-
PDF
- Split View
-
Views
-
Cite
Cite
Sean L. Goh, Ramesh P. De Silva, Kumud Dhital, Rohan M. Gett, Is low serum albumin associated with postoperative complications in patients undergoing oesophagectomy for oesophageal malignancies?, Interactive CardioVascular and Thoracic Surgery, Volume 20, Issue 1, January 2015, Pages 107–113, https://doi.org/10.1093/icvts/ivu324
Close - Share Icon Share
Abstract
A best evidence topic was written according to a structured protocol. The question addressed was: in patients undergoing oesophagectomy for oesophageal malignancy, is low serum albumin associated with postoperative complications? Altogether, 87 papers were found using the reported search, of which 16 demonstrated the best evidence to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. This paper includes 2 level 2 papers, 12 level 3 papers and 2 level 4 papers. All the papers compared either all or some of the following postoperative complications: mortality, morbidity, anastomotic leak, respiratory and non-respiratory complications, and length of hospital stay. Eleven of the 16 papers found an association between low serum albumin and postoperative complications. Of these, one study showed that low serum albumin combined with low fibrinogen levels (FA score) was predictive of postoperative recurrence of oesophageal cancer. Another study showed that when combined with white cell count and C-reactive protein (CRP, NUn score), serum albumin had a high diagnostic accuracy for major complications after postoperative day 3. The largest study compared the in-hospital mortality in 7227 patients who underwent oesophageal surgery for malignancy. The percentage of in-hospital mortality was associated with low serum albumin (<15.0 vs >35.0 g/l, 21.0 vs 11.3%, P <0.001). Five of the 16 papers found no significant association between low serum albumin and postoperative complications. Of these papers, one showed that low serum albumin was not an independent risk factor, while four others found no association between low serum albumin with respiratory complications, anastomotic leak and postoperative mortality. Instead, these studies found other factors responsible for postoperative complications such as: CRP, smoking, disease duration, malnutrition and low T-cell levels. Taken together, while low serum albumin is associated with postoperative complications, opinion regarding the prognostic value of low serum albumin and nutritional support remains conflicted. Because of the confounding factors encountered in these studies, the clinician should consider the finding of low serum albumin in patients, together with disease and surgical factors to provide optimal care for these patients.
INTRODUCTION
A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].
THREE-PART QUESTION
In patients undergoing [oesophagectomy] for oesophageal cancer, is [low serum albumin] associated with [postoperative complications].
CLINICAL SCENARIO
You are running a preoperative assessment clinic and see a middle-aged male with T1N0M0 oesophageal cancer. His routine investigations were unremarkable except for a low serum albumin (<4.0 g/dl). You have recently read reports that low serum albumin may lead to postoperative complications. To provide optimal perioperative management, you turn to the literature and review the best evidence available.
SEARCH STRATEGY
Medline 1980 to July 2014 using OVID interface [exp albumins/ OR albumin.mp] AND [esophagectomy.mp OR exp Esophagectomy OR oesophagectomy.mp] AND [esophageal cancer.mp OR exp esophageal neoplasms OR oesophageal cancer.mp] AND [exp Post-operative Complications/ OR postoperative complications.mp OR complications.mp].
SEARCH OUTCOME
Eighty-seven papers were found using the reported search. From these, 16 were identified to have provided the best evidence to answer the question. These are presented in Table 1.
Table 1:
Best evidence papers
| Author, date, journal and country Study type (level of evidence) . | Patient group . | Outcomes . | Key results . | Comments . |
|---|---|---|---|---|
| Matsuda et al. (2014), J Clin Oncol, Japan [2] Retrospective single-centre cohort study (level 3) | 215 patients who underwent transthoracic oesophagectomy | Postoperative disease-free survival (DFS) | Univariate: low serum albumin (<4.0 g/dl) and fibrinogen (<350 mg/dl) associated with lower DFS Multivariate: low serum and fibrinogen levels predict postoperative DFS | Fibrinogen and albumin scores able to predict postoperative occurrence in oesophageal cancer patients |
| Yoshida et al. (2013), Surg Today, Japan [3] Retrospective single-centre cohort study (level 3) | 299 patients undergoing elective subtotal oesophagectomy Group 1: patients with respiratory complications, n = 53 Group 2: patients without respiratory complications, n = 246 | Respiratory complications | Serum albumin <4.0 g/dl group 1: group 2 = 25 : 129, P = 0.486 | No significant association was identified between low serum albumin preoperatively and incidence of postoperative pulmonary complications |
| Noble et al. (2012), J Gastrointest Surg, UK [4] Retrospective single-centre review (level 3) | 258 patients undergoing undergoing oesophagogastric resection for high-grade dysplasia | Major complications (Clavin–Dindo III–V) | Lower serum albumin in the major complication group from postoperative day 3 (P <0.008) and postoperative day 7 (P <0.0001) | This study examined the combined use of blood-borne markers including C-reactive protein (CRP), white cell count and albumin levels to develop a novel ‘NUn’ score to guide perioperative management and improve outcomes |
| Anastomotic leak | Serum albumin significantly low on multivariate analysis (P = 0.008) | |||
| Poziomyck et al. (2012), Nutr Cancer, Brazil [5] Prospective single-centre cohort study (level 3) | 74 patients undergoing foregut surgery for malignancy AJCC 2010 stage II or III: oesophagectomy (n = 19), gastrectomy (n = 43) and Whipple (n = 12) Group 1: patients who had postoperative complications and died, n = 50 Group 2: Patients who had postoperative complications and did not die, n = 24 | Mortality | A statistical significance was found for albumin levels (3.9 vs 3.5 g/dl, P <0.04) between patients who had postoperative complications and who lived or died, while other markers (haemoglobin, haematocrit and total lymphocyte count) did not | While malnutrition is a risk for adverse outcomes, new comparative studies are needed for nutritional assessment to elucidate an approach that better demonstrates nutritional risk of patients |
| Length of hospital stay | Length of hospital stay was inversely correlated with serum albumin (r = 0.277, P = 0.017) | |||
| Shiozaki et al. (2012), Oncol Lett, Japan [6] Retrospective single-centre cohort study (level 3) | 96 patients undergoing oesophagectomy (76 thoracotomy and 20 thoracoscopic surgery) Group 1: Presence of postoperative respiratory complications (n = 20) Group 2: Absence of postoperative complications (n = 76) | Postoperative respiratory complications: pneumonia and acute respiratory distress syndrome (ARDS) | The patients with and without postoperative complications did not differ significantly with respect to preoperative serum albumin levels Preoperative serum levels were not effective predictors of postoperative respiratory complications (multiple logistic regression analysis) | Forced expired volume in 1 second (FEV1.0%) serum CRP and smoking history were reliable predictors of the risk of respiratory complications following oesophageal resection compared with serum albumin levels |
| Khan et al. (2010), Saudi J Gastroenterol, Pakistan [7] Retrospective single-centre cohort study (level 3) | 284 patients undergoing subtotal oesophagectomy for malignant disease Group 1: patients who had undergone subtotal oesophagectomy and were alive on completion of 12 months (n = 185) Group 2: patients who had undergone subtotal oesophagectomy and died by the completion of 12 months (n = 99) | Mortality at 12 months | Preoperative serum albumin levels in Group 1 were higher compared with Group 2 (3.35 ± 0.49 vs 2.99 ± 0.51, P <0.05) | Preoperative low serum albumin levels have a strong predictive relation to mortality on patients who undergo oesophagectomy for oesophageal cancer |
| Marin et al. (2010), Arq Gastroenterol, Brazil [8] Retrospective single-centre cohort study (level 3) | 100 patients undergoing oesophagectomy for malignant disease Group 1: major surgery (oesphagectomy with gastroplasty by transhiatal or transthoracic route, n = 25) Group 2: minor/palliative surgery (gastrostomy or jejunostomy, n = 75) | Postoperative complications: pleuropulmonary of infectious/non-infectious origins, cardiovascular, surgical site infection, sepsis, anastomosis dehiscence, stenosis and tumour recurrence | Lower frequency of hypoalbuminaemia (>3.5 g/dl) was associated with lower weight loss and less postoperative complications (P = 0.041) Surgical mortality in minor surgery patients was higher in the presence of hypoalbuminaemia, with postoperative survival more than 30 days significantly higher in patients with serum albumin >3.5 g/dl (P = 0.041) | Authors conclude that disease severity (e.g. advanced stage) causes a worse nutritional state, which complicates surgery and is associated with mortality, and thus recommended early diagnosis and nutritional treatment to support surgical action |
