Intravenous ganciclovir prophylaxis for cytomegalovirus in heart, heart-lung, and lung transplant recipients

Cytomegalovirus (CMV) disease has had a significant clinical impact on the heart, heart-lung and lung transplant recipients in our centre. CMV disease has been so severe with CMV antibody-negative heart-lung transplant patients receiving organs from CMV antibody-positive donors (CMV-mismatched patients) that in 1986 we adopted the policy of not transplanting CMV-positive organs into CMV-negative heart-lung or lung recipients. In December 1992, we instituted a policy of providing intravenous ganciclovir (5 mg/kg twice a day for 28 days) during the immediate postoperative period for CMV-mismatched heart recipients and CMV antibody-positive heart-lung and lung patients, who have been the patients at greatest risk of severe CMV disease in our centre. A placebo group was not employed because of ethical considerations, ganciclovir having been shown to be effective for the treatment of CMV infections among transplant patients. Compared with a historical control group of patients receiving no prophylaxis, prophylactic ganciclovir reduced the incidence of CMV infection (39 % vs 91 %, P = 0.0006) and CMV disease (17 % vs 74 %, P = 0.0004) among CMV antibody-positive heart-lung recipients. Prophylactic ganciclovir did not significantly reduce the incidence of CMV infection or disease among heart or isolated lung recipients. Ganciclovir was well tolerated, with few adverse reactions. In the case of heart-lung transplant patients, one month of intravenous prophylactic ganciclovir significantly reduced the incidence of both CMV infection and disease when compared with patients who received no prophylaxis. With the lung transplant and heart transplant patients, there were no significant differences between the prophylaxis and nonprophylaxis groups, although there was a consistent trend towards less infection and disease in the prophylaxis groups. Received: 14 April 1998 Received after revision: 24 September 1998 Accepted: 18 December 1998     

Cytomegalovirus‐induced thrombotic microangiopathy after renal transplant successfully treated with eculizumab: case report and review of the literature

De novo thrombotic microangiopathy (TMA) after renal transplant is rare. Cytomegalovirus (CMV)‐related post‐transplant TMA has only been reported in 6 cases. We report an unusual case of a 75‐year‐old woman who developed de novo TMA in association with CMV viremia. The recurrence of TMA with CMV viremia, the resolution with treatment for CMV, and the lack of correlation with a calcineurin inhibitor (CNI) in our case support CMV as the cause of the TMA. What is unique is that the use of eculizumab without plasmapheresis led to prompt improvement in renal function. After a failure to identify a genetic cause for TMA and the clear association with CMV, eculizumab was discontinued. This case provides insight into the pathogenesis and novel treatment of de novo TMA, highlights the beneficial effects of complement inhibitors in this disease, and shows that they can be safely discontinued once the inciting etiology is addressed.     

Cytomegalovirus-related disease and risk of acute rejection in renal transplant recipients: a cohort study with case-control analyses

The relationship between a cytomegalovirus (CMV) infection and the acute rejection of a renal transplant is not well established. The aim of the study was to document whether the clinical presentation of a CMV infection as a diffuse inflammatory disease or as a clinically asymtomatic illness is a risk factor of acute renal transplant rejection. One hundred and ninety-two consecutive renal transplant recipients were included in a historical cohort study for exposed – non exposed analyses. CMV infection after transplantation was the exposure factor. Before transplantation, 113 patients had antibodies against CMV and 79 were seronegative. The patients were divided into three groups: Group 1 consisted of 64 patients who had neither clinical signs of CMV disease nor CMV serological changes after transplantation, Group 2 consisted of 77 seropositive patients with asymptomatic viremia, and Group 3 consisted of 51 seropositive patients with clinical signs of diffuse inflammation that included fever, neutropenia, and various visceral involvements (CMV disease). Groups 2 and 3, the seropositive patients, were paired with Group 1 patients. Acute rejection was considered as CMV-induced when it occurred within one month following viremia, during the first year after transplantation. Transplant patients with CMV disease, had a significant likelihood of developing acute rejection after CMV infection or reactivation (P < 0.01). The odds ratio for developing rejection was 5.98, 95 % confidence interval: 1.21–29.40. Such a link was not documented for recipients with asymptomatic CMV infection. In conclusion, CMV disease, but not asymptomatic viremia, is a risk factor of acute renal transplant rejection. On epidemiological grounds, these results support the hypothesis that factors controlling both the viral replication and the diffuse inflammatory process are implicated in acute graft rejection. Received: 20 July 1999 Revised: 22 February 2000 Accepted: 9 June 2000     

Cytomegalovirus infection and colonic perforation in renal transplant patients

Cytomegalovirus (CMV) infection in immunocompromised patients is a major cause of morbidity and mortality. A well-documented manifestation of gastrointestinal CMV infection is gastrointestinal haemorrhage. In contrast, CMV-associated intestinal perforation has rarely been reported after transplantation, although it is well documented in AIDS patients. Three patients are reported who received their first cadaveric renal transplant in 1994 and subsequently developed CMV disease. During the course of their CMV illness, which was treated with ganciclovir, each presented with clinical suspicion of peritonitis and proceeded to laparotomy. All three were found to have sigmoid colon perforations with histological evidence of CMV infection. Following bowel resection and defunctioning, two patients made an uneventful recovery and have had the continuity of their bowel restored, but one died of overwhelming sepsis within hours of surgery. The explanation for the apparent clustering of this rare condition in transplant patients is uncertain.     

Definitions of cytomegalovirus disease after heart transplantation: Antigenemia as a marker for antiviral therapy

In this prospective study, cytomegalovirus (CMV) antigenemia was defined as the marker for initiation and episodes of antigenemia as the indicator for the duration of antiviral therapy (CMV hyperimmune globulin and ganciclovir). The CMV antigenemia assay and CMV-specific IgM and IgG antibody tests were used to monitor CMV infection in 22 heart transplant recipients who, between October 1992 and July 1994, were followed up for 6 months. A total of 178 out of 627 antigenemia assays tested positive. The highest number of positive cells was greater after primary infection than after either reactivation (43.3 vs 0.3; P<0.01) or reinfection (43.3 vs 9.3; P=NS). Sixty episodes of antigenemia were observed. More episodes of antigenemia were seen after primary infection than after either reactivation (4.6 vs 0.2; P<0.01) or reinfection (4.6 vs 2.2; P=NS). The detection of antigenemia indicated the initiation of antiviral therapy within 24 h after the blood sample was harvested. Therapy was stopped immediately after a subsequent negative result became available. Our experience indicates that antigenemia directed antiviral therapy prevents CMV disease after primary and secondary infection in heart transplant recipients.     

Serum neopterin/creatinine values correlate with severity of symptoms caused by cytomegalovirus infection in renal transplant recipients

Serum neopterin/creatinine ratios were longitudinally measured in 86 renal transplant recipients from the day before transplantation until 4 months after transplantation, and the relationship to the clinical symptoms of cytomegalovirus (CMV) infection was studied. Infection with cytomegalovirus occurred in 23 patients, 11 cases of which were due to primary infection. Symptoms caused by CMV infection were more severe in male patients, in patients who had received prior antirejection treatment, and in patients with primary CMV infection. The measurement of serum neopterin/creatinine ratios proved to be a marker for the severity of symptoms caused by CMV infection, as peak values were significantly higher in eight patients with CMV disease than in patients with no or only mild symptoms of CMV infection (P<0.05). Moreover, in seven out of eight cases of CMV disease, serum neopterin/creatinine ratios started to rise up to 2 weeks before CMV infection was proven by serology.     

Prospective cytomegalovirus surveillance in paediatric renal transplant patients

Fifty-eight consecutive paediatric recipients of 65 renal transplants were prospectively studied for up to 72 months for evidence of cytomegalovirus (CMV) infection. Blood, urine and saliva were screened for CMV pre transplant and then weekly for the first 6 weeks, and monthly thereafter, by conventional cell culture and by detection of early antigen flourescent foci. Donor CMV serostatus was available in 51 cases. All patients received triple therapy as immunosuppression and none had CMV prophylaxis. Twenty-six episodes of CMV infection (40% of transplants) and 10 of CMV disease (15%) were identified. Donor CMV seropositivity, regardless of recipient CMV serostatus, was significantly associated with CMV infection (P=0.04). First CMV excretion was significantly earlier in patients who subsequently became symptomatic than in those who did not (P=0.04), and a positive blood culture was significantly associated with CMV disease (P=0.04). We found no association between extra anti-rejection treatment or recipient's age and CMV infection or disease. CMV disease had no influence on renal function as assessed by glomerular filtration rate at 6, 12 and 24 months post transplant, nor on graft loss. CMV disease was fatal in 1 patient. We conclude that trials of CMV prohpylactic agents are indicated, particularly in recipients of CMV seropositive grafts. A positive blood culture should lead to consideration of anti-CMV therapy.     

