Objectives: Pneumonia is a significant cause of
morbidity and mortality in solid-organ transplant recipients. We studied the demographic characteristics, respiratory management, and outcomes of solid-organ transplant recipients with pneumonia in an intensive care unit. Materials and Methods: There have been 2857 kidney, 687 liver, and 142 heart transplants performed between October 16, 1985, and February 28, 2021, at our center. We retrospectively analyzed records for 51 of 193 recipients with pneumonia during the posttransplant period between January 1, 2016, and December 31, 2018. Results: Fifty-one of 193 recipients were followed in the intensive care unit. Mean age was 45.4 ± 16.6 years among 42 male (82.4%) and 9 female (17.6%) recipients. Twenty-six patients (51%) underwent kidney transplant, 14 (27.5%) liver transplant, 7 (13.7%) heart transplant, and 4 (7.8%) combined kidney and liver transplant. Most pneumonia episodes occurred 6 months after transplant (70.6%) with acute hypoxemic respiratory failure. Mean Acute Physiology and Chronic Health Evaluation System II score was 18.9 ± 7.7, and the Sequential Organ Failure Assessment score was 8.5 ± 3.9 at intensive care unit admission. Whereas 66.7% of pneumonia cases were nosocomial acquired, 33.3% were community acquired. The intensive care unit and 28-day mortality rates were 39.2% and 64.7%, respectively. Conclusions: Solid-organ transplant recipients with pneumonia have been associated with poor prognosis. Our cohort followed in the intensive care unit comprised mostly patients with nosocomial pneumonia with acute hypoxemic respiratory failure, hospitalized 6 months after transplant with high Acute Physiology and Chronic Health Evaluation System II scores predictive of mortality. In this high-risk patient group, careful follow-up, early discovery of warning signs, and rapid treatment initiation could improve the outcomes in the intensive care unit.
Key words : Heart transplant, Kidney transplant, Liver transplant
Pneumonia is an inflammation of the alveolar space, most commonly triggered by bacteria but also caused by other classes of pathogens and less frequently by autoimmune processes. Pneumonia is characterized by the infiltration of the alveolar space by leukocytes and a fibrinous exudate that leads to lung dysfunction; in its more severe forms, it may require invasive mechanical ventilation and admission to an intensive care unit (ICU).1 Estimated mortality from pneumonia in European countries ranges from 1% to 48%. Pneumonia is the most common infectious cause of admission to the ICU, with a mortality rate of 35% to 70%.2-4
Immunosuppression treatment regimens and prophylactic strategies for patients after solid-organ transplant (SOT) have improved with a target of minimizing toxicity and adverse effects while optimizing organ function. However, infections, especially pneumonia, are still life-threatening complications in SOT recipients under treatment with chronic immunosuppression therapy.5,6 Pneumonia is a significant cause of morbidity and mortality in this population, with an incidence of 19.4 episodes per 1000 population on follow-up per year and may lead to death in SOT patients. The occurrence of posttransplant pneumonia adversely affects both graft and recipient survival, as well as the cost of care for SOT recipients. Many microorganisms may cause pneumonia in SOT recipients. Some infectious etiologies are self-limited, whereas others may cause significant morbidity and mortality.7 Therefore, these patients may need mechanical ventilation and follow-up in ICU.
In the present study, we aimed to determine the demographic characteristics, comorbidities, management of respiratory complications, intubation and mechanical ventilation requirements, length of stay (LOS) in the ICU and hospital, and outcomes of early-onset and late-onset pneumonia in SOT recipients who were followed in the ICU.
Materials and Methods
This study was approved by the Baskent University Institutional Review Board (project No. KA20/65). A total of 2857 kidney, 687 liver, and 142 heart transplants have been performed at our center between October 16, 1985, and February 28, 2021. We retrospectively analyzed the medical records of SOT recipients with pneumonia who were followed in the ICU during the posttransplant period between January 1, 2016, and December 31, 2018.
Solid-organ transplant recipients who were admitted to the hospital with relevant symptoms after transplant (such as fever, cough, sputum production, chest pain, dyspnea, and respiratory failure), those with new-onset or progressed pulmonary infiltration as shown by chest radiography accompanied by a physical examination, and those with similar symptoms and signs that occurred during the stay in hospital or ICU were considered to have pneumonia. Patients younger than 15 years and those for whom data were not available were excluded.
Diagnosis of pneumonia was defined as the onset of pulmonary infiltrate as shown in chest radiographs and/or computed tomography scans of the thorax plus at least 2 of the following criteria: fever >38 °C, cough, purulent sputum, dyspnea or >20 breaths/min, pleuritic chest pain, and leukocyte count of >10 000/mm3 or <4000/mm3. Clinical, laboratory, pathologic, and radiologic findings, positive microbiology test results from respiratory specimens, and details of effective clinical treatment approaches were obtained from the data recorded at the time of diagnosis of pneumonia. Chest radiographs were independently reviewed by 1 pulmonologist and 1 radiologist, both of whom were blinded to each other and blinded to each patient’s clinical course. Diagnoses were classified as community-acquired pneumonia (CAP), hospital-acquired pneumonia (HAP), ventilator-associated pneumonia (VAP), or health care-associated pneumonia (HCAP), according to the criteria of the Infectious Diseases Society of America/American Thoracic Society applied to non-SOT patients, to investigate whether this classification system was applicable to SOT recipients. All episodes diagnosed within 48 hours of admission in patients who did not fulfill the criteria for HCAP, HAP, or VAP were considered CAP.6,7 Diagnoses were also categorized as either nosocomial-acquired pneumonia (HAP, VAP, and HCAP) or community-acquired pneumonia.8
The ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (Pao2/FIo2 ratio) is a parameter that evaluates hypoxemic respiratory failure. If this ratio is below 300, then there is hypoxemic respiratory failure, which is classified as mild, moderate, or severe.9,10
The following data were obtained from patient medical records: age, sex, transplant organ type, comorbidity, immunosuppression regimen, number and time of pneumonia episodes, Acute Physiology and Chronic Health Evaluation System (APACHE II) score, Sequential Organ Failure Assessment (SOFA) score, Glasgow Coma Scale (GCS) score, microbiologic diagnostic method, microorganism type, respiratory support option, need for mechanical ventilation (MV) and tracheotomy, mode of MV, level of FIo2, Pao2/FIo2 ratio, level of positive end-expiratory pressure (PEEP), laboratory values, LOS in hospital and ICU, and ICU and 28-day mortality rates.
Statistical analyses were performed with SPSS (version 25.0). Variables are expressed as mean values ± SD. Frequencies are expressed as number
of patients and percentage (of total). Categorical variables between the groups were analyzed with the chi-square test or the Fisher exact test. For significant differences between 3 or 4 groups, post hoc multiple comparison tests were applied. The independent samples median test was applied for nonparametric data. Factors associated with mortality were determined by univariate and multiple logistic regression analyses. P < .05 was considered statistically significant.
Fifty-one of 193 SOT recipients were followed in
the ICU. The mean age was 45.4 ± 16.6 years (range, 15-73 years) including 42 male (82.4%) and 9 female (17.6%) patients. Eight patients (15.7%) had smoking history. Diabetes mellitus and chronic heart disease were the most common comorbidities (19.6%). Twenty-six patients underwent kidney transplant (51%), 14 patients underwent liver transplant (27.5%), 7 patients underwent heart transplant (13.7%), and 4 patients underwent combined kidney and liver transplant (7.8%). Eighteen of the transplanted organs were from living donors (35.3%), and 33 were from deceased donors (64.7%). Most of the pneumonia episodes occurred more than 6 months after transplant (70.6%) with acute hypoxemic respiratory failure (Table 1). There were 28 patients who had more than 2 episodes of pneumonia (54.9%). Dyspnea was the most common symptom (64.7%).
The mean APACHE II score was 18.9 ± 7.7, the mean GCS score was 12.5 ± 3.5, and the mean SOFA score was 8.5 ± 3.9 at ICU admission (Table 1).
Whereas 66.7% of pneumonia cases were nosocomial acquired (HAP, VAP, and HCAP), 33.3% were community acquired. The incidence varied according to the type of nosocomial pneumonia, for which 37.3%, 21.6%, and 7.8% of SOT recipients were diagnosed with HAP, VAP, and HCAP, respectively. The microbiologic diagnostic method was tracheal aspirate in 52.9% of patients and bronchoalveolar lavage in 21.6% of patients. The main pneumonia etiologies were bacterial (52.9%, n = 27), viral (23.5%, n = 12), and fungal infections (25.5%, n = 13). No microorganism was isolated in 29.4% of pneumonia episodes. The most common immunosuppression regimen was the combination of prednisolone, mycophenolate mofetil, and tacrolimus (54.9%) (Table 2).
The Pao2/FIo2 ratio was below 300 in 41 patients (80.4%). The FIo2 was ≥50 in 35 patients. Forty-one patients had acute hypoxemic respiratory failure (80.4%), classified as mild in 15 patients, moderate in 25 patients, and severe in 1 patient. Nine patients received only low-flow oxygen therapy (nasal/mask; 17.6%), and 6 patients received high-flow oxygen therapy (11.8%). Thirty-three (64.7%) patients were intubated, and 16 patients (31.4%) required tracheotomy. Twenty patients were treated with only invasive mechanical ventilation (39.2%), and 7 patients were treated with noninvasive mechanical ventilation (15%). The most common PEEP value was 8 cm H2O (range, 5-15 cm H2O). The most common mode of invasive mechanical ventilation was pressure-synchronized intermittent mandatory ventilation (Table 3).
The mean levels of leukocytes, C-reactive protein (CRP), and procalcitonin were, respectively, 11.6 ± 8.8 103/μL, 150 ± 119.5 mg/L, and 31.4 ± 128 μg/L at the diagnosis of pneumonia. The mean hemoglobin level (10.8 ± 1.9 mg/dL) was higher in renal transplant recipients than in other SOT recipients (Table 4).
The mean duration of MV was 22.6 ± 35.8 days. The mean values for LOS in the ICU and hospital after pneumonia were 26.4 ± 74.7 and 24.1 ± 26.8 days, respectively. The mean LOS in ICU (30.5 ± 34.5 days) and hospital (63.8 ± 58.9 days) before VAP and LOS at ICU (38 ± 47.3 days) after VAP were longer than for other pneumonia types. The mean values for LOS at ICU (18 ± 38.7 days) and hospital (5.3 ± 17 days) before the diagnosis of pneumonia were the shortest in the SOT recipients who had pneumonia more than 6 months after transplant (Table 5).
The ICU mortality rate was 39.2%, and the 28-day mortality rate was 64.7%, both of which were higher than the expected mortality rate of 32.2% as calculated from the mean APACHE II score.
There were no statistically significant differences for mortality compared with transplant organ type and time to occurrence of posttransplant pneumonia. Among the risk factors that may affect mortality, the pneumonia type, level of PEEP, and endotracheal intubation were first evaluated with univariate analysis and then with multiple logistic regression analysis. Accordingly, endotracheal intubation alone increased mortality by 50 times (odds ratio, 50; 95% CI, 8.98-278.14). The level of PEEP was also considered as a factor associated with mortality (odds ratio, 1.53; 95% CI, 1.21-1.94) because endotracheal intubation is associated with mortality. No difference was found between pneumonia types by logistic regression analysis (Table 6). However, VAP had a significantly higher mortality rate (30.3%) than other pneumonia types. The mean level of PEEP was higher (8.7 ± 3.3 cm H2O) in nonsurvivors compared with survivors (Table 7).
In this study, we investigated ICU follow-up of SOT recipients with pneumonia during the posttransplant period. Fifty-one patients with pneumonia in the ICU during the study period were evaluated. Our patients were admitted at least 6 months after transplant, and more than half were nosocomial-acquired cases with bacterial etiology. Invasive mechanical ventilation was used to treat acute hypoxemic respiratory failure in 64.7% of SOT recipients. The mean level of PEEP was significantly higher in nonsurvivors compared with survivors. Endotracheal intubation and the high level of PEEP were associated with poor clinical prognosis. We also documented that LOS at hospital and ICU before pneumonia were associated with the development of pneumonia. Ventilator-associated pneumonia had a significantly higher mortality rate compared with other pneumonia types. The ICU mortality rate was 39.2%, and 28-day mortality rate was 64.7%.
