Methods
Between January 2000 and August 2008, 56 consecutive patients with symptomatic nonparasitic hepatic cysts were evaluated and treated at our institution.
The decision for surgery was generally based on a minimum cyst size of 5 cm and abdominal pain, dyspnea and/or vegetative symptoms caused by adjacent organ compression or increased abdominal pressure and, therefore, causing gastrointestinal transport impairments, resulting in nausea, vomiting, oesophageal reflux, loss of appetite and/or early satiety. Mainly cysts located on the liver margin or underneath the surface within the parenchyma of the segments II, III, IVb, V and VI were selected for laparoscopic surgery.
There were 49 women and 7 men, with a mean age of 59 years (range 39 to 81 years). 4 patients had PCLD. All patients underwent pre-operative sonography. The echinococcus serology was negative in all cases.
These were the criteria for sonographic evaluation as benign / non-parasitic liver cyst: smooth margins, no septum, watery / clear cyst fluid, no contrast enhancement and calcifications. Further diagnostic procedures were initiated in cases of even the slightest suspicion of parasitic or neoplastic cyst.
All operations were performed as deroofing. The decision for open or laparoscopic surgery was made by the respective surgeon depending on his/her experience, cyst location and relevant co-morbidities (e.g. vena cava thrombosis being a contra-indication against the laparoscopic technique). The decision whether to perform an omentoplasty was also made exclusively by the operating surgeon. For open surgery, access was gained via a J incision. Laparoscopic surgery was performed via a 3-trocar-technique in the region of the right hypochondrium (in individual cases, another trocar was used as auxiliary trocar). For deroofing, we attempted to remove a maximum of the cyst roof ("wide deroofing"). The operating surgeon decided on the dissection method and in most cases, ultrasonic dissection was performed (Ultracision harmonic scalpel, Ethicon Endo Surgery, Norderstedt, Germany).
A specific questionnaire with regard to medical history (including a detailed history of the medication, paying special attention to female sexual hormones), symptoms and quality of life was developed. The patients were interviewed post-operatively and were asked to compare their current clinical complaints with the pre-operative complaints. All patients had additionally been invited for a sonographic and clinical follow-up examination. The pre- and post-operative sonographic examination was standardized on the basis of the Couinaud segment classification. Statistics were performed with SPSS 18 (chi-square test, Kruskal-Wallis test, paired t-test, Wilcoxon test). Patients were analysed regarding demographics, cyst morphology, symptoms and complaints, therapeutic procedures, postoperative complications according to Clavien-Dindo classification and longterm results (synopsis of relevant demographic and clinical characteristics in Table
1).
Table 1
Demographic and clinical characteristics
Male : Female | 7:49 |
Simple liver cysts : Polycystic liver disease | 52:4 |
Age, years (range) | 59 (39–81) |
BMI, kg/m2 (range) | 27 (19–41) |
ASA I, n (%) | 9 (16) |
ASA II, n (%) | 42 (75) |
ASA III, n (%) | 5 (9) |
Diameter of dominant cyst, cm (range) | 12 (6–20) |
Laparoscopic/open deroofing, n (%) | 47 (84)/9 (16) |
Conversion rate (%) | 6.4 |
Simultaneous cholecystectomy, n (%) | 16 (29)* |
Postoperative complication rate (%) | 16** |
Recurrence rate (%) | 8.7 |
An approval of the local ethic’s committee was not needed in this design, since all patients give their consent to follow-up investigations and further use of their data when entering our department. Furthermore, the study design was non-invasive.
Discussion
For SLCs cyst deroofing (laparoscopic wherever possible) is considered as being the therapeutic gold standard. The symptoms in patients with liver cysts are non-specific. Die Zystengröße allein scheint eine untergeordnete Rolle zu spielen, da selbst bei großen kongenitalen Leberzysten asymtpmatische Verläufe keine Seltenheit sind [
3]. It is certain that other factors play a role, in particular the location of the cysts. So far, in the literature, there are hardly any considerations of psychosomatic factors that may play a role in a certain patient proportion which may be difficult to identify; these factors are complex and difficult to measure.
For reasons of clarity of the terminology and because of the fact that too small a fenestration may favour recurrence, the term complete cyst
deroofing should be used in order to emphasize this aspect [
3,
12].
The value of the current study is in its rather concise follow-up (clinical examination, sonography, questionnaire) in a sufficiently large patient population and long follow-up period. Our patient population is comparable to those of other groups on the topic with respect to age, gender distribution, cyst size, symptoms and complication rate [
2,
10,
13]. Studies with more than 50 patients and a specific follow-up are scarce [
2,
10,
11,
13,
14]. A sufficiently long follow-up period is of critical importance in patients with liver cysts because there is evidence that a relatively large patient proportion (5-50% according to the literature) develops recurrent clinical symptoms from cysts within 3–5 years [
2]. The mean follow-up period in this current study is 43 months. This duration was reached or surpassed in only half of the studies published so far [
2]. There are only scarce exact comparisons of pre- and post-operative symptoms in the literature [
2,
11].
Our altogether recurrence rate is as high as 28.3%. For principal reasons in our opinion the type of recurrence (i.e. a "true surgical recurrence" at the former surgical site vs. a "recurrence of the disease" in terms of completely new or enlarged pre-existing cysts) should be considered as different entities in an analysis of surgical outcome. This discrimination can only be made by an accurate imaging during the follow-up and will not be done in the daily routine. Split like this our surgical recurrence rate was 8.7% and recurrent disease occurred in 19.6%, both rates being independent of the surgical approach.
Our study includes 4 patients with PCLD. Cyst deroofing (possibly laparoscopically) in Gigot stage I may provide a long lasting improvement. The recurrence rate in PCLD is generally higher, and more frequently liver resection is the treatment of choice [
2]. Thus, there is a fundamental difference between SLC/PCLD stage I and PCLD stage II/III. In cases of diffuse disease, reduced nutritional status, extreme hepatomegaly and ascites, liver transplantation may be considered (in cases of GFR < 40 ml/min with simultaneous kidney transplantation) [
15].
As in selected patients with PCLD the decision for resection or deroofing may be made, these cases were primarily included in the analysis. However, numerous reports in the literature suggest that patients with PCLD should be analyzed separately. Even though our 4 patients with PCLD should not have falsified our results significantly, we would strictly separate these two entities in future studies.
In all studies, females predominate (1:3 – 1:9) [
2,
10,
12,
13]. The age peak is around 60 years, as it is in our patient population. Between the age of 30 and 60 years, there is a progressive growth of cysts for reasons that are not known so far [
16]. Several authors have linked this to the influence of estrogens [
17]. In our patient population, we did not observe a significant association between female sexual hormones and symptoms or cyst size.
16/56 (29%) of our patients underwent simultaneous cholecystectomy for concomitant cholecystolithiasis or immediate proximity of the gallbladder to the cyst (8 open, 8 laparoscopically; 6 concomitant cholecystolithiasis, 9 immediate proximity, 1 concomitant cholecystolithiasis and immediate proximity).
The decision for surgery was based on the liver cyst even though it is not possible to clarify definitively if in case of concomitant gall stones the reported symptoms may be attributed to these (none of these patients did show typical signs of gall stone related disease, such as relevant elevated laboratory parameters, sonographic criteria or typical symptoms in the medical history). In case of gall stones, we generally perform a cholecystectomy simultaneously at our institution, except in cases where the patients decides otherwise.
In 84% of the patients, the cyst deroofing was started as a laparoscopic procedure, the conversion rate was 6.4%. There was previous abdominal or gynaecological surgery in 73% and 49%, respectively, while only in one case the previous abdominal surgery (Billroth-II resection) was the reason for a primary open procedure. In a total of 9 cases, the surgical procedure was primarily open (3 unfavorable cyst locations, 2 vena cava thromboses, 1 PCLD, 1 recurrence, 1 abscess, 1 Billroth-II resection). As reported in the literature, this approach was reserved for particular situations. In both cases of a concomitant vena cava thrombosis, successful thrombectomy was performed.
Three fourths of the patients were ASA stage II and almost one tenth were stage III. This explains the comparatively long hospital stay of 6 and 8 days, respectively (laparoscopic and open group, respectively).
In the literature, there are occasional reports about predominant affection of the right or left lobe. We could not confirm that observation in our patient population. In one third of our patients with laparoscopic surgery, segments IVa, VII and VIII were affected; access to these is considered to be difficult. This is as such not a contraindication against the laparoscopic approach, neither did it affect the conversion rate in a negative way. We believe that this confirms the experience of other authors, that this problem is met by an on-going learning curve on one hand and by specific patient positioning (modified left sided position) as required on the other hand.
In the laparoscopy group, the intra-operative complication rate was 17% (possibly related to the learning curve) while it was 0% in the laparotomy group (difference not significant). Those were mainly bleeding complications which led to conversion in 2 cases. The post-operative complication rate of 16% was within the data reported in the literature. 3 cases (Clavien-Dindo grade IIIa) required intervention for bile leakage (sonographically guided drainage and/or ERC). The difference in the post-operative complication rate (laparoscopy vs. laparotomy 13% vs. 33%) was significant but does reflect in general the selection of "more difficult" patients for open surgery.
The use of omentoplasty is being discussed controversially in the literature. We have employed it in 14% of our patient population (8/56; 5 laparoscopically, 3 open). None of our patients with omentoplasty developed a recurrence. However, in one case, there was intra-operative bleeding from the omentum majus during laparoscopic omentoplasty; this did not necessitate conversion. Because of the few cases of omentoplasty in our patient population, we cannot make a definitive statement. There are reports about recurrence in the literature despite omentoplasty [
10,
18]. Furthermore, there is evidence that omentoplasty may cause additional complications or favour recurrence [
10,
19]. The data in the literature are controversial and there as many proponents as opponents of omentoplasty [
2,
9,
10,
18‐
21] while in more recent publications the value of omentoplasty is being viewed critically [
10,
18,
19,
21]. Currently, a falciform ligament pedicle graft is being proposed as an alternative prophylaxis against recurrence, but there are no long-term data available so far [
22].
Half of our patient population (23/46, 50%) were completely symptom-free after cyst deroofing. The other half reported persisting symptoms. Only in some of these patients a new cyst is the explanation for the reported complaints.
This high rate may be explained by the design of the interview. 11 symptoms, possibly related to liver cysts, have been included in the interview. As soon as even only one of the 11 possible symptoms was reported in the post-operative interview, this was declared as persistent clinical complaints. In retrospect, this interview design is un-specific. Finally, the liberal decision for surgery may be an explanation as symptoms which may have had other causes may have not been decisively clarified and may have been falsely attributed to liver cysts. It may be beneficial to assess the psychosomatic aspects separately in order to relate them to the respective somatic complaints.
Our work also suggests that un-specific symptoms may have been judged too un-critically as being the cause of clinical complaints and, thus, have been judged to be relevant for the decision making (Table
3). The association between cyst location and the specific complaints should possibly have been established more precisely and only a very obvious association should lead to surgery (e.g. a relatively large cyst exerts pressure on the gastro-duodenal junction). We will, thus, perform more intensive diagnostics (in particular endoscopically) of the vicinity in the future in order to rule out concomitant disease. Furthermore, we consider the introduction of the cyst aspiration test in the future for better patient selection.
Irrespective of the accuracy of the follow-up, a recurrence rate of approximately 5% and an associated reoperation rate of less than 10% may be assumed [
2,
23,
24]. As opposed to earlier reports in the literature [
3,
25], our data and numerous other studies show, that the laparoscopic procedure as such is no longer a risk factor for a higher probability of recurrence [
2,
11,
13]. Carrying a low rate of complications and conversions, it is, thus, to be considered the treatment of choice for the vast majority of cases. In our study, we have mainly used the ultrasound dissector in the laparoscopically treated patients. Several authors see a potential association of the recurrence rate with the dissection technique. However, the development of recurrence is supposedly more affected by the extent of the fenestration/deroofing (and, thus, only indirectly by the dissection technique) [
10,
26].
Several groups report the avoidance of cyst recurrence after an ablation of the cyst lining in the residual fenestrated cystic cavity with argon beam coagulation [
25,
27]. We have used this technique so far only for liver resections. The results with argon beam coagulation to the cyst lining in the literature appear to be promising so that we will use it in the future in order to avoid cyst recurrence.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
HS collected the data, analysed the data and wrote the manuscript; FR analysed the data and wrote the manuscript; JF collected the data, analysed the data and wrote the manuscript; KJ analysed the data and wrote the manuscript; YD analysed the data and wrote the manuscript; GT performed the ultrasound investigations, collected the data and wrote the manuscript; US analysed the data and wrote and revised the manuscript. All authors read and approved the final manuscript.