Objectives: The safety of living liver donors is considered a high priority. In this study, we aimed to highlight the incidence and risk factors of respiratory complications among living liver transplant donors at our institute.
Materials and Methods: We evaluated data of 178 related living liver donors who were seen from January 2014 to December 2018. We recorded significant respiratory complications, such as pulmonary embolism, pleural effusion, pneumothorax, pneumonia, acute lung injury, acute respiratory distress syndrome, and transfusion-related acute lung injury. Complications were noted as clinically evident and/or needing intervention. We also recorded the frequency of nonrespiratory complications and duration of intensive care unit and hospital stays.
Results: Ten donors (5.6%) developed significant respiratory complications: 2 (1.1%) had pulmonary embolisms, 3 (1.7%) developed symptomatic pleural effusion that required thoracentesis, and 4 (2.25%) had chest infections. The remaining donor (0.6%) had unexplained respiratory insufficiency. Logistic regression analyses identified age ≥ 35 years and previous surgery as the main risk factors of significant respiratory complications. There were no recorded cases of pneumothorax, acute lung injury, acute respiratory distress syndrome, and transfusion-related acute lung injury. Raw surface collection (14.6%) and biliary leakage (7.9%) were the most frequent nonrespiratory complications. There was no significant difference between patients with and without significant respiratory complications with regard to intensive care unit and hospital stays.
Conclusions: Despite the low incidence of significant respiratory complications among our living liver donor cohort, close monitoring and early management are essential to achieve better prognosis, especially in donors older than 35 years or those with previous surgery.
Key words : Donor care, End-stage liver disease, Liver transplantation, Pulmonary embolism
In Egypt, living donors represent the only option for patients with end-stage liver disease who require liver transplant due to failure to implement legal deceased-donor liver transplant programs.1
Despite the widespread use of living-donor liver transplant (LDLT) with satisfactory results, there is still concern for healthy donors who undergo a complicated surgery for liver graft retrieval. Studies have shown increased numbers of cases of morbidity and mortality among living liver donors after hepatectomy.2,3 Recently, there are increasing reports about postoperative respiratory complications within this population.4,5 In this retrospective study, we highlighted the incidence and risk factors of respiratory complications among living liver donors in our center.
Materials and Methods
After approval of the study protocol by the research ethical committee of the Faculty of Medicine, Ain Shams University (code number: FMASU FAMSU R37/2018) and registration in the ClinicalTrials.gov database (registration number: NCT03862664), we retrieved medical records of 178 related donors who underwent hepatic lobectomy for liver transplant between January 2014 and December 2018 in Ain Shams Specialized Hospital. Our aim was to evaluate respiratory complications that occurred in living liver donors in the first 6 months postoperatively.
All donors underwent thorough preoperative evaluations to ensure eligibility criteria for donation. Evaluations comprised documentation of comprehensive history, physical examination, and full laboratory and radiologic investigations, including computed tomography (CT), venography, arteriography portography, magnetic resonant cholangiopancreatography, electrocardiogram, echocardiography, pulmonary function testing, and liver biopsy as indicated.
The main donor selection criteria in our center included the following: excellent general physical and mental condition, age between 18 and 50 years of either sex, free of chronic disease or previous major abdominal surgery, body mass index (BMI) ≤ 28 kg/m2, and ABO blood group identical to recipients and negative for hepatitis B, hepatitis C, and human immunodeficiency viruses. Patients with major illnesses (ASA physical status > II), BMI > 28 kg/m2, psychiatric diseases or mental retardation, pregnant and nursing women, and significant laboratory abnormalities were excluded.
Informed consent was obtained from all living liver donors after full explanation of the procedure. This was done on 2 occasions: the first time in the presence of another family member and the second time in front of the ethical committee. The donor was reassured that he/she had the right to withdraw at any time until just before the operation.
A standardized general anesthetic technique was used for all patients. Measures to reduce intraoperative bleeding included maintaining a low positive central venous pressure during the process of piggyback and liver resection using a combination of the Cavitron Ultrasonic Surgical Aspirator/bipolar electrocautery (CUSA System 200 Macrodissector; Cavitron Surgical Systems, Stamford, CT, USA) and the harmonic scalpel (Johnson & Johnson, Highland Park, NJ, USA). Accessory hepatic veins (segment 5, segment 8, and right inferior hepatic vein) that crossed the resection line were either securely closed with surgical clips and sutures or clamped and reconstructed on back table according to size, flow, and graft-to-recipient weight ratio.
With regard to fluids given to the donor, we used a restrictive fluid strategy before graft removal to preserve the low positive central venous pressure (around 5 cm H2O) in order to decrease blood loss. Ringer acetate was mainly used as the solution. After resection and before the end of surgery, intravenous fluids were increased to restore the central venous pressure to between 6 and 12 cm H2O. Intraoperative blood loss was collected and salvaged by a cell saver machine and used when needed. Heparin 5000 IU was given intravenously to the donor before removal of the graft to avoid thrombosis in its vasculature.
