| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2008;7:1079-1083. doi:10.1510/icvts.2008.176271 © 2008 European Association of Cardio-Thoracic Surgery
Is prompt exploratory laparotomy the best attitude for mesenteric ischemia after cardiac surgery?
a Department of General Surgery, Hotel Dieu de France Hospital, Faculty of Medicine, Saint-Joseph University, Alfred Naccache Street, Beirut, Lebanon Received 22 January 2008; received in revised form 19 August 2008; accepted 22 August 2008
Corresponding author. Tel.: +961 1 615300; fax: +961 1 615295.
Mesenteric ischemia following cardiac surgery is a life-threatening complication. Early identification of patients may help optimizing management and improving outcome. Between January 2000 and July 2007, surgical exploration was realized when mesenteric ischemia was suspected after coronary-artery bypass grafts (CABG). Patients were divided in two groups according to diagnosis confirmation upon laparotomy. Peri-operative predictors of complication and death were analyzed. Of 1634 consecutive patients, 13 (0.8%) developed acute abdomen with suspicion of mesenteric ischemia. Seven (0.4%) underwent resection for ischemic lesions (group 1), of whom two were during a second look laparotomy. The other six patients had normal bowel (group 2). Both groups were comparable according to preoperative status, clinical signs, biological and radiological findings. Delays to laparotomy were 13.7±19.0 and 51.4±29.0 h in group 1 and 2, respectively (P=0.02). Mortality rates were 46.1% (6/13) overall, 42.8% for group 1 and 50% for group 2. All deaths occurred within the first nine postoperative days. Mesenteric ischemia following CABG is a fatal complication in almost half the cases. Diagnostic tools and timely laparotomy still need to be optimized. Low threshold-based strategy for prompt surgical intervention is efficient for both diagnosis and treatment.
Key Words: Mesenteric ischemia; Exploratory laparotomy; Cardiopulmonary bypass
Gastrointestinal (GI) complications following cardiac surgery include mesenteric ischemia, pancreatitis, gastroduodenal ulceration, acute cholecystitis, hollow viscus perforation and hepatic failure. Although rarely encountered, their occurrences significantly increase morbidity and mortality. Among these, acute mesenteric ischemia is the most dreaded, owing to extremely high mortality rate reaching 68–100% [1, 2]. The need for early diagnosis and treatment based on a high index of suspicion [3, 4] is a commonly approved concept. However, the current diagnostic tools lack specificity and sensitivity, resulting in delayed diagnosis and treatment. A bedside color-flow duplex of celiac and superior mesenteric axis can be unreliable in obese and uncooperative patients with inflated bowel. Arteriogram is often impractical in an unstable post-cardiac surgery patient. Several retrospective series [5, 6] attempted to define risk stratification models based on preoperative and intraoperative variables in order to build management algorithms. Risk factors included older age, intraoperative hypoperfusion, emergency operation, longer cardiopulmonary bypass (CPB) times, need for high-dose vasopressors, intra-aortic balloon pumps, and valve operations. Despite these schemes, management of mesenteric ischemia following cardiac surgery remains controversial [7]. This paper aims: (1) to assess incidence and mortality of mesenteric ischemia in a series of consecutive patients undergoing coronary-artery bypass grafting (CABG) under CPB, (2) to compare peri-operative factors according to laparotomy findings, and (3) to present the institutional management strategy in this setting.
2.1. Patients' selection This is an observational retrospective study conducted between 1 January, 2000 and 31 July, 2007, including all patients who underwent isolated or valvular-combined CABG under CPB. The institutional cardiac surgery morbidity and mortality database was screened for patients who developed a clinical syndrome suggestive of mesenteric ischemia within the first postoperative month. Suspicion was based on suggestive clinical presentation and/or positive biological and/or radiological findings (Table 1). The local hospital Ethical Committee approved database review. Patients presenting with non-ischemic pathology or transient gastrointestinal symptoms or signs were excluded.
