Infectious complications more than 1 year after liver transplantation: a 3-decade nationwide experience.

Because few reports have addressed infections late (≥1 year) after liver transplantation (LT), we evaluated the incidence, risk factors and pathogens involved. Infection data were from the Finnish LT registry, with starting date, type and relevant pathogens for 501 Finnish adult LT patients surviving ≥1 year post-transplant. Follow-up end points were end of study, death or retransplantation. Logistic regression to assess risk factors was adjusted for age, gender and follow-up time. With 3923 person-years of follow-up, overall infection incidence was 66/1000 person-years; 155 (31%) suffered 259 infections, and two-thirds experienced only one infection. Cholangitis (20%), pneumonia (19%) and sepsis (14%) were most common. The most frequent bacteria were Enterococcus spp. and Escherichia coli, and the most frequent viruses cytomegalovirus and varicella zoster virus. Fungal infections were rare (n = 7). With 13 fatal infections, 17% of all late deaths involved infection. Primary sclerosing cholangitis (PSC) and Roux-en-Y-type biliary anastomosis were associated with cholangitis; 18% of PSC patients suffered late cholangitis. Late acute rejection was associated with sepsis. Age, gender or cytomegalovirus did not significantly influence late infections. In conclusion, although infection risk under maintenance immunosuppression therapy is relatively low, particular vigilance regarding cholangitis, pneumonia and sepsis seems appropriate.

Long-term results of liver transplantation.

Liver transplantation (LT) is an established therapy associated with a dramatic improvement in patients life expectancy. With improved early-term management, current 10-year patient survival rates in many indications exceed 70%. Life-long immunosuppressive therapy may, however, be accompanied by considerable longterm toxicity: most importantly, renal dysfunction, cardiovascular disease, and cancer, which, in addition to recurrence of the primary liver disease, emerge as key contributors to late mortality. Chronic kidney disease cumulatively affects up to 28% of patients by ten years after LT. Various factors can contribute to renal impairment, but perioperative acute kidney injury, calcineurin inhibitor toxicity, hypertension, and diabetes are considered most important. LT patients demonstrate 3-fold risk for cardiovascular events, which seems to result mostly from an excess of traditional risk factors, mainly hypertension and diabetes. The cumulative cancer incidence reaches 16-42% by 20 years after LT, and cancer rates are 2- to 4-fold higher among LT patients than among matched controls. Highest rates are for nonmelanoma skin cancer (3- to 70-fold) and lymphoma (8- to 29-fold). The liver graft usually displays uncomplicated function in the long term. Most common causes for chronic graft dysfunction include disease recurrence and biliary problems. LT generally restores patients quality of life to a level comparable with that of the general population, with only minor deficits in some areas. Thus, long-term survival after LT is impressive, and despite these long-term complications, patients quality of life remains comparable with that of the general population.

Quantitative HHV-6B antigenemia test for the monitoring of transplant patients.

Human herpesvirus-6 (HHV-6) infection, mostly caused by variant B, is common after transplantation. Here, we report a new modified method using an HHV-6B glycoprotein IgG antibody, OHV-3, and attempt to quantify the HHV-6 antigenemia after liver transplantation. Twenty-four liver transplant recipients were frequently monitored by the HHV-6 antigenemia test, which detects the HHV-6B virion protein in peripheral blood mononuclear cells (PBMC). HHV-6B antigens were now retrospectively demonstrated using a glycoprotein OHV-3 IgG antibody in the immunoperoxidase staining from the same specimens and quantified as positive cells/10,000 PBMC. The results were confirmed and quantified by DNA hybridization in situ. Altogether, 206 blood specimens were analyzed. During the first six months, HHV-6 antigenemia was detected in 17/24 (71%) recipients by using the HHV-6B virion antibody. In total, 37% (77/206) of specimens were positive with the virion antibody and 39% (78/201) by the OHV-3 antibody. The peak number of OHV-3-positive cells in the PBMC varied from 5 to 750/10,000 (mean 140/10,000). The OHV-3 antibody was useful to quantify the HHV-6B antigenemia. The findings of the HHV-6B quantitative antigenemia using the OHV-3 antibody correlated well with the previous qualitative HHV-6 antigenemia assay, and can be used as an alternative quantitative method in the monitoring of HHV-6 in transplant patients.

Hepatic neutrophil activation during reperfusion may not contribute to initial graft function after short cold ischemia in human liver transplantation.

