Does Obesity Affect Outcomes in Patients Undergoing Esophagectomy for Cancer? a Meta-Analysis
Total Page:16
File Type:pdf, Size:1020Kb
World J Surg (2012) 36:1785–1795 DOI 10.1007/s00268-012-1582-4 Does Obesity Affect Outcomes in Patients Undergoing Esophagectomy for Cancer? A Meta-analysis Babar Kayani • Koji Okabayashi • Hutan Ashrafian • Leanne Harling • Christopher Rao • Ara Darzi • Yuko Kitagawa • Thanos Athanasiou • Emmanouil Zacharakis Published online: 24 April 2012 Ó Socie´te´ Internationale de Chirurgie 2012 Abstract impact on the risk of anastomotic leakage (coefficient = Background The incidence of esophageal carcinoma and -7.94 [-15.24–0.65, P = 0.03) and atrial fibrillation the global prevalence of obesity are both increasing. As a (coefficient =-6.94 [-12.79–1.10], P = 0.02). Overall, result, there is an increased number of esophagectomies obese patients had significantly better long-term survival being performed on obese patients. The identification than non-obese patients (Hazard Ratio = 0.78 [0.64–0.96], of specific complications in obese patients undergoing P = 0.02). esophagectomy may allow improved risk assessment and Conclusions In patients who are eligible for surgery, postoperative management to reduce morbidity and mor- obesity alone does not increase risk of postoperative tality. This meta-analysis aimed to determine whether complications or mortality and should not be an indepen- obese patients are at increased risk of postoperative com- dent contraindication for esophagectomy. However, the plications, mortality, and compromised survival compared presence of diabetes mellitus in conjunction with obesity to non-obese patients following esophageal resection. may be associated with increased risk of anastomotic Methods A Medline, Embase, Ovid, and Cochrane data- leakage and atrial fibrillation. Because of the adverse base search was performed on all articles between January physiological remodeling in obesity, surgeons should 1980 and January 2012 comparing post-esophagectomy maintain a low threshold for the investigation and man- outcomes between obese and non-obese patients. This agement of complications and ensure meticulous manage- study was conducted in accordance with the recommen- ment of co-morbidities. Obesity may also improve long- dations of the Cochrane Collaboration and the Quality of term postoperative survival after esophageal surgery, Reporting of Meta-Analyses guidelines. although further studies with higher levels of evidence are Results There was no significant difference between necessary to fully determine any advantageous effects of obese and non-obese patients with respect to extent of obesity following oncological esophageal surgery. tumor resection, cardiorespiratory complications, anasto- motic leakage, reoperation rates, wound infection, or post- operative mortality. Meta-regression analysis showed that Introduction diabetes in obese patients was associated with a significant Esophageal cancer is the eighth most common cancer worldwide, and the incidence of esophageal adenocarci- noma is the fastest growing of any gastrointestinal malig- B. Kayani Á H. Ashrafian Á L. Harling Á C. Rao Á A. Darzi Á nancy [1, 2]. Because of the paucity of early stage T. Athanasiou Á E. Zacharakis (&) Department of Surgery and Cancer, Imperial College London, symptoms, most patients present with advanced disease, St. Mary’s Hospital, Praed Street, London W2 1NY, UK and therefore the overall 5-year survival of esophageal e-mail: [email protected] cancer is reported between 5 and 16 % [3]. Obesity is the excessive accumulation of fat that predisposes an individual K. Okabayashi Á Y. Kitagawa Department of Surgery, Graduate School of Medicine, to health risks and is defined by the World Health Organi- 2 Keio University, Tokyo, Japan sation (WHO) as Body Mass Index (BMI) C 30 kg/m [4]. 123 1786 World J Surg (2012) 36:1785–1795 Obesity has become a global epidemic, with over 10 % of articles’’ function was used to identify further studies for the world’s adult population being affected, and it is pre- review. Two of the authors (B.K. and H.A.) independently dicted that the incidence of this disease will continue to rise. performed the literature search, and any disagreements At present, up to 82 % of patients with adenocarcinoma of between them were settled by a third author (E.Z.). The the esophagus or gastroesophageal junction are overweight search was limited to articles published between January or obese [5]. 