https://orcid.org/0000-0003-2156-254X
Figures
Abstract
Objective
Febrile neutropenia (FN) is a serious clinical event, associated with significant morbidity and mortality. Imaging has a central role in the identification of the fever cause. The study objectives were to assess abnormalities of potential infectious origin on paranasal sinus, chest and abdomen/pelvis CT scans performed during an episode of FN, in patients with or without specific clinical signs, and to evaluate their impact on the frequency of changes in treatment.
Materials and methods
This retrospective study included 306 patients with FN from a single academic center between October 2018 and December 2019. Clinical and biological data, final clinical diagnosis for the FN episode, and change in treatment following CT scans were collected from medical records. CT images were reviewed for positive findings in each anatomical region.
Results
Overall, 188 patients (61.4%) had at least one positive CT finding on paranasal sinus, chest and/or abdomen/pelvis regions, and 153 patients (50.0%) received a final clinical diagnosis of infection, based on microbiological and/or radiological findings. There were more chest and abdomen/pelvis CT positive findings in patients with specific clinical signs than in patients without (65.7% vs. 25.8%, and 59.0% vs. 22.7% respectively, p <0.001), with a higher frequency of therapeutic modifications in positive CTs. There were significantly more paranasal sinus CT abnormalities in symptomatic patients (33.3% vs. 8.5%, p = 0.03), but without any impact on treatment.
Citation: Tran C, de Kerviler É, Bergeron A, Raffoux E, Xhaard A, de Bazelaire C, et al. (2025) Contribution of paranasal sinus, chest, and abdomen/pelvis computed tomography in patients with febrile neutropenia. PLoS ONE 20(1): e0316459. https://doi.org/10.1371/journal.pone.0316459
Editor: Cho-Hao Howard Lee, National Defense Medical Center, TAIWAN
Received: October 14, 2024; Accepted: December 11, 2024; Published: January 2, 2025
Copyright: © 2025 Tran et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: According to the current French regulation, authorization to access patient data is given individually to the researcher who has made the request, if this request is judged as acceptable, through the IRB. Therefore, this researcher cannot openly share the data, even anonymized. However, access to data is possible for other researchers on request (aphp-accesdonnees@aphp.fr).
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist
Introduction
Febrile neutropenia (FN) is a critical yet common condition affecting onco-hematology patients [1]. It is defined as an oral temperature of >38.3°C or two consecutive readings of >38.0°C for 2 hours associated with an absolute neutrophil count (ANC) below 500/mm3 or expected to fall below this threshold [2]. FN is associated with prolonged hospital stay, delay in the overall chemotherapy treatment plan, and mortality that can reach 40% in high-risk patients [2]. Immediate mortality is secondary to uncontrolled infections of bacterial, fungal or viral origins [3].
Patients with FN usually receive empirical broad-spectrum antibiotic therapy [2]. However, identifying the precise cause of fever is important to select the most adequate treatment, while limiting potential side effects of unnecessary drugs. It is a challenging task because signs and symptoms of inflammation are typically attenuated in the context of neutropenia. Moreover, non-infectious causes of fever such as certain medication use, blood transfusions, radiation, and neoplastic fever are not rare [4]. Infections are clinically documented in only 20%–30% of febrile episodes [5]. Along with extensive microbiological laboratory testing, imaging plays a central role, contributing to identify potential sources of infection, to approach the final appropriate diagnosis, and to guide therapeutic strategies. The contribution of chest computed tomography (CT) has already been investigated in FN for the early detection and management of invasive fungal infections such as aspergillosis [6–8]. However, the role of paranasal sinus and abdomen/pelvis CT, sometimes combined with chest CT in clinical routine to explore FN, has scarcely been studied.
Therefore, the main objective of our study was to identify abnormalities of potential infectious origin detected on paranasal sinus, chest and abdomen/pelvis CT scans performed in the setting of FN episode, in patients with or without specific clinical signs. The secondary objective was to assess the frequency of treatment changes following CT scan report.
Materials and methods
Study protocol and patient selection
This monocentric retrospective study was conducted after approval from the French National Ethics Committee on Medical Imaging Research (CRM-2107-182) that waived the need for informed consent in Saint-Louis Hospital (Assistance Publique-Hôpitaux de Paris), Paris, France. This hospital is a 500-bed academic center largely dedicated to treatment of cancers, including hematological malignancies and solid tumors. The study spanned over a 15-month period from October 1st, 2018 to December 31st, 2019. For the needs of the study, data were accessed from August 1st, 2021 to October 31st 2021. Authors had access to data that could identify patients during the data collection step. Data were then deidentified for the data analysis step.
The patient selection process consisted of reviewing the reports of all CT scans covering paranasal sinus and/or chest and/or abdomen/pelvis regions performed in adult patients in the radiology department during the study period. The keywords "febrile aplasia" and "febrile neutropenia" were searched, as well as mentions of a fever in patients undergoing chemotherapy, immunosuppressive therapies, followed-up for hematological diseases or solid cancer and/or who had undergone stem cell transplantation.