BACKGROUND: Experimental models of hepatic ischemia/reperfusion injury have implicated a pathophysiologic role for neutrophils in subsequent hepatocellular damage. In human liver transplantation, however, the effect of reperfusion-induced neutrophil activation on initial graft function is not clear. METHODS: In 38 patients undergoing liver transplantation, neutrophil CD11b and L-selectin expression, neutrophil count, and plasma lactoferrin levels were measured. To assess changes within the graft during initial reperfusion, samples of blood entering and leaving the graft were obtained simultaneously, and transhepatic ratio calculated (hepatic vein/portal vein; 1 denotes no change, <1 a decrease, and >1 an increase across the liver). Graft steatosis, postoperative liver function, and outcome were recorded. Associations between neutrophil activation markers and outcome measures were evaluated. RESULTS: Substantial hepatic neutrophil activation occurred during initial reperfusion, demonstrated by concomitant L-selectin shedding and CD11b upregulation (transhepatic ratios 0.9 [0.7-1.0]; 1.4 [0.9-1.9]; both P < .001; portal vs hepatic vein]. Simultaneously, hepatic neutrophil sequestration and lactoferrin release occurred (0.3 [0.2-0.5]; 1.7 [1.3-3.4]; both P < .001). Neither cold ischemic time (CIT; median 5 hours 36 minutes) nor hepatic neutrophil activation during reperfusion predicted early graft function, nor was there any association between CIT and neutrophil activation. CONCLUSIONS: Despite short CIT, extensive graft neutrophil activation and sequestration occurred. This, however, was not associated with impaired early graft function, suggesting short CIT may protect against severe neutrophil-mediated injury.

Inhibition of semicarbazide-sensitive amine oxidases decreases lymphocyte infiltration in the early phases of rat liver allograft rejection.

Vascular adhesion protein-1 (VAP-1) has been shown to mediate lymphocyte adhesion to endothelia at sites of inflammation in vitro and in vivo. VAP-1 is also an ectoenzyme with semicarbazide-sensitive amine oxidase (SSAO) activity. In this study we investigated whether inhibition of SSAO influences the inflammatory infiltration in acute rat liver allograft rejection. BN recipients of DA liver allografts were treated with 50 mg/kg/d semicarbazide, an inhibitor of SSAO, or similar volumes of saline. 10 rats/group were followed for graft survival, and 10 rats/group were sacrificed on day 7 post-transplantation for histology and T-lymphocyte isolation. The area percentage of portal inflammatory infiltrates in the grafts was assessed from digital photomicrographs. The proportion of CD4-, CD8- and IL2-receptor positive lymphocytes in the graft was quantified with flow cytometry. On day 7, semicarbazide treatment significantly decreased the inflammatory infiltrate area in the grafts. CD4-, CD8- and IL2-receptor positive cells were equally affected. However, animal survival was not affected. Blockade of the enzymatic activity of VAP-1 has a significant effect on lymphocyte infiltration early in acute liver rejection. Later, activation of other adhesion pathways can by-pass the blockade caused by VAP-inhibition.

High serum soluble CD30 does not predict acute rejection in liver transplant patients.

Increased pre- and posttransplantation values of soluble CD30 (sCD30) have been shown to be associated with acute kidney transplant rejection. We sought to study whether high sCD30 could predict rejection early after liver transplantation. The study population included 54 consecutive liver transplant patients, whose samples were collected before liver transplantation and at discharge, which was at a mean time of 3 weeks after transplantation. During the first 6 months posttransplantation, 22 patients experienced an acute rejection episode. Serum sCD30 concentrations were measured by an enzyme-linked immunoassay; changes in serum sCD30 levels posttransplantation were also expressed as relative values compared with pretransplantation results. Liver patients before transplantation displayed higher serum sCD30 values compared with healthy controls: mean values +/- SD were 93 +/- 58 IU/mL vs 17 +/- 8 IU/mL, respectively. At 3 weeks after transplantation the mean sCD30 concentration in liver transplant patients decreased to 59 +/- 42 IU/mL (P = .005). The mean pretransplantation serum sCD30 value was slightly lower among rejecting vs nonrejecting patients: 78 +/- 43 IU/mL vs 104 +/- 65 IU/mL (P = NS). Posttransplantation values in both groups decreased significantly: 47 +/- 34 IU/mL in patients with rejection (P = .014) vs 69 +/- 45 IU/mL in patients without rejection (P = .012). The relative value at 3 weeks posttransplantation decreased slightly more among patients with vs without rejection (70% vs 88%; NS). No correlation was found between serum sCD30 and anti-HLA class I antibodies or crossmatch positivity. In conclusion, neither pre- nor posttransplantation sCD30 levels were associated with acute rejection in liver transplant patients.

