Introduction

In Western countries, acute calculous cholecystitis is a common disease with a high socio-economic impact. Cholecystectomy is considered the treatment of choice in patients deemed fit for surgery [1–4]. In surgical practice, cholecystectomy is one of the most frequently performed procedures, mostly performed by laparoscopy nowadays. The rationale for cholecystectomy is based on the old adage that an inflammatory focus should be eliminated immediately from the body to prevent clinical deterioration.

Even though cholecystectomy is the treatment of choice, the benefit of surgery in case of acute calculous cholecystitis has never been properly researched. Prospective, let alone randomized studies demonstrating superiority of surgical over conservative treatment are lacking. Performing cholecystectomy in case of acute cholecystitis prevents further episodes of gallstone-related disease, but the relatively high complication rate associated with both early and delayed cholecystectomy (i.e. 15 and 30 %, respectively [5]), especially in high-risk patients, should not go unnoticed. The decision to perform surgery should therefore be well considered. Conservative treatment prevents the surgical risk; however, leaving the gallbladder in situ may cause recurrent gallstone-related disease.

To assess whether it is safe (or even safer) to leave the gallbladder in situ in patients with acute calculous cholecystitis, the feasibility and safety of conservative treatment during index admission should be assessed. Subsequently, the recurrence rate of gallstone-related disease during long-term follow-up should be evaluated. It has been demonstrated that delayed cholecystectomy is associated with significant higher complication rates than early cholecystectomy [5], and therefore conservative treatment is only feasible if a delayed cholecystectomy is not required.

The aim of this study was to assess the available evidence concerning the feasibility of conservative treatment of acute calculous cholecystitis. The question for this review is thus twofold: (1) Is conservative treatment for acute calculous cholecystitis effective and safe during index admission? (2) What is the recurrence rate of gallstone-related disease during long-term follow-up?

Methods

A systematic review was conducted following the guidance of the Centre for Reviews and Dissemination concerning undertaking reviews in health care and reported in accordance with the PRISMA statement [6, 7].

Literature search

In May 2015, two authors (C. L. and J. O.) independently performed a literature search to identify studies reporting on conservative treatment for adults suffering from acute calculous cholecystitis. MEDLINE, Embase and Cochrane Library databases were searched for papers using the keywords: “acute cholecystitis” in combination with “conservative” or “antibiotic” or “anti-bacterial” or “non-invasive” or “non-surgical” or “non-operative” or “observation” or “drain” or “cholecystostomy” or “delayed/interval/planned/elective/late cholecystectomy”. The search was limited to articles published in English and Dutch and published after January 1990.

Study selection

All titles and abstracts of publications were independently identified and reviewed for relevance by two authors (C. S. and J. O.), with referral to a third author (D. B.) to resolve queries in case of discordant opinions. Subsequently, full-text papers were retrieved and checked. A cross-reference search of the remaining articles was performed to identify other studies previously missed. In case of unclear methods or results, authors were contacted to seek clarification. If the same data were reported in two or more publications, the most comprehensive paper was selected. Reviews, case reports, unpublished data as well as articles of which no full text was available were excluded (Fig. 1).

Fig. 1
figure 1

Flow chart of study selection process

Full size image

Conservative treatment during index admission was defined as non-invasive management of acute calculous cholecystitis, i.e. supportive care with or without antibiotics. Uneventful resolution of symptoms without the need for intervention was considered as a successful conservative treatment. To reduce the risk of bias, only prospective studies were included.

To evaluate the long-term success of non-surgical management of acute calculous cholecystitis, the inclusion criteria were adjusted. Only studies with at least 12-month follow-up and explicit information on the recurrence of gallstone-related disease were included. Studies describing patients who initially underwent percutaneous cholecystostomy were also eligible, since these patients still have their gallbladder in situ after initial successful non-surgical treatment of acute calculous cholecystitis and may therefore develop recurrent gallstone disease. Due to the limited amount of prospective studies with long-term follow-up, retrospective studies were also included.

Data extraction

Data were extracted by two independent authors (C. S. and J. O.) with referral to a third author (D. B.) in case of any disagreements. The characteristics of the included studies are visualized in Tables 1 and 2.

Table 1 Demographic information and findings of included studies concerning short-term follow-up
Full size table
Table 2 Demographic information and findings of included studies concerning long-term follow-up
Full size table

Validity assessment

The methodological quality of the articles was independently assessed by both authors. The MINORS scoring scale was used to assess the risk of bias in non-randomized studies, with a global ideal score of 16 (Table 3) [8]. The Cochrane Collaboration’s tool was used to assess the risk of bias in randomized controlled trials (Table 4) [9].

