Ann Thorac Surg 2001;71:766-768
© 2001 The Society of Thoracic Surgeons
a Surgical Metabolism and Nutrition Laboratory, Department of Surgery, State University of New York (SUNY) Upstate Medical University, Syracuse, New York, USA
Address reprint requests to Dr Meguid, State University of New York (SUNY) Upstate Medical University, University Hospital, 750 East Adams St, Syracuse, NY 13210
Two remarkable articles appear in this issue of The Annals of Thoracic Surgery by Jagoe and colleagues [1, 2] that reinforce the well-known fact that preoperative nutritional status affects the outcome of patients undergoing curative lung cancer operation. This is remarkable because, as Jagoe and colleagues references indicate, surgeons and physicians worldwide from a variety of clinical disciplines together with their "statistical pundits" have grappled for several decades with the same two questions: (1) Does a patients preoperative nutritional status have an impact on surgical outcome? (2) What are the preoperative risk indicators?
The evolution of the investigations relevant to these questions and their resultant data proceeded independently but in parallel over the past 120 years . Emotionally, we understand that famine and malnutrition worldwide are causes of disease and death. Intellectually, we understand the deleterious biochemical effects of long-term fasting leading to significant clinical malnutrition, data generated in normal volunteer(s) by Benedict  in 1915 and Keys and colleagues  in 1950. On the human side, no doubt exists concerning the cause-effect relationship between malnutrition, morbidity, and mortality, based on the documentary evidence of Jewish doctors in the Warsaw ghetto during World War II . However, a lingering doubt somehow still appears to exist concerning such a link in our surgical patients, even though every practicing surgeon intuitively knows that operating on a debilitated or malnourished patient can spell disaster and often becomes a rueful and costly experience.
Clinically, Warren  in 1932 noted that more than 20% of cancer deaths were from disease-related malnutrition. In 1936, Studley  documented that in patients operated on for chronic peptic ulcer, if preoperative weight loss was 20% or more the postoperative complications including mortality were 33.5% compared with 3.5% in those who had lost less weight. Malnutrition reduces skeletal muscle bulk and function. This impairs not only gastrointestinal, but more importantly respiratory muscles leading to reduced vital capacity and resting minute ventilation, contributing to bronchopneumonia and to a significantly hypoxic ventilatory response. Cardiac muscle mass and contractility are reduced. A decrease in circulatory volume increases thrombogenicity and impairs renal function leading to electrolyte abnormalities and cardiac arrhythmias. Impaired immune function leads to increased susceptibility to opportunistic infections and slower wound healing.
These points are all borne out to varying degrees by Jagoe and colleagues , and are characterized as respiratory fatigue necessitating reintubation, nonhealing intercostal incisions with persistent drainage, and postoperative readmission to hospital. Anorexia and malnutrition also lead to apathy and depression because deficient protein intake preoperatively and postoperatively adversely affects production of neurotransmitter precursors, leading to impaired emotional/behavioral responses that hamper self-motivation and enhance fatigue. No doubt these symptoms were evident in those patients who remained in the intensive care unit (ICU) for prolonged periods, some of whom ultimately died. Thus, the answer to the first question, based on the wealth of these data , is clearly yes!
How to identify the patient at risk? Over the years, every conceivable protein or nutrition-related index has been used to assess nutritional status of both medical and surgical patients . No single gold standard has emerged; all techniques have limitations. With the help of "statistical pundits," several indices have been combined in an effort to increase sensitivity and specificity. Several predictive regression equations have been used, as evidenced by Jagoe and colleagues , to determine which combination of tests is most useful in identifying the high-risk preoperative surgical patient suffering from a specific pathophysiologic disease process, eg, lung cancer. The field of nutritional support emanated from the application of nutritional assessment techniques (weight, triceps skinfold (TSF), and midarm muscle circumference [MAMC]) borrowed from the public health domain. The techniques of TSF measurement, reflecting fat mass, and MAMC, reflecting muscle mass, are suitably sensitive as nutritional screening indicators when applied to defining gross nutritional characteristics of populations. However, they are overly sensitive and nonspecific when applied to sick hospitalized patients, and thus are not dependable. For many years physicians have used a critical eye when consulting with a patient, which translates into what we now recognize as subjective global assessment (SGA). As used by Jagoe and colleagues , SGA correlates well with other more objective and sophisticated indicators. Nevertheless, controversies still exist as to which are the best preoperative risk indicators .
An ideal nutritional indicator would be one that can be used readily by a surgeon whether in the United Kingdom, the United States, or elsewhere. At first glance the body mass index (BMI) as used by Jagoe and colleagues  seems such an index. However, it is a static measurement and hence its interpretation is questionable. Based on a BMI criterion of 25.0 to 29.9 kg/m2, approximately 20% to 33% of the U.S. population is overweight, whereas another 30% are obese. An increasing BMI is also the trend throughout the industrialized world. Could Jagoe and colleagues mean BMI of 25.1 include some patients who initially had a higher BMI and were catabolic with lung cancer-related body weight loss? In addition, could the reliance on such indices give the falsely low impression of malnutrition in this patient population with lung cancer? An elevated BMI is linked to increased coronary heart disease, hypertension, and diabetes, which increase the risk factors for patients undergoing major thoracic operations. Information concerning comorbidities weigh against a better outcome.
However, the authors also had patients in whom the BMI was less than 18.5 kg/m2. Such patients are clearly malnourished or catabolic and consequently have longer ICU stays, are unable to tolerate full treatment regimens, and experience a 40% to 60% greater frequency of complications in response to surgical/medical treatments. They also take longer to recover from operations, have higher readmission rates, and a twofold to threefold higher death rate than their well-nourished counterparts, thereby increasing costs by an estimated 75%. Complications, whether septic or nonseptic, significantly delay the resumption of oral intake and prolong hospitalization [11, 12]. Thus, a BMI of less than 18.5 kg/m2 spells trouble for the patient, the surgeon, the hospital administrator, and the insurance companys CEO . Alas, the finding of a normal BMI in the presence of an illness is therefore not as dependable a nutritional indicator as one would have wished.
Somewhat more useful preoperative predictable risk factors are those dependent on functional tests such as the inadequate oral nutrient intake period  or handgrip dynamometry . Although more predictive of poorer outcome, these tests are not always readily available compared with the ease of obtaining albumin. Unfortunately, albumin has an undeservedly bad reputation as not being a useful indicator, because of its long half-life and changes secondary to intrinsic hepatic disease. However, the practicing surgeon should be reassured of albumins practical usefulness in chronically ill patients. In 54,215 patients undergoing major noncardiac operations, a serum albumin less than 21 g/L was associated with a morbidity rate of 65% and a mortality of 29% . Albumin levels, it seems, can be accepted once more as a reliable predictor of surgical outcome: it is a relatively low-cost test and should be used more often as a prognostic tool to detect malnutrition and to identify the patient at risk of adverse surgical outcome. Thus, after 40 years of research the answer to the second question is that a constellation of clinical and biochemical indices, including albumin, are useful and informative preoperative tests in identifying the high-risk preoperative patient. Unfortunately, predicting risk still remains an art based on clinical experience, rather than an absolute science.
The two related questions discussed above are closely linked to a third and more controversial issue, one not addressed by the authors. How can outcome be improved? Numerous studies done during the past 30 years show both the beneficial and detrimental effects of preoperative and postoperative nutritional support on reducing surgical complications including death . These seemingly conflicting results arise from a number of assumptions concerning the properties and the use of nutritional support, particularly total parenteral nutrition (TPN). The majority of data upon which we now tend to base our nutritional support practice were generated over a decade ago. At that time the concept of "more is better" generally prevailed. Thus, not only was the predominant caloric component one of high glucose concentration, wherein hypoglycemia inhibits immune function, but nutritional support was given inappropriately to well-nourished patients. Furthermore, most controlled studies focused on the risk-laden TPN and provided too many calories. Repeating such studies today would be cost-prohibitive and would face the ethical dilemma of withholding nutritional support to the control group.
Other studies have shown that nutritional support, either enteral or parenteral, is cost-effective, although expensive, and will reduce the length of ICU stay  and prevent readmission, particularly when extended into the rehabilitation period . Whereas the data by Jagoe and colleagues do not include nutritional support in their patients with complications, it seems inconceivable that this group of patients did not receive nutritional support. Malnutrition, after all, is a problem of physician awareness . Data clearly show that administration of nutritional support enhances survival and diminishes postoperative complications . It is evident that patients who are malnourished according to the criteria of Jagoe and colleagues [1, 2] would have benefited from being given nutritional support before or in conjunction with recovery from lung operation. Thus the answer to the third question is also an unequivocal yes! Particularly if you are the patient.
The appropriate use of nutritional support has other well-recognized features: complications (mechanical, metabolic, and infection) and cost. Unfortunately, these reasons are all too often used as an excuse why it is not instituted. In addition, given the current climate of fiscal constraint, abolishing or diminishing institutionalized nutritional assessment and dedicated professional nutritional support treatment programs, in the name of cost savings, is short-sighted. Furthermore, the failure by the medical profession to recognize nutritional/metabolic support as a medical board subspecialty or as an added certifiable specialty is regrettable. Not only will this lead to future lack of suitably trained professionals but also the lack of career opportunities. This spells the demise of a subspecialty and potentially heralds future patient-related problems with the use of nutritional support.
The field of nutritional support is rapidly moving toward pharmaconutrition. This is the provision of current knowledge-based quantitative and qualitative nutrients to reverse the catabolic biochemical processes in response to stress and malnutrition . With an understanding of the peripheral and central mechanisms controlling food intake during disease, we will be able to better treat anorexia, reduced food intake, weight loss, low BMI, and low fat free mass index, which occur with cancer. In addition, a number of putative mediators are currently being explored, with encouraging results. The use of a branched-chain amino acids supplement given to anorectic cancer patients led to an increase in their food intake before operation . The use of fish oil as a supplement to stabilize weight loss and promote weight gain in patients with pancreatic cancer is currently being tested in multicenter clinical trials based in the United Kingdom. Preliminary data support its efficacy . Ultimately, the understanding of hypothalamic and peripheral mechanisms regulating appetite and weight loss in cancer patients  will enhance surgeons ability to improve their patients metabolic status. This involves giving not only the essential nutrients required to meet the extra energy needs of the disease and the operation, but the manipulation of the neuroimmunochemical responses underlying the onset of complications in patients undergoing thoracic operations.
|HOME||HELP||FEEDBACK||SUBSCRIPTIONS||ARCHIVE||SEARCH||TABLE OF CONTENTS|
|ANN THORAC SURG||ASIAN CARDIOVASC THORAC ANN||EUR J CARDIOTHORAC SURG|
|J THORAC CARDIOVASC SURG||ICVTS||ALL CTSNet JOURNALS|