Pericardial effusion as a rare complication of a perforated appendicitis

IntroductionWhilst pericardial effusion is a known complication of abdominal pathology, it is rarely reported following ruptured appendicitis and even more rarely requires drainage in that situation. This work has been reported in line with the SCARE criteria (Agha et al., 2016).Presentation of caseWe report a 14-year-old male who developed extensive right hepatorenal and right paracolic abscesses, bilateral pleural effusions and a large pericardial effusion following laparoscopic appendicectomy. Due to the size of the effusion, thoracoscopic pericardotomy was required.DiscussionPericardial effusion is a very rare complication of advanced appendicitis despite a demonstrable connection between the retroperitoneum and the mediastinum. Only two cases were reported in our literature search. There is no consensus as to whether percutaneous drainage or pericardiotomy is the treatment of choice.ConclusionThe report is presented as a reminder of a rare complication of a common general surgical condition.     

Left side thoracoscopically assisted gastroplasty

Laparoscopic antireflux surgery is the procedure of choice for gastroesophageal reflux disease (GERD). However, many clinicians have reservations about its application in patients with complicated GERD, notably those with esophageal shortening. In this report, we present our experience with the laparoscopic management of the shortened esophagus. A total of 235 patients with primary GERD underwent laparoscopic antireflux procedures, 38 of whom were suspected preoperatively to have a shortened esophagus. Of the 235 patients, 8 (3.4%) needed a left thoracoscopically assisted gastroplasty in addition to laparoscopic Toupet repair (n= 4) or Nissen fundoplication (n= 4). Complications included pleural effusion (n= 1), pneumothorax (n= 2), and minor atelectasis (n= 1). The average hospital stay was 3 days. Results were satisfactory in 7 of 8 patients, with a mean follow-up of 20.2 months (range, 9–34 months). The surgical management of the shortened esophagus is difficult. However, the role of minimally invasive techniques is justified. Early results are appealing, with less morbidity, satisfactory control of GERD related symptoms, and a shortened hospital stay. Received: 3 August 1999/Accepted: 10 November 1999/Online publication: 17 April 2000     

Thoracoscopic decortication in infants and children

Background: The treatment of pediatric empyemas remains controversial. While thoracentesis and tube thoracostomy appear adequate for relatively benign organisms, virulent bacteria cause thick fibrinous pleural peels entrapping the lung. Open thoracotomies have been effectively used for decortication but are painful. Methods: We report the use of minimally invasive thoracoscopic decortication in 12 patients (mean age 5 years). All failed conventional management with persistent fever, increasing oxygen requirement, recurrent effusion, and pleural consolidation; 5- and 10-mm trocars were used and complete decortication was accomplished. Results: Ten of 12 patients were afebrile by 72 h and discharged 4–12 days after surgery. Eleven of 12 patients had clear chest x-rays by 1 month. Conclusion: Thoracoscopic decortication is a safe and effective means of treating pediatric empyemas. Received: 25 March 1996/Accepted: 28 May 1996     

Thoracoscopic splanchnicectomy for pain control in patients with unresectable carcinoma of the pancreas

Background: Intractable pain is the most distressing symptom in patients suffering from unresectable pancreatic carcinoma. Palliative interventions are justified to relieve the clinical symptoms with as little interference as possible in the quality of life. The purpose of this study was to examine the efficacy and safety of thoracoscopic splanchnicectomy for pain control in patients with unresectable carcinoma of the pancreas. Methods: Between May 1995 and April 1998, 24 patients (14 men and 10 women) with a mean age of 65 years (range, 30–85) suffering from intractable pain due to unresectable carcinoma of the pancreas underwent 35 thoracoscopic splanchnicectomies. All patients were opiate-dependent and unable to perform normal daily activities. Subjective evaluation of pain was measured before and after the procedure by a visual analogue score. The following parameters were also evaluated: procedure-related morbidity and mortality, operative time, and length of hospital stay. Results: All procedures were completed thoracoscopically, and no intraoperative complications occurred. The mean operative time was 58 ± 22 min for unilateral left splanchnicectomy and 93.5 ± 15.6 min for bilateral splanchnicectomies. The median value of preoperative pain intensity reported by patients on a visual analogue score was 8.5 (range, 8–10). Postoperatively, pain was totally relieved in all patients, as measured by reduced analgesic use. However, four patients experienced intercostal pain after bilateral procedures, even though their abdominal pain had disappeared. Complete pain relief until death was achieved in 20 patients (84%). Morbidity consisted of persistent pleural effusion in one patient and residual pneumothorax in another. The mean hospital stay was 3 days (range, 2–5). Conclusions: We found thoracoscopic splanchnicectomy to be a safe and effective procedure of treating malignant intractable pancreatic pain. It eliminates the need for progressive doses of analgesics, with their side effects, and allows recovery of daily activity. The efficacy of this procedure is of major importance since life expectancy in these patients is very short. Received: 23 December 1999/Accepted: 6 January 2000/Online publication: 12 July 2000     

Limited axillary thoracotomy vs video-assisted thoracoscopic surgery for spontaneous pneumothorax

Background: Video-assisted thoracoscopic surgery (VATS) has been reported to have a higher pneumothorax recurrence rate than limited axillary thoracotomy (LAT). We investigated the cause of pneumothorax recurrence after VATS by comparing surgical results for VATS and LAT. Methods: Ninety-five patients with spontaneous pneumothorax underwent resection of pulmonary bullae by VATS (n= 51) or LAT (n= 44). Operating duration, bleeding during surgery, number of resected bullae, duration of postoperative chest tube drainage, postoperative hospital stay, postoperative complication, and pneumothorax recurrence were analyzed to compare VATS and LAT in a retrospective study. Results: The duration of surgery, postoperative chest tube drainage, and postoperative hospital stay was significantly shorter in VATS than in LAT cases (p < 0.0005 and p < 0.005). Bleeding during surgery was significantly less in VATS than in LAT cases (p < 0.005). Numbers of resected bullae were significantly lower in VATS (2.7 ± 2.1) than in LAT cases (3.9 ± 2.7) (p < 0.05). Postoperative pneumothorax recurrence was more frequent in VATS (13.7%) than in LAT cases (6.8%), but there was no significant difference. Conclusions: VATS has many advantages over LAT in treating spontaneous pneumothorax, although the pneumothorax recurrence rate in VATS cases was double that in LAT cases. The lower number of resected bullae in VATS than in LAT cases suggests that overlooking bullae in operation could be responsible for the high recurrence rate in VATS cases. We recommend additional pleurodesis in VATS for spontaneous pneumothorax to prevent postoperative pneumothorax recurrence. Received: 13 August 1997/Accepted: 15 December 1997     

Complications in thoracoscopic spinal surgery

Background: The literature contains few reports on negative outcomes after thoracoscopic spinal surgery. Methods: From November 1995 to February 1998, 90 patients underwent minimally invasive spinal surgery by thoracoscopic assistance as treatment for their anterior spinal lesions. The diagnoses included 41 spinal metastases, 13 cases of scoliosis, 12 burst fractures, 10 cases of tuberculous spondylitis, 8 cases of pyogenic spondylitis, 2 thoracic disc herniations, 2 cases of ankylosing spondylitis with discitis, 1 osteoporotic compression fracture, and 1 case of thoracolumbar kyphosis. The procedures included biopsy only (3 patients); thoracic discectomy (3 patients); multilevel anterior releases, discectomy, and fusion (14 patients); corpectomy for decompression (6 patients); corpectomy and interbody fusion (32) patients; and internal instrumentation (28 patients). Results: A total of 30 complications were noted in 22 patients (24.4%). Two fatal complications occurred, resulting from massive blood transfusion in one case and postoperative pneumonia in another. Other nonfatal complications included four cases of transient intercostal neuralgia, three superficial wound infections, three cases of pharyngeal pain, two cases of lung atelectasis, two cases of residual pneumothorax, two cases of subcutaneous emphysema, one inadvertent pericardial penetration due to adhesion, one chylothorax that resolved after conservative management, one vertebral screw malposition, and one graft dislodgement that needed late revision surgery. Three patients required ventilatory support for longer than 72 hours. Five patients with spinal metastases had an estimated intraoperative blood loss of more than 2,000 ml. No injury to the internal organs or spinal cord was observed. There were four conversions to open procedures due to two cases of severe pleural adhesions and two poorly tolerated one-lung ventilation. At the latest follow-up, nine patients had died as a result of cancer dissemination. Conclusions: (a) Well-selected patients and attention to details are essential to optimizing surgical results. (b) A refined technique for less invasive tumor surgery has been developed. (c) Surgeons had better experience with the standard anterior spinal approach and showed no hesitation in converting to an open procedure when necessary. A procedure failure does not mean a treatment failure. Received: 14 May 1998/Accepted: 25 August 1998     

Perforating appendicitis

Background: This pilot study was performed to reassess the widespread postulate that laparoscopic surgery is contraindicated in cases of perforating appendicitis. Methods: A total of 75 children (appendiceal perforation: n= 67; perityphlic abscesses and peritonitis: n= 8) were analyzed in a prospective nonrandomized trial. Ten of them were treated by laparoscopic appendectomy. Results: In the laparoscopy group, both pain and hospitalization were less time-consuming (i.e., by 50% and 19%, respectively). Antibiotics use was down from 2.6 over 6 days to 2.2. over 5.5 days, while the duration of surgery was up by 52%. Wound healing disturbances occurred in 10% (n= 1) and postoperative fever in 50% (n= 5) of patients, compared to 14% (n= 9) and 15% (n= 10) in the group treated by open surgery. All severe complications requiring reintervention (10%; massive subcutaneous abscess, n= 3; retrocolic abscess, n= 2; adhesion-related ileus, n= 3; appendicular stump, n= 1) were associated with open surgery. Conclusions: There was not a single major complication in the laparoscopy group. These unexpected results are in contrast to previous reports and have prompted us to initiate a prospective randomized trial. Received: 27 August 1998/Accepted: 20 January 1999     

Multicentric experience of the Belgian Group for Endoscopic Surgery (BGES) with endoscopic adrenalectomy

Background: Adrenalectomy is not a frequent operation. Therefore the newly developed laparoscopic approach is sporadically performed by surgeons dealing with endocrine disorders. Methods: Some 54 videoendoscopic adrenalectomies performed on 52 patients by five surgical teams between October 1993 and December 1996 were prospectively evaluated. Results: Indications for endoscopic adrenalectomy were pheochromocytoma (n= 17), primary hyperaldosteronism (n= 15), Cushing's adenoma or disease (n= 7), nonsecreting adenoma (n= 7), single metastasis from adenocarcinoma (n= 2), adenoma with dehydroepiandrostenedione (DHEAS) hypersecretion (n= 3), and ACTH-secreting metastases from a thymoma (n= 1). Of the 54 adrenalectomies performed, 31 were of the left gland, 19 of the right and two bilateral. Laparoscopic adrenalectomy was successful in 50 patients (96%). Median tumor size was 4 cm (range 1.5–12), median operation duration was 80 min (range 59–360), and median postoperative stay was 4 days (range 2–13). One patient required blood transfusion. Conclusions: Endoscopic adrenalectomy can safely be performed—even sporadically—by surgeons well versed in adrenalectomy techniques for endocrine disorders and trained in endoscopic surgery. Received: 25 March 1997/Accepted: 16 May 1997     

Does laparoscopy increase bacteremia and endotoxemia in a peritonitis model?

Background: Laparoscopy is increasingly used in patients with intraabdominal bacterial infection although pneumoperitoneum may increase bacteremia by elevated intraabdominal pressure. Methods: The influence of laparotomy and laparoscopy on bacteremia, endotoxemia, and postoperative abscess formation was investigated in a rat model. Rats received intraperitoneally a standardized fecal inoculum and underwent laparotomy (n= 20), or laparoscopy (n= 20), or no further manipulation in the control group (n= 20). Results: Bacteremia and endotoxemia were higher after laparotomy and laparoscopy compared to the control group (p= 0.01) 1 h after intervention. One hour after intervention, aerobic and anaerobic bacterial species were detected in the laparotomy group while only anaerobic bacteria were found in the other two groups. Although bacteremia and endotoxemia did not differ among the three groups after 1 week, the mean number of intraperitoneal abscesses was significantly higher (p < 0.05) after laparotomy (n= 10) compared with laparoscopy (n= 6) and control group (n= 5). Conclusion: Laparoscopy does not increase bacteremia and intraperitoneal abscess formation compared to laparotomy in an animal model of peritonitis. Received: 28 May 1996/Accepted: 25 July 1996     

心脏外科术后心包积液的危险因素与治疗

目的 探讨心脏外科手术后心包积液的危险因素和治疗方法.方法 回顾分析22 462例患者临床资料,定义心包积液诊断标准.观察心包积液患者与无积液患者的临床表现,对症治疗,分析危险因素.结果 509例(2.3％)患者有心包积液262例有临床特殊症状,其中51例有心包压塞的临床表现.有、无心包积液的患者年龄、性别、冠心病史等因素差异无统计学意义(P＞0.05);而大体重、瓣膜病、主动脉阻断和体外循环时间差异有统计学意义(P＜0.05).结论 心包积液的危险因素有大体重,术前心功能Ⅲ、Ⅳ级,瓣膜病,先天性心脏病,大血管疾病,体外循环和主动脉阻断时间延长.超声引导下的心包积液穿刺引流是安全有效的.     

A prospective randomized controlled trial of sclerotherapy vs ligation in the prophylactic treatment of high-risk esophageal varices

Background: Endoscopic ligation (EVL) and endoscopic sclerotherapy (EIS) are both effective in the treatment of bleeding esophageal varices, but the efficacy of the two techniques in the prophylaxis of first variceal bleeding has not been investigated. The aim of this study was to investigate the frequency of first variceal bleeding, the recurrence of varices, and survival after treatment with the two techniques, as compared to a nontreated control group. Methods: A total of 157 patients with liver cirrhosis and advanced esophageal varices with no previous history of upper gastrointestinal bleeding were randomly assigned to either an EIS group (n= 55), an EVL group (n= 52), or a nontreated control group (n= 50). After the eradication of esophageal varices in the EIS and in EVL groups and in all control patients, the endoscopic examination was performed at 3-month intervals. Results: There were no significant differences between EIS and EVL in the eradication rate of esophageal varices (85% in the EIS group versus 81% in the EVL group). The mean number of sessions required to obtain eradication was lower in the EVL group than in the EIS group (4.8 ± 1.8 versus 6.2 ± 2.0; p= 0.0003), but the recurrence of esophageal varices was higher in the EVL group (31% versus 11%; p= 0.01). Total mortality was significantly lower in the EIS patients than in the controls (20% versus 38%; p= 0.04). It was also lower, but not significantly, in the EVL patients than in the controls (23% versus 38%; p= 0.10). A significant decrease in variceal bleeding was observed both in sclerotherapy cases (20%) and controls (54%; p= 0.0005) and in ligation cases and controls (29%; p= 0.01). No significant difference in bleeding episodes was observed between the sclerotherapy and ligation cases (p= 0.29). No serious complications were observed either in the EIS or EVL groups. Conclusions: EIS and EVL are similarly effective in the prevention of first variceal bleeding. The choice between EIS and EVL depends on the skill of the endoscopic unit. For highly experienced surgeons facing no complications, sclerotherapy seems to be preferable; for all others, it is technically easier to perform ligation. Received: 29 June 1998/Accepted: 18 September 1998     

Minimally invasive surgical staging for esophageal cancer

Background: The incidence of esophageal adenocarcinoma is increasing in the United States, and the 5-year survival rate is dismal. Preliminary data suggest that conventional imaging is inaccurate in staging esophageal cancer and could limit accurate assessment of new treatments. The objective of this study was to compare minimally invasive surgical staging (MIS) with conventional imaging for staging esophageal cancer. Methods: Patients with potentially resectable esophageal cancer were eligible. Staging by conventional methods used computed tomography (CT) scan of the chest and abdomen, and endoscopic ultrasound (EUS), whereas MIS used laparoscopy and videothoracoscopy. Conventional staging results were compared to those from MIS. Results: In 53 patients, the following stages were assigned by CT scan and EUS: carcinoma in situ (CIS; n= 1), I (n= 1), II (n= 23), III (n= 20), IV (n= 8). In 17 patients (32.1%), MIS demonstrated inaccuracies in the conventional imaging, reassigning a lower stage in 10 patients and a more advanced stage in 7 patients. Conclusions: In 32.1% of patients with esophageal cancer, MIS changed the stage originally assigned by CT scan and EUS. Therefore, MIS should be applied to evaluate the accuracy of new noninvasive imaging methods and to assess new therapies for esophageal cancer. Received: 5 April 1999/Accepted: 15 March 2000/Online publication: 12 July 2000     

Laparoscopic creation of stomas

Background: Some indications for laparoscopic bowel surgery are still controversial. However, the use of laparoscopic techniques for the treatment of benign disorders is less often challenged. Moreover, the morbidity of nonresectional procedures is less than that encountered with resectional cases. Therefore, stoma creation seems ideally suited to laparoscopy. The aim of our study was to assess the outcome of laparoscopic stoma creation. Methods: All patients who underwent laparoscopic intestinal diversion were evaluated; parameters included age, gender, indication for the procedure, history of previous surgery, operative time, length of hospitalization, recovery of bowel function, and postoperative complications. Results: Between March 1993 and January 1996, 32 patients of a mean age of 42.2 (range 19–72) years (14 males, 18 females) underwent elective laparoscopic fecal diversion (25 loop ileostomy, four loop colostomy, three end colostomy). Indications for fecal diversion were fecal incontinence (n= 11), Crohn's disease (n= 6), unresectable rectal cancer (n= 4), pouch vaginal fistula (n= 3), rectovaginal fistula (n= 2), colonic inertia (n= 2), radiation proctitis (n= 1), anal stenosis (n= 1), Kaposi's sarcoma of the rectum (n= 1), and tuberculous fistula (n= 1). Conversion was required in five patients (15.6%) due to the presence of adhesions (three), enterotomy (one), or colotomy (one). All of these five patients had undergone previous abdominal surgery and were operated on early in our experience. Major postoperative complications occurred in two patients (6%) and in both cases consisted of stoma outlet obstruction after construction of a loop ileostomy. One of the two patients had undergone prior surgery. This patient required reoperation, at which time a rotation of the terminal ileum at the stoma site was found. The other patient had a narrow fascial opening which was successfully managed with 2 weeks of self-intubation of the stoma. The mean operative time was 76 (range 30–210) min; mean length of hospitalization was 6.2 (range 2–13) days; stoma function started after a mean of 3.1 (range 1–6) days. Patients with previous abdominal surgery had a longer mean operative time (14/32; 117 min) compared to patients who had no previous surgery (18/32; 55 min) (p < 0.0002). These longer operative times and hospital stay were attributable to extensive enterolysis, which was required in some cases. Conclusion: In conclusion, laparoscopic creation of intestinal stomas is safe, feasible, and effective. Although the length of the procedure is longer in patients who have had prior surgery, previous surgery is not a contraindication, and even in these cases, a laparotomy can be avoided in the majority of patients. Lastly, care must be taken to ensure adequate fascial opening and correct limb orientation. Received: 25 March 1996/Accepted: 21 May 1996     

Thoracoscopic Pericardial Window Creation and Thoracic Duct Ligation in Neonates

Objective:

We describe 2 newborn infants with persistent pericardial effusion treated with thoracoscopic pericardial window and thoracic duct ligation.

Methods:

Patient 1 was a premature female newborn who presented with severe cardiac anomalies, including dextrocardia. She was treated with pulmonary artery banding and pacemaker placement for complete cardiac block. Postoperatively, she developed pericarditis with persistent symptomatic pericardial effusion. She did not improve despite pericardial drain placement. She was treated with a thoracoscopic pericardial window. Patient 2 was a newborn male who presented with cardiac tamponade secondary to congenital chylopericardium. He did not respond to pericardial drain placement or medical management with fasting, total parenteral nutrition, and octreotide. He was treated with thoracoscopic pericardial window and thoracic duct ligation.

Results:

Patient 1 improved rapidly. The pericardial effusion disappeared. The chest tube was removed 5 days following surgery. She died 6 weeks later of a cardiac arrhythmia secondary to pacemaker failure. The pericardial effusion had resolved. Patient 2 responded to the pericardial window and thoracic duct ligation. He was discharged 10 days following the procedure.

Conclusions:

Thoracoscopy provides an excellent approach to the pericardium. Pericardial windows and biopsy can be safely performed with this approach. The thoracic duct can be easily identified and ligated even in small babies. Recovery can be fast with minimal postoperative discomfort. Cosmetic results are excellent and length of hospitalization is minimized.     