All patients were extubated in the operating theater and admitted to the intensive care unit (ICU) immediately postoperatively.
Donors were admitted to the ICU at least overnight for monitoring. During the ICU stay, the nasogastric tube and Foley catheter were removed and early oral nutrition was encouraged. Postoperative medications included prophylactic antibiotic coverage (cefepime 2 g every 12 h intravenously for the first day as a prophylactic measure together with intravenous metronidazole 500 mg every 8 h) and proton pump inhibitors in the form of intravenous pantoprazole 40 mg/24 hours as prophylaxis against stress ulceration.
Prophylactic measures, like intermittent calf compression during surgery and in the first 24 hours after surgery, were applied to reduce the risk of deep vein thrombosis (DVT). Standard DVT prophylaxis with low-molecular-weight (LMW) heparin was implemented by the second day if there were no contraindications. Chest physiotherapy and early mobilization were part of routine immediate postoperative care.
Postoperative follow-up plan
Follow-up included daily monitoring of vital data, wounds, and drains. Postoperative laboratory investigations were done every day during the first week, every other day in the next week, and on the last day before discharge. Investigations included measurements of complete blood count, C-reactive protein, blood urea, serum creatinine, sodium, potassium, magnesium, phosphate, alanine transaminase, aspartate transaminase, total and direct bilirubin, alkaline phosphatase, and gamma-glutamyltransferase. The standard coagulation tests included prothrombin time, international normalized ratio (INR), activated partial thromboplastin time, fibrinogen, and platelets. In addition, abdominal and chest ultrasonography examinations were done daily during the first week and then twice per week until discharge.
Data from hospital medical records were reviewed. Patient demographics and intraoperative and postoperative data were retrieved and analyzed. We collected data on baseline characteristics (age, sex, BMI, smoking history) and on intraoperative variables (type of surgical procedure, duration of surgery, graft demography, and administration of blood products). Respiratory complications, including pulmonary embolism, pleural effusion, pneumothorax, pneumonia, acute lung injury (ALI), acute respiratory distress syndrome (ARDS), and transfusion-related acute lung injury (TRALI), were considered when they were clinically evident. Postoperative chest radiography, chest CT (if it was done), need for postoperative ventilation, ICU stay, and total hospital stay data were collected.
The primary outcome was the development of respiratory complications in donors. The secondary outcomes were frequency of nonrespiratory complications, ICU stay duration, and hospital stay duration.
Using the PASS program (NCSS, Kaysville, UT, USA) and setting the alpha error at 5% and confidence interval width of 0.1, we considered the previous results from Ulubay and colleagues,4 which showed that the incidence of respiratory complications among living donors was 8.5%. Based on this, the needed sample size was at least 124 cases.
Collected data were coded, tabulated, and statistically analyzed using IBM SPSS statistics software version 18.0 (IBM Corporation, Chicago, IL, USA).
Descriptive statistics were performed for quantitative data (presented as minimum and maximum of the range), for quantitative normally distributed data (presented as mean ± standard deviation), and for qualitative data (presented as number and percent).
Inferential analyses were done for quantitative variables using the Shapiro-Wilk test for normality testing and independent t tests in cases of 2 independent groups with normally distributed data. For qualitative data, inferential analyses for independent variables were done using chi-square test for differences between proportions and the Fisher exact test for variables with small expected numbers. Logistic regression was used to find independent factors that affected certain conditions. P values < .05 were considered significant.
Of 178 analyzed donors, 10 (5.6%) developed significant respiratory complications (SRCs) that were clinically evident. Table 1 presents the demographic characteristics of analyzed donors. All donors were related to their recipient: 125 (70.2%) were first-degree relatives and 53 (29.8%) were second-degree relatives. With regarding to donors with SRCs, all were first-degree relatives. All donors with SRCs underwent right hepatectomy. Donors with SRCs were nonsignificantly older. In addition, previous surgery was nonsignificantly more frequent among donors with SRCs. Logistic regression analysis identified age ≥ 35 years and previous surgery as the main risk factors for SRCs (Table 2). All cases of SRC were recorded during the first month postoperatively.
Two donors (1.1%) developed pulmonary embolisms. Both were male donors and nonsmokers without relevant medical or surgical history. Patients complained of sudden dyspnea, tachycardia, and chest pain. They were promptly diagnosed with elevated D-dimer, and pulmonary CT angiography revealed mild changes. Electrocardiogram and echocardiography results were normal. No patient deteriorated to respiratory or circulatory failure. There was no correlation between pulmonary embolism and the development of DVT. Both donors received the optimum anticoagulation therapy.