1980 and January 2012. A data extraction form to identify Obesity is associated with the development of a number variables and potential biases in the eligible studies was of chronic diseases, including cardiovascular disease, dia- designed. The following details from each study were betes mellitus, hypertension, dyslipidemia, and cancer of recorded where possible: first author, publication date, the colon, breast, and pancreas [6, 7]. The excess central study design, patient characteristics (number of patients, adipose tissue promotes gastroesophageal reflux disease and sex ratio, age, co-morbidities), BMI of patients, type of a fourfold increase in the risk of adenocarcinoma [5]. Sur- esophageal resection (transhiatal, transthoracic, minimally gery on obese patients is also associated with significant invasive), histology (squamous cell carcinoma, adenocar- stress on an already compromised cardiorespiratory system, cinoma), tumor stage, preoperative adjuvant treatment, with reduced expiratory reserve volume and adverse ven- completion of resection margins on pathological exami- tricular remodeling [8, 9]. However, there is no consensus nation, postoperative complications, mortality, and long- about the impact of obesity on perioperative complications term survival. after esophageal surgery. Some studies have shown that surgery on obese patients leads to an increased risk of Study selection wound complications, pulmonary infections, atelectasis, renal failure, thromboembolic events, and prolonged anes- The meta-analysis included all published studies between thesia complications [10–17]. Other studies have not shown January 1980 and January 2012 comparing outcomes in any significant difference in major postoperative compli- obese and non-obese patients undergoing esophagectomy cations or mortality between obese and non-obese patients for esophageal cancer. [18–23]. Furthermore, studies on critically ill patients and Inclusion criteria for the study were these: esophageal patients undergoing specific surgical interventions, such cancer as the sole gastrointestinal pathology; clear defini- as percutaneous coronary intervention, have shown that tion of obesity based on BMI (obesity defined as overweight and obese patients have better outcomes than BMI C 30 kg/m2); obesity as the only variable between patients of normal body weight [24–27]. This survival the cohorts; clear documentation of the operative resection advantage of overweight and obese patients over patients of method used; report on at least one of the outcomes of normal BMI has become known as the ‘‘obesity paradox.’’ interest. The aims of the present study were to quantitatively Exclusion criteria for the meta-analysis were these: combine comparative studies in order to determine the studies that included patients with previous partial esoph- impact of obesity on esophagectomy with regard to: (1) R0 ageal surgery or gastric surgery; studies that did not define resection—i.e., complete tumor resection with no micro- obesity according to the WHO classification and instead metastatic residual tumor; (2) cardiovascular complica- used other definitions based on arm circumference or chest- tions; (3) respiratory complications; (4) anastomosis and to-waist-size ratio; studies on patients with esophageal wound complications; (5) reoperation rates; and (6) mor- resection without esophageal or gastroesophageal carci- tality and long-term survival. noma (for example surgery for esophageal atresia or achalasia); studies from which it was not possible to quantify or report in a standardized manner the outcome of Materials and methods interest; studies that investigated the impact of weight loss prior to surgery (to minimize bias). Data sources and extraction Outcomes of interest An electronic search was performed in the Medline, Em- base, Ovid, and Cochrane databases to identify clinical The following outcome measures were used to compare studies comparing postoperative outcomes between obese outcomes between obese and non-obese patients after and non-obese patients after esophageal resection. The esophageal resection for cancer: (1) R0 resection—i.e., following Mesh search headings were used: ‘‘Esophagec- complete tumor resection with no micrometastatic residual tomy,’’ ‘‘Obesity,’’ ‘‘Body Mass Index,’’ ‘‘Complications,’’ tumor; (2) cardiovascular complications—myocardial and ‘‘Mortality.’’ A second search was performed to extract infarction, atrial fibrillation; (3) respiratory complications— references from the retrieved articles, and the ‘‘related respiratory failure, pneumonia, empyema, chylothorax, 123 World J Surg (2012) 36:1785–1795 1787 tracheal perforation, recurrent laryngeal nerve injury, pul- are depicted in classical Forrest plots, with point estimates monary embolism/deep vein thrombosis; (4) anastomosis and the 95 % CI for each trial and overall; size of the and wound complications: anastomotic leak, gastric tip squares is proportional to effect size. necrosis, wound infection; (5) reoperation rates; (6) mor- tality and long-term survival. Results Statistical analysis Study selection Analysis of combined data was performed with RevMan 5.0 (Review Manager Version 5.0,