Evaluation of procalcitonin (PCT) as a marker of infection in early post living donated liver transplant period.

BACKGROUND AND AIM
Procalcitonin (PCT) has been increasingly used as a biomarker of bacterial infection and as a tool to guide antimicrobial therapy. Despite its increased use, data in patients with solid organ transplants are limited. The study aimed to assess the frequency of rising PCT associated with infectious complications in immunosuppressed living donated liver transplantation.


METHODS
A single-center, retrospective observational study. Preoperative patients' demographic data, operative, anesthetic data, and postoperative clinical course were analyzed post-liver transplant (LT) till discharge from the intensive care unit.


RESULTS
Sixty patients were classified according to the culture results' into a positive culture group & a negative one and then followed up the sepsis variables in each group. Total leukocyte count (TLC) was elevated in the positive culture group in comparison to the negative culture one and was statistically significant (P-value <0.05) till the fourth day postoperative. Procalcitonin was higher in the positive culture group than in the negative one on days 1, 3, and 5 postoperative and was statistically significant (P-value <0.05). The cutoff values in the receiver operating characteristic curve (ROC) with >90% specificity to infection post LT were PCT of ≥9 ng/ml and TLC of ≥17.3/mm3 on day one.


CONCLUSIONS
Following up PCT level on day one with TLC is essential and will help to detect sepsis and guide early antimicrobial initiation post-liver transplantation. Combined measurements of PCT and TLC with cutoff values of <9 ng/ml and < 17.3/mm3 respectively will help to exclude infections in 83.7% of patients, thus avoiding unnecessary usage of higher generations empiric antimicrobials.

limited. Even without the presence of infection, PCT increases as a result of surgical procedures during transplantation, implantation of devices, and use of induction immunosuppressive therapy. 4 Aim Of The Work: The study aim is to determine the association between rising serum procalcitonin and the occurrence of infectious complications in immunosuppressed LDLTRx.

Patients & Methods:
Study type: This is a single center, retrospective study in ICU-NHTMRI (Clinicaltrials.gov registration number: NCT03389360).

Inclusion criteria:
All patients that were included were adults aged between 18 to 65 years old who are recipients of living donated liver transplantation at the Department of Liver Transplant, between January 2015 and January 2018 and with no contraindications for early (immediately postoperative) immunosuppression.

Exclusion criteria:
Patients were excluded from the study if they were repeatedly admitted to ICU after rst admission.

*Early post-transplant infections:
Early infections will be de ned from 0-30 days after transplant. Bacteria and yeast are the most frequent pathogens in the rst 30 days after transplant. While post-transplant period will be de ned as the period since transplant till less than 4 weeks. 4 Symptoms & signs of infection post liver transplant may include generally; malaise, chills, low grade fever, and rigor. However because of their immunosuppressed state, transplant recipients may not manifest fever as readily as the general population. Speci c signs according to the site of infection include e.g. pulmonary infections; cough, dyspnea, and chest pain. Urinary tract infections; dysuria, pyuria, and hematuria, while intra-abdominal & surgical site infections; abdominal pain, erythema at incision site, and/or elevated liver enzymes. 5 Study Design: The data in this study was collected from the medical records; showing that all patients transferred from operative theatre to the ICU were sedated, intubated, and ventilated. They received empiric antimicrobial prophylaxis (piperacillin-tazobactam;ICU\ICU Protocol\Liver transplant protocol\Adult LT\Protocol for ICU management of.docx) upon admission and early immunosuppression (from day zero) according to the clinical practice guidelines in our center; they received Tacrolimus with Methyl Prednisolone with/without Mycophenolate Mofetil. Preoperative patients' demographic data were obtained. Also, Child-Pugh score prior transplant, and the primary cause of liver transplant.
Operative and anesthetic details as; operation time, units of blood, blood products transfusion, ischemic time, type of preservatives used, back table procedures, extra-hepatic procedure, and graft to recipient weight ratio (GRWR) were recorded.
Postoperative patient evaluation that included; Sequential Organ Failure Assessment (SOFA) on admission and /48h, hemodynamic monitoring, including hourly measurement of heart rate, mean arterial pressure (MAP), temperature, central venous pressure (CVP), arterial oxygen saturation (SaO 2 ), daily total volume of uid infused, daily urine output, uid balance daily, blood gas analysis / 6h, and daily mean values were recorded till hospital discharge.
Routine laboratory work-up included biochemical markers of liver, kidney function and hematological parameters (complete blood count and coagulation pro le) were obtained.
Also routine patient evaluation for infection was obtained from the medical records through scheduled measurements of PCT every other day during ICU stay and upon needed during hospital stay. Other markers of infections were recorded till hospital discharge as C-reactive protein (CRP) every other day, daily total leukocyte count (TLC) and band cells %.
Microbiological evidence of infection was con rmed by cultures that were followed from the medical records as they were regularly sampled every 48-72 hours during ICU stay and upon any clinical, laboratory biomarker (TLC, band cells, CRP, and PCT), and/or radiological ndings (e.g. pulmonary in ltrates in chest X-ray, cholangitis or hepatic abscess by CT abdomen) suggestive of infection. Management of suspected infection as shown by the medical records; when PCT value was elevated with clinical, radiological and/or laboratory evidence of infection (TLC, CRP, band %), culture from the suspected site of infection was withdrawn. If the source of infection was not evident, cultures from blood, urine, sputum, surgical wound, drains and nasal swab were obtained and empiric antimicrobial against gram positive bacteria was added (as indicated by the local antibiogram). While the management of proven infection will be a culture based antimicrobial initiation.

Primary outcome measures:
The primary outcome of the study is to determine the association between rising procalcitonin and the occurrence of infectious complications in immunosuppressed LDLTR.

Secondary outcome measures:
To detect the non-infectious causes of rising procalcitonin post living donated liver transplant, its relation to the length of ICU and hospital stay.

Statistical Analysis:
Data were analyzed using SPSS (Statistical Package for Social Sciences; SPSS Inc., Chicago, IL, USA) version 21 for Microsoft Windows. Numerical data were presented as mean ±SD, median and range as appropriate. Categorical data were presented as frequency and percentage. Numerical data were explored for normality using Kolmogrov-Smirnov test and Shapiro-Wilk test. Comparisons between the two groups for normally distributed numerical variables were done using the Student's t-test while for non-parametric numeric variables Mann-Whitney test was used. Associations between qualitative data were done using Chi-square test or Fisher's exact test as appropriate. The receiver operating characteristic (ROC) curve was conducted for PCT to calculate sensitivity, speci city, positive predictive value (PPV), and negative predictive value (NPV). Probability (p-value) equal or less than 0.05 is considered signi cant.

Results:
Patients' demographic variables: Sixty patients enrolled in the study underwent liver transplantation. They were categorized mainly according to the culture results withdrawn in the early postoperative care (in the rst ICU admission) into; the positive culture group and negative culture group. Seventeen patients had positive culture results and the remainder had negative results.
Comparison of the remaining demographic data (Table 1) could not be performed due to either large number of groups or small number of patients per group.
In table 2 (anesthetic and operative variables), only the amount of blood loss and blood transfused intraoperative had the association with positive culture results and they were statistically signi cant (Pvalue <0.05).

Laboratory variables:
Regarding the laboratory variables (shown in table 3) used to help diagnose infection; only total leukocyte count (TLC) was elevated in the positive culture group in comparison to the negative culture one and was statistically signi cant (P-value <0.05) till the fourth day postoperative. While band cells% & CRP was not statistically signi cant in both groups.
Procalcitonin assessed in both groups (Table 4); it was higher in the positive culture group than in the negative one on days 1, 3, and 5 postoperative and was statistically signi cant (P-value <0.05). Table 5, showed the cutoff value of PCT in (ROC) on day one postoperative was ≥ 9ng/ml, it had the sensitivity of 52.9% while the speci city of 83.7%. The positive predictive value (PPV) was 56.3%, the negative predictive value (NPV) was 81.8%, and the overall accuracy of the test was 75%.
According to the cutoff of PCT ≥ 9ng/ml, TLC cutoff value on day one was determined to be of ≥ The need for re-ventilation after extubation was (N= 1/43, 2.3%) in the negative culture group and was (N= 1/17, 5.9%) in the positive culture group. One patient (2.3%) developed renal impairment requiring renal replacement therapy in the negative culture group while it was (0.0%) in a positive culture group. Therefore, the noninfectious complications developed and recorded to be studied in this research was low enough to allow the statistical analysis.
Regarding the impact of PCT at the cutoff value of ≥9 and < 9ng/ml on ICU stay, was signi cant only on day three as expressed by median (Min-Max), were 7(4-17) days and 6(2-13) days respectively and it was statistically signi cant (P-value= 0.05). While the impact of PCT at the cutoff value of ≥9ng/ml on hospital stay, was not statistically signi cant (P-value > 0.05).

