|Home | About | Journals | Submit | Contact Us | Français|
Sixteen recipients of neonatal liver grafts were compared with 114 contemporaneous pediatric recipients of grafts from older donors. Graft and patient survival were worse in the neonatal group although the differences were not statistically significant. Patients with neonatal livers who had no technical complications required a longer time postoperatively to correct jaundice and a prolonged prothrombin time. These functional differences were limited to the 1st postoperative month and the end result was the same as with liver transplantation from older donors.
The available liver donor pool has been expanded with the use of grafts from older donors [13, 19]. In liver transplantation in infants and small children, the number of candidates far exceeds organ availability, resulting in significant mortality of patients on waiting lists [4, 23]. Possible options include the use of reduced-size or split liver grafts [1, 2, 7, 8,12] and the use of livers from neonatal donors. Results of our experience with the use of neonatal livers for transplantation in very small pediatric recipients is herein reported. A minimum of 14 months follow-up is provided.
The charts of all patients who underwent primary liver transplantation at the Children’s Hospital of Pittsburgh under cyclosporine-steroid therapy and had liver grafts preserved with University of Wisconsin solution  were reviewed. These transplants took place between October 1987, when University of Wisconsin solution became available to us for clinical use, and September 1989, when FK 506 was introduced as the primary immunosuppressive agent . Follow-up is provided to December 31, 1990, and consequently ranged from 14 to 39 months (median 25 months).
One hundred and thirty patients were studied: their primary liver disease is shown in Table 1. The recipient age ranged from 1 to 204 months (mean 51.1 ± 4.9 months) and the male/female ratio was 7/6.
Two groups were identified depending on the age of the donor: group A were recipients of neonatal (age 28 days or less) donor grafts (n = 16) and group B recipients of grafts from older donors (n = 114). In case of retransplantation, the age of only the primary donor was considered.
Donor age, body weight, prothrombin time, serum total bilirubin and alanine aminotransferase (ALT) were recorded. These laboratory data were missing for 10 donors in group B.
An accounting of the donors and recipients in group A is given in Table 2. Anoxic injury due to birth asphyxia was the most common cause of donor death (75%). Thirteen donors (81%) were less than 2 weeks old and all weighed less than 5 kg. All but two recipients were less than 1 year old and all weighed less than 10 kg. Biliary atresia was the underlying disease in 11 patients. Nine of them (82%) had had previous failed portoenterostomies.
A summary of the donor and recipient demographic data in group B is shown in Table 3.
The surgical techniques employed for graft recovery and transplantation have been reported previously [14, 16, 18]. Organ preservation period (cold ischemic time) ranged from 6.7 h to 17.9 h (9.4 ± 1.4 h) in group A and from 5.1 h to 29.1 h (8.5 ± 0.5 h) in group B. No statistical differences were noted between the two groups.
Heparin (50 IU/kg subcutaneously twice a day), aspirin (20 mg for body weight < 8 kg, 40 mg for 8–15 kg, or 80 mg for >15 kg, orally, once a day) and Persantine (dipyridamole, 6.25 mg for body weight < 12 kg or 12.5 mg for ≥ 12 kg, orally, twice a day) were administered, unless bleeding occurred, when the prothrombin time was less than 20 s throughout the hospitalization. All of the patients in group A received parenteral hyperalimentation including vitamin K immediately postoperatively until they tolerated oral feeding well.
Doppler ultrasonography was performed routinely within 24 h postoperatively and was repeated when a complication was suspected. Absence of arterial pulsations on two consecutive ultrasound examinations established the diagnosis of arterial thrombosis. In some patients, this diagnosis was made by arteriography or during surgical exploration.
Body weight of the patients in group A was recorded preoperatively and followed for the duration of the study. Control values were obtained from the National Center for Health Statistics percentiles .
The patients in each group who survived and had a technically uncomplicated postoperative course were identified (group A1 and group B1, respectively). Ten patients in group B1 were excluded due to incompleteness of data. Body weight, prothrombin time, serum total bilirubin and ALT preoperatively and postoperatively at days 1 through 7 and days 14 and 30 were recorded. The prothrombin time values were not in all cases available for postoperative days 14 and 30. In these cases, the mean of the immediately preceding and the following available values were entered.
All data are presented as mean ± the standard error of the mean ( + SEM); differences in the mean were tested using Student’s t test or χ2 for comparison procedure. Actuarial survival rates were calculated by the method of Kaplan-Meier. A P value of less than 0.05 was considered to be statistically significant.
Out of all the 130 patients, 35 died (27%), 6 in group A and 29 in group B. The primary causes of death in these 35 patients are shown in Table 4.
Actuarial survival rates at 1, 2, and 3 years postoperatively were 68 %, 68 %, and 51 % respectively in group A and 76%, 74% and 74% in group B (Fig. 1, top). Actuarial 1-, 2-, and 3-year graft survival rates were 56%, 56%, and 38% respectively in group A and 70%, 68%, and 67% in group B (Fig. 1. bottom). None of the differences in patient or graft survival achieved statistical significance (P = 0.37 and 0.19, respectively).
Five patients (31%) in group A underwent re transplantation, due in three patients to hepatic artery thrombosis, in one to primary nonfunction, and in one to hepatitis (Table 2). Two of them were alive at the end of follow-up, including one patient who had a third graft. Nineteen (16%) patients in group B underwent retransplantation, due in ten patients to hepatic artery thrombosis, in six patients to primary nonfunction, and in three patients to rejection. Of them nine were alive at the end of follow-up. Three patients received a third graft; all three died.
Perioperative complications are presented in Table 4. Hepatic artery thrombosis was the most common in group A and its incidence was significantly higher than in group B (31%, n = 5 vs 9%, n = 10, P = 0.027). Of the 15 patients who developed hepatic artery thrombosis nine survived, three in group A, of them one without retransplantation, and six in group B, of them two without retransplantation. Three patients developed portal vein thrombosis: one in group A and two in group B. The patient in group A died as a result of this complication. Both patients in group B were alive, one after successful thrombectomy and one other after distal splenorenal shunt. One patient in group B developed portal vein rupture due to infection. Biliary complications not associated with hepatic artery thrombosis were more common in group B, but the difference was not significant (P = 0.4).
The biochemical profile of the donors in the two groups was similar (mean ALT: 64.2 ± 22.2 vs 79.3 ± 12.1, mean prothrombin time 10.5 ± 0.9 vs 12.8 ± 0.4, respectively) except for the total bilirubin which was 1.7 + 0.5 (0.4–8.4)mg/dl in group A and 0.7 ± 0.1 (0.1–6.7) mg/dl in group B (P<0.05).
Perioperative biochemical values were compared between groups A1 (n = 7) and B1 (n = 51. Fig. 2). Serum total bilirubin levels in days 1 through 14 and prothrombin time for days 1 through 4 were significantly higher in group A1. Serum ALT levels were not significantly different.
Eight of the patients in group A were alive with their primary grafts. The body weight at the time of transplantation was below the 5th percentile in six of them, at the 5th–10th percentile in one, and at the 25th–50th percentile in the eight patient. All patients gained weight during the observation period. While three patients remained at the 5th percentile, all others upgraded (Fig. 3).
There are two features of neonatal liver transplantation requiring special consideration: technical difficulties due to the small size of the grafts , and functional performance, because neonatal livers are thought to be immature [3, 9, 11].
The increased incidence of hepatic artery thrombosis in the patients with neonatal transplants can be seen as a reflection of the technical difficulties encountered [10, 17, 21]. Interestingly, other complications related to the venous or biliary reconstructions did not occur with a greater frequency in the neonatal group. The possible role of immaturity of the graft and delayed function may also be related to hepatic artery thrombosis because of decreased or absent protein C synthesis by the newly transplanted liver .
Of the liver function tests referred to here, the prothrombin time is representative of the synthetic function of the liver and the total bilirubin of its ability to clear metabolic waste products. It appears that neonatal liver grafts gain normal synthetic function within 4–5 days after transplantation. Older donor grafts normally function immediately after transplantation. The ability to clear bilirubin is even more slowly regained: total bilirubin did not return to normal for 1 month, as compared to 1 week in grafts from older donors. Interestingly, after this slow recovery, which appears to be characteristic of the neonatal donor livers, the ultimate result is indistinguishable from that of livers from older donors.
In conclusion, if successful, neonatal liver transplantation can have equally good results as transplantation with livers from older donors.
Supported by research grants from the Veterans Administration and project grant no. DK 29961 from the National Institutes of Health, Bethesda, Md., USA.