Research articles

By Dr. Teinye G Douglas
Corresponding Author Dr. Teinye G Douglas
James Paton Memorial Hospital, 125 TransCanada Highway - Canada A1V 1P7
Submitting Author Dr. Teinye G Douglas

BMI and Serum Albumin, Predictors of outcome

Douglas TG. Body Mass Index and Serum Albumin as Predictors of Outcome in Colorectal Cancer Patients. WebmedCentral COLOPROCTOLOGY 2013;4(2):WMC003982
doi: 10.9754/journal.wmc.2013.003982

This is an open-access article distributed under the terms of the Creative Commons Attribution License(CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Submitted on: 01 Feb 2013 12:29:51 AM GMT
Published on: 01 Feb 2013 06:27:46 AM GMT


Background: Colorectal cancer is predominantly a disease of the elderly in whom the incidence of malnutrition has been estimated at about 5-6% between the ages of 70 and 80 and 8-12% over the age of 80. Most patients with colorectal cancer admitted for elective surgery appear well nourished on admission but leave hospital looking under-nourished. We attempt to quantify this in an effort to understand which direction further studies and even clinical intervention might take.

Study design: Case notes of fifty patients who have undergone resection surgery for colorectal cancer at The James Cook University Hospital over a five year period were reviewed and data extracted. The prevalence of malnutrition and the effects of preoperative BMI and serum albumin on length of hospital stay, days off food and complications were studied.

Results: The prevalence of malnutrition (BMI < 20kg/m2) was 6.0%. More patients were obese/overweight (48%) than were malnourished. There was no correlation between BMI on admission and length of hospital stay (P = 0.88).

Conclusion: There were more overweight/obese than undernourished patients on admission. The BMI on admission did not affect the length of hospital stay. The drop in albumin concentration had a negative correlation with the number days off food. Hypoalbuminaemia did not increase the rate of wound infection.


A 40% prevalence of malnutrition (body mass index (BMI) < 20 kg/m2) has been reported amongst hospitalised patients on admission1. About two-thirds of these patients lose weight whilst in hospital, protein calorie malnutrition (PCM) being the most prevalent amongst surgical patients2. Colorectal cancer patients are susceptible to protein calorie malnutrition and may be malnourished at the time of presentation as a result of poor nutrient intake and increased nutrient requirement as part of the cancer cachexia syndrome or develop malnutrition in course of treatment although some may even paradoxically appear overweight.3, 4. Serum albumin and body weight are the most commonly measured indices of nutritional assessment. The body mass index [BMI = Weight (kg)/height (m2)] or percentage weight loss are simple and sensitive clinical measures of gross nutritional status5. The ideal BMI is 20-25 kg/m2. A BMI of less than 20 kg/m2 indicates the possibility of undernutrition while a BMI of less than 18 kg/m2 is suggestive of undernutrition.5 Pre-treatment serum albumin level has been shown to be an independent prognostic factor in a number of malignant diseases6, 7. Reinhardt et al demonstrated a linear correlation between the degree of hypoalbuminaemia and the thirty-day mortality in hospitalised patients. They found a thirty day mortality rate of 24.6% in hypoalbuminaemic (albumin < 3.4gm/dl) patients compared with 1.7% in patients with normal albumin. They showed that a one gram drop in serum albumin level (3.5 to 2.5 gm/dl) translated into a 35% increase in mortality8. Serum albumin level is said to probably be the best indicator of survival in cancer patients9. Traditionally, serum albumin has been used as an index of nutritional assessment10, 11, 12, 13. and is useful if estimated serially14. Serum albumin concentrations are influenced by numerous factors including the state of hydration of the patients, its distribution between extra-cellular and intra-cellular fluid compartments, rate of synthesis and breakdown. Nutrient intake (total calories and specific amino acids) is the primary determinant of the rate of synthesis. The oncotic pressure of the hepatocytes also influence albumin synthesis – a decrease in pressure increasing rate of albumin synthesis. Synthesis is decreased during periods of decreased food intake15. Thirty gram per litre (30gm/litre) is the level below which it is said to indicate nutritional deficiency. However its rate of fall is said to be a determinant of the nature the nature of the illness and its outcome16. The main criticism of albumin as a nutritional marker is its long life of twenty one days. This makes it not a very sensitive indicator of early protein deficiency and therefore said to be inappropriate as a marker of nutritional status in short-term starvation. Nevertheless, it still remains a reliable common biochemical parameter in nutritional assessment11. Hypoalbuminaemia is a poor prognostic sign in cancer patients17. Hermann et al studied serum albumin levels within 48 hours of hospitalisation for acute illness to predict in hospital death, length of hospitalisation and re-admission rate in 15,511 patients older than forty years. Patients with serum albumin less than 34g/l (21% of the study population), were more likely to die (in hospital mortality of 14%), stayed longer in hospital when compared with those with normal serum albumin (in hospital mortality of 4%). They concluded that serum albumin level was independent for age and was a stronger predictor of outcome than age or other laboratory indices18. Surgery in the undernourished results in impairment of wound healing19, 20 increases wound infection rate21, 22, increased incidence of anastomotic dehiscence23 and delays recovery.

Patients and Methods

The aim of this retrospective study was to establish the effect of pre-operative serum albumin and body mass index on post-operative outcome, including morbidity, mortality and hospital stay in colorectal cancer patients who have undergone resection surgery.

Medical notes of fifty patients who had undergone resection colorectal surgery for a five year period from 1994 to 1999 for colorectal cancer with histological confirmation, grading and staging were studied and the data in table 1 were extracted.

Patients operated upon for benign colorectal conditions such as polyps with severe dysplasia, diverticular disease, Crohn’s, ulcerative colitis and patients with intercurrent medical conditions - diabetes mellitus and pre-existing chronic renal failure were excluded from the study as these patients already had impaired metabolism and  patients who had emergency operations were excluded from the study .

Astler and Coller’s modification24 of Dukes’ classification   was routinely used by the pathology department for staging and tumours were graded as well, moderately or poorly differentiated. Patients were grouped into four groups based on their BMI on admission (Table 2).

Definition Of Malnutrition: For the purpose of this study, malnutrition was defined as BMI < 20kg/m2, (1.,5)  or serum albumin concentration < 35gm/l25. BMI >25kg/m2 was defined as overweight/obese.

Major complications were defined as untoward events that threatened life or prolonged hospital stay3.

1. Postoperative septicaemia was defined as body temperature > 37.5°C with a positive blood culture, leucocytosis or leucopenia,
2. Intra-abdominal sepsis as intra-abdominal purulent collection requiring ultrasound or CT guided or open drainage.
3. A fistula was defined as an enterocutaneous or colocutaneous connection confirmed radiologically,
4. pneumonia as chest infection with positive sputum culture with an abnormal chest X-ray requiring antibiotic treatment.
5. Wound infection was defined as purulent wound discharge with positive swab culture and was scored using the ASEPSIS scoring system26
6. Post operative ileus as abnormal bowel function with or without vomiting requiring nasogastric intubation and cessation of oral fluids,
7. Urinary tract infection (UTI) as significant bacteriuria (105),
8. Pseudomembranous colitis as post-operative diarrhoea with isolation of Clostridium diffcille with or without endoscopic confirmation,
9. Sinus as a connection of skin or any epithelial surface to a deep seated abscess and
10. Post operative mortality as death that occurs following operation in patients who have been hospitalised continually or discharged home since operation until thirty days.

Statistical analysis: The Statistical Package for Social Science (SPSS,Chicago,Illinois,USA) software was used for data analysis. Normally distributed (parametric) data such as albumin concentration were calculated as mean (SD, standard deviation) and 95% confidence interval of the mean (CI) and analysed using variance (ANOVA). Continuous non-parametric data such as days in hospital were calculated as medians and analysed with Kruskal Wallis test. T-test and Paired T-test were used to test difference in means were appropriate. The effects of hypoalbuminaemia on length of hospital stay were analysed with Mann-Whitney test and Fisher’s test was used to test effects of hypoalbuminaemia on wound infection. Wound infection scores were analysed with Chi square test. T – test and paired T- tests were used to test differences in mean were appropriate. Statistical difference was set at P < 0.05.


There were 33 males and 17 females (male: female ratio of 1.9: 1) with a mean age of 68.4±9.8 years (range 40-88 years).

BMI At Presentation: The mean BMI at presentation was 25.5(± 4.1, range 17.3-35.4)kg/m2.   (Table 2) See Illustration 2

Albumin On Admission: The mean albumin on admission and discharge were 42.8gm/l (SD 4.9, range 30-52gm/dl) and 35.1(SD5.3, range 21-44) gm/l respectively. Four patients (8%) presented with albumin of <35gm/l) (Two with 33 and one each with 30 and 32gm/l). All four patients had normal body mass index at presentation. Twenty four (48%) patients were discharged with serum albumin concentration of < 35g/l. The mean drop in albumin concentration between admission and discharge was 7.7 (± 4.2, range 0.0 to 17.0) gm/l. The drop had a significant negative correlation with the number of days off food after surgery (P = 0.04). The longer the length of hospital stay and length of postoperative starvation, the larger the drop in albumin concentration.

Days In Hospital and Days Off Food:

See Illustration 3

The BMI on admission did not affect the length of hospital stay (P=0.88) or the median days off food after surgery (P=0.69)

Complications: 11 of 24 patients (45.8%) with BMI of 20-25 kg/m2 developed a total of 27 complications, 6 of 17 patients (35%) of those with BMI of 25-30kg/m2 developed a total of 10 complications while 4 of 6 patients (66.7%) of patients with BMI > 30 developed 7 complications. 1 of 3 patients with BMI < 20kg/m2 developed a total of 2 complications. Two patients presented with normal serum albumin and BMI < 20kg/m2. One patient developed pseudomembranous colitis (antibiotic associated colitis) while the second patient developed no complications. Two other patients presented who with low serum albumin but BMI > 30kg/m2. Both developed wound infection, wound dehiscence and oedema post-operatively (Table 4).

Deaths: There were two deaths (4%) one in the nourished group (BMI 20-25kg/m2) and the other in obese group (BMI > 30kg/m2). One of the deaths was an 82 year old male with a BMI of 23.1 kg/m2 with a pre-admission albumin of 44gm/dl who developed faecal peritonitis from anastomotic leak and pneumonia. Re-laparatomy showed ischaemic segment of bowel at the anastomotic site. He was fed parenterally on the 13th post-operative day and died four days later in the Intensive Care Unit. The second death was a 72-year- old female who presented with a BMI of 33.5kg/m2 and a serum albumin of 30gm/dl. She developed wound dehiscence, vomiting, sepsis, wound infection, hypoproteinaemia, hypoalbuminaemia and oedema and died ten days post surgery.


Forty six percent (46%) of patients in this study presented with BMI of >25+ kg/m2 (either over weight or obese) and 6 % were under-nourished with BMI < 20kg/m2. There were more overweight/obese patients than malnourished patients.

Hickman et al3 in their series from 1958 to 1978 also found more obese/overweight patients with colorectal cancer (34%) than patients with low weight (22%) at the time of admission. Malnutrition and obesity pose problem in the post-operative period. The over weight and obese are particularly at risk if oral intake is delayed as their nutritional requirements tend to be overlooked because of their weight. They tend to burn off their muscle bulk but still appear fat and overweight27. In this study, 33.3% of patients with BMI < 20kg/m and 66.7% of patients with BMI >30kg/m developed postoperative complications. At the time of this study, patients were not routinely weighed on admission and whilst in hospital and as such inferences could not be made on weight changes or BMI changes between admission and discharge.

The prevalence of malnutrition by albumin concentration of < 35g/l was 8% on admission and 48% on discharge. Bistrian et al28 found a 44% incidence of malnutrition based on serum albumin of < 35g/l amongst 251 medical patients with varying diagnosis which included alcoholism, gastrointestinal, liver, metabolic and infectious diseases. A Scandinavian study found a 5% prevalence of protein energy malnutrition using parameters including BMI, hypoalbuminaemia in 1206 randomly selected adults29. Mullen et al30 however found 16% prevalence of malnutrition based on hypoalbuminaemia. Patients in all these studies had various diagnosis which included oesophageal and gastric cancers and had varying degrees of eating problems unlike the patients in this study none of whom had any eating problems and all had the same diagnosis.

The drop in serum albumin concentration was 7.4 ± 4.2gm/l and had a significant negative correlation with the number of days off food after surgery (P=0.04). Silkman et al31 in a randomised study of patients, who had had colonic surgery, showed that infusion of nutrient containing supplementary glucose resulted in an impressive and statistically significant increase in the rate of albumin synthesis. It has been shown that during starvation muscle proteins are broken down for gluconeogenesis and for the albumin synthesis. Albumin is normally manufactured in the liver mainly from exogenous precursors and the rate of synthesis is dependent on nutrient intake. Fasting, specifically omitting protein from the diet has also been shown to reduce albumin synthesis32. Prolonged post-operative starvation decreases albumin synthesis and nutritional supplementation in the peri-operative period is therefore crucial. Albumin concentration is however also affected by its redistribution in the body’s compartments.

One patient in each study had generalised oedema due to severe hypoalbuminaemia. Muscle protein breakdown for gluconeogenesis and reduced food intake are the main causes of hypoalbuminaemia in the cancer patient. Hypoalbuminaemia also did not influence the incidence of wound infection. Kudsk K A et al33 found a low incidence of complications for patients with colorectal cancer with the same level of albumin when compared with oesophageal or pancreatic cancer (29% versus 100%). They also found that there was a low incidence of mortality if the serum albumin levels were above 32gm/l but as the serum albumin fell below 32gm/l, the incidence of mortality progressively increased regardless of the site of the surgery.

Serum albumin levels correlate well with dietary protein intake30, therefore, delayed exogenous protein intake could explain the fall in albumin levels although it could be argued that the levels could have been affected by the state of hydration of these patients. However, all fifty patients in this study were well hydrated and had normal renal and liver functions. The change in albumin in this study correlated with the level of albumin on admission-the higher the level on admission the larger the difference on discharge. The patients with lower serum albumin on admission conserving enough serum albumin required to maintain plasma oncotic pressure necessary to sustain life

There were no significant differences in the complications in the nourished and malnourished groups. This could be explained by the small sample size. It has recently been shown that weight loss is a basic indicator of surgical risk in modern surgical patients only when it is associated with clinically obvious impairment of organ function34. None of the patients in this study showed any impairment of organ function on admission.

The change in albumin concentration could have been reduced by providing patients very early in the post-operative period with exogenous protein enterally as this reduces the period of starvation and may reduce hospital stay.


The prevalence of malnutrition (BMI < 20kg/m2) was 6.0%. More patients (48%) were obese/overweight (BMI >25kg/m2) than were malnourished. There was no correlation between BMI on admission and length of hospital stay (P = 0.88). The BMI on admission did not affect the length of hospital stay. The drop in albumin concentration had a negative correlation with the number days off food. Hypoalbuminaemia did not increase the rate of wound infection.

In the treatment of colorectal cancer patients, nutritional assessment should form part of the pre-assessment process. This will identify those at risk of post-operative complication even if they look physically fit or overweight. Serious thought should be given to nutritional support in the elderly and if the period of starvation exceeds five days especially in institutions where fast-track recovery is not practised after colorectal resection, as this will minimise serious complications and reduce hospital stay.


I wish to thank Mrs Vicky Whittaker of Teesside University for statistical advice and Prof R.G. Wilson MD, FRCS for giving me the opportunity to perform this study.


1. McWhirter JP, Pennington CR   Incidence and recognition of malnutrition in hospital. BMJ 1994; 308: 945-948
2. Butterworth EF, Blackburn GC Hospital malnutrition and how to assess the nutritional status of a patient. Nutrition Today (March- April) 1976 Teaching Aid No.18
3. Hickman DM, Miller RA, Rombeau JL et al Serum Albumin and Body Weight as Predictors of Postoperative course in Colorectal Cancer.  Journal of Parenteral and Enteral nutrition. 1980; 4 (3): 314-316
4. Rombeau JL, Goldman SL, Apelgren KM et al Protein Calorie Malnutrition in patients with colorectal cancer. Disease of colon & Rectum 1978; 21; 8: 587- 589
5. Drug & Therapeutic Bulletin, Malnourished Inpatients: Overlooked and Undertreated: DTB Vol. 34, No. 8 August 1996
6. Falcorner JJ, et al Acute Phase Protein Response and Survival duration of patients with pancreatic cancer. Cancer 1995; 75: 2077- 2082
7. Tayek JA Nutrition in biochemical aspects of cancer patients in David Heber, George L Blackburn & Vay Liang W. Go. Editors.NutritionalOncologyAcademypress 1999, 519 536
8. Reinhardt FG, Myscofski RW, Wilkins DB et al Incidence and Mortality of Hypoalbuminaemic patients In Hospitalised Veterans. Journal of Parenteral & Enteral. Nutrition. 1980, 4: 357-359
9. Tayek KJ Nutritional and biochemical aspect of the cancer patient in Haber D, Blackburn GL, & Vay Liang W. Go Editors NutritionalOncologyAcademyPress 1999: 519-536
10. James WPT, Hay AM. Albumin metabolism: effect of nutritional state and the dietary protein intake. Journal of Clinical Investigation 1968; 47(9): 1958 – 1972
11. Raman VG. General principles of parental nutritional assessment. In: Lee HA, Raman VG, editors. A Handbook of Parental Nutrition.London: Chapman & Hall; 1990. p. 7 – 51
12. Hermann FR, Safran C, Levkoff SE, Minaker FR. Serum albumin level on admissions predictor of death, length of hospital stay and readmission. Archive of Internal Medicine. 1992; 152: 125 -130.
13. Stack JA, Babineau TI, Bistrian BR. Assessment of nutritional status in clinical practice. Gastroenterologist 1996; 4(suppl. 1): S8 S15.
14. Tweedle D. Fluid and nutritional replacement. In: Burnard KG, Young AE, editors. The New Aird’s Companion in Surgical Studies. 2nd edition,London: Churchill Livingston; 1998. p 37.
15. D’Angio RG. Is there a role for albumin administration in nutritional support? The Annals of Pharmacology 1994; 28: 478 – 482
16. Doweiko JP, Nompleggia DJ. The role of albumin in human physiology and pathophysiology, Part 111: Albumin and disease states. Journal of Parental & Enteral Nutrition. 1991; 15(4): 476 – 483.
17. Margarson MP, Soni N. Serum albumin: Touchstone or totem. Anaesthesia 1998; 53(3): 789 – 803.
18. Hermann FR, Safran C, Levkoff SE, Minaker FR. Serum albumin level on admission as predictor of death, length of hospital stay and readmission. Archive of Internal Medicine. 1992; 152: 125 – 130.
19. Goodson WH 3rd, Lopez-Sarmiento A, Jensen JA, West J, Granja-Mena L, Chavez-Estralla J. The influence of a brief preoperative illness on postoperative healing. Annals of Surgery 1978; 205(3): 250 – 255.
20. Haydock DA, Hill GL Improved wound healing response in surgical patients receiving intravenous nutrition. British Journal of Surgery 1987; 74(4): 320 32321.
21. MacLean LD. Host resistance in surgical patients. Journal of trauma 1979; 19(5): 279 – 304.
22. Lennard TW, Shenton BK, Borzotta A, Donnelly PK, White M, Gerrie LM, Proud G, Taylor RM. The influence of surgical operations on components of human immune system. British Journal of Surgery 1985; 72 (10): 77- 776.
23. Haydock DA, Hill GL. Impaired wound healing in surgical patients with varying degrees of malnutrition. Journal of Parental and Enteral Nutrition 1986; 10(6):550 - 554
24.  Astler VB & Coller FA (1954):  The Prognostic Significance of Direct Extension of Carcinoma of the colon & Rectum. Ann Surg. 1954; 139: 846 – 857.
25.  Hickey MS, Arbeit JM, & Way LW Surgical Metabolism & Nutrition In Way LW Edit. Current Surgical Diagnosis & Treatment 19th Edition , Norwalk, Connecticut, Appleton & Lange 1994: 143 – 173
26. WilsonAP, Treasure T, Sturridge MF, Gruneberg RNA Scoring Method (ASEPSIS) for postoperative wound infections for use in clinical trials of antibiotic prophylaxis. Lancet 1986 Feb 8; 1 (8476): 311 - 3
27. Lennard- Jones JE. A Positive Approach to Nutrition as Treatment: King’s Fund Centre. January 1992: 10 -12.
28. Bistrian BR, Blackburn GL, Vitale J, Cochran D, Naylor J. Prevalence of malnutrition in general medical patients. Journal of American Medical Association 1976; 235 (15): 1567 – 1570.
29. Stefan Thorslund, Goran Toss, Inger Nilsson, Henning v Schenck, Tommy Symreng & Hanns Zetterqvist. Prevalence of Protein Energy Malnutrition in a Large Population of Elderly People at Home. Scandinavian Journal of Primary Health 1990. vol 8, No. 4, pages 243 – 248.
30. Mullen LJ, Buzby, GP, Gertner MH, Stein TP, Hargrove WC, Oram-Smith J, Rosato  EF. Protein synthesis dynamics in human gastrointestinal malignancies. Surgery 1908; 87 (3): 331 - 338
31. Silkman JJ, Rosenoer VM, Young YB, Young VR, Long PC, Munro HN. Energy intake can determine albumin synthesis in man after surgery. Surgery 1985; 97 (3): 271 – 277.
32. Nicholson JP, Wolmarans MR, Park GR. The role of albumin in critical illness. British Journal of Anaesthesia 2000; 85 (4): 599 – 610.
33. Kudsk KA, Tolley EA, DeWitt RC, Janu PG, Blackwell AP, Yearly S, King BK. Pre-operative albumin and surgical site identify surgical risk for major postoperative complications, Journal of Parental and Enteral Nutrition. 2003; 27 (1): 1 -9.
34. Windsor JA, Hill GL. Weight Loss with Physiologic Impairment. A Basic Indicator of Surgical Risk. Ann Surg. 1988 March; 207(3): 290 - 296

Source(s) of Funding


Competing Interests



This article has been downloaded from WebmedCentral. With our unique author driven post publication peer review, contents posted on this web portal do not undergo any prepublication peer or editorial review. It is completely the responsibility of the authors to ensure not only scientific and ethical standards of the manuscript but also its grammatical accuracy. Authors must ensure that they obtain all the necessary permissions before submitting any information that requires obtaining a consent or approval from a third party. Authors should also ensure not to submit any information which they do not have the copyright of or of which they have transferred the copyrights to a third party.
Contents on WebmedCentral are purely for biomedical researchers and scientists. They are not meant to cater to the needs of an individual patient. The web portal or any content(s) therein is neither designed to support, nor replace, the relationship that exists between a patient/site visitor and his/her physician. Your use of the WebmedCentral site and its contents is entirely at your own risk. We do not take any responsibility for any harm that you may suffer or inflict on a third person by following the contents of this website.

Please use this functionality to flag objectionable, inappropriate, inaccurate, and offensive content to WebmedCentral Team and the authors.


Author Comments
1 comment posted so far


What is article Popularity?

Article popularity is calculated by considering the scores: age of the article
Popularity = (P - 1) / (T + 2)^1.5
P : points is the sum of individual scores, which includes article Views, Downloads, Reviews, Comments and their weightage

Scores   Weightage
Views Points X 1
Download Points X 2
Comment Points X 5
Review Points X 10
Points= sum(Views Points + Download Points + Comment Points + Review Points)
T : time since submission in hours.
P is subtracted by 1 to negate submitter's vote.
Age factor is (time since submission in hours plus two) to the power of 1.5.factor.

How Article Quality Works?

For each article Authors/Readers, Reviewers and WMC Editors can review/rate the articles. These ratings are used to determine Feedback Scores.

In most cases, article receive ratings in the range of 0 to 10. We calculate average of all the ratings and consider it as article quality.

Quality=Average(Authors/Readers Ratings + Reviewers Ratings + WMC Editor Ratings)