The most commonly encountered complication was pleural effusion, which represented 29.2% (n = 52) among the studied cases. Daily postoperative chest ultrasonography was done routinely for all donors. Different grades of pleural effusion were observed, with minimal grade in 15 donors (29.4%), mild grade in 29 donors (56.9%), moderate grade in 6 donors (11.8%), and severe grade in 1 donor (2.0%). Among donors with and without pleural effusion, there were significantly more female donors (42.3% [n = 22] vs 18.3% [n = 23]; P = .001). Logistic regression analysis identified female sex as a significant risk factor for effusion (odds ratio = 2.84; 95% confidence interval, 1.36-5.92; P = .006). Most of these donors were treated conservatively, and pleural fluid was spontaneously absorbed. Only 3 cases were clinically relevant (1.7% of all analyzed cases). These 3 donors were further examined with chest CT, and ultrasonographic-guided tapping of pleural effusion was performed.
Four donors (2.25%) had chest infections, manifested by cough and fever, although chest radiography results were free of clinical abnormalities. The donors received proper antibiotic therapy according to culture and sensitivity tests, and infections did not progress into pneumonia. One case of unexplained respiratory insufficiency was detected. This donor experienced sudden dyspnea and chest pain. Computed tomography angiography and D-dimer tests were done, which revealed no abnormalities. In addition, cardiac examination, electrocardiogram, and echocardiography tests were normal. No donor required intubation, mechanical ventilation, or readmission to the ICU.
There were no reported donors with pneumothorax, ALI, ARDS, or TRALI. No incidences of mortality were recorded during our period of analyses. Table 3 shows the demographic results of donors with SRCs.
With regarding to nonrespiratory complications, the most commonly encountered complication was raw surface collection (14.6% [n = 26]), with only one case among patients with SRCs. All incidences were minimal or mild collection on the liver cut surface and were managed conservatively. Only 3 donors required pigtail insertion to evacuate the collection. The second most commonly encountered complication was biliary leakage (7.9% [n = 14]) with 1 incidence in the SRC group. These nonrespiratory complications were managed conservatively. Table 4 shows the distribution of nonpulmonary complications among the analyzed donors.
There were no significant differences between patients with and without SRCs with regard to ICU and hospital stays. There were no cases of reoperation during the period of analysis (Table 5).
As early as 1989, LDLT has given hope to many patients with end-stage liver disease; unfortunately, this procedure subjects healthy donors to major surgery with a relatively high risk for mortality and morbidity.6,7
In our present study, the frequency of SRCs was 5.6% in 178 living liver donors. Results from other studies showed variable incidences of respiratory complications among donors ranging from 0.7% to 38.5%.8-11 This variation may be related to the absence of consensus on the definition of reported respiratory complications. In our study, we excluded minor complications discovered accidentally during chest ultrasonography examinations.
All SRCs in our study occurred in patients who underwent right hepatectomy. Dondero and associates5 and Ates and associates12 reported no respiratory complications after left hepatectomy. In addition, Iwasaki and his colleagues10 reported a lower incidence of pulmonary complications with left hepatectomy compared with right hepatectomy (2.1% vs 6.6%).
Our study revealed that age ≥ 35 years and previous surgery were risk factors for SRCs among our donor cohort. Candido and colleagues2 identified right lobectomy, smoking, and intraoperative blood transfusion as risk factors for overall postoperative complications in living liver donors. In our center, preoperative smoking cessation strategy is one of the necessary steps in the donor preparation program. It has been demonstrated to be exceedingly powerful in diminishing the incidence of postoperative respiratory complications.13
Despite the low incidence of postoperative pulmonary embolism in our present study (2 donors [1.1%]), it is considered a critical event that needs prompt diagnosis and management. Several studies have reported variable incidences of postoperative pulmonary embolism in donors of LDLT, with at least 4 cases of mortality.14-17
Cerutti and colleagues18 studied postoperative coagulation status in living liver donors using thromboelastography. They found that more than half of the donors demonstrated a transient hypercoagulable state despite routine LMW heparin prophylaxis. This may be due to release of huge amounts of tissue factor, factor VIII, and/or von Willebrand factor during cutting of the liver tissue with subsequent enhancement of the coagulation cascade over the body, which is followed by fibrinolysis.
In our center, there is a strict policy to prevent thromboembolic events after hepatectomy for LDLT, which start with the preoperative evaluation of donors. This evaluation involves obtaining the history of thromboembolic diseases and screening for coagulation abnormalities with prothrombin time, partial thromboplastin time, INR, factor V Leiden mutations, proteins C and S, and antithrombin III. Factor V Leiden mutations have been considered a common cause for rejection of donors (accounting for 23% of all causes of rejection).19
In our center, we apply intermittent calf compression during surgery and in the first 24 hour postoperatively. We use LMW heparin for prophylaxis against DVT by the second day if there are no contraindications. Patients are also encouraged to achieve early mobilization. However, close monitoring with thromboelastography and/or laboratory investigations is still required for early detection of pulmonary embolism and to tailor the dose of LMW heparin in highly susceptible individuals. In addition, we recommend correlating the incidence of pulmonary embolism with perioperative fluid therapy and drugs.
In our study cohort, pleural effusion was frequently recorded (52 donors [29.2%]). However, only 3 donors (1.7%) required thoracocentesis. In our study, female sex was a risk factor for development of pleural effusion. Previous studies have also revealed a high frequency of pleural effusion among living liver donors, occurring in 20.4% to 75% of donors.5,20 Most cases were mild and required no medical intervention. The pathology behind the development of pleural effusion after liver resection is still not fully understood. Some investigators have suggested that hypoalbuminemia and the operative technique may be reasons for development.21
After hepatectomy, low albumin levels may occur because of either dilution by intraoperative fluid therapy with subsequent alteration in vascular permeability or transient impairment of synthetic function of the liver due to low residual liver volume.22 In addition, the extended operative technique, especially with right hepatectomies and division of ligaments, may add to severe damage to the lymphatic vessels with secondary creation of abnormal peritoneal-pleural pathways. This could result in the increased frequency of postoperative pleural effusion.23,24
Chest infection was a common SRC (2.25%) in our study. Chest infections usually carry significant morbidity as it can progress to pneumonia if not treated properly. In general, a chest infection was usually seen 3 to 5 days postoperatively. Multiple factors contribute to postoperative chest infection or pneumonia.25 Dondero and associates5 reported 3 cases of empyema (2.7%) and 3 cases of bacterial pneumonia (2.7%) among living liver donors, with Ates and associates12 reporting pneumonia in 4 cases (0.3%).
The low incidence of pneumonia in our study may be attributed to preoperative preventive measures like smoking cessation and exclusion of obese patients. Our center also employs postoperative measures, such as effective pain management, chest physiotherapy, and early mobilization.
In our study, 1 donor experienced sudden respiratory distress. Ates and associates12 reported acute respiratory insufficiency in 5 donors (0.4%). In our analyses, there were no cases of pneumothorax. In contrast, Dondero and associates5 identified 4 cases of iatrogenic pneumothorax and Ates and associates12 reported 2 cases (0.2%) of pneumothorax. Most pneumothorax cases are related to central venous line application; the low incidence in this study may be due to the use of ultrasonography for central venous line insertion. In our study, we recorded no cases of ALI, ARDS, and TRALI.
Raw surface collection (14.6%) and biliary leakage (7.9%) were the most common nonrespiratory complications among analyzed patients in our study. In their study of 3565 Japanese living liver donors, Hashikura and colleagues reported bile leakage and wound infection as the most common nonrespiratory complications.8
There was no significant difference between patients with and without SRCs with regard to duration of ICU and hospital stays. This may be related to the prompt management of complications, resulting in most donors not requiring an extended hospital stay. No cases of reoperation were recorded during our analyzed period, and we had no cases of mortality. Previous studies have reported mortality rates of 0.18% to 0.2% among donors for LDLT.26,27
A limitation of this study was that we did not correlate frequency of SRCs with postoperative laboratory investigations, such as albumin and INR levels.
Donor safety is a major issue in LDLT; however, complications are still inevitable in some cases. Close monitoring with early diagnosis and prompt management are essential to achieve a better prognosis. Age ≥ 35 years and previous surgery could be identified as risk factors for SRCs. More prospective and retrospective studies are needed to obtain further information on pulmonary complications and to investigate proper management strategies.
DOI : 10.6002/ect.2019.0394
From the 1Department of Anesthesiology, Intensive Care and Pain Management,
Faculty of Medicine, Ain Shams University, and the 2Department of General
Surgery, Faculty of Medicine, Ain Shams University, Cairo, Egypt
Acknowledgements: The authors received no sources of support for this study and have no conflicts of interest to declare.
Corresponding author: Marwa Ahmed Khairy Elbeialy, Department of Anesthesiology, Intensive Care and Pain Management, Faculty of Medicine, Ain-Shams University, Abbassia, Postal Code 11591, Cairo, Egypt
Phone: +2 01006518802
Table 1. Demographic Characteristics of Donors
Table 2. Logistic Regression for Factors Affecting Significant Respiratory Complications
Table 3. Demographics of Donors With Significant Respiratory Complications
Table 4. Comparison of Nonrespiratory Complications
Table 5. Outcomes Among Study Cohort