Once the eligible patients were identified by scanning the CT reports, medical files of each patient were studied, and the following inclusion criteria were applied (Fig 1): i) availability of clinical, biological, and imaging data; ii) history of hematological malignancy, solid cancer, immunosuppressive therapies, stem cell transplantation, or other cause of immunosuppression (aplastic anemia, inflammatory disease treatment with immunosuppressive drug, others), iii) neutropenia defined as ANC < 500/mm3, iv) mention of fever (oral temperature of >38.3°C or two consecutive readings of >38.0°C) concomitant to neutropenia in the hospitalization reports. On the opposite, patients with missing clinical, biological and/or imaging data, no history of immunosuppression, and neutropenia without fever or fever without neutropenia were excluded. For patients who had undergone more than one CT within a month during the same episode of FN, only the first CT was considered.
CT: Computed tomography.
Out of 478 eligible patients identified from the CT reports, 172 were excluded based on the above-mentioned criteria (Fig 1), and 306 patients were included.
Data collection
Clinical data.
Sex, age and body-mass index (BMI) were recorded for all patients. The primary conditions for which patients were being treated when FN occurred were obtained from the clinical records and categorized as follows
- Acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), not treated with HSCT nor CAR T-cell therapy
- Hematopoietic stem cell transplant (HSCT) recipients (autologous or allogeneic), whatever the underlying cause of transplantation
- Lymphoma, chronic lymphocytic leukemia (CLL), and multiple myeloma (MM), not treated with HSCT nor CAR T-cell therapy
- Myeloproliferative neoplasm (MPN) and myelodysplastic syndrome (MDS), not treated with HSCT
- Patients undergoing CAR T-cell therapy (for ALL or lymphoma)
- Solid cancer patients
- Others
The duration of neutropenia was recorded in days and categorized as short or prolonged using a 14-day cutoff [9]. The duration of FN was also recorded in days.
The presence or absence of clinical signs, categorized by system (nose/paranasal sinus, respiratory, digestive/urological), was documented from the clinical files and the CT orders. The final clinical diagnosis, as determined by physicians in the hospitalization departments and documented in the hospitalization report, was collected and categorized as infectious or non-infectious. Neutropenic enterocolitis, although being initially an inflammatory disease, was categorized as infectious because of the high risk of associated bacteremia and the need for broad-spectrum antibiotics [10]. Any treatment changes following CT scan results, including alterations in antibiotic, antifungal, antiviral or corticosteroid therapies, were recorded from the clinical files.
Biological data.
White blood cell counts (WBC) and, if available, C-reactive protein (CRP) levels closest to the CT scan date, were recorded. The center where the study was performed does not report neutrophil counts when WBC ≤ 200/mm3. Therefore, ANC were categorized in a binary way based on this threshold.
Imaging data.
CT scan data included the examination date, the scanned anatomical regions scanned (paranasal sinus, chest, and abdomen/pelvis areas) and the use of contrast medium.
When abnormalities compatible with infection were noted in CT scan reports, CT images were reviewed jointly by a 5th year radiology resident and a board-certified radiologist with 8 years of experience, on a PACS workstation and categorized based on their main features. Paranasal sinus CT scans were assessed for sinusitis signs defined as gas-fluid levels, gas bubbles, or soft-tissue opacification of sinus with or without local invasion. Chest abnormalities were assessed for signs of bacterial, viral or fungal pneumonia, including nodules (that could be solid, ground glass, with halo sign, with cavitation), micronodules, parenchymal consolidations (with or without nodules or cavitation), ground glass opacities, and for pleural effusion. Abdomen/pelvis CT scans were examined for digestive wall thickening, hypo enhancements and nodules in solid organs, gallbladder thickening and distension, pyelic wall thickening, peritoneal effusions, and soft tissue collections. In each anatomical region, the above-described findings were defined as CT positive findings.
Statistical analysis
Categorical data were presented as numbers and percentages. Continuous variables, all deviating from a normal distribution (confirmed by Shapiro-Wilk and Kolmogorov-Smirnov tests), were expressed as median with interquartile range (IQR). Chi2 tests, Fisher tests, Mann-Whitney tests, or Kruskal-Wallis tests were used as appropriate to compare both continuous and categorical variables.
All tests were two-tailed, with statistical significance set at p < 0.05. We conducted all analyses using MedCalc software for Windows, version 20.009 (MedCalc Software, Ostend, Belgium).
Results
Study population
Clinical characteristics of patients.
The study population consisted of 306 patients. The most prevalent patient groups were acute leukemia (n = 117, 38.2%), HSCT (n = 71, 23.2%), and lymphoid malignancies (n = 52, 17.0%). Details on the patients’ underlying conditions are available in S1 Table.
Clinical characteristics are displayed in Table 1. They varied between groups. The male/female ratio was 1.17:1, with 53.9% (165) men. However, there was a significantly higher percentage of women (84.6%, p = 0.001–0.04) in the solid cancer group. The median age was 55 years, with significant differences between groups (p<0.001): for instance, median age was 48 years in the HSCT group, and 65 in the MPN/MDS group (p<0.001).
Overall, clinical signs (excluding fever) were reported in 180 (58.8%) patients, including nose/paranasal sinus signs in 9 (2.9%) patients, respiratory signs in 99 (32.4%), and digestive/urological signs in 105 (34.3%). Additionally, 237 (77.5%) patients were explored in at least one anatomical region in which they had no corresponding clinical sign (for instance explored with a chest CT without respiratory signs). One hundred and twenty-six patients (41.2%) had no clinical signs other than fever.
In total, 153 patients out of 306 finally received a clinical diagnosis of infection (50%), supported by radiological and/or microbiological evidence.
Changes in treatment following CT scans were reported in 84 (27.5%) patients.
Biological characteristics of patients.
Biological characteristics of patients are detailed in Table 1. Data on neutropenia duration categorized as short or prolonged was available for 274 patients (89.5% of the total cohort), and 19.3% of patients (53/274) experienced prolonged FN (>14 days). The prevalence of prolonged FN was significantly higher in the MPN/MDS group (52.9% (9/17, p<0.001–0.02).
The median WBC count at the time of CT was 340/mm3 (IQR 200–960). More than half of the patients (162/306, 52.9%) had an ANC ≤ 200/mm3, with a notable overrepresentation in the HSCT group (50/71, 70.4%, p = 0.01–0.03).
CT findings and relations to clinical signs, changes in treatment and final diagnoses
CT characteristics.
Areas explored by CT included the chest region in 93.1% (285/306) of patients, the abdomen/pelvis region in 80.4% (246/306), and the paranasal sinus region in 22.2% (68/306). For 17.7% of patients (54/306), CT included the 3 regions. Intravenous contrast medium was used in 76.1% (233/306) patients, predominantly during examinations that included the abdominopelvic region (96.6%, 225/233).
CT findings and clinical signs.
Overall, 188 patients (61.4%) had at least one positive CT finding, compatible with infection, on nose/paranasal sinus, chest and/or abdomen/pelvis regions. In patients with at least one clinical sign (in addition to fever), 126 out of 180 (66.7%) had at least one positive CT finding in the corresponding anatomical region (paranasal sinus CT finding for nose/paranasal sinus signs, chest CT finding for respiratory signs and abdomen/pelvis CT finding for digestive/urological signs). In patients without clinical signs in an explored anatomical region, 84 out of 237 (35.4%) had at least one positive CT finding in those regions. Finally, among the 126 patients without any clinical sign apart from fever, 51 (40.5%) had at least one positive CT finding.
The frequencies of CT positive findings for each anatomical region and the presence or absence of clinical signs in the corresponding system are detailed in Table 2. For the paranasal sinus region, 33.3% (3/9) of scans were positive in symptomatic patients versus 8.5% (5/59) in asymptomatic patients (p = 0.03). Chest and abdomen/pelvis CTs with positive findings were significantly more prevalent in patients with clinical signs compared with those without (65.7% vs. 25.8%, and 59.0% vs. 22.7% respectively, p<0.001). However, a notable percentage of positive CTs occurred in asymptomatic patients: 25.8% (48/186) for chest CTs and 22.7% (32/141) for abdomen/pelvis CTs. There was no significant difference in the distribution of positive CT findings among symptomatic and asymptomatic patients across the 7 patient groups.
CT findings and final clinical diagnosis.
Among the 8 sinus CTs with positive findings, 5 resulted in a clinical diagnosis of acute sinusitis (3/5 in asymptomatic patients, 2/3 in symptomatic patients), without any proven case of invasive fungal infection. Illustrative examples are presented in Fig 2.
Axial sections of paranasal sinus CT in 2 different patients (A, B). A: Partial filling of the right maxillary sinus with gas-fluid level and tiny bubbles on the surface of the fluid (arrow); final diagnosis of acute right maxillary sinusitis without microbiological evidence. B: Partial filling of the left portion of the sphenoidal sinus, with peripheral mucosal thickening, gas-fluid level, and some bubbles on the surface of the fluid; punctate hyper attenuations within the filling (arrow); final diagnosis of acute left sphenoidal sinusitis without microbiological evidence.
As shown in Table 3, 38 CT scans in the chest region (33.6%) out of 113 with positive findings did not lead to a clinical diagnosis of pneumonia. Among these, 25 (65.8%) were in patients without respiratory signs. Chest CT scans with positive findings in patients without respiratory symptoms (48/186, 25.8%) resulted in various diagnoses, including 4 cases (8.3%) of aspergillosis and 2 cases (4.2%) of mucormycosis.
Table 4 summarizes the causative agents found for the 75 pneumonias diagnosed in the cohort and Fig 3 illustrates the CT appearance of some of these infections.
Axial chest CT sections in 6 different patients (A-F). A: Parenchymal condensation of the right upper lobe, with irregular contours and aerated bronchogram, associated with bilateral pleural effusion. Microbiological diagnosis of Pseudomonas aeruginosa pneumonia. B: Parenchymal condensation of the right lower lobe with peripheral ground glass and intralobular reticulations (crazy paving). Microbiological diagnosis of Staphylococcus warneri pneumonia. C: Pseudonodular condensation area of the culmen, with peripheral ground glass halo. Microbiological diagnosis of Aspergillus fumigatus pneumonia. D: Rounded ground-glass area of the apicodorsal segment of the culmen, with irregular contours, associated with pleural effusion. Microbiological diagnosis of Rhizomucor and Rhizopus pneumonia. E: Bilateral nodules with blurred contours (arrows). Microbiological diagnosis of septic embolisms due to Stenotrophomonas maltophilia. F: Bilateral micronodules and ground glass nodules with blurred contours (arrows). Microbiological diagnosis of Parainfluenza virus type 3 pneumonia.
Table 5 shows the semiology of CT abnormalities and corresponding final diagnoses in the abdomen/pelvis region. Twenty-eight CT scans (29.8%) out of 94 with positive CT findings did not lead to a diagnosis of infection in this region, mostly in patients with no clinical signs for the abdomen/pelvis region (20/28, 71.4%). Among patients with abdomen/pelvis positive CT findings, those related to a clinical diagnosis of neutropenic enterocolitis predominated, constituting 54.3% (51/94) of positive CT findings. There was one case each of hepatosplenic candidiasis and abdominal invasive mucormycosis (1.1% each). Fig 4 illustrates the CT appearance of some of these infections.
Abdomen/pelvis CTs in coronal (A) and axial (B-F) sections in 6 different patients. Abnormalities indicated by arrows. A: Wall thickening of the cecum: Diagnosis of neutropenic colitis. B: Newly appeared hypoenhancing hepatic and splenic micronodules: Diagnosis of hepatosplenic invasive candidiasis. C: Appendiceal dilation with wall thickening, mucosal hyperenhancement and submucosal edema: Diagnosis of acute appendicitis. D: Segmental wall thickening of the sigmoid colon with edematous thickening of a dilated diverticulum, and pericolic fat stranding: Diagnosis of uncomplicated acute diverticulitis. E: Gallbladder wall thickening, with mucosal hyperenhancement and perivesicular fat stranding: Diagnosis of acute cholecystitis. F: Hypo enhancing renal collection: Diagnosis of renal abscess.
Finally, 39 patients out 118 (33.1%) were diagnosed with infection despite normal CTs, including bacteriemia or fungemia (catheter-related or not), musculo-skeletal, urinary and cutaneous infections. A summary of clinical signs, CT findings and final diagnoses is provided in S1 and S2 Figs.
CT findings and treatment changes.
Table 6 presents the incidence of treatment changes following CT findings. Overall, treatment changes occurred in 84 instances of positive CT scan regions out of 215 (39.1%). A statistically significant increase in changes was observed after positive CT findings in chest (41/113 or 36.3% versus 0/172 or 0.0%, p<0.001) and abdomen/pelvis (42/94 or 44.7% versus 0/152 or 0.0%, p<0.001) CTs. However, there was no significant difference when a paranasal sinus positive finding was found (1 positive CT finding out of 7, p = 0.12).
Discussion
Our single-center retrospective study, including 306 patients undergoing CT scans during a FN episode, offers an overview of imaging findings and associated clinical diagnoses in an immunocompromised population. Overall, 188 patients (61.4%) had at least one positive CT finding. There were significantly more chest and abdomen/pelvis CT positive findings in symptomatic patients (65.7% vs. 25.8%, and 59.0% vs. 22.7% respectively, p <0.001), but also a significant number of positive CTs in asymptomatic patients (25.8% and 22.7% for chest and abdomen/pelvis, respectively). Treatment changes following chest and/or abdomen/pelvis CT were only noted in cases of positive CT findings, reinforcing the crucial role of imaging in the management of these patients.
Most patients in our study had hematological conditions, and solid cancer cases constituted only 4.2% of the cohort. The low representation of solid tumors is first due to an institutional factor, the center where the study was performed being an academic center with 23% of the 500 inpatient beds dedicated to hematology departments. Secondly, neutropenia in patients with solid tumors, usually attributable to chemotherapy, is less severe and less prolonged than in patients with hematological conditions receiving myeloablative treatments, with likely lower frequencies of CT use in their management [11].