Table 3 Methodological quality of the included non-randomized studies (MINORS)
Full size table
Table 4 Methodological quality of the included randomized controlled trials (Cochrane Collaboration’s tool)
Full size table

Data analysis

The pooled success and mortality rate of conservative treatment during index admission and the pooled recurrence rate of gallstone-related disease during long-term follow-up were calculated with a random-effects model, using meta-analysis software version 3.1. Statistical heterogeneity among the included studies was determined by using forest plots, by performing a χ 2 (“Chi-squared”) heterogeneity test and by calculating the I 2-index. A high I 2-index represents a high suspicion of heterogeneity. All pooled event rates were shown in forest plots despite the level of heterogeneity.

Results

Figure 1 shows the article selection according to the PRISMA statement [6, 7]. A total of 1343 references were identified from medical journal databases. No new articles were identified by cross-reference search. After removing duplicates, 1169 potentially relevant studies were screened based on title and abstract, leaving 60 full manuscripts to be assessed for eligibility concerning efficacy of conservative treatment during index admission and 71 manuscripts concerning long-term outcome of non-surgical treatment of acute calculous cholecystitis. As a result of this assessment, inclusion criteria were met in 14 and 10 studies, respectively. There was total agreement among the authors regarding the inclusion of studies.

Conservative treatment of acute calculous cholecystitis during index admission

Fourteen prospective studies, either randomized [10–17] or non-randomized [18–22], were included. Seven trials [10, 12, 13, 16, 17, 21, 23] were designed to compare emergency cholecystectomy with delayed cholecystectomy. Patients assigned to delayed cholecystectomy were treated conservatively during index admission, followed by elective laparoscopic cholecystectomy several weeks after discharge. Data of these patients were used to assess the success rate of conservative treatment during index admission. Three studies [14, 15, 22] were designed to compare conservative management with another treatment for acute calculous cholecystitis (e.g. sphincterotomy or cholecystostomy), whereas in the remaining studies [11, 18–20], all patients were treated by means of conservative management.

Table 1 shows the characteristics of the included studies. A total of 1315 patients with acute calculous cholecystitis were analysed. The severity of cholecystitis was explicitly reported in three studies [10, 11, 18]; all three concerned patients with mild cholecystitis. One study [14] excluded patients that required urgent surgical or percutaneous management, thus probably severely ill patients. In the remaining studies, the severity of cholecystitis was not mentioned [12, 13, 16, 17, 19–23].

Outcome

A total of 1315 patients, included in 14 studies, were treated conservatively for acute calculous cholecystitis, of whom 1137 (87 %) showed an uneventful resolution of symptoms without the need for intervention. Conservative treatment included bowel rest and intravenous administration of broad-spectrum antibiotics and fluids. Table 1 shows specific information about the antibiotic treatment, as far as described in detail in the articles. Failure of conservative treatment was determined at the discretion of the attending surgeon based on subjective findings, such as lack of improvement, or objective findings, such as worsening of clinical signs and laboratory results. In case of failure of conservative treatment, either emergency cholecystectomy [13, 15–18, 21–23] or percutaneous cholecystostomy [11, 19, 20] was performed.

The mortality rate associated with conservative treatment during index admission was 0.8 % (11 of 1315) [10–23]. The highest mortality rate (17 %) was reported by Hatzidakis et al. [15]. This study included 42 high-surgical-risk patients with an APACHE-II score ≥12, of whom seven died during index admission.

A total of 245 patients diagnosed with explicitly mentioned mild cholecystitis were included in three studies [10, 11, 18]. In this specific group, conservative treatment was successful in 96 % (236 of 245). Only nine patients required intervention due to failure of conservative treatment, of whom seven underwent percutaneous cholecystostomy and two cholecystectomy. Mortality in this group was nil. Mean length of hospital stay was mentioned in only one study and was 4 days [11].

The pooled success rate of conservative treatment during index admission was 86 % (95 % CI 0.8–0.9) (Fig. 2). There was a strong heterogeneity among the included studies (I 2 = 95 %). A pooled analysis of only randomized controlled trials showed a success rate of 91 % (95 % CI 0.9–1.0) with an I 2-index of 82 % (forest plot not shown). The pooled mortality rate of conservative treatment during index admission was 0.5 % (95 % CI 0.001–0.009) with an I 2-index of 0 % (Fig. 3).

Fig. 2
figure 2

Forest plot of success rate of conservative treatment during index admission

Full size image
Fig. 3
figure 3

Forest plot of mortality rate of conservative treatment during index admission

Full size image

Long-term outcome of non-surgical treatment of acute calculous cholecystitis

To evaluate the long-term outcome of conservative treatment, ten studies were included: two randomized controlled trials [15, 24], three retrospective studies of prospectively collected data [25–27] and five retrospective studies [28–32]. The characteristics of the selected studies are summarized in Table 2. A total of 526 patients, who were initially treated conservatively for acute calculous cholecystitis, and in whom cholecystectomy was not electively planned, were analysed. Follow-up ranged from 12 months to 5 years. One randomized controlled trial [24] was specifically designed to examine the long-term efficacy of complete conservative treatment versus immediate surgery in non-high-risk patients with acute calculous cholecystitis.

Outcome

During long-term follow-up, 117 of the 526 patients (22 %) developed recurrence of gallstone-related disease after initial non-surgical treatment of acute calculous cholecystitis. The recurrence rate varied substantially across the included studies and ranged from 3 to 47 %. The definition of recurrence of gallstone disease differed between studies. Eight studies [25, 27–32] described recurrent gallstone disease only as recurrence of acute calculous cholecystitis, whereas two studies [24, 26] included all gallstone-related problems. The randomized controlled trial specifically designed to examine the long-term efficacy of complete conservative treatment included 33 patients, of whom 10 (30 %) experienced gallstone-related events during a median follow-up of 14 years [24].

The time from initial treatment to recurrence varied between the included studies, but the recurrence mainly occurred within 2 years after initial cholecystitis: three studies [15, 26, 32] reported a range from 2 to 24 months, one study reported a mean of 14 months [29], and two studies reported a median of 2 [28] and 15 months [24]. Recurrent diseases were successfully controlled by conservative medical measurements [26, 27, 32], cholecystectomy [24, 30, 31] or percutaneous cholecystostomy [25, 28].

The pooled recurrence rate of gallstone-related disease during long-term follow-up was 19.7 % (95 % CI 0.1–0.3) (Fig. 4). There was a strong heterogeneity among the included studies (I 2 = 90 %). When dividing the included studies in two groups based on duration of follow-up, a pooled recurrence rate of 22 % (95 % CI 0.06–0.04) was found for studies with a follow-up of <2 years (I 2 = 87 %) and a recurrence rate of 18 % (95 % CI 0.04–0.3) for studies with a follow-up of >2 years (I 2 = 93 %) (forest plot not shown).

Fig. 4
figure 4

Forest plot of recurrence rate of gallstone-related disease during long-term follow-up

Full size image

Quality assessment of the included studies

The results of the methodological quality assessment of the included studies are shown in Tables 3 and 4. Ten studies [10–17, 23, 24] were randomized controlled trials, eight studies [18–22, 25–27] collected data according to a protocol established before the commencement of the study, whereas five studies [28–32] did not have such protocol and identified all patients retrospectively. None of the studies had blinded evaluation of the endpoint due to the nature of the intervention and study. Overall, the included studies were of an estimated moderate quality.

Discussion

This systematic review demonstrates that conservative treatment during index admission is successful in 87 % of all patients with acute calculous cholecystitis. Especially in mild acute cholecystitis, conservative treatment appears safe and effective: 96 % of the patients showed uneventful resolution of symptoms without the need for intervention. After initial non-surgical treatment, 22 % of the patients developed recurrent biliary symptoms, mainly within 2 years after initial cholecystitis. Pooled analysis shows comparable results. According to a randomized controlled trial [24] with a median follow-up of 14 years, the likelihood of recurrent gallstone disease was slightly higher (30 %), but this study included only 33 patients.

When comparing surgical with non-surgical treatment, and determining the feasibility of the latter, not just the likelihood of recurrence but also the surgical risk of the patient should be regarded. In patients with advanced age and/or severe comorbidities, the risk of recurrence is anyhow reduced, due to a relatively limited survival time [33]. In these patients, non-surgical management can be considered as a definitive treatment. Younger, non-high-risk patients would probably easier withstand conservative treatment compared to their older counterparts, but in the same time, are exposed to an increased risk of recurrence due to their longer life expectancy. In view of prevention of future episodes of gallstone disease, cholecystectomy might be a reasonable choice [24]. Nevertheless, the risk–benefit profile shifts towards non-operative management, considering the fact that a second episode of cholecystitis might never occur, as well as the relatively high complication rate associated with cholecystectomy [23].

In medical practice, the tendency to remove an inflamed gallbladder is deeply rooted, even though high-quality evidence is lacking. As in case of cholecystitis, in other acute gastrointestinal inflammations the benefit of surgery over conservative care is not always clearly demonstrated, and therefore, treatment algorithms (in select cases) are slowly shifting towards conservative management. In case of uncomplicated acute diverticulitis, conservative treatment seems feasible and safe [34, 35]; so surgery should be reserved for cases with significant complications, unresponsive to medical treatment [36]. Also in acute appendicitis, randomized trials have shown feasibility and safety of initial non-surgical management [37]. A comparable management algorithm for (mild) acute calculous cholecystitis is plausible [30].

Is it possible to identify patients for whom conservative treatment is most suitable? Concerning the severity of cholecystitis, it seems clear that, without any doubt, conservative treatment during index admission is feasible in mild cases. Concerning the long-term outcome of conservatively treated mild cholecystitis, no definitive conclusions could be drawn, since all (but one) studies with long-term follow-up concerned high-risk patients and none reported on the initial severity of cholecystitis. But given the fact that the overall recurrence rate of biliary symptoms never exceeded one-third of the patients, a conservative approach during follow-up is justified. From the available data of the included studies, it was not possible to identify other factors that might determine the feasibility of conservative treatment.

The success rate of conservative treatment found in this systematic review may have been influenced by several factors. Firstly, studies not reporting on the failure rate of conservative treatment were not included, since it was unclear whether failure had not been mentioned by the authors or did not occur at all. Since the latter is most likely, the success rate of conservative treatment in this review might be underestimated. Secondly, the definition of recurrent gallstone disease differed per study. Some studies described recurrent gallstone-related disease as acute cholecystitis, not reporting whether other complications (e.g. gallstone attack) did not occur or had not been reported. Therefore, the recurrence rate might have been underestimated. Lastly, the duration of follow-up varied substantially between the studies, ranging from one to 14 years. In studies with a relatively short follow-up, a recurrence after follow-up could have been developed, and therefore the recurrence rate might be underestimated. However, the majority of recurrences occur within 2 years after initial cholecystitis, and the risk of new gallstone-related disease decreases over time. Vethrus et al. [14] showed that more than 70 % of the events occurred within 20 months after acute cholecystitis. Similar results regarding the development of complications have been found in patients with symptomatic uncomplicated gallstones [38].

There was a strong heterogeneity among the included studies, demonstrated by the high I 2-values. Possible explanations include differences in sample size, patients characteristics (e.g. age, comorbidities, duration of symptoms, severity of cholecystitis) and antibiotic regimens (e.g. type of antibiotics and duration of treatment) between the included studies. The pooled mortality rate of 0.5 %, on the other hand, was associated with an I 2-index of 0 %, enabling us to conclude that during index admission, conservative treatment of acute cholecystitis is indeed associated with a low mortality.

Cancer may be present in a gallbladder complicated with acute cholecystitis. A large retrospective study of 2700 patients with acute calculous cholecystitis managed with cholecystectomy showed that malignant pathologies were found in 2.3 % of the patients [39]. Gallbladder cancer was most frequently diagnosed in women and patients with advanced age. In our hospital, 590 patients with acute calculous cholecystitis underwent laparoscopic cholecystectomy between 2002 and 2015; pathological examination of the gallbladder showed malignant pathologies in 2 patients (0.3 %). Imaging modalities such as endoscopic ultrasonography, computed tomography and magnetic resonance imaging are useful to diagnose gallbladder cancer [40, 41]. With these figures, however, it is debatable whether additional imaging studies to diagnose malignancies should be performed in all conservatively treated patients.

This systematic review implicates that conservative treatment in case of acute cholecystitis is a feasible treatment option. There is, however, insufficient evidence to demonstrate actual superiority of conservative treatment over cholecystectomy for this indication. Currently, we are designing a prospective randomized controlled trial comparing both treatment options in patients with mild acute calculous cholecystitis.

To determine superiority of one or the other treatment strategy, not only the technical aspects but also the expenditures of both strategies should be evaluated. The costs of emergency cholecystectomy have been subject to many studies, whereas studies focussing on the economic aspects of conservative treatment are lacking. When comparing early with delayed cholecystectomy for acute cholecystitis, total hospital costs are significantly lower for early cholecystectomy [23, 42, 43]. Since our study, however, shows that only 22 % of the patients need to be readmitted for recurrent gallstone-related disease, of whom only some need a surgical re-intervention, the total costs of conservative treatment may be reduced compared to emergency cholecystectomy. A randomized controlled trial and a complementary research using economic and public health approaches including assessment of quality of life, direct and indirect costs are needed.

To the best of our knowledge, this is the first systematic review examining short- and long-term outcome of conservative treatment of patients with acute calculous cholecystitis. Based on the best available evidence, conservative treatment of acute calculous cholecystitis during index admission seems feasible and safe, especially in patients with mild disease. During long-term follow-up, about a quarter of the patients seem to develop recurrent gallstone disease, although this is based on limited data.