Is conservative management a safe approach for patients with acute acalculous cholecystitis presenting with an acute abdomen?

Abstract

Patients with acute acalculous cholecystitis (AAC) often present with acute abdominal symptoms. However, recent clinical studies have suggested that some patients with AAC and an acute abdomen, especially when caused by viruses or rheumatic disease, may not require cholecystectomy and that conservative treatment is adequate. Whether cholecystectomy is superior to conservative treatment for patients with AAC presenting with a severe acute abdomen is still uncertain. This was a case series study of AAC-related literature published between 1960 and 2022. In total, 171 cases (104 viral infection-associated AAC and 67 rheumatic disease-associated AAC) were included. The prognoses of patients receiving cholecystectomy or conservative treatment were compared. To account for confounding factors, etiological stratification and logistic regression were performed. The prognosis was similar for patients undergoing cholecystectomy and conservative treatment (P value .364), and virus infection-associated AAC had a better prognosis than rheumatic disease-associated AAC (P value .032). In patients with AAC caused by viruses or rheumatic disease, the acute abdomen can be adequately managed by conservative treatment of the underlying etiology and does not mandate surgical intervention.

1. Introduction

Acute acalculous cholecystitis (AAC) describes gallbladder inflammation in the absence of gallstones. AAC accounts for 5% to 10% of acute cholecystitis cases in adults, whereas in children it accounts for 50% to 70% of all cases.^[1] AAC has many causes including severe trauma, viral infection, rheumatic disease, metabolic abnormalities, malignancy, and total parenteral nutrition.^[2] The diagnosis of AAC mainly relies on the recognition of acute abdominal symptoms and imaging such as ultrasound and computed tomography, with a gallbladder wall thickness over 3 mm, gallbladder distension >5 cm in diameter, and pericholecystic fluid accumulation in the absence of gallstones the main diagnostic criteria.^[3] AAC is considered to be more life-threatening than typical calculous cholecystitis due to a higher risk of necrosis and perforation.^[4] Cholecystectomy is a treatment option for AAC in critically-ill patients who fail conservative management. Therefore, about 30% of laparoscopic cholecystectomies are for AAC,^[5] but it is still uncertain whether cholecystectomy improves the prognosis of patients with AAC. Definitive evidence supporting the benefit of cholecystectomy in AAC patients is lacking.

Viral infection and rheumatic disease are 2 major etiologies of AAC.^[6–9] In children, 40% to 70% of AAC cases are caused by viral infections and 20% to 30% of AAC cases are caused by rheumatic disease.^[9,10] These 2 etiologies share several characteristics. First, both present with acute abdominal symptoms including abdominal pain, abdominal distension, epigastric guarding, and/or positive Murphy sign. Second, both derange liver function, as indicated by elevated bilirubin and hepatobiliary enzyme levels. Moreover, imaging reveals hepatosplenomegaly and a thickened gallbladder in both conditions but without gallbladder calculi. However, it is still unclear whether these different etiologies influence outcomes in AAC patients, and there is little evidence to suggest that surgery is effective in these cases.

It would of course be ideal to examine these issues in a prospective clinical study. However, viral and rheumatic AAC are relatively uncommon, so a prospective study is not feasible. Nevertheless, to provide new insights into the best way to treat viral and rheumatic AAC and to determine whether cholecystectomy is appropriate in these cases, we conducted a case series analysis investigating outcomes in viral and rheumatic AAC patients managed with conservative therapy or cholecystectomy.

2. Materials and methods

2.1. Search strategy

The PubMed and Web of Science databases were searched for studies published between 1960 and 2022 using the search terms “acalculous cholecystitis,” “virus,” “infection,” “rheumatic disease,” and “systemic vasculitis.” Relevant articles from the reference lists of review articles were also identified.

2.2. Ethical review

The study protocol was approved by the Ethics Committee of Peking Union Medical College Hospital (Ref: K2045). The ethics committee waived the requirement for informed consent because the data used in this study were obtained from the PubMed and Web of Science databases, so no patients or public were involved in this research.

2.3. Inclusion and exclusion criteria

Articles containing the following data were included: age, sex, symptoms, imaging findings, treatment, laboratory test results (white blood cell [WBC] counts, total bilirubin [TBil], hepatobiliary enzyme levels, C-reactive protein, and erythrocyte sedimentation rate), and outcomes. Articles not written in English or Chinese, or where the full text was unavailable, were excluded. Cohort studies, systematic reviews, imaging studies, and expert opinions were excluded.