2.2. Patients' management A systematic immediate surgical laparotomy was performed in all patients hypothetically having mesenteric ischemia. Group 1 consisted of patients with intra-operatively confirmed diagnosis according to bowel and visceral discoloration and necrosis. Necrotic bowel was resected and proximal stoma was performed. Group 2 represented patients with normal bowel and viscera on exploratory laparotomy. Patients with borderline bowels were reassessed in a 48-h second look operation and were included in group 1 if ischemia was present. When embolic cause was likely, superior mesenteric artery embolectomy was considered. In case of arterial mesenteric thrombosis, thrombectomy or arterial bypass was planned. These situations were not encountered in this series. Forty-eight variables were studied (Table 1) and grouped in six main categories: demographic data and peri-operative parameters, clinical presentation, biological findings, radiological findings, laparotomy findings and outcome. Data were collected from patients' charts. Primary outcomes were the presence of mesenteric ischemia and death. The independent factors' distribution was compared between both groups. Continuous variables are presented as mean±S.D. and proportions are shown for nominal and dichotomous variables. The Student t-test and the Mann–Whitney U-tests were used as appropriate (non-normal distribution) to compare continuous data, and  2 statistic was used to compare qualitative data, corrected by the Fisher exact test when appropriate. All tests were two-sided and a P<0.05 was considered statistically significant. All computations were done using SPSS® v13 (Chicago, IL) statistical software.
3.1. Demographic and perioperative factors Out of 1634 patients, thirteen (0.8%) underwent open laparotomy for suspicion of mesenteric ischemia. Group 1 represented 0.43% of the total population and consisted of six males and one female, while group 2 included the remaining 0.37% and consisted of five males and one female. Demographic and perioperative factors are summarized in Table 2. Mean age was higher in group 2 (71.8±6.9 vs. 66.8±7.8 years, P=0.34) without reaching statistical significance.
Smoking, hypertension and hyperlipidemia were more frequent in group 1 with no statistical significance. Peripheral vascular disease and previous myocardial infarction were more frequently encountered in group 1 (P=0.29 and P=0.59, respectively). Left ventricular ejection fraction was similar in both groups (P=0.24). The mean CPB and aortic cross-clamping times and minimal postoperative cardiac output were comparable. Only one patient from group 2 required an intra-aortic balloon pump and hemodialysis postoperatively. The number of inotropes and vasopressors used were similar in both groups. De novo atrial fibrillation occurred in four patients in each group (P=0.99). Treatment with full-dose unfractionated heparin was promptly initiated after the diagnosis was established. Table 3 compares the two groups for clinical signs suggestive of mesenteric ischemia. All patients with confirmed mesenteric ischemia presented with abdominal pain and guarding while these signs were inconstant in group 2 (P=0.19 and P=0.07, respectively). Prolonged ileus was noted in five patients in group 1 and in three in group 2 (P=0.59). Cyanosis was not found in group 2 but only in four patients in group 1 (P=0.07). Abdominal distension occurred in four patients of each group (P=0.99), and signs of digestive hemorrhage were noted only once in a group 2 patient.
3.3. Biological findings Relevant biological parameters are summarized in Table 4. Elevations of creatinine and LDH were constant in group 1. Metabolic acidosis and leucocytosis were not different between the two groups. In fact, all other biological parameters (transaminases, lipase, CK and PT) were evenly distributed between the two groups. Only one patient in group 1 showed all biological disturbances concomitantly.
3.4. Radiological findings accuracy Table 5 shows accuracy of radiological findings according to final diagnosis. Plain X-ray of the abdomen was realized in one patient in each group; it showed dilated bowel with no further diagnostic information. Ultrasound of the abdomen was performed in five cases in group 1. It was falsely normal in two, non-conclusive in two, and reported gallbladder disease in the last one. In group 2, the abdominal ultrasound was performed in three cases, considered as normal in one and inconclusive in two. Mesenteric vessels Doppler was only performed in group 2 patients. It showed normal mesenteric flow in one patient and was inconclusive in the other. An abdominal CT-scan was obtained in five patients in group 1. It confirmed the diagnosis in three cases despite that creatinine level elevation forbids intravenous contrast injection. Rectosigmoidoscopy was done in two cases in each group, showing ischemic lesions in the two group 1 patients.
3.5. Laparotomy characteristics Laparotomy characteristics are shown in Table 6. Mean laparotomy delay (considered from the first clinical, biological and/or radiological finding suggestive of mesenteric ischemia to the time of exploratory laparotomy) is significantly shorter for group 1 than for group 2, 13.7±19.0 and 51.4±29.0 h, respectively (P=0.02). Intra-operatively, all group 1 patients showed diffuse bowel lesions compatible with a hypoperfusion state. No focal lesions related to arterial emboli nor venous thrombi were noted and superior mesenteric pulse was palpated in all patients. Five patients of group 1 underwent primary bowel resection while two others required a 48-h second-look laparotomy with limited resection of ischemic portions. Necrosis was limited to the colon in only two cases and extended to the small-bowel in the other five. Only one group 2 patient, initially undergoing cholecystectomy for acute cholecystitis, required a second-look laparotomy for persistent abdominal pain.
3.6. Outcome Six of the thirteen patients (46.1%) died within the first postoperative month. Three patients died in each group, corresponding to mortality rates of 42.8% and 50% for group 1 and 2, respectively. The causes of death were multi-organ failure and septicemia. All deaths occurred during the first nine postoperative days (Table 7).
Previous reports have clearly recognized mesenteric ischemia as the most serious digestive complication after CPB. It occurs in <1% of cases [1, 6, 8, 9] and carries a mortality rate of 70–100% [1, 6, 8, 10, 11]. The present series of seven confirmed cases out of 1634 consecutive CABG (0.43%) showed a lower mortality rate (42.8%) compared to the worldwide literature. This suggests that early laparotomy for suspected mesenteric ischemia after CPB could reduce mortality rate. Most studies that dealt with mesenteric ischemia following CPB aimed to identify high-risk patients [1, 8, 11, 12]. Cardiovascular risk factors and intraoperative hypoperfusion were the most commonly stated, but still controversial [9]. In the actual series, no evident differences were noted among both groups. Furthermore, relying on clinical signs to evoke mesenteric ischemia can be misleading in cardiac surgery patients. CPB triggers a systemic inflammatory response and vasodilation so that hemodynamic instability can no longer be interpreted as a suggestive sign for mesenteric ischemia. While the diagnosis usually relies on symptoms-to-signs mismatch, patients' sedation and postoperative status make physical examination unreliable to confirm or rule out the diagnosis. Cyanosis, a classical finding in ischemic patients, remains a valuable sign. On the other hand, blood tests could not be predictive of mesenteric ischemia since values were constantly abnormal after CPB. As stated by Edwards et al. [13], neither routine clinical investigations nor biological tests are discriminatory for mesenteric ischemia when the diagnosis is suspected. As for radiological tests, they lack specificity and sensitivity, being accurate in almost 35% of cases. Angiography, which is the gold standard tool for diagnosing peripheral splanchnic disease [14], and advocated by some authors as the first examination to perform, was deliberately disregarded in the current strategy, because unstable patients could not bear a time consuming and invasive procedure. Although it is well established that prompt diagnosis is imperative [1, 8], therapeutic and management guidelines have not yet been defined. Many authors claim the advantage of conservative treatment with selective angiography and local vasodilators or thrombolytic injections [7, 14, 15]. This procedure can reduce the need for laparotomy. Nevertheless, mortality rates remain high, and in failed cases, the procedure may delay early laparotomy and increase mortality [3]. In the setting of diffuse ischemic lesions due to non-occlusive disease as noted in all group 1 patients, a technically-demanding and time-consuming procedure was to be avoided. The authors of this paper believe that prompt surgery should be considered in the context of suspected mesenteric ischemia. This attitude helps breaking the vicious circle that leads to multiple organ failure and death. Close monitoring of patients by a multidisciplinary team and low threshold for suspicion of mesenteric ischemia are the key points to reduce mortality. This is suggested by shortened mean delay before surgery to 13.68 h in group 1, allowing a mortality rate of 42.8%. Assuming that the group 2 represents high-risk patients after CPB, a prompt surgical intervention does maintain a comparable mortality rate in ischemic patients. Knowing that care-giving and follow-up were similar in both groups, we consider that the longer delay in group 2 was due to more insidious clinical presentation. In these settings, exploratory laparotomy puts an end to a doubtful situation that could have lasted for several days. This actual series has several limitations. First, this is a retrospective descriptive study with consequently inherent biases. Although data retrieval was optimal by carefully scrutinizing the electronic databases, the retrospective collecting of data per se is biased. Second, the small size of the series underpowered the statistical comparisons, leading to inflated type II error. A larger sample size could have allowed more meaningful statistical comparisons and multivariate analysis so that risk models could be built up. Third, a unique approach was considered for all patients. Although this attitude constitutes the key element of this paper, comparing two or more attitudes could have resulted in better understanding of the different clinical situations. Several conclusions can be drawn from this series: (1) mesenteric ischemia following cardiac surgery is rare but associated to high hospital mortality; (2) clinical, biological and radiological explorations are not accurate for diagnosing mesenteric ischemia in the setting of CABG patients; (3) efficient patient management relies on multidisciplinary team collaboration; and (4) very low threshold for prompt laparotomy remains the most reliable diagnostic mean with potential mortality rate reduction.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction