* ”Al. Trestioreanu” Oncological Institute, Bucharest, Romania* ”Al. Trestioreanu” Oncological Institute, Bucharest, Romania* Caritas Hospital, Bucharest, Romania* CH Albertville, France* Department of Surgery, CH Pitesti, Romania
The suppurative pathology of the pancreatic chamber represents the final term for necrotic-hemmorhagic pancreatitis. Pancreatic, peripancreatic and distance suppurations are accompanied by a mortality rate of 8-30%, and surgical reinterventions represent a common situation, sometimes even a planned one. Both the morbidity related to the 3rd-6th septenaries of evolution and the mortality are the consequences of incomplete draining of the septic foci situated in the pancreatic chamber or at a distance. (1,2,3,4)
It is certain that, from the statistical point of view, the mortality peaks in severe acute pancreatitis are found in the 1st septenary and in the 3rd-4th septenary of evolution. The first peak corresponds to the deaths caused by multiple system organ failure (MSOF) and by the systemic inflammatory response syndrome (SIRS) in patients who are in the pre-suppurative, non-surgical stage. Approximately 50% of the necroses become superinfected by microbial lymphatic contamination or by translocation. Obviously, not all contaminations evolve towards a suppurative process. (5,6,7)
The decision for a surgical approach of the pancreatic suppuration is based on clinical, morphological and bacteriological criteria. From the clinical point of view, the surgical approach is recommended in the following situations: acute abdomen, obvious signs of sepsis, organ failure which either persists for more than 3 days or worsens under conservatory treatment. The most important criterion is the clinical evolution. The CT scan provides morphological arguments, by documenting the topography of the septic foci, and the percutaneous puncture provides certainty by means of the bacteriological examination. (8,9,10)
The pancreatic compartment
The pancreas is located deeply retroperitoneally, having anterior relations mediated by the omental bursa, a retrogastric adnexa of the peritoneal cavity. Posterior to the pancreas there are a series of vascular and lymphatic structures together with nervous plexuses of acknowledged importance. From an anatomical point of view, the pancreatic space is separated in the sagital plane into two compartments, right and left, the delimitation being made by the superior arteriovenous mesenteric axis. These compartments communicate through the aortic-mesenteric angle. On the right side, there are the pancreatic head and the retroportal band, and on the left side there are the pancreatic body and tail. In a straight line with the superior mesenteric axis there are 3 nervous plexuses, anastomosed into each other, forming the nervous hilum of the pancreas: the celiac plexus, the superior mesenteric one and the aortic-renal plexus, in the order of their cranial-caudal positioning.
The enzymatic aggression upon these plexuses is directly responsible for the “vasomotor storm” from the whole splanchnic area, including the renal part.
Finally, it should be noted that within the retropancreatic space there is a fan-shaped “curtain” made up of fibro-nervous tissue. The right band joins the posterior side of the head of the pancreas and of the uncinate process with the superior mesenteric pedicle. The left band, less expressed, extends from the neck of the pancreas to the left semilunar ganglion. Therefore, in the frontal plane, there is a “screen” separating the retropancreatic plane, more richly represented by vascular, lymphatic and nervous ganglionic structures, from the prepancreatic retroperitoneal tissue, more poorly represented as far as the above mentioned components are concerned. (11) (Fig. 1)
This leads to the idea according to which it is difficult to completely approach retropancreatic suppurations by way of the omental bursa. Likewise, the diffusion of the retro-pancreatic suppurations will take place mainly in pre and retrorenal directions, towards the colic angles, retrocolically, and less towards the pancreatic spaces which can be approached through the omental bursa, but separated from the retropancreatic plane by the fibro-nervous “screen”.
Draining of the Septic Collections
The directions of the retropancreatic diffusion in pancreatic suppurations cannot be systematized, as there is a wide variety of unpredictable septic spindle progression. This is why the most valuable source of information for the surgical stage is represented by imaging: CT or MRI. Consequently, we must insist on the obligation of oral and parenteral contrast media in order to be able to obtain the most objective documents possible. The purpose is to assess the limits of the septic necrosis as accurately as possible, as well as their relation to the major anatomical landmarks in the area. By drawing up an accurate map of the pancreatic region, using imaging reconstruction techniques as well, we will be able to select the best approach so as to access the pancreatic space in order to void and to drain the suppurative processes. (1,6,10,11) (4,5,7, 10,12)
The surgical objective is to perform the necrosectomy as completely as possible, and this depends on the way the pancreatic chamber is approached. Obvioulsy, insufficient access to the suppurative area leads to the incomplete draining, necrotic tissues being left behind, to the continuation and progression of the sepsis, irrespective of the method used: open, half-open or closed approach. Therefore, in order to perform an adequate necrosectomy, it is necessary to have wide access to the pancreatic space, both prepancreatically and retropancreatically, both to the right and to the left compartments, as against the superior mesenteric vascular axis.
In most publications dealing with the approach of the pancreatic area, the emphasis is placed on the access through the omental bursa. Surgeons have endeavoured to find the most efficient therapeutic solution. Necrosectomy is followed by the draining of the suppurative space: closed draining with lavage, open abdomen (laparotomy, planned interventions). All the tactical approaches aim to remove the necrotic tissues as completely as possible, so as to avoid leaving behind suppurative processes or developing suppurations after the initial operation. Obviously, the surgical therapeutic failure resides in an insufficient necrosectomy, followed by inefficient draining, both being tributary to an inadequate, limited approach of the retroperitoneal space. It is useless to perform planned interventions or laparotomy if the subsequent access follows the same path which has proved to be inefficient during the first draining intervention.
Access route to the pancreas
Classical monographies present two main access pathways: through the omental bursa and through the retroperitoneum. (13) As far as the first access path is concerned, there are four entrances into the omental bursa: the gastrocolic ligament, the gastrohepatic ligament, the gastrosplenic ligament and the Winslow hiatus. Usually, all these access paths, except for the last one, are used to access the omental bursa. But are we in the pancreatic space? Of course not. We are inside one of the adnexa of the peritoneal cavity, which does not offer the best conditions to perform a complete necrosetomy and to drain efficiently. The insufficiency of the approach using this route is more obvious at the level of the retropancreatic plane, more consistently represented by adipocellular tissue, lymphoid and vascular structures than the prepancreatic tissue. We have to underline the fact that the omental bursa provides difficult and insufficient access to the retropancreatic space. The lesser sac does not provide wide access to the whole pancreatic space, irrespective of the type of drainage used (open, closed). The retropancreatic area remains partially accessible. What results is an incomplete sequestrectomy and a retropancreatic drainage of doubtful quality.
The other access pathway is the direct retroperitoneal one: duodenopancreatic décollement (Kocher), spleno-pancreatic décollement, right and left coloparietal décollement, transmesocolic approach. (14,15) (Fig. 2)
Unfortunately, these means of access are more tentatively used by surgeons, who prefer the isolated access through the omental bursa. The Kocher path is appropriate for the retropancreatic collections on the right of the superior mesenteric vessels and it is used more frequently. The splenopancreatic décollement, exceptionally presented and illustrated in monographies, has few followers due to the risks of splenic bleeding, sanctioned by a splenectomy performed in full septic process, with all the postoperative morbidity attached to this procedure. This access path would provide direct access to the retropancreatic plane on the left side of the superior mesenteric vessels. Unfortunately, the presence of the spleen at the site of the initial access is discouraging. This is why Warshaw (16) and Shi (15) recommended and indicated the use of the transmesocolic path, passing through the thickness of the root of the transverse mesocolon .However, this path leads the surgeon more towards the anterior side of the pancreas than towards the retropancreatic area. The value of the coloparietal décollements at the level of the colic angles remains to be specified. They enable the retropancreatic access, but rather long way have to be covered to reach the area we are interested in – the retropancreatic plane. This last path is valuable when the suppuration has retrocolic expansions.
The Submesocolic Approach – Treitz’s Angle
Since there is a constant and precise anatomical landmark, Treitz’s duodenojejunal angle, the retroperitoneal space can be easily accessed, more specifically the retropancreatic space, and all that it represents on the left of the superior mesenteric axis. Two duodenojejunal fossettes are delimited at this level: the superior and the inferior one. (13) The sectioning of the posterior parietal peritoneum immediately to the left of Treitz’s angle allows the décollement and reclination to the right of this intestinal segment. Thus, we create an opening wide enough to allow retropancreatic access. To the left of this opening, as a landmark, lies the inferior mesenteric vein with its tributary, the left colic vein. In order to widen the access path to the left, the mesenteric vein can be ligated and sectioned along its passage without any inconvenience at all. This procedure is not compulsory and it is rarely useful. Thus, after dividing the peritoneum to the left and intercepting the suspensory ligament of Treitz’s angle, Treitz’s angle can easily be reclined to the right until the superior mesenteric axis. Thus, we obtain a wide path to the retropancreatic space in order to perform the necrosectomy and the drainageof the space situated posterior to the body and tail of the pancreas! (Fig. 3)
This approach is not routinely applicable. It can be recommended based on the information provided by the CT anatomy of Treitz’s angle, in relation with the pancreatic and peripancreatic collections. (17) So as to have a correct indication, Treitz’s angle must be identified on the CT images, as well as on the images reconstructed on the sagittal and frontal planes. (Fig. 4) The use of oral contrast media is compulsory, and so is the use of intravenous contrast media. Thus, the following can be visualized: Treitz’s angle, the superior mesenteric vessels, the inferior mesenteric vein, the relations between the collections and the major vessels, the renal vascular pedicles and the kidneys. An accurate map of the suppurative process in relation to the omental bursa and to the above mentioned retropancreatic plane will enable us to select the best access paths in order to perform a necrosectomy as complete as possible as well as the necessary drainage. (Fig. 5)
We associated the use of this access route when we had indications from the CT regarding the retropancreatic existence of a remarkable component of the suppurative process, in an unmediated relation to Treitz’s angle, situated in the left pararenal area. This approach was used in 12 cases of pancreatitis with suppurative peripancreatitis. We obtained a good quality draining of the septic collections, followed by multiple closed drainage. We never exteriorized a drain through the submesocolic breach, which we usually obturated with the greater epiploon. We did not record any reinterventions or deaths, but we will certainly be confronted with them during the evolution of future patients who will develop severe pancreatic suppurations.
At present, surgeons unanimously believe that the basic lesion, which predicts possible future surgery, is represented by the sterile pancreatic and peripancreatic necrosis. In the obsolete terminology, this lesion was referred to as the “pancreatic phlegmon” (Warshaw 1974), being actually a non-infected necrosis. According to Beger (1,2), 40% of the necroses become contaminated, and, therefore, are infected. Subsequently, suppuration appears, better said suppurative necrosis, leading to the development of sphacelus or pancrea-tic sequestra situated within the pancreatic or peripancreatic abscessed area. It is not our purpose to discuss the nomenclature used to refer to the suppurations of the pancreatic space. They all have a common denominator: hematomas and tissue necrosis located in purulent areas in various evolutional stages, hosted either by the pancreas or by the retroperitoneal plane, more or less close to the pancreas. Even nowadays, the delimitation of necroses and of suppurative processes remains a challenge for imaging exploration. The septic clinical dominant is decisive as against the other criteria when choosing the time for the surgical approach. Nevertheless, imaging remains the fundamental indicator in establishing the most direct and the widest access paths towards the pancreatic area. (Fig. 6, 7, 8)
Once the surgical indication has been decided, necrosectomy is the only solution in the treatment of these suppurations. Pancreatic resections and total pancreatectomy are very radical attitudes, whose enormous risks have no connection with the minor benefits. Percutaneous draining remains a solution for exceptional cases for the circumscribed, fluid and small suppurations, which do not represent the subject of the current discussion. The experience of some expert authors, found in prestigious publications, insists on the approach through the omental bursa after the division of the gastro-colic and duodenocolic ligament (2,3,4). This access enables the debridement of necrotic and suppurative foci, the intervention being completed by the insertion of drains into the omental bursa and by constant lavage of this peritoneal adnexa with physiological saline (3-6l/day). The fact that the surgeon limits himself only to drain the omental bursa, using tubes or the open-abdomen technique, does not offer any guarantee of success, since there are frequent cases of persistent retropancreatic suppurative remains.
The retropancreatic space, the site of the suppuration, cannot be voided and drained through the isolated access of the omental bursa. The sensible use of imaging data (CT, MRI), using precise anatomical landmarks in relation to the pancreatic space and the suppuration area can lead to a decrease of morbidity and a decrease of the rate of reinterventions. There are statistics indicating that mortality in pancreatic suppurations ranks below 10%,on patients managed only by omental bursa approach. (1, 2) It is possible to obtain such results, but they can probably be explained by a lower, less extended component of the suppurative process. We cannot include in the same category of postoperative assessment patients with focal necroses together with patients with extended necroses and patients with massive local or distance suppurations.
We state from the very beginning that our purpose is mainly to present a preliminary note, based on a limited number of cases in which we have used this retropancreatic access path. Our conclusions can be only fragile, but they do represent a subject to be reflected upon.
1. In the suppurations of the pancreatic area, the limited approach through the lesser sac may be insufficient. The debridement and draining of retropancreatic necrosis cannot be performed, irrespective of the method used for the draining of the omental bursa (open or closed). Sometimes it’s not enough.
2. The superior mesenteric vascular axis and the retro-pancreatic fibro-nervous bands compartment the retropancreatic space into two distinct areas: right and left, which have to be drained separately if they both host suppurative processes.
3. The retropancreatic approach to the right of the superior mesenteric vessels is easy if it is performed through duodeno-pancreatic decollement.
4. The left retropancreatic approach is both problematic and difficult using the classical access paths: spleno-pancreatic decollement, transmesocolic route.
5. It may be useful to resort to the mesocolic path suggested by us, in the area of Treitz’s angle because:
- it offers a short and direct way in the retropancreatic area of the body and tail of the pancreas;
- it allows the efficient debridement and draining of the retropancreatic space to the left of the superior mesenteric axis.
6. The indication for the additional use of this path is determined based on CT criteria (the ratio between the septic collection and Treitz’s angle) and on intraoperative criteria (exploratory puncture to the left of Treitz’s angle).
7. The compulsory CT examination with oral and venous contrast media in order to evaluate accurately the relation between the suppurative foci and Treitz’s angle and the superior mesenteric axis. The use of CT reconstruction is of great help.
8. In most cases it is necessary to perform the decollement and the reclining to the right of Treitz’s angle in order to widen the access path.
9. The submesocolic route can be an additional path to be used in order to perform a more complete retropancreatic necrosectomy, with the purpose of avoiding the septic remnents and, implicitly, reinterventions.
1. Beger, H.G. - Surgical management of necrotic pancreatitis. Surg. Clin. North Amer., 1989, 69:529.
2. Beger, H.G., Ran, B., Mayer, J., PRALLE, U. - Natural course of acute pancreatitis. World J. Surg., 1997, 21:130.
3. Popa, Fl., Girolteanu, H., Strimbu, V., Constantin, V. - Particularitãti evolutive şi optiuni terapeutice în pancreatitele supurative. Chirurgia (Bucur.), 1996, 45:183.
4. Popescu, I., Georgescu, S., Maher, Al. - Atlas – Supuratiile pancreatice şi peripancreatice, Ed. Med., 1990.
5. Balthazar, E.J. - Complications of acute pancreatitis: clinical and CT evaluation. Rad. Clin. North American 2002, 40:6.
6. Georgescu, I., Georgescu, E., Surlin, V. - Acute pancreatitis. În: “Tratat de chirurgie”, vol IX, partea a II-a, sub redactia lui Popescu I., Ed. Acad. Române 2009, pag. 405-442.
7. Heinrich, S., Schafer, M., roussom, u., claviem, p.D. - Evidenced based treatment of acute pancreatitis. Amer. Surg., 2006, 243:154.
8. Popa, Fl., Girolteanu, H., Bãlãlãu, C., Cuibac, A. - Propunere de nou scor prognostic de gravitate în pancreatitele acute. Chirurgia (Bucur.) 2003, 98:127.
9. Marincaş, M., Brãtucu, E., TOBA, M., CIRIMBEI, C., PAUN, L. - Surgical approach in acute necrotizing pancreatitis. Chirurgia (Bucur.) 2006, 101:237.
10. Mercadier, M. - Surgical treatment of acute pancreatitis: Tactics, techniques and results. World J. Surg., 1981, 5:393.
11. VanDamme, J.P. - Behavioral anatomy of the abdominal arteries. Surg. Clin. North Amer., 1993, 73:699.
12. Popescu, I. - Management of acute severe pancreatitis. Chirurgia (Bucur.) 2006, 101:225.
13. TESTUT, L. - Traité d’anatomie humaine, tome 4, Ed. G. Doin, 1931.
14. Branum, G., Salloway, J. - Pancreatic necrosis: results of necrosectomy, packing and ultimate closure over drains. Am. Surg., 1998, 227:870.
15. Shi, E.C.P., Yeo, B.W., Ham, J.M. - Pancreatic abscesses. Brit. J. Surg., 1984, 71:689.
16. Warshaw, A.L., Gonliang, J. – Improved survival in 45 patients with pancreatic abscess. Amer. Surg., 1935, 202:408.
17. Rotman, N., Bonnet, F., Larde, D., fagmiez, p.l. - Computed tomography in the evaluation of the late complication of acute pancreatitis. Am. J. Surg., 1986, 152:286.