded by a greater proportion of recipients younger than 2 years and who required greater pressor support (specifically dopamine) among the deceased donor partial graft recipients, and a greater incidence of pre-operative mechanical ventilation among female donor graft recipients. Neither of these two variables were independently associated with extubation timing in the multivariable model.
Recipients who successfully underwent IE had a significantly shorter ICU and hospital length of stay in our series, as has been described in previous adult and pediatric reports (1–3, 6–9). By performing a multivariable linear regression for length of stay, however, we were able to demonstrate that IE itself is not independently associated with shorter ICU and hospital length of stay, but rather may be a proxy for lower disease severity and better clinical status. Specifically, pre-operative mechanical ventilation, pRBC transfusion volume, age < 2 years, dopamine and vasopressin infusion volume, MELD/PELD > 30 were all found to be associated with longer ICU or overall hospital length of stay, and all had a lower incidence among patients that were extubated in the OR. Additionally, deceased donor partial liver graft transplantation was associated with longer ICU and overall hospital length of stay, consistent with previous studies on predictors of hospital length of stay following pediatric liver transplantation (13, 14).
This study is limited by its retrospective nature, small number of recipients relative to adult liver transplant series, and relatively short follow-up time. Several of the variables in the regression models have wide confidence intervals and therefore should be validated in a larger, ideally multi-institutional series. Furthermore, there is considerable center-specific variation in anesthetic, surgical, and medical care of pediatric liver transplant recipients and decision regarding timing of extubation may be very center and team dependent. Therefore, the generalizability of our results to other centers is limited. However, our study adds to the limited number of studies describing implementation of IE in a high percentage of pediatric liver transplant recipients.
A majority of pediatric liver transplant recipients can be successfully extubated in the operating room. Recipient age younger than 2 years, pre-operative mechanical ventilation, and amount of intra-operative epinephrine and dopamine infused were risk factors against immediate extubation. Recipients that were immediately extubated had short ICU and overall hospital length of stay, though this was primarily driven by lower rates of pre-operative mechanical ventilation, deceased donor partial liver grafts, pRBC transfusion volume, pressor infusions, and MELD/PELD scores. Immediate extubation can be safely and effectively implemented in pediatric liver transplantation, similar to the current standard in adult liver transplantation.
Funding:
Supported by NIH/NCATS Colorado CTSA Grant Number TL1 TR002533. Contents are the authors’ sole responsibility and do not necessarily represent official NIH views.
Abbreviations:
- adjOR
adjusted odds ratio
- CI
confidence interval
- DE
delayed extubation
- ECMO
extra-corporeal membrane oxygenation support
- FFP
fresh frozen plasma
- ICU
intensive care unit
- IE
immediate extubation
- OR
odds ratio
- MELD
model for end-stage liver disease
- PELD
pediatric end-stage liver disease
- pRBC
packed red blood cells
Footnotes
Disclosure: The authors declare no conflicts of interest.
Data Availability:
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References:
- 1.Mandell MS, Lockrem J, Kelley SD. Immediate tracheal extubation after liver transplantation: experience of two transplant centers. Anesth Analg. 1997;84(2):249–53. [DOI] [PubMed] [Google Scholar]
- 2.Acho C, Morita Y, Fernandez V, Safwan M, Galusca D, Abouljoud M, et al. Immediate Postoperative Extubation Decreases Pulmonary Complications in Liver Transplant Patients. Transplantation. 2021;105(9):2018–28. [DOI] [PubMed] [Google Scholar]
- 3.Suphathamwit A, Pongraweewan O, Lakkam S, Tovikkai C. Predictive score for immediate extubation after liver transplantation. Clin Transplant. 2021;35(3):e14212. [DOI] [PubMed] [Google Scholar]
- 4.Chae MS, Kim JW, Jung JY, Choi HJ, Chung HS, Park CS, et al. Analysis of pre- and intraoperative clinical for successful operating room extubation after living donor liver transplantation: a retrospective observational cohort study. BMC Anesthesiol. 2019;19(1):112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Glanemann M, Busch T, Neuhaus P, Kaisers U. Fast tracking in liver transplantation. Immediate postoperative tracheal extubation: feasibility and clinical impact. Swiss Med Wkly. 2007;137(13–14):187–91. [DOI] [PubMed] [Google Scholar]
- 6.Gurnaney HG, Cook-Sather SD, Shaked A, Olthoff KM, Rand EB, Lingappan AM, et al. Extubation in the operating room after pediatric liver transplant: A retrospective cohort study. Paediatr Anaesth. 2018;28(2):174–8. [DOI] [PubMed] [Google Scholar]
- 7.Fullington NM, Cauley RP, Potanos KM, O’Melia L, Zurakowski D, Bae Kim H, et al. Immediate extubation after pediatric liver transplantation: a single-center experience. Liver Transpl. 2015;21(1):57–62. [DOI] [PubMed] [Google Scholar]
- 8.O’Meara ME, Whiteley SM, Sellors JM, Luntley JM, Davison S, McClean P, et al. Immediate extubation of children following liver transplantation is safe and may be beneficial. Transplantation. 2005;80(7):959–63. [DOI] [PubMed] [Google Scholar]
- 9.Sahinturk H, Ozdemirkan A, Yilmaz O, Zeyneloglu P, Torgay A, Pirat A, et al. Immediate Tracheal Extubation After Pediatric Liver Transplantation. Exp Clin Transplant. 2021;19(10):1063–8. [DOI] [PubMed] [Google Scholar]
- 10.Unlukaplan A, Torgay A, Pirat A, Arslan G, Haberal M. Predictors of Immediate Tracheal Extubation in the Operating Room After Pediatric Liver Transplantation. Exp Clin Transplant. 2020. [DOI] [PubMed] [Google Scholar]
- 11.Ulukaya S, Arikan C, Aydogdu S, Ayanoglu HO, Tokat Y. Immediate tracheal extubation of pediatric liver transplant recipients in the operating room. Pediatr Transplant. 2003;7(5):381–4. [DOI] [PubMed] [Google Scholar]
- 12.Huang J, Yoeli D, Sundaram SS, Carpenter T, Annam A, Pahlavan S, et al. Extracorporeal membrane oxygenation as rescue therapy in a pediatric liver transplant recipient with very severe hepatopulmonary syndrome. Pediatr Transplant. 2022;26(2):e14185. [DOI] [PubMed] [Google Scholar]
- 13.Zhang T, Barrett S, Cotton R, Galvan NTN, O’Mahony C, Moore Vierling J, et al. Pediatric length-of-stay index following liver transplantation. Pediatr Transplant. 2020;24(7):e13779. [DOI] [PubMed] [Google Scholar]
- 14.Covarrubias K, Luo X, Massie A, Schwarz KB, Garonzik-Wang J, Segev DL, et al. Determinants of length of stay after pediatric liver transplantation. Pediatr Transplant. 2020;24(4):e13702. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.