Materials and methods

A non-randomized, prospective, consecutive and monocenter study with deceased donor LTs was carried out during the period of January 2014 through December 2017. We only excluded live donor LTs.

We sent specimens of PF as well as a tissue sample for culture of fungi, aerobic and anaerobic bacteria once back-table was finished prior to implantation.

We considered positive PF the development of one or more germs in the samples and negative PF when there was no growth of germs after 5 days of cultures in the case of bacteria and 42 days for fungi. The isolation of methicillin-sensitive (MS) Staphylococcus, negative coagulase Staphylococcus, polymicrobial flora, corinebacterias and Streptococcus viridans were considered contamination. On the other hand, the fol lowing were classified as pathogenic: Staphylococcus aureus, methicillin-resistant Staphylococcus (MRSA), Streptococcus pyogenes, enterobacterias, Enterococcus, Gram-negative an aerobic bacilli, Gram-negative aerobic bacteria (Pseudomonas aeruginosa) and any fungi.

Broad-spectrum prophylaxis with 3 g of intravenous ampicillin-sulbactam was routinely indicated in all re cipients, 30 minutes before skin incision and four times a day up to 48 hours after surgery. In case of positivity in PF or in other donor cultures (routinely performed blood cultures or any other tissue or fluid considered potentially infected), antibiotic therapy was indicated in all cases ac cording to the corresponding antibiogram (except in cas es of positive cultures of PF considered contamination). No scheduled cultures were performed in recipients during their hospital stay in absence of suspected infection. When fever (axillar temperature ≥ 38° C) or hypothermia (axillar tem perature ≤ 36°), combined with any other sign or symptom of infection were present, such as tachypnea (> 20 breaths per minute), pCO₂ > 32mmHg, tachycardia (> 90 bpm), leukocy tosis (> 12 000 WBC/ml) or leukopenia (< 4000 cells/ml), or more than 10% immature neutrophils⁵, blood cultures and urine cultures were performed. Sputum, bronchoalveolar lavage, abdominal fluid or collections cultures were only performed after clinical or imaging suspicion of these localized infections. Percentage of positive cultures of PF were determined, distin guishing contamination from pathogenic germs. Isolated germs in each case were grouped by frequency of appearance.

We identified as “direct correlation” when the same germ with the same sensitivity and resistance to antimicrobials ac cording to antibiogram grew up in PF and in any culture from the recipient. No genotypic study was done to identify germs.

Demographic data of both groups were analyzed, consid ering age of the recipients, gender, body mass index (BMI), Model of End Stage Liver Disease (MELD) score, days on waiting list, blood transfusions during surgery and early orotra cheal extubation. Ninety days post-LT infections in both groups (positive PF, PF+; and PF-) were compared regardless of the isolated germ in the recipient. We classified infections as bacteremia, urinary tract infections (UTI), lower respiratory infections (tracheobronchitis, pneumonia and pleural effusion), superficial and deep surgical site infections (SSI) and catheter-associated infections according to WHO criteria⁶.

ICU and Hospital stay were analyzed in days. Donor risk factors for PF+ were analyzed, including age, sex, BMI, days of hospitalization and of mechanical ventilation and a history of positive blood cultures.

Statistical analysis was performed with SPSS 21 software. Continuous variables were expressed as mean and range and were analyzed between groups with Student and Mann- Whitney test. The categorical variables were presented by frequency and percentage and were compared with the chi square method and Fisher’s test. Survival was analyzed with Kaplan Meyer curve using the Long Rank for survival com parison. In all cases, the level of statistical significance was considered with a p < 0.05.

Results

Eighty eight LT were included in the observed period. Donors, PF cultures and postoperative evolution were analyzed. The mean age of recipients was 49 years (range 1-70). Demographic data is presentad in Table 1. PF cultures were positive in 33 cases (38%): twenty eight cases of PF+ (85%) were identified as contamination and 5 as infection (15%). These 5 cases received directed antibiotic therapy once the germ was isolated. Germs are detailed by frequency in Table 2.

Table 1 Demographic data of both groups, preservation fluid (PF) positive and negative 

Table 2 Frequency of isolated germs in preservation fluids 

PF+ group did not present higher infection rate in general within the first 90 days (38% vs. 35.5%, p 0.84). No case of direct correlation was identified.

ICU stay between study groups did not show significant differences (11.1 vs 8.17, p: 0.374) as well as hospital stay in days (19.4 vs 15.8, p 0.427).

Finally, no significant differences were found in donor risk factors in both groups (described in Table 3). Same germ in PF and donor blood culture was identified in only 1 case without direct correlation in recipient.

Table 3 Donor risks factors for PF positivity 

Discussion

Modern immunosuppressive regimens among other ad vances have increased LT graft survival by reducing the in cidence of acute and chronic rejection, with an counterpart increased risk of opportunistic infections in recipients². At the same time, the organ scarcity and thus unproportioned growth of the organ transplant list compared to available organs has increased the use of marginal donors⁷, includ ing those with an increased risk of infection transmission. However, very few reports have focused on the post-trans plant impact of PF cultures^3,4, being completely unknown in our region. Having post-LT infections a significative impact on both patient and graft survival^8-10, and the lack of pro spective studies on this subject, we consider ours relevant. For this purpose, the authors carried out this prospective study that showed a positivity rate of 33% in PF cultures, very similar to previous reports^3,4 requiring specific antibi otic treatment only in 15% of cases (pathogenic germs). We collected culture samples after back-table was fin ished. Although some groups described collecting cultures at different times³, we considered our approach practical for our aim to analyze impact in recipients outcomes re gardless the moment of contamination, to contemplate all moments of possible contamination within manipulation of the graft prior to implantation.

Antibiotic profilaxis prior to transplantation and up to 48 hours later would prevent transmition of germs con sidered as contamination in both, PF and surgical site¹¹. When pathogenic germs were isolated in PF, directed antibiotic therapy was supplied, without registering any “direct correlation” in recipients who presented any positive culture, in contrast to the results shown by Ce rutti et al, that registered a correlation in 4% of cases⁸. The treatment may have prevented the appearance of any transmition and therefore may have incurred a bias per se. However, in his work Cerutti does not specify if direct correlation was secondary to isolated germs in the preservation fluid or in any other donor cultures. Whenever a positive correlation is found, a genotypic study of isolated germs should be done if we want to confirm the same origin^12,13. However, the authors consider that in terms of costs and clinical relevance this study would not be necessary.

Regardless of correlation, positivity in PF (whether it was contamination or pathogenic, in which case targeted antibiotic treatment was implemented) did not incur in a higher rate of infections in recipient as well as in differ ences in ICU and total hospital stay, so we can conclude that should not be considered as a risk factor. Even so, a prospective work with higher number of patients is needed to confirm these.

We found no significate correlation between donor risks factors and PF positivity. In our study, mechanical ventila tion days and donor age did not increase the incidence, as described by Ruiz et all³. However, authors identified almost significant correlation in PF positivity when donors had a history of positive blood cultures (p = 0.06). Per haps, with a higher N of patients, such differences could become significant. A blood culture in one case devel oped the same germ as the preservation fluid (Klebsiella pneumoniae), antibiotic therapy was done in recipient and no direct correlation was observed, that’s mean, that the patients didn’t present that germen in any culture.

In conclusion, in our experience, when there was a positive PF for contamination germs, no treatment was required for the recipient, without a negative consequence in terms of infection. And when the PF fluid culture was positive for a pathogenic germ, if directed treatment was conducted, there were no infections in the recipient as well. We recommend routine culture of PF and management of positive PF cultures as outlined before.