Acute cholecystitis in neutropenic patients.

BACKGROUNDS/AIMS: The frequency of acute cholecystitis reported in neutropenic patients is between 0.4-1.65%. Clinical manifestations differ from general population as well as clinical approach, diagnosis and treatment. The aim of this work is to describe clinical characteristics, diagnostic approach, and outcomes of patients with hematological diseases that presented with neutropenia and fever associated with acute cholecystitis in a tertiary referral hospital.
METHODS: We performed a retrospective analysis of patients with diagnosis of neutropenia and fever associated with acute cholecystitis in the period between January 2000 and January 2017. Quantitative variables were analyzed with mean and standard deviation, and qualitative variables with frequency and percentage.
RESULTS: During the study period, 2007 patients presented with neutropenia and fever. Twelve of them (0.59%) had associated acute cholecystitis. The most common hematologic disease among these patients was lymphoblastic leukemia. Acute acalculous cholecystitis was diagnosed in 6 cases (50%). Eleven patients (91.6%) had a severe presentation and cholecystostomy was performed in 9 (75%) cases. The main cause of mortality was septic shock (33.3%).
CONCLUSIONS: Treatment of acute cholecystitis in patients with neutropenia must be individualized. Cholecystostomy should be considered as a bridge therapy for an interval cholecystectomy.

INTRODUCTION

Abdominal infections in patients with neutropenia are usually severe and life-threatening.1 Neutropenia may be due to a primary hematologic disorder or secondary to chemotherapy. The frequency of acute cholecystitis in this population is between 0.4–1.65% with a 30-day mortality of 26–48%,2 being acalculous cholecystitis the most common presentation.34

Neutropenic enterocolitis is the most common intra abdominal infections associated with fever in neutropenic patients.56 However, acute cholecystitis may have an atypical presentation in this patients and should always be considered as a differential diagnosis.78

Unfortunately there is scarce information on this topic on the literature and as a result there are no specific recommendations to guide diagnosis and treatment of acute cholecystitis in hematologic diseases. The aim of this study was to describe the clinical characteristics, diagnostic approach, treatment, and outcomes of patients with hematological disorders, specifically neutropenia and fever associated with acute cholecystitis in a tertiary referral hospital.

PATIENTS AND METHODS

A search was performed at our institution's electronic system to identify patients admitted to the hospital with the diagnosis of neutropenia and fever between January 2000 and January 2017. We included patients with diagnosis of acute cholecystitis. Neutropenia was defined according to the American Society of Hematology as an absolute neutrophil count below 1,500 cells/µl.9 Diagnosis and severity of acute cholecystitis were established based on the Tokyo guidelines.1011 Quantitative variables were described with mean and standard deviation and qualitative variables with absolute frequencies and percentages.

RESULTS

General characteristics

Out of 2,007 patients hospitalized for neutropenia and fever 12 cases (0.59%) were associated with acute cholecystitis (Table 1). The median age was 35 years (21–70 years old), 5 patients were women (41.6%).

Table 1

Clinical characteristics in patients with hematologic malignancies and acute cholecystitis

MDS, myelodysplastic syndrome; ALL, acute lymphoblastic leukemia; NHL, Non-Hodgkin lymphoma; AML, myeloid acute leukemia; US, Ultrasound scan; CT, Computed tomography; Hyper-CVAD, hyperfractionated therapy of cyclophosphamide, vincristine sulfate, doxorubicin hydrochloride (Adriamycin), and dexamethasone; SMILE, regimen of steroid (dexamethasone), methotrexate, ifosfamide, L-asparaginase, and etoposide

Acute cholecystitis diagnosis and severity

All patients, except one who had no abdominal pain, met diagnostic criteria for acute cholecystitis according to Tokyo 2013 criteria. Acute acalculous cholecystitis was diagnosed in 6 individuals (50%). The most common local sign of inflammation was Murphy's sign in 7 patients (58.3%). Regarding systemic inflammatory response, fever and tachycardia were present in 9 patients (75%), and 5 patients (41.6%) had elevated C reactive protein. Abdominal ultrasound was performed in 10 patients (91.6%), and computed tomography in 2 (16.2%); findings were compatible with acute cholecystitis. Eleven patients (91.6%) presented with severe acute cholecystitis. Organ failures included hematological dysfunction (platelets <100,000/mm3) in 11 patients (91.6%), cardiovascular dysfunction (hypotension requiring vasopressors) in 5 patients (41%), renal dysfunction (creatinine >2.0 mg/dl) in 4 patients (33.3%), and respiratory dysfunction (PaO2/FiO2 ratio <300) in 2 patients (16.6%) Table 2.

Table 2

Acute cholecystitis diagnosis and grading

US, Ultrasound; CT, Computed tomography; CRP, C reactive protein

Hematologic characteristics

Concerning the hematologic diagnosis, seven patients (58.3%) had lymphoblastic leukemia, and acute myeloid leukemia, non-Hodgkin lymphoma, NK lymphoma, myelodysplastic syndrome (dendritic cell neoplasia), and aplastic anemia were present 1 patient (8.3%) each. Seven patients (58%) were under chemotherapy with Hyper-CVAD (hyperfractionated cyclophosphamide, vincristine sulfate, doxorubicin hydrochloride and dexamethasone), and the others with regimens that included cytarabine/rituximab, cyclosporine/bisulfan, or the SMILE scheme (dexamethasone, methotrexate, ifosfamide, L-asparaginase and etoposide). Regarding chemotherapy adverse effects, grade 1–2 neurotoxicity was observed in two patients (16.6%) and grade 1–4 hematotoxicity in 11 (91.6%). Among those patients who received Hyper-CVAD therapy, grade 3–4 neutropenia and thrombocytopenia developed in all patients, and grade 1–2 anemia in five of them (71%).

