Customize your JAMA Network experience by selecting one or more topics from the list below.
Kayaalp C, Sengul N, Akoglu M. Importance of Cyst Content in Hydatid Liver Surgery. Arch Surg. 2002;137(2):159–163. doi:10.1001/archsurg.137.2.159
Cyst content in hydatid liver surgery is a determinant of postoperative cavity-related complications.
Cohort analytic study.
Referral public hospital.
Sixty-seven patients who had conservative surgery for hydatid liver disease were analyzed prospectively. Cysts were grouped as to the contents. Cysts containing bile and/or pus were considered complicated, whereas others were uncomplicated. After partial cystectomy, cavities were managed with external drainage, omentopexy, or introflexion, determined by the choice of the surgeons.
Main Outcome Measures
Outcomes were measured by postoperative cavity-related complications, such as biliary leakage (leakage for 10 days or less), biliary fistula (leakage for more than 10 days), cavity infections, and postoperative hospitalization.
Patient and cyst characteristics were comparable among the groups. External drainage technique had slightly more frequent cavity-related complications than omentopexy or introflexion (1.5-fold to 2-fold difference). More importantly, complications were significantly more frequent in the complicated cysts than in the uncomplicated cysts (7-fold to 30-fold difference). Mean postoperative hospitalization time for uncomplicated cysts was 6.5 days, and it was similar whether they were treated by external drainage, omentopexy, or introflexion (7.0, 6.2, and 5.8 days, respectively). Mean postoperative hospitalization time for complicated cysts was longer than for uncomplicated cysts (17.5 days; P = .008).
Uncomplicated cysts have lower complication rates and short hospital stay with each cavity management technique. Complicated cysts have higher complication rates and longer hospital stay regardless of the management technique. Therefore, complicated and uncomplicated cysts should be considered different forms of the disease and evaluated differently.
HYDATID DISEASE of the liver can be managed surgically by 2 methods, radical or conservative. The radical method involves total excision of the cyst by pericystectomy or hepatectomy, with its attendant increase in operative risk for a benign disease. The conservative method includes removal of cyst contents, inactivation of scolices, and management of the residual cavity. Although the radical method has a lower recurrence risk, the conservative method is safer and easier.1-3
The most controversial aspect of the conservative method is the management of the residual cavity. A wide variety of techniques, such as marsupialization, external drainage, capitonnage, introflexion, and omentopexy, have been recommended for prevention of bile leakage and secondary cavity infections. Most studies that compare these management techniques do not take into consideration cyst contents and only focus on surgical techniques.2,4-7 Our group previously showed that cyst content was an important factor for bile leakage and cavity infection after hydatid liver surgery.8
The aim of this study was to compare the different cavity management methods according to cyst contents after partial cystectomy and to evaluate the importance of cyst content in hydatid liver surgery.
This was a single-center study in which all patients who had surgery for hydatid liver disease between January 1, 1999, and December 31, 2000, were included. The diagnosis was made in all patients by abdominal ultrasonography and computed tomography. The radiologic types of the cysts were classified as unilocular (Gharbi types I and II), multilocular (Gharbi type III), or degenerated (Gharbi type IV).9 Surgery was performed for cysts larger than 4 cm; smaller cysts were treated medically and were observed as to whether they reached the liver capsule.10 Preoperative chemotherapy with albendazole (10 mg/kg per day) was begun 3
to 7 days before surgery.
The liver was exposed through a right subcostal incision. Radical surgery was performed when appropriate, and total cystectomy was considered only for cysts located peripherally in the liver. All other cysts were treated by conservative methods. The area around the cysts was carefully isolated by gauze packs soaked in a scolicidal agent (0.5% cetrimide and 0.05% chlorhexidine gluconate combination; Savlex; Drogsan, Ankara, Turkey). The cyst was first aspirated, and if bilious or purulent contents were found, the cyst was classified as complicated. Purulent contents were sampled for microbiologic analyses and the aspirate was replaced with the scolicidal agent. If there was no bile staining or pus, the cyst was considered uncomplicated. After the cyst wall was opened, all contents were removed and the cavity was then wiped with scolicide-soaked swabs. The cavity was examined carefully for sites of biliary leakage for 5 minutes and, if present, they were sutured with 3-0 silk. After partial cystectomy, cavities were managed with external drainage, omentopexy, or introflexion, according to the choice of the surgeons. The cavities treated with omentopexy and introflexion were also drained to control biliary leakage.