Paranasal sinus CT scans (68, 22.2% of patients) were mostly performed in asymptomatic patients (54, 79.4%) and only 5 diagnoses of non-fungal acute sinusitis were made (3 in asymptomatic patients). Paranasal sinus involvement is not uncommon in invasive fungal infections [12] and mortality is high [13]. Paranasal sinus CT is therefore recommended by some guidelines in case of neutropenia and evocative clinical signs. However, there is no guidance on the use of imaging in patients without localizing symptoms [14,15]. In this subgroup, our results, which underline the low diagnostic yield of paranasal sinus CT in asymptomatic patients, are in line with previously published data. In a study that included 56 paranasal sinus CT, Gurram et al found that the source of febrile neutropenia was attributed to the CT sinus findings in 9 cases, all of whom occurring in symptomatic patients [16]. Another work by Chang et al included 142 non targeted paranasal sinus CT in patients with febrile neutropenia and identified positive examinations in only 13 cases (9.1%) [17]. Moreover, it has been shown that sinusitis if much less likely to be a source of infection, compared to pneumonia [17,18]. Of note, in our study, none of the patients diagnosed with pulmonary or abdominal invasive fungal infection had concomitant positive paranasal sinus CT findings. To our knowledge, there is no relevant study evaluating the impact of paranasal sinus CT on morbidity and mortality in patients with FN, therefore its role in asymptomatic patients and in the absence of a prior diagnosis or suspicion of invasive fungal infection remains to be proven. Even in this case, it must be reminded that paranasal sinus CT has a suboptimal sensitivity and specificity for fungal infection at the early stage [19].
Lung is known as the first site of infection in immunocompromised patients [20] and chest was the most explored area (93.1%). Positive findings were found in 113 (39.6%) but approximately one-third of these abnormalities (38/113, 33.6%) did not translate into an infectious diagnosis, mostly in asymptomatic patients (25/38, 65.8% of this subgroup). Some chest abnormalities were not specific and could notably indicate features of disease specific lesions or of pulmonary edema, a common finding in patients with hematological malignancies [21]. It is also possible that microbiological sampling limitations did not allow to perform a diagnosis, particularly in case of minor abnormalities, and pneumonia diagnosis was therefore not retained [22]. Interestingly, 25.8% (48/186) of patients without respiratory signs had positive chest CTs, contributing to a clinical diagnosis of infection in nearly half of them (23/48, 45.8%), including 6 cases of fungal infections. In a recent article, Chang et al found an even higher 33.3% (46/138) rate of positive chest CT scans in patients with febrile neutropenia without localizing signs. The authors did not provide the number of fungal infections among this population [17]. However, in a cohort of patients recently diagnosed with acute myeloid leukemia, Bitterman et al. showed that the diagnosis of invasive pulmonary aspergillosis was not predicted by the presence or absence of respiratory symptoms. This underlines the poor predictive performances of clinical signs in this population. On the other hand, the role of early chest CT for the detection of pulmonary fungal infection is well established as a prompt diagnosis is critical to improve the disease outcome through the initiation of antifungal therapy [23,24]. Of note, it is now clear that chest X-ray has suboptimal performances for infection detection in this specific context [6,25,26]. Overall, current national and international guidelines recommend the use of chest CT in FN, with some variations: in cases of respiratory signs, or after 72h of appropriate antibiotics, or in patients at risk for invasive aspergillosis even without respiratory symptoms [2,14,15]. Our results certainly support the use of routine chest CT in the context of FN, including in asymptomatic patients. Chest CT is also valuable because of its excellent negative predictive value, allowing to rule out a lung infection as the cause of the fever. Indeed, in our cohort, we did not observe any change in the anti-infectious treatment when the CT showed no abnormality. Still, it should be repeated if this fever persists, as abnormalities related to pneumonia are likely to appear rapidly [27]. Low-dose protocols should be used to limit radiation exposure [28].
Finally, 246 abdomen/pelvis CTs were performed, i.e., in 80.4% of included patients, with 94 cases of positive findings (38.2%). As in chest CTs, there was a significantly higher frequency of positive findings in symptomatic cases compared with asymptomatic ones (59.0% vs. 22.7%, p<0.001) and a substantial proportion of positive CT scans (28/94, 29.8%) that did not lead to an infectious diagnosis. It is essential to emphasize the nonspecific nature of CT abnormalities in leading to these diagnoses, underscoring the pivotal role of clinical signs and microbiological sampling. Interestingly, 22.7% (32/141) of abdomen/pelvis CT scans in asymptomatic patients were positive. The most common CT findings were intestinal abnormalities, as identified by previous studies that included patients with febrile neutropenia and CT of the abdominal region, with or without localizing clinical signs [17,29]. The most common clinical diagnosis associated with digestive wall thickening was neutropenic enterocolitis, identified in 54.3% (51/94) of abdomen/pelvis CT with positive findings. Neutropenic enterocolitis has previously been identified as the first cause of radiologic bowel abnormality in a cohort of neutropenic patients [30]. It is a severe complication of neutropenia, whose pathophysiology is still not fully understood, that favors infection; for this reason, broad-spectrum antibiotics are usually required, and antifungal therapy can be discussed in severe cases [10,31,32]. The diagnosis of neutropenic enterocolitis generally involves association of fever, abdominal pain, neutropenia, and thickening of the abdominal wall on abdominal studies (ultrasound and CT) [33–35]. Nevertheless, clinical signs may be absent, and imaging therefore plays a central role [36]. Abdominal ultrasound is a radiation-free, contrast-free method; however, it may be less performant than CT for the identification of complications that may require surgical intervention or for alternative diagnoses [10]. Aside from neutropenic enterocolitis, several cases of common acute abdominal infections were identified, including diverticulitis, appendicitis and cholecystitis. Such diagnoses may require specific interventions, beyond antimicrobial treatment. However, neutropenic patients may have unusual presentations, with attenuated clinical signs [37]. Finally, invasive fungal infections in the abdominopelvic region were uncommon in our cohort (2 cases). However, given their severity, early detection is important, and CT may have a role to play in it, in association with clinical and biological data, as well as in monitoring the response to treatment. Overall, and despite the lack of current recommendations on this topic, our study suggests that abdomen/pelvis CT may be useful in asymptomatic patients with febrile neutropenia.
The analysis of therapeutic modifications following CT scans revealed a significant association between the presence of abnormalities on chest and abdomen/pelvis CTs and a higher frequency of treatment changes in affected patients compared with those without abnormalities. We did not find any significant difference regarding paranasal sinus CT scans, possibly due to the smaller number of patients in that group. However, it is important to acknowledge that the absence of therapeutic modification does not necessarily indicate a lack of impact of imaging in patient management. In cases where radiological abnormalities with infectious features are identified, maintaining an appropriate antibiotic therapy may be prioritized over de-escalation. This also underscores the multifaceted role of CT findings in patient care. The radiological insights from CT scans can contribute to the prognostic evaluation of patients who already have a clinical diagnosis. Additionally, early detection of disease progression despite chemotherapy may prompt consideration of alternative therapies before scheduled oncologic reassessment. Still, caution is advised in interpreting the results related to post-CT therapeutic changes. The retrospective nature of clinical data, sourced solely from hospitalization reports, may not capture the nuances of complex therapeutic discussions, potentially oversimplifying the decision-making process and the impact of CT.
This study has several limitations, primarily due to its retrospective nature. Patient selection relied on indications mentioned in CT scan reports, and collection of clinical data, as stated above, may have been incomplete. Consequently, clinical signs that were present but not mentioned in clinical files can have been missed. Additionally, we could not control with precision the criteria on which clinical diagnoses of infection were made and had to rely on the conclusion of the hospitalization reports. For this reason, it is also challenging to understand why some positive CT findings did not translate into an infection diagnosis. A second limitation pertains to the cohort study design, excluding FN patients who did not undergo CT scans. A control population would allow comparison of prognoses between those who underwent CT scans and those who did not. However, precise matching of all clinical and biological characteristics poses challenges, and ethical considerations also preclude such a study due to existing literature supporting the value of chest CT in this patient group. Finally, the single-center nature of the study further limits generalizability, with most patients having hematological malignancies. Applicability to patients with solid neoplasia remains uncertain, given the specificities of our hospital’s practices and the lack of clear guidelines for non-chest CT scans in FN. Nonetheless, this study provides insights into the use of paranasal sinus and abdomen/pelvis CT scans in FN cases.
To conclude, our retrospective study showed that while chest and abdomen/pelvis CT abnormalities in patients with FN were more common in patients with clinical symptoms, a significant number of asymptomatic patients had positive findings in these regions that could lead to a clinical diagnosis of infection. These results support the existing recommendations for the large use of chest CT in patients with FN and may encourage the use of contrast-enhanced abdomen/pelvis CT in this indication, even in the absence of symptoms. Data regarding paranasal sinus CT, although more limited, do not provide a strong argument in favor of its systematic recommendation in patients without clinical signs. Because of the lack of specificity of some CT findings, communication between radiologists and referring physicians is key to interpret imaging data in challenging cases, to select the optimal management of these findings. However, questions remain unanswered. These questions include the actual impact of the imaging detection of abnormalities—particularly on abdomen/pelvis—on the prognosis of patients, as well as the cost-effectiveness of performing targeted versus non-targeted CT scans.
Supporting information
S1 Fig. Summary of clinical signs, CT findings and final diagnoses, per region.
https://doi.org/10.1371/journal.pone.0316459.s002
(TIF)
S2 Fig. Summary of clinical signs, CT findings and final diagnoses, per patient.
https://doi.org/10.1371/journal.pone.0316459.s003
(TIF)
References
- 1. Klastersky J. Management of fever in neutropenic patients with different risks of complications. Clin Infect Dis. 2004;39 Suppl 1:S32–7. pmid:15250018
- 2. Klastersky J, Naurois Jd, Rolston K, Rapoport B, Maschmeyer G, Aapro M, et al. Management of febrile neutropaenia: ESMO Clinical Practice Guidelines †. Annals of Oncology. 2016;27:v111–v8.
- 3. Viscoli C, Varnier O, Machetti M. Infections in patients with febrile neutropenia: epidemiology, microbiology, and risk stratification. Clin Infect Dis. 2005;40 Suppl 4:S240–5.
- 4. Pasikhova Y, Ludlow S, Baluch A. Fever in Patients With Cancer. Cancer Control. 2017;24(2):193–7. pmid:28441374
- 5. Freifeld AG, Bow EJ, Sepkowitz KA, Boeckh MJ, Ito JI, Mullen CA, et al. Clinical practice guideline for the use of antimicrobial agents in neutropenic patients with cancer: 2010 update by the infectious diseases society of america. Clin Infect Dis. 2011;52(4):e56–93. pmid:21258094
- 6. Heussel CP, Kauczor H-U, Heussel GE, Fischer B, Begrich M, Mildenberger P, et al. Pneumonia in Febrile Neutropenic Patients and in Bone Marrow and Blood Stem-Cell Transplant Recipients: Use of High-Resolution Computed Tomography. Journal of Clinical Oncology. 1999;17(3):796–. pmid:10071269
- 7. Ho DY, Lin M, Schaenman J, Rosso F, Leung ANC, Coutre SE, et al. Yield of diagnostic procedures for invasive fungal infections in neutropenic febrile patients with chest computed tomography abnormalities. Mycoses. 2011;54(1):59–70. pmid:19793207
- 8. Jin J, Wu DP, Liu Y, Pan SS, Yan JL, Aram JA, et al. Utility of CT assessment in hematology patients with invasive aspergillosis: a post-hoc analysis of phase 3 data. Bmc Infectious Diseases. 2019;19:9.
- 9. L. F. Strategies for prevention of infection in short-duration neutropenia. Infect Control Hosp Epidemiol. 1998;18(8):590–2. pmid:9758061
- 10. Nesher L, Rolston KV. Neutropenic enterocolitis, a growing concern in the era of widespread use of aggressive chemotherapy. Clin Infect Dis. 2013;56(5):711–7. pmid:23196957
- 11. Klastersky J, Paesmans M. The Multinational Association for Supportive Care in Cancer (MASCC) risk index score: 10 years of use for identifying low-risk febrile neutropenic cancer patients. Support Care Cancer. 2013;21(5):1487–95. pmid:23443617
- 12. Klimko N, Khostelidi S, Shadrivova O, Volkova A, Popova M, Uspenskaya O, et al. Contrasts between mucormycosis and aspergillosis in oncohematological patients. Med Mycol. 2019;57(Supplement_2):S138–S44. pmid:30816977
- 13. Craig JR. Updates in management of acute invasive fungal rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg. 2019;27(1):29–36. pmid:30585877
- 14. Heinz WJ, Buchheidt D, Christopeit M, von Lilienfeld-Toal M, Cornely OA, Einsele H, et al. Diagnosis and empirical treatment of fever of unknown origin (FUO) in adult neutropenic patients: guidelines of the Infectious Diseases Working Party (AGIHO) of the German Society of Hematology and Medical Oncology (DGHO). Ann Hematol. 2017;96(11):1775–92. pmid:28856437
- 15. Ullmann AJ, Aguado JM, Arikan-Akdagli S, Denning DW, Groll AH, Lagrou K, et al. Diagnosis and management of Aspergillus diseases: executive summary of the 2017 ESCMID-ECMM-ERS guideline. Clin Microbiol Infect. 2018;24 Suppl 1:e1–e38. pmid:29544767
- 16. Gurram P, Almeida N, Vijayvargiya P, Grimont C, Garrigos Z, Khalil S, et al. Utility of Sinus CT in the Evaluation of Patients with Febrile Neutropenia. 2020 Dec 31;7(Suppl 1):S61–2. Open Forum Infect Dis. 2020;7:S61–2.
- 17. Chang VKO, Liang ES, Schmidt P. The diagnostic utility of computed tomography scans performed for febrile neutropenia in a single centre. Curr Probl Diagn Radiol. 2024;53(3):341–5. pmid:38309990
- 18. Choi WK, D’Sa A, Holman JN, Javed S, Thompson S, Nadgir RN. Chest CT Has Higher Yield for Infection than CT Sinus in Febrile Neutropenic Patients. Curr Probl Diagn Radiol. 2022;51(3):340–3.
- 19. Sanguinetti M, Posteraro B, Beigelman-Aubry C, Lamoth F, Dunet V, Slavin M, et al. Diagnosis and treatment of invasive fungal infections: looking ahead. The Journal of antimicrobial chemotherapy. 2019;74(Suppl 2):ii27–ii37. pmid:31222314
- 20. Cheng GS, Crothers K, Aliberti S, Bergeron A, Boeckh M, Chien JW, et al. Immunocompromised Host Pneumonia: Definitions and Diagnostic Criteria: An Official American Thoracic Society Workshop Report. Ann Am Thorac Soc. 2023;20(3):341–53. pmid:36856712
- 21. Cornetto MA, Chevret S, Abbes S, de Margerie-Mellon C, Hussenet C, de Fontbrune FS, et al. Early Lung Computed Tomography Scan after Allogeneic Hematopoietic Stem Cell Transplantation. Biol Blood Marrow Transplant. 2016;22(8):1511–6. pmid:27189110
- 22. Lewis RE, Cahyame-Zuniga L, Leventakos K, Chamilos G, Ben-Ami R, Tamboli P, et al. Epidemiology and sites of involvement of invasive fungal infections in patients with haematological malignancies: a 20-year autopsy study. Mycoses. 2013;56(6):638–45. pmid:23551865
- 23. Heussel CP, Kauczor HU, Heussel GE, Fischer B, Begrich M, Mildenberger P, et al. Pneumonia in febrile neutropenic patients and in bone marrow and blood stem-cell transplant recipients: Use of high-resolution computed tomography. Journal of Clinical Oncology. 1999;17(3):796–805. pmid:10071269
- 24. Caillot D, Couaillier JF, Bernard A, Casasnovas O, Denning DW, Mannone L, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. Journal of Clinical Oncology. 2001;19(1):253–9. pmid:11134220
- 25. El Majzoub I, El Zakhem A, Cheaito R, Cheaito MA, Kaddoura R, Alkozah M, et al. The utility of chest X-ray vs. computed tomography in febrile neutropenia patients presenting to the emergency department. J Infect Dev Ctries. 2020;14(10):1178–84. pmid:33175714
- 26. Gerritsen MG, Willemink MJ, Pompe E, Bruggen Tvd, Rhenen Av, Lammers JWJ, et al. Improving early diagnosis of pulmonary infections in patients with febrile neutropenia using low-dose chest computed tomography. PloS one. 2017;12(2):e0172256. pmid:28235014
- 27. Heussel CP, Kauczor H-U, Ullmann AJ. Pneumonia in neutropenic patients. European Radiology. 2004;14:256–71. pmid:12898175
- 28. de Margerie-Mellon C, de Bazelaire C, Montlahuc C, Lambert J, Martineau A, Coulon P, et al. Reducing Radiation Dose at Chest CT: Comparison Among Model-based Type Iterative Reconstruction, Hybrid Iterative Reconstruction, and Filtered Back Projection. Academic Radiology. 2016;23:1246–54. pmid:27346234
- 29. Al-Ward R, Rizvi S, Hamdi A, Farid S, Sohail M. Utility of CT Abdomen in Evaluation of Neutropenic Fever in Patients with Hematological Malignancies. Open Forum Infect Dis. 2018;26(5):S476.
- 30. Kirkpatrick ID, Greenberg HM. Gastrointestinal complications in the neutropenic patient: characterization and differentiation with abdominal CT. Radiology. 2003;226(3):668–74. pmid:12601214
- 31. Xia R, Zhang X. Neutropenic enterocolitis: A clinico-pathological review. World J Gastrointest Pathophysiol. 2019;10(3):36–41. pmid:31692935
- 32. Duceau B, Picard M, Pirracchio R, Wanquet A, Pene F, Merceron S, et al. Neutropenic Enterocolitis in Critically Ill Patients: Spectrum of the Disease and Risk of Invasive Fungal Disease. Crit Care Med. 2019;47(5):668–76. pmid:30741755
- 33. Rodrigues FG, Dasilva G, Wexner SD. Neutropenic enterocolitis. World J Gastroenterol. 2017;23(1):42–7. pmid:28104979
- 34. McCarville MB, Adelman CS, Li C, Xiong X, Furman WL, Razzouk BI, et al. Typhlitis in childhood cancer. Cancer. 2005;104(2):380–7. pmid:15952190
- 35. Mullassery D, Bader A, Battersby AJ, Mohammad Z, Jones EL, Parmar C, et al. Diagnosis, incidence, and outcomes of suspected typhlitis in oncology patients—experience in a tertiary pediatric surgical center in the United Kingdom. J Pediatr Surg. 2009;44(2):381–5. pmid:19231539
- 36. Itani M, Menias CO, Mellnick VM, El Zakhem A, Elsayes K, Katabathina V, et al. Imaging of abdominal and pelvic infections in the cancer patient. Abdominal radiology. 2021;46(6):2920–41. pmid:33386914
- 37. White MG, Morgan RB, Drazer MW, Eng OS. Gastrointestinal Surgical Emergencies in the Neutropenic Immunocompromised Patient. J Gastrointest Surg. 2021;25(12):3258–64. pmid:34506017