Transcript
Clinical Science Surgical treatment of hydatid disease of the liver:25 years of experience Sandro Tagliacozzo, M.D., Michelangelo Miccini, Ph.D.,Stefano Amore Bonapasta, M.D., Matteo Gregori, M.D., Adriano Tocchi, M.D.* Division of Gastroenterologic and Hepato-Bilio Pancreatic Surgery, First Department of Surgery, Sapienza University Medical School, Rome, Italy AbstractBACKGROUND: The aim of this study was to evaluate the results of conservative and radicaltreatment of liver hydatid disease. METHODS: Recordsofpatientswhounderwentsurgeryforliverhydatiddiseasebetween1980and2005were reviewed. Outcomes measured were operative morbidity and mortality, hospital stay, and recurrence. RESULTS: Two hundred fourteen patients underwent conservative treatment (external drainage,marsupialization, omentoplasty), and 240 had radical surgery (hepatic resection, cystopericystectomy).Operative morbidity was 79.9% and 16.2% for conservative and radical procedures, respectively ( P .001). Operative mortality was 6.5% for conservative procedures and 9.2% for radical procedures ( P .3). The recurrence rate was 30.4% in patients having conservative surgery and 1.2% in patientsundergoing radical surgery ( P .001). No recurrences occurred in patients with clear cysts afterconservative surgery. CONCLUSIONS: Cystopericystectomy was a safe and effective procedure that achieved excellentimmediate and long-term results. Hepatic resection should be considered only in exceptional cases,because it involves the unnecessary sacrifice of healthy hepatic parenchyma. Conservative surgery andalternative procedures should be restricted to the treatment of clear cysts and to patients who cannotundergo radical surgery.© 2011 Elsevier Inc. All rights reserved. KEYWORDS: Liver hydatid disease;Cystopericystectomy;Hepatic resection Hydatid disease (HD) is a zoonotic disease caused by thelarval form of Echinococcus . The most common form is E granulosus , which gives rise to cysts, primarily in the liver.HD has a worldwide distribution and is endemic in manycountries in the Mediterranean region, the Middle East andFar East, and South Africa. 1 Because of increasing immi-gration, the disease is becoming more frequent outside en-demic areas. In recent years, knowledge of its etiology andpathogenesis has largely contributed to improved preven-tion and reduced the incidence of HD. 2,3 Clarifying someaspects of the biology of the parasite and of its interactionwith the human host represents the scientific basis for amore rational approach to surgery of liver HD (LHD). 4–6 Pathophysiologic basis of surgical management The adult tape worm consists of a head (scolex) and 3following segments (proglottides). The scolex has 4 suckersand a prominent rostellum armed with a double row of 30 to36 hooks. Sexual, mature organs and countless eggs arecontained in the more distal of the 3 proglottides. Each egg * Corresponding author. Tel.: 39-6-4462127; fax: 39-6-4959357.E-mail address:
[email protected] received December 2, 2009; revised manuscript February8, 20100002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.amjsurg.2010.02.011 The American Journal of Surgery (2011) 201, 797–804 consists of a shell containing 6 hook-armed embryohexacanth (6 hooks), as shown inFigure 1.The adult tape- worm lives in the small bowel of its most common defini-tive host, the dog, attached to the mucosa by its hooklets.Infected eggs pass out with dog feces and contaminate soil,water, and plants. Humans become infected via accidentalingestion of eggs. Once inside the human intestine, the eggsopen, and the embryo hexacanth attaches to and then crossesthe intestinal mucosa and via the portal system migrates tothe liver, where the parasite develops into the larval stage,which is the hydatid cyst. The growth of the cyst leads to theformation of a connective lamina in the surrounding paren-chyma (ectocyst or pericyst). The parasite-derived endocystmay consist of either 1 or 2 layers. The outer one, thelaminated layer, is a totally acellular membrane permeableto water and electrolytes, which protects the cyst from hostenzymes, bile, and bacteria. The inner layer, the germinallayer, is the living element of the parasite (Fig. 2). Cysts with the sole laminated layer are sterile cysts, also calledunivesicular or clear cysts, whereas cysts provided withboth laminated and germinal layers are fertile or multive-sicular cysts. Invaginations of the germinal layer form broodcapsules, each containing 5 to 10 protoscolices. When broodcapsules open, protoscolices are released into the cysticfluid, giving rise to daughter cysts, a process called endo-genic