Drains were removed on the third postoperative day, provided no biliary drainage was seen. Patients who continued to have biliary drainage in the first 10 days postoperatively were considered to have biliary leakage. Longer biliary drainage was classified as biliary fistula. If the bile output of the fistula was more than 100 mL, we planned endoscopic retrograde cholangiopancreatography and nasobiliary drainage. If the bile output was less than 100 mL, patients were treated conservatively. All cavity infections were diagnosed by purulent drainage from cavity drains or clinical findings of infection (temperature >37.8°C and leukocyte count >10 000/µL) combined with imaging of the cavities with abdominal ultrasound and computed tomography.
Patients and cysts were grouped as to cavity management techniques (external drainage, omentopexy, and introflexion) and cyst contents (complicated and uncomplicated). Because patient and cyst numbers are not equal, patient and cyst characteristics were examined separately. Patient characteristics such as age, sex, and primary or recurrent disease were analyzed in each group. Patients who had multiple cysts were categorized according to the largest cyst in the liver. Cyst characteristics such as radiologic type, location, and diameter were analyzed in each group. Postoperative outcomes were measured by biliary leakage, biliary fistula, cavity infections, and postoperative hospitalization. For statistical analysis, the χ2 test was used. Fisher exact test was used when a value lower than 5 was expected. Age, diameter of the cyst, and hospital stay were compared by Kruskal-Wallis variance analysis. SPSS for Windows software (Version 10.0; SPSS Inc, Chicago, Ill) was used for the statistical analysis.
Eighty-five consecutive patients who had hydatid liver disease were treated during the period of study. Two patients who had cysts smaller than 4 cm were treated medically. Thirteen patients who had a total pericystectomy were excluded from the analysis, along with 3 more patients who had common bile duct exploration and biliary drainage. The remaining 67 patients (22 men and 45 women) were studied prospectively. The age range was 18 to 87 years (mean ± SD age, 43 ± 15 years). Fourteen patients (21%) had had previous hydatid liver surgery. Fifty-six patients (84%) had a single cyst and 11 patients (16%) had multiple liver cysts. Patient characteristics of the groups were comparable, and there were no differences between the groups according to age, sex, and previous hydatid liver surgery (Table 1).
A total of 82 cysts were treated in 67 patients. The diameter of the cysts ranged from 5 to 23 cm (mean ± SD, 10.2 ± 5.2 cm). Radiologic appearance was unilocular, multilocular, and degenerated in 27 (33%), 25 (30%), and 30 (37%), respectively. Fifty-four cysts (66%) were located in the right lobe. Cyst characteristics according to radiologic type, location, and diameter were comparable among the groups (Table 2).
While 56 cysts (68%) were uncomplicated, 26 (32%) had bilious and/or purulent contents. Only 9 cysts were sampled for microbiologic analysis; 6 of them were infected by Escherichia coli or Klebsiella pneumoniae. The cyst cavity was managed by omentopexy in 16 cysts (20%) and by introflexion in 21 cysts (26%). External drainage was performed in 45 cysts (55%).
In 17 patients (25%), postoperative biliary leakage occurred through the abdominal drain. Ten (15%) closed spontaneously within 10 days, with the amount of drainage varying between 50 and 400 mL. The external drainage technique resulted in more frequent biliary leakage than omentopexy and introflexion (24.4% vs 18.7% and 14.3%, respectively; P = .17) (Table 3). More importantly, biliary leakage was significantly more frequent in the complicated cysts than in the uncomplicated cysts (61% vs 2%; P<.001).
Only 7 patients (10%) had postoperative persistent biliary leakage, classified as biliary fistula; the amount of bile drainage was between 50 and 200 mL. Four of them required endoscopic retrograde cholangiopancreatography and nasobiliary drainage, and the remaining 3 patients were treated conservatively. All fistulas closed within 8 weeks. External drainage resulted in more frequent biliary fistulas than did omentopexy and introflexion (11.1% vs 6.3% and 4.8%, respectively; P = .29) (Table 3). All biliary fistulas resulted in cavity infections. Biliary fistula was again significantly more frequent in complicated cysts than in uncomplicated cysts (23% vs 2%; P = .003).
Cavity infections occurred in 12 patients (18%), mainly in the complicated cysts (5% vs 35%; P = .001). Two patients required laparotomy and abscess drainage. The remaining patients were treated medically. External drainage had a higher risk of cavity infection than did omentopexy and introflexion (18% vs 12% and 10%, respectively; P = .32) (Table 3). One patient (1.5%) died of cavity infection, intra-abdominal sepsis, and pulmonary insufficiency after treatment of a complicated cyst by external drainage.
Mean postoperative hospitalization time for uncomplicated cysts was 6.5 days, and it was similar whether they were treated by external drainage, omentopexy, or introflexion (7.0, 6.2, and 5.8 days, respectively). However, the mean postoperative hospitalization time for complicated cysts (17.5 days) was longer than that for uncomplicated cysts (P = .008), and it was longer in the external drainage group (24.5 days) than in the omentopexy (13.5 days) and introflexion (9.0 days) groups (P = .04 and P = .01, respectively).
Surgery remains the main treatment for hydatid liver disease. Although there is no randomized study to compare radical and conservative surgery, some surgeons favor pericystectomy or hepatectomy in most cases,11,12 and some surgeons, especially in endemic areas, prefer conservative surgery.3-6 We performed conservative surgery in most cases (84%), and radical procedures were reserved for cysts located peripherally in the liver. We believe that radical methods in hydatid liver surgery should not be used when the operation may be complicated or risky for the patient, and a policy like ours that performs radical methods selectively may provide short operative time, less blood loss, and low operative risk for this benign disease. In addition, most hydatid liver cysts can be treated successfully by conservative surgery.1-3
Despite these advantages, conservative surgery has a higher recurrence rate than radical excisions, and postoperative cavity-related complications are not uncommon.4,5,11,12
Most studies that compare the cavity management techniques in hydatid liver surgery do not take into consideration the cyst contents and only focus on the surgical techniques. Management of the cyst cavity is a step in the conservative approach for protection against postoperative cavity-related complications. Several techniques are advocated, including marsupialization, external drainage, capitonnage, introflexion, and omentopexy. Selection of the appropriate method is still controversial, and each has advantages and disadvantages. Because marsupialization had a high morbidity and required long hospital stays,4,5 we discarded its use in the 1980s. External drainage via a tube has been reported as a simple and effective method,3,13 and we performed it in most of cases (55%). Capitonnage and introflexion can be used only selectively, because the rigid cyst wall resulting from fibrosis and calcification restricts the use of these techniques. Although omentum has a natural absorptive capacity, omentopexy can hinder repeated operation on the liver.
Biliary leakage is a common postoperative complication that is associated with increased risk of morbidity. The incidence is variable, from 7.9%14 to 28.6%,15 depending largely on the criteria used for defining the leakage and its duration. In this study, the incidence of biliary leakage was 25%. The apparently increased incidence in our study is due to our criteria, which allowed for the inclusion of patients with temporary leakage in addition to persistent cases. The latter occurred in only 10% of the patients. All fistulas ceased spontaneously or with the help of endoscopic retrograde cholangiopancreatography and nasobiliary drainage within 8 weeks.
We found that treatment of uncomplicated cysts by omentopexy or intraflexion resulted in a lower cavity infection rate than did external drainage (0% vs 10%; P = .16), but biliary leakage and fistula rates were similar among the groups (0% vs 3%; P = .55). Hospital stay was also similar in each group for uncomplicated cysts. Previously reported complication rates for uncomplicated cysts ranged between 0% and 20.8% (Table 4), and there are some conflicting results for the treatment of these cysts. Ozacmak and colleagues16 concluded that partial cystectomy followed by introflexion and omentopexy should be the treatment of choice for uncomplicated cysts, but they had an unexpectedly high biliary fistula rate (14%). Dziri and associates17 reported a lower rate of deep abdominal cavity infections in the uncomplicated cysts when treated by omentopexy (13% vs 4.8%), but they did not find lower rates of biliary leakage and fistula, which supports our results. Rakas and coworkers3 found that external drainage was the simplest and safest form of management of the residual cavity in the uncomplicated cysts, and obliteration was unnecessary. Because complication rates were lower in this group and the differences were small in each technique, on the basis of the present findings, it is still very difficult to recommend any one of the management methods in uncomplicated cysts. There is a need for more studies that focus especially on uncomplicated cysts, and the results of these studies will also form the guidelines for the use of a newly popular treatment modality, laparoscopic surgery.19,20 Most studies on laparoscopic hydatid liver surgery include a limited number of patients, usually manage the cavity by external drainage without cavity obliteration, and report high success rates. Therefore, before laparoscopic hydatid cyst surgery is widely accepted, the need for obliteration with the open surgical technique in uncomplicated cysts must be clarified. We conclude that less frequently observed complications (3%-10%) in the uncomplicated cysts can be almost completely eliminated by omentopexy or introflexion. However, this results in more than 90% of patients undergoing a possibly unnecessary obliteration procedure that may cause problems itself.21,22
