ORIGINAL ARTICLE. Adenocarcinoma of the Third and Fourth Portions of the Duodenum

Transcription

1 Adenocarcinoma of the Third and Fourth Portions of the Duodenum Results of Surgical Treatment ORIGINAL ARTICLE Adriano Tocchi, MD; Gianluca Mazzoni, MD; Francesco Puma, MD; Michelangelo Miccini, MD; Diletta Cassini, MD; Elia Bettelli, MD; Sandro Tagliacozzo, MD Hypothesis: To verify the adequacy of duodenal segmentectomy after intestinal derotation in the treatment of primary adenocarcinoma of the third and fourth portions of the duodenum. Design: A retrospective review of the surgical management of patients who underwent derotation of the third and fourth portions of the duodenum was undertaken to determine long-term outcome. Setting: Departments of surgery in 3 university hospitals. Patients: Between January 1, 1980, and December 31, 2000, 47 patients with primary adenocarcinoma of the third and fourth portions of the duodenum were surgically treated at 3 different institutions. Main Outcome Measures: Details of primary surgery were abstracted from clinical records of the original hospital referral. Postoperative clinical course and long-term outcome were evaluated by a review of the hospital records and follow-up. Results: The results of a barium swallow test series was positive in 38 cases (80.8%) and esophagogastroduodenoscopy was primarily diagnostic in 30 patients (63.8%). In all cases duodenal segmentectomy was attempted. Twenty-two patients underwent palliative gastrojejunal bypass and in 9 patients pancreaticoduodenectomy was performed. In 16 cases duodenal segmentectomy was performed after intestinal derotation. Anastomoses were performed manually in all cases. Fifteen of the resected patients died of recurrent disease. A median (SD) diseasefree survival of 36 (23.6) months (range, 6-85 months) was observed. The median (SD) overall survival was 37.5 (23.9) months (range, months), the overall 5-year survival rate was 23% (11 patients), and the actuarial 5-year survival rate was 51% (24 patients). Conclusions: Duodenal segmentectomy associated with intestinal derotation was shown to be a straightforward, safe procedure for the treatment of the primary adenocarcinoma of the third and fourth portions of the duodenum. This surgical procedure should be preferred to pancreaticoduodenectomy because it is associated with negligible rates of morbidity and mortality, while allowing for satisfactory margin clearance and adequate lymphadenectomy. Arch Surg. 2003;138:80-85 From the First Department of Surgery of the University of Rome La Sapienza Medical School, Rome, Italy (Drs Tocchi, Mazzoni, Miccini, Cassini, Bettelli, and Tagliacozzo); Department of General and Thoracic Surgery of the University of Perugia Medical School, Perugia, Italy (Dr Puma); and the Department of Surgery of the University of Cagliari Medical School, Cagliari, Italy (Dr Tagliacozzo). ADENOCARCINOMA OF the duodenum is an exceedingly rare condition representing not more than 0.3% to 0.4% of all gastrointestinal tract cancers. 1-3 About 45% of all carcinomas of the duodenum arise in the third and fourth portions, a distribution pattern that parallels the relative lengths of each portion of this organ. 1,4,5 The most common symptom of cancer of the third and fourth portions of the duodenum is upper abdominal pain related to partial duodenal obstruction. 6 Anemia with hemoccult-positive stools is frequent, although frank upper gastrointestinal tract hemorrhage is unusual. 1,7,8 The diagnosis may be difficult to make because the symptoms are vague and similar to those found in many benign conditions. When identified, appropriate treatment of the third and fourth portions of the duodenum (3-4 DC) remains somewhat controversial. However, adequate early surgery is the only method by which longterm survival may be obtained. 9,10 In previous reports by other authors 1,6,7,11-13 pancreaticoduodenectomy (PD) and duodenal segmentectomy (DS) have been both used; PD as a curative procedure and DS mainly for palliative purposes. The goal of this study was to assess whether DS can also be considered a curative treatment for primary adenocarcinoma of the 3-4 DC. METHODS PATIENTS Between January 1, 1980, and December 31, 2000, 47 patients with primary adenocarci- 80

2 Figure 1. The greater omentum is detached from the transverse colon along the avascular plane. noma of the 3-4 DC were referred for treatment to the departments of surgery at 3 different Italian institutions (ie, University of Cagliari Medical School, Cagliari; University of Perugia Medical School, Perugia; and University of Rome La Sapienza Medical School, Rome). These institutions were recruited for this study because they have a similar surgical approach to this disease. The medical records of all patients were reviewed for demographic and clinical data. The type and number of preoperative studies were evaluated. The locations of the tumors were determined in all cases by preoperative endoscopy apart from contrast study reports when available. The 1997 Union Internationale Contre le Cancer classification was used to stage the tumors and histological features were established by surgical specimen or endoscopic biopsy specimen. SURGERY Background In all cases a bowel preparation was performed and prophylactic antibiotic therapy was begun intravenously 1 hour before the scheduled operations. Twenty-two patients underwent palliative gastrojejunal surgery, either on the anterior or the posterior gastric wall. Pancreaticoduodenectomy, consisting of a Whipple procedure, was performed in 9 patients because of extensive involvement of the pancreatic tissue. Duodenal segmentectomy was performed following intestinal derotation in the 16 remaining cases. In patients undergoing curative procedure (PD or DS) adequate resection margins were assessed at the time of surgery. Description of Intestinal Derotation Procedure The abdominal cavity is entered through a midline vertical incision. The greater omentum is detached from the colon along the avascular plane (Figure 1), the duodenojejunal flexure is identified, and the proximal jejunum retracted caudally while the peritoneum is incised along its left side. The large intestine from the cecum to the midpoint of the transverse colon is extensively mobilized to allow complete rotation of the ileal loops (Figure 2). The secondary root of the small-bowel mesentery is totally mobilized upward as far as the third portion of the duodenum (Figure 3). The duodenum is exposed by division of the peritoneum lateral to its Figure 2. The large intestine is extensively mobilized. second and third part and the ligament of Treitz is divided along the anterior cranial aspect of the 3-4 DC (Figure 4). The 3-4 DC and related mesentery can then be easily moved to the right of the superior mesenteric artery (Figure 5). Intestinal continuity is restored by an end-to-end duodenojejunal anstomosis. Drainage of the anastomotic site is always provided. EARLY OUTCOME The postoperative period was defined as the hospital stay, regardless of duration. Morbidity and mortality were assessed. Drainage was removed 24 hours after the passage of stools, except when there was evidence of clinical dehiscence. Between the 8th and 14th postoperative days a water-soluble contrast medium was administered orally to all patients to assess the integrity of the anastomosis; oral feeding was then started. FOLLOW-UP Follow-up was obtained from medical records, interviews with the patients, and endoscopic reports in all cases. All patients were assessed for recurrence monthly for 6 months and then once a year. Radiological follow-up consisted of gastrointestinal tract series, as well as computed tomographic and ultrasound scans when available. STATISTICAL ANALYSIS The results of parametric data are expressed as mean (SD). Nonparametric data are expressed as median (SD). Statistical significance was estimated using the Kruskal-Wallis, Spearman rank correlation, and stepwise logistic regression tests as appropriate. Statistical significance was set at P

3 Figure 4. The duodenum is exposed by division of the lateral peritoneum and the ligament of Treitz. Figure 3. The secondary root of the small bowel is totally mobilized. RESULTS PATIENTS Demographic and clinical characteristics of all patients are listed in Table 1. Gardner syndrome and duodenal polyposis were diagnosed in 1 and 3 patients, respectively. All patients underwent multiple preoperative studies. A barium swallow test series of the upper gastrointestinal tract was positive in 38 cases (80.8%). Esophagogastroduodenoscopy was primarily diagnostic in 30 (63.8%) of 47 patients and confirmatory of the results of a previous barium swallow test series in 14 (82.3%) of 17 patients. SURGERY The median surgical time was 90 (11.8) minutes (range, minutes) for the gastrojejunal procedure, 290 (32.6) minutes (range, minutes) for PD, and 130 (12.8) minutes (range, minutes) for DS. Anastomoses were performed manually in all cases. The median number of transfused blood units was 0.36 (0.73) (range, 0-3 units) (Table 2). EARLY OUTCOME Figure 5. The third and fourth portions of the duodenum are easily moved to the right of the superior mesenteric artery. Hospital stay (median length, 14 days) and the mortality (overall rate, 8.5%) were significantly lower in patients who underwent DS. One patient in this group died of cardiopulmonary failure after myocardial infarction. Mortality in patients who underwent PD was always related to the surgical procedure; 3 patients died of multiple organ failure after septic shock (2 cases) or hemorrhagic shock (1 case), all following the dehiscence of 82

4 Table 1. Demographic and Clinical Data Variable Value Age, mean (SD) [range], y 57.6 (7.5) [42-73] Sex, No. Male 26 Female 21 Length of disease, mean (SD) [range], mo 8 (3.1) [2-16] Symptoms Anemia 13 (27.7) Loss of weight 24 (51.1) Pain 25 (53.2) Obstruction 11 (23.4) Preoperative diagnostic technique, No. (%) of patients/total No. of patients Computed tomography 19/33 (57.6) Barium swallow test series 38/47 (80.9) Endoscopy First 30/47 (63.8) Second 14/17 (82.4) Location of adenocarcinoma Third portion 37 (78.7) Fourth portion 9 (19.1) Ligament of Treitz 1 (2.1) Tumor staging T2 14 (29.8) T3 19 (40.4) T4 14 (29.8) Lymph node staging N0 15 (31.9) N1 10 (21.3) NX 22 (46.8) Metastasis staging M0 37 (78.7) M1 10 (21.3) TNM staging I 10 (21.3) II 17 (36.2) III 10 (21.3) IV 10 (21.3) Grading Poor 16 (34.0) Fair 19 (40.4) Good 12 (25.5) *Percentages may not total 100% because patients may report more than 1 option or patients may not have provided this information. the pancreaticojejunal anastomoses (Table 2). All postoperative complications were managed conservatively. FOLLOW-UP Eight patients, 3 in the gastrojejunal and 5 in the DS group, were lost to follow-up. The median follow-up time was 24 (24.2) months (range, 4-85 months). LONG-TERM OUTCOME None of the patients who could not be resected lived longer than 11 (2.3) months (median, 9 months). In patients treated with a curative procedure, 7 showed local recurrence, 5 experienced only distant metastases, and 3 had local and distant recurrent disease. No statistically significant difference was found between patients who underwent DS and patients who underwent PD (4 vs 6, P=.13) The median disease-free survival was of 36 Table 2. Surgical Procedures and Early Outcome* Variable Value P Value Treatment, No. (%) of patients Gastrojejunal bypass 22 (46.8) Pancreaticoduodenectomy 9 (19.1)... Duodenal segmentectomy 16 (34.0) Surgical time, min Gastrojejunal bypass 90 (11.8) [80-120] Pancreaticoduodenectomy 290 (32.6) [ ]... Duodenal segmentectomy 130 (12.8) [ ] Blood transfusion, units Gastrojejunal bypass 0 Pancreaticoduodenectomy 2 (1) [0-3].001 Duodenal segmentectomy 0 (0.4) [0-1] Hospital stay, d Gastrojejunal bypass 13 (2) [11-19] Pancreaticoduodenectomy 18 (6.4) [12-34].007 Duodenal segmentectomy 14.5 (4.9) [11-29] Morbidity, No. (%) of sick patients/total No. of patients Gastrojejunal bypass 1/22 (4.5) Pancreaticoduodenectomy 3/9 (33.3).01 Duodenal segmentectomy 3/16 (18.7) Mortality, No. (%) of deaths/total No. of patients Gastrojejunal bypass 0 Pancreaticoduodenectomy 3/9 (33.3).01 Duodenal segmentectomy 1/16 (6.2) *Nonparametric data are expressed as median (SD) [range] unless otherwise indicated. Ellipses indicate not applicable. (23.6) months (range, 6-85 months). The median overall survival was 37.5 (23.9) months (range, 11-85), the overall 5-year survival rate was 23% (11 patients), and the actuarial 5-year survival rate was 51% (24 patients). Univariate analysis showed TNM stage, tumor stage, and nodal stage as factors influencing survival. No correlation was found between survival and the length of symptoms (P=.71) (Table 3). Multivariate analysis failed to confirm all the other correlations, showing only nodal stage (P=.01) as a factor influencing survival. COMMENT Causative factors of duodenal cancer have not been clearly identified. 4,5 Patients with familial polyposis and Gardner syndrome have been assumed to have a higher likelihood of developing duodenal cancer Patients who have duodenal polyps without a predisposing family history are also at an increased risk of developing a malignant lesion. 14 One of the patients in this study was affected by Gardner syndrome and duodenal adenomas had been previously diagnosed in 3 patients. Nonspecific presentation and symptoms that frequently mimic benign conditions (often treated by the patients themselves) are the main pitfalls for an early diagnosis of duodenal cancer. 8,10 Symptoms and signs observed in our series compare well with those reported in the literature (Table 1). All patients underwent multiple investigations and a variable number of falsenegative results was reported for each diagnostic procedure (Table 1). The disease was discovered most commonly by a barium swallow test series of the upper 83

5 Table 3. Univariate Analysis Variable Survival (SD), [range], mo P Value Sex Male 36.0 (22.5) [12-78] Female 48.0 (26.9) [11-85].73 Symptoms Anemia Yes 39.0 (26.0) [12-64] No 36.0 (24.4) [11-85].83 Loss of weight Yes 48.0 (25.5) [17-85] No 36.0 (22) [11-64].32 Pain Yes 36.0 (19.9) [11-64] No 56.0 (26.9) [12-85].17 Obstruction Yes 39.0 (21.6) [21-64] No 36.0 (24.9) [11-85].71 Location of adenocarcinoma Third portion 48.0 (23.6) [17-85].16 Fourth portion 19.0 (22.6) [11-64] Tumor staging T (21.2) [19-85] T (25.9) [11-78].12 T (18.4) [38-64] Lymph node staging N (15.4) [36-85] N (17.2) [11-69].002 TNM staging I 57.5 (15.7) [39-85] II 62.5 (17.5) [36-78].007 III 18.5 (17.2) [11-69] Grading Poor 36.0 (17.3) [17-61] Fair 48.0 (24.6) [12-69].7 Good 47.5 (28.3) [11-85] gastrointestinal tract; these images should always be carefully examined because lesions in this region seem to be easily overlooked. 8,9,17 Esophagogastroduodenoscopy had a 36% false-negative result rate in our series, in all cases because the study was terminated before reaching the site of the lesion. Our experience, similar to that of other authors, 8,13 suggests that whenever upper gastrointestinal endoscopy is performed, the entire duodenum should be examined, particularly when no abnormality is noted in the first and the second portions. Ultrasound and computed tomographic scan were performed in only a few cases, both providing little help in the diagnosis of the primary disease, yet revealing a high accuracy in detecting metastatic spread. Endoscopic ultrasound was never used, but this diagnostic tool may become useful for the diagnosis and the staging of these tumors. Surgery is judged the only way to attempt a curative treatment, and PD has been considered the procedure that satisfies the principles of an adequate curative cancer operation. 9-12,17-19 Duodenal segmentectomy has been suggested to represent a less morbid option. However, the use of this procedure is a debated topic. When attempting a DS, primary operative steps such as assessment of tumor resectability, search for an adequate cancer-free proximal margin, and attainment of a tension-free anastomosis have been described as heavily hindered by the anatomy of the duodenum, which is embedded in the retroperitoneum and anteriorly bound by the superior mesenteric artery. Furthermore, lymphadenectomy associated with this procedure is thought to be inadequate, given the fact that the duodenal mesentery seems short because it is fused in the retroperitoneum, and dissection must be carried out close to the bowel. 