Abdominal, laparoscopic, and vaginal hysterectomy with bilateral salpingo-oophorectomy

Background: The abdominal route is the traditional method of performing hysterectomy with bilateral salpingo-oophorectomy. In a feasibility study, we compared a nonconventional (vaginal) route for bilateral salpingo-oophorectomy at the time of vaginal hysterectomy (VH + BSO) to similar forms of hysterectomy performed abdominally or with operative laparoscopy. Methods: Fifty-nine patients were subject to either total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH + BSO; n= 19), or laparoscopic-assisted vaginal hysterectomy and bilateral salpingo-oophorectomy (LAVH + BSO; n= 19) or VH + BSO (n= 21). Results: VH + BSO resulted in a shorter operating time compared to LAVH + BSO (p < 0.001), shorter hospital stay compared with TAH + BSO (p= 0.001), and quicker long-term recovery compared to the other two operations. Conclusion: This preliminary but significant study shows that the vaginal route for salpingo-oophorectomy at the time of vaginal hysterectomy is superior to other methods of hysterectomy. A randomized trial is needed to confirm these initial findings. Received: 20 February 1998/Accepted: 18 August 1998     

Eleven years' experience with pericardial-peritoneal window in the management of malignant and benign pericardial effusions

Background: Before 1983 we routinely used subxiphoid drainage for the management of pericardial effusions. Pericardial-pleural window through a left anterior thoracotomy was used in selected patients. Due to frustration over the rate of recurrent pericardial effusions with subxiphoid drainage alone and concern over the higher morbidity with thoracotomy, the creation of a 3-cm pericardial-peritoneal window in the fused portion of the pericardium and diaphragm overlying the left lobe of the liver was added to subxiphoid drainage in 1983. Methods: This study is a retrospective chart review of the 33 patients undergoing pericardial-peritoneal window from 1983 through 1993. Eighteen patients had malignancies, mainly lung and breast, and 15 had benign pericardial effusions. Results: The procedure was well tolerated, with a 30-day mortality of 9%; however, no deaths were directly related to the pericardial effusion or the procedure. No patient developed peritoneal carcinomatosis or diaphragmatic hernia. One patient developed recurrent pericardial effusion during follow-up, and two required pericardiectomy for constrictive disease. Among those with malignancies, patients with breast cancer had the longest survival after pericardial-peritoneal window. Conclusions: Pericardial-peritoneal window is a simple, safe, and effective procedure and applicable to most patients with malignant and noninfectious benign pericardial effusion, including those with tamponade. Presented at the 47th Annual Cancer Symposium of The Society of Surgical Oncology, Houston, Texas, March 17–20, 1994.     

Impact of hematological diagnosis on early and late outcome after laparoscopic splenectomyrid=""

Background: Laparoscopic splenectomy (LS) is now regarded as the treatment of choice for autoimmune thrombopenia (ITP). However, there have been few reports describing the application of LS to other splenic diseases, such as malignant entities and conditions associated with splenomegaly. Hematological diseases have specific clinical features that can influence immediate outcome after LS. Although the long-term effects of LS are unknown, a risk of splenosis has been suggested. Therefore, we designed a study to analyze the impact of primary hematological disease on immediate and late outcome in a prospective series of LS patients. Methods: We performed a prospective analysis of 111 LS done between February 1993 and March 1999. The patients were classified by hematological indications into the following four groups: (a) group 1, low platelet count. This group was further subdivided into group 1A, idiopathic thrombocytopenic purpura (ITP) (n= 48) and group 1B, HIV-related ITP (n= 8); (b) group 2, anemia. This group was further subdivided into group 2A, autoimmune hemolytic anemia (n= 8), and group 2B, spherocytosis (n= 11); (c) group 3, malignancy (n= 28); and (d) group 4, others (n= 8). Immediate outcomes were recorded prospectively. Hematological status and late complications were reviewed after a mean follow-up of 24 ± 18 months. Results: There were no significant differences between the groups in terms of conversion, transfusion requirements, and morbidity, although transfusion and morbidity were slightly higher in group 3. However, hospital stay was significantly longer in groups 3 and 4 than in groups 1 and 2. Long-term follow-up showed satisfactory hematological results in ≥75% of patients (group 1A, 82%; group 1B, 88%; group 2A, 88%; group 2B, 100%; group 3, 75%; group 4, 88%). Overall, late morbidity was 8.3% and mortality was 6.2%, mainly due to deaths in group 4 (six of 22 patients). Conclusion: LS is a safe and reproducible procedure for most hematological indications, with a similar immediate outcome for benign diseases and a long-term hematological response comparable to the standard results that have been observed in open series. Received: 1 April 1999/Accepted: 22 November 1999/Online publication: 8 May 2000     