Discussion:
Diagnosis of infection post solid organ transplantation should be as early as possible, as the delay of antimicrobial administration will lead to potentially long term effects on the morbidity and mortality.
While initiation of antimicrobials to treat colonization will expose both patients & centers to resistance, increase cost, and drug side effects. 4 PCT is used as a marker of early bacterial translocation in some group of patients. 6 However, in solid organ transplantation (SOT), PCT levels may be affected by other factors e.g. the surgical procedure, underlying disease, and immunosuppression used that interfere with its interpretation & makes it more challenging in the transplant group. [7][8][9] Procalcitonin is the precursor molecule of human calcitonin, its molecular weight is about 13KD & with no hormonal activity. 10 In normal individuals, it is present in very low concentration about 0.5ng/ml. However, during bacterial infection & multi-organ failure, it will exceed this value to reach 100ng/ml. 11 Procalcitonin has half-life about 24hours while calcitonin has a shorter half-life about 10-20 minutes. 10 Regarding the primary outcome of the study is to determine the association between rising serum procalcitonin and infectious complications in immunosuppressed LDLTRx.
So, the patients were classi ed into the positive culture group & the negative one then retrospectively following up the sepsis laboratory variables in each group (mainly TLC, Band cells%, CRP and PCT) to allow nding simple, rapid marker that would be suitable post LDLTX to diagnose clinically signi cant infections.
PCT levels were higher with statistically signi cant (P-value < 0.05) in the positive culture group from day one till day 5 postoperative. The cutoff value that was highly speci c to infection, ≥ 9ng/ml (speci city of 83.7% on day one postoperative and speci city was highest on day three 90.7%).
Early post LTx, the cutoff value of PCT that would be associated with infection has not been de ned, depending on the type of allograft and the extent of the surgery, so PCT levels may increase or remain stable after transplantation. 12 Perrakis et al. found that post LTx, PCT values > 5ng/mL increased 11.7 times the odds of developing complications e.g. infectious complications, renal failure, bleeding, and respiratory failure. 13 As the literature in the setting of liver or kidney transplantation is not extensive, so, in heart transplant recipients, PCT with cutoff values > 0.6ng/ml was correlated with local infection while in multiple infections & sepsis PCT levels were substantially elevated to (7.3-22.4ng/ml). 14 Chen et al. showed that PCT can be used to predict catheter-related bloodstream infections post liver transplantation (LTx), where the PCT values > 3.1ng/ml had a sensitivity of 72% and speci city of 87% to predict the infection. 15 Study of lung transplant patients, using the PCT to differentiate between colonization & infection where PCT was mildly elevated in colonization with cutoff value <2 and >0.5ng/ml. 16,17 Kunz et al. measured PCT immediately before and daily after liver transplantation in 22 patients and found that all patients had increased PCT levels without clinical signs of infection. 18 Prieto et al. reported a cutoff value of 1.92ng/mL as a predictor for both infectious and non-infectious complications with a sensitivity of 95.6% and speci city of 89.5%, although those who suffered complications had worse preoperative criteria and higher Child-Pugh scores. 19 Regarding to PCT dynamics, in the study of heart and lung recipients, PCT level was high in the rst 24hours postoperative & it remained high in those patients who suffered infectious complications, [20][21][22] PCT Level reached (mean ±SD),(54.6 ±8.8ng/ml) in patients with complications, compared to (9.1 ±9.3ng/ml) in patients without complications. At the same time TLC and CRP could not differentiate between patients with infections from those without. 20,21 Thus, it is advised to follow PCT level after 24h reading post-transplant.
In this study, TLC was higher in the positive culture group and statistically signi cant (P-value < 0.05) from day one till the fourth day postoperative. TLC cutoff value of ≥ 17.3/mm 3 on day one; had the speci city of > 90%. Therefore & according to our sample, both PCT & TLC can help early rapid diagnosis of infection till culture results will be available. While both CRP and band cells% were not differ in between the positive culture group and the negative one.
Although CRP results were high in both groups of culture results but were not statistically signi cant, this rise could be related to the surgery, underlying disease, or blood transfusion.
CRP level and leukocyte counts could not be able to differentiate between infectious and non-infectious complications at any time point. 19,20 Regarding outcome at PCT cutoff value 0f ≥9ng/ml, ICU stay on day three was correlated with that cutoff & was statistically signi cant (P-value= 0.05). While hospital stay was not differ at either PCT cutoff value of ≥ 9ngml or < 9ng/ml & it might be due to the small number of patients enrolled.
Study done for heart transplant recipients, found that increased PCT level to > 10ng/mL was associated with poor outcomes. 22 While following the demographic, anesthetic and operative variables; the amount of blood loss and amount of blood transfused were correlated with the positive culture results and they were statistically signi cant (P-value= 0.001 & 0.029 respectively). These results matches another study done where dominant elective abdominal surgeries were included and showed that the number of units of blood transfused was directly proportional to the incidence of complications and mortality; infectious complications represented the highest incidence 36.3%. 23 Therefore, the surgical and anaesthetic techniques that aid to limit blood loss and blood transfusion intraoperative will positively affect the incidence of postoperative infectious complications.

Conclusion:
In conclusion, according to the sample size in this study; following up both PCT level and TLC on day one is essential and will help to detect patients that have sepsis or multiple infections to guide the early antimicrobial initiation till culture results available.
According to the speci city of cutoff values of both PCT and TLC, we conclude that in low income countries and in the same population we can use TLC without PCT level on day one post LTx with a cutoff value ≥ 17.3/mm 3 to help early diagnosis of infection and administration of antimicrobials.
Recommendations is for further researches with larger sample size in liver transplant recipients to investigate any difference in the cutoff value in comparison to this research and to search about the impact of using PCT on hospital stay, correlation with noninfectious complications, and mortality.

Declarations
Ethics approval and consent to participate: The study le submitted for approval by NHTMRI IRB. Due to the retrospective nature of the research, the research participants' medical data were collected anonymously from patients' le by a third party. Reconstructing of patients will be di cult and will hinder research conduction, so we request the waiving of informed consent from the IRB. The study had the approval on February 2018.
This study was conducted in accordance with the declaration of Helsinki.