2.4. Data extraction and quality assessment

We applied the PRISMA approach to screen for relevant literature (Fig. 1). Two investigators (CC and YW) independently evaluated the literature and extracted the required information from eligible studies. All discrepancies were resolved through discussion with a third investigator (WS). After identifying 671 articles, 153 articles involving 171 cases were eligible: 104 viral infection and 67 rheumatic disease AAC patients.^[8,11–162] Viral infection types included: Epstein Barr virus, hepatitis virus, dengue virus, SARS-CoV-2 (COVID-19), human immunodeficiency virus, and cytomegalovirus. Rheumatic diseases included: systemic lupus erythematosus, adult-onset Still disease, Henoch-Schönlein purpura, polyarteritis nodosa, Churg-Strauss syndrome (eosinophilic granulomatosis with polyangiitis), microscopic polyangiitis, temporal arteritis, Wegener granulomatosis (granulomatosis with polyangiitis), Kawasaki syndrome, systemic juvenile idiopathic arthritis, and juvenile dermatomyositis (Table 1).

Figure 1.

PRISMA flow diagram of included studies. Studies related to acute acalculous cholecystitis (AAC) caused by viral infection or rheumatic disease were identified in the PubMed and Web of Science databases and the reference lists of review articles. Articles not containing age, sex, symptoms, imaging findings, treatment, laboratory test results, and outcomes were excluded. Articles written in languages other than English or Chinese were excluded, and articles, where the full text was unavailable, were also excluded. Only case reports were included.

Table 1.

Etiologies of acute acalculous cholecystitis cases (n = 171).

Etiology	Number of cases
Virus infection	104 (total)
Epstein Barr virus	49
Hepatitis virus	22
Dengue virus	10
Coronavirus	11
Human immunodeficiency virus	5
Cytomegalovirus	7
Rheumatic diseases	67 (total)
Systemic lupus erythematosus	25
Adult-onset Still disease	3
Henoch-Schönlein purpura	8
Polyarteritis nodosa	14
Churg-Strauss syndrome	9
Microscopic polyangiitis	3
Temporal arteritis	1
Wegener granulomatosis	1
Kawasaki syndrome	1
Systemic juvenile idiopathic arthritis	1
Juvenile dermatomyositis	1

2.5. Statistical analysis

All statistical analyses were conducted in the R language environment (version 4.2.1). Conservative treatment was defined as treatment other than cholecystectomy. Treatment outcomes were classified into an “ameliorate” group and a “deteriorate” group according to the following criteria: patients who experienced death, hypovolemic or septic shock, unconsciousness, respiratory distress, neurological impairment, cardiac or renal failure, hematological involvement, gastrointestinal tract hemorrhage, or who were transferred to the intensive care unit due to severe complications during treatment were classed as the “deteriorate” group. The variable “acute abdomen” was defined as the presence of abdominal tenderness, positive Murphy sign, rebound tenderness, or epigastric guarding.

Continuous, parametric variables are presented as means ± standard deviation if the data were normally distributed (as assessed by the Shapiro–Wilk test). Nonparametric variables are presented as medians and interquartile ranges. Categorical variables are expressed as numbers and percentages. Student t test was used for comparisons of parametric data, whereas Wilcoxon test was used for non-parametric data. The chi-squared test was used to compare categorical variables. All statistical tests were conducted using the R stats package. Statistical significance was defined as a P value < .05. Logistic regression was carried out using the R glm package. Missing values in continuous variables in the logistic model were imputed with the R mice package using the multivariate chained equation. Scaling and centering of the continuous variables were performed prior to fitting the regression model. The receiver operating characteristic curve for the logistic regression model was established with the ROCR package.

3. Results

3.1. Differences in clinical characteristics of AAC in the different treatment groups

Table 2 shows the overall clinical characteristics of AAC patients in the different treatment groups. There were significant differences in age (P value < .001) and sex (P value < .001) between the 2 groups, with more young, female patients receiving conservative treatment. The median age of the cholecystectomy group was twice that of the conservative treatment group (38 years old vs 19 years old), and the sex ratio was more balanced in the cholecystectomy group (male number/female number: cholecystectomy group 36/32; conservative treatment group 22/81). AAC attributable to rheumatic disease was more common in the cholecystectomy group, while virus infection-associated AAC was more commonly treated conservatively (P < .001).

