Objective. Anastomotic fistulas in digestive surgery is a serious complication that can lead to patient’s death. The identification of paraclinical laboratory samples to allow early diagnosis of fistulas would allow an optimal patient’s management. Materials and method. We conducted a prospective study on 41 cancer patients operated for gastric tumors between 2016 and 2017 in the First Clinic of General Surgery and Oncology of the “Prof. Dr. Alexandru Trestioreanu” Institute of Oncology, Bucharest. All patients were subjected to gastric resections or gastrojejunal anastomoses. Therefore, digestive sutures were performed. Postoperative follow-up included the determination of C-reactive protein (CRP), monitoring of leukocyte counts in the peritoneal cavity (from a biological product harvested from the chest tube), associated with the number of leukocytes in the blood. Depending on the values of these biological samples and the clinical evolution, other paraclinical investigations were used to confirm the diagnosis in selected cases. Results. The data obtained showed that patients with digestive fistulas had a rapid increase and maintained high values of CRP and leukocytes in the peritoneal cavity, which was associated with an increase in leukocytes’ blood count. Changes occur approximately two days prior to clinical signs and their confirmation by imaging (ultrasound, CT scan). Conclusion. Regular and standardized monitoring of CRP, leukocytes in the peritoneal cavity and leukocytosis on days 1, 3 and 5 allows the early identification of patients at risk for fistula and the rapid selection of patients requiring additional explorations and/or surgery.
Obiectiv. Fistulele anastomotice în chirurgia digestivă reprezintă o complicaţie gravă, care poate duce la decesul pacientului. Identificarea unor probe paraclinice de laborator care să permită diagnosticul precoce al fistulelor ar ajuta la optimizarea managementului pacientului. Materiale şi metodă. Am efectuat un studiu prospectiv la 41 de pacienţi cu cancer, operaţi pentru tumori gastrice, între 2016 şi 2017, în Clinica I de Chirurgie Generală şi Oncologică a Institutului Oncologic „Prof. Dr. Alexandru Trestioreanu” din Bucureşti. Toţi pacienţii au fost supuşi unor rezecţii gastrice sau au fost efectuate anastomoze gastrojejunale. Aşadar, s-au practicat suturi digestive. Urmărirea postoperatorie a inclus determinarea proteinei C reactive (PCR) şi monitorizarea numărului de leucocite din cavitatea peritoneală (din produs biologic recoltat de pe tubul de dren), asociat cu numărul de leucocite din sânge. În funcţie de valorile acestor probe biologice şi de evoluţia clinică, în cazuri selecţionate, s-au asociat alte explorări paraclinice pentru confirmarea diagnosticului. Rezultate. Datele obţinute au arătat că la pacienţii cu fistule digestive are loc o creştere rapidă şi o menţinere la valori ridicate a PCR şi a leucocitelor din cavitatea peritoneală, la care s-a asociat şi o creştere a leucocitelor din sânge. Modificările apar cu aproximativ două zile înainte de apariţia semnelor clinice şi a confirmării lor prin probe imagistice (ecografie, computer-tomograf). Concluzii. Monitorizarea regulată şi standardizată în zilele 1, 3 şi 5 postoperatorii a valorii PCR, a numărului leucocitelor din cavitatea peritoneală şi a leucocitozei permite identificarea precoce a pacienţilor cu risc de fistulă şi selectarea rapidă a celor care necesită explorări suplimentare şi/sau intervenţie chirurgicală.
Anastomotic fistula in digestive surgery is a severe complication that can lead to patient’s death. The success of the anastomosis involves two factors, namely surgical technique and biological potential (local and general).
The current definition of colorectal anastomotic fistula is given by Müller in 1994 and includes: expression of faeces through the wound or vagina, fever over 38°C, radiographic or endoscopic signs of fistula, intraperitoneal abscess, or peritonitis in the presence of a fistula(1). All these signs appear late, when the septic intraperitoneal process is established. The surgeons’ concerns about the risk of anastomotic fistulas have led to studies on the relationship between a number of clinicopathological factors and the occurrence of anastomotic dehiscence.
From a scientific point of view, it is difficult to diagnose the appearance of fistulas early by clinical and paraclinical methods. The most important complication of the fistula is the infection manifested as intraperitoneal abscess, generalized peritonitis and/or abscess and phlegmon of the wound.
From a socioeconomic point of view, these serious septic complications increase the mortality (6-22%) and the general morbidity with 56%(2), as well as health care costs(3). Under these conditions, it is necessary to establish biological criteria that can postoperatively identify patients with a potential risk of fistulas. Taking therapeutic measures as quickly as possible could reduce morbidity and mortality.
Problems raised by the fistulas that occur in digestive surgery have always concerned the surgical world, and all studies conducted over the past 30 years have shown that there has been no significant reduction in the incidence of anastomotic fistulas. The elements related to the surgical technique regarding the compliance with the technical principles, such as well-vascularized endings, lack of tension in the anastomosis, non-seismic correct suture, and the elimination of the septic factor, are well known and applied. The implementation in recent years of new methods, such as laparoscopic and robotic surgery, that provide better visibility in a narrow space such as the pelvis, and the use of mechanical suture tools have failed to reduce the incidence of fistulas. Once established, it leads to the emergence of a septic state that forces energetic and costly therapeutic measures that cannot always save the patient.
The aim of the paper is to try to establish a fast diagnostic pathway for fistulas in the preclinical and pre-imagistic phases and, in this way, to suggest a therapeutic attitude as fast and efficient as possible, in order to prevent the risk of stercoral peritonitis, the dreaded complication of fistulas, which can lead to a 56% morbidity rate (Lindgren 2011)(4) and to a mortality rate of up to 22% (Rullier, 1998)(5). Finding biological markers that are not costly and can be included in a high predictability score could allow the application of early and effective measures(6).
Our concern has been extended to fistulas that occur after gastrointestinal surgery, meaning after a surgery where the septic factor is considerably more modestly expressed. Gastric and small intestine neoplasms are not excluded from postoperative fistulous complications (30%). The unpredictable behaviour of the oesophagus and the duodenal duct after the esojejunal anastomosis and the suture of the duodenum is well known. Mechanical or manual, the anastomotic suture and the clogging of the duodenal duct still pose a notable risk for fistulae.
Materials and method
The lots studied were as follows:
26 patients with gastric cancer (26%)
15 with other types of digestive tumours.
The objectives of the paper are:
1. Postoperative identification of the C-reactive protein in the blood.
2. Quantitative dosing of the number of leukocytes present in postoperative drainage fluid from the chest tube in patients with digestive anastomoses operated for gastric cancer.
3. Determining leukocytosis in the blood.
4. Establishing a predictive score to identify the anastomotic fistula before it establishes a septic state secondary to the occurrence of abscess or peritonitis.
C-reactive protein (CRP) and the study of leukocytes in peritoneal fluid are the early response of the peritoneum to germ aggression at the time of the production of the fistula.
We have prospectively studied the CRP and leukocyte values in the peritoneal drainage fluid associated with blood leucocytosis on days 1, 3 and 5, postoperatively. We compared these values in patients who had a favourable progression with data from those who had unfavourable development (fistula, death).
The risk of anastomotic fistula is increased between the 5th and the 8th postoperative day. Through this study, we aim to capture the appearance of the fistula before the emergence of clinical signs (peritonitis) and imaging (ultrasound, CT).
The RCP is measured using the Konelab600 device, and the leukocytes are quantitatively measured by the CYBOW 720 device. The devices are owned by the “Prof. Dr. Alexandru Trestioreanu” Institute of Oncology, Bucharest (IOB), and are state-of-the-art readers with simple and secure use, as well as features that ensure the performance and speed of the diagnosis.
CRP is the most studied marker of inflammation, being used in the assessment of the anastomotic fistula score (Dutch Leakage Score – Dulk et al., 2009)(7). Other authors practice microdialysis on drain tubes to determine oxygenation from anastomosis (Matthiessen, 2007)(8). The most recent approach on the subject, the use of the fistula score and its usefulness in assessing the possibility of peritonitis (the dreadful complication of the fistula), is presented by Freek Daams, in World J. Gastroenterology, in 2013, and again in 2014 (Colorectal anastomotic leakage: Aspects of prevention, detection and treatment, and in Prediction and diagnosis of colorectal anastomotic leakage: A systematic review of literature)(2,6).
We have registered 26 patients with gastric cancer in which the stomach-level localization was as follows: 11 (42.30%) in the gastric body (nine with adenocarcinoma and two with GIST); 13 with tumors at the antrum (50%; one patient at the histopathological examination on paraffin was diagnosed with benign lesion – ulcer); two (7.69%) experienced esogastric lesions. Postoperative fistula was reported by a total of 8 patients (30.76%), and the localizations were: seven in the duodenal duct and one in the esojejunal anastomosis. Three (37.5%) of these patients had surgical reinterventions (drainage, suture), and five fistulas were treated conservatively. The number of deaths recorded immediately after surgery was three (11.53%).
The fistula was clinically and imagistically objectified (native CT or contrast substance – collection or fistulous trajectory), on average, on day 6.62 postoperatively (approximately on day 7). The patients with fistula stayed, on average, 17.25 days in the intensive care unit (ICU) and accounted for 22.37 postoperative days of hospitalization.
Table 1 shows the average values of CRP in all patients and, separately, the mean values of patients with and without fistula, on harvest days 1, 3 and 5. On day 7, only 13 samples were processed, of which six were fistulas. On this day, patients either evolved favourably, or developed fistulas, detectable clinically and by imaging (CT – fistulography, collection), and the therapeutic behaviour was established: intervention or conservative treatment.
The number of leukocytes in the peritoneal fluid harvested from the drain tube for days 1, 3 and 5 is presented in Table 1. The mean values for the entire batch and, separately, their average number are shown for fistulous and non-fistulous patients. On the seventh day, only 11 patients were registered, because the rest of 15 were either having their drain tube removed, or the diagnosis of fistula or collection (localized peritonitis) no longer justified the continuation of the monitoring.
The mean value of the number of white blood cells (leukocytosis) on days 1, 3 and 5 is presented in Table 1, both for fistulous patients and postoperative fistula patients. On the fifth day, blood sampling was performed for 25 patients. On the seventh day, only 16 patients were harvested, while others with a favourable evolution were not harvested.
Fifteen patients with pancreatic cancer, major biliary tract (CBP) or other types of tumors that invaded the small intestine were monitored for entero-enteral, collagen and gastrojejunal anastomoses. The location was as follows: two tumors (13.33%) in the CBP, seven tumors in the pancreas (46.66%) and six with ileum invasion (40%). Three patients presented postoperative fistulas (20%), and the locations were: two in the ileum (they occurred in the same patient) and one after a cerebral duodenopancreatectomy (DPC) for a pancreatic head neoplasm. All three patients (100%) were surgically reinitiated (drainage, suture). One immediate postoperative death (6.66%) was registered in the patient with CPD. The fistula was clinically and imagistically detected (CT), on average, on day 6.33 postoperatively (two on the seventh day, and one on the fifth day). The patients with fistula had an average of 29 days in ICU and 33.33 days of postoperative hospitalization.
Table 2 shows the average values for CRP, the number of leukocytes in the peritoneal fluid collected from the chest tube and, respectively, the number of white blood cells (leukocytosis) in all patients and, separately, the mean values of patients with and without fistula on harvest days 1, 3 and 5. On the seventh day, only 12 patients were monitored, including two with fistula. On this day, patients either evolved favourably, or had a fistula.
Thus, we have registered 41 patients with tumors (26 with gastric tumors, and 15 with other types of digestive tumors). Postoperative fistula was present in 11 patients (26.8%) and surgical reinterventions were performed on six of them (14.6% of all patients, 54.54% of those with fistulas), the remainder of the fistulae being treated conservatively. There were four immediate postoperative deaths (9.75%). The fistula was clinically and imagistically objectified (native CT or contrast substance – collection or fistulous trajectory), on average, on day 6.47 postoperatively (approximately on the seventh day). The patients with fistula stayed, on average, 23.12 days in ICU, and accounted for 27.68 postoperative days of hospitalization.
The issue of early detection of anastomotic fistula was debated in 2008 by Cornel Iancu and collaborators, who performed a retrospective study on a group of 993 patients operated for colorectal neoplasm and analyzed several clinical and biological factors that could indicate the risk of preoperative fistula. The result was that a protein level below 5.5 g/dl and a haemoglobin level below 9.4 g/dl were predictive factors for the risk of fistulas(9).
The risk of fistulas in colorectal surgery remains high, without being significantly influenced by purchase of surgical technique. Dekker et al. in 2010 published a preoperative fistula prediction score on which to judge if anastomosis or colostomy is indicated for the patient(13). Applying this score does not completely eliminate the risk of fistulas, and therefore, in a study by Dulk et al. (2009), several clinical factors (fever, heart rate, ileus, gastric retention) and biological factors (CRP level, leukocyte count, renal function) were analyzed in an attempt to create a postoperative prediction score for fistulas. High-score patients underwent a more sustained clinical and paraclinical observation. This score reduced the fistula diagnosis delay from 4 to 1.5 days(7).
This score includes a series of clinical signs, some of which have a high subjective burden that in our opinion may negatively affect the results. Starting from the principle of evidence-based medicine, we consider that the exclusion of clinical criteria from the score calculation would be beneficial. As Freek Daams shows in 2013, the C-reactive protein is a good marker for postoperative infectious complications, but is not specific for fistula(2). In other studies, a series of biochemical markers from peritoneal fluid have been analyzed, which have partially proven their value in the detection of anastomotic fistula. The measurement of cytokines in peritoneal drainage fluid is promising and could justify the routine placement of juxta-anastomotic drainage(11,12), while peritoneal microdialysis(8) could detect local oxygenation changes in the vicinity of anastomosis, but did not show statistically significant predictive value in patients without clinical symptomatology(13,14). Due to the limitations of current knowledge, the authors of the aforementioned articles assert that further studies are needed to confirm the clinical value of this score, by introducing the parameters obtained in the current clinical treatment protocols and allowing early reoperation of these patients prior to the occurrence of sepsis(15,16).
Although routine peritoneal drainage tube placement is not recommended in literature, we believe that it is advantageous to mount perforated polyurethane silicone tubes in the Douglas space, as it allows us to establish a faster diagnosis of fistulas (leukocyte count from the peritoneal fluid, purulent or enteral liquid exteriorization, bile, fistulography or conservative treatment)(17).
In some patients, we found that from the third to the fifth postoperative day, when besides the enteral alimentation (per bone), the parenteral nutrition also begins, there is an increase in leukocytes from the peritoneal fluid (on average, 125.25/µl, reaching values of 500/µl of leukocytes for some patients), returning to low values or even zero. Clinically, patients did not show signs of peritoneal irritation, were afebrile, and did not show digestive fistulas or abdominal collection on images (CT, ultrasound). Blood leukocytes remained within normal limits, and CRP was decreasing, indicating the patient’s favourable progression.
High value of CRP after the third postoperative day, with low leukocyte counts in peritoneal drainage, but with blood leukocytosis, indicated other complications: bronchopneumonia, wound infection etc., but not digestive fistula.
Compared to other studies, we monitored the same parameters in anastomoses performed for the surgical treatment of gastric and esogastric tumors. CRP on day 1 in fistulous patients showed lower mean values than for colon interventions (132.82 versus 169.66 mg/dl), but the values increased progressively on days 3, 5 and 7 (188.43 – 224.23 – 254.24 mg/dl).
The patients with favourable progression showed mean values on the first day of 109.80 mg/dl and then a downward progression of RCP over the same days (99.75 – 60.15 – 36.21 mg/dl).
The number of leukocytes in the peritoneal fluid was higher on the first day for fistulous patients (340.62/µl), and after a slight decrease on the third day (218.75 µl), there was a significant increase on the fifth day (307.14/µl), when the fistula was likely to occur, although clinically and paraclinically, this was recorded on postoperative day 6.62, on average. After the development of the fistula, on the seventh day, there was a decrease (150/µl) of the fistulas. For patients without fistulae, the trend was a progressive decline (180.27 – 145.93 – 35).
Blood leukocytes followed the same trend as peritoneal leukocytes, until the fifth day (14,571.25 – 10,515.00 – 12,785/µl), but with a significant increase (16,243.75/µl) on the seventh day, when the fistula occurred clinically and on imaging.
In small intestine interventions, CRP increased progressively on days 3 and 5, and decreased slightly on the seventh day of fistula expulsion (219.06 – 221.70 – 159.40 mg/dL) and progressively declined for those without fistula (from 108.61 on day 1 to 59.24 mg/dl on day 7). Leukocytes from the peritoneal fluid followed the same trend (200 – 287.50 – 375/µl). The fistula was clinically and imagistically detected, on average, on day 6.65 postoperatively. Blood leukocytes were within normal limits on the first day, had a downward trend in fistula-free patients, and for those with fistula increased on the third day, decreased on the fifth day and increased on the seventh day (9,466.66 – 9,773.33 – 8,211.33 – 12,936.66/µl).
On all types of surgical procedures performed for gastric tumors and consecutive anastomoses, we found that in patients without fistulas, the correlated evolution of monitored parameters (CRP, Ld, Ls) was progressively downward, and in those with fistulas, an alarm signal appeared on the third day, when CRP and Ld remained high as mean values, and blood leukocytes were maintained within normal range.
We consider that, in addition to the literature, the association to CRP of the leukocytes count in the peritoneal fluid could trigger an alarm signal on the third postoperative day regarding the possibility of anastomotic fistula and the establishment of active measures: monitoring (ultrasound, CT with contrast substance), broad spectrum antibiotics or combination of antibiotics, oral suppression and parenteral feeding, along with hydro-electrolytic rebalancing, CT drainage, or early (laparoscopic, classic) reintervention. In fact, we could say that the fistula can be diagnosed 2-3 days before it is clinically manifested and imagistically highlighted, and the leukocyte count in peritoneal drainage is considered to be a more accurate criterion than measuring blood leukocytes.
The originality of the study consists in analyzing two biological blood markers (CRP, leukocytosis) and a peritoneal biologic marker (the number of leukocytes) in an attempt to establish a prediction score for anastomotic fistulas in patients with digestive cancer. The marker in the peritoneal fluid exteriorized on juxta-anastomotic drainage tubes will increase the specificity of the results from the singular analysis of the reactive C protein, which is known to be a nonspecific inflammatory marker. At the same time, eliminating subjective clinical factors will increase the objectivity of the results. We mention that digestive sutures were made both manually and mechanically, and were not subjected to a separate analysis.
Regular and standardized monitoring of CRP, leukocytes in the peritoneal cavity and leukocytosis on days 1, 3, and 5 allows the early identification of patients at risk for fistula and the rapid selection of patients requiring additional exploration and/or surgery.
Although the problem of colic fistula was a frequent reason for analysis due to the serious septic complications it generates, with high mortality and morbidity, so far the studies conducted have failed to bring considerable improvements in the evolution of these patients. Starting from the known severity of septic intraperitoneal complications, we can say that we should do everything in our power to prevent them, and the purpose of this paper is to do so.
We believe that the article is also of social and economic significance, since the early identification of fistulas in anastomoses in digestive cancer surgery can avoid the occurrence of intraperitoneal septic complications and reduces the costs of patient care in the hospital and the costs associated with the temporary incapacity to work. Another element of economic impact is that testing our proposed markers does not require costly equipment and supplies, and can also be performed in hospitals with minimal facilities.
Conflict of interests: The authors declare no conflict of interests.
Müller FP, Schmidt WU, Hesterberg R, Roher HD. Treatment of anastomotic leakage after colon and rectum resection. Br. J. Surg. 1994; v.81, suppl., p. 33.
Daams F, Luyer M, Lange JF. Colorectal anastomotic leakage: Aspects of prevention, detectionandtreatment. World J Gastroenterol. 2013 Apr 21; 19(15): 2293–2297.
Straatman J, Cuesta MA, de Lange-de Klerk ES, van der Peet DL. Hospital cost-analysis of complications after major abdominal surgery. Dig Surg. 2015; 32(2):150-6.
Lindgren R, Hallböök O, Rutegård J, Sjödahl R, Matthiessen P. What is the risk for a permanent stoma after low anterior resection of the rectum for cancer? A six-year follow-up of a multicenter trial. Dis Colon Rectum. 2011; 54: 41–47.
Rullier E, Laurent C, Garrelon JL, Michel P, Saric J, Parneix M. Risk factors for anastomotic leakage after resection of rectal cancer. Br J Surg.1998; 85:355–358.
Daams F, Wu Z, Lahaye MJ, Jeekel J, Lange JF. Prediction and diagnosis of colorectal anastomotic leakage: A systematic review of literature. World J Gastrointest Surg. 2014 February 27; 6(2): 14-26.
den Dulk M, Noter SL, Hendriks ER, Brouwers MA, van derVlies CH, Oostenbroek RJ, Menon AG, Steup WH, van de Velde CJ. Improved diagnosis and treatment of anastomotic leakage after colorectalsurgery. Eur J Surg Oncol. 2009; 35:420–426.
Matthiessen P, Strand I, Jansson K, Törnquist C, Andersson M, Rutegård J, Norgren L. Is early detection of anastomotic leakage possible by intraperitoneal microdialysis and intraperitoneal cytokines after anterior resection of the rectum for cancer? Dis Colon Rectum. 2007; 50:1918–1927.
Iancu C, Mocan LC, Todea-Iancu D, et al. Host-related predictive factors for anastomotic leakage following large bowel resections for colorectal cancer. J Gastrointestin Liver Dis. 2008 Sept; vol. 17, 299-303.
Dekker JW, Liefers GJ, de Mol van Otterloo JC, Putter H, Tollenaar RA. Predicting the risk of anastomotic leakage in left-sided colorectal surgery using a colon leakage score. J Surg Res. 2011; 166:e27–e34.
Yamamoto T, Umegae S, Matsumoto K, Saniabadi AR. Peritoneal cytokines as early markers of peritonitis following surgery for colorectal carcinoma: a prospective study. Cytokine. 2011; 53:239–242.
Tsujinaka S, Kawamura YJ, Konishi F, Maeda T, Mizokami K. Pelvic drainage for anterior resection revisited: use of drains in anastomotic leaks. ANZ J Surg. 2008; 78:461–465.
Karliczek A, Benaron DA, Baas PC, Zeebregts CJ, Wiggers T, van Dam GM. Intraoperative assessment of microperfusion with visible lightspectroscopy for prediction of anastomotic leakage in colorectal anastomoses. Colorectal Dis. 2010; 12:1018–1025.
Sheridan WG, Lowndes RH, Young HL. Tissue oxygen tension as a predictor of colonic anastomotic healing. Dis Colon Rectum. 1987; 30:867–871.
Hirano Y, Omura K, Tatsuzawa Y, Shimizu J, Kawaura Y, Watanabe G. Tissue oxygen saturation during colorectal surgery measured by near-infraredspectroscopy: pilot study to predict anastomotic complications. World J Surg. 2006; 30:457–461.
Ellebaek Pedersen M, Qvist N, Bisgaard C, Kelly U, Bernhard A, Møller Pedersen S. Peritoneal microdialysis. Early diagnosis of anastomotic leakage after low anterior resection for rectosigmoid cancer. Scand J Surg. 2009; 98:148–154.
Thomas HA. Radiologic investigation and treatment of gastrointestinal fistulas. Surgical Clinics of North America. 1996 Oct 1; Volume 76, Issue 5, pages 1081-1094.
Lagoutte N, Facy O, Ravoire A, Chalumeau C, Jonval L, Rat P, Ortega-Deballon P. C-reactive protein and procalcitonin for the early detection of anastomotic leakage after elective colorectal surgery: pilot study in 100 patients. J Visc Surg. 2012; 149:e345-e349.
Garcia-Granero A, Frasson M, Flor-Lorente B, Blanco F, Puga R,Carratalá A, Garcia-Granero E. Procalcitoninand C-reactive protein asearlypredictors of anastomotic leak in colorectal surgery: a prospective observational study. Dis Colon Rectum. 2013; 56:475-483.
Giaccaglia V, Salvi PF, Cunsolo GV, Sparagna A, Antonelli MS, Nigri, G, Balducci G, Ziparo V. Procalcitonin, as an early biomarker of colorectal anastomotic leak, facilitates enhanced recovery after surgery. J Crit Care. 2014; 29:528-532.