Albumin dialysis has no clear effect on cytokine levels in patients with life-threatening liver insufficiency.

Cytokines play a important role in life-threatening liver insufficiency. They are released within the liver in response to hepatic injury and inflammation. To study cytokine clearance during albumin dialysis treatment (Molecular Adsorbent Recirculating System [MARS]), we monitored proinflammatory (tumor necrosis factor alpha [TNF-alpha] and interleukin-8 [IL-8]) and anti-inflammatory (IL-10 and IL-6) cytokines and the lymphocyte activation marker IL-2sRalpha in 81 consecutive ICU patients displaying serious hepatic decompensation. Cytokine levels were measured before treatment and after the last MARS treatment in 49 acute liver failure (ALF) and 32 acute decompensation of chronic liver disease (AcOChr) patients who were mainly considered for liver transplantation. No significant change in cytokines was observed before versus after the last MARS treatment in the AcOChr group, and only IL-10 decreased significantly in the ALF group. Baseline levels of IL-8 and IL-6 were significantly lower and IL-10 was higher in the ALF group compared with those in the AcOChr group. TNF-alpha and IL-2sRalpha levels did not differ between the groups. After treatment, IL-8 was also significantly lower in ALF patients compared with the levels in AcOChr patients. In this study, MARS therapy did not show a clearly identifiable efficacy at removing circulating cytokines. However, the results revealed that ALF and AcOChr patients displayed different profiles of circulating cytokines.

Changes in liver graft rejections over time.

Incidence and possible risk factors of acute rejection, time to acute rejection, graft rejection within 3 months, multiple rejections within 1 year, steroid-resistant rejection, and graft lost to chronic rejection or to chronic dysfunction were evaluated in 388 liver transplantations. HLA matches, anti-HLA class I antibodies, positive crossmatch test, or positive cytomegalovirus serology did not have an effect on the occurrence of acute or chronic rejection. Increased total bleeding diminished occurrence of acute rejection, lengthened the time to acute rejection, and reduced the risk of steroid-resistant rejection. Immunological pretransplant factors did not have a major effect on the occurrence of rejection after liver transplantation. Different types of rejections diminished over time and the time period to the first acute rejection increased, although the basic immunosuppression stayed mainly the same over 20 years in our center.

Substantial increase in cadaveric organ donors in hospitals implementing the donor action program in Finland.

The results of solid organ transplantation have improved during the last decade. Five-year patient survivals over 80% and graft survivals over 70% are common in many transplant centers. Also, quality-of-life assessments show that not only adults but even small children have a good quality of life after successful organ transplantation. Furthermore, transplantation programs have proved to be cost-effective. However, the organ shortage is a worldwide problem, which has in many countries led to prolonged waiting times, deaths on the waiting list, increased living related donations, acceptance of lower-quality organs, and in some instances even commercialization of the organ supply. Thus, it is extremely important to find strategies that increase the number of cadaveric organs for donation. In Finland organ transplantation is concentrated in one center with about 250 transplantations of different organs performed annually. The number of patients needing a new cadaveric organ is steadily increasing, but the number of donors has remained the same during the last decade. To improve cadaveric organ procurement the Donor Action (DA) program, which consists of a Hospital Attitude Survey and a medical records review performed by the donor hospital, has proved to increase the number of cadaveric donors. We introduced the DA program in Finland in 2000. Here in we report the results of this program in terms of its impact on the availability of cadaveric donors.

The effect of albumin dialysis on cytokine levels in acute liver failure and need for liver transplantation.

In acute liver failure (ALF), detoxification capacity of liver cells is reduced and a variety of cytokines, immune modulators, and toxic substances are accumulating. Multiple organ failure in ALF has been associated with increased blood cytokine levels. We have used a blood purification system, molecular adsorbent recirculating system (MARS), which is based on removal of both protein bound and water-soluble substances and toxins in liver failure. In this study, we measured the effect of MARS therapy on plasma cytokine levels in 49 patients with ALF. Interleukin 6 (IL-6), IL-8, IL-10 and tumor necrosis factor (TNF)alfa were determined immediately before and after the first MARS therapy and after the last session using enzyme-linked immunosorbent assays. The overall survival of these ALF patients was 82% at 6 months; the native liver recovered in 26 cases, and 14 were successfully transplanted. All three interleukins were increased before the MARS treatment but only anti-inflammatory IL-10 was reduced significantly during therapy, which in this setting could be interpreted as a positive effect. We were not able to show constant decreases in proinflammatory cytokines, but only transient effects on IL-8 and IL-6. Surprisingly TNFalfa level was normal and did not change during therapy. In theory, MARS albumin dialysis may remove toxic substances from the blood circulation and thereby improve the possibilities of the liver to recover; however, of the measured cytokines only IL-10 decreased significantly.