Table 2.

Clinical and demographic characteristics of acute acalculous cholecystitis patients by treatment group.

Variable	Conservative treatment*	Cholecystectomy†	P value‡
Total	103	68
Demographics
Age	19 (12–30)	38 (22.3–54.5)	<.001
Male	22 (21.4%)	36 (53%)	<.001
Female	81 (78.6%)	32 (47%)
Etiologies
Viral infection	79 (76.7%)	25 (36.8%)	<.001
Rheumatic disease	24 (23.3%)	43 (63.2%)
Symptoms
Abdominal pain	94 (91.3%)	58 (85.3%)	.33
Acute abdomen	88 (85.4%)	57 (83.8%)	.94
Fever	85 (82.5%)	44 (64.7%)	.014
Hypotension	15 (14.6%)	8 (11.8%)	.76
Vomiting	49 (47.6%)	21 (30.9%)	.044
Laboratory test results
WBC (×109/L) (3.5–9.5)	8.9 (5.8–14.7)	14.9 (8.8–20.2)	.009
Hb (g/dL) (11–15)	11.5 ± 2.2	11.7 ± 2.0	.76
AST (U/L) (10–40)	294 (134.8–561.5)	130.5 (50.0–429.2)	.076
ALT (U/L) (9–50)	268.0 (103.5–609.5)	157.0 (71.7–391.0)	.18
ALP (U/L) (100–250)	379.1 (237.5–583.2)	290.5 (161.2–586.8)	.50
GGT (U/L) (8–55)	165.0 (122.0–243.0)	118.0 (67.7–230.0)	.37
TBil (mg/dL) (0.2–1.2)	2.5 (1.0–5.9)	1.8 (0.7–4.2)	.12
ESR (mm/h) (0–20)	35.0 (23.0–66.0)	75.0 (47.7–98.7)	.037
CRP (mg/L) (<8.2)	4.6 (1.2–10.5)	6.6 (2.6–16.9)	.21
Imaging
Gallbladder wall thickness (mm) (<3 mm)	9.0 (7.0–12.0)	6.2 (4.7–9.3)	.065
Gallbladder distension (cm) (<5 cm in diameter)	18 (17.5%)	20 (29.4%)	.099
Pericholecystic fluid	57 (55.3%)	21 (30.9)	.003
Prognosis
Ameliorate	89 (86.4%)	50 (73.5%)	.056
Deteriorate	14 (13.6%)	18 (26.5%)
Death	1	3
Shock and unconscious	8	2
Other complications‡	5	13

Alb = albumin, ALP = alkaline phosphatase, ALT = alanine aminotransferase, AST = aspartate transaminase, CRP = C-reactive protein, ESR = erythrocyte sedimentation rate, GGT = gamma-glutamyl transferase, Hb = hemoglobin, TBil = total bilirubin, WBC = white blood cell.

^*Categorical variables are presented as number (percentage), while continuous variables are presented as mean ± SD if normally distributed (examined by the Shapiro–Wilk test) and presented as median (interquartile range) if not normally distributed.

^†Categorical variables were evaluated with the chi-squared test, while continuous variables were evaluated by Student t test if normally distributed (examined by the Shapiro–Wilk test) and the Wilcoxon test if not normally distributed.

^‡Other complications include respiratory distress, neurological deficit, cardiac failure, renal dysfunction, hematological involvement, gastrointestinal tract perforation, or hemorrhage.

AAC symptoms were similar between the 2 treatment groups, except for fever and vomiting, which were more common in the conservative treatment group (P = .014 and .044, respectively). Interestingly, hypotension, which is often considered an indication for surgery, was similar between groups (P = .76). Inflammatory indicators such as the WBC count and erythrocyte sedimentation rate were higher in the cholecystectomy group (P = .009 and .037, respectively). There were no significant differences in hemoglobin, hepatobiliary enzymes, TBil, or C-reactive protein levels between groups. Intriguingly, the gallbladder wall seemed to be thicker and pericholecystic fluid was more common on imaging in the conservative treatment group (P = .065 and .003, respectively), but gallbladder distension was more prevalent in the cholecystectomy group (P = .099).