Although early posttransplant infections are common findings, reports that argue the opposite, including reports from our group, are numerous in the recent literature.5-7,11,12 The present study detected that 70.6% of pneumonia episodes occurred more than 6 months after transplant. The reason for this incidence rate may be the more frequent empirical use of antibiotics, shorter duration of postoperative mechanical ventilation, and improved attention to prevention of infections. As in our study, other recent studies have reported that more than half of the episodes occurred late after SOT (>6 months).6,7
After 6 months, community-acquired infections are common because by that time the level of immunosuppression may be reduced in response to sufficient allograft function.7,9 Contrary to this classic knowledge, we report here that more than half of pneumonia cases were nosocomial acquired (HAP, VAP, and HCAP) and only a third of cases were community acquired, similar to the results reported by Giannella and colleagues.6 The reason for this incidence rate may be that our patient population had comorbid diseases that required frequent hospital visits or hospitalization. It has been documented that nosocomial infections and the need for mechanical ventilation significantly increase the mortality of pneumonia in SOT recipients,14 which may be the result of more aggressive, treatment-resistant microorganisms. Herein, we documented that VAP was associated with increased mortality, which confirms previous results.14,15
Although incidence rates of HAP and VAP may be as high as 22% in the general population, this ratio may vary in SOT recipients according to organ type.4 In heart transplant recipients, this incidence rate has been reported to be 15%, in liver transplant recipients 5% to 48%, and in kidney transplant recipients 4% to 16%.16-20 We found the CAP incidence to be 33.3%. Therefore, our total combined incidence rate of HAP, VAP, and HCAP reached 66.7%, possibly because we have included all SOT recipients in our study. Both the mixed patient population and the calculation of all HCAP values together may be the cause of the higher results shown in our study.
Although bacterial agents are observed most frequently with microbiologic etiology, viral and fungal factors play an important role. Early discovery of opportunistic infections is crucial for successful treatment of these immunosuppressed patients. Flexible bronchoscopy is an invaluable tool for assessment of lung infections in SOT recipients.6,8 In this study, microbiologic agents were detected in 82.4% of the study population, with test samples obtained via deep tracheal aspirate in 27 patients and via fiberoptic bronchoscopy in 11 patients. Most of our patients had a low Pao2/FIo2 ratio, that is, less than 300, and so fiberoptic bronchoscopy, which is the gold standard for assessment of lung infections, was a crucial tool for this procedure to avoid additional respiratory distress in these 11 patients. The main pneumonia etiologies were bacterial (52.9%), followed by viral and fungal infections. No microorganisms were detected in 29.4% of pneumonia episodes in our study. Consistent with this study, Giannella and colleagues reported that the most common etiology for pneumonia was bacterial, followed by viral etiology.6 The incidence of tuberculosis in SOT recipients was 20 to 75 times higher than the rate in the general population based on endemicity. Although the study was conducted in an endemic country, incidence of tuberculosis was not determined. Although Dorschner and colleagues reported that the causative pathogens of pneumonia were different according to organ type, we found no difference in the causative pathogens for pneumonia that developed in recipients of heart, liver, and kidney transplants in our study. The reason for the difference between our findings and those of Dorschner and colleagues may that we did not include lung transplant recipients in our study. Another possible reason may be the short postoperative time span (only 1 month) in the study of Dorschner and colleagues.21
Cases of respiratory insufficiency due to pneumonia are frequently treated with endotracheal intubation or tracheotomy when prolonged mechanical ventilation is needed. Tracheotomy has gained popularity as a method to facilitate weaning from the ventilator by reduction of the pulmonary dead space, which subsequently allows greater access to clear pulmonary secretions and thereby improve patient comfort. Pirat and colleagues have proposed percutaneous dilational tracheotomy as a method of choice for prolonged airway treatment in SOT recipients, with the conditions that patients are properly selected and the procedure is performed by an experienced operator with endoscopic guidance.22 Giannella and colleagues found that 7.4% of patients with pneumonia required mechanical ventilation.6 In our study, 64.7% of patients were intubated and received invasive mechanical ventilation, and 31.4% of patients required tracheotomy. The reason for this difference between the studies is that both outpatients and inpatients were included in the study of Giannella and colleagues.6
In the present study, we detected higher PEEP levels in nonsurvivors than in survivors. Klompas reported that high or nonreduction PEEP levels were associated with ventilator-related events in intubated patients for at least 2 days.23 Similarly, a high PEEP value in our study was associated with mortality and VAP development. We attribute this association to the fact that patients with high PEEP had more severe episodes of acute respiratory distress syndrome. Also, this severity of acute respiratory distress syndrome generally requires longer duration of mechanical ventilation, and so these patients have higher risk for VAP, which may lead to higher mortality.