Postural Epigastric Pain as a Sign of Cytomegalovirus Gastritis in Renal Transplant Recipients: A Case-Based Review

Background

Cytomegalovirus (CMV) infection is a common cause of morbidity and mortality among patients receiving chronic maintenance immunosuppression and is often considered the most important infection in renal transplantation. CMV gastritis has been reported in transplant recipients. Symptoms are usually considered nonspecific, and gastroscopy with biopsy is usually performed to establish the diagnosis.

Methods

We report a case of primary CMV gastritis in a renal transplant recipient. A 34-year-old man presented 4 months after renal transplantation with a 1-week history of epigastric pain that decreased in supine position, increased while sitting, and further increased when standing or walking. The immunosuppressive regimen consisted of tacrolimus, mycophenolate mofetil, and prednisone. Evaluation revealed CMV viremia with a high viral load and CMV gastritis was confirmed using gastroscopy and histopathologic examination. Intravenous ganciclovir was started and continued for 3 weeks. The epigastric pain completely resolved after treatment with ganciclovir.

Conclusions

Postural epigastric pain as a sign of CMV gastritis is fairly rare in renal transplant recipients. To our knowledge this is the third article presented in the literature so far.     

Evaluation of a Cytomegalovirus Prophylaxis Protocol in Cytomegalovirus-IgG Positive Renal Transplant Recipients (R+)

BackgroundThe aim of the study was to evaluate the efficacy of a unique cytomegalovirus- (CMV) prophylaxis protocol in terms of CMV infection and disease progression in CMV IgG positive kidney transplant recipients.MethodsAchievement of negative CMV load, using concurrent prophylactic intravenous ganciclovir therapy during induction immunosuppression, combined with a 6-month prophylactic course of acyclovir, would yield a reduced incidence of early CMV infection and disease. CMV DNA was tested for at discharge, at the third, and sixth post-op months, and at the occurrence of any event that could be associated with CMV infection. CMV DNA positive patients received ganciclovir treatment until the viral load became negative. CMV replication was monitored using a quantitative PCR method capable of detecting as few as 42.5 copies/mL. All patients were given a maintenance dose of acyclovir.ResultThe file data of 267 patients who had undergone kidney transplantation between 2007 to 2016 were examined. Thirty-four patients were excluded from the study for various reasons, unrelated to the protocol. Of the remaining 233 patients, 42 (18%) had CMV DNA infection. Three patients had CMV disease (1.3%), 1of whom died of pneumonia. Diabetes mellitus (DM) was a risk factor for CMV DNA positivity (P < .004).ConclusionThe incidence of CMV infection and disease is low in renal transplant recipients whose CMV viral load is eliminated after concurrent ganciclovir administration with induction immunosuppression.     

Recurrent cytomegalovirus disease, visceral leishmaniosis, and Legionella pneumonia after liver transplantation: a case report

PURPOSE: Recurrent cytomegalovirus (CMV) disease is a frequent complication of liver transplantation. Visceral leishmaniosis in a transplant recipient is, on the other hand, extremely rare and only two cases of kala-azar have been described after liver transplantation. Immunosuppressed patients are known to be at risk of Legionella infection and the relationship between infection with this organism and hospital water supplies has been well described. These three diseases carry a high mortality rate. Our report examines the potential relationship between these complications. CLINICAL FEATURES: We describe the case of a liver transplant recipient who presented the three complications successively and survived. After reviewing the literature, we explore hypotheses linking these infections and discuss treatment strategies. CONCLUSIONS: In the patient described, infection with leishmania probably occurred months prior to the clinical presentation, a delay that matches the incubation period of kala-azar. The simultaneous onset of leishmaniosis and of a high CMV viremia may have been a coincidence. However, CMV infection has been shown to be an independent predictor of invasive fungal infection in liver transplant recipients. CMV does indeed have a suppressive effect on the humoral and cellular immune response in vitro as well as in vivo. The clinical manifestations of leishmaniosis may, therefore, have been precipitated in this patient by the additive immunosuppressive effect of antirejection drugs and CMV.     

Impact of cytomegalovirus infection,year of transplantation,and donor age on outcomes after liver transplantation for hepatitis C

Recurrence of hepatitis C virus (HCV) infection after liver transplantation (LT) is almost universal. However, variables that hasten the progression of allograft injury have not been fully defined. Cytomegalovirus (CMV) is a common infection post-LT, and its impact on the course of post-LT HCV infection remains unclear. We investigated the impact of CMV infection on patient and graft outcomes in 93 consecutive HCV-infected liver transplant recipients. Data were collected prospectively, with surveillance cultures for CMV and protocol liver biopsies. CMV infection (defined as isolation of CMV from blood and treatment with ganciclovir) occurred in 25 patients (26.9%). Graft failure (defined as cirrhosis, relisting for LT, re-LT, or death) was significantly more common in CMV-positive compared with CMV-negative patients (52% v 19.1%; P = .002). Fibrosis stage 2 or greater on the 4-month liver biopsy specimen was more common in CMV-infected patients (45% v 16.4%; P = .01). Patients who underwent LT in more recent years had an increased risk for graft failure. Donor and recipient age, CMV infection, and mycophenolate mofetil use were significantly associated with graft failure in a stepwise multivariate analysis. CMV infection occurs in approximately one quarter of HCV-infected liver transplant recipients and is an independent risk factor for graft failure in these patients. Whether CMV mediates this by inducing increased immunosuppression or directly enhancing HCV replication requires further study. (Liver Transpl 2002;8:362-369.)     

Evaluation of clinical outcomes of prophylactic versus preemptive cytomegalovirus strategy in liver transplant recipients

Cytomegalovirus (CMV) is a major cause of morbidity and mortality following solid organ transplantation (SOT). Two strategies, prophylactic, and preemptive have emerged for the prevention of CMV infection and disease after SOT. This retrospective chart review of two liver transplant cohorts: prophylactic and preemptive, compares the clinical impact of transitioning from prophylactic to preemptive strategy. The primary outcome is the incidence of CMV viremia at 3‐and 6‐months post‐transplant. Secondary outcomes include: incidence of CMV tissue‐invasive disease, acute cellular rejection, leukopenia and neutropenia, opportunistic infection rates, hospital readmission rates, and mortality at 3‐and 6‐months post‐transplant. A total of 109 patients were included in the analysis. The incidence of CMV viremia was 4.9% and 50.0% (P < 0.001) in the prophylactic versus preemptive cohort, respectively, at 3 months post‐transplant. The incidence of CMV viremia was 24.6% and 8.3% (P = 0.026) in the prophylactic versus preemptive cohort, respectively, at 6 months post‐transplant. There were no statistical significant differences in the secondary outcomes between both cohorts. In conclusion, there is a statistical significant difference in time to onset of CMV viremia; however, the use of either prophylactic or preemptive strategy was not associated with significant negative clinical outcomes of CMV.     

The Impact of Infection on Chronic Allograft Dysfunction and Allograft Survival After Solid Organ Transplantation

Infectious diseases after solid organ transplantation (SOT) are a significant cause of morbidity and reduced allograft and patient survival; however, the influence of infection on the development of chronic allograft dysfunction has not been completely delineated. Some viral infections appear to affect allograft function by both inducing direct tissue damage and immunologically related injury, including acute rejection. In particular, this has been observed for cytomegalovirus (CMV) infection in all SOT recipients and for BK virus infection in kidney transplant recipients, for community‐acquired respiratory viruses in lung transplant recipients, and for hepatitis C virus in liver transplant recipients. The impact of bacterial and fungal infections is less clear, but bacterial urinary tract infections and respiratory tract colonization by Pseudomonas aeruginosa and Aspergillus spp appear to be correlated with higher rates of chronic allograft dysfunction in kidney and lung transplant recipients, respectively. Evidence supports the beneficial effects of the use of antiviral prophylaxis for CMV in improving allograft function and survival in SOT recipients. Nevertheless, there is still a need for prospective interventional trials assessing the potential effects of preventive and therapeutic strategies against bacterial and fungal infection for reducing or delaying the development of chronic allograft dysfunction.     

Cytomegalovirus infection accelerates cardiac allograft vasculopathy: correlation between angiographic and endomyocardial biopsy findings in heart transplant patients

In order to determine the impact of cytomegalovirus (CMV) infection on cardiac allograft vasculopathy (CAV), we quantitated angiograms and endomyocardial biopsy (EMB) specimens obtained from 53 heart transplant recipients. CMV infection was particularly associated with the development of discrete stenosis in major branch vessels (P<0.03). Also, the number of diffusely affected vessel segments was significantly higher in CMV patients than in CMV-free recipients after the 2nd post-operative year (P<0.05). The EMB histology correlated well with angiography. Significantly higher levels of arteriolar endothelial cell proliferation and intimal thickness were recorded in biopsies of CMV patients than in those of CMV-free recipients during the 1st postoperative year (P<0.02 and P<0.005, respectively). The CMV-associated vascular changes in EMB histology clearly preceded angiographically detectable CAV findings. Taken together, CMV infection accelerated heart allograft arteriosclerosis. The histological changes appeared prior to changes detected by coronary angiography. The CMV effect was particularly pronounced during the first 2 post-transplant years but leveled off thereafter. Thus, CMV-accelerated allograft arteriosclerosis may be linked in particular with early graft loss of CMV-infected heart transplant recipients.     

Hyperimmunoglobulin prophylaxis,monitoring and preemptive ganciclovir treatment eliminate the risk of CMV infection to improve patient and renal allograft survival

Abstract This study was designed to investigate whether the introduction of ganciclovir to clinical use for anti‐CMV treatment changes the risk of CMV infection in renal transplant patients. A total of 1545 cases who had received cadaveric renal transplants were divided into two groups: group 1 (n = 721) was made up of patients who received their transplants within 6 years before the introduction (1991) of ganciclovir and group 2 (n = 824), of individuals transplanted thereafter. Patient and graft survival of CMV D+/R‐ patients was uni‐ and multivariately compared with non‐CMV D+/R‐ patients. In CMV D+/R‐ patients in group 1, survival was significantly lower, and their relative risk for graft loss was 1.32‐fold (P = 0.0483) that of non‐CMV D+/R‐ patients. In group 2 patient and graft survival was identical regardless of whether the patients were at risk for CMV infection or not. The risk of CMV infection can be eliminated by hyperimmunoglobulin prophylaxis, CMV monitoring and preemptive ganciclovir treatment in renal transplant patients.     

Association between mannose-binding lectin deficiency and cytomegalovirus infection after kidney transplantation

Mannose-binding lectin (MBL) deficiency has been suggested as a risk factor for certain viral infections. However, there is no data about the possible association between MBL deficiency and cytomegalovirus (CMV) infection, especially after organ transplantation. We measured plasma MBL levels in 16 kidney transplant recipients with high-risk CMV serostatus (donor-positive/recipient-negative). MBL deficiency was diagnosed if MBL levels were <500 ng/mL. Of these 16 patients, seven developed CMV disease, four developed asymptomatic CMV infection, and in five patients CMV replication was not detected. Overall, 9 of 16 patients (56%) had MBL deficiency: five of seven (71%) patients with CMV disease, four of four (100%) patients with asymptomatic CMV infection, and zero of five (0%) patients without CMV infection (P=0.005; CMV infection/disease versus no infection). MBL deficiency may be a significant risk factor for the development of CMV infection in kidney transplant recipients, suggesting a role for innate immunity in the control of CMV infection after organ transplantation.     

Cytomegalovirus pneumonitis after kidney transplantation is not caused by plugging of cytomegalic endothelial cells only

In addition to life-threatening pneumonia, cytomegalovirus (CMV) may also cause subclinical pulmonary dysfunction after kidney transplantation. To investigate the role of plugging of cytomegalic endothelial cells in the pulmonary capillary bed, we prospectively determined specific carbon monoxide diffusion capacity (KCOc) and its components: the pulmonary diffusing membrane factor (Dm) and pulmonary capillary blood volume (Vcap) before and during CMV infection in 13 kidney transplant recipients and 13 controls. During CMV infection, mean KCOc decreased significantly by 28 % of the initial value (mean KCOc 79 vs 109; P < 0.005 ) due to a decrease in both Vcap and Dm. The KCOc in controls showed a significantly smaller decrease due to a slightly lower Vcap. We conclude that kidney transplant recipients with CMV infection have significant pulmonary diffusion disturbances due to a combination of lower Vcap and lower Dm. The most likely explanation for this phenomenon is a local inflammatory process due to CMV and not plugging of cytomegalic endothelial cells only. Received: 25 February 1998 Received after revision: 26 June 1998 Accepted: 22 September 1998     

Cause of End-Stage Renal Disease Is Not a Risk Factor for Cytomegalovirus Infection After Kidney Transplant

BackgroundCytomegalovirus infection (CMV) after kidney transplantation leads to increased morbidity and mortality. Whether the cause of end-stage renal disease (ESRD) influences the risk of CMV infection post-transplant is not known.MethodsWe analyzed data from 2741 adult kidney transplant recipients from January 1993 through December 2014. The causes of ESRD included diabetes mellitus (n = 947), hypertension (n = 442), polycystic kidney disease (n = 549), and glomerulonephritis (GN) (n = 803). The primary outcome was incidence of CMV infection, defined as the first episode of detectable CMV DNA in the blood following transplant.ResultsThree hundred and thirty patients developed a CMV infection over a median follow-up of 4.5 years. Patients with diabetes mellitus (DM) as the cause of ESRD had a higher incidence of CMV infection post-transplant compared to patients with GN (2.37 vs 1.58/100 person-years, P < .005) whereas hypertension (HTN) and autosomal dominant polycystic kidney disease (PKD) were similar (2.17 and 2.07/100 person-years). DM was associated with a 35% higher risk of CMV infection compared to GN in unadjusted analyses [hazard ratio=1.35 [95% confidence interval 1.02–1.78], P = .04). However, after adjustment for age, the risk of CMV infection was similar in all groups (DM: age-adjusted hazard ratio 1.02 [0.78–1.39]; HTN: 0.96 (0.67–1.36); PKD: 1.08 [0.78–1.48]; compared to GN). The risk of CMV infection increased with age (adjusted hazard ratio=1.32 [1.18–1.47] for every decade of life, P < .001).ConclusionsOur study demonstrates that the cause of ESRD is not a significant risk factor for CMV infection in kidney transplant recipients once adjusted for age. Future studies are needed to identify risk factors for CMV infection to define patient-centered monitoring and prevention.     

CMV specific T cell immunity predicts early viremia after liver transplantation

BackgroundCytomegalovirus (CMV) infection is one of the most important factors affecting liver transplant with direct and indirect effects. However, CMV disease after transplant remains poorly predicted.ObjectiveIn this study, preoperative CMV-specific cell-mediated immunity was evaluated in recipients of liver transplant in Korea, where most people are seropositive.MethodsA total of 32 patients were enrolled in a prospective study, and blood samples were collected before liver transplant to determine CMV-specific cell-mediated immunity. Testing using ELiSpot IFN-γ (CMVspot) and CMV serology were performed simultaneously.ResultsCMVspot results showed that 30 recipients had CMV-specific cell-mediated immunity, of which 29 were positive for phosphoprotein 65 and 14 for immediate early protein 1 (IE-1). All patients were positive for CMV IgG before transplantation, and 17 patients had a CMV viremia episode after transplantation. CMVspot showed 100% specificity and positive predictive value, and 11.76% sensitivity to predict CMV viremia. Patients with positive or borderline results for IE-1 did not show viremia two months after transplantation (p = .041).ConclusionCMVspot may be helpful in establishing a treatment strategy that includes regular monitoring for risk stratification of CMV reactivation.     

Intraoperative hypothermia is an independent risk factor for early cytomegalovirus infection in liver transplant recipients

BACKGROUND: Early cytomegalovirus (CMV) reactivation infections have been found to be associated with a number of stress responses such as severe bacterial infection. Intraoperative hypothermia is known to be a significant physiological stressor. Hence, we sought to evaluate a relationship between intraoperative hypothermia during liver transplantation and early CMV infection. METHODS: Relationships between intraoperative hypothermia and early CMV infection were assessed using univariate and multivariate analysis. RESULTS: There were 11 of 100 patients who had CMV infection detected within the first 30 days after the transplant and 16 of 100 in the next 3 months after the transplant. The median intraoperative temperature (34.4 degrees C) of those who subsequently developed early CMV infection was significantly lower than that in patients who did not develop early CMV infection (35.3 degrees C; P = 0.032). Multivariate analysis showed that only hypothermia was an independent predictor of early CMV infection (odds ratio 2.8, 95% confidence intervals 1.2-6.4; P = 0.047). CONCLUSIONS: Intraoperative hypothermia during liver transplantation increases the risk of CMV infection in the 1st month postoperatively and active warming seems to reduce this risk.