The impact of obesity on esophagectomy: a meta-analysis

The impact of obesity on esophagectomy: a meta-analysis SUMMARY The impact of body mass index (BMI) on postoperative outcomes after curative resection for esophageal cancer has been assessed in many studies worldwide with conflicting conclusions. The aim of this meta-analysis is to evaluate the influence of preoperative BMI on surgical and oncologic outcomes after radical surgery for esophageal cancer, in Western studies. A comprehensive electronic search was performed to identify Western publications reporting BMI and outcomes following surgery for esophageal cancer. Articles that did not report preoperative BMI, postoperative morbidity, and early mortality were excluded. Statistical analysis was performed using the OpenMetaAnalyst software (Version 10.10). One hundred and ninety records were examined and 8 studies were included with a total of 2838 patients. The study population was stratified into two groups: a nonobese group (BMI < 30 kg/m2), containing 2199 patients, and an obese group (BMI ≥ 30 kg/m2), with 639 patients. In the obese group, there was an increased risk (up to 35%) of anastomotic leak (P = 0.003; RR: 0.857, 95% CI: 0.497, 0.867). The obese group showed a significantly more favorable five-year overall survival (P = 0.011). Although there was a significant association between anastomotic leak and obesity, patients with obesity also have a better overall 5-year survival. This meta-analysis demonstrates that patients with obesity should be counseled regarding the specific risks of surgery but they can be reassured that despite these risks overall outcome is satisfactory. INTRODUCTION Several studies have shown that adiposity and high body mass index (BMI) are associated with increased morbidity and mortality following general and esophagogastric cancer surgery.1,2 The BMI (the weight in kilograms divided by the square of the height in meters) is the World Health Organisation's3 international index used to classify underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5–24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2), and obese (BMI ≥ 30 kg/m2) patients. Over the last 30 years, obesity rates have been steadily increasing among Western populations while its prevalence in Eastern regions has remained consistently low. Currently, Asia shows the lowest prevalence rate (3.2% in males, 6.8% in females) while USA records the highest (24% in males, 29.6% in females) and Europe presents an intermediate rate (21.5% in males, 24.5% in females).4–6 Obesity is associated with a number of chronic conditions including diabetes mellitus, hypertension, ischemic heart disease, and cerebrovascular disease, all of which can increase anesthetic and surgical risk.5,7,8 These comorbidities can contribute to postoperative complications such as wound infections, anastomotic leak, and respiratory disease.9,10 Moreover, surgery in patients with obesity can result in increased blood loss and a longer operation time.11 It is now well known that obesity plays a role in the etiology of esophagogastric junctional adenocarcinomas.12 Studies on the impact of obesity on the surgical management of esophageal cancer are inconsistent with some reporting a higher incidence of postoperative complications. Outcomes such as operative time, postoperative bleeding, and lymph node harvest rate are variable between studies. Surprisingly, several studies confirm obesity is associated with better long-term overall survival.13-15 Eastern studies often include patients with a BMI > 25 as obese and their study populations (early stage squamous cell carcinoma, SCC) are quite different to Western groups.16 The aim of this meta-analysis is to evaluate the effect of preoperative BMI on outcome following radical surgery for esophageal cancer in a Western series. MATERIALS AND METHODS Search strategy and study selection A comprehensive electronic search from 2004 to December 2016 was performed to identify Western publications relating to preoperative BMI and outcomes after esophageal surgery for cancer. The search was performed using PubMed search engine and the following MeSH terms (Medical Subject Headings) were combined: ‘Body mass index’, ‘Overweight’, ‘Obesity’, ‘Esophageal Neoplasm's’, ‘Esophageal Neoplasms/surgery’, ‘Esophagectomy’, ‘Surgical Procedures, Operative’ and ‘Upper Gastrointestinal Tract Surgery’. A further search of the reference lists of all relevant articles was also undertaken to identify any additional studies. In order to exclude overlapping publications or publications sharing the same population, the authors only included the most pertinent article in terms of clarity and completeness of information. Selected articles included all case-control and cohort studies according to the following criteria: (1) series from Western centers, (2) clear definition of overweight and obesity according to WHO classification, (3) homogeneous cohorts of patients with BMI as a lone variable between groups, (4) analysis of at least one of the outcomes of interest (surgical or oncological), (5) reporting on multimodal treatment details and (6) English language journals published since 2004. The exclusion criteria for the meta-analysis were as follows: reviews or case reports, studies sharing the same population, reports from Eastern countries, studies of the nutritional effect of esophagogastric surgery and those reporting or examining postoperative outcomes after esophagectomy for benign or traumatic conditions. Outcomes of interest and data extraction The following variables were used for comparison between the different BMI groups: anastomotic leak, perioperative bleeding, recurrent laryngeal nerve injury, chyle leak, abdominal and wound infection, cardiac complications, pulmonary complications, re-operation rate, in-hospital and 30 day mortality. Long-term survival was analyzed as an oncologic outcome. Data extraction included first author, publication date, study design, demographic characteristics (population size and inclusion period, and BMI cuts-off), multimodal treatment (neoadjuvant and adjuvant therapies, type of surgical procedures and whether a minimally invasive approach was used), and disease characteristics (histological subtype, pathological stage, and surgical margins). Groups identified According to the WHO definition, obesity is defined as a BMI ≥ 30 kg/m2. This was used to stratify patients into two groups: nonobese (BMI < 30 kg/m2) and obese (BMI ≥ 30 kg/m2). The studies that were selected had constructed the groups according to the WHO definitions as per Table 1. Patients with BMI > 35 (7%; 187 patients) and BMI < 20 (10%; 295 patients) were included in these groups as they were not separated in the original studies. We have identified that this is a source of potential error in interpretation of the overall results. Table 1 Studies of esophageal resection: patient characteristics Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  †Number of patients; ‡Nonobese versus Obese, number of patients. Some underweight (BMI < 20–18.5 kg/m2) and Severely Obese (Obesity Class II-III, BMI > 35 kg/m2) people were included in nonobese and obese groups respectively. View Large Table 1 Studies of esophageal resection: patient characteristics Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  †Number of patients; ‡Nonobese versus Obese, number of patients. Some underweight (BMI < 20–18.5 kg/m2) and Severely Obese (Obesity Class II-III, BMI > 35 kg/m2) people were included in nonobese and obese groups respectively. View Large Quality assessment and statistical analysis Quality assessment and statistical methods were based on the Cochrane Handbook for Systematic Reviews of Interventions.17 All retrieved articles were qualitatively evaluated in terms of patients selection, adequate reporting of outcomes and presence of clear variable definitions. Data from different