vesiculation. The process of vesiculation occurs notonly inside but also outside the cyst within the pericyst(exogenous vesiculation).We have retrospectively reviewed and analyzed thechanges in the approach and results of surgical treatment of LHD according to our evolving, personal knowledge of thepathophysiology of the parasite. Methods Patients Patients were identified by a computer-based indexingsystem containing data on all patients admitted for LHD atthe First Department of Surgery of the University of Rome“La Sapienza” Medical School between January 1, 1980,and January 1, 2005. The overall study group was dividedinto 2 subgroups: group A (1980–1992) and group B(1993–2005). Criteria for exclusion from this study wereprevious medical or surgical treatment for HD, concomitantextrahepatic HD, and the presence of a bronchobiliary fis-tula. No laparoscopic procedure was performed in the cur-rent series. Suitable records of all patients were extractedfrom the computer database and retrospectively analyzed.Variables considered for analysis were age, gender, lengthof clinical history, number of hepatic cysts, and site. Ac-cording to their location, cysts were divided into superficialand deep or vasculobiliary cysts. Cysts extrinsic to thehepatic parenchyma, however distant from main intrahe-patic vessels and vena cava, were defined as superficialcysts. Cysts deeply located in the hepatic parenchyma, nearthe pedicles and stems of the main intrahepatic vessels andvena cava, were defined as vasculobiliary cysts. Further- Figure 1 Echinococcus granulosus . Adult form of tapeworm.E egg; H hook; P proglottides; S head or scolex; SO sexual organs. Figure 2 Schematic drawing of a liver hydatid cyst. BC broodcapsule; DC daughter cyst; GL germinal layer; L liver;LL laminated layer; P pericyst; S protoscolex. 798 The American Journal of Surgery, Vol 201, No 6, June 2011 more, cysts were classified according to their content asclear or sterile cysts and multivesicular or fertile cysts.Comorbidity, surgical procedures used, operative morbidityand mortality, length of postoperative hospital stay, andrecurrence were other variables considered. Surgery Surgical access consisted of laparotomy (median or sub-costal incision) and, mostly in the first period, thoracoph-renolaparotomy. Surgical procedures were classified as rad-ical or conservative. Liver resection, cystopericystectomy(CP), and subtotal pericystectomy were grouped as radicalprocedures and partial cystectomy with tube drainage,omentoplasty, or “capitonage” as conservative approaches.All clear cysts were treated with conservative procedures.Radical and conservative procedures were never done to-gether. Intraoperative cholangiography was always associ-ated with both types of procedures. Cholecystectomy andeither surgical papillotomy or T-tube biliary drainage wereassociated when cystobiliary communication or elevatedpressure in the biliary tract were detected by intraoperativecholangiography and in case lesions of the intrahepaticbiliary system were determined during the course of sur-gery. More recently, in case of preoperative diagnosis of cystobiliary communication, the biliary tract has beencleared preoperatively by endoscopy. All surgical proce-dures were performed by 2 of the authors (S.T., A.T.). Outcomes The early outcome was determined from the medicalrecords. Postoperative complications were classified as lo-cal when they occurred at the site of surgery or as generalcomplications otherwise. After discharge, patients were fol-lowed periodically in the 3rd, 5th, and 12th postoperativemonths during the first year and thereafter every year. Thefollow-up procedures included physical examination, sero-logic tests, plain abdominal radiography, and ultrasound andcomputed tomography after these diagnostic procedures be-came available. The median follow-up period was 87months (range, 36–190 months). The last control was per-formed in January 2009. Reappearance of live cysts at thesite of a previously treated cyst was defined as local recur-rence. Histology Histology of the cystic wall and pericyst was performedin all specimens from radical procedures. Statistical analysis Comparisons of the continuous variable were performedusing Student’s t test, and categorical variables were com-pared using Pearson’s 2 test. Statistical analyses were per-formed with Statistica 5’97 (StatSoft, Inc, Tulsa, OK). P values .05 were regarded as statistically significant. Results The present study included 454 patients.Table 1liststheir demographic and clinical characteristics. Results of serologic tests (hemagglutination, complement fixation)were positive in 69% of 454 patients. A total of 695 hepaticcysts were treated.Table 2shows the characteristics of thecysts and their distribution. Cysts were solitary in 296 pa- Table 1 Clinical and demographic dataVariable ValueAge (y) 51 (15–77)GenderMale 152 (33.5%)Female 302 (66.5%)Clinical presentationAbdominal pain 328 (72.2%)History of jaundice 315 (69.3%)Abdominal mass 281 (61.8%)History of fever 124 (27.3%)Nausea and vomiting 106 (23.3%)Asymptomatic 43 (9.4%)Duration of symptoms (mo) 16 (2–72)ComorbidityObesity* 32 (7.0%)Cardiovascular disease 28 (6.1%)Lung disease 11 (2.4%)Diabetes mellitus 10 (2.2%)Neurologic disease 7 (1.5%) Data are expressed as mean (range) or as number (percentage).