The median time that elapsed between the last chemotherapy administration and the event of acute cholecystitis was 36.58 days (range 6–150). In preparation for cholecystostomy, 8 (88.8%), 1 (11.1%), and 1 (11.1%) patients were transfused with platelets, fresh frozen plasma, and red blood cells, respectively.

Microbiology characteristics

Escherichia coli with extended-spectrum beta-lactamase (ESBL) was isolated from blood and bile cultures in 3 (25%), and 2 patients (16.6%), respectively. Other species isolated in bile included Aeromonas hydrophila, Klebsiella pneumonie and Staphylococcus aureus.

Treatment

Initially all patients received broad spectrum antibiotics, including carbapenems and vancomycin in most cases, as shown in Table 1. Cholecystostomy was performed in 9 patients (75%): the approach was percutaneous in 8 cases and open in one. Furthermore, 2 patients (16.6%) underwent laparoscopic cholecystectomy, and one (8.3%) received only medical treatment. It is worth to mention that 4 of the 5 patients with acalculous cholecystitis were treated with cholecystostomy tube and one of them was treated only with antibioitcs. After four weeks three patients that were initially treated with cholecystostomy underwent interval laparoscopic cholecystectomy, and another patient had the cholecystostomy tube removed, with no complications. One patient still has a cholecystostomy in place.

Outcomes

Four patients died (33.3%), three of them had undergone cholecystostomy and the patient that received only medical treatment died too. The cause of death was sepsis with multiorgan failure, one patient had concomitant neutropenic enterocolitis. Three out of 6 patients (50%) with acalculous cholecystitis died.

DISCUSSION

Abdominal pain in patients with neutropenia might require an exhaustive approach due to its broad differential diagnosis and atypical presentation among these patients,12 who may not present an inflammatory response. Despite neutropenic enterocolitis being the most common etiology in this scenario,13 acute cholecystitis should be considered as a differential diagnosis.14 It is thought that the smaller mucosal surface of the gallbladder compared with the colon might explain the low frequency of acute cholecystitis as etiology of abdominal pain in neutropenic patients with fever.1 Our study showed a frequency of 0.59% without male or female predominance, similar to the reported in other series.15 Acalculous cholecystitis was documented in 50% of the cases, compared to 55–65% in other series.16 In non-neutropenic patients acalculous cholecystitis is much less frequent, comprising roughly 5% of the cases of acute cholecystitis. This may be because patients with neutropenia have several risk factors for this presentation such as bile stasis, prolonged fasting, and increased intraluminal pressure.28

Treatment decision between surgical cholecystectomy or percutaneous cholecystostomy in these patients is troublesome since most of them fulfill criteria for severe acute cholecystitis and might require the former according to Tokyo recommendations,11 however, there is not enough evidence to recommend one procedure over the other in this specific population. All our patients met Tokyo criteria for acute cholecystitis, and 11 patients (91%) were classified as severe based on hematological failure resulting either from their underlying disease or from chemotherapy induced myelosuppression, but probably not because of severe sepsis, which questions the validity of these criteria in this specific population. As a matter of fact, hematologic failure was the only criteria of severity in 5 of 11 patients. Hematologic disease is an important factor to consider when planning treatment because most patients will need transfusion of blood products prior to any invasive procedure. In our study ten patients (83.3%) required blood products including platelets transfusion in 8 patients (88.8%), fresh frozen plasma in 1 patient (11.1%) and red blood cells transfusion in 1 patients (11.1%).

Concerning treatment (Table 3), cholecystostomy was performed in 9 patients (75%), laparoscopic cholecystectomy in 2 (16.6%), and one patient received only medical treatment. Mortality in our series was 33.3%, which is in line with the 26–48% reported in previous studies.17 Septic shock was the main cause of death. There were no deaths in the laparoscopic cholecystectomy group, while three patients in the cholecystostomy group (37%), and one in the antibiotic treatment group died. According to these results, surgical cholecystectomy or percutaneous cholecystostomy may be more beneficial than antibiotic treatment alone. Mortality was higher in acalculous presentation when compared to calculous cholecystitis (50% versus 16.6%, respectively) probably explained by the fact that the former represents a group of patients with a more severe disease or with a delayed diagnosis.

Table 3

Treatment, complications and transfusion requirements

In conclusion, acute cholecystitis in neutropenic patients is a rare diagnosis, accounting for less than 1% of the neutropenic fever admissions. Among this group of patients, acalculous cholecystitis is more frequent than in general population. Blood and bile cultures should be obtained so it can help to guide antimicrobial therapy. Since there are no guidelines for acute cholecystitis in neutropenic patients, their treatment should be individualized, for those with acalculous cholecystitis cholecystostomy tube placement is preferred; those presenting with calculous cholecystitis can be treated either with cholecystostomy tube if the patient is unstable or laparoscopic cholecystectomy if the patient's general condition allows it.