Infections Drive a Heavier Inpatient Burden During Liver Transplant Hospitalization
Infections complicated nearly one-third of adult liver transplant hospitalizations in a large national cohort, and their presence carried a clear inpatient penalty. Among 50,630 transplant-related admissions captured from the National Inpatient Sample between 2016 and 2022, 15,130 hospitalizations, or 29.9%, included at least one infection. The most frequent diagnoses were pneumonia, urinary tract infection, and cholangitis, a distribution that fits the clinical vulnerability of transplant candidates and recipients whose physiology is already strained by advanced liver disease, portal hypertension, and perioperative immune dysfunction.
The signal was not subtle. After adjustment for demographic factors, hospital characteristics, comorbidity burden, liver disease etiology, decompensation status, and year of hospitalization, infected admissions had higher odds of all-cause mortality than noninfected admissions. The adjusted odds ratio was 1.47, with a 95% confidence interval of 1.14 to 1.90 and a p value of 0.003. In the same adjusted models, infection also tracked with more frequent non-home discharge and a substantially higher complication burden, with adjusted odds ratios of 1.89 and 1.65, respectively.
Pneumonia, Urinary Tract Infection, and Cholangitis Accounted for Most Cases
Three infection categories dominated the inpatient picture. Pneumonia accounted for 8.5% of infected hospitalizations, urinary tract infection for 8.3%, and cholangitis for 7.5%. Each of these conditions can destabilize an already fragile transplant admission, but they do so through different pathways. Pneumonia threatens oxygenation and ventilatory reserve during and after surgery. Urinary tract infection can progress rapidly in immunosuppressed patients and may seed bacteremia. Cholangitis is especially concerning in the liver transplant setting because biliary complications and bacterial contamination of the hepatobiliary tree can amplify systemic inflammation at exactly the wrong moment.
A single infection can alter recovery. Multiple infections can reshape the admission.
This study also assessed outcomes by the number of documented infection types during the index hospitalization, adding an important clinical dimension. Even without granular bedside chart review, the pattern is biologically plausible: as infection burden accumulates, so does physiologic stress, organ dysfunction risk, and the likelihood that a patient cannot return directly home after discharge. In a population already selected for extreme illness severity, the additive burden of infection likely magnifies the consequences of anemia, renal dysfunction, vasopressor exposure, mechanical ventilation, and immune dysregulation.
Longer Stays and Higher Costs Followed the Infection Signal
Resource utilization rose sharply when infection was present. Adjusted predicted length of stay was 30 days in the infected group versus 16.2 days in the noninfected group. Adjusted predicted hospitalization cost was $235,953 compared with $156,336. Those differences are clinically meaningful because they capture more than administrative spending; they reflect prolonged exposure to hospital-acquired complications, delayed rehabilitation, more intensive monitoring, and a more difficult path to discharge readiness.
The cost gap also reflects the nature of liver transplantation itself. These admissions often involve complex preoperative optimization, intraoperative hemodynamic instability, and postoperative surveillance for graft function, rejection, bleeding, renal injury, and biliary complications. When infection is layered on top, the admission becomes harder to compress into a predictable recovery window. Broad-spectrum antibiotics, diagnostic imaging, repeat cultures, respiratory support, and renal replacement therapy can all extend the stay and increase the cost curve. The administrative burden is visible in the numbers; the bedside burden is the reason those numbers move.
Why Infection Hits Liver Transplant Patients So Hard
The pathophysiology behind these findings is straightforward. Patients undergoing liver transplantation often live with impaired immunity, gut-barrier disruption, bacterial translocation, and dysbiosis before surgery even begins. Those features create an internal environment in which infection is easier to acquire and harder to contain. Once transplantation occurs, immunosuppressive therapy preserves the graft but also lowers the host response needed to clear microbial threats. The resulting balance is precarious: too little immune control risks rejection, while too much suppression risks invasive infection.
The clinical implications extend beyond the immediate admission. Early post-transplant infections remain a major cause of mortality, and infections during the index hospitalization may be an early marker of a patient destined for a complicated postoperative course. In practice, that means transplant teams need heightened vigilance for pulmonary, urinary, biliary, and bloodstream infections as soon as the patient enters the transplant pathway. It also argues for aggressive source control, careful antimicrobial selection, and rapid recognition of decompensation in patients with recent or active infection.
A National Dataset, but Not a Bedside Chart
The analysis drew on a 20% stratified sample of U.S. hospital discharges from the Healthcare Cost and Utilization Project National Inpatient Sample, using ICD-10-CM procedure code 0FY00Z0 to identify adult liver transplant hospitalizations. This design provides scale and national breadth, which are major strengths when studying a complication as common and consequential as infection. It also allowed adjustment for a broad set of covariates, including age, sex, race, insurance, income quartile, hospital characteristics, liver disease cause, liver-related decompensations, and comorbidity burden measured by the modified Charlson comorbidity index.
Administrative data, however, cannot replace bedside detail. The database cannot fully resolve infection timing, severity, microbiology, antimicrobial exposure, graft function, or whether an infection began before admission or emerged during the hospitalization. It also cannot capture the clinical judgment behind proceeding with transplantation in the setting of recent infection. Those gaps matter when interpreting causality. The associations are strong and biologically credible, but they still describe population-level risk rather than an individual patient trajectory.
Preventive Strategy Needs to Start Before the Operative Date
The findings support a practical message for transplant programs: infection prevention around liver transplantation is not a peripheral quality metric. It is central to survival, discharge disposition, and resource use. Preoperative screening for active infection, prompt treatment of bacterial sources, pulmonary optimization, strict line care, and early postoperative surveillance all fit within that imperative. For patients with cholangitis, pneumonia, urinary infection, or bloodstream infection, the threshold for escalation should remain low because delay can quickly translate into prolonged hospitalization and higher mortality risk.
In this cohort, infection did not behave like a minor complication. It aligned with more deaths, more complications, less frequent discharge home, longer stays, and higher costs. The unanswered question is not whether infection matters in liver transplantation; it is how best to reduce it across the entire transplant pathway while preserving timely access to surgery for patients whose liver disease leaves little room for delay.
- • PLOS ONE study DOI 10.1371/journal.pone.0363042: https://doi.org/10.1371/journal.pone.0363042