Pancreatic carcinoma

Background: The aim of this study was to establish the implications of a normal pancreatogram in patients with pancreatic cancer. Methods: We reviewed all ERCP done at our institution for the period 1983–92 and studied 62 of 727 patients who had a diagnosis of pancreatic (n= 416) or biliary (n= 311) cancer but a normal pancreatic duct. Results: Thirty of the 62 patients had pancreatic cancer. In 15 cases, the ERCP diagnosis was incorrect, and in 19 cases, Santorini's duct was not visualized. Other imaging revealed a pancreatic head mass in 25 patients (2.5–>7 cm). Only three patients had resectable tumors; another eight underwent laparotomy. Five had bypass surgery, 10 required biliary stenting, and nine had no treatment. Four patients died in hospital, and eight were lost to follow-up. In the remaining 18 patients, median survival was 7 months (range, 1–30 months). Conclusion: A normal pancreatogram does not exclude the diagnosis of pancreatic cancer, nor does it confer a better prognosis. Received: 4 September 1997/Accepted: 24 October 1997     

The thoracoscopic approach for internal mammary nodes in breast cancer

Background: This pilot study was conducted to estimate the use of the thoracoscopic surgery as a new approach for the internal mammary nodes (IMN) in breast cancer. Patients and methods: For this study, 21 women with breast cancer who underwent the approach for nodes IMN were enrolled. All the women had suspicious IMN metastasis and no distant metastasis. Results: Thoracoscopic IMN dissection was performed safely for 20 of the women, with an average operative time of 44 min. One woman was excluded from the procedure because of pleural adhesion. The patients were restricted from walking for 1.3 days because of chest drainage, but no patients had severe complication or chest wall deformity after the operation. Six patients had positive IMN outcomes. After surgery, 10 of the 20 patients had a lower tumor node metastases (TNM) staging. Two patients who tested positive for IMN and three who tested negative experienced a relapse, but none had pleural dissemination in a median follow-up period of 24 months. Conclusion: Thoracoscopic surgery may be useful in managing patients with IMN. Received: 2 June 1999/Accepted: 17 January 2000/Online publication: 29 August 2000     

First clinical experience with an endoscopic retroperitoneal approach for anterior fusion of lumbar spine fractures from levels T12 to L5

Background: Recent experience indicates that unstable spine fractures should be stabilized dorsoventrally. To avoid the high morbidity associated with the common anterior approach—i.e., thoraco-phreno-lumbotomy—we developed a technique that allows the anterior fusion of lumbar spine fractures using an endoscopic retroperitoneal (lumboscopic) approach. Methods: Lumboscopic anterior fusion was performed a few days after the initial dorsal stabilization. The retroperitoneal space was accessed endoscopically via a suprailic incision and enlarged using a ballon spacer and CO₂ insufflation. The peritoneum and the kidney were gently pushed ventrally. Mobilization of the psoas muscle dorsally then allowed exposure of the fractured spine bodies. Via two additional trocars placed opposite the fractured level, the damaged disc and bone were removed, and anterior spondylodesis was performed with an iliac crest bone block and a titanium plate. Results: The technique was applied successfully in 12 patients with fractures of L1 (n= 6), L2 (n= 4), L3 (n= 1), and L4 (n= 1) as a mono- or bisegmental fusion, requiring instrumentation from T12 to L5. No major complications (including neurological problems) were encountered. Blood loss was minimal. None of the patients required conversion to open surgery. Patients were mobilized early, starting regularly at the second postoperative day. Conclusions: Lumboscopic instrumentation of the lumbar spine is a safe, minimally invasive method for the treatment of spine fractures. The patients benefit from reduced pain, low morbidity, and excellent cosmetic results. Received: 30 April 1999/Accepted: 2 July 1999