3.2. Prognosis in AAC patients receiving surgery or conservative management

We next investigated the impact of the management approach on prognosis. Deterioration was defined when a patient experienced death, hypovolemic or septic shock, complications including unconsciousness, respiratory distress, neurological impairment, cardiac or renal failure, hematological involvement, gastrointestinal tract hemorrhage, or requiring intensive care unit monitoring. Eighteen of 68 patients undergoing cholecystectomy had symptom deterioration, while only fourteen of 103 patients who received conservative treatment deteriorated, although this difference did not quite reach statistical significance (chi-squared test P = .056; Fig. 2A). Patients were further divided into subgroups by etiology. Virus infection-associated AAC patients received more conservative treatment, while rheumatic disease-associated AAC patients more commonly underwent cholecystectomy. However, there was no difference in prognosis according to treatment decision in either the viral infection AAC group (P = .34) nor the rheumatic disease AAC group (P = .86; Fig. 2B and C).

Figure 2.

Prognosis in acute acalculous cholecystitis (AAC) patients managed with conservative treatment or cholecystectomy. (A) The prognosis of patients with virus-associated AAC or rheumatic disease-associated AAC treated conservatively or by cholecystectomy. The chi-squared test P value is 0.056 (n = 171); (B) The prognosis of virus-associated AAC patients treated conservatively or by cholecystectomy. The chi-squared test P value is 0.34 (n = 104); (C) The prognosis of rheumatic disease-associated AAC patients treated conservatively or by cholecystectomy. The chi-squared test P value is 0.863 (n = 67).

3.3. Logistic regression model of prognosis

To explore the risk factors that may influence prognosis, we applied logistic regression to determine any factors that might predict the clinical outcome (deteriorate vs ameliorate) (Fig. 3A). The variables in the model included sex, treatment type (cholecystectomy or conservative treatment), etiology (viral infection or rheumatic disease), acute abdomen symptoms, age, TBil, gamma-glutamyl transferase, and WBC count. The logistic regression model revealed that the etiology was significantly associated with the outcome, with patients with virus infection-related AAC having a better prognosis than those with rheumatic disease (P = .032). Treatment type and having an acute abdomen were not associated with prognosis (P = .364 and .308, respectively). Sex, age, TBil level, gamma-glutamyl transferase, and WBC were not associated with the prognosis. The receiver operating characteristic curve for the logistic regression model had an AUC of .678 (95% CI 0.575–0.781) (Fig. 3B).

Figure 3.

Logistic regression model for acute acalculous cholecystitis (AAC) prognosis. (A) Forest plot depicting the logistic regression results assessing potential factors influencing the prognosis. Logistic regression results are presented as odds ratios (OR) with 95% confidence intervals; (B) Receiver operating characteristic (ROC) curve analysis for the AAC prediction model. The area under the ROC curve (AUC) was 0. 678, with a 95% confidence interval of 0.575 to 0.781.

4. Discussion

AAC is a multifactorial disease that is traditionally considered to occur during critical illness, after surgery, with parenteral nutrition, and in patients with congestive heart failure, diabetes mellitus, and atherosclerosis.^[2] AAC is generally considered to have a poor prognosis, with approximately 50% of patients developing gangrene and 10% perforation.^[163] Therefore, it is believed that AAC requires timely diagnosis and immediate surgical intervention.^[164] However, recent reports have shown that AAC can arise from other causes such as viral infections and rheumatic diseases, especially in childhood.^[6] Patients with these etiologies mostly present with acute abdominal symptoms, fever, or jaundice, with AAC confirmed by ultrasonography or computed tomography. Interestingly, some studies have suggested that AAC caused by a viral infection or rheumatic disease may have a good prognosis without cholecystectomy.^[141] Thus, whether AAC presenting with an acute abdomen mandates cholecystectomy is still a matter of debate. However, due to the scarcity of AAC cases, there is currently no clinical cohort that can resolve this question.

To address this problem, here we searched for published literature on AAC cases caused by viral infection or rheumatic disease and systematically analyzed the data. The prognosis of patients undergoing cholecystectomy or conservative management was similar. Interestingly, we found that patients with AAC caused by rheumatic diseases were more likely to undergo cholecystectomy than their virus-infected counterparts. Moreover, patients with rheumatic disease-related AAC had a generally worse prognosis than those with AAC caused by viruses. Of note, a combined analysis of all viral and rheumatic disease AAC patients seemed to suggest that choosing cholecystectomy was related to a poor outcome. This confounding effect was eliminated when stratified by etiology and logistic regression. Our work suggests that a management algorithm for AAC patients should include screening for viral infections and rheumatic disease via viral serology (e.g., HAV-IgM) and autoimmune antibodies (e.g., antinuclear antibodies). Although it takes time to screen for etiologies, this can provide important clinical information to guide treatment and might improve life quality by avoiding unnecessary surgical intervention and preserving functional gallbladders. Moreover, AAC patients suffering from acute abdominal symptoms caused by active rheumatic disease might not be fully relieved by cholecystectomy, instead requiring steroid treatment.^[38,53,77]