| Park et al. (2009), Crit Care, UK [9] Retrospective multicentre review (level 4) | 7227 patients who were admitted to 181 critical care units following elective oesophageal surgery for malignancy | In-hospital mortality | Multiple logistic regression showed that percentage of in-hospital mortality was associated with low serum albumin (<15.0 vs >35.0 g/l, 21.0 vs 11.3%, P <0.001) Preoperative albumin levels did not predict postoperative levels or increased risk of death | Postoperative serum albumin was confirmed as an additional prognostic factor regarding the risk of death. More work is needed to determine how this may improve clinical management |
| Ryan et al. (2007), J Gastrointest Surg, USA [10] Retrospective single-centre review (level 4) | 200 patients undergoing oesophagectomy (thoracotomy with three- or two-stage exploration) and two-field lymphadenectomy for malignant disease | Morbidity and mortality | Patients with albumin <20 g/l were more likely to develop complications than those with albumin >20 g/l (54 vs 28%, respectively, P <0.011) Patients with albumin <20 g/l had higher rates of ARDS (22 vs 5%, P <0.001), respiratory failure (27 vs 8%, P <0.01), and were more likely to die in the hospital (27 vs 6%, P <0.001) than those with albumin >20 g/l Multivariate logistic regression analysis showed that day 1 albumin level was independently related to postoperative complications [odds ratio (OR) = 0.85; 95% confidence interval (CI) 0.85—0.77; P <0.001] | Authors suggest that albumin less than 20 g/l on the first postoperative day may identify a cohort postoperatively that should continue to be monitored in the high dependency unit (HDU) or the intensive care unit (ICU) |
| Tomimaru et al. (2006), J Surg Oncol, Japan [11] Retrospective single-centre cohort study (level 3) | 762 patients who underwent oesophagectomy (right thoracotomy and three- or two-field lymph node dissection) Group 1: salvage oesophagectomy after definitive chemoradiation (n = 24) Group 2: oesophagectomy after neoadjuvant chemoradiation (n = 26) | Morbidity and mortality | The serum albumin and lymphocyte count in the salvage group were significantly lower than those in the neoadjuvant group Univariate analysis performed for age (<60/<60), clinical T (T1–2/T3–4), reason for salvage surgery, curability, preoperative serum albumin (<4.0 /<4.0 g/dl) and lymphocyte did not identify preoperative serum albumin as an independent prognostic factor The lower preoperative serum albumin level in the salvage group may be due to a more malnourished and immunosuppressed condition than those in the neoadjuvant group | Because salvage oesophagectomy is performed on immunocompromised hosts, it therefore important to accurately assess the preoperative T factor, in addition to serum albumin levels, when deciding on the indication for salvage surgery |
| Kudsk et al. (2003), J Parenter Enteral Nutr, USA [12] Retrospective multicentre cohort study (level 3) | 706 patients undergoing elective surgery on the oesophagus (n = 59), stomach (n = 40) and pancreas (n = 221) | Incidence of complications: pneumonia, fasciitis, anastomosis or wound dehiscence, intra-abdominal abscess, renal or respiratory failure, decubitus ulcer formation or death | With albumin <3.75 g/dl (Groups 1–5), oesophageal surgery resulted in significantly more complications than stomach (P = 0.04) or colon (P = 0.005) Complications increased significantly as albumin levels dropped (Group 1 vs Group 2–7, P <0.02; 1–2 vs 3–7, 1–3 vs 4–7, 1–4 vs 5–7, 1–5 vs 6–7, all P <0.0001), 1–6 vs 7, P = 0.035) | Both preoperative serum albumin and operative site affect the complication rate, which subsequently influenced postoperative stay (POS), ICU stay and nil per oral (NPO) days |
| Mortality | Mortality was low (1–5%) with a preoperative albumin >3.25 g/dl, regardless of operative site. Mortality increased up to 20–30% in the two lowest albumin groups | |||
| Length of postoperative stay (POS) | Complications significantly increased POS in all groups except for the highest and lowest albumin groups (P <0.0001) | |||
| Intensive care unit stay | Mean ICU stay gradually increased with a decreasing preoperative albumin level | |||
| Nil per oral days | Delay to resumption of diet was prolonged in patients surviving oesophagectomy at all albumin levels <4.25 g/dl | |||
| Nakamura et al. (2004), Am J Surg [13], Japan Retrospective single-centre cohort study (level 3) | 55 patients undergoing oesophagectomy for oesophageal malignancy Salvage group: patients who underwent oesophagectomy after definitive chemoradiation (n = 27) Neoadjuvant group: patients who underwent oesophagectomy after neoadjuvant chemoradiation (n = 28) | Mortality, morbidity, need for mechanical ventilation, ICU stay, postoperative hospital stay, anastomotic leakage, wound infection, pleural effusion, residual tumours and pathological effect | No differences in overall postoperative survival/mortality between the salvage and neoadjuvant groups | Low serum albumin level in the salvage group could be related to the long duration of disease |
| Rentz et al. (2003), J Thorac Cardiovasc Surg, USA [14] Prospective multicentre cohort study (level 2) | 945 patients undergoing oesophagectomy. Group 1: transthoracic oesophagectomy (n = 562) Group 2: transhiatal oesophagectomy (n = 383) | Serious morbidity: pneumonia, unplanned intubation, respiratory failure, pulmonary oedema, cardiac arrest requiring cardiopulmonary resuscitation, myocardial infarction, wound dehiscence, coma for >24 h, acute renal failure, sepsis, perioperative bleeding >4 units and graft or prosthesis failure | Risk stratification model applied to both groups showed an association between serum albumin <3.5 g/dl with postoperative mortality (β coefficient: 0.56, P = 0.0135) and morbidity (β coefficient: 0.37, P = 0.0102) | Although no differences in postoperative mortality and morbidity were found between either techniques, identifiable risk factors, such as serum albumin levels, that might reduce mortality and morbidity may assist in the selection of the procedure best suited for the patient |
| Bailey et al. (2003), Ann Thorac Surg [15], USA Prospective multicentre outcomes cohort study (level 2) | 1777 patients who had undergone oesophagectomy for malignancy (n = 1509) and benign disease (n = 268) | Mortality and morbidity | Lower preoperative albumin level was an independent risk factor for overall morbidity on multivariable analysis (β coefficient = −0.21, SE = 0.09, P = 0.02, OR = 0.81) | Use of preoperative serum albumin levels, in addition to other independent risk factors, can be used to better stratify patients before oesophagectomy |
| Takagi et al. (2001), Nutrition, Japan [16] Retrospective single-centre cohort study (level 3) | 103 patients undergoing oesophagectomy for thoracic oesophageal cancer via thoracotomy and three-field lymphadenectomy Patients with complications, n = 27 (6 received preoperative chemoradiation) Patients without complications, n = 76 (2 received preoperative chemoradiation) | Infectious complications: pneumonia, abscess (neck, mediastinum and abdomen), enterocolitis, sepsis and viral infection | At 21 days postoperatively, serum albumin level was significantly higher in patients without postoperative infectious complications compared with those who had (3.55 ± 0.08 vs 3.20 ± 0.14 g/dl, P <0.05) | Although postoperative serum albumin levels may predict infectious complications, to prevent infectious complications, preoperative treatment that may adversely affect the host’s immune system, such as heavy chemoradiation, should be avoided |
| Nishi et al. (1988), Ann Surg, Japan [17] Retrospective single-centre cohort study (level 3) | 364 patients who underwent oesophagectomy, oesophagogastrectomy and/or partial gastrectomy 62 patients admitted for cholecystectomy were used as controls | Anastomotic leak and respiratory complications | There were significantly lower serum albumin levels in elderly patients (>70 years old) compared with controls An extremely high correlation was seen between serum albumin levels and total lymphocyte count in patients with oesophageal cancer (r = 0.56, P <0.001) | Although serum albumin levels did not correlate with the study outcomes, improvement in the nutritional status of patients with cancer can reduce operative morbidity and mortality rates. The authors suggested that cancer patients required between 40 and 50 kcal/kg/day with simultaneous administration of 150–200 non-protein calories/g nitrogen |
| Author, date, journal and country Study type (level of evidence) . | Patient group . | Outcomes . | Key results . | Comments . |
|---|---|---|---|---|
| Matsuda et al. (2014), J Clin Oncol, Japan [2] Retrospective single-centre cohort study (level 3) | 215 patients who underwent transthoracic oesophagectomy | Postoperative disease-free survival (DFS) | Univariate: low serum albumin (<4.0 g/dl) and fibrinogen (<350 mg/dl) associated with lower DFS Multivariate: low serum and fibrinogen levels predict postoperative DFS | Fibrinogen and albumin scores able to predict postoperative occurrence in oesophageal cancer patients |
| Yoshida et al. (2013), Surg Today, Japan [3] Retrospective single-centre cohort study (level 3) | 299 patients undergoing elective subtotal oesophagectomy Group 1: patients with respiratory complications, n = 53 Group 2: patients without respiratory complications, n = 246 | Respiratory complications | Serum albumin <4.0 g/dl group 1: group 2 = 25 : 129, P = 0.486 | No significant association was identified between low serum albumin preoperatively and incidence of postoperative pulmonary complications |
| Noble et al. (2012), J Gastrointest Surg, UK [4] Retrospective single-centre review (level 3) | 258 patients undergoing undergoing oesophagogastric resection for high-grade dysplasia | Major complications (Clavin–Dindo III–V) | Lower serum albumin in the major complication group from postoperative day 3 (P <0.008) and postoperative day 7 (P <0.0001) | This study examined the combined use of blood-borne markers including C-reactive protein (CRP), white cell count and albumin levels to develop a novel ‘NUn’ score to guide perioperative management and improve outcomes |
| Anastomotic leak | Serum albumin significantly low on multivariate analysis (P = 0.008) | |||
| Poziomyck et al. (2012), Nutr Cancer, Brazil [5] Prospective single-centre cohort study (level 3) | 74 patients undergoing foregut surgery for malignancy AJCC 2010 stage II or III: oesophagectomy (n = 19), gastrectomy (n = 43) and Whipple (n = 12) Group 1: patients who had postoperative complications and died, n = 50 Group 2: Patients who had postoperative complications and did not die, n = 24 | Mortality | A statistical significance was found for albumin levels (3.9 vs 3.5 g/dl, P <0.04) between patients who had postoperative complications and who lived or died, while other markers (haemoglobin, haematocrit and total lymphocyte count) did not | While malnutrition is a risk for adverse outcomes, new comparative studies are needed for nutritional assessment to elucidate an approach that better demonstrates nutritional risk of patients |
| Length of hospital stay | Length of hospital stay was inversely correlated with serum albumin (r = 0.277, P = 0.017) | |||
| Shiozaki et al. (2012), Oncol Lett, Japan [6] Retrospective single-centre cohort study (level 3) | 96 patients undergoing oesophagectomy (76 thoracotomy and 20 thoracoscopic surgery) Group 1: Presence of postoperative respiratory complications (n = 20) Group 2: Absence of postoperative complications (n = 76) | Postoperative respiratory complications: pneumonia and acute respiratory distress syndrome (ARDS) | The patients with and without postoperative complications did not differ significantly with respect to preoperative serum albumin levels Preoperative serum levels were not effective predictors of postoperative respiratory complications (multiple logistic regression analysis) | Forced expired volume in 1 second (FEV1.0%) serum CRP and smoking history were reliable predictors of the risk of respiratory complications following oesophageal resection compared with serum albumin levels |
| Khan et al. (2010), Saudi J Gastroenterol, Pakistan [7] Retrospective single-centre cohort study (level 3) | 284 patients undergoing subtotal oesophagectomy for malignant disease Group 1: patients who had undergone subtotal oesophagectomy and were alive on completion of 12 months (n = 185) Group 2: patients who had undergone subtotal oesophagectomy and died by the completion of 12 months (n = 99) | Mortality at 12 months | Preoperative serum albumin levels in Group 1 were higher compared with Group 2 (3.35 ± 0.49 vs 2.99 ± 0.51, P <0.05) | Preoperative low serum albumin levels have a strong predictive relation to mortality on patients who undergo oesophagectomy for oesophageal cancer |
| Marin et al. (2010), Arq Gastroenterol, Brazil [8] Retrospective single-centre cohort study (level 3) | 100 patients undergoing oesophagectomy for malignant disease Group 1: major surgery (oesphagectomy with gastroplasty by transhiatal or transthoracic route, n = 25) Group 2: minor/palliative surgery (gastrostomy or jejunostomy, n = 75) | Postoperative complications: pleuropulmonary of infectious/non-infectious origins, cardiovascular, surgical site infection, sepsis, anastomosis dehiscence, stenosis and tumour recurrence | Lower frequency of hypoalbuminaemia (>3.5 g/dl) was associated with lower weight loss and less postoperative complications (P = 0.041) Surgical mortality in minor surgery patients was higher in the presence of hypoalbuminaemia, with postoperative survival more than 30 days significantly higher in patients with serum albumin >3.5 g/dl (P = 0.041) | Authors conclude that disease severity (e.g. advanced stage) causes a worse nutritional state, which complicates surgery and is associated with mortality, and thus recommended early diagnosis and nutritional treatment to support surgical action |
| Park et al. (2009), Crit Care, UK [9] Retrospective multicentre review (level 4) | 7227 patients who were admitted to 181 critical care units following elective oesophageal surgery for malignancy | In-hospital mortality | Multiple logistic regression showed that percentage of in-hospital mortality was associated with low serum albumin (<15.0 vs >35.0 g/l, 21.0 vs 11.3%, P <0.001) Preoperative albumin levels did not predict postoperative levels or increased risk of death | Postoperative serum albumin was confirmed as an additional prognostic factor regarding the risk of death. More work is needed to determine how this may improve clinical management |
| Ryan et al. (2007), J Gastrointest Surg, USA [10] Retrospective single-centre review (level 4) | 200 patients undergoing oesophagectomy (thoracotomy with three- or two-stage exploration) and two-field lymphadenectomy for malignant disease | Morbidity and mortality | Patients with albumin <20 g/l were more likely to develop complications than those with albumin >20 g/l (54 vs 28%, respectively, P <0.011) Patients with albumin <20 g/l had higher rates of ARDS (22 vs 5%, P <0.001), respiratory failure (27 vs 8%, P <0.01), and were more likely to die in the hospital (27 vs 6%, P <0.001) than those with albumin >20 g/l Multivariate logistic regression analysis showed that day 1 albumin level was independently related to postoperative complications [odds ratio (OR) = 0.85; 95% confidence interval (CI) 0.85—0.77; P <0.001] | Authors suggest that albumin less than 20 g/l on the first postoperative day may identify a cohort postoperatively that should continue to be monitored in the high dependency unit (HDU) or the intensive care unit (ICU) |
| Tomimaru et al. (2006), J Surg Oncol, Japan [11] Retrospective single-centre cohort study (level 3) | 762 patients who underwent oesophagectomy (right thoracotomy and three- or two-field lymph node dissection) Group 1: salvage oesophagectomy after definitive chemoradiation (n = 24) Group 2: oesophagectomy after neoadjuvant chemoradiation (n = 26) | Morbidity and mortality | The serum albumin and lymphocyte count in the salvage group were significantly lower than those in the neoadjuvant group Univariate analysis performed for age (<60/<60), clinical T (T1–2/T3–4), reason for salvage surgery, curability, preoperative serum albumin (<4.0 /<4.0 g/dl) and lymphocyte did not identify preoperative serum albumin as an independent prognostic factor The lower preoperative serum albumin level in the salvage group may be due to a more malnourished and immunosuppressed condition than those in the neoadjuvant group | Because salvage oesophagectomy is performed on immunocompromised hosts, it therefore important to accurately assess the preoperative T factor, in addition to serum albumin levels, when deciding on the indication for salvage surgery |
| Kudsk et al. (2003), J Parenter Enteral Nutr, USA [12] Retrospective multicentre cohort study (level 3) | 706 patients undergoing elective surgery on the oesophagus (n = 59), stomach (n = 40) and pancreas (n = 221) | Incidence of complications: pneumonia, fasciitis, anastomosis or wound dehiscence, intra-abdominal abscess, renal or respiratory failure, decubitus ulcer formation or death | With albumin <3.75 g/dl (Groups 1–5), oesophageal surgery resulted in significantly more complications than stomach (P = 0.04) or colon (P = 0.005) Complications increased significantly as albumin levels dropped (Group 1 vs Group 2–7, P <0.02; 1–2 vs 3–7, 1–3 vs 4–7, 1–4 vs 5–7, 1–5 vs 6–7, all P <0.0001), 1–6 vs 7, P = 0.035) | Both preoperative serum albumin and operative site affect the complication rate, which subsequently influenced postoperative stay (POS), ICU stay and nil per oral (NPO) days |
| Mortality | Mortality was low (1–5%) with a preoperative albumin >3.25 g/dl, regardless of operative site. Mortality increased up to 20–30% in the two lowest albumin groups | |||
| Length of postoperative stay (POS) | Complications significantly increased POS in all groups except for the highest and lowest albumin groups (P <0.0001) | |||
| Intensive care unit stay | Mean ICU stay gradually increased with a decreasing preoperative albumin level | |||
| Nil per oral days | Delay to resumption of diet was prolonged in patients surviving oesophagectomy at all albumin levels <4.25 g/dl | |||
| Nakamura et al. (2004), Am J Surg [13], Japan Retrospective single-centre cohort study (level 3) | 55 patients undergoing oesophagectomy for oesophageal malignancy Salvage group: patients who underwent oesophagectomy after definitive chemoradiation (n = 27) Neoadjuvant group: patients who underwent oesophagectomy after neoadjuvant chemoradiation (n = 28) | Mortality, morbidity, need for mechanical ventilation, ICU stay, postoperative hospital stay, anastomotic leakage, wound infection, pleural effusion, residual tumours and pathological effect | No differences in overall postoperative survival/mortality between the salvage and neoadjuvant groups | Low serum albumin level in the salvage group could be related to the long duration of disease |
| Rentz et al. (2003), J Thorac Cardiovasc Surg, USA [14] Prospective multicentre cohort study (level 2) | 945 patients undergoing oesophagectomy. Group 1: transthoracic oesophagectomy (n = 562) Group 2: transhiatal oesophagectomy (n = 383) | Serious morbidity: pneumonia, unplanned intubation, respiratory failure, pulmonary oedema, cardiac arrest requiring cardiopulmonary resuscitation, myocardial infarction, wound dehiscence, coma for >24 h, acute renal failure, sepsis, perioperative bleeding >4 units and graft or prosthesis failure | Risk stratification model applied to both groups showed an association between serum albumin <3.5 g/dl with postoperative mortality (β coefficient: 0.56, P = 0.0135) and morbidity (β coefficient: 0.37, P = 0.0102) | Although no differences in postoperative mortality and morbidity were found between either techniques, identifiable risk factors, such as serum albumin levels, that might reduce mortality and morbidity may assist in the selection of the procedure best suited for the patient |
| Bailey et al. (2003), Ann Thorac Surg [15], USA Prospective multicentre outcomes cohort study (level 2) | 1777 patients who had undergone oesophagectomy for malignancy (n = 1509) and benign disease (n = 268) | Mortality and morbidity | Lower preoperative albumin level was an independent risk factor for overall morbidity on multivariable analysis (β coefficient = −0.21, SE = 0.09, P = 0.02, OR = 0.81) | Use of preoperative serum albumin levels, in addition to other independent risk factors, can be used to better stratify patients before oesophagectomy |
| Takagi et al. (2001), Nutrition, Japan [16] Retrospective single-centre cohort study (level 3) | 103 patients undergoing oesophagectomy for thoracic oesophageal cancer via thoracotomy and three-field lymphadenectomy Patients with complications, n = 27 (6 received preoperative chemoradiation) Patients without complications, n = 76 (2 received preoperative chemoradiation) | Infectious complications: pneumonia, abscess (neck, mediastinum and abdomen), enterocolitis, sepsis and viral infection | At 21 days postoperatively, serum albumin level was significantly higher in patients without postoperative infectious complications compared with those who had (3.55 ± 0.08 vs 3.20 ± 0.14 g/dl, P <0.05) | Although postoperative serum albumin levels may predict infectious complications, to prevent infectious complications, preoperative treatment that may adversely affect the host’s immune system, such as heavy chemoradiation, should be avoided |
| Nishi et al. (1988), Ann Surg, Japan [17] Retrospective single-centre cohort study (level 3) | 364 patients who underwent oesophagectomy, oesophagogastrectomy and/or partial gastrectomy 62 patients admitted for cholecystectomy were used as controls | Anastomotic leak and respiratory complications | There were significantly lower serum albumin levels in elderly patients (>70 years old) compared with controls An extremely high correlation was seen between serum albumin levels and total lymphocyte count in patients with oesophageal cancer (r = 0.56, P <0.001) | Although serum albumin levels did not correlate with the study outcomes, improvement in the nutritional status of patients with cancer can reduce operative morbidity and mortality rates. The authors suggested that cancer patients required between 40 and 50 kcal/kg/day with simultaneous administration of 150–200 non-protein calories/g nitrogen |
Table 1:
Best evidence papers
| Author, date, journal and country Study type (level of evidence) . | Patient group . | Outcomes . | Key results . | Comments . |
|---|---|---|---|---|
| Matsuda et al. (2014), J Clin Oncol, Japan [2] Retrospective single-centre cohort study (level 3) | 215 patients who underwent transthoracic oesophagectomy | Postoperative disease-free survival (DFS) | Univariate: low serum albumin (<4.0 g/dl) and fibrinogen (<350 mg/dl) associated with lower DFS Multivariate: low serum and fibrinogen levels predict postoperative DFS | Fibrinogen and albumin scores able to predict postoperative occurrence in oesophageal cancer patients |
| Yoshida et al. (2013), Surg Today, Japan [3] Retrospective single-centre cohort study (level 3) | 299 patients undergoing elective subtotal oesophagectomy Group 1: patients with respiratory complications, n = 53 Group 2: patients without respiratory complications, n = 246 | Respiratory complications | Serum albumin <4.0 g/dl group 1: group 2 = 25 : 129, P = 0.486 | No significant association was identified between low serum albumin preoperatively and incidence of postoperative pulmonary complications |
| Noble et al. (2012), J Gastrointest Surg, UK [4] Retrospective single-centre review (level 3) | 258 patients undergoing undergoing oesophagogastric resection for high-grade dysplasia | Major complications (Clavin–Dindo III–V) | Lower serum albumin in the major complication group from postoperative day 3 (P <0.008) and postoperative day 7 (P <0.0001) | This study examined the combined use of blood-borne markers including C-reactive protein (CRP), white cell count and albumin levels to develop a novel ‘NUn’ score to guide perioperative management and improve outcomes |
| Anastomotic leak | Serum albumin significantly low on multivariate analysis (P = 0.008) | |||
| Poziomyck et al. (2012), Nutr Cancer, Brazil [5] Prospective single-centre cohort study (level 3) | 74 patients undergoing foregut surgery for malignancy AJCC 2010 stage II or III: oesophagectomy (n = 19), gastrectomy (n = 43) and Whipple (n = 12) Group 1: patients who had postoperative complications and died, n = 50 Group 2: Patients who had postoperative complications and did not die, n = 24 | Mortality | A statistical significance was found for albumin levels (3.9 vs 3.5 g/dl, P <0.04) between patients who had postoperative complications and who lived or died, while other markers (haemoglobin, haematocrit and total lymphocyte count) did not | While malnutrition is a risk for adverse outcomes, new comparative studies are needed for nutritional assessment to elucidate an approach that better demonstrates nutritional risk of patients |
| Length of hospital stay | Length of hospital stay was inversely correlated with serum albumin (r = 0.277, P = 0.017) | |||
| Shiozaki et al. (2012), Oncol Lett, Japan [6] Retrospective single-centre cohort study (level 3) | 96 patients undergoing oesophagectomy (76 thoracotomy and 20 thoracoscopic surgery) Group 1: Presence of postoperative respiratory complications (n = 20) Group 2: Absence of postoperative complications (n = 76) | Postoperative respiratory complications: pneumonia and acute respiratory distress syndrome (ARDS) | The patients with and without postoperative complications did not differ significantly with respect to preoperative serum albumin levels Preoperative serum levels were not effective predictors of postoperative respiratory complications (multiple logistic regression analysis) | Forced expired volume in 1 second (FEV1.0%) serum CRP and smoking history were reliable predictors of the risk of respiratory complications following oesophageal resection compared with serum albumin levels |
| Khan et al. (2010), Saudi J Gastroenterol, Pakistan [7] Retrospective single-centre cohort study (level 3) | 284 patients undergoing subtotal oesophagectomy for malignant disease Group 1: patients who had undergone subtotal oesophagectomy and were alive on completion of 12 months (n = 185) Group 2: patients who had undergone subtotal oesophagectomy and died by the completion of 12 months (n = 99) | Mortality at 12 months | Preoperative serum albumin levels in Group 1 were higher compared with Group 2 (3.35 ± 0.49 vs 2.99 ± 0.51, P <0.05) | Preoperative low serum albumin levels have a strong predictive relation to mortality on patients who undergo oesophagectomy for oesophageal cancer |
| Marin et al. (2010), Arq Gastroenterol, Brazil [8] Retrospective single-centre cohort study (level 3) | 100 patients undergoing oesophagectomy for malignant disease Group 1: major surgery (oesphagectomy with gastroplasty by transhiatal or transthoracic route, n = 25) Group 2: minor/palliative surgery (gastrostomy or jejunostomy, n = 75) | Postoperative complications: pleuropulmonary of infectious/non-infectious origins, cardiovascular, surgical site infection, sepsis, anastomosis dehiscence, stenosis and tumour recurrence | Lower frequency of hypoalbuminaemia (>3.5 g/dl) was associated with lower weight loss and less postoperative complications (P = 0.041) Surgical mortality in minor surgery patients was higher in the presence of hypoalbuminaemia, with postoperative survival more than 30 days significantly higher in patients with serum albumin >3.5 g/dl (P = 0.041) | Authors conclude that disease severity (e.g. advanced stage) causes a worse nutritional state, which complicates surgery and is associated with mortality, and thus recommended early diagnosis and nutritional treatment to support surgical action |
| Park et al. (2009), Crit Care, UK [9] Retrospective multicentre review (level 4) | 7227 patients who were admitted to 181 critical care units following elective oesophageal surgery for malignancy | In-hospital mortality | Multiple logistic regression showed that percentage of in-hospital mortality was associated with low serum albumin (<15.0 vs >35.0 g/l, 21.0 vs 11.3%, P <0.001) Preoperative albumin levels did not predict postoperative levels or increased risk of death | Postoperative serum albumin was confirmed as an additional prognostic factor regarding the risk of death. More work is needed to determine how this may improve clinical management |
| Ryan et al. (2007), J Gastrointest Surg, USA [10] Retrospective single-centre review (level 4) | 200 patients undergoing oesophagectomy (thoracotomy with three- or two-stage exploration) and two-field lymphadenectomy for malignant disease | Morbidity and mortality | Patients with albumin <20 g/l were more likely to develop complications than those with albumin >20 g/l (54 vs 28%, respectively, P <0.011) Patients with albumin <20 g/l had higher rates of ARDS (22 vs 5%, P <0.001), respiratory failure (27 vs 8%, P <0.01), and were more likely to die in the hospital (27 vs 6%, P <0.001) than those with albumin >20 g/l Multivariate logistic regression analysis showed that day 1 albumin level was independently related to postoperative complications [odds ratio (OR) = 0.85; 95% confidence interval (CI) 0.85—0.77; P <0.001] | Authors suggest that albumin less than 20 g/l on the first postoperative day may identify a cohort postoperatively that should continue to be monitored in the high dependency unit (HDU) or the intensive care unit (ICU) |
| Tomimaru et al. (2006), J Surg Oncol, Japan [11] Retrospective single-centre cohort study (level 3) | 762 patients who underwent oesophagectomy (right thoracotomy and three- or two-field lymph node dissection) Group 1: salvage oesophagectomy after definitive chemoradiation (n = 24) Group 2: oesophagectomy after neoadjuvant chemoradiation (n = 26) | Morbidity and mortality | The serum albumin and lymphocyte count in the salvage group were significantly lower than those in the neoadjuvant group Univariate analysis performed for age (<60/<60), clinical T (T1–2/T3–4), reason for salvage surgery, curability, preoperative serum albumin (<4.0 /<4.0 g/dl) and lymphocyte did not identify preoperative serum albumin as an independent prognostic factor The lower preoperative serum albumin level in the salvage group may be due to a more malnourished and immunosuppressed condition than those in the neoadjuvant group | Because salvage oesophagectomy is performed on immunocompromised hosts, it therefore important to accurately assess the preoperative T factor, in addition to serum albumin levels, when deciding on the indication for salvage surgery |
| Kudsk et al. (2003), J Parenter Enteral Nutr, USA [12] Retrospective multicentre cohort study (level 3) | 706 patients undergoing elective surgery on the oesophagus (n = 59), stomach (n = 40) and pancreas (n = 221) | Incidence of complications: pneumonia, fasciitis, anastomosis or wound dehiscence, intra-abdominal abscess, renal or respiratory failure, decubitus ulcer formation or death | With albumin <3.75 g/dl (Groups 1–5), oesophageal surgery resulted in significantly more complications than stomach (P = 0.04) or colon (P = 0.005) Complications increased significantly as albumin levels dropped (Group 1 vs Group 2–7, P <0.02; 1–2 vs 3–7, 1–3 vs 4–7, 1–4 vs 5–7, 1–5 vs 6–7, all P <0.0001), 1–6 vs 7, P = 0.035) | Both preoperative serum albumin and operative site affect the complication rate, which subsequently influenced postoperative stay (POS), ICU stay and nil per oral (NPO) days |
| Mortality | Mortality was low (1–5%) with a preoperative albumin >3.25 g/dl, regardless of operative site. Mortality increased up to 20–30% in the two lowest albumin groups | |||
| Length of postoperative stay (POS) | Complications significantly increased POS in all groups except for the highest and lowest albumin groups (P <0.0001) | |||
| Intensive care unit stay | Mean ICU stay gradually increased with a decreasing preoperative albumin level | |||
| Nil per oral days | Delay to resumption of diet was prolonged in patients surviving oesophagectomy at all albumin levels <4.25 g/dl | |||
| Nakamura et al. (2004), Am J Surg [13], Japan Retrospective single-centre cohort study (level 3) | 55 patients undergoing oesophagectomy for oesophageal malignancy Salvage group: patients who underwent oesophagectomy after definitive chemoradiation (n = 27) Neoadjuvant group: patients who underwent oesophagectomy after neoadjuvant chemoradiation (n = 28) | Mortality, morbidity, need for mechanical ventilation, ICU stay, postoperative hospital stay, anastomotic leakage, wound infection, pleural effusion, residual tumours and pathological effect | No differences in overall postoperative survival/mortality between the salvage and neoadjuvant groups | Low serum albumin level in the salvage group could be related to the long duration of disease |
| Rentz et al. (2003), J Thorac Cardiovasc Surg, USA [14] Prospective multicentre cohort study (level 2) | 945 patients undergoing oesophagectomy. Group 1: transthoracic oesophagectomy (n = 562) Group 2: transhiatal oesophagectomy (n = 383) | Serious morbidity: pneumonia, unplanned intubation, respiratory failure, pulmonary oedema, cardiac arrest requiring cardiopulmonary resuscitation, myocardial infarction, wound dehiscence, coma for >24 h, acute renal failure, sepsis, perioperative bleeding >4 units and graft or prosthesis failure | Risk stratification model applied to both groups showed an association between serum albumin <3.5 g/dl with postoperative mortality (β coefficient: 0.56, P = 0.0135) and morbidity (β coefficient: 0.37, P = 0.0102) | Although no differences in postoperative mortality and morbidity were found between either techniques, identifiable risk factors, such as serum albumin levels, that might reduce mortality and morbidity may assist in the selection of the procedure best suited for the patient |
| Bailey et al. (2003), Ann Thorac Surg [15], USA Prospective multicentre outcomes cohort study (level 2) | 1777 patients who had undergone oesophagectomy for malignancy (n = 1509) and benign disease (n = 268) | Mortality and morbidity | Lower preoperative albumin level was an independent risk factor for overall morbidity on multivariable analysis (β coefficient = −0.21, SE = 0.09, P = 0.02, OR = 0.81) | Use of preoperative serum albumin levels, in addition to other independent risk factors, can be used to better stratify patients before oesophagectomy |
| Takagi et al. (2001), Nutrition, Japan [16] Retrospective single-centre cohort study (level 3) | 103 patients undergoing oesophagectomy for thoracic oesophageal cancer via thoracotomy and three-field lymphadenectomy Patients with complications, n = 27 (6 received preoperative chemoradiation) Patients without complications, n = 76 (2 received preoperative chemoradiation) | Infectious complications: pneumonia, abscess (neck, mediastinum and abdomen), enterocolitis, sepsis and viral infection | At 21 days postoperatively, serum albumin level was significantly higher in patients without postoperative infectious complications compared with those who had (3.55 ± 0.08 vs 3.20 ± 0.14 g/dl, P <0.05) | Although postoperative serum albumin levels may predict infectious complications, to prevent infectious complications, preoperative treatment that may adversely affect the host’s immune system, such as heavy chemoradiation, should be avoided |
| Nishi et al. (1988), Ann Surg, Japan [17] Retrospective single-centre cohort study (level 3) | 364 patients who underwent oesophagectomy, oesophagogastrectomy and/or partial gastrectomy 62 patients admitted for cholecystectomy were used as controls | Anastomotic leak and respiratory complications | There were significantly lower serum albumin levels in elderly patients (>70 years old) compared with controls An extremely high correlation was seen between serum albumin levels and total lymphocyte count in patients with oesophageal cancer (r = 0.56, P <0.001) | Although serum albumin levels did not correlate with the study outcomes, improvement in the nutritional status of patients with cancer can reduce operative morbidity and mortality rates. The authors suggested that cancer patients required between 40 and 50 kcal/kg/day with simultaneous administration of 150–200 non-protein calories/g nitrogen |
| Author, date, journal and country Study type (level of evidence) . | Patient group . | Outcomes . | Key results . | Comments . |
|---|---|---|---|---|
| Matsuda et al. (2014), J Clin Oncol, Japan [2] Retrospective single-centre cohort study (level 3) | 215 patients who underwent transthoracic oesophagectomy | Postoperative disease-free survival (DFS) | Univariate: low serum albumin (<4.0 g/dl) and fibrinogen (<350 mg/dl) associated with lower DFS Multivariate: low serum and fibrinogen levels predict postoperative DFS | Fibrinogen and albumin scores able to predict postoperative occurrence in oesophageal cancer patients |
| Yoshida et al. (2013), Surg Today, Japan [3] Retrospective single-centre cohort study (level 3) | 299 patients undergoing elective subtotal oesophagectomy Group 1: patients with respiratory complications, n = 53 Group 2: patients without respiratory complications, n = 246 | Respiratory complications | Serum albumin <4.0 g/dl group 1: group 2 = 25 : 129, P = 0.486 | No significant association was identified between low serum albumin preoperatively and incidence of postoperative pulmonary complications |
| Noble et al. (2012), J Gastrointest Surg, UK [4] Retrospective single-centre review (level 3) | 258 patients undergoing undergoing oesophagogastric resection for high-grade dysplasia | Major complications (Clavin–Dindo III–V) | Lower serum albumin in the major complication group from postoperative day 3 (P <0.008) and postoperative day 7 (P <0.0001) | This study examined the combined use of blood-borne markers including C-reactive protein (CRP), white cell count and albumin levels to develop a novel ‘NUn’ score to guide perioperative management and improve outcomes |
| Anastomotic leak | Serum albumin significantly low on multivariate analysis (P = 0.008) | |||
| Poziomyck et al. (2012), Nutr Cancer, Brazil [5] Prospective single-centre cohort study (level 3) | 74 patients undergoing foregut surgery for malignancy AJCC 2010 stage II or III: oesophagectomy (n = 19), gastrectomy (n = 43) and Whipple (n = 12) Group 1: patients who had postoperative complications and died, n = 50 Group 2: Patients who had postoperative complications and did not die, n = 24 | Mortality | A statistical significance was found for albumin levels (3.9 vs 3.5 g/dl, P <0.04) between patients who had postoperative complications and who lived or died, while other markers (haemoglobin, haematocrit and total lymphocyte count) did not | While malnutrition is a risk for adverse outcomes, new comparative studies are needed for nutritional assessment to elucidate an approach that better demonstrates nutritional risk of patients |
| Length of hospital stay | Length of hospital stay was inversely correlated with serum albumin (r = 0.277, P = 0.017) | |||
| Shiozaki et al. (2012), Oncol Lett, Japan [6] Retrospective single-centre cohort study (level 3) | 96 patients undergoing oesophagectomy (76 thoracotomy and 20 thoracoscopic surgery) Group 1: Presence of postoperative respiratory complications (n = 20) Group 2: Absence of postoperative complications (n = 76) | Postoperative respiratory complications: pneumonia and acute respiratory distress syndrome (ARDS) | The patients with and without postoperative complications did not differ significantly with respect to preoperative serum albumin levels Preoperative serum levels were not effective predictors of postoperative respiratory complications (multiple logistic regression analysis) | Forced expired volume in 1 second (FEV1.0%) serum CRP and smoking history were reliable predictors of the risk of respiratory complications following oesophageal resection compared with serum albumin levels |
| Khan et al. (2010), Saudi J Gastroenterol, Pakistan [7] Retrospective single-centre cohort study (level 3) | 284 patients undergoing subtotal oesophagectomy for malignant disease Group 1: patients who had undergone subtotal oesophagectomy and were alive on completion of 12 months (n = 185) Group 2: patients who had undergone subtotal oesophagectomy and died by the completion of 12 months (n = 99) | Mortality at 12 months | Preoperative serum albumin levels in Group 1 were higher compared with Group 2 (3.35 ± 0.49 vs 2.99 ± 0.51, P <0.05) | Preoperative low serum albumin levels have a strong predictive relation to mortality on patients who undergo oesophagectomy for oesophageal cancer |
| Marin et al. (2010), Arq Gastroenterol, Brazil [8] Retrospective single-centre cohort study (level 3) | 100 patients undergoing oesophagectomy for malignant disease Group 1: major surgery (oesphagectomy with gastroplasty by transhiatal or transthoracic route, n = 25) Group 2: minor/palliative surgery (gastrostomy or jejunostomy, n = 75) | Postoperative complications: pleuropulmonary of infectious/non-infectious origins, cardiovascular, surgical site infection, sepsis, anastomosis dehiscence, stenosis and tumour recurrence | Lower frequency of hypoalbuminaemia (>3.5 g/dl) was associated with lower weight loss and less postoperative complications (P = 0.041) Surgical mortality in minor surgery patients was higher in the presence of hypoalbuminaemia, with postoperative survival more than 30 days significantly higher in patients with serum albumin >3.5 g/dl (P = 0.041) | Authors conclude that disease severity (e.g. advanced stage) causes a worse nutritional state, which complicates surgery and is associated with mortality, and thus recommended early diagnosis and nutritional treatment to support surgical action |
| Park et al. (2009), Crit Care, UK [9] Retrospective multicentre review (level 4) | 7227 patients who were admitted to 181 critical care units following elective oesophageal surgery for malignancy | In-hospital mortality | Multiple logistic regression showed that percentage of in-hospital mortality was associated with low serum albumin (<15.0 vs >35.0 g/l, 21.0 vs 11.3%, P <0.001) Preoperative albumin levels did not predict postoperative levels or increased risk of death | Postoperative serum albumin was confirmed as an additional prognostic factor regarding the risk of death. More work is needed to determine how this may improve clinical management |
| Ryan et al. (2007), J Gastrointest Surg, USA [10] Retrospective single-centre review (level 4) | 200 patients undergoing oesophagectomy (thoracotomy with three- or two-stage exploration) and two-field lymphadenectomy for malignant disease | Morbidity and mortality | Patients with albumin <20 g/l were more likely to develop complications than those with albumin >20 g/l (54 vs 28%, respectively, P <0.011) Patients with albumin <20 g/l had higher rates of ARDS (22 vs 5%, P <0.001), respiratory failure (27 vs 8%, P <0.01), and were more likely to die in the hospital (27 vs 6%, P <0.001) than those with albumin >20 g/l Multivariate logistic regression analysis showed that day 1 albumin level was independently related to postoperative complications [odds ratio (OR) = 0.85; 95% confidence interval (CI) 0.85—0.77; P <0.001] | Authors suggest that albumin less than 20 g/l on the first postoperative day may identify a cohort postoperatively that should continue to be monitored in the high dependency unit (HDU) or the intensive care unit (ICU) |
| Tomimaru et al. (2006), J Surg Oncol, Japan [11] Retrospective single-centre cohort study (level 3) | 762 patients who underwent oesophagectomy (right thoracotomy and three- or two-field lymph node dissection) Group 1: salvage oesophagectomy after definitive chemoradiation (n = 24) Group 2: oesophagectomy after neoadjuvant chemoradiation (n = 26) | Morbidity and mortality | The serum albumin and lymphocyte count in the salvage group were significantly lower than those in the neoadjuvant group Univariate analysis performed for age (<60/<60), clinical T (T1–2/T3–4), reason for salvage surgery, curability, preoperative serum albumin (<4.0 /<4.0 g/dl) and lymphocyte did not identify preoperative serum albumin as an independent prognostic factor The lower preoperative serum albumin level in the salvage group may be due to a more malnourished and immunosuppressed condition than those in the neoadjuvant group | Because salvage oesophagectomy is performed on immunocompromised hosts, it therefore important to accurately assess the preoperative T factor, in addition to serum albumin levels, when deciding on the indication for salvage surgery |
| Kudsk et al. (2003), J Parenter Enteral Nutr, USA [12] Retrospective multicentre cohort study (level 3) | 706 patients undergoing elective surgery on the oesophagus (n = 59), stomach (n = 40) and pancreas (n = 221) | Incidence of complications: pneumonia, fasciitis, anastomosis or wound dehiscence, intra-abdominal abscess, renal or respiratory failure, decubitus ulcer formation or death | With albumin <3.75 g/dl (Groups 1–5), oesophageal surgery resulted in significantly more complications than stomach (P = 0.04) or colon (P = 0.005) Complications increased significantly as albumin levels dropped (Group 1 vs Group 2–7, P <0.02; 1–2 vs 3–7, 1–3 vs 4–7, 1–4 vs 5–7, 1–5 vs 6–7, all P <0.0001), 1–6 vs 7, P = 0.035) | Both preoperative serum albumin and operative site affect the complication rate, which subsequently influenced postoperative stay (POS), ICU stay and nil per oral (NPO) days |
| Mortality | Mortality was low (1–5%) with a preoperative albumin >3.25 g/dl, regardless of operative site. Mortality increased up to 20–30% in the two lowest albumin groups | |||
| Length of postoperative stay (POS) | Complications significantly increased POS in all groups except for the highest and lowest albumin groups (P <0.0001) | |||
| Intensive care unit stay | Mean ICU stay gradually increased with a decreasing preoperative albumin level | |||
| Nil per oral days | Delay to resumption of diet was prolonged in patients surviving oesophagectomy at all albumin levels <4.25 g/dl | |||
| Nakamura et al. (2004), Am J Surg [13], Japan Retrospective single-centre cohort study (level 3) | 55 patients undergoing oesophagectomy for oesophageal malignancy Salvage group: patients who underwent oesophagectomy after definitive chemoradiation (n = 27) Neoadjuvant group: patients who underwent oesophagectomy after neoadjuvant chemoradiation (n = 28) | Mortality, morbidity, need for mechanical ventilation, ICU stay, postoperative hospital stay, anastomotic leakage, wound infection, pleural effusion, residual tumours and pathological effect | No differences in overall postoperative survival/mortality between the salvage and neoadjuvant groups | Low serum albumin level in the salvage group could be related to the long duration of disease |
| Rentz et al. (2003), J Thorac Cardiovasc Surg, USA [14] Prospective multicentre cohort study (level 2) | 945 patients undergoing oesophagectomy. Group 1: transthoracic oesophagectomy (n = 562) Group 2: transhiatal oesophagectomy (n = 383) | Serious morbidity: pneumonia, unplanned intubation, respiratory failure, pulmonary oedema, cardiac arrest requiring cardiopulmonary resuscitation, myocardial infarction, wound dehiscence, coma for >24 h, acute renal failure, sepsis, perioperative bleeding >4 units and graft or prosthesis failure | Risk stratification model applied to both groups showed an association between serum albumin <3.5 g/dl with postoperative mortality (β coefficient: 0.56, P = 0.0135) and morbidity (β coefficient: 0.37, P = 0.0102) | Although no differences in postoperative mortality and morbidity were found between either techniques, identifiable risk factors, such as serum albumin levels, that might reduce mortality and morbidity may assist in the selection of the procedure best suited for the patient |
| Bailey et al. (2003), Ann Thorac Surg [15], USA Prospective multicentre outcomes cohort study (level 2) | 1777 patients who had undergone oesophagectomy for malignancy (n = 1509) and benign disease (n = 268) | Mortality and morbidity | Lower preoperative albumin level was an independent risk factor for overall morbidity on multivariable analysis (β coefficient = −0.21, SE = 0.09, P = 0.02, OR = 0.81) | Use of preoperative serum albumin levels, in addition to other independent risk factors, can be used to better stratify patients before oesophagectomy |
| Takagi et al. (2001), Nutrition, Japan [16] Retrospective single-centre cohort study (level 3) | 103 patients undergoing oesophagectomy for thoracic oesophageal cancer via thoracotomy and three-field lymphadenectomy Patients with complications, n = 27 (6 received preoperative chemoradiation) Patients without complications, n = 76 (2 received preoperative chemoradiation) | Infectious complications: pneumonia, abscess (neck, mediastinum and abdomen), enterocolitis, sepsis and viral infection | At 21 days postoperatively, serum albumin level was significantly higher in patients without postoperative infectious complications compared with those who had (3.55 ± 0.08 vs 3.20 ± 0.14 g/dl, P <0.05) | Although postoperative serum albumin levels may predict infectious complications, to prevent infectious complications, preoperative treatment that may adversely affect the host’s immune system, such as heavy chemoradiation, should be avoided |
| Nishi et al. (1988), Ann Surg, Japan [17] Retrospective single-centre cohort study (level 3) | 364 patients who underwent oesophagectomy, oesophagogastrectomy and/or partial gastrectomy 62 patients admitted for cholecystectomy were used as controls | Anastomotic leak and respiratory complications | There were significantly lower serum albumin levels in elderly patients (>70 years old) compared with controls An extremely high correlation was seen between serum albumin levels and total lymphocyte count in patients with oesophageal cancer (r = 0.56, P <0.001) | Although serum albumin levels did not correlate with the study outcomes, improvement in the nutritional status of patients with cancer can reduce operative morbidity and mortality rates. The authors suggested that cancer patients required between 40 and 50 kcal/kg/day with simultaneous administration of 150–200 non-protein calories/g nitrogen |
RESULTS
Matsuda et al. [2] examined the between serum fibrinogen and albumin (FA) score with postoperative disease-free survival. Patients with low serum albumin (<4.0 g/dl) and fibrinogen (<350 mg/dl), and thus a higher FA score, had a higher incidence of cancer recurrence following uni- and multivariate analysis.
Yoshida et al. [3] compared postoperative complications and found no significant predictive relationship between low preoperative serum albumin (<40 g/l) and postoperative pulmonary complications.
Noble et al. [4] examined major complications and anastomotic leak in 258 patients undergoing oesophagogastric resection for high-grade dysplasia. They found a significant association between low serum albumin and major complications on postoperative days 3 and 7, and anastomotic leak. Furthermore, when albumin was combined with C-reactive protein (CRP) and white cell count (WCC) to create a NUn score, they found that a score >10 was predictive of an anastomotic leak and death on postoperative day 4.
Poziomyck et al [5] examined mortality in patients who developed postoperative complications following foregut surgery. Lower serum albumin levels were associated with mortality and longer hospital stay.
Shiozaki et al. [6] analysed respiratory complications in patients who underwent oesophagectomy. Serum albumin levels were not significantly different between patients with or without postoperative respiratory complications.
Khan et al. [7] compared postoperative 12-month mortality in patients who underwent subtotal oesophagectomy for malignant disease. Patients who were alive after 12 months had higher preoperative serum albumin levels compared with those who did not.
Marin et al. [8] reviewed postoperative outcomes in patients who underwent major or minor/palliative oesophageal surgery for malignant disease. Patients with higher serum albumin levels (>3.5 g/dl) had less postoperative complications. In a subset of patients undergoing minor surgery (gastrostomy or jejunostomy), lower serum albumin was associated with mortality and 30-day postoperative survival.
Park et al. [9] performed a retrospective multicentre study in 7227 patients who underwent oesophageal surgery for malignancy. They found a positive association between in-hospital mortality and low postoperative serum albumin levels following multiple logistic regression.
Ryan et al. [10] examined morbidity and mortality in patients who underwent oesophagectomy for malignant disease. Patients with lower serum albumin (<20 g/l) on the first postoperative day were more likely to develop respiratory failure or acute respiratory distress syndrome (ARDS), or to die in the hospital. Multiple regression modelling showed that postoperative day 1 albumin level was independently related to postoperative complications.
Tomimaru et al. [11] compared morbidity and mortality between 762 patients who underwent salvage oesophagectomy with or without prior chemoradiation therapy (CRT). The authors showed that while serum albumin and lymphocyte counts were significantly lower in the salvage group, there was no association between preoperative serum albumin and post-oesophagectomy outcomes.
Kudsk et al. [12] performed a study in 706 patients undergoing elective foregut surgery. Patients with lower serum albumin levels had a higher incidence of postoperative complications, mortality, length of intensive care unit stay, postoperative stay and nil per oral days.
Nakamura et al. [13] analysed morbidity and mortality in 55 patients undergoing oesophagectomy with or without prior CRT. They found no statistically significant difference in overall mortality between either groups.
Rentz et al. [14] examined morbidity in 945 patients who underwent transthoracic and transhiatal oesophagectomy. They found an association between low serum albumin (<35 g/dl) and postoperative mortality and morbidity in both surgical types.
Bailey et al. [15] compared mortality and morbidity in a prospective cohort study of 1777 patients who underwent oesophagectomy for malignant and benign disease. They showed that lower preoperative albumin level was an independent risk factor for overall morbidity following multivariate analysis.
Takagi et al. [16] examined the presence of infectious complications in 103 patients who underwent oesophagectomy for thoracic oesophageal cancer. They showed that postoperative day 21 serum albumin level was significantly higher in patients without infectious complications compared with those who had infection.
Nishi et al. [17] analysed anastomotic leak and respiratory complications in 364 patients who underwent oesophagectomy, oesophagogastrectomy, and/or partial gastrectomy, and found that serum albumin levels were not correlated with study outcomes.
CLINICAL BOTTOM LINE
Taken together, several studies reviewed in this paper suggest an association between low serum albumin and postoperative complications. However, there are conflicting opinions regarding the prognostic value of low serum albumin, and thus, the role of nutritional support to optimize these patients preoperatively. One limitation of this study is the narrow focus on serum albumin and postoperative complications. Several confounding factors were encountered in some of these papers. For example, chyle leak would decrease in the number of circulating T-cells and lower serum albumin levels, thus causing postoperative infections that may be mistakenly attributed to lowered serum albumin levels. Other confounding factors such as preoperative chemoradiotherapy, CRP levels and smoking may contribute to postoperative complications. It is important, in clinical practice, to consider the finding of low serum albumin in the context of the patient as well as, disease and surgical factors to provide optimal care for these patients. In addition, further studies are warranted to determine the role of preoperative nutritional support in this group of patients.
Conflict of interest: none declared.
REFERENCES
1, , .
Dunning
J
Prendergast
B
Mackway-Jones
K
Towards evidence-based medicine in cardiothoracic surgery: best BETS
, Interact CardioVasc Thorac Surg
, 2003
, vol. 2
(pg. 405
-9
)2, , , , , , et al.
Matsuda
S
Takeuchi
H
Fukuda
K
Nakamura
R
Takahashi
T
Wada
N
Clinical significance of fibrinogen and albumin score for postoperative recurrence in esophageal cancer patients
, J Clin Oncol
, 2014
32(Suppl 3):Abstr 14
3, , , , , , et al.
Yoshida
N
Watanabe
M
Baba
Y
Iwagami
S
Ishimoto
T
Iwatsuki
M
Risk factors for pulmonary complications after esophagectomy for esophageal cancer
, Surg Today
, 2013
, vol. 44
(pg. 526
-532
)4, , , , , , et al.
Noble
F
Curtis
N
Harris
S
Kelly
JJ
Bailey
IS
Byrne
JP
Risk assessment using a novel score to predict anastomotic leak and major complications after oesophageal resection
, J Gastrointest Surg
, 2012
, vol. 16
(pg. 1083
-95
)5, , , , , .
Poziomyck
AK
Weston
AC
Lameu
EB
Cassol
OS
Coelho
LJ
Moreira
LF
Preoperative nutritional assessment and prognosis in patients with foregut tumors
, Nutr Cancer
, 2012
, vol. 64
(pg. 1174
-81
)6, , , , , , et al.
Shiozaki
A
Fujiwara
H
Okamura
H
Murayama
Y
Komatsu
S
Kuriu
Y
Risk factors for postoperative respiratory complications following esophageal cancer resection
, Oncol Lett
, 2012
, vol. 3
(pg. 907
-12
)7, , .
Khan
N
Bangash
A
Sadiq
M
Prognostic indicators of surgery for esophageal cancer: a 5 year experience
, Saudi J Gastroenterol
, 2010
, vol. 16
(pg. 247
-52
)8, , , .
Marin
FA
Lamonica-Garcia
VC
Henry
MA
Burini
RC
Grade of esophageal cancer and nutritional status impact on postsurgery outcomes
, Arq Gastroenterol
, 2010
, vol. 47
(pg. 348
-53
)9, , , , , , et al.
Park
DP
Welch
CA
Harrison
DA
Palser
TR
Cromwell
DA
Gao
F
Outcomes following oesophagectomy in patients with oesophageal cancer: a secondary analysis of the ICNARC Case Mix Programme Database
, Crit Care
, 2009
, vol. 13
Suppl 2
pg. S1
10, , , , , .
Ryan
AM
Hearty
A
Prichard
RS
Cunningham
A
Rowley
SP
Reynolds
JV
Association of hypoalbuminemia on the first postoperative day and complications following esophagectomy
, J Gastrointest Surg
, 2007
, vol. 11
(pg. 1355
-60
)11, , , , , , et al.
Tomimaru
Y
Yano
M
Takachi
K
Miyashiro
I
Ishihara
R
Nishiyama
K
Factors affecting the prognosis of patients with esophageal cancer undergoing salvage surgery after definitive chemoradiotherapy
, J Surg Oncol
, 2006
, vol. 93
(pg. 422
-8
)12, , , , , , et al.
Kudsk
KA
Tolley
EA
DeWitt
RC
Janu
PG
Blackwell
AP
Yeary
S
Preoperative albumin and surgical site identify surgical risk for major postoperative complications
, JPEN J Parenter Enteral Nutr
, 2003
, vol. 27
(pg. 1
-9
)13, , , , , , et al.
Nakamura
T
Hayashi
K
Ota
M
Eguchi
R
Ide
H
Takasaki
K
Salvage esophagectomy after definitive chemotherapy and radiotherapy for advanced esophageal cancer
, Am J Surg
, 2004
, vol. 188
(pg. 261
-6
)14, , , , , , et al.
Rentz
J
Bull
D
Harpole
D
Bailey
S
Neumayer
L
Pappas
T
Transthoracic versus transhiatal esophagectomy: a prospective study of 945 patients
, J Thorac Cardiovasc Surg
, 2003
, vol. 125
(pg. 1114
-20
)15, , , , , , et al.
Bailey
SH
Bull
DA
Harpole
DH
Rentz
JJ
Neumayer
LA
Pappas
TN
Outcomes after esophagectomy: a ten-year prospective cohort
, Ann Thorac Surg
, 2003
, vol. 75
(pg. 217
-22
) discussion 22
16, , , , , .
Takagi
K
Yamamori
H
Morishima
Y
Toyoda
Y
Nakajima
N
Tashiro
T
Preoperative immunosuppression: its relationship with high morbidity and mortality in patients receiving thoracic esophagectomy
, Nutrition
, 2001
, vol. 17
(pg. 13
-7
)17, , , , , , et al.
Nishi
M
Hiramatsu
Y
Hioki
K
Kojima
Y
Sanada
T
Yamanaka
H
Risk factors in relation to postoperative complications in patients undergoing esophagectomy or gastrectomy for cancer
, Ann Surg
, 1988
, vol. 207
(pg. 148
-54
)© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
Issue Section:
BEST EVIDENCE TOPICS
Comments
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved
I read with interest the excellent paper from Goh et al [1]. It comprehensively addressed the question of whether serum albumin acts as a predictor of complications in the postoesophaectomy patient. The authors took into account the important fact that low serum albumin may occur secondary to a number of other possible confounding factors. They also briefly address the fact that little is known about the role of preoperative nutritional support in these patients.
This paper leads to a number of further questions that should be asked. Firstly, is there a difference in the predictive value of albumin if it is low in the pre-operative setting, but restored using nutritional support to a normal level postoperatively, compared to the patient with a normal preoperative albumin which becomes low in the postoperative setting? Secondly, is there an ideal time and rate at which to improve a patient's albumin level in order to reduce the chances of further complications? Finally, is there a difference in outcomes between patients who have their albumin restored using parenteral nutrition vs those patients who have this done using enteral nutrition?
References
[1] Goh SL, De Silva RP, Dhital K, Gett RM. Is low serum albumin associated with postoperative complications in patients undergoing oesophagectomy for oesophageal malignancies? Interact CardioVasc Thorac Surg 2014; doi: 10.1093/icvts/ivu324.
Conflict of interest:
none declared
© The Author 2014. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved
I read with great interest the best evidence topic by Goh et al. [1] on the relationship between preoperative serum albumin levels and postoperative complications in patients undergoing oesophageal resection for oesophageal malignancies. This is a sensitive group in which malnutrition is caused not only by the metabolic perturbations from the neoplastic disease, but also from dysphagia and alimentary tract dysfunction, further aggravating the situation. Indeed, it has been demonstrated that up to 32% of oesophageal cancer patients being considered for oesophagectomy present with severe weight loss, underlining the fact that they are prone to malnutrition [2]. Given the well-established association between malnourishment and poor surgical outcomes, it seems reasonable that patients scheduled for major surgery, such as oesophagectomy, should undergo formal nutritional assessment, and if necessary nutritional support. Among the various indices used for the evaluation of the nutritional status, hypoalbuminaemia defined as serum albumin level < 3.0 g/dL, has been identified as the most valuable predictor of adverse postoperative outcomes, while to date, the only validated assessment method for surgical patients is the NRS-2002, which is based on weight loss, reduced intake and disease severity [3]. It is important to note that nutritional assessment of the surgical patient is not used to "clear" him for surgery, but rather to optimise surgical outcomes in high-risk groups.
When the oesophagectomy candidate is found to be malnourished, the options to proceed with nutritional support include oral supplements, parenteral nutrition (PN) and enteral tube feeding. The potential benefits of the first option are frequently unattainable due to gastrointestinal intolerance, anorexia and non-compliance, while the major drawbacks to preoperative initiation of PN include glycaemic variability, central line complications and the risk of infection. Nevertheless, PN has been found particularly useful for patients undergoing upper gastrointestinal surgery, given that it can be provided for at least 7 days during the preoperative period [3]. A meta-analysis of various nutritional support methods in patients undergoing gastrointestinal surgery demonstrated that enteral nutrition (EN) is advantageous over PN in terms of infectious complications, anastomotic leaks and duration of hospital stay. Moreover, EN is especially beneficial for patients with malnutrition or malignancy [4]. Feeding jejunostomy tubes (JT) inserted percutaneously, endoscopically or laparoscopically can provide the desired route for EN before oesophagectomy or neoadjuvant therapy, although their routine use has been associated with the development of major complications, including reoperation, bowel ischemia and even death [5]. However, in the malnourished patient undergoing oesophageal resection apart from preoperative support, JT will also provide the postoperative nutritional bridge to oral intake.
References
[1] Goh SL, De Silva RP, Dhital K, Gett RM. Is low serum albumin associated with postoperative complications in patients undergoing oesophagectomy for oesophageal malignancies? Interact CardioVasc Thorac Surg 2014; doi: 10.1093/icvts/ivu324.
[2] Bailey SH, Bull DA, Harpole DH, Rentz JJ, Neumayer LA, Pappas TN, et al. Outcomes after esophagectomy: a ten-year prospective cohort. Ann Thorac Surg 2003;75:217-22.
[3] Evans DC, Martindale RG, Kiraly LN, Jones CM. Nutrition optimization prior to surgery. Nutr Clin Pract 2014;29:10-21.
[4] Mazaki T, Ebisawa K. Enteral versus parenteral nutrition after gastrointestinal surgery: a systematic review and meta-analysis of randomized controlled trials in the English literature. J Gastrointest Surg 2008;12:739-55.
[5] Ben-David K, Kim T, Caban AM, Rossidis G, Rodriguez SS, Hochwald SN.Pre-therapy laparoscopic feeding jejunostomy is safe and effective in patients undergoing minimally invasive esophagectomy for cancer. J Gastrointest Surg 2013;17:1352-8.
Conflict of interest: none declared