ARTICLE IN PRESS

Available at www.sciencedirect.com

Clinical Nutrition (2007) 26, 440–443

0261-5614/$ - see frdoi:10.1016/j.clnu.

�Corresponding aMedicine, Lister HoTel.: +44 1438 78164

E-mail address:

http://intl.elsevierhealth.com/journals/clnu

ORIGINAL ARTICLE

The incidence and impact of low body mass index onpatients with operable lung cancer

Thida Wina,�, Andrew J. Ritchieb, Francis C. Wellsb, Clare M. Larochea

aThoracic Oncology Unit, Papworth Hospital, Papworth, Cambridge, UKbDepartment of Cardiothoracic Surgery, Papworth Hospital, Papworth, Cambridge, UK

Received 22 March 2006; accepted 25 January 2007

KEYWORDSBMI;Operable lung cancer;Postoperative compli-cation

ont matter & 20072007.01.009

uthor. Departmentspital, Stevenage S9; fax: +44 1438 78

thida.win@nhs.uk

SummaryBackground & aims: To prospectively assess the nutritional status of patients referred forlung cancer surgery, as well as to assess the prognostic value of nutritional status indetermining the surgical outcome.Methods: One hundred and forty-six patients with potentially operable lung cancer wererecruited. Loss of appetite and weight loss were recorded. All patients had serum albuminlevels and body mass index (BMI) measured. Surgical outcome were noted.Results: Mean age was 69 (range 42–85) years; 29/146 were not referred for surgery. Eightpatients underwent failed thoracotomy. In the remaining 109 patients, mean BMI was 26.Seven patients had BMI of 19 or less. Forty-four patients had ideal body weight. Themajority of patients (n ¼ 58) were overweight. Mean serum albumin was 37 g/l and lowerthan 30 g/l in 5 cases. There were 4% postoperative deaths and 32% with poor surgicaloutcome. There was no statistical difference in mean BMI, serum albumin, loss of appetiteor weight loss between the two outcome groups.Conclusion: BMI is usually well preserved in patients with operable lung cancer. There wasno association between low BMI, low serum albumin, loss of appetite or weight loss, andpostoperative death or poor surgical outcome in this study.& 2007 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rightsreserved.

Elsevier Ltd and European Societ

of General and RespiratoryG1 4AB, UK.1199.

(T. Win).

Introduction

Previous epidemiological studies have suggested a signifi-cant inverse relationship between body mass index (BMI) andthe incidence of lung cancer.1–3 However the clinicalrelevance of these findings is unclear. Although manystudies1,2,4–6 have suggested that leanness is a risk for lung

y for Clinical Nutrition and Metabolism. All rights reserved.

ARTICLE IN PRESS

Body mass index and operable lung cancer 441

cancer, Jagoe et al.7 found that severe nutrition depletionwas uncommon in lung cancer patients.

Nutritional assessment may not be given enough attentionin patients undergoing pulmonary surgery as most patientshave smoked, and because both malignancy8 and chronic airway limitation9 predispose to malnutrition. Nutritionalstatus may be important in the management of cancerpatients, as nutritional depletion has previously beenreported as showing an association with poor outcome andincreased risk of complications, particularly postoperativeinfections.10 Some previous studies have confirmed thatspecifically lung cancer patients who have undergonesurgery are often nutritionally depleted, which may predis-pose them to postoperative complications.11,12

Nutritional indices such as serum albumin concentrationmay predict complications after an operation, especiallypulmonary.12 One study13 suggested that impaired nutritionwas an important predictor of death and need for re-ventilation after operation for lung cancer. History ofsignificant weight loss (410% of body weight) was alsofound previously to be a significant predictor of complica-tion in lung cancer surgery.12 As enhanced recovery ofpatients after surgery has become an important focus ofperioperative management, ESPEN issued guidelines onEnteral Nutrition: Surgery.14

We have therefore prospectively assessed the nutritionalstatus of patients referred for lung cancer surgery in amodern population, to establish the prevalence of malnutri-tion, as well as to assess the prognostic value of nutritionalstatus in determining outcome.

Materials and methods

One hundred and forty-six patients with potentially oper-able lung cancer were recruited for the study at theRegional Cardiothoracic Centre, Papworth Hospital, Pap-worth, Cambridge, over a 2-year period. Routine stagingprocedures (PET scan and CT head) and fitness tests (lungfunction test, exercise test and quantitative V/Q scan) wereperformed for each patient. Before surgery, we assess all ourpatients nutritionally as per ESPEN guidelines14 exceptSubjective Global Assessment questionnaire. A nutritionalhistory was taken including loss of appetite and weight. Allpatients were measured for body weight (kg) and height(meter). The BMI was obtained from Quetelet’s index(weight in kilograms divided by the square of height inmeters). Preoperative serum albumin and total proteinlevels were measured for all patients.

The decision to perform lung resection was made by thesurgeon concerned using clinical, radiological and lungfunction criteria. All operations were performed at PapworthHospital by one of the three designated cardio-thoracicsurgeons. The routine anaesthetic procedure included singlelung ventilation using a double lumen endobronchial tube.Standard posterio-lateral thoracotomy was performed for allpatients. At the end of the procedure, two intercostal tubeswere placed in the pleural cavity and attached to underwaterseals. Post-operative physiotherapy was performed usingbreathing exercises and early ambulation.

The duration of hospital stay, 30 days surgical mortalityand complication rate were recorded. A ‘poor surgical

outcome’ was defined as patients with either post-operativehospital death or major complication (myocardial infarc-tion, cardiac failure, respiratory failure, pulmonary emboli,pneumonia, renal failure).

All patients gave written informed consent, and the localEthics Committee approved the study.

The ability of nutritional status (BMI, serum albumin, lossof appetite and weight loss) to predict poor operativeoutcome was assessed using Student’s unpaired t-test.

Results

Mean age was 69 (42–85) years. Of the 146 patients, 29 werenot referred for surgery, either due to upstaging afterfurther investigations (PET scan/CT head) or because ofmedical co-morbidity. Eight patients underwent failedthoracotomy (open and closed thoracotomy). The remaining109 patients made up the study sample. There were 66 maleand 43 female. Sixty-two patients had lobectomy, 37patients had pneumonectomy, 6 patients had bi-lobectomyand 4 patients had wedge resection.

The histology of the resected tumours showed 48squamous cell carcinomas, 37 adenocarcinomas, 15 unspe-cified non-small cell lung cancers, 4 large cell carcinomasand 5 miscellaneous. Post-operative pathological stagingshowed that 68 patients were stage I, 24 were stage II and 17were stage III. There was no postoperative pathologicalstage obtained for the patients undergoing failed thoracot-omy.

Mean BMI was 25.7 (17–38). Seven patients had BMI of 19or less. Forty-four patients had ideal body weight (BMI20–25). However the majority of patients (n ¼ 58) had ahigh BMI of more than 25. The mean serum albumin was 37and lower than 30 g/l in only 5 cases (4.5%). Total proteinwas normal in all cases except 3. Only one patient with lowBMI had low albumin. Twenty-five per cent of patientsreported loss of appetite and 12% had weight loss at the timeof diagnosis. Only 2 patients had weight loss of more than10% of body weight.

According to ESPEN guidelines, 11 (3 with BMIo19, 8 withserum albumin o30 g/l) patients had severe nutritional risk.

There were 4 postoperative deaths in 109 patients (4%),one with BMI of 20, one with 22 and two with 28. Twopatients died of cardiac causes and two died of a respiratorycause. One patient who died at day 23 had a low albuminlevel of 27 g/l but high BMI of 28. All patients who died weremale and had undergone pneumonectomy (1—left,3—right).

Satisfactory outcome was achieved in 81 (74%) patientsand a ‘poor surgical outcome’ in 28 (26%) patients (4 deathsand 24 major complication) although there were no patientswith bronchopleural fistula. The mean BMI for satisfactoryoutcome was 26.4 (SD ¼ 4.5) and for poor outcome was 25.3(SD ¼ 4.1). The mean serum albumin for satisfactory out-come was 37.5 g/l and for poor outcome was 35.8 g/l.Thirteen per cent patients had reported weight loss and 25%patients loss of appetite in the satisfactory outcome group,whilst 9% patients had reported weight loss and 22% patientsloss of appetite in the ‘poor surgical outcome’ group. Fourout of 11 patients (36%) with severe nutritional risk (as perESPEN guidelines) had poor surgical outcome. Student’s

ARTICLE IN PRESS

T. Win et al.442

unpaired t-test showed no statistical significant differencesin mean BMI, serum albumin, loss of appetite and weight lossbetween the ‘poor outcome’ or the ‘satisfactory outcome’group. There was no statistical difference (po0.05)between the group undergoing surgery (mean BMI 25.9)and the patients who were not referred for surgery (meanBMI 25.5).

Discussion

Despite reports of frequent nutritional depletion andsubsequent nutrition-related complications after lung can-cer operations, preoperative nutritional assessment is rarelyperformed. It is well established that nutritional status hasan important influence on postoperative morbidity inpatients undergoing gastrointestinal surgery, oesophagealsurgery and lung cancer surgery. Malnutrition is associatedwith an increased risk of poor tissue healing, impairedimmune function and respiratory muscle dysfunction.Although various measurements have been used in theassessment of nutritional deficiencies15 there is no generalagreement of which is the most useful. Amongst them, theBMI, the serum albumin level, and a history of weight lossand loss of appetite are the most easily available and arefamiliar to most clinicians.

Busch et al.12 reported that weight loss and reducedserum albumin level were the most significant nutritionalfactors associated with pulmonary risk. Seltzer et al.16 andBellantone et al.17 found that a total lymphocyte count ofless than 1500/mm3 was associated with increased compli-cations. They also found that a reduced serum albumin levelof 35 g/l was associated with a 4.6 fold increased rate ofcomplications in surgical patients.

Although several investigators have shown that weight andrelative weight are inversely associated with the risk of lungcancer1,2,4,18 our study found a low incidence of low BMI inoperable lung cancer, consistent with findings previouslyreported by Jagoe et al.7 However, in our cohort, only 7(4.8%) patients had low BMI, compared to the 13.3% reportedby Jagoe et al. Low serum albumin was also rare (4.5%) in ourstudy, which is in agreement with other studies.11,19 Ourresults contradict other previous findings1,2,4–6 and suggestthat the incidence of low BMI in patients with operable lungcancer is low. In fact, the majority of our patients were overrather than under weight.

These findings are most likely due to our patient group,which only involved patients with operable lung cancer.Most, therefore, were in the early stages of lung cancerrather than the late cachectic stage, although the BMI ofpatients who were not referred for surgery due to advancedstage was not significantly different from that in patientswith early (operable) lung cancer. It is possible that recentchanges in the dietary habit of people in the UK might alsopartly explain our results.

Bashir et al.11 found that a low serum albumin level wasassociated with an increased risk of bronchopleural fistula ina small study of 39 patients. He suggested a largerprospective study to examine the relationship betweenpreoperative nutritional state and surgical outcome. How-ever, none of our patients developed a bronchopleuralfistula. Jagoe et al.13 studied 52 patients undergoing surgical

resection of lung cancer and showed that impaired nutritionwas an important predictor of death and the need forreventilation after an operation for lung cancer. However,our prospective study of 109 patients found no associationbetween low BMI or low serum albumin and postoperativedeath or poor surgical outcome. This might be partly due torecent advances in surgical technique, better postoperativecare and/or early postoperative nutritional intervention.

Loss of appetite was reported in about one quarter of ourpatient group although weight loss was found in only around10% of patients. Busch et al.12 found that weight loss andserum albumin level were the most significant nutritionalfactors associated with pulmonary risk. In their multivariateanalysis, serum albumin was the only nutritional factor toremain significant although our study did not confirm theirresult.

Four out of 11 patients with severe nutritional risk (as perESPEN guidelines) had poor surgical outcome. Although 36%poor surgical outcome in this group is higher than 26% ofpoor surgical outcome in whole group, it was difficult to beanalyzed statistically as very different in sample size.

Our study has few limitations. We only have very fewmortality and small number of morbidity in our large cohortof operable lung cancer patients. Although we use BMI,serum albumin level, loss of appetite and weight loss asthese measures are easily available to physicians, thesevariables may not be sufficiently adequate for a carefulpatient assessment except in the presence of extremederangements. In addition, we are not able to compare fullywith other authors as we did not examine nutritional intakeby nutritional risk score20 nor use subjective GlobalAssessment Questionnaire.14

In conclusion, our results suggest that the incidence oflow BMI in patients with operable lung cancer is low. Therewas no association between low BMI, low serum albumin,loss of appetite or weight loss and postoperative death orpoor surgical outcome in our study group. Prospective studyusing Nutritional risk score and ESPN guidelines on EnteralNutrition: Surgery14 would be helpful to determine theassociation between surgical outcome and nutritional statusin lung cancer surgery.

Acknowledgements

Dr. Thida Win received the funding from NHS Eastern regionresearch grant.

Contribution of authors:Design study: Thida Win, Andrew Ritchie, Francis Wells,

Clare LarochePractical performance: Thida Win, Andrew Ritchie,

Francis WellsData analysis: Thida WinPreparation manuscript: Thida WinCritical review: Clare LarocheConflict of interest: None

References

1. Knekt P, Heliovaara M, Rissanen A, et al. Leanness and lung-cancer risk. Int J Cancer 1991;49:208–13.

ARTICLE IN PRESS

Body mass index and operable lung cancer 443

2. Kabat GC, Wynder EL. Body mass index and lung cancer risk. AmJ Epidemiol 1992;135:769–74.

3. Wannamethee G, Shaper AG. Body weight and mortality inmiddle aged British men: impact of smoking. Br Med J1989;299:1497–502.

4. Nomura A, Heilbrun LK, Stemmermann GN. Body mass indexas a predictor of cancer in men. J Natl Cancer Inst 1985;74:319–23.

5. Goodman MT, Wilkens LR. Relation of body size and the risk oflung cancer. Nutr Cancer 1993;20:179–86.

6. Kark JD, Yaari S, Rasooly I, Goldbourt U. Are lean smokers atincreased risk of lung cancer? The Israel Civil Servant CancerStudy. Arch Intern Med 1995;155:2409–16.

7. Jagoe RT, Goodship TH, Gibson GJ. Nutritional status of patientsundergoing lung cancer operations. Ann Thorac Surg2001;71:929–35.

8. Holroyde CP, Reichard Jr GA. General metabolic abnormalitiesin cancer patients: anorexia and cachexia. Surg Clin North Am1986;66:947–56.

9. Wilson DO, Rogers RM, Hoffman RM. Nutrition and chronic lungdisease. Am Rev Respir Dis 1985;132:1347–65.

10. Dempsey DT, Mullen JL, Buzby GP. The link between nutritionalstatus and clinical outcome: can nutritional intervention modifyit? Am J Clin Nutr 1988;47:352–6.

11. Bashir Y, Graham TR, Torrance A, Gibson GJ, Corris PA.Nutritional state of patients with lung cancer undergoingthoracotomy. Thorax 1990;45:183–6.

12. Busch E, Verazin G, Antkowiak JG, Driscoll D, Takita H.Pulmonary complications in patients undergoing thoracotomyfor lung carcinoma. Chest 1994;105:760–6.

13. Jagoe RT, Goodship TH, Gibson GJ. The influence of nutritionalstatus on complications after operations for lung cancer. AnnThorac Surg 2001;71:936–43.

14. Weimann A, Braga M, Harsanyi L, Laviano A, Ljungqvist O, SoetersP. DGEM (German Society for Nutritional Medicine). In: Jauch KW,Kemen M, Hiesmayr JM, Horbach T, Kuse ER, Vestweber KH,editors. ESPEN Guidelines on Enteral Nutrition: Surgery includingorgan transplantation. Clin Nutr 2006; 25: 224–44.

15. Grant JP, Custer PB, Thurlow J. Current techniques ofnutritional assessment. Surg Clin North Am 1981;61:437–63.

16. Seltzer MH, Fletcher HS, Slocum BA, Engler PE. Instantnutritional assessment in the intensive care unit. J ParenterEnteral Nutr 1981;5:70–2.

17. Bellantone R, Doglietto GB, Bossola M, Pacelli F, Negro F,Crucitti F. Preoperative parenteral nutrition in the high risksurgical patient. J Parenter Enteral Nutr 1988;12:195–7.

18. Lew EA, Garfinkel L. Variations in mortality by weight among750,000 men and women. J Chronic Dis 1979;32:563–76.

19. Mullen JL, Gertner MH, Buzby GP, Goodhart GL, Rosato EF.Implications of malnutrition in the surgical patient. Arch Surg1979;114:121–5.

20. Kondrup J, Johansen N, Plum LM, et al. Incidence of nutritionalrisk and causes of inadequate nutritional care in hospitals. ClinNutr 2002;21:461–8.