studies reporting similar outcomes were combined and since only binary variables were examined the results were expressed as relative ratio (RR). Heterogeneity among selected articles was checked using χ2 tests where admitted cuts-off were P-value < 0.05 for statistical significance and 95% as Confidence Interval (CI). After data extraction, forest plot analyses were conducted for all observed outcome variables. All statistical analyses were performed using Open Meta-Analyst software (Version 10.10) by the Center for Evidence-Based Medicine (CEBM) of Brown University.18 RESULTS Search results and included trials The search produced a total of 190 records matching all MeSH terms on the PubMed search engine (Fig. 1). After the initial selection, 58 articles were removed due to age of publication or overlapping data. One hundred and thirty-two studies were examined by comparing all of the abstracts to the inclusion criteria: 104 articles were excluded due to a lack of information on BMI, patients' treatment, and if they were an Eastern series, reviews, or case reports. Twenty-eight papers were assessed for eligibility and 8 Western studies were included in the meta-analysis. All the main characteristics of the selected articles are summarized in Tables 1 and 2.13-15,19-23 Fig. 1 View largeDownload slide Selection process for articles chosen. Fig. 1 View largeDownload slide Selection process for articles chosen. Table 2 Studies of esophageal resection: treatment characteristics     Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775        Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775    †Histopathology: EAC, esophageal adenocarcinoma; SCC, squamous cell carcinoma; ‡Note: Staging information missing on 99 patients; §Perioperative oncological treatments: AC, adjuvant chemotherapy; NAC, neoadjuvant chemotherapy; NS, not stated; ¶TT, trans thoracic (including 2-stage, 3-stage and thoraco-abdominal approaches); MI, minimally invasive; TH, transhiatal; ††Follow-up: Median in months, (Range if possible); ‡‡Data not available on 24 patients. View Large Table 2 Studies of esophageal resection: treatment characteristics     Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775        Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775    †Histopathology: EAC, esophageal adenocarcinoma; SCC, squamous cell carcinoma; ‡Note: Staging information missing on 99 patients; §Perioperative oncological treatments: AC, adjuvant chemotherapy; NAC, neoadjuvant chemotherapy; NS, not stated; ¶TT, trans thoracic (including 2-stage, 3-stage and thoraco-abdominal approaches); MI, minimally invasive; TH, transhiatal; ††Follow-up: Median in months, (Range if possible); ‡‡Data not available on 24 patients. View Large This amounted to a comprehensive population of 2838 patients who were stratified into two groups according to WHO definition of obesity: 2199 (77%) patients were nonobese (BMI < 30 kg/m2) and 639 (23%) were obese (BMI ≥ 30 kg/m2). The results of the analyses on postoperative outcomes between the obese and nonobese groups are summarized in Table 3. Table 3 Summary of outcomes after esophagectomy (includes overall survival) Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  *Significant results; †Values were calculated with Fixed Effects MH. View Large Table 3 Summary of outcomes after esophagectomy (includes overall survival) Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  *Significant results; †Values were calculated with Fixed Effects MH. View Large Analysis of surgical outcomes Information on the type of operation performed was available for the majority of the studies (Table 2). There were 1131 (40%) transthoracic esophagectomies (including 2-stage, 3-stage, and thoracoabdominal approach), 1328 (47%) transhiatal esophagectomies, and 355 (13%) minimally invasive procedures. Information on the anastomotic type was incomplete but where available, there were 761 (30%) thoracic anastomoses compared to 1775 (70%) cervical anastomoses (Table 2). Histopathological stage was documented in 6 studies (1912 patients). For early stage disease (Stage 0–2) 55 (7%) patients were obese and 693 (93%) were nonobese. In late stage disease (Stage 3–4) 332 (29%) patients were obese, and 832 (71%) nonobese. Both postoperative surgical and medical complications were analyzed. Most articles did not follow a standardized classification of complications. Only one study classified complications19 according to Clavien-Dindo.24 There was no significant association between all complications and BMI (Supplementary Fig. S1) (P = 0.578). The obese group was found to have a significantly increased incidence of anastomotic leak (RR = 0.657 [0.497, 0.867], P = 0.003) (Fig. 2). All other postoperative complications such as postoperative bleeding, early mortality, chyle leak, recurrent laryngeal nerve injury, re-operation, wound infection, cardiac, or respiratory complications did not show any significant association with obesity (Supplementary Figs S2–S9). Fig. 2 View largeDownload slide Anastomotic leak: There was a significant association between anastomotic leak and BMI ≥ 30 Subgroup 1 comprised studies with mixed histopathology (SCC and EAC). Subgroup 2 contained studies with EAC only. As subgroup 2 were small in number, this separate analysis was not significant. The total group is referred to in the results section. Fig. 2 View largeDownload slide Anastomotic leak: There was a significant association between anastomotic leak and BMI ≥ 30 Subgroup 1 comprised studies with mixed histopathology (SCC and EAC). Subgroup 2 contained studies with EAC only. As subgroup 2 were small in number, this separate analysis was not significant. The total group is referred to in the results section. Analysis of oncological outcomes The rate of surgical margin positivity after esophagectomy was compared in 5 studies and this was not significantly associated with BMI (RR = 1.021 [0.745, 1.399], P = 0.899) (Supplementary Fig. S10). Only three of the studies reported results on long-term survival. According to the data, the higher BMI group was significantly associated with a better 5 year overall survival (RR = 1.172 [1.038, 1.323], P = 0.011) (Fig. 3). Fig. 3 View largeDownload slide 5 Year Overall Survival: There was a significant association between 5-year overall survival and BMI ≥ 30. Fig. 3 View largeDownload slide 5 Year Overall Survival: There was a significant association between 5-year overall survival and BMI ≥ 30. DISCUSSION This meta-analysis, of 2838 patients undergoing esophagectomy, has evaluated perioperative morbidity and mortality in obese patients (BMI ≥ 30). Obesity was strongly associated with an anastomotic leak. The severity of anastomotic leak (ranging from subclinical, found on imaging to gastric necrosis requiring reoperation) is unclear from the 8 studies. A transhiatal esophagectomy with a cervical anastomosis was often performed in patients with SCC, whereas an Ivor Lewis esophagogastrectomy with a thoracic anastomosis was the operation of choice for junctional adenocarcinomas. The operation type may be the reason for this significant effect on anastomotic integrity. This study also demonstrates that the obese group had a better 5-year overall survival rate after esophageal resection for cancer. This result is not as easy to interpret and may reflect staging differences or less cachexia in the obese group. The study did not reveal any relationship between obesity and wound infection, cardiac complications, postoperative bleeding, or respiratory complications, which have been associated with obesity in other studies.9–11 Preoperative optimization and prophylactic measures such as prehabilitation, optimization of cardiac function and perioperative antibiotics in this group may explain this result. In the last 20 years, many authors have focused on obesity and its effect on surgical procedures.25-27 It is well