AAC is heterogenous, with various pathophysiologies, prognoses, and complications requiring different treatment strategies. The pathogenesis of AAC is usually attributed to bile stasis, gallbladder ischemia, or inflammation.^[3] It is believed that the central lesion leading to AAC is gallbladder microcirculation dysfunction together with cellular hypoxia.^[165] Different from the pathophysiology of calculous cholecystitis, which mainly involves increased intraluminal pressure and endothelial inflammation caused by cystic duct obstruction, the fundamental pathological change of AAC is arterial occlusion that leads to microcirculatory disruption, resulting in focal ischemia and consequent necrosis. Gallbladder microcirculation dysfunction can be caused by hypovolemic shock or cardiac arrest,^[166] thrombocytopathy,^[167] atherosclerosis, surgery, or trauma.^[168] These classical AAC risk factors usually overlap with severe illness, especially in older, male patients. In such patients, cholecystitis is but one of many manifestations of systemic inflammatory response syndrome. However, some other AAC populations are much younger. Infection-associated AAC patients are usually children or teenagers,^[1] while most rheumatic disease-associated AAC patients are female. Immunohistopathological analyses have revealed virus antigens and lymphocytic infiltrates in the biliary and gallbladder epithelium in hydropic virus infection-associated AAC gallbladder tissue.^[148] Moreover, gallbladders in rheumatic disease-associated AAC are also infiltrated with mononuclear cells, mainly T lymphocytes, at the site of gallbladder perforation.^[56] These studies suggest a cell-mediated inflammatory response in AAC, rather than pure bile stasis or hypovolemic-induced ischemia. Therefore, the pathogenesis of AAC is diverse and driven by a broad spectrum of etiologies that may require individualized treatment strategies to improve outcomes.

A curious finding from our literature search was that, in the older literature, both viral and rheumatic disease-associated AAC were mainly treated by cholecystectomy rather than conservatively. Of the 171 cases in our review, 29 cases were published before 2000, and, in these cases, 86.2% (25/29) underwent cholecystectomy, much higher than the 30.3% of AAC cases referred for cholecystectomy over the last 22 years. This suggests that the clinical approach to surgical intervention is changing over time, with the concept that surgical intervention should be reserved only for when conservative treatment fails to manage the patient illness gradually becoming the mainstream view. The goal now is to prevent unnecessary invasive treatment and avoid iatrogenic harm.

This study has several limitations. First, we could not eliminate publication bias, which may have impacted our analysis. Second, some data were missing from some papers, and AAC symptoms may not have been fully recorded. Nevertheless, the information gained from previous studies is valuable since AAC is still poorly understood. Third, AAC caused by viral infections or rheumatic disease is not representative of the entire AAC population, and AAC caused by other etiologies, such as parenteral nutrition and congestive heart failure, are still a management challenge. Further studies are needed to improve the management algorithm for other forms of AAC.

5. Conclusion

Our study suggests that when patients with AAC caused by viruses or rheumatic disease present with acute abdomen symptoms, conservative treatment is appropriate and has similar outcomes to cholecystectomy. Furthermore, AAC is heterogenous and different etiologies lead to distinct prognoses. Managing the primary cause of AAC and providing supportive care to patients with severe acute abdomen symptoms is sufficient to relieve patients’ symptoms.

Author contributions

Conceptualization: Yang Jiao.

Data curation: Chuheng Chang, Youyang Wang, Wen Shi, Haifeng Xu.

Formal analysis: Chuheng Chang, Youyang Wang, Wen Shi, Haifeng Xu.

Funding acquisition: Yang Jiao.

Methodology: Chuheng Chang, Yang Jiao.

Investigation: Youyang Wang, Yang Jiao.

Project administration: Chuheng Chang.

Resources: Xiaoming Huang, Yang Jiao.

Writing – original draft: Chuheng Chang.

Writing – review & editing: Youyang Wang, Wen Shi, Haifeng Xu, Xiaoming Huang, Yang Jiao.