Objectives: Weaning from mechanical ventilation after liver transplant can be demanding. In selected cases, tracheostomy is helpful. The optimal timing for tracheostomy in ventilator-dependent liver transplant recipients is not well known.
Materials and Methods: We retrospectively analyzed data of 447 patients who had undergone liver transplant in our hospital. Thirty-nine patients who had high risk of prolonged mechanical ventilation according to the Respiratory Risk Score were identified from 95 patients who received tracheostomy after liver transplant.
Results: When compared with tracheostomy performed > 3 days after transplant, early tracheostomy (≤ 3 days) had a higher likelihood of a brief duration of mechanical ventilation (62.5% vs 9.7%; P = .001). Accordingly, time spent in an intensive care unit was shorter with early tracheostomy.
Conclusions: This study provides a retrospective analysis of a small study cohort; therefore, validation of our findings requires a prospective randomized multicenter study on early tracheostomy in respiratory high-risk liver transplant recipients.
Key words : Percutaneous dilatational tracheostomy, Pneumonia, Respiratory Risk Score
Prolonged mechanical ventilation (PMV) has a significant negative effect on survival of critically ill patients; this has also been accordingly shown for patients after liver transplant (LT).1 Ventilator-associated pneumonia may contribute to decreased survival with prolonged ventilation and may be addressed by a faster weaning from invasive mechanical ventilation.2 Although tracheostomy may facilitate weaning from mechanical ventilation, the optimal timing for tracheostomy in ventilator-dependent intensive care patients remains controversial.3-5 The indications for tracheostomy and timing of the intervention and the patient population have significant effects on study outcomes. This makes the indication for tracheostomy challenging, as adequate scores for the prediction of duration of ventilation for specific patient populations are often missing.6
The safety of percutaneous dilatational tracheos-tomy after LT has been shown,7 but nothing is known about the effects of early tracheostomy after LT in critically ill recipients on duration of PMV, time in an intensive care unit (ICU), or survival.
We have previously developed a respiratory risk score (RRS) to identify LT recipients at high risk for PMV. This RRS can predict on the first postoperative day if a critically ill LT patient will require PMV.8
We hypothesized that, in patients with a high RRS, early tracheostomy would reduce the duration of mechanical ventilation.
Materials and Methods
We retrospectively analyzed 447 consecutive LT procedures performed between February 2006 and April 2015 at our center. Liver transplant was performed as orthotopic full-size transplants in all cases. Patients > 18 years old were included. Exclusion criteria included acute retransplant and preexisting tracheostomy. Two patients missing more than one RRS parameter and 2 recipients receiving tracheostomy later than postoperative day 35 were excluded before analyses (Figure 1).
The RRS scores for 95 patients who underwent tracheostomy after transplant were calculated as previously described (Table 1).8 Using literature definitions of early and late tracheostomy,5,9,10 we then categorized patients according to the timing of tracheostomy as early (≤ 3 days after transplant) or late (> 3 days). All tracheostomies were performed using a dilatational technique with bronchoscopic assistance. Our primary endpoint was duration of mechanical ventilation < 15 days after transplant. The duration of mechanical ventilation included all periods of mechanical ventilation after tracheostomy, even if patients returned to mechanical ventilation after a period of spontaneous breathing. Secondary study endpoints were time in an ICU and mortality at month 3.
Collected data were implemented and analyzed using SPSS statistical software (version 23; SPSS Inc., IBM Corporation, Somers, NY, USA). Continuous variables are reported as medians with quartiles; RRS and Simplified Acute Physiology Score II (SAPS-II) values are given as means. The associations between timing of tracheostomy (< 3 vs > 3 d after transplant) and the status of “free of mechanical ventilation” before day 15 and mortality at month 3 were analyzed by chi-square test. By linear regression test, we studied the associations between timing of tracheostomy (≤ 3 vs > 3 d after transplant) and ICU length of stay.
Overall, 95 LT recipients received a tracheostomy after LT. The median time to tracheostomy was 8 days (range: 0-35 d), and the median time on mechanical ventilation was 20 days (range: 0-228 d). There were no episodes of bleedings or other complications related to tracheostomy.
To identify patients at high risk for PMV, we used the RRS. We found that 25 of the 95 LT patients with tracheostomy were at low risk (mean 1.13 points), 31 patients were at intermediate risk (mean 5.69 points), and 39 patients were at high risk (mean 9.69 points). Critical illness of the patients at high risk (39 patients) is reflected by the high median Model for End-Stage Liver Disease (MELD) score of 36 points (range: 6-40), low oxygenation ratio of 223 (range: 108-520), and high incidence of preoperative dialysis (41%).
Most likely due to this risk profile, PMV (32 d; range: 5-231 d), prolonged time on ICU (43 d), and high mortality at month 3 (28.2%) were observed.
The median age of the 39 patients with high risk according to the RRS was 55 years (16 patients were male, 23 patients were female). The most frequent diagnosis was cirrhosis due to alcohol (25.2%), hepatitis C, and primary biliary cirrhosis (both 10.3%).
Eight of the 39 patients were free of respirator use before day 15 after LT. Patients who underwent early tracheostomy were more likely to be free of mechanical ventilation before day 15 (62.5%) compared with LT recipients with late tracheostomy (9.7%; P = .001; Table 2). In addition, patients with early tracheostomy spent less time in an ICU (29 d; range, 5-199 d) than patients who received late tracheostomy (45 d; range, 10-158 d; P = .804), although this finding did not reach statistical significance. However, there were no significant associations between early versus late tracheostomy with regard to mortality at month 3 (25% vs 29%; P = .821; Table 2).
In patients with respiratory failure after LT, little evidence exists to guide the timing of tracheostomy. We found that patients who underwent tracheostomy ≤ 3 days after transplant had a shorter duration of mechanical ventilation than patients who underwent tracheostomy > 3 days after transplant.
Observed mortality at month 3 in the high respiratory risk group of LT patients who received a tracheostomy was high (28.2%); although the proportion of these patients in the total cohort was low at 8.7%, this factor influenced the overall outcome dramatically. Time on a respirator in these patients was high (mean of 32 days), and PMV was shown to have a major effect on survival by us and others.11 Although we found no difference in mortality between early and late tracheostomy in our cohort, early tracheostomy might help to reduce mortality4 as it decreases time on a respirator.
Since the introduction of MELD-based allocation, the rate of LT candidates with a pretransplant MELD score higher than 30 has increased.12 In the present study, an increase from 4.1% in 2006 to 19.5% in 2014 was observed. Because dialysis and ascites especially favor the development of PMV,11 this observation will likely lead to increasing the numbers of high respiratory risk LT recipients in the future. The relevance of evidence-based strategies for optimizing the intensive care of these high-risk LT patients will be of particular importance.
Because of the retrospective design of this study and the small number of LT recipients at high respiratory risk, our preliminary findings warrant a prospective randomized multicenter study on early tracheostomy in LT recipients with regard to a potential extension of indication to tracheostomy after LT. To achieve comparability in a multicenter setting, a commonly accepted operating procedure especially for volume and blood management and a weaning strategy must be included. Even with such an approach, it will be difficult to identify possible confounders because LT recipients with high respiratory risk are rare and regularly have a broad range of secondary diagnoses.
The practice of early tracheostomy after LT in patients at high risk of postoperative respiratory failure revealed a significant association with the duration of mechanical ventilation in our patients. Other findings are possibly limited by several confounding variables. Further work is needed to best identify the timing of tracheostomy after liver transplant.
Volume : 16
Issue : 5
Pages : 631 - 634
DOI : 10.6002/ect.2017.0113
From the 1General, Visceral, and Transplant Surgery, Hannover Medical School,
Hannover, Germany; and the 2Core Facility Quality Management & Health Technology
Assessment in Transplantation, Integrated Research and Treatment Center for
Transplantation, Hannover Medical School, Hannover, Germany
Acknowledgements: The work of H. Schrem was supported by a grant from the German Federal Ministry of Education and Research (reference number: 01EO1302). The authors have no conflicts of interest to declare. SC and MK contributed to study design and manuscript preparation; SC, KT, and FWRV contributed to data collection; HS and SC contributed to statistical analysis; MK, WK, JK, and FL contributed to data interpretation; HS, FL, JK, and WK contributed to critical revision of the manuscript; HS contributed to funds collection.
Corresponding author: Moritz Kleine, General, Visceral, and Transplantation Surgery, Hannover Medical School, Carl-Neuberg-Str. 1, 30625 Hannover, Germany
Phone: +49 511 532 6534
Figure 1. Study Group Definitions
Table 1. Calculation of the Respiratory Risk Score
Table 2. SAPS II Score and Clinical Outcomes of Early Versus Late Tracheostomy in Patients With High Risk for Prolonged Mechanical Ventilation According to Respiratory Risk Score