*Body mass index 30 kg/m 2 . Table 2 Characteristics of cystsVariable ValueTotal number 695TypeMultivesicular 639 (87.7%)Clear 56 (12.3%)Size (cm) 19 (3–41)Number of cysts per patient1 296 (65.1%)2 103 (22.6%)3 34 (7.4%) 3 21 (4.6%)LocationRight lobe 458 (65.8%)Deep 314 (68.5%)Left lobe 203 (29.2%)Deep 72 (35.4%)Bilobar 34 (4.8%)Deep 31 (91.1%) Data are expressed as number (percentage) or as mean (range). 799S. Tagliacozzo et al. Surgical treatment of hepatic hydatidosis tients and multiple in 158 (one lobe in 85, both lobes in 73).A total of 417 cysts were classified as deep: 314 were sitedin the right lobe, 72 in the left lobe, and 31 in both lobes.Fifty-six cysts were clear (sterile cysts), and 639 were mul-tivesicular. Migration of daughter cysts into the commonbile duct was documented in 71 patients. A conservativeapproach was adopted in 214 patients (319 cysts), whereas240 patients (376 cysts) were treated with radical proce-dures (Table 3). No differences were found in age and gender between these groups.The local complication rates and hospital stays weresignificantly superior in patients who underwent conserva-tive procedures. No main differences in general complica-tions, with the exception of pleural effusion, were observedin patients treated with conservative or radical procedures.Operation time was significantly shorter for conservativetreatment, which, moreover, required fewer blood transfu-sions. Operative mortality was not significantly differentbetween conservative (6.5%) and radical (9.2%) procedures(Table 4). The number and type of surgical associated pro- cedures were similar in the 2 groups (Table 5). The presence of daughter cysts outside the cystic wall was observed onhistology in 67 of the 376 specimens (17.8%) from radicalprocedures. The incidence of recurrence was 30.4% in pa-tients who underwent conservative surgery and 1.2% inpatients treated with radical procedures (Table 4). No re- currence was observed in patients with clear cysts.Table 6shows a comparison between the 2 subgrouptime periods. Local complication and recurrence rates de-creased from 62% to 32% and from 15% to 8%, respec-tively, between the 2 time periods. Comments The desired goals for the treatment of LHD includecomplete elimination of the parasite and prevention of re-current disease with minimum mortality. 7 However, selec- Table 3 Surgical treatmentsTreatment Number of PatientsRadical treatment 240Hepatic resection 73Right extended 3Right 8Left 18Sectoriectomy 44CP 167Total 138Subtotal 29Conservative treatment 214Marsupialization 75External drainage 87Omentoplasty 52 Table 4 Outcomes of radical and conservative proceduresRadical Procedures Conservative ProceduresOutcome (n 240) (n 214) P Local operative complications 39 (16.2%) 171 (79.9%) .001Biliary leaks 11 (4.6%) 65 (30.4%) .001Residual cavity infection 11 (4.6%) 80 (37.4%) .001Residual cavity hematoma 17 (7.1%) 26 (12.1%) .065 (NS)General complications 18 (7.5%) 29 (13.6%) .034Pleural effusion 3 (1.2%) 17 (7.9%) .001Pulmonary embolism 2 (.8%) 1 (.5%) .630 (NS)Cardiac failure 5 (2.1%) 4 (1.9%) .870 (NS)Respiratory failure 2 (.8%) 6 (2.8%) .111 (NS)Renal Failure 0 1 (.5%) .289 (NS)Hepatic failure 6 (2.5%) 0 .019Hospital stay (d) 13.8 33.7 .001Recurrences 3 (1.2%) 48 (30.4%)* .001Operation time (min) 281 (240–420) 200 (110–261) .001Blood transfusion (L) 3.53 (2–6) r1.35 (0–4) .001Operative mortality 22 (9.2%) 14 (6.5%) .301 (NS) Data are expressed as number (percentage) or as mean (range).*Fifty-six of 214 patients treated with conservative procedures had solitary clear cysts. The rate of recurrence was calculated in 158 patients withmultivesicular cysts. Table 5 Associated proceduresAssociated ProcedureRadical TreatmentConservativeTreatmentCholecystectomy 45 38Cholecystectomy papillotomy 17 19Cholecystectomy T-tube 9 17Papillotomy 3 5Choledochotomy T-tube 2 4Endoscopic sphincterotomy 3 2 800 The American Journal of Surgery, Vol 201, No 6, June 2011