Experience of Mars therapy with and without transplantation in 101 patients with liver insufficiency.

Liver support devices are used to treat life-threatening organ dysfunction until a hepatic graft is available or recovery of the native liver. We used a blood purification system--molecular adsorbent recycling system (MARS)--that is based on removal of both protein-bound and water-soluble substances and toxins. Within 2.5 years we treated 101 patients, who were stratified to three subgroups: acute liver failure (ALF; n = 56), acute decompensation in chronic liver failure (AcOCh; n = 35) and liver graft failure (n = 10). MARS seems to be a promising therapy for ALF, allowing the patient's own liver to recover or to gain enough time to find a liver graft. The most promising results, namely the highest number of livers to recovery were observed among acute patients with liver failure due to a toxic etiology. However, we did not discover much benefit of MARS for patients with AcOCh without liver transplantation.

Risk factors predicting survival of liver transplantation.

Prognostic models were developed for analyzing graft survival in a single-center study consisting of all 388 adult liver transplantations performed during 20 years. Proportional hazard models and generalized linear models were used to assess which risk factors, related to donor and recipient characteristics as well as graft preservation and operation, had an effect on graft survival. The prognostic modeling evidenced favorable trends in graft survival time during the successive quinquennials 1982-1987, 1988-1992, and 1993-1997, in comparison to the referent time period 1998-2002. Significant predictors of graft survival time were donor's age, recipient-donor gender compatibility, recipient's blood group, intraoperative blood transfusion, size of the transplanted organ, and indication for transplantation. Conventional histocompatibility matching did not correlate with graft outcome.

DNA microarray-based gene expression profiles of cytomegalovirus infection and acute rejection in liver transplants.

An association between cytomegalovirus (CMV) infection and alloresponse has been suggested. CMV increases inflammation and adhesion molecule expression in graft, and induces cytokines and growth factors, linked with transplant vasculopathy and chronic rejection. We have investigated the gene expression of various inflammatory factors in the CMV-associated immune response and compared this with the immune response of acute rejection in liver transplants by using DNA microarray technology. Gene expression was studied at mRNA level in biopsies from liver transplant patients experiencing CMV infection or acute rejection. RNA extracted from liver grafts after reperfusion was used as control material. Among the strongly upregulated genes in the specimens obtained from liver transplants during CMV infection were IFN-gamma, caspases 1 and 3, granzymes A and B, TGF-beta receptors II and III, IL-10 receptor alpha, VCAM-1, TNF receptor, IL-4, TNF-alpha, IL-10, IL-2 receptor beta, IL-1beta, PDGF-receptor beta, vascular adhesion protein-1, TGF-beta2, and ICAM-1. In biopsies with acute liver allograft rejection, the most significantly upregulated genes were MHC class II, IFN-gamma, caspases 1 and 3, IL-2R beta and gamma, granzymes A and B, VLA-4, L-selectin, E-selectin, VCAM-1, and IL-1beta. Upregulated genes common for CMV and alloresponse were granzyme A and B, E-selection, IFN-gamma, VCAM-1, VLA-4, TNF, caspases 1, 3, and 8, and PDGF. Microarray analysis defined different entities in the immune responses of CMV infection and acute rejection. The differences and similarities of the gene expression profiles related to those in CMV infection and rejection may help to understand the intragraft immunologic events.

HHV-6-DNAemia related to CMV-DNAemia after liver transplantation.

In addition to cytomegalovirus (CMV), activation of other betaherpesviruses, especially human herpesvirus 6 (HHV-6), has been reported in liver transplant patients. The purpose of this study was to investigate the posttransplant HHV-6-DNAemia in relation to CMV-DNAemia in liver transplant patients. Thirty-one adult liver allograft recipients were regularly monitored for CMV and HHV-6 during the first 3 months after transplantation. For the diagnosis of CMV infections, pp65-antigenemia assay and quantitative DNA-PCR were used. HHV-6 was demonstrated by using quantitative DNA-PCR and HHV-6 antigenemia test. Altogether 253 blood specimens of 31 recipients were analyzed. In addition, CMV and HHV-6 specific antigens were demonstrated by immunohistochemistry in liver biopsy specimens in the case of graft dysfunction. Thirteen patients (40%) developed a clinically significant CMV infection, at a mean of 33 days (range 5 to 62 days) after transplantation and were treated with intravenous ganciclovir. The peak viral loads of these symptomatic CMV infections were high (CMV-DNA 34210 +/- 37557 copies/mL plasma). Six additional asymptomatic patients demonstrated significantly lower CMV-DNAemia levels (1020 +/- 1008 copies/mL, P < .05), and were not treated. Concurrently with CMV, HHV-6 DNAemia and antigenemia were detected in 17 of 19 patients, mean 11 days (range 6 to 24 days) after transplantation. HHV-6 appeared prior to CMV in most cases (12 of 17). However, the peak viral loads were low (HHV-6-DNA <1500 copies/mL blood), even in the five patients who demonstrated HHV-6 antigens on liver biopsy. All CMV infections were successfully treated with ganciclovir and the CMV DNAemia/antigenemia subsided. HHV-6 also responded to the antiviral treatment, but more slowly and less clearly. In conclusion, HHV-6 activations were common and usually associated with CMV infection in liver transplant patients. Further investigation of the clinical significance of HHV-6 DNAemia/antigenemia is necessary.

Pharmacokinetics and ventilatory effects of oxycodone before and after liver transplantation.

The pharmacokinetics and ventilatory effects of oxycodone were studied in six volunteer patients with end-stage liver cirrhosis before and after orthotopic liver transplantation. Plasma samples and urine were collected for 24 hours after intravenous administration of 0.05 mg/kg oxycodone hydrochloride. Concentrations of oxycodone and its metabolites, noroxycodone and oxymorphone, were measured in plasma and urine. THe median elimination half-life of oxycodone was 13.9 hours (range, 4.6 to 24.4 hours) in patients with cirrhosis before transplantation and 3.4 hours (range, 2.6 to 5.1 hours) after transplantation (p < 0.05). Correspondingly, oxycodone clearance increased from 0.26 L/min (range, 0.15 to 0.73 L/min) before transplantation to 1.13 L/min (range, 0.71 to 3.98 L/min) after transplantation (p < 0.05). Oxycodone depressed ventilation more strongly before transplantation than after transplantation (p < 0.05). Care should be exercised when oxycodone is used in patients with end-stage disease.

ICAM-1 induction on hepatocytes as a marker for immune activation of acute liver allograft rejection.

Intercellular adhesion molecule-1 (ICAM-1) induction on hepatocytes was investigated in relation to acute liver allograft rejection, CMV infection, and systemic bacterial infections. Twenty-four liver transplant recipients underwent an episode of acute rejection, 13 developed a symptomatic clinical CMV infection, and 7 had bacterial sepsis. Seven recipients without rejection or infection complications were used as controls. All rejection episodes monitored by frequent FNABs were reversible, and lymphocyte and lymphoid blast-dominated with a with peak of inflammation (7.2 +/- 3.9 corrected increment units [CIU]). The rejections were treated with high-dose steroids, and the inflammation subsided within one week. ICAM-1 was demonstrated from fine needle aspiration biopsy (FNAB) preparations by a monoclonal antibody and immunoperoxidase staining. ICAM-1 was not detected on the hepatocytes immediately after transplantation or in control patients, but was always seen during rejection. ICAM-1 appeared 1-5 days before the onset of inflammation in FNAB. The intensity of ICAM-1 expression increased toward the peak of inflammation and subsided together with inflammation. During CMV infection a mild immune activation was seen in FNAB (peak 2.5 +/- 0.8 CIU) and in blood. An intense ICAM-1 induction also preceded the immune activation caused by CMV, and subsided slowly with successful antiviral treatment. In addition, a slight ICAM-1 induction on the hepatocytes was recorded during bacterial sepsis. ICAM-1 induction on hepatocytes appears to be linked with an early phase of immune response, and it even precedes the lymphoid activation of rejection. However, several infections, such as CMV and bacterial infections, raise an immune response and may also induce ICAM-1. In conclusion, ICAM-1 induction on hepatocytes can be considered an early, though unspecific, marker for acute liver allograft rejection.

Human herpesvirus-6 antigenemia after liver transplantation.

BACKGROUND: Human herpesvirus (HHV)-6 has recently been reported in liver transplant patients. It infects and causes dysfunction in hepatic transplants, which provides serious differential diagnostic problems between allograft rejection and viral infection. The diagnosis of posttransplantation HHV-6 infection is usually based on serology or on polymerase chain reaction detection of viral DNA in peripheral blood specimens. However, serology does not tell the exact time of the infection, and detection of viral DNA by polymerase chain reaction may also indicate a latent infection in seropositive patients. Here we report the diagnostic use of frequent monitoring of HHV-6 antigenemia after liver transplantation. METHODS: Altogether 622 blood specimens from 51 consecutive adult liver transplant patients were analyzed. The diagnosis was based on demonstration of HHV-6-specific antigens in peripheral blood mononuclear cells using immunoperoxidase staining and monoclonal antibodies and on serology. RESULTS: During the first year (7-280 days) after transplantation, HHV-6 infection was diagnosed in 11 (22%) of 51 patients. HHV-6 early antigens, as well as HHV-6 variant B antigens, were detected in all 11 patients. HHV-6 diagnosis was confirmed by serology. The episode of HHV-6 antigenemia usually lasted for several weeks together with mild, if any, clinical signs of the infection. A significant graft dysfunction was associated with HHV-6 antigenemia in 8 of 11 patients, and viral antigens were also detected in the liver biopsy specimens of 3 of these patients. CONCLUSIONS: An active HHV-6 infection can be diagnosed from peripheral blood by detection of virus-specific antigens in mononuclear cells. HHV-6 antigenemia correlated with seroresponse.

Hepatocellular integrity in liver donors and recipients indicated by glutathione transferase alpha.

Glutathione transferase Alpha (GSTA) is a sensitive indicator of hepatocellular integrity. Its reference range is low (0.7-14 microgram/L) and its half-life is short (1 hr) in serum. We evaluated the changes in GSTA concentration in 18 recipients during and after liver transplantation. The respective liver donors were also included in 13 cases. The baseline GSTA concentrations were normal or slightly elevated in all donors, 1.2-79 micrograms/L (median 5.1 micrograms/L) and recipients, 1.1-34 micrograms/L (median 6.4 micrograms/L). Surgical dissection of donor liver caused a moderate or even large increase in GSTA concentration, peak 80-6500 microgram/L (median 800 micrograms/L). In the recipients the peak of GSTA concentrations varied from 1400 to 47000 micrograms/L (median 5000 micrograms/L), and it was always observed within 45 min after reperfusion of the graft. The highest GSTA values were observed after long cold ischemia and in patients transplanted for acute liver failure. However, they were not associated with early graft dysfunction. There was a correlation between the AUC of GSTA and cold ischemia time in the recipients with chronic nonalcoholic liver failure (r=0.94). There was no correlation between GSTA values in the donors and recipients (r=0.14). The apparent half-life of GSTA in serum was 56 min (median). Perioperative GSTA concentrations in the donors had no obvious predictive value. In the recipients an exceptionally long apparent half-life of GSTA immediately after transplantation or a large second increase in GSTA were predictors of postoperative complications.

Infections in pediatric kidney and liver transplant patients after perioperative hospitalization.

BACKGROUND: Infectious complications are a major cause of morbidity and mortality after organ transplantation. There are several reports on infections during the first months after transplantation, but there are very few data regarding infections in long-term survivors of pediatric organ transplantation. METHODS: The incidence and type of infections were retrospectively analyzed in 56 children who underwent 59 liver or renal transplantations. Follow-up was begun when the patient was sent home after a successful operation. All of the children received triple immunosuppression. RESULTS: During a mean follow-up of 4.8 years (total, 286 patient years), 1540 episodes of infection were recorded. The median incidence was 4.8 episodes/patient year. The greatest number was seen in the smallest children, 3 to 6 months after transplantation. Viral upper respiratory tract infections were the most common problem, accounting for half of the episodes (2.7 episodes/patient year). Gastroenteritis was the second most common viral infection. Only 45 episodes of infection with herpesviruses were recorded, and seven of those were caused by cytomegalovirus. Otitis media and sinusitis were the most common bacterial infections and complicated upper respiratory infection in 23% of episodes. Thirty-nine episodes of urinary tract infections were diagnosed, thirty-one in children with renal transplants. Other bacterial infections were rare, and only three episodes of verified bacterial sepsis were diagnosed. CONCLUSION: The frequency and type of infections in children with liver and renal transplants who are on triple immunosuppression are quite similar to those in age-matched healthy children.