Several studies have evaluated the utility of procalcitonin measurement for detection and differentiation of bacterial infection versus other infectious SOT complications. Although few studies have evaluated procalcitonin in the setting of SOT recipients with pneumonia, an elevated procalcitonin level may indicate the presence of an infection.7,24 As expected, the mean levels of leukocytes, CRP, and procalcitonin were high in our cohort. Although we could not find a statistically significant relationship with these high levels, we believe these high levels had clinical relevance.
In our study, SOT recipients with VAP had longer hospital and ICU LOS before the diagnosis of pneumonia, and the ICU LOS after diagnosis of pneumonia was longer than for other pneumonia types. Dizdar and colleagues reported that death was a more frequent outcome in patients with nosocomial pneumonia who had longer hospital LOS than in patients with a shorter hospital LOS.25 The cause of this difference in outcomes could be that nosocomial pneumonia such as VAP is seen after hospitalization and the subsequent prolonged hospital LOS may increase the risk of pneumonia. As in our results, VAP is known to prolong LOS in hospital and ICU.8
Many studies have shown that pneumonia in SOT recipients is associated with higher mortality than in other populations.15,26 It has been reported that the mortality associated with nosocomial pneumonia is between 20% and 50%, with a rate of up to 50% among patients who were mechanically ventilated.27 In SOT recipients the mortality rate increased up to the range of 50% to 70%.4,19,28 Our ICU mortality rate of 39.2% confirmed the data from previously published articles. The higher mortality rates among SOT recipients who had pneumonia with hypoxemic respiratory failure may be associated with the higher incidence of nosocomial pneumonia in our study.
Limitations of the present study include the retrospective, single-center design and the limited, small size. Also, protected specimen brush technique (one of the most reliable techniques for assessment of pneumonia in mechanically ventilated patients) and transbronchial biopsy were not performed in our study. Finally, we had no patients who underwent lung, pancreas, and small intestine transplant.
Both diagnosis and treatment of pneumonia in SOT recipients are challenging because of varied clinical presentations, numbers of potential etiologies, and breadth of management options and outcomes. Our cohort of SOT recipients with pneumonia comprised mostly patients with nosocomial pneumonia and who were hospitalized 6 months after transplant. Among those patients with acute respiratory failure who were followed in the ICU, (1) endotracheal intubation and high levels of PEEP were associated with mortality and (2) VAP had poor prognosis.
In conclusion SOT recipients with pneumonia are associated with poor prognosis and, in this high-risk patient group, more careful follow-up, early disclosure of warning signs, and starting the treatment quickly can be effective on the poor outcomes in the ICU.
DOI : 10.6002/ect.2021.0215
From the 1Department of Anesthesiology and Critical Care Unit, and the 2Department of General Surgery, Division of Transplantation, Baskent University Faculty of Medicine, Ankara, Turkey
Acknowledgements: Abstract presented at the American Transplant Congress, May 30 to June 1, 2020 (oral presentation), the 28th International Congress of The Transplantation Society, September 13-16, 2020 (poster presentation), and the International Liver Transplant Society Congress, May 5-8, 2021 (poster presentation). This study was supported by the Baskent University Research Fund. Other than stated, the authors have not received any funding or grants in support of the presented research or for the preparation of this work and have no declarations of potential conflicts of interest.
Corresponding author: Fatma Irem Yesiler, Fevzi Çakmak Road,10th Street, Apt. 45 Bahçelievler, Ankara 06490, Turkey
Phone: +90 312 203 6868 4818
Table 1. Demographic and Clinical Characteristics of the Patients (N = 51)
Table 2. Pneumonia Types and Immunosuppression Regimens
Table 3. Respiratory Management
Table 4. Laboratory Parameters of the Patients at the Time of Diagnosis
Table 5. Association of Duration of Mechanical Ventilation and Length of Stay at Hospital and Intensive Care Unit, Between Pneumonia Type and Transplant Month
Table 6. Logistic Regression Model Results for Mortality
Table 7. Comparisons of Survivors (N = 18) and Nonsurvivors (N = 33)