8,9,17 Duodenal segmentectomy has been considered a less radical procedure than PD while carrying as high of an operative morbidity and mortality and, thus, has been considered unsuitable for the elective treatment of this disease. 1,6,7,11-13,20 In the current experience a satisfactory proximal margin clearance together with an adequate excision of the mesentery related to the resected duodenum were easily achieved, thanks to the restored straight embryonal shape of the distal duodenum and duodenojejunal junction. A relevant number of lymph nodes, 16 much higher than that reported in literature, was available for examination. Intestinal continuity was restored in each case, on the right side of the superior mesenteric axis, by a straight, tensionfree, end-to-end duodenojejunal anastomosis. Operative mortality was low and complications following this procedure were infrequent. Tumor diameter, histological grading, and invasion into the periduodenal tissues have been strongly suggested as the main factors affecting the outcome of these patients. 8,21 Our findings do not support this statement. This lack of agreement probably reflects that most case series report the condition of a few patients; most ignored nodal involvement because lymphadenectomy had not been performed at all or was limited. The 5-year survival for resected patients in the current series was 51%. In a multivariate analysis, tumor staging and grading did not show any relation to long-term outcome, but nodal status significantly affected survival (P =.01). In support of these findings we cite the case of 2 patients undergoing duodenal and transverse colon resection because of neoplastic infiltration of the colonic mesentery. Despite tumor staging they are free from disease at 64 and 38 months after surgery; in both cases regional lymph nodes showed no abnormality at surgery. CONCLUSIONS Two main conclusions can be drawn from our results. First, PD should no longer be considered the appropriate elective therapy for primary adenocarcinoma of the third and fourth portions of the duodenum, because DS associated with intestinal derotation represents an approach that is radical, but is associated with negligible rates of morbidity and mortality. Second, when regional lymphadenectomy is routinely performed, the presence of nodal metastasis, a main unfavorable prognostic factor, is no different from that found with most gastrointestinal cancers. Accepted for publication September 7, In memoriam we thank Sandro Tagliacozzo, MD, for his research, from which our research was carried out and this manuscript was written. Corresponding author and reprints: Adriano Tocchi, MD, Via Bruno Bruni 94, Rome 00189, Italy ( adriano.tocchi@uniroma1.it). 84

6 REFERENCES 1. Spira IA, Ghazi A, Wolff WI. Primary adenocarcinoma of the duodenum [review]. Cancer. 1977;39: Barclay TH, Kent HP. The diagnosis of primary tumors of the duodenum. Gut. 1962;3: Iovine VM, Tsangaris N. Primary carcinoma of the duodenum. Am J Surg. 1961; 27: Lowenfels AB. Why are small-bowel tumours so rare? Lancet. 1973;1: Sabiani P, Le Treut YP, Maillet B, et al. Les adenocarcinomes de l angle duodenojejunal. J Chir (Paris). 1987;124: Joesting DR, Beart RW, Van Heerden JA, Weiland LH. Improving survival in adenocarcinoma of the duodenum. Am J Surg. 1981;141: Moss WM, McCart M, Juler G, Miller DR. Primary adenocarcinoma of the duodenum. Arch Surg. 1974;108: Lowell JA, Rossi RL, Munson JL, Braasch JW. Primary adenocarcinoma of third and fourth portions of duodenum. Arch Surg. 1992;127: Lillemoe K, Imbembo AL. Malignant neoplasm of the duodenum. Surg Gynecol Obstet. 1980;150: Heniford BT, Iannitti DA, Evans P, et al. Primary nonampullary/periampullary adenocarcinoma of the duodenum. Am Surg. 1998;64: Cortese AF, Cornell GN. Carcinoma of the duodenum. Cancer. 1972;29: Scott-Coombs DM, Williamson RC. Surgical treatment of primary duodenal carcinoma: a personal series. Br J Surg. 1994;81: Delacore R, Thomas JH, Forster J, et al. Improving resectability and survival in patients with primary duodenal carcinoma. Am J Surg. 1993;166: Schnur PL, David E, Brown PW Jr, et al. Adenocarcnoma of the duodenum and the Gardner syndrome. JAMA. 1973;223: Yao T, Iida M, Ohsato K, et al. Duodenal lesions in familial polyposis of the colon. Gastroenterology. 1977;73: Jones TR, Nance FC. Per-ampullary malignancy in Gardner s syndrome. Ann Surg. 1977;185: Fronticelli CM, Borghi F, Gattolin A, et al. Primary adenocarcinoma of the angle of Treitz: case report. Arch Surg. 1996;131: Alwark A, Anderson A, Lasson A. Primary carcinoma of the duodenum. Ann Surg. 1980;191: Kaminski N, Shaham D, Eliakim R. Primary tumours of the duodenum. Postgrad Med J. 1993;69: Pulcini G, Biasca F, Lanzi S, et al. Tumore maligno primitivo della IV porzione duodenale esteso all angolo di Treitz. Ann Ital Chir. 1999;70: Ryder NM, Ko CY, Hines OJ, et al. Primary duodenal adenocarcinoma: a 40-year experience. Arch Surg. 2000;135: IN OTHER AMA JOURNALS ARCHIVES OF NEUROLOGY The Role of Hypothermia in the Management of Severe Brain Injury: A Meta-analysis Odette A. Harris, MD, MPH; John M. Colford, Jr, MD, PhD; Matthew C. Good; Paul G. Matz, MD Context: Hypothermia is utilized in the management of severe traumatic brain injury despite the lack of unequivocal evidence supporting its use. Because of its widespread use, the effects of hypothermia are a concern. Objective: To determine the effectiveness of hypothermia in the management of severe brain injury. Data Sources: Two investigators working independently abstracted data in a blinded fashion from studies identified using multiple literature databases, including MEDLINE, Ovid, PubMed, the Cochrane Database of Systematic Reviews, EMBASE, and the abstract center for the American Association of Neurological Surgery and the Congress of Neurological Surgery, as well as the bibliographies of these articles. Additionally, experts in the field of hypothermia and neurotrauma provided additional references. Study Selection: Seven studies met predetermined inclusion criteria: (1) the study was a randomized clinical trial comparing the efficacy of hypothermia vs normothermia in patients with posttraumatic head injury, (2) only subjects aged 10 years or older were included in the study, and (3) relative risks (odds ratios [ORs], cumulative incidence, or incidence density measures) and 95% confidence intervals (CIs) or weighted mean differences and 95% CIs could be calculated from the data presented in the article. These criteria were applied in a blinded fashion by 2 independent investigators. Data Extraction: No single outcome variable was evaluated in all studies. The following outcome variables were assessed: intracranial pressure, Glasgow Outcome Scale score, pneumonia, cardiac arrhythmia, prothrombin time, and partial thromboplastin time. Either ORs or weighted mean differences (when the data provided did not permit calculation of an OR) comparing the effects of hypothermia vs normothermia were calculated from the data provided. Data Synthesis: The weighted mean difference (hypothermia normothermia) for intracranial pressure was 2.98 mm Hg (95% CI, 7.58 to 1.61; P=.2). The OR (hypothermia vs normothermia) for Glasgow Outcome Scale score was 0.61 (95% CI, ; P=.3). The OR for pneumonia was 2.05 (95% CI, ; P=.14). The OR for cardiac arrhythmia was 1.27 (95% CI, ; P=.7). The weighted mean difference for prothrombin time was 0.02 seconds (95% CI, 0.07 to 0.10; P=.7). The weighted mean difference for partial thromboplastin time was 2.22 seconds (95% CI, ; P.001). Conclusions: This meta-analysis of randomized controlled trials suggests that hypothermia is not beneficial in the management of severe head injury. However, because hypothermia continues to be used to treat these injuries, additional studies are justified and urgently needed. (2002;59: ) Corresponding author and reprints: Odette A. Harris, MD, MPH, Department of Neurosurgery, Stanford University Medical Center, 300 Pasteur Dr, Edwards Bldg, Second Floor, Stanford, CA ( odette@leland.stanford.edu). 85