established that patients with obesity have a higher incidence of comorbidities such as diabetes, cardiac and pulmonary conditions, as well as increased girth, which leads to technical difficulty during surgery, worse postoperative recovery and, therefore, an impact on health care expenditure. Studies pertaining to obesity and esophageal cancer resections have shown conflicting results. Some have described an association between obesity and severe complications including anastomotic leak,22 others have described no difference in complications between obese and nonobese in the postoperative period.21 Some have documented that patients with obesity show better disease free and overall survival,13-15 others discount this and show the same outcome for both groups.19-23 Eastern studies analyzing the influence of high BMI on patients undergoing esophagectomy use different classifications than the WHO definition, for example, Zogg et al. included patients with a BMI > 27.5 as obese.28 Indeed a WHO expert consultation has published recommendations on BMI in Asian populations indicating that a BMI > 23 represented increased risk and a BMI > 27.5 represented high risk.29 Eastern study populations can be quite different to Western series. Zhang et al. reported a series of 2031 patients, which consisted mainly of patients with SCC (87.4%) with an early pathological stage (mostly ≤ stage II).16 Our study sought to clarify the risk of obesity (BMI > 30) in Western studies. Our study population had a higher number of cervical anastomoses (1775) than thoracic (762). We found that obesity is a risk factor for a postoperative anastomotic leak. This could be explained by impaired vascularity of the gastric conduit, by microvascular disease and diabetes, which are more common in patients with obesity.19 Pierie et al. demonstrated, in 1994, that assessment of blood perfusion of the gastric tube is a significant predictor of anastomotic stricture rates and impaired healing in cervical anastomoses.30 A more recent paper has shown that BMI > 25 kg/m2 was significantly associated with development of a cervical stricture on univariate analysis.31 Increased tension on the conduit due to a cervical anastomosis may also contribute to this result. Neck dissection itself can be more challenging in a patient with obesity with a short neck.20-23 Long-term survival is an interesting outcome of this meta-analysis. Several studies previously reported no influence of obesity on overall survival32 while some considered obesity as a negative prognostic factor33-35 and others reported that a higher BMI could be a protective factor after surgery.14 In this study, the obese group had a significantly better 5-year overall survival even though the two groups were comparable in terms of surgical radicality. A number of factors may account for this apparent benefit. Obesity may confer a protective nutritional effect. Multiple hormonal, endocrine, or nutritional factors could contribute to the benefit of obesity in this situation.14,15 Grotenhuis et al. reported a reduced rate of circumferential margin involvement reflecting a greater volume of fat around the esophagus; this study however, did not show an overall difference in survival.22 A number of studies have reported an association of raised BMI with earlier stage disease.15,20,22,23,36 This may reflect higher rates of endoscopic surveillance in view of the higher rates of gastro-esophageal reflux symptoms in obese patients.37 This however was not borne out in this analysis as the patients with early stage disease comprised 292 (25%) obese and 872 (75%) nonobese. In the group with late stage disease the proportions were similar (409 (25%) were obese and 1237 (75%) were nonobese). The improved survival in obese patients may be a relative difference reflecting the poorer outcome with normal or low BMI in whom weight loss may have a greater effect.14,36 We excluded eastern studies to minimize heterogeneity between articles, specifically, regarding patient characteristics, and differences in terms of diagnostic and multimodal approach. The number of available western series that fit the inclusion criteria has limited this meta-analysis. Despite this, significant differences in operative technique (ranging from totally minimally invasive to open approach with cervical anastomosis) and histological characteristics (SCC and esophageal adenocarcinoma, EAC) persisted (Table 2). Another limitation, that previous meta-analyses have mentioned, is the influence of a phenomenon named ‘obesity paradox’: an intrinsic selection bias against the obese group due to multiple comorbidities leading to better mean performance status by more careful selection for surgery.38 As standardized classifications and definitions of all variables of interest were not used in these studies, data extraction was limited due to inconsistency in variables such as specific patient characteristics (comorbidities, ASA score or tumor location), other treatment data and oncological features (type of lymphadenectomy and surgical radicality). As some studies did not independently define underweight patients (BMI < 25) or super obese patients, these were included in the nonobese and obese group, respectively, and may have confounded the results. Many studies do not undertake to separate out these diverse groups. As both extremes can equally contribute to significant morbidity in their own right there is an inaccurate portrayal of the extent of the problem when groups are combined. Only one of the eight studies reported complications according to the Clavien-Dindo grade.24 This greatly impacts the quality of the studies and as a result the quality of the meta-analysis. The findings of this study have highlighted further areas to assess to improve outcome for this group of patients. The higher rate of anastomotic leak with a cervical anastomosis is well known but does seem to be a particular problem in the obese population. There is good evidence to support prehabilitation in terms of postoperative morbidity, length of stay, nutritional and physical status.39 A trial examining a prehabilitation program (PREHAB) in patients undergoing esophagogastric surgery is underway40 and it will be interesting to see if such an approach would be beneficial to the obese group. In summary, this meta-analysis demonstrates that in Western series of patients with esophageal cancer 23% of patients had obesity. This is associated with an increase in morbidity, specifically anastomotic leak but this did not translate into early mortality or a reduction in long term survival. While a raised BMI should not preclude multimodal treatment including surgical resection in Western patients, we would suggest it is important to mitigate risk by careful preoperative planning, optimization and a higher index of suspicion for anastomotic leak postoperatively. SUPPLEMENTARY DATA Supplementary data are available at DOTESO online. References 1 Clinical Oncology Group. Identification of risk factors for the development of complications following extended and superextended lymphadenectomies for gastric cancer. Br J Surg  2005; 92: 1103– 9. CrossRef Search ADS PubMed  2 Bamgbade O A, Rutter T W, Nafiu O O, Dorje P. Postoperative complications in obese and non-obese patients. World J Surg  2007; 31: 556– 60. Google Scholar CrossRef Search ADS PubMed  3 World Health Organization. Obesity: preventing and managing the global epidemic. Report of WHO consultation. World Health Organ Tech Rep Ser  2000; 894: 1– 253. 4 World Health Organization. World Health Statistics 2015. Retrieved on 13/12/2017 from http://www.who.int/gho/publications/world_health_statistics/EN_WHS2015_Part2.pdf?ua=1. 5 National Obesity Observatory (NHS), UK. Adult Obesity International Comparisons data factsheet. 2016. Retrieved 13/12/2017 from http://webarchive.nationalarchives.gov.uk/20170110170001/https://www.noo.org.uk/NOO_pub/Key_data. 6 Flegal K M, Carroll M D, Ogden C L, Curtin L R. Prevalence and trends in obesity among US adults, 1999–2008. JAMA  2010; 303: 235– 41. Google Scholar CrossRef Search ADS PubMed  7 Dority J, Hassan Z U, Chau D. Anesthetic implications of obesity in the surgical patient. Clin Colon Rectal Surg  2011; 24: 222– 8. Google Scholar CrossRef Search ADS PubMed  8 Dindo D, Muller M K, Weber M, Clavien P A. Obesity in general elective surgery. Lancet  2003; 361: 2032– 5. Google Scholar CrossRef Search ADS PubMed  9 Flancbaum L, Choban P S. Surgical implications of obesity. Annu Rev Med  1998; 49: 215– 34. Google Scholar CrossRef Search ADS PubMed  10 Mullen J T, Davenport D L, Hutter M M et al.   Impact of body mass index on perioperative outcomes in patients undergoing major intra-abdominal cancer surgery. Ann Surg Oncol  2008; 15: 2164– 72. Google Scholar CrossRef Search ADS PubMed  11 Tjeertes E K, Hocks S E, Beks S B et al.   Obesity-a risk factor for postoperative complications in general surgery? BMC Anaesthesia  2015; 15: 112. Google Scholar CrossRef Search ADS   12 Ryan A M, Duong M, Healy L et al.   Obesity, metabolic syndrome and esophageal adenocarcinoma: epidemiology, etiology and new targets. Cancer Epidemiol  2011; 35: 309– 19. Google Scholar CrossRef Search ADS PubMed  13 Scarpa M, Cagol M, Bettini S et al.   Overweight patients operated on for cancer of the esophagus survive longer than normal-weight patients. J Gastrointest Surg  2013; 17: 218– 27. Google Scholar CrossRef Search ADS PubMed  14 Melis M, Weber J M, McLoughlin J M et al.   An elevated body mass index does not reduce survival after esophagectomy for cancer. Ann Surg Oncol  2011; 18: 824– 31. Google Scholar CrossRef Search ADS PubMed  15 Madani K, Zhao R, Lim H J, Casson S M, Casson A G. Obesity is not associated with adverse outcome following surgical resection of oesophageal adenocarcinoma. Eur J Cardiothorac Surg  2010; 38: 604– 8. Google Scholar CrossRef Search ADS PubMed  16 Zhang S S, Yang H, Luo K J et al.   The impact of body mass index on complication and survival in resected oesophageal cancer: a clinical-based cohort and meta-analysis. Br J Cancer  2013; 109: 2894– 903. Google Scholar CrossRef Search ADS PubMed  17 Higgins J P T, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. www.cochrane-handbook.org. 18 Wallace B C, Dahabreh I J, Trikalinos T A, Lau J, Trow P, Schmid C H. Closing the gap between methodologists and end-users: R as a computational back-end. J Stat Software  2012; 5: 1– 15. 19 Blom R L, Lagarde S M, Klinkenbijl J H, Busch O R, van Berge Henegouwen M I. A high body mass index in esophageal cancer patients does not influence postoperative outcome or long-term survival. Ann Surg Oncol  2012; 19: 766– 71. Google Scholar CrossRef Search ADS PubMed  20 Kilic A, Schuchert M J, Pennathur A et al.   Impact of obesity on perioperative outcomes of minimally invasive esophagectomy. Ann Thorac Surg  2009; 87: 412– 5. Google Scholar CrossRef Search ADS PubMed  21 Scipione C N, Chang A C, Pickens A, Lau C L, Orringer M B. Transhiatal esophagectomy in the profoundly obese: implications and experience. Ann Thorac Surg  2007; 84: 376– 82. Google Scholar CrossRef Search ADS PubMed  22 Grotenhuis B A, Wijnhoven B P, Hötte G J, van der Stok E P, Tilanus H W, van Lanschot J J. Prognostic value of body mass index on short-term and long-term outcome after resection of esophageal cancer. World J Surg  2010; 34: 2621– 7. Google Scholar CrossRef Search ADS PubMed  23 Healy L A, Ryan A M, Gopinath B, Rowley S, Byrne P J, Reynolds J V. Impact of obesity on outcomes in the management of localized adenocarcinoma of the esophagus and esophagogastric junction. J Thorac Cardiovasc Surg  2007; 134: 1284– 91. Google Scholar CrossRef Search ADS PubMed  24 Clavien P A, Barkun J, de Oliveira M L et al.   The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg  2009; 250: 187– 96. Google Scholar CrossRef Search ADS PubMed  25 Wigfield C H, Lindsey J D, Munoz A, Chopra P S, Edwards N M, Love R B. Is extreme obesity a risk factor for cardiac surgery? An analysis of patients with a BMI > 40. Eur J Cardiothorac Surg  2006; 29: 434– 40. Google Scholar CrossRef Search ADS PubMed  26 Williams T K, Rosato E L, Kennedy E P et al.   Impact of obesity on perioperative morbidity and mortality after pancreaticoduodenectomy. J Am Coll Surg  2009; 208: 210– 7. Google Scholar CrossRef Search ADS PubMed  27 Duan X F, Tang P, Shang X B, Jiang H J, Zhao Q, Yu Z T. High body mass index worsens survival in patients with esophageal squamous cell carcinoma after esophagectomy. Dig Surg  2017; 34: 319– 27. Google Scholar CrossRef Search ADS PubMed  28 Zogg C K, Mungo B, Lidor A O et al.   Influence of body mass index on outcomes after major resection for cancer. Surgery  2015; 158: 472– 85. Google Scholar CrossRef Search ADS PubMed  29 WHO expert consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet  2004; 363: 157– 63. CrossRef Search ADS PubMed  30 Pierie J P, De Graaf P W, Poen H et al.   Impaired healing of cervical oesophagogastrostomies can be predicted by estimation of gastric serosal blood perfusion by laser Doppler flowmetry. Eur J Surg  1994; 160: 599– 603. Google Scholar PubMed  31 van Heijl M, Gooszen J A, Fockens P et al.   Risk factors for development of benign cervical strictures after esophagectomy. Annal Surg  2010; 251: 1064– 9. Google Scholar CrossRef Search ADS   32 Takemoto K, Shiozaki A, Fujiwara H et al.   High BMI does not influence short- and long-term outcomes of patients with esophageal cancer treated with esophagectomy. Gan To Kagaku Ryoho  2014; 41: 1991– 3. Google Scholar PubMed  33 Yoon H H, Lewis M A, Shi Q et al.   Prognostic impact of body mass index stratified by smoking status in patients with esophageal adenocarcinoma. J Clin Oncol  2011; 29: 4561– 7. Google Scholar CrossRef Search ADS PubMed  34 Watanabe M, Ishimoto T, Baba Y et al.   Prognostic impact of body mass index in patients with squamous cell carcinoma of the esophagus. Ann Surg Oncol  2013; 20: 3984– 91. Google Scholar CrossRef Search ADS PubMed  35 Cheng Y, Wang N, Wang K et al.   Prognostic value of body mass index for patients undergoing esophagectomy for esophageal squamous cell carcinoma. Jpn J Clin Oncol  2013; 43: 146– 53. Google Scholar CrossRef Search ADS PubMed  36 Hayashi Y, Correa A M Y, Hofstetter W L et al.   The influence of high body mass index on the prognosis of patients with esophageal cancer after surgery as primary therapy. Cancer  2010; 116: 5619– 27. Google Scholar CrossRef Search ADS PubMed  37 Tytgat G N, Van Sandick J W, Lanschot J J, Obertop H. Role of surveillance in intestinal metaplasia of the esophagus and gastroesophageal junction. World J Surg  2003; 27: 1021– 5. Google Scholar CrossRef Search ADS PubMed  38 STARSurg Collaborative. Multicenter prospective cohort study of body mass index and postoperative complications following gastrointestinal surgery. BJS  2016; 103: 1157– 72. CrossRef Search ADS   39 Valkenet K, van de Port I G, Dronkers J J, de Vries W R, Lindeman E, Backx F J. The effects of preoperative exercise therapy on postoperative outcome: a systematic review. Clin Rehabil  2011; 25: 99– 111. Google Scholar CrossRef Search ADS PubMed  40 Le Roy B, Pereira B, Bouteloup C et al.   Effect of prehabilitation in gastrooesophageal adenocarcinoma: study protocol of a multicentric, randomised, control trial-the PREHAB study. BMJ  2016; 6: e012876. Google Scholar CrossRef Search ADS   © The Author(s) 2017. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diseases of the Esophagus Oxford University Press

The impact of obesity on esophagectomy: a meta-analysis

Loading next page...
 
/lp/ou_press/the-impact-of-obesity-on-esophagectomy-a-meta-analysis-snsHenAs9J
Publisher
The International Society for Diseases of the Esophagus
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1120-8694
eISSN
1442-2050
D.O.I.
10.1093/dote/dox149
Publisher site
See Article on Publisher Site

Abstract

SUMMARY The impact of body mass index (BMI) on postoperative outcomes after curative resection for esophageal cancer has been assessed in many studies worldwide with conflicting conclusions. The aim of this meta-analysis is to evaluate the influence of preoperative BMI on surgical and oncologic outcomes after radical surgery for esophageal cancer, in Western studies. A comprehensive electronic search was performed to identify Western publications reporting BMI and outcomes following surgery for esophageal cancer. Articles that did not report preoperative BMI, postoperative morbidity, and early mortality were excluded. Statistical analysis was performed using the OpenMetaAnalyst software (Version 10.10). One hundred and ninety records were examined and 8 studies were included with a total of 2838 patients. The study population was stratified into two groups: a nonobese group (BMI < 30 kg/m2), containing 2199 patients, and an obese group (BMI ≥ 30 kg/m2), with 639 patients. In the obese group, there was an increased risk (up to 35%) of anastomotic leak (P = 0.003; RR: 0.857, 95% CI: 0.497, 0.867). The obese group showed a significantly more favorable five-year overall survival (P = 0.011). Although there was a significant association between anastomotic leak and obesity, patients with obesity also have a better overall 5-year survival. This meta-analysis demonstrates that patients with obesity should be counseled regarding the specific risks of surgery but they can be reassured that despite these risks overall outcome is satisfactory. INTRODUCTION Several studies have shown that adiposity and high body mass index (BMI) are associated with increased morbidity and mortality following general and esophagogastric cancer surgery.1,2 The BMI (the weight in kilograms divided by the square of the height in meters) is the World Health Organisation's3 international index used to classify underweight (BMI < 18.5 kg/m2), normal weight (BMI 18.5–24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2), and obese (BMI ≥ 30 kg/m2) patients. Over the last 30 years, obesity rates have been steadily increasing among Western populations while its prevalence in Eastern regions has remained consistently low. Currently, Asia shows the lowest prevalence rate (3.2% in males, 6.8% in females) while USA records the highest (24% in males, 29.6% in females) and Europe presents an intermediate rate (21.5% in males, 24.5% in females).4–6 Obesity is associated with a number of chronic conditions including diabetes mellitus, hypertension, ischemic heart disease, and cerebrovascular disease, all of which can increase anesthetic and surgical risk.5,7,8 These comorbidities can contribute to postoperative complications such as wound infections, anastomotic leak, and respiratory disease.9,10 Moreover, surgery in patients with obesity can result in increased blood loss and a longer operation time.11 It is now well known that obesity plays a role in the etiology of esophagogastric junctional adenocarcinomas.12 Studies on the impact of obesity on the surgical management of esophageal cancer are inconsistent with some reporting a higher incidence of postoperative complications. Outcomes such as operative time, postoperative bleeding, and lymph node harvest rate are variable between studies. Surprisingly, several studies confirm obesity is associated with better long-term overall survival.13-15 Eastern studies often include patients with a BMI > 25 as obese and their study populations (early stage squamous cell carcinoma, SCC) are quite different to Western groups.16 The aim of this meta-analysis is to evaluate the effect of preoperative BMI on outcome following radical surgery for esophageal cancer in a Western series. MATERIALS AND METHODS Search strategy and study selection A comprehensive electronic search from 2004 to December 2016 was performed to identify Western publications relating to preoperative BMI and outcomes after esophageal surgery for cancer. The search was performed using PubMed search engine and the following MeSH terms (Medical Subject Headings) were combined: ‘Body mass index’, ‘Overweight’, ‘Obesity’, ‘Esophageal Neoplasm's’, ‘Esophageal Neoplasms/surgery’, ‘Esophagectomy’, ‘Surgical Procedures, Operative’ and ‘Upper Gastrointestinal Tract Surgery’. A further search of the reference lists of all relevant articles was also undertaken to identify any additional studies. In order to exclude overlapping publications or publications sharing the same population, the authors only included the most pertinent article in terms of clarity and completeness of information. Selected articles included all case-control and cohort studies according to the following criteria: (1) series from Western centers, (2) clear definition of overweight and obesity according to WHO classification, (3) homogeneous cohorts of patients with BMI as a lone variable between groups, (4) analysis of at least one of the outcomes of interest (surgical or oncological), (5) reporting on multimodal treatment details and (6) English language journals published since 2004. The exclusion criteria for the meta-analysis were as follows: reviews or case reports, studies sharing the same population, reports from Eastern countries, studies of the nutritional effect of esophagogastric surgery and those reporting or examining postoperative outcomes after esophagectomy for benign or traumatic conditions. Outcomes of interest and data extraction The following variables were used for comparison between the different BMI groups: anastomotic leak, perioperative bleeding, recurrent laryngeal nerve injury, chyle leak, abdominal and wound infection, cardiac complications, pulmonary complications, re-operation rate, in-hospital and 30 day mortality. Long-term survival was analyzed as an oncologic outcome. Data extraction included first author, publication date, study design, demographic characteristics (population size and inclusion period, and BMI cuts-off), multimodal treatment (neoadjuvant and adjuvant therapies, type of surgical procedures and whether a minimally invasive approach was used), and disease characteristics (histological subtype, pathological stage, and surgical margins). Groups identified According to the WHO definition, obesity is defined as a BMI ≥ 30 kg/m2. This was used to stratify patients into two groups: nonobese (BMI < 30 kg/m2) and obese (BMI ≥ 30 kg/m2). The studies that were selected had constructed the groups according to the WHO definitions as per Table 1. Patients with BMI > 35 (7%; 187 patients) and BMI < 20 (10%; 295 patients) were included in these groups as they were not separated in the original studies. We have identified that this is a source of potential error in interpretation of the overall results. Table 1 Studies of esophageal resection: patient characteristics Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  †Number of patients; ‡Nonobese versus Obese, number of patients. Some underweight (BMI < 20–18.5 kg/m2) and Severely Obese (Obesity Class II-III, BMI > 35 kg/m2) people were included in nonobese and obese groups respectively. View Large Table 1 Studies of esophageal resection: patient characteristics Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  Study  Country  Study type  Inclusion period  Sample size†  BMI of patients  Groups‡  Scarpa et al.13  Italy  Prospective  2000 - 2008  278  >30  61  BMI ≥ 30 61 (22%)      Cohort study      25–29.9  121  vs            20–24.9  81  BMI < 30 217 (78%)            <19.9  15    Blom et al.19  The Netherlands  Prospective  1993–2010  736  >35  15  BMI ≥ 30 76 (10%)      Cohort study      >30  61  vs            25–29.9  308  BMI < 30 660 (90%)            20–24.9  304              <19.9  48    Melis et al.14  USA, NY–FL  Retrospective  1994–2008  490  >35  64  BMI ≥ 30 166 (34%)      Cohort study      >30  102  vs            25–29.9  176  BMI < 30 324 (66%)            20–24.9  148    Grotenhuis et al.22  The Netherlands  Prospective  1991–2007  556  >30  58  BMI ≥ 30 58 (10%)      Cohort study      25–29.9  214  vs            18.5–24.9  244  BMI < 30 498 (90%)            <18.5  40    Madani et al.15  Canada  Prospective  1991–2006  142  >30  56  BMI ≥ 30 56 (39%)      Cohort study      20–29.9  85  vs            <19.9  1  BMI < 30 86 (61%)  Kilic et al.20  USA, PA  Retrospective  1999 - 2004  282  >30  84  BMI ≥ 30 84 (30%)      Cohort study      18.5–29.9  198  vs                BMI < 30 198 (70%)  Healy et al.23  UK, Ireland  Retrospective  1998–2005  150  >35  6  BMI ≥ 30 36 (24%)      Cohort study      >30  30  vs            25–29.9  71  BMI < 30 114 (76%)            20–24.9  43    Scipione et al.21  USA, CA  Retrospective  1977–2006  204  >35  102  BMI ≥ 35 102 (50%)      Case control study      <30  102  vs                BMI < 30 102 (50%)  †Number of patients; ‡Nonobese versus Obese, number of patients. Some underweight (BMI < 20–18.5 kg/m2) and Severely Obese (Obesity Class II-III, BMI > 35 kg/m2) people were included in nonobese and obese groups respectively. View Large Quality assessment and statistical analysis Quality assessment and statistical methods were based on the Cochrane Handbook for Systematic Reviews of Interventions.17 All retrieved articles were qualitatively evaluated in terms of patients selection, adequate reporting of outcomes and presence of clear variable definitions. Data from different studies reporting similar outcomes were combined and since only binary variables were examined the results were expressed as relative ratio (RR). Heterogeneity among selected articles was checked using χ2 tests where admitted cuts-off were P-value < 0.05 for statistical significance and 95% as Confidence Interval (CI). After data extraction, forest plot analyses were conducted for all observed outcome variables. All statistical analyses were performed using Open Meta-Analyst software (Version 10.10) by the Center for Evidence-Based Medicine (CEBM) of Brown University.18 RESULTS Search results and included trials The search produced a total of 190 records matching all MeSH terms on the PubMed search engine (Fig. 1). After the initial selection, 58 articles were removed due to age of publication or overlapping data. One hundred and thirty-two studies were examined by comparing all of the abstracts to the inclusion criteria: 104 articles were excluded due to a lack of information on BMI, patients' treatment, and if they were an Eastern series, reviews, or case reports. Twenty-eight papers were assessed for eligibility and 8 Western studies were included in the meta-analysis. All the main characteristics of the selected articles are summarized in Tables 1 and 2.13-15,19-23 Fig. 1 View largeDownload slide Selection process for articles chosen. Fig. 1 View largeDownload slide Selection process for articles chosen. Table 2 Studies of esophageal resection: treatment characteristics     Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775        Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775    †Histopathology: EAC, esophageal adenocarcinoma; SCC, squamous cell carcinoma; ‡Note: Staging information missing on 99 patients; §Perioperative oncological treatments: AC, adjuvant chemotherapy; NAC, neoadjuvant chemotherapy; NS, not stated; ¶TT, trans thoracic (including 2-stage, 3-stage and thoraco-abdominal approaches); MI, minimally invasive; TH, transhiatal; ††Follow-up: Median in months, (Range if possible); ‡‡Data not available on 24 patients. View Large Table 2 Studies of esophageal resection: treatment characteristics     Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775        Early stage‡ 0–2  Late stage 3–4  T stage 0–2  T stage 3–4    TT/TH/MI¶  Type of anastomosis  Follow-up  Study  Histopathology†  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  (obese vs nonobese)  Chemotherapy§  Oesophagectomy  (thoracic vs cervical)  period††  Scarpa et al.13  EAC (50%)  32 vs 110  29 vs 102  –  –  NAC: 130  278/0/0  NS  NS    SCC (47%)          AC: 50        Blom et al.19  EAC (74%)  31 vs 285  43 vs 374  –  –  NAC: 0  271/465/0  271 vs 465  NS    SCC (22%)          AC: NS        Melis et al.14  EAC (55%)  25 vs 36  134 vs 205  –  –  NAC: 273  ‡‡344/49/73  359 vs 107  25    SCC (45%)          AC: 97        Grotenhuis et al.22  EAC (85%)  –  –  23 vs 158  35 vs 340  NAC: NS  15/541/0  0 vs 556  NS    SCC (15%)          AC: NS        Madani et al.15  EAC (100%)  –  –  14 vs 21  42 vs 65  NAC: NS  77/65/0  0 vs 142  62 (15–137)              AC: NS        Kilic et al.20  EAC (81%)  65 vs 126  19 vs 72  –  –  NAC: 139  0/0/282  0 vs 282  NS    SCC (19%)          AC: NS        Healy et al.23  EAC (100%)  27 vs 61  9 vs 52  –  –  NAC: 81  146/4/0  131 vs 19  39              AC: NS        Scipione et al.21  EAC (91%)  75 vs 75  27 vs 27  –  –  NAC: NS  0/204/0  0 vs 204  NS    SCC (9%)          AC: NS        Total    55 vs 693  332 vs 832  37 vs 179  77 vs 405      761 vs 1775    †Histopathology: EAC, esophageal adenocarcinoma; SCC, squamous cell carcinoma; ‡Note: Staging information missing on 99 patients; §Perioperative oncological treatments: AC, adjuvant chemotherapy; NAC, neoadjuvant chemotherapy; NS, not stated; ¶TT, trans thoracic (including 2-stage, 3-stage and thoraco-abdominal approaches); MI, minimally invasive; TH, transhiatal; ††Follow-up: Median in months, (Range if possible); ‡‡Data not available on 24 patients. View Large This amounted to a comprehensive population of 2838 patients who were stratified into two groups according to WHO definition of obesity: 2199 (77%) patients were nonobese (BMI < 30 kg/m2) and 639 (23%) were obese (BMI ≥ 30 kg/m2). The results of the analyses on postoperative outcomes between the obese and nonobese groups are summarized in Table 3. Table 3 Summary of outcomes after esophagectomy (includes overall survival) Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  *Significant results; †Values were calculated with Fixed Effects MH. View Large Table 3 Summary of outcomes after esophagectomy (includes overall survival) Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  Outcomes  Number of studies included  Association with  RR (C.I.)  P-value  Any complications  6  None  0.972 (0.879, 1.074)  0.578  Anastomotic leak  7  Higher BMI†  0.657 (0.497, 0.867)  0.003*  Wound infections  5  None  1.047 (0.672, 1.632)  0.839  Chyle leak  4  None  1.701 (0.363, 7.957)  0.501  Vocal cord paresis  4  None  1.406 (0.746, 2.650)  0.292  Cardiac complications  5  None  1.018 (0.604, 1.716)  0.945  Pulmonary complications  8  None†  1.015 (0.775, 1.330)  0.913  Bleeding  3  None  0.759 (0.528, 1.090)  0.136  Reoperation  4  None  1.407 (0.874, 2.266)  0.161  Early mortality  8  None†  0.847 (0.501, 1.433)  0.536  Surgical radicality  5  None  1.021 (0.745, 1.399)  0.899  Five year overall survival  3  Higher BMI  1.172 (1.038, 1.323)  0.011*  *Significant results; †Values were calculated with Fixed Effects MH. View Large Analysis of surgical outcomes Information on the type of operation performed was available for the majority of the studies (Table 2). There were 1131 (40%) transthoracic esophagectomies (including 2-stage, 3-stage, and thoracoabdominal approach), 1328 (47%) transhiatal esophagectomies, and 355 (13%) minimally invasive procedures. Information on the anastomotic type was incomplete but where available, there were 761 (30%) thoracic anastomoses compared to 1775 (70%) cervical anastomoses (Table 2). Histopathological stage was documented in 6 studies (1912 patients). For early stage disease (Stage 0–2) 55 (7%) patients were obese and 693 (93%) were nonobese. In late stage disease (Stage 3–4) 332 (29%) patients were obese, and 832 (71%) nonobese. Both postoperative surgical and medical complications were analyzed. Most articles did not follow a standardized classification of complications. Only one study classified complications19 according to Clavien-Dindo.24 There was no significant association between all complications and BMI (Supplementary Fig. S1) (P = 0.578). The obese group was found to have a significantly increased incidence of anastomotic leak (RR = 0.657 [0.497, 0.867], P = 0.003) (Fig. 2). All other postoperative complications such as postoperative bleeding, early mortality, chyle leak, recurrent laryngeal nerve injury, re-operation, wound infection, cardiac, or respiratory complications did not show any significant association with obesity (Supplementary Figs S2–S9). Fig. 2 View largeDownload slide Anastomotic leak: There was a significant association between anastomotic leak and BMI ≥ 30 Subgroup 1 comprised studies with mixed histopathology (SCC and EAC). Subgroup 2 contained studies with EAC only. As subgroup 2 were small in number, this separate analysis was not significant. The total group is referred to in the results section. Fig. 2 View largeDownload slide Anastomotic leak: There was a significant association between anastomotic leak and BMI ≥ 30 Subgroup 1 comprised studies with mixed histopathology (SCC and EAC). Subgroup 2 contained studies with EAC only. As subgroup 2 were small in number, this separate analysis was not significant. The total group is referred to in the results section. Analysis of oncological outcomes The rate of surgical margin positivity after esophagectomy was compared in 5 studies and this was not significantly associated with BMI (RR = 1.021 [0.745, 1.399], P = 0.899) (Supplementary Fig. S10). Only three of the studies reported results on long-term survival. According to the data, the higher BMI group was significantly associated with a better 5 year overall survival (RR = 1.172 [1.038, 1.323], P = 0.011) (Fig. 3). Fig. 3 View largeDownload slide 5 Year Overall Survival: There was a significant association between 5-year overall survival and BMI ≥ 30. Fig. 3 View largeDownload slide 5 Year Overall Survival: There was a significant association between 5-year overall survival and BMI ≥ 30. DISCUSSION This meta-analysis, of 2838 patients undergoing esophagectomy, has evaluated perioperative morbidity and mortality in obese patients (BMI ≥ 30). Obesity was strongly associated with an anastomotic leak. The severity of anastomotic leak (ranging from subclinical, found on imaging to gastric necrosis requiring reoperation) is unclear from the 8 studies. A transhiatal esophagectomy with a cervical anastomosis was often performed in patients with SCC, whereas an Ivor Lewis esophagogastrectomy with a thoracic anastomosis was the operation of choice for junctional adenocarcinomas. The operation type may be the reason for this significant effect on anastomotic integrity. This study also demonstrates that the obese group had a better 5-year overall survival rate after esophageal resection for cancer. This result is not as easy to interpret and may reflect staging differences or less cachexia in the obese group. The study did not reveal any relationship between obesity and wound infection, cardiac complications, postoperative bleeding, or respiratory complications, which have been associated with obesity in other studies.9–11 Preoperative optimization and prophylactic measures such as prehabilitation, optimization of cardiac function and perioperative antibiotics in this group may explain this result. In the last 20 years, many authors have focused on obesity and its effect on surgical procedures.25-27 It is well established that patients with obesity have a higher incidence of comorbidities such as diabetes, cardiac and pulmonary conditions, as well as increased girth, which leads to technical difficulty during surgery, worse postoperative recovery and, therefore, an impact on health care expenditure. Studies pertaining to obesity and esophageal cancer resections have shown conflicting results. Some have described an association between obesity and severe complications including anastomotic leak,22 others have described no difference in complications between obese and nonobese in the postoperative period.21 Some have documented that patients with obesity show better disease free and overall survival,13-15 others discount this and show the same outcome for both groups.19-23 Eastern studies analyzing the influence of high BMI on patients undergoing esophagectomy use different classifications than the WHO definition, for example, Zogg et al. included patients with a BMI > 27.5 as obese.28 Indeed a WHO expert consultation has published recommendations on BMI in Asian populations indicating that a BMI > 23 represented increased risk and a BMI > 27.5 represented high risk.29 Eastern study populations can be quite different to Western series. Zhang et al. reported a series of 2031 patients, which consisted mainly of patients with SCC (87.4%) with an early pathological stage (mostly ≤ stage II).16 Our study sought to clarify the risk of obesity (BMI > 30) in Western studies. Our study population had a higher number of cervical anastomoses (1775) than thoracic (762). We found that obesity is a risk factor for a postoperative anastomotic leak. This could be explained by impaired vascularity of the gastric conduit, by microvascular disease and diabetes, which are more common in patients with obesity.19 Pierie et al. demonstrated, in 1994, that assessment of blood perfusion of the gastric tube is a significant predictor of anastomotic stricture rates and impaired healing in cervical anastomoses.30 A more recent paper has shown that BMI > 25 kg/m2 was significantly associated with development of a cervical stricture on univariate analysis.31 Increased tension on the conduit due to a cervical anastomosis may also contribute to this result. Neck dissection itself can be more challenging in a patient with obesity with a short neck.20-23 Long-term survival is an interesting outcome of this meta-analysis. Several studies previously reported no influence of obesity on overall survival32 while some considered obesity as a negative prognostic factor33-35 and others reported that a higher BMI could be a protective factor after surgery.14 In this study, the obese group had a significantly better 5-year overall survival even though the two groups were comparable in terms of surgical radicality. A number of factors may account for this apparent benefit. Obesity may confer a protective nutritional effect. Multiple hormonal, endocrine, or nutritional factors could contribute to the benefit of obesity in this situation.14,15 Grotenhuis et al. reported a reduced rate of circumferential margin involvement reflecting a greater volume of fat around the esophagus; this study however, did not show an overall difference in survival.22 A number of studies have reported an association of raised BMI with earlier stage disease.15,20,22,23,36 This may reflect higher rates of endoscopic surveillance in view of the higher rates of gastro-esophageal reflux symptoms in obese patients.37 This however was not borne out in this analysis as the patients with early stage disease comprised 292 (25%) obese and 872 (75%) nonobese. In the group with late stage disease the proportions were similar (409 (25%) were obese and 1237 (75%) were nonobese). The improved survival in obese patients may be a relative difference reflecting the poorer outcome with normal or low BMI in whom weight loss may have a greater effect.14,36 We excluded eastern studies to minimize heterogeneity between articles, specifically, regarding patient characteristics, and differences in terms of diagnostic and multimodal approach. The number of available western series that fit the inclusion criteria has limited this meta-analysis. Despite this, significant differences in operative technique (ranging from totally minimally invasive to open approach with cervical anastomosis) and histological characteristics (SCC and esophageal adenocarcinoma, EAC) persisted (Table 2). Another limitation, that previous meta-analyses have mentioned, is the influence of a phenomenon named ‘obesity paradox’: an intrinsic selection bias against the obese group due to multiple comorbidities leading to better mean performance status by more careful selection for surgery.38 As standardized classifications and definitions of all variables of interest were not used in these studies, data extraction was limited due to inconsistency in variables such as specific patient characteristics (comorbidities, ASA score or tumor location), other treatment data and oncological features (type of lymphadenectomy and surgical radicality). As some studies did not independently define underweight patients (BMI < 25) or super obese patients, these were included in the nonobese and obese group, respectively, and may have confounded the results. Many studies do not undertake to separate out these diverse groups. As both extremes can equally contribute to significant morbidity in their own right there is an inaccurate portrayal of the extent of the problem when groups are combined. Only one of the eight studies reported complications according to the Clavien-Dindo grade.24 This greatly impacts the quality of the studies and as a result the quality of the meta-analysis. The findings of this study have highlighted further areas to assess to improve outcome for this group of patients. The higher rate of anastomotic leak with a cervical anastomosis is well known but does seem to be a particular problem in the obese population. There is good evidence to support prehabilitation in terms of postoperative morbidity, length of stay, nutritional and physical status.39 A trial examining a prehabilitation program (PREHAB) in patients undergoing esophagogastric surgery is underway40 and it will be interesting to see if such an approach would be beneficial to the obese group. In summary, this meta-analysis demonstrates that in Western series of patients with esophageal cancer 23% of patients had obesity. This is associated with an increase in morbidity, specifically anastomotic leak but this did not translate into early mortality or a reduction in long term survival. While a raised BMI should not preclude multimodal treatment including surgical resection in Western patients, we would suggest it is important to mitigate risk by careful preoperative planning, optimization and a higher index of suspicion for anastomotic leak postoperatively. SUPPLEMENTARY DATA Supplementary data are available at DOTESO online. References 1 Clinical Oncology Group. Identification of risk factors for the development of complications following extended and superextended lymphadenectomies for gastric cancer. Br J Surg  2005; 92: 1103– 9. CrossRef Search ADS PubMed  2 Bamgbade O A, Rutter T W, Nafiu O O, Dorje P. Postoperative complications in obese and non-obese patients. World J Surg  2007; 31: 556– 60. Google Scholar CrossRef Search ADS PubMed  3 World Health Organization. Obesity: preventing and managing the global epidemic. Report of WHO consultation. World Health Organ Tech Rep Ser  2000; 894: 1– 253. 4 World Health Organization. World Health Statistics 2015. Retrieved on 13/12/2017 from http://www.who.int/gho/publications/world_health_statistics/EN_WHS2015_Part2.pdf?ua=1. 5 National Obesity Observatory (NHS), UK. Adult Obesity International Comparisons data factsheet. 2016. Retrieved 13/12/2017 from http://webarchive.nationalarchives.gov.uk/20170110170001/https://www.noo.org.uk/NOO_pub/Key_data. 6 Flegal K M, Carroll M D, Ogden C L, Curtin L R. Prevalence and trends in obesity among US adults, 1999–2008. JAMA  2010; 303: 235– 41. Google Scholar CrossRef Search ADS PubMed  7 Dority J, Hassan Z U, Chau D. Anesthetic implications of obesity in the surgical patient. Clin Colon Rectal Surg  2011; 24: 222– 8. Google Scholar CrossRef Search ADS PubMed  8 Dindo D, Muller M K, Weber M, Clavien P A. Obesity in general elective surgery. Lancet  2003; 361: 2032– 5. Google Scholar CrossRef Search ADS PubMed  9 Flancbaum L, Choban P S. Surgical implications of obesity. Annu Rev Med  1998; 49: 215– 34. Google Scholar CrossRef Search ADS PubMed  10 Mullen J T, Davenport D L, Hutter M M et al.   Impact of body mass index on perioperative outcomes in patients undergoing major intra-abdominal cancer surgery. Ann Surg Oncol  2008; 15: 2164– 72. Google Scholar CrossRef Search ADS PubMed  11 Tjeertes E K, Hocks S E, Beks S B et al.   Obesity-a risk factor for postoperative complications in general surgery? BMC Anaesthesia  2015; 15: 112. Google Scholar CrossRef Search ADS   12 Ryan A M, Duong M, Healy L et al.   Obesity, metabolic syndrome and esophageal adenocarcinoma: epidemiology, etiology and new targets. Cancer Epidemiol  2011; 35: 309– 19. Google Scholar CrossRef Search ADS PubMed  13 Scarpa M, Cagol M, Bettini S et al.   Overweight patients operated on for cancer of the esophagus survive longer than normal-weight patients. J Gastrointest Surg  2013; 17: 218– 27. Google Scholar CrossRef Search ADS PubMed  14 Melis M, Weber J M, McLoughlin J M et al.   An elevated body mass index does not reduce survival after esophagectomy for cancer. Ann Surg Oncol  2011; 18: 824– 31. Google Scholar CrossRef Search ADS PubMed  15 Madani K, Zhao R, Lim H J, Casson S M, Casson A G. Obesity is not associated with adverse outcome following surgical resection of oesophageal adenocarcinoma. Eur J Cardiothorac Surg  2010; 38: 604– 8. Google Scholar CrossRef Search ADS PubMed  16 Zhang S S, Yang H, Luo K J et al.   The impact of body mass index on complication and survival in resected oesophageal cancer: a clinical-based cohort and meta-analysis. Br J Cancer  2013; 109: 2894– 903. Google Scholar CrossRef Search ADS PubMed  17 Higgins J P T, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. www.cochrane-handbook.org. 18 Wallace B C, Dahabreh I J, Trikalinos T A, Lau J, Trow P, Schmid C H. Closing the gap between methodologists and end-users: R as a computational back-end. J Stat Software  2012; 5: 1– 15. 19 Blom R L, Lagarde S M, Klinkenbijl J H, Busch O R, van Berge Henegouwen M I. A high body mass index in esophageal cancer patients does not influence postoperative outcome or long-term survival. Ann Surg Oncol  2012; 19: 766– 71. Google Scholar CrossRef Search ADS PubMed  20 Kilic A, Schuchert M J, Pennathur A et al.   Impact of obesity on perioperative outcomes of minimally invasive esophagectomy. Ann Thorac Surg  2009; 87: 412– 5. Google Scholar CrossRef Search ADS PubMed  21 Scipione C N, Chang A C, Pickens A, Lau C L, Orringer M B. Transhiatal esophagectomy in the profoundly obese: implications and experience. Ann Thorac Surg  2007; 84: 376– 82. Google Scholar CrossRef Search ADS PubMed  22 Grotenhuis B A, Wijnhoven B P, Hötte G J, van der Stok E P, Tilanus H W, van Lanschot J J. Prognostic value of body mass index on short-term and long-term outcome after resection of esophageal cancer. World J Surg  2010; 34: 2621– 7. Google Scholar CrossRef Search ADS PubMed  23 Healy L A, Ryan A M, Gopinath B, Rowley S, Byrne P J, Reynolds J V. Impact of obesity on outcomes in the management of localized adenocarcinoma of the esophagus and esophagogastric junction. J Thorac Cardiovasc Surg  2007; 134: 1284– 91. Google Scholar CrossRef Search ADS PubMed  24 Clavien P A, Barkun J, de Oliveira M L et al.   The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg  2009; 250: 187– 96. Google Scholar CrossRef Search ADS PubMed  25 Wigfield C H, Lindsey J D, Munoz A, Chopra P S, Edwards N M, Love R B. Is extreme obesity a risk factor for cardiac surgery? An analysis of patients with a BMI > 40. Eur J Cardiothorac Surg  2006; 29: 434– 40. Google Scholar CrossRef Search ADS PubMed  26 Williams T K, Rosato E L, Kennedy E P et al.   Impact of obesity on perioperative morbidity and mortality after pancreaticoduodenectomy. J Am Coll Surg  2009; 208: 210– 7. Google Scholar CrossRef Search ADS PubMed  27 Duan X F, Tang P, Shang X B, Jiang H J, Zhao Q, Yu Z T. High body mass index worsens survival in patients with esophageal squamous cell carcinoma after esophagectomy. Dig Surg  2017; 34: 319– 27. Google Scholar CrossRef Search ADS PubMed  28 Zogg C K, Mungo B, Lidor A O et al.   Influence of body mass index on outcomes after major resection for cancer. Surgery  2015; 158: 472– 85. Google Scholar CrossRef Search ADS PubMed  29 WHO expert consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet  2004; 363: 157– 63. CrossRef Search ADS PubMed  30 Pierie J P, De Graaf P W, Poen H et al.   Impaired healing of cervical oesophagogastrostomies can be predicted by estimation of gastric serosal blood perfusion by laser Doppler flowmetry. Eur J Surg  1994; 160: 599– 603. Google Scholar PubMed  31 van Heijl M, Gooszen J A, Fockens P et al.   Risk factors for development of benign cervical strictures after esophagectomy. Annal Surg  2010; 251: 1064– 9. Google Scholar CrossRef Search ADS   32 Takemoto K, Shiozaki A, Fujiwara H et al.   High BMI does not influence short- and long-term outcomes of patients with esophageal cancer treated with esophagectomy. Gan To Kagaku Ryoho  2014; 41: 1991– 3. Google Scholar PubMed  33 Yoon H H, Lewis M A, Shi Q et al.   Prognostic impact of body mass index stratified by smoking status in patients with esophageal adenocarcinoma. J Clin Oncol  2011; 29: 4561– 7. Google Scholar CrossRef Search ADS PubMed  34 Watanabe M, Ishimoto T, Baba Y et al.   Prognostic impact of body mass index in patients with squamous cell carcinoma of the esophagus. Ann Surg Oncol  2013; 20: 3984– 91. Google Scholar CrossRef Search ADS PubMed  35 Cheng Y, Wang N, Wang K et al.   Prognostic value of body mass index for patients undergoing esophagectomy for esophageal squamous cell carcinoma. Jpn J Clin Oncol  2013; 43: 146– 53. Google Scholar CrossRef Search ADS PubMed  36 Hayashi Y, Correa A M Y, Hofstetter W L et al.   The influence of high body mass index on the prognosis of patients with esophageal cancer after surgery as primary therapy. Cancer  2010; 116: 5619– 27. Google Scholar CrossRef Search ADS PubMed  37 Tytgat G N, Van Sandick J W, Lanschot J J, Obertop H. Role of surveillance in intestinal metaplasia of the esophagus and gastroesophageal junction. World J Surg  2003; 27: 1021– 5. Google Scholar CrossRef Search ADS PubMed  38 STARSurg Collaborative. Multicenter prospective cohort study of body mass index and postoperative complications following gastrointestinal surgery. BJS  2016; 103: 1157– 72. CrossRef Search ADS   39 Valkenet K, van de Port I G, Dronkers J J, de Vries W R, Lindeman E, Backx F J. The effects of preoperative exercise therapy on postoperative outcome: a systematic review. Clin Rehabil  2011; 25: 99– 111. Google Scholar CrossRef Search ADS PubMed  40 Le Roy B, Pereira B, Bouteloup C et al.   Effect of prehabilitation in gastrooesophageal adenocarcinoma: study protocol of a multicentric, randomised, control trial-the PREHAB study. BMJ  2016; 6: e012876. Google Scholar CrossRef Search ADS   © The Author(s) 2017. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Diseases of the EsophagusOxford University Press

Published: Dec 22, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off