We believe that all purulent hydatid liver cysts have cystobiliary communications and should be evaluated in the same group with the cysts that contain bilious components and treated in a similar fashion. A few reports have examined specifically the complicated cysts, and in these trials, complications ranged from 13.5% to 89.4% (Table 4). Most authors recommend external drainage in complicated cysts and reserve introflexion and omentopexy for uncomplicated cysts.5,23,24
Aktan and coworkers23 considered infection an absolute contraindication and intrabiliary rupture a relative contraindication for omentopexy. We performed omentopexy and introflexion in the complicated cysts, and we did not observe any increase in cavity-related complications; on the contrary, complicated cysts treated by omentopexy and introflexion had lower rates of biliary complications than those treated by external drainage. Hospitalization was also shorter in the groups that underwent obliteration rather than external drainage. Our findings support those of Dziri and associates,17 who performed omentopexy even in the complicated cysts and found fewer complications than in the no-omentopexy group. Although treatment of complicated cysts with external drainage may result in a higher rate of cavity-related complications and longer hospital stay than obliteration techniques, the incidence of complications was still unacceptably high in both groups (50% vs 71%). Therefore, we conclude that preexisting omentopexy and introflexion techniques seem partly effective in the treatment of the complicated cysts, and additional surgical techniques are required. The addition of biliary drainage,7,18 testing for bile leakage by injection of air or dye into the bile tract,17 preoperative or intraoperative cholangiography,14,18 and internal drainage4,23 may be useful in lowering the complication rates in these cysts.
Although we did not clearly demonstrate that the type of cavity management had a strong preventive role in cavity-related complications, we observed a 1.5-fold to 2-fold lower frequency of biliary leakage, fistula, and cavity infections with obliteration of the cyst cavity by omentopexy or introflexion than with external drainage. This may be due to obliteration of the small biliary orifices in the cyst cavity by omentopexy or introflexion. Mean hospital stay was also shorter in the obliteration groups than with external drainage, especially in the complicated cysts. Although several retrospective studies reported that external drainage had a high cavity-related morbidity and long hospital stay,1,4,5,16 it was not clear how many cysts had complicated content in each cavity management group. There is no doubt that high morbidity rates were most often related to the contents of these cysts rather than to external drainage. We found that the major determinant of postoperative morbidity in hydatid liver surgery was the cyst content. Complicated cysts had higher rates of biliary leakage (30-fold), biliary fistula (11-fold), and cavity infection (7-fold) than did uncomplicated cysts. Therefore, we believe that complicated and uncomplicated hydatid liver cysts should be evaluated separately.
Conservative surgical methods can be used to treat most hydatid liver cysts, with low operative risk and low mortality. We believe that hydatid liver cysts have special characteristics, and these data strongly suggest that these cysts should be separated as to their contents. Uncomplicated cysts have lower complication rates with each management technique. Conversely, complicated cysts have 7-fold to 30-fold higher complication results regardless of the management technique. The course, results, and treatment options for the 2 categories of the same disease are quite different. Therefore, uncomplicated and complicated cysts should be considered different forms of the same disease and evaluated differently. Omentopexy and introflexion can be used to decrease complications in both uncomplicated and complicated cysts. Nevertheless, when compared with the importance of cyst content, the management type had a minor role in the prevention of cavity-related complications (1.5-fold to 2-fold differences). In uncomplicated cysts, cavity obliteration may resolve the risks of cavity complications after evidence from new studies shows better results with obliteration techniques than with external drainage. In complicated cysts, obliteration of the cavity alone is not sufficient for the prevention of complications, and additional procedures seem to be necessary.
Corresponding author: Cuneyt Kayaalp, MD, Kennedy caddesi, Yalim sokak, 8/6 Kavaklidere, 06660 Ankara, Turkey (e-mail: email@